Number 1 Manual of Molecular and Clinical Laboratory Immunology (PDFDrive)

EIGHTH EDITION
MANUAL OF
MOLECULAR
AND CLINICAL
LABORATORY
IMMUNOLOGY
E DI TOR S
Barbara Detrick
John L. Schmitz
Robert G. Hamilton
MANUAL OF
MOLECULAR
AND CLINICAL
LABORATORY
IMMUNOLOGY
EIGHTH EDITION
EIGHTH EDITION
MANUAL OF
MOLECULAR
AND CLINICAL
LABORATORY
IMMUNOLOGY
EDITED BY
BARBARA DETRICK
Johns Hopkins University, School of Medicine, Baltimore, Maryland
JOHN L. SCHMITZ
University of North Carolina, School of Medicine, Chapel Hill, North Carolina
ROBERT G. HAMILTON
Johns Hopkins University, School of Medicine, Baltimore, Maryland
WASHINGTON, DC
Copyright © 1976, 1980, 1986, 1992, 1997, 2002, 2006, 2016 by ASM Press. ASM Press is a
registered trademark of the American Society for Microbiology. All rights reserved. No part of
this publication may be reproduced or transmitted in whole or in part or reutilized in any form
or by any means, electronic or mechanical, including photocopying and recording, or by any
information storage and retrieval system, without permission in writing from the publisher.
Disclaimer: To the best of the publisher’s knowledge, this publication provides information
concerning the subject matter covered that is accurate as of the date of publication. The
publisher is not providing legal, medical, or other professional services. Any reference herein
to any specific commercial products, procedures, or services by trade name, trademark, manu-
facturer, or otherwise does not constitute or imply endorsement, recommendation, or favored
status by the American Society for Microbiology (ASM). The views and opinions of the
author(s) expressed in this publication do not necessarily state or reflect those of ASM, and
they shall not be used to advertise or endorse any product.
Library of Congress Cataloging-in-Publication Data
Names: Detrick, Barbara, editor. | Schmitz, John L. (John Leo), editor. |

Hamilton, Robert G., editor.
Title: Manual of molecular and clinical laboratory immunology / edited by
Barbara Detrick, John L. Schmitz, and Robert G. Hamilton.
Description: 8th edition. | Washington, DC : ASM Press, [2016] | ?2016
Identifiers: LCCN 2016012270 (print) | LCCN 2016014573 (ebook) | ISBN
9781555818715 | ISBN 9781555818722 ()
Subjects: LCSH: Immunodiagnosis—Handbooks, manuals, etc. |
Immunology—Handbooks, manuals, etc. | Molecular immunology—Handbooks,
manuals, etc.
Classification: LCC RB46.5 .M36 2016 (print) | LCC RB46.5 (ebook) | DDC
616.07/9—dc23
LC record available at http://lccn.loc.gov/2016012270
doi:10.1128/9781555818722
Printed in Canada
10 9 8 7 6 5 4 3 2 1
Address editorial correspondence to: ASM Press, 1752 N St., N.W.,

Washington, DC 20036-2904, USA.
Send orders to: ASM Press, P.O. Box 605, Herndon, VA 20172, USA.
Phone: 800-546-2416; 703-661-1593. Fax: 703-661-1501.
E-mail: books@asmusa.org
Online: http://estore.asm.org
Contents
Editorial Board / xi 6 Immunoglobulin Genes / 51

THOMAS J. KIPPS, EMANUELA M. GHIA, AND
Contributors / xiii LAURA Z. RASSENTI
Foreword: How It Began / xxiii 7 Immunoglobulin Quantification and Viscosity
Measurement / 65
Preface / xxv
JEFFREY S. WARREN
Author and Editor Conflicts of Interest / xxvii 8 Clinical Indications and Applications of Serum
and Urine Protein Electrophoresis / 74
A
DAVID F. KEREN AND RICHARD L. HUMPHREY
9 Immunochemical Characterization of
s ec tion Immunoglobulins in Serum, Urine, and
Cerebrospinal Fluid / 89
General Methods / 1 ELIZABETH SYKES AND YVONNE POSEY
VOLUME EDITOR: ROBERT G. HAMILTON 10 Cryoglobulins, Cryofibrinogenemia, and
SECTION EDITOR: THOMAS A. FLEISHER Pyroglobulins / 101
PETER D. GOREVIC AND DENNIS GALANAKIS
1 Introduction / 3
THOMAS A. FLEISHER 11 Strategy for Detecting and Following
Monoclonal Gammopathies / 112
2 Molecular Methods for Diagnosis of Genetic JERRY A. KATZMANN AND DAVID F. KEREN
Diseases Involving the Immune System / 5
AMY P. HSU
3 The Human Microbiome and Clinical

Immunology / 19
FREDERIC D. BUSHMAN
s ec tion C
4 Protein Analysis in the Clinical Immunology Complement / 125
Laboratory / 26
VOLUME EDITOR: ROBERT G. HAMILTON
ROSHINI SARAH ABRAHAM AND
DAVID R. BARNIDGE SECTION EDITOR: PATRICIA C. GICLAS
12 Introduction / 127
B
PATRICIA C. GICLAS
13 The Classical Pathway of Complement / 129
s ec tion PATRICIA C. GICLAS
14 Analysis of Activity of Mannan-Binding Lectin,
Immunoglobulin an Initiator of the Lectin Pathway of the
Methods =/ 47 Complement System / 133
STEFFEN THIEL
15 The Nature of the Diseases That Arise from
SECTION EDITOR: DAVID F. KEREN Improper Regulation of the Alternative Pathway
5 Introduction / 49 of Complement / 138
DAVID F. KEREN RICHARD J. H. SMITH
v
vi ■ Contents
s ec tion D s ec tion E
Flow Cytometry / 145
Functional Cellular
VOLUME EDITOR: JOHN L. SCHMITZ
SECTION EDITOR: MAURICE R. G. O’GORMAN Assays / 259
16 Introduction / 147 VOLUME EDITOR: BARBARA DETRICK
MAURICE R. G. O’GORMAN SECTION EDITOR: STEVEN D. DOUGLAS
17 Polychromatic Flow Cytometry / 149 26 Introduction / 261
ANGÉLIQUE BIANCOTTO AND STEVEN D. DOUGLAS
J. PHILIP McCOY, JR.
27 Cryopreservation of Peripheral Blood
18 High-Sensitivity Detection of Red and Mononuclear Cells / 263
White Blood Cells in Paroxysmal Nocturnal ADRIANA WEINBERG
Hemoglobinuria by Multiparameter Flow
Cytometry / 168 28 Lymphocyte Activation / 269
ANDREA ILLINGWORTH, MICHAEL KEENEY, ROSHINI SARAH ABRAHAM
AND D. ROBERT SUTHERLAND 29 Functional Assays for B Cells and
19 Standardized Flow Cytometry Assays for Antibodies / 280
Enumerating CD34+ Hematopoietic Stem MOON H. NAHM AND ROBERT L. BURTON
Cells / 182 30 Methods for Detection of Antigen-Specific
D. ROBERT SUTHERLAND AND T Cells by Enzyme-Linked Immunospot Assay
MICHAEL KEENEY (ELISPOT) / 290
20 Functional Flow Cytometry-Based Assays BARBARA L. SHACKLETT AND
of Myeloid and Lymphoid Functions for DOUGLAS F. NIXON
the Diagnostic Screening of Primary 31 Regulatory T Cell (Treg) Assays: Repertoire,
Immunodeficiency Diseases / 199 Functions, and Clinical Importance of Human
MAURICE R. G. O’GORMAN Treg / 296
21 Acute Lymphoblastic Leukemia/Lymphoma: THERESA L. WHITESIDE
Diagnosis and Minimal Residual Disease 32 Measurement of NK Cell Phenotype and Activity
Detection by Flow Cytometric in Humans / 300
Immunophenotyping / 207 SAMUEL C. C. CHIANG AND
JOSEPH A. DiGIUSEPPE YENAN T. BRYCESON
22 Acute Myeloid Leukemia: Diagnosis and 33 Functional Assays for the Diagnosis of Chronic
Minimal Residual Disease Detection by Flow Granulomatous Disease / 310
Cytometry / 217 DEBRA LONG PRIEL AND DOUGLAS B. KUHNS
BRENT WOOD AND LORI SOMA
F
23 Chronic Lymphocytic Leukemia, the Prototypic
Chronic Leukemia for Flow Cytometric
Analysis / 226 s ec tion
HEBA DEGHEIDY, DALIA A. A. SALEM,
CONSTANCE M. YUAN, AND Cytokines and
MARYALICE STETLER-STEVENSON
Chemokines / 321
24 Plasma Cell Disorders / 235
JUAN FLORES-MONTERO, LUZALBA SANOJA, VOLUME EDITOR: BARBARA DETRICK
JOSÉ JUAN PÉREZ, FANNY POJERO, SECTION EDITOR: JOHN J. HOOKS
NOEMÍ PUIG, MARÍA BELÉN VIDRIALES, AND
ALBERTO ORFAO 34 Introduction / 323
JOHN J. HOOKS
25 Future Cytometric Technologies and
Applications / 251 35 Multiplex Cytokine Assays / 324
HOLDEN T. MAECKER ELIZABETH R. DUFFY AND DANIEL G. REMICK
36 Cytokine Measurement by Flow
Cytometry / 338
HOLDEN T. MAECKER
37 Chemokine and Chemokine Receptor
Analysis / 343
SABINA A. ISLAM, BENJAMIN D. MEDOFF, AND
Contents ■ vii
ANDREW D. LUSTER 48 Immunological Tests in Tuberculosis / 433

38 Cytokines: Diagnostic and Clinical CHRISTINE M. LITWIN
Applications / 357 49 Mycoplasma: Immunologic and Molecular
PRIYANKA VASHISHT AND TIMOTHY B. NIEWOLD Diagnostic Methods / 444
39 Detection of Anticytokine Autoantibodies and KEN B. WAITES, MARY B. BROWN, AND
Clinical Applications / 365 JERRY W. SIMECKA
SARAH K. BROWNE 50 Chlamydia and Chlamydophila
Infections / 453
G
ROSEMARY SHE
51 The Rickettsiaceae, Anaplasmataceae, and
s ec tion Coxiellaceae / 461
LUCAS S. BLANTON AND DAVID H. WALKER
Immunohistology and 52 The Bartonellaceae, Brucellaceae, and
Immunopathology / 373 Francisellaceae / 473
CHRISTINE M. LITWIN, BURT ANDERSON,
VOLUME EDITOR: ROBERT G. HAMILTON RENEE TSOLIS, AND AMY RASLEY
SECTION EDITOR: R. NEAL SMITH
I
ROBERT G. HAMILTON
41 Immunofluorescence Methods in the Diagnosis s ec tion
of Renal and Cardiac Diseases / 376
A. BERNARD COLLINS, JAMES R. STONE, AND Mycotic and Parasitic
R. NEAL SMITH Diseases / 483
42 Western Blot Analysis for the Detection
of Anti-Glomerular Basement Membrane
Antibodies and Anti-Phospholipase A2 Receptor SECTION EDITOR: THOMAS B. NUTMAN
Antibodies / 385 53 Introduction / 485
A. BERNARD COLLINS AND R. NEAL SMITH THOMAS B. NUTMAN
54 Immunological and Molecular Approaches for
s ec tion H
the Diagnosis of Parasitic Infections / 486
PATRICIA P. WILKINS AND THOMAS B. NUTMAN
55 Serological and Molecular Diagnosis of Fungal
Infectious Diseases Caused Infections / 503
MARK D. LINDSLEY
by Bacteria, Mycoplasmas,
Chlamydiae, and
Rickettsiae / 391
VOLUME EDITOR: JOHN L. SCHMITZ
s ec tion J
SECTION EDITOR: CHRISTINE M. LITWIN Viral Diseases / 535
43 Introduction / 393 VOLUME EDITOR: JOHN L. SCHMITZ
CHRISTINE M. LITWIN SECTION EDITORS: RICHARD L. HODINKA AND
44 Diagnostic Methods for Group A Streptococcal JOHN L. SCHMITZ
Infections / 394 56 Introduction / 537
CHRISTINE M. LITWIN, SHELDON E. LITWIN, JOHN L. SCHMITZ
AND HARRY R. HILL
57 Immunologic and Molecular Methods for Viral
45 Diagnosis of Helicobacter pylori Infection and Diagnosis / 538
Assessment of Eradication / 404 MARIE LOUISE LANDRY AND YI-WEI TANG
BRUCE E. DUNN AND SUHAS H. PHADNIS
58 Herpes Simplex Virus / 550
46 Laboratory Diagnosis of Syphilis / 412 D. SCOTT SCHMID
JOHN L. SCHMITZ
59 Varicella-Zoster Virus / 556
47 Lyme Disease, Relapsing Fever, and
Leptospirosis / 419 D. SCOTT SCHMID
GUIQING WANG AND 60 Epstein-Barr Virus and Cytomegalovirus / 563
MARIA E. AGUERO-ROSENFELD HENRY H. BALFOUR, JR., KRISTIN A. HOGQUIST,
AND PRIYA S. VERGHESE
viii ■ Contents
61 Human Herpesviruses 6, 7, and 8 / 578 78 Neutropenia and Neutrophil Defects / 765

RICHARD L. HODINKA STEVEN M. HOLLAND
62 Parvovirus B19 / 591 79 Evaluation of Natural Killer (NK) Cell
STANLEY J. NAIDES Defects / 774
63 Respiratory Viruses / 598 KIMBERLY RISMA AND REBECCA MARSH
DAVID J. SPEICHER, MOHSIN ALI, AND
L
MAREK SMIEJA
64 Measles, Mumps, and Rubella Viruses / 610
DIANE S. LELAND AND RYAN F. RELICH s ec tion
65 Viral Hepatitis / 620
HUBERT G. M. NIESTERS,
Allergic Diseases / 781
ANNELIES RIEZEBOS-BRILMAN, AND VOLUME EDITOR: ROBERT G. HAMILTON
CORETTA C. VAN LEER-BUTER
SECTION EDITOR: PAMELA A. GUERRERIO
66 Viral Agents of Gastroenteritis / 639
GABRIEL I. PARRA AND KIM Y. GREEN
PAMELA A. GUERRERIO
67 Arboviruses / 648
81 Quantitation and Standardization of
ROBERT S. LANCIOTTI AND JOHN T. ROEHRIG
Allergens / 784
68 Diagnosis of Hantavirus Infections / 658 RONALD L. RABIN, LYNNSEY RENN, AND
WILLIAM MARCIEL DE SOUZA AND JAY E. SLATER
LUIZ TADEU MORAES FIGUEREIDO
82 Immunological Methods in the Diagnostic
69 Rabies Virus / 665 Allergy Clinical and Research Laboratory / 795
D. CRAIG HOOPER ROBERT G. HAMILTON
70 Human T-Cell Lymphotropic Virus 83 Assay Methods for Measurement of Mediators
Types 1 and 2 / 674 and Markers of Allergic Inflammation / 801
BREANNA CARUSO, RAYA MASSOUD, AND JOHN T. SCHROEDER, R. STOKES PEEBLES, JR.,
STEVEN JACOBSON AND PAMELA A. GUERRERIO
71 Diagnosis of Prion Diseases / 682 84 Tests for Immunological Reactions to
RICHARD RUBENSTEIN, ROBERT B. PETERSEN, Foods / 815
AND THOMAS WISNIEWSKI CARAH B. SANTOS, DAVID M. FLEISCHER, AND
72 Principles and Procedures of Human ROBERT A. WOOD
Immunodeficiency Virus Diagnosis / 696 85 Diagnosis of Rare Eosinophilic and Mast Cell
KELLY A. CURTIS, JEFFREY A. JOHNSON, AND Disorders / 825
S. MICHELE OWEN CEM AKIN, CALMAN PRUSSIN, AND
AMY D. KLION
K
M
s ec tion
Immunodeficiency s ec tion
Diseases / 711 Systemic Autoimmune
VOLUME EDITOR: BARBARA DETRICK Diseases / 839
SECTION EDITORS: KATHLEEN E. SULLIVAN AND
HOWARD M. LEDERMAN VOLUME EDITOR: BARBARA DETRICK
73 The Primary Immunodeficiency Diseases / 713 SECTION EDITOR: WESTLEY H. REEVES
HOWARD M. LEDERMAN 86 Introduction / 841
74 Severe Combined Immune Deficiency: Newborn WESTLEY H. REEVES
Screening / 715 87 Antinuclear Antibody Tests / 843
JAMES W. VERBSKY AND JOHN M. ROUTES ALESSANDRA DELLAVANCE,
75 Combined Immunodeficiencies / 721 WILSON DE MELO CRUVINEL,
PAULO LUIZ CARVALHO FRANCESCANTONIO,
CHRISTINE SEROOGY AND MELISSA ELDER
AND LUIS EDUARDO COELHO ANDRADE
76 Antibody Deficiencies / 737 88 Detection of Autoantibodies by Enzyme-Linked
KIMBERLY C. GILMOUR, ANITA CHANDRA, AND Immunosorbent Assay and Bead Assays / 859
D. S. KUMARARATNE
EDWARD K. L. CHAN, RUFUS W. BURLINGAME,
77 Hereditary and Acquired Complement AND MARVIN J. FRITZLER
Deficiencies / 749 89 Immunodiagnosis and Laboratory Assessment of
PATRICIA C. GICLAS
Contents ■ ix
Systemic Lupus Erythematosus / 868 101 Detection of Antimitochondrial

WESTLEY REEVES, SHUHONG HAN, Autoantibodies in Primary Biliary Cholangitis
JOHN MASSINI, AND YI LI and Liver Kidney Microsomal Antibodies in
90 Immunodiagnosis of Autoimmune Autoimmune Hepatitis / 966
Myopathies / 878 PATRICK S. C. LEUNG, MICHAEL P. MANNS,
ROSS L. COPPEL, AND M. ERIC GERSHWIN
MINORU SATOH, ANGELA CERIBELLI,
MICHITO HIRAKATA, AND EDWARD K. L. CHAN 102 Cardiovascular Diseases / 975
91 Immunodiagnosis of Scleroderma / 888 CHERYL L. MAIER, C. LYNNE BUREK,
NOEL R. ROSE, AND AFTAB A. ANSARI
MASATAKA KUWANA
92 Antibody and Biomarker Testing in Rheumatoid 103 Celiac Disease and Inflammatory Bowel
Arthritis / 897 Disease / 983
MELISSA R. SNYDER
ANN DUSKIN CHAUFFE AND
MICHAEL RAYMOND BUBB 104 Autoantibodies Directed against Erythrocytes
93 Antiphospholipid Antibody Syndrome: Clinical in Autoimmune Hemolytic Anemia / 990
Manifestations and Laboratory Diagnosis / 905 R. SUE SHIREY AND KAREN E. KING
MARTINA MURPHY AND NEIL HARRIS 105 Immune Thrombocytopenia / 995
94 Antineutrophil Cytoplasmic Antibodies (ANCA) THOMAS S. KICKLER
and Strategies for Diagnosing ANCA-Associated 106 Monitoring Autoimmune Reactivity within the
Vasculitides / 909 Retina / 998
R. W. BURLINGAME, C. E. BUCHNER, J. G. HANLY, JOHN J. HOOKS, CHI-CHAO CHAN,
AND N. M. WALSH H. NIDA SEN, ROBERT NUSSENBLATT, AND
95 IgG4-Related Disease: Diagnostic BARBARA DETRICK
Testing by Serology, Flow Cytometry, and
O
Immunohistopathology / 917
JOHN H. STONE
96 Future Perspectives for Rheumatoid Arthritis s ec tion
and Other Autoimmune Diseases / 922 Cancer / 1005
JEREMY SOKOLOVE
N
SECTION EDITORS: DANIEL CHAN AND
LORI J. SOKOLL
s ec tion 107 Introduction / 1007
ROBERT G. HAMILTON
Organ-localized 108 Immunoassay-Based Tumor Marker
Autoimmune Diseases / 927 Measurement: Assays, Applications, and
Algorithms / 1008
VOLUME EDITOR: JOHN L. SCHMITZ ELIZABETH A. GODBEY, LORI J. SOKOLL, AND
SECTION EDITORS: C. LYNNE BUREK AND ALEX J. RAI
PATRIZIO CATUREGLI 109 Malignancies of the Immune System: Use of
97 Introduction / 929 Immunologic and Molecular Tumor Markers in
C. LYNNE BUREK Classification and Diagnostics / 1015
ELAINE S. JAFFE AND MARK RAFFELD
98 Endocrinopathies: Chronic Thyroiditis, Addison
Disease, Pernicious Anemia, Graves’ Disease, 110 Monitoring of Immunologic Therapies / 1036
Diabetes, and Hypophysitis / 930 THERESA L. WHITESIDE
C. LYNNE BUREK, N. R. ROSE, 111 Circulating Tumor Cells as an Analytical
GIUSEPPE BARBESINO, JIAN WANG, Tool in the Management of Patients with
ANDREA K. STECK, GEORGE S. EISENBARTH, Cancer / 1051
LIPING YU, LUDOVICA DE VINCENTIIS, DANIEL C. DANILA, HOWARD I. SCHER, AND
ADRIANA RICCIUTI, ALESSANDRA DE MARTIN FLEISHER
REMIGIS, AND PATRIZIO CATUREGLI
P
99 Myasthenia Gravis / 954
ARNOLD I. LEVINSON AND ROBERT P. LISAK
100 Autoantibodies to Glycolipids in Peripheral s ec tion
Neuropathy / 961
HUGH J. WILLISON Transplantation
Immunology / 1063
VOLUME EDITOR: BARBARA DETRICK
x ■ Contents
SECTION EDITORS: ELAINE F. REED AND THANGAMANI MUTHUKUMAR,

QIUHENG JENNIFER ZHANG CHOLI HARTONO, MINNIE M. SARWAL, AND
112 Histocompatibility and Immunogenetics MANIKKAM SUTHANTHIRAN
Testing in the 21st Century / 1065 119 Killer Cell Immunoglobulin-Like Receptors in
QIUHENG JENNIFER ZHANG AND Clinical Transplantation / 1150
ELAINE F. REED RAJA RAJALINGAM, SARAH COOLEY, AND
113 Molecular Methods for Human Leukocyte JEROEN VAN BERGEN
Antigen Typing: Current Practices and Future 120 Chimerism Testing / 1161
Directions / 1069 LEE ANN BAXTER-LOWE
MARK KUNKEL, JAMIE DUKE,
Q
DEBORAH FERRIOLA, CURT LIND, AND
DIMITRI MONOS
114 Evaluation of the Humoral Response in s ec tion
Transplantation / 1091
PAUL SIKORSKI, RENATO VEGA, Laboratory
DONNA P. LUCAS, AND ANDREA A. ZACHARY
Management / 1169
115 Non-Human Leukocyte Antigen Antibodies in
Organ Transplantation / 1103 VOLUME EDITOR: ROBERT G. HAMILTON
ANNETTE M. JACKSON AND BETHANY L. DALE SECTION EDITOR: RONALD J. HARBECK
116 Evaluation of the Cellular Immune Response in 121 Clinical Immunology Laboratory
Transplantation / 1108 Accreditation, Licensure, and
DIANA METES, NANCY L. REINSMOEN, AND Credentials / 1171
ADRIANA ZEEVI LINDA COOK AND RONALD J. HARBECK
117 Complement in Transplant Rejection / 1123 122 Validation and Quality Control: General
CARMELA D. TAN, E. RENE RODRIGUEZ, AND Principles and Application to the Clinical
WILLIAM M. BALDWIN III Immunology Laboratory / 1180
118 Molecular Characterization of Rejection in VIJAYA KNIGHT AND TERRI LEBO
Solid Organ Transplantation / 1132
DARSHANA DADHANIA, TARA K. SIGDEL, Author Index / 1193
Subject Index / 1195

Editorial Board
C. LYNNE BUREK (section N) HOWARD M. LEDERMAN (section K)

Johns Hopkins University, Department of Pathology, SOM, Pediatric Allergy & Immunology, Johns Hopkins Hospital -
720 Rutland Ave., Baltimore, MD 21205 CMSC 1102, 600 N Wolfe St, Baltimore, MD 21287-3923
PATRIZIO CATUREGLI (section N) CHRISTINE M. LITWIN (section H)

Johns Hopkins University, Department of Pathology, SOM, Department of Pathology and Laboratory Medicine,
720 Rutland Ave., Baltimore, MD 21205 Medical University of South Carolina, 171 Ashley Ave.,
Charleston, SC 29425
DANIEL CHAN (section O)
Department of Pathology, Johns Hopkins University, SOM, THOMAS B. NUTMAN (section I)
Clinical Chemistry, CRB 11 3M 05, Baltimore, MD 21287 Department of Pathology and Laboratory Medicine,
Medical University of South Carolina, 171 Ashley Ave.,
STEVEN D. DOUGLAS (section E) Charleston, SC 29425
The Children’s Hospital of Philadelphia, University of
Pennsylvania, Suite 1208 Abramson Research Building, 34th MAURICE R. G. O’GORMAN (section D)
& Civic Center Blvd., Philadelphia, PA 19104 Keck School of Medicine, University of Southern California,
and the Children’s Hospital of Los Angeles, Pathology and
THOMAS A. FLEISHER (section A) Pediatrics, 4650 Sunset Blvd #43, Los Angeles, CA 90027
Department of Laboratory Medicine, Clinical Center,
National Institutes of Health, Bldg. 10 Rm. 2C306, 10 Center ELAINE F. REED (section P)
Drive, Bethesda, MD 20814 UCLA, Pathology, Rehab 1520, 1000 Veteran Avenue,
Immunogenetics Center, Los Angeles, CA 90095
PATRICIA C. GICLAS (section C)
National Jewish Health, Diagnostic Complement Laboratory, WESTLEY H. REEVES (section M)
1400 N. Jackson St., Denver, CO 80206 University of Florida, Division of Rheumatology & Clinical
Immunology, PO Box 100221, Gainesville, FL 32610-0221
PAMELA A. GUERRERIO (section L)
Food Allergy Research Unit, Laboratory of Allergic Diseases, R. NEAL SMITH (section G)
National Institute of Allergy and Infectious Diseases, Massachusetts General Hospital, Pathology, 501B Warren
4 Memorial Dr., Building 4, Room 228B, MSC0430, Bldg., 14 Fruit St., Boston, MA 02114
Bethesda, MD 20892
LORI J. SOKOLL (section O)
RONALD J. HARBECK (section Q) Department of Pathology, The Johns Hopkins University
National Jewish Health, 1400 Jackson Street, School of Medicine, Baltimore, MD 21205
Denver, CO 80206
KATHLEEN E. SULLIVAN (section K)
RICHARD L. HODINKA (section J) University of Pennsylvania, Division of Allergy and
University of South Carolina School of Medicine Greenville Immunology, Children’s Hospital of Philadelphia, 3615 Civic
and Greenville Health System, Room 210, Health Science Center Blvd., Philadelphia, PA 19104
Administration Building, 701 Grove Rd., Greenville, SC 2960
QIUHENG JENNIFER ZHANG (section P)
JOHN J. HOOKS (section F) UCLA Immunogenetics Center, Department Pathology &
National Institutes of Health, Immunology & Virology Laboratory Medicine, 15-20 Rehab, 1000 Veteran Ave.,
Section, NEI, Bldg. 10 Rm. 10N248, 10 Center Dr, Los Angeles, CA 90024
Bethesda, MD, 20814
DAVID F. KEREN (section B)

University of Michigan, 5228 Medical Science I, 1301
Catherine, Ann Arbor, MI 48109
xi
Contributors
ROSHINI SARAH ABRAHAM GIUSEPPE BARBESINO

Mayo Clinic, Laboratory Medicine and Pathology, Hilton Thyroid Unit, Massachusetts General Hospital – Harvard
210e, 200 1st St. SW, Rochester, MN 55905 Medical School, 15 Parkman St., Boston, MA 02114
MARIA E. AGUERO-ROSENFELD DAVID R. BARNIDGE

NYU Langone Medical Center, Rm. H374A, 560 First Ave., Department of Laboratory Medicine and Pathology, Mayo
New York, NY 10016 Clinic, Rochester, MN 55905
CEM AKIN LEE ANN BAXTER-LOWE

Brigham and Women’s Hospital, Department of Medicine, Children’s Hospital Los Angeles, 4650 Sunset Blvd., #32,
Rheumatology, Immunology, 75 Francis Street, Los Angeles, CA 90027
Boston, MA 02115
ANGELIQUE BIANCOTTO
MOHSIN ALI CHI/NHLBI, National Institutes of Health, 10 Center Drive,
Icahn School of Medicine at Mount Sinai, Department Bldg. 10 Room 7N110a, Bethesda, MD 20892
of Medical Education, One Gustave L. Levy Place,
New York, NY 10029 LUCAS S. BLANTON
University of Texas Medical Branch-Galveston,
BURT ANDERSON Department of Internal Medicine, 301 University Blvd.,
Department of Molecular Medicine, Morsani College of Galveston, TX 77555
Medicine, University of South Florida, 12901 Bruce B. Downs
Blvd., Tampa, FL 33612 MARY B. BROWN
Department of Infectious Diseases and Pathology, College of
Veterinary Medicine, University of Florida, P.O. Box 110880,
LUIS EDUARDO COELHO ANDRADE
2015 S.W. 16th Ave., Gainesville, FL 32611
Escola Paulista de Medicina, Universidade Federal de
Sao Paulo, Rheumatology Division, Rua Botucatu 740,
SARAH K. BROWNE
Vila Clementino, Sao Paulo, SP 04023-062, and Fleury
NIAID, NIH, Immunopathogenesis Section, Bldg. 10 - CRC
Laboratories, Immunology Division, Av. Valdomiro de Lima
Rm. B3-4233, 10 Center Drive, Bethesda, MD 20014
508, São Paulo, SP 04344-070, Brazil
YENAN T. BRYCESON
AFTAB A. ANSARI Center for Hematology and Regenerative Medicine,
Department of Pathology and Laboratory Medicine, Emory
Department of Medicine, Karolinska Institutet, Karolinska
University School of Medicine, Atlanta, GA 30322
University Hospital Huddinge, S-14186 Stockholm, Sweden,
and Institute of Clinical Sciences, University of Bergen,
WILLIAM M. BALDWIN III N-5021 Bergen, Norway
Department of Immunology, 9500 Euclid Ave.,
Cleveland, OH 44022 MICHAEL RAYMOND BUBB
Division of Rheumatology, University of Florida, 1600 S.W.
HENRY H. BALFOUR, JR. Archer Rd D2-39, P.O. Box 100221, Gainesville, FL 32610
University of Minnesota Medical School, Laboratory
Medicine & Pathology, and Pediatrics, MMC 609, 420 C. E. BUCHNER
Delaware St. SE, Minneapolis, MN 55455 Genalyte, Inc., 10520 Wateridge Circle, San Diego, CA 92121
xiii
xiv ■ Contributors
C. LYNNE BUREK SARAH COOLEY

Johns Hopkins University, Department of Pathology, SOM, University of Minnesota, Hematology, Oncology and
720 Rutland Ave., Baltimore, MD 21205 Transplantation, 420 Delaware St. SE, Mayo Mail Code 806,
Minneapolis, MN 55455
RUFUS W. BURLINGAME
Genalyte, Inc., Diagnostic Assay Development, 10520 ROSS L. COPPEL
Wateridge Circle, San Diego, CA 92121 Faculty of Medicine, Nursing and Health Sciences, Monash
University, Clayton, Victoria, Australia 3800
ROBERT L. BURTON
University of Alabama at Birmingham, 845 19th St. S, WILSON DE MELO CRUVINEL
BBRB612, Birmingham, AL 35294 Pontifícia Universidade Católica de Goiás, School of
Medical, Pharmaceutical and Biomedical Sciences,
FREDERIC D. BUSHMAN Avenida Universitária 1440, Setor Universitário, Goiânia,
Perelman School of Medicine, University of Pennsylvania, GO, 74.605-010, Brazil
Department of Microbiology, 3610 Hamilton Walk,
Philadelphia, PA 19104 KELLY A. CURTIS
Division of HIV/AIDS Prevention, National Center for HIV/
BREANNA CARUSO AIDS, Viral Hepatitis, STD, and TB Prevention, Centers for
National Institute of Neurological Disorders and Stroke, Disease Control and Prevention, Atlanta, GA 30329
National Institutes of Health, 9000 Rockville Pike, Rockville,
MD 20892 DARSHANA DADHANIA
Weill Cornell Medical College, Division of Nephrology &
PATRIZIO CATUREGLI Hypertension, 525 E. 68th St., Box 3, New York, NY 10065
Johns Hopkins University, Department of Pathology, SOM,
720 Rutland Ave., Baltimore, MD 21205 BETHANY L. DALE
Immunogenetics Laboratory, Johns Hopkins University School
ANGELA CERIBELLI of Medicine, 2041 E. Monument St., Baltimore, MD 21205
Rheumatology and Clinical Immunology, Humanitas Clinical
and Research Center, Via A. Manzoni 56, 20089, Rozzano DANIEL C. DANILA
(Milan), Italy Memorial Sloan Kettering Cancer Center, 1275 York Ave.,
New York, NY 10065
CHI-CHAO CHAN
Laboratory of Immunology, National Eye Institute, National ALESSSANDRA DE REMIGIS
Institutes of Health, Bldg. 10 Rm 10N109, 10 Center Drive, Johns Hopkins University, Department of Pathology, Rutland
Bethesda, MD 20814 Ave., Baltimore, MD 21205
EDWARD K. L. CHAN WILLIAM MARCIEL DE SOUZA

Department of Oral Biology, University of Florida, P.O. Box Virology Research Center, School of Medicine of Ribeirao Preto
100424, Gainesville, FL 32610 of University of Sao Paulo, Ribeirao Preto, São Paulo, Brazil
ANITA CHANDRA LUDOVICA DE VINCENTIIS

Department of Clinical Biochemistry and Immunology, Box Johns Hopkins University, Department of Pathology, Rutland
109, Addenbrooke’s Hospital, Hills Road, Cambridge CB2 Ave., Baltimore, MD 21205
0QQ, United Kingdom
HEBA DEGHEIDY
ANN DUSKIN CHAUFFE FDA, Center for Biologics Evaluation and Research,
Division of Rheumatology, University of Florida, WO52/72 RM 3209, 10903 New Hampshire Ave., Silver
1600 S.W. Archer Rd D2-39, P.O. Box 100221, Spring, MD 20993
Gainesville, FL 32610
ALESSANDRA DELLAVANCE
SAMUEL C. C. CHIANG Fleury Laboratories, Research and Development
Center for Hematology and Regenerative Medicine, Department, Avenida Valdomiro de Lima 508, São Paulo,
Department of Medicine, Karolinska Institutet, SP 04344-070, Brazil
Karolinska University Hospital Huddinge, S-14186
Stockholm, Sweden BARBARA DETRICK
Immunology Laboratory, Department of Pathology, Johns
A. BERNARD COLLINS Hopkins University, School of Medicine, 600 N. Wolfe St.,
Massachusetts General Hospital, Pathology, 503 Warren Bldg., Baltimore, MD 21287
14 Fruit St., Boston, MA 02114
JOSEPH A. DiGIUSEPPE
LINDA COOK Hematopathology and Special Hematology Laboratory,
University of Washington, Laboratory Medicine, 1616 Department of Pathology & Laboratory Medicine, Hartford
Eastlake Ave. E, Suite 320, Seattle, WA 98102 Hospital, 80 Seymour St., Hartford, CT 06102
Contributors ■ xv
STEVEN D. DOUGLAS DENNIS GALANAKIS

The Children’s Hospital of Philadelphia, University of State University of New York, Stony Brook, NY 11794
Pennsylvania, Suite 1208 Abramson Research Building,
34th & Civic Center Blvd., Philadelphia, PA 19104 M. ERIC GERSHWIN
Division of Rheumatology/Allergy and Clinical Immunology,
ELIZABETH R. DUFFY Genome and Biomedical Sciences Facility Suite 6510,
Boston University School of Medicine, Pathology and School of Medicine, University of California at Davis,
Laboratory Medicine, 670 Albany St., Boston, MA 02118 Davis, CA 95616
JAMIE DUKE EMANUELA M. GHIA

The Children’s Hospital of Philadelphia, 3401 Civic Center UCSD, Moores Cancer Center, 3855 Health Science Drive,
Blvd., Philadelphia, PA 19104 M/C 0820, La Jolla, CA 92093
BRUCE E. DUNN PATRICIA C. GICLAS

Medical College of Wisconsin, 8701 Watertown Plank Road, National Jewish Health, Diagnostic Complement Laboratory,
Milwaukee, WI 53226 1400 N. Jackson St., Denver, CO 80206
GEORGE S. EISENBARTH KIMBERLY C. GILMOUR

[Deceased] Immunology, Camelia Botnar Laboratories, Great Ormond
Street Hospital for Children NHS Foundation Trust, London
MELISSA ELDER WC1N 3JH, United Kingdom
University of Florida, Pediatrics, 1600 S.W. Archer Road,
Gainesville, FL 32610 ELIZABETH A. GODBEY
Department of Pathology, Columbia University Medical
DEBORAH FERRIOLA Center, New York, NY 10032
The Children’s Hospital of Philadelphia, 3401 Civic Center
Blvd., Philadelphia, PA 19104 PETER D. GOREVIC
Division of Rheumatology, The Mount Sinai Medical
LUIZ TADEU MORAES FIGUEREIDO Center, Annenberg Building; Room 21-056, Box 1244,
Virology Research Center, School of Medicine of New York, NY 10029
Ribeirao Preto of University of Sao Paulo, Ribeirao Preto,
São Paulo, Brazil KIM Y. GREEN
Caliciviruses Section, Laboratory of Infectious Diseases,
DAVID M. FLEISCHER National Institute of Allergy and Infectious Diseases,
Children’s Hospital Colorado, Pediatrics, Aurora, 9000 Rockville Pike, Building 50, Room 6318,
CO 80045 Bethesda, MD 20892
MARTIN FLEISHER PAMELA A. GUERRERIO

Memorial Sloan Kettering Cancer Center, 1275 York Ave., Food Allergy Research Unit, Laboratory of Allergic
New York, NY 10065 Diseases, National Institute of Allergy and Infectious
Diseases, 4 Memorial Dr., Building 4, Room 228B,
THOMAS A. FLEISHER MSC0430, Bethesda, MD 20892
Department of Laboratory Medicine, Clinical Center,
National Institutes of Health, Bldg. 10 Rm. 2C306, ROBERT G. HAMILTON
10 Center Drive, Bethesda, MD 20814 Johns Hopkins University School of Medicine, Dermatology,
Allergy and Clinical Immunology Reference Library, 5501
JUAN FLORES-MONTERO Hopkins Bayview Circle, Baltimore, MD 21224
Centro de Investigación del Cáncer (Instituto de Biología
Molecular y Celular del Cáncer, CSIC-USAL), Instituto SHUHONG HAN
Biosanitario de Salamanca (IBSAL), Servicio General University of Florida, Division of Rheumatology & Clinical
de Citometría (NUCLEUS-Universidad de Salamanca), Immunology, PO Box 100221, Gainesville, FL 32610-0221
Salamanca, 37007, Spain
J. G. HANLY
PAULO LUIZ CARVALHO FRANCESCANTONIO Dalhousie University and Nova Scotia Health Authority
Pontifícia Universidade Católica de Goiás, School of (Central Zone), Departments of Medicine and Pathology,
Medical, Pharmaceutical and Biomedical Sciences, Avenida Nova Scotia Rehabilitation Center, 1341 Summer St.,
Universitária 1440, Setor Universitário, Goiânia, GO, Halifax, NS B3H 4K4, Canada
74.605-010, Brazil
RONALD J. HARBECK
MARVIN J. FRITZLER National Jewish Health, 1400 Jackson Street, Denver,
University of Calgary, Cumming School of Medicine, Calgary, CO 80206
Alberta T2N 4N1, Canada
xvi ■ Contributors
NEIL HARRIS STEVEN JACOBSON

University of Florida, Department of Pathology, 1600 SW National Institute of Neurological Disorders and Stroke,
Archer Rd, Gainesville, FL 32610 National Institutes of Health, 9000 Rockville Pike,
Rockville, MD 20892
CHOLI HARTONO
Weill Cornell Medical College, Nephrology, 505 E. 70th St., ELAINE S. JAFFE
Helmsley 2nd Floor, New York, NY 10021 Laboratory of Pathology, Center for Cancer Research,
National Institutes of Health, 10 Center Dr./Rm. 3S235,
HARRY R. HILL MSC-1500, Bethesda, MD 20892
University of Utah, Department of Pathology, Pediatrics and
Medicine, 50 N. Medical Drive, Room 5B-114, Salt Lake City, JEFFREY A. JOHNSON
UT 84132 Division of HIV/AIDS Prevention, National Center for HIV/
AIDS, Viral Hepatitis, STD, and TB Prevention, Centers for
MICHITO HIRAKATA Disease Control and Prevention, Atlanta, GA 30329
Medical Education Center, Graduate Medical Education
Center, Keio University School of Medicine, Tokyo, Japan JERRY A. KATZMANN
Mayo Clinic and Mayo Foundation, Laboratory Medicine and
RICHARD L. HODINKA Pathology, 200 First St. SW, Rochester, MN 55905
University of South Carolina School of Medicine Greenville
and Greenville Health System, Room 210, Health Science MICHAEL KEENEY
Administration Building, 701 Grove Rd., Greenville, SC 2960 Hematology/Flow Cytometry, London Health Sciences
Centre, Victoria Hospital, 800 Commissioners Road E,
KRISTIN A. HOGQUIST London, Ontario, N6A5W9 Canada
Center for Immunology, University of Minnesota, 2-186 MBB,
2101 6th St. SE, Minneapolis, MN 55455 DAVID F. KEREN
University of Michigan, 5228 Medical Science I, 1301
STEVEN M. HOLLAND Catherine, Ann Arbor, MI 48109
National Institutes of Health, LCID, CRC B3-4141, MSC
1684, Bethesda, MD 20892 THOMAS S. KICKLER
Johns Hopkins University School of Medicine, 1800 Orleans
JOHN J. HOOKS Street, Sheikh Zayed B2-120Q, Baltimore, MD 21287
National Institutes of Health, Immunology & Virology
Section, NEI, Bldg. 10 Rm. 10N248, 10 Center Drive, KAREN E. KING
Bethesda, MD 20814 Johns Hopkins Hospital, Transfusion Medicine, 1800 Orleans
St., Baltimore, MD 21287
D. CRAIG HOOPER
Thomas Jefferson University, Jefferson Center for THOMAS J. KIPPS
Neurovirology, 1020 Locust St, Philadelphia, PA 19107 UCSD, Moores Cancer Center, 3855 Health Science Drive,
M/C 0820, La Jolla, CA 92093
AMY P. HSU
National Institutes of Health, Laboratory of Clinical AMY D. KLION
Infectious Disease, National Institute of Allergy and Infectious National Institutes of Health, Laboratory of Parasitic Diseases,
Diseases, Bldg. 10 CRC Rm B3-4233, 10 Center Drive, NIAID, Bldg. 4, Rm. B1-28, Bethesda, MD 20892
Bethesda, MD 20892
VIJAYA KNIGHT
RICHARD L. HUMPHREY National Jewish Health, National Jewish Health Advanced
Johns Hopkins Hospital, Pathology, 600 North Wolfe St., Diagnostic Laboratories, Division of Pathology, Department of
Baltimore, MD 21287 Medicine, 1400 Jackson St., Denver, CO 80206
ANDREA ILLINGWORTH DOUGLAS B. KUHNS

Dahl Chase Diagnostic Services, 417 State St., Suite 540, Clinical Services Program, P.O. Box B, Leidos Biomedical
Bangor, ME 04401 Research, Inc., Frederick National Laboratory for Cancer
Research, Frederick, MD 21702
SABINA A. ISLAM
Center for Immunology and Inflammatory Diseases, Division D.S. KUMARARATNE
of Rheumatology, Allergy and Immunology, Massachusetts Department of Clinical Biochemistry and Immunology, Box
General Hospital, Boston, MA 02114 109, Addenbrooke’s Hospital, Hills Road, Cambridge CB2
0QQ, United Kingdom
ANNETTE M. JACKSON
Immunogenetics Laboratory, Johns Hopkins University MARK KUNKEL
School of Medicine, 2041 E. Monument Street, Baltimore, The Children’s Hospital of Philadelphia, 3401 Civic Center
MD 21205 Blvd., Philadelphia, PA 19104
Contributors ■ xvii
MASATAKA KUWANA DONNA P. LUCAS

Department of Allergy and Rheumatology, Nippon Medical Johns Hopkins University, Immunogenetics Laboratory,
School, 1-1-5 Sendagi, Bunkyo-ku, Tokyo 113-8602, Japan 2041 E. Monument St., Baltimore, MD 21205
ROBERT S. LANCIOTTI ANDREW D. LUSTER

Arbovirus Diseases Branch, Centers for Disease Control & Center for Immunology and Inflammatory Diseases, Division
Prevention, 3150 Rampart Road (CSU Foothills Campus), of Rheumatology, Allergy and Immunology, Massachusetts
Fort Collins, CO 80521 General Hospital, Boston, MA 02114
MARIE LOUISE LANDRY HOLDEN T. MAECKER

Yale University, Laboratory Medicine and Internal Medicine, Stanford University, Institute for Immunity, Transplantation,
P.O. Box 208035, New Haven, CT 06520 & Infection, Stanford University Medical School, 299
Campus Drive, Stanford, CA 94305
TERRI LEBO
National Jewish Health, Advanced Diagnostic Laboratories, CHERYL L. MAIER
1400 Jackson St., Denver, CO 80206 Department of Pathology and Laboratory Medicine, Emory
University School of Medicine, Atlanta, GA 30322
HOWARD M. LEDERMAN
Pediatric Allergy & Immunology, Johns Hopkins Hospital - MICHAEL P. MANNS
CMSC 1102, 600 N Wolfe St, Baltimore, MD 21287-3923 Department of Gastroenterology and Hepatology, Zentrum
Innere Medizin, Medizinische Hochschule Hannover,
DIANE S. LELAND Hannover, Germany
Department of Pathology and Laboratory Medicine, Indiana
University School of Medicine, IU Health Pathology REBECCA MARSH
Laboratory Building, Room 6027F, 350 W 11th St, Division of Bone Marrow Transplantation and Immune
Indianapolis, IN 46202 Deficiency, Cincinnati Children’s Hospital Medical Center,
3333 Burnet Ave., Cincinnati, OH 45229
PATRICK S. C. LEUNG
Division of Rheumatology/Allergy and Clinical Immunology, JOHN MASSINI
Genome and Biomedical Sciences Facility Suite 6510, University of Florida, Division of Rheumatology & Clinical
School of Medicine, University of California at Davis, Immunology, PO Box 100221, Gainesville, FL 32610-0221
Davis, CA 95616
RAYA MASSOUD
ARNOLD I. LEVINSON National Institute of Neurological Disorders and Stroke,
Perelman School of Medicine, University of Pennsylvania National Institutes of Health, 9000 Rockville Pike, Rockville,
School of Medicine, Room 316 Blockley Hall, 423 Guardian MD 20892
Drive, Philadelphia, PA 19104
J. PHILIP McCOY, JR.
YI LI National Institutes of Health, NHLBI, 10 Center Drive,
University of Florida, Division of Rheumatology & Clinical Bethesda, MD 20892
Immunology, PO Box 100221, Gainesville, FL 32610-0221
BENJAMIN D. MEDOFF
CURT LIND Center for Immunology and Inflammatory Diseases,
The Children’s Hospital of Philadelphia, 3401 Civic Center Pulmonary and Critical Care Unit, Massachusetts General
Blvd., Philadelphia, PA 19104 Hospital, Harvard Medical School, Boston, MA 02114
MARK D. LINDSLEY DIANA METES

Mycotic Diseases Branch, Centers for Disease Control and University of Pittsburgh Medical Center, Thomas E Starzl
Prevention, 1600 Clifton Road, Mailstop G-11, Atlanta, Transplantation Institute, BST E1549, 200 Lothrop St.,
GA 30333 Pittsburgh, PA 15213
ROBERT P. LISAK DIMITRI MONOS

Wayne State University Medical Center, Neurology, 4201 St. The Children’s Hospital of Philadelphia, 3401 Civic Center
Antoine St., Detroit, MI 48201 Blvd., Philadelphia, PA 19104
CHRISTINE M. LITWIN MARTINA MURPHY

Department of Pathology and Laboratory Medicine, Medical University of Florida, Hematology/Oncology, 1600 SW
University of South Carolina, 171 Ashley Ave., Charleston, Archer Rd., Gainesville, FL 32610
SC 29425
THANGAMANI MUTHUKUMAR
SHELDON E. LITWIN Weill Cornell Medical College, Division of Nephrology &
Department of Medicine, Medical University of South Hypertension, 525 E. 68th St., Box 3, New York, NY 10065
Carolina, 114 Doughty St., Charleston, SC 29425
xviii ■ Contributors
MOON H. NAHM Centro de Investigación del Cáncer (Instituto de Biología

University of Alabama at Birmingham, 845 19th St. S, BBRB Molecular y Celular del Cáncer, CSIC-USAL), Salamanca,
614, Birmingham, AL 35294 37007, Spain
STANLEY J. NAIDES ROBERT B. PETERSEN

Immunology, Quest Diagnostics Nichols Institute, 33608 Case Western Reserve University, Department of Pathology,
Ortega Highway, San Juan Capistrano, CA 92675 5-126 Wolstein Building, 2103 Cornell Road, Cleveland,
OH 44106
HUBERT G. M. NIESTERS
University Medical Centre Groningen, Department of SUHAS H. PHADNIS
Medical Microbiology, Division of Clinical Virology, Medical College of Wisconsin, Pathology, 9200 W. Wisconsin
Hanzeplein 1, Groningen, The Netherlands Ave., Milwaukee, WI 53005
TIMOTHY B. NIEWOLD FANNY POJERO

