Available online http://arthritis-research.

com/content/5/6/279

Review
Microarray analysis of gene expression in lupus
Mary K Crow1 and Jay Wohlgemuth2
1Mary Kirkland Center for Lupus Research, Hospital for Special Surgery, New York, NY, USA
2Expression Diagnostics, Inc, South San Francisco, CA, USA

Corresponding author: Mary K Crow (e-mail: crowm@hss.edu)

Received: 26 Aug 2003 Revisions requested: 5 Sep 2003 Revisions received: 22 Sep 2003 Accepted: 1 Oct 2003 Published: 13 Oct 2003

Arthritis Res Ther 2003, 5:279-287 (DOI 10.1186/ar1015)

© 2003 BioMed Central Ltd (Print ISSN 1478-6354; Online ISSN 1478-6362)

Abstract
Recent advances in the study of global patterns of gene expression with the use of microarray
technology, coupled with data analysis using sophisticated statistical algorithms, have provided new
insights into pathogenic mechanisms of disease. Complementary and reproducible data from multiple
laboratories have documented the feasibility of analysis of heterogeneous populations of peripheral
blood mononuclear cells from patients with rheumatic diseases through use of this powerful
technology. Although some patterns of gene expression, including increased expression of immune
system cell surface activation molecules, confirm previous data obtained with other techniques, some
novel genes that are differentially expressed have been identified. Most interesting is the dominant
pattern of interferon-induced gene expression detected among blood mononuclear cells from patients
with systemic lupus erythematosus and juvenile dermatomyositis. These data are consistent with long-
standing observations indicating increased circulating interferon-α in the blood of patients with active
lupus, but draw attention to the dominance of the interferon pathway in the hierarchy of gene
expression pathways implicated in systemic autoimmunity.

Keywords: gene expression, interferon, microarray, statistical algorithms, systemic lupus erythematosus

Introduction: the dawning of the microarray RNA-derived material from a cell population is hybridized
era to the gene array, is an innovative technology that has
The concept that the identification of genes that are differ- already changed our understanding of the mechanisms
entially expressed in a disease state will elucidate disease that underlie disease [1].
mechanisms has driven the development of new technol-
ogy. Earlier approaches, including Northern blotting, poly- The utility of microarray analysis of gene expression was
merase chain reaction (PCR), and RNase protection, have demonstrated impressively in 2000 when Alizadeh and
permitted analysis of small numbers of gene transcripts, colleagues used this technique to study the malignant cell
but the value of characterizing a broad spectrum of gene population of patients with diffuse large B cell leukemia
products expressed in a cell population or in a disease [2]. Although individual patient samples were not readily
state has stimulated the invention of more sophisticated differentiated on the basis of traditional cell surface phe-
tools. Subtractive hybridization and representational differ- notypic markers, microarray analysis discerned two dis-
ence analysis, comparing gene expression in two cell pop- crete tumor groups: those with a gene expression profile
ulations, are time-intensive approaches used in the late similar to that of germinal center B cells from healthy indi-
1990s to assist in gene discovery and to identify molecu- viduals and those with a profile similar to that of activated
lar pathways relevant to a disease. Microarray analysis, a mature B cells. Significantly, these two groups were char-
system in which thousands of oligonucleotide sequences acterized by markedly different clinical courses: B cell lym-
are spotted on a solid substrate, usually a glass slide, and phomas of the germinal center type had a 5-year survival
CNTFR = ciliary neurotrophic factor receptor; IFN = interferon; JCA = juvenile chronic arthritis; JDM = juvenile dermatomyositis; PBEF = pre-B-cell
colony-enhancing factor; PBMC = peripheral blood mononuclear cells; PCR = polymerase chain reaction; RA = rheumatoid arthritis; SAM = signifi-
cance analysis of microarrays; SLE = systemic lupus erythematosus. 279
 Arthritis Research & Therapy Vol 5 No 6 Crow and Wohlgemuth

of 76%, whereas lymphomas of the activated B cell type expressed or underexpressed in one population compared
were associated with a 16% 5-year survival. As striking as with another.
this study was, at that time confidence was not high that
microarray analysis of gene expression could be success- Although a comparative analysis of the current statistical
fully applied to heterogeneous populations of cells, cells approaches to microarray data is beyond the scope of this
that were not monoclonal. review, we will direct the reader to several of the most
useful algorithms that we and others have used to derive a
Numerous investigations over the past several years have comprehensive data set that includes genes most signifi-
demonstrated that significant and useful microarray data cantly differentially expressed between cell populations
can be derived from more complex cell samples, including (Table 1). When analyzing microarray data from two sets
cell populations from peripheral blood. Although such of patient-derived samples, we first use the SAM algorithm
studies face challenges in data interpretation, several lab- with parameters set to determine the genes significantly
oratories have used microarray analysis to study mononu- overexpressed or underexpressed in the patient group of
clear cells from patients with autoimmune diseases. When interest compared with the control group [9]. The false
studying mixed cell populations, gene expression profiling detection rate is set at approximately 5% (indicating that
can successfully detect differential gene expression in in only 5 cases out of 100 will a gene be erroneously iden-
specific cell types present in the samples under compari- tified as significantly differentially expressed). Additional
son. However, it can also measure and reflect the cellular algorithms, including supervised harvesting classification
composition of the sample. The contribution of variable and a method termed ‘shrunken centroids’, are then
enrichment of a cell population in a sample can be sorted applied to the data set [10–16]. Genes that either are
out by combining cell sorting or histology with expression identified in several of the algorithms or are most highly
profiling experiments. Cell sorting can also be used as an ranked in one of the algorithms are then used in a hierar-
initial step in sample preparation to enrich for specific cell chical clustering approach to identify additional genes that
types in a cell mixture to overcome sensitivity and speci- are coexpressed with the most significantly differentially
ficity limitations of arrays. expressed genes.

