CA2072356A1 - Diagnosis and treatment of diseases - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE

Methods for diagnosing the onset or monitoring of the course of a disease comprising obtaining a T-cell nucleic acid or polypeptide test sample from an animal and detecting in such test sample a first target nucleic acid or polypeptide sequence correlated with the disease. The first target nucleic acid or sequence corresponds to a DNA
sequence encoding the variable region of a chain of a T-cell receptor. The first target polypeptide sequence corresponds to a variable region of a chain of a T-cell antigen receptor. Methods for treating an animal afflicted with a disease correlated with a T-cell clone containing a specific variable region. The method comprises treating the animal with at least one antibody which is reactive with the specific variable region of the T-cell clone.
Antibodies for detecting or treating a disease correlated with at least one T-cell clone containing a specific variable region wherein said antibody is specific for said variable region.

Description

WO91/0~23 1 PCT/US~/076~
DIAGNosIs hND TREATMENT OF DIS~ASE~ 2 0 7 2 3 5 6 t0 The invention described herein was made in the course of work under grants from the National Institutes of 15 Health, U.S. Department of Health and Human Services.
The V.S. Government has certain rights in this invention.

FIELD OF THE INVENTION
2G The invention relates to methods for diagnosing the predisposition onset and course of disease. The invention also relates to methods for treating disease.
Further, the invention relates to reagents which are useful for the diagnosis and treatment of such 25 diseases.

BACKGROUND OF ~HE INVENTION
The human T-cell receptor ~-chain (TCR~) locus has been extensively studied since the cloning of the first cDN~
30 encoding the ~-chain (Yanagi, `f. et al. (1984) Nature 30~, 145-149; Hedrick et al. t'984~ Nature 308, 149-152). This locus is a gene complex containing varia~le ~V), diversity (n;~ and joining (J) gene segments which participate in somatic cell 35 rearrangement with a constant (C) reyion gene segment WO9l/09~23 PCT/US90/076~

- 2 - 2~72~

to encode the ~-chain of the T-cell receptor (Chien et al. (1984) Nature 309, 322-326). By in situ hybridization, the TCR~ locus resides at 7q35 (Isobe et al. (1985) Science 228, 580). By current estimates, 5 this complex spans more than 600 kb and contains 70 to 80 variable region segments (Concannon et al. (lsB6) Proc. Natl. Acad. Sci. USA 83, 6598-6602; Tillinghast et al. (1986) Science 233, 879-883; Kimura et al.
(1987) Eur. J. Immunol. 17, 375-383; Lai et al. (1988) 10 Nature 331, 543-546). These V region genes are adjacent to two tandemly organized regions each of which include a D and a C gene segment separated by a cluster of six or seven J region gene segments (Tunnacliffe et al. (1985) Nucleic Acids Res. 13, 15 6651-6661; Toyonaga et al. (1985) Proc. Natl. Acad.
Sci. USA 82, 8624-8628). Following successful DNA
rearrangement of the V, D, and J gene segments, the translated ~-chain polypeptide pairs with a T-cell receptor ~-chain and can be expressed on the surface of 20 the T-cell (reviewed by Kronenberg et al. (1986) Annu.
Rev. Immunol. 4, 529-591). The T-cell receptor then functions in recognizing antigen in the context of a self major histocompatibility molecule (Dembic et al.
(1986) Immunol. Today 7, 308). The co-recognition of 25 antigen and major histocompatibility molecules by a T-cell must be specific and carefully regulated, since improper immune regulation fosters autoimmunity.

T-cells play a pivotal role in the differentiation and 30 regulation of effector mechanisms within the immune system (Paul et al. (1977) Science 195, 1293-1300).
Several laboratories have studied diseases in which there appears to be improper immune regulation, such as autoimmunity, anergy, and some forms of 35 immunodeficiency and have implicated T-cells in the WO9~/09623 PCT/US90/076~
~ 3 ~ 2072356 pathogenesis of such diseases. Such theories have recently been t~sted by studying germline DNA
polymorphisms of the ~- and ~ chain in normal and disease populations (reviewed by Kumar et al. (1989) 5 Ann. Rev. Immunol . 7 657-682~. Positive associations have been report~d between specific TCR~ restriction fragment length polymorphisms (RFLPs) and insulin dependent diabetes mellitus (Hoover et al. (1986) Cold S~ring Harb. Symp. Ouant. Biol. 51, 803-808; Milward et 10 al. (lg87) Clin. Exp. Immunol. 70, 152-157; Ito et al.
(1988~ Diabetes 37, 1633-1636), Graves' disease (Demaine et al. (1987) J. Clin. Endo. and Metab. 65, 643-646; Ito et al. (1989) J. Clin. Endrocrinol. Metab.
69, 100-104), membranous nephropathy (Demaine et al.
15 (1988) Immunoqenetics 27, 19-23), and rheumatoid arthritis (Gaô et al. (1988) Am. J. Med. 85, 14-16).
All of these positive associations are based on TCR~
RFLP frequency comparisons between populations. In each of these studies, the RFLP was found to be 20 correlated with one of the two C gene segments associated with the ~-chain.

One laboratory has reported an association between an RFLP linked to a V region gene and autoimmune 25 hypothyroidism as compared to a disease rather than normal population. Weetman et al. (1987) Human Immunology 20, 167-172. The validity of this conclusion, and its relationship to detecting such an association in a normal population or one at risk for 30 hypothyrodism is suspect.

An association between an RFLP linked to either an ~-variable or ~-constant chain segment and susceptibility to multiple sclerosis (MS) has also been 35 reported. (Martell, N. et al. (1987) C.R. Acad. Sc.

WO9l/09623 PCT/US90/07699 - ~ - 2~723~

Paris t. 304, Series III, no. 5, 105-110). This laboratory also investigated the possibility of an RFLP
in the ~-chain of the TCAR in MS patients but did not find a correlation between control and disease samples.
5 Based on an independent study of RFLPs of the ~-chain of the human T-cell receptor with the same ~-chain cDNA
probe cleaved to form variable and constant region probes, it appears that the above ~chain RFLP is linked to a constant rather than variable region. See 10 Hoover et al. (1985) J. Exp. Med. 162, 1087-1092.

Another laboratory performed similar experiments using a variety of restriction endonucleases and apparently the same TCR~ probe used by Martell et al. (Oksenberg, 15 J.R. et al. (1988) Human Immunol. 22, 111-121). This laboratory reported that no significant differences in the frequencies of the polymorphisms in the TCAR genes were found between the MS and control groups.

20 Recently, a laboratory reported T-cell receptor ~-chain haplotypes in patients with MS as defined by various RFLPs. Ciulla et al. (1988) Ann. N.Y. Acad. Sci. 540, 271-276. Two haplotypes reportedly had a tendency to be overrepresented in the MS group studied. The 25 authors concluded, however, that the results were not statistically significant.

Although the etiology of MS is unknown (McFarlin et al.
(1982) N. Enal. J. Med. 307, 1183-1188 and 1~46-1251), 30 it is believed that both environmental (Cook et al.
(1980) Neuroloqy 30~2), 80-91; Kurtzke (1980) Neuroloqy 30, 61-79) and genetic factors (Spielman et al. (1982) Epidemiol. Rev. 4, 45-65; McFarland et al. (1984) Ann.
NY Acad. Sci. 436, 118-124; Ebers et al. (1986) N.
35 Enql. J. Med. 315, 1638-1642) contribute and that 3 P~/US90/076~
~ 5 ~ ~072356 immunological mechanisms are involved in the pathogenesis. Studies in families (Spielman et al.
(1982) su~ra) particularly twins (McFarland et al.
(1984) supra; Ebers at al. (1986) su~ra) have led to 5 the hypothesis that two or more genes may be involved in susceptibility to MS. Previous genetic studies have focused on genes coding for HLA molecules that are co-recognized with foreign antigens by T lymphocytes (reviewed in Kronenberg et al. (1986) Annu. Rev.
10 Immunol. 4, 529-591). More recently, considerable progress has been made in the definition of T-cell antigen receptor (TCAR) molecules involved in this recognition process and the genes encoding these structures (Sim et al. (1984) Nature 312, 771-775;
15 Yanagi et al. (1984) Nature 308, 145-149; Ikuta et al.
(1985) Pro. Natl. Acad. Sci. USA 82=, 7701-7705;
Yoshikai et al. (1985) Nature 316, 837-840; Concannon et al. t1986) Proc. Natl. Acad. Sci._USA 83, 6598-6602;
Kimura et al. (1986) J. Exp. Med. 164, 739-750; Kimura 20 et al. (1987) Eur. J. Immunol. 17, 375-383; Tillinghast et al. (1986) Science 233, 879-883; Beall et al. (1987) J.Immunol. 139, 1320-1325).

TCR genes may be candidates for genetic elements 25 contributing to the susceptibility to MS and other immunologically mediated disorders. Support for this possibility has been obtained from animal studies which have identified associations between experimental demyelinating disorders and TCR genes. For example, 30 chronic-relapsing experimental encephalomyelitis (EAE) (Mokhtarian et al. (1984) Nature 309, 356-358), and Theiler's murine encephalomyelitis virus (TMEY)-induced demyelinating disease (Lipton (1975) Infect. Immun. 11, 1147-1155; Melvold et al. (1987) J._Immunol. 138, 35 1429-1433) can be induced in SJL mice which have WO91/09623 PCT/US90/076~
- 6 - 20723~S' deleted approximately 50% of the V~ genes of the TCR
(Behlke et al. (1985) Science 229, 566-570; ~ehlke et al. (1986) Proc. Natl. Acad sci. USA 33, 767-771; Lai et al. (1987) supra). In addition, differential 5 susceptibility to TMEV-induced demyelinating disease has been mapped near the constant region genes of the TCR (Melvold et al. (1987) supra).

It is believed that genetic background and an 10 environmental agent, possibly a virus, are involved in the etiology of MS (Spielman et al. (1982) supra;
McFarland et al. (1984) su~ra; Ebers et al. (1986) supra). Support for a significant genetic contribution to the pathogenesis of MS has been obtained from twin 15 studies which have shown higher concordance in monozygotic than dizygotic twins (McFarland et al.
(1984) supra; Ebers et al. (1986) supra). However, the disease is not entirely genetically controlled because the concordance in monozygotic twins is considerably 20 less than 100%. There is increased representation of DR2 in the Caucasian MS population which indicates that one genetic element may reside in the genes encoding the HLA complex (Tiwari et al. (1980) Histocom~atibility Testinq 1980 P. Teraski, ed., UCLA
25 Tissue Typing Laboratory, Los Angeles, CA, 687-692;
Batchelor (1978) Bri. Med. Bull. 34, 279-284).

Additional MS susceptibility genes have been postulated to exist, and recently, it has been suggested (Goodman 30 et al. (1987) Current Neuroloqy, S.H. Appel, ed., Year Book Medical Publishers, Chicago, IL, 91-127) that genetic variations in the TCR genes may play a role in the pathogenesis of MS.

WO 91/Og623 PCr/USgO/07699 2~7235~

The references discussed above are provided solely for their disclosure prior to the filing date of the instant case, and nothing herein is to be construed as an admission that the inventor(s~ are not entitled to 5 antedate such disclosure by virtue of prior invention or priority based on earlier-filed applications.

Based on the foregoing, it is apparent that a need exists for methods and reagents for diagnosing and 10 treating diseases such as ~S.

Accordingly, it is an object herein to provide methods and reagents to diagnose predisposition to a disease such as multiple sclerosis.
Further, it i~ an object herein to provide methods and reagents to diagnose the onset and course of such diseases.

20 Still further, it is an object herein to provide methods and reagents for treating disease.
SUMMARY OF THE INVENTION
In accordance with the foregoing objects, the invention includes methods for diagnosîng predisposition to a 25 disease. Such methods include obtaining a DNA test sample from an animal, such as a human, and detecting in that sample a target DNA sequence correlated with a susceptibility gene for the disease. The target DNA
sequence comprises a DNA sequence contained within or 30 in close proximity to a genomic DNA sequence encoding a variable region of a T-cell antigen receptor ~-chain.
In specific embodiments of the invention, the target DNA sequence comprises a restriction fragment length polymorphism (RFLP~ linked to one or more V~ segments 35 of a T-cell antigen receptor ~-chain~ The detection of WO91/~9623 PCT/US90/076~9 ~ 0 7 2 3 5 ~

such an RFLP provides an indication that a first susceptibility gene for the disease is present in the genome of the individual.

5 The detection of a first target DNA sequence associated with a T-cell antigen receptor chain may be combined with the detection of a second target DNA sequence associated with the variable region of a different T-cell antigen receptor chain. Further, such detection 10 of target DNA sequences, which may, for example, define and ~-chain haplotypes, may be combined with determining the Major Histocompatibility Complex (~HC) haplotype as a further indication of the predisposition of that individual to a particular disease.
The invention~also includes methods for diagnosing the onset or monitoring the course of a disease. Such methods include obtaining a T-cell nucleic acid sample from an animal such as a human and detecting in such a 20 sample a target nucleic acid sequence correlated with the disease. The first target nucleic acid sequence is formed of DNA or ~A which corresponds to a DN~
sequence contained within or in close proximity to a rearranged genomic ~NA sequence encoding a variable 25 region of a T-cell antigen receptor. Such variable regions include ~ and ~ variable regions of the T-cell antigen receptor. The detection of a first target nucleic acid sequence for a specific variable region or gene segment of a particular T-cell antigen receptor 30 chain may be combined with the detection of a second target nucleic acid sequence correlated with the same disease. This second target nucleic acid sequence is also formed of an RNA or DNA and corresponds to a DNA
sequence contained within or in close proximity to a 35 rearranged genomic DNA sequence encoding a specific wO91~0s623 PCT/US90/076~
2~723~

variable region or gene segment of a different T-cell antigen receptor chain.