Mayo Clinic, Department of Immunology and Division of Centro de Investigación del Cáncer (Instituto de Biología
Rheumatology, 200 1st Street SW, Rochester, MN 55905 Molecular y Celular del Cáncer, CSIC-USAL), Instituto
Biosanitario de Salamanca (IBSAL), Servicio General
DOUGLAS F. NIXON de Citometría (NUCLEUS-Universidad de Salamanca),
Dept. of Microbiology, Immunology and Tropical Medicine, Salamanca, 37007, Spain
School of Medicine & Health Sciences, The George
Washington University, Ross Hall 736, 2300 Eye Street, NW, YVONNE POSEY
Washington, D.C. 20037 Beaumont Hospital – Royal Oak, Clinical Pathology,
3601 W. 13 Mile Road, Royal Oak, MI 48073
ROBERT NUSSENBLATT
Laboratory of Immunology, National Eye Institute, National DEBRA LONG PRIEL
Institutes of Health, Bldg. 10 Rm 10N109, 10 Center Drive, Clinical Services Program, P.O. Box B, Leidos Biomedical
Bethesda, MD 20814 Research, Inc., Frederick National Laboratory for Cancer
Research, Frederick, MD 21702
THOMAS B. NUTMAN
Laboratory of Parasitic Diseases, National Institute for Allergy CALMAN PRUSSIN
and Infectious Diseases, National Institutes of Health, Laboratory of Allergic Diseases, National Institute of Allergy
4 Center Drive, Room B1 03, Bethesda, MD 20892 and Infectious Diseases, National Institutes of Health,
Building 10, Room 11N238, Bethesda, MD 20892-1881
MAURICE R. G. O’GORMAN
Keck School of Medicine, University of Southern California, NOEMÍ PUIG
and the Children’s Hospital of Los Angeles, Pathology and Departmento de Hematología, Hospital Universitario
Pediatrics, 4650 Sunset Blvd #43, Los Angeles, CA 90027 de Salamanca, Instituto Biosanitario de Salamanca
(IBSAL); Centro de Investigación del Cáncer (Instituto de
ALBERTO ORFAO Biología Molecular y Celular del Cáncer, CSIC-USAL),
Centro de Investigación del Cáncer (Instituto de Biología Salamanca, 37007, Spain
Molecular y Celular del Cáncer, CSIC-USAL), Instituto
Biosanitario de Salamanca (IBSAL), Servicio General
RONALD L. RABIN
Center for Biologics Evaluation and Research, U.S. Food and
de Citometría (NUCLEUS-Universidad de Salamanca),
Drug Administration, 10903 New Hampshire Avenue, Silver
Salamanca, 37007, Spain
Spring, MD 20993
S. MICHELE OWEN
MARK RAFFELD
National Center for HIV/AIDS, Viral Hepatitis, STD, and
Laboratory of Pathology, Center for Cancer Research,
TB Prevention, Centers for Disease Control and Prevention,
National Institutes of Health, 10 Center Dr./Rm. 3S235,
Atlanta, GA 30329
MSC-1500, Bethesda, MD 20892
GABRIEL I. PARRA ALEX J. RAI
Caliciviruses Section, Laboratory of Infectious Diseases,
Department of Pathology, Columbia University Medical
National Institute of Allergy and Infectious Diseases,
Center, New York, NY 10032
9000 Rockville Pike, Building 50, Room 6316, Bethesda,
MD 20892 RAJA RAJALINGAM
University of California at San Francisco, Immunogenetics
R. STOKES PEEBLES, JR. and Transplantation Laboratory, Department of Surgery, 45
Vanderbilt University, Medicine, T-1218 MCN, Vanderbilt Castro St., Main Hospital Level B, CPMC Davis Campus, San
University Medical Center, Nashville, TN 37232 Francisco, CA 94114
JOSÉ JUAN PÉREZ AMY RASLEY
Departmento de Hematología, Hospital Universitario de Host-Pathogen Laboratory Group, Lawrence Livermore
Salamanca, Instituto Biosanitario de Salamanca (IBSAL); National Laboratory, Livermore, CA 94550
Contributors ■ xix
LAURA Z. RASSENTI RICHARD RUBENSTEIN

UCSD, Moores Cancer Center, 3855 Health Science Drive, SUNY Downstate Medical Center, Departments of Neurology
M/C 0820, La Jolla, CA 92093 and Physiology/Pharmacology, 450 Clarkson Ave., Brooklyn,
NY 11203
ELAINE F. REED
UCLA, Pathology, Rehab 1520, 1000 Veteran Avenue, DALIA A. A. SALEM
Immunogenetics Center, Los Angeles, CA 90095 CCR, NCI, NIH, Laboratory of Pathology, Building 10, Mail
Stop 1500, Room 3S 241, Bethesda, MD 20892
WESTLEY H. REEVES
University of Florida, Division of Rheumatology & Clinical LUZALBA SANOJA
Immunology, PO Box 100221, Gainesville, FL 32610-0221 Centro de Investigación del Cáncer (Instituto de Biología
Molecular y Celular del Cáncer, CSIC-USAL), Instituto
NANCY L. REINSMOEN Biosanitario de Salamanca (IBSAL), Servicio General
HLA and Immunogenetics Laboratory, Comprehensive de Citometría (NUCLEUS-Universidad de Salamanca),
Transplant Center, Cedars-Sinai Health Systems, HLA Salamanca, 37007, Spain
and Immunogenetics Lab-SSB 197, 8723 Alden Drive, Los
Angeles, CA 90048 CARAH B. SANTOS
National Jewish Health, 1400 Jackson St., K731A, Denver,
RYAN F. RELICH CO 80206
Department of Pathology and Laboratory Medicine, Indiana
University School of Medicine, IU Health Pathology MINNIE M. SARWAL
Laboratory Building, Room 6027E, 350 W 11th St, University of California San Francisco, Division of
Indianapolis, IN 46202 Transplant Surgery, G893, 513 Parnassus Ave., San Francisco,
CA 94143
DANIEL G. REMICK
Boston University School of Medicine, 670 Albany St., MINORU SATOH
Boston, MA 02118 Department of Clinical Nursing, School of Health Sciences,
University of Occupational and Environmental Health,
LYNNSEY RENN Japan, 1-1 Isei-ga-oka, Yahata-nishi-ku, Kitakyushu, Fukuoka,
Center for Biologics Evaluation and Research, U.S. Food and 807-8555, Japan
Spring, MD 20993 HOWARD I. SCHER
Memorial Sloan Kettering Cancer Center, 1275 York Ave.,
ADRIANA RICCIUTI New York, NY 10065
Johns Hopkins University, Department of Pathology, Rutland
Ave., Baltimore, MD 21205 D. SCOTT SCHMID
Centers for Disease Control and Prevention, NCIRD/DVD/
ANNELIES RIEZEBOS-BRILMAN MMRHLB, 1600 Clifton Rd NE, Atlanta, GA 30333
University Medical Centre Groningen, Department of
Medical Microbiology, Division of Clinical Virology, JOHN L. SCHMITZ
Hanzeplein 1, Groningen, The Netherlands University of North Carolina, Department of Pathology &
Laboratory Medicine, School of Medicine, Rm. 1035 East
KIMBERLY RISMA Wing, UNC Hospitals, Chapel Hill, NC 27514
Division of Allergy/Immunology, Cincinnati Children’s
Hospital Medical Center, 3333 Burnet Ave., Cincinnati, JOHN T. SCHROEDER
OH 45229 Johns Hopkins University, Medicine, Division of Allergy and
Immunology, Unit Office 2, 5501 Hopkins Bayview Circle,
E. RENE RODRIGUEZ Baltimore, MD 21224
Department of Pathology, 9500 Euclid Ave., Cleveland,
OH 44022 H. NIDA SEN
Laboratory of Immunology, National Eye Institute, National
JOHN T. ROEHRIG Institutes of Health, Bldg. 10 Rm 10N109, 10 Center Drive,
Centers for Disease Control and Prevention, Atlanta, Bethesda, MD 20814
GA (Retired)
CHRISTINE SEROOGY
NOEL R. ROSE University of Wisconsin, Pediatrics, 1111 Highland Ave.,
Johns Hopkins University, Department of Pathology, SOM, 4139 WIMR, Madison, WI 53705
720 Rutland Avenue, Baltimore, MD 21205
BARBARA L. SHACKLETT
JOHN M. ROUTES Dept. of Medical Microbiology and Immunology, School of
Department of Pediatrics and Department of Microbiology Medicine, University of California at Davis, 3146 Tupper
and Molecular Genetics, Medical College of Wisconsin, Hall, 1 Shields Ave., Davis, CA 95616
Milwaukee, WI 53226
xx ■ Contributors
ROSEMARY SHE ANDREA K. STECK

Keck Medical Center of USC, Pathology, 1441 Eastlake Ave., Barbara Davis Center for Childhood Diabetes, University of
Suite 2424, Los Angeles, CA 90089 Colorado School of Medicine, Aurora, CO 80045
R. SUE SHIREY MARYALICE STETLER-STEVENSON

Johns Hopkins Hospital, Transfusion Medicine, 1800 Orleans CCR, NCI, NIH, Laboratory of Pathology, Building 10, Mail
St., Baltimore, MD 21287 Stop 1500, Room 3S 235G, Bethesda, MD 20892
TARA SIGDEL JAMES R. STONE

University of California San Francisco, Division of Transplant Massachusetts General Hospital, Pathology, 185 Cambridge
Surgery, 513 Parnassus Avenue, S-1268 Medical Sciences Street, Boston, MA 02114
Building, San Francisco, CA 94143
JOHN H. STONE
PAUL SIKORSKI Harvard Medical School, Division of Rheumatology, 25
Johns Hopkins University, Immunogenetics Laboratory, 2041 Shattuck St, Boston, MA 02115
E. Monument St., Baltimore, MD 21205
MANIKKAM SUTHANTHIRAN
JERRY W. SIMECKA Weill Cornell Medical College, Division of Nephrology &
Department of Cell Biology and Immunology, University of Hypertension, 525 E. 68th St., Box 3, New York, NY 10065
North Texas Health Science Center, RES 402A 3500 Camp
Bowie Blvd., Fort Worth, TX 76107 D. ROBERT SUTHERLAND
Laboratory Medicine Program, Toronto General Hospital/
JAY E. SLATER University Health Network, 200 Elizabeth St., Room 11E416,
Center for Biologics Evaluation and Research, U.S. Food and Toronto, Ontario, M5G2C4 Canada
Spring, MD 20993 ELIZABETH SYKES
Beaumont Hospital – Royal Oak, Clinical Pathology, 3601 W.
MAREK SMIEJA 13 Mile Road, Royal Oak, MI 48073
McMaster University, Department of Pathology & Molecular
Medicine, L424-St. Joseph’s Healthcare Hamilton, 50 CARMELA D. TAN
Charlton Ave E, Hamilton, ON L8N 4A6, Canada Department of Pathology, 9500 Euclid Ave., Cleveland,
OH 44022
RICHARD J. H. SMITH
Iowa Institute of Human Genetics, Molecular Otolaryngology YI-WEI TANG
and Renal Research Laboratories, University of Iowa, Memorial Sloan-Kettering Cancer Center, Clinical
Iowa City, IA 52242 Microbiology Service, 1275 York Ave., S328, New York, NY
10065
R. NEAL SMITH
Massachusetts General Hospital, Pathology, 501B Warren STEFFEN THIEL
Bldg., 14 Fruit St., Boston, MA 02114 Aarhus University, Department of Medicine, Bartholin
Building, Wilhelm Meyers Allé 4, Aarhus, 8000, Denmark
MELISSA R. SNYDER
Mayo Clinic, Hilton 2-10D, 200 First St. SW, Rochester, RENEE TSOLIS
MN 55905 Department of Medical Microbiology and Immunology,
University of California, Davis, CA 95616
LORI J. SOKOLL
Department of Pathology, The Johns Hopkins University JEROEN VAN BERGEN
School of Medicine, Baltimore, MD 21205 Department of Immunohematology and Blood Transfusion,
Leiden University Medical Center, 2333 ZA Leiden, The
JEREMY SOKOLOVE Netherlands
VA Palo Alto Health Care System, 3801 Miranda Ave, Palo
Alto, CA 94304-1207, and Division of Immunology and CORETTA C. VAN LEER-BUTER
Rheumatology, Stanford University School of Medicine, University Medical Centre Groningen, Department of
Stanford, CA 94305 Medical Microbiology, Division of Clinical Virology,
Hanzeplein 1, Groningen, The Netherlands
LORI SOMA
University of Washington, Department of Laboratory PRIYANKA VASHISHT
Medicine, NW120, Box 357110, 1959 Pacific St., Mayo Clinic, Department of Immunology and Division of
Seattle, WA 98195-7110 Rheumatology, 200 1st St. SW, Rochester, MN 55905
DAVID J. SPEICHER RENATO VEGA

Griffith University, Menzies Health Institute Queensland, Johns Hopkins University, Immunogenetics Laboratory,
Gold Coast Campus, Queensland 4222, Australia 2041 E. Monument St., Baltimore, MD 21205
Contributors ■ xxi
JAMES W. VERBSKY THERESA L. WHITESIDE

Department of Pediatrics and Department of Microbiology University of Pittsburgh Cancer Institute, Research Pavilion
and Molecular Genetics, Medical College of Wisconsin, at the Hillman Cancer Center, 5117 Centre Ave. Suite 1.27,
Milwaukee, WI 53226 Pittsburgh, PA 15213
PRIYA S. VERGHESE PATRICIA P. WILKINS

Pediatric Kidney Transplantation, University of Minnesota, Division of Parasitic Diseases & Malaria, Center for Global
Children’s Hospital, 2450 Riverside Ave., MB 687, Health, Centers for Disease Control and Prevention, 1600
Minneapolis, MN 55454 Clifton Road, Atlanta, GA 30333
MARÍA BELÉN VIDRIALES HUGH J. WILLISON

Departmento de Hematología, Hospital Universitario de B330, GBRC, 120 University Place, University of Glasgow,
Salamanca, Instituto Biosanitario de Salamanca (IBSAL); Glasgow, Scotland, G12 8TA, United Kingdom
Centro de Investigación del Cáncer (Instituto de Biología
Molecular y Celular del Cáncer, CSIC-USAL), Salamanca, THOMAS WISNIEWSKI
37007, Spain New York University School of Medicine, Department of
Neurology, Psychiatry and Pathology, Alexandria ERSP, Rm.
KEN B. WAITES 802, 450 E. 29th St., New York, NY 10016
Department of Pathology, University of Alabama at
Birmingham, WP 230, 619 S. 19th St., Birmingham, BRENT WOOD
AL 35249 University of Washington Medical Center, Hematopathology,
Seattle, WA 98109
DAVID H. WALKER
University of Texas Medical Branch-Galveston, Department ROBERT A. WOOD
of Pathology, 301 University Blvd., Galveston, TX 77555 Johns Hopkins University, Baltimore, MD 21287
NOREEN M. WALSH LIPING YU

Dalhousie University and Nova Scotia Health Authority Barbara Davis Center for Childhood Diabetes, University of
(Central Zone), Department of Pathology and Laboratory Colorado School of Medicine, Aurora, CO 80045
Medicine, Mackenzie Building, Room 721, 5788 University
Ave., Halifax, Nova Scotia B3H1V8, Canada CONSTANCE M. YUAN
CCR, NCI, NIH, Laboratory of Pathology, Building 10, Mail
GUIQING WANG Stop 1500, Room 2A33, Bethesda, MD 20892
New York Medical College, Department of Pathology, 100
Woods Road, Westchester Medical Center Rm. 1J-04, ANDREA A. ZACHARY
Valhalla, NY 10595 Johns Hopkins University, Immunogenetics Laboratory,
2041 E. Monument St., Baltimore, MD 21205
JIAN WANG
Department of Endocrinology, Jinling Hospital, ADRIANA ZEEVI
Nanjing, China University of Pittsburgh Medical Center, Clinical Laboratory
Building, Room 4033, 3477 Euler Way, Pittsburgh, PA 15213
JEFFREY S. WARREN
University of Michigan, Pathology, 5242 MSI, 1301 Catherine QIUHENG JENNIFER ZHANG
St., Ann Arbor, MI 48109 UCLA Immunogenetics Center, Department Pathology &
Laboratory Medicine, 15-20 Rehab, 1000 Veteran Ave.,
ADRIANA WEINBERG Los Angeles, CA 90024
Department of Pediatrics, Medicine and Pathology, University
of Colorado Health Sciences Center, 4200 E. Ninth Ave.,
Campus Box C 227, Denver, CO 80262
Acknowledgment of Previous Contributors

The Manual of Molecular and Clinical Laboratory Immunology is by its nature a continuously
revised work which refines and extends the contributions of previous editions. Since its first
edition in 1976, many eminent scientists have contributed to this important reference work.
The American Society for Microbiology and its Publications Board gratefully acknowledge
the contributions of all of these generous authors over the life of this Manual.
Foreword:
How It Began
In 1971, I was working at the University of Oxford’s Sir he understood the practice of laboratory immunology
William Dunn School of Pathology in the laboratory of and was one of the few immunologists who actually re-
James Gowans, the investigator who first definitively searched the immunology of infection. Herman readily
showed that the lymphocyte was the source of specific agreed to partner with me on the Manual, and so began
adaptive immunity. I was busily cannulating the thoracic a close collaboration that continued for three subse-
ducts of rats in order to harvest T lymphocytes when I was quent editions of the Manual, ended only by his untimely
informed that a transatlantic telephone call was coming death. The AAI also accepted an offer of collaboration
in. My first reaction was fear of bad news. Rather, it was and appointed a liaison committee to work with us.
a phone call from Earle Spaulding. I knew Earle as the We were off and running, but we had no idea of how
chairman of microbiology at Temple and active in the to proceed. There had never been a manual describing
Eastern Pennsylvania branch of the American Society for the entire laboratory practice of immunology. Part of our
Microbiology (ASM). He explained that he was calling as mission was to include the many applications of immu-
a member of the editorial group of the Manual of Clinical nology devoted to detection and analysis of a wide variety
Microbiology (MCM), at that time in its first edition. His of diseases, not only those induced by microorganisms.
particular concern was the chapter on immunology, which Should we approach the subjects disease by disease or
devoted 100 pages to various serologic tests for infectious method by method? We finally decided to compromise
organisms with no mention of noninfectious diseases. by beginning the book with invited chapters on the com-
Earle felt strongly that the field of immunologic diagnosis mon methods used in the immunology laboratory, then
was growing exponentially and deserved a separate, com- continuing with sections covering their application to
panion manual. The MCM editorial board agreed, provid- the main categories of disease. We included a final sec-
ing I was willing to accept the position of Editor-in-Chief. tion on laboratory administration and quality control.
I was delighted to receive the invitation. I had re- Having developed particular sections, we then sought
cently chaired a “blue ribbon” committee of the Amer- the most experienced and highly qualified individuals
ican Association of Immunologists (AAI) on the future to serve as section editors. Because of the cross-cutting
of clinical immunology. We concluded that there was no matrix arrangement, there was major concern that some
space for a new patient-centered clinical specialty, but topics would be dealt with twice or even three times. We
great need for improved, expanded laboratory support. A therefore decided to organize a “stakeholders meeting,” at
comprehensive manual would serve as a great stimulus which all of the section editors met at ASM in Washing-
to the whole field of laboratory-based clinical immunol- ton, DC, with proposed outlines of their sections. Going
ogy. I accepted the offer with two qualifications. First, I through each one systematically, we identified topics
needed a co-editor, particularly someone well versed at a where overlap occurred and ensured that everything im-
practical level in immunology related to infectious dis- portant was included once, but not more. We also made
eases. Second, I asked that such a manual be cosponsored a fundamental decision that the book would be complete
by the AAI. Both qualifications were agreed to by the and free-standing. The methods would be described in
ASM Publications Board. sufficient detail that the laboratory worker could actually
The person I had in mind as co-Editor-in-Chief was prepare the materials, perform the tests, and interpret the
Herman Friedman. I knew Herman from contacts aris- results without consulting other references. It should be
ing from our joint interest in allergy research. I knew understood that, at that time, most laboratory reagents
xxiii
xxiv ■ Foreword: How It Began
were prepared within the laboratory and were generally care of patients with infectious malignant inflammatory
not available as commercial kits. This format required that and immune-mediated disorders. With the ready avail-
we keep descriptions terse and the reference lists short. ability of validated kits, the job of the clinical laboratory
When the first edition of the Manual of Clinical Im- immunologist has shifted toward working with clinical
munology was published in 1976, we felt it warranted colleagues on the significance and interpretation of lab-
some type of celebration. Herman suggested that we oratory tests.
should organize a meeting to mark the birth of the book I’m proud to have been involved in the genesis of
and to bring together the leaders in clinical laboratory this Manual. It would not have been possible without
immunology, including our authors and section editors. the continued support of ASM, the cooperation of
Eventually, this led to the formation of the Association AAI, the persistence of succeeding volume and sec-
of Medical Laboratory Immunologists and the American tion editors, the contributions of hundreds of practic-
Board of Medical Laboratory Immunology. ing clinical laboratory immunologists, and the foresight
The Manual continues to be published at regular in- of a few visionary microbiologists of the 1970 era who
tervals to the present, as the editorial lineup has evolved. realized that immunology had become a discipline and
Barbara Detrick and Robert G. Hamilton joined me as specialty of its own. It never would have happened if
Editors for the Sixth Edition, and Dr. Detrick has con- Herman Friedman had not joined with me in accepting
tinued to lead the Manual for the Seventh and the pres- the challenge. I hope that he will long be remembered
ent Eighth Edition. I hope the series will go on for many for his numerous contributions to immunology.
years. Although the Manual’s name has changed and the NOEL R. ROSE, MD, Ph.D.
format is altered, the overall aim is still to improve the
Preface
For over 40 years, the Manual of Clinical Laboratory Im- New chapters have been introduced to highlight these
munology has been the leading reference source, both in changes. For example, section D, Flow Cytometry, de-
the United States and abroad, to advance the field of lab- scribes the latest applications of these techniques, such as
oratory immunology, to foster the best contemporary and polychromatic flow cytometry and mass cytometry; sec-
most cutting-edge methodologies, and to translate basic tion F reviews fresh information on the clinical applica-
immunologic principles into appropriate laboratory tests. tions of cytokines and chemokines; the infectious disease
Since the publication of the 7th edition of this Man- sections H, I, and J include the newest strategies used
ual, remarkable progress has been made in the field of in infectious disease diagnosis and treatment, including
immunology, and these notable advancements have been the HIV and syphilis algorithms; section K, Immunode-
reflected in the clinical immunology arena as well. The ficiency Diseases, presents the recent newborn screening
scope of clinical immunology is exceptionally broad and programs for severe combined immune deficiency; and
encompasses nearly every medical specialty, including section P, Transplantation Immunology, outlines the
such areas as transplantation, rheumatology, oncology, usefulness of next-generation sequencing in the human
infectious disease, allergy, hematology, and neurology, to leukocyte antigen (HLA) laboratory.
name a few. Because of its strategic position in the hos- Once again, this Manual is offered not just in print but
pital setting, it is critical that the clinical immunology also electronically as either an EPUB file or a PDF. This
laboratory should have a guide to follow with regard to special feature will allow a larger audience to review and
accurate and appropriate laboratory procedures. As the use the Manual.
field of clinical immunology continues to expand, we As we produce the 8th edition of this Manual, it is
look to the laboratory director as a key person to gather appropriate to celebrate its success. Noel Rose, the Man-
the new basic information and integrate it into useful ual’s first Editor-in-Chief, has provided a foreword reflect-
clinical procedures as well as to serve as a pivotal contact ing on how the field has changed over the past 5 decades.
for communication with the various disciplines. In ad- Since the publication of this Manual is a joint effort
dition to keeping abreast with the most updated testing of many dedicated individuals, I wish to acknowledge the
systems, the goal for this Manual is that it must not only outstanding commitment and invaluable support of our
serve the needs of today’s clinical immunology laboratory volume editors, section editors, and chapter authors, all
but also look to the future, where even more dramatic of whom, as internationally renowned experts in their
progress in diagnosis and treatment can be anticipated. areas, have contributed their extraordinary experience,
In an effort to capture the new dimensions in this field energy, and time to the success of this edition. Also, I
and to reflect the continuous evolution of clinical im- would like to extend my appreciation to the ASM edi-
munology, significant changes have been introduced into torial staff, in particular Ellie Tupper, Senior Production
the 8th edition of the Manual of Molecular and Clinical Editor, and Christine Charlip, Director, ASM Press, who
Laboratory Immunology. Several sections of the Manual have provided their valuable experience and support to
have been notably updated to reflect the latest labora- complete this edition.
tory approaches in molecular assays as well as the shift to BARBARA DETRICK, Ph.D.
automated testing, kit-based diagnostics, and new tech- Editor in Chief
nical tools: themes that are carried throughout the book.
xxv
Author and Editor
Conflicts of Interest
Cem Akin (coauthor on chapter 85) has consultancy agree- the Children’s Hospital of Philadelphia and is an employee of
ments with Novartis and Patara Pharma and receives research Thermo Fisher Scientific, Transplant Diagnostics.
funding from Dyax.
Robert P. Lisak (coauthor on chapter 99) is on an advisory
Barbara Detrick (Editor in Chief, coauthor on chapter 106) board for Syntimmune.
serves as a consultant to Siemens Healthcare Diagnostics, Inc.,
Abbott Laboratories, and INOVA Diagnostics, Inc. Dimitri Monos (coauthor on chapter 113) receives royalties
from Omixon. Omixon has licensed the protocol we developed
Deborah Ferriola (coauthor on chapter 113) receives royal- for HLA typing by NGS from the Children’s Hospital of Phila-
ties from Omixon. Omixon has licensed the protocol we de- delphia and makes it available as a commercial product named
veloped for HLA typing by NGS from the Children’s Hospital “Holotype HLA.” Omixon is mentioned in this chapter as a
of Philadelphia and makes it available as a commercial prod- company that provides software analysis tools for the genotyp-
uct named “Holotype HLA.” Omixon is mentioned in this ing of HLAs using NGS data. It is not mentioned as a company
chapter as a company that provides software analysis tools for that commercializes HLA typing products/kits, because at the
the genotyping of HLAs using NGS data. It is not mentioned time of writing Omixon had not developed this activity.
as a company that commercializes HLA typing products/kits,
because at the time of writing Omixon had not developed this Stanley J. Naides (chapter 62) is a full-time employee of Quest
activity. Diagnostics Nichols Institute and receives a salary, stock, and
stock options from Quest Diagnostics.
Marvin J. Fritzler (coauthor on chapter 88) has been a consul-
Timothy Niewold (coauthor on chapter 38) has received re-
tant to or received research gifts in kind from Inova Diagnostics
search grants from Janssen Inc. and EMD Serono Inc.
Inc., Euroimmun GmbH, Mikrogen GmbH, Dr. Fooke Labora-
torien GmbH, ImmunoConcepts, GSK Canada, Amgen, Roche,
Maurice R. G. O’Gorman (chapter 20) is a BD Biosciences
and Pfizer. He is the Director of Mitogen Advanced Diagnostics
consultant and contractee.
Laboratory.
Paul Sikorski (coauthor on chapter 114) is an employee of
Andrea Illingworth (coauthor on chapter 18) has received
One Lambda, Inc., a Thermo Fisher Scientific brand.
unrestricted Educational Grant funding and speaker honoraria
from Alexion Pharmaceuticals.
Marek Smieja (coauthor on chapter 63) has done small studies
with Copan and GenMark.
Michael Keeney (coauthor on chapters 18 and 19) is a consul-
tant for Beckman Coulter, Canada, and Alexion Pharma, Can- Melissa R. Snyder (chapter 103) participates on the Strategic
ada. He has received unrestricted Educational Grant funding Advisory Committee with INOVA Diagnostics.
and speaker honoraria from Alexion Pharmaceuticals.
Kathleen E. Sullivan (section editor) is a Baxter grant recipi-
Masataka Kuwana (chapter 91) holds a patent on an anti- ent and an Immune Deficiency Foundation consultant.
RNA polymerase III antibody measuring kit.
D. Robert Sutherland (coauthor on chapters 18 and 19) has
Curt Lind (coauthor on chapter 113) receives royalties from received speaker fees and consulting fees from Alexion Phar-
a licensing agreement between Omixon Biocomputing and maceuticals.
xxvii
xxviii ■ LABORATORY MANAGEMENT
Yi-Wei Tang (coauthor on chapter 57) has received research Seattle Genetics and Amgen and honoraria from Abbvie for
funds from Roche Molecular Dignostics and the Luminex Cor- Advisory Board participation.
poration.
Andrea A. Zachary (coauthor on chapter 114) is a consultant
Brent Wood (coauthor on chapter 22) has received research for BiologicTx and Genentech and is a Scientific Advisory
funding and honoraria for Advisory Board participation from Board member for Immucor.
General Methods
A
section
SECTION EDITOR: THOMAS A. FLEISHER
1 Introduction / 3
THOMAS A. FLEISHER
2 Molecular Methods for Diagnosis of Genetic Diseases Involving the
Immune System / 5
AMY P. HSU
3 The Human Microbiome and Clinical Immunology / 19
FREDERIC D. BUSHMAN
4 Protein Analysis in the Clinical Immunology Laboratory / 26
ROSHINI SARAH ABRAHAM AND
DAVID R. BARNIDGE
Introduction
THOMAS A. FLEISHER
1
The “-omics” revolution has begun. Since the publication a number of newly characterized immune disorders. This
of the previous edition of Manual of Molecular and Clin- trend will continue: currently based primarily on genomic
ical Laboratory Immunology, there have been major tech- applications focused on exome sequencing, but within the
nological advances that are facilitating the application of very near future the approach will move to whole-genome
genomics, proteomics, and microbiomics to better under- sequencing. The technology for next-generation sequenc-
stand human health and disease. These disciplines are truly ing continues to be refined, and there is every reason to
complementary, as genomics is directly linked to proteom- believe that faster, less expensive, and more accurate whole-
ics, and these new technologies are providing an improved genome sequencing will result. Chapter 2 by Hsu provides
understanding of the functional consequences related to an overview of the standard Sanger-based sequencing, with
alterations in the genome that lead to changes in the pro- clinical examples to help exemplify successful application
tein products of genes. The major advancements in next- of this method. This chapter also provides an overview of
generation sequencing have made characterization of the next-generation sequencing and reviews clinical applica-
human microbiome a reality. Major efforts in microbiomics tions of this methodology. More-extensive discussions of
currently are focused on evaluation of the human bacterial next-generation sequencing applied in health and disease
community in health and disease. These three disciplines can be found in reviews by Gonzaga-Jaurequi et al. (1) and
find common ground in the field of immunology, in which Manolio et al. (2). A newer component of the technolog-
the genome provides the triggering instructions for protein ical revolution in biologic systems is focused on evaluating
products that are critically important for immune function the microbial constituents—the microbiome—that are the
and host defense. Alterations in the genetic code are now commensal “partners” of every living host. Chapter 3 by
known to contribute to an ever-growing list of defects in Bushman provides a clear overview of the basic principles
immune function that result in susceptibility to microbial that have propelled the field of microbiomics and how this
disease. In addition, altered immune function leading to is likely to provide insight into understanding human health
inflammatory disease will be further clarified with these and disease. A recent review focusing on the microbiome
powerful new tools. Early lessons in the study of the human of the human gastrointestinal tract provides additional in-
microbiota have revealed that commensal bacteria interact sights into this important and emerging field (3).
directly with the immune system to aid and model immu- The combination of genomics, proteomics, and micro-
nity. They also provide advantage to the host in settings of biomics, three complementary disciplines, has substantial
health and contribute to pathogenesis in circumstances of impact on the field of clinical immunology. Discovery
disease. The future of clinical immunology at a diagnostic of new genes provides insight that is further clarified by
and therapeutic level increasingly will rely on these tech- studying the protein products of these genes both at a ba-
niques and further advances that will evolve more and sic characterization and a functional level. The develop-
more rapidly. ments in the field of microbiomics are revealing that the
The laboratory methods available to evaluate proteins environment of each individual makes significant contri-
and the nucleotide code form the substance of chapters butions to the immunologic system and that perturbations
2 and 3. Chapter 4 by Abraham and Barnidge focuses on among our microbial “partners” can significantly influence
specific protein quantitation using current qualitative and human health and disease. Familiarity with these devel-
quantitative methods as well as the applications of mass oping areas is necessary as they move more and more into
spectrometry to generate a more accurate reflection of the mainstream laboratory testing linked to patient diagnosis
constituent protein components in a sample. The promise and care.
of new biomarkers that may aid in diagnostic and/or ther-
apeutic monitoring has yet to be realized, but advances
in proteomics technology suggest that these applications REFERENCES
should become a reality in the near future. The field of ge- 1. Gonzaga-Jaurequi C, Lupski JR, Gibbs RA. 2012. Hu-
nomics has exploded with readily available next-generation man genome sequencing in health and disease. Annu Rev
sequencing techniques that have clarified genetic causes for Med 63:35–61.
doi:10.1128/9781555818722.ch1
3
4 ■ GENERAL METHODS
2. Manolio TA, Chisholm RL, Ozenberger B, Roden DM, genomic medicine in the clinic: the future is here. Genet
Williams MS, Wilson R, Bick D, Bottinger EP, Brilliant Med 15:258–267.
MH, Eng C, Frazer KA, Korf B, Ledbetter DH, Lupski 3. Tyler AD, Smith MI, Silverberg MS. 2014. Analyzing the
JR, Marsh C, Mrazek D, Murray MF, O’Donnell PH, human microbiome: a “how to” guide for physicians. Am J
Rader DJ, Relling MV, Shuldiner AR, Valle D, Weinshil- Gastroenterol 109:983–993.
boum R, Green ED, Ginsburg GS. 2013. Implementing
Molecular Methods for Diagnosis of Genetic
Diseases Involving the Immune System
AMY P. HSU
2
Disorders of the immune system affect a significant number PCR
of individuals, with prevalence estimates (per 100,000) de- One of the fundamental techniques in molecular genetics
termined by registries ranging from 5.6 in Australia (1) to is the use of PCR. This technique allows the specific am-
4.979 in France, 2.6 in The Netherlands, and down to 1.377 plification of a segment of DNA defined by the forward
in Germany (2). Accurate diagnosis of immune system dis- and reverse primer pair used in the reaction. The double-
orders may allow early intervention prior to extensive ill- stranded DNA template is denatured, and oligonucleotide
ness for severe disease, or more specific treatment in the case primers in the 20-to 35-bp range anneal to the template at
of later-diagnosed or milder disease. This chapter explores the location complementary to the sequence of the primer.
some of the current molecular methodologies for detecting A polymerase enzyme is then used to add individual deoxy-
disorders of the immune system and highlights specific pit- nucleotides to the 3′ end of the primer to continue copying
falls that may hinder accurate diagnosis. the template molecule. After a specified time, the tempera-
ture rises to denature the template-copy strand, and both
then become available for primer binding. The cycle then
SAMPLES repeats itself, resulting in an exponential amplification of
Study of the immune system is aided by the accessibility of the target region.
an appropriate sample. A single 8-cc tube of blood may be Sanger sequencing has been the gold standard for genetic
lysed to extract DNA and/or RNA from all the blood cells diagnosis in recent decades. The technique is similar to stan-
together. Alternatively, whole blood may be processed to dard PCR; however, only one primer is used, either forward
separate peripheral blood mononuclear cells (PBMCs) from or reverse, and included in the PCR mix is a percentage of
granulocytes, followed by DNA or RNA isolation from a fluorescently labeled dideoxynucleotides (ddNTPs). When
subset of cells. If obtaining a blood sample is not possible, the polymerase incorporates a ddNTP, the extension ceases,
then small amounts of DNA may be isolated from buccal resulting in a fragment labeled with a fluorochrome depen-
swabs, tissue biopsy samples, saliva, or cell lines such as fi- dent on the identity of the final nucleotide. This is repeated
broblasts or Epstein-Barr virus-transformed B-cell lines. A through multiple cycles. The resulting product is cleaned
secondary sample may be utilized if the patient has already to remove unincorporated, labeled ddNTPs and then elec-
undergone hematopoietic stem cell transplant, in which trophoresed through a polymer-filled capillary to separate
case a non-hematopoietic-cell sample is required. Buccal the fragments based on size. The resulting chromatogram
swab samples are easily obtained, are room temperature is actually a display of time on the x axis versus intensity
stable, and may be shipped to a testing lab. DNA from a on the y axis, with the color of the peak being the third
buccal sample is often used for diagnosing family members dimension. Because smaller fragments migrate through the
of an affected individual. Care must be taken, however, if a polymer faster, they are detected first and longer fragments
patient has already received a bone marrow transplant be- are seen later (Fig. 1).
cause the cells obtained from a buccal sample may contain Peak heights from one base to the next are not necessar-
a large percentage of neutrophils and thus produce a false- ily even; there is some variability within the local sequence,
negative result. DNA extraction involves lysing of the nu- and certain base combinations result in higher or lower
cleated cells either mechanically or with a detergent-based peaks. In general, however, the shorter fragments seen at
lysis solution, removal of lipids and proteins, followed by the beginning of the sequence readout have higher signal
DNA purification through alcohol precipitation or column strengths and well-defined peaks (Fig. 1A), while larger,
capture. RNA isolation is similar, although care must be later fragments have shorter, broader peaks and less reso-
taken to avoid RNA degradation by RNases. There are lution in homopolymer runs, seen in the three red peaks in
numerous off-the-shelf kits available for small-scale DNA the middle panel. Good-quality sequence should have well-
and RNA isolation, as well as instruments for medium-to defined peaks with little to no baseline noise compared with
large-scale isolation. the lower tracing in Fig. 1A, in which the baseline is high
doi:10.1128/9781555818722.ch2
5
FIGURE 1 Sanger sequence chromatograms. (A) Normal sequence displaying a single base call at each
location. The top row demonstrates sequence from early in the chromatogram, ~150 bp into the sequence,
while the middle row is the same location but near the end of the chromatogram, at ~750 bp. The bottom
chromatogram shows readable sequence but significant nonspecific baseline noise, making interpretation less
than ideal. (B) An example of wild-type sequence compared to a heterozygous mixed base. The wild-type
blue (C) peak has decreased in height, and the red (T) allele overlays the same location. (C) An example of
wild-type sequence (top tracing) compared to 4-bp duplication (bottom tracing). The bases listed below the
tracing demonstrate how to determine the variation causing the heterozygous peaks. Bases in black are wild
type, while bases in red signify the mutant allele. Boxed bases correspond between wild type and mutant,
allowing identification of the inserted bases indicated by a bracket.
with apparent nonspecific sequence. Poor sequence quality nucleotides in the heterozygous patient versus two copies
renders interpretation difficult. of the same nucleotide in the homozygous patient. Figure
1B shows the control sequence on top, while the lower
sequence has a decreased blue peak corresponding to the
SANGER SEQUENCE ANALYSIS wild-type C allele and the presence of a red T peak as well
The product of a Sanger sequence reaction is a series of at the same location. This is a mutation in the coding re-
fragments of varying size with the final base labeled with an gion of GATA2 at cDNA 1081 that results in an amino acid
identifying fluorochrome. Each fragment is the product of change. It is named according to the Human Genome Vari-
one allele, while the resulting chromatogram is the sum of ation Society nomenclature (3) as c.1081C>T. It indicates
the fragments. Patients heterozygous for a single nucleotide a change at base 1081 of the cDNA using the initiating
polymorphism (SNP), a variant occurring in the general ATG as base 1, and that the wild-type or major allele has
population that is usually not pathogenic, or a point muta- cytosine while the mutation or minor allele has thymidine
tion, a change from one nucleotide to another that is patho- at that site. The protein nomenclature, p.R361C (preceded
genic and often specific to a patient or related individuals, with a “p”), shows that the wild-type arginine at amino acid
have a chromatogram showing individual peaks with one position 361 has been replaced with a cysteine.
site containing two peaks of roughly equal height. When In addition to single nucleotide changes, Sanger sequenc-
compared with a chromatogram from an individual who is ing also detects small insertions or deletions. These may
homozygous at the variant site, the peaks have decreased range from one to hundreds of nucleotides, but they must be
amplitude, because the signal is being split between two wholly contained within the PCR product used as a sequence
2. Molecular Methods for Diagnosis of Genetic Diseases ■ 7
template to be detected. In the case of Fig. 1C, two sequences sequencing. The library preparation allows for amplified
align, with the left-hand portion being identical between the input of numerous target regions compared with the single
control and patient chromatograms. The tracings change amplified fragment for Sanger sequencing. Coupled with
from clean, individual peaks to double peaks in the patient. the high-volume capture of the three instruments, it is easy
This has a very different appearance from the tracing in Fig. to see the appeal for clinical usage. As of September 2013,
1A with poor sequence quality. In this case, the peaks are the capital costs of the instruments ranged between $65,000
still well defined, with no additional baseline. To determine and $125,000, with the cost increased for a laboratory in-
the mutation resulting in the double sequence, it is helpful to terested in higher throughput needing ancillary instrumen-
write the bases on two lines, using the top line for the wild- tation such as robotics to take advantage of the capacity
type sequence seen in the top tracing and the mutant bases of the machines (4). Each instrument has its own advan-
underneath seen in red. When the two sequences are com- tages. Which platform is optimal for any given laboratory
pared, it becomes apparent that they differ by the insertion is based on the currently available configurations and the
of the bracketed red bases, GACC, which is a duplication of goals of the laboratory. Capacity will continue to increase at
the four preceding bases. After the duplication, the sequence an astonishing rate, making comparisons virtually obsolete
matches the wild type, as shown by the boxed bases that are by the time they are published; however, recently reported
the same between wild type and mutant. Occasionally when performances for the three main platforms (5) are shown
the mutant allele is written out, it is determined to be a com- in Table 1. The broad conclusions drawn at the time of the
plex mutation containing both an insertion and deletion. analysis are that the MiSeq has the highest throughput and
the lowest substitution error rate, at 0.1 per 100 bases; the
454 GS Junior has the longest reads; and the Ion Torrent is
the fastest, with the highest reads per run, although it is less
NEXT-GENERATION SEQUENCING accurate when calling homopolymers of any length, with
While Sanger sequencing is still the gold standard for mo- accuracy declining as the homopolymer stretch increases.
lecular genetic diagnostics, decreasing cost and increased Data obtained from NGS need to undergo stringent
throughput are driving the use of next-generation sequenc- analysis to obtain variant calls. There are numerous soft-
ing (NGS). A further impetus for the use of NGS is the ware packages available online or as part of genomics soft-
identification of disease genes in rare diseases in which only ware suites (reviewed in reference 6 and elsewhere). While
a few individuals are diagnosed with mutations in the gene. the choice of software is individual, all follow the same basic
With targeted gene sequencing, the patient’s phenotype pipeline. For this discussion, the publicly available Genome
or infection is used to suggest candidate genes to evaluate Analysis Toolkit (GATK) pipeline is used as a model (7, 8).
for mutations. For narrow, well-defined phenotypes such as Because the reads are much shorter (100 to 500 bp) than
infection with a mycobacteria species, this approach has those obtained by Sanger sequencing (800 to 1,000 bp) and
identified ~80 to 90% of patient disease genes (S. M. Hol- the read outputs are pooled so they come from mixed loca-
land, personal communication), while in large-scale new- tions, the reads must be mapped to a framework. In the case
born screening, targeted sequencing has identified specific of patient samples, this would be the current human genome
genetic lesions for T-B+ SCID in ~85% of patients (J. M. reference sequence. Each read aligns to a region, resulting
Puck, personal communication). In other cases with un- in a pileup of fragments per region—this provides the read
usual presentations, pursuing Sanger sequencing of candi- depth or coverage of a given target sequence. Once the data
date genes can become expensive and time-consuming and are mapped, the alignment is converted to the standard
still not identify a causative genetic lesion. Many labora- Sequence Alignment/Map (SAM) format (9), containing
tories are moving to NGS platforms as research tests, with chromosome, start and stop bases of the alignment, strand,
suspected mutations confirmed by Sanger sequencing. mismatch information, and probability scores. The SAM
There are currently three main benchtop platforms format can be converted to a compressed, indexed binary
available: the Illumina MiSeq, the Ion Torrent PGM, and format called BAM, allowing for fine-tuning of the align-
the Roche 454 GS Junior. Each instrument is capable of ment. The mapped reads are filtered to remove duplicate
generating sufficient data to perform exome sequencing or reads, preventing overrepresentation of a variant or wild
targeted resequencing of a defined panel of multiple genes, type due to repeat sequencing of one clone from the library
ranging from tens to a few hundred. All of the platforms preparation. Next, insertions and deletions (indels), which
have a workflow that includes fragmentation of the input may align differently depending on the size of the indel and
sample, size selection, cleanup, and adapter ligation. The the strand sequenced, can be made consistent across all
MiSeq then undergoes a cluster generation and sequencing the reads in a region through the use of a realignment tool.
on the instrument, while the 454 GS Junior and PGM have This allows more confident detection of small indel varia-
an emulsion PCR, bead recovery, and enrichment prior to tions. Lastly, just as individual peak heights vary in Sanger
TABLE 1 Comparison of three major NGS platforms