The recent advances in the analysis of broad gene expres- Array data can be internally validated by, for example, car-
sion patterns have rapidly led to new insights into patho- rying multiple probes for the same gene. When data from
genic mechanisms of rheumatic diseases and are all such probes in a particular experiment are identified as
supporting new initiatives in therapeutic drug develop- correlating in an array experiment, the probability of the
ment. Most striking are microarray data that have refo- finding being real is increased. Further, when data from
cused attention on the interferon (IFN) pathway in multiple genes that are known to be coexpressed in a
systemic lupus erythematosus (SLE) [3–6]. cluster or cellular pathway are correlated, the result has
greater significance than data from a single gene. Micro-
Analysis of microarray data array analysis cannot be relied on to develop a definitive
A statistical analysis of appropriately normalized micro- rank order of the most significant gene products associ-
array data is as important as the cell preparation, initial ated with a particular disease or cell state. Rather, the
hybridization, and data extraction in deriving valuable infor- technology is most useful in drawing attention to path-
mation from this technology. Arrays are useful because ways, and sometimes individual genes, that are most rele-
they allow large-scale screening for differential gene vant to the recent in vivo experience of the cell population
expression. However, the thousands of variables in the being studied.
analyses represent a significant statistical problem. Theo-
retically, when using an array with 1000 genes and a stan- As revealing as microarray data can be, confirmation of the
dard t-test with a significance at the P < 0.05 level, one data using more accurate, quantitative, and less variable
would expect to identify 50 genes that are ‘differentially approaches, such as real-time PCR, and validation in a
expressed’ by chance alone. Corrections for multiple com- second patient population are essential for drawing mean-
parisons, such as the Bonferroni method, can be used to ingful conclusions [17].
address this problem but tend to lose efficacy with very
large numbers of comparisons, and some differentially Early microarray data from the use of SLE
expressed genes might not appear as significant [7,8]. peripheral blood
Significance analysis of microarrays (SAM) is an approach In early 2003, Rus and colleagues reported data from a
that uses an estimate of the false detection rate to make study of gene expression in peripheral blood mononuclear
an estimate of significance under conditions in which cells (PBMC) from 21 lupus patients and 12 controls [18].
thousands of data points are being compared [9]. We The microarray assay used (Panorama Cytokine Gene
have learned that no one statistical algorithm is sufficient Array membranes; Sigma Genosys, Inc) included
280 to derive an accurate view of the genes significantly over- 375 genes enriched in cytokines, chemokines, cell surface
 Available online http://arthritis-research.com/content/5/6/279

Table 1

Statistical algorithms used in analysis of microarray data

Statistical algorithms Characteristics References Sources

Significance analysis of microarrays Identifies differentially expressed genes between [9] Stanford University,
(SAM) sample sets; estimates significance for genes; http://www-stat.stanford.edu/~tibs/
considers large numbers of genes in array x-mine, Brisbane, CA, http://www.x-mine.com/
experiments
Hierarchal clustering Unsupervised clustering; clusters genes with [15] University of California, Berkeley,
similar expression patterns; clusters samples http://rana.lbl.gov/EisenSoftware.htm
with similar expression patterns
Supervised harvesting classification Class prediction; identifies subset of genes that [10] x-mine, Brisbane, CA, http://www.x-mine.com/
best classify samples as gene sets; estimates
accuracy of gene set on prospective population
Classification and regression trees Class prediction; develops decision trees to [12,14] CART: Salford Systems,
(CART), multiple additive regression classify samples using the expression of a http://www.salford-systems.com/
trees (MART) subset of genes; estimates accuracy of the MART: Stanford University,
gene panel on a prospective set http://www-stat.stanford.edu/~jhf/R-MART.html
or Salford Systems
Shrunken centroids (prediction Class prediction; identifies subset of genes that [11] Stanford University,
analysis for microarrays, PAM) best classify samples as gene sets; estimates http://www-stat.stanford.edu/~tibs/PAM/
accuracy of gene set on prospective population index.html
Affymetrix MAS 5.0 Affymetrix
GeneSpring Silicon Genetics
Pathways 3.0 Research Genetics

receptors, and other immune-system cell surface mole- tors 3 and 4), IL1RAP (interleukin-1 receptor accessory
cules, including adhesion molecules. Data analysis com- protein), TGFBR3 (transforming growth factor-β receptor
prised several approaches: selection of genes with mean type III), and CCR7 (a T cell chemokine receptor important
expression in the SLE group that was more than 2.5-fold for the recruitment of CD4 T cells to the periarteriolar lym-
that observed in the healthy control group; genes that phoid sheath). Although the Rus study did not provide an
were significantly different between groups with the use of opportunity to detect the expression of genes that were
the Mann–Whitney U-test at a significance level of not obviously directly related to immune system function, it
P < 0.05; and SAM analysis with a false detection rate of drew attention to several genes that had not been studied
8%. Fifty genes were identified as differing by more than previously in detail in SLE. Overall, the data presented in
2.5-fold, and 20 genes differed between groups on the this study supported the value of the microarray approach
basis of the Mann–Whitney U-test, of which 15 differed by for detecting genes differentially expressed among PBMC
more than 2.5-fold between groups. The importance of from lupus patients.
confirming microarray data with an additional technique
was clearly illustrated in this report. The gene that showed A second early report came from Maas and colleagues,
the greatest fold difference between SLE and control who showed PBMC microarray data from a small number
groups by microarray, that encoding ciliary neurotrophic of patients with SLE (n = 9), rheumatoid arthritis (RA;
factor receptor α (CNTFR), could not be confirmed by n = 9), type I diabetes mellitus (n = 5) or multiple sclerosis
reverse transcriptase PCR (RT–PCR). Increased expres- (n = 4), along with nine control subjects before and after
sion of two other genes that were also studied by immunization with influenza vaccine [19]. The array used
RT–PCR, CXCR2 and that encoding pre-B-cell colony- (Research Genetics GF-211 membrane) included more
enhancing factor (PBEF), was confirmed, although the than 4000 genes, and the statistical analysis was based
fold increase as assessed by RT–PCR was less than esti- on Eisen’s Cluster and Treeview software, as well as the
mated by microarray. Of the genes significantly overex- Research Genetics Pathways 3.0 program used to locate
pressed, the products of some, including MMP3, differentially expressed genes in pathways related to the
TNFRSF1B (TNF receptor 2), IL1B, and FCGRIA, have immune system [15]. As in the Rus study, microarray
been documented to be increased in SLE. Among other analysis of unfractionated PBMC provided data that
genes with increased expression were TNFSF10 (encod- seemed to be reproducible and statistically significant,
ing TRAIL), TNFRSF10C and TNFRSF10D (TRAIL recep- and characterization of the cell composition of the PBMC 281
 Arthritis Research & Therapy Vol 5 No 6 Crow and Wohlgemuth