Methods fox diagnosing the onset or monitoring of the 5 course of a disease also include obtaining a suitable sample containing T-cell antigen receptor from an animal, such as a human, and detecting in such a T-cell receptor sample a first target polypeptide sequence correlated with an disease. This polypeptide sequence 10 is contained within a variable region of a T-cell antigen receptor. Such variable regions include ~ and ~ variable regions of the T-cell antigen receptor. As with the a~ove-described methods, this detection of a f irst target polypeptide sequence may be combined with 15 the detection of a 5econd target polypeptide sequence, also correlated with the disease, wherein the second polypeptide sequence comprises a specific variable region of a different T-cell antigen receptor chain correlated with the disease.
The detection of target nucleic acid and polypeptide sequences correlated with disease, which include for example specific ~ and ~ variable regions or segments, may be combined with determining the MHC expression of 25 the individual as a further indication of the onset or cause of the disease.

The invention also includes methods and reagents for treating disease. In such methods, the disease is 30 correlated with at least one T-cell clone containing a specific variable region in the T-cell receptor of the T-cell clone. The method comprises treating an afflicted animal, such as a human, with at least one antibody which is reactive with the specific variable 35 region of that T-cell receptor.

WOgl/09623 PCT/US~/U7699 ~0723~

Further, the invention includes antibodies for detecting and treating disease correlated with at least one T-cell clone containing a specific variable region or specific variable segment in the T-cell receptor of 5 the T-cell clone.

BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 is a schematic representation of the four types of myelin basic protein (MBP) T-cell receptors in 10 B10.PL mice. The types are defined on the basis of the combination of ~ and ~ T-cell receptor chains. The bars represent T-cell receptor hete~odimers with ~
chains on the left and ~ chains on the right. Numbers beside each bar indicate variable (V) and joini~g (J) 15 gene segments (V on top, J on the bottom) constituting complete genes encoding each change (diversity gene segments for the b chain are not shown). The circles with different shading represent T-cells with different profiles of antigen fine specificity. The numbers 20 below each type indicate the total number and percentage of T-cells determined to belong to each category.

Figure 2 identifies the nucleotide sequences of human 25 chain cDNAs. The clones from which nucleotide sequences were determined are indicated at the left.
Clones designated PL were derived from a peripheral lymphocyte cDNA library. Other clones derived from established T-cell lines are labelled with the name of 30 the cell line except for V~YT35, which is derived from a MOLT-3 cell line (Eyanagi (1984) Nature 308, 145-149). Clone ATL21 (Ikuta, et al. (1985) Proc. Nat.
Acad. Sci. U.S.A. 82 7701-7705) was derived from a genomic sequence and introns have been edited out in 35 this alignment. Additional germ-line V~ gene segment WO~1/09~23 PC~/US~/076~
`` - 11 - 2~723~

sequences homologous to PL2.13, PL~.9 and PL5.10 have been published, but are not shown here ~Ikuta, supra).
The sequences are listed in the order V~1-V~15, as in Figure 3.
Figure 3 is the amins acid sequence of murine and human V~ genes. Human (h) and murine (m) V~ gene-segment nucleotide sequences were translated and aligned to maximize homology at the amino acid level. Amino acids 10 conserved at the 75~ level are indicated by asterisks.
V~2.1 and V~2.2 differ only in the leader sequence and hence appear identical in this comparison. Amino acids are identified by the single-letter code.

Figure 4 is a Southern blot analysis displaying the most common restriction patters for T-cell receptor V~
gene segment subfamilies 1 through 14. Two different digests are shown for each subfamily. Vertical lines indicate the estimated numbers of gene segments within 20 individual subfamilies. In some cases, estimates of subfamily sizes were made based on additional hybridization data not shown. Subfamilies V~12 and V~13 appear to share common members as indicated by nucleotide sequence data.
Figure 5 depicts the segregation of polymorphic V~ and C~ alleles. Each chromosome indicates the alleles detected at the V~8, V~11 and C~ loci ordered top to bottom, respectively. Numbers below schematic 30 chromosomes are for subject identification.

Figure 6 demonstrates linkage disequilibrium between the alleles at the v~8 and V~ll loci in humans.
Southern blots display the hybridization patters for 35 both homozygotes and heterozygotes at each locus.

WO91/09623 PCT/US~/07699 - 12 - ~ ~723~6 Allelic frequencies are indicated to the left.
Fragment size is indicated to the right. The distribution of haplotypes in a population of 52 unrelated individuals generated by genotyping at the 5 V~8, V~11 and C~ loci are indicated in the lower part of the figure. The distribution of haplotypes expected was calculated from the allelic frequencies.

Figure 7 depicts a restriction enzyme map for the 10 murine V~ locus.

Figure 8A depicts the nucleotide sequence of TCR ~ cDNA
clones derived from 70 hybridomas specific for the MVP
peptide l-9NAc.
Figure 8B depicts the translated protein sequence of two distinct V~ sequences from Figure 8A.

Figure 9 depicts the cell surface expression of a V~8 20 chain by T-cells specific for the MBP peptide 1-9NAc and the in vivo elimination of V~8 plus lymphocytes following i.p. injection of F23.1 antibody. In panels A through C, TH PL172 .10 and PL212.6 and TH clone 2C6 were analyzed for T-cell cell surface expression using 25 quantitative fluorescence of flow cytometry. The T-cells were stained with the monoclonal antibody of 23.1 (anti-V~8) or 500.A.A2 all by fluorescein isothyoceinate (FITC)-conjugated goat anti-mouse IgG.
Only PL172.10 and 2C6 Tu cells express surface V~8 30 chains (PL212.6 cells express the V~13 gene segment).
Background fluorescence (Bkgrd) represents staining by the FITC antibody alone. Each plot represents the analysis of 20,000 cells. In panels D and E of Figure 9, the B10.PL mice were injected i.p. with 500 35 micrograms protein A-purified F23.1 antibody and 0.5 ml W091/09623 PCT/US9~/0~6 20~2~
normal saline. Spleen cells were removed in 0, 24 and 72 hours after injection (D) and 5 and 7 days after injection (E). They were thereafter passed over nylon wool, stained with F23.l antibody and FITC-conjugated 5 goat anti-mouse IgG and analyzed by fluorescence flow cytometry. Each plot represents the analysis of 20,000 cells.

Figure lO depicts the inhibition of MBP response with 10 anti-TCR monoclonal antibodies.

Figure ll is a histogram demonstrating the down modulation of v~8.2 T-cells following _n vivo antibody injections of an EAE mouse.
Figure 12 deplcts the effect of in vivo monoclonal antibody treatment on the response of EAE mice to MBP.

Figure 13 depicts the comparison of the response to MBP
20 and monoclonal antibody treated mice with and without EAE.

Figure 14 demonstrates treatment with anti-V~
monoclonal antibody results in reversal of MBP-induced 25 paralysis in EAE mice.

Figure 15 depicts the allelic frequencies of 38 caucasian, unrelated MS patients compared to lO0 normal individuals. The hybridization pattern at each locùs 30 is displayed by Southern blot for both homozygotes and heterozygotes. Allelic frequencies are indicated to the 1eft of each pattern and fragment size to the right. Allelic frequencies of MS patients are indicated first followed by those determined for normal WO91~09623 PCT/~90/076~
- 14 - 20~235~

individuals in parentheses. Asterisks (*) indicate invariant bands.

DETAILED DESCRIPTION
5 T-cells play an important role in human diseases involving the immune system, such as, for example, autoimmune diseases. In autoimmune diseases, T-cells are believed to act as the causative agents that incorrectly recognize the body itself as foreign. The 10 action of T-cells is mediated through antigen recognition by the cell antigen receptors present on the T-cell surface. These T-cell antigen receptors (TCARs) (also sometimes referred to as T-cell receptors or TCRs) are unique to each different type of T-cell 15 clone and can be used to specifically identify a particular cell line.

The present invention is based, in part, on the discovery that certain restriction fragment length 20 polymorphisms (RFLPs) linked to specific V~ gene segments are in linkage disequilibrium with at least one putative susceptibility gene for multiple sclerosis. The invention is also based, in part, on the discovery that clonal populations of T-cells 25 containing specific variable regions in both the ~ and ~-chain of the TCAR are present in a mouse model system for multiple sclerosis. See Figure l. In this mouse model system, treatment with a monoclonal antibody specific for a particular V~ segment in the TCAR of a 30 clonal population of T-cells associated with murine experimental encephalomyelitis (EAE) reduced incidence of the disease. This work is described in more detail in co-pending patent application Serial No. 229,288, filed August 8, 1988, which is expressly incorporated 35 herein by reference. The invention is also based on WO91/09623 PCl/US90/076~
- 15 - ~ ~72~6 the discovery that treatment of EAE with a second monoclonal antibody specific for a different V~ segment contained in the TCAR of other T-cell clones identified with that disease leads to nearly complete protection 5 against the induction of EAE. This treatment also resulted in a dramatic reversal of paralysis in diseased animals.

In addition to the foregoing, the invention is based upon the discovery that a multiple sclerosis susceptibility gene is located in the human genome centromeric to V~11 and likely between V~8 and V~11.
Since there are at least six TCAR V~ genes between V~8 and V~11 (Lai et al. (1988) Nature331, 543-546, it is 15 believed that a gene encoding a V~ segment in this chromosomal region will be found to be a multiple sclerosis susceptibility gene. Given the discovery of clonal T-cells containing specific T-cell variable regions in EAE in mice and the above discoveries 20 relating to the localization of the multiple sclerosis susceptibility gene in the V~ locus in humans, it is believed that diseases such as multiple sclerosis are also caused, in part, by clonal populations of T-cells containing specific TCARs.
Given the success in preventing EAE in mice by the administration of a monoclonal antibody and the significant reversal of paralysis observed when multiple antibodies are administered in mice afflicted 30 with EAE, it is believed that similar reagents can be used to diagnose and treat disease in other animals such as humans.

Any disease that involves T cells can be studied to 35 identify TC~R disease correlations. These correlations WO9l/09623 PCT/U~a/0769~
- 16 - 2~23~

can then be used to produce diagnostic, therapeutic or disease monitoring procedures. Relevant diseases include but are not limited to autoimmune diseases, neoplastic diseases, infectious diseases, 5 hypersensitivity, transplantation and graft-versus-host disease, and degenerative nervous system diseases.
Autoimmune diseases include but are not limited to arthritis such as rheumatoid arthritis, type I
diabetes, juvenile diabetes, multiple sclerosis, 10 thyroiditis, myasthenia gravis, systemic lupus erythematosis, Sjogren's syndrome, Grave's disease, Addison's disease, Goodpasture's syndrome, scleroderma, dermatomyositis, myxoedema, pernicious anemia, autoimmune atrophic gastritis, and autoimmune hemolytic anemai. Neoplastic diseases include but are not limited to ly~phoproliferative diseases such as leukemias, lymphomas, Non-Hodgkinls lymphoma, and Hodgkin's lymphoma, and cancers such as cancer of the breast, colon, lung, liver, pancreas, etc. Infectious 20 diseases include but are not limited to viral infections caused by viruses such as HIV, HSV, EBV, CMV, Influenza, Hepatitis A, B, or C; fungal infections such as those caused by the yeast genus Candida;
parasitic infections such as those caused by 25 schistosomes, filaria, nematodes, trichinosis or protozoa such as trypanosomes causing sleeping sickness, plasmodium causing malaria or leishmania causing leischmaniasis; and bacterial infections such as those caused by mycobacterium, corynebacterium, or 30 staphylococcus. Hypersensitivity diseases include but are not limited to Type I hypersensitivities such as contact with allergens that lead to allergies, Type II
hypersensitivities such as those present in Goodpasture's syndrome, myasthenia gravis, and 35 autoimmune hemolytic anemia, and Type I~

WO91/09623 PCT/US90/076~
` - 17 - 2~7~3~6 hypersensitivities such as those manifested in leprosy, tuberclosis, sarcoidosis and schistosomiasis.
Degenerative ~ervous system diseases include but are not limited to multiple sclerosis and Alzheimer's 5 disease.

As used herein~ a "disease" is any immunological disease which is mediated, at least in part, by a T-cell clonal population. Such diseases are capable of 10 afflicting any animal containing an immune system. The preferred animal of interest, of course, is the human being~ Other preferred animals include domesticated animals such as equine, bovine, ovine, porcine, canine, feline and murine species. Diseases in such other species may be analogous to those identified in humans or be uniquel~ characterized for a particular species or group of species. Thus, the methods and reagents of the present invention to diagnose and treat disease are useful to the medical and veterinary disciplines.
T-cell antigen receptor specific reagents, such as antibodies or nucleic acids, are used herein for the diagnosis, treatment, and monitoring of disease. The first step in this process is to correlate the presence 25 of a specific TCAR with a disease state. This correlation may either involve the expression of a particular TCAR on a subset of disease related cells or may involve an analysis of the TCAR genes that are inherited by the individual and their relationship to ~0 predisposing that individual to developing a particular disease. Once such a correlation is made, it is used to detect or diagnose that disease state in patients.