Price comparison Run comparison
Minimum Mean Indel error
Platform Assembled contig
Cost predicted Run Mb per read frequency
Cost/Mb Total bases alignment to
per run throughput time (h) h length (per 100
reference
(Mb) (bases) bases)
454 GS Junior $1,100 1,135 18 $31.00 14.4 171 Mb 522 0.38 3.72% unmapped
Ion Torrent PGM $1,425 1,100 13 $14.25 33.3 300 Mb 122 1.5 4.6% unmapped
(316 chip)
MiSeq $1,750 1,500 27 $10.50 55.5 301.6 Gb 141 <0.001 3.95% unmapped
sequencing due to nucleotide neighbors, there is a similar qPCR

variation in sequencing by synthesis. The quality scores of An alternative method for measuring RNA transcripts is the
the alignment need to be recalibrated using base quality use of quantitative RT-PCR (qPCR), in which the cDNA is
score recalibration to more accurately reflect the proba- amplified with a set of primers and the production of product
bility of mismatch to the reference. These scores are used is measured in “real time” via a scanning detector. This can
in downstream variant calling that may identify a disease- be accomplished with the use of an intercalating dye such as
causing mutation. At this point, the data are a BAM file SYBR green that will bind to double-stranded product and
that can be utilized for variant calling. produce a fluorescent emission that accumulates over succes-
The goal of sequencing is to identify regions differing sive cycles. Alternatively, a fluorescent reporter probe can be
from the reference sequence. In Sanger sequencing, this is used that binds between the two amplification primers. The
a visual process in which a peak is altered in height or fluo reporter probe contains both a reporter and quencher mole-
rochrome detected, as seen in Fig. 1, and the investigator cule, which prevents fluorescence while the two are in close
can decide if the variation is real or an artifact of the se- proximity. As extension from the amplification primer occurs,
quence reaction. In a similar manner, the BAM file needs to the exonuclease activity of the polymerase causes degrada-
be analyzed to determine appropriate variants and sensitive tion of the probe separating the quencher from the reporter,
enough not to miss calls, but specific enough to weed out allowing emission that is detected during scanning. As with
most artifacts. There are several online tools available for SYBR green, as subsequent cycles of PCR occur, the inten-
variant detection, with new methods being published fre- sity of the signal from the released fluorophore increases. The
quently (reviewed in reference 10). Here we use GATK’s output of a qPCR assay is a logarithmic plot of fluorescence
HaplotypeCaller as an example. The BAM file is analyzed intensity versus cycle number (Fig. 2). Analysis of a qPCR
for local regions of possible variation, and those regions un- assay is performed by quantifying each well to determine at
dergo a de novo alignment to fine-tune the site. The local re- which cycle the fluorescence exceeds a threshold value (CT).
alignment can determine both single nucleotide variations Quantifying input copy number may be performed through
as well as indels. The output of a variant detection script the use of a standard curve with a titrated sample of known
is a VCF file that provides a list of all changes identified concentration and plotting of the experimental concentra-
and verified in the data set. As should be apparent at this tion (CT) on the standard curve. Alternatively, a relative
point, unlike Sanger sequencing, the labor-intensive por- quantitation may be obtained by comparing the CT of the
tion of generating NGS is not the bench science but the target gene to a reference gene, usually a housekeeping gene.
computational and bioinformatics demand of data analysis. The difference in threshold cycles between target and ref-
As pointed out by McCourt et al. (11), moving forward, mo- erence transcripts provides a measure termed the ∆CT. The
lecular diagnostic laboratories will have a basic requirement difference in the ∆CT between the experimental (treated or
for molecular diagnostic bioinformaticians. disease) and reference (untreated or healthy control) sam-
ples provides the ∆∆CT. Because PCR provides exponential
RT-PCR amplification, the final calculation for quantification of input
becomes 2–∆∆CT. The derivation of this formula is well cov-
Some mutations result in alteration of the primary RNA
ered by Livak and Schmittgen (12).
transcript, leading to errors in splicing, polyadenylation, or
mRNA stability. Not all of these variants are detected by
genomic sequence of the coding exons. In cases in which a
patient’s phenotype strongly suggests a specific gene, anal- TRECs
ysis of the RNA transcript may provide an answer. There T-cell excision circles (TRECs) are formed during T-cell re-
is an extra step, however, in processing the patient sample. ceptor V(D)J recombination and are biomarkers for newly
Because genomic DNA is double stranded, the use of two produced naive T cells (13). Once this residual, circular
primers, one forward and one reverse, leads to exponential DNA fragment is produced in an immature thymic T cell, it
amplification of the target region. RNA, on the other hand, is not replicated during cell division. This dilutes the TREC
is single stranded and must first be made double stranded concentration as naive T cells migrate to the periphery
by reverse transcriptase PCR (RT-PCR). This is performed and expand in response to antigen. While relative quan-
from total RNA using an oligo(dT) primer that will bind titation is useful for looking at expression levels of target
to the poly(A) tail of the mRNA. This process results in transcripts, genomic qPCR was adopted to quantitate total
double-stranded mRNA, with random hexamer primers TRECs from newborn babies (14). It has become the gold-
that will bind to any RNA molecule with the appropriate standard method for performing this analysis (15). Low or
complementary hexamer or with a gene-specific primer that absent TRECs are a phenotypic indicator of poor or declin-
will bind only to a specific target. Once the RNA has been ing thymic output and usually severe T-cell lymphopenia,
converted to double- stranded cDNA, amplification of a a marker of SCID or other severe defects in T-cell devel-
transcript of interest proceeds as in genomic DNA. opment. Many states are now screening for the presence
One measure of RNA is message abundance. This has of TRECs using dried blood spots obtained from all infants
been examined by the use of Northern blots, which entails in the United States at birth. Reports from Wisconsin (16,
running the RNA on an agarose gel to separate it by size. 17) and California (18) have documented genetic diagno-
It is then transferred to a membrane support and probed sis of infants who were initially discovered during newborn
with a complementary probe and detection label (either by screening for TRECs, allowing medical intervention prior
radioactivity or a biotin-streptavidin reaction). The blot is to the onset of illness.
then exposed to X-ray film or a capture screen, respectively.
Densitometry may be used to obtain a relative quantitation
of a specific transcript compared to a housekeeping gene, if ARRAYS
the blot is stripped and reprobed, or compared to a healthy Microarray-based comparative genomic hybridization was
individual’s control on the same blot. The initial gel or blot initially utilized to screen for variations within cancer sam-
image can also provide information regarding the size of the ples. It has been adapted by clinical laboratories, however, to
transcript or the presence of multiple splice forms. screen for variations in DNA copy number, both deletions
FIGURE 2 Real-time qPCR plot. (Left) Amplification plot of duplicate dilutions of a known standard TREC plasmid. The x axis is PCR
cycle number; the y axis is ∆Rn, which is Rn (the ratio of the reporter signal normalized to a constant fluorescent signal) minus the value of
Rn at baseline; the green line is threshold value. Threshold cycle (CT) is defined as the cycle number at which the amplification plot crosses
the designated ∆Rn threshold (blue dots). (Right) Logarithmic plot of the same standards as well as two duplicate unknown samples, x and
y, for which the CT value, and therefore copy number or concentration, can be derived by comparison with the standard curve. (Figure
courtesy of Jennifer M. Puck.)
and insertions of 200 to 500 bp or larger. High-resolution analysis software will highlight the region as an insertion or
comparative genomic hybridization, first described by Urban deletion, respectively, as shown in Fig. 3. Similar to DNA
et al. (19), uses oligonucleotide probes spanning targeted arrays, which look for genomic variation, microarrays may
genomic regions and spotted to a glass slide. DNA samples be used with total RNA from a specific tissue to examine
from a control and patient are fluorescently labeled, each differences in RNA transcript levels between two tissue
with a different dye, and the resulting labeled DNA is hy- types or healthy versus diseased tissues. Mapping increased
bridized to the array. After hybridization, the slide is scanned or decreased RNA transcripts may aid in defining pathways
and the fluorescence intensity of each dye signal is calcu- involved in disease pathogenesis.
lated for each oligonucleotide probe on the slide. In regions
where each sample hybridizes equally, the ratio of the signal
between the two samples will be 1 because each has the same ANALYSIS OF VARIATIONS
number of copies of the target region. If the patient sample Regardless of whether a variant is identified by Sanger se-
contains a deletion on one chromosome, then the signal quencing or NGS, once a change is recognized, then the
from that sample will be 0.5 times the signal of the control. consequence of the variant on the transcript or protein
Likewise, if there is a duplication of a region, the signal will needs to be evaluated to determine if the change may be
be increased. As the probes are tiled across the target re- disease related. Single base changes can be mapped to the
gion, the ratios are plotted on a log 2 scale, so equivalent consensus coding sequence to determine if an amino acid
ratios will be plotted on the zero axis. If multiple consecutive codon is changed. SNPs resulting in an altered codon for
probes have an aberrant signal, either high or low, then the the same amino acid as the consensus are referred to as
FIGURE 3 High-resolution comparative genomic hybridization array spanning 375,000 bases of the region surrounding CYBB on the X
chromosome. Plots are log 2 of the intensity of the patient sample compared to all samples run on the array. Panel A demonstrates one fe-
male patient with a large deletion encompassing LANCL3, XK, CYBB, and DYNLT3 as evidenced by the contiguous −0.5 intensity signal,
which then resolves to 0, indicating the presence of two alleles.
“synonymous,” while those causing amino acid substitutions results in only one in eight receptors being fully wild type.
are “nonsynonymous” SNPs. If a SNP is disease causing, This is insufficient to transduce the apoptosis signal (22, 23).
then a nonsynonymous SNP is termed a “missense mu- Transcripts containing a premature termination codon, such
tation,” while a base change that results in a termination that splicing occurs more than 50 to 55 bp after the mutation,
codon is referred to as a “nonsense mutation.” At a given usually result in nonsense-mediated decay, a mechanism by
nucleotide, patients may be homozygous, meaning their two which the cell degrades erroneous mRNA molecules (24; re-
alleles have the same nucleotide at the site; or heterozygous, viewed in reference 25). Loss of transcript or protein product
meaning they have two different bases at the site, indicating from one allele may be seen in autosomal recessive diseases,
the presence of two identifiable alleles. A third possibility is in which case a mutation should be seen on the second allele
that they may be hemizygous, meaning they have only one as well. Alternatively, in tightly regulated genes, loss of one
copy at that location. This is most often seen in the case allele may result in haploinsufficiency, in which half the level
of X or Y chromosome genes in males, although a deletion of protein product is not sufficient. Haploinsufficiency in im-
of one allele also results in hemizygosity. Sanger sequence mune disorders has been reported in autoimmune lympho
chromatograms from homozygous and hemizygous individ- proliferative syndrome due to insufficient FAS protein on the
uals are indistinguishable, so care must be taken especially cell surface (26), as well as in GATA2 deficiency leading to
when reporting X chromosome genetic variations. MonoMAC syndrome (27).
Identified variants need to be examined for suggested In addition to changes within the coding sequence,
deleterious effects in light of the phenotype of the patient many mutations occur at consensus splice sites immedi-
and possible inheritance patterns. Missense changes may ately flanking the coding exons. Traditionally, splicing has
be analyzed by examining the sequence homology within a been thought of in terms of the three core splicing signals:
protein family as well as across species, predicted structure- 5′ splice sites (5′ss), 3′ splice sites (3′ss), and the intronic
based alterations, mRNA and protein features, as well as branch point near the 3′ end of the intron. Figure 4 shows
conservation of the region identified by multiple alignments the conserved motifs associated with intron removal. Most
and phylogenetics. Historically, this was performed by align- easily identified are those changes altering the invariant
ing protein sequences across species or within the protein GT or AT splice sites immediately flanking the exon. How
family to estimate conservation of the specific amino acid. ever, the entire motif may be critical to appropriate splice
Examining amino acid properties of the substituted amino osome formation. Wang and Burge (28) and Ward and
acid versus the wild type could strengthen the prediction. Cooper (29) provide reviews of the motifs for recognition
Currently, there are numerous bioinformatics analysis pro- of splice sites and exon identity as well as components of
grams available online to predict whether a change is dele- the spliceosome. For example, a mutation at the first or
terious. Several of these programs are listed in Table 2. The last base of an exon, even if predicted to be synonymous,
left-hand section of the table lists analyses that may be per- often leads to loss of correct splicing of that exon, as do
formed on single changes, while the two columns farthest to mutations at the +5 site of an intron. For introns with
the right highlight two programs, wANNOVAR (20) and short polypyrimidine tracts, a mutation that changes one
dbNSFP (21), for higher-throughput data such as NGS. The of the pyrimidines to a purine may result in exon skipping,
table shows which analyses the programs include. The sec- as seen in DOCK8 (dedicator of cytokinesis 8) deficiency
ond group is especially helpful for filtering the large number (30). In addition to the three core splicing signals, there
of variations identified with NGS. is an increased recognition of splicing regulatory elements
Nonsense mutations, those resulting in a premature ter- (SREs) that occur in both exons and introns. SREs are
mination codon, can have different effects depending on the composed of elements that enhance or inhibit splicing and
location within the transcript. Late stop codons, occurring in are named by their location and effect on the transcript;
the last exon or near the end of the penultimate exon, usually exonic splicing enhancers (ESEs), exonic splicing silencers
result in a stable mRNA transcript, which may be translated (ESSs), intronic splicing enhancers (ISEs), and intronic
into a truncated protein. The shortened protein may cause splicing silencers (ISSs) all act by recruiting factors to in-
disease in a dominant-negative fashion by interfering with crease or decrease exon recognition and spliceosome re-
partner chains. This has been seen in autoimmune lymph- cruitment. Many mutations predicted to result in missense
oproliferative syndrome, in which the shortened FAS pro- or nonsense substitutions are being shown to affect SRE
tein is capable of assembling into the stable trimer receptor. sequences and alter proper splicing of the exon (31). The
However, it is lacking the intracellular death domain critical third section of Table 2 indicates current websites available
for downstream signaling. The presence of one mutant allele to screen for splicing mutations.
FIGURE 4 Conserved splicing motifs. Gray boxes indicate exons; solid lines are introns. 3′ss begins
with the branch point located between 15 and 40 bases 5′ to the exon, noted in brackets, with the loose
motif of YNYURAY. Y represents pyrimidines (C or T); N is any base; U is uracil, the RNA equivalent of
T, R is A or G. Following the branch point sequence is the polypyrimidine tract, Y(n), where n ranges from
5 to 20 bases of mostly pyrimidines, followed by the 5′ss invariant AG (bold) as the last two bases. The
first base of the exon is usually a G. At the end of the exon, the final base is again usually a G, followed by
the invariant GT (bold) at the start of the intron. The 5′ss motif is GTRNG.
TABLE 2 Online resources for variation analysis

Contained in
Resource wANNOVAR
dbNSFP (21)
(20)
A. Analysis of missense changes
Program (reference) Information used Prediction model
PolyPhen2 (43) Eight sequence-based and three Naive Bayes classifier Yes Yes
structure-based predictive
features
SIFT (44) Sequence homology based on Position-specific scoring matrix Yes Yes
PSI-BLAST
MutationTaster (45) Conservation, splice site, mRNA Naive Bayes classifier Yes Yes
features, protein features
LRT (likelihood ratio test) (46) Sequence homology Likelihood ratio test of codon Yes Yes
neutrality
Mutation Assessor (47) Sequence homology of protein Combinatorial entropy formalism No Yes
families and subfamilies within
and between species
FATHMM (48) Sequence homology Hidden Markov models No Yes
SiPhy (49) Multiple alignments Inferring nucleotide substitution No Yes
pattern per site
GERP++ (50) Multiple alignments and Maximum likelihood Yes Yes
phylogenetic tree evolutionary rate estimation
phyloP (51) Multiple alignments and Distributions of the number Yes Yes
phylogenetic tree of substitutions based on
phylogenetic hidden Markov
model
B. Variation databases
Database Contents Website
dbSNP Reported genomic variations http://www.ncbi.nlm.nih.gov/ Yes No
SNP/
ESP6500 Allele frequency in 6,500 http://evs.gs.washington.edu/ Yes No
NHLBI ESPa exomes EVS/
1000G (52) Allele frequency in 1000 http://www.1000genomes.org/ Yes No
Genomes Project
C. Splicing signals analysis
Program Splicing signals analyzed Website
Human Splicing Finder (32) 5′ss, 3′ss, ESE, ESS, ISE, ISS, http://www.umd.be/HSF3 No No
mutation analysis
SROOGLE (53) 5′ss, 3′ss, ESE, ESS http://sroogle.tau.ac.il/ No No
Automated Splice Site and 5′ss, 3′ss, prediction of exon http://ossify.sg.csd.uwo.ca No No
Exon Definition Analyses definition
(ASSEDA)
ESE Finder (54, 55) ESE http://rulai.cshl.edu/cgi-bin/tools/ No No
ESE3/esefinder.cgi?
process=home
RESCUE-ESE (56) ESE http://genes.mit.edu/burgelab/ No No
rescue-ese/
PESX (57) ESE, ESS http://cubio.biology.columbia No No
.edu/pesx/pesx/
NHLBI ESP, National Heart, Lung, and Blood Institute Exome Sequencing Project.
a
was examined by flow cytometry. While the healthy control