indicated that a variable presence of mononuclear cell striking for its enrichment in those that have been linked to
populations could not account for differential gene expres- IFN. MX1, MX2, G1P3, IRF7, and C1ORF29 were among
sion. However, the analysis was not able to distinguish a the IFN-induced genes overexpressed in the JDM muscle
gene expression profile that was distinct for SLE as com- samples. Increased expression of several genes, including
pared with RA or other autoimmune diseases. Ninety-five G1P3 (encoding IFN-induced 6-16) and CDKN1A (p21
genes were identified that distinguished the samples from cyclin kinase inhibitor), were confirmed by real-time PCR,
all four autoimmune diseases from healthy controls, includ- and G1P3 expression was also identified in JDM periph-
ing those encoding the cell surface receptors TGFBR2, eral blood. It is of interest that there was no significant
CSF3R, and BMPR2, which were overexpressed in the increase in mRNA for either IFN-α or IFN-γ, although IFN-γ
autoimmune patients, and several genes implicated in protein was seen in JDM muscle by using immunohisto-
apoptosis (TRADD, TRAF2, CASP6, CASP8), which chemistry. The authors interpreted their data as consistent
were underexpressed. It is of interest that ADAR, an IFN- with effects of both IFN-α/β and IFN-γ on gene expression.
induced gene encoding the RNA-specific adenosine In addition to this set of IFN-induced genes, gene expres-
deaminase, was highly overexpressed in most patients sion profiles indicating ischemia and myofiber degenera-
with autoimmune disease. Increased RNA editing, depen- tion and regeneration were detected.
dent on the ADAR protein, has been identified in T cells in
patients with SLE [20]. The patterns seen in the patients Data supporting a pathogenic role for type I IFN (predomi-
with autoimmune disease were distinct from those nantly IFN-α) in SLE have been available for 25 years, with
observed in samples from healthy subjects who had type II IFN (IFN-γ) also being implicated in murine models
received influenza vaccination, suggesting that the path- of lupus [22–31]. It has only been recently that microarray
ways involved in the immunopathogenesis of autoimmune data from several laboratories have refocused attention on
disease do not simply reflect an active immune response this important cytokine and its downstream targets.
to an antigen. This study relied mainly on clustering algo- Studies from our laboratory and from others have used
rithms to identify genes that were differentially expressed more extensive microarrays than those used in the two
among the study groups and also eliminated from analysis SLE studies described above to characterize the broad
genes whose expression levels did not vary by more than gene expression profile operative in the peripheral blood
3SD from their means, supporting our view that multiple of patients with SLE [3–6]. Most striking is an ‘IFN signa-
approaches, including algorithms such as SAM, are useful ture’, a prominent overexpression of mRNAs encoded by
in discerning gene expression relevant to disease-specific genes regulated by IFNs and similar to that detected in
molecular pathways. JDM muscle.

Interferon-induced gene expression in Detection of an interferon signature in SLE by

rheumatic diseases using microarray technology
A gene expression profile identified by microarray analysis Ambitious projects aimed at characterizing broad gene
and consistent with IFN-mediated gene transcription in a expression profiles in large numbers of SLE PBMC
rheumatic disease was first reported by Tezak and col- samples have come to fruition in 2003. Baechler and col-
leagues in a study of muscle biopsy tissue from four leagues have reported microarray data from 48 SLE
patients with juvenile dermatomyositis (JDM) [21]. Data patients and 42 healthy controls with Affymetrix U95A
from biopsies were compared with gene expression data GeneChips [4]. After eliminating genes that were highly
from two healthy control peripheral blood samples as well sensitive to induction ex vivo, 4566 genes remained for
as previously obtained microarray data from muscle biopsy analysis. An initial set of genes was selected on the basis
samples from patients with Duchenne muscular dystrophy of an unpaired Student’s t-test (apparently without correc-
by using Affymetrix HuFL GeneChips. Genes showing at tion for multiple comparisons) and further selection based
least a twofold difference in comparisons of JDM samples on a more than 1.5-fold difference in expression between
with each of two control samples were used to develop a lupus samples and controls, a difference of at least 100
list of differentially expressed genes. Ninety-one genes units in the expression value, and P < 0.001 by t-test. SAM
showed more than twofold increased expression, and analysis was not used in this study. In the full data set,
87 genes showed more than twofold lower expression. It 161 genes fulfilled all of these criteria. Hierarchical clus-
should be noted that in studies using small numbers of tering of these genes was then performed to identify gene
patient samples, although genes might be identified that expression patterns among the study samples. As in the
are truly differentially expressed between the samples, the JDM study, a striking increase in the expression of a group
variable expression patterns might be unrelated to the of genes previously reported to be induced by IFN was
disease state. In this study, fold differences between observed in about half of the SLE subjects. The authors
patient samples and controls were highly variable (ranging identified 23 of the 161 differentially expressed genes as
from 3.8-fold to 96.4-fold higher in patients than in con- targets of IFN by determining gene expression of PBMC
282 trols), but the list of differentially expressed genes was cultured for 6 hours either with IFN-α plus IFN-β or with
 Available online http://arthritis-research.com/content/5/6/279