TCAR reagents may further be used to treat the disease 35 state, in their capacity to specifically modulate the WO91/09623 P~T/US~/076~
- 18 - 20723~6 action of the T-cells they identify. They may also be conjugated with various toxins to further modulate the targeted T-cells. In addition, throughout or following the treatment of a particular disease, TCAR reagents 5 are used to monitor the disease treatment to determine whether the treatment is effective and the disease is in remission, or to determine whether the patient has relapsed with a return of the disease state, or to determine whether the patient remains in a stable, 10 disease free condition. During the monitoring period, the TCAR reagents are useful both diagnostically to indicate the presence or absence of disease, as well as therapeutically to treat the disease.

15 Diagnosis of Disease Based U~on TCARs The diagnosis~of disease in humans or other animals as described herein requires a correlation between specific TCARs with each disease. This correlation is determined by the methods and reagents described 20 herein.

The diagnosis, treatment and monitoring of disease is based upon disease correlations determined by comparing the TCARs in disease related samples with suitable 25 baseline samples. The disease correlations and resulting clinical procedures can be based upon quantitative as well as qualitative differences in the TCARs expressed in the different samples. Thus, a relative change in the value of a correlating TCAR in 30 serially obtained samples during the course of a treatment is of value; as is the absolute value at each point in the series.

In a clinical setting, disease diagnosis is made using 35 a variety of procedures and TCAR reagents. Such WO91/09623 PCT/US~/07699 - 19 - ~723~6 diagnosis involves the use of TCAR specific reagents that are labeled with detectable moieties such as biotin, radiolabels, fluorescent labels, or metal ions to name a few. These reagents include either nucleic 5 acid probes or antibodies and can be used in a variety of clinical procedures including but not limited to imaging analysis, RFLP analysis, PCR analysis, fluorescent cytometry, fluorescent microscopy, in sltu hybridization, nuclear magnetic resonance analysis, 10 ELISA analysis, etc. Diagnostic procedures can be either performed in vitro or in vivo.

Diagnostic kits are useful in performing the diagnosis.
Such kits include the necessary TCAR reagents coupled 15 with the appropriate detectable marker for the analysis, suitable standards, solid phase components such as microscope slides, microtiter dishes, or beads as necessary, other active components such as enzymes and substrates useful for detection, etc. If multiple 20 TCAR reagents are required to perform the diaqnosis, kits include each such reagent; as for example, MHC, V~
or V~ specific reagent(s).

Treatment of Disease Based_Upon_TCARs 25 The treatment of disease based upon TCARs in humans or animals is most effective when only those T-cell subsets involved in the disease are specifically modulated. Treatment of larger T-cell subsets can also be effective, but has the potential disadvantage of 30 modulating non-disease specific T-cells as well. In practice, it may be necessary to modulate a subset of normal T-cells having variable region sequences common to those used by the T-cell clone(s) which mediate the disease.
~5 WO91~09~23 P~/US90/076~
2072~6 Effective immunotherapies based upon TCA~ reagents involve either the ablation or deletion of disease specific T-cells, or alternatively, the induced proliferation of specific T-cells. Thus, in some 5 diseases it is desira~le to remove a deleterious population of T-cells that are causing the disease. In other cases, it may be desirable to specifically stimulate a disease related subpopulation of T-cells to proliferate, if these T-cells are beneficial in 10 treating the disease or in suppressing it. In each of the cases, the dose of the TCAR reagent, for example, an antibody specific for a particular TCAR, should be predetermined by in vitro techniques known in the art, such that the cells are either correctly stimulated or 15 else correctly targeted for removal.

The mode of treatment can involve either acute or chronic treatment conditions. This in turn will lead to treatment regimens involving either one time bolus 20 administrations, continuous administrations or repeated administrations. The dose forms can include, but are not limited to injectable forms applied by either intraperitoneal, intramuscular, or intravenous injection, slow release forms, such as by those 25 delivered in transplantable forms, on patches, or in other colloidal forms. The reagent itself needs to be properly formulated, as for example, a humanized antibody combined with various sugars, buffers or stability causing compounds that extend the stability 30 or half-life of the reagent. To enhance the half-life, the reagent can first be modified to increase or decrease the amount of carbohydrates complexed to it, or alternatively to complex it with reagents such as polyethylene glycol (PEG). Finally, pharmaceutical WO91/09623 PCT/US90/076~
- 21 ~ ~ ~723~6 compositions comprising the therapeutic reagent in the appropriate buffers, salts and pH are required.

Monitoring Disease Therapy or 5 Progression Based Upon TCARs The same reagents that are used to diagnose or treat disease can also be used to monitor the effectiveness of a disease therapy or to monitor the progression of the disease throughout phases of remission, relapse or stable periods. In arthritis, for example, the initial disease correlation is made on those TCARs that are expressed by the arthritic T-cell clone(s). TCAR
reagentts) based on this correlation can then be used to diagnose this arthritis and to treat it. Further, 15 during the treatment course, as the arthritic cells are being elimina$ed, the TCAR reagent(s) can be used to track the extent of the elimination of the arthritic cell subset to the point where all the cells are gone and the disease is in remission. Following remission, 20 the TCAR reagent(s) can be periodically used to test for the reappearance of the arthritic cells. Thus, the patient is periodically rediagnosed to determine whether the individual is in a stable remission, has relapsed and needs to be retreated, or is undergoing a flareup of acute disease. Thus, the TCAR reagents are also useful in combined diagnostic and therapeutic procedures.

Correlation of T-cell Antiqen Receptors with Disease 30 The diagnosis, treatment and monitoring of T-cell related diseases depends upon the correlation of specific TCARs with disease states. An infinite spectrum of T-cells exists in the body that are involved in recognizing an equally infinite spectrum of 35 disease related antigens. The specificity of this recognition is accomplished by the interaction of WO91/09623 PCT/US90/076~
- 22 - 2~7235~

specific TCARs with disease antigens. Thus, the diversity of the TCARs themselves is also very high.
This diversity is created by the multiplicity of TC~R
germline segments, the combinatorial joining of these 5 different segments, the junctional flexibility and N-region diversity at the site~ of the joined segments, and the use or different translational reading frames or multiple D region segments (U.S. patent application, Serial Number 726,502, filed April 24, 1985, now U.s.
10 Patent No. 4,886,743). Any given disease correlation is determined by identifying which of these diverse TCARs is indicative of the disease, as described in more detail herein.

Suitable samples There are many different acceptable samples that can be used to detect a correlation of TCARs with different disease states or to diagnose disease or to monitor disease therapy. Depending upon the disease, these 20 disea5e specific samples include cells from the peripheral blood, synovial fluid, pleural fluid or spinal fluid as well as those that have infiltrated tissues, synovium, skin lesions or components of the nervous system. Such infiltrated cells can be studied 25 either directly in tissue sections or else allowed to diffuse from the tissues and studied directly or expanded in tissue culture.

Depending upon the amount of compartmentalization of 30 the immune system cells in a disease, some samples will be better than others. For instance, the cells that have actually infiltrated a synovium are preferable to the whole population of cells present in the peripheral blood when trying to analysis rheumatoid arthritis 35 specific TCARs. In other diseases, however, the WOgl/09623 PCT/U~90/076~
- 23 - 2~723~6 peripheral blood or spinal fluid is the sample of choice.

Correlations and diagnosis of a predisposition to 5 disease can be made using restriction fragment length polymorphisms (RFLPs) of inherited TcAR genes. Any DNA
test sample containing genomic DNA can be used for this analysis. However, the analysis is much less complicated if the DNA test sample does not contain a 10 substantial amount of T-cells. This is due to the fact that the rearranged TCAR yenes observed in T-cells, complicate the analysis of the presence or absence of different TCAR alleles. For RFLP analysis the samples of choice are peripheral blood B cells or fibroblasts.
To identify disease correlations or to diagnosis, treat or monitor disease, additional samples are also useful.
These include samples that can be used to obtain comparison values. One such comparison sample is a 20 similar sample taken from a normal individual who does not have the disease of interest. The normal individual's sample can be used as a measure of the relative increase or decrease of the TCAR in the diseased patient's sample. Other useful samples 25 include samples taken from the same patient before the onset of disea~e or before the initiation of therapy.
For rheumatoid arthritis patients, for example, a sample taken from the same patient during a stable period is extremely useful to obtain a value to compare 30 with a sample taken during an acute flareup period.
Similarly, when monitoring the efficacy of a disease treatment for arthritic patients for instance, samples taken from the same patient before the therapy begins, during the therapy, and during periods of remission and 35 relapse are all valuable. These longitudinal or serial WO 91/09623 PCI /USgO/07699 20723~
samples can be used to show the amount of change observed in the relevant TCAR marker.

Methodology to Identify Autoimmune 5 TCAR Disease Correlation Any of the TCAR reagents described herein can be used to identify disease correlations useful for disease diagno~is and prognosis. The disease correlations fall into two categories. The first category involves correlations between the expression of certain TCARs and the presence of certain diseases. The second category involves prognosis based upon the analysis of the inherited TCARs in an individual's genomic DNA to determine whether the actual TCAR qenes that were 15 inherited can predispose that individual to develop a disease. In the later case, the disease itself might not yet be manifested and the TCARs t~at correlate with the disease might not yet be expressed. The methodologies described infra. illustrate some of the 20 ways in which disease correlations with TCAR reagents can be made; others will be known by those skilled in the art.

l. Disease Correlations Based Upon the Expression of TCARs 25 In a normal individual, there is a diverse population of T-cells each expressing a unique TCAR heterodimer.
These cells are not activated until they come in contact with their cognate disease specific antigen presented as a complex with major histocompatibility 30 antigens. once the antigen is recognized, the T cell becomes activated and clonally expands resulting in a subset of T-cells each of which expresses the same TCAR. This TCAR is an identifying characteristic of the disease specific T-cell subset. If more than one, but a limited number of T-cells, become activated by WO91/09623 PCT/~S90/07699 2~72356 the disease specific antigen(s), then multiple clones expand resulting in an oligoclonal subset of T-cells.
The oligoclonal subset is still only expressing a limited number of TCMs which in turn can be used to 5 identify the oligoclonal subset.

Techniques that are useful for identifying disease correlations must be able to detect the disease specific subset or oligoclonal subset of T-cells in the 10 presence of all of the other T-cells that are present in the sample. Some of the techniques that can be used are described infra.

a) cDNA Libraries 15 For this technique, the RNA is isolated from the T-cells in a sample and used to synthesize complementary DNA (cDNA) which is in turn cloned into bacterial cells to produce the library. The library represents the complete set of TCARs that were 20 expressed in the sample at approximately the same proportions as expressed. An analysis of the clonPs in the library, by sequencing the TCAR cDNAs or hybridizing ~hem to known TCAR probes, for example, indicates which TCARs are expressed most often. This 25 is a characteristic indicative of a disease expanded subset of cells each expressing the same TCAR. One advantage of a cDNA library is that it is possible to analyze each V, D, J, and C region expressed by the dominantly expressed TCARs to determine their 30 significance in the correlation to disease.

b) Polymerase Chain Reaction Technologies Instead of producing a cDNA library, it is possible to 35 primarily amplify TCARs by using TCAR specific primers in a polymerase chain reaction (PCR) (U.S. Patent WO91/Q9~23 PC~/US9~/07699 ~ 26 - ~ ~723~6 4,683,195 by Mullis et al.). As an example of this technology, the RNAs obtained from a disease specific sample are first used to synthesize cDNA. The cDNA is in turn amplified with TCAR specific primers by PCR to 5 produce highly TCAR enriched DNAs. These DNAs can then be sequenced or labelled and then used to probe a nitrocellulose filter that has been spotted with a panel of known TCAR genes. In either case, the most abundantly identified TCAR sequences or genes are 10 deemed to be indicative of the oligoclonal or disease specific subset of T-cells. Alternatively, a one-sided PCR procedure can be used which is analogous to that described by Loh E.Y. et al. (1989) Science 243, 217-220.
c) In Situ Hybridization The disease sample itself can be analyzed directly by in situ hybridization. In this procedure, the TCAR RNA
being expressed inside each individual T-cell can be 20 detected by hybridizing those cells to TCAR specific gene probes. Generally, the sample is divided into multiple samples that are in turn probed with a panel of TCAR genes. This panel can include V, J DJ, or C
reg.ion specific probes. If a cell is expressing a 25 particular TCAR RNA, it will be detected as a positive when hybridized to the corresponding TCM gene probeO

d) Southern Analvsis Genomic DNA can be prepared from the disease sample and 30 analyzed by Southern analysis using a panel of TCAR
gene probes. Each probe will detect the unrearranged and rearranged TCAR genes in each T-cell of the sample.
Since most of the non-disease T-cells will be present at very low percentages, the rearranged bands 35 corresponding to the~e T-cells will be too weak to see WO91~09623 ~CT/US90/076~
- 27 - 207~3~6 above background by Southern blot analysis. For an expanded disease T-cell subset or oligoclonal subset, however, the rearranged TCARs will be present at higher levels, and are detectable as band(s) distinct from the 5 unrearranged bands.

