JP2001510027A - LO-CD2a antibody inhibiting T cell activation and proliferation and use thereof - Google Patents

Abstract

  (57) [Summary] The present invention relates to LO-CD2a antibodies and also to prevent an immune response in a human patient, especially where the immune response is mediated by the activation and proliferation of T cells or natural killer cells; and A method of using such an antibody or molecule that binds to the same epitope (or portion thereof) that inhibits it. Administration of an effective amount of a LO-CD2a antibody to a human patient will prevent or inhibit graft rejection, graft versus host disease or an autoimmune disease.

Description

DETAILED DESCRIPTION OF THE INVENTION

TECHNICAL FIELD [0001] The present invention relates to an antibody (or a fragment or a derivative thereof), and preferably to an antibody (or a fragment or a derivative thereof) that binds to human lymphocytes. More particularly, the invention relates to preventing and / or inhibiting an ongoing immune response in a patient through the administration of such an antibody (or fragment or derivative thereof) to the patient. Desirably, the invention relates to preventing or inhibiting T cell activation and proliferation through administration of such antibodies or fragments or derivatives thereof to a patient.

BACKGROUND OF THE INVENTION The prior art has disclosed the possibility of using antibodies against the CD2 antigen to inhibit graft rejection. In general, the prior art discloses the use of antibodies that bind to the CD2 antigen, possibly as useful in inhibiting graft rejection. Ortho Pharmaceutical Corporation, U.S. Patent No. 4,364,973, 4,
614,720, 4,515,893, 4,743,681, and 4,
See 798,806.

[0003] The cited references also did not show that such antibodies were useful for inhibiting transplant rejection in human patients or non-human primates. The following reference, J. et al. V. Jioruji other, monkey allograft OKT11A monoclonal antibody of immunosuppressive effects and immunogenicity in receptor transplantation Bulletin, Volume 15, No. 1, 1
March 983; J. Sirlow et al., Monoclonal anti-chimpanzee T cell antibody , Transplantation, 36, 3, 293-298.

[0004] Accordingly, an antibody against a CD2 antigen that inhibits graft rejection inhibits T cell activation and proliferation or stimulates T cell activation and proliferation in human or non-human primates by administering an agent that stimulates T cell activation and proliferation. No disclosure is made regarding inhibiting graft rejection even before or after antibody addition.

[0005] In view of the foregoing, the present invention provides a humanized antibody from the monoclonal antibody LO-CD2a produced from the cell line ATCC Deposit No. HB11423,
It is another object of the present invention to provide a method for inhibiting an immune response and a method for inhibiting transplant rejection.

(Disclosure of the Invention) The present invention can solve the above-mentioned problems by providing the following configuration.

(1) One humanized antibody consisting of CDRs from LO-CD2a, wherein the humanized antibody is a rat as shown in FIGS. 55 and 56 in the framework of the heavy chain variable region of the humanized antibody. Amino acids 47, 67, 70 of the LO-CD2a heavy chain variable region
, 72, 76, 85 and 87 and one of amino acids 12, 13, 28 and 48
A humanized antibody comprising: 2, 3, 4 or 4.

(2) The humanized antibody according to the above (1), wherein the humanized antibody is a rat LO-LOA as shown in FIGS. 55 and 56 in the framework of the heavy chain variable region of the humanized antibody. Amino acids 12, 13, 28, 47, 48, 67 of CD2a heavy chain variable region
, 70, 72, 76, 85 and 87.

(3) The humanized antibody according to (1) above, wherein the humanized antibody is a rat LO-C as shown in FIG. 39 in the framework of the light chain variable region of the humanized antibody.
One or more amino acids 9, 12, 41, 42 of the D2a light chain variable region;
A humanized antibody, further comprising 50, 51 and 82.

(4) The humanized antibody according to (3) above, wherein the humanized antibody is a rat LO-C as shown in FIG. 39 in the framework of the light chain variable region of the humanized antibody.
A humanized antibody comprising amino acids 9, 12, 41, 42, 50, 51 and 82 of the D2a light chain variable region.

(5) The humanized antibody according to the above (1), wherein the heavy chain variable region of the humanized antibody has the amino acid sequence of FIGS. 55 and 56.

(6) The humanized antibody according to the above (3), wherein the light chain variable region of the humanized antibody has the amino acid sequence of FIG. 39.

(7) One method for inhibiting an immune response in a patient, the method comprising:
A method comprising treating with the antibody of any of the preceding claims.

(8) One method of inhibiting rejection of a graft in a patient, the method comprising:
1) A method comprising treating with the antibody according to the above.

BEST MODE FOR CARRYING OUT THE INVENTION In order to explain the present invention in more detail, the present invention will be described below with reference to the accompanying drawings.

In accordance with one aspect of the present invention, a molecule (preferably a monoclonal antibody or a monoclonal antibody thereof) that binds to the same human lymphocyte epitope (or portion thereof) as the monoclonal antibody produced by the cell line deposited under ATCC Deposit No. HB11423. Fragment) is provided. Antibodies produced in the deposited cell line are sometimes referred to as L
Quoted as O-CD2a. The term "molecule" or "antibody that binds to the same epitope as LO-CD2a" includes LO-CD2a. "LO-CD
The term "2a" includes antibodies produced by the deposited cell line ATCC HB11423 and those identical to, for example, those produced by recombinant techniques.

The molecules or antibodies of the invention inhibit the activation and proliferation of human T cells, and Applicants have added the molecules or antibodies either before or after the agent that stimulates T cell activation. Sometimes it has been found that such inhibition can be achieved.

The molecules or antibodies of the invention have the property of binding to the epitope of the CD2 antigen (CD2 positive human T cells), but the ability of such molecules or antibodies to inhibit T cell activation or proliferation. Applicant now believes that may or may not be effected through binding to CD2 positive cells and that the mechanism of action involves the binding of a molecule or antibody to CD2 positive cells.

In accordance with one aspect of the invention, an antibody, hereinafter referred to as LO-CD2a (or a fragment or derivative thereof), or any molecule that mimics such an antibody or derivative or fragment thereof, is administered to a human patient. Through doing
One method is provided for preventing and / or inhibiting an ongoing immune response in a human patient.

A cell line that produces LO-CD2a has been deposited with the American Type Culture Collection at 12301 Parklone Drive, Rockville, Md., 20852, July 28, 1993 and has ATCC access number ATC.
CHB11423 was provided. This antibody is a rat monoclonal antibody.

Applicants do not intend to limit the invention to any theoretical analogy, but use the monoclonal antibodies of the invention to prevent or reduce the severity of an immune response and to inhibit the activity of T cells. The mechanism that allows the inhibition of proliferation and proliferation is that the LO-CD2a antibody reduces the density of CD2 expressed on the surface of T cells and thus reduces the number of CD2-T lymphocytes, and / or It is believed that this is a fact that affects marked transduction. These mechanisms of action are thought to be responsible not only for the prevention of the immune response, but also for the ongoing reduction of the immune response. In addition, L
O-CD2a antibody as is exemplified herein, inhibit natural killer (NK) cells test tube activity. This is appropriate for the present invention. This is because non-MHC-restricted cytotoxic mechanisms such as NK cell activity are thought to have influenced graft-versus-host disease.

According to one aspect of the invention, by administering to a human patient an effective amount of a molecule (preferably an antibody) that binds to the same epitope on human lymphocytes (or any portion thereof), such as an LO-CD2a antibody. One method is provided for inhibiting the initial or further activation and proliferation of T cells in a patient. Preferred molecules are LO-CD2a or chimeric and / or humanized forms thereof.
Such a molecule may, for example, have the same complementarity determining region (CDR) as the LO-CD2a antibody.
Will be included.

As used herein throughout this application, the term “inhibit” is intended to mean to prevent or inhibit graft rejection, or reduce disease severity, or induce or reverse resistance. I have. The term "transplant" as used herein for the purpose of this application is intended to mean any or all transplants, including but not limited to allograft and xenograft transplants. Such transplants include, but are not necessarily limited to, examples, including cells, bone marrow, tissue, solid organs, bone, and the like.

The term “immune response” as used herein is intended to mean an immune response that depends on the activation and proliferation of T cells, including both cellular effects and T cell dependent antibodies. , Which is not limiting in the examples, and (i) the implant,
(Ii) graft-versus-host disease, and (iii) self-antigens that result in autoimmune diseases such as, but not limited to, rheumatoid arthritis, systemic lupus, multiple sclerosis, diabetes mellitus, etc. Is derived in response to

The molecule used in the present invention binds to the same epitope (or a part of this epitope) as the LO-CD2a monoclonal antibody. "LO-CD
2a binds to the same epitope as the monoclonal antibody "
It is intended to describe not only the D2a monoclonal antibody, but also other antibodies, fragments or derivatives or molecules thereof that bind to the same such epitopes as the LO-CD2a monoclonal antibody.

[0026] Such other antibodies include, but are not limited to, rats, mice, pigs, cows, humans, chimeric, humanized antibodies, or fragments or derivatives thereof, by way of example and not limitation.

As used herein, the term “derivative” refers to the same epitope (or portion thereof) that is recognized by a chimeric or humanized antibody, a single chain antibody, a bispecific antibody, or a LO-CD2a monoclonal antibody. It refers to other antibodies that bind.

As used herein, the term “fragment” refers to a portion of an antibody, and such portions of the antibody are not necessarily limited to CDRs, Fabs, or LO-CD2a, depending on the embodiment. And other such portions that bind to the same epitope or any portion thereof.

The term “antibody” as used herein refers to an antibody prepared by recombinant techniques, such as polyclonal, monoclonal antibodies, also antibody fragments and derivatives, such as chimeric or humanized antibodies, monoclonal antibody LO-CD2a. And single-chain or bispecific antibodies that bind to the same epitope or portion thereof.
The term "molecule" is not necessarily limited by example, but may be any peptide, oligonucleotide, or other such derivative derived from any source that mimics an antibody or binds to the same epitope or portion thereof as an antibody fragment or derivative thereof. Such compounds are included.

Another embodiment of the invention is selected from the group consisting of a LO-CD2a antibody, or an antibody or derivative, or a fragment thereof or a molecule that binds to the same epitope (or portion thereof) as the LO-CD2a antibody. One method is provided for treating a patient to be or has undergone graft implantation with an effective amount of at least one member. This treatment is preferably performed with whole LO-CD2a antibody or intact LO-C.
Performed with the D2a antibody.

As described above, the monoclonal antibody of the present invention can be prepared by a method similar to the technology disclosed herein, as described in Koehler and Milstein (Nature 256, 495-497, 1975). It can be produced by technology. The preparation of the monoclonal LO-CD2a antibody is described in more detail in Example 1 of this application. As indicated previously, LO-CD2a antibodies can also be produced recombinantly using procedures known in the art. Recombinant antibodies may also take the form of chimeric antibodies, where the variable region of the LO-CD2a rat antibody is combined with the constant region of one species of antibody. Thus, for example, a monoclonal antibody is humanized by combining the CDR regions of a rat LO-CD2a monoclonal antibody with the V regions, framework regions and constant regions of a human antibody to provide a chimeric human-rat monoclonal antibody.

In one embodiment, the antibody is a humanized form of a LO-CD2a antibody constructed from the constant region of a human antibody, and the framework and CDR regions of the light and heavy chain variable regions. The chain and heavy chain variable region framework regions are derived from the human antibody light and heavy chain variable region framework regions and
DR's are rat LO-CD2a, CDR's. In one embodiment,
One or more amino acid residues in the framework regions of the light and heavy chain variable regions are from the rat LO-CD2a framework region. Residues from such rat framework regions are retained in the humanized antibody,
This is because such residues maintain the binding specificity of LO-CD2a.
Thus, in producing humanized antibodies, and in accordance with a desirable aspect of the invention, the CDR's of human antibodies include LO-CD2a from FR1, FR2 and FR3, among others.
Certain amino acids in the light chain variable portion of L-CD2a and especially those in the heavy chain variable portion of LO-CD2a from FR-2 and FR-3 have additional factors retained in constructing humanized antibodies. Replaced by CDR's of LO-CD2a. That is, the corresponding amino acids of the human framework are replaced with the indicated amino acids from the rat LO-CD2a framework. As pointed out in connection with FIG. 39, amino acids 9, 1 in the framework of the light chain variable region of rat LO-CD2a
Amino acids 4 in the framework of the heavy chain variable region of rat LO-CD2a are retained, and as indicated in FIG. 43, 2, 41, 42, 50, 51 and 82.
7, 67, 70, 72, 76, 85 and 87 are retained by the humanized antibody. A specific example of the construction of such a humanized antibody is given in Example 7 below.

In another embodiment, in the heavy chain variable region of the humanized antibody, FR1, FR2
And certain amino acids of the FR3 region are retained in constructing humanized antibodies.
55 and 56, amino acids 47, 67, 70, 72, 76, 85 and 87 and also amino acids 12, 13, 28 and 48 in the framework of the heavy chain variable region of rat LO-CD2a. 1, 2, 3, or 4 are retained by the humanized antibody. In a preferred embodiment, rat LO-CD2
amino acids 12, 13, 28, 47, 48 in the framework of the heavy chain variable region of a
, 67, 70, 72, 76, 85 and 87 are retained by the humanized antibody. In another preferred embodiment, in addition to retaining the amino acids in the framework of the heavy chain variable region of rat LO-CD2a in the humanized antibody, the humanized antibody further comprises a framework of the light chain variable region of the humanized antibody. As shown in FIG.
One or more amino acids 9, 12, 41, 4 of the CD2a light chain variable region
2, 50, 51 and 82. Desirably, each amino acid 9, 12, 41, 42, 50, 51 and 82 in the light chain variable region of rat LO-CD2a is retained by the humanized antibody.

In the most preferred embodiment, the humanized antibody contains rat LO-CD2a as shown in FIGS. 55 and 56 within the framework of the heavy chain variable region of the humanized antibody.
Amino acids 12, 13, 28, 47, 48, 67, 70, 72 of the heavy chain variable region of
76, 85 and 87. The humanized antibody also incorporates amino acids 9, 12, 41, 42, 50, 51 and 82 of the rat LO-CD2a light chain variable region within the framework of the humanized antibody light chain variable region as shown in FIG. Contains.
Such an antibody is sometimes referred to hereinafter as MEDI-507, the structure of which is described in Example 11 below.

[0035] In another embodiment, the invention relates to a human constant region and rat LO-CD2a.
And a use thereof.

The antibodies or molecules of this invention desirably (i) bind to all T lymphocytes and null cells but show B lymphocytes, as indicated by two-color staining of lymphocytes analyzed by flow cytometry. Does not bind to the sphere (FIGS. 29, 30 and 31)
, 32); (ii) (as determined by staining with anti-CD3 antibody Leu4) all T cells, all CD4 and CD8 positive cells as defined by respectively Leu3a and Leu2b antibodies, and CD ₃ negative Binds to some lymphocytes that are (null cells); (iv) Binds to null cells as evidenced by staining of CD16 positive cells detected with Leu11, a marker for NK cells. (FIG. 29
, 30); staining of B cells as defined by anti-CD19 binding
Not seen with 2a. (FIGS. 31, 32). The LO-CD2a antibody also desirably has the property of binding to human null cells, and further stains human cells that are both CD2 + and CD4 + by bicolor staining rather than human cells that are both CD2 + and CD16 +. And has a higher intensity of staining of human cells that are both CD2 + and CD8 + than to human cells that are both CD2 + and CD16 +.

[0037] LO-CD2a binding to CD2 was confirmed by transiently expressing CD2 in COS cells.

COS cells are derived from Peterson A. And seed B. , Nature 329, 19
Transfected transiently with the πH3MCD2 plasmid containing the gene encoding the entire CD2 molecule, as described on Oct. 29, 1987, 842-846.

Transfection was achieved with the DEAE-dextran method. Cells were harvested and stained with the anti-CD2 monoclonal antibodies Leu5b (Becton-Dickinson) and LO-CD2a, along with murine W632, an antibody to MHC class I, and the corresponding isotype-matched control as positive controls for staining. The specificity of the reactivity was demonstrated by assessing the binding of the same panel of monoclonal antibodies to COS cells transfected with an unrelated plasmid.

The staining pattern of these monoclonal antibodies against the transiently expressed native CD2 (FIG. 33) shows that transfection with CD2 led to binding of both antibodies and LO-CD2a
Supported the ability to bind to CD2.

The preparation of an LO-CD2a monoclonal antibody suitable for the purposes of the present invention must be apparent to those skilled in the art from the teachings herein.

Antibodies or fragments or derivatives or molecules of the type described above inhibit T cell activation and proliferation and also reduce the density of CD2 expression on the cell surface, thereby reducing CD2 ⁺ it can be administered in vivo to reduce the number of T lymphocytes.

Thus, for example, in an in vivo procedure, such LO-CD2a antibodies are administered to prevent and / or inhibit an immune response, thereby inhibiting T cell activation and proliferation.

[0044] Antibodies or fragments or derivatives or molecules thereof of the type described above may be used in vivo to reduce the density of CD2 ⁺ expression on the cell surface and thus the number of CD2 ⁺ cells of donor cells in accordance with the invention. It can be administered outside . By way of example and without limitation, in an in vitro procedure, such an antibody or fragment or derivative or molecule thereof may be administered to a donor bone marrow prior to transplantation to prevent the development of graft-versus-host disease during transplantation. Will be injected.

In such in vivo or in vitro techniques, the antibody or fragment or derivative or molecule thereof will be administered in a pharmaceutically acceptable carrier. Representative examples of such carriers include isotonic saline solutions, buffers and the like. Such pharmaceutically acceptable carriers are well known in the art, and the selection of a suitable carrier is deemed to be within the skill of the art from the lessons contained therein.

The LO-CD2a antibody or other molecule of the present invention is administered by intravenous or intramuscular administration in vivo .

As indicated above, the LO-CD2a antibody or other molecule of the invention is administered in vivo in an amount effective to inhibit graft rejection. The term "effective amount" for the purposes of this application shall mean the amount of a monoclonal antibody capable of producing the desired effect, i.e., inhibition of graft rejection or inhibition of T cell activation. . Generally, such antibodies will be administered in an amount of at least 1 mg. In addition, after the first treatment, the amount described above is reduced during subsequent treatments if none. Thus, the scope of the invention is not limited by such amounts.