IFN-γ. It should be noted that the patterns of gene expres- SLE patients. Data were analyzed with a correction for
sion induced in PBMC by type I and type II IFNs can vary multiple comparisons. With this approach, 15 genes were
with time after initial stimulation. The data from Baechler, found differentially expressed between SLE patients and
then, provide only a partial assessment of IFN-regulated healthy controls, and 14 of those 15 were identified as
genes at a single point in time. targets of IFN. Several of the most significantly differen-
tially expressed genes were among those identified in the
The pattern of expression of the IFN-regulated genes was Baechler study (C1ORF29, MX1, LY6E, PLSCR1, and
complex (Table 2). Eleven of the IFN-regulated genes that APOBEC3B) and others identified with less stringent sta-
were differentially expressed between SLE and control tistical criteria were also found in the Baechler study
PBMC clustered together and were preferentially induced (Table 2) or are closely related to genes identified by
by a combination of IFN-α and IFN-β. However, two genes Baechler (OAS1, OAS2, MX2). In addition, Bennett identi-
preferentially induced by IFN-γ (SERPING1 and fied IFI44 (encoding hepatitis C-associated microtubular
FCGR1A) were also significantly increased in the SLE aggregate protein), IFIT4 (termed CIG49 by those
patients and clustered with the IFN-α/β-induced genes. authors), CIG5 (viperin), and C1ORF29 as nearly univer-
Additional genes, preferentially induced by IFN-α/β but not sally expressed in their SLE subjects. The JCA samples did
clustering with the major IFN-induced group (CD69, not demonstrate overexpression of IFN-induced genes.
RGS1, IL1RN, and AGRN), were also increased in the
SLE group, and two genes repressed by IFN-α/β were Although neither the Baechler study nor the Bennett study
significantly underexpressed in the SLE patients. Three confirmed the IFN-regulated gene expression signature by
genes significantly overexpressed in SLE were decreased using more quantitative techniques, the similarity of results
in expression by both IFN-γ and IFN-α/β (EREG, THBS1, derived from the two studies is remarkable, strongly sup-
and ETS1). These are decreased somewhat more by IFN-γ porting the significance of the IFN pathway in SLE and
than by IFN-α/β. Taken together, these SLE data, along also demonstrating the power of microarray technology,
with the very useful information about genes induced by even when applied to heterogeneous populations of
type I and type II IFNs, support the type I IFNs as being peripheral blood cells.
particularly important in the gene expression pattern that
distinguishes SLE PBMC from those of healthy controls. Additional observations by Bennett and colleagues
Confirmation of these data by a more quantitative tech- included the significant overexpression of DEFA3, encod-
nique will be essential to determine more precisely the rel- ing the neutrophil-specific α3 defensin that is predomi-
ative expression of this gene set in patients with SLE, as nantly expressed in precursors of mature
well as those with other autoimmune diseases. polymorphonuclear cells [5]. Sorting of granular cells iden-
tified by flow cytometry confirmed a population of early
In addition to the IFN-regulated genes, the Baechler study granulocyte cells not present in mononuclear cell popula-
documented an increased expression of genes encoding tions from controls. DEFA3 and another neutrophil gene
immune system activation antigens, including TNFR6 FPRL1 (encoding formyl peptide receptor-like-1), along
(encoding Fas), CD54 (ICAM-1), and CD69 in the SLE with several of the IFN-induced genes, were highly corre-
samples, along with FCGRIA (as observed in the Rus lated with disease activity as measured by the Systemic
study) and FCGRIIA (Table 3) [4]. Other genes detected Lupus Erythematosus Disease Activity Index.
by Rus and colleagues were also identified (IL1B, IL1RB).
Genes with decreased expression in the SLE samples Our laboratories have performed microarray analyis on
included TCF3 (encoding transcription factor E2 α), PBMC samples from 22 SLE, 15 RA, 8 osteoarthritis,
important in B cell development; TCF7 (transcription 2 JCA, and 9 control PBMC samples and have detected a
factor 7 [T-cell specific, HMG box]), a polymorphism of gene expression signature virtually identical to that
which has been associated with type I diabetes; and LCK described by the other groups (Table 2) [3]. Importantly,
(lymphocyte-specific tyrosine kinase), which mediates our data were derived from a microarray (proprietary to
T cell activation. Expression Diagnostics, Inc) distinct from that used by
Baechler and Bennett and included more than 8000 gene
A second extensive gene expression study by Bennett and sequences, most of which were identified from subtracted
colleagues also used Affymetrix U95AV2 microarrays to and normalized cDNA libraries isolated from resting and
analyze PBMC from 30 pediatric lupus patients ranging in activated leukocytes. We have found that, in contrast to the
age from 6 to 18 years (mean age 13), 12 patients with RA, osteoarthritis, and JCA patients, and healthy controls,
juvenile chronic arthritis (JCA), and 9 healthy control chil- most adult SLE patients express the IFN gene signature
dren [5]. Of the 30 SLE patients, 18 were studied no (example data shown in Fig. 1). Moreover, we have con-
more than 1 year after diagnosis, reflecting the gene firmed the increased expression of several of those overex-
expression profile at a point in time closer to the onset of pressed genes by using real-time PCR analysis and in a
symptoms than is usually possible to document in adult second cohort of SLE patients [3]. Of great interest is the 283
 Arthritis Research & Therapy Vol 5 No 6 Crow and Wohlgemuth