~ ) Flow Cytometry Analysis Anti-TCAR antibodies can be used to screen disease specific T-cell samples for the expression of specific 10 TCARs. The antibodies can either react directly with the intact TCAR on live cells, or the cells can be pre-treated with ethanol or other suitable reagents, so that antibodies reactive with denatured TCAR can be used. The sample is divided into multiple portions and 15 each portion in analyzed by FLOW with one antibody or an antibody Fool. The antibodies that react with the greatest proportion of cells in the sample are deemed to represent the disease specific cells.

20 Using the same panel of antibodies, samples can also be analyzed by fluorescent microscopy instead of fluorescent cytometry.

2. Disease Correlations Based Upon the I~heritance of TCA~ plotypes TCARs are inherited as a multigene family. For each TCAR gene, one allele will be inherited maternally, and one allele will be inherited paternally. Differences in the alleles inherited for each TCAR gene can be 30 defined by restriction fragment length polymorphisms (RFLPs). Several of these have been described (copending U.S. patent application, Serial Number 07/176,706, filed April 24, 1985). TCAR haplotypes can be defined for each individual by identifying which 35 alleles (or RFLPs) are observed in the genomic DNA of that individual. This is preferably done by Southern WO91/09623 PCTtUS90/076~
- 28 - 2 0723~ 6 analysis, where the individual's genomic DNA is digested with various restriction enzymes known to define a RFLP of interest, and then hybridized to a panel of TCAR gene probes. Once the DNA sequence 5 surrounding the restriction enzyme site responsible for the RFLP is known, then oligonucleotides can be synthesized corresponding to ~lanking sequences around the site and the technique of PCR can also be used.

10 A disease correlation is then made by determining whether the inheritance of particular RFLPs for TCARs (or TCAR gene haplotypes) occur more often in patients manifesting the disease than in those where the disease is not present. Family studies are especially useful 15 in this type of analysis, as the inheritance of a TCAR
gene haplotype can be traced throughout the generations of the family and correlations made between affected and unaffected individuals. From this analysis it becomes possible to predict that an individual, 20 although not currently manifesting the disease, is predisposed to developing the disease, based upon that person's inheritance of the disease indicating TCAR
haplotype.

25 For disease correlations, a RFLP involving the variable regions of TCARs will be preferred over those involving the constant regions~ It is the variable regions of TCARs that interact with antigens and are, thus, disease specific. There are only 2 ~ constant regions ~0 and l ~ constant regions of TCARs that are used by the majority of T-cells.

Disease correlations based upon inherited TCAR alleles can also be identified by using suitable anti-TCAR
35 antibodies or fragments thereof containing the binding WO9~/09623 PCT/~S90/076 20723~6 domain. TAus, if an antibody reacts with the epitope that is unique to a specific inhexited allele, that antibody can be used to detect the presence or absence of that allele.

Diagnosls of Predisposition to Disease In those aspects of the invention wherein a disease has an etiology associated, in part, with an inheritable qenetic predisposition, at least one putative 10 "susceptibility gene" for the disease may be present in the animal at risk for the disease. Such susceptibility genes include, but are not limited to, T-cell receptor genes encoding the variable region of and/or ~-chains of the TCAR. Such variable regions in 15 the human ~-chain of the T-cell receptor include V~
gene segments, J~ gene segments and D~ gene segments.
In the case of human TCAR ~-chains such variable genes include those encoding V~ gene ~egments and Ju gene segments. Such V, J or D gene segment may comprise 20 normal or abnormal genes. Thus, for example, a V~ gene segment may be normal in that when it is found in a population, there is little or no variation in its nucleotide and amino acid sequence. In such a case, the inheritance of such a "normal" gene may result in a 25 haplotype indicative of a predisposition to a disease.
In the other case, an abnormal gene segment, l.e. one in which there i5 substantial allelic variation between individuals, may define a haplotype wherein the inheritance of the "abnormal" gene segment causes a 30 predisposition to a disease.

Even if a susceptibility gene for a particular disease is not identified, a predisposition to such a disease can be detected by analyzing the inheritance of one or 35 more "target DNA sequences" correlated with such a 2~723$6 susceptibility gene. Such target DNA sequences are typically contained within or in close physical proximity to a genomic DNA sequence encoding a variable region of a TCAR ~ or ~-chain. Target DNA sequences 5 are "correlated" with a susceptibility gene if it is demonstrated that each is in linkage disequilibrium with such a susceptibility gene. Likewise, a target DNA sequence is in close proximity to a variable region if it is in linkage disequilibrium with a DNA sequence 10 encoding a variable region or segment of a TCAR chain.
Linkage disequilibrium is determined by methods well known to those skilled in the art and as described herein.

15 In some diseases, more than one susceptibility gene may be responsible for a particular disease. In those cases, one or more second target DNA sequences correlated with a second susceptibility gene are also detected as an indication of the predisposition of an 20 individual to that disease. In addition, the Major Histocompatibility Complex (MHC) haplotype may provide a further indication of predisposition to a particular disease. For example, in some populations, approximately 80% of the individuals afflicted with 25 multiple sclerosis have an MHC haplotype of DR2+.

The presence of a first target DNA sequence, for example, within the variable region of a T-cell receptor ~-chain correlated to a first susceptibility 30 gene defi.nes a ~-chain haplotype indicative of the predisposition of the animal to that disease. When a second susceptibility gene for a particular disease is also identified as a part of the etiology of a particular disease, e.q. a susceptibility gene located 35 within the T-cell receptor ~-chain variable region WO91/09623 PCTJUS90/~7699 20723~6 locus, an ~-chain haplotype is defined by a second target DNA sequence as a further indication of predisposition to such a disease.

5 of course, it is to be understood that predisposition is not to be construed as an absolute indication that a particular disease will develop. Rather, such a determination identifies the risk of an individual for developing such a disease. Thus, an individual may 10 have haplotypes indicative of a predisposition to MS
but may not develop symptoms of such a disease unless exposed to other environmental factors required to precipitate the onset of the disease, _.~., a viral infection, etc.
When diagnosis of predisposition to a disease is desired, a preferred DNA test sample is genomic DNA
from an individual animal. This is especially preferred ~hen the target DNA sequence correlated with 20 a susceptibility gene lies outside the structural region encoding a segment of a variable region of a T-cell antigen receptor chain such as the ~ or ~-chain.
In such a situation, the target DNA sequence may comprise a unique seguence which encodes or destroys a 25 DNA sequence recognized by a particular restriction endonuclease. In such a situation, the target DNA
sequence is detected by digesting genomic DNA with the appropriate restriction endonuclease. Thereafter, the digested genomic material may be separated on the basis 30 of size by electrophoresis and probed with a labelled oligonucleotide, e.a., a labelled V~ or V~ cDNA to determine if the cleavage site of the restriction endonuclease correlated with the susceptibility gene is present in the genomic DNA. Alternatively, the 35 presence or absence of a target DNA sequence encoding WO9l/09623 PCT/US~/0~6~
- 32 - 207~3~6 an RFLP may be detected by using an oligonuleotide ligase assay procedure. Landegren U. et al. (1988) Science 241, 1077-1080; Landegren, U. et al. (1988) Science 242, 229-347.
s Target DNA sequences other than those encoding RFLPs may also be used. For example, a unique sequence in close proximity with a variable region gene may be in Iinkage disequilibrium with a susceptibility gene. To 10 the extent that such a sequence does not encode an RFLP, such a unique sequence may be detected in genomic DNA by the polymerase chain reaction (PCR) (Sakai et al. (1986) Nature 329, 166). In this method, two oligonucleotide primers complementary to opposite 15 strands each located at or near each of the ends of the unique DNA se~uence defining the target DNA are used in conjunction with a DNA polymerase to exponentially synthesize the target DNA sequence. Such amplified target DNA sequences may thereafter be detected by an 20 appropriately labelled probe. Numerous modifications to this basic PCR technique have been made by those skilled in the art and such modified PCR procedures may be readily adapted to detect the target DNA sequences as defined herein.
In some instances, the target DNA sequence may be contained within a susceptibility gene for a particular disease. In such cases, the sequence of all or part of the susceptibility gene will be known. To the extent 30 that such a susceptibility gene has sufficient sequence divergence from other genomic sequences, it may be detected by standard hybridization techniques.
Alternatively, the above described PCR techniques may be used to detect the target sequence contained within 35 the susceptibility gene. Further, to the extent the WO91/09623 PCT/US90~076 207~3~

susceptibility gene itself encodes an RFLP, the presence of the susceptibility gene may be detected or confirmed by RFLP analysis with an appropriate probe for the susceptibility gene or other gene sequences 5 linked to the RFLP.

In many cases, the target sequence correlated to a susceptibility gene is linked to one or more gene segments encoding the variable region of a ~ or ~ TCAR
10 polypeptide chain. When such is the case and the target sequence comprises an RFLP, appropriate probes for such variable region gene segments may be used in conjunction with an appropriate restriction endonuclease. For example, in the case of the various 15 subfamilies of V~ gene segments, a panel of cDNA probes for the V~ subfamilies are used. See Figure 2 and Concannon et al. (1986) Proc. Natl. Acad. Sci. USA 83, 6589-6602. For V~ gene segments, cDNA probes for the V~ subfamilies are used. See Klein M. et al. (1987) 20 Proc. Natl. Acad Sci. U$A 84, 6884-6888.

In a specific embodiment of the invention, specific RFLPs associated with 15 subfamilies of V~ gene segments were used to determine if a correlation 25 existed between such RFLPs and a predisposition to multiple sclerosis in humans. As described in more detail in the examples, various haplotypes defined by RFLPs associated with V~8 and V~ll were used to detect a correlation between such RFLPs and a putative 30 multiple sclerosis susceptibility ~ene. Based on the extent of linkage disequilibrium measured between these RFLPs and MS, it was determined that this putative MS
susceptibility gene is located between gene segments V~8 and V~ll in the human genome. Although the 35 susceptibility gene has yet to be completely WO91/09623 PCT/USgO/076 characterized, it is believed to be a V~ gene segment which is used in a T-cell antigen receptor of a T-cell clone which mediates the onset and course of MS. Thus, the identification of individuals having a haplotype 5 indicative of the presence of one or more of these RFLPs provides an indication of the presence of the multiple sclerosis susceptibility gene in that individual.

10 As previously indicated, multiple genetic components are suspected of causing MS. In this regard, since the T-cell receptor comprises an ~ and ~-chain which recognizes antigen in the context of an MHC molecule, such other genetic components may reside in the MHC
15 locus and/or a variable region locus encoding the T-cell receptor u-chain.

In a second embodiment of the invention, an RFLP within or in close proximity to the V~16 gene segment has been 20 correlated with susceptibility to MS based upon the linkage disequilibrium between this RFLP and MS. A
putative second MS susceptibility gene is therefore located within the a chain variable region gene locus.
Thus, in addition to defining a ~-chain haplotype for 25 predisposition to MS, an ~-chain haplotype may also be defined for an individual which provides by itself or in combination with an indication of an increased risk of that individual for developing MS.

30 Diagnosis of Onset or Monitoring of Course of Disease In another embodiment of the invention, methods are provided for diagnosing the onset or monitoring of the course of a disease. In these embodiments, a test 35 sample comprises either a T-cell nucleic acid sample or a T-cell polypeptide test sample. As used herein, a WO91/09623 PCT/~S90/076 - 35 ~
2~23~6 "T-cell nucleic acid test sample" comprises DNA or RNA
corresponding to rearranged genomic DNA from T-cells.
Such rearranged genomic DNA encodes the ~-chains and ~-chains used by the population of differentiated T-cell 5 to form TCAR RNAs and polypeptides. A T-cell nucleic acid test sample typically is an appropriate tissue or fluid sample containing T-cells. T-cells in such a sample may if necessary, be further purified by standard methods known to those skilled in the art to 10 facilitate analysis of the nucleic acids present in the T-cell population. Such T-cell nucleic acid sample contain rearranged genome DNA encoding ~ and ~ TCAR
chains used by the numerous T-cell clones present in the individual. It is the detection of a particular 15 nucleic acid sequence correlated with the disease encoding eith-er a specific variable region of a T-cell receptor ~ and/or ~-chain of the TCAR within the T-cell population which allows for the diagnosis of the onset or course of the disease.
In those situations where a V, D or J segment is used not only by the T-cell correlated with the disease but by other T-cells, it is possible to detect T-cells which mediate the disease by determining a correlation 25 between the different variable region segments used by the T-cells mediating the disease. For example, in the case of EAE in murine species, it has been determined that specific V~, J~ and v~, J~ segments of the TCAR
for different T-cell clones are correlated with EAE.
30 Figure l is a schematic representation of four types of T-cell receptors and BlO.PL mice afflicted with EAE.
As can be seen, in the ~-chain, V~2.3 and V~4.2 are each associated with a J~39 gene segment. Thus, Northern analysis of mRNA extracted from T-cells with 35 probes corresponding to the VJ region encoded by V~2.3 W O 91/09623 PC~r/US90/07699 ~ - 36 - 2~723~6 /J~3s and V~4.2 /J~39 can be used to detect the presence of two of the four T-cell clones responsihle for EAE in mice. A similar analysis can be performed for the various segments defining the ~-chain. In 5 addition, the EAE specific T-cell clones can be detected with even greater accuracy by detecting expression of the specific V~ and V~ variable regions of the TCARs of these clones. By analogy, other diseases may be diagnosed in a similar manner once the 10 particular variable region segments are identified for each T-cell population responsible for the particular disease.