According to this example, such antibodies are administered to maintain T cell activation and inhibition of graft rejection. Thus, by way of example and not by way of limitation, the antibodies may be administered in a physiologically acceptable carrier suspension in an amount of about 1 mg / dose to about 50 mg / dose once or twice daily, if necessary, for 1-2 hours. It is administered by intravenous infusion over a period of about 8 days or more. Such treatment for graft rejection is desirably at the beginning of the transplant, or just before or immediately after the transplant, or when graft rejection occurs. Treatments could be given once or twice daily for as little as one or two days when started at the time of transplantation to introduce a selective low response state to the graft. In accordance with the present invention, such treatment for an autoimmune disease associated with the administration of the antibody or fragment or derivative or molecule thereof is initiated when the treating physician determines that it is desirable to inhibit the pathological immune response.

Thus, according to one aspect of the present invention, by administering an antibody according to the present invention at the time of transplantation and most often for a short period thereafter, it is possible to introduce a hypo-responsiveness into the transplanted tissue or organ, Can prevent or inhibit the appearance of repetitive symptoms of further rejection.

The techniques of this invention for inhibiting T cell activation may be used alone or in other ways to inhibit T cell activation or inhibit graft rejection or graft versus host disease. It can be used in combination with technology, drugs or compounds.

The present invention is further described in connection with the following examples, which are illustrative but not intended to limit the scope of the invention.

The cells, cultures, mAbs and mitogens used in the examples are prepared and used according to methods and procedures known and practiced by those of ordinary skill in the art. The following are used in the preparation and use of cells, cultures, mAbs and mitogens used in subsequent examples.

Cells and cultured PBMC were obtained by Ficoll-Hypaque (Pharmacia, Sweden) sedimentation of heparinized blood obtained from a local blood donation center. The isolated PBMCs were enriched in a medium: penicillin 100 U / ml, streptomycin 100 μl
g / ml, L-glutamine 20 mM, and 20% pooled human AB serum or 1
Resuspended in RPMI 1640 medium (Jibco, Belgium) supplemented with 5% heat-inactivated fetal calf serum. PBMC were cultured at 1 × 10 ⁵ cells / well in a 96 U-well microplate (Falcon) to a final volume of 200 μl culture medium / well. Two-way MLC was performed with 1 × 10 ⁵ cells in each donor / well with the same volume of culture medium as described above. All cultures were made in triplicate. Eight hours before the time point indicated in the results, the culture was 2.0 μCi / T of 3 _H- T (Amersham, Belgium; 247.9 GBq / mmol; 6.7 Ci / mmol).
Radioisotopes that were pulse-labeled in the wells and incorporated into the culture were weighed by liquid scintillation on a beta counter (Beckman L5 6000SE). The percentage of inhibition was calculated as follows:% inhibition = [1- (mean cp of test culture)
m / mean cpm of control culture)] x 100. All results are expressed as the average of three individual cultures. Standard deviations were always less than 15% of the mean, except where those values were shown graphically.

[0054] Cytofluorometric analysis was performed using a Faxscan cytofluorograph (Becton Dickinson) on Hewlett-Packard hardware with the Consort 30 program. Separate analysis for lymphocyte and blast staining was enabled using differential gating defined by size and granularity. 25,000 events were analyzed in each sample. In these experiments, the final concentration of LO-CD2a was 200 ng / ml except where indicated.

Mabs and mitogen LO-DRA and LO-Tact-1 (both FITC-labeled)
Are rat mAbs produced in our laboratory (public reference, H. Bazan (
Edited in 1990, p. 287). LO-Tact-1 is the IL-2 receptor p
Directed to the 55 chain (public reference, H. Bazan, Immunology, 1984, and Jansen, M., Buck, D. and Maynow, VC, leukocyte type IV).
, A leukocyte differentiation antigen; Nap (ed.), Oxford University Press, 1989, p. 403). Mouse anti-human-CD2 and anti-CD3 mAbs
s (Leu-5b and Leu-4a-FITC labels) were obtained from Becton Dickinson (Belgium). Mouse anti-human CD4 or anti-human CD8 mAb
s (phycoerythrin label), and mouse IgG FITC- or phycoerythrin label (negative control) were obtained from Coulter. OKT3 (Ortho-Syrag, Belgium) was used at a final concentration of 100 ng / ml. Phytohemagglutinin A (phytohemagglutinin) (PHA: Welcome Labs, UK) and Concanavalin A (ConA; Calbiochem Company, USA) were used at final concentrations of 1 and 10 μg / ml, respectively.

Biotinylation of LO-CD2a . The concentration of purified LO-CD2a was adjusted to 1 mg / ml with 0.1 M sodium bicarbonate buffer, pH 8.4. NHS-biotin (Boehringer Mannheim 1008 960) at a concentration of 1.5 mg / ml D
Dissolved in MSO. For each MAB, 0.1 ml of NHS biotin solution was added. The mixture was rotated at ambient temperature for 2 hours. Reaction was 2M for each ml of antibody
Completed by adding 0.1 ml of Tris-HCl, pH 8.0 (10 minutes at ambient temperature), and then adding 1 ml of 1% BSA in phosphate buffered saline (PBS) to each ml of antibody. . To remove free biotin, the solution is incubated overnight at 4 ° C.
Dialyzed against 000 volumes of PBS. Both the biotinylation reaction and the conjugated mAb were wrapped in aluminum foil and hidden from light.

Lysis of red blood cells (RBC) . RBC was removed from whole blood by lysis with ammonium chloride. 10X stock solution _{NH 4 C1,90g, KHCO 3, 10g} , E
It was prepared to consist of 370 mg of DTA, and water to give an amount of 100 mls.
40 mls of 1X ammonium chloride was added to each 10 mls of blood and incubated for 10 minutes at room temperature. The mixture was then centrifuged at 1200 rpm for 10 minutes and the pellet was resuspended in 0.1 ml of azide in 10 ml of PBS.

Staining of peripheral blood Staining was performed at 4 ° C. on a 96-well cluster plate with round bottom (Costar # 3790).
Made in For single color staining, 10 μl of the mAb was appropriately diluted in PBS containing 0.2 mg of human immunoglobulin and added to each well. Erythrocyte depleted blood was distributed on plates in a volume of 90 μl per well. Cells and mAb were mixed by gentle tapping and incubated for 30 minutes. 50 μl of cold PBS was added to each well, and the plates were centrifuged at 1900 rpm for 2 minutes. The supernatant was removed by inversion and gentle tapping of the plate. Cells were dispersed by tapping the plate with a counter. The washing procedure was repeated twice with the addition of 200 μl of cold PBS. 10 μl of a 1/20 dilution of goat F (ab ′) ₂ anti-rat 1 g-FITC was added to the dispersed cells in each well and incubated for 30 minutes in the dark. Cells are 180 μl of cold PBS
Was added to each well and then centrifuged at 1900 rpm for 2 minutes.
The supernatant was removed, the cells were dispersed, and 200 μl of 0.5% cold paraformaldehyde was added to each well. Cells were transferred to tubes (Falcon # 2054) and diluted to about 0.5 mls with 0.5% paraformaldehyde. Samples were evaluated on a Becton-Dickinson Facscan machine using lysis II software.

Two-color staining was performed with a similar experimental record. Cells contain primary mAb and FITC
-Incubated with conjugated anti-rat reagent, 1/5 dilution of standard mouse serum was added to block any remaining sites of the anti-rat reagent. Following a 15 minute incubation (no washing), 20 μl of PE-labeled mAb specific for known CD determinants was added,
It was kept warm for 30 minutes. Cells were washed and fixed as described for single staining.

Example 1 LO-CD2a is a rat (IgG2b-kappa) anti-CD2 monoclonal antibody produced and characterized in our lab, as indicated elsewhere in the literature (cited below). See: Thea, H., Lavette, A.M., Ratanne, D., Ninanne, J., De Bruyere, M., Socal, G. and Bazan, H., -H. ed.), rat hybridomas and Rattomonoku polyclonal antibody, CRC Press ,, Inc., Boca Raton, 19
90 Florida, p. 309; M. , Latanne, D.E. ,
Segars, J.M. Manu Breeze, P.M. Ninanne, J .; , De Bruyere, M.E. Bazan, H .; And Socal, G .; , -H. Bazin (eds), rat hybridomas and rat monoclonal antibodies , CRC Press, Inc., Boca Leighton, 1990 Florida, p. 287). LO-CD
2a was purified from ascites by immunoaffinity chromatography, taking advantage of the allotypic differences that exist between the immunoglobulin of the rat receiving the produced hybridoma and the mAb secreted by the latter (Bazan, H., et al.
Cormont, F. And Declark, L.A. Journal of Immunology Methodology, 1984
Year 71: 9). This is the mouse mAb Leu-5b (FITC label)
, Approximately 90% of the population labeled with the mouse T11 (rhodamine-labeled) mAb (data not shown). CD2
The epitope recognized by LO-CD2a on the molecule is different from the epitope recognized by anti-CD2 mouse mAbs Leu-5b and T11 (FIGS. 2 and 3).
).

Example 2 LO-CD2a Modulates PBMC but Does Not Mitogen In order to determine the effect of rat mAb LO-CD2a on resting lymphocytes, PBMC is expressed in increasing concentrations of this mAb. Incubated in the presence of As seen in Appendix 1, PBMC which were incubated for 6 days in the presence of LO-CD2a show no significant changes in the rate of ³ H-T tori inclusive as compared to the control cultures. Cell viability at the end of this period varied, but averaged about 80% as assessed by trypan blue exclusion. When resting PBMCs are incubated in the presence of LO-CD2a,
As assessed by flow cytometry, there was no significant change in phenotypic expression of some membrane markers. Cell markers of resting mature T cells (eg, CD3, C
D4 and CD8) show the same pattern of change in the presence or absence of LO-CD2a after 6 days of culture, and activating molecules such as CD25 (IL-2R / p55) are expressed in these experimental conditions. It is not or is not modified by LO-CD2a as in the case of the DR antigen determinant (FIG. 4).

When PBMC were incubated in the presence of LO-CD2a for 6 days, a significant decrease in the percentage of Leu-5b + gated lymphocytes was observed (FIG. 5). CD4- and CD
The proportion of 8-lymphocytes was unaffected by the presence of LO-CD2a during 6 days in culture, since the observed decrease in CD2-bearing lymphocytes could not be attributed to the elimination of these cells, but rather the disappearance of CD2 molecules. Alternatively, it indicates that this is due to a conformational change in this glycoprotein produced by the binding of LO-CD2a.

The observed decrease in Leu-5b-lymphocytes was due to a decrease in C after LO-CD2a binding.
To confirm whether this was due to a conformational change in D2 or to the loss (internalization or release) of this molecule, PBMC were cultured in the presence of 500 ng / ml of LO-CD2a and Leu-5b (FITC label), T11
-Analyzed by flow cytometry using RD1 (rhodamine label) and MARK-3 (FITC label) for 6 days. As shown in FIG.
-5b or T11 mAbs cannot bind to PBMC after 2-4 days of culture in the presence of LO-CD2a. Under these conditions, the mouse anti-rat kappa chain mAb
MARK-3 labeled 50% of the cells on day 6 of culture, which showed that only 35% of the original CD2-bearing cells showed no LO-CD2a on their surface, but Leu-
5b-FITC and T11-RD1 staining was significantly reduced on day 2. This suggests that a conformational change in CD2 supplying the Leu5B and T11 epitopes unavailable for binding occurs in response to LO-CD2a.

Analysis of the mean fluorescence of CD2 + cells indicates that the expression density of this marker is LO-CD2a
Decreased over time in the presence of. A similar phenomenon was observed with Leu-5b FITC labeling or LO (as revealed by MARK-1 FITC labeling).
-It was observed whether any of CD2a was used to detect CD2 + lymphocytes. An aliquot of the same PBMC is Leu-5b (final concentration 1 in culture medium).
: 2 in the presence of commercially available mAb) dialyzed against PBS. As shown in FIG. 5b, under these experimental conditions, all CD2-bearing cells were Leu-5b mAb (as revealed by goat anti-mouse-FITC).
Coated. Staining with T11-RD1 was significantly reduced, while a smaller and slower decrease was observed in the percentage of cells submitting the epitope recognized by the LO-CD2a-FITC mAb. Summarizing these results, C
The D2 molecule partially changes its conformation in response to LO-CD2a and
This shows that gradual regulation of D2 / LO-CD2a occurs.

LO-CD2a is measured by (3H-thymidine ( ³ H-T) incorporation when MLC is performed (for a period of 6 days or more) in the presence of increasing concentrations of rat mAb that inhibits MLR. MLR)
Was observed at mAb concentrations as low as 125 ng / ml. FIG.
In, we show a typical example of a dose-response curve of MLR inhibition by LO-CD2a. As can be seen in FIG. 6a, LO-CD2a is 250 ng.
/ Ml induces 80% inhibition of the MLR (day 6 of culture), with the rate of inhibition remaining almost constant or over 80% over a wide range of concentrations (mAbs from 0.25 to 5.0 μg / ml). . FIG. 7 shows LO-C for MLR from day 0 to day 6 of culture.
The time course of the inhibitory effect of different concentrations of D2a is shown. A typical example of ³ H-T incorporation on MLC (with or without LO-CD2a) is shown in FIG. 6c, where LO-CD2a was added at a final concentration of 200 ng / ml.

FIG. 7 shows the effect of LO-CD2a on MLR when this mAb (at 200 ng / ml) is added at different times after the start of MLC. ( ³ H-
More than 90% inhibition of the MLR was obtained when the mAb was added at day 0, and the inhibitory effect was 4 days after the start of MLC, as measured by T incorporation).
-There is still an inhibitory effect (45% in this example) when CD2a is added. Similar results (not shown) were obtained for LO-CD2a (0.20-5.
Higher concentrations (of 0 μg / ml).

LO-CD2a Blocks IL-2R Expression Pathway When cytofluorographic analysis was performed on lymphoblast subsets of MLC (
8a and 8b), the following observations were made: a) The number of blasts already present at the start of the MLC (approximately 300-500 blasts of 25,000 events analyzed) was 4 days in the control culture Increased sharply from day 2 to day 6 (more than 1200 blasts from 25,000 events analyzed); b) MLC performed in the presence of LO-CD2a, marked changes in the number of blasts during the entire culture And on day 6 the blast count is constantly lower or almost the same as the initial blast count on day 0 (FIG. 8a); c) CD
The percentage of 25 blasts increased sharply in cells incubated without LO-CD2a (FIG. 8b); d) This percentage increased among a small number of blasts from MLC incubated in the presence of mAb. Stayed below 20% (FIG. 8b) and the mean fluorescence (as a measure of CD25 expression) was reduced by 75% compared to blasts present in control cultures (results not shown); e) mAb In the absence of, the percentage of CD3-blasts remained constant during the first 4 days of culture (FIG. 8b) and at day 6 the percentage of CD3 cells was 90%.
%, While in the presence of LO-CD2a, the proportion of C3- slowly increased and reached only about 45% at 6 days. These results indicate that the presence of LO-CD2a inhibits the entry of these cells through a pathway of activation characterized by expression of the IL-2 receptor (CD25). The number of CD2 + blasts is LO-CD2
a remains constant or decreases in the presence of a, and the expression density of this membrane marker decreases strongly under these conditions (data not shown).

When phenotypic analysis was performed on resting ( non-blast ) lymphocyte subsets of MLC over 6 days, results similar to those shown in FIG. 4 were obtained: in the presence of LO-CD2a No significant change was detected in the proportion of CD3 +, CD4 + or CD8 + lymphocytes compared to control cultures; LO-CD2a
CD25 expression (activation marker) could not be detected irrespective of the presence or absence of (Fig. 9). These results suggest that LO-CD2a has very little, if any, effect on the resting subset of T lymphocytes in MLC; that is, T cells are not committed during activation. At the same time, LO-CD2a induces a significant decrease in the proportion of CD2 + lymphocytes during MLC, as shown in FIG. 8b. These results indicate that LO in both unstimulated culture (FIG. 6) and MLC
The action of -CD2a is to reduce the expression of CD2 and / or to induce a conformational change in its structure.

LO-CD2a can block the TcR / CD3 complex or mitogen receptor-dependent T cell activation pathway.

[0070] When LO-CD2a was added to mitogen-activated PBMC, a significant inhibition of lump tori ³ H-T was observed. In the presence or absence of LO-CD2a added either 0 hours or 1 hour after the start of culture, mitogen (OKT3, CoT3
nA and PHA) was one of three experiments with PBMC incubated. In the first case, the mitogen was added after one hour. Mitogen was added at 0 hour when LO-CD2a was added 1 hour after the start of culture. This is mitogen or LO
-Pre-incubation of PBMC with CD2a was performed to see if the addition of secondary reagents could trigger events that could be affected. Pulsed cultures with ³ H-T (6 hours)
After 96 hours. ³ H-T 50% or higher inhibition of lump tori whether it is added after the first or mitogen, was observed in the presence of LO-CD2a (Figure 11). The same effect was observed when cells were harvested and exposed to mitogen 4 days after the start of MLC (data not shown). Compared to the same culture accept mitogen only, in cultures receiving both the mitogen and LO-CD2a, a dramatic reduction in the included tori ³ H-T was observed after the start 2 days MLC (
Results are not shown). Mitogen insulation previous MLC of additional previous LO-CD2a is lowered to narrow ³ H-T-taken up in comparable values at the MLC without OKT3.

If LO-CD2a was also added one day after the start of mitogen-induced proliferation,
Mitogen-induced proliferation could be inhibited. The results of an experiment performed with two donors are shown in FIG. PBMC is mitogen (OKT3, ConA and PHA)
Was kept warm. In these experiments, LO-CD2a was added at 0 (day 0), 24 hours (1 day) or 48 hours (2 days) before the start of culture. LO-C
The inhibition of proliferation in response to OKT3 and ConA by D2a was significant if added at 24 hours after an additional 24 hours of mitogen.

Example 3 Inhibition of Natural Killer Cell (NK) Activity PBMC were separated from heparinized blood by Ficoll-Hypaque sedimentation. After washing, the effector cells suspended in the enrichment medium were incubated overnight at 1 × 10 ⁶ / ml on Falcon plates to remove mononuclear cells (by attachment).