Table 2

Interferon-induced genes identified in large-scale microarray analyses of SLE PBMC

Response of PBMC to IFN-α/β

(type I)/IFN-γ (type II) Expression in SLE compared with control PBMC

Gene Protein Reference: [4]* [4]† [5]‡ [3]§ [6]||

IFIT1 Interferon-induced protein with tetratricopeptide repeats-1 18.3 Up Up Up

OASL 2′-5′-oligoadenylate synthetase-like 16.3 Up Up Up
LY6E Lymphocyte antigen 6 complex, locus E 14.9 Up Up Up Up
OAS2 2′-5′-oligoadenylate synthetase 12.7 Up Up
OAS3 2′-5′-oligoadenylate synthetase NA NA Up
IFI44 Hepatitis C microtubular aggregate protein 10.7 Up Up
MX1 Myxovirus resistance 1 9.8 Up Up Up
G1P3 Interferon, alpha-inducible protein (IFI-6-16) 7.1 Up Up
PRKR Protein kinase, interferon-α-inducible double-stranded 6.9 Up Up
RNA-dependent¶
IFIT4 Interferon-induced protein with tetratricopeptide repeats 4 6.8 Up Up Up
PLSCR1 Phospholipid scramblase 1 6.6 Up Up Up
C1ORF29 Hypothetical protein expressed in osteoblasts; similar to IFI44 6.1 Up Up Up
HSXIAPAF1 XIAPassociated factor-1 6.1 Up Up Up
G1P2 Interferon, alpha-inducible protein (IFI-15K) 5.9 Up Up
Hs. 17518 Viperin 5.5 Up
(Cig5)
IRF7 Interferon regulatory factor 7 4.6 Up Up
CD69 Early T-cell activation antigen 4.1 Up
LGALS3BP Lectin, galactoside-binding, soluble, 3 binding protein 3.8 Up Up
IL1RN Interleukin-1 receptor antagonist 3.5 Up Up
APOBEC3B Phorbolin 1-like 2.0 Up Up Up
RGS1 Regulator of G-protein signaling 1 1.8 Up
AGRN Agrin > 71.2 (γ → < 0) Up Up
EREG Epiregulin 1.3 (α/β and γ → < 0) Up
THBS1 Thrombospondin 1 1.3 (α/β and γ → < 0) Up
ETS1 v-ets erythroblastosis virus E26 oncogene 1.2 (α/β and γ → < 0) Up
homolog 1
ADAM9 A disintegrin and metalloproteinase domain 9 1.1 (α/β and γ → < 0) Up
SERPING1 Serine (or cysteine) proteinase inhibitor (C1 inhibitor) 0.85 Up Up
USP20 Ubiquitin specific protease 20 0.25 (α/β and γ → < 0) Down
MATK Megakaryocyte-specific tyrosine kinase < 0.13 (α/β → < 0) Down
FCGR1A Fc fragment of IgG, high-affinity Ia receptor 0.08 Up Up

The genes and corresponding proteins listed were identified as significantly differentially expressed in SLE and healthy control PBMC in the study
by Baechler and colleagues [4], or were among the most commonly overexpressed transcripts among SLE PBMC in the study by Bennett and
colleagues [5]. The significant overexpression of these genes in SLE compared with control PBMC in microarray data sets from Crow and
colleagues [3] or Han and colleagues [6] is also noted. In addition, genes identified by Baechler and colleagues as both regulated by type I (IFN-
α/β) or type II (IFN-γ) and differentially expressed by SLE PBMC are noted. A ratio of gene expression induced in healthy PBMC by IFN-α/β (1000
U/ml for 6 hours) compared with IFN-γ (1000 U/ml for 6 hours) for each gene was calculated by determining the net microarray score for each of
four control samples studied by Baechler and colleagues (stimulated microarray score minus unstimulated microarray score) for both IFN-α/β and
IFN-γ stimulation, determining the average of the four scores, and dividing the IFN-α/β score by the IFN-γ score for each gene. In some cases,
scores for both IFN-α/β- and IFN-γ-induced gene expression were less than background [indicated as (α/β and γ → < 0)]. When the score for
either IFN-α/β-induced or IFN-γ-induced gene expression was less than the background, the score for the lower value was replaced with a score of
50, to permit the calculation of an approximate ratio. Abbreviations: JCA, juvenile chronic arthritis; NA, not available; OA, osteoarthritis; RA,
rheumatoid arthritis. *The microarray system used was an Affymetrix U95A GeneChip (Affymetrix, Santa Clara, CA); 4566 genes were analyzed.
Four healthy donors of PBMC were studied. †The microarray system used was an Affymetrix U95A GeneChip; 4566 genes were analyzed. The
donors of PBMC studied were 48 with SLE and 42 healthy. ‡The microarray system used was an Affymetrix U95AV2 GeneChip; about 4600
genes were analyzed. The donors of PBMC studied were 30 pediatric with SLE, 12 with JCA, and 9 healthy children. §The microarray system used
was a proprietary microarray from Expression Diagnostics, Inc; 8143 oligonucleotides are represented. The donors of PBMC studied were 22 with
SLE, 15 with RA, 8 with OA, 2 with JCA, and 9 healthy. ||The microarray system used was a Mergen ExpressChip DNA Microarray (Mergen, Ltd,
San Leandro, CA); 3002 genes are represented. The donors of PBMC studied were 10 with SLE and 18 healthy. ¶Identified as capicua homolog
284 in some studies.
 Available online http://arthritis-research.com/content/5/6/279