In addition, the rearranged genomic DNA contained within one or more T-cell clones which mediate a particular d~sease may contain an RFLP generated upon rearrangement or which is a remnant of an RFLP from the undifferentiated DNA. If this is the case, detection in a T-cell nucleic acid sample of such an RFLP as a 20 target sequence correlated with the T-cell variable region may be used to diagnose the onset or course of the T-cell mediated disease.

The diagnosis of the onset or course of a T-cell 25 mediated disease may also be performed by analyzing a T-cell antigen receptor test sample. As used herein, a "T-cell antigen receptor test sample" is any sample from an animal which contains T-cells. As with those test samples for analyzing T-cell nucleic acids, the 30 sample containing T-cells may be treated according to standard procedures known to those skilled in the art to enrich for T-cells. In addition, the test sample or T-cell enriched sample, may be treated with appropriate detergents and purified by methods well known to those WO91/09623 P~/VS90/076~
~ 37 ~ 207~35~

in the art to obtain purified or partially purified T~cell antigen receptor.

The T-cell antigen receptor test sample is analyzed to 5 determine if a first target polypeptide sequence comprising the variable region of a TCAR chain correlated with a disease is present by a specific variable region within ~ and/or ~-chains. As used herein, a first target polypeptide sequence is 10 "correlated" with disease if a specific variable region or segment within the variable region of a T-cell antigen receptor polypeptide chain is found to be common amongst a population or subpopulation afflicted with the same malady. In addition, an analogous second 15 target polypeptide sequence also correlated with the particular disease and comprising a specific variable region or segment in a different TCAR chain may be used to detect the onset and course of the disease.

20 Thus, for example, as shown in Figure l, the BlO.PL
T-cells associated with EAE in murine species express a limited repertoire of TCAR a and ~-chains. As such, the onset and course of EAE in such an animal is monitored by assaying, for example, for V~2.3, V~4.2, 25 V~8.2 and/or V~13. Such detection is typically with monoclonal antibodies for these particular segments of the variable region and may also include antibodies directed to particular J~, J~ and D~ gene segments associated with the variable region of T-cell receptors 30 that are responsible for this disease.

It is to be understood that the onset and course of some diseases may be detected by the methods described herein for analyzing T-cell nucleic acids or 35 polypeptide samples. For example, it is contemplated - 38 - 20723~6 that some diseases are acquired based on e~posure to environmental factors such as various pathogens including viruses, bacteria and parasites. In such cases, there may be no genetic predisposition to such 5 diseases. Such a situation, however, does not preclude a predictable T-cell response to a particular antigen to produce one or more clonal populations of T-cells which in addition to mediating an immunological response to the foreign antigen also mediate an 10 response.

Reagents for Diagnosis and Treatment of Disease In other embodiments of the invention, reagents are provided for diagnosing or treating a disease characterized.by a clonal population of T-cells containing one or more specific variable regions or segments in the TCAR. Such reagents include nucleic acid probes for diagnosis and antibodies for diagnosis 20 and treatment. Such antibodies may be polyclonal antibodies but are preferably monoclonal antibodies which are reactive with the specific variable regions or segments correlated with the disease. When used in a diagnostic assay as described above, the antibody is 25 typically labelled so that its binding with a particular TCAR may be detected. Such antibodies can also be used to correlate a particular TCAR with a disease. Polyclonal or monoclonal antibodies can be produced by procedures known in the art, but the 30 characteristics of the antibody will depend upon the immunogen and screening procedures used to produce it.

When used to treat disease, antibodies specific to particular T-cell antigen receptor segments can be used 35 in unmodified form or conjugated to radionuclides or toxins by means well known in the art, and used to WO91/09623 PCT/US~/076~
~0723~6 deliver the conjugated substances to targeted T-cells.
For example, F23.1 antibodies can be conjugated to a toxin, such as Ricin, by the method of Bumol (1983) Proc. Natl. Acad. Sci. 80, 529. Briefly, monoclonal 5 antibodies reactive with the desired T-cell receptor segment, such as V~8, are prepared by conventional and well-known means. The antibodies are purified and combined with excess (6 mol/mol) N-succinimydyl 3-(2-pyridyldithio) propionate (Pharmacia, Uppsala, t0 Sweden) in PBS. After 30 minutes incubation at room temperature, the solution is dialyzed against PBS. The modified antibodies are conjugated with an appropriate toxin, such as diphtheria toxin A chain. Other toxins such as ricin A can also be employed. The diphtheria 15 toxin A chain is isolated as detailed in Bumol, supra.
The modified~antibodies are mixed with excess (3 mol/mol) reduced diphtheria toxin A chain (10% of the total volume), allowed to react for 36 hours at 4OC, and concentrated by chromatography on Sephadex 20 G-200. The product is applied to a Sephadex G200 column (1.0 x 100 cm), allowed to equilibrate and eluted with PBS.

The toxin-conjugated antibodies are administered to the 25 ani~al, preferably by inter-peritoneal injections of approximately 40 ~g of purified antibody conjugate or as determined to be appropriate on the basis of patient weight, severity of disease and other such factors.
Injections are repeated at intervals, preferably 30 approximately every three days.

1. NUCLEIC ACIDS
Depending upon the technique to be used, either DNA or RNA sequences encoding TCARs can be used to diagnose 35 predisposition or onset of a disease. Since the WO~1/09623 PCT/U~90/076 2~7~356 discovery of TCARs, many of the genes of this multigene family have been identified, cloned and sequenced. It is currently estimated that there are 50-100 V~s, 2 D~s, 13 J~s, 2 C~s, over 100 V~s, over 50 J~s, and 1 C~
5 gene segments that are combined to make TCAR
heterodimers. In addition, gene segments encoding the variable an constant regions of ~ and ~ chains of TCARs used by a small class of T cells are known. Any of these sequences, whether cDNA, genomic DNA or RNA, can 10 be used to determine disease correlations and diagnose predisposition, onset or course of a disease. In addition, synthetic DNAs or RNAs can be produced by procedures well known in the art to produce nucleic acid sequences that correspond to portions of the 15 complete gene or RNA sequences. These synthetic sequences may be either short oligonucleotide sequences, complete gene sequences or sequences corresponding to all or part of any combination of the V, D, J, or C gene seqments.

2. ANTIBODIES
Either polyclonal or monoclonal antibodies can be produced by procedures known in the art. However, the characteristics of the antibody will depend on the 25 immunogen and screening procedure used to produce it.

a) Immunoqens Animals can be immuni~ed to produce antibodies with a variety of TCAR containing samples. These include (1) 30 peptide sequences that correspond to portions of a TCAR
(U.S. pending patent applications, Serial Number 726,502 filed April 24, 1985 and Serial Number 176,706 filed April 1, 1988), (2) whole T-cells that express a unique TCAR heterodimer on their surface e.~ or 35 ~, ~ (3) TCAR polypeptide that has been purified from WO91/og623 PCT/VS90/07699 T-cells by well known procedures such as immunoprecipitation, (4) recombinantly expressed and purified TCAR protein produced from expression systems, including but not limited to yeast, eukaryotic, or 5 prokaryotic cells, and (5) either whole cells or TCAR
produced by them, where the cells have been transfected with specific TC~R genes of interest and are expressi~g the transfected genes in a soluble or membrane bound form.

b) Scrçeninq Procedures The characteristics of the final antibodies are dependent upon the screening procedures used to obtain them. If the screening procedure is designed to 15 identify only those antibodies that recognize denatured protein, the~ the resulting antibodies that are selected by the screening procedure may not generally react with the intact TCAR polypeptide present on live cells. If the antibodies are to be used for diagnostic 20 procedures, this may or may not make any difference, as a diagnostic procedure that relies on detecting denatured TCAR protein can be chosen. If the antibodies are to be effective therapeutically, however, it is important that they recognize the intact 25 TCAR present on the patient's immune system T-cells.

Screening procedures that can be used to screen hybridoma cells expressing different anti-TCAR
antibodies include among others ~l) enzyme linked 30 immunosorbent essays (ELISA), (2) flow cytometry (FLOW) analysis, (3) immunoprecipitation, and (4) the ability to comodulate the CD3 antigen (part of the TCAR-CD3 complex present on the surface of T-cells) off of the surface of cells. The comodulation and FLOW screening 35 procedures are preferred for the selection of W091/09623 PCT/I)S9~/07~
42 - ~07~35~

antibodies that are able to recognize intact TCAR on live cells due to the inherent properties of these techniques. The immunoprecipitation and ELISA
procedures are preferred for the identification of 5 antibodies to inactive or denatu~ed TCAR.

l) ELISA Screening Assay Many different formats of an ELISA that can be used to screen for anti-TCAR antibodies can be envisioned by 10 one skilled in the art. These include, but are not limited to, formats comprising purified, synthesized or recombinantly expressed TCAR polypeptide attached to the solid phase or bound by antibodies attached to the solid phase or formats comprising the use of whole 15 T-cells or T-cell lysate membrane preparations either attached to the solid phase or bound by antibodies attached to the solid phase. Samples of hybridoma supernatants would be reacted with either of these two formats, followed by incubation with, for instance, a 20 goat anti-mouse immunoglobulin complexed to an enzyme-substrate that can be visually identified.

2) FLOW Screening Assay Supernatants of antibody producing hybridomas can be 25 screened by FLOW in a number of different ways as is known by one skilled in the art. one screening procedure involves the binding of potential antibodies to a panel of T-cells that express well-known TCARs on their surface. Generally antibodies that react with 30 intact TCAR are detected by this analysis, but the T-cells can be fixed slightly with ethanol, in which case antibodies reacting with denatured TCAR
polypeptide can be identified. FLOW assays can also be formatted where potential antibodies are screened by WO91/09~23 PCT/US90/076~
~ 43 ~ 2072356 their ability to compete with the binding of a known antibody fsr the TCAR present on a T-cell.

3) Immunoprecipitation Screeninq Assav If the antibodies to be screened do not react with intact TCAR, they can be screened by their ability to immunoprecipitate a known TC~R as analyzed by SDS polyacrylamide gel electrophoresis or Western blot analysis. Hybridoma cell supernatants can be pooled and then screened by this technique. Using this assay, it is possible to identify the chain of the TCAR
heterodimer that the anti-TCAR antibodies are recognizing.

4) Screening by Comodulation of the CD3 Antigen The TCAR normally exists on the surface of T-cells as a complex with the chains of the CD3 complex. When an antibody binds to this TCAR:CD3 complex, the complex in 20 turn becomes internalized in the T-cell and disappears from the cell surface. Thus, if T-cells are reacted with an antibody specific to the TCAR and the complex becomes internalized, a further reaction with an anti-CD3 specific antibody will result in substantially no 25 detection of CD3 bearing cells by FLOW analysis. This comodulation screening procedure detects antibodies that are able to interact with intact TCAR on the surface of T-cells, and thus, is preferred for the selection of anti-TCAR antibodies that have the most 30 useful characteristics for TCAR therapeutic reagents for treating disorders.

5) Additional Screening Assays Many additional screening assays, such as those based upon competition with anti-TCAR antibodies o known 2072~6 specificity or the ability to cause T-cells expressing known TCARs to proliferate in culture, will be known to those skilled in the art.
;

WO91/09623 PCT/US90/076~
~ 45 ~ 2~7~356 C) Molecular Characterization of Anti-TCAR Antibody The TCAR heterodimer comprises a b chain subunit made up of V~, D~, J~, and C~ gene segments and an ~-chain subunit made up of V~, J~, and C~ gene segments.
Antibodies to the TCAR can recognize epitopes in either the b chain or the a chain regions, as well as combined epitopes made up of some ~-chain and some ~-chain sequences. The epitopes can consist of linear stretches, amino acids from either chain or alternatively can arise from conformational sequences where the amino acids in the epitope come together from different parts of the receptor sequence.

TCAR specific antibodies can be subclassified as follows. Some antibodies will be anti-C~, anti-V~, anti-J~, anti-DJ~, anti-V~, anti-J~, or anti-C~
specific to name a few of the possible subclassifications. In addition, V region antibodies, 20 for example, can be further divided into thcse anti-V
region antibodies that are specific for each of the subfamilies of V~ or V~, such as anti-V~l specific or anti-V~8 specific antibodies. These subfamily specific antibodies can be subdivided even further into 25 antibodies that will recognize only individual members of each subfamily, such as for example anti-V~8.l specific antibodies. Antibodies can also be produced that react with shared regions of the TCAR, such as anti-V~2J~2.3 specific antibodies. A similar approach 30 can be used to characterize antibodies to TCARs comprising ~ and 8 subunits.

D) Functional Characterization of Anti-TCAR Antibody 35 TCAR specific antibodies can also be classified as anti-idiotypic, anti-clonotypic, anti-major framework, WO91/0~6~3 PC~/US90/076~

or anti-minor framework antibodies. This classification is based more on the functional characteristics of the antibody than on the molecular site that the antibody reacts with. Historically anti-5 idiotypic antibodies are characterized by theantibody's ability to react with the "idiotypic"
region, i.e., the region of the receptor that makes it unique from other receptors. This is analogous to the idiotypic regions of immunoglobulins, which are regions of the immunoglobulin heterodimer that interact with antigen. Anti-clonotypic antibodies are so defined, because they reacted only with the TCAR present on an individual T-cell clone. Anti-major a framework antibodies, for example, are distinguished by their 15 ability to recognize all a bearing TCARs, such as would also be true ~f anti-Ca region antibodies. Anti-minor framework antibodies react with a subset of T-cells, such that they can be used to subdivide the total population of T-cells detected by anti-major framework 20 antibodies into subsets. These subsets can then be even further subdivided by anti-idiotypic antibodies.