The target cells (K562 cell line) were incubated overnight with ⁵¹ chromium ⁵¹ Cr (0.9 ml of cell suspension at 3 × 10 ⁶ /ml+0.02 ml from a solution of 5 mCi / ml ⁵¹ Cr, Amersham). Was labeled by

After incubation for 16 hours, the effector cells and target cells were washed four times, counted and different E / T ratios (effector cell / target cell ratio): 200/1 (4 × 1
0 ⁶ / ml effector cells suspension 100ul to 2 × 10 ⁴ / ml target cells 1
00 ul), 100/1, 50/1 and 25/1, incubated in 96 V bottom microplates.

After incubating for 4 hours, ⁵¹ Cr release was measured using a gamma counter to remove supernatant 1 from each well.
It was determined by counting 00 μl.

The maximum (target cells + HCLIN) and spontaneous release (target cells + enriched media) were used to calculate the specific solubility:

[0077]

(Equation 1)

[0078] In the NK assay with two normal donors, 5 ug / ml, 1 ug / ml, and 0.
The cell content of LO-CD2a at 5 ug / ml (FIGS. 10a and 10b) was approximately 50% over all tested concentrations of antibody and all tested E / T ratios.
% Inhibition of cytotoxicity. This compares to essentially complete inhibition of growth at 0.25 ug / ml and higher doses of MLR.

[0079] In vivo studies Materials and Methods Monoclonal Antibodies MARK3-FITC in Example 4 nonhuman primate is a mouse mAb directed against the rat Ig kappa 1b allotype joined with FITC. MARG2b-Biotin is a mouse anti-rat IgG2b immunoglobulin mAb conjugated with biotin. These two
MAbs were produced and labeled in our lab. For immunofluorescence studies, these mAbs were used at a final concentration of 2.5 μg / ml. Leu-5b-F
ITC (Becton-Dickinson) and T11 Rhodamine (Coulter) are 2
Mouse anti-human CD2 mAbs. T4- and T8-rhodamine labels (
Coulter) are mouse anti-human CD4 and CD8 mAbs, respectively.

Phenotypic Analysis Anti-human T-cell mAbs (anti-CD2, -CD4, -CD8, see above) were added to a 100 μl sample of whole blood and incubated at 4 ° C. for 45 minutes. Erythrocytes are in Tris-buffered ammonium chloride lysis buffer (ammonium chloride 144 mM / Tris 17 mM)
, PH 7.2) and the lymphocytes were PBS / FCS2% / NaN3, 0.2
Washed in%. For detection of unlabeled mAbs, a second mAb (FITC- or biotin-conjugate) was added to a final concentration of 2.5 μg / ml. 4 at 4 ° C
After incubation for 5 minutes, the cells were washed with PBS / FCS / NaN3. Biotinylated mA
For bs, a warming (15 minutes) with a streptavidin-phycoerythrin conjugate was performed. Labeled human or monkey lymphocytes were resuspended in a 2% formalin solution and analyzed on a Faxon cytofluorometer (Becton Dickinson) equipped with a lysis II program to gate lymphocytes as a function of size versus particle size. . Aliquots of cells were incubated with FITC- or phycoerythrin-conjugated mouse Igs (Coulter) as controls for non-specific staining.

Levels of Circulating Abs Serum LO-CD2a was modified with horseradish peroxidase with mouse anti-rat 1gG2b mAb (MARG2b-8 produced in our laboratory) as the first layer (coating) and for detection. Bound mouse anti-rat kappa chains (MAR
K-3) Quantified by ELISA using mAb. Briefly, microtiter plates (Falcon) were incubated overnight at 100 μL / well of MARG2b-8 (5 μg / ml) and the unoccupied area of the plastic was saturated with PBS containing 5% milk powder (bovine). After incubation for 1 hour at room temperature, the plates were washed with PBS containing Tween-20, 0.1% and incubated with diluted monkey or human serum at 100 μl / well. After washing out the unbound material, the plates were washed for 1 hour at 100 μl / well MAR.
Incubated with K3-peroxidase (2 μg / ml in PBS). After washing again, the plates are OPD (citrate-phosphate buffer containing 0.03% hydrogen peroxide).
(O-phenylenediamine-dihydrochloride, 0.4 mg / ml, Sigma Chemicals). The staining reaction product was detected at 492 nm. A standard curve was constructed in parallel with known concentrations of purified LO-CD2a serially diluted in a pool of control monkey or human serum.

For detection of monkey or human anti-LO-CD2a antibody, LO-CD2a (5 μg /
ml) coated with a 96-well microtiter plate. Anti-LO-CD2a human or monkey antibodies binding to the plate were revealed by horseradish peroxidase labeled with rat anti-human IgM (LO-HM-7) or IgG (LO-HG-22) mAbs.

A. Cynomolgus monkey One Cynomolgus monkey received LO-CD2a at 10 mg / day for three consecutive days. The monoclonal antibody was well resistant.

Lymphocyte loss was observed after the first injection, but there was very little additional loss after the second and third injections.

A second monkey received 20 mg / day for 10 days. The mAbs were also well tolerated. No side effects were observed after dose administration, in which the animals were active, cautious and had good food with no evidence of nausea or gastrointestinal upset.

The total monocyte lymphocyte count and cell population of the second monkey are summarized in FIGS. NK activity decreased slightly after 10 injections (FIG. 17). The circulating levels of MAb were very high (FIG. 18) and immunization occurred at the end of treatment (FIG. 19).

B. Baboon Determines baboon resistance to LO-CD2a, analyzes the effect of this mAb on some of the membrane markers of baboon lymphocytes, and
The experiment described here was started to determine the half-life of

Staining of baboon cells with LO-CD2a results in less than 20% positive cells at a significantly lower average fluorescence intensity than that of the stained human cells. This staining pattern reflects weak cross sensitivity or binding via Fc interaction with baboon cells.

This study was performed on a baboon (Papio mormon) weighing 8.8 kg. Prior to each injection of LO-CD2a, monkeys were anesthetized. Anesthesia was first with keteller (2 ml) and prazin (0.5 ml), and the second with keteller only, followed by keteller and prazin (0.3 ml). LO-CD2a was diluted in 100 ml of saline and injected intravenously (iv 10 minutes).
Phenotypic analysis of lymphocytes and circulating antibodies (injected LO-CD2a, newly formed anti-LO-CD2a antibodies, and pre-existing cross-reactive baboon anti-LO
Blood samples (10 ml) were taken in two tubes for the measurement of (-CD2a antibody). Tubes for lymphocyte typing contained EDTA. Samples were taken before the first treatment to determine baseline levels.

The first dose of LO-CD2a (10 mg) was administered on day 0 of the study. 4 following
Individual doses (10 mg / dose) were administered on days 7, 8, 9, and 10. Blood samples were taken 2-3 minutes after each LO-CD2a dose. On the 7th and 9th (
Supplemental blood samples were also collected prior to LO-CD2a injection (in tubes containing EDTA). Blood samples were taken on days 1, 2, 11, 12, 13, 16 and 24.

No abnormal activity response or dietary habits were observed during the LO-CD2a injection or throughout the study. Animal weight was approximately 8.8 measured on day 0.
Kg (see table below).

Baboon Weight from Day 0 to Day 24 (Kg) Day 07 89 9 10 12 13 16 24 Body Weight 8.8 8.9 9.1 9.1 8.8 9.0 9.1 8.9 8.9 Phenotype and Analysis of Circulating mAbs Peripheral Blood Lymphocytes of this Baboon Fluorescent staining revealed several interesting properties: a) Under the action of LO-CD2a, the CD2-positive subset of lymphocytes
-At the end of the fifth dose of CD2a, i.e. at the time of maximum accumulation of mAbs in the blood (as revealed by the two different anti-CD2 mAbs) (Fig. 20a and 20b Please see). CD4 + and C in lymphocytes
Assuming that the D8 + subset does not decrease during this period (FIG. 22),
Because + and CD8 + cells contain most CD2-bearing lymphocytes, a decrease in CD2 ⁺ cells indicates that the conformation is reduced or altered in membrane marker expression over lymphocytes.

As can be seen in (FIG. 20a), after the first dose of LO-CD2a, C
A slight decrease in D2 + (Leu-5b ⁺ or T11 ⁺ ) positive lymphocytes is observed. Two days after this first dose, the levels of CD2 positive cells (T11 ⁺ Leu5b ⁺ ) rose to their starting values.

End of 4 24-hour spaced doses of LO-CD2a (7-10 days)
The proportion of CD2 positive cells (Leu5b ⁺ or T11 ⁺ ) sharply decreased and started to rise slowly 3 days after the end of LO-CD2a administration.

B) At the same time, the proportion of LO-CD2a positive cells, ie the proportion of cells bound by circulating mAb, rises to 22% after the second dose of LO-CD2a (7 days),
Then the CD revealed by anti-CD2 mAbs Leu5b and T11
²⁺ cells decreased as they did (Figure 20). Reduction of LO-CD2a + cells was revealed by the MARK-3 FITC mAb (FIG. 21).

[0096] LO-CD2a + cell reduction was due to MARK3-FITC or MA
LO-C present in cells detected by RG2b-8-biotin conjugated mAbs
Determined by detection of D2a (FIG. 21a). The same phenomenon occurs if cells circulate
It was observed if the site was first incubated with LO-CD2a at 2.5 μg / ml to saturate all sites not occupied by Ab. LO-CD2a is MA
It was detected by RK-3-FITC or MARG2b-8-biotin (FIG. 21).

C) As can be seen in (FIG. 22), the T4 positive subset of baboon lymphocytes shows a modest rise from day 9 to day 12, after which the percentage of T4 ⁺ lymphocytes is reduced to the original value Returned to. Concomitant with the rise in T4 ⁺ cells, the proportion of T8 normal lymphocytes rose from day 9 to day 11. After that date, the percentage returned to its original value.

D) The level of circulating mAb LO-CD2a decreased to inconspicuous values 3 days after the first injection (see FIG. 20b). When LO-CD2a is applied in four short-interval doses (7 to 10 days), serum LO-CD2a levels (approximately 3.7 mg)
/ Ml (maximum during this period) decreased slowly after the last dose (10 to 16 days), indicating a relatively long Ab half-life in this animal model. No baboon anti-LO-CD2a antibody was detected in blood samples taken on days 11, 12, 13, 16 and 24.

Conclusion LO-CD2a appears to be more tolerated by non-human primates, as indicated by the absence of an apparent response in cynomolgus baboons. LO-CD2a appears to have a relatively long half-life in baboons. Twenty-four hours after the first dose of LO-CD2a (day 1 in FIG. 27), 50% of the maximum detected level of MAb was still present in the serum. 3 days after the last dose of LO-CD2a (day 13 in FIG. 27)
, 50% of the maximum detectable level of the mAb was still present in the serum.

Although the staining pattern in non-human primates is not consistent with that observed for CD2 on human cells, a gradual increase in this percentage of cells following a decrease in the percentage of CD2 normal lymphocytes indicates that LO-CD2a Similar to that observed in human PBMC mononuclear cells cultured in the presence.

Example 5 Patients Treated with LO-CD2a Patients Treated with LO-CD2a on a Compassionate Basis Patient # 1 (Mb.E.) This is renal allogeneic due to end-stage renal failure A female patient with chronic pyelonephritis who was treated for a systemic transplant. A rejection episode occurred and was treated with OKT3 for 10 days. Creatinine levels dropped from 2 to 1.4 mg / dl. Approximately four months later, the rejection episode was diagnosed by a biopsy showing a creatinine level of 2 mg / dl and moderate rejection. The patient was treated with 1.5 g of Solmedrol and a course of ATG for the next 8 days, at which time creatinine levels were 1.65 mg / dl. 7 days after treatment
A biopsy was performed, which showed cell rejection and moderate vascular rejection. Two days after the biopsy (day 0), the patient became anuria with a creatinine level of 2.4 mg / dl. On the same day, the patient was LO-CD2a, 10 mg, corticosteroid / solmedrol 1.5.
g, plus, 1 g of polarimin (antihistamine) and 1 g of dafalgan (acetaminophen). Treatment with corticosteroids, dexchlorophenylamine and acetaminophen is "covered" by the transplant community (covera
ge). No side effects were observed. By the end of 23 hours,
The patient excreted 700 ml of urine and creatine was 2.72 mg / dl. In the next 9 days she received 10 mg / day of LO-CD2a. The patient left the hospital without a follow-up biopsy at 11 days.

Measurement of serum creatinine levels during ATG treatment and subsequent LO-CD2a treatment showed that creatinine levels increased despite ATG treatment and then decreased to stabilize with LO-CD2a. (FIGS. 31 and 32).

The leukocyte count dropped from a peak of 10,000 to 2000 during treatment with LO-CD2a and continued to drop until the final measurement on day 21 (FIG. 23). Lymphocyte counts were low and changed during the observation period.

The serum levels of LO-CD2a were immediately 2.0-3.0 μg / ml following each treatment.
2 and decreased to a minimum of about 1.0 μg / ml between treatments (FIG. 2).
4). With the last treatment on day 9, this level dropped by 50% in 24 hours and fell to zero on day 14. The patient returned to the outpatient forty days.

The patient's creatinine level was 2.27 at 40 days, 2.48 at 50 days, and 6
At day 6, it had risen to 3.11, at which point a biopsy was obtained with the first report consistent with severe cell rejection and interstitial bleeding (see below). Patient has 150 kidneys
Irradiation, treatment with 3 × 125 mg solmedrol, followed by drug maintenance therapy with cyclosporine and 12.5 mg / day steroid. Creatinine levels continued to rise during the following period. At 70 days the creatinine level was 3.3,
It was 5.63 on the 80th and 8.35 on the 84th. At day 86, creatinine levels were 10.8, and a transplant nephrectomy was performed at day 88. The compliance of this patient with maintenance immunosuppression for the period between the 10-day excretion and the 66-day biopsy is questionable and a factor in compliance with uncertain renal loss despite apparently successful rescue. Must be.

(Ii) Patient # 2 This patient was a 38 year old man with hepatitis C. He received a renal allograft for the treatment of end-stage renal failure due to chronic interstitial nephropathy (nephropathy). 1
Three months later, he underwent a transplant nephrectomy for a series of acute cells resistant to OKT3 and vascular rejection.

One year and ten months after the transplant nephrectomy, he underwent a second renal allograft. After three days, his creatinine level was 1.4 mg / dl. Three days later he received 500 mg of Solmedrol. After that day the patient's creatinine was 1.8 mg / dl
Met. The next day he received 500 mg of Solmedrol. Creatinine is 3
. It was 25 mg / dl. The next day he received 500 mg Solmedrol. His creatinine was 2.95 mg / dl. Three days later his creatinine level was 2.3 mg / dl and he underwent a biopsy, which showed 3 plus cellular rejection. Three days later he received LO-CD2a, 10 mg, and solmedrol 200 mg polaramine and dalfagan. The observed side effects were limited to drowsiness. No high fever or hypertension was noted. For the next 9 days he will be on LO-CD2a every day
, Received 10 mg of treatment. The next day after the end of such treatment, the biopsy showed no rejection.

Patients were well tolerated in the course of LO-CD2a, had no fever or hypertension at any dose and did not show any signs of clinical side effects. Routine hematology and clinical chemistry laboratory tests (including LFTs) obtained during the course of treatment indicate that lymphocyte counts have decreased from 290 / cubic mm to a minimum of 100 / cubic mm and that rejection has occurred. It also shows any changes attributable to the administration of the antibody, except for a decrease in creatinine levels associated with inflammation (from 2.7 mg / dl at the start of treatment with LO-CD2a to 1.10 at the end of the course). Did not.

FIG. 25 shows the serum creatinine level of this patient which decreased from 2.5 to about 1.0 on the days following treatment with LO-CD2a. The patient had lymphopenia before and during the treatment, and the leukocyte count did not show any dramatic change with the treatment (FIG. 26). In this patient, the serum level of LO-CD2a was 2.0 μg / m after each treatment.
2 and decreased from 1.0 to the lowest value of 0.25 μg / ml (FIG. 2).
7). After 8 months of treatment with LO-CD2a, the patient was healthy with normal renal function and no signs of recurrent rejection.

Patient 2-Biopsy # 1-Diagnosis: Indeterminate The biopsy contained about 20 glomeruli that were inconspicuous. There were diffuse mononuclear infiltrates and a small degree of interstitial edema. Only a small degree of tubular penetration was found, but no vascular trauma. These findings are inadequate for the diagnosis of acute cellular rejection.
Rarely mononuclear cells were present in the small arteries, which was suspect but did not meet the criteria for diagnosing rejection.

Patient 2—Biopsy # 2, about 2 weeks after the first biopsy—No diagnostic abnormalities were noted.

This biopsy appeared similar to the previous biopsy and contained about 10 glomeruli. Invasion was very sparse and no vascular trauma was identified.

(Iii) Patient # 3 This patient was a 19-year-old man with von Willebrand disease who underwent renal allograft for treatment of end-stage renal failure due to chronic pyelonephritis. The graft was removed after 17 days due to acute vascular rejection in secondary hypertension after 6 days course of OKT3 failure.

Five and a half months later, he underwent a second renal allograft. After 10 days his creatinine level was 6 mg / dl. The next day the creatinine level was 7 mg / d
The biopsy showed 3 plus cell rejection and vascular rejection (proliferative endarteritis without necrosis or thrombosis). On the same day, he received LO-CD2a, 10 mg, solmedrol and polaramine and 40 mg of dafalgan. No side effects were observed. For the next 9 days he will be on a daily basis without any other drugs or side effects.
-Received 10 mg treatment of CD2a. Two days after completion of the treatment, his creatinine level was 1.75 mg / dl, and biopsy showed focal necrosis and one focal lesion of chronic rejection, but no signs of acute rejection.

No clinical side effects (BP or temperature changes) were observed. Routine hematology and clinical chemistry laboratory tests (including LFTs) show that creatinine levels are reduced (from 7.10 mg / dl at the start of treatment of LO-CD2a)
Except for 1.75 mg / dl at the end of the 0-day course, no changes were attributed to antibody administration. Lymphocyte count was 340 / cubic mm before treatment and 220 during treatment.
/ Cubic mm, and rose to 690 / cubic mm 9 days after discontinuation of treatment with LO-CD2a and to 1000 / cubic mm 23 days after the end of treatment.

The white blood cell count of this patient did not change significantly with the treatment (FIG. 28). Serum creatinine levels dropped dramatically with treatment (FIG. 28). Seven months after the initial treatment with LO-CD2a, the patient became stiff with normal renal function and there were no signs of recurrent rejection.