DIAGNOSIS had a fluorescence index (FI) of 30.6, the patient had an FI
Bioinformatics tools may be utilized to screen variants iden- of only 12.6, demonstrating reduced levels of intracellular
tified by sequence. However, on some level, the effect of the NEMO protein. Full-length IKBKG cDNA was generated by
change needs to be demonstrated and correlated with the RT-PCR, revealing a smaller fragment in the patient than in
patient’s phenotype. If a male infant presents with SCID, the control sample. Sequencing of the PCR product revealed
lacking T cells and NK cells but having adequate numbers a 153-bp in-frame deletion corresponding to exon 5 contain-
of B cells, then the most likely diagnosis would be X-linked ing the c.597G>A change (Fig. 5). The 153-bp deletion is
SCID and IL2RG would be the first gene sequenced. A mu- expected to result in a stable mRNA that should be trans-
tation identified within the coding region that is predicted lated into a protein missing 51 amino acids. Using the Hu-
to be deleterious would fit the phenotype of the patient and man Splicing Finder website (32), the original c.597G>A
correlate with other, previously identified mutations. There mutation was predicted to cause both the loss of an exonic
are some special cases that need to be considered and will be splicing enhancer and the creation of an hnRNP A1 silencer
demonstrated by specific case studies below. motif. The exonic binding of a silencer in place of a splicing
A young male patient presenting with granulomas and enhancer has been recognized to mediate silencing of exon
conical teeth, often seen with NF-κB essential modulator sequencing (33). Mutations such as this, in which the protein
(NEMO) deficiency, was sequenced for IKBKG. Genomic se- is present but not fully functional, are termed “hypomorphic.”
quencing revealed an apparent heterozygous change in exon A second patient with infections related to NEMO de-
5, c.597G>A, which did not alter the encoded amino acid. ficiency, Haemophilus influenza pericarditis in childhood
Given the phenotype of the patient, which was strongly sug- and disseminated Mycobacterium avium as an adult, was
gestive of an IKBKG mutation, intracellular NEMO protein screened by genomic PCR. The patient was reported to be
A. B.
FIGURE 5 Analysis of NEMO variants. (A) Intracellular flow cytometry for NEMO protein. The dot-
ted line is isotype control; the solid line is specific NEMO staining. The top panel is the normal control,
showing an FI of 30.6. The middle panel is patient 1, with all cells showing specific staining but with a
reduced intensity and an FI of 12.6. The bottom histogram is patient 2, again demonstrating that all cells
contain NEMO protein, but less than both the control and patient 1, with an FI of 5.4. (B) The top panel
shows the genomic organization of IKBKG, the gene encoding NEMO. Gray boxes are exons, narrow
boxes are untranslated, and small white box insets are uORFs. Locations of patient genomic mutations
are noted, with wild-type sequence above the genomic structure, mutant motif below, and the single base
mutation in bold. The lower panel depicts the structure of the cDNA, showing aberrant splicing. Patient 1
has a 153-bp deletion of exon 5, noted by the thin line in the middle of the cDNA. Patient 2 has a deletion
of a portion of the 5′ untranslated region, including one of the two uORFs.
wild type for the coding exons of IKBKG. Again, due to the With the explosion of whole-exome sequencing, every varia-
phenotype of the patient, intracellular NEMO staining was tion may now be deposited into the NCBI curated database,
performed, which demonstrated reduced, but not absent, dbSNP, and be assigned an “rs” identifier number. Addition-
NEMO protein levels. The patient had an FI of 5.4, com- ally, there is an active effort by NCBI to extract variants from
pared with the 30.6 seen in the healthy control. Full-length the literature and enter them, which may result in errors due
cDNA was generated by RT-PCR, and Sanger sequencing to mutations later being retracted or corrected. Currently,
revealed a 110-bp deletion at the 3′ end of the noncoding many mutant alleles are present in dbSNP, albeit with a
exon 1. Genomic sequencing of the exon revealed a G>C low or absent population frequency, and automated analysis
mutation at the last base of the exon. This caused a loss of NGS variants returns dbSNP identifiers. Assuming that
of the 5′ss of intron 1 and usage of an alternative, cryptic presence in the “healthy” population excludes an allele from
5′ss contained within exon 1, resulting in the deletion seen being pathogenic, this may result in a disease-causing muta-
in the cDNA. Contained within the untranslated exon are tion being filtered out. Large-scale data analysis is a balance
two upstream open reading frames (uORFs). Disruption of between sensitivity and specificity, and care must be taken
uORFs has been shown to alter uORF-mediated control if a variant is identified in a logical candidate gene to not
of protein expression (34), suggesting that the deletion in dismiss it based solely on in silico predictions.
the 5′ untranslated region seen in this patient leads to his Sometimes patients are seen with mild phenotypes rem-
decreased intracellular protein level and that his disease is iniscent of a recognized disease. In this case, sequencing of
caused by insufficient levels of NEMO protein. the target gene should be performed and examined carefully.
A third male patient with possible NEMO deficiency There are numerous cases in the literature in which a mu-
was screened for mutations in the X-linked gene IKBKG. tation is identified in an affected individual and screening
Genomic DNA was isolated, and PCR amplification of the of the family members reveals one parent as being a mosaic
10 IKBKG exons was performed. Sanger sequencing demon- for the mutation, having a small percentage of cells with
strated a single nucleotide change, c.761C>T, causing the the mutation and the remaining cells being wild type. This
substitution of glutamine for arginine at amino acid posi- has been reported in hyper-IgE syndrome due to mutations
tion 254, p.R254Q. This mutation has been seen previously in STAT3 in two unrelated families. In each case, the mu-
in an unrelated family and is predicted to be deleterious by tation seen in the children was identified in PBMCs from
PolyPhen2 (unpublished data). Due to the genomic struc- the father but at low levels (37). In this case, the mutation
ture of the IKBKG locus, which is flanked by an inverted was identified by Sanger sequencing and the mutant peak,
pseudogene of IKBKG exons 4 to 10, the genomic mutation while heterozygous in the children, appeared as a minor
appeared to be heterozygous. The sequencing was repeated peak in the father (Fig. 6A). Had the father been sequenced
using RT-PCR performed on RNA isolated from PBMCs. in isolation, the mutation may have been overlooked as
The primers were designed to amplify the full coding se- nonspecific background. This is even though the wild-type
quence of IKBKG. When the resulting PCR product was peak is diminished in comparison with the normal control,
sequenced, only wild-type bases were seen throughout the as seen by the decreased height of the wild-type G (black)
coding region. This demonstrated that the mutation had peak when compared with the C (blue) peak preceding it.
occurred in the pseudogene and not in the transcript- The change in peak height is one sign that a portion of the
producing IKBKG locus. Changes in the coding region that sequence is derived from an altered nucleotide sequence
appear silent, mutations in the untranslated exons, and the rather than background nonspecific sequence.
presence of a pseudogene demonstrate the importance of A similar case is seen in a male patient with a presen-
performing cDNA analysis for IKBKG mutation screening. tation of leaky SCID, with reduced numbers of T cells and
There are some mutations that may be predicted to be NK cells but normal numbers of B cells. Mutation of IL2RG
benign. However, given the location within the protein or is the most common genetic defect in SCID. Screening
the clinical phenotype of the patient, they may warrant fur- revealed two peaks at c.260T>C, which caused the sub-
ther analysis. A patient presented with an IPEX (immune stitution of proline for leucine at amino acid 87 (p.L87P).
dysregulation, polyendocrinopathy, enteropathy, X-linked)- This mutation is predicted to be deleterious, and the pa-
like phenotype but without an identified mutation in tient’s mother is heterozygous for it, indicating germ line
FOXP3. Because there were other patients in this cohort inheritance. Closer examination of the sequence tracing,
with identified STAT1 mutations, all exons of STAT1 were however, reveals a small wild-type peak in the patient (Fig.
sequenced and a single base change was seen that resulted 6B). Because he has only one X chromosome, this change is
in an amino acid substitution within the coiled-coil domain suggestive of a reversion mutation—a change occurring in
of STAT1, p.V266I. All of the algorithms from Table 1 pre- somatic cells that results in correction or modification of the
dicted the amino acid change to be benign or neutral; how- mutation. In this case, it returned to the wild-type nucleo-
ever, there is a known mutation at the adjacent amino acid, tide. The most common single nucleotide change seen in
A267V, causing an autosomal dominant gain-of-function the human genome is a C>T substitution that occurs in the
mutation associated with disseminated coccidioidomyco- context of a CG dinucleotide. In this patient, the mutation
sis (35). Signal transduction studies performed using cells created a CG dinucleotide that subsequently underwent a
from the patient with V266I demonstrated that, similar to C>T reversion event in a lymphocyte precursor. This re-
other mutations in the coiled-coil domain, this patient had version allowed for the development of T cells and the leaky
increased and prolonged phosphorylation of STAT1 in re- phenotype observed (unpublished data). Both the mosaic
sponse to cytokine stimuli (36). and reversion patients demonstrate the importance of con-
Another class of mutations that may be initially dismissed sidering patient phenotype and family inheritance when-
by in silico analysis involves variations detected within a gene ever possible. Given errors in amplification by PCR enzymes
with autosomal recessive inheritance. To present with a dis- and natural occurrences of somatic variations, the low-level
ease phenotype, the patient requires two mutant alleles. Of- presence of the mosaic mutations and wild-type reversion
ten the parents will be healthy because they have one normal may be filtered out in exome sequence data.
allele and one mutant allele. Deleterious variations therefore The phenotype of the patient, those characteristics or
may be seen at low frequency in the general population. infections that bring him to seek medical attention, is a key
FIGURE 6 (A) Sanger sequence chromatograms from a patient demonstrating mosaicism for a STAT3
mutant allele. The left column shows coding sequence; the right column, the intronic SNP 132 bp 3′ to
the mutation but contained within the same PCR product. The top row is from a healthy control, demon-
strating the wild-type G at c.1145 and homozygous G at the intronic SNP. The middle row is from the
affected son, with heterozygous peaks at the site of the mutation, c.1145G>A, and homozygous G at the
SNP. The bottom row is from the father, showing a reduced peak height of the wild-type G allele (black)
compared with the healthy control and the presence of a small green peak corresponding to the mutant
allele (arrow) but at less than heterozygous levels. This is in contrast to his intronic SNP, which demon-
strates heterozygosity (arrow). (B) Sequence tracings of the X-linked IL2RG gene. The top tracing is from
a healthy control, showing only wild-type sequence. The middle tracing is from the patient’s mother, who
is heterozygous for the c.260C>T mutation, showing the reduced T peak (red) and the presence of the
mutant C peak (blue). The bottom tracing, from a male patient with the inherited mutant C allele from
his mother, shows a small, wild-type T peak (red) underneath (arrow).
factor when evaluating genes for mutations. The patient’s was generated by RT-PCR. Sequencing of the PCR prod-
clinical presentation often suggests a need for specific, tar- uct revealed an insertion in the cDNA between exons 3
geted gene or pathway evaluation using Sanger sequencing and 4 in one patient, between exons 5 and 6 in a second,
or whole-exome sequencing to discover variants. However, and between exons 6 and 7 in the third. Sequencing of
there are patients with a disease phenotype in whom no the genome between the two exons in each case revealed
mutation can be identified, even when all exons and cod- that the inserts were derived from intronic sequences and
ing and noncoding regions are sequenced. In this case, if were cryptic exons included due to a point mutation that
the phenotype of the patient matches a recognized dis- created a strong 5′ss. As seen in Fig. 7, the cryptic exons in
ease gene, then screening for variations in cDNA may be the first two patients resulted in a frameshift, which would
helpful. Three unrelated patients presented with infections indicate a loss of mRNA stability and subsequent protein
characteristic of chronic granulomatous disease (CGD). production. The third patient had an insertion of 120 bp
By the standard dihydrorhodamine flow cytometric assay that was not predicted to disrupt mRNA stability and re-
for CGD, all three patients had reduced activity. This sulted in a 40-amino-acid in-frame insertion (unpublished
was reflected by an abnormal neutrophil oxidative burst. data). Two additional patients with similar mutations have
Sequencing of CYBB, the most frequently mutated gene been reported by Noack et al. (38) and Rump et al. (39).
causing CGD, failed to reveal a causal variation. RNA was None of these mutations would be identified by standard
isolated from patients’ neutrophils, and full-length cDNA Sanger or whole- exome sequencing because the deep
FIGURE 7 CYBB genomic structure showing exons (black boxes) and introns (black line). Below is
intronic sequence from three patients, demonstrating the intronic point mutation (bold) in each that cre-
ates a cryptic 5′ss allowing for inclusion of a cryptic exon (white boxes). Wild-type sequence for each site
is shown below the patients’ mutant sequences. The insertion of the cryptic exon results in a frameshift
for the first two patients, while the third, with an insertion of 120 bp, is predicted to have a 40-amino-acid
insertion.
introns would not be screened again. This emphasizes the reduced level of GATA2 transcript or protein is sufficient to
need for cDNA analysis. cause disease. Among patients with phenotypic MonoMAC
One final example comes from a group of patients with syndrome, there were several without identified mutations
MonoMAC syndrome (40), which is an autosomal domi- after all exons and flanking splice sites had been sequenced.
nant disease caused by mutations in the hematopoietic stem Similar to the patients without identified CYBB mutations,
cell transcription factor GATA2. Missense mutations have full-length GATA2 cDNA screening was performed. In two
been identified in the C-terminal region of the protein en- patients, although they displayed heterozygosity by genomic
coding the highly conserved second zinc finger involved sequence for known SNPs found within the exons, cDNA
in binding to the GATA motif on DNA. Additional mu- sequence demonstrated the presence of one predominant
tations result in the loss of mRNA transcript from one al- allele (Fig. 8A). Further work needs to be performed to
lele, which is caused by a frameshift insertion or deletion elucidate the genetic lesion leading to loss of expression of
mutation or premature stop codons. This indicates that a the allele in these patients. Haploinsufficiency of GATA2
A. B.
C.
FIGURE 8 Reduced GATA2 allelic expression in MonoMAC syndrome. (A) The top row shows genomic sequence from two patients
demonstrating heterozygosity for a SNP within the cDNA transcript of GATA2. The bottom row shows cDNA sequence from full-length
RT-PCR. In patient 1, only the T allele is present (arrow), indicating loss of the transcript containing the C allele, while patient 2’s se-
quence contains the C allele with only a small peak of the G allele (arrow). These sequences demonstrate uniallelic expression leading to
haploinsufficiency of GATA2. (B) At the top is genomic sequence from two affected sisters; the proband is heterozygous for three known
SNPs found within the cDNA, while her sister is homozygous, allowing determination of the shared mutant allele haplotype (boxed bases).
At the bottom is GATA2 cDNA sequence from the proband’s CD3+ and CD3− PBMCs, revealing decreased levels of the mutant allele as
represented by peak height (arrows). (C) GATA2 genomic locus showing three reported isoform structures. The dashed box indicates a con-
served intron 5 region with a high GERP score, DNase I hypersensitivity, and strong transcription factor binding. Zoom of bases highlights
the E-box/GATA composite element (boxed); the 28-base deletion seen in the first patient (underlined sequence); and the ETS motif with
a C>T mutation seen in five patients (asterisk), including the sisters shown in panel B.
has been demonstrated in patients with mutations causing in the other patients. Functional constructs demonstrated
mRNA instability. This suggests that the molecular basis of reduced expression caused by each mutation (27, 42) and
the disease in these two patients is likely a loss of expres- correlated with the reduced allelic expression seen in the
sion of one GATA2 allele. In another group of MonoMAC patients. The data from these patients demonstrate another
patients, reduced but not absent expression from one allele class of mutations, those resulting in reduced expression in a
was recognized by cDNA analysis. Figure 8B shows genomic protein in which haploinsufficiency may cause disease. This
sequence from one patient and her sister, both affected with condition is recognized primarily by screening for uniallelic
MonoMAC. The proband is heterozygous for three known expression by cDNA.
SNPs found within the cDNA, while her sister is homozy- Figure 9 provides a framework for analysis of patients
gous for the same SNPs. This allows determination of the with immune disorders. First, the clinical phenotype needs
mutant allele haplotype, as both sisters must have inherited to be defined. Clinical presentation including age, gender,
the mutation. Subsequent screening of GATA2 cDNA in TREC levels in infants, and medical history provides cer-
the proband shows decreased peak height of the mutant tain clues. The infection history of the patient also can
allele compared with the heterozygous peaks seen in the suggest a target gene or pathway. Lastly, clinical laboratory
genomic sequence. Examination of the genomic structure assays such as lymphocyte phenotyping and functional flow
of GATA2 from the UCSC Genome Browser reveals a re- cytometric assays can further refine a working diagnosis.
gion within intron 5 notable for high levels of conservation, If there is a specific pathway or gene suggested based on
DNase I sensitivity, and evidence of multiple transcription clinical presentation, then targeted Sanger sequencing of
factor binding (Fig. 8C). Located with this is a composite that gene may identify a genetic lesion. In the absence of a
element recognized as critical for GATA2 expression (41). limited candidate gene list, then NGS via a medium-scale,
Screening of the conserved intronic region revealed a 28- targeted approach, such as using a panel of genes screened
base deletion encompassing a portion of the composite el- by Ion Torrent, or a large-scale whole-exome analysis may
ement in one patient, while a single point mutation at the be performed. After bioinformatics processing of the data, a
nearby ETS transcription factor family motif was identified filtered list of candidate variations is obtained, which then
FIGURE 9 Framework for molecular and genetic diagnosis in immunocompromised patients, beginning
with defining of patient phenotype, proceeding to performing genomic analysis, and concluding with
inferred or demonstrated pathogenicity of mutation.
are examined to see if they are consistent with regard to the standards for BRAF, EGFR and KRAS mutational analysis.
patient’s phenotype. Variations resulting in loss of canonical PLoS One 8:e69604. doi:10.1371/journal.pone.0069604.
splice sites or nonsense or missense changes should be eval- 12. Livak KJ, Schmittgen TD. 2001. Analysis of relative gene
uated for the predicted protein effect and confirmed with expression data using real-time quantitative PCR and the
functional studies of the protein. In the case in which a del- 2–∆∆CT method. Methods 25:402–408.
eterious variant is not identified, if a target gene is suspected, 13. Hazenberg MD, Verschuren MC, Hamann D, Miedema
then screening for altered protein size or levels or aberrant F, van Dongen JJ. 2001. T cell receptor excision circles as
cDNA caused by splicing defects or uniallelic expression markers for recent thymic emigrants: basic aspects, techni-
may be useful. Recognition of the possibility of mutations cal approach, and guidelines for interpretation. J Mol Med
(Berl) 79:631–640.
in the untranslated exons and conserved regions within and 14. Chan K, Puck JM. 2005. Development of population-
flanking the gene is also important for defining genetic le- based newborn screening for severe combined immunode-
sions. While methods to detect variations will continue to ficiency. J Allergy Clin Immunol 115:391–398.
improve in both sensitivity and throughput, understanding 15. Puck JM. 2012. Laboratory technology for population-
how to analyze variations and their potential effect on tran- based screening for severe combined immunodeficiency
script and protein will continue to be necessary to diagnose in neonates: the winner is T-cell receptor excision circles.
genetic lesions involved in defects of the immune system. J Allergy Clin Immunol 129:607–616.
16. Routes JM, Grossman WJ, Verbsky J, Laessig RH,
Hoffman GL, Brokopp CD, Baker MW. 2009. Statewide
REFERENCES newborn screening for severe T-cell lymphopenia. JAMA
1. Kirkpatrick P, Riminton S. 2007. Primary immunodefi- 302:2465–2470.
ciency diseases in Australia and New Zealand. J Clin Im- 17. Verbsky JW, Baker MW, Grossman WJ, Hintermeyer M,
munol 27:517–524. Dasu T, Bonacci B, Reddy S, Margolis D, Casper J, Gries
2. Gathmann B, Binder N, Ehl S, Kindle G, ESID Registry M, Desantes K, Hoffman GL, Brokopp CD, Seroogy
Working Party. 2012. The European internet-based pa- CM, Routes JM. 2012. Newborn screening for severe
tient and research database for primary immunodeficien- combined immunodeficiency; the Wisconsin experience
cies: update 2011. Clin Exp Immunol 167:479–491. (2008–2011). J Clin Immunol 32:82–88.
3. Taschner PE, den Dunnen JT. 2011. Describing structural 18. Kwan A, Church JA, Cowan MJ, Agarwal R, Kapoor N,
changes by extending HGVS sequence variation nomen- Kohn DB, Lewis DB, McGhee SA, Moore TB, Stiehm
clature. Hum Mutat 32:507–511. ER, Porteus M, Aznar CP, Currier R, Lorey F, Puck JM.
4. Quail MA, Smith M, Coupland P, Otto TD, Har- 2013. Newborn screening for severe combined immuno-
ris SR, Connor TR, Bertoni A, Swerdlow HP, Gu Y. deficiency and T-cell lymphopenia in California: results of
2012. A tale of three next generation sequencing plat- the first 2 years. J Allergy Clin Immunol 132:140–150.
forms: a comparison of Ion Torrent, Pacific Biosciences 19. Urban AE, Korbel JO, Selzer R, Richmond T, Hacker A,
and Illumina MiSeq sequencers. BMC Genomics 13:341. Popescu GV, Cubells JF, Green R, Emanuel BS, Gerstein
doi:10.1186/1471-2164-13-341. MB, Weissman SM, Snyder M. 2006. High-resolution
5. Loman NJ, Misra RV, Dallman TJ, Constantinidou C, mapping of DNA copy alterations in human chromosome
Gharbia SE, Wain J, Pallen MJ. 2012. Performance com- 22 using high-density tiling oligonucleotide arrays. Proc
parison of benchtop high- throughput sequencing plat- Natl Acad Sci U S A 103:4534–4539.
forms. Nat Biotechnol 30:434–439. 20. Chang X, Wang K. 2012. wANNOVAR: annotating ge-
6. Lee HC, Lai K, Lorenc MT, Imelfort M, Duran C, Ed- netic variants for personal genomes via the web. J Med
wards D. 2012. Bioinformatics tools and databases for Genet 49:433–436.
analysis of next-generation sequence data. Brief Funct Ge- 21. Liu X, Jian X, Boerwinkle E. 2013. dbNSFP v2.0: a
nomics 11:12–24. database of human nonsynonymous SNVs and their
7. McKenna A, Hanna M, Banks E, Sivachenko A, Cibul- functional predictions and annotations. Hum Mutat
skis K, Kernytsky A, Garimella K, Altshuler D, Gabriel S, 34:E2393–E2402.
Daly M, DePristo MA. 2010. The Genome Analysis Tool- 22. Fisher GH, Rosenberg FJ, Straus SE, Dale JK, Middle-
kit: a MapReduce framework for analyzing next-generation ton LA, Lin AY, Strober W, Lenardo MJ, Puck JM. 1995.
DNA sequencing data. Genome Res 20:1297–1303. Dominant interfering Fas gene mutations impair apoptosis
8. DePristo MA, Banks E, Poplin R, Garimella KV, Ma- in a human autoimmune lymphoproliferative syndrome.
guire JR, Hartl C, Philippakis AA, del Angel G, Rivas Cell 81:935–946.
MA, Hanna M, McKenna A, Fennell TJ, Kernytsky AM, 23. Jackson CE, Fischer RE, Hsu AP, Anderson SM, Choi
Sivachenko AY, Cibulskis K, Gabriel SB, Altshuler D, Y, Wang J, Dale JK, Fleisher TA, Middelton LA, Sneller
Daly MJ. 2011. A framework for variation discovery and MC, Lenardo MJ, Straus SE, Puck JM. 1999. Auto-
genotyping using next-generation DNA sequencing data. immune lymphoproliferative syndrome with defective
Nat Genet 43:491–498. Fas: genotype influences penetrance. Am J Hum Genet
9. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Ho- 64:1002–1014.
mer N, Marth G, Abecasis G, Durbin R, 1000 Genome 24. Nagy E, Maquat LE. 1998. A rule for termination-codon
Project Data Processing Subgroup. 2009. The Sequence position within intron-containing genes: when nonsense
Alignment/Map (SAM) format and SAMtools. Bioinfor- affects RNA abundance. Trends Biochem Sci 23:198–199.
matics 25:2078–2079. 25. Isken O, Maquat LE. 2007. Quality control of eukaryotic
10. Pabinger S, Dander A, Fischer M, Snajder R, Sperk M, mRNA: safeguarding cells from abnormal mRNA func-
Efremova M, Krabichler B, Speicher MR, Zschocke J, tion. Genes Dev 21:1833–1856.
Trajanoske Z. 2014. A survey of tools for variant analysis 26. Hsu AP, Dowdell KC, Davis J, Niemela JE, Anderson
of next-generation genome sequencing data. Brief Bioin- SM, Shaw PA, Rao VK, Puck JM. 2012. Autoimmune
form 15:256–278. lymphoproliferative syndrome due to FAS mutations out-
11. McCourt CM, McArt DG, Mills K, Catherwood MA, side the signal-transducing death domain: molecular mech-
Maxwell P, Waugh DJ, Hamilton P, O’Sullivan JM, Salto- anisms and clinical penetrance. Genet Med 14:81–89.
Tellez M. 2013. Validation of next generation sequencing 27. Hsu AP, Johnson KD, Falcone EL, Sanalkumar R, San-
technologies in comparison to current diagnostic gold chez L, Hickstein DD, Cuellar-Rodriguez J, Lemieux JE,
Zerbe CS, Bresnick EH, Holland SM. 2013. GATA2 hap- Hickstein DD, Holland SM. 2011. Mutations in GATA2
loinsufficiency caused by mutations in a conserved intronic are associated with the autosomal dominant and sporadic
element leads to MonoMAC syndrome. Blood 121:3830– monocytopenia and mycobacterial infection (MonoMAC)
3837, S1–S7. syndrome. Blood 118:2653–2655.
28. Wang Z, Burge CB. 2008. Splicing regulation: from a parts 41. Wozniak RJ, Boyer ME, Grass JA, Lee Y, Bresnick
list of regulatory elements to an integrated splicing code. EH. 2007. Context- dependent GATA factor function:
RNA 14:802–813. combinatorial requirements for transcriptional con-
29. Ward AJ, Cooper TA. 2010. The pathobiology of splicing. trol in hematopoietic and endothelial cells. J Biol Chem
J Pathol 220:152–163. 282:14665–14674.
30. Zhang Q, Davis JC, Lamborn IT, Freeman AF, Jing H, 42. Johnson KD, Hsu AP, Ryu MJ, Wang J, Gao X, Boyer
Favreau AJ, Matthews HF, Davis J, Turner ML, Uzel ME, Liu Y, Lee Y, Calvo KR, Keles S, Zhang J, Holland
G, Holland SM, Su HC. 2009. Combined immunodefi- SM, Bresnick EH. 2012. cis-Element mutated in GATA2-
ciency associated with DOCK8 mutations. N Engl J Med dependent immunodeficiency governs hematopoiesis and
361:2046–2055. vascular integrity. J Clin Invest 122:3692–3704.
31. Sterne-Weiler T, Howard J, Mort M, Cooper DN, Sanford 43. Adzhubei IA, Schmidt S, Peshkin L, Ramensky VE,
JR. 2011. Loss of exon identity is a common mechanism of Gerasimova A, Bork P, Kondrashov AS, Sunyaev SR.
human inherited disease. Genome Res 21:1563–1571. 2010. A method and server for predicting damaging mis-
32. Desmet FO, Hamroun D, Lalande M, Collod-Béroud G, sense mutations. Nat Methods 7:248–249.
Claustres M, Béroud C. 2009. Human Splicing Finder: an 44. Sim NL, Kumar P, Hu J, Henikoff S, Schneider G, Ng
online bioinformatics tool to predict splicing signals. Nu- PC. 2012. SIFT web server: predicting effects of amino
cleic Acids Res 37:e67. doi:10.1093/nar/gkp215. acid substitutions on proteins. Nucleic Acids Res 40(Web
33. Zhu J, Mayeda A, Krainer AR. 2001. Exon identity es- Server issue):W452–W457.
tablished through differential antagonism between exonic 45. Schwarz JM, Rödelsperger C, Schuelke M, Seelow D.
splicing silencer-bound hnRNP A1 and enhancer-bound 2010. MutationTaster evaluates disease-causing potential
SR proteins. Mol Cell 8:1351–1361. of sequence alterations. Nat Methods 7:575–576.
34. Barbosa C, Peixiero I, Romão L. 2013. Gene expression 46. Chun S, Fay JC. 2009. Identification of deleterious
regulation by upstream open reading frames and human mutations within three human genomes. Genome Res
disease. PLoS Genet 9:e1003529. doi:10.1371/journal. 19:1553–1561.
pgen.1003529. 47. Reva B, Antipin Y, Sander C. 2011. Predicting the func-
35. Sampaio EP, Hsu AP, Pechacek J, Bax HI, Dias DL, tional impact of protein mutations: application to cancer ge-
Paulson ML, Chandrasekaran P, Rosen LB, Carvalho nomics. Nucleic Acids Res 39:e118. doi:10.1093/nar/gkr407.
DS, Ding L, Vinh DC, Browne SK, Datta S, Milner 48. Shihab HA, Gough J, Cooper DN, Stenson PD, Barker
JD, Kuhns DB, Long Priel DA, Sadat MA, Shiloh M, GL, Edwards KJ, Day IN, Gaunt TR. 2013. Predicting
De Marco B, Alvares M, Gillman JW, Ramarathnam V, the functional, molecular, and phenotypic consequences
de la Morena M, Bezrodnik L, Moreira I, Uzel G, John- of amino acid substitutions using hidden Markov models.
son D, Spalding C, Zerbe CS, Wiley H, Greenberg DE, Hum Mutat 34:57–65.
Hoover SE, Rosenzweig SD, Galgiani JN, Holland SM. 49. Garber M, Guttman M, Clamp M, Zody MC, Friedman
2013. Signal transducer and activator of transcription 1 N, Xie X. 2009. Identifying novel constrained elements
(STAT1) gain- of-
function mutations and disseminated by exploiting biased substitution patterns. Bioinformatics
coccidioidomycosis and histoplasmosis. J Allergy Clin Im- 25:i54–i62.
munol 131:1624–1634. 50. Davydov EV, Goode DL, Sirota M, Cooper GM, Sidow
36. Uzel G, Sampaio EP, Lawrence MG, Hsu AP, Hackett M, A, Batzoglou S. 2010. Identifying a high fraction of the
Dorsey MJ, Noel RJ, Verbsky JW, Freeman AF, Janssen human genome to be under selective constraint using
E, Bonilla FA, Pechacek J, Chandrasekaran P, Browne GERP++. PLoS Comput Biol 6:e1001025. doi:10.1371/
SK, Agharahimi A, Gharib AM, Mannurita SC, Yim JJ, journal.pcbi.1001025.
Gambineri E, Torgerson T, Tran DQ, Milner JD, Holland 51. Pollard KS, Hubisz MJ, Rosenbloom KR, Siepel A. 2010.
SM. 2013. Dominant gain-of-function STAT1 mutations Detection of nonneutral substitution rates on mammalian
in FOXP3WT IPEX-like syndrome. J Allergy Clin Immunol phylogenies. Genome Res 20:110–121.
131:1611–1623. 52. Clarke L, Zheng-Bradley X, Smith R, Kulesha E, Xiao C,
37. Hsu AP, Sowerwine KJ, Lawrence MG, Davis J, Hender- Toneva I, Vaughan B, Preuss D, Leinonen R, Shumway
son CJ, Zarember KA, Garofalo M, Gallin JI, Kuhns DB, M, Sherry S, Flicek P, 1000 Genomes Project Consor-
Heller T, Milner JD, Puck JM, Freeman AF, Holland tium. 2012. The 1000 Genomes Project: data management
SM. 2013. Intermediate phenotypes in patients with au- and community access. Nat Methods 9:459–462.
tosomal dominant hyper-IgE syndrome caused by somatic 53. Schwartz S, Hall E, Ast G. 2009. SROOGLE: web-
mosaicism. J Allergy Clin Immunol 131:1586–1593. server for integrative, user-friendly visualization of splic-
38. Noack D, Heyworth PG, Newburger PE, Cross AR. ing signals. Nucleic Acids Res 37(Web Server issue):
2001. An unusual intronic mutation in the CYBB gene W189–W192.
giving rise to chronic granulomatous disease. Biochim Bio- 54. Smith PJ, Zhang C, Wang J, Chew SL, Zhang MQ,
phys Acta 1537:125–31. Krainer AR. 2006. An increased specificity score matrix
39. Rump A, Rösen-Wolff A, Gahr M, Seidenberg J, Roos C, for the prediction of SF2/ASF-specific exonic splicing en-
Walter L, Günther V, Roesler J. 2006. A splice-supporting hancers. Hum Mol Genet 15:2490–2508.
intronic mutation in the last bp position of a cryptic exon 55. Cartegni L, Wang J, Zhu Z, Zhang MQ, Krainer AR.
within intron 6 of the CYBB gene induces its incorpora- 2003. ESEfinder: a web resource to identify exonic splicing
tion into the mRNA causing chronic granulomatous dis- enhancers. Nucleic Acids Res 31:3568–3571.
ease (CGD). Gene 371:174–181. 56. Fairbrother WG, Yeh RF, Sharp PA, Burge CB. 2002.
40. Hsu AP, Sampaio EP, Khan J, Calvo KR, Lemieux JE, Predictive identification of exonic splicing enhancers in
Patel SY, Frucht DM, Vinh DC, Auth RD, Freeman AF, human genes. Science 297:1007–1013.
Olivier KN, Uzel G, Zerbe CS, Spalding C, Pittaluga S, 57. Zhang XH, Chasin LA. 2004. Computational definition
Raffeld M, Kuhns DB, Ding L, Paulson ML, Marciano of sequence motifs governing constitutive exon splicing.
BE, Gea-Banacloche JC, Orange JS, Cuellar-Rodriguez J, Genes Dev 18:1241–1250.
The Human Microbiome and Clinical Immunology
FREDERIC D. BUSHMAN
3
Humans, in many respects, are not single organisms, but the goal is often to compare a set of healthy control samples
more like coral reefs, complex assemblages of myriad diverse to a set of samples from a disease state. For example, one
creatures. The human microbiota, the bugs living in associ- might compare stool samples from a cohort of healthy chil-
ation with humans, contains more cells than comprise the dren to samples from a cohort of immunocompromised chil-
human body itself (1). Like the fish on a reef, the microbes dren. It is typically not feasible to complete the sequences
interact with each other, sometimes competing and some- of all microbes in each sample, but it is possible to collect
times collaborating, to form a complex and resilient food a deep sequence sample of each community. Sequences can
web. Some of the population promote the growth and devel- then be aligned to databases and the types of microbes pres-
opment of the reef, while others consume the reef material, ent inferred from database annotation. If total DNA is an-
turning the hard corals to sand. alyzed, then the gene types present (e.g., genes for adenine
This chapter reviews the nature of the human microbi- biosynthetic enzymes and virulence factors) can be assessed
ome, with emphasis on methods for microbiome research as well. One can then compare results for the collection of
and the pitfalls that have ruined many studies in the field. healthy control samples to the immunocompromised sub-
However, despite the challenges, the output from the field ject samples and identify differences associated with disease.
has been spectacular (2–7), and the technology emerging Such an analysis, based on large-scale but incomplete char-
from microbiome research holds tremendous promise for acterization of mixed communities by sequencing, has come
translation to clinical applications. A few of the possible to be called “metagenomics.”
future directions are described at the end of the chapter.
METHODS FOR METAGENOMICS RESEARCH
WHAT IS METAGENOMIC ANALYSIS? In this section, we review the DNA isolation, library prepa-
Microorganisms often live in mixed communities, and it ration, and sequencing approaches used in metagenomic
is clear that the global composition and function of mixed studies. The overall lesson is that details of the analytical
communities can be relevant to human health. For exam- approach used are driven quite strongly by the exact ques-
ple, actions of human microbial communities have been tion posed—relatively small differences in the goals of the
implicated in all the major causes of human mortality (1), study can mean dramatic differences in the metagenomic
including heart disease (8, 9), cancer (10), and diabetes methods needed for analysis. Several examples of types of
(11). Loss of normal control of the microbiome in immuno- questions and approaches to their investigation are pre-
deficient states is associated with pathogenesis of opportu- sented toward the end of this chapter.
nistic infections, and the tools of metagenomic analysis can
report the composition and nature of those communities in Nucleic Acid Purification
unprecedented detail. First, a set of samples must be acquired and nucleic acid
Complicating analysis of the human microbiome, only a purified for sequencing. Many nucleic acid purification kits
small minority of all the microbes on earth have been sub- are available from commercial suppliers, and many can be
jected to culture in the laboratory, so traditional culture- used successfully (13). A key control for any study is to run
based methods are often not suitable for studies of mixed DNA-free water samples through the purification procedure
microbial communities. A new approach is made possible side by side with the samples of interest. Some commercial
by the remarkable new “deep sequencing” methods, which kits contain substantial quantities of contaminating 16S
allow determination of millions or even billions of bases of DNA, which can be ruinous if the starting amount of nu-
DNA sequence in a single instrument run. cleic acid is low in the samples of interest. Commercial kits
Bacterial genomes range in size from ~500 kb to many are also available that yield both DNA and RNA from the
megabases, and the genomes of fungi are larger still (12). same sample, allowing assay of both.
Thus, sequencing all the genomes in a mixed community, Methods for lysis are worked out to allow capture of many
which may contain hundreds of types of microbes, is still a organisms in a sample, but the methods represent a compro-
daunting task. Furthermore, in a typical biomedical study, mise between convenience and complete sampling. Some
doi:10.1128/9781555818722.ch3
19
Gram-positive bacteria and fungi have tough cell walls and to mark primers in this fashion. Barcode systems are now
are known to be harder to lyse than typical bacteria. Spores available for most of the major sequencing platforms.
are tougher still, and any study for which data on spores are
important requires targeted evaluation of methods for lysis. Tag Sequencing
Unfortunately, it is possible that harsh lysis methods may Several approaches are commonly used to acquire deep se-
degrade nucleic acid from the first-to-lyse organisms before quence data on microbiome populations. In one popular im-
the toughest are cracked open. plementation, PCR is used to amplify conserved sequences
Stepping back, a general lesson is that it is important to present in bacteria or fungi, and then the collections of se-
carefully formulate the question under study, and if possible quence tags are used to characterize a sample. The bacterial
to check the performance of the methods chosen in address- 16S rRNA gene contains highly conserved regions, which
ing the question posed. can be used as binding sites for PCR primers (19–22). An
advantage to the use of 16S data is that well-curated data-
DNA Sequencing bases are available containing 16S sequences together with
The development of the astounding new DNA-sequencing phylogenetic annotation, and sophisticated pipelines are
methods has made possible the explosion of work on the in place (QIIME, mothur) that can take in such data and
human microbiome. For example, using the new Illumina output finished analysis (23, 24). For fungi, 18S rRNA can
HiSeq technology, one can acquire hundreds of billions of be used similarly as a tag, as well as the internal transcribed
bases of sequence information in a single instrument run. spacer region (25, 26).
The availability of such vast amounts of sequence data al- A variety of methods have been worked out for compar-
lows characterization of mixed microbial communities in ing sets of communities with one another. Given a set of
unprecedented detail. samples from, for example, healthy controls and immuno-
The automation of the Sanger sequencing chemistry was compromised subjects, one might want to ask whether there
a breakthrough in its day, driving the completion of the first are general trends in the data that distinguish one from the
drafts of human genome sequence in 2001. These instru- other. Several methods have been developed for calculating
ments could generate 384 reads of ~500 bases each, or ~200 pairwise distances between communities, so that all pairs
kb of sequence per run, under favorable circumstances. can be calculated to form a matrix of distances between
The first of the massively parallel deep sequencing meth- communities. Such matrices can then be used for statisti-
ods to be commercialized was the Roche 454 pyrosequencing cal analysis of possible patterns in the data. For example,
method (14). This allowed collection of ~500,000 sequence one can ask whether the mean distance between all pairs of
reads of ~350 bases per instrument run, or close to 200 mil- samples of the same group is smaller than the mean distance
lion bases. All of these next-generation sequencing methods for pairs of samples between groups. If yes, this is evidence
have higher error rates than the original Sanger sequencing for separate clustering, and allows simple statistical evalua-
method. The Roche 454 platform is particularly error prone tion. The UniFrac method is one popular tool for generat-
at homopolymer sequences, so that the numbers of bases in ing distances in a phylogenetic framework (27–29) and is
the homopolymer runs are commonly miscounted. implemented in the QIIME pipeline (23). An example is
While impressive, the newer Solexa/Illumina method is, shown in Fig. 1, illustrating the difference in microbial com-
at this writing, the most widely used and produces the most munities at different human body sites analyzed by UniFrac.
data per instrument run (15). The HiSeq instruments can
generate >250 billion bases of sequence information in a sin- Shotgun Sequencing
gle run in the form of paired 250-base reads. The ABI SOLiD Tag sequencing yields information on the organisms rep-
system also delivers amounts of sequence in this range. resented in a microbial population but does not report on
Additional companies are also developing still more re- the full complement of genes present. Bacteria of even the
markable technologies. Pacific Biosciences has marketed same species can vary by up to 30% of their genes (30), so
a third-generation instrument capable of generating long naming a strain using tag sequence data provides only part
reads on single molecules (16) (in the jargon, the “first gen- of the picture. Thus, shotgun sequencing of total DNA fol-
eration” was instruments automating Sanger sequencing lowed by alignment to databases adds information not just
and the “second generation” the Roche 454, Illumina, and on the organisms present but on the types of genes pres-
ABI SOLiD technologies). Ion Torrent has marketed an in- ent in the community as well. The methods for analyzing
expensive instrument that yields reads in the 350-base range shotgun metagenomic data are less well worked out than for
at relatively low cost. Both Roche 454 and Illumina have tag sequencing data (31), and the data volumes can be very
also marketed benchtop machines—the 454 Junior and Il- large, presenting challenges in data transfer and manage-
lumina MiSeq—that can yield sequence samples of about ment that are not as severe with tag sequencing. However,
one-tenth the output of the larger instruments at reduced methods are improving with each passing year. For exam-
cost and increased convenience. ple, MEGAN allows collection of aligned sequences to yield
taxonomic attribution (32), MetaPhlAn uses a sequence tag
DNA Barcoding approach over shotgun reads to annotate bacteria in an effi-
Often it is not desirable to collect billions of bases of se- cient fashion (33), and multiple databases allow assignment
quence information on a single sample, but rather it is of genes detected to functional classes or pathways (34, 35).
more cost-effective and efficient to collect much less data Thus, although methods development is ongoing, meth-
per sample over many more samples. A popular approach ods for metagenomic research are in place and available to
to this is to engineer DNA barcodes into synthetic DNA address many questions of interest in clinical immunology.
primers used in preparation of sequencing libraries, and use
separate primer sets to amplify each sample in a set. The am-
plification products can then be pooled together after library ARTIFACTS IN MICROBIOME RESEARCH
preparation, sequenced together, and separated out compu- Essentially all the steps described for microbiome analysis
tationally after the run based on the barcode information in above are imperfect. Carrying out a successful microbiome
each sequence read (17, 18). Multiple strategies can be used study involves understanding the limitations of the methods
3. The Human Microbiome and Clinical Immunology ■ 21
FIGURE 1 Clustering of bacterial communities by body site analyzed using UniFrac. Each dot rep-
resents one human microbiome sample, characterized by deep sequencing of 16S tags from the V1V2
region of the gene. Each sample was characterized using a few thousand sequence reads. PC indicates
principal coordinates.
used and aligning the question posed with the strengths of the primers used. Careful studies are starting to appear com-
methods and not the weaknesses. Some relevant reports paring the fungi queried using different primer sets that pro-
have been published (4, 36, 37). vide a starting point for experimental design (25, 26).
Biases in Sequence Tag Analysis Challenges with Low-Biomass Samples
All of the 16S tag systems used to analyze bacteria are bi- Much of the work on the human microbiome has focused on
ased in the bacterial types they recover (4, 38, 39). This is the lower gastrointestinal tract, where bacteria are present
unavoidable because the highly conserved sites in the 16S in the range of 1011 per g of intestinal contents (1). Given
rRNA gene are not perfectly conserved, just well conserved, such gigantic amounts of bacteria, one doesn’t usually worry
so that primers anneal to some target sequences better than about other sources of bacterial DNA in a gut sample. How-
others. Careful comparisons over defined communities can ever, the situation is much different for many other sample
yield startlingly different results with different primer sets types. Bronchoalveolar lavage, skin scrapings, blood plasma,
when relative abundance is queried, though differences are and multiple other sample types typically contain low levels
less when data are scored for presence and absence of rela- of bacteria, so that admixture of environmental DNA from
tively high-level taxonomic groups. dust or contamination in commercial reagents can be a sub-
How much this matters depends entirely on the question stantial fraction of the total.
posed. If the goal is to identify exactly the types and pro- Thus, it is absolutely critical to surround low-biomass
portions of bacteria in a complex mixture, these biases are samples with controls to assess the types and proportions of
significant concerns. Tag sequence data provide an overview contaminating sequences. Never carry out such an experi-
of much of the community, but further experimentation is ment without working up DNA-free water through the full
needed to characterize abundance convincingly (e.g., quan- DNA purification procedure. Blank PCR reactions are an-
titative PCR). However, if the goal is to compare two groups other must. More-sophisticated controls include instrument
of samples—perhaps fecal samples from immunocompro- washes and matched tissues from germ-free animals. Control
mised children to healthy controls— then the sequence samples are put through the DNA purification procedure
sample returned in a typical 16S tag study may be fine for side by side with the samples of interest and then sequenced
documenting a difference in communities such as separate with dedicated barcodes in the final deep sequencing run.
clustering. As with any measurement in experimental sci- A handful of reads can sometimes be obtained even from
ence, it is always best to verify an important conclusion us- PCR reactions that appear blank after analysis by gel elec-
ing multiple forms of measurement. For example, if a single trophoresis and staining with ethidium bromide, so it is im-
bacterium is judged to be functionally important in distin- portant to sequence everything. In addition, it is valuable to
guishing sample sets, then quantitative PCR could be used use quantitative PCR to quantify the total number of 16S-
to verify the difference between groups. complementary sequences per unit volume. Estimates of to-
Tag systems for analyzing microeukaryotes such as fungi tal 16S gene copies can then be compared between samples
present an even more extreme case. For such studies, re- of interest and contamination controls to check whether
searchers need to expect from the start that they are only the experimental samples indeed contain sequence counts
looking at the slice of the fungal community accessible with above background.
These issues are particularly pronounced in studies of major challenges. For one, genomes of all metazoans har-
the fungal microbiome. Here the absolute amounts of fun- bor integrated sequences from endogenous retroviruses and
gal DNA in samples are commonly low, and environmental retrotransposons that are clearly related to exogenous retro-
organisms are abundant. Achieving credibility for any study viruses (12), obstructing clear assessment of this viral group.
of the fungal microbiome requires reporting extensive con- Additional gene families are known to be common between
tamination controls and a careful discussion of how these viruses and bacterial sequences, so that alignment of total
controls were incorporated in the interpretation. One ap- shotgun sequence data to databases of viruses often falsely
proach to dissecting the contributions of different sources is pulls out matches.
the Bayesian SourceTracker software (40). If feasible, one approach to studying viruses is to first pu-
A further challenge comes in comparisons among sam- rify viral particles—if samples are clean enough, one can be
ples with different starting numbers of 16S copies. Suppose confident that each sequence analyzed likely came from a
that two sets of samples contain indistinguishable bacterial virus. A disadvantage of this approach is that any purifica-
communities, but one set has a lower biomass. Now add the tion method likely recovers some viruses better than others,
same number of contaminating environmental 16S DNA and some not at all, so only a portion of the viral population
copies to each. The contaminating DNA comprises a larger is queried.
proportion in the samples with fewer starting 16S DNA This method is well suited for analysis of bacterial vi-
copies. A sensitive measure comparing communities can ruses. Global populations of bacterial viruses are thought to
then falsely call the two sample sets as different when in be gigantic—up to 107 particles per ml in rich seawater (43).
reality they differed only in the amount of starting material. These viruses are thus necessarily poorly represented in se-
Papers making extreme claims but lacking sufficient con- quence databases, and so simply aligning newly acquired se-
trol of the above artifacts are common in the literature. quence to a database does a poor job in identifying bacterial
viruses. In sequence surveys of highly purified natural viral
Cage Effects in Studies of Mouse Models communities, most of the reads look like nothing you have
Another underappreciated artifact in microbiome studies ever seen before, consistent with this picture.
using mouse models is the strength of cage effects. Mice are As for bacteria and fungi, the method used is dictated by
copraphagic, so all the mice in a cage quickly come to share the question posed. For example, for hunting for new types
a common set of gut microorganisms. Mice of the same of viruses in immunocompromised subjects, one can take
treatment group or genotype in different cages can differ database hits as candidates for validation by quantitative
just because of cohousing. The effects can be so strong as PCR and other methods. As long as downstream validation
to dominate over what would otherwise seem to be extreme is part of the experimental plan, then metagenomic data can
treatments. be simply treated as hypothesis generating, obviating many
However, this is a solvable problem. The solution is to of the concerns above.
design the experiment so as to allow the cage effect to be
treated explicitly as a variable in the final statistical analysis
of the experiment. For two treatment groups, one can set up METAGENOMIC ANALYSIS TO INVESTIGATE
multiple cages for each group. It is okay to have relatively DISEASE STATES
few mice per cage, but the number of cages needs to be size- Despite the above challenges, the microbiome field has
able. Then in the statistical analysis of sequence data, one yielded much striking data, some of which bears on prob-
can ask whether the microbial communities differ between lems in clinical immunology. The Human Microbiome
groups given the measured effect of the cage variable. Project has completed draft sequences of thousands of bac-
Another approach involves longitudinal analysis, such teria, providing a rich resource for analysis of new sequences
as a drug treatment study, in which each mouse can serve by alignment (52). Studies of immunocompromised states
as its own control. Thus the changes can be tabulated and such as lung transplant (53, 54) or lentiviral infection (5)
compared between treatment groups. Such an analysis pre- have shown characteristic changes in microbial community
supposes that differences in the starting communities due composition. These are likely to lead directly to improved
to cage effects do not influence the subsequent response to methods for diagnosis of microbial colonization in immu-
the treatment. nocompromised patients. As the methods for sequence ac-
Cage effects are extremely strong for fungal populations. quisition become less expensive, focus turns to collection of
In one recent study (41), fungal populations were moni- appropriate samples for analysis and development of analyt-
tored longitudinally during an antibiotic intervention. Ep- ical tools for interpreting data. Many groups are presently
isodic increases in specific types of fungi were seen in all developing tools to allow molecular diagnostics based on
groups, including the untreated controls, and these waves of deep sequencing data.
colonization differed between cages. Fungal colonization of Another area of interest centers on understanding
mice has recently been reported to influence inflammatory patterns in healthy and dysbiotic communities based on
bowel disease models (42), which are known to show differ- microbiome sequence data. One pattern characteristic of
ent results in different mouse facilities, raising the possibil- immunocompromised states is higher variance in compo-
ity that fungal colonization and cage effects may represent sition. You could say that this parallels the famous Bum-
an underappreciated confounding variable. pus study of sparrow mortality in 1898 (http://fieldmuseum
.org/explore/hermon-bumpus-and-house-sparrows), which
may be the first application of multivariate mathematics to
METAGENOMIC ANALYSIS OF VIRUSES a problem in ecology. Bumpus studied sparrows that were
A number of compelling studies have examined the viruses knocked down by a violent storm. Some survived and oth-
of the human microbiome (43–51)—however, their anal- ers died. Bumpus took many anatomical measurements of
ysis requires careful design. For example, it may be of in- each bird and asked how the birds that died differed from
terest to query shotgun sequence samples from total stool those that survived. Bumpus did find a pattern, but there
DNA to identify the virome present, but this presents was no single trait associated with mortality. Instead, the
birds that died tended to diverge more from the average REFERENCES
“perfect” bird than did the survivors, but each fatality di- 1. Garrett WS, Gordon JI, Glimcher LH. 2010. Homeosta-
verged in its own way. That is, the fatal cases were distin- sis and inflammation in the intestine. Cell 140:859–870.
guished by higher variance, not by any single feature. A 2. Song SJ, Lauber C, Costello EK, Lozupone CA, Hum-
similar pattern is seen for microbial communities in immu- phrey G, Berg-Lyons D, Caporaso JG, Knights D, Cle-
nocompromised subjects. mente JC, Nakielny S, Gordon JI, Fierer N, Knight
One example is provided by studies of lung transplant re- R. 2013. Cohabiting family members share microbiota
cipients. In bronchoalveolar lavage samples, the variance was with one another and with their dogs. eLife 2:e00458.
higher for samples from lung transplant recipients than from doi:10.7554/eLife.00458.
healthy subjects. The lung transplant subjects also showed 3. Faith JJ, Guruge JL, Charbonneau M, Subramanian S,
lower richness, which is the total number of species detected Seedorf H, Goodman AL, Clemente JC, Knight R, Heath
in the sample (53). All of this is consistent with outgrowth of AC, Leibel RL, Rosenbaum M, Gordon JI. 2013. The
specific organisms in each immunocompromised subject, but long-term stability of the human gut microbiota. Science
different organisms in each case. The drop in richness can 341:1237439. doi:10.1126/science.1237439.
be understood as a single microbial lineage proliferating and 4. Human Microbiome Project Consortium. 2012. A frame-
“shouldering out” other community members. Consistent work for human microbiome research. Nature 486:215–221.
with this picture, the total amount of 16S DNA was higher in 5. Ridaura VK, Faith JJ, Rey FE, Cheng J, Duncan AE,
Kau AL, Griffin NW, Lombard V, Henrissat B, Bain JR,
lung transplant recipients than in healthy controls by PCR, Muehlbauer MJ, Ilkayeva O, Semenkovich CF, Funai K,
and individual recipients showed outgrowths of Pseudomo- Hayashi DK, Lyle BJ, Martini MC, Ursell LK, Clemente
nas, Staphylococcus, Achromobacter, and other known species JC, Van Treuren W, Walters WA, Knight R, Newgard
of concern. In one case, an organism not previously consid- CB, Heath AC, Gordon JI. 2013. Gut microbiota from
ered a lung resident, Snethia, was found to be predominant. twins discordant for obesity modulate metabolism in mice.
This organism would not normally be tracked in transplant Science 341:1241214. doi:10.1126/science.1241214.
recipients, presenting an example of how the new sequenc- 6. Arumugam M, Raes J, Pelletier E, Le Paslier D, Yamada
ing methods can provide information that is unavailable with T, Mende DR, Fernandes GR, Tap J, Bruls T, Batto JM,
other commonly used diagnostic methods. Similar patterns of Bertalan M, Borruel N, Casellas F, Fernandez L, Gautier
high variance are seen in metagenomic analysis of stool sam- L, Hansen T, Hattori M, Hayashi T, Kleerebezem M,
ples from pediatric subjects with primary immunodeficiencies Kurokawa K, Leclerc M, Levenez F, Manichanh C, Niel-
(unpublished data). sen HB, Nielsen T, Pons N, Poulain J, Qin J, Sicheritz-
The human microbiome has been implicated in multiple Ponten T, Tims S, Torrents D, Ugarte E, Zoetendal EG,
further aspects of human health and disease. Studies from Wang J, Guarner F, Pedersen O, de Vos WM, Brunak S,
Gordon and coworkers have emphasized that obesity can Doré J; MetaHIT Consortium, Antolín M, Artiguenave
be associated with distinctive patterns of gut microbial col- F, Blottiere HM, Almeida M, Brechot C, Cara C, Cher-
onization (5, 55–60). Strikingly, the obese phenotype in hu- vaux C, Cultrone A, Delorme C, Denariaz G, Dervyn
mans can be transferred to gnotobiotic (germ-free) mice by R, Foerstner KU, Friss C, van de Guchte M, Guedon
inoculation from obese but not lean individuals, suggesting E, Haimet F, Huber W, van Hylckama-Vlieg J, Jamet A,
Juste C, Kaci G, Knol J, Lakhdari O, Layec S, Le Roux
that the gut microbiome plays a causal role in the obese phe- K, Maguin E, Mérieux A, Melo Minardi R, M’rini C,
notype. Going the other way, a recent study suggested that Muller J, Oozeer R, Parkhill J, Renault P, Rescigno M,
long-term diet dictates the composition of the gut microbi- Sanchez N, Sunagawa S, Torrejon A, Turner K, Van-
ome (7), providing a potential mechanism connecting diet demeulebrouck G, Varela E, Winogradsky Y, Zeller G,
and the many health outcomes linked to dietary choices. Weissenbach J, Ehrlich SD, Bork P. 2011. Enterotypes of
Another striking series of studies have recently linked the human gut microbiome. Nature 473:174–180.
metabolic products of the gut microbiome to heart disease 7. Wu GD, Chen J, Hoffmann C, Bittinger K, Chen YY,
(8, 9, 61). Hazen and coworkers began by analyzing metab- Keilbaugh SA, Bewtra M, Knights D, Walters WA,
olites associated with adverse cardiac outcome in a large Knight R, Sinha R, Gilroy E, Gupta K, Baldassano R,
cohort of subjects. Comparison of discovery and validation Nessel L, Li H, Bushman FD, Lewis JD. 2011. Linking
cohorts revealed phosphatidylcholine, choline, and trime- long-term dietary patterns with gut microbial enterotypes.
thylamine oxide as metabolites linked to adverse outcome. Science 334:105–108.
Further studies added carnitine and trimethylamine to the 8. Wang Z, Klipfell E, Bennett BJ, Koeth R, Levison BS,
pathway. Very extensive follow-up studies outlined a path- Dugar B, Feldstein AE, Britt EB, Fu X, Chung YM, Wu
way involving the gut microbiome as a key step. Foods such Y, Schauer P, Smith JD, Allayee H, Tang WH, DiDonato
as eggs and meat are rich sources of phosphatidylcholine JA, Lusis AJ, Hazen SL. 2011. Gut flora metabolism of
and carnitine. The compounds are broken down to yield phosphatidylcholine promotes cardiovascular disease. Na-
trimethylamine in a step dependent on the gut microbi- ture 472:57–63.
ome. Trimethylamine is then metabolized by liver oxidases 9. Koeth RA, Wang Z, Levison BS, Buffa JA, Org E, Sheehy
BT, Britt EB, Fu X, Wu Y, Li L, Smith JD, DiDonato JA,
to yield trimethylamine oxide, which is directly proath- Chen J, Li H, Wu GD, Lewis JD, Warrier M, Brown JM,
erogenic. The microbial genes for production of trimethyl- Krauss RM, Tang WH, Bushman FD, Lusis AJ, Hazen
amine have recently been identified by Craciun and Balskus SL. 2013. Intestinal microbiota metabolism of l-carnitine,
(62), opening up the possibility of modulating this pathway a nutrient in red meat, promotes atherosclerosis. Nat Med
to minimize heart disease. 19:576–585.
In summary, the new technology and new insights of 10. Hu B, Elinav E, Huber S, Strowig T, Hao L, Hafemann
microbiome research are tremendously potent, but with in- A, Jin C, Wunderlich C, Wunderlich T, Eisenbarth SC,
creased sensitivity comes new types of artifacts that are not Flavell RA. 2013. Microbiota-induced activation of epi-
to be underestimated. Nevertheless, microbiome analysis is thelial IL-6 signaling links inflammasome-driven inflam-
becoming an important contributor to diagnostics and re- mation with transmissible cancer. Proc Natl Acad Sci U S A
search in clinical immunology and many other fields. 110:9862–9867.
11. Qin J, Li Y, Cai Z, Li S, Zhu J, Zhang F, Liang S, Zhang phylogenetic analysis of Cyclospora, the human intestinal
W, Guan Y, Shen D, Peng Y, Zhang D, Jie Z, Wu W, Qin pathogen, suggests that it is closely related to Eimeria spe-
Y, Xue W, Li J, Han L, Lu D, Wu P, Dai Y, Sun X, Li Z, cies. J Infect Dis 173:440–445.
Tang A, Zhong S, Li X, Chen W, Xu R, Wang M, Feng 22. McKenna P, Hoffmann C, Minkah N, Aye PP, Lackner
Q, Gong M, Yu J, Zhang Y, Zhang M, Hansen T, Sanchez A, Liu Z, Lozupone CA, Hamady M, Knight R, Bushman
G, Raes J, Falony G, Okuda S, Almeida M, LeChate- FD. 2008. The macaque gut microbiome in health, lentivi-
lier E, Renault P, Pons N, Batto JM, Zhang Z, Chen H, ral infection, and chronic enterocolitis. PLoS Pathog 4:e20.
Yang R, Zheng W, Li S, Yang H, Wang J, Ehrlich SD, doi:10.1371/journal.ppat.0040020.
Nielsen R, Pedersen O, Kristiansen K, Wang J. 2012. A 23. Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K,
metagenome-wide association study of gut microbiota in Bushman FD, Costello EK, Fierer N, Peña AG, Good
type 2 diabetes. Nature 490:55–60. rich JK, Gordon JI, Huttley GA, Kelley ST, Knights
12. Bushman FD. 2001. Lateral DNA Transfer: Mechanisms D, Koenig JE, Ley RE, Lozupone CA, McDonald D,
and Consequences. Cold Spring Harbor Laboratory Press, Muegge BD, Pirrung M, Reeder J, Sevinsky JR, Turn-
Cold Spring Harbor, NY. baugh PJ, Walters WA, Widmann J, Yatsunenko T,
13. Wu GD, Lewis JD, Hoffmann C, Chen YY, Knight R, Zaneveld J, Knight R. 2010. QIIME allows analysis of
Bittinger K, Hwang J, Chen J, Berkowsky R, Nessel L, high-throughput community sequencing data. Nat Meth-
Li H, Bushman FD. 2010. Sampling and pyrosequenc- ods 7:335–336.
ing methods for characterizing bacterial communities in 24. Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann
the human gut using 16S sequence tags. BMC Microbiol M, Hollister EB, Lesniewski RA, Oakley BB, Parks DH,
10:206. doi:10.1186/1471-2180-10-206. Robinson CJ, Sahl JW, Stres B, Thallinger GG, Van
14. Margulies M, Egholm M, Altman WE, Attiya S, Bader Horn DJ, Weber CF. 2009. Introducing mothur: open-
JS, Bemben LA, Berka J, Braverman MS, Chen YJ, source, platform-independent, community-supported soft-
Chen Z, Dewell SB, Du L, Fierro JM, Gomes XV, God- ware for describing and comparing microbial communities.
win BC, He W, Helgesen S, Ho CH, Irzyk GP, Jando Appl Environ Microbiol 75:7537–7541.
SC, Alenquer ML, Jarvie TP, Jirage KB, Kim JB, Knight 25. Dollive S, Peterfreund GL, Sherrill-Mix S, Bittinger
JR, Lanza JR, Leamon JH, Lefkowitz SM, Lei M, Li J, K, Sinha R, Hoffmann C, Nabel CS, Hill DA, Artis
Lohman KL, Lu H, Makhijani VB, McDade KE, Mc D, Bachman MA, Custers-Allen R, Grunberg S, Wu
Kenna MP, Myers EW, Nickerson E, Nobile JR, Plant GD, Lewis JD, Bushman FD. 2012. A tool kit for quan-
R, Puc BP, Ronan MT, Roth GT, Sarkis GJ, Simons JF, tifying eukaryotic rRNA gene sequences from human
Simpson JW, Srinivasan M, Tartaro KR, Tomasz A, Vogt microbiome samples. Genome Biol 13:R60. doi:10.1186/
KA, Volkmer GA, Wang SH, Wang Y, Weiner MP, Yu gb-2012-13-7-r60.
P, Begley RF, Rothberg JM. 2005. Genome sequencing 26. Schoch CL, Seifert KA, Huhndorf S, Robert V, Spouge
in microfabricated high-density picolitre reactors. Nature JL, Levesque CA, Chen W, Fungal Barcoding Consor-
437:376–380. tium, Fungal Barcoding Consortium Author List. 2012.
15. Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Nuclear ribosomal internal transcribed spacer (ITS) region
Huntley J, Fierer N, Owens SM, Betley J, Fraser L, as a universal DNA barcode marker for Fungi. Proc Natl
Bauer M, Gormley N, Gilbert JA, Smith G, Knight R. Acad Sci U S A 109:6241–6246.
2012. Ultra-high-throughput microbial community anal- 27. Lozupone C, Knight R. 2005. UniFrac: a new phyloge-
ysis on the Illumina HiSeq and MiSeq platforms. ISME J netic method for comparing microbial communities. Appl
6:1621–1624. Environ Microbiol 71:8228–8235.
16. Eid J, Fehr A, Gray J, Luong K, Lyle J, Otto G, Peluso P, 28. Lozupone C, Hamady M, Knight R. 2006. UniFrac—an
Rank D, Baybayan P, Bettman B, Bibillo A, Bjornson K, online tool for comparing microbial community diversity
Chaudhuri B, Christians F, Cicero R, Clark S, Dalal R, in a phylogenetic context. BMC Bioinformatics 7:371.
Dewinter A, Dixon J, Foquet M, Gaertner A, Hardenbol doi:10.1186/1471-2105-7-371.
P, Heiner C, Hester K, Holden D, Kearns G, Kong X, 29. Hamady M, Lozupone C, Knight R. 2010. Fast UniFrac:
Kuse R, Lacroix Y, Lin S, Lundquist P, Ma C, Marks P, facilitating high-throughput phylogenetic analyses of mi-
Maxham M, Murphy D, Park I, Pham T, Phillips M, Roy crobial communities including analysis of pyrosequencing
J, Sebra R, Shen G, Sorenson J, Tomaney A, Travers K, and PhyloChip data. ISME J 4:17–27.
Trulson M, Vieceli J, Wegener J, Wu D, Yang A, Zacca- 30. Rasko DA, Rosovitz MJ, Myers GS, Mongodin EF,
rin D, Zhao P, Zhong F, Korlach J, Turner S. 2009. Real- Fricke WF, Gajer P, Crabtree J, Sebaihia M, Thomson
time DNA sequencing from single polymerase molecules. NR, Chaudhuri R, Henderson IR, Sperandio V, Ravel
Science 323:133–138. J. 2008. The pangenome structure of Escherichia coli: com-
17. Hoffmann C, Minkah N, Leipzig J, Wang G, Arens MQ, parative genomic analysis of E. coli commensal and patho-
Tebas P, Bushman FD. 2007. DNA bar coding and pyrose- genic isolates. J Bacteriol 190:6881–6893.
quencing to identify rare HIV drug resistance mutations. 31. Liu B, Gibbons T, Ghodsi M, Treangen T, Pop M.
Nucleic Acids Res 35:e91. doi:10.1093/nar/gkm435. 2011. Accurate and fast estimation of taxonomic profiles
18. Binladen J, Gilbert MT, Bollback JP, Panitz F, Bendixen from metagenomic shotgun sequences. BMC Genomics
C, Nielsen R, Willerslev E. 2007. The use of coded PCR 12(Suppl 2):S4. doi:10.1186/1471-2164-12-S2-S4.
primers enables high-throughput sequencing of multiple 32. Huson DH, Auch AF, Qi J, Schuster SC. 2007. MEGAN
homolog amplification products by 454 parallel sequenc- analysis of metagenomic data. Genome Res 17:377–386.
ing. PLoS One 2:e197. doi:10.1371/journal.pone.0000197. 33. Segata N, Waldron L, Ballarini A, Narasimhan V, Jous-
19. Pace NR, Olsen GJ, Woese CR. 1986. Ribosomal RNA son O, Huttenhower C. 2012. Metagenomic microbial
phylogeny and the primary lines of evolutionary descent. community profiling using unique clade-specific marker
Cell 45:325–326. genes. Nat Methods 9:811–814.
20. Sogin ML, Morrison HG, Huber JA, Mark Welch D, 34. Natale DA, Galperin MY, Tatusov RL, Koonin EV. 2000.
Huse SM, Neal PR, Arrieta JM, Herndl GJ. 2006. Micro- Using the COG database to improve gene recognition in
bial diversity in the deep sea and the underexplored “rare complete genomes. Genetica 108:9–17.
biosphere.” Proc Natl Acad Sci U S A 103:12115–12120. 35. Kanehisa M, Goto S, Kawashima S, Okuno Y, Hattori
21. Relman DA, Schmidt TM, Gajadhar A, Sogin M, Cross M. 2004. The KEGG resource for deciphering the genome.
J, Yoder K, Sethabutr O, Echeverria P. 1996. Molecular Nucleic Acids Res 32(Database issue):D277–D280.
36. Navas-Molina JA, Peralta-Sánchez JM, González A, human gut. PLoS One 7:e42342. doi:10.1371/journal.
McMurdie PJ, Vázquez-Baeza Y, Xu Z, Ursell LK, Lau- pone.0042342.
ber C, Zhou H, Song SJ, Huntley J, Ackermann GL, 50. Minot S, Grunberg S, Wu GD, Lewis JD, Bushman FD.
Berg-Lyons D, Holmes S, Caporaso JG, Knight R. 2013. 2012. Hypervariable loci in the human gut virome. Proc
Advancing our understanding of the human microbiome Natl Acad Sci U S A 109:3962–3966.
using QIIME. Methods Enzymol 531:371–444. 51. Wylie KM, Mihindukulasuriya KA, Sodergren E, Wein-
37. Charlson ES, Bittinger K, Haas AR, Fitzgerald AS, stock GM, Storch GA. 2012. Sequence analysis of the
Frank I, Yadav A, Bushman FD, Collman RG. 2011. human virome in febrile and afebrile children. PLoS One
Topographical continuity of bacterial populations in the 7:e27735. doi:10.1371/journal.pone.0027735.
healthy human respiratory tract. Am J Respir Crit Care Med 52. Gevers D, Knight R, Petrosino JF, Huang K, McGuire
184:957–963. AL, Birren BW, Nelson KE, White O, Methé BA,
38. Liu Z, Lozupone C, Hamady M, Bushman FD, Knight Huttenhower C. 2012. The Human Microbiome Proj-
R. 2007. Short pyrosequencing reads suffice for accurate ect: a community resource for the healthy human mi-
microbial community analysis. Nucleic Acids Res 35:e120. crobiome. PLoS Biol 10:e1001377. doi:10.1371/journal.
doi:10.1093/nar/gkm541. pbio.1001377.
39. Schloss PD, Westcott SL. 2011. Assessing and improving 53. Charlson ES, Diamond JM, Bittinger K, Fitzgerald AS,
methods used in operational taxonomic unit- based ap- Yadav A, Haas AR, Bushman FD, Collman RG. 2012.
proaches for 16S rRNA gene sequence analysis. Appl Envi- Lung-enriched organisms and aberrant bacterial and fun-
ron Microbiol 77:3219–3226. gal respiratory microbiota after lung transplant. Am J Respir
40. Knights D, Kuczynski J, Charlson ES, Zaneveld J, Mozer Crit Care Med 186:536–545.
MC, Collman RG, Bushman FD, Knight R, Kelley ST. 54. Charlson ES, Bittinger K, Chen J, Diamond JM, Li H,
2011. Bayesian community-wide culture-independent mi- Collman RG, Bushman FD. 2012. Assessing bacterial
crobial source tracking. Nat Methods 8:761–763. populations in the lung by replicate analysis of samples
41. Dollive S, Chen YY, Grunberg S, Bittinger K, Hoffmann from the upper and lower respiratory tracts. PLoS One
C, Vandivier L, Cuff C, Lewis JD, Wu GD, Bushman 7:e42786. doi:10.1371/journal.pone.0042786.
FD. 2013. Fungi of the murine gut: episodic variation 55. Ley RE, Bäckhed F, Turnbaugh P, Lozupone CA, Knight
and proliferation during antibiotic treatment. PLoS One RD, Gordon JI. 2005. Obesity alters gut microbial ecology.
8:e71806. doi:10.1371/journal.pone.0071806. Proc Natl Acad Sci U S A 102:11070–11075.
42. Iliev ID, Funari VA, Taylor KD, Nguyen Q, Reyes CN, 56. Ley RE, Turnbaugh PJ, Klein S, Gordon JI. 2006. Micro-
Strom SP, Brown J, Becker CA, Fleshner PR, Dubinsky bial ecology: human gut microbes associated with obesity.
M, Rotter JI, Wang HL, McGovern DP, Brown GD, Un- Nature 444:1022–1023.
derhill DM. 2012. Interactions between commensal fungi 57. Turnbaugh PJ, Ley RE, Mahowald MA, Magrini V,
and the C-type lectin receptor Dectin-1 influence colitis. Mardis ER, Gordon JI. 2006. An obesity-associated gut
Science 336:1314–1317. microbiome with increased capacity for energy harvest.
43. Rohwer F. 2003. Global phage diversity. Cell 113:141. Nature 444:1027–1031.
44. Edwards RA, Rohwer F. 2005. Viral metagenomics. Nat 58. Turnbaugh PJ, Bäckhed F, Fulton L, Gordon JI. 2008.
Rev Microbiol 3:504–510. Diet-induced obesity is linked to marked but reversible
45. Zhang T, Breitbart M, Lee WH, Run JQ, Wei CL, Soh alterations in the mouse distal gut microbiome. Cell Host
SW, Hibberd ML, Liu ET, Rohwer F, Ruan Y. 2006. Microbe 3:213–223.
RNA viral community in human feces: prevalence of plant 59. Turnbaugh PJ, Gordon JI. 2009. The core gut micro
pathogenic viruses. PLoS Biol 4:e3. biome, energy balance and obesity. J Physiol 587:
doi:10.1371/journal.pbio.0040003. 4153–4158.
46. Reyes A, Haynes M, Hanson N, Angly FE, Heath AC, 60. Henao-Mejia J, Elinav E, Jin C, Hao L, Mehal WZ,
Rohwer F, Gordon JI. 2010. Viruses in the faecal mi- Strowig T, Thaiss CA, Kau AL, Eisenbarth SC, Jurczak
crobiota of monozygotic twins and their mothers. Nature MJ, Camporez JP, Shulman GI, Gordon JI, Hoffman
466:334–338. HM, Flavell RA. 2012. Inflammasome-mediated dysbi-
47. Minot S, Sinha R, Chen J, Li H, Keilbaugh SA, Wu GD, osis regulates progression of NAFLD and obesity. Nature
Lewis JD, Bushman FD. 2011. The human gut virome: 482:179–185.
inter-individual variation and dynamic response to diet. 61. Tang WH, Wang Z, Levison BS, Koeth RA, Britt EB, Fu
Genome Res 21:1616–1625. X, Wu Y, Hazen SL. 2013. Intestinal microbial metabo-
48. Minot S, Bryson A, Chehoud C, Wu GD, Lewis JD, lism of phosphatidylcholine and cardiovascular risk. New
Bushman FD. 2013. Rapid evolution of the human gut Engl J Med 368:1575–1584.
virome. Proc Natl Acad Sci U S A 110:12450–12455. 62. Craciun S, Balskus EP. 2012. Microbial conversion of
49. Minot S, Wu GD, Lewis JD, Bushman FD. 2012. Con- choline to trimethylamine requires a glycyl radical enzyme.
servation of gene cassettes among diverse viruses of the Proc Natl Acad Sci U S A 109:21307–21312.
Protein Analysis in the Clinical
Immunology Laboratory
ROSHINI SARAH ABRAHAM AND DAVID R. BARNIDGE
4
THE ROLE OF PROTEIN ANALYSIS IN of interest, which defines the optimal method for analysis.
DIAGNOSTIC IMMUNOLOGY Most proteins measured in the immunology laboratory are
The diagnostic immunology laboratory relies heavily on soluble and present in serum or plasma, urine, or body flu-
protein measurements, especially with the explosion of ids including ascites, cerebrospinal fluid (CSF), and perito-
clinically relevant biomarker analysis. Of particular import neal and bronchial fluid. However, other important sources
are the tremendous advances that have been made in the of nonsoluble proteins include cells present either in fluid
technology for protein detection, and while not all of it matrices (blood, urine, CSF, or other body fluids) or tissue
has gained traction in the clinical immunology laboratory, matrices, obtained primarily from biopsy samples (2). Fur-
this remains an area of huge growth. However, regulatory ther, proteins can be studied in vivo, either for diagnostic or
processes have not kept up with the burgeoning research therapeutic purposes, either in the context of cells or soluble
in the area of protein analysis, and new diagnostic tests for protein, using magnetic resonance and other imaging mo-
protein analytes are approved for clinical testing at a gla- dalities that assess labeled proteins (3, 4). A notable exam-
cial pace. Nonetheless, it is critical for the clinical immu- ple of in vivo protein analysis for determining diagnosis and
nologist to understand these advances and determine how monitoring response to treatment is the use of 123I-labeled
they can best be utilized in the clinical laboratory. Besides serum amyloid P component in systemic amyloidosis (5).
keeping pace with the rapidly changing technology, the age- Also, the use of labeled bisphosphonate bone tracers has
old fundamental principles of analytical validation of new been shown to be specific and sensitive in identifying trans-
tests, protein based or not, are still applicable. This chapter thyretin amyloid protein deposits in the heart (6).
covers the basic principles of protein testing in the clinical
laboratory and provides special emphasis on the role of mass Sample Collection and Its Impact on Protein
spectrometry (MS) in diagnostic protein analysis. Analysis
Many factors influence the measurement of proteins. These
can be subdivided into four major categories: (i) physiolog-
PREANALYTICAL ISSUES IN PROTEIN ical processes, (ii) collection processes, (iii) sample process-
ANALYSIS ing, and (iv) therapeutic intervention (2).
Preanalytical parameters determine the validity of any diag-
nostic test result; therefore, the importance of considering
preanalytical issues in clinical test development and vali- TYPES OF PROTEIN ANALYSIS
dation cannot be overemphasized. The components of the Diagnostic evaluation of proteins is dependent not only on
preanalytical phase include both generic elements and those the source and collection factors but also on how the result is
specific for protein analysis. Generic preanalytical measures used for interpretation or correlation with clinical and other
include relevant test selection with appropriate analytical findings. Therefore, protein analyses can be categorized into
sensitivity and specificity to detect the biomarker (protein) qualitative, semiquantitative, and quantitative measure-
of choice and the probability it will aid in diagnosing the ments. Qualitative analyses are used largely in contexts in
disease or clinical condition (clinical sensitivity/specificity). which the presence or absence of the protein of interest is
Other factors to consider include preparation of patient and sufficient for diagnostic utility, and there are typically large
sample collection and handling (1). Each of these has to differences between positive and negative responses. Most
be individually assessed and validated so that the analysis qualitative protein tests use visual assessment to determine
is performed under predetermined conditions that will con- if the protein biomarker is present or absent. Other qualita-
tribute to a reliable test result and interpretation. tive measurements include contexts of structural alteration
of proteins necessitating some form of physical separation
Sources of Protein for Diagnostic Analysis using chromatography, electrophoresis, or MS. The most
Diagnostically relevant proteins can be studied from a vari- common examples of this approach include monoclonal
ety of sources, and sample collection depends on the protein protein analysis by serum or urine standard electrophoresis
doi:10.1128/9781555818722.ch4
26
4. Protein Analysis in Clinical Immunology ■ 27
followed by immunofixation electrophoresis or assessment (11). Under conditions of antibody excess, the quantity of
of α1- antitrypsin variants by isoelectric focusing, which the precipitate is directly related to the quantity of antigen in
overcomes the imprecision of electrophoresis by determin- the test sample. The major limitations include the duration
ing protein migration based on its charge (isoelectric point) of time required; the relative imprecision of the assay, with
in a pH gradient (7). coefficients of variation often >10%; the relative insensitiv-
Protein analyses can also be semiquantitative, whereby ity; and dependence on antigen quantity and structure.
the absolute amount of the analyte in the sample is not pro-
vided but the relative proportion of the analyte of interest Nephelometry
to other proteins in the mixture can be obtained. A com- RID has been supplanted by nephelometry in most labora-
mon example of semiquantitative protein measurement is in tories as the method of choice for measuring most proteins
the use of point-of-care dipsticks that measure urine protein. in body fluids, including immunoglobulins and intact or free
Newer dipsticks are available that can report albumin-to- light and heavy chains (12–14). Nephelometry is a method
creatinine or total protein-to-creatinine ratios, and though that relies on light scattering by soluble immune complexes
the analytical sensitivity is reasonable, there is significant in solution. The amount of light scatter produced by solu-
variability in the specificity of these dipstick reagents. There- ble immune complexes is directly proportional to the con-
fore, the most definitive test for detecting proteinuria is the centration of antigen. In contrast to the standard precipitin
quantitative measurement of urine protein. Similarly, most reaction, which requires that the concentrations of anti-
diagnostic measurements of clinically relevant proteins re- gen and antibody be at equivalence, nephelometry often
quire quantitative evaluation and comparison with age-and performs best with excess antibody. The two main types of
gender-matched reference values derived from healthy con- nephelometric reactions include endpoint (fixed time) and
trols. Further, confounding factors in interpretation of many rate (kinetic) nephelometry, which measure light scatter af-
diagnostically relevant proteins include diurnal, seasonal, ter equilibration of the antigen-antibody reaction and the
and exercise-related variability, and these elements have to peak rate of immune complex generation, respectively (12–
be incorporated into clinical interpretation of results, besides 14). The primary disadvantages of nephelometry include
being relevant in formulating reference values. increased costs for instruments and reagents, although these
can be offset by replacing manual methods with automa-
tion; variability in reagent antisera, especially to structurally
METHODS OF PROTEIN ANALYSIS altered proteins; and interfering substances that can alter
The above-discussed types of protein measurements can uti- light-scattering properties.
lize a variety of methodologies, depending on whether it is
relevant to quantify the amount of protein or the functional Electrophoresis
activity of the protein. Similarly, qualitative assays can use The migration of proteins on agarose electrophoresis is used
different methods for determining the presence or absence most commonly for identification of monoclonal proteins
of protein or normal versus abnormal function without in patients with monoclonal gammopathies (13, 15). In
quantifying the magnitude of functional activity. the serum protein electrophoresis depicted in Fig. 1, the
Methods for detecting protein function range from bind- stained protein bands representing albumin and α1-, α2-,
ing assays to enzyme activity assays measured by flow cy- and β-globulins can be easily visualized. While the most
tometry, immunoassays, or MS. Examples include NADPH abundant protein bands, such as albumin, are clearly visu-
oxidase activity as assessed by dihydrorhodamine flow cy- alized with definition, immunoglobulins migrating in the
tometry (8–10) and C1 esterase inhibitor function as deter- γ region are often broad and diffuse. This lack of resolu-
mined by an enzyme immunoassay using a labeled, activated tion in the γ region is due to the presence of many electro-
target protein. This target protein facilitates complex for- phoretically heterogeneous immunoglobulins that migrate
mation between the “test” protein (enzyme inhibitor) and in the same region. Serum protein electrophoresis can be
the target protein (enzyme substrate). Examples of these used to qualitatively or semiquantitatively identify, by vi-
different methods are discussed below. sual and densitometric analysis, the presence of monoclo-
nal immunoglobulins, hypogammaglobulinemia, nephrotic
Immunoassays syndrome, and polyclonal hypergammaglobulinemia. Serum
There are many immunoassay formats used in the clinical electrophoresis is not a particularly sensitive method for the
immunology laboratory for the measurement of proteins or detection of hypogammaglobulinemia, and there should
peptides. Most immunoassays offer quantitative analysis of be additional confirmation with quantitation of immuno-
proteins, although the degree of accuracy of quantitation globulins by nephelometry. Monoclonal proteins should
may vary depending on the specific method. Immunoassays be further assessed by immunofixation electrophoresis.
can be divided into those that use heterogeneous or homo- Densitometric analysis of stained proteins after total serum
geneous formats, with the former having either antibody or electrophoresis is an acceptable method for following the
antigen immobilized on a solid substrate and a separate step response to therapy of diseases characterized by monoclonal
and the latter having antigen-antibody complex formation proteins after an initial diagnosis.
taking place in the solution phase with no separation of Most proteins display a net charge in an electrical field,
bound and free ligand. Immunoassays were described exten- which allows for their electrophoretic separation. The size
sively in the previous edition of this book (2); therefore, and shape of a protein molecule, the ionic strength of the
only a synopsis of key methodologies is provided herein. buffer, the frictional resistance of the supporting medium,
the applied current, the temperature of the reaction, and
RID the duration of the applied current all influence the migra-
Radial immunodiffusion (RID) is a relatively old method tion of proteins in a typical electrophoresis reaction (13,
that has been replaced in most laboratories. It is based on 16). As many as 12 discrete regions may be visualized with
the classic precipitin reaction, in which antigen and antibod- higher-resolution protein electrophoresis using agarose and
ies react to form precipitates in liquid or semifluid gel media controlled voltage and temperatures (Fig. 1).
FIGURE 1 Agarose gel electrophoresis separation of plasma proteins based on their electrophoretic
mobility. α1Ac, α1-antichymotrypsin; α1Ag, α1-acid glycoprotein; Alb, albumin, α1At, α1-antitrypsin;
AT3, antithrombin III; C1Inh, C1 esterase inhibitor; Cer, ceruloplasmin; CRP, C-reactive protein; FB,
factor B; Fibr, fibrinogen; Gc, Gc-globulin (vitamin D-binding protein); Hpt, haptoglobin; Hpx, hemo-
pexin; α-Lp, α-lipoprotein; β-Lp, β-lipoprotein; α2-M, α1-macroglobulin; Pl, plasminogen; Pre A, preal-
bumin; Tf, transferrin. Reproduced from reference 11 with permission from Elsevier Ltd.
Besides serum analysis, other body fluids are evaluated amount of unlabeled antigen increases, the amount of la-
in the clinical laboratory by electrophoresis. Urine samples beled antigen bound to antibody decreases, resulting in a
are routinely tested in patients with serum monoclonal pro- decrease in detection signal if antibody-bound antigen is the
teins to identify the monoclonal protein components, par- readout or, conversely, increase in detection signal if labeled
ticularly light chains, and CSF can be tested for oligoclonal free antigen is the readout. In the noncompetitive format,
immunoglobulins that may be present in neurological disor- antigens bind to an excess of labeled antibody, forming a
ders such as multiple sclerosis. complex, which increases with increase in antigen in the
sample. ELISA is a highly sensitive method capable of mea-
IFE suring proteins in nanogram or picogram amounts, without
Immunofixation electrophoresis (IFE) is the most common the use of radioactive tags. While ELISA methods may be
method used for establishing the presence and isotype of a used to detect specific antibodies qualitatively or to make
monoclonal immunoglobulin protein in serum and urine semiquantitative measurements of specific antibodies by
(13, 17). The use of a marker dye facilitates visual analysis use of a standard curve calibrated in arbitrary units, there
of the monoclonal protein, and the location of the immuno- are limitations on the accuracy and precision of semiquan-
precipitate depends on the electrophoretic mobility of the titative tests performed by ELISA. Accurate and true quan-
specific monoclonal protein. Most clinical laboratories use titative applications of the basic ELISA method are less
IFE after screening samples by serum electrophoresis and/ common.
or quantitative immunoglobulin measurements. Although More recently, detection of single protein molecules by
serum is the most common sample type, urine or CSF may ELISA has been reported using a novel modification of the
also be examined to detect and/or characterize monoclo- standard ELISA technique (19). This method involves the
nal proteins or their fragments. IFE has several advantages, use of a capture antibody immobilized to a matrix of thou-
including a shorter time to obtain results, increased sensi- sands of microspheres. This assay format results in the mea-
tivity, better resolution, and greater ease of interpretation. surement of individual target molecules, in contrast to the
However, the IFE method is not without its disadvantages, measurement of a collective signal generated in a conven-
which include the enhanced quality, increased quantity, and tional ELISA, and can be used to quantify low-abundance
greater cost of the antisera required. Optimal resolution ne- proteins. However, there are other bead-based, multiplex
cessitates that the exact amount of antisera be determined assays that have greater utility in the clinical laboratory,
for each sample to avoid either excess dilution or excess an- which are discussed in further detail in this chapter. Other
tibody. In addition, because of the increased sensitivity of protein detection methods include agglutination and im-
IFE, smaller bands are more frequently identified, which can munofluorescence; these have been used for identification
confound the clinical interpretation. of pathogenic antibodies typically associated with autoim-
mune diseases.
ELISA
The enzyme-linked immunosorbent assay (ELISA) is prob- Multiplex Methods for Protein Detection and
ably the most widely used immunoassay (13, 18), and it is Quantitation
available in either competitive or noncompetitive formats. The demand for simultaneous, quantitative measurement
In the competitive format, unlabeled antigens compete with of several proteins of diagnostic interest has been growing,
labeled antigens for limited antibody-binding sites. If the leading to a focus on multiplex technology (20–23). While
the benefits of multiplex methodology are apparent, analyt- represents a significant advance in the detection of proteins
ical concerns include artifacts in quantitation due to matrix in biological samples. Studies with standard ELISA methods
effects. Also, cross-reactivity between the capture and de- reveal comparable sensitivity between the two techniques.
tecting reagents, which often are antibodies, can confound Alternatively, sensitivity can be enhanced by using bright
quantitation and data interpretation (24). fluorochromes for antibody detection, as well as optimizing
Among multiplex methods, protein arrays and bead- surface chemistry and engineering antibodies to increase
based technologies are most widely used at present, with valency, or using intermediate detection agents that have
rapidly evolving advances in both their chemistries and for- multiple binding sites, such as streptavidin, or nanocrystals,
mats. There are five types of protein arrays most commonly such as quantum dots (Qdots) (27, 28). Multiplex methods
used. These include planar glass or silicon chips; flow cy- for autoantibody detection are of particular relevance in the
tometric bead arrays; multiplex, microplate immunoassays; field of autoimmune diseases.
and nitrocellulose and three-dimensional microarrays (25); Considerations during selection and validation of a mul-
each has its advantages and disadvantages and specific appli- tiplex assay in the laboratory include evaluation of detection
cations. There are also antibody arrays available for protein reagent and analyte cross-reactivity, stability of reagents,
analysis, which utilize either direct labeling, single-capture, precision across different lots of multiplex kits, multianalyte
or dual-antibody formats (Fig. 2) (25, 26). Alternatively, interactions, the ability to achieve both the dynamic range
antigen or peptide capture arrays can be used with single and relevant sensitivity and specificity, and the impact of
readout antibodies (Fig. 3). The substrates for both of these diverse matrices and biological states on the accuracy of the
microarrays and the signal readout are numerous. They in- final result. The pros and cons of each array approach have
clude microplates, gels, slides, suspension arrays with col- to be carefully considered before selecting the appropriate
orimetry, fluorescence, radioactivity, chemiluminescence, format relevant to the goal of protein quantitation in a spe-
enzyme- linked, nanoparticle, light scattering, and other cific diagnostic context and the analytes being measured
signal-generation and -amplification procedures. (Table 1). Multiplex cytokine analysis offers a good example
To improve the sensitivity (limit of detection) of these of the benefits and disadvantages of multianalyte analysis.
microarrays to detect extremely low concentrations of pro- The most common biological sample for cytokine analysis is
teins, a robust signal-enhancement method called rolling plasma or serum, although conceivably any body fluid ma-
circles amplification is used (27). This modification allows trix could be used. Multiplex cytokine assay formats include
the limit of detection of a protein array assay to be in- planar protein arrays or suspension arrays. Several cytokines
creased into the picomolar and subpicomolar ranges, which can be simultaneously detected and quantified using either
FIGURE 2 Representation of experimental formats for antibody microarrays. (I) Direct labeling of
proteins and detection using a single antibody captured on the microarray. (II) Indirect assessment using a
two-antibody system. The capture and the first detection antibody are matched, and the detector antibody
is measured using a labeled second (readout) antibody.
FIGURE 3 Representation of antigen microarray. Antigens are “caught” on the microarray through
many different processes, which can be generalized as “printing.” Antibodies against specific antigen are
detected using a labeled second detector antibody. TTG, tissue transglutaminase.
the flow cytometry approach— Cytometric Bead Array various multiplex reagent kits, there are instrumentations
(Becton Dickinson, San Jose, CA)—or the specialized in- that offer multiplex technology besides the flow cytometric
strument approach—the Luminex xMAP technology. and Luminex instruments described above. They include
The accuracy of multiplex cytokine analysis methods the Meso Scale Discovery, which uses a microplate format
has been compared to standard ELISA methods. Although with electrochemiluminescence detection using labeled
overall correlations are relatively good, there is loss of such tags that emit light on electrochemical stimulation, and the
correlation for quantitative data (24). A study comparing Pierce Endogen SearchLight platform, which utilizes ELISA
three commercial multiplex cytokine assays revealed vari- technology in a similar 96-well format. A study comparing
ability between kits from different manufacturers for various multiplex cytokine analysis using these two platforms (31)
analytical parameters, such as accuracy, quantitation in dif- revealed that while solid-phase multiplex assays can per-
ferent matrices, recovery, and reproducibility. This indicates form consistently within a platform, optimization and val-
that the introduction of a multiplex cytokine assay necessi- idation experiments should include recombinant cytokine
tates careful optimization and validation (29, 30). Besides spike recovery, linearity based on dilution analysis, stability
TABLE 1 Advantages and disadvantages of different protein arraysa