Antibodies that are useful as diagnostic or therapeutic reagents can consist of all of these types of TCAR
25 specific antibodies, but some types are expected to be more effective than others depending upon the circumstances under which they are used. For example, anti-clonotypic antibodies are used therapeutically in disease states which are characterized by the over 30 proliferation of a single T-cell clone. ~owever, anti-clonotypic reagents are not the preferred reagents for treating diseases that arise from not just one T-cell clone, but from a handful of different clones. In addition, anti-clonotypic reagents are generally useful 35 for treatment protocols that are effective only in a WO91/09623 PC~/US90/07699 ~ 47 ~ X072356 single patient. Thus, anti-V region specific or anti-minor framework specific reagents that react with a handful of disease specific clones that exist in multiple patients with the same disease state (but 5 perhaps heterogeneous MHC antigens) are preferred therapeutic reagents. Such an antibody is used to treat the whole subgroup of patients, not just individual patients.

10 Since there are a limited number of gene segments for the variable regions of the various subunit chains used in TCARs, a finite repertoire of antibodies specific for the polypeptides encoded by each of these segments may be generated and used for diagnosis and/or 15 treatment. The choice of antibody depends primarily on the gene segments used by the TCAR correlated with a particular disease. The use of such antibodies however, may also result in cross reactivity with normal TCARs utilizing the peptide encoded by such gene 20 se~ments. In diagnostic assays, such cross-reactivity should not impede the accuracy of the assay since disease correlated T-cell clones are present in relatively high concentration. In the case of antibody therapy, such cross-reactivity may deplete some of the 25 normal T-cells. However, such depletion is not expected to adversely compromise the patient's immune system since only a small sub-population of T-cells would be affected.

30 Any TCAR specific antibody can also be characterized by its ability to recognize either denatured TCAR
polypeptide or nondenatured polypeptide as would exist on live cells. Depending upon the diagnostic procedure to be used, either type of antibody is useful, but for 35 antibodies to be therapeutically effective, they WO9]/09623 PCT/US90/~76~
- 48 - 20723~6 preferably react with the intact, nondenatured receptor that exists on the cells circulating throughout the patients immune system.

e) Isotypic Characterization In addition to the properties described supra, the isotype of the anti-TCAR specific antibody is also important. For different clinical applications, an antibody of a specific ~ 50type may be preferable to one 10 of a different isotype. For example, the IgG2a isotype reacts with Fc receptors on cells of the reticuloendothelial system and is more readily removed from the circulation and sequestered in the spleen than other isotypes. Such an antibody that has reacted with a target cell may result in the more efficient removal of the target~cell from the site of active disease. In addition, some isotypes (such as IgG2a) are more effective in antibody dependent cell cytotoxicity reactions than others. In general, antibodies of the 20 IgG isotype are preferable to those of the IgM isotype because they have higher binding affinities.

The desired isotype of an antibody may be ~elected by scr~ening potential antibodies by an ELISA assay 25 designed to select the isotype of interest. For example, the solid phase can be coated with goat anti-mouse IgG Fc specific antibodies, if it is desired to select for antibodies having the IgG isotype. Given an antibody of one isotype, it is also possible to switch 30 the isotype to a different isotype. Many methods for accomplishing this switch are known to those skilled in the art. For example, the isotype switch can be done by repeatedly selecting for the isotype of interest using magnetic beads (super paramagnetic iron oxide 35 particles, Biomag~ beads purchased from Advanced WO91/09623 PCr/US90/076~
~ 49 ~ 2~72~5~

Magnetics, Inc.) coated with a goat anti-mouse antibody preparation including all isotype classes. In switching the isotype from IgGl to IgG2a, for instance, the IgG2a binding sites on the coated magnetic beads sare first blocked with an irrelevant antibody of the IgG2a isotype. All cells producing antibodies of differing isotypes will then be bound by the beads and removed magnetically, resulting in an enrichment of cells producing the IgG2a isotype. These cells can then 10 be cloned by limiting dilution, and using commercially available anti-isotypic reagents in an ELISA assay, the IgG2a producing clones can be identified.

f) Humanizing Antibodies 15Antibodies can be produced in many different animal hosts, including but not limited to, rabbits, mice, and rats. When these antibodies are used therapeutically in humans, they are recognized to varying degrees as foreign and an immune response is generated in the 20 patient. One approach to overcome this problem is to immunosuppress the patient in addition to the antibody treatment. Another approach is to produce chimeric antibody molecules that combine a mouse variable region domain and a human constant region domain. SUCh 25 chimeras produce a less marked immune response than non-chimeric antibodies. A further approach is to use the techniques of recombinant DNA technology to "humanize" the antibodies. Such humanized antibodies retain the hypervariable regions required for the 30 antigen recognition sites, but all other regions including the framework regions of the variable domain are of human origin. These humanized antibodies are preferable for immunotherapy in that they minimize the effects of an immune response. This in turn leads to a 35 lowering of any concomitant immunosuppression and to W091/09623 PCT/US90/076~
20723~6 increased long term effectiveness in, for instance, chronic disease situations or situations requiring repeated antibody treatments.

g) Antibody Fragments and Conjuaates In addition to the use of the whole anti-TC~R antibody in diagnosis or immunotherapy of diseases, fragments of the antibodies may also be used~ Fra~ments such as F(ab' )2' Fab' or Fab fragments can be generated by 10 enzymes such as pepsin or papain and reducing agents, as is routinely known in the art. These fragments retain the antigen binding domains present in the parent molecules. These types of molecules can also be produced by recombinant DNA technology where the variahle region domains are ligated and expressed as single chain polypeptides. Whole antibody molecules or their fragments can also be produced as immunoconjugates. In addition to the antibody portion, these conjugates contain another active group that may 20 be a toxin, cytotoxic compound or radionuclide to name but a few (Copending U.S. Patent Application, Serial Number 07/229,288, filed Augu~t 5, 1988). The antibody can thus be used to target the active toxin or cytotoxic compound to the specific disease site 25 involved.

TREATMENT OF DISEASE
One level of treatment of diseases that involve the immune system is to modulate the entire system with 30 immunosuppressive drugs, such as cyclosporin A. This type of treatment is not specific to the actual defect in the system, however, and problems of cytotoxicity and long term effectiveness arise. Since T-cells have been implicated in many different T-cell mediated 35 diseases, another level of treatment involves WO 91/Og623 PCr/llS90/07699 modulating the entire T-cell population. Although this is more specific than broad i~munosuppression therapies, it is still a non-specific therapy that results in the loss of the beneficial functions of some 5 T-cell subsets along with the loss of the deleterious effects of the disease subset of T-cells. This type of therapy is illustrated by the use of the antibody, OKT3, which interacts with the CD3/T3 complex present on all T-cells, in the trPatment of allograft 10 rejection. The preferred immunotherapy herein is one that affects substantially only disease related T-cells and not any other T-cells.

disease specific T-cells are identified by the 15 expression of specific TCARs. Thus, various levels of immunotherapies can be developed by subdividing the whole T-cell population into thsse subsets of T-cells that are disease related by virtue of the expression of disease related TCARs. Immunotherapies based upon 20 anti-constant region TCARs are no more specific than the anti-CD3 immunotherapies, as both therapies affect the whole population of T-cells. However, immunotherapies based upon anti-variable TCAR subfamily reagents (anti-V~6, for instance) target only that 25 subset of T-cells expressing that TcAR variable region subfamily and as a consequence are more specific.
Greater specificity is obtained by further subdividing the whole T-cell population into those expressing a particular V region gene, such as V~18.l, as only those 30 T-cells expressing this particular V~ gene segment are affected. Additional specificity is obtained by using other V, D, J, DJ, VJ, etc. specific immunotherapies.
Thus, the preferred immunotherapies are those that substantially affect only the subset of T-cells that 35 are disease related and which do not impair non-disease WO91/09623 PC~/US~0/076~

related T-cells and their functions. Although less preferable, immunotherapies are acceptable which do adversely affect some non~disease related T-cells provided the patient's immune sys em is not 5 compromised. The discussion presented herein for immunotherapy is also applicable for the preferred diagnostic procedures using antibodies and procedures designed to monitor disease therapy and progression using antibodies.

To develop the most specific disease diagnostic, therapeutic and monitoring methods, it is necessary to subdivide the population of T~cells into those that are disease specific. This is done by analyzing the TCAR
expression and detecting which ones are disease related. A ~referred way to accomplish this is by developing a TCAR expression matrix. The first element in the matrix involves the major histocompatibility antigens (MHC). Many T-cell mediated diseases, such as 20 diseases, are correlated with the expression of certain MHC molecules. Thus, disease can be subdivided into subsets expressing different MHC molecules. The next level in the matrix is to determine which ~ chain variable region TCARs are being expressed in the 25 different MHC subgroups. This will result in the further subdivision of each MHC subgroup into further divisions. The next level in the matrix is to determine which ~-chain variable region TCARs are being expressed. The combination of V~, V~ and MHC results in 30 further subdivisions of the (MHC + V~) groups. Even further subdivisions are obtained by analyzing the expression of specific D and/or J TCAR gene segments.
Disease correlations can be made at any of the levels of the matrix, but the more levels that are analyzed, 35 the more specific the correlation will be. Diagnostic, 207~3~6 therapeutic and monitoring strategies can also be developed at any level in the matrix, but again, using more levels results in more specific methods.

5 There is a disadvantage in oreating too many different levels within the matrix, in that the ultimate degree of subdivisions leads to identifying correlations that are only effective at a single individual level, resulting in therapeutics that would benefit only one 10 individual. Thus, intermediate levels in the matrix are identified that result in a fully adequate correlation with disease and that will still be effective in diagnosing, treating or monitoring various subgroups of patients having that disease.
The following is presented by way of example and is not to be construed as a limitation of the scope of the invention.

Human T-cell Rece~tor V~_Gene PolymQrphism This example describes the identification of various restriction fragment length polymorphisms (RFLPs) associated with V~ segments of the human genome. Some 25 of the work described in this example is also disclosed in copending U.S. Patent Application Serial No. 176,706, filed April 1, 1988 and in Concannon et al. (1~87) J. E~. Med. 165, 1130-1140, each of which are incorporated by reference.
DNA Samples High-molecular-weight DNA was isolated from a panel of 30 lymphoblastic cell lines (LG series) derived from the offspring of consanguineous marriages and ~5 homozygous for class I and class II MHC antigens W091/09623 PCr/US~/076~
_ 54 - 207235 6 (Gatti, et al. (1979) Tissue Antiqens 13, 35). DNA
samples from lymphoblastoid cell lines representing 69 additional unrelated individuals were supplied through thQ courtesy of CEPH (Centre pour L'Etude du 5 Polymorphisme Humain), a Paris-based gene-mapping consortium. DNAs derived from 60me additional cell lines and from some tissue samples were also examined.
Rearrangement of the C~ genes was not observed in any of these DNA samples.
Probes Nucleic acid probes corresponding to members of human V~ gene segment subfamilies V~l through V~14 were isolated from subclones described by Concannon et al.
15 (1986) Proc. Natl. Acad. Sci. USA 83, 6598 and correspond t~ that set forth in Figure 2 (nucleotide sequence) and Figure 3 (amino acid sequence). For subfamilies 5 and 7, probes corresponding to two different subfamily members were used. Gel-purified 20 inserts containing V~-specific sequences were labeled by random priming with ~-[32P]triphosphates (Feinberg, et al. (1984) Anal. Biochem. 137, 266) to specific activities ranging from 5 x 10~ to 1 x 109 cpm/~g and were used without further purification.
Sou~hern B~
Restriction digests of genomic DNAs were carried out under conditions specified by the enzyme manufacturers.
In the case of DNAs from the cell lines, the 30 completeness of digestion was monitored by testing the ability of a l-~g aliquot of the reaction mixture to digest an additional l~g of ~ DNA simultaneously. 5-10 ~g of DNA was loaded per lane on gels to be transferred. Blots were prepared either by the method 35 of Gatti et al. (1984) Biotechniques l, 148 or Reed et WO9l/0~623 PCT/~S~/076~
- 55 - 2~723~ ~

al. (1985) Nuclelc Acids Res. 13, 7207 on Zeta-Probe nylon membranes (Bio-Rad Laboratories, Richmond, CA).
The sa~e blots were used for all v~ gene hybridizations so that the results of each round of hybridization 5 could be directly compared. some blots were useful for hybridizations with more than 20 probes.

Hybridizations were carried out for 12-15 hours at 37C
in 50~ formamide, 5 x SSC (l X SSC is 0.15 M NaCl/0.015 10 M sodium citrate), 0.02 M sodium phosphate, pH 6.7, 100 ~g/ml sheared denatured salmon sperm DNA, 0.5% nonfat powdered dry milk, 10% dextran sulfate, 1% SDS, and 1-2 ng/ml of probe. Filters were washed in 2 X SSC and 0.1% SDS at 60OC and exposed to Kodak XAR-5 X-ray film 15 for 12 to 120 h. Probe was removed for subsequent rounds of hybridization by washing twice in boiling .01 X SSC for 15 minutes on a room temperature rocker platform.