Table 3

Selected gene families overexpressed or underexpressed in SLE PBMC

Gene families overexpressed Examples Gene families underexpressed Examples

IFN target genes See Table 2 Transcription factors TCF3, TCF7

TNF and TNF receptor families TNFSF10 (TRAIL), TNFRSF10C (TRAIL receptor 3), Kinases LCK
TNFR6 (Fas)
Chemokines and chemokine
receptors CCR7, CXCR2 T cell receptors TCRB, TCRD
Cell surface activation antigens CD69 C-type lectins KLRB1
Fc receptors FCGR1A, FCGR2A
Metalloproteinases MMP3, MMP9
Defensins DEFA3, F2RPA

Genes listed have been identified in microarray studies described in [3–6,8,19].

Figure 1 Human HO4) in a study of 10 SLE patients and 8 healthy

controls. The Han study also detected an increased
Healthy RA JCA SLE expression of several of the genes identified in the other
three studies (Table 2). In contrast to the Baechler and
OAS3
Bennett reports, which note that type I and II IFNs are
PRKR undetectable in serum from most SLE patients, we have
OAS2
IFI44 measured plasma IFN-α protein by ELISA and are currently
IFI44
IRF7 analyzing those data in relation to the IFN target gene
IRF7
IFIT1
expression data, as well as measures of clinical disease
CCL2 activity. Importantly, our data provide direct support for
LYL1
AQP3 IFN-α in the gene expression profile observed in SLE
MX2
IFIT1
PBMC (KA Kirou, C Lee, S George, K Louca, M Peterson,
MX2 MK Crow, unpublished observations). However, it should
HSXIAPAF1
STAT1 be noted that the IFN signature is complex, as described in
G1P3
CCL3 detail for the Baechler study, and additional work will be
ADA
IFITM2
required to determine the relative roles of type I and type II
Hs.76853 IFN in the gene expression profile observed in SLE periph-
CCR1
CD1a eral blood. Analysis of the age of study patients, time from
Hs.17481
ADAR
diagnosis, organ system involvement, and current therapy,
HIST2H4 along with gene expression data, will be essential for an
understanding of the place of IFN family members in
Exemplary gene sequences that cluster with PRKR and OAS3. disease pathogenesis.
Hierarchical clustering was performed on the total study population to
determine genes that cluster with PRKR and OAS3. A visual
Beyond the impressive IFN gene signature, each of the
demonstration of the expression of a selection from those genes,
comprising a partial IFN signature, is shown. Data are shown from a studies identified additional genes that were differentially
subset of SLE samples tested (n = 14) and from rheumatoid arthritis expressed in SLE and control samples. In view of the bio-
(RA) (n = 11), juvenile chronic arthritis (JCA) (n = 2), and control logical, patient, and assay variation encountered in gene
samples (n = 8). Relative expression compared with an internal control profiling of SLE, a diagnostic gene signature might prove
ranged from approximately –0.5 (bright green) to 0.5 (bright red).
most useful clinically. An algorithm such as shrunken cen-
troids can be used to develop a robust multigene classifier
identity of the stimulus for the IFN-induced gene expression that can be tested in clinical studies as a measure of diag-
signature. It is noteworthy that none of the SLE or JDM nosis or clinical outcome [11].
studies identified significantly increased expression of
type I or type II IFN mRNA, with the exception of a recent Microarray analysis of gene expression at
report by Han and colleagues [6] that detected IFN-ω, sites of tissue damage
another type I IFN species, by using a distinct microarray The next generation of reports describing global gene
system (Mergen ExpressChip DNA Microarray System expression patterns based on microarray assays will prob- 285
 Arthritis Research & Therapy Vol 5 No 6 Crow and Wohlgemuth