Types of protein arrays Positive features Negative features
Peptide (antigen capture) Multiplex capability for several Primarily useful for autoantibody
hundred analytes detection and microbial serological
Requires a single detection diagnosis
antibody Extent of quantitation is variable
Sensitivity is variable
Specificity is variable
Single capture antibody Multiplex capability for several Sensitivity is low
hundred analytes Specificity is low
Requires a single antibody for Analysis is semiquantitative and not
antigen capture absolute
Dual antibodies (capture and Absolute quantitation Requires separate antibodies for
readout) Sensitivity is high antigen capture and detection
Specificity is high Limited number of analytes that can
be multiplexed in a single assay
format
a
Adapted from reference 25.
TABLE 2 Factors to consider in selecting multiplex assay colorimetric detection; this provides greater sensitivity.
platforms Also, the capture reagents are prepared by covalent cou-
Factor Relative importance pling of antibodies or antigens to beads, in contrast to pas-
sive adherence by adsorption in the ELISA method, which
Antibody cross-reactivity High results in greater density of capture molecules and firmer
(other detecting reagents) adherence. Therefore, multiplex bead assays have become
Accuracy of calibration curve High popular in the diagnostic setting, though they have not
Source of antibodies used for High completely supplanted ELISA-based methods.
capture and detection Cytokines are not the only analytes that can be mea-
Sample source (serum, High
sured by multiplex analysis. The function of anti-vaccine
plasma, body fluid)
antibodies can be assessed using fluorescent flow cytometric
beads conjugated with antigen. An example is the multi-
Alternate assay reagents and plex opsonophagocytic assay (OPA) that is used to detect
effect on: functional antibody responses to capsular polysaccharides of
Spike recovery High Streptococcus pneumoniae (32) (Fig. 4 and 5). While most
Accuracy Low antibody detection formats offer multiplex quantitation of
Analytical measuring range High IgG antibodies to the 23 serotypes of the pneumococcal
polysaccharides, functional antibody assessment using flow
Nonspecific binding due to Variable, depending on cytometry cannot be practically achieved for 23 serotypes
heterophile antibodies clinical context and sample due to the limitations related to time and complexity of
Consistency of detection of High analysis. Therefore, a clinical OPA can assess at the most
analyte six serotypes in a multiplex combination on a three-laser
flow cytometer, which necessitates careful selection of sero-
types for analysis, primarily based on their immunogenicity
and their incorporation in either the protein conjugate or
of samples, inter-and intraobserver variability, and con- polysaccharide pneumococcal vaccine.
sistent measurement of analytes of interest. In the clinical
laboratory, the decision to choose an appropriate platform Flow Cytometry and Mass Cytometry for Protein
for multiplex cytokine analysis must take into account the Detection
various factors that contribute to variability in analyte mea- This section is not meant to be a comprehensive treatise
surement (Table 2). on the subject of flow cytometry but specifically discusses
The multiplex bead- based assays have a distinct ad- its use in relation to protein detection. Flow cytometry has
vantage over ELISA in that they use direct fluorescent de- been used in the clinical laboratory for a variety of applica-
tection of the analyte, in contrast to an enzyme-mediated tions, including diagnostic immunology for identifying cell
FIGURE 4 OPA for detection of functional antipneumococcal antibodies in serum after vaccination.
The use of fluorescently labeled pneumococcal polysaccharide-coated beads permits flow cytometric as-
sessment of the opsonic capability of antipneumococcal antibodies in serum, in the presence of exogenous
complement, by measuring phagocytic uptake using a differentiated granulocyte cell line.
FIGURE 5 (A) Fluorescent microscopy showing differentiated granulocyte cell line with phagocytosed
(opsonized) fluorescent beads coated with pneumococcal polysaccharide. (B) Flow cytometric analysis
of fluorescent signal from phagocytosed beads. Each bead is coated with a unique polysaccharide, and a
multiplex mixture of beads is used to measure functional antibodies produced for each specific pneumo-
coccal serotype. (C) The reciprocal dilution that demonstrates 50% maximal uptake of labeled beads is
calculated as the phagocytic titer.
populations of interest for both malignant and nonmalig- allows analysis of up to 45 simultaneous parameters without
nant conditions. However, flow cytometry can also be used the use of fluorescent agents or interference from spectral
to look at the expression of particular cell surface receptors overlap (35). Mass cytometry uses stable, nonradioactive iso-
(proteins and glycoproteins) that are either constitutive or topes of rare earth elements (lanthanides) as reporters. Mass
induced by cell activation. CD64, for instance, is expressed cytometry is a modification of inductively coupled plasma
on monocytes. It is constitutive but also induced at high MS for single-cell analysis. Similar to flow cytometry, cells
levels on neutrophils as a result of infection (Fig. 6A). Flow are stained with antibodies labeled with metal isotopes.
cytometry can also be used to perform intracellular protein Cells can be stained with DNA intercalators, which pro-
analysis, including identification of proteins for a diagnos- vide information on DNA content and cell viability. In the
tic purpose (present or absent). Illustrative examples are specialized mass cytometer instrument, the stained cells are
Bruton’s tyrosine kinase (Btk) for the diagnosis of X-linked nebulized into single-cell droplets, and the charged atomic
agammaglobulinemia (XLA) (Fig. 6B); functional alter- ions are subsequently introduced into the mass spectrometer.
ation of proteins, including phosphorylation (Fig. 6C) (33); While there are distinct advantages of the mass cytometry
and cytokine production on cellular activation (see chap- technique, there are certain disadvantages compared with
ter 28). The distinct advantage of using flow cytometry for standard fluorescent polychromatic flow cytometry. The pri-
these applications is the ability to correlate the protein of mary limitation is that the new method does not allow for
interest with its cellular counterpart or association with spe- cell sorting of viable cells for additional analysis. Also, the
cific function or expression under specific conditions. These standard light-based measurements of forward and size scat-
relationships cannot be achieved in a comprehensive man- ter analysis for cell size and complexity, as well as assessment
ner by static in vitro protein analysis. Flow cytometry has of cell proliferation (see chapter 28) and Ca2+ flux, among
also been used to study protein-protein interactions in situ in other analyses, cannot yet be studied by mass cytometry since
live or intact cells using the fluorescence resonance energy metal-reporter equivalents are not available. Also, the sen-
transfer technique (34). While these approaches may sound sitivity of lanthanide-labeled antibodies is lower than that
somewhat futuristic in the clinical laboratory, phosphopro- of most fluorescent labels due to the chelating polymer used,
tein analysis and ex vivo measurement of cytokine-producing and this could pose difficulty for analysis of cellular targets
T cells is already part of the diagnostic armamentarium of with low signal-to-noise ratios. Furthermore, the throughput
specialized reference laboratories. of most commercial flow cytometers is much higher than that
Similar to multiplex immunoassays, multiplex pro- of the mass cytometer, which is limited to ~1,000 cells/s. At
tein analysis by flow cytometry has made major advances. present, the best use of the mass cytometer is for the anal-
Fluorescence-based cytometry can quantify ~18 proteins at ysis of intracellular regulatory molecules (cytokines and
a rate of >10,000 cells/s. However, this can be exponentially phosphoproteins) for which autofluorescence can pose a con-
increased using the new technology of mass cytometry, which founder or in contexts of numerous simultaneous measure-
combines flow cytometry and MS and can measure >36 pro- ments (35). Complex immunophenotyping and rare-event
teins at a rate of 1,000 cells/s. This advanced methodology analysis, on the other hand, may currently be best suited
A.
A.
A.
B.
B.
B.
C.
C.
C.
FIGURE 6 (A) Flow cytometric expression of CD64 on “resting” neutrophils (left panel), which is essentially absent because it is not
a constitutive marker in this cellular subset. CD64 expression on neutrophils from a patient with bacterial sepsis (blue peak, right panel),
as it is a marker expressed on neutrophil activation specifically in the context of infectious stimuli. (B) Flow cytometric analysis of an in-
tracellular protein, Btk, in B cells from a healthy donor (left panel) and in monocytes (Monos) from a healthy donor (middle panel), and
the absence of Btk protein in monocytes from a patient with XLA (right panel). Red peak, isotype control; blue peak, specific anti-human
Btk antibody. (C) Flow cytometric analysis of modification (e.g., phosphorylation) of intracellular cell signaling proteins. Alterations in
function of cell signaling pathways can be assessed after cell stimulation and activation. The assay format can be singleplex (stimulus 1 or
stimulus 2 only) or multiplex (stimulus 1 and stimulus 2 in combination), assessing different cell signaling proteins simultaneously (i.e.,
single experiment). Panel C reproduced from reference 11 with permission from Elsevier Ltd.
for flow cytometry. However, since there is rapid evolution laboratory, shared samples between laboratories at different
of instrumentation, reagents, and analytical software, these sites, or chart review for protein tests that are used as screen-
boundaries may blur with time and permit use of each tech- ing or diagnostic tests for a single clinical condition. In all
nology interchangeably or for highly specific applications in cases, whether regular PT or AAP is performed, there must
the clinical laboratory. be appropriate documentation of PT and test results.
QA AND QC ISSUES IN PROTEIN ANALYSIS PROTEIN ANALYSIS: PROTEOMICS USING MS

Every clinical laboratory must have a robust quality assur-
ance (QA) and quality control (QC) program to ensure Ionization Techniques
that diagnostic testing meets regulatory standards and is Mass spectrometers can now be found in many laboratories
optimized for use in patient care. The QA program ensures actively providing molecular mass data for proteins and
that appropriate checks are present for the routine ana- peptides from a wide variety of biological matrices. Every
lytical process of testing but also for the elements related mass spectrometer consists of three parts: an ion source, an
to analytical and clinical validation prior to introduction analyzer, and a detector, all of which are linked to a vacuum
of a new test. These include preanalytical issues such as system. Ions created in the source are separated in the an-
sample stability, anticoagulant, and interfering substances; alyzer region and then detected. The two most widely used
analytical components including accuracy and precision, ion sources for MS-based proteomics are electrospray ioniza-
reportable range, and reference values; and postanalytical tion (ESI) (37) and matrix-assisted laser desorption ioniza-
parameters of test reporting, interpretation, and test utili- tion (MALDI) (38). These two ionization techniques have
zation. For appropriate use of clinical laboratory tests, there played a key role in the acceptance of mass spectrometers as
must be guidelines that define the clinical utility, strengths, practical protein chemistry lab tools. ESI and MALDI cre-
and limitations of the assay. Additionally, implementation ate intact molecular ions from high-molecular-mass proteins
of a new test has to take into account appropriate training of without destroying the polypeptide chain and are therefore
lab personnel; documents detailing instrument evaluation referred to as “soft” ionization techniques. Once in the gas
and maintenance for robust performance; and strategies for phase, protein ions can then be separated by their mass-to-
handling poor- quality results, improperly handled speci- charge ratio (m/z) in the analyzer region.
mens, and inaccurate data entry or reporting of test results.
In the case of protein analysis, instrumentation and reagents ESI
are a critical aspect of test maintenance and performance in The ESI process transfers ions in solution to ions in the gas
the laboratory. Therefore, protocols have to be developed phase at atmospheric pressure. ESI sources continuously
to monitor both instruments and reagents, including reg- create ions as proteins and peptides arrive at the source
ular instrument validation and lot-to-lot comparison of re- through a solution running from a syringe pump or a liq-
agents, using predefined acceptance criteria for clinical use uid chromatograph to an electrically conductive capillary
of a new lot of a reagent. Certain methodologies also can held at a high electric potential (3,000 to 6,000 V). A spray
pose unique QA challenges, particularly multiplex assays is created as the solution reaches the exit of the capillary
(36). These have to be considered in detail before selecting (often referred to as a Taylor cone) due to the electrostatic
the assay as well as in developing QA parameters for moni- field between the end of the capillary and the entrance to
toring assay performance. QA programs also involve the use the mass spectrometer. The ions in the solution are trans-
of QC materials, which can include calibrators, standards, ferred to the gas phase as spray droplets burst due to charge
and other well-characterized QC reagents available through repulsion and are further reduced in size due to heated gas
national or international agencies. Different levels of QC that evaporates the solvent surrounding the ions. When
assessment have to be performed for both instruments and the ESI emitter is positively charged, proteins and peptides
assays. QC reagents for assays should include those that as- are positively charged via protonation, while a negatively
sess the analytical measuring range, e.g., low, medium, and charged ESI emitter results in negatively charged, deproton-
high controls. The types of QCs used may vary depending ated proteins and peptides. The mechanism of ESI creates
on whether the assay is qualitative or quantitative. For the multiple ions from the same protein, each with a different
former, it may be sufficient to have materials that include number of charges and therefore a different m/z value (39).
those with negative and positive results, while for quantita- For a molecule such as an immunoglobulin light chain with
tive assays, a more graded QC assessment may be required, a molecular mass of 23,000 Da, a series of charges ranging
with measurement of the high and low end of the range, and from +10 to +25 is observed. Because of this, ESI spectra
possibly an intermediate level. can be complex due to the fact that a single protein has
Another part of a clinical QA system is regular perfor- multiple peaks. The molecular mass is determined after
mance of proficiency testing (PT), which evaluates the the multiply charged peaks are converted from the mass-
entire process within the laboratory. This includes sample to-charge (m/z) domain to the uncharged molecular mass
assessment for stability, interfering materials, analytical (Da) domain, through the aid of a computer program (40).
measurements, and postanalytical data reporting and inter- Figure 7 shows an ESI mass spectrum acquired on a quad-
pretation. For some protein analytes, there are standardized rupole time-of-flight mass spectrometer (Q-TOF) showing
PT materials available through regulatory agencies or pro- the κ light chain of the therapeutic monoclonal antibody
fessional societies (e.g., College of American Pathologists) adalimumab after reduction with dithiothreitol. The mass
in which multiple laboratories participate. For other more spectrum at the top of the figure shows all the multiply
esoteric protein-based assays, especially those using flow cy- charged ions from the monoclonal light chain of adalim-
tometry or MS, standardized PT materials are not available. umab, while the bottom figure shows a single peak after the
In these cases, laboratories must develop their own alterna- multiply charged ions were converted to molecular mass.
tive assessment of proficiency (AAP), which is performed The average molecular mass observed was 23,412.19 Da,
similarly to PT, at least twice a year, by the clinical lab which matches closely with the known average molecular
staff. The AAP process can include split samples within the mass of 23,412.13 Da. The spectrum in Fig. 7 was acquired
FIGURE 7 Mass spectrum of a monoclonal immunoglobulin light chain protein showing the multiply
charged ions produced by ESI (top). Each peak represents the same light chain protein with a different
number of protons attached, which changes each ion’s mass/charge ratio (m/z). The molecular mass of the
light chain is determined by converting each peak to the uncharged state through an algorithm performed
by a computer program (bottom).
at a liquid chromatography (LC) flow rate of 25 μl/min with same sample shown in Fig. 7 of adalimumab monoclonal light
a 2-μl injection, but ESI can be performed at many different chains after reduction with dithiothreitol collected on a TOF
flow rates ranging from nanoliters per minute up to a milli- mass spectrometer. The difference in the number of charge
liter per minute. Injection volumes for a sample can range states produced for the light chain of adalimumab by ESI
from a few microliters to a few hundred microliters. versus MALDI is easy to see when the two spectra are com-
pared. Because MALDI uses a laser to produce ions, it can be
MALDI focused on a particular spot to produce an ion image from a
MALDI, on the other hand, produces ions that have much tissue sample. MALDI MS imaging is a relatively new tech-
lower charge states compared with ESI. For example, the pre- nique that is becoming more prevalent in clinical pathology
dominant ions created for an immunoglobulin light chain are laboratories (42). There are also new ionization techniques
typically +1 and +2. MALDI typically creates ions under that combine dried samples and ESI, such as desorption ESI,
vacuum instead of atmospheric pressure using a saturated that are gaining ground as tools for monitoring tissue samples
solution of an organic acid mixed with the analyte solution in real time (43). Together, ESI and MALDI make up the
and then dried to form a crystalline matrix. The plate con- vast majority of ion sources on instruments that analyze pro-
taining the dried spots is transferred into a vacuum via an teins and peptides for clinical assays.
airlock system. Then a laser is fired at the matrix to form
ions. The organic acid matrix absorbs the light energy from Mass Analyzers
the laser and rapidly heats, creating a plume where protons The triple-quadrupole mass spectrometer is the most com-
from the matrix are transferred to the analyte in positive-ion mon mass spectrometer encountered in clinical laborato-
mode (41). Figure 8 shows a MALDI mass spectrum from the ries due to its ruggedness and reliability. The instruments
instruments can be operated in scanning mode, which al-

lows ions with different m/z to be transmitted through the
detector, or they can be set to transmit only ions with a spe-
cific m/z. Triple-quadrupole mass spectrometers are used in
the clinical laboratory due to the high level of sensitivity and
specificity obtained using three different quadrupole analyz-
ers in tandem. Ions entering a triple-quadrupole instrument
first encounter a single quadrupole (Q1), followed by a col-
lision cell (this is not a true quadrupole but is often referred
to as Q2), followed by another single quadrupole (Q3). The
first quadrupole is typically set to transmit a specific m/z,
which is transferred to the collision cell, where neutral gas
molecules (usually nitrogen) collide with the ions to form
fragment ions. The fragment ions produced can be scanned
in the third quadrupole, or the third quadrupole can be set
to transmit a specific m/z that is unique to a fragment ion for
the analyte. This series of events is called tandem MS, or
MS/MS. When the first quadrupole is set to transmit a spe-
cific m/z to be fragmented in the collision cell and the third
quadrupole is set to transmit a specific fragment ion m/z, the
experiment is called selected reaction monitoring, or SRM
(also referred to as multiple reaction monitoring, or MRM).
This type of experiment is the mainstay of quantitative MS
performed in clinical laboratories. Figure 9 shows a graphi-
cal representation of the SRM process.
Quadrupole ion trap mass spectrometers are another
commonly used mass spectrometer in clinical laboratories.
An ion trap is designed as either a three-dimensional ion
trap where the ions oscillate in a circular fashion or a two-
FIGURE 8 Mass spectrum of the same monoclonal immuno- dimensional linear ion trap where the ions move back and
globulin light chain protein shown in Fig. 7 ionized using MALDI. forth. Both geometries use the same principles of modifying
The spectrum clearly shows the prevalence of the +1 and +2 DC polarity and radio-frequency AC voltages to keep ions
charge states as compared with the highly charged ions created by in a quadrupole trap. Instead of being transmitted down
ESI. The molecular mass of the light chain with the +1 charge a linear path as in a triple- quadrupole mass spectrome-
state is determined by subtracting the mass of a proton. ter (45), ions are injected into the ion trap, excited, and
then allowed to oscillate in the trap before being ejected
and detected. The resolution of a quadrupole ion trap can
be upwards of 4,000 with a mass measurement accuracy of
typically run at a resolution of 1,000 with mass measurement ~40 ppm. The benefit of quadrupole ion traps is their scan-
accuracy of ~40 ppm. The geometry of a quadrupole instru- ning speed, which can be orders of magnitude faster than a
ment is just as its name implies; it is made up of a set of four quadrupole instrument. They are often used for structural
identical parallel rods (~20 cm in length and 1 cm in diame- analysis of peptides because they have the capacity to do
ter). Rods that are opposite each other have the same polar- MS/MS/MS. A specific peptide ion can be trapped and then
ity and the same DC voltage. An additional radio-frequency fragmented multiple times all in the same experiment and
AC voltage is applied to all rods. As the polarity and the at scanning rates much faster than with a triple-quadrupole
voltages are changed on the rod pairs, the electrical field instrument. Quadrupole ion traps also have the added ben-
oscillates and ions with the correct m/z remain stable and efit of being able to use different dissociation techniques to
are transmitted through the quadrupole (44). Quadrupole fragment ions, such as electron transfer dissociation (46).
FIGURE 9 A graphical representation of the transmission through a triple-quadrupole mass spectrom-

eter for a specific proteotypic peptide ion quantified using SRM. The intact peptide ion is created by ESI
and is represented by the green balls. The peptide ion is selected in Q1 and fragmented in the collision cell
(Q2), and then a proteotypic peptide-specific fragment ion is transmitted through Q3 on to the detector.
Such new methods can generate fragment ions containing

posttranslational modifications that are not observed using
other fragmentation techniques. One issue with ion traps
is the space-charging effect, which can limit the linear dy-
namic range of the instrument and is not observed in triple-
quadrupole mass spectrometers. Instrument manufacturers
have employed techniques such as automatic gain control
to limit the number of ions entering the trap, allowing new
instruments to better manage space charging.
TOF mass spectrometers can also be found in the clinical
laboratory and are well suited for measuring the molecular
mass of large proteins (47). TOF instruments consist of a
flight tube in which the m/z of an ion is determined by the
time it takes to travel a set distance between a start posi-
tion and the detector. This is called the flight time, and it is
dependent on the ion’s mass and the voltage on the pulser
that forces the ions into the flight tube (48). There are two
common types of TOF instruments, linear and reflectron.
Linear TOF instruments measure the time it takes an ion
to travel between two set points in a linear flight path from
the ion source to the detector. A reflectron instrument con-
tains a flight tube that includes a series of lenses that slow
the ions and then force them to reverse direction on to the
detector. The ions travel farther in a reflectron instrument
and are more focused compared with a linear instrument,
resulting in better resolution. There are differences in the
resolution and mass measurement accuracy of ESI-TOF ver-
sus MALDI-TOF instruments due to the kinetic energy of
the ions after they are created. In an ESI-TOF instrument,
the ions enter the TOF region at a 90° angle, creating the
orthogonal-TOF geometry. The steady stream of ions from
the ESI source is pulsed into the flight tube in a tight beam,
making their mass measurement highly accurate. MALDI-
TOF instruments have the ions enter the flight tube after
the laser pulse when the ions are dispersed throughout the
ablated matrix plume. To compensate for the ion dispersion,
delayed extraction is used to make a tighter ion beam before
the ion packet is accelerated through the flight tube to the
detector. Advances in the speed and data transfer rates of
the electronics used in TOF instruments have made them
faster and more reliable, enabling resolution of upwards of
30,000 with a mass measurement accuracy of 2 ppm for most
high-end ESI-TOF instruments.
The newest generation of mass spectrometer is the Or-
bitrap, based on a new design developed by Makarov (49).
The Orbitrap measures an ion’s m/z by collecting the image
current produced while the ions oscillate around an elec- FIGURE 10 Diagram of an Orbitrap analyzer showing how ions
trode assembly. The mass of each ion is found by converting are injected into the Orbitrap via the C-trap and then allowed
the image current to m/z using the Fourier transform. The to orbit the isolated electrodes. The induction current made by
instrument is capable of attaining resolution of 240,000 the ions is detected and then converted to m/z using the Fourier
with a mass measurement accuracy of <1 ppm. The high transform.
resolution is a function of the high precision and accuracy
with which the frequency of the ions orbiting in the cell
can be measured via the image current. The Orbitrap it- transmitted, or as an ion selector, in which only ions with
self is <10 cm in length and is kept at very low pressure. a specific m/z are transmitted. The first generation of Or-
The ions are injected into the Orbitrap (Fig. 10) using a bitraps was hybrid instruments with a linear ion trap (LIT)
C-trap that focuses the ions to a point within the cell where in front of the Orbitrap, while the second generation com-
they begin their orbit around the center electrode. Orbitrap bined a single quadrupole with the Orbitrap. Instruments
instruments were first introduced in research laboratories, with a LIT most often generate ion fragments in the LIT,
but they are becoming more popular in clinical laboratories while instruments with a quadrupole have a separate colli-
with the advent of new instruments that combine the robust sion cell called a higher-energy collisional dissociation cell
attributes of quadrupole instruments with the Orbitrap. for generating fragment ions for structural analysis. In Q-
Orbitrap and TOF mass spectrometers are often coupled TOF instruments, a quadrupole is coupled with a collision
with quadrupole mass spectrometers, resulting in a hybrid cell in the same manner as in a triple-quadrupole instrument
instrument. The reason for combining the two instruments and the third quadrupole is replaced by a TOF analyzer.
stems from the fact that a quadrupole can function as an ion Both Orbitrap and TOF hybrid instruments offer superior
guide, in which all ions within a certain m/z range can be scanning speed, resolution, and mass measurement accuracy
compared with triple-quadrupole and quadrupole ion trap are then analyzed by LC-MS/MS to obtain amino acid se-
mass spectrometers. While triple-quadrupole instruments quence information. The mass of the peptide combined
with ESI sources have been the primary instrument for per- with the amino acid sequence found by MS/MS is used to
forming absolute quantification in the clinical laboratory, reconstruct each protein in the sample from the bottom up.
Orbitrap and Q-TOF mass spectrometers with ESI sources The first b ottom-up experiments in the early 1990s used the
are being used more often for both qualitative and quanti- accurate mass of the tryptic peptide without the use of MS/
tative analysis. These hybrid instruments have a promising MS amino acid sequence. The tryptic peptide masses ob-
future in the clinical laboratory as they have the ability to served were compared to the masses of the tryptic peptides
provide high resolution and excellent mass measurement expected from the protein database (61). This approach led
accuracy to address specific clinical questions related to un- to a high level of uncertainty due to large numbers of peptides
known compounds and interferences. with nearly the same intact mass but different amino acid
sequences. Many of the first bottom-up experiments relied on
LC Coupled with MS SDS-PAGE to separate proteins followed by cutting out of
High-performance LC has been used in clinical laboratories the stained bands followed by in-gel digestion (61). Tryptic
for decades, typically coupled with optical detectors such peptides were extracted from the gel and subjected to MS/
as UV-visible spectrophotometers. Because ESI is a flowing MS to obtain peptide sequence information to compare to a
liquid-based ionization technique, it is perfectly suited to be protein database. The selection of the tryptic peptides to be
coupled with LC (50). Manufacturers of mass spectrometers sequenced by MS/MS was done manually at first, as was the
used in clinical laboratories now provide ESI sources that protein database searching (62, 63).
are robust and can easily handle running large volumes of By the turn of the millennium, advancements in auto-
samples using standard-flow LC methods. These ESI sources mated peptide MS/MS acquisition combined with auto-
can be directly coupled to an LC with flow rates ranging mated protein database search engines enabled Yates and
from 10 to 1,000 μl/min and incorporate high temperature coworkers to coin the term “shotgun proteomics.” This re-
(up to 600°C) and heated gas to remove excess solvent to fers to the broad swath of protein identifications that could
optimize the number of ions that enter the mass spectrom- be generated in one LC-MS/MS experiment (63, 64). At
eter (51). High-volume clinical laboratories often use mul- present, there are any number of mass spectrometers cou-
tiplexed LC systems in which up to four columns running pled with commercially available protein database search
the same method take turns eluting into the ESI source in engines that can provide tens of thousands of MS/MS spec-
a “leapfrog” approach (52). These multiplexed systems may tra, and thousands of protein identifications, from a single
also incorporate two-dimensional LC, or LC/LC, in which analysis, all performed in a matter of hours (65). A diagram
columns are set in tandem to further purify the sample in of the shotgun proteomics process is shown in Fig. 11. While
a completely automated fashion. Between standard- flow the majority of shotgun proteomics has been performed in
and nanoflow LC systems are the micro LC systems that a research setting, some clinical laboratories have adopted
run at 10 to 50 μl/min. A micro LC uses less solvent and the approach to identify proteins present in biopsy samples,
less sample and can be easily connected to a standard ESI in which specific regions of the tissue are removed using
source. Micro LC systems are not currently multiplexed, laser capture dissection and then analyzed using shotgun
and the number of LC columns commercially available is proteomics (66, 67). The laser capture dissection-shotgun
limited compared with high-flow systems. In practice, one- proteomics approach is essentially equivalent to immuno-
dimensional LC is still the most commonly used LC setup histochemistry using thousands of protein-specific antibod-
in the clinical laboratory setting. Regardless of the LC flow ies. However, the mass spectrometer is a nonbiased detector,
rate, the majority of peptide and protein LC separations are and it provides information on all the proteins present in
done using reverse-phase LC columns with water and ace- the sample via tryptic peptide sequences. This results in
tonitrile, or methanol, gradients containing formic acid for more-sensitive and -specific results.
ion pairing when coupled with ESI (53).
MS/MS Analysis of Proteolytic Peptides To Quantify
Using MS to Analyze Peptides and Proteins Proteins Using SRM
Mass spectrometers have played a central role in the iden- Currently, the majority of MS-based clinical laboratory as-
tification, structural characterization, and quantification of says that quantify peptides quantify a protein-specific pep-
proteins and peptides in research laboratories for decades. tide created by digesting a sample with a protease. This
Since the early 1980s, MS has been used to quantify en- approach was first described by Barr and coworkers (68) and
dogenous peptides of clinical importance, with many of the has since been widely adopted by clinical laboratories to
first studies focusing on neuropeptides (54, 55). Currently, quantify a wide range of proteins (69–73). The peptide rep-
clinical laboratories are beginning to use mass spectrome- resenting the protein to be quantified is commonly referred
ters for quantifying endogenous peptides such as insulin-like to as a proteotypic peptide (74). A proteotypic peptide rep-
growth factor 1 and insulin (56–58), as well as endogenous resents the stoichiometric equivalent of the intact protein
peptide biomarkers (59). However, the majority of work is in the sample. This approach works when the sample is
done using peptides obtained by digesting the sample with complex, such as serum, and the intact protein is present at
an enzyme before using MS. a low concentration because the response by ESI is higher
for peptides than for proteins.
MS/MS Analysis of Tryptic Peptides To Identify The best proteotypic peptide to use for quantification
Proteins can be found empirically by digesting a pure form of the
Mass spectrometers routinely identify and quantify peptides protein if it is available. If purified protein is not available,
from the proteolysis of protein mixtures. This is accom- then the best proteotypic peptides can be identified in silico
plished by what is often referred to as “bottom-up proteom- using computer software (75). The software uses the known
ics” (60). The term gets its name from the fact that an sequence of the protein and the enzyme that will be used to
enzyme with known cleavage specificity, such as trypsin, is generate a list of the proteotypic peptides along with sug-
first used to digest proteins into peptides. The tryptic peptides gested collision energies. In addition, databases are being
FIGURE 11 Stepwise depiction of a shotgun LC-MS/MS experiment. The figure shows how a tryptic
peptide mixture is first separated by LC, ionized, and then scanned in the mass spectrometer. Tryptic pep-
tides with sufficient signal are automatically selected for fragmentation, and the observed MS/MS data are
automatically compared to a protein database. A list is provided of the best matches to tryptic peptides
derived from proteins in the database.
created that give the proteotypic peptides that have been concentration in every sample to serve as an internal stan-
successful using publicly available data. There are numerous dard. Peak areas from selected fragment ions are then used
examples of assays being developed to quantify upwards of to calculate the concentration in unknown samples in the
50 proteins in serum and urine (76) using SRM on a triple- same manner as other LC-MS/MS assays quantifying small
quadrupole mass spectrometer with an ESI source and a molecules.
high-flow LC.
Research groups continue to push the number of pro- MS/MS of Intact Proteins: Top-Down MS
teins that can be quantified in a digest of plasma, serum, The bottom-up LC-MS/MS approach to protein identifica-
and urine by means of relative quantification. However, tion and quantification has a proven track record. However,
absolute quantification is the norm in clinical laborato- there is the argument that information that could prove use-
ries, which typically involves the use of a traceable protein ful in clinical diagnostics is lost by cleaving the intact pro-
standard. The ideal standard for protein quantification by tein into peptides. Researchers continue to create new ways
SRM is a 13C and 15N stable isotope-labeled protein with to identify proteins by LC-MS/MS without digesting them
the same amino acid sequence as the native protein. Stable first. This approach is referred to as “top-down proteomics”
isotope-labeled standards have the same solution chemis- and involves the fragmentation of intact proteins followed
try as the native protein but have a different mass due to by piecing together the fragments in the same manner as
the added neutrons in the carbon and nitrogen nuclei. This protein database searching in shotgun proteomics (78, 79).
allows the mass spectrometer to distinguish them from the There are a number of technical challenges that are faced
native protein. Limited studies have been done using sta- when trying to identify a protein using fragment ions pro-
ble isotope-labeled protein standards because the proteins duced from the intact protein. The majority of top-down
are made using recombinant expression systems (Escherichia experiments use ESI to generate protein ions to be analyzed
coli, mammalian cell culture, etc.) and are not always easy in the mass spectrometer. Because ESI generates multiple
to produce. Stable isotope-labeling reagents may also be charge states for the intact protein, it also creates fragment
cost-prohibitive. Examples of clinical studies using a sta- ions with multiple charge states. Most tryptic peptides have
ble isotope-labeled recombinant standard protein include +2, +3, or +4 charge states with fragment ions that primar-
the absolute quantification of urinary albumin (73), serum ily have a +1 charge state. In the case of proteins, there may
parathyroid hormone (72), and C-reactive protein (77). Al- be upwards of 20 different charge states for the same pro-
ternatively, a stable isotope-labeled synthetic peptide with tein, and when a specific charge state is fragmented, it can
the same amino acid sequence as the proteotypic peptide generate hundreds of multiply charged fragment ions. Most
can be produced at a reduced cost and can serve as an in- top-down proteomics is done on instruments that have high
ternal standard to monitor retention time and fragment ion resolution (>30,000) and high mass measurement accuracy
ratios. Quantification is most often done using a standard (<2 ppm). This aids in the interpretation of the fragment
curve and QC samples made using purified protein diluted ions using software specifically designed to identify a protein
in a similar matrix that is digested in the same way as the from the multiply charged intact protein ions and the large
samples. Stable isotope-labeled peptide is added at the same collection of multiply charged fragment ions (80). Another
challenge when doing top-down analysis of proteins is sepa- Intact Proteins—Phenotyping Proteins Using MS
rating the proteins prior to ESI. LC is the method of choice; MS has been used in the clinical laboratory to identify
however, additional separation techniques such as prepara- phenotypic changes in proteins by observing the shift in the
tive electrophoresis, in which proteins are first separated by molecular mass of the protein compared to the normal wild-
their pI, have been reported (81). Clinical applications of type molecular mass. One of the first examples reported
top-down protein analysis include determining the muta- right after the discovery of ESI was the detection of muta-
tion sites in transthyretin and hemoglobin from dried blood tions in hemoglobin (85). Since that time, many clinical
spots (82, 83). Top-down analysis is also being done on im- groups have described using LC-ESI-MS to identify muta-
munoglobulin light chains from patients with multiple my- tions in hemoglobin, and the assay is routinely performed
eloma, whereby the isotype of the light chain is determined in clinical labs because hemoglobin can be obtained easily
by the singly charged fragment ions that originate from the in high concentrations from red blood cells (86). Another
constant region of the κ or λ light chain (84). Figure 12 high-abundance protein that has been monitored for mu-
shows an example of the top-down fragment ion spectrum tations using MS is transthyretin, as certain mutations are
acquired for a κ light chain. Although advances are being known to be associated with amyloid deposition (87). An
made in top-down protein analysis due to advances in high- online immunoaffinity- LC-ESI-
MS assay was developed
resolution mass spectrometers with high mass measurement to purify transthyretin from serum to obtain an accurate
accuracy, the technique has yet to be used routinely in a molecular mass for transthyretin. Another protein pheno-
clinical laboratory setting. type assay developed for the clinical laboratory is an online
FIGURE 12 Top-down MS of adalimumab spiked into normal serum. The ion at m/z = 1,233 in the
top spectrum matches the +19 charge state ion from the κ light chain of adalimumab and was selected for
top-down MS. The arrow points to the fragment ion mass spectrum shown below. The labeled fragment
ions match the expected masses for fragment ions from the C-terminal portion of the κ light chain, which
contains the constant region. The calculated y ion masses for the κ light chain constant region-specific
amino acid sequence are shown in the table.
immunoaffinity- LC-ESI-MS test for evaluating carbohy- proteins or digested for analysis of proteotypic peptides. The
drate on transferrin (88). All of these examples use the mass reverse holds true for small proteins and peptides, for which
spectrometer to provide an accurate molecular mass of the the supernatant from an ammonium sulfate protein precip-
patient sample compared to a normal wild type. If the mo- itation can be analyzed instead of the pellet. Other meth-
lecular mass of the patient sample does not match the nor- ods of precipitating large proteins can also be used, such as
mal sample, it indicates that a mutation is present. In the adding organic solvent to create a protein pellet. There is
case of transferrin, a mass shift matching a change in car- also the option of using molecular weight cutoff filters to
bohydrate structure indicates an alteration in metabolism. separate proteins and peptides. Various sizes are available
and work due to centripetal force, preferentially forcing an-
Clinical Samples Analyzed by MS alytes of a specific size through a filter. Alternatively, solid-
The higher the concentration of the protein or peptide in a phase extraction can be used to capture peptides or proteins
sample, the less sample preparation is needed prior to anal- to remove low-molecular-weight compounds or to partition
ysis. It is not surprising that proteins such as hemoglobin analytes with certain hydrophobicity, basicity, or acidity
require minimal sample preparation due to the high concen- attributes.
tration of the protein in red blood cells. When evaluating Immunopurification can also be used to purify a protein
the preparation needed for proteins, one must also consider or peptide prior to analysis using MS. When an immunocap-
molecular mass and posttranslational modifications. As a ture method is used, reagents used to remove the protein or
general rule, the higher the mass, the less response, due to a peptide from the antibody need to be cleared from the sample
number of physical factors in the ionization process whether before ionization. One of the more successful uses of immu-
using ESI or MALDI. If a protein has a mix of possible post- nocapture for clinical proteomics is stable isotope standards
translational modifications, this reduces the response for any and capture by antipeptide antibodies (SISCAPA) (90).
one species. The three-dimensional structure of the protein This method is based on capturing proteotypic peptides us-
also alters the ionization efficiency, and denaturation may be ing antibodies raised against a specific proteotypic peptide.
needed to obtain the best signal (89). If a protein or peptide The technique dramatically lowers the level of detection
needs to be purified prior to analysis, there are a number of for low-abundance proteins (91) and can be automated for
methods that are often used to prepare clinical samples. One high-throughput clinical laboratories. Ideally, a purification
of the fastest and most inexpensive ways to isolate proteins method should end with the final step having the analyte
is to use saturated ammonium sulfate to bring proteins out in a solvent matrix that will allow for the best separation
of solution. The pellet containing the sample can then be when using LC-ESI-MS (92) or efficient ionization when
resolubilized, desalted, and then analyzed directly for intact doing MALDI MS (93). As a general rule, volatile buffers
TABLE 3 Key concepts in clinical MS