Patient 3-Biopsy # 1-Diagnosis: Dangerous cell rejection affecting small arteries and to a lesser extent gaps and glomeruli. Arched arteries are marked mononuclear to intima with destruction of elastic fibers. Showed infiltration. There were occasional sparse infiltration foci in the gaps that penetrated the tubules. The interstices showed diffuse light interstitial edema. There were about 7 glomeruli. These showed hypercytosis and endothelial swelling in mononuclear cells. Overall, this pattern was indicative of severe acute cell rejection.

Patient 3-Biopsy # 2, Approximately 2 Weeks After the First Biopsy-Diagnosis: Consistent With Treated Rejection A few arterioles are shown. The gap showed fine diffuse fibrosis and minimal mononuclear infiltrates. The tubules were locally atrophic but otherwise inconspicuous. There were no signs of active cell rejection.

After 28 months of follow-up treatment of the second and third patients who remained in compliance with the patient's immunosuppressive therapy, no subsequent appearance of rejection symptoms was reported.

(Iv) Patient # 4 Suffering from severe versus host graft disease (heavy skin, gastrointestinal, renal and central nervous system toxicity resistant to high dose prednisone administration) after allogeneic bone marrow transplantation 12 patients
Received 10 mg / day of LO-CD2a for a day. His condition improved. Kidney function returned to normal, diarrhea ceased, skin improved and confusion resolved. Four days later, the antibody was stopped, the symptoms recurred, and the patient died despite the start of the second course of antibody.

LO-CD2a thus reverses the ongoing immune response to external tissues (allogeneic and xenogeneic, because it inhibits xenogeneic MLRs as well as allogeneic MLRs). Could be used for. Antibodies will be given by intravenous infusion once or twice daily over 10 to 14 days. It could also be used prophylactically to inhibit T cell activation immediately following organ transplantation as part of the induction protocol.

In a second experiment, Phase I safety, pharmacokinetic and dosing clinical trials begin with the first biopsy renal allograft receptor to demonstrate the appearance of acute rejection symptoms Was done. The antibody preparation used in this test was L-cell produced in cell culture.
O-CD2a.

The compassionate use of the antibody showed no side effects and was suggested to be effective at reversing acute graft rejection at a dose of 10 mg / day for 10 days. Preliminary clinical studies with chimpanzees have shown similar in vivo effects with 0.1-10
This suggested that it was observed at a dose equal to the human dose over 0 mg / day. The purpose of this Phase I study is therefore to investigate reducing the dose level of LO-CD2a starting at 10 mg / day for 10 days (additional 5 days as an extension of the option) to give a measure of the minimum effective dose. Was that. Since no side effects were observed in the "scope" of the steroid in the compassionate use patients described above, it was decided to administer only minimal analgesics and antihistamines without steroid coverage. This was done to predictively characterize the side effects induced by LO-CD2a.

[0124] Eleven patients have been enrolled under this protocol to date. 4 patients 1
0 mg / day was treated, 5 patients were treated with 5 mg / day, and 2 patients were treated with 2.5 mg / day.

[0125] Among the 11 patients enrolled, 9 patients experienced a reversal or partial reversal of acute rejection confirmed by biopsy. Nine of them were LO-CD2a at 10 mg / day.
All 4 patients treated with 5 mg / day, 3 patients treated with
. There were two patients treated at 5 mg / day. One of the patients treated at 5 mg / day (see below, patient 8) voluntarily withdrew from the study after the first treatment and the second patient (patient 9) showed a poor response.

Table 1 summarizes the results obtained in renal rejection patients treated with LO-CD2a under this protocol.

[0127]

[Table 1]

Side effects were observed during treatment with LO-CD2a without steroid coverage including nausea, vomiting, fever, chills, and hypertension, possibly as a result of the transient release of cytokines. The majority (> 70%) of these events were observed during the administration of the first dose, with the events being of limited scope and duration. For example, no fever above 40 ° C. was observed and most of the events resolved within the time of onset. No hypertension or severe diarrhea was observed. There was no apparent relationship between the intensity and incidence of these events and antibody dose. Despite the apparent discomfort of these symptoms, there were no events requiring emergency resuscitation.

In a third experiment, ten patients with acute renal allograft rejection had L as follows.
Treated on a compassionate basis without steroid coverage in O-CD2a.

Two were treated with 10 mg / day, four with 5 mg / day, and four with 2.5 mg / day. All compassionate patients were treated at 10 mg / day and 5 mg / day,
All but one patient treated at 2.5 mg / day showed evidence of complete or partial resolution of rejection by biopsy and other clinical signs. The adverse event profile of these compassionate use patients was similar to that predominantly seen here with a limited duration and range of first dose symptoms.

In a fourth experiment, six patients with steroid resistant versus host graft disease and one liver transplant recipient received LO-CD2a on a caring basis. No side effects were reported for these patients. A short story about these patients follows.

All 6 GvHD (graft versus host disease) patients had LO-CD for 10 consecutive days.
A dose of 2a, 10 mg was treated daily and administered intravenously over 1 hour. Accompanying this was administration of cyclosporine and steroids. All but one showed improvement in GvHD symptoms. Resolution of GvHD symptoms began 3-6 days after the start of mAb therapy. Improved symptoms of hepatic GvHD under mAb therapy were observed in two patients. Recurrent signs of GvHD were seen in two of the three patients evaluated.

The seventh patient received the donor liver, but unfortunately suffered from sepsis as a result of the surgical complications associated with renal failure due to cyclosporin toxicity, and significantly reduced bone marrow function. Because of her disease state, surgeons can take patients at risk in their assigned second liver by using currently available immunosuppressants (OKT3, mofetil, FK506, cyclosporine and ATG) for induction Did not want. The patient underwent liver transplantation and received LO-CD2a treatment at 5 mg / day for 7 days as the first dose to be infused during the start of surgery prior to the use of forceps in the transplanted organ.
Subsequent doses were given lower doses of steroids. During the treatment period, the patient's renal function was significantly improved over conventional immunosuppressive regimens to be initiated. The patient did not show any signs of rejection 8 weeks after transplantation.

Although the present invention is directed to inhibiting graft rejection in a preferred embodiment, the scope of the present invention should not be limited thereto, and may be generally directed to inhibiting T cell activation for any and all of the purposes. It should be appreciated that it is useful.

Example 6 Construction and Expression of Chimeric Antibodies Cloning and Sequencing of LO-CD2a _VH and _VL Total RNA was obtained from cell line LO-CD2a (ATCC HB11423) according to Churgwin's method ( Biochemistry , 18, page 5294, 1979).
). The mRNA is then followed by the Oligotex-dT mRNA kit (
(Chatsworth, CA). About 200
-300 ng of mRNA was reverse transcribed using the RNA-PCR kit from Perkin-Elmer Cetus (Norfolk, CT). The reaction was performed at 42 ° C. for 1 hour. The oligonucleotide primers required for the amplification of the _VH and _VL genes were selected using the following references: 1) Sequence of immune-related proteins , Kabat et al., 5th edition, 1991, 2) Aulandy. , Et al., Bulletin of the National Academy of Sciences (USA) , 86: 3833-3837 (1989).

[0136]

(Equation 2)

The numbers refer to amino acid residues set forth in Kabat et al., 1991.

The polymerase chain reaction (PCR) was performed using the following conditions for Parkin-Elmer D
Performed on NA Thermal Cycler 480: 5 minutes at 94 ° C. 1 minute at 94 ° C, 6
Thirty cycles consisting of 2 minutes at 0 ° C. and 2 minutes at 72 ° C., followed by 5 minutes at 72 ° C. DNA fragments were gel purified from agarose 1% using the Kiex Gel Extraction Kit (Qiagen, Chatsworth, CA). Fragment is then Kanango and Pandey, bio-technics, Vol. 14: 912-913 been blunt in accordance with the method of the page (1993), Blue script KSII ^- (
(Stratagene, La Jolla, Calif.).
The multiple clones were sequenced by the dideoxy chain termination method using the Sequenase ^™ T7 polymerase kit (US Biochemical, Ohio, Cleveland).

Due to the potential error rate inherent in PCR, at least three different reactions were performed.
The sequence most commonly observed in the LO-CD2a _VL and _VH genes is shown in FIG.
4, 35, 36 and FIGS. 37, 38, wherein FIGS. 34, 35, 36 show the nucleotide and amino acid sequences of the LO-CD2a _VL chain including the native leader sequence. FIGS. 37 and 38 show the nucleotide and amino acid sequences of the LO-CD2a V _H chain including the native leader sequence.

As shown in FIGS. 34, 35, and 36, the leader sequence is amino acid residues -20 to -1. Framework 1 is amino acid residues 1 to 23. CDR1 is amino acid residues 24-39. Framework 2 consists of amino acid residues 40 to 5
4. CDR2 is amino acid residues 55-61. Framework 3 has 62 to 93 amino acid residues. CDR3 is amino acid residues 94-102. Framework 4 is amino acid residues 103-112.

As shown in FIGS. 37 and 38, the leader sequence has amino acid residues -19 to -1.
It is. Framework 1 is amino acid residues 1 to 30. CDR1 is amino acid residues 31-35. Framework 2 is amino acid residues 36 to 49. CDR2 is from amino acid residue 50 to 66. Framework 3 is from amino acid residues 67 to 98. CDR3 is amino acid residues 99-107. Framework 4 is from amino acid residues 108 to 118.

B. Insertion into Vectors for Transient Expression Two vectors were licensed from the Medical Research Council (MRC) in London for the expression of each of the chimeric light and heavy chains of LO-CD2a. 9.2 kilobase light chain vector (hcmv-v11ys-kr-neo)
Contains the genomic clone of the humanized V _L region of human kappa constant region and anti-lysozyme as a HindIII-BamHI fragment. 8.6 kilobase heavy chain vector (
hcmv-VhLys-gammal-neo) is Hind III-Bam
The HI fragment includes the genomic clone of the human γ1 constant region and the humanized V _H region of anti-lysozyme. These vectors are based on Maeda et al., Human Antibody Hybridoma , Volume 2:
It is described in more detail on pages 124-134 (1991).

Since DNA fragments containing the native signal peptide are not available, LO-CD
The 2a V region was cloned behind a signal already present in the MRC vector. The light chain V region with the signal fragment was constructed from two fragments each derived from a separate PCR reaction as follows: Reaction 1 : The DNA template was an MRC light chain vector. The amplified fragment contained part of framework (FR) 1 in addition to the signal peptide. The two peptides used were as follows:

[0144]

(Equation 3)

The antisense primer contained the LO-CD2a FR1 sequence not found in the anti-lysozyme MRC vector. The PCR reaction produced a 0.15 kilobase Hind III-Tth III fragment.

Reaction 2 : The DNA template was the LO-CD2a _VL clone in Bluescript. The amplified fragment is ligated to the end of FR4 (from the Tth III site).
-Contained CD2a FR1. The 3 'untranslated region found in the MRC light chain vector was added to the 3' end of LO-CD2a using an antisense oligonucleotide. The two oligonucleotides used were:

[0147]

(Equation 4)

This reaction produced a 0.35 kilobase TthIII-BamHI fragment. Both PCR products were gel purified using QIAX and restricted using the appropriate enzymes. HindIII-TthIII fragment plus TthIII-Bam
The HI fragment was then ligated in three directions with Bluescript Hind III and Ba.
Linked between the mHI sites. This construct containing the entire _VL region of LO-CD2a plus the MRC signal peptide was then sequenced.

The heavy chain LO-CD2a V region construct contains an MRC signal sequence at its 5 'end, and a long 3' untranslated region also derived from an MRC heavy chain vector.
The final construct was made from three different PCR reactions described below: Reaction 1 : The DNA template was an MRC heavy chain vector. Because the anti-lysozyme _VL and _VH genes use the same signal, the sense primer was the same as that used for the LO-CD2a _VL construct, ie, 5 ' _VL lyssig . The antisense primer was 3'V _H lyssig :

[0150]

(Equation 5)

This reaction contains 0.1 parts of FR1 of LO-CD2a in addition to the MRC signal.
A 6 kilobase Hind III-Pst I fragment was produced. Fragments were gel purified, restricted, and ligated to Hind III-Pst I cut Bluescript for sequencing.

Reaction 2 : The DNA template was the LO-CD2a _VH region of Bluescript. The reaction is a 0.3 kilobase Pst I-Sty I containing most of the _VH region.
Fragments were produced. Because of the internal Pst I site in FR3 of LO-CD2a, the Pst I-Syt I fragment had to be constructed from the following two PCR reactions.

[0153]

(Equation 6)

The template DNA shown above is LO-CD2a V _H in Bluescript, clone 82-8.

Reaction A : Produce a 0.2 kilobase fragment using the oligonucleotides referred to as oligos 1 and 2 as primers.

[0156]

(Equation 7)

Reaction B : A 0.1 kb fragment is generated using oligos 3 and 4 as primers.

[0158]

(Equation 8)

Oligos 2 and 3 above contain changes in the nucleotide sequence that eliminate the internal Pst I site without changing the amino acid sequence of LO-CD2a. Aliquots (2-5 μl) of the overlap products of reactions A and B above were combined and served as template for a third PCR reaction. Oligonucleotide primers for this reaction were numbered 1 and 4 in the diagram. The 0.3 kilobase product was gel purified by QIAX and restricted with Pst I and Sty I. The internal Pst I site was successfully mutated because the fragment remained intact.

Reaction 3 : The final _VH fragment was produced using the MRC heavy chain vector as a template.
This 0.23 kilobase Sty I-Bam HI fragment is FR of LO-CD2a.
4 and the entire 3 'untranslated region from the MRC vector. The primers used were:

[0161]

(Equation 9)

The resulting fragment was gel purified and restricted with Sty I and Bam HI. Ps
The tI-StyI and StyI-BamHI fragments were then ligated into a PstI-BamHI cut bluescript for sequencing.

[0163] All oligonucleotides were synthesized on an Applied Biosystems synthesizer. All sequence reactions are performed using the Sequenase ^™ T7 Polymerase Kit (Ohio,
Cleveland, US Biochemical). All PCR _S were performed using the following protocol: 95 ° C for 5 minutes. 1 at 94 ° C
35 cycles of 1 minute, 50 ° C for 1 minute, 72 ° C for 2 minutes, final extension at 72 ° C for 5 minutes.

The LO-CD2a _VL and _VH fragments containing the correct sequence were removed from Bluescript and Hind III and B of the MRC light and heavy chain vectors, respectively.
cloned between the am HI sites. For heavy chains, 5 'Hind III
-The Pst I fragment is first constructed in Bluescript (Pst I-Bam
HI). The entire Hind III-Bam HI fragment was then M
Cloned into RC vector.

C. N-terminal amino acid sequencing of _VH and _VL N-terminal amino acid sequence analysis was performed on samples of LO-CD2a heavy and light chains to confirm the sequence obtained using RNA-PCR, Cambridge, Mass. Performed by Harvard Microchemistry Laboratory, Inc. Samples were prepared as follows: 200 μg of LO-CD2a was 12% S in the presence of B-mercaptoethanol.
DS was applied throughout the polyacrylamide gel run. Following electrophoresis, proteins were transferred to PVDF membrane using a Western transfer instrument. The membrane was stained briefly with Ponceau S, decolorized in 1% acetic acid, and the light and heavy chain bands were dried under vacuum and sent for amino acid analysis and N-terminal sequencing.

The amino acid sequence of the first 20 residues of LO-CD2a V _H completely matched the clone sequence. However 2,3,7 residues FR1 is, that differ from those encoded by the cloned gene shown sequence of V _L is. All of these differences
Based on the best deduced sequences obtained from the literature cited above, there are PCR primers used for cloning purposes.

D. Correcting the DNA sequence confirmed and its correct this sequence of N-terminal amino acid sequence In order to clone the V _L and V _H both natural signal peptide of the LO-CD2a simultaneously, RACE-PCR (c DNA, the E nds terminal R apid rapid A Nplification amplification -PCR) was adopted.
MRNA from LO-CD2a cells was reverse transcribed and the resulting cDNA was tethered to the G tail at its 3 'end using terminal transferase in the presence of dGTP. The cDNA was then amplified using specific 3 'and 5' oligonucleotides complementary to the G tail. Appropriate restriction sites were added at the 5 'end of each oligonucleotide to simplify subcloning.

The oligonucleotides used for the preparation of the cDNA were as follows.

[0169]

(Equation 10)

The RACE-PCR oligonucleotides were as follows:

[0171]

[Equation 11]

The RACE-PCR reaction was performed using the following protocol: 5 minutes at 94 ° C.
40 cycles of 94 ° C. for 30 seconds, 50 ° C. for 30 seconds, and 72 ° C. for 50 seconds, followed by a 5 minute extension at 72 ° C.

[0173] LO-CD2a, the PCR product obtained in the V _L and V _H were gel extracted using Kiakkusu. The V _H fragment was restricted with Xho I and Stu I,
Ligated into I-SmaI cleavage bluescript. The _VL fragment was blunt-ended and ligated into SmaI cut Bluescript. Numerous clones were sequenced in both light and heavy chain V regions, confirming the signal sequence.

These will be important for expression because the signal sequences found in immunoglobulin genes generally have introns. Genomic clones containing the _VL and _VH leader sequences were similarly identified. Genomic DNA was prepared as follows: 4 × 10 ⁷ LO-CD2a cells were spun, washed with cold PBS, spun, and washed again with PBS. Cells were resuspended in 0.4 ml of digestion buffer (with fresh additional proteinase K). The mixture is shaken at 50 ° C for 12-
Incubated for 15 hours, extracted with an equal volume of phenol / chloroform / isoamyl alcohol and spun at 1700 × g. The aqueous phase was transferred to a clean tube and 1/2 volume of 7.5M ammonium acetate and 2 volumes of 95% ethanol were added. DNA was pelleted at 1700 xg spin for 2 minutes. The pellet was washed with 70% ethanol and air dried. The pellet was resuspended in 80 ml TE, pH 8.0.

[0175] Using genomic DNA obtained from a cell line LO-CD2a as a template, the genomic leader sequences of oligonucleotides both V _L and V _H below, the Sph I, V _H at same unique restriction site (V _L Designated to amplify the portion of the framework region that ends with Pst I).