ably derive from studies of tissue samples from sites of Acknowledgements

disease activity. Unpublished data presented at scientific We thank our collaborators at the Hospital for Special Surgery and
Expression Diagnostics, Inc, particularly Dr Kyriakos Kirou and Ms
meetings demonstrate the feasibility of microarray analysis Christina Lee and Ms Sandhya George, for their contributions to the
of glomeruli from lupus nephritis kidneys and from skin work described in this review. MKC is supported by a Target Identifica-
tion in Lupus Grant from the Alliance for Lupus Research, by a Novel
lesions of lupus patients. As valuable as those data will be, Research Grant from the Lupus Research Institute, and by the Mary
we are encouraged that peripheral blood seems to Kirkland Center for Lupus Research.
provide a reasonable sampling of those gene pathways
that are activated and relatively disease specific. References
1. Brown PO, Botstein D: Exploring the new world of the genome
with DNA microarrays. Nat Genet 1999, 21:33-37.
Conclusion 2. Alizadeh AA, Eisen MB, Davis RE, Ma C, Lossos IS, Rosenwald A,
Since the dramatic illustration of the clinical utility of Boldrick JC, Sabet H, Tran T, Yu X, Powell JI, Yang L, Marti GE,
Moore T, Hudson J Jr, Lu L, Lewis DB, Tibshirani R, Sherlock G,
microarray technology in patients with malignant disease Chan WC, Greiner TC, Weisenburger DD, Armitage JO, Warnke
3 years ago, efforts to study mixed mononuclear cell popu- R, Staudt LM: Distinct types of diffuse large B-cell lymphoma
lations in patients with autoimmune diseases have been identified by gene expression profiling. Nature 2000, 403:503-
511.
remarkably successful. Not only are microarray-derived 3. Crow MK, George S, Paget SA, Ly N, Woodward R, Fry K,
data interpretable and significant, but they have drawn our Wohlgemuth J: Expression of an interferon-alpha gene
program in SLE. Arthritis Rheum 2002, 46:S281.
attention back to a key cytokine pathway. Increased 4. Baechler EC, Batliwalla FM, Karypis G, Gaffney PM, Ortmann
expression of IFN in patients with active SLE was first WA, Espe KJ, Shark KB, Grande WJ, Hughes KM, Kapur V,
reported in 1979, but relatively few investigators have Gregersen PK, Behrens TW: Interferon-inducible gene expres-
sion signature in peripheral blood cells of patients with
pursued the role of IFN in SLE [22]. An important excep- severe lupus. Proc Natl Acad Sci USA 2003, 100:2610-2515.
tion is the group led by Ronnblom and Alm, who noted the 5. Bennett L, Palucka AK, Arce E, Cantrell V, Borvak J, Banchereau
J, Pascual V: Interferon and granulopoiesis signatures in sys-
induction of lupus autoantibodies and clinical lupus in temic lupus erythematosus blood. J Exp Med 2003, 197:711-
patients receiving therapeutic IFN-α [26]. That group has 723.
gone on to characterize the IFN-producing cells as well as 6. Han GM, Chen SL, Shen N, Ye S, Bao CD, Gu YY: Analysis of
gene expression profiles in human systemic lupus erythe-
some of the properties of immune complexes that induce matosus using oligonucleotide microarray. Genes Immun
the production of IFN by plasmacytoid dendritic cells 2003, 4:177-186.
7. Benjamini Y, Hochberg Y: Controlling the false discovery rate: a
[32–37]. In addition, Bennett’s collaborators, led by practical and powerful approach to multiple testing. J R Statist
Banchereau, have presented functional data showing the Soc 1995, 57:289–300.
induction of efficient stimulators of allogeneic T cell 8. Dudoit S, Gentleman RC, Quackenbush J: Open source soft-
ware for the analysis of microarray data. Biotechniques 2003,
responses by IFN-α in SLE sera [28]. Now, the repeated Mar Suppl:45-51.
appearance of IFN-induced genes among the most signifi- 9. Tusher VG, Tibshirani R, Chu G: Significance analysis of
microarrays applied to the ionizing radiation response. Proc
cantly overexpressed genes in data derived from multiple Natl Acad Sci USA 2001, 98:5116-5121.
laboratories using distinct microarrays raises the profile of 10. Hastie T, Tibshirani R, Botstein D, Brown P: Supervised harvest-
IFN as a pathogenic mediator of the myriad alterations to ing of expression trees. Genome Biol 2001, 2:research0003.1-
0003.12
the immune system seen in SLE. 11. Tibshirani R, Hastie T, Narasimhan B, Chu G: Diagnosis of multi-
ple cancer types by shrunken centroids of gene expression.
Proc Natl Acad Sci USA 2002, 99:6567-6572.
In view of the important effects of IFN on immune system 12. Friedman JH: Stochastic Gradient Boosting. Technical report.
function, including activities that could contribute to the Stanford, CA: Department of Statistics, Stanford University; 1999.
development of systemic autoimmunity, this cytokine 13. Golub TR, Slonim DK, Tamayo P, Huard C, Gaasenbeek M,
Mesirov JP, Coller H, Loh ML, Downing JR, Caligiuri MA, Bloom-
system might represent an excellent target for therapeutic field CD, Lander ES: Molecular classification of cancer: class
modulation [38–44]. At the same time, both type I and discovery and class prediction by gene expression monitor-
type II IFNs are important, if not essential, for effective host ing. Science 1999, 286:531-537.
14. Breiman L, Friedman JH, Olshen RA, Stone CJ: Classification and
defense against pathogenic microbes. The weight of data Regression Trees. Belmont, CA: Wadsworth; 1984.
are consistent with the action of type I IFNs as primary 15. Eisen MB, Spellman PT, Brown PO, Botstein D: Cluster analysis
and display of genome-wide expression patterns. Proc Natl
mediators of the observed gene expression signature in Acad Sci USA 1998, 95:14863-14868.
SLE, with significant consequences for the development 16. Sherlock G: Analysis of large-scale gene expression data. Curr
of clinical disease. However, additional potential mediators Opin Immunol 2000, 12:201-205.
17 Kirou K, Lee C, Crow MK: Measurement of cytokines in autoim-
of the IFN gene signature, including CpG DNA or double- mune disease. Methods Mol Med, in press.
stranded RNA, are candidates that should be explored. 18. Rus V, Atamas SP, Shustova V, Luzina IG, Selaru F, Magder LS,
Via CS: Expression of cytokine- and chemokine-related genes
in peripheral blood mononuclear cells from lupus patients by
Competing interests cDNA array. Clin Immunol 2002, 102:283-290.
MKC is a scientific advisor to Expression Diagnostics, Inc, 19. Maas K, Chan S, Parker J, Slater A, Moore J, Olsen N, Aune TM:
Cutting edge: molecular portrait of human autoimmune
and JW is Vice-President, Research and Development, disease. J Immunol 2002, 169:5-9.
Expression Diagnostics, Inc. Research conducted by MKC 20. Laxminarayana D, Khan IU, Kammer G: Transcript mutations of
the alpha regulatory subunit of protein kinase A and up-regu-
was supported through a contract between Expression lation of the RNA-editing gene transcript in lupus T lympho-
286 Diagnostics, Inc. and Hospital for Special Surgery. cytes. Lancet 2002, 360:842-849.
 Available online http://arthritis-research.com/content/5/6/279