Test type Common instrument format Examples and references
Peptide quantitation Standard-flow LC system coupled to an ESI source on a triple-quadrupole Insulin-like growth factor
mass spectrometer. The LC system can be multiplexed for high and human growth
throughput. Sample quantification is done by using a standard curve hormone (97); see also
using a purified peptide standard. reference 59
Protein quantification using Standard-flow LC system coupled to an ESI source on a triple-quadrupole Zinc-α2 glycoprotein
a proteotypic peptide mass spectrometer. The LC system can be multiplexed for high and prostate-specific
throughput. Sample quantification is done by using a standard curve antigen (70, 71); see
using a protein standard. The sample is digested and a protein- also reference 98
specific proteotypic peptide is monitored by SRM. A stable isotope-
labeled peptide or stable isotope-labeled intact protein may be used
for quantification. Multiple proteins can be quantified at the same
time. Proteotypic peptides can be enriched using immunoaffinity
(SISCAPA).
Protein phenotyping—intact Standard-flow LC system coupled to an ESI source coupled to a References 87, 88, and 99
quadrupole, Q-TOF, or TOF mass spectrometer. MALDI can also be
used. Protein can be immunopurified offline or online. The mutation
is identified via a mass shift observed in the intact protein molecular
mass.
Protein phenotyping— Standard-flow LC system coupled to an ESI source coupled to a triple References 85, 86, and
mutation-specific peptides quadrupole, Q-TOF, or TOF mass spectrometer. MALDI can also be 100
used. The LC system can be multiplexed for high throughput. The
sample is digested and a mutation-specific peptide is monitored by
SRM. Phenotype is determined using stable isotope-labeled peptides
and mutation-positive samples.
Proteomic profiling of tissue- Nanoflow LC system coupled to an ESI source coupled to an Orbitrap References 67, 101, and
derived samples mass spectrometer. The sample is extracted from a slide containing 102
the tissue using laser capture dissection. The sample is digested and
shotgun proteomics is performed with protein database searching.
Phenotype is determined by the protein identifications from the
database search.
such as ammonium acetate and ammonium bicarbonate at variability in flow cytometric evaluation of reduced nico-
concentrations of <50 mM are acceptable. Inorganic salts tinamide adenine dinucleotide phosphate oxidase function
and acids can interfere with LC separation and ionization in patients with chronic granulomatous disease. J Pediatr
and should be avoided. Ionic detergents such as SDS are 128:104–107.
also problematic and should be removed or exchanged with 10. Vowells SJ, Sekhsaria S, Malech HL, Shalit M, Fleisher
a nonionic detergent. Two-dimensional LC systems can also TA. 1995. Flow cytometric analysis of the granulocyte re-
be used to reduce matrix effects from the sample buffer by spiratory burst: a comparison study of fluorescent probes.
trapping proteins and peptides on a trap column, where they J Immunol Methods 178:89–97.
are held while the matrix is flushed to waste before eluting 11. Abraham RS, Barnidge DR, Lanza IR. 2013. Assessment
of proteins of the immune system, p 1145–1159. In Rich
the sample onto the analytical column (94).
RR, Fleisher TA, Shearer WT, Schroeder HW Jr, Frew AJ,
LC-ESI-MS assays, as with any other clinical assay, have Weyand CM (ed), Clinical Immunology: Principles and Prac-
their level of quantification based on the analyte concentra- tice, 4th ed. Elsevier Saunders, Philadelphia, PA.
tion that is clinically relevant. In general, proteins or pep- 12. Bradwell AR, Harding SJ, Fourrier NJ, Wallis GL,
tides at or above 1 μM in concentration are relatively easy Drayson MT, Carr-Smith HD, Mead GP. 2009. Assess-
to quantify by LC-ESI-MS. For SRM methods, much of the ment of monoclonal gammopathies by nephelometric
work is still trial and error. A digest is done on the sample measurement of individual immunoglobulin κ/λ ratios.
and the expected proteotypic peptides are monitored. If a Clin Chem 55:1646–1655.
response is observed, then the specificities of the transitions 13. Homburger HA, Singh R. 2008. Assessment of proteins
are examined. The most effective way to determine speci- of the immune system, p 1419–1434. In Rich RR, Fleisher
ficity is to have a stable isotope-labeled peptide synthesized TA, Shearer WT, Schroeder HW Jr, Frew AJ, Weyand CM
with the same amino acid sequence as the proteotypic pep- (ed), Clinical Immunology: Principles and Practice, 3rd ed.
tide. If the LC retention time and fragment ion ratios of Mosby Saunders, Philadelphia, PA.
the stable isotope-labeled peptide match the native peptide, 14. Whicher JT, Price CP, Spencer K. 1983. Immunonephe-
then specificity is established. This process becomes difficult lometric and immunoturbidimetric assays for proteins. Crit
to manage as the number of proteotypic peptides increases; Rev Clin Lab Sci 18:213–260.
however, it is becoming easier to manage due to advances 15. Keren DF. 2003. Protein Electrophoresis in Clinical Diagnosis.
in software and tryptic peptide databases that provide infor- Arnold, London, United Kingdom.
mation on the proteotypic peptides that have already been 16. Keren DF, Humphrey R. 2006. Clinical indications and
applications of serum and urine protein electrophoresis,
observed by others to give the lowest level of detection (75, p 75–87. In Detrick B, Hamilton RG, Folds JD (ed), Man-
95, 96). See Table 3 for a summary of the key concepts in ual of Molecular and Clinical Laboratory Immunology, 7th ed.
clinical MS. ASM Press, Washington, DC.
17. Katzmann JA, Kyle RA. 2006. Immunochemical char-
acterization of immunoglobulins in serum, urine and ce-
REFERENCES rebrospinal fluid, p 88–100. In Detrick B, Hamilton RG,
1. Livesey JH, Ellis MJ, Evans MJ. 2008. Pre-analytical re- Folds JD (ed), Manual of Molecular and Clinical Laboratory
quirements. Clin Biochem Rev 29(Suppl 1):S11–S15. Immunology, 7th ed. ASM Press, Washington, DC.
2. Remaley AT, Hortin GL. 2006. Protein analysis for diag- 18. Lequin RM. 2005. Enzyme immunoassay (EIA)/enzyme-
nostic applications, p 7–21. In Detrick B, Hamilton RG, linked immunosorbent assay (ELISA). Clin Chem 51:
Folds JD (ed), Manual of Molecular and Clinical Laboratory 2415–2418.
Immunology, 7th ed. ASM Press, Washington, DC. 19. Rissin DM, Kan CW, Campbell TG, Howes SC, Fournier
3. Ahrens ET, Bulte JW. 2013. Tracking immune cells in DR, Song L, Piech T, Patel PP, Chang L, Rivnak AJ, Fer-
vivo using magnetic resonance imaging. Nat Rev Immunol rell EP, Randall JD, Provuncher GK, Walt DR, Duffy
13:755–763. DC. 2010. Single- molecule enzyme- linked immunosor-
4. Hawkins PN. 2002. Serum amyloid P component scintig- bent assay detects serum proteins at subfemtomolar con-
raphy for diagnosis and monitoring amyloidosis. Curr Opin centrations. Nat Biotechnol 28:595–599.
Nephrol Hypertens 11:649–655. 20. Earley MC, Vogt RF Jr, Shapiro HM, Mandy FF, Kellar
5. Sachchithanantham S, Wechalekar AD. 2013. Imaging in KL, Bellisario R, Pass KA, Marti GE, Stewart CC, Han-
systemic amyloidosis. Br Med Bull 107:41–56. non WH. 2002. Report from a workshop on multianalyte
6. Perugini E, Guidalotti PL, Salvi F, Cooke RM, Petti- microsphere assays. Cytometry 50:239–242.
nato C, Riva L, Leone O, Farsad M, Ciliberti P, Bacchi- 21. Hsu HY, Joos TO, Koga H. 2009. Multiplex microsphere-
Reggiani L, Fallani F, Branzi A, Rapezzi C. 2005. based flow cytometric platforms for protein analysis and
Noninvasive etiologic diagnosis of cardiac amyloidosis us- their application in clinical proteomics—from assays to
ing 99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid results. Electrophoresis 30:4008–4019.
scintigraphy. J Am Coll Cardiol 46:1076–1084. 22. Jun BH, Kang H, Lee YS, Jeong DH. 2012. Fluorescence-
7. Massi G, Fabiano A, Ragusa D, Auconi P. 1979. Char- based multiplex protein detection using optically encoded
acterization of alpha-1-antitrypsin by isoelectric focusing microbeads. Molecules 17:2474–2490.
on an ultrathin polyacrylamide gel layer. An economic 23. Kettman JR, Davies T, Chandler D, Oliver KG, Fulton
high-resolution system for determining PiM subtypes. Hum RJ. 1998. Classification and properties of 64 multiplexed
Genet 53:91–95. microsphere sets. Cytometry 33:234–243.
8. Mauch L, Lun A, O’Gorman MR, Harris JS, Schulze I, 24. Elshal MF, McCoy JP. 2006. Multiplex bead array assays:
Zychlinsky A, Fuchs T, Oelschlägel U, Brenner S, Kut- performance evaluation and comparison of sensitivity to
ter D, Rösen-Wolff A, Roesler J. 2007. Chronic granu- ELISA. Methods 38:317–323.
lomatous disease (CGD) and complete myeloperoxidase 25. Kingsmore SF. 2006. Multiplexed protein measurement:
deficiency both yield strongly reduced dihydrorhodamine technologies and applications of protein and antibody ar-
123 test signals but can be easily discerned in routine test- rays. Nat Rev Drug Discov 5:310–320.
ing for CGD. Clin Chem 53:890–896. 26. Ellington AA, Kullo IJ, Bailey KR, Klee GG. 2010.
9. Vowells SJ, Fleisher TA, Sekhsaria S, Alling DW, Ma- Antibody- based protein multiplex platforms: technical
guire TE, Malech HL. 1996. Genotype- dependent and operational challenges. Clin Chem 56:186–193.
27. Wilson DS, Nock S. 2003. Recent developments in 47. Ens W, Standing KG. 2005. Hybrid quadrupole/time-
protein microarray technology. Angew Chem Int Ed Engl of-flight mass spectrometers for analysis of biomolecules.
42:494–500. Methods Enzymol 402:49–78.
28. Hultschig C, Kreutzberger J, Seitz H, Konthur Z, Büs- 48. Brown RS, Gilfrich NL. 1992. Optimizing signal and
sow K, Lehrach H. 2006. Recent advances of protein mi- mass resolution for matrix-assisted laser desorption utiliz-
croarrays. Curr Opin Chem Biol 10:4–10. ing a linear time-of-flight mass spectrometer. Rapid Com-
29. Djoba Siawaya JF, Roberts T, Babb C, Black G, Golakai mun Mass Spectrom 6:697–701.
HJ, Stanley K, Bapela NB, Hoal E, Parida S, van Helden 49. Hu Q, Noll RJ, Li H, Makarov A, Hardman M, Graham
P, Walzl G. 2008. An evaluation of commercial fluorescent Cooks R. 2005. The Orbitrap: a new mass spectrometer. J
bead-based luminex cytokine assays. PloS One 3:e2535. Mass Spectrom 40:430–443.
doi:10.1371/journal.pone.0002535. 50. Ikonomou MG, Blades AT, Kebarle P. 1990. Investigations
30. Leng SX, McElhaney JE, Walston JD, Xie D, Fedarko of the electrospray interface for liquid- chromatography
NS, Kuchel GA. 2008. ELISA and multiplex technologies mass-spectrometry. Anal Chem 62:957–967.
for cytokine measurement in inflammation and aging re- 51. Covey T. 1996. Analytical characteristics of the elec-
search. J Gerontol A Biol Sci Med Sci 63:879–884. trospray ionization process, p 21–59. In Snyder AP (ed),
31. Toedter G, Hayden K, Wagner C, Brodmerkel C. 2008. Biochemical and Biotechnological Applications of Electrospray
Simultaneous detection of eight analytes in human serum Ionization Mass Spectrometry. American Chemical Society,
by two commercially available platforms for multiplex cy- Washington, DC.
tokine analysis. Clin Vaccine Immunol 15:42–48. 52. Jemal M, Ouyang Z, Powell ML. 2000. Direct-injection
32. Martinez JE, Clutterbuck EA, Li H, Romero-Steiner S, LC- MS-MS method for high- throughput simultaneous
Carlone GM. 2006. Evaluation of multiplex flow cytomet- quantitation of simvastatin and simvastatin acid in human
ric opsonophagocytic assays for determination of functional plasma. J Pharm Biomed Anal 23:323–340.
anticapsular antibodies to Streptococcus pneumoniae. Clin 53. Wang J, Aubry AF, Cornelius G, Caporuscio C, Sleczka
Vaccine Immunol 13:459–466. B, Ranasinghe A, Wang-Iverson D, Olah T, Jemal M.
33. Schulz KR, Danna EA, Krutzik PO, Nolan GP. 2012. 2010. Importance of mobile phase and injection solvent
Single-cell phospho- protein analysis by flow cytometry. selection during rapid method development and sample
Curr Protoc Immunol 8.17.11–18.17.20. analysis in drug discovery bioanalysis illustrated using con-
34. Vereb G, Nagy P, Szöllosi J. 2011. Flow cytometric venient multiplexed LC-MS/MS. Anal Methods 2:375–381.
FRET analysis of protein interaction. Methods Mol Biol 54. Desiderio DM, Katakuse I. 1983. Fast atom bombardment-
699:371–392. collision activated dissociation-linked field scanning mass
35. Bjornson ZB, Nolan GP, Fantl WJ. 2013. Single-cell mass spectrometry of the neuropeptide substance P. Anal Bio-
cytometry for analysis of immune system functional states. chem 129:425–429.
Curr Opin Immunol 25:484–494. 55. Desiderio DM, Katakuse I, Kai M. 1983. Measurement
36. Kricka LJ, Master SR. 2009. Quality control and protein of leucine enkephalin in caudate nucleus tissue with fast
microarrays. Clin Chem 55:1053–1055. atom bombardment-collision activated dissociation-linked
37. Fenn JB, Mann M, Meng CK, Wong SF, Whitehouse field scanning mass spectrometry. Biomed Mass Spectrom
CM. 1989. Electrospray ionization for mass spectrometry 10:426–429.
of large biomolecules. Science 246:64–71. 56. Bystrom CE, Sheng S, Clarke NJ. 2011. Narrow mass
38. Karas M, Hillenkamp F. 1988. Laser desorption ionization extraction of time-of-flight data for quantitative analysis
of proteins with molecular masses exceeding 10,000 dal- of proteins: determination of insulin-like growth factor-1.
tons. Anal Chem 60:2299–2301. Anal Chem 83:9005–9010.
39. Wilm M. 2011. Principles of electrospray ionization. 57. Chen Z, Caulfield MP, McPhaul MJ, Reitz RE, Tay-
Mol Cell Proteomics 10:M111.009407. doi:10.1074/mcp. lor SW, Clarke NJ. 2013. Quantitative insulin analysis
M111.009407. using liquid chromatography–tandem mass spectrome-
40. Mann M, Meng CK, Fenn JB. 1989. Interpreting try in a high-throughput clinical laboratory. Clin Chem
mass-spectra of multiply charged ions. Anal Chem 61: 59:1349–1356.
1702–1708. 58. Van Uytfanghe K, Rodríguez-Cabaleiro D, Stöckl D,
41. Kafka AP, Kleffmann T, Rades T, McDowell A. 2011. Thienpont LM. 2007. New liquid chromatography/elec-
The application of MALDI TOF MS in biopharmaceutical trospray ionisation tandem mass spectrometry measure-
research. Int J Pharm 417:70–82. ment procedure for quantitative analysis of human insulin
42. Norris JL, Caprioli RM. 2013. Analysis of tissue speci- in serum. Rapid Commun Mass Spectrom 21:819–821.
mens by matrix-assisted laser desorption/ionization imag- 59. Romanova EV, Dowd SE, Sweedler JV. 2013. Quantita-
ing mass spectrometry in biological and clinical research. tion of endogenous peptides using mass spectrometry based
Chem Rev 113:2309–2342. methods. Curr Opin Chem Biol 17:801–808.
43. Pirro V, Eberlin LS, Oliveri P, Cooks RG. 2012. Interac- 60. Zhang Y, Fonslow BR, Shan B, Baek MC, Yates JR III.
tive hyperspectral approach for exploring and interpreting 2013. Protein analysis by shotgun/bottom-up proteomics.
DESI-MS images of cancerous and normal tissue sections. Chem Rev 113:2343–2394.
Analyst 137:2374–2380. 61. Shevchenko A, Wilm M, Vorm O, Jensen ON, Podtelej
44. Watson JT, Sparkman OD. 2007. Introduction to Mass nikov AV, Neubauer G, Shevchenko A, Mortensen P,
Spectrometry: Instrumentation, Applications and Strategies for Mann M. 1996. A strategy for identifying gel-separated
Data Interpretation, 4th ed. John Wiley & Sons Ltd, Chich- proteins in sequence databases by MS alone. Biochem Soc
ester, United Kingdom. Trans 24:893–896.
45. Jonscher KR, Yates JR III. 1997. The quadrupole ion trap 62. Eng JK, McCormack AL, Yates JR. 1994. An approach to
mass spectrometer—a small solution to a big challenge. correlate tandem mass spectral data of peptides with amino
Anal Biochem 244:1–15. acid sequences in a protein database. J Am Soc Mass Spec-
46. Earley L, Anderson LC, Bai DL, Mullen C, Syka JE, trom 5:976–989.
English AM, Dunyach JJ, Stafford GC Jr, Shabanow- 63. Yates JR III, Eng JK, McCormack AL, Schieltz D. 1995.
itz J, Hunt DF, Compton PD. 2013. Front-end electron Method to correlate tandem mass spectra of modified pep-
transfer dissociation: a new ionization source. Anal Chem tides to amino acid sequences in the protein database. Anal
85:8385–8390. Chem 67:1426–1436.
64. Link AJ, Eng J, Schieltz DM, Carmack E, Mize GJ, Mor- Intact mass detection, interpretation, and visualization to
ris DR, Garvik BM, Yates JR III. 1999. Direct analysis of automate Top-Down proteomics on a large scale. Proteom-
protein complexes using mass spectrometry. Nat Biotechnol ics 10:3589–3597.
17:676–682. 79. Tran JC, Zamdborg L, Ahlf DR, Lee JE, Catherman AD,
65. Jones KA, Kim PD, Patel BB, Kelsen SG, Braverman A, Durbin KR, Tipton JD, Vellaichamy A, Kellie JF, Li M,
Swinton DJ, Gafken PR, Jones LA, Lane WS, Neveu JM, Wu C, Sweet SM, Early BP, Siuti N, LeDuc RD, Comp-
Leung HC, Shaffer SA, Leszyk JD, Stanley BA, Fox TE, ton PD, Thomas PM, Kelleher NL. 2011. Mapping intact
Stanley A, Hall MJ, Hampel H, South CD, de la Cha- protein isoforms in discovery mode using top-down pro-
pelle A, Burt RW, Jones DA, Kopelovich L, Yeung AT. teomics. Nature 480:254–258.
2013. Immunodepletion plasma proteomics by TripleTOF 80. Taylor GK, Kim YB, Forbes AJ, Meng FY, McCarthy R,
5600 and Orbitrap Elite/LTQ-Orbitrap Velos/Q Exactive Kelleher NL. 2003. Web and database software for identi-
mass spectrometers. J Proteome Res 12:4351–4365. fication of intact proteins using “top down” mass spectrom-
66. de Groot CJ, Güzel C, Steegers-Theunissen RP, de Maat etry. Anal Chem 75:4081–4086.
M, Derkx P, Roes EM, Heeren RM, Luider TM, Steegers 81. Kellie JF, Catherman AD, Durbin KR, Tran JC, Tipton
EA. 2007. Specific peptides identified by mass spectrom- JD, Norris JL, Witkowski CE II, Thomas PM, Kelleher
etry in placental tissue from pregnancies complicated by NL. 2012. Robust analysis of the yeast proteome under 50
early onset preeclampsia attained by laser capture dissec- kDa by molecular-mass-based fractionation and top-down
tion. Proteomics Clin Appl 1:325–335. mass spectrometry. Anal Chem 84:209–215.
67. Vrana JA, Gamez JD, Madden BJ, Theis JD, Bergen 82. Edwards RL, Griffiths P, Bunch J, Cooper HJ. 2012. Top-
HR 3rd, Dogan A. 2009. Classification of amyloidosis down proteomics and direct surface sampling of neonatal
by laser microdissection and mass spectrometry-based dried blood spots: diagnosis of unknown hemoglobin vari-
proteomic analysis in clinical biopsy specimens. Blood ants. J Am Soc Mass Spectrom 23:1921–1930.
114:4957–4959. 83. Théberge R, Infusini G, Tong W, McComb ME, Costello
68. Barr JR, Maggio VL, Patterson DG Jr, Cooper GR, CE. 2011. Top- down analysis of small plasma proteins
Henderson LO, Turner WE, Smith SJ, Hannon WH, using an LTQ-Orbitrap. Potential for mass spectrometry-
Needham LL, Sampson EJ. 1996. Isotope dilution– based clinical assays for transthyretin and hemoglobin. Int
mass spectrometric quantification of specific proteins: J Mass Spectrom 300:130–142.
model application with apolipoprotein A- I. Clin Chem 84. Barnidge DR1, Dasari S, Botz CM, Murray DH, Sny-
42:1676–1682. der MR, Katzmann JA, Dispenzieri A, Murray DL. 2014.
69. Barnidge DR, Dratz EA, Martin T, Bonilla LE, Moran Using mass spectrometry to monitor monoclonal immuno-
LB, Lindall A. 2003. Absolute quantification of the G globulins in patients with a monoclonal gammopathy. J.
protein-coupled receptor rhodopsin by LC/MS/MS using Proteome Res 13:1419–1427.
proteolysis product peptides and synthetic peptide stan- 85. Falick AM, Shackleton CH, Green BN, Witkowska HE.
dards. Anal Chem 75:445–451. 1990. Tandem mass spectrometry in the clinical analysis
70. Barnidge DR, Goodmanson MK, Klee GG, Muddiman of variant hemoglobins. Rapid Commun Mass Spectrom
DC. 2004. Absolute quantification of the model biomarker 4:396–400.
prostate-specific antigen in serum by LC-MS/MS using 86. Troxler H, Kleinert P, Schmugge M, Speer O. 2012. Ad-
protein cleavage and isotope dilution mass spectrometry. vances in hemoglobinopathy detection and identification.
J Proteome Res 3:644–652. Adv Clin Chem 57:1–28.
71. Bondar OP, Barnidge DR, Klee EW, Davis BJ, Klee GG. 87. Bergen HR III, Zeldenrust SR, Naylor S. 2003. An
2007. LC-MS/MS quantification of Zn-α2 glycoprotein: a on-line assay for clinical detection of amyloidogenic
potential serum biomarker for prostate cancer. Clin Chem transthyretin variants directly from serum. Amyloid
53:673–678. 10:190–197.
72. Kumar V, Barnidge DR, Chen LS, Twentyman JM, Cra- 88. Lacey JM, Bergen HR, Magera MJ, Naylor S, O’Brien JF.
dic KW, Grebe SK, Singh RJ. 2010. Quantification of 2001. Rapid determination of transferrin isoforms by im-
serum 1–84 parathyroid hormone in patients with hyper- munoaffinity liquid chromatography and electrospray mass
parathyroidism by immunocapture in situ digestion liquid spectrometry. Clin Chem 47:513–518.
chromatography–tandem mass spectrometry. Clin Chem 89. Kuprowski MC, Boys BL, Konermann L. 2007. Analy-
56:306–313. sis of protein mixtures by electrospray mass spectrometry:
73. Seegmiller JC, Barnidge DR, Burns BE, Larson TS, effects of conformation and desolvation behavior on the
Lieske JC, Kumar R. 2009. Quantification of urinary albu- signal intensities of hemoglobin subunits. J Am Soc Mass
min by using protein cleavage and LC-MS/MS. Clin Chem Spectrom 18:1279–1285.
55:1100–1107. 90. Anderson NL, Anderson NG, Haines LR, Hardie DB,
74. Craig R, Cortens JP, Beavis RC. 2005. The use of pro- Olafson RW, Pearson TW. 2004. Mass spectrometric
teotypic peptide libraries for protein identification. Rapid quantitation of peptides and proteins using stable isotope
Commun Mass Spectrom 19:1844–1850. standards and capture by antipeptide antibodies (SIS-
75. Bereman MS, MacLean B, Tomazela DM, Liebler DC, CAPA). J Proteome Res 3:235–244.
MacCoss MJ. 2012. The development of selected reaction 91. Anderson L, Hunter CL. 2006. Quantitative mass spec-
monitoring methods for targeted proteomics via empirical trometric multiple reaction monitoring assays for major
refinement. Proteomics 12:1134–1141. plasma proteins. Mol Cell Proteomics 5:573–588.
76. Anderson L, Hunter CL. 2006. Quantitative mass spec- 92. Taylor PJ. 2004. Matrix effects: the Achilles heel of
trometric multiple reaction monitoring assays for major quantitative high- performance liquid chromatography–
plasma proteins. Mol Cel Proteomics 5:573–588. electrospray–tandem mass spectrometry. Clin Biochem
77. Kilpatrick EL, Liao WL, Camara JE, Turko IV, Bunk 38:328–334.
DM. 2012. Expression and characterization of 15N-labeled 93. Gusev AI, Wilkinson WR, Proctor A, Hercules DM.
human C-reactive protein in Escherichia coli and Pichia pas- 1995. Improvement of signal reproducibility and ma-
toris for use in isotope-dilution mass spectrometry. Protein trix/comatrix effects in MALDI analysis. Anal Chem
Expr Purif 85:94–99. 67:1034–1041.
78. Durbin KR, Tran JC, Zamdborg L, Sweet SM, Cather- 94. Pascoe R, Foley JP, Gusev AI. 2001. Reduction in matrix-
man AD, Lee JE, Li M, Kellie JF, Kelleher NL. 2010. related signal suppression effects in electrospray ionization
mass spectrometry using on-line two-dimensional liquid 99. Zanella-Cleon I, Joly P, Becchi M, Francina A. 2009.
chromatography. Anal Chem 73:6014–6023. Phenotype determination of hemoglobinopathies by mass
95. Farrah T, Deutsch EW, Kreisberg R, Sun Z, Campbell spectrometry. Clin Biochem 42:1807–1817.
DS, Mendoza L, Kusebauch U, Brusniak MY, Hütten- 100. Chen Y, Snyder MR, Zhu Y, Tostrud LJ, Benson LM,
hain R, Schiess R, Selevsek N, Aebersold R, Moritz RL. Katzmann JA, Bergen HR III. 2011. Simultaneous phe-
2012. PASSEL: the PeptideAtlas SRMexperiment library. notyping and quantification of α-1-antitrypsin by liquid
Proteomics 12:1170–1175. chromatography–tandem mass spectrometry. Clin Chem
96. Reiter L, Rinner O, Picotti P, Hüttenhain R, Beck M, Brus- 57:1161–1168.
niak MY, Hengartner MO, Aebersold R. 2011. mProphet: 101. D’Souza A, Theis J, Quint P, Kyle R, Gertz M, Zelden-
automated data processing and statistical validation for rust S, Vrana J, Dogan A, Dispenzieri A. 2013. Explor-
large-scale SRM experiments. Nat Methods 8:430–435. ing the amyloid proteome in immunoglobulin- derived
97. Thevis M, Bredehöft M, Kohler M, Schänzer W. 2010. lymph node amyloidosis using laser microdissection/
Mass spectrometry- based analysis of IGF- 1 and hGH. tandem mass spectrometry. Am J Hematol 88:577–580.
Handb Exp Pharmacol 195:201–207. 102. Maleszewski JJ, Murray DL, Dispenzieri A, Grogan M,
98. Shi TJ, Su D, Liu T, Tang K, Camp DG II, Qian WJ, Pereira NL, Jenkins SM, Judge DP, Caturegli P, Vrana
Smith RD. 2012. Advancing the sensitivity of selected re- JA, Theis JD, Dogan A, Halushka MK. 2013. Relation-
action monitoring-based targeted quantitative proteomics. ship between monoclonal gammopathy and cardiac amy-
Proteomics 12:1074–1092. loid type. Cardiovasc Pathol 22:189–194.
Immunoglobulin
Methods
section B
SECTION EDITOR: DAVID F. KEREN
5 Introduction / 49 9 Immunochemical Characterization of

DAVID F. KEREN Immunoglobulins in Serum, Urine, and
6 Immunoglobulin Genes / 51 Cerebrospinal Fluid / 89
THOMAS J. KIPPS, EMANUELA M. GHIA, AND ELIZABETH SYKES AND YVONNE POSEY
LAURA Z. RASSENTI 10 Cryoglobulins, Cryofibrinogenemia, and
7 Immunoglobulin Quantification and Viscosity Pyroglobulins / 101
Measurement / 65 PETER D. GOREVIC AND DENNIS GALANAKIS
JEFFREY S. WARREN 11 Strategy for Detecting and Following
8 Clinical Indications and Applications of Serum Monoclonal Gammopathies / 112
and Urine Protein Electrophoresis / 74 JERRY A. KATZMANN AND DAVID F. KEREN
DAVID F. KEREN AND RICHARD L. HUMPHREY
Introduction
DAVID F. KEREN
5
The section on immunoglobulin methods covers the basic Serum and urine protein electrophoreses are still the
genetic background for immunoglobulin production, mea- principal assays for the detection of monoclonal gammop-
surement of immunoglobulins, identification of monoclo- athies and other immunoglobulin-based clinical abnormal-
nal protein products by serum protein electrophoresis and ities. In chapter 8, Keren and Humphrey review both the
immunofixation, and detection of oligoclonal bands in ce- technical details and the clinical applications of serum and
rebrospinal fluid. In addition, this section covers the char- urine protein electrophoresis. Because of the expanded use
acterization of cryoglobulins and cryofibrinogens and an of capillary electrophoresis since the previous edition, the
overall strategy for using all the techniques in this section to chapter discusses both the classic gel and newer capillary
detect, stratify risks of progression of, and monitor patients electrophoresis patterns along with problems and pitfalls
with monoclonal gammopathies. that may lead to false-positive or false-negative serum and
Since the 7th edition of the Manual of Molecular and urine findings. In addition, the chapter provides useful de-
Clinical Laboratory Immunology (2006), there have been tails for specimen processing, quality control, and quality
significant advances in the design and implementation of assurance (5). The chapter illustrates the ability of capillary
analytical tools to study immunoglobulins and interpret electrophoresis to detect polyclonal increases in IgG4 as
their results in a clinical context. The background for the part of IgG4-related systemic disease (6).
molecular basis of immunoglobulin complexity and antigen In cases where monoclonal gammopathies are detected
specificity has evolved rapidly. Kipps, Ghia, and Rassenti or suspected, Sykes and Posey (chapter 9) present a thor-
(chapter 6) present a cogent discussion of the immunoglob- ough review of their characterization in serum and urine by
ulin heavy-chain and light-chain complexes together with immunofixation and immunosubtraction. Their presenta-
a detailed explanation of the lack of uniform expression tion provides useful examples of both immunosubtraction
of each functional immunoglobulin heavy-chain variable (immunotyping) patterns, when capillary electrophoresis is
gene. This is accompanied by a comprehensive overview used for the characterization, and immunofixation patterns,
of the molecular basis for immunoglobulin gene rearrange- when gel techniques are employed. In addition, their chap-
ment and its relationship to the heterogeneity of the hu- ter presents illustrative examples of isoelectric focusing with
moral immune response. This is illustrated using a detailed immunofixation in the evaluation of cerebrospinal fluid for
figure of the light-and heavy-chain gene complexes. They the presence of oligoclonal bands.
also describe how the detection of immunoglobulin gene re- Gorevic and Galanakis (chapter 10) discuss the state-
arrangements using PCR can aid in detecting clonal popula- of-the-art methodology in the detection and measurement
tions of B cells and how these highly sensitive methods may of cryoglobulins. They also provide a thorough review of
be deployed to investigate minimal residual disease. cryofibrinogenemia and pyroglobulins. This comprehensive
Warren (chapter 7) reviews how IgG, IgA, IgM, immu- chapter provides recent information on the importance of
noglobulin subclasses, serum free light chains, and even hepatitis C virus in both type II and type III cryoglobulins
IgD and IgE are measured by nephelometric or immunotur- (7). Many practical suggestions are made about the impor-
bidimetric assay systems. Former gel diffusion and Laurell tance of proper handling in the identification and charac-
rocket techniques have fallen out of use. He makes note of terization of these temperature-sensitive specimens (8). The
their problem with the antigen excess effect and occasional section on cryofibrinogens has an up-to-date discussion of
nonlinearity in the serum free-light-chain test (1, 2). He the effect of heparin and citrate anticoagulation on cryofi-
highlights the recently FDA-approved heavy/light-chain as- brinogen detection. Lastly, the authors discuss the peculiar
says. Using a nephelometric method similar to that used in phenomenon of pyroglobulins, which lack the clinical man-
quantifying serum free light chains, one may now separately ifestations of cryoglobulins but can present confounding
measure IgG-kappa (IgGκ) from IgG-lambda (IgGλ), IgAκ laboratory findings in heat-based assays such as those used
from IgAλ, and IgMκ from IgMλ (3, 4). to inactivate complement.
This analysis may be especially useful in following the To coordinate the use of the above-mentioned tech-
beta-region M proteins that are cloaked by transferrin and niques in the context of current clinical guidelines,
C3 bands. Katzmann and Keren (chapter 11) first provide a handy
doi:10.1128/9781555818722.ch5
49
50 ■ IMMUNOGLOBULIN METHODS
categorization of plasma cell proliferative disorders. This 2. Daval S, Tridon A, Mazeron N, Ristori JM, Evrard B.
is followed by a diagnostic testing strategy that uses an ef- 2007. Risk of antigen excess in serum free light chain mea-
ficient electrophoretic and nephelometric assay analysis. surements. Clin Chem 53:1985–1986.
They recommend a basic diagnostic screening panel con- 3. Bradwell AR, Harding SJ, Fourier NJ, Wallis GL,
sisting of serum protein electrophoresis, immunofixation, Drayson MT, Carr-Smith HD, Mead GP. 2009. Assess-
and serum free-light-chain testing. However, the chapter ment of monoclonal gammopathies by nephelometric
provides caveats and illustrations of special cases to prevent measurement of individual immunoglobulin kappa/lambda
missing low tumor burden conditions such as light-chain- ratios. Clin Chem 55:1646–1655.
type amyloidosis and light-chain deposition disease. They 4. Katzmann JA, Rajkumar SV. 2013. A window into immu-
noglobulin quantitation and plasma cell disease: antigen
show how these measurements are useful in stratifying the epitopes defined by the junction of immunoglobulin heavy
risk of progression, in survival estimates, and in monitor- and light chains. Leukemia 27:1–2.
ing of monoclonal gammopathies. The authors discuss the 5. Keren DF. 2012. Protein Electrophoresis in Clinical Diagnosis.
use of immunosubtraction and the new heavy/light assays American Society for Clinical Pathology Press, Chicago, IL.
to monitor IgA and IgM monoclonal gammopathies that 6. Jacobs JFM, van der Molen RG, Keren DF. 2014. Rela-
may be obscured by comigrating beta-region proteins. tively restricted migration of polyclonal IgG4 may mimic
a monoclonal gammopathy in IgG4-related disease. Am J
Clin Pathol 142:76–81.
7. Russi S, Sansonno D, Mariggio MA, Vinella A, Pavone F,
REFERENCES Lauletta G, Sansonno S, Dammacco F. 2014. Assessment
1. Tate JR, Mollee P, Dimeski G, Carter AC, Gill D. 2007. of hepatitis core protein in HCV-related mixed cryoglobu-
Analytical performance of serum free light- chain assay linemia. Arthritis Res Ther 18:16.
during monitoring of patients with monoclonal light-chain 8. Warren JS. 2013. Clinically unsuspected cryoglobuline
diseases. Clin Chim Acta 376:30–36. mia. Am J Clin Pathol 139:352–359.
Immunoglobulin Genes
THOMAS J. KIPPS, EMANUELA M. GHIA, AND LAURA Z. RASSENTI
6
IMMUNOGLOBULIN MOLECULES an intact immunoglobulin molecule. Soon after synthesis, the
antibody light-chain constant region associates with the na-
Introduction scent immunoglobulin heavy chain, releasing the latter from
Immunoglobulins are a heterogeneous group of glycopro- the immunoglobulin-binding protein. In the absence of anti-
teins produced by B lymphocytes and plasma cells. A single body light chain, the immunoglobulin-binding protein binds
person can synthesize 10 million to 100 million different the first constant-region domain of the newly synthesized
immunoglobulin molecules, each having distinct antigen- heavy chain, thereby retaining the heavy-chain polypeptide
binding specificities. This great diversity in the so-called in the cell’s endoplasmic reticulum (9).
humoral immune system allows us to generate antibodies
specific for a variety of substances, including synthetic mol- Heavy-Chain Isotypes
ecules not naturally present in our environment. Despite Five major classes of immunoglobulin molecules—
the diversity in the specificities of antibody molecules, the immunoglobulin G (IgG), IgA, IgM, IgD, and IgE—
binding of an antibody to an antigen initiates a limited correspond to the five classes of heavy-chain isotypes (γ, α,
series of biologically important effector functions, such as μ, δ, and ε). The immunoglobulin molecule of each isotype
complement activation and/or adherence of the immune can contain either a κ or a λ light chain but not both. The
complex to receptors on leukocytes (1). Resolution of the physical properties of each of these classes of immunoglobu-
immunoglobulin structure has revealed how these mole- lin molecules are summarized in Table 1.
cules can have such great diversity in antigen-binding activ-
ities while maintaining conserved effector functions, such as IgG
complement activation. IgG is the most abundant of immunoglobulins found in
adult plasma, accounting for approximately 80% of the to-
Basic Immunoglobulin Structure tal immunoglobulin. IgG is the predominant antibody pro-
The basic unit of the immunoglobulin molecule is composed duced during a secondary immune response. IgG molecules
of two identical heavy chains and two identical light chains. can penetrate extravascular spaces and cross the placental
These four polypeptides are held together by disulfide bonds barrier to provide immunity to the fetus. These molecules
and noncovalent interactions (2–5). The amino-terminal have a four-chain 150-kDa immunoglobulin structure with
domains (110 to 120 amino acids) of the heavy and light a hinge region that can be attacked by proteolytic enzymes
chains are designated the variable regions, because their such as papain and pepsin, allowing for separation of the
primary structures vary markedly among different immu- antigen-binding fragment(s), Fab or F(ab)2, from the crys-
noglobulin molecules (6). The carboxy-terminal domains, tallizable or constant fragment (Fc) of the antibody mol-
however, are referred to as constant regions, because their ecule. Receptors for the Fc (FcR) allow effector cells to
primary structures are the same among immunoglobulins of recognize target cells coated with a specific antibody (10).
the same class or subclass. The amino acids in the light-and There are four subclasses of IgG: IgG1, IgG2, IgG3, and
heavy-chain variable regions interact to form an antigen- IgG4. Each subclass has a particular heavy-chain constant
binding site (2, 7). Each four-chain immunoglobulin basic region and has different effector functions (8). The most
unit has two identical binding sites. Stability for the im- abundant class is IgG1, which accounts for approximately
munoglobulin molecule is provided by the constant-region 65% of the total IgG in the plasma. Of the IgG subclasses,
domains of the heavy and light chains. The specific effector IgG1 binds best to FcRI (CD64) and FcRII (CD32), with
functions of the different immunoglobulin classes are me- affinities (Kd) of 10−8 M and 5 × 10−7 M, respectively. IgG1
diated by the heavy-chain constant regions (Table 1) (8). and IgG3 bind equally well to FcRIII (CD16), with a Kd of
There are two classes of immunoglobulin light chains, the 2 × 10−6 M. FcRIII is the FcR expressed by natural killer
κ and λ light chains, which differ in the amino acid sequences cells (NK cells, or K cells) that mediate antibody-dependent
of the constant-region domains. The ratio of κ to λ chains in cell-mediated cytotoxicity. Proteins of the IgG4 or IgG2
adult plasma is 2:1. The light-chain constant region’s main subclass bind poorly to FcRI (CD64) or FcRII (CD32) and
purpose may be to allow for proper assembly and release of do not bind to FcRIII (CD16) at all. The average half-life
doi:10.1128/9781555818722.ch6
51
TABLE 1 Physical properties of human immunoglobulins