20 Relative Sizes_of Human V~ Gene Segment Subfamilies Probes representing each of the human T-cell antigen receptor V~ gene subfamilies V~1 through V~14 were hybridized, to germline DNA from -100 unrelated individuals. For 30 of these individuals (the LG
25 series cell lines), this hybridization analysis was carried out with four different restriction enzymes.
This allowed for estimation of the size of the various V~ gene segment subfamilies with greater accuracy because it avoided problems of interpretation caused by 30 polymorphism in outbred human populations. As shown in Figure 4, the 14 V~ subfamilies contain at least 48 V~
gene segments. As in the mouse (Patten, et al. (1984) Nature (Lond.) 312, 40; Barth et al. (1985) Nature (Lond.) 316, 517; and Behlke, et al. (1985) Science 35 (Wash. DC), 229-566) the subfamilies are small in size, - 56 - 2~723~6 the largest containing seven members; but unlike the mouse, fewer are single membered. See in general Wilson R. et al (1988) Immunolo~i_ 1 Reviews 101, 149-172.

By statistical analysis of V~ gene representation in cDNA libraries, the expressed human V~ gene repertoire has been estimated to be <59 genes with 95% confidence (Concannon et al. ~1986) Proc. Natl. Acad. Sci. USA 83, 10 6598). The 48 genes identified herein by hybridization approach this number and yet do not include three additional members of the V~6 subfamily described by IXuta et al. (1985) Proc. Natl. Acad. Sci. USA 82, 7701, and those members contained in subfamilies V~15, 15 V~16, V~17 and V~18 (a total of at least eight members) (Tillinghast,^et al. (1986) Science (Wash. DC) 223, 879 and Kimura et al. (1986) J. Exp. Med. 164, 739).
Therefore, it is likely the total number of V~ genes in the human genome exceeds that estimated by statistical 20 analysis of expressed genes. Some of the V~ gene segments detected by hybridization will correspond to pseudogenes that cannot be productively rearranged, or are not transcribed in populations of mature T-cells (Siu, et al. (1986) Proc. Natl. Acad. Sci. USA
25 164,1600). Therefore, these gene segments would not be represented in cDNA libraries from peripheral lymphocytes.

Polymorphism Associated with V~ Gene Seaments 30 To further assess the amount of polymorphism associated with V~ gene segments, DNA from each of the 30 cell lines digested with four different restriction enzymes was hybridized to probes for each of the V~l through V~14 gene segment subfamilies. Table I describes the 35 probe/enzyme combinations that detected polymorphic WO91/09623 PCT/US~/07699 ~ 57 ~ 2~72356 differences between individuals within this panel~
These polymorphisms were specific for the restriction enzymes used in the analysis and recurring patterns were observed with DNAs from different individuals.
5 Such polymorphism was found to be associated with 12 of the 14 V~ gene segment subfamilies tested.

WO91/09623 PCT/US90/~76~
- 58 - 2072~56 TABLE_I
Incidence of Polymorphism in Human V~ Gene Segment Subfamilies ~
Subfamily Eco RI Bam HI Hind III BglII

; V~l - - +
V~2 - - + +
10 V~3 + + + +
V~4 + - _ _ V~5 - - _ +
: V~6 ~ ~ ~ +
V~7 - _ + +
V~8 - + + +
v,~g V~10 15 V~ll - + - +
V~12 - + +
V~13 - - ~ +
V~14 - - +

(+) The probe/enzyme combinations that detected polymorphism when DNA from 30 unrelated individuals was digested with four different restriction enzymes and blotted with probes representing 14 V~ gene segment subfamilies. Polymorphism in the V~3 and V~6 subfamilies are specific to single cell lines, LG-22 25 and LG-41, as described in the text. (-) Probe/enzyme combinations that failed to detect polymorphism.
Polymorphic restriction sites for several different enzymes were detected with some probes, (e.q., V~2 with Hind III and Bgl II; V~8 with Bam HI, Hind III, and Bgl 30 II; and V~ll with Bam HI and Bgl II; and V~12 with Bam HI and Hind III). Three polymorphisms involved alleles that were relatively common in the panel. For most V~
probes, one predominant hybridization pattern was observed with only a few individuals displaying a 35 second pattern. Because of the infrequency of these WO91~09623 PCT~US90/07699 ~ 59 -variant forms in the small panel, it proved difficult in many cases to observe the minimal three hybridization patterns required to define a biallelic polymorphism (l.e., AA, AS, and BB) or to find 5 informative families to demonstrate segregation of these less common polymorphisms. For these polymorphisms, the frequencies of the various hybridizati~n patterns observed for each of the polymorphic V~ gene segment subfamilies are set forth 10 in Table II.

WO 91~09623 PCI /US90/07699 -60- ~072356 ~,c ~
:~ ~ ~a c ~ ~ a ~ ~ ~ a ~ X
~ ~ ;~
~ cq v D C~ cn ~ ca a~

a U ~ :r: ~ æ z c @~ ~e c e ~ ~ ~ ~
v~ r` C~ a ~s ~ ~V ~ Z ~ ~ Z
~ ~)-3 ~D o :r: C , L~ ~ a~
~ bD ~ C~

o ~ ~ o o c ~ ~' ~
h~ oC

Z
C ~ ~ S~; cq az Z az n U~ O ~ ~ U~O V~ OD CD O ~

WO 91/09623 ~61- PCI/US90/07699 u ~ 207235 f~
~ 3~ ~ ~ z ~ ~ e c c ~
~, > ~ _ CQ ~ _ ~ O
C ~ 7c~ ~ rSJ _ ~ " e , ~ ~ ~ ~ ~4 ~ r4 ~ ~q r~ r D C --brJ _ rf~ N u trJ I LC~ O O ^
tL~ _ ~ V r-l ~
a~ ._ . J- ~ C

~ e ~; ~ ~ v ~rs~ _ r~ ~4 D C `~
z c,. c a b~ ~ ~ ~ C

~ a~ ~ C Ui ~ ~ .

~ OD ~ ca v C
~ z; c ~ ~e c ~ 7a V '~ J' 1:~ ~ ~ r~ oo 0~ ~ U O ~ ~ O
~ ~ ~ V ,C: ~ O
~ v ~ I~ ~ ~ ~ r~ ~ ~ _ ,~ v C
.~ C ~1 ~ u~ ~ L~
~ o _ L ~ ~ 1 ~ ~ 3 W~91~09623 PCT/U~90/07699 20723~

Two of the probe/enzvme combinations (V8/Bam HI and V~12/Bam HI) appeared to detect the same polymorphic Bam HI site. Subseyuent restriction mapping of a cosmid clone which hybridized to both the V~8 and V~12 5 probes indicated that these probes were hybridizing to v~ gene segments which flanked a single polymorphic Bam HI restriction site.

Se~reaation of VB Gene PolymorDhisms 10 Two of the polymorphisms which were identified and displayed the minimal three hybridization patterns consistent with the segregation of two alleles at each locus (V~/Bam HI and V~11/Bam HI) were tested for segregation. The V~8 and V~11 probes were hybridized 15 sequentially to Bam HI-digested genomic DNA from members of several three-generation families, one of which is shown in Figure 5. To increase the informativeness of markers in this genomic area, the segregation of a previously reported polymorphism 20 detected with a C~ probe (Robinson et al. (1985j Proc.
Natl. Acad. Sci. USA 82, 3804) in Bgl II-digested DNA
from this family was also tested. ~he V~8 probe hybridized to two or three bands when washed under high stringency (1 x SSC/0.1% SDS at 65C), an invariant 25 band of 3.3 kb containing the V~8.2 gene segment, and a polymorphic band of either 23 or 2.0 kb containing the V~8.1 gene segment. In the analysis of 100 unrelated individuals, the 23-kb allele occurred at a frequency of 46.4S and the 2.0-kb allele at a frequency of 53.5%.
30 The V~11 probe hybridized to an invariant band cf 12 kb and polymorphic bands of either 25 and/or 20 kb in loo unrelated individuals. The 25-kb allele occurred at a frequency of 47.4% and the 20-kb allele at a frequency of 52.6%. The C~ probe hybridized to bands of either 35 lO or g.0 kb in the same panel of individuals. The 10-~ W~9lfO9~23 PCT~US90/07699 - 63 - 2~72356 kb allele occurred at a frequency of 55.9% and the 9-kb allele at a fre~uency of 44.1%. A diagram of this family and the results obtained by hybridization with V~8, V~ll, and C~ probes are shown in Figure 5.

Linkage Disequilibrium Between Alleles at the y~8 and VB11 Gene Loci The haplotypes created by the three probes described above, V~8, V~ll, and C~ are potentially informative markers for following genetic segregation because the alleles at each locus are approximately equal in frequency, and hence one might expect to see each of the theoretically possible nine haplotypes in outbred human populations with high frequencies.
15 However, analysis of 70 unrelated individuals homozygous f~r at least one of the three loci indicated that certain alleles at the V~8 and V~ll loci were preferentially associated and hence were in linkage disequilibrium. Fifty-two of these subjects were 20 doubly homozygous, allowing the assignment of haplotypes. Figure 6 indicates both the distribution of haplotypes expected in this sampling, based on allelic frequencies at these three loci, and the actual distribution observed. The observed distribution of 25 haplotypes is highly biased toward those haplotypes in which the V~8 2.0-kb allele is associated with the V~ll 25-kb allele and those in which the V~8 23-kb allele is associated with the V~11 20-kb allele (p~O.OO1).

30 Alleles at the V~8 and V~ll loci were in strong linkage disequilibrium as evidenced by the fact that of the 30 individuals who were homozygous for the 2.0-kb allele at the V~8 locus, 26 (87%) were also homozygous for the 25-kb allele at the V~ll locus, compared with an 35 expected 7 (Z3%) individuals (p<O.OO1). Conversely, of 24 individuals who were homozygous for the 23-kb allele WO9l/09623 P~/US9~/076 20 t2356 at the V~8 locus, 22 (~2%) were homozygQus for the 20-kb allele at the V~11 locus, compared with an expected number of 7 (30%) (pcO.OOl). This nonrandom distribution also mildly affect~d alleles at the c~
5 locus. For example, only one individual homozygous for the 9-kb allele at C~ was seen among those people who were homozygous for the 23-kb allele at the V~ locus, instead of the 5 expected, and of the 12 individuals in the panel who were homozygous for the 9-kb allele at 10 the C~ locus, 6 were homozygous for the 25-kb allele at the V~ll locus as opposed to the three expected.
However, these latter figures were not statistically significant (O.l>p>0.05).

15 There are several possible explanations for this disequilibrium. First, the loci could be very close from a genetic standpoint, either because of close physical linkage, or recent generation of alleles. The finding herein of allelic frequencies approaching 50%
20 in a variety of populations in a panel of unrelated individuals containing representatives of a number of different populations, and the finding of the same alleles among these representatives, provides support for the argument that this effect is probably not due 25 to recent mutation. Second, there could be a suppression of recombination in the region containing these loci that would cause them to appear linked.
Studies of the recombination rates between polymorphic V~ gene segment alleles and alleles at flanking loci 30 could test this possibility. Third, there could be some selective pressure favoring the existence of certain haplotypes over others. Preliminary results suggest that at least the first explanation plays a role i.n the observed disequilibrium because the V~8, 35 V~ll, and C~ loci all appear to map to a restriction WO 9l/09623 PCr/~S90/07~9 ~ 65 --20723~6 fragment of <600 kb in size. Despite the fact that many of these linJced polymorphic alleles were of low frequency in the panel, among the 30 individuals which were examined, 27 different haplotypes were identified.

EX~PLE 2 Restricted Use of T~Cell Receptor V Genes in Murine Autoimmune Ence~halomYelitis and Antibody TheraPY
This example describes the TCARs found on clonal populations of T-cells correlated with EAE in mouse.
It also describes monoclonal antibody therapy directed against a sub-set of these T-cells which mediate EAE.
This work is also described in U.S.Patent Application 15 229,288 filed August 8, 1988 and by Urban et al. (1988) Cell 54, 577-592, each of which are expressl.y incorporated herein by reference.

T-cell Lives and Clones 20 T-cells responding to MBP were isolated from the draining popliteal, inguinal, and paraaortic lymph nodes of eight-week-old BlO.PL mice ten days after subcutaneous injection of 200,~g MBP emulsified in complete Freund's adjuvant according to the method of 25 Kimoto et al. (1980) J. EXP. Med. 152, 759-770. MBP
was prepared from rat brains using the method of Smith (1969) J. Neurochem. 16, 83-92. Lymph node cells were initially cultured at 3 x 106 cells/ml in serum-free HL-1 Ventrex medium containing 20 ~g/ml rat MBP. Five 30 days later responding T-cell blasts were cultured at 5 x 105 cells/ml in complete culture medium containing 20 ,ug/ml rat MBP, 2 x 106 cells/ml syngeneic irradiated t3000 rad) spleen cells as a source of antigen presenting cells (APC) and 10% supernatant from 3~ concanavalin A-stimulated rat spleen cells according to the method of Gillis et al. (1978) J. Immunol. 120, W091/09623 P~T/US90/076~
- 66 - 20723~

2027-2032. T-cells wQre routinely restimulated with MBP, APC and CAS every 12-14 days. Clones were derived by limiting dilution in 96-well flat bottom microtiter rates at 0.3 cells per well. Complete culture medium 5 consisted of Dulbecco's Modified Eagle's Media supplemented with 4.5 g/l glucose, 2mM glutamine, lOO~g/ml streptomycin, lOO~g/ml penicillin, 5 105 M 2-mercaptoethanol and 10% fetal bovine serum (cDMEM).