[0176]

(Equation 12)

The PCR reaction was performed as follows: Genomic DNA 1 from LO-CD2a cells
00ng, 200 pmol each of oligo LVL and BKA (for _VL fragment) or 200 pmol each of LVH and PVHA (for _VH fragment), 1 mm, d
10 μl of NTPs, 10 μl of 10 × Pfu buffer. 1 μl (2.5 units) of Pfu DNA polymerase (Stratagene, La Jolla, Calif.), 10
Deionized water to 0 μl. Pfu was used because it was more accurate than Taq polymerase.

The reaction conditions were as follows: 94 ° C. for 5 minutes, 50 ° C. for 5 minutes. 94 ° C for 1 minute, 50
35 ° C. for 1 minute at 72 ° C., then 5 minutes at 72 ° C. PCR products are gel
Purified, restricted and ligated into Bluescript for sequencing. Correct array
Once clones containing these clones are identified, Bluescript containing these clones
The vectors are Hind III and Sph I (V_L ) Or Hind II
I and Pst I (V_H ) And the fragments were gel separated. 0.75 kg
The base Hind III-Sph I fragment was then used to restore the original LO-CD2a V _L Linked to Bluescript containing the construct, and then Hind III-Sp
The hI fragment was removed. The new construct is a native LO-CD2a signal plasmid.
And the corrected FR1 sequence (consistent with the N-terminal sequence). 0
. The 16 kilobase Hind III-Pst I fragment was derived from the original LO-CD2a
V_H Linked to Bluescript containing the construct, and then Hind III-P
The stI fragment was removed. New construct contains natural signal + intron
Was. Newly constructed V_L And V_H The fragments were then Hind III and B
am HI digested from Bluescript and expressed in COS cells
Were cloned into the MRC light and heavy chain vectors, respectively.

E. Transient expression in COS cells COS7 cells were obtained from the ATCC (accession number CRL-1651) and were grown in Dulbecco's Minimum Essential Medium (DMEM) with 10% fetal bovine serum (FBS). Optimal transfection was achieved at approximately 50% confluence of adherent cells. In preparing transfections, plasmid DNA was added to DMEM containing New Serum and DEAE-Dextran / chloroquine diphosphate. The COS cell medium is removed and the DNA
The mixture was added and the cells were incubated at 37 ° C. for 3 hours. The medium was then removed and 10% DMSO in PBS was added to the cells for 2 minutes and then removed. FB
DMEM with S10% was added to the cells. After overnight incubation, the medium was changed and the cells were incubated at 37 ° C for 2 days. Supernatants were collected for assay by ELISA for secretion of the chimeric antibody.

F. Detection of Secreted Chimeric Antibodies by Eliza Chimeric antibody secretion was confirmed by assaying supernatants from transfected COS cells with Eliza designed to detect the presence of human antibodies (or portions thereof). Goat anti-human IgG (H + L) was added to phosphate buffered saline (PBS) at 5 μg / ml.
And attached to the wells of an Eliza microtiter plate overnight at 4 ° C. Plates were washed three times using an ELISA plate washer.

The remaining free sites were 200 μl of PBS containing 1% fetal bovine serum albumin (PB
(S-BSA) was added for 1/2 hour at room temperature. Two-fold dilutions of the supernatant and a positive control reference (purified human IgG1K) in PBS-BSA were prepared. Medium alone and / or PBS-BSA alone constituted the negative control. Antibody dilutions and controls were added to the wells and incubated at room temperature for 1.5 hours. The plates were then washed three times in a plate washer in PBS containing 0.05% Tween 20. An appropriate dilution of goat anti-human IgG (gamma chain specific) -horseradish peroxidase (HRP) conjugated antibody or goat anti-human kappa light chain-HRP conjugated antibody was added to each well and incubated at room temperature for 1 hour. The plate is P as described above.
Washed with BS-Tween 20 and then grown substrate containing hydrogen peroxide (ABTS
) Was added. Bound antibodies were detected by reading the absorbance at a wavelength of 405 nm.

G. Binding specificity of secreted chimeric antibody Chimeric antibody binding specificity is CD2 expressing mutant Jurkat cell line JRT3
-Assessed by flow cytometry analysis of antibodies binding to T3-5. The binding profile of the chimeric antibody (human IgG1) was determined by the native rat antibody (IgG2b) and the isotype-matched control MAB _S (unrelated (non-CD-2) binding specificity).
It was compared to the binding profiles of human IgG1 and rat IgG2b).

Preparation of JRT3-T3-5 (Jurkat) Cell Line The Jurkat cell line was obtained from the ATCC (Accession No. TIB-153),
Fetal bovine serum (FBS) 10%, amino acid supplement (NCTC) 10%, and L
-Grown in D-MEM (complete medium) containing 6 mM glutamine. Cells are at 37 ° C
, Maintained at 10% carbon dioxide, at a ratio of 1: 4 (cell concentration at passage is about 3 × 10 ⁶
/ Ml) three times per week. Jurkat cells are harvested,
Centrifuged to remove spent medium and washed with DMEM. The cells were then resuspended in phosphate buffered saline (PBS) with 0.1% sodium azide (NaAz) and aliquoted for cell counting. Viable cell count was determined by trypan blue exclusion.

Indirect Staining of Jurkat Cells Cell surface staining was performed on 96-well U-bottom microtiter plates. Approximately 6 × 10 ⁵ cells in a volume of 90 μl were dispensed into each well of a microtiter plate. Dilutions of the antibodies to be tested were prepared in PBS with NaAz, 0.1% and distributed in appropriate wells in a volume of 10 μl. The cells are incubated with the antibodies for 15 minutes at room temperature, after which the cells are added to each well with NaAz, 0.1% PBS for 2 minutes at 1900 rpm (
The cells were washed three times by centrifugation with Sorvall RT6000D). Resuspension of the cells was achieved by gently tapping the plate. Suitable fluorescein isothiocyanate (F
ITC) 10u aliquot of conjugated secondary antibody (anti-human Ig or anti-rat Ig)
1 was added to the appropriate wells and incubated at room temperature in the dark for 15 minutes. Plates were washed three times with PBS with 0.1% NaAz as previously described. The stained cells were fixed by adding 200 μl of 0.5% paraformaldehyde in PBS and fixed at 4 ° C. (
(Up to 1 week).

Flow cytometric analysis of stained Jurkat cells Stained cells were transferred to 12x17 mm polystyrene tubes for data acquisition using Becton-Dickinson Facscan. Data acquisition and analysis
Performed using II software. Jurkat cells expressing CD2 are LO-CD
Incubated with 2a (rat IgG2b) MAB, chimeric version of LO-CD2a (human IgG1), and corresponding isotype-matched controls. Bound antibody was detected using the appropriate FITC-conjugated secondary antibody according to the protocol described previously. Analysis shows similar binding patterns of native rat LO-CD2a and chimeric human-rat LO-CD2a.

H. Stable expression in NSO cells To express chimeric antibodies in stable transfectants, a glutamine synthase gene amplification system was obtained from CellTech Limited (Berkshire, UK). This system is described by Bebington, et al., Biotechnology, 10, 169-17.
5 (1992). The expression vector used was pEE6
hCMV-B and pEE12. Such vectors are described in published PCT application numbers WO86 / 05807, WO87 / 04462, WO89 / 01036, and WO89 / 10404. Genomic clones of these genes were obtained from the MRC light and heavy chain vectors, respectively, because neither of these vectors contained Ckappa or Cgamma1. Both constant clones were sequenced to obtain a restriction map. Two constructs were made in pEE12. The first contained from the light chain (V + C) 5 'to the heavy chain (V + C). The second construct contained the heavy chain 5 'to the light chain.

The strategy with the light chain was as follows.

[0188] 1. pEE6hCMV-B and pEE12 are Xma I and Ec, respectively.
o Digested with RI.

[0189] 2. The 5 ', 1.93 kilobase portion of the chimeric light chain comprises Hind III and E
It was removed from the MRC vector using coRI. This fragment was used as a template for PCR mutagenesis as described below.

[0190]

(Equation 13)

[0191] Restriction sites are underlined.

PCR was performed to change the 5 'restriction site from Hind III to Xma I and add a Kozak consensus sequence to the 5' end of the construct. This is essential for effective translation (Kozak, M. Journal of Cell Biology , 108:
229, 1989). The 3 'PCR oligo is used to remove internal Bam HI and Eco RI sites that interfere with subsequent cloning steps. P
The final product of CR mutagenesis is a 0.85 kilobase Xma I-Msc I fragment. PCR _S was performed according to the instructions provided with the TA Cloning Kit (Invitrogen, San Diego, CA). The following conditions were used for PCR: 2 minutes at 94 ° C. Followed by 1 minute at 94 ° C., 2 minutes at 55 ° C., and 72 ° C.
With 30/2 cycles. This was followed by a 5 minute extension at 72 ° C. Ligation and transformation were performed according to kit instructions. A number of clones were sequenced by the dideoxy chain termination method as described previously. Correction clones were removed from the TA cloning vector by digestion with Xma I and Msc I. This fragment was gel purified using QIAX.

The 3.2.7 kilobase C kappa fragment was removed from the MRC vector by digestion with Msc I and Eco RI. This fragment was gel purified using QIAX.

Using two different three-way ligation reactions, the complete chimeric light chain, ie, 0.85
The kilobase Xma I / Msc I fragment + 2.7 kilobase Msc I / Eco RI fragment was ligated into both pEE6hCMV-B and pEE12, each of which was cut with Xma I / Eco RI.

The strategy with heavy chains was as follows.

[0196] 1. Both pEE6hCMV-B and pEE12 were transfected into E. coli strain DM1. Both vectors were digested with Eco RI and Bc1I (Bc1I
If the plasmid grows in methylase minus bacteria, it will only cut).

[0197] 2. The chimeric heavy chain was digested with Hind III and Eco RI to give MRC.
Removed from the vector. The resulting 2.7 kilobase fragment was gel purified using QIAX. This fragment was digested with Nhe I and Bgl II. This is 0
. 7 kb Hind III / Nhe I fragment and 2 kb Nhe
Produces an I / Bg1 II fragment. Both fragments were gel purified. The 0.7 kilobase fragment was then used as a template for PCR mutagenesis.

[0198]

[Equation 14]

Restriction sites are underlined.

This PCR was performed to change the 5 'restriction site from Hind III to Eco RI and to add a Kozak consensus sequence. The 3 'oligo is used to remove an internal Bam HI site that would interfere with subsequent cloning steps. PCRs, ligation reactions and transformations were performed as previously described.

Using two different three-way ligation reactions, the complete chimeric heavy chain, the 0.7 kb Eco RI / Nhe I fragment + the 2.0 kb Nhe I / Bg1 II fragment, was pEE6hCMV-B and pEE12. Both were ligated, each cut with Eco RI / Bc1 I.

(Bc1 I and Bg1 II are compatible restriction sites).

The final construct on pEE12 containing both the chimeric light and heavy chains was made as follows.

From the light chain 5 'to the heavy chain : pEE6hCMV-B carrying the chimeric heavy chain is Bg1 I
Digested with I / Bam HI. 5.1 including heavy chain plus hCMV promoter
The kilobase fragment was gel purified and ligated into the BamHI site of pEE12 containing the chimeric light chain. Correct orientation was verified by digestion with Sal I / Bam HI. The presence of the 0.28 kilobase fragment indicates the correct orientation.

From heavy chain 5 'to light chain : pEE6hCMV-B carrying the chimeric light chain was
Digested with I / Bam HI. 5.9 with light chain plus hCMV promoter
The kilobase fragment was gel purified and ligated into the BamHI site of pEE12 containing the chimeric heavy chain. The orientation was verified by digestion with Sal I / Bam HI as described previously.

NS / O cells (Galfer et al., Methodology in Enzymology, 73 (B), 3-46
ECACC deposited with the European Animal Cell Culture Collection (Page 1981)
Catalog number 85110503) was transfected by electroporation. Transfected cells were selected by growth on glutamine-free medium. Antibody production and binding activity against CD2-expressing Jurkat cells was confirmed as described below.

A functional analysis of the human-rat chimeric antibody shows that its functionality is similar to that of the rat LO-CD2a antibody. Both antibodies inhibit the primary mixed leukocyte reaction (MLR) when up to 120 ng / ml of the antibody is added to the culture. In addition, addition of the chimeric antibody to the primary MLR induces a poorly responsive state in the responding population to attack the original or third-party alloantigen. Hyporesponsiveness is specific for allogeneic antigens, as an attack on mitogen or tetanus toxoid induces a proliferative response.

[0208] Example 7 of humanized antibody constructs and expression A. Construction of humanized light chain HUM5400 (EMBL access X55) sharing homology with LO-CD2a
A framework region from the human V kappa gene, designated as 400), was chosen for humanization of the light chain V region. The following is a comparison between LO-CD2a and the HUM5400 framework:

[0209]

(Equation 15)

Rat LO-CD2a light chain variable region sequence, homologous human variable region, HUM54
A comparison of 00 and humanized LO-CD2a is shown in FIG. The complete amino acid sequence is given for the LO-CD2a variable region and the residues are numbered according to the rat sequence. Residues identical to those of the rat at the corresponding position in the HUM5400 sequence with the humanized sequence are indicated by horizontal dashes, while non-identical residues are given in letter code. The humanized LO-CD2a light chain variable region is designated by the HUM5400 framework region, rat LO-CD2a CDR's (underlined), and seven rat LO-CD2a framework residues (* designated above the rat sequence). These were chosen because such residues are relevant for maintaining the binding specificity of LO-CD2a.

As shown in FIG. 39, framework 1 is from amino acid residues 1 to 23. CDR1 is amino acid residues 24-39. Framework 2 is amino acid residues 40-54. CDR2 is amino acid residues 55-61. Framework 3 is amino acid residues 62 to 93. CDR3 is at amino acid residue 94
To 102. Framework 4 is amino acid residues 103-112.
The leader sequence is amino acid residues -20 to -1 (FIGS. 40, 41, 42). Rat amino acid residues retained in the framework region are amino acid residues 9 and 12 of framework 1, amino acid residues of framework 2 are 41, 42, 50 and 51, and amino acid residues of framework 3 are 82 It is.

The humanized light chain was constructed to include the CDRs of LO-CD2a and the variable region framework of HUM5400, but only seven unusual residues (*)
Retained from the D2a framework. The 5 'region was obtained from the chimeric light chain construct. This 0.43 kilobase Hind III / Hph I fragment contains the native signal plus intron and sequences encoding the first three amino acid residues of framework 1 that are identical in the rat and human frameworks. The remainder (i.e., the 3 'end) of the construct (0.37 kilobases), which contains nucleotides encoding all but the first three amino acids of the variable region, contains seven nucleotides over a size of 63-81 nucleotides. From overlapping oligonucleotides to P
Synthesized by CR. These longer oligonucleotides have a shorter 5 '
And served as a template for 3 'external PCR oligonucleotides (21-26 nucleotides in length). In all cases, 5 pmol of template was used for each external PCR
Used with 100 pmol of oligonucleotide. All PCRs were performed with Pfu polymerase to achieve greater fitness. The procedure was as follows: 95 ° C. for 5 minutes. 25 cycles of 2 minutes at 94 ° C, 2 minutes at 55 ° C and 2 minutes at 72 ° C. This was followed by an additional extension of 5 minutes at 72 ° C. All syntheses were achieved in four steps. In the first step, the first long oligonucleotide was added to a 0.43 kilobase Hind III-Hph H fragment. The next three sets of overlapping oligonucleotides were then
Added continuously using R. After the synthesis of the entire 0.8 kilobase construct was completed, it was gel purified using QIAX, Hind III and Bam
Restricted with HI, gel purified again with QIAX, and further with Hind III / Ba
mHI cleaved Bluescript KS II. A number of clones were sequenced until the correct version was obtained. The clone is then Hind III
And was removed from Bluescript by digestion with Bam HI. The resulting fragment was gel purified on QIAX and cut with Hind III / Bam HI
It was ligated to the RC light chain vector. The nucleotide and amino acid sequences of the humanized LO-CD2a light chain V region are shown in FIGS.

A description of the overlapping oligonucleotides used in the synthesis of the humanized light chain and their uses follows.

[0214]

(Equation 16)

[0215] Oligonucleotides 1, 3, 5, and 7 are reverse complement sequences.

Oligonucleotides 2, 4, and 6 are sense strand sequences.

Oligonucleotide # 1 overlaps the 0.43 kilobase Hind III / Hph I fragment derived from the chimeric light chain construct. This oligonucleotide was added to the 0.43 kilobase fragment by PCR using the following PCR oligos:

[0218]

[Equation 17]

In a similar manner, oligonucleotides # 2 and # 3 were threaded together by PCR using the following PCR oligos:

[0220]

(Equation 18)

After PCR, both products were gel purified using QIAX, followed by a third round of PC
Joined together using R. The third round of PCR required the following oligos:

[0222]

[Equation 19]

The resulting fragments were gel purified by QIAX. Oligonucleotides # 4 and # 5 were then ligated together by PCR using the following oligos:

[0224]

(Equation 20)

The fragments were gel purified and added to the previous construct using PCR oligos:

[0226]

(Equation 21)

The final fragment was constructed using oligonucleotides # 6 and # 7 and a PCR oligo:

[0228]

(Equation 22)

After gel purification, this fragment was added to the rest of the humanized light chain construct by PCR using oligos.

[0230]

(Equation 23)

B. Construction of humanized heavy chain Human antibody clone Amu5-3 (Genbank accession number U00562)
Framework regions were used for the generation of humanized heavy chains. Below is LO-C
Comparison with the framework of D2a and that of Amu5-3:

[0232]

(Equation 24)

FIG. 43 shows the rat LO-CD2a heavy chain variable region sequence, homologous human variable region Am.
1 shows u5-3 and humanized LO-CD2a (humanized Vh). The complete amino acid sequence is given in LO-CD2a and the residues are numbered according to the rat sequence. Residues identical to those of the rat at the corresponding position in the humanized sequence and the Amu5-3 sequence are indicated by horizontal dashes, while non-identical residues are given in letter code. Humanized LO-CD2a Vh is in the Amu5-3 framework region, rat L
Consists of O-CD2a CDRs, and seven rat LO-CD2a framework residues (designated by * on rat residues), which are LO-CD2
a was selected because it is relevant to maintain the binding specificity. The vertical line at the CDR3 of the rat and humanized sequences represents the space required to align the three sequences because Amu5-3 had a longer CDR3 than the rat and humanized regions. is there.