Heavy- Antigen-
Heavy-chain No. of heavy- Secretory Molecular Concn (mg/ % of total
chain class binding
subclasses chain domains form(s) mass (Da) ml) in serum immunoglobulins
(isotype) valency
IgG (γ) γ1, γ2, γ3, γ4 4 Monomer 150,000 2 1.8–16 80
IgA (α) α1, α2 4 Monomer, 160,000 2 (monomer) 1.4–4.0 13
dimer (monomer)
400,000 4 (secretory
(secretory protein)
protein)
IgM (μ) 5 Pentamer 900,000 10 0.5–2.0 6
IgD (δ) 4 Monomer 180,000 2 1.0–0.4 1
IgE (ε) 5 Monomer 190,000 2 .17–450 ng/ml 0.002
of circulating IgG molecules is about 21 days. The response The IgM monomer represents the ligand-binding part of the
to a given antigen can result in a skewed IgG subclass re- receptor. The component that is responsible for signal trans-
sponse, and this is frequently a source of investigation as duction consists of two glycoproteins, CD79a and CD79b.
to correlates of protection or for the design of vaccines. The cytoplasmic domains of CD79a and CD79b contain
Specific subclasses can be associated with individual disease tyrosine motifs responsible for transduction of signal from
processes. For example, in patients with pemphigus vulgaris, the receptor (19).
a mucocutaneous blistering disease, IgG4 antibodies to des-
moglein 3 are pathogenic (11, 12).
IgD
The IgD molecules constitute only 1% of the plasma im-
IgA munoglobulins, and they are expressed on B cells with IgM.
The IgA molecules constitute 13% of the total plasma im- These immunoglobulins do not cross the placenta and do
munoglobulins. There are two major classes of IgA mole- not easily penetrate extravascular spaces. However, IgD
cules, designated IgA1 and IgA2, with IgA1 being the more molecules are found in relatively high concentrations in um-
abundant (85% of total IgA in plasma). The half-life of IgA bilical cord blood. IgD molecules are thought to function as
molecules is about 6 days. B-cell membrane receptors for antigens and may help in the
IgA antibodies are synthesized during a secondary im- recruitment of B cells for specific antigen-driven responses
mune response and contribute to mucosal immunity (13– (20). Similar to IgM, membrane-bound IgD is associated
15). IgA antibodies are the primary antibodies in saliva, with CD79a and Cd79b for signaling. IgD is expressed on
tears, breast milk, and the fluids of the gastrointestinal, the membranes of B cells when they leave the marrow and
respiratory, and urinary tracts (16). These secreted immu- populate secondary lymphoid organs. Most IgD-positive B
noglobulins consist of an IgA dimer bound to the joining cells also coexpress IgM, and both participate in B-cell re-
(J)-chain polypeptide and a secretory protein with a molec- ceptor signaling through CD79a and CD79b. It has been
ular mass of 70 to 80 kDa. The J chain is required for proper proposed that membrane-bound IgD regulates B-cell fate
hepatic transport of IgA (17). The secretory component is at specific developmental stages through changes in acti-
actually part of an FcR for dimeric IgA that is synthesized vation status (21). B-cell stimulation and activation can
not by B cells but rather by epithelial cells of organs such result from the binding of specific bacterial proteins to the
as the intestine. This protein facilitates the transport of the constant region of IgD. For example, circulating IgD can
IgA protein across the epithelial cell and may protect the se- react with IgD-binding protein of Moraxella catarrhalis, in-
creted IgA molecule from proteolytic digestion by enzymes dependently of the variable regions of the antibody (22).
in the intestinal lumen. Since these molecules do not cross
the placenta barrier and do not easily bind to cell surfaces,
their main role may be to prevent foreign substances from IgE
binding to mucosal surfaces and entering the blood. Finally, IgE has been called a reaginic antibody to denote its as-
it has been proposed that secretory IgA might also act as a sociation with immediate hypersensitivity. IgE antibodies
potentiator of the immune response in intestinal tissue by constitute a very small percentage of the total plasma im-
means of uptake of antigen to dendritic cells (18). munoglobulins (0.002%). Although four human IgE iso-
forms can be produced by alternative splicing of the epsilon
IgM primary transcript (23), the isoforms appear to have similar
The IgM immunoglobulins comprise about 6% of the immu- functions. Plasma IgE levels may increase (5 to 20 times the
noglobulins in adult plasma. These molecules have very high baseline) in parasitic infections and children with atopic
molecular weights (thus, they are called macroglobulins), diseases. The Fc portion of the IgE molecule can bind with
and they are formed by the linking of five identical immu- high affinity to receptors on the surfaces of basophils and
noglobulin units by disulfide bonds and a J chain. IgM is the mast cells. The cross-linking of IgE antibody by an aller-
predominant class found during a primary immune response. gen can induce the release of vasoactive amines, proteases,
The IgM molecules do not cross the placenta and do not lipid-derived inflammatory mediators, and cytokines, such
enter into extravascular spaces; however, they fix comple- as tumor necrosis factor alpha, gamma interferon, or inter-
ment more efficiently than the monomeric IgG molecules. leukins 1, 3, 4, 5, and 6. Studies indicate that the microen-
The half-life of IgM molecules in plasma is approximately 6 vironment of mucosal tissues in allergic disease favors class
days. In addition, monomeric IgM is the main component switching to IgE (24). Anti-IgE antibodies have been devel-
of an antigen receptor expressed on B cells (B-cell receptor). oped as therapy for allergy and asthma (25).
6. Immunoglobulin Genes ■ 53
IMMUNOGLOBULIN GENE COMPLEXES locus IGHV3-30. The second type of allelic variation results
from duplications, insertions, and/or deletions of whole seg-
Immunoglobulin Heavy-Chain Gene Complex ments of IGHV genes within the immunoglobulin heavy-
chain gene complex. Duplication of an IGHV gene(s) results
Immunoglobulin Heavy-Chain Constant-Region in some haplotypes having identical IGHV genes belonging
Exons to distinct loci, each possibly differing from their respective
The heavy- chain gene complex is located at band q32 alleles by one or more nucleotide base substitutions. For
on the long arm of chromosome 14 (26). This complex example, there may be an insertion in and about the locus
is composed of 38 to 46 functional heavy-chain variable- for IGHV3-30, in part consisting of another copy of this
region genes (IGHV genes), over 80 nonfunctional IGHV gene. As a result, alleles of IGHV3-30, e.g., IGHV3-30*18
pseudogenes, 23 functional diversity (IGHD) segments, 6 (1.9III) and IGHV3-30*1 (hv3005), also may be alleles of
functional IGHJ minigenes, and exons encoding the con- locus IGHV3-30.5 in haplotypes containing this insertion.
stant regions for each of the immunoglobulin heavy-chain On the other hand, some haplotypes are missing gene loci
isotypes (27–29). The order (5′→3′) of the genes encod- altogether. For example, allele frequencies for IGHV3-30*1
ing each of the immunoglobulin heavy-chain isotypes is Cμ (hv3005) and IGHV3-30*18 (1.9III) in the Caucasian pop-
(M), Cδ (D), Cγ3 (G3), Cγ1 (G1), Cε1 (EP1, a nonfunc- ulation are 0.19 and 0.72, respectively (36). An additional
tional pseudogene), Cα1 (A1), Cγ (GP, an open reading haplotype(s) with an allele frequency of 0.08 lacks either
frame lacking the switch region), Cγ2 (G2), Cγ4 (G4), Cε IGHV3-30*18 (1.9III) or IGHV3-30*1 (hv3005) and thus
(E), and Cα2 (A2) (Fig. 1). The genes encoding the heavy- apparently is a blank haplotype for this locus. Genetic dis-
chain constant regions and each associated intronic switch equilibrium also is noted for certain groups of IGHV genes
region are as depicted in Fig. 1. These genes are labeled, and on a given haplotype. For example, IGHV3-30-3*01 (56p1
pseudogenes are also indicated in the figure. or 3d216) is an insertion or deletion element that has been
observed only with haplotypes carrying one or two copies of
Immunoglobulin Heavy-Chain Variable-Region IGHV3-30*18 (1.9III).
Genes The relative expression level of each functional IGHV
The IGHV gene segments map within a region of approxi- gene is not uniform. Certain IGHV genes, e.g., IGHV3-23,
mately 1,100 kb that is telomeric to the IGHJ and constant- IGHV4- 34, or IGHV1- 69, are overexpressed relative to
region genes (Fig. 1). Each IGHV gene can be assigned to their physical representation among other IGHV genes (36,
one of seven IGHV gene subgroups. Each subgroup com- 37). Each of nine IGHV3 genes (IGHV3-23, IGHV3-30,
prises IGHV genes, which share more than 80% nucleic IGHV3-7, IGHV3-33, IGHV3-21, IGHV3-48, IGHV3-11,
acid sequence homology. Genes of the IGHV1, IGHV5, and IGHV3-15, and IGHV3-30-3) accounts for 5 to 19% of
IGHV7 subgroups have similarities in the primary structure, the IGHV3 gene rearrangements, whereas the remaining
suggesting a common ancestral origin in evolution, whereas functional IGHV3 genes contribute to less than 4% of the
the IGHV2, IGHV4, and IGHV6 families share similarities rearrangements. Some of the IGHV genes that encode a
that allow them to be classified into a different clan (30, disproportionate share of the immunoglobulin expressed
31). The IGHV3 genes constitute their own discrete clan. by normal adults also are polymorphic. For example, three
The IGHV genes of each subgroup, except IGHV6, are deletion/insertion polymorphisms (Del I to Del III) with
interspersed throughout the immunoglobulin heavy-chain deletion allele frequencies ranging from 0.1 to 0.3 were
locus. By convention, the loci encoding each of the vari- identified. Del I is a polymorphism affecting three IGHV
ous IGHV genes are assigned a number corresponding to genes (IGHV1-8, IGH3-9, and IGHV2-10), of which two
the IGHV gene subgroup followed by a hyphen and then are functional. Greater than 10% of individuals in the hu-
the rank order distance from the heavy-chain D segments man population may not have these gene segments in their
on chromosome 14 (Fig. 1). The immunoglobulin IGHV6 genome, and ~44% may have only one copy of these gene
subgroup has only one functional IGHV gene. Since this segments (38). In Del II, a 50-kb insertion containing five
is the first IGHV gene telomeric to the D segments, this IGHV genes (IGHV3-30-5, IGHV4-30-4, IGHV3-30-3,
gene is called IGHV6-1 (Fig. 1). There are an additional IGHV4-30-2, and IGHV4-30-1) was identified; this inser-
50 loci that have been identified as functional IGHV genes tion has been observed in 45% of Caucasoids (39). Del III
(32) (Fig. 1). The largest subgroup is IGHV3, with 25 func- spans ~21 to 53 kb, involving four contiguous genes, among
tional genes. The next largest are the IGHV1 and IGHV4 which only the functional gene IGHV4-39 has an identi-
subgroups, each with 11 functional genes. The IGHV2, fied null allele (40, 41). In addition, some persons have a
IGHV5, and IGHV7 subgroups each have one to four func- deletion of the IGHV gene IGHV4-31, which otherwise
tional genes. Interspersed among the functional IGHV may encode a significant proportion of the heavy-chain rep-
genes are several nonfunctional pseudogenes (Fig. 1). ertoire (36). The International Immunogenetics database
The extent of identified genetic polymorphism varies be- (http://www.imgt.org/) provides an Internet listing of IGHV
tween the different IGHV gene loci. Some immunoglobulin gene maps and alleles (42).
IGHV gene loci, e.g., IGHV6-1, IGHV5-51, or IGHV4-34,
are highly conserved (33, 34). Indeed, the single-copy im-
munoglobulin IGHV6 gene, IGHV6-1, is conserved even Immunoglobulin Light-Chain Gene Complexes
among higher primates (35). Other loci have been used to
identify genetic polymorphic variations. These allelic varia- κ Light-Chain Complex
tions are of two different types. The first type of genetic poly- The κ light-chain gene complex is contained within band
morphism is the classic form, in which there are two or more p11.2 on the short arm of chromosome 2 (Fig. 2). This gene
alleles at a single locus, each differing from one another in complex consists of approximately 34 to 38 (4 genes may
one or more nucleotide bases. For example, IGHV3-30*18 be functional or pseudogenes depending on the alleles)
(1.9III) and IGHV3-30*1 (hv3005) differ from each other functional kappa light-chain variable-region genes (IGKV
at several nucleotide bases and share only 98.8% overall ho- genes), 5 functional IGKJ genes, and 1 constant-region gene
mology (98.3% coding sequence homology) but are alleles of (Fig. 2) (43, 44).
FIGURE 1 Immunoglobulin heavy-chain gene complex. The heavy-chain genes encoding the constant regions are represented by blue
boxes. Switch regions are represented by a filled circle upstream of the IGHC genes. Enhancers are represented by light blue circles. Each
IGHV, IGHD, and IGHJ gene is labeled on the right of each symbol. Functional IGHV, IGHD, and IGHJ genes are represented by green
boxes, blue lines, and yellow lines, respectively. IGHV, IGJH, and IGHC pseudogenes are represented by red boxes, orange boxes, and blue
open boxes, respectively. IGHV and IGHC open reading frames are represented by yellow boxes and blue dashed boxes, respectively. Un-
related pseudogenes are represented by purple open boxes. Colors are according to the international ImMunoGeneTics information system
(IMGT) color menu for genes. Reproduced with the kind authorization of Marie-Paule Lefranc (IMGT [http://www.imgt.org]).
The IGKV genes in the κ light-chain gene complex The p region contains 40 IGKV gene segments, while the d
are found in two regions: an IGKC-proximal region, desig- region contains 36 gene segments (Fig. 2). The d region ap-
nated the p region, and an IGKC-distal region, designated parently arose through duplication of a large portion of the
the d region. Each region spans approximately 500 kb. p region. Consequently, there are 33 pairs of IGKV genes
Approximately 800 kb separates the two regions (Fig. 2). that share 95 to 100% nucleic acid sequence homology,
55
FIGURE 2 Immunoglobulin light-chain gene complexes. (Left) κ light-chain gene complex on chro-
mosome 2p11-12. The blue box represents the functional IGLC gene. The IGKJ gene segments are indi-
cated by yellow lines labeled “J1 to 5.” The κ light-chain enhancers are represented by light blue circles.
IGKV functional genes, pseudogenes, and open reading frame are indicated by green, red, and yellow
boxes, respectively. The IGKV genes of the p region are designated by a number for the subgroup, followed
by a hyphen and a number for the localization from 3′ to 5′ in the locus. The IGKV genes of the d region
are designated by the same numbers as the corresponding genes in the p region, with the letter D added.
Arrows show the IGKV genes whose orientation is opposite to that of the IGKJ gene segments. (Right) λ
light-chain gene complex on chromosome 22q11.2. The blue boxes represent functional IGLJ and IGLC
gene segments, whereas blue open boxes represent IGLJ and IGLC pseudogenes. IGLV functional genes,
pseudogenes, and open reading frame are indicated by green, red, and yellow boxes, respectively. IGLV
pseudogenes that could not be assigned to subgroups with functional genes are represented by red boxes
and designated by a roman number in parentheses, corresponding to the clans, followed by a dash and a
number for the localization from 3′ to 5′ in the locus. The IGLV genes are organized into three clusters,
designated A, B, and C, which are indicated to the left of each cluster. Unrelated pseudogenes are repre-
sented by purple open boxes. The λ light-chain enhancer is represented by a light blue circle. Colors are
according to the IMGT color menu for genes. Reproduced with the kind authorization of Marie-Paule
Lefranc (IMGT, the international ImMunoGeneTics information system [http://www.imgt.org]).
accounting for 66 of the 76 IGKV genes in the κ light- (3r, DPL23) through IGLV3-27 (V2-19) in Fig. 2. The next
chain complex (45, 46). cluster, cluster B, contains 12 functional IGLV genes of the
The IGKV genes in the κ light-chain gene complex can IGLV1, IGLV5, IGLV7, and IGLV9 gene subgroups, desig-
be categorized into three main subgroups (1 to 3) and several nated IGLV1-36 (1a, DPL1) through IGLV5-52 (5b). The
smaller subgroups (4, 5, 6, and 7), based on nucleotide se- third cluster, cluster C, contains five functional IGLV genes
quence homology (47, 48). The largest subgroup is IGKV1, of the IGLV4, IGLV6, IGLV8, IGLV10, and IGLV11 gene
with 20 functional genes (Fig. 2). The next largest subgroups subgroups, designated VpreB through IGLV4-69 (4b) (Fig.
are IGKV2, with 10 functional genes, and IGKV3, with 7 2). As with the other immunoglobulin complexes, there are
functional genes. There are three open reading frame genes multiple nonfunctional pseudogenes interspersed between
in the IGKV6 subgroup and one functional gene in each of these functional IGLV genes in all three clusters.
the IGKV4 and IGKV5 subgroups. The IGKV7 subgroup As noted for the relative expression of individual IGHV
has only one nonfunctional pseudogene. and IGKV genes, the expression of individual IGLV genes
Similar to the IGHV locus, there are several prominent appears to be nonrandom. IGLV genes of the IGLV1 and
alleles identified in the IGKV gene locus. The IGKV genes IGLV3 subgroups are used most frequently. These subgroups
coding for segments IGKV1-13 (L4) and IGKV3D-15 (L16) respectively encode approximately 44 or 40% of the λ light-
each have several alleles, some having stop codons (49, 50). chain immunoglobulins in normal adult sera (61). This pro-
Moreover, the IGKV2D-29 (A2) gene also is polymorphic, portionate representation apparently is not observed in B-cell
with some alleles having defective promoters that preclude plasmacytic disorders. Although the IGLV2 subgroup was
their expression into protein. Inheritance of defective IGK- identified on 3% of the λ light-chain immunoglobulins in
V2D-29 (A2) alleles has been associated with an increased normal adult sera, it accounted for 40% of the λ Bence Jones
risk for serious infection with type b Haemophilus influenzae, proteins and 60% of the λ macroglobulins from patients with
suggesting that the polymorphic variations in the germ line Waldenström macroglobulinemia in one survey (61).
repertoire can influence the susceptibility to infectious dis-
ease (51).
As with the immunoglobulin IGHV genes, the expres- IMMUNOGLOBULIN GENE
sion of IGKV genes is not uniform. Eleven of the approx- REARRANGEMENT
imately 34 to 38 known functional IGKV genes encode
most of the κ light-chain variable regions expressed in the Immunoglobulin Gene Rearrangement and
normal adult repertoire (49). Moreover, of the 44 genes Expression in Ontogeny
that are potentially functional, only 28 commonly encode As B cells develop, they generally first rearrange their im-
κ light-chain variable regions (Fig. 2) (49, 50). This raises munoglobulin heavy-chain genes (62). One or more IGHD
the possibility that some of the potentially functional IGKV segments rearrange to become juxtaposed with a single
genes have defects that preclude their ability to undergo IGHJ element. This generates an IGHD- IGHJ complex
light-chain gene rearrangement or to be expressed into pro- that then may rearrange with any one of approximately 38
tein. Alternatively, these genes may have an extremely low to 46 functional IGHV genes.
expression frequency. Subsequently, gene rearrangements occur within the
light-chain complexes. One of the 40 functional IGKV
λ Light-Chain Gene Complex genes rearranges with any one of five IGKJ segments.
The λ light-chain gene complex is located at band q11.2 on Should these gene rearrangements fail to generate a func-
the long arm of chromosome 22. These λ constant-region tional IGKV-IGKJ exon, the Kde generally rearranges to a
genes are telomeric to the λ variable-region genes (IGLV). site in or immediately downstream of the IGKV-IGKJ exon,
Originally, the isotypes that they encoded were designated thus deleting the κ light-chain constant-region exon (63).
Mcg+, Ke− Oz−, Ke− Oz+, and Ke+ Oz−, depending on their Many of the IGKV genes in the p region are in the opposite
reactivity with Mcg, Kern, and Oz antisera that were raised orientation of the IGKJ segments, thus requiring that the
against λ Bence Jones proteins of patients with multiple V exons in this region undergo inversion during immuno-
myeloma (52). These isotypes are now designated IGLC1, globulin gene rearrangement. The processes of Ig gene lo-
IGLC2, IGLC3, and IGLC7, respectively. In total, there are cus contraction and looping during V(D)J recombination
7 to 11 IGLC genes telomeric to the IGLV genes, depending are essential for creating a diverse antibody repertoire.
on the haplotype (53). Each IGLC gene is associated with Recently, a 650-bp sequence corresponding to DNase I hy-
its own IGLJ segment. The most prevalent haplotype con- persensitive sites HS1 and HS2 within the mouse IGKV-
tains four functional IGLC genes (IGLC1, IGLC2, IGLC3, IGKJ intervening region was described. This sequence binds
and IGLC7, encoding the Mcg, Ke− Oz−, Ke− Oz+, and Ke+ CCCTC-binding factor and specifies locus contraction and
Oz− isotypes, respectively) and three pseudogenes (IGLC4, long-range IGKV gene usage. Its deletion results in a 7-fold
IGLC5, and IGLC6) (Fig. 2) (54, 55). increase in proximal IGKV gene usage along with an ~50%
There are approximately 29 to 33 functional IGLV genes reduction in overall locus contraction (64). Subsequent to
and over 40 nonfunctional IGLV pseudogenes that are ar- a failed κ light-chain gene rearrangement, one of the func-
ranged into 10 subgroups (56–58). Each subgroup comprises tional IGLV genes can rearrange with any one of the four
IGLV genes sharing more than 75% nucleotide sequence or five functional IGLJ-IGLC exons to generate a gene that
homology (57, 59) (Fig. 2). Like that of the κ light-chain can encode a λ light chain (55).
locus, the sequence organization suggests that large DNA The term progenitor B cells (or pro-B cells) is reserved
duplications contributed to the generation of the germ line for precursor B cells that have only rearranged IGHD and
repertoire of IGLV gene segments (56). The IGLV genes are IGHJ elements, whereas precursor B cells that have com-
clustered into three large DNA segments located within 860 pleted immunoglobulin heavy-chain gene rearrangement
kb of the IGLJ and IGLC genes (56, 58, 60). The cluster and have a functional VH-D-JH complex are referred to as
most proximal to the IGLJ-IGLC exons, designated cluster pre-B cells. The immunoglobulin light-chain loci of both
A, comprises 16 functional IGLV genes, mostly belonging to pro-B cells and pre-B cells are generally in the germ line
the IGLV2 and IGLV3 gene subgroups, designated IGLV3-1 configuration.
Nevertheless, a small amount of immunoglobulin μ chain Although each spacer varies in sequence, the length of each
in association with “surrogate” light chains is expressed by spacer is conserved and corresponds to one or two turns of
pre-B cells. The surrogate light chain is made up of two the DNA double helix. Each spacer serves to bring the hep-
noncovalently associated proteins called lambda-5 (λ5) and tamer and nonamer sequences to the same side of the DNA
VpreB, which together form a molecule having structural helix, where they both can be bound by a protein complex
homology with conventional light chains (65). Thus, in the that catalyzes recombination.
surrogate light chain, λ5 replaces a light-chain constant re- This process of somatic DNA recombination is initiated
gion and VpreB the variable part although bearing an extra when the recombination activating gene (RAG) endonucle-
N-terminal protein sequence. Both are located on chromo- ase introduces DNA double-strand breaks (DSBs) at the bor-
some 22: the λ5 gene is telomeric to the true light-chain der of two recombining gene segments and their flanking RSSs
locus, and the VpreB gene is located within the cluster of (78–80). DNA cleavage by RAG leads to four broken DNA
IGLV genes defined by breakpoints of chromosomal trans- ends that are repaired and joined through a process called
locations found in a few leukemias and lymphomas (66). nonhomologous DNA end joining (NHEJ) to form coding
During the early stages of B-cell development, VpreB and λ5 and signal joints (81, 82). Occasionally these DSBs can be
are covalently associated with the μ heavy chains to form a repaired aberrantly, leading to the formation of chromosomal
primitive immunoglobulin receptor referred to as pre-B-cell lesions such as translocations, deletions, or inversions (83,
receptor (pre-BCR). Pre-B cells may be expressed on the 84). If the breakpoints of these chromosomal lesions lie near
surface membrane of such a receptor together with CD79a potential oncogenes or tumor suppressor genes, they can lead
and CD79b (67). Monoclonal antibodies that recognize λ5 to cellular transformation and lymphoid tumors. The mech-
or VpreB specifically bind to pre-B cells and can react with anism of DNA rearrangement is similar for the heavy-and
B-lineage acute lymphocytic leukemias (68). light-chain loci. However, only one joining event is needed
Expression of the surrogate light chains plays a critical to generate a light-chain gene, whereas two are needed to
role in normal B-cell development. This is underscored by generate a complete heavy-chain gene. The most common
studies on transgenic mice that lack functional λ5 genes mode of rearrangement involves the looping out and dele-
(69). In these mice, B-cell development in the marrow is tion of the DNA between two gene segments on the same
blocked at the pre-B-cell stage, thereby markedly reducing chromosome; this occurs when the coding sequences of the
the numbers of functional mature B lymphocytes in the two gene segments are in the same orientation in the DNA
blood and lymphoid tissues (70). Similarly, humans that (85). The 12-and 23-mer-spaced RSSs are brought together
have inactivating mutations in the λ5 genes on both alleles by interactions between proteins that specifically recognize
of chromosome 22 have agammaglobulinemia and markedly the length of the spacer between the heptamer and nonamer
reduced numbers of B cells (71). When immunoglobulin μ signals, thus accounting for the 12/23 joining rule (75, 76).
chains form a complex with the “surrogate” λ light chains, The two DNA molecules then are broken and rejoined in a
the complementarity-determining region 3 (CDR3) of the different configuration (86). By joining precisely in a head-
“surrogate” λ light chain covers the CDR3 of the heavy to-head configuration, the ends of the heptamer sequences
chain in the pre-B-cell receptor, allowing the pre-B cell to form a signal joint in a circular piece of extrachromosomal
avoid antigen-specific selection (72). DNA that then is lost from the genome when the cell divides.
Even though the cell has two different sets of each of However, the DNA that lies between the two gene segments
the immunoglobulin gene complexes that initially undergo is retained in an inverted orientation when a second mode
seemingly independent immunoglobulin gene rearrange- of recombination occurs between two gene segments with
ments, under normal conditions, immunoglobulin expres- opposite transcriptional orientations. Although this mode
sion by each B lymphocyte or plasma cell is limited to only of recombination is less common, such rearrangements ac-
one immunoglobulin heavy-chain allele and one light-chain count for about one-half of all IGKV-to-IGKJ joins, as the
allele through a process called allelic exclusion. Allelic ex- transcriptional orientation of one-half of the human IGKV
clusion generally is observed in most B-cell tumors; how- gene segments is opposite to that of the IGKJ gene segments.
ever, rare cases of B-cell leukemia may lack allelic exclusion RAG-1 and RAG-2 are coexpressed normally only in
and express both immunoglobulin alleles (73). developing lymphocytes that are undergoing receptor gene
rearrangement. During B-cell development, the appearance
Genetic Basis for Immunoglobulin Gene of cytoplasmic immunoglobulin μ chains defines the pre-B
Rearrangement cell. Pre-B cells, whose immunoglobulin μ chains can asso-
Recombination signal sequences (RSSs) flanking each germ ciate with the “surrogate” λ light chains, begin to express
line V gene, D element, and J segment serve as the recog- a pre-B-cell receptor. Its appearance turns off RAG1 and
nition site for the V(D)J recombinase and direct the site of RAG2, preventing further immunoglobulin heavy-chain re-
DNA recombination. The RSSs consist of two conserved arrangement (allelic exclusion). This is followed by four to
sequence elements of different lengths, one of 7 bp (the six cycles of cell division (87). Late pre-B daughter cells re-
heptamer; consensus, 5′ CACAGTG 3′) and one of 9 bp activate RAG1 and RAG2 and begin to undergo light-chain
(the A/T-rich nonamer; consensus, 5′ ACAAAAACC 3′), rearrangement. Mice with either of these genes knocked out
separated by either 12 bp or 23 bp of nominally conserved cannot undergo immunoglobulin or T-cell receptor gene
intervening DNA (12-RSS or 23-RSS, respectively) (74). rearrangements and consequently fail to produce mature B
V(D)J recombination is directed between distinct gene seg- or T lymphocytes (88). A form of combined immune de-
ments whose flanking RSSs contain different-length spacers ficiency called Omenn syndrome in humans results from
(62, 75, 76). This is referred to as the 12/23 joining rule mutations that impair, but do not completely abolish, the
(77). Because all segments of a particular type (e.g., seg- function of RAG-1 or RAG-2 (89). It is interesting that
ments of IGKV genes) are flanked by one type of signal se- patients with Omenn syndrome harbor point mutations in
quence and all the segments to which they should be joined the RAG1 RING domain (see below) that reduces the effi-
(e.g., IGKJ segments) are flanked by the opposite type of ciency of V(D)J recombination (90, 91).
signal sequence, the 12/23 rule ensures that the joining will After the RAG-1/RAG-2 endonuclease recognition of
be restricted to events that could be biologically productive. either the 12-mer-spaced or the 23-mer-spaced RSS, such
endonuclease introduces DSBs and remains bound to the Junctional diversity is generated in the sequences of the
DNA. The presence of mutations affecting the ability of rearranged gene segments during the recombination pro-
the RAG proteins to bind and to maintain the broken ends cess. Each coding joint originating from the hairpinned
in a stable postcleavage complex can lead to misrepair of termini of gene segments is cleaved at random sites by an
the DSBs, thereby enhancing the risk for oncogenic chro- endonuclease. An overhanging flap is generated by the
mosomal aberrations (92, 93). Involved in the processing cleavage of a hairpin away from its apex and, if incorpo-
and juxtaposition of these DSBs are several proteins, in- rated into the joint, results in the addition of palindromic
cluding the high-mobility-group proteins 1 and 2 (HMG1 nucleotides that contribute to junctional diversity. Further
and HMG2). HMG1 and HMG2 play an important role in modifications on the opened hairpin ends by nucleases can
the assembly of nucleoprotein complexes involved in DNA remove a self-complementary overhang or cut further into
repair and transcription because they are widely expressed, the original coding sequence. In addition, the terminal de-
abundant nuclear proteins that bind and bend DNA with- oxynucleotidyl transferase, a lymphocyte-specific enzyme,
out sequence specificity (94). HMG1, facilitating the bend- can add non-template-encoded nucleotides (113). Finally,
ing of the DNA, allows the components of one DSB-RAG nucleolytic activities that remove potential coding end nu-
complex to bind and to cleave the DNA at a different RSS, cleotides prior to the final ligation of the DNA breaks into
therefore, in accordance with the 12/23 joining rule, bring- one intact recombination joint can provide additional junc-
ing together two disparate RSSs (75, 76). tional diversity (86).
The RAG complex introduces DSBs immediately adja- Lineage, stage, order, and feedback regulation of V(D)J
cent to immune receptor gene segments, resulting in blunt recombination are mediated epigenetically by modulating
phosphorylated signal ends and covalently sealed hair- the accessibility of particular loci or regions of loci to RAG
pinned coding termini that are joined by NHEJ (80). Coding (114). Numerous epigenetic accessibility markers, such as
and signal ends are joined at very different rates, rapid for the histone H3 lysine 4 trimethylation, are enriched around
coding ends while slower for signal ends (95). This is likely IGHJ in early pro-B cells in association with germ line tran-
explained by the fact that the RAG complex remains tightly scription (115). In addition, ubiquitination events can regu-
associated with signal ends after cleavage in a postcleavage late recombination of immunoglobulin gene segments. The
complex, whereas coding ends are released from this post- zinc finger region A of RAG1 includes an N-terminal RING
cleavage complex and need to be brought together again for domain that acts as an E3-ubiquitin ligase, which can ubiq-
repair (96). Seven NHEJ components are required for VDJ uitinate a panel of targets for various downstream events (77,
recombination: Artemis, Ku70, Ku80, DNA- dependent 116–121) and can interact with other E2 enzymes to ubiq-
protein kinase (DNA-PK), XRCC4, XLF, and DNA ligase uitinate substrates involved in V(D)J recombination, such
IV (Lig4) (97). Although all seven are required for efficient as histone 3 (116, 117). Ubiquitination of RAG2 allows
coding end joining, the dependence of signal end joining for rapid degradation of the protein upon entering S phase,
on these NHEJ factors is less absolute. The serine-threonine thereby halting any potential off-target activities of RAG and
protein kinase (DNA-PK) is activated by DNA DSBs and limiting its capacity to induce V(D)J recombination during
is essential for the normal repair of DNA breaks induced by inappropriate phases of the cell cycle. DNA breaks during
ionizing radiation, chemical agents, or VDJ recombination S phase are potentially harmful for cells, as such breaks can
(98–100). DNA-PK-deficient mice can make only trivial lead to deleterious translocations when misrepaired by ho-
amounts of immunoglobulin or T-cell receptors and are called mologous recombination. It is therefore crucial to limit V(D)
severe combined immunodeficiency mice (101). Artemis- J recombination activity within G1 phase; this restriction ap-
deficient mice have a “leaky” severe combined immunode- pears to be controlled by RAG2 degradation (122–124).
ficiency phenotype and develop some T and B cells in later
life (100). Artemis is not required for signal end joining but Heavy-Chain Class Switching
instead plays a specific role in opening the sealed hairpinned A single B lymphocyte, during differentiation, can synthe-
termini of cleaved coding ends (102). The dependence of size heavy chains with the same variable region coupled
signal end joining on DNA-PK is variable. Artemis’s endo- to different constant regions (125). As pre-B cells develop
nuclease activity requires physical interaction with DNA-PK into mature B cells in early development, intact IgM mono-
as well as DNA-PK’s enzymatic activity. Because Artemis is mers are inserted into the plasma membrane, followed by
dispensable for signal end joining, it is reasonable to assume IgD molecules with the same antigen-binding specificity.
that signal end joining also might progress normally in the The IgM and IgD constant-region genes generally are tran-
absence of DNA-PK (102). Indeed, signal end joining pro- scribed together. Alternative splicing permits simultaneous
ceeds fairly efficiently in developing murine lymphocytes (10 coproduction of IgM and IgD from a single species of RNA.
to 50% of wild-type levels) and in some murine cell lines and Later during development, these variable domains can
one human cell line deficient in DNA-PK (103, 104). How- associate with other isotypes (IgG, IgA, or IgE) through
ever, DNA-PK deficiency in other species (hamster, horse, a controlled process called class switch recombination
dog) results in a more severe reduction in signal end joining (CSR). CSR requires active transcription of the down-
(30-to 1,000-fold reduced) (105–107). Interestingly, mice stream constant-region exons encoding the switched im-
deficient in Ku not only are deficient in T and B cells but munoglobulin isotype. Such active transcription generally
also have small stature and other nonimmunologic defects, is triggered by stimulation of the B cell with antigen, mi-
suggesting that the Ku proteins also play important roles in togens, and/or ligation of CD40 via the ligand for CD40
normal development (108, 109). Mice defects resulting from (CD154), which is expressed by activated T cells. Patients
mutations in either Ku, XRCC4, Lig4, Artemis, or DNA-PK with X-linked or autosomal recessive inheritance of muta-
genes predispose to lymphoma (110, 111). The initial recog- tions in CD154 or CD40, respectively, can have an immune
nition and recruitment of repair proteins to DSBs is heavily deficiency syndrome (hyper-IgM syndrome type I), which is
dependent on phosphorylation events (e.g., kinase activity characterized by normal to high levels of IgM and extremely
and autophosphorylation of DNA-PK), but the downstream low levels of other immunoglobulin isotypes in serum (126,
repair events appear to be more dependent on E3-ubiquitin 127). Antigen-reactive T lymphocytes also provide inter-
ligases (112). leukins that can influence whether the B cells continue to
express IgM or switch their immunoglobulin heavy-chain themselves (62). The last of these occurs through a process
isotype to IgG, IgA, or IgE (125, 128). called somatic hypermutation (SHM). This process occurs
Immunoglobulin CSR occurs in or near the switch region in antigen-activated B cells and cannot be triggered merely
upstream of the gene and any one of the switch regions of the by mitogen-induced B-cell activation.
other heavy-chain isotype genes. Switch regions in Fig. 1 are The SHM process primarily introduces single nucleo-
represented by filled circles upstream of the IGHC genes. The tide exchanges into the rearranged immunoglobulin vari-
μ switch region (Sμ) consists of approximately 150 repeats of able region, at a rate of 10−3 per base pair per generation.
the sequence (GAGCT)n(GGGGGT), where n is generally However, during discrete stages of B-cell differentiation and
three but can be as many as seven. Other switch regions (Sγ, in particular during the secondary immune response to an
Sα, and Sε) consist of similar sequences that contain repeats antigen, expressed immunoglobulin variable genes can in-
of the GAGCT and GGGGGT sequences. DNA recombi- cur new mutations at rates as high as 10−3 base changes per
nation between Sμ and Sγ, Sα, or Sε accompanied by the de- base pair per generation over several cell divisions (139).
letion of intervening DNA segments results in the apposition Mutations generated by SHM are clustered in the region
of the previously rearranged variable-region gene next to the spanning from 300 bp 5′ of the rearranged variable-region
new constant-region gene, thus switching the heavy-chain exon to approximately 1 kb 3′ of the rearranged minigene J
class. While VDJ recombination occurs mostly in the G0 and/ segment, frequently around hot spots defined by the primary
or G1 stage of the cell cycle, CSR seems to require DNA rep- DNA sequence, such as the sequence RGYW (R, purine [A
lication (129). or G]; Y, pyrimidine [C or T]; W, A or T) and its comple-
Unlike VDJ recombination, CSR requires expression of ment. Such sequences are hot spots for mutation that are
activation-induced deaminase (AID), an enzyme expressed conserved among species (140, 141).
in activated B cells that also is required for somatic hyper- SHM is dependent on the expression and activity of AID
mutation (see “Generation of Antibody Diversity” below) through a process that has some similarity to CSR (142).
(130–132). Inherited defects in AID can result in an im- AID is responsible for initiating mutations at RGYW motifs.
mune deficiency syndrome called hyper-IgM syndrome type Mutations spread as a consequence of downstream events or-
II, which is characterized by relatively high levels of IgM and chestrated by the mismatch repair and base excision repair
negligible levels of other immunoglobulin isotypes in serum pathways (143–147). In addition to having the hyper-IgM
(133). Specific inactivation of the C-terminal AID domain, immune deficiency syndrome type II, patients with inher-
encoded by exon 5 (E5), allows very efficient deamination ited defects in AID have B cells that lack the capacity to
of the AID target regions but greatly impacts the efficiency undergo SHM (131). As with CSR, SHM requires active
and quality of subsequent DNA repair. Specifically eliminat- transcription of the genes undergoing mutation. In the re-
ing E5 not only precludes CSR but also causes an atypical, gion encompassing the rearranged variable-region gene, AID
enzymatic-activity-dependent, dominant-negative effect on most likely deaminates the dC, converting the dC into dU,
CSR. This explains the autosomal dominant inheritance of which is converted into T after DNA replication, giving
AID variants with truncated E5 in patients with hyper-IgM rise to CG-to-T or -A transitions. Alternatively, UNG may
syndrome type II and establishes that AID, through the E5 remove the dU, resulting in staggered nick cleavage of the
domain, provides a link between DNA damage and repair DNA. Such abasic sites are subsequently cleaved by apurinic/
during CSR (134). The sites where CSR takes place in B cells apyrimidinic endonuclease. Low-fidelity DNA synthesis may
activated in response to an antigen are germinal centers of repair these staggered nicks, resulting in frequent mutations.
the lymph nodes and spleen where AID is expressed (135). A complex called the mutasome, formed by DNA-cleaving
Uracil-DNA glycosylase (UNG) is an enzyme that re- enzymes and DNA repair enzymes (e.g., mismatch repair
moves the uracils (dU) generated from the deamination of enzymes, base excision repair enzymes, proteins involved in
cytosines (dC) by AID. Inherited defects in this enzyme un- NHEJ, etc.), also apparently binds the target DNA to reduce
derlie an autosomal recessive form of the hyper-IgM immune its tendency to incur complete DNA DSBs. Through this
deficiency syndrome, which is similar to that of patients with process, mostly transitional mutations are introduced at high
inherited defects in AID (6, 136). Apurinic/apyrimidinic frequencies into the expressed immunoglobulin V genes as
endonuclease cleaves the abasic sites generated by UNG; well as other transcriptionally active genes within hot spots,
the staggered nicks in the DNA are closely positioned and which serve as substrates for AID, UNG, and the mutasome
may result in DNA DSBs (137). Such DNA DSBs are end (140, 143, 148). Subsequently, “affinity maturation” of the
processed, repaired, and joined by mechanisms and proteins antibodies expressed during the immune response to antigen
similar to those involved in NHEJ used for VDJ recombina- takes place by selection of the B cell and its daughter cells
tion. Since CSR occurs in the intron between the variable- that express mutated V genes encoding an immunoglobulin
region exon and the exon encoding the first constant-region variable region with improved fitness for binding antigen
domain, no mutations in the regions encoding the variable (149). Such selection enhances the frequency of noncon-
or constant regions of the newly generated immunoglobulin servative base substitutions in the DNA sequences encoding
heavy chain are generated by this process (138). the complementarity-determining regions that serve as the
contact sites for antigen binding (150).
Immunoglobulin enhancers may account in part for the
GENERATION OF ANTIBODY DIVERSITY preferential somatic hypermutation of immunoglobulin
The diversity among immunoglobulin variable regions genes. Combinations of immunoglobulin enhancers target so-
can be generated by several mechanisms: (i) the germ line matic mutation to immunoglobulin genes by recruiting AID
presence of multiple V, J, and D gene segments; (ii) the and/or by making the immunoglobulin genes better substrates
DNA segments randomly joining to produce a complete for mutation (151). In addition, posttranslation AID ubiq-
variable-region exon; (iii) uncorrected errors made during uitination has an important regulatory role during CSR and
the recombination process; (iv) the heavy-and light- SHM (152, 153). Unlike RAG2, AID protein stability is not
chain polypeptides pairing to produce a complete immu- associated with phases of the cell cycle but rather with subcel-
noglobulin monomer capable of binding an antigen; and lular localization. In mouse B cells, nuclear AID is subjected
(v) somatic mutations in the rearranged DNA segments to rapid turnover upon polyubiquitination (152).
DETECTION OF IMMUNOGLOBULIN GENE the J gene segment(s) will fail to amplify any immunoglobulin
REARRANGEMENTS genes unless the genes have first undergone rearrangement.
Analysis for immunoglobulin gene rearrangements can detect However, after immunoglobulin gene rearrangement, the
clonal immunoglobulin gene rearrangements. Immunoglobu- variable-region gene leader sequence primers and J primers
lin gene rearrangement irreversibly alters the genomic DNA anneal to sites that are separated by fewer than 400 bp, mak-
of the developing B cell and its descendant daughter cells. ing PCR straightforward (154). On the other hand, a PCR
Because of the many possible distinct immunoglobulin gene assay on genomic DNA with primers specific for a V gene sub-
rearrangements possible, a B cell’s particular type of immu- group and a constant-region exon does not amplify rearranged
noglobulin gene rearrangement can serve as a clonal marker. immunoglobulin genes because the distance between the J
segment and the immunoglobulin constant-region exon is too
PCR Assays large. Such primer pairs are better suited for amplifying cDNA
Sensitive methods for detecting immunoglobulin gene re- derived from the immunoglobulin transcripts from which the
arrangements, such as PCR, can be used to examine for intron separating the J region and the constant-region exon
residual cells of a malignant B-cell clone following antitu- are deleted through RNA processing and splicing.
mor therapy. PCR, using sense strand oligonucleotide prim- Genomic DNA is the most commonly used for PCR in
ers corresponding to an immunoglobulin V gene subgroup clinical laboratories, as this does not require synthesis of
(Table 2) together with antisense strand oligonucleotide cDNA from isolated RNA. However, care must be taken
primers corresponding to the relevant J-region or constant- when using genomic DNA for PCR, which can amplify both
region exon, can amplify the rearranged V gene in genomic the productively rearranged immunoglobulin gene and its
DNA or cDNA, respectively. unproductively rearranged allele. The use of downstream
Because these gene segments are separated by large stretches primers located in the constant region (isotype- specific
of intervening DNA in germ line DNA, a PCR assay with ge- primers) allows definition of the isotype and may be advan-
nomic DNA and primers specific for a V gene subgroup and tageous in the case of mutated sequences, since it is not tar-
geted by the somatic hypermutation process (155).
TABLE 2 Sense and antisense strand oligonucleotide Resolution of the junctional sequences in the rearranged
primers to amplify the rearranged V gene in genomic DNA or immunoglobulin genes expressed by a tumor can provide a
cDNAa specific tumor marker. This marker can be used to examine
for any tumor-derived immunoglobulin gene fragments am-
Gene Primer sequenceb plified by PCR performed on genomic DNA of lymphoid
VH genes tissue. Such methods are highly sensitive, allowing for detec-
VH1/7 ATGGACTGGACCTGGAGVATCC tion of minimal residual disease of small numbers of lymphoid
tumor cells that otherwise cause no detectable pathologic or
VH2 CTCCACRCTCCTGCTGCTGAC immunophenotypic trace of residual disease after therapy.
VH3 GCTGGGTTTTCCTTGTTGC
VH4 ATGAAACACCTGTGGTTCTTCCT
REFERENCES
VH5 ATGGGGTCAACCGCCATCCTYG
1. Joiner KA, Fries LF, Frank MM. 1987. Studies of anti-
VH6 CTGTCTCCTTCCTCATCTTCC body and complement function in host defense against
JH CTYACCTGARGAGACRGTGACC bacterial infection. Immunol Lett 14:197–202.
Cμ AACGGCCACGCTGCTCG 2. Edelman GM. 1991. Antibody structure and molecular im-
Vκ genes munology. Scand J Immunol 34:1–22.
3. Schroeder HW Jr, Cavacini L. 2010. Structure and func-
Vκ1/2 ATGAGGSTCCCYGCTCAGCTGCTGG
tion of immunoglobulins. J Allergy Clin Immunol 125(2
Vκ3 CTCTTCCTCCTGCTACTCTGGCTCCCAG Suppl 2):S41–S52.
Vκ4 ATTTCTCGTTGCTCTGGATCTCTG 4. Torres RM, Imbonden JB, Schroeder HW Jr. 2008. An-
Vκ5 GACATCGTGATGACCCAG tigen receptor genes, gene products, and co-receptors, p
53–77. In Rich RR, Fleisher TA, Shearer WT, Schroeder
Vκ6a GAAATTGTGCTGACTCAG HW Jr, Frew AJ, Weyand CM (ed), Clinical Immunology:
Vκ6b GATGTTGTGATGACACAG Principles and Practice, 3rd ed. Mosby Elsevier, London,
Cκ GTTTCTCGTAGTCTGCTTTGCTCA United Kingdom.
5. Virella G, Wang AC. 1993. Immunoglobulin structure.
Vλ genes Immunol Ser 58:75–90.
Vλ1 GGTCCTGGGCCCAGTCTGTGCTG 6. Kavli B, Andersen S, Otterlei M, Liabakk NB, Imai K,
Vλ2 GGTCCTGGGCCCAGTCTGCCCTG Fischer A, Durandy A, Krokan HE, Slupphaug G. 2005.
B cells from hyper-IgM patients carrying UNG mutations
Vλ3 GCTCTGTGACCTCCTATGAGCTG
lack ability to remove uracil from ssDNA and have ele-
Vλ4/5 GGTCTCTCTCSCAGCYTGTGCTG vated genomic uracil. J Exp Med 201:2011–2021.
Vλ6 GTTCTTGGGCCAATTTTATGCTG 7. Harris LJ, Larson SB, Hasel KW, Day J, Greenwood A,
Vλ7/8 TTCYCAGRCTGTGGTGACYCAG McPherson A. 1992. The three-dimensional structure of
an intact monoclonal antibody for canine lymphoma. Na-
Vλ9 CWGCCTGTGCTGACTCAG ture 360:369–372.
Vλ10 CAGGCAGGGCTGACTCAG 8. Jefferis R, Lund J, Goodall M. 1995. Recognition sites on
Cλ CACCAGTGTGGCCTTGTTGGCTTG human IgG for Fc gamma receptors: the role of glycosyla-
tion. Immunol Lett 44:111–117.
a
Sense-strand oligonucleotide primers corresponding to the leader sequences
of each of the major V gene subgroups and antisense strand oligonucleotide prim-
9. Lee YK, Brewer JW, Hellman R, Hendershot LM. 1999.
ers corresponding to the relevant J-region or constant-region exon. BiP and immunoglobulin light chain cooperate to control
b
V, R, Y, S, and W represent degenerate bases (V is A or C or G; R is A or G; the folding of heavy chain and ensure the fidelity of immu-
Y is C or T; S is C or G; and W is A or T). noglobulin assembly. Mol Biol Cell 10:2209–2219.
10. Ravetch JV, Bolland S. 2001. IgG Fc receptors. Annu Rev sequences reveals about fifty groups of VH segments with
Immunol 19:275–290. different hypervariable loops. J Mol Biol 227:776–798.
11. Parlowsky T, Welzel J, Amagai M, Zillikens D, Wygold 33. Pascual V, Capra JD. 1991. Human immunoglobulin
T. 2003. Neonatal pemphigus vulgaris: IgG4 autoantibod- heavy-chain variable region genes: organization, polymor-
ies to desmoglein 3 induce skin blisters in newborns. J Am phism, and expression. Adv Immunol 49:1–74.
Acad Dermatol 48:623–625. 34. Sanz I, Kelly P, Williams C, Scholl S, Tucker P, Capra
12. Yeh SW, Cavacini LA, Bhol KC, Lin MS, Kumar M, Du- JD. 1989. The smaller human VH gene families display
val M, Posner MR, Ahmed AR. 2006. Pathogenic human remarkably little polymorphism. EMBO J 8:3741–3748.
monoclonal antibody against desmoglein 3. Clin Immunol 35. Meek K, Eversole T, Capra JD. 1991. Conservation of the
120:68–75. most JH proximal Ig VH gene segment (VHVI) through-
13. Corthesy B, Kraehenbuhl JP. 1999. Antibody-mediated out primate evolution. J Immunol 146:2434–2438.
protection of mucosal surfaces. Curr Top Microbiol Immunol 36. Milner EC, Hufnagle WO, Glas AM, Suzuki I, Alexan-
236:93–111. der C. 1995. Polymorphism and utilization of human VH
14. Lamm ME, Nedrud JG, Kaetzel CS, Mazanec MB. 1995. genes. Ann N Y Acad Sci 764:50–61.
IgA and mucosal defense. APMIS 103:241–246. 37. Sasso EH, Johnson T, Kipps TJ. 1996. Expression of the
15. Mestecky J, Lue C, Russell MW. 1991. Selective trans- immunoglobulin VH gene 51p1 is proportional to its germ
port of IgA. Cellular and molecular aspects. Gastroenterol line gene copy number. J Clin Invest 97:2074–2080.
Clin N Am 20:441–471. 38. Pramanik S, Li H. 2002. Direct detection of insertion/de-
16. Woof JM, Mestecky J. 2005. Mucosal immunoglobulins. letion polymorphisms in an autosomal region by analyzing
Immunol Rev 206:64–82. high-density markers in individual spermatozoa. Am J Hum
17. Niles MJ, Matsuuchi L, Koshland ME. 1995. Polymer Genet 71:1342–1352.
IgM assembly and secretion in lymphoid and nonlymphoid 39. Lefranc MP, Lefranc G. 2001. The Immunoglobulin Facts-
cell lines: evidence that J chain is required for pentamer book. Academic Press, San Diego, CA.
IgM synthesis. Proc Natl Acad Sci U S A 92:2884–2888. 40. Cui X, Li H. 2000. Human immunoglobulin VH4 sequences
18. Corthesy B. 2007. Roundtrip ticket for secretory IgA: role resolved by population-based analysis after enzymatic am-
in mucosal homeostasis? J Immunol 178:27–32. plification and denaturing gradient gel electrophoresis. Eur
19. Brezski RJ, Monroe JG. 2008. B-cell receptor. Adv Exp J Immunogenet 27:37–46.
Med Biol 640:12–21.
41. Chimge NO, Pramanik S, Hu G, Lin Y, Gao R, Shen L,
20. Roes J, Rajewsky K. 1993. Immunoglobulin D (IgD)-
Li H. 2005. Determination of gene organization in the hu-
deficient mice reveal an auxiliary receptor function for
man IGHV region on single chromosomes. Genes Immun
IgD in antigen-mediated recruitment of B cells. J Exp Med
6:186–193.
177:45–55.
21. Geisberger R, Lamers M, Achatz G. 2006. The rid- 42. Ruiz M, Giudicelli V, Ginestoux C, Stoehr P, Robinson
dle of the dual expression of IgM and IgD. Immunology J, Bodmer J, Marsh SG, Bontrop R, Lemaitre M, Le-
118:429–437. franc G, Chaume D, Lefranc MP. 2000. IMGT, the inter
22. Riesbeck K, Nordstrom T. 2006. Structure and immuno- national ImMunoGeneTics database. Nucleic Acids Res
logical action of the human pathogen Moraxella catarrha- 28:219–221.
lis IgD-binding protein. Crit Rev Immunol 26:353–376. 43. Kawasaki K, Minoshima S, Nakato E, Shibuya K, Shin-
23. Lyczak JB, Zhang K, Saxon A, Morrison SL. 1996. Ex- tani A, Asakawa S, Sasaki T, Klobeck HG, Combriato
pression of novel secreted isoforms of human immunoglob- G, Zachau HG, Shimizu N. 2001. Evolutionary dynam-
ulin E proteins. J Biol Chem 271:3428–3436. ics of the human immunoglobulin kappa locus and the
24. Gould HJ, Sutton BJ, Beavil AJ, Beavil RL, McCloskey germline repertoire of the Vkappa genes. Eur J Immunol
N, Coker HA, Fear D, Smurthwaite L. 2003. The biology 31:1017–1028.
of IGE and the basis of allergic disease. Annu Rev Immunol 44. Tomlinson IM, Cook GP, Walter G, Carter NP, Rieth-
21:579–628. man H, Buluwela L, Rabbitts TH, Winter G. 1995. A
25. Chang TW, Wu PC, Hsu CL, Hung AF. 2007. Anti-IgE complete map of the human immunoglobulin VH locus.
antibodies for the treatment of IgE-mediated allergic dis- Ann N Y Acad Sci 764:43–46.
eases. Adv Immunol 93:63–119. 45. Zachau HG. 1993. The immunoglobulin kappa locus-or-
26. Kirsch IR, Morton CC, Nakahara K, Leder P. 1982. Hu- what has been learned from looking closely at one-tenth of
man immunoglobulin heavy chain genes map to a region a percent of the human genome. Gene 135:167–173.
of translocations in malignant B lymphocytes. Science 216: 46. Schable K, Thiebe R, Flugel A, Meindl A, Zachau
301–303. HG. 1994. The human immunoglobulin kappa locus:
27. Kipps TJ. 1997. Human B cell biology. Int Rev Immunol pseudogenes, unique and repetitive sequences. Biol Chem
15:243–264. Hoppe Seyler 375:189–199.
28. Matsuda F, Ishii K, Bourvagnet P, Kuma K, Hayashida 47. Huber C, Huber E, Lautner-Rieske A, Schable KF,
H, Miyata T, Honjo T. 1998. The complete nucleotide Zachau HG. 1993. The human immunoglobulin kappa lo-
sequence of the human immunoglobulin heavy chain vari- cus. Characterization of the partially duplicated L regions.
able region locus. J Exp Med 188:2151–2162. Eur J Immunol 23:2860–2867.
29. Ruiz M, Pallares N, Contet V, Barbi V, Lefranc MP. 48. Lautner-Rieske A, Huber C, Meindl A, Pargent W,
1999. The human immunoglobulin heavy diversity Schable KF, Thiebe R, Zocher I, Zachau HG. 1992. The
(IGHD) and joining (IGHJ) segments. Exp Clin Immuno- human immunoglobulin kappa locus. Characterization of
genet 16:173–184. the duplicated A regions. Eur J Immunol 22:1023–1029.
30. Cook GP, Tomlinson IM. 1995. The human immunoglob- 49. Cox JP, Tomlinson IM, Winter G. 1994. A directory of
ulin VH repertoire. Immunol Today 16:237–242. human germ-line V kappa segments reveals a strong bias in
31. Kirkham PM, Mortari F, Newton JA, Schroeder HW. their usage. Eur J Immunol 24:827–836.
1992. Immunoglobulin VH clan and family identity pre- 50. Klein R, Zachau HG. 1995. Expression and hypermu-
dicts variable domain structure and may influence antigen tation of human immunoglobulin kappa genes. Ann N Y
binding. EMBO J 11:603–609. Acad Sci 764:74–83.
32. Tomlinson IM, Walter G, Marks JD, Llewelyn MB, 51. Feeney AJ, Atkinson MJ, Cowan MJ, Escuro G, Lugo
Winter G. 1992. The repertoire of human germline VH G. 1996. A defective Vkappa A2 allele in Navajos which
may play a role in increased susceptibility to Haemophilus 69. Kitamura D, Kudo A, Schaal S, Muller W, Melchers F,
influenzae type b disease. J Clin Invest 97:2277–2282. Rajewsky K. 1992. A critical role of lambda 5 protein in B
52. Hilschmann N. 1987. The immunoglobulin receptor. Beh- cell development. Cell 69:823–831.
ring Inst Mitt 81:98–99. 70. Corcos D, Dunda O, Butor C, Cesbron JY, Lores P,
53. Ghanem N, Dariavach P, Bensmana M, Chibani J, Le- Bucchini D, Jami J. 1995. Pre-B-cell development in the
franc G, Lefranc MP. 1988. Polymorphism of immuno- absence of lambda 5 in transgenic mice expressing a heavy-
globulin lambda constant region genes in populations chain disease protein. Curr Biol 5:1140–1148.
from France, Lebanon and Tunisia. Exp Clin Immunogenet 71. Minegishi Y, Coustan-Smith E, Wang YH, Cooper MD,
5:186–195. Campana D, Conley ME. 1998. Mutations in the human
54. Dariavach P, Lefranc G, Lefranc MP. 1987. Human im- lambda5/14.1 gene result in B cell deficiency and agamma-
munoglobulin C lambda 6 gene encodes the Kern+Oz- globulinemia. J Exp Med 187:71–77.
lambda chain and C lambda 4 and C lambda 5 are 72. Bankovich AJ, Raunser S, Juo ZS, Walz T, Davis MM,
pseudogenes. Proc Natl Acad Sci U S A 84:9074–9078. Garcia KC. 2007. Structural insight into pre-B cell recep-
55. Vasicek TJ, Leder P. 1990. Structure and expression of tor function. Science 316:291–294.
the human immunoglobulin lambda genes. J Exp Med 73. Rassenti LZ, Kipps TJ. 1997. Lack of allelic exclu-
172:609–620. sion in B cell chronic lymphocytic leukemia. J Exp Med
56. Kawasaki K, Minoshima S, Nakato E, Shibuya K, Shin- 185:1435–1445.
tani A, Schmeits JL, Wang J, Shimizu N. 1997. One- 74. Lewis SM. 1994. The mechanism of V(D)J joining: lessons
megabase sequence analysis of the human immunoglobulin from molecular, immunological, and comparative analyses.
lambda gene locus. Genome Res 7:250–261. Adv Immunol 56:27–150.
57. Pallares N, Frippiat JP, Giudicelli V, Lefranc MP. 1998. 75. Gellert M. 1997. Recent advances in understanding V(D)
The human immunoglobulin lambda variable (IGLV) J recombination. Adv Immunol 64:39–64.
genes and joining (IGLJ) segments. Exp Clin Immunogenet 76. Steen SB, Gomelsky L, Speidel SL, Roth DB. 1997. Ini-
15:8–18. tiation of V(D)J recombination in vivo: role of recombi-
58. Williams SC, Frippiat JP, Tomlinson IM, Ignatovich O, nation signal sequences in formation of single and paired
Lefranc MP, Winter G. 1996. Sequence and evolution double-strand breaks. EMBO J 16:2656–2664.
of the human germline V lambda repertoire. J Mol Biol 77. Schatz DG, Swanson PC. 2011. V(D)J recombination:
264:220–232. mechanisms of initiation. Annu Rev Genet 45:167–202.
59. Chuchana P, Blancher A, Brockly F, Alexandre D, Le- 78. Fugmann SD, Lee AI, Shockett PE, Villey IJ, Schatz
franc G, Lefranc MP. 1990. Definition of the human im- DG. 2000. The RAG proteins and V(D)J recombination:
munoglobulin variable lambda (IGLV) gene subgroups. complexes, ends, and transposition. Annu Rev Immunol
Eur J Immunol 20:1317–1325. 18:495–527.
60. Frippiat JP, Williams SC, Tomlinson IM, Cook GP, 79. Oettinger MA. 1999. V(D)J recombination: on the cut-
Cherif D, Le Paslier D, Collins JE, Dunham I, Winter ting edge. Curr Opin Cell Biol 11:325–329.
G, Lefranc MP. 1995. Organization of the human im- 80. Gellert M. 2002. V(D)J recombination: RAG proteins, re-
munoglobulin lambda light-chain locus on chromosome pair factors, and regulation. Annu Rev Biochem 71:101–132.
22q11.2. Hum Mol Genet 4:983–991. 81. Lieber MR. 2010. The mechanism of double-strand DNA
61. Abe M, Ozaki S, Wolfenbarger D, deBram-Hart M, break repair by the nonhomologous DNA end- joining
Weiss DT, Solomon A. 1994. Variable-region subgroup pathway. Annu Rev Biochem 79:181–211.
distribution among lambda-type immunoglobulins in nor- 82. Rooney S, Chaudhuri J, Alt FW. 2004. The role of the
mal human serum. J Clin Lab Anal 8:4–9. non-homologous end-joining pathway in lymphocyte de-
62. Tonegawa S. 1993. The Nobel Lectures in Immunology. velopment. Immunol Rev 200:115–131.
The Nobel Prize for Physiology or Medicine, 1987. So- 83. Gostissa M, Alt FW, Chiarle R. 2011. Mechanisms that
matic generation of immune diversity. Scand J Immunol promote and suppress chromosomal translocations in lym-
38:303–319. phocytes. Annu Rev Immunol 29:319–350.
63. Graninger WB, Goldman PL, Morton CC, O’Brien SJ, 84. Nussenzweig A, Nussenzweig MC. 2010. Origin of
Korsmeyer SJ. 1988. The kappa-deleting element. Germ- chromosomal translocations in lymphoid cancer. Cell
line and rearranged, duplicated and dispersed forms. J Exp 141:27–38.
Med 167:488–501. 85. Helmink BA, Sleckman BP. 2012. The response to and
64. Xiang Y, Park SK, Garrard WT. 2013. Vkappa gene reper- repair of RAG-mediated DNA double-strand breaks. Annu
toire and locus contraction are specified by critical DNase I Rev Immunol 30:175–202.
hypersensitive sites within the Vkappa-Jkappa intervening 86. Grawunder U, West RB, Lieber MR. 1998. Anti-
region. J Immunol 190:1819–1826. gen receptor gene rearrangement. Curr Opin Immunol
65. Martensson IL, Ceredig R. 2000. Review article: role of 10:172–180.
the surrogate light chain and the pre-B-cell receptor in 87. Matthias P, Rolink AG. 2005. Transcriptional networks
mouse B-cell development. Immunology 101:435–441. in developing and mature B cells. Nat Rev Immunol
66. Melchers F, Karasuyama H, Haasner D, Bauer S, Kudo 5:497–508.
A, Sakaguchi N, Jameson B, Rolink A. 1993. The sur- 88. Shinkai Y, Rathbun G, Lam KP, Oltz EM, Stewart V,
rogate light chain in B-cell development. Immunol Today Mendelsohn M, Charron J, Datta M, Young F, Stall AM,
14:60–68. et al. 1992. RAG-2-deficient mice lack mature lympho-
67. Ten Boekel E, Yamagami T, Andersson J, Rolink A, cytes owing to inability to initiate V(D)J rearrangement.
Melchers F. 1999. The formation and selection of cells ex- Cell 68:855–867.
pressing preB cell receptors and B cell receptors. Curr Top 89. Villa A, Santagata S, Bozzi F, Imberti L, Notarangelo
Microbiol Immunol 246:3–9; discussion, 246:10. LD. 1999. Omenn syndrome: a disorder of Rag1 and Rag2
68. Tsuganezawa K, Kiyokawa N, Matsuo Y, Kitamura F, genes. J Clin Immunol 19:87–97.
Toyama-Sorimachi N, Kuida K, Fujimoto J, Karasuyama 90. Villa A, Sobacchi C, Notarangelo LD, Bozzi F, Abinun M,
H. 1998. Flow cytometric diagnosis of the cell lineage and Abrahamsen TG, Arkwright PD, Baniyash M, Brooks
developmental stage of acute lymphoblastic leukemia by EG, Conley ME, Cortes P, Duse M, Fasth A, Filipov-
novel monoclonal antibodies specific to human pre-B-cell ich AM, Infante AJ, Jones A, Mazzolari E, Muller SM,
receptor. Blood 92:4317–4324. Pasic S, Rechavi G, Sacco MG, Santagata S, Schroeder
ML, Seger R, Strina D, Ugazio A, Valiaho J, Vihinen M, Russell terriers: a new animal model of DNA-PKcs defi-
Vogler LB, Ochs H, Vezzoni P, Friedrich W, Schwarz K. ciency. J Immunol 167:2142–2150.
2001. V(D)J recombination defects in lymphocytes due 108. Gu Y, Sekiguchi J, Gao Y, Dikkes P, Frank K, Ferguson
to RAG mutations: severe immunodeficiency with a spec- D, Hasty P, Chun J, Alt FW. 2000. Defective embryonic
trum of clinical presentations. Blood 97:81–88. neurogenesis in Ku- deficient but not DNA- dependent
91. Simkus C, Anand P, Bhattacharyya A, Jones JM. protein kinase catalytic subunit-deficient mice. Proc Natl
2007. Biochemical and folding defects in a RAG1 Acad Sci U S A 97:2668–2673.
variant associated with Omenn syndrome. J Immunol 109. Ouyang H, Nussenzweig A, Kurimasa A, Soares VC,
179:8332–8340. Li X, Cordon-Cardo C, Li W, Cheong N, Nussenzweig
92. Huye LE, Purugganan MM, Jiang MM, Roth DB. 2002. M, Iliakis G, Chen DJ, Li GC. 1997. Ku70 is required
Mutational analysis of all conserved basic amino acids for DNA repair but not for T cell antigen receptor gene
in RAG- 1 reveals catalytic, step arrest, and joining- recombination in vivo. J Exp Med 186:921–929.
deficient mutants in the V(D)J recombinase. Mol Cell Biol 110. Bassing CH, Suh H, Ferguson DO, Chua KF, Manis J,
22:3460–3473. Eckersdorff M, Gleason M, Bronson R, Lee C, Alt FW.
93. Tsai CL, Drejer AH, Schatz DG. 2002. Evidence of a 2003. Histone H2AX: a dosage-dependent suppressor of
critical architectural function for the RAG proteins in oncogenic translocations and tumors. Cell 114:359–370.
end processing, protection, and joining in V(D)J recom- 111. Celeste A, Difilippantonio S, Difilippantonio MJ,
bination. Genes Dev 16:1934–1949. Fernandez-Capetillo O, Pilch DR, Sedelnikova OA,
94. Thomas JO, Travers AA. 2001. HMG1 and 2, and re- Eckhaus M, Ried T, Bonner WM, Nussenzweig A.
lated ‘architectural’ DNA-binding proteins. Trends Bio- 2003. H2AX haploinsufficiency modifies genomic stabil-
chem Sci 26:167–174. ity and tumor susceptibility. Cell 114:371–383.
95. Schlissel MS. 1998. Structure of nonhairpin coding-end 112. Al-Hakim A, Escribano-Diaz C, Landry MC, O’Don-
DNA breaks in cells undergoing V(D)J recombination. nell L, Panier S, Szilard RK, Durocher D. 2010. The
Mol Cell Biol 18:2029–2037. ubiquitous role of ubiquitin in the DNA damage response.
96. Jones JM, Gellert M. 2001. Intermediates in V(D)J re- DNA Repair (Amsterdam) 9:1229–1240.
combination: a stable RAG1/2 complex sequesters cleaved 113. Komori T, Okada A, Stewart V, Alt FW. 1993. Lack of N
RSS ends. Proc Natl Acad Sci U S A 98:12926–12931. regions in antigen receptor variable region genes of TdT-
97. Lieber MR. 2008. The mechanism of human nonhomol- deficient lymphocytes. Science 261:1171–1175.
ogous DNA end joining. J Biol Chem 283:1–5. 114. Sleckman BP, Oltz EM. 2012. Preparing targets for V(D)
98. Goodarzi AA, Jeggo PA. 2013. The repair and signal- J recombinase: transcription paves the way. J Immunol
ing responses to DNA double-strand breaks. Adv Genet 188:7–9.
82:1–45. 115. Subrahmanyam R, Sen R. 2012. Epigenetic features that
99. Ma Y, Schwarz K, Lieber MR. 2005. The Artemis:DNA- regulate IgH locus recombination and expression. Curr
PKcs endonuclease cleaves DNA loops, flaps, and gaps. Top Microbiol Immunol 356:39–63.
DNA Repair (Amsterdam) 4:845–851. 116. Grazini U, Zanardi F, Citterio E, Casola S, Goding CR,
100. Taccioli GE, Amatucci AG, Beamish HJ, Gell D, Xiang McBlane F. 2010. The RING domain of RAG1 ubiquityl-
XH, Torres Arzayus MI, Priestley A, Jackson SP, Mar- ates histone H3: a novel activity in chromatin-mediated
shak Rothstein A, Jeggo PA, Herrera VL. 1998. Tar- regulation of V(D)J joining. Mol Cell 37:282–293.
geted disruption of the catalytic subunit of the DNA-PK 117. Jones JM, Bhattacharyya A, Simkus C, Vallieres B,
gene in mice confers severe combined immunodeficiency Veenstra TD, Zhou M. 2011. The RAG1 V(D)J recombi-
and radiosensitivity. Immunity 9:355–366. nase/ubiquitin ligase promotes ubiquitylation of acetylated,
101. Khanna KK, Jackson SP. 2001. DNA double- strand phosphorylated histone 3.3. Immunol Lett 136:156–162.
breaks: signaling, repair and the cancer connection. Nat 118. Jones JM, Gellert M. 2003. Autoubiquitylation of the
Genet 27:247–254. V(D)J recombinase protein RAG1. Proc Natl Acad Sci
102. Ma Y, Pannicke U, Schwarz K, Lieber MR. 2002. Hair- U S A 100:15446–15451.
pin opening and overhang processing by an Artemis/ 119. Kassmeier MD, Mondal K, Palmer VL, Raval P, Kumar
DNA- dependent protein kinase complex in nonho- S, Perry GA, Anderson DK, Ciborowski P, Jackson S,
mologous end joining and V(D)J recombination. Cell Xiong Y, Swanson PC. 2012. VprBP binds full-length
108:781–794. RAG1 and is required for B-cell development and V(D)J
103. Bogue MA, Jhappan C, Roth DB. 1998. Analysis of vari- recombination fidelity. EMBO J 31:945–958.
able (diversity) joining recombination in DNAdependent 120. Simkus C, Bhattacharyya A, Zhou M, Veenstra TD,
protein kinase (DNA-PK)-deficient mice reveals DNA- Jones JM. 2009. Correlation between recombinase acti-
PK-independent pathways for both signal and coding joint vating gene 1 ubiquitin ligase activity and V(D)J recom-
formation. Proc Natl Acad Sci U S A 95:15559–15564. bination. Immunology 128:206–217.
104. Kulesza P, Lieber MR. 1998. DNA-PK is essential only 121. Yurchenko V, Xue Z, Sadofsky M. 2003. The RAG1
for coding joint formation in V(D)J recombination. Nu- N-terminal domain is an E3 ubiquitin ligase. Genes Dev
cleic Acids Res 26:3944–3948. 17:581–585.
105. Ding Q, Reddy YV, Wang W, Woods T, Douglas P, 122. Jiang H, Chang FC, Ross AE, Lee J, Nakayama K, De-
Ramsden DA, Lees-Miller SP, Meek K. 2003. Auto- siderio S. 2005. Ubiquitylation of RAG-2 by Skp2-SCF
phosphorylation of the catalytic subunit of the DNA- links destruction of the V(D)J recombinase to the cell cy-
dependent protein kinase is required for efficient end cle. Mol Cell 18:699–709.
processing during DNA double-strand break repair. Mol 123. Lee J, Desiderio S. 1999. Cyclin A/CDK2 regulates V(D)
Cell Biol 23:5836–5848. J recombination by coordinating RAG-2 accumulation
106. Wiler R, Leber R, Moore BB, VanDyk LF, Perryman and DNA repair. Immunity 11:771–781.
LE, Meek K. 1995. Equine severe combined immuno- 124. Mizuta R, Mizuta M, Araki S, Kitamura D. 2002.
deficiency: a defect in V(D)J recombination and DNA- RAG2 is down- regulated by cytoplasmic sequestra-
dependent protein kinase activity. Proc Natl Acad Sci U S A tion and ubiquitin-dependent degradation. J Biol Chem
92:11485–11489. 277:41423–41427.
107. Meek K, Kienker L, Dallas C, Wang W, Dark MJ, Venta 125. Stavnezer J, Amemiya CT. 2004. Evolution of isotype
PJ, Huie ML, Hirschhorn R, Bell T. 2001. SCID in Jack switching. Semin Immunol 16:257–275.
126. Durandy A, Revy P, Fischer A. 2004. Human models 140. Michael N, Martin TE, Nicolae D, Kim N, Padjen K,
of inherited immunoglobulin class switch recombina- Zhan P, Nguyen H, Pinkert C, Storb U. 2002. Effects of
tion and somatic hypermutation defects (hyper-IgM syn- sequence and structure on the hypermutability of immu-
dromes). Adv Immunol 82:295–330. noglobulin genes. Immunity 16:123–134.
127. Ferrari S, Plebani A. 2002. Cross-talk between CD40 141. Rogozin IB, Sredneva NE, Kolchanov NA. 1996. So-
and CD40L: lessons from primary immune deficiencies. matic hypermutagenesis in immunoglobulin genes. III.
Curr Opin Allergy Clin Immunol 2:489–494. Somatic mutations in the chicken light chain locus. Bio-
128. Calame KL, Lin KI, Tunyaplin C. 2003. Regulatory chim Biophys Acta 1306:171–178.
mechanisms that determine the development and func- 142. Dudley DD, Chaudhuri J, Bassing CH, Alt FW. 2005.
tion of plasma cells. Annu Rev Immunol 21:205–230. Mechanism and control of V(D)J recombination versus
129. Yamane A, Robbiani DF, Resch W, Bothmer A, Na- class switch recombination: similarities and differences.
kahashi H, Oliveira T, Rommel PC, Brown EJ, Nus- Adv Immunol 86:43–112.
senzweig A, Nussenzweig MC, Casellas R. 2013. RPA 143. Di Noia JM, Neuberger MS. 2007. Molecular mecha-
accumulation during class switch recombination rep- nisms of antibody somatic hypermutation. Annu Rev Bio-
resents 5′-3′ DNA- end resection during the S- G2/M chem 76:1–22.
phase of the cell cycle. Cell Rep 3:138–147. 144. Di Noia J, Neuberger MS. 2002. Altering the pathway
130. Honjo T, Kinoshita K, Muramatsu M. 2002. Molecular of immunoglobulin hypermutation by inhibiting uracil-
mechanism of class switch recombination: linkage with DNA glycosylase. Nature 419:43–48.
somatic hypermutation. Annu Rev Immunol 20:165–196. 145. Neuberger MS. 2008. Antibody diversification by so-
131. Muramatsu M, Kinoshita K, Fagarasan S, Yamada S, matic mutation: from Burnet onwards. Immunol Cell Biol
Shinkai Y, Honjo T. 2000. Class switch recombination 86:124–132.
and hypermutation require activation-induced cytidine 146. Peled JU, Kuang FL, Iglesias-Ussel MD, Roa S, Kalis
deaminase (AID), a potential RNA editing enzyme. Cell SL, Goodman MF, Scharff MD. 2008. The biochem-
102:553–563. istry of somatic hypermutation. Annu Rev Immunol
132. Muramatsu M, Nagaoka H, Shinkura R, Begum NA, 26:481–511.
Honjo T. 2007. Discovery of activation-induced cytidine 147. Odegard VH, Schatz DG. 2006. Targeting of somatic hy-
deaminase, the engraver of antibody memory. Adv Immu- permutation. Nat Rev Immunol 6:573–583.
nol 94:1–36. 148. Storb U, Shen HM, Michael N, Kim N. 2001. So-
133. Revy P, Muto T, Levy Y, Geissmann F, Plebani A, Sanal matic hypermutation of immunoglobulin and non-
O, Catalan N, Forveille M, Dufourcq-Labelouse R, immunoglobulin genes. Philos Trans R Soc Lond B Biol Sci
Gennery A, Tezcan I, Ersoy F, Kayserili H, Ugazio AG, 356:13–19.
Brousse N, Muramatsu M, Notarangelo LD, Kinoshita 149. Ollila J, Vihinen M. 2005. B cells. Int J Biochem Cell Biol
K, Honjo T, Fischer A, Durandy A. 2000. Activation- 37:518–523.
induced cytidine deaminase (AID) deficiency causes the 150. Dorner T, Foster SJ, Brezinschek HP, Lipsky PE. 1998.
autosomal recessive form of the Hyper- IgM syndrome Analysis of the targeting of the hypermutational ma-
(HIGM2). Cell 102:565–575. chinery and the impact of subsequent selection on the
134. Zahn A, Eranki AK, Patenaude AM, Methot SP, Fifield distribution of nucleotide changes in human VHDJH re-
H, Cortizas EM, Foster P, Imai K, Durandy A, Larijani arrangements. Immunol Rev 162:161–171.
M, Verdun RE, Di Noia JM. 2014. Activation induced 151. Buerstedde JM, Alinikula J, Arakawa H, McDonald JJ,
deaminase C-terminal domain links DNA breaks to end Schatz DG. 2014. Targeting of somatic hypermutation by
protection and repair during class switch recombination. immunoglobulin enhancer and enhancer-like sequences.
Proc Natl Acad Sci U S A 111:E988–E997. PLoS Biol 12:e1001831.
135. MacLennan IC. 2005. Germinal centers still hold secrets. 152. Aoufouchi S, Faili A, Zober C, D’Orlando O, Weller
Immunity 22:656–657. S, Weill JC, Reynaud CA. 2008. Proteasomal degra-
136. Imai K, Slupphaug G, Lee WI, Revy P, Nonoyama S, dation restricts the nuclear lifespan of AID. J Exp Med
Catalan N, Yel L, Forveille M, Kavli B, Krokan HE, 205:1357–1368.
Ochs HD, Fischer A, Durandy A. 2003. Human uracil- 153. Delker RK, Zhou Y, Strikoudis A, Stebbins CE, Pa-
DNA glycosylase deficiency associated with profoundly pavasiliou FN. 2013. Solubility- based genetic screen
impaired immunoglobulin class- switch recombination. identifies RING finger protein 126 as an E3 ligase for
Nat Immunol 4:1023–1028. activation-induced cytidine deaminase. Proc Natl Acad
137. Chen X, Kinoshita K, Honjo T. 2001. Variable deletion Sci U S A 110:1029–1034.
and duplication at recombination junction ends: implica- 154. Sonnenberg A, Marciniak JY, Rassenti L, Ghia EM,
tion for staggered double-strand cleavage in class-switch Skowronski EA, Manouchehri S, McCanna J, Widhopf
recombination. Proc Natl Acad Sci U S A 98:13860–13865. GF II, Kipps TJ, Heller MJ. 2014. Rapid electrokinetic
138. Wang CL, Wabl M. 2004. DNA acrobats of the Ig class isolation of cancer-related circulating cell-free DNA di-
switch. J Immunol 172:5815–5821. rectly from blood. Clin Chem 60:500–509.
139. Kaji T, Furukawa K, Ishige A, Toyokura I, Nomura M, 155. Langerak A. 2009. From the patient to the sequence:
Okada M, Takahashi Y, Shimoda M, Takemori T. 2013. selection of material, primers and PCR protol, clonality
Both mutated and unmutated memory B cells accumulate analysis, and sequencing protocol, p 23–31. In Ghia P,
mutations in the course of the secondary response and de- Rosenquist R, Davi F (ed), Immunoglobulin Gene Analysis
velop a new antibody repertoire optimally adapted to the in Chronic Lymphocytic Leukemia. Wolters Kluwer Health
secondary stimulus. Int Immunol 25:683–695. Italy Ltd., Milan, Italy.
Immunoglobulin Quantification
and Viscosity Measurement
JEFFREY S. WARREN
7
Quantification of intact serum immunoglobulins has immunoglobulin and related measurements. Developments
proven useful in the evaluation of patients with suspected since the 7th edition of this Manual include the advent of
immunodeficiency disorders, lymphocyte and plasma cell free-light-chain assays and heavy/light-chain or junctional
neoplastic diseases, allergic conditions, and some chronic epitope assays. The third section addresses fundamental
inflammatory and autoimmune disorders. Since the advent technical aspects of clinically useful viscosity measurement
of quantitative immunoglobulin assays nearly 50 years ago and important interpretive considerations.
(1), increasingly robust analytical methods have been de-
veloped. The armamentarium of intact immunoglobulin
assays was first substantively expanded by the development IMMUNOGLOBULIN STRUCTURE
of immunoglobulin light-chain measurements in which the Knowledge of immunoglobulin structure is important for un-
light chains are bound to heavy chains (intact immunoglob- derstanding measurements of intact immunoglobulins and
ulins). The last decade has seen the development of both free light chains as well as in clinical viscometry. Basic an-
free (unbound) immunoglobulin light- chain assays and tibody function (i.e., antigen binding) was recognized years
heavy/light-chain (HLC) or junctional epitope assays, the before the elucidation of immunoglobulin structure (6). Early
latter of which allows for individual measurements of IgGκ, studies of antibody binding to highly purified carbohydrate
IgGλ, IgAκ, IgAλ, IgMκ, and IgMλ (2, 3). Despite these antigens led to the deduction that antibody was composed
advances in immunoglobulin and light-chain quantification of protein (6). The development of clinical electrophoresis
methods, there remain technical complexities that can in- led to the recognition that antibody migrated largely within
fluence accuracy and, in turn, proper clinical application. the so-called gamma region of serum proteins, hence the
Equally important, even with robust assays, is the need for names “gamma globulin” and “immunoglobulin” (Fig. 1)
thorough understanding of the clinical indications for, and (7). In 1952, Colonel Ogden Bruton, a pediatrician, used
limitations to, immunoglobulin and related measurements. serum protein electrophoresis (SPEP) to evaluate a boy suf-
Viscosity is the internal friction of a fluid, caused by in- fering from recurrent bacterial infections (8). Recognition
termolecular attraction, leading to resistance of the fluid that the gamma fraction of protein was absent, coupled with
to flow (4). Viscosity measurements can be very complex the boy’s favorable response to injections of gamma globu-
(4). By extension, rigorous whole-blood, plasma, and serum lin from healthy donors, supported the conclusion that this
measurements are also very complex. Fortunately, clinically serum fraction was important in host defense and led to the
useful measurements of viscosity are relatively straightfor- clinical definition of a specific immunodeficiency syndrome
ward. A variety of clinical conditions can affect the viscosity (Bruton’s agammaglobulinemia). Recognition of monoclonal
of blood (5). While abnormalities of the formed elements of immunoglobulin (also known as M-protein or “paraprotein”)
blood (e.g., polycythemia, some hemoglobinopathies, and as the structurally uniform product of a clonal proliferation
extreme leukocytosis) can result in clinically significant hy- of neoplastic B lymphocytes or plasma cells was important in
perviscosity, overwhelmingly the most common causes are the recognition and understanding of such diseases as mul-
either immunoglobulin concentration or abnormal immu- tiple myeloma and Waldenström’s macroglobulinemia.
noglobulin structure. There are several analytical methods In contrast to serum alpha and beta globulins, which
for measurement of whole-blood, plasma, and serum viscos- are composites of structurally homogeneous proteins (e.g.,
ity. It is important to understand the uses and limitations of alpha-1 antiproteinase, transferrin, and complement protein
these measurements and how each is reported. 3), polyclonal immunoglobulins are structurally heteroge-
This chapter is divided into three sections. The first pro- neous (9). This structural heterogeneity is accounted for at
vides a brief review of immunoglobulin structure, important several levels. There are five major immunoglobulin classes
because of its direct relevance to quantitative immuno- (isotypes), each determined by heavy-chain type (see be-
globulin and related measurements. The second provides low) and each different in structure and function (Table 1).
an overview of assay methods and highlights issues related Subclasses within immunoglobulin G (IgG), IgA, and IgM
to quality control, test validation, and quality assurance, (Table 2) also contribute to immunoglobulin heterogeneity,
as well as a brief discussion of the clinical application of as do allotypic variations within both heavy-and light-chain
doi:10.1128/9781555818722.ch7
65
TABLE 2 Characteristics of IgG subclasses