21. Tezak Z, Hoffman EP, Lutz JL, Fedczyna TO, Stephan D, Bremer 42. Radvanyi LG, Banerjee A, Weir M, Messner H: Low levels of
EG, Krasnoselska-Riz I, Kumar A, Pachman LM: Gene expression interferon-alpha induce CD86 (B7.2) expression and acceler-
profiling in DQA1*0501+ children with untreated dermato- ates dendritic cell maturation from human peripheral blood
myositis: a novel model of pathogenesis. J Immunol 2002, mononuclear cells. Scand J Immunol 1999, 50:499-509.
168:4154-4163. 43. Luft T, Pang KC, Thomas E, Herzog P, Hart DN, Trapani J, Cebon
22. Hooks JJ, Moutsopoulos HM, Geis SA, Stahl NI, Decker JL, J: Type I IFNs enhance the terminal differentiation of dendritic
Notkins AL: Immune interferon in the circulation of patients cells. J Immunol 1998, 161:1947-1953.
with autoimmune disease. N Engl J Med 1979, 301:5-8. 44. Crow MK: Interferon-alpha: a new target for therapy in SLE?
23. Hooks JJ, Jordan GW, Cupps T, Moutsopoulos HM, Fauci AS, Arthritis Rheum 2003, 48:2396-2401.
Notkins AL: Multiple interferons in the circulation of patients
with systemic lupus erythematosus and vasculitis. Arthritis
Rheum 1982, 25:396-400. Correspondence
24. Preble OT, Black RJ, Friedman RM, Klippel JH, Vilcek J: Systemic Mary K Crow MD, Hospital for Special Surgery, 535 East 70th Street,
lupus erythematosus: presence in human serum of an New York, NY 10021, USA. Tel: +1 212 606 1397; fax: +1 212 774
unusual acid-labile leukocyte interferon. Science 1982, 216: 2337; e-mail: crowm@hss.edu
429-431.
25. Rich SA: Human lupus inclusions and interferon. Science
1981, 213:772-775.
26. Ronnblom LE, Alm GV, Oberg KE: Possible induction of sys-
temic lupus erythematosus by interferon-alpha treatment in a
patient with a malignant carcinoid tumour. J Intern Med 1990,
227:207-210.
27. Wandl UB, Nagel-Hiemke M, May D, Kreuzfelder E, Kloke O,
Kranzhoff M, Seeber S, Niederle N: Lupus-like autoimmune
disease induced by interferon therapy for myeloproliferative
disorders. Clin Immunol Immunopathol 1992, 65:70-74.
28. Blanco P, Palucha AK, Gill M, Pascual V, Banchereau J: Induction
of dendritic cell differentiation by IFN-alpha in systemic lupus
erythematosus. Science 2001, 294:1540-1543.
29. Peng SL, Moslehi J, Craft J: Roles of interferon-gamma and
interleukin-4 in murine lupus. J Clin Invest. 1997, 99:1936-
1946.
30. Lawson BR, Prud’homme GJ, Chang Y, Gardner HA, Kuan J,
Kono DH, Theofilopoulos AN: Treatment of murine lupus
with cDNA encoding IFN-γγR/Fc. J Clin Invest 2000, 106:
207-215.
31. Santiago-Raber ML, Baccala R, Haraldsson KM, Choubey D,
Stewart TA, Kono DH, Theofilopoulos AN: Type-I interferon
receptor deficiency reduces lupus-like disease in NZB mice. J
Exp Med 2003, 197:777-788.
32. Svensson H, Johannisson A, Nikkila T, Alm GV, Cederblad B: The
cell surface phenotype of human natural interferon-α α produc-
ing cells as determined by flow cytometry. Scand J Immunol
1996, 44:164-172.
33. Vallin H, Blomberg S, Alm GV, Cederblad B, Ronnblom L:
Patients with systemic lupus erythematosus (SLE) have a cir-
culating inducer of interferon-alpha (IFN-α α) production acting
on leukocytes resembling immature dendritic cells. Clin Exp
Immunol 1999, 115:196-202.
34. Vallin H, Peters A, Alm GV, Ronnblom L: Anti-double-stranded
DNA antibodies and immunostimulatory plasmid DNA in com-
bination mimic the endogenous IFN-alpha inducer in systemic
lupus erythematosus. J Immunol 1999, 163:6306-6313.
35. Bave U, Alm GV, Ronnblom L: The combination of apoptotic
U937 cells and lupus IgG is a potent IFN-alpha inducer. J
Immunol 2000, 165:3519-3526.
36. Bave U, Vallin H, Alm GV, Ronnblom L: Activation of natural
interferon-alpha producing cells by apoptotic U937 cells com-
bined with lupus IgG and its regulation by cytokines. J Auto-
immun 2001, 17:71-80.
37. Magnusson M, Magnusson S, Vallin H, Ronnblom L, Alm GV:
Importance of CpG dinucleotides in activation of natural IFN-
alpha-producing cells by a lupus-related oligodeoxynu-
cleotide. Scand J Immunol 2001, 54:543-550.
38. Ronnblom L, Alm GV: A pivotal role for the natural interferon α-
producing cells (plasmacytoid dendritic cells) in the patho-
genesis of lupus. J Exp Med 2001, 194:59-63.
39. Aman MJ, Tretter T, Eisenbeis I, Bug G, Decker T, Aulitzky WE,
Tilg H, Huber C, Peschel C: Interferon-alpha stimulates pro-
duction of interleukin-10 in activated CD4+ T cells and mono-
cytes. Blood 1996, 87:4731-4736.
40. Brinkmann V, Geiger T, Alkan S, Heusser CH: Interferon alpha
increases the frequency of interferon gamma-producing
human CD4+ T cells. J Exp Med 1993, 178:1655-1663.
41. Kirou KA, Vakkalanka RK, Butler MJ, Crow MK: Induction of Fas
ligand-mediated apoptosis by interferon-alpha. Clin Immunol
2000, 95:218-226. 287