As shown in FIG. 43, framework 1 is from amino acid residues 1 to 30. CDR1 is amino acid residues 31-35. Framework 2 is amino acid residues 36 to 49. CDR2 is from amino acid residue 50 to 66. Framework 3 is from amino acid residues 67 to 98. CDR3 is amino acid residues 99-107. Framework 4 is from amino acid residues 108 to 118. The leader sequence is amino acid residues -19 to -1 (FIGS. 44, 45, 46).

The rat LO-CD2a amino acid residue retained in the framework region is amino acid residue 47 of framework 2 and amino acid residue 6 of framework 3
7, 70, 72, 76, 85 and 87.

A single humanized heavy chain construct was created. This construct contains the LO-CD2a CDRs and the Amu5-3 framework, with the exception of the seven residues (*) retained from LO-CD2a. This construct was produced in the same manner as that of the human light chain. In this case, there were 12 overlapping template oligonucleotides spanning the size of 69-88 nucleotides. Twelve external PCR oligonucleotides ranged in size from 21-26. Synthesis was accomplished in six steps with the addition of a pair of overlapping template oligonucleotides at each step. The final construct (0.7 kb) was gel purified using QIAX as previously described, digested with Hind III and Bam HI, gel purified again, and bluescripted for sequencing as previously described. Was linked to. Once the correct clone was identified, it was removed from Bluescript by restriction with Hind III / Bam HI and then ligated into an MRC heavy chain vector that had been digested with the same enzymes. The nucleotide and amino acid sequences of the humanized LO-CD2a heavy chain V region are shown in FIGS.

A description of the overlapping oligonucleotides used in the synthesis of the humanized heavy chains and their uses follows.

[0238]

(Equation 25)

[0239] Overlapping oligonucleotides # 1 and # 2 were joined together by PCR using the following PCR oligos:

[0240]

(Equation 26)

The resulting fragment was gel purified.

The overlapping oligonucleotides # 3 and # 4 were joined together by PCR using the following PCR oligos:

[0243]

[Equation 27]

The fragments were gel purified.

Oligonucleotides # 5 and # 6 were joined together by PCR using the following PCR oligos:

[0246]

[Equation 28]

After gel purification, this fragment was ligated with oligonucleotide # 3 by PCR with the following oligos:
And joined to the fragment produced in # 4:

[0248]

(Equation 29)

The resulting fragment was gel purified.

Oligonucleotides # 7 and # 8 were joined together by PCR using the following PCR oligos:

[0251]

[Equation 30]

The resulting fragment was gel purified and added to the construct made by joining oligonucleotides # 3 to # 6. This was achieved using a PCR oligo.

[0253]

(Equation 31)

The product fragment generated by joining oligonucleotides # 3 to # 8 was gel purified. The 5 'end of the construct (oligonucleotides # 1 + # 2) was then added using PCR oligos.

[0255]

(Equation 32)

This fragment was gel purified and the next fragment (3 ′) was added using oligonucleotides # 9 and # 10.

Oligonucleotides # 9 and # 10 were joined by PCR using the following PCR oligos:

[0258]

[Equation 33]

After gel purification, the 3 'fragment was added to the rest of the construct using PCR oligos.

[0260]

(Equation 34)

The resulting fragments were gel purified and added to the rest of the construct.

[0262] Oligonucleotides # 11 and # 12 were synthesized using the following PCR oligos.
Joined by:

[0263]

(Equation 35)

After gel purification, this final 3 'fragment was added to the rest of the construct using oligo (5') PCR4H (sense) and (3 ') PCR6J' (antisense). The resulting 0.7 kilobase final construct was gel generated, sequenced, and cloned into an MRC heavy chain vector as previously indicated.

Transient expression of COS cells and detection of secreted antibodies was performed as described previously for the chimeric antibodies. The humanized antibody was purified by affinity chromatography (protein A). Binding studies on Jurkat cells were performed using LO-CD2
A similar binding pattern is shown between the humanized and rat forms of a (FIG. 47).
. Jurkat cell lines expressing CD2 include rat LO-CD2a, humanized LO
-CD2a (LO-CD2aHu) or rat IgG2b or human Ig
Stained with G control. Antibody concentration was 0-4 μg / ml. LO-CD
The rat and humanized forms of 2a bind with a similar binding pattern to Jurkat cells, whereas the rat and human isotype control antibodies do not. Bound antibodies were detected with isotype-specific FITC-conjugated antisera. The results, shown in FIG. 47, are expressed as the percentage of cells positively stained with the antibody over the concentration range described above. Functional studies have shown that humanized antibodies can further inhibit the primary MLR. Cultures in which the nanogram amount of LO-CD2a, LO-CD2aHu, rat IgG2b control, or human IgG1 control contains an equal number of human peripheral blood mononuclear cells (PBMC) from designated responders and stimulators (irradiated cells) Added to the well. Control wells did not contain any antibody. Cultures were incubated for 5 days, then pulsed overnight tritiated thymidine ⁽³ HT). Proliferation is detected by the ³ HT uptake. As given in Table 2 below, the results are expressed as the average CPM (radioactivity counts per minute) recorded on a beta plate reader.

[0266]

[Table 2]

[0267] Humanized antibodies also show that these T cells are not allogeneic and humanized LO-CD, but rather when the isotype control (with unrelated specificity) is replaced with a human antibody.
In the presence of 2a to induce a low response state for attacking alloantigens in T cells (as measured by ³ HT uptake in average CPM) when cultured in primary MLR (FIG. 48) . The functional properties of the humanized antibody are similar to those of rat LO-CD2a.

Example 8 LO-CD2a elicits Alloantigen-Specific Low Responsiveness The ability of T cells to recognize alloantigen in the secondary MLR following addition of LO-CD2a only to primary cultures Was tested. Primary MLR cultures containing responding individual cells and irradiated stimulator cells in the presence of LO-CD2a, control antibody (LO-DNP11, Mass., Charlestown, Biotransplant, Inc.) or without antibody for 7 days It was kept warm. Cells were then washed and rested with medium only for an additional 3 days. After a rest period, the cultured cells were re-challenged with the original stimulator cells or cells obtained from a different donor ("third party" cells).

Representative experimental results from these studies are illustrated in FIG. In Panel A,
A primary response was observed when responding individual cells were cultured at 200 ng / ml in the presence of the control antibody, but no response was observed when LO-CD2a was cultured in the presence of 200 ng / ml, This was consistent with previously reported data. The kinetics of the response was measured by harvesting the cultures on days 3, 5, and 7. Data was provided as the average of triplicate well runs at each data point where cpm from individual wells was within 10% of the average.

As depicted in FIG. 49b, responding individual cells from cultures treated with either LO-CD2a or control antibodies had a 1: 1 ratio without any antibody present in the secondary MLR. Alleles were re-attacked by allogeneic antigens carrying stimulator cells. The kinetics of the response was measured by harvesting the cultures on days 3, 5, and 7. Responses of cells cultured in the primary MLR with either LO-CD2a or control antibodies are included as controls. Data is presented as the average of triplicate well runs at each data point when cpm from individual wells was within 10% of the average. Data is from the same experiment depicted in FIG. 49a.

As shown in FIG. 50C, cells stimulated in the primary MLR with a specific allogeneic antigen in the presence of LO-CD2a or a control antibody will then cause the presence of any antibody present in the secondary culture. Without attack with allogeneic antigens carrying cells from third party donors in a 1: 1 ratio. The kinetics of the response was measured by harvesting the cultures on days 3, 5, and 7. Cell responses to autologous stimulator cells cultured in the primary MLR with control antibody or LO-CD2a are included as controls. Data is presented as the average of triplicate well runs at each data point where cpm from individual wells was within 10% of the average. Data is from the same experiment depicted in FIGS. 49a and b.

Cells cultured in primary culture in the presence of control antibodies are responsive to re-challenge by the same allogeneic stimulator cells in primary culture (panel b) and to stimulation by third party cells. Again, it was responsive (panel c). In contrast, cells cultured in the presence of LO-CD2a between primary MLRs were either re-challenged with primary allogeneic stimulator cells (panel B) or stimulated by third party cells (panel c). Again, it was not responsive. Cells in culture containing LO-CD2a were viable and responsive to stimuli other than alloantigens. For example, stimulation with either PHA or soluble OKT3 caused an equivalent proliferative response in cells cultured with LO-CD2a control antibody, or fresh autologous PBMC (data not shown). Flow cytometric analysis of these cells after resting failed to detect LO-CD2 at the cell surface (data not shown). Thus, these data are
This shows that exposure to O-CD2a and allogeneic antigens can induce allogeneic antigen hyporesponsiveness, ie, a state of resistance.

Cells obtained from culture after alloantigen stimulation to address the apparent low alloresponsive alloantigen specificity induced by LO-CD2a during alloantigen stimulation Was attacked to respond to a soluble protein antigen, a tetanus toxoid. A soluble antigen response requires the presence of viable antigen presenting cells (APCs) lost in allogeneic antigen-stimulated culture after 7 days. Therefore, in these studies,
A fresh source of APC was provided to cultured cells by adding irradiated autologous PBMC to a secondary assay culture. As depicted in FIG. 38A, responding individual cells from cultures treated with either LO-CD2a or control antibodies were stimulated at a 1: 1 ratio without the presence of any antibodies present in the secondary MLR. Re-attacked with allogeneic antigens carrying cells. The kinetics of the response was measured in harvest cultures at 3, 5, and 7 days. LO
-The response of the cells to autologous stimulator cells cultured in the primary MLR with either CD2a or a control antibody is also included as a control. Data are presented as the average of triplicate well runs at each data point where cpm from individual wells was within 10% of the average.

As depicted in FIG. 51b, responding individual cells from cultures treated with either LO-CD2a or control antibodies had 7.5 μg / ml without any antibodies present in the subculture. Reattacked with tetanus toxoid. The kinetics of the response was measured by harvest culture at 3, 5, and 7 days. The response of cells cultured in primary culture without antibody and autologous stimulator cells served as a control for response to tetanus toxoid.

The results shown in FIGS. 51a and b show that cells cultured with LO-CD2a plus allogeneic antigen in the primary MLR were poorly responsive to allogeneic antigen in the secondary MLR. This indicates that the cells were responsive to tetanus toxoid when presented by fresh APC.

Example 9 To address the role of the Fc portion of LO-CD2a, studies were performed comparing the functional effects of the F (ab ') ₂ fragment with whole antibodies.

As shown in FIG. 52, PBMCs were irradiated with Epstein-Barr virus (EB
V) Responding individual cells to stimulator cells at a 2: 1 ratio with the transformed B cell line
The cells were cultured for 6 days with increasing doses of CD2a or its F (ab ') ₂ fragment. This graph depicts the effect of the F (ab ') ₂ fragment on the response of four different donors, and each time point is the average of triplicate well motion at each data time point.
The results are reported as a percentage of the control response. To make the graph clear, all LO-
Data from the addition of the CD2a antibody is not shown. All LO-CD2a
At a dose of 30 ng / ml, the donor response was: 1-18% of donor, 2-6.6% of donor, 3-3.7% of donor, and 4-12% of donor relative to control. there were. Cell responses without antibodies present from the tested individuals averaged 56,690 cp.
m from 404,843 cpm and cpm from individual wells were within 10% of the mean.

F (ab ′)_Two To further evaluate the potential inhibitory potential of the fragment, F (ab ')_Two fragment
Titration of unfractionated PBMC stimulated with soluble OKT3 (APC-dependent response)
Was added to As shown in FIG. 53, PBMC was soluble OKT3 (100 ng).
/ Ml) for 3 days. LO-CD2a intact antibody or F (ab ') _Two Either fragment was added in increasing doses on day 0. The result is an isotype
Table: Percentage of proliferative response of cells to OKT3 in the presence of control monoclonal antibody
Will be revealed. At each data point, the cpm from the individual well was within 10% of the average.
This is the average of the triple wells. Mean for stimulated PBMC without antibody
cpm was 60,117.

The results in FIG. 53 show that proliferation of T cells against OKT3 was not inhibited when the F (ab ′) ₂ fragment was used.

Example 10 APC Requires Inhibition of LO-CD2a in In Vitro Cultures To address the issue of the role of APC in inhibiting LO-CD2a in PBM,
The T cell population of C was CD14, CD56, CD19 and HL by immunomagnetic sorting.
A-DR positive cells were lost. Analysis of the purified cells by flow cytometry technique showed that the APC population was lost by over 95% (data not shown).
. The proliferation of these purified T cells against soluble OKT3 (APC-dependent response) was reduced by this loss to less than 1% of the proliferation of unfractionated PBMC (data not shown). Purified T cells or unfractionated PBMC were stimulated by plate-bound OKT3 (APC-independent method of T cell stimulation) in the presence or absence of LO-CD2a. The ability of LO-CD2a to inhibit T cell activation was abolished when APC was removed (FIG. 54). As shown in FIG. 54, PBMC or PBMC from which APC was lost by the immunomagnetic technique were plated on a 96-well plate coated with OKT3 (10 μg / ml) and cultured for 3 days. LO-CD2a was added at the beginning of the culture. The results are expressed as the percentage of the proliferative response of the cells to OKT3 in the presence of the isotype-matched control monoclonal antibody. The plotted data represents three experiments, where each data point is cp from an individual well.
m is the average of triplicate wells within an average of 10%.

Example 11 Construction and Analysis of Second Generation Humanized LO-CD2a (MEDI-507) Construction of Vectors First generation additional mutagenesis of humanized LO-CD2a (Example 7B) heavy chain This was done to improve the binding properties of the antibody. Four additional rat framework residues were retained in addition to the seven retained in the molecule described in Example 7. These four
Amino acid changes were performed using three PCR mutagenesis experiments (PCRs
One led to two amino acids being changed).

All PCRs were performed using Pfu DNA polymerase (La Jolla, CA).
(Strada Gene) and the following program: 95 ° C. for 5 minutes.
5 minutes at 72 ° C. 25 cycles of 1 minute at 94 ° C, 1 minute at 55 ° C, and 1 minute at 72 ° C. Then extended at 72 ° C. for 5 minutes. The template DNA is the first generation humanized LO-CD2aV _H
It was a 701 base pair (confirmed) HindIII / BamHI DNA fragment isolated from the region. The oligonucleotides used are listed in Table 3 below.

[0283]

[Table 3]

Reaction # 1 generates an 185 base pair DNA fragment containing oligonucleotides 877 and 1007, which replaces two additional rat LO-CD2a antibody residues with amino acids (AA
) 12 and 13 (Gln12 and Arg13) were retained in the framework FR1.
(Residues are numbered according to Kabat et al., Sequence of Immunology-Related Proteins, Fourth Edition, 1987, Ministry of Health and Human Services, Public Health Administration, National Institutes of Health).

Reaction # 2 generated a 515 base pair DNA fragment containing oligonucleotides 1006 and 875, which also retained two rat residues at AAsGln12 and Arg13.

Reaction # 3 generated a 290 base pair DNA fragment containing oligonucleotides 877 and 1012, which retained the rat LO-CD2a residue at position 48.

Reaction # 4 generated a fragment containing oligonucleotides 1011 and 875 and retaining rat residues at the same position 48.

All reaction products were purified using QIAX (Chatsworth, CA, Qiagen). The reaction product then underwent additional PCR to generate a 701 base pair _VH fragment as follows. Reaction # 5 contained the reaction products of reactions # 1 and # 2, and oligonucleotides 877 and 875. Reaction # 6 is reaction # 3
And # 4 reaction products, and oligonucleotides 877 and 875.

To obtain the mutation at position 28, (isoleucine) PCR was set up using template from reaction # 5 (ie containing Gln12, Arg13). Reaction # 7
Generated oligonucleotide fragments 877 and 1020, a 230 base pair DNA fragment, which retained the rat residue at position 28. Reaction # 8 is oligonucleotide 1
470 base pair DN, including 008 and 875, also retaining the rat residue at position 28
A fragment was generated. Following purification of the reaction product using QIAX (Qiagen, Chatsworth, CA), the reaction product was then subjected to additional PCR to generate a 701 base pair _VH fragment, followed by P. Bluescript / Sma (Stratagene, La Jolla, CA).

Following sequence analysis, the two fragments were gel separated using Qiax (Qiagen, Chatsrose, CA): (1) a 270 base pair HindIII / containing three of the four new amino acid changes. EcoNI (Hu Vh mut
1 + 2A), and (2) a 430 base pair EcoNI / containing the fourth amino acid change.
BamHI (Hu Vh mut 3D). The fragment was then ligated into the MRC heavy chain vector described in Example 6B, which contained the human heavy chain constant region between the HindIII and BamHI sites to produce the final clone, clone # 257-20. . The light chain expression begda is a first generation expression vector (202-2)
It was the same as that used in.

FIG. 57 shows the amino acid sequence of the rat antibody used in the modeling process and the second generation humanized LO-CD2a VH fragment (MEDI-507) aligned in parallel with the human heavy chain variable sequence.

Transient Expression in COS Cells Using MRC Vectors
Licensed from the National Council (MRC, London, UK) (Figure 57)
. The 9.2 kilobase light chain vector (hcmv-vlys-kr-neo) is H
Humanizing the human kappa constant region of anti-lysozyme as an indIII-BamHI fragment
V_L Includes genomic clones of the region. 8.6 kilobase heavy chain vector (hcmv-
VhLys-gammal-neo) is a HindIII-BamHI fragment.
Human γ1 constant region of anti-lysozyme and humanized V_H Including genomic clones of the region (Maeda
, Et al., 1991,Human antibody hybridoma2, Vol. 124-134).
Anti-lysozyme V_L And V_H The region was removed by restriction digestion and humanized V_L And V _H The construct was inserted in its place. The two plasmids are then transferred to COS-7 cells
Co-transfected.

COS-7 cells were obtained from the ATCC (accession number CRL-1651) and
Individuals were grown in FBS / DME medium to 40-60% confluence in 1700 cm ² corrugated plastic roller bottles (Corning, New York). All media was removed and replaced at 37 ° C. with transfection media [Nuceram 10% (Massachusetts, Waltham, Collaborative Research, Inc.), DEAE-dextran chloroquine diphosphate, humanized L
DM with O-CD2a heavy chain DNA, 0.25 mg, light chain DNA, 0.25 mg
EM]. Roller bottle is returned for warmer and roller bottle equipment for 123 hours,
Thereafter, it was removed from all media and replaced with 10% DMSO in PBS at 37 ° C.
After 2-5 minutes of contact, the DMSO solution was removed and replaced with FBS / DMEM.
The roller bottles were returned to the incubator and roller bottle facility at 37 ° C. per day, after which the media was removed and discarded (protein accumulation is now considered too low for harvest), and then replaced with fresh, identical media. Was replaced. After incubating at 37 ° C. for a few days, half of the media in each roller bottle was removed and replaced with fresh identical media. This process was repeated until about three harvests were obtained (after that point the yield had decreased). Harvested supernatants were assayed for the presence of humanized IgG in ELISA.