Property IgG1 IgG2 IgG3 IgG4
Concn in serum 1.8–7.8 1.0–4.6 0.3–1.4 0.08–1.8
(g/liter)
Half-life (days) 14–23 14–23 7–8 14–23
Complement fixation Strong Weak Strong None
Phagocyte binding Yes No Yes No
(via Fc receptors)
Associated with No No No Yes
allergies
FIGURE 1 Serum protein electrophoresis: densitometric display
of serum proteins separated by agarose gel electrophoresis using a
pH 8.6 buffer. Capillary zone electrophoresis is liquid based; thus, fragments (Fab) and one fragment that could be crystallized
no permanent gel is produced. (Fc). Fab fragments had sedimentation coefficients of 3.5 S
(Svedberg units). That the Fc fragment could be crystallized
provided evidence that it was biochemically homogeneous.
isotypes. Immunoglobulins are also subject to posttransla- Digestion of purified antibody with pepsin yielded a differ-
tional modification. Overwhelmingly, however, the great ent set of fragments. The largest, with a sedimentation coef-
heterogeneity of structure reflects the vast array of different ficient of 5 S, had an antigen-binding valence of 2, hence its
amino acid sequences within the variable and hypervariable designation, F(ab′)2. Reduction of the disulfide bonds that
regions of immunoglobulin molecules. One of the pivotal hold an F(ab′)2 fragment intact resulted in the formation of
scientific advances in the history of immunology was the two fragments that each resembled (but were not identical
elucidation of the molecular mechanisms responsible for the to) papain-generated Fab. In 1961, Edelman and Poulik re-
generation of antibody (immunoglobulin) diversity (chapter ported that chemical reduction (of disulfide bonds in intact
6) (10). It is important to remember these different levels as immunoglobulin molecules) leads to the formation of two
contributors to the structural heterogeneity of immunoglobu- “heavy” chains and two “light” chains (12).
lins because they warrant consideration in both the quantifi- As noted above, the structurally variable (and hyper-
cation of polyclonal immunoglobulins by class (e.g., total IgG variable) Fab domains of immunoglobulin molecules are
or IgM) and in the quantification of monoclonal immuno- responsible for antigen binding. In contrast, the Fc domain
globulins (e.g., in patients with multiple myeloma). The gen- determines the biological function of a given immuno-
eral structure of an intact immunoglobulin molecule includes globulin molecule. As summarized in Table 1, the biolog-
two heavy chains (gamma, alpha, mu, delta, or epsilon), two ical functions of immunoglobulins are diverse, including
light chains (kappa or lambda), and several bridging disul- such characteristics as complement fixation (e.g., IgG and
fide bonds (Fig. 2). Within each heavy and light chain are IgM), placental transfer (IgG), and high-affinity cytophilic
N-terminal variable and hypervariable (antigen-binding) do- binding to mast cells (IgE) (13). IgM generally circulates
mains and C-terminal constant domains. Antibody structure as a pentamer that consists of five covalently linked IgM
is best understood in the context of the genetic and molecular monomers with an antigen-binding valence of 10. Serum
bases of structural and functional diversity (chapter 6). IgA occurs as a monomer that weighs nearly 160 Da with a
Pioneering studies of antibody structure, carried out in valence of 2. In secretions, IgA commonly occurs as a mul-
the 1950s and 1960s, provided important insight into anti- timer, usually a dimer.
body function and resulted in the immunoglobulin subunit Light chains (kappa and lambda) each possess an N-
nomenclature still in use. In 1958, Porter (11) digested rabbit terminal 110- amino-acid variable (V) region and a C-
gamma globulin with the enzyme papain. Separation of the terminal 107-to 110- amino-acid constant (C) domain.
digestion products revealed two identical antigen-binding By analogy, heavy chains (gamma, alpha, mu, epsilon, and
TABLE 1 Characteristics of immunoglobulinsa

Nomenclature IgG IgA IgM IgD IgE
Heavy-chain class Gamma Alpha Mu Delta Epsilon
Heavy-chain subclasses 1, 2, 3, 4 1, 2 1, 2
Light-chain types Kappa and lambda Kappa and lambda Kappa and lambda Kappa and lambda Kappa and lambda
Physical characteristics
Molecular mass (Da) 143,000–160,000 159,000–447,000 900,000 177,000–185,000 187,000–200,000
Sedimentation 6.7–7.0 7.5–9.0 18–19 6.9–7.0 7.9–8.0
coefficient (Svedberg
units)
Functional characteristics
Serum half-life (days) 7–23 5–6 5 2–8 1–5
Complement fixation + − ++ − −
Placental transfer + − − − −
Reaginic activity ± − − − ++
a
±, borderline; +, present; ++, strong activity; −, no activity.
7. Immunoglobulin Quantification and Viscosity Measurement ■ 67
IgA is the most important immunoglobulin in the mu-

cosal host defense system. Mucosal IgA exists chiefly as
paired monomeric subunits linked by J (“joining”) chains
to form dimers (13). A 60-kDa peptide called the “secretory
piece” is necessary for IgA to be secreted by various types
of epithelial cells that line mucosal surfaces. IgA, like all
immunoglobulins, is produced by B lymphocytes and plasma
cells. Mucosal IgA is actively transported across mucosal ep-
ithelial layers (from the abluminal to the luminal surface).
As noted in Table 1, IgA exists as IgA1 and IgA2 subclasses,
the former present in higher concentration in serum and
each having a half-life of 4 to 5 days.
Serum IgM circulates chiefly in the form of pentamers.
Monomeric subunits include two mu heavy chains, which
in turn each possess one more constant-region domain than
most gamma heavy-chain species. As a result, pentameric
IgM has a molecular mass of approximately 900 kDa (J
chain plus 5 times nearly 180 kDa). IgM is the initial im-
munoglobulin class expressed on B cells during lymphocyte
FIGURE 2 Schematic of an intact immunoglobulin G molecule. development and is the predominant immunoglobulin class
Antigen binding occurs between light and heavy chains in the N- in a primary humoral response (13).
terminal variable-hypervariable binding regions. IgG possesses an IgD circulates in very low concentrations. Like IgM, B
antigen-binding valence of 2. Monomeric immunoglobulin mole- cell surface IgD is expressed early in lymphocyte develop-
cules, whether the heavy chains are gamma, alpha, mu, delta, or ment (13). Monomeric IgD consists of two heavy chains
epsilon, possess the same general structure. Individual immunoglob- (each weighs approximately 60 kDa) and two identical
ulin molecules include either kappa or lambda light chains, not both. kappa or lambda light chains. The biological role of serum
IgD is not well understood, and clinical measurement of
IgD has no value except for the rare patient with an IgD-
delta) each possess a 110-amino-acid N-terminal variable secreting neoplasm (e.g., multiple myeloma) or the equally
region and, depending on the particular class, three or four rare patient with a familial hyper-IgD fever syndrome.
110-amino-acid constant-region domains. Individual vari- The clinical utility of IgE measurements is discussed in
able-or constant-region domains are folded into globular greater detail in section M, which describes allergic diseases.
motifs called “Ig domains.” Ig domains are held in conforma- Serum IgE molecules circulate as monomers that consist
tion by disulfide bonds (Fig. 2). Structurally apposed light- of two epsilon chains (each approximately 70 kDa) and
and heavy-chain variable regions form an antigen-binding two identical kappa or lambda light chains (14). Like mu
domain. It is useful to think of the antigen-binding (Fab) heavy chains, epsilon heavy chains contain an additional
region of an immunoglobulin molecule as a hand that grasps constant-region domain. While IgE is normally present in
a uniquely shaped doorknob (antigen). Finally, as alluded to low concentrations, elevated polyclonal IgE immunoglob-
above, the genetic and molecular bases for immunoglobulin ulin is seen in a wide variety of allergic and autoimmune
structure, antibody diversity, and immunoglobulin assembly diseases. Antigen-specific IgE measurements are widely used
and secretion have been studied intensively (chapter 6) (10, in the evaluation and management of patients with specific
13). Kappa light chains are encoded by chromosomal band atopic disorders (14). For instance, one may quantitatively
2p11, lambda light chains are encoded by 22q11, and the measure IgE antipollen or IgE antipeanut antigen. These
heavy chain is encoded by the loci on chromosome 14. A measurements require very sensitive analytical immuno-
highly regulated series of DNA recombination, nucleic acid assay systems that rely on detection of specific binding of
splicing, and subunit assembly steps is required to generate IgE immunoglobulin molecules to their defined cognate
intact immunoglobulin molecules (10, 13). Immunoglobu- antigens. Cytophilic IgE binds to mast cells and basophils
lins are produced only by B lymphocytes and plasma cells. via very-high-affinity Fcε receptors. Activation of such cells
via divalent IgE-specific antigen binding results in the rapid
release of vasoactive and chemotactic mediators that can
IMMUNOGLOBULIN CLASSES cause localized swelling and/or a generalized, sometimes
IgG is the predominant class of serum immunoglobulins. life-threatening reaction (anaphylaxis) (15). Measurement
More than 60% of circulating IgG is IgG1, with IgG2, IgG3, of total IgE and antigen-specific IgE antibodies is discussed
and IgG4, in that order, present in decreasing concentra- in section M.
tions (Table 2) (13, 14). Except for IgG3, which has a serum
half-life of 1 week, the IgG subclasses have half-lives of 2 to
3 weeks. IgG antibodies comprise the most important effec- MEASUREMENT OF IMMUNOGLOBULINS
tor class of molecules in a secondary or anamnestic humoral Currently, quantification of intact IgG, IgA, and IgM is al-
immune response. Among large populations of B lympho- most exclusively performed by automated rate nephelomet-
cytes that express different cell surface IgG molecules, high- ric or immunoturbidimetric assay systems (16). Application
affinity antigen binding provides a selective advantage for of nephelometry and immunoturbidimetry to immunoglob-
subsequent clonal expansion (13). As a result, as a humoral ulin measurement represents a trend away from more labor-
immune response “matures” over time, increasing propor- intensive, less sensitive, and slower manual methods such
tions of IgG antibodies exhibit higher degrees of antigen- as radial immunodiffusion (RID) and the Laurell rocket
binding affinity (affinity maturation). This is in part why technique (17, 18). Likewise, most serum-bound kappa and
repetitive exposures to a given antigen result in increasingly lambda light-chain measurements and total IgE measure-
effective humoral immunity. ments are also carried out using automated nephelometric
or immunoturbidimetric assay systems (16). More than 20% unknown analyte, in this case the serum IgG subclass, can be
of clinical laboratories that participated in the 2004 College calculated by comparing the diameter of the precipitin ring
of American Pathologists (CAP) proficiency testing survey to those precipitin rings of the standard curve generated by
program used a RID method for IgG subclass measurements, the serially diluted known concentrations of antigen added to
and nearly all laboratories that measured total IgD employed different wells. Higher concentrations of antigen must diffuse
a RID assay (17, 18). Currently, nearly all laboratories (that further to reach concentrations low enough to form antigen-
participate in the CAP proficiency testing program) utilize antibody complexes at equivalence. Typically, a 5 log10 di-
nephelometric or immunoturbidimetric assays (16). The ex- lution standard curve is employed (1, 22). The diameters
ception is for total IgE, which is measured by nonisotopic measured from precipitin rings formed by patient samples can
immunoassay. be interpolated from the set of standard curve diameters. RID
Nephelometric assays are based on rate reactions in offers the advantages that it is simple and requires very little
which antigen, in this case an immunoglobulin such as IgG, equipment. The major disadvantages are that RID assays are
is injected into a reaction chamber with an antigen-specific manual, take 16 to 48 hours to run because of the diffusion
antibody (19). As antigen-antibody complexes form in sus- time required, and often employ very expensive reagents.
pension, they form a precipitin “cloud.” (“Nephos” is derived Technical artifacts attributable to lipids, monomeric IgM or
from the Greek word for cloud.) Antigen-antibody com- IgA, or cross-reacting antibodies to animal proteins can lead
plexes are detected with an intense light source. As immune to inaccurate results. When reagent antibody is monoclonal,
complexes form, increasing numbers of photons are deflected as in many IgG subclass assays, the per-test cost can be ex-
at 30° to 90° angles. They are then quantified with a photo- traordinarily high. One decade ago, when more than 20% of
multiplier tube located on the inner side surface of the re- laboratories that participated in the CAP proficiency testing
action chamber. As the antigen-antibody precipitin “cloud” program used RID assays for IgG subclass measurement (17,
grows in size and density, more photons are deflected and a 18), a single set of RID plates (enough to test nine patients)
greater signal is recorded. The rise in the number of light sig- cost $400 (J. Warren, unpublished observation)! Inherent in
nal events (photons) as a function of time (in milliseconds) RID immunoglobulin assays are higher coefficients of vari-
is determined by the concentrations of antigen and antibody. ation than those observed for nephelometric and immuno-
Modern rate nephelometers are sophisticated instruments turbidimetric methods (greater than 10% versus less than or
that employ parallel duplicate assay chambers, thus ensuring equal to 8%) (16–18). Because of high relative expense and
precision; they exhibit high throughput and are programmed long analytical times, the use of RID assays in clinical labo-
to run through a series of dilutions of antigen or antibody, ratories has decreased dramatically during the past 2 decades
thus allowing for the “finding of ” a near-optimal precipitin- (16).
forming rate reaction. While rate nephelometers are robust, The Laurell rocket technique is very similar to RID,
wildly aberrant results can occur as the result of extreme an- except that the diffusion phase of the assay is accelerated
tigen excess (the high-dose “hook” effect) (20). This problem from hours to minutes by the application of an electrical
may be particularly pronounced when measuring an unusu- field (23). While the endpoint is still the formation of a
ally high concentration of an analyte (outside the expected precipitin line (antigen-antibody complexes at equivalence
dynamic range of the assay) otherwise expected to have a in agar), the geometry is changed from a circle to that of
relatively low concentration. Such antigen excess can escape a rocket-shaped arc, hence the designation “rocket.” With
detection via antigen excess checks (a series of repeated mea- the Laurell technique, the height of the precipitin rocket is
surements with more-dilute antigen or antibody solutions). proportional to the antigen concentration. As in the RID
When measuring IgG subclasses, it is a useful quality check to technique, a standard curve is constructed by running a se-
sum the concentrations of all four subclasses to ascertain that ries of different known concentrations of a standard anti-
their additive total approximates the separately measured gen preparation. Again, immunoglobulin concentrations in
total IgG concentration. Immunoturbidimetric assays are patient samples are calculated based on interpolation from
formatted similarly, but rather than relying upon the quanti- the standard curve. While Laurell rocket electrophoresis is
tation of deflected light, they rely on the interruption of light faster than RID, the remaining shortcomings still hold, and
transmission (180°) through the suspension of immune com- use in clinical laboratories is very uncommon (16). Both
plexes being formed by antigen and reagent antibody (21). RID and Laurell rocket assay systems may be most appropri-
Nephelometric and immunoturbidimetric test systems ate for very-low-volume laboratories, where a nephelometer
have had a large economic impact on clinical laboratories may be too expensive.
because they are automated, can be used to measure many
different analytes with a single platform, and allow rapid
and high-volume throughput. As a result, employment of MEASUREMENT OF FREE LIGHT CHAINS
these methods yields a lower cost per test in laboratories In the past decade, an analytical approach to quantification
that process significant volumes. of free (dissociated from heavy chain) immunoglobulin light
RID assays entail addition of reagent antibody to warm chains in serum and/or urine has been developed, and the
(50°C) agar while it is in the liquid state (1, 22). Liquefied assay has been tested in a variety of clinical settings. Mature
antibody-containing agar is poured into a shallow, flat plastic B lymphocytes and plasma cells typically produce approx-
dish or plate and allowed to cool to room temperature. The imately 40% excess of free light chains (13). Thus, early
resulting semisolid agar has the consistency of cold Jell-O. after the emergence of a single (disordered or neoplastic)
Cylindrical vertical wells are then cut into the solidified agar. clone, the “signal-to-noise” ratio of monoclonal light chains
Standard dilutions of known concentrations of antigen, in to “background” polyclonal light chains is low—making
this case, immunoglobulin such as a specific IgG subclass, are monoclonal populations difficult to detect. Identification
placed into multiple wells. Serum samples that contain pa- of free monoclonal light chains in the sera of patients with
tient IgG to be measured are placed in other wells. As the an- multiple myeloma, other B cell disorders, and light-chain
tigen (IgG subclass) diffuses in all directions from each well, a (amyloid light-chain [AL])-type amyloidosis is also limited,
visible antigen-antibody precipitin line forms where antigen especially early in the course of the disease, because these
and antibody achieve equivalence. The concentration of the proteins, by virtue of their low molecular mass (22 to 25
kDa), are filtered by glomeruli and subsequently reabsorbed cell leukemias and lymphomas, and AL-type amyloidosis is
by proximal renal tubular epithelial cells. Accordingly, an exciting prospect. As in other nephelometric and im-
SPEP may reveal no apparent abnormality. The diagnostic munoturbidimetric immunoglobulin assays, sFLC measure-
yield is increased by also testing urine for such Bence Jones ments are prone to the high-dose “hook” effect (19, 20).
proteins (monoclonal urinary free light chains). Again, low This is of particular concern in measurements of serum free
concentrations of Bence Jones protein may not be seen early light chains in patients with B lymphocyte and plasma cell
in disease because proximal tubular reabsorption of such neoplasms because of the extreme variation in concentra-
clonal proteins is relatively efficient until the quantities of tions that can be seen in a single patient over time. There
protein increase and exceed the reabsorptive capacity of the are several analytical limitations to sFLC assays. These in-
tubular epithelium or until tubular damage (acquired Fan- clude lot-to-lot variability and the fact that some sFLCs do
coni syndrome) occurs and there is increased excretion of not dilute in a linear fashion (27). It will be important for
Bence Jones protein. The advent of the nephelometric se- additional clinical studies to address the specific potential
rum free-light-chain (sFLC) assay has increased the ability applications and limitations of these assays. Particular at-
to detect monoclonal proteins (43) (see below). tention will need to be paid to comparisons between free-
In 2001, Bradwell et al. developed a sensitive immunoas- light-chain assays and optimally used “traditional” assays.
say for the quantification of free immunoglobulin light chains For example, it will be appropriate to compare diagnostic
in both urine and serum (2). Affinity-purified antibodies that sensitivities and specificities of free-chain-assays to those
react specifically with epitopes accessible only on free light of serum and urine immunofixation assays, not merely se-
chains (Fig. 3), but unable to reach the same epitopes seques- rum or urine immunofixation assays. In addition, it will be
tered in bound light chains, are linked to the surface of latex important to assess the impact of sensitive free-light-chain
particles. These antibody-coated particles form aggregates assays on clinical outcomes in patients. Serum and urine
when incubated with free light chains. Employment of such free-light-chain assays are Food and Drug Administration
particles in automated nephelometric or immunoturbidimet- cleared for use in the diagnosis and monitoring of patients
ric analyzers has resulted in a robust test system. Because free with B cell and plasma cell neoplasms and collagen vascular
kappa and lambda (polyclonal) light chains are normally also diseases such as systemic lupus erythematosus (25).
present in serum and urine (and measured), the clinically Many important utilization and cost-effectiveness ques-
useful assay for detection of monoclonality necessitates cal- tions remain to be answered for these expensive tests. For
culation of the ratio of free kappa to free lambda light chain. instance, what are appropriate testing frequencies in the con-
The occurrence of a significant clonal increase in either free text of various disease processes and therapies? How should
kappa or free lambda light chains distorts this ratio. Thus, a sFLC assays be utilized in the context of other laboratory test-
very high or a very low ratio of kappa to lambda free light ing? Any studies of cost-effectiveness of sFLC testing versus
chains is reflective of a kappa monoclonal or lambda mono- “traditional approaches” (SPEP, 24-hour Bence Jones protein
clonal process, respectively. Individual sFLC measurements, quantification, etc.) should take into account not only the
kappa or lambda, reflect the total circulating respective free- cost of “traditional” testing but also inconvenience (e.g.,
light-chain concentration. 24-hour urine collection) and technical limitations (e.g.,
Clinical studies since the widespread deployment of quantification of urine Bence Jones protein and M-protein
sFLC assays have led to a better understanding of when and quantification by SPEP). Clearly, assessments of optimal uti-
how to utilize this test. Much of what has been learned was lization and cost-effectiveness encompass a broad set of ques-
distilled into a set of practice guidelines published by the tions and consideration of many factors.
International Myeloma Working Group in 2009 (see be-
low) and has since been reiterated, refined, and extended in
more-recent studies (24–26). JUNCTIONAL EPITOPE (HEAVY/LIGHT-
The addition of sensitive and specific free-light-chain as- CHAIN) ASSAYS
says to the armamentarium of laboratory tests available for In 2009, Bradwell et al. described the development of an as-
the diagnosis and management of patients suspected to have say based on reagent antisera that specifically recognize and
multiple myeloma, Waldenström’s macroglobulinemia, B bind to unique epitopes displayed at the junction between
FIGURE 3 The serum free-light-chain (sFLC) assay is based on purified polyclonal antibodies that
specifically bind epitopes that are accessible on free light chains but hidden (and inaccessible) in intact
immunoglobulin molecules.
FIGURE 4 The heavy-light chain (HLC) assay, a junctional epitope assay, is based on purified poly-
clonal antibodies that specifically bind unique epitopes that occur where heavy and light chains are struc-
turally apposed. Antibodies specifically bind IgGκ, IgGλ, IgAκ, IgAλ, IgMκ, and IgMλ (six reagents).
the heavy-and light-chain constant regions of IgGκ, IgGλ, immunodeficiency all exhibit markedly subnormal concen-
IgAκ, IgAλ, IgMκ, and IgMλ molecules (six reagents), trations of IgG, IgA, and IgM in serum. The most common
hence the name “junctional epitope” or so-called “heavy/ congenital humoral immunodeficiency disorder is selective
light” chain (HLC) assays (Fig. 4) (3). The quantitation of IgA deficiency, which ranges, depending on the ethnicity of
immunoglobulins by isotype-specific light chain type (e.g., the population, from 1 in 400 to 1 in 700 adults (29). Many
IgGκ and IgGλ) may provide diagnostically useful infor- patients with selective IgA deficiency are asymptomatic, but
mation related to clonal proliferation, and HLC-pair ra- some suffer from recurrent mucosal infections and/or auto-
tios (e.g., IgGκ/IgGλ) coupled with individual heavy/light immune phenomena. A subset of selective IgA deficiency
chain measurements (e.g., IgGκ and IgGλ) may be useful patients exhibit concomitant IgG subclass deficiency. Ag-
in disease monitoring (28). As noted with sFLC assays, gressive treatment of patients with various neoplastic and
demonstration of cost-effectiveness and impact on clinical autoimmune diseases has resulted in many individuals with
outcomes in patients will be important considerations. acquired humoral immunodeficiencies. Like patients with
congenital humoral immunodeficiency disorders, these
individuals are at risk of high-grade bacterial and other
CLINICAL ASPECTS OF IMMUNOGLOBULIN infections. Likewise, multiple myeloma is frequently accom-
MEASUREMENT panied by both the presence of a monoclonal immunoglob-
Serum immunoglobulin concentrations vary widely be- ulin spike (M-protein) and the suppression of polyclonal
tween children and adults (13, 29). Immunoglobulin G “background” gamma globulin (see below).
concentrations in a newborn are nearly equal to those ob- A vast array of chronic inflammatory and infectious dis-
served in adults as a result of transplacental passage from the eases are accompanied by polyclonal increases in immuno-
mother. (Maternal IgA and IgM do not pass into the fetal globulin concentration. In these settings, there is usually
circulation.) Immunoglobulin G concentrations then de- little to no value in specific quantitative measurements of
cline to a nadir at approximately 6 months of age. After this IgG, IgA, or IgM. Serum protein electrophoresis, accom-
nadir in the IgG concentration, it rises progressively until panied by densitometric measurement of the total gamma
adult concentrations are reached during early adolescence. globulin concentration, usually suffices as a “first- tier”
It is important to apply age-specific reference ranges when means to assess the humoral immune system.
one interprets quantitative immunoglobulin measurements A substantial majority of patients with multiple my-
in prepubertal children. Likewise, it is important to recog- eloma and essentially all patients with Waldenström’s
nize that there is some individual-to-individual variation in macroglobulinemia have monoclonal increases in immu-
this pattern. noglobulins in serum, urine, or both (30). Serum protein
Many congenital and acquired immunodeficiency dis- electrophoresis and both serum and urine immunofixation
orders are characterized by hypogammaglobulinemia (29). electrophoresis (IFE) remain key assays used in the diagnosis
Depending upon the underlying etiology and accompanying and monitoring of disease progression/response to therapy
abnormalities, humoral immunodeficiency syndromes are in patients with multiple myeloma and related plasma cell/B
often accompanied by specific clinical manifestations. In lymphocyte proliferative diseases (30). Increasingly, clini-
general, patients with humoral immunoglobulin deficien- cal laboratories have shifted from agarose gel to capillary
cies are at risk for a variety of infections, including some zone electrophoresis as the primary method for SPEP (31).
caused by high-grade encapsulated bacteria such as Strepto- Secreted monoclonal immunoglobulins (M- proteins) or
coccus pneumoniae, Haemophilus influenzae, etc. (29). A de- immunoglobulin components (e.g., free light chains) are de
tailed discussion of humoral immunodeficiency syndromes facto tumor markers. Major limitations of SPEP and IFE in
is beyond the scope of this chapter, but many such disorders the evaluation of patients with myeloma and other plasma
have been elucidated and are currently best classified based cell disorders have been lack of analytical sensitivity in oli-
on identification of specific molecular defects. For example, gosecretory lesions and the unwieldiness of 24-hour urine
children suspected of suffering from Bruton’s (X- linked) collections. “Traditional” SPEP and IFE studies are particu-
agammaglobulinemia or a severe combined immunodefi- larly insensitive in the evaluation of patients with AL am-
ciency and adults suffering from suspected common variable yloidosis (32).
Studies conducted since the development of the quan- apposed sample cup/plate and the rotating needle/cone.)
titative kappa and lambda sFLC assays have led to the cre- The greater the viscosity of the fluid, the greater the torque
ation of expert guidelines for the clinical applications of generated. In turn, the torque is measured with a torque me-
these tests (24). The combination of the sFLC assays, SPEP, ter and translated into “viscosity.” Falling-drop viscometry
and serum IFE is a highly sensitive screen for myeloma and exploits the fact that particulates fall more slowly through a
related plasma/B cell proliferative diseases and, except in viscous fluid than through a less viscous fluid.
the case of suspected AL amyloidosis, negates the need for The viscosity of biological samples is typically reported
24-hour urine studies (32–38). (Tissue biopsy is still required in SI (International System of Units) unit centipoise (cP);
for definitive diagnosis of all of these entities.) Baseline normal plasma viscosity is approximately 1.35 to 1.85 cP.
sFLC measurements offer prognostic value in essentially all (Water exhibits a viscosity of 1.0 cP at 37°C). Whole-blood
plasma cell disorders (including monoclonal gammopathy viscosity can also be expressed as “equivalent hematocrit of
of undetermined significance [MGUS]) (39–42). The sFLC whole-blood viscosity,” which is the viscosity equivalent to
assay provides utility in quantitative monitoring of patients that exhibited by anticoagulated whole blood at the given
with “oligosecretory” plasma cell diseases including oligose- hematocrit. Clinical interpretation of “equivalent hemato-
cretory myeloma, nonsecretory myeloma (when there is an crit” requires an actual hematocrit measurement for com-
abnormal sFLC kappa or lambda concentration), and light- parison. For example, a patient with hyperviscosity due to a
chain (AL) amyloidosis. (Elevations of sFLC are not ob- high concentration of a monoclonal paraprotein might ex-
served in all patients with nonsecretory myeloma.) Several hibit a hematocrit equivalent of 55% and an actual hemato-
studies support the use of absolute sFLC measurements (i.e., crit of 40%. In this case, the “excess” viscosity is attributable
free kappa and/or free lambda) in monitoring with AL am- to the paraprotein. Conversely, a patient with polycythemia
yloidosis, oligosecretory myeloma, nonsecretory myeloma and hyperviscosity might exhibit a hematocrit equivalent of
(applicable cases), and solitary plasmacytoma (24). Cur- whole-blood viscosity of 55% and an actual hematocrit of
rently, as emphasized by the 2009 International Myeloma 55%. In this case, the elevated viscosity is wholly attribut-
Working Group guidelines, the data do not justify the use of able to the increase in red blood cell mass. Tables or figures
serial sFLC measurements rather than serial measurements of that display the mathematical relationship between hemat
otherwise measurable M-proteins by SPEP, serum IFE, and/or ocrit and viscosity of whole blood are useful in clinical
quantitative 24-hour urine Bence Jones protein quantifica- practice (4), particularly given the nonlinear relationship
tion studies in patients with multiple myeloma (24). between hematocrit and whole-blood viscosity (see below).
Katzmann and Rajkumar recently summarized a series
of studies that addressed the clinical utility of junctional
epitope-HLC assays (28). First, Bradwell et al. (44) and Lud- CLINICAL ASPECTS OF VISCOSITY
wig et al. (45) provided data that suggest that the diagnostic MEASUREMENT
sensitivity of HLC-pair ratios (e.g., IgGκ/IgGλ) in multiple The most common cause of hyperviscosity is the presence of
myeloma is similar to that of SPEP plus serum IFE. These a paraprotein. More than 80% of cases of hyperviscosity are
investigators also observed that highly abnormal HLC-pair due to an IgM paraprotein associated with Waldenström’s
ratios were prognostic (44, 45). Finally, Katzmann et al. macroglobulinemia. IgM molecules possess a high molecular
(46) used the junctional epitope assay to reveal that heavy/ mass (pentamers are approximately 900 kDa) as well as a
light-chain-pair suppression (e.g., suppression of IgGλ in shape that increases their intrinsic viscosity. IgA parapro-
patients with IgGκ MGUS) is an independent predictor of teins, less commonly IgG paraproteins, and rarely, very high
disease progression (MGUS to multiple myeloma). The last concentrations of polymerized or aggregated monoclonal
observation may refine risk assessment for MGUS patients light chains can also result in hyperviscosity (5). Increases
beyond that available by simple estimation of serum M- in red blood cell mass (e.g., polycythemia), reduced deform-
protein concentration (30). ability of red blood cells (e.g., hemoglobinopathies), and
very high white blood cell counts (e.g., chronic lympho-
cytic leukemia) can also increase whole-blood viscosity. It
VISCOSITY MEASUREMENT is important to recognize that the arithmetic relationship
Viscosity is the intrinsic resistance of a fluid to flow. Ab- between hematocrit and whole-blood viscosity is nonlinear
normalities of the formed elements of blood (e.g., increased (5). There is a steep upward inflection in viscosity near a he-
red blood cell mass, decreased red blood cell deformability, matocrit of 60%. As a result, a patient may be asymptomatic
and changes in numbers and/or properties of white blood with a hematocrit of 55% and dramatically affected at 60%.
cells) and abnormalities of plasma proteins (e.g., increased As alluded to above, hyperviscosity attributable to a para-
immunoglobulin concentration) can cause hyperviscosity. protein increases both whole-blood and plasma (or serum)
While emerging technologies such as laser Doppler veloci- viscosity, while abnormalities of formed elements affect
metry and mass-detecting sensors can be used in clinical vis- whole-blood viscosity but not plasma (or serum) viscosity.
cometry (47), most laboratories continue to employ either Symptoms and signs of hyperviscosity are primarily rhe-
Ostwald, Wells-Brookfield, or falling-drop viscometers. ologic but quite varied. They include fatigue, blurred vision,
Ostwald viscometers are based on the relationship be- headache, tinnitus, decreased hearing, vertigo, paresthesias,
tween fluid viscosity and the rate of flow through a fixed wall segmental dilation of retinal veins, retinal hemorrhage, mu-
narrow-bore tube. The greater the viscosity of either antico- cosal bleeding, nystagmus, ataxia, somnolence, stupor, and
agulated whole blood, plasma, or serum, the lower the rate coma (5). Individual patients exhibit different thresholds and
of flow. Typically, the flow rate of the sample of interest is findings, but symptoms and signs of hyperviscosity “typically”
compared to a standard (e.g., water). The Wells-Brookfield occur when whole-blood viscosity exceeds 4 cP (2 to 3 times
viscometer employs a stationary plate (sample cup) and a the normal level). For example, an individual with advanced
cone (needle) that is rotated within the cup by a motor run atherosclerosis involving the cerebral vasculature might ex-
at a constant speed under standard temperature conditions hibit neurologic symptoms and signs of hyperviscosity with
(37°C). The fluid sample of interest is placed into the cup a plasma viscosity of 3 cP, while an otherwise healthy per-
in which the cone rotates. (The fluid spreads between the son (with patent and distensible vessels) might not become

Number 1 Manual of Molecular and Clinical Laboratory Immunology (PDFDrive)

Uploaded by

Copyright:

Available Formats

Number 1 Manual of Molecular and Clinical Laboratory Immunology (PDFDrive)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Number 1 Manual of Molecular and Clinical Laboratory Immunology (PDFDrive)

Uploaded by

Copyright:

Available Formats

What are some of the causes of increased blood viscosity?

What are some of the causes of increased blood viscosity?

What are some common symptoms of hyperviscosity?

What are some common symptoms of hyperviscosity?

EIGHTH EDITION

Library of Congress Cataloging-­­in-­­Publication Data

Names: Detrick, Barbara, editor. | Schmitz, John L. (John Leo), editor. |

Address editorial correspondence to: ASM Press, 1752 N St., N.W.,

Editorial Board / xi 6 Immunoglobulin Genes / 51

ANDREW D. LUSTER 48 Immunological Tests in Tuberculosis / 433

61 Human Herpesviruses 6, 7, and 8 / 578 78 Neutropenia and Neutrophil Defects / 765

Systemic Lupus Erythematosus / 868 101 Detection of Antimitochondrial

SECTION EDITORS: ELAINE F. REED AND THANGAMANI MUTHUKUMAR,

Subject Index / 1195

C. LYNNE BUREK (section N) HOWARD M. LEDERMAN (section K)

PATRIZIO CATUREGLI (section N) CHRISTINE M. LITWIN (section H)

DAVID F. KEREN (section B)

ROSHINI SARAH ABRAHAM GIUSEPPE BARBESINO

MARIA E. AGUERO-ROSENFELD DAVID R. BARNIDGE

CEM AKIN LEE ANN BAXTER-LOWE

C. LYNNE BUREK SARAH COOLEY

EDWARD K. L. CHAN WILLIAM MARCIEL DE SOUZA

ANITA CHANDRA LUDOVICA DE VINCENTIIS

STEVEN D. DOUGLAS DENNIS GALANAKIS

JAMIE DUKE EMANUELA M. GHIA

BRUCE E. DUNN PATRICIA C. GICLAS

GEORGE S. EISENBARTH KIMBERLY C. GILMOUR

MARTIN FLEISHER PAMELA A. GUERRERIO

NEIL HARRIS STEVEN JACOBSON

ANDREA ILLINGWORTH DOUGLAS B. KUHNS

MASATAKA KUWANA DONNA P. LUCAS

ROBERT S. LANCIOTTI ANDREW D. LUSTER

MARIE LOUISE LANDRY HOLDEN T. MAECKER

MARK D. LINDSLEY DIANA METES

ROBERT P. LISAK DIMITRI MONOS

CHRISTINE M. LITWIN MARTINA MURPHY

MOON H. NAHM Centro de Investigación del Cáncer (Instituto de Biología

STANLEY J. NAIDES ROBERT B. PETERSEN

TIMOTHY B. NIEWOLD FANNY POJERO

LAURA Z. RASSENTI RICHARD RUBENSTEIN

ROSEMARY SHE ANDREA K. STECK

R. SUE SHIREY MARYALICE STETLER-STEVENSON

TARA SIGDEL JAMES R. STONE

DAVID J. SPEICHER RENATO VEGA

JAMES W. VERBSKY THERESA L. WHITESIDE

PRIYA S. VERGHESE PATRICIA P. WILKINS

MARÍA BELÉN VIDRIALES HUGH J. WILLISON

NOREEN M. WALSH LIPING YU

Acknowledgment of Previous Contributors

TABLE 1 Comparison of three major NGS platforms

sequencing due to nucleotide neighbors, there is a similar qPCR

TABLE 2 Online resources for variation analysis

was examined by flow cytometry. While the healthy control

TABLE 1 Advantages and disadvantages of different protein arraysa

QA AND QC ISSUES IN PROTEIN ANALYSIS PROTEIN ANALYSIS: PROTEOMICS USING MS

instruments can be operated in scanning mode, which al-

FIGURE 9 A graphical representation of the transmission through a triple-­quadrupole mass spectrom-

Such new methods can generate fragment ions containing

TABLE 3 Key concepts in clinical MS

Library of Congress Cataloging-in-Publication Data

FIGURE 9 A graphical representation of the transmission through a triple-quadrupole mass spectrom-