10 The method of Zamvil et al. (1985) Nature 317, 355-358, was used to test T-cells for their ability to adoptively induce EAE. Recipient mice received low-dose whole-body irradiation (~50 rad) just prior to the i.v. injection of T-cells. Mice were assessed for 15 clinical severity of EAE according to Zamvil et al.
(1985) supra.~

Most TH Cells in BlO.PL Mice immunized with MBP react against the N-Terminal peptide 1-9NAc. Three T-cell 20 lines specific for MBP (BML-l, 1, and 3) were established from three separate groups of BlO.PL mice (3-4 individual mice per group) immunized with rat MBP
emulsified in complete Freund's adjuvant (200~g/mouse).
In agreement with previous results obtained for PL/J
25 H-2U mice (Zamvil et al. (1986) Nature 324, 258-260), the lines were found to proliferate specifically to an acetylated N-terminal peptide, 1-9NAc, indicating the immunodominance of this epitope in the BlO.PL response to MBP. These Tl~ cells were also uniformly restricted 30 to the IAu class II molecule as their proliferation was specifically bloc~ed with an anti-I-AUb monoclonal antibody (10-2.16, Oi et al. (1978) Clin. Top.
Microbiol. Immunol. 81, 115-129). The TH lines induced EAE in naive irradiated (350 R) mice after 15-19 days Wo~l/09623 PCT/US90/07~
- 67 - 207~356 when given intravenously at doses as low as 5 x 105 cells per mouse.

Production and Assay_of T-Cell Hvbridomas 5 Hybridomas were generated by fusing MBP-reactive BlO.PL
helper T-cell blasts with the AKR thymoma BW5147 as previously described in Kappler et al. (1981) J. Exp.
Med. 153, 1198-1214. Three separate fusions were performed involving independent lines of T-cells (BML
10 1, 2 and 3), each derived from 3-4 individual mice and passaged for 2-8 weeks prior to fusion. Hybrids were screened for their ability to secrete IL-2 in response to both native MBP and a synthetic acetylated N-terminal MBP derivative tl-9NAc) previously shown to induce EAE in Pl/J mice according to the method of Zamvil et al: (1986) Nature 324, 258-260. Briefly, 105 hybridoma cells were added to 106 BlO.PL APC and serial dilutions of antigen in a total volume of 250 ~L.
Twenty-four hours later 100 ~1 of super~atant was 20 transferred to 5 x 103 IL-2 dependent HT-w cells in 100 ~1 cDMEM. After an additional 24 hours, the cells were pulsed for 6 hours with I mCi 3H-thymidine per well.
The cpm thymidine incorporation was calculated as the mean for triplicate cultures. The standard deviation 25 from replicate cultures was always less than 10% of the mean. Hybrids testing positive in this assay were subcloned twice by limiting dilution at 0.2 cells per well.

30 A large number of MBP-specific TH hybridomas were produced by using each of the three cell lines to the AER thymoma BW5147 (Kappler et al. (1981) su~ra). A
total of 71 MBP-specific hybridomas from three independent fusions were isolated and the majority (59 35 out of 71 TH cells; 83%) were found to be specifically W091~09623 PCT/US~/076 2~723~6 responsive to the peptide 1-9 NAc; of these, 37 were randomly selected for further analysis of their T cell receptor genes, as discussed below. Each fusion involved a separate T-cell line (BMI,-1, 2 and 3) 5 passaged in the presence of native rat MBP (100 ~g/ml) for 2-8 weeks prior to fusion. In agreement with previous work by Zamvil et al. ~1986) supra, it was found the other 12 hybridomas, which failed to recognize the peptide l-9NAc, to be directed against 10 one or more epitopes present on rat, but not mouse, MBP. The T cells directed against such epitopes have previously been shown to be nonencephalitogenic in mice (Zamvil et al. (1986) su~ra).

15 Southern and Northern Blots l. Va Genes DNA for Southern blot analysis was prepared from liver and T-cell lines, clones and hybridomas with the Applied Biosystems Model 340A Nucleic Acid Extractor 20 (Foster City, CA). Ten micrograms of restriction enzyme-digested DNA was separated by electrophoresis on an 0.7% agarose gel, blotted onto nylon membrane and hybridized using conventional methods (Reed et al.
(1985) Nucle~c ~cids Res. 13, 7207-7221). Final washes 25 were for 30 minutes at 65c in 2X SSC/0.1% NaDodSO4.
RNA for Northern blot analysis was prepared by resuspension of the cells in guanidinium thiocyanate followed by centrifugation through a CsCl cushion according to the method of Chirgwin et al. (1979) 30 Biochemistry ~8, 5294-5299. Poly(A)+ RNA was selected on oligo(dT)-cellulose columns and lO~g was electrophoresed on a 1% agarose formaldehyde gel.
Conditions for blotting hybridization and washing were the same as those for Southern blots.

WO91/09623 PCT/US9~/07fi99 2~72356 To determine the relative usage in MBP TH hybridomas of each of the gene segments described above, 36 additional TH cells were analyzed by Northern and Southern blots. Southern blot analyses with a C~ probe 5 of the DNA of these TH cells yielded a unique rearranged bands of -3.5 kb upon digestion with the restriction enzyme Hind III and -l.7 kb upon digestion with Eco RI (data not shown) in all the hybridomas found to express the V~2 gene. These DNA fragments 10 hybridize with both a V~2 cDNA probe and a J~39 oligonucleotide probe (data not shown). Likewise, the rearranged V~4.2-V~39 genes were identified as a unique -14 kb bands upon digestion with Hind III and -0.7 kb bands with Eco RI when yenomic blots were probed with a 15 V~4 cDNA probe and a V~39 oligonucleotide probe (data not shown). Although the genomic arrangement of the V~
and J~ gene segments are not yet known, the fact that all the V~2 expressing hybridomas (including those whose V~2.3-J~30 genes were sequenced) exhibit 20 identical sized V-J DNA fragments upon digestion with the Hind III restriction enzyme (likewise for Eco RI
and Bam HI) suggests that all these TH cells utilize the same V~2.3 and J~39 gene segments. A similar rationale holds for the V~4 expressing hybridomas and 25 hence it was concluded that all are probably expressing the V~4.2 and J~39 gene segments. Thus, all of these MBP TH hybridomas appear to have rearranged either V~2.3 or V~4.2 gene segments to the J~39 gene segment.
Northern blot analyses demonstrate that these T-cells 30 appropriately express either V~2-J~39 or V~4-J~39 mRNA's about l.7 kb in size. The cumulative results of Northern and Southern blot and sequence analyses lead to the conclusion that only two V~ and one J~ gene segments encode the functional T-cell receptors in 35 these 43 MBP-specific TH cells (see Figure l).

W~91/09623 PCTtUS~0/07 2. VB Genes The V~ genes of 37 additional TH cells were analyzed by Southern and Northern blots using the methods described for the V~ genes (results not shown). Northern blot 5 analyses demonstrate a total of 26 TH cells positive for the V~8 and J~26 gene segments. South~rn blot analyses of these cells with the restriction enzymes Eco RI and Hind III identify common 1.6 and 9.3 kb DNA
bands, respectively, cohybridizing to probes for the 1~ V~8 gene segment and J~2 gene segment cluster ~results not shown). These DNA bands also cohybridize to an oligonucleotide probe for the J~2.6 gene segment (data not shown). Since the T-cell hybridization partner BW5147 has deleted all three of its V~8 gene segments 15 and since detailed restriction maps of cosmid clones including alr of the V~9 gene segments are available (Chou et al. (1987) Natl. Acad. Sci~ USA 84, 1992-1996;
Lai et al. (1987) Proc. Natl. Acad. Sci. USA 84, 3~46~3850), it was concluded that all these rearranged 20 V~8 bands employ the V~8.2 gene segment. For example, the restriction enzyme map in Figure 7 shows the V~8.1 gene to ~e located on a 1.3 kb Hind III fragment and the V~8.2 and V~8.3 genes on the same 9.5 kb Hind III
fragmentr Rearrangement to the J~2 locus wo~ld produce 25 fragments of the following approximate sizes: V~8.1, 2-3.5kb; V~8.2, 9-10.5kb; and V~8.3, 3-4.5 kb. The rearranged 9.3 kb Hind III band exhibited by all 4 of the TH hybridomas when probes for the V~8 and J~2 gene segments are employed (results not shown) is therefore 30 consistent with the use of the V~8.2 gene segment. All 43 TH cells were tested on the fluorescence-activated cell sorter with a monoclonal antibody specific for V~8 chains (F23.1; Staerz et al. (1985) J. Immunol. 134, 3994-4000). Since V~8 chains can be detected on the 35 surfaces of all V~8-rearranged TH cells but not on WO91/09623 PCT/US~/07 2ai~356 V~13-rearranged cells (data not shown), and since Northern blots from these T~ hybridomas hybridize to both the V~8 and J~2.6 oligonucleotide probes, it was concluded that the V~8.2 gene segment is productively 5 rearranged to the J~2.6 gene segment. The V~13-V~2.2 rearrangement has heen identified by Southern blot analysis as a 2.7 kb Hind III band cohybridizing to probes for the V~13 and the J~2.2 gene segments (data not shown). This 2.7 kb band was present in all 10 hybridomas expressing the V~13 and J~2.2 gene segments by Northern blot analyses. It was concluded that the 43 MBP-specific TH cells employ just two V~ and two J~
gene segments in productive rearrangements ~see Figure 1).
Flow CytometrY
Cells were resuspended in phosphate-buffered saline (PBS) at 2 x 107 cells/ml and seeded at 50 ~l/well into conical-bottom 96-well plates containing 50 ~1 primary 20 antibody. The anti-V~8 murine monoclonal antibody F23.1 (Staerz et al. (1985) J. Immunol. 134, 3994-4000) and the anti-CD3 murine monoclonal antibody 500A.A2 (Richie et al. (1988) J. Immunol. 140, 4115-4122) were used at 10~g/ml. Following incubation with primary 25 antibody for 30 minutes at 37C, the cells were washed with PBS and resuspended in 50 ~l/well 1:50 fluorescein isothiocyaniate (FITC)-conjugated goat anti-mouse IgG.
After incubation at room temperature for 20 minutes, the cells were washed with PBS and resuspended in 1 ml 30 PBS containing 1% FCS. Histograms relating the log of fluorescence intensity to cell number were based on an analysis of 2 x 104 cells using an Ortho 50H
Cytofluorograph (5~5 nm green band pass filter with an argon laser excitation of 488 nm).

6~3 PCT/U~90/076 2072~56 Isolation of_cDNA Clones and DNA Sequencin~
1. v~ Genes Double-stranded cDN~ was synthesized from 5 ~g polyA+
RNA using the Amersham cDNA Synthesis System 5 (RPN.1256). cDNA libraries were constructed by adding syn~hetic Eco RI linkers and cloning into the Eco RI
site of lambda gtlO with the Amersham cDNA Cloning System (RPN.1257). Libraries contained 1-3 x 107 total recombinants and were screened with C~, C~ and specific 10 V~-region probes at a concentration of 105 cpm/ml.
Clones with the longest cDNA inserts were cloned into the Eco RI site of M13mpl8 for DNA sequencing. Dideoxy sequencing was carried out according to the method of sanger et al. (1977) Pro. Natl. Acad. Sci. USA 74, 15 5463-5467, using either the M13 universal sequencing primer or specific oligonucleotide primers to the Ca or C~ genes. All sequences were determined on both strands.

20 cDNA libraries were constructed from nine TH hybridomas from one cell line specific for the l-9NAc encephalitogenic peptide and screened these with probes for the constant region of the a (Ca) or b (C~) genes of the T-cell receptor. Fifteen C~ and 15 C~ cDNA
25 clones were isolated from each library. Clones were sequenced which (1) possessed inserts greater in length than 1.0 kb and (2) failed to hybridize to probes for the rearranged V~ and V~ genes of the BW517 tumor parent, V~1 (Chien et al. (1984) Nature 312, 31-35, 30 V~16, V~ arth et al. (1985) Nature 316, 1-7) and V~5 (Lee et al. (1988) J. Immunol. 140, 1665-1675). The sequenced inserts fell into two categories: (1) productive rearrangements, involving joining of V~-J~
or V~-D~p-J~ gene segments in open reading frames, 35 presumably encoding functional V~ or V~ regions, and WO91/09623 PCT/US90/07~99 - 7~ -2~723~6 (2) nonproductive rearrangements, involving out-of-frame joining of these same gene segments or germline 3 gene segment transcripts lacking as associated V gene segment. Productive cDNA's from seven libraries (1-7 5 clones per library) and productive ~ cDNA's from six libraries (1-5 clones per library) were isolated. All productive clones isolated from the same number of clones analyzed was presumably insufficient to detect the productive rearrangement.
The nucleotide sequences of the seven productive V~
genes are shown in Figure 8A. The results demonstrate that only two V~ gene segments, V~4.2, and a single J~
gene segment, J~39, were employed in these TH cells.
15 All six sequenced V~2 genes are identical at the nucleotide level (see Figure 8A). The V~2 gene segment is newly described and is designated as V~2.3 because it is highly similar to two other previously sequenced members of