The resulting antibodies were purified from transfected COS cells by affinity chromatography (Protein A). An experimental system with three in vitro activities of this humanized antibody: binding to Jurkat cells, inhibition of the primary mixed lymphocyte reaction (MLR), and low response of the secondary MLR to alloantigens (allogeneic antigens) Sex induction, using rat L
Compared to the activity of O-CD2a. The functionality of the humanized antibody produced by transient expression was similar to that of rat LO-CD2a, and was later confirmed by stable antibody expression in NSO cells.

Stable Expression in NSO Cells Using Celltech Vectors The biological production of recombinant mammalian gene products for clinical use requires higher eukaryotes to achieve proper processing of proteins and safe biological activity. Generally requires the use of cells. Immortalized mycobacterial cells are currently the system of choice for this purpose due to their ability to grow at high density in inexpensive media, high productivity, and the suitability of post-transplantation processing. Two desirable systems are: 1) Glutamine synthetase (GS) in mouse myeloma NSO cells, selection (Bebington et al., 1992,
Bio / Technology, Vol. 10, 169-175 pages), and 2) Chinese hamster ovary (CHO-DHFR minus) dihydrofolate reductase (dhfr) amplification in cells (Kaufman and Sharp, 1982, Molecular Biology journal, 159, pp. 601-621).

The glutamine synthetase system in NSO cells is a second generation humanized LO-CD
2a (MEDI-507) was utilized for stable expression and resulting production. N
SO cells are glutamine auxotrophs essential for the addition of additional glutamine for survival or for the introduction of an exogenous GS gene (Bevington, et al., 1992). The expression system takes advantage of this property to select cells that carry the GS and immunoglobulin genes in a glutamine-deficient medium (Coquette et al., 1990, Biotechnology (New York), 8, 662-667). One way in which cells can increase the level of glutamine synthetase is to increase the gene copy number of DNA amplification using concomitant amplification of copy number of adjacent genes. Potential for an increased gene dose is an increase in the production of the gene product, in this case both glutamine synthetase and immunoglobulin.

[0297] The Celltech expression vectors used (Figure 44) were pEE6CMV-B and pEE.
12 (Published PCT application numbers WO86 / 05807, WO87 / 0446).
2, WO 89/01036, and WO 89/10404).
Since none of these vectors contained Ckappa or Cgamma1, genomic clones for these genes were obtained from MRC light and heavy chain vectors, respectively. Both constant clones were sequenced to obtain restriction rounds.

Due to the difference between the MRC vector and the CellTech vector (the MRC vector for the light and heavy chains is distinct, the heavy chain is separated by the hCMV promoter region behind the light chain in the CellTech vector) The following changes were required to the light and heavy chain constructs.

The following changes were made to clone the transiently expressed humanized BTI-322 light chain construct into a CellTech vector. The light chain clone, clone # 202-2, is H
After digestion with indIII / EcoHI, the 1.93 kilobase fragment was gel purified. This fragment served as a template for PCR mutagenesis to change the 5 'restriction site to XmaI and to add a translation initiation consensus sequence (Kozak, 1998,
Point mutations define sequences adjacent to the AUG start codon that regulate translation by eukaryotic ribosomes. Cells 44, 283-292). This PCR reaction was also used to remove internal BamHI and EcoRI sites to allow for a later cloning step. The resulting 0.85 kilobase PCR product is sequenced,
It was then digested with XmaI / MscI. Clone # 202-2 is then Msc
Digested with I / EcoRI and the 2.7 kilobase fragment was gel separated. 0.85
A three-way continuation of the kilobase fragment and the 2.7 kilobase fragment was performed by inserting them into pEE12 between the XmaI and EcoRI sites of the polylinker. The resulting clone was named clone # 237-4.

Insertion of the heavy chain construct into the Celltech vector required the following changes: Clone # 257-20 was digested with HindIII / EcoRI and the entire 2.7 kilobase heavy chain was gel separated. This fragment was then further 0.7 kilobase HindII
The I / NheI fragment was digested with NheI / BglII to produce a 2.0 kilobase Nhel / BglII fragment (the BglII site was 3% of the original clone).
Very close to the 'end, i.e. 20 base pairs away from the EcoRI site). 0.
The 7 kilobase HindIII / NheI fragment then added a 5 'restriction site to HindI.
To change from II to EcoRI and to add a translation initiation consensus sequence,
It was used as a template for CR mutagenesis (Kozak, 1986, cell , 44
Vol., Pages 283-292). The 3 'end was also modified by removing the internal BamHI site. The resulting fragments were sequenced. When the correct clone is found, 3
Orientation was performed as follows: the 0.7 kb EcoRI / NheI fragment and the 2.0 kb NheI / Bg1II fragment were ligated into pEE6hCMV-B between the EcoRI and Bc1I sites of the polylinker (Bc1I). And Bg1II
Is a matching site). After successful bacterial transformation was confirmed by PCR, one of these clones, designated clone # 6, was used to construct the final expression vector.

To do this, a 6 kilobase BglII / BamHI fragment derived from clone # 6 containing the humanized heavy chain plus its hCMV promoter was cloned into pEE12 /
It was inserted into the BamHI site of the humanized light chain vector 237-4. After transformation into E. coli, to confirm that the clone contains both the light and heavy chains in the correct orientation, 5
PCR was performed using template DNA obtained from a bacterial colony (colony PCR) boiled for minutes. The two oligonucleotide PCR primers are 3
It was designed to amplify a region extending from the 5 'end to the 5' end of the heavy chain. clone#
12 was confirmed to be correct. To further confirm that this clone contained the light and heavy chain inserts in the proper orientation, plasmid DNA was isolated from this clone and the DNA was digested with BamHI / SaII. The release of the small 0.25 kilobase fragment from clone # 12 shows that the heavy chain and hCMV promoter-containing 6
The kilobase BglII / BamHI fragment indicated that it was in the correct 5 'to 3' orientation (if the orientation was reversed, the BamHI / Sa1I digestion had produced a 6.3 kilobase fragment). It would have been.

Stable Expression in NSO Cell Lines Clone # 12 DNA linearized with Sa1I was used for the transformation of NSO cells according to CellTech's Glutamine Synthetase Gene Amplification System-Operating Procedure Manual (second edition: bone marrow) Revised version 5, 6 for use in tumor cells
-14, November 5, 1992). Cells were electroporated with 3 μF, 1500 volts, 2 pulses (40 μg transfected per 10 ⁷ NSO cells). After electroporation, cells were plated on 4-6 plates of 16,000, 4,000, and 1000 cells per well at 50 μL / well of FBS / DME 10%. The next day 150 of glutamine-free IMDM with 10% dialyzed FBS
μL / well was added to the plate. When it was confirmed that the colony was growing on the 96-well plate in the glutamine-free medium, the colony was labeled (MEDI
(Indicating that they have taken up the gene for glutamine synthetase linked to the gene for the -507 antibody gene). A total of 204 colonies were screened with ELISA and selected with the highest titer since transfection was performed with the DNA described previously. Approximately one third of the colonies were selected for tension in 48 well plates, incubated for 5 days, and then tensioned in T-25 flasks. When significant cell growth occurs in any particular T-25 flask (and more usually within 3-6 days), the flask is further filled with 5 × 10 ⁴ viable cells in each of a few T-25 flasks. / M
l. One of these flasks may be 2-3
Used to determine antibody production by allowing them to grow for a week. Other flasks measure cell growth rates and use different cell lines until you can determine which cell lines are worth further testing and cloning based on their suitability for antibody production, cell growth, and serum-free medium growth Was used to grow. At this point, for approximately 5 weeks, the cells were used for growth so that the cell line simply passed from flask to flask, without necessarily performing cell counting prior to each passage. At the end of this period, subcloning was started on some of the most promising cell lines, as determined by the criteria described previously. MEDI-507hu 2.2.404 was from a 16K cell / well slab plate. Subcloning was performed at 0.5 and 5 cells / well on each of two plates with the clone MEDI-507hu2.
2.204. Performed on X and allowed to incubate for approximately 3 weeks.

A subclone labeled clone # 204.11 was generated on a 5 cell / well plate (originally calculating the starting viable cell density). This specific plate is 96
Grew in 33 of the wells. Twenty-nine clones (from all plates) were labeled and screened with ELISA. Five of the 29 clones were selected for 48-well plates and then tensioned into subsequent 5-7 day T-25 flasks for incubation. A total of 5 clones were incubated for an additional 5-7 days, then tensioned in T-75 flasks. After 5-7 days, cells from these flasks are used to prepare frozen vials for each clone, and the rest of the cells are used to test antibody production, growth, and adaptability to serum-free media. Was done. After the screening process was completed, the frozen glass bottles of the best clones of the best cell line were thawed, and several of each vial were expanded for settlement and further evaluated in culture.

Comparison of in vitro functionality and biological properties of humanized MEDI-507 and rat LO-CD2a Relative binding affinity, inhibition of mixed lymphocyte reaction (MLR), low response to alloantigen in secondary MLR Ability to induce sex and C in Hu-CD2 transgenic mice
Humanized MEDI-50 against that of rat LO-CD2a due to D2 cell loss
Studies comparing the functional and biological activities of 7 were conducted. The results show that the two antibodies behave similarly.

Relative Binding to Jurkat Cells The Jurkat cell-based binding assay was performed using human MEDI-507 (lot, PA1
12895) was tested for competitive binding of rat LO-CD2a (lot SO82
/ 83) for comparison. 300 μL of Jurkat cells at a cell density of 1.3 × 10 ⁶ cells / ml were treated with ¹²⁵ I-labeled rat LO-CD2a at 4 ° C.
Was incubated with 20 μL of a 4 mM solution of E. coli and 80 μL of cold (unlabeled) rat LO-CD2a or humanized MEDI-507. The effective concentration of labeled rat LO-CD2a was 0.2 nM. Unlabeled rat LO-CD2a and humanized ME
DI-507 varied from 2 nM to 0 nM. Tubes containing an effective concentration of chilled LO-CD2a at 20 nM served as a control for non-specific binding.

After incubation at 4 ° C. for 30 minutes, 240 μL of the reaction mixture was placed in a thin polypropylene tube on top of 50 μL of the silicone oil mixture (2: 1 (v / v) of AR-200: Thomas Silicone Oil). Placed in layers, 14,000 × g in a microcentrifuge,
Centrifuged for 5 minutes. Separation of the labeled antibody from the bound antibody was achieved by allowing the cells to precipitate through an oil plug. The oil layer was cut and counted. 100 μL of the reaction mixture before separation was counted directly to determine the total input cpm. Analysis of the assay shows a similar binding pattern between rat LO-CD2a and humanized MEDI-507.

Inhibition of MLR Rat LO-CD2a inhibits mixed lymphocyte reaction (lot SO82 / 83)
And humanized MEDI-507 from two different study lots (lots QAS and ASQ
) Ability was tested. Irradiated human stimulator PBMC was 10-60 ng / mL for rat LO-CD2a, humanized MEDI-507, and appropriate isotype control
Were incubated with human responders PBMC from different donors in the presence of a range of concentrations. Incubation for 5 days at 37 ° C., the culture was pulsed with ³ H-thymidine and thymidine incorporation was determined. The results (FIG. 46) show that both rat LO-CD2a and humanized MEDI-507 similarly inhibit MLR in a dose-dependent manner.

Induction of Hyporesponsiveness An important biological property of rat LO-CD2a is its ability to induce in vitro alloantigen-specific hyporesponsiveness. The ability of humanized MEDI-507 (Lot QAS and ASQ) to induce low responsiveness was compared to that of rat LO-CD2a (Lot SO82 / 83).

The primary MLRs were humanized MEDI-507, rat LO-CD2a, or appropriate isotype control, and 10 ng / ml, 60 ng / ml in antibody-free control cultures.
It was performed in the presence of ml and 200 ng / ml. After incubation for 7-8 days, cells were separated from the culture, placed in Ficoll, washed, resuspended in culture medium and allowed to "pause" at 37 ° C for 3-4 days. Cells from these cultures were used as responsive solid cells in secondary MLR cultures containing irradiated stimulator cells from the original donor. Cultures were then incubated for an additional 48 hours, pulsed with ³ H-thymidine, and thymidine incorporation was determined. Analysis of the results (FIG. 61) was performed using humanized MEDI-507.
Also induce a state of low responsiveness to alloantigen challenge.

CD2 cell loss The effects of humanized MEDI-507 and rat LO-CD2a on human CD2 receptors transplanted into mice (Hu-CD2 transgenic mice) were compared. 25
Hu-CD2 transgenic mice were assigned to one of five groups and
Received EDI-507 (0.12 mg / kg or 0.006 mg / kg), rat LO-CD2a (0.12 mg / kg or 0.06 mg / kg), or buffer control (5 mice / group) . Mice were bled on day 0 (prior to receiving the dose), on days 1, 3, 7, 14, 21, 28, 36 and 44, and the blood was analyzed for CD2 lymphocytes.

[0311] Whole blood was collected in tubes containing heparin. Four 20 μL aliquots were dispensed into 96-well round bottom tissue culture plates, stained with the appropriate antibodies, and treated with FACS rice to lyse red blood cells and fix stained cells. After washing to remove unbound antibody and cell debris, the remaining lymphocyte population was analyzed by flow cytometry.

FIG. 62 shows that low doses of humanized MEDI-507 and rat LO-CD2a (0.0
06 mg / kg, which in these mice resulted in a 0.15 μg dose) showed no detectable effect on the total number of CD2 cells. These mice are 1
It was not examined after the 4th day. High doses of humanized MEDI-507 and rat LO-CD2a (0.12 mg / kg, in these mice it would be a 3.0 μg dose) had a significant loss effect on the total number of CD2 cells. Recruitment of CD2-bearing cells in transgenic mice treated with high doses of humanized and rat antibodies could also be compared.

(Industrial Applicability) As described above, the present invention provides a human or non-human primate that inhibits the activation and proliferation of T cells and thereby inhibits transplant rejection. Also disclosed are methods of producing conjugated antibodies and, in combination, inhibiting the immune response and inhibiting graft rejection.

[0314] Numerous modifications and variations of the present invention are possible in light of the above teachings, and thus the invention may be practiced other than as specifically described within the scope of the claims set forth above.

[Brief description of the drawings]

FIG. 1: Two-color staining of peripheral blood mononuclear cells (PBMC) with biotinylated LO-CD2a and Leu-5bPE. The following parameters were followed for this staining.

FIG. 2. Human PBMCs are stained with LO-CD2a-FITC and then a
A) T11-PE conjugated to phycoerythrin (PE) (Coulter antibody against CD2), or b) Leu-5B conjugated to phycoerythrin (PE)
-Stained with PE (Becton-Dickinson antibody to CD2). In each case, there was no staining by the secondary antibody altered by pretreatment with LO-CD2a.

FIG. 3. Human PBMCs are stained with LO-CD2a-FITC followed by a
A) T11-PE conjugated to phycoerythrin (PE) (Coulter antibody against CD2), or b) Leu-5B conjugated to phycoerythrin (PE)
-Stained with PE (Becton-Dickinson antibody to CD2). In each case, there was no staining by the secondary antibody altered by pretreatment with LO-CD2a.

FIG. 4. Effect of LO-CD2a on membrane markers. 2 × 10 ⁶ cells / m
The PBMC at 1 were cultured in the absence (solid line) or presence (dashed line) of LO-CD2a (200 ng / ml). At the times indicated in the figures, cells were harvested and processed for cytofluorometer analysis. a) and b); PBMC is anti-CD
3 (Leu-4a-FITC), anti-CD4 (T4-RD) mAbs, anti-class I
Labeled with I antigen (LO-DRa-FITC) or anti-CD8 (T8-RD) monoclonal antibody (mAbs). Negative controls for commercial mouse mAbs were aliquots of the same cells stained with FITC or rhodamine-labeled mouse IgGs. The negative control for rat mAbs was standard rat serum followed by cells incubated with FITC-labeled mouse anti-rat mAb (MARK-FITC). Results are shown as percentage of positive cells.

FIG. 5. Effect of LO-CD2a on membrane markers and human blood lymphocyte cultures with and without the addition of LO-CD2a. Lymphocyte culture at 1 × 10 ⁶ cells / ml was performed using a) anti-CD2 (Leu-5b-FITC), anti-CD4 (T ₄
-RD1) mAb or anti-CD8 (T8-RD) mAb at indicated time points. The negative control for the commercial mouse mAb was an aliquot of the same cells stained with FITC or rhodamine-labeled mouse IgGs. Rat mA
The negative control for bs was standard rat serum followed by cells incubated with FITC-labeled mouse anti-rat mAb (MARK-FITC). Results are shown as percentage of positive cells.

FIG. 6. Effect of LO-CD2a and Leu-5b on CD2 expression.
The human PBMC is a) LO-CD2a (200 ng / ml) or b) Leu
-5b (dialyzed against PBS, diluted 1: 2) and incubated at the indicated times, a) CD2 (Leu-5b-FITC and T11-RD1) expression, and LO-CD2a ( Mark-3-FITC) or b) C
Stained for expression of D2 (LO-CD2a-FITC, T11-RD1) and Leu-5b goat anti-mouse (GAM-FITC).

FIG. 7: Effect of LO-CD2a on MLR. a) Added at time 0
Mixed lymphocyte cultures incubated for 6 days in the presence of increasing concentrations of LO-CD2a
Of MLR in Cultures were harvested at 6 days. b) LO-C added at time 0
Inhibition of MLR in mixed lymphocyte cultures incubated at different concentrations of D2a. Culture
Nutrition was collected at 24 hour intervals. c) LO-CD2a (200 ng / ml)
Mixed lymphocyte culture in the presence (solid line) or in the presence (dashed line)^ThreeH-thymidine
(HT) Uptake (cpm). d) LO added at different times after the start of heat retention
-Inhibition of MLR by CD2a (200 ng / ml). Cultures were harvested on day 6
. All cultures were performed in triplicate at a final volume of 200 μg / well (1 × of each donor / ml).
10⁶ Cells) were made. 8 hours before harvesting the culture^ThreeH-thymidine was added.
The result of c) is cpm x 10 incorporated per well sampled at the indicated time points ^Three As shown. The results of a), b) and d) show that the control culture (LO-CD2a
Percentage of MLR in triplicate cultures (mean ± standard deviation) compared to
Shown.

FIG. 8: Effect of LO-CD2a on blasts during MLC. Peripheral blood mononuclear cells were cultured in mixed lymphocyte cultures with or without the addition of 200 ng / ml of LO-CD2a. At the indicated time points, cells were removed and CD2 (Leu5b
-FITC) and analyzed by flow cytometry after staining with an antibody to (FITC). The blasts were gated by forward and side scatter, and the expression of the indicator marker was quantified on the blasts.

FIG. 9: Effect of LO-CD2a on resting cells during MLC. Human lymphocytes were cultured with or without LO-CD2a (200 ng / ml). At the indicated time points, cells were removed and CD3 (Leu4a-FITC),
CD4 (T4-RD1), CD8 (T8-RD1) or CD25 (LO-T
ACT-1-FITC). Resting lymphocytes were confirmed by differential gating on size and granularity, and results are shown as a percentage of total resting lymphocyte staining with the indicated antibodies (FIG. 9). Leu-5 in culture with and without LO-CD2a
The percentage of resting cells positively stained by b is shown in FIG.

FIG. 10. Effect of LO-CD2a on resting cells during MLC. Human lymphocytes were cultured with or without LO-CD2a (200 ng / ml). At the indicated time points, cells were removed and CD3 (Leu4a-FITC)
, CD4 (T4-RD1), CD8 (T8-RD1) or CD25 (LO-
(TACT-1-FITC). Resting lymphocytes were confirmed by differential gating on size and granularity, and results are shown as a percentage of total resting lymphocyte staining with the indicated antibodies (FIG. 9). Leu- in culture with and without LO-CD2a
The percentage of resting cells positively stained by 5b is shown in FIG.

FIG. 11: Effect of LO-CD2a on mitogen-stimulated lymphocytes. PBM
C is OKT3 (100 ng / ml), Con-A (10 μg / ml) and PH
The cells were cultured for 96 hours in the absence or presence of A (1 μg / ml). In parallel cultures, LO-CD2a (200 ng / ml) was added one hour after the mitogen (grey bar) or one hour before the mitogen (white bar). Lattice pattern bars are LO-C
Represents culture performed in the presence of D2a alone. Cultures (in triplicate) were pulse labeled with ³ H thymidine during the last 8 hours of incubation.

FIG. 12: Effect of LO-CD2a on mitogen-driven activation of PMBC. PMBC _S from two donors was OKT3 (100 ng / ml), CON-A
(10 μg / ml) and PHA (1 μg / ml) for 96 hours. In the parallel culture, LO-CD2a (200 ng / ml) was 0 days after the start of the culture (
0 hours), 1 day (24 hours), or 2 days (48 hours). The graph shows the percentage of inhibition of mitogen-induced proliferation by LO-CD2a in each donor.

FIG. 13. Effector cells and target cells in the presence of LO-CD2a ( ⁵¹ Inhibition of NK activity by incubation of CR-labeled K562 cells). Three types of LO-CD2a
Concentrations of 5 μg / ml, 1 μg / ml and 0.5 μg / ml were tested. NK activity
Effector cells, which were peripheral peripheral lymphocytes, had a solubility percentage of labeled target cells.
Indicated by Two standard test materials have three E / T ratios: 200/1, 100 /
Tested at 1,50 / 1.

FIG. 14. Effector cells and target cells in the presence of LO-CD2a ( ⁵¹ Inhibition of NK activity by incubation of CR-labeled K562 cells). Three types of LO-CD2a
Concentrations of 5 μg / ml, 1 μg / ml and 0.5 μg / ml were tested. NK activity
Effector cells, which were peripheral peripheral lymphocytes, had a solubility percentage of labeled target cells.
Indicated by Two standard test materials have three E / T ratios: 200/1, 100 /
Tested at 1,50 / 1.

FIG. 15: LO-CD2a at 20 mg / day for 10 days (day 0-9).
Total lymphocytes per μl of peripheral blood of Cynomolgus monkey received on day.

FIG. 16. LO-CD2 at 20 mg / day for 10 days (day 0-9).
The PBMC of cynomolgus monkeys receiving CD2a (CD2 (Leu-5b), CD4 (L
eu3a), CD8 (Leu2a), natural killer cells (CD8 and CD
11b), and monoclonal antibodies to B cells (anti-IgM) were stained on the indicated days and analyzed by flow cytometry. The results are expressed as the ratio of the total number of stained cells per microliter of blood.

FIG. 17. LO-CD2 at 20 mg / day for 10 days (day 0-9).
NK activity of cynomolgus monkeys receiving a. NK activity was assayed at 11 and 22 days and was expressed as% solubility at 25/1, 50/1 and 100/1 E / T.

FIG. 18. Serum concentration of LO-CD2a in cynomolgus monkeys receiving antibody at 20 mg / day for 10 days (day 0-9). Monoclonal antibodies were measured by ELISA as described herein and expressed in μg / ml.

FIG. 19: LO-CD2a at 20 mg / day for 10 days (day 0-9).
Transition of IgG antibodies against LO-CD2a in cynomolgus monkeys that received sun.
Antibodies to the monoclonal antibodies are measured by serial dilutions of sera drawn on the designated days by the sandwich ELISA described herein and are indicated by optical density (absorbance) at 492 nm.

FIG. 20: Effect of LO-CD2a on baboon lymphocytes. a) Blood was obtained from baboons on indicated days, and cells were detected using anti-CD2 antibodies T11-RD1 and Leu-5b-FITC, LO to detect bound LO-CD2a.
-Stained with mouse anti-rat kappa 1b antibody conjugated to CD2a and MARK3-FITC, FITC. b) Serum samples taken on indicated days were assessed for LO-CD2a levels by ELISA.

FIG. 21: Effect of LO-CD2a on baboon lymphocytes. MARK3-FITC and MAR where blood is collected on indicated days and cells are detected with PE-conjugated streptavidin to detect bound LO-CD2a
K2b-8-biotin (mouse monoclonal anti-rat I conjugated to biotin
gG2b antibody). a) No pretreatment of cells. b) Incubation with LO-CD2a, 2.5 μg / ml prior to staining to detect any sites not occupied by circulating antibodies.

FIG. 22: Effect of LO-CD2a on baboon lymphocytes. Blood samples were taken on indicated days and T4-RD1 (CD4), T8-RD
1 (CD8) or MARK3-FITC (bound to LO-CD2a).

FIG. 23. Leukocytes, lymphocytes and creatinine of patient # 1 treated with ATG then LO-CD2a for allograft rejection.

FIG. 24. Serum levels of LO-CD2a during and after treatment in patient # 1.

FIG. 25. Creatinine in patient # 2 before, during and after treatment with LO-CD2a.

FIG. 26. Leukocyte and lymphocyte counts in patient # 2 before, during and after treatment with LO-CD2a.

FIG. 27. Serum levels of LO-CD2a in patient # 2 removed immediately before and 2.5 hours after each injection.

FIG. 28. Patient # 3 receiving LO-CD2a for renal allograft rejection
White blood cell count, lymphocyte count and serum creatinine level in rats.

FIG. 29: LO-CD2a and (2) Leu5b, (3) Leu4 (C
D3), (4) Leu3a (CD4), (5) Leu2b (CD8) and (6)
) Two-color staining of Leu11 (anti-CD16) NK cells with markers. LO-CD2a
Binding was detected with goat anti-rat IG-FITC. The upper set of two-color histograms (1-6) shows two-color staining. The lower set (7-12) shows a single staining with each antibody.

FIG. 30 shows LO-CD2a and (2) Leu5b, (3) Leu4 (C
D3), (4) Leu3a (CD4), (5) Leu2b (CD8) and (6)
) Two-color staining of Leu11 (anti-CD16) NK cells with markers. LO-CD2a
Binding was detected with goat anti-rat IG-FITC. The upper set of two-color histograms (1-6) shows two-color staining. The lower set (7-12) shows a single staining with each antibody.

FIG. 31. Rat isotype control for LO-CD2a (Pharmingen, purified rat IgG2b, kappa) or LO-CD2a and CD4 (c
, D), CD8 (e, f), CD16 (g, h), CD19 (i, j) and C
Two-color staining of human PBL with phycoerythrin conjugated antibody to D2 (k, l).
LO-CD2a and isotype controls were FITC-conjugated affinity purified F (
ab ') 2 detected with anti-rat immunoglobulin (Southern Biotechnology). All antibodies against the CD antigen were phycoerythrin conjugated antibodies obtained from Becton Dickinson [CD4 (Leu3a), CD8 (Leu2a), CD
16 (Leu-11b), CD19 (Leu12) and CD2 (LeuSb)
]. In each case, the staining with the isotype control is shown in the first histogram and also in the LO-
CD2a is shown by the second histogram. Histogram a shows the pattern for the isotype control and b shows the pattern for LO-CD2a.

FIG. 32. Rat isotype control for LO-CD2a (Pharmingen, purified rat IgG2b, kappa) or LOCD2a and CD4 (c,
d), CD8 (e, f), CD16 (g, h), CD19 (i, j) and CD
Two-color staining of human PBL with phycoerythrin conjugated antibody to 2 (k, l). L
O-CD2a and the isotype control were FITC-conjugated affinity purified F (a
b ') detected with 2 anti-rat immunoglobulins (Southern Biotechnology). All antibodies against the CD antigen were phycoerythrin conjugated antibodies obtained from Becton Dickinson [CD4 (Leu3a), CD8 (Leu2a), CD1
6 (Leu-11b), CD19 (Leu12) and CD2 (LeuSb)]
. In each case the staining with the isotype control is shown in the first histogram and also in the LO-C
D2a is shown by the second histogram. Histogram a shows the pattern for the isotype control and b shows the pattern for LO-CD2a.

FIG. 33. Cytofluoroscopy analysis of staining of COS cells transfected with wild-type CD2. The left panel shows CO transfected with a control vector without CD2.
Histogram of staining of S cells, set of panels on the right show histogram of staining of COS cells transiently transfected with vector containing complete CD2 molecule. In each set, the top histograms show murine W632 (an antibody to class I known to be expressed by COS cells), and 76-2-11 (murine W
(Isotype control for 632). The middle panel is Leu5b (
Anti-CD2 from Becton Dickinson and 76-2-11, staining with isotype-matched control for Leu5b staining, lower panel showing LO-CD2a
And staining with rat isotype-matched controls for LO-CD2a.

FIG. 34. Nucleotide and amino acid sequence of chimeric LO-CD2a _VL chain.

FIG. 35. Nucleotide and amino acid sequence of chimeric LO-CD2a _VL chain.

FIG. 36. Nucleotide and amino acid sequence of chimeric LO-CD2a _VL chain.

FIG. 37. Nucleotide and amino acid sequence of chimeric LO-CD2a V _H chain.

FIG. 38. Nucleotide and amino acid sequence of chimeric LO-CD2a V _H chain.

FIG. 39. Rat LO-CD2a, human HUM5400 and humanized LO
-The amino acid sequence of the light chain variable region of CD2a.

FIG. 40. Nucleotide and amino acid sequence of humanized LO-CD2a variable region.

FIG. 41. Nucleotide and amino acid sequence of a humanized LO-CD2a variable region.

FIG. 42. Nucleotide and amino acid sequence of humanized LO-CD2a variable region.

FIG. 43. Rat LO-CD2a, human Amu5-3, and humanized LO
-The amino acid sequence of the heavy chain variable region of CD2a.

FIG. 44. Nucleotide and amino acid sequence of the heavy chain variable region of humanized LO-CD2a.

FIG. 45. Nucleotide and amino acid sequence of the heavy chain variable region of humanized LO-CD2a.

FIG. 46. Nucleotide and amino acid sequence of the heavy chain variable region of humanized LO-CD2a.

FIG. 47. Binding of rat LO-CD2a, humanized LO-CD2a to Jurkat cells.

[Figure 48] rat LO-CD2a, humanized LO-CD2a, and the induction of in vitro hyporesponsiveness by control rat and human immunoglobulins.

FIG. 49. Inhibition of primary MLR by LO-CD2a and response of T cells cultured on LO-CD2a in the primary MLR to antigen in the secondary MLR or to a third party stimulator in the MLR.

FIG. 50. Inhibition of primary MLR by LO-CD2a and response of T cells cultured on LO-CD2a in the primary MLR to antigen in the secondary MLR or to a third party stimulator in the MLR.

FIG. 51. Secondary ML of T cells cultured with LO-CD2a in the primary MLR.
Response to antigen in R or tetanus toxoid in subculture.

FIG. 52: Effect of F (ab ′) ₂ fragment of LO-CD2a on MLR.

FIG. 53. Comparison of the inhibition of PBMC proliferation by soluble OKT3 between intact LO-CD2a antibody and F (ab ′) ₂ fragment of LO-CD2a.

FIG. 54. Plate binding OKT of APC _S for inhibition of LO-CD2a.
Effect on proliferation induced by 3.

FIG. 55. Amino acid sequences of the heavy chain variable region of rat LO-CD2a, humanized antibody MEDI-507, and human Amu5-3.

FIG. 56. Amino acid sequences of the heavy chain variable region of rat LO-CD2a, humanized antibody MEDI-507, and human Amu5-3.

FIG. 57. hcmv-V11ys-kr-neo and hcmv-VhL
Plasmid map of ys-gammal-neo.

FIG. 58. Plasmid map of pEE6hCMV-B and pEE12.

FIG. 59. Comparison of rat LO-CD2a with humanized MEDI-507 antibody in Jurkat cells based on binding assay.

FIG. 60. LO-CD2a, humanized M for inhibiting mixed lymphocyte reaction
Comparison of potency of EDI-507 antibody and human IgG. The MLR response of one responder / stimulator pair is shown as an example of the relative inhibition of the response by rat LO-CD2a or MEDI-507. Rat IgG2b or human IgG are isotype controls.

FIG. 61. Rat LO-CD2a and humanized MEDI- in low response assay
Comparison of 507 antibodies. In summary, a total of 7 days of MLR was found in rat LO-CO2a,
Performed in 6-well plates in the presence of MEDI-507 or control rat or human IgG. Cultures were ficolled to remove dead cells and 3-4
Placed on day medium. Cells were then evaluated for their ability to proliferate against secondary challenge with the original atypical stimulus. The percentage of growth control associated with the re-challenged culture of the primary MLR that did not receive the exogenous antibody is shown. This "medium" treated culture measures 100% response.

FIG. 62. Rat LO-CD2a and humanized M versus total number of CD2 cells.
Comparison of the effects of the EDI-507 antibody.

──────────────────────────────────────────────────続 き Continuation of front page (81) Designated country EP (AT, BE, CH, DE, DK, ES, FI, FR, GB, GR, IE, IT, LU, MC, NL, PT, SE), OA (BF, BJ, CF, CG, CI, CM, GA, GN, ML, MR, NE, SN, TD, TG), AP (GH, KE, LS, MW, SD, SZ, UG, ZW) , EA (AM, AZ, BY, KG, KZ, MD, RU, TJ, TM), AL, AM, AT, AU, AZ, BB, BG, BR, BY, CA, CH, CN, CZ, DE, DK, EE, ES, FI, GB, GE, HU, IL, IS, JP, KE, KG, KP, KR, KZ, LK, LR, LS, LT, LU, LV, MD , MG, MK, MN, MW, MX, NO, NZ, PL, PT, RO, RU, SD, SE, SG, SI, SK, TJ, TM, TR, TT, UA, UG, US, UZ, VN (72) Inventor Ratanne, Dominique Belgium, Brussels Be-1150, Bois Postal 17, Abounu General de Longville, 16 (72) Inventor Caplan, Ruth United States, 01876 Massachusetts, Chucksbury, Danvegan Road 10 (72) Inventor Keaver-Emmons, Thomas United States, 19073 Ben Sylvania, Newton Square, Saw Mill Road 3231 (72) Inventor Postema, Christina, E. United States, 02129 Massachusetts, Charlestown, # 2 Russell 48 ( 72) Inventor White-Sharf, Mary, E. United States, 01890 Chuse' Tsu, Winchester, Johnson loaded 19 F-term (reference) 4B024 AA01 AA20 BA44 CA07 GA11 HA01 4C085 AA13 BB11 CC02 CC32 DD62 DD63 EE01 4H045 AA10 AA11 BA10 CA40 DA76 EA20 FA72 FA74

Claims (8)

[Claims] 1. A humanized antibody consisting of CDRs from LO-CD2a, wherein the humanized antibody comprises a rat LO as shown in FIGS. 55 and 56 in the framework of the heavy chain variable region of the humanized antibody. -Amino acid 47 of the CD2a heavy chain variable region,
A humanized antibody comprising 67, 70, 72, 76, 85 and 87 and one, two, three or four of amino acids 12, 13, 28 and 48. 2. The humanized antibody of claim 1, wherein the humanized antibody is a rat LO-CD2a heavy chain as shown in FIGS. 55 and 56 in the framework of the heavy chain variable region of the humanized antibody. Amino acids 12, 13, 28, 47, 4 of chain variable region
A humanized antibody comprising 8, 67, 70, 72, 76, 85 and 87. 3. The humanized antibody of claim 1, wherein said humanized antibody is a rat LO-CD2a light chain variable as shown in FIG. 39 in the framework of the light chain variable region of the humanized antibody. One or more amino acids 9, 12, 41 of the region
, 42, 50, 51 and 82. 4. The humanized antibody of claim 3, wherein said humanized antibody is a rat LO-CD2a light chain variable as shown in FIG. 39 in the framework of the light chain variable region of the humanized antibody. A humanized antibody comprising the amino acids 9, 12, 41, 42, 50, 51 and 82 of the region. 5. The humanized antibody according to claim 1, wherein the variable region of the heavy chain of the humanized antibody has the amino acid sequence of FIGS. 55 and 56. 6. The humanized antibody according to claim 3, wherein the light chain variable region of the humanized antibody has the amino acid sequence of FIG. 39. 7. A method of inhibiting an immune response in a patient, the method comprising treating the patient with the antibody of claim 1. 8. A method of inhibiting rejection of a graft in a patient, the method comprising treating the patient with the antibody of claim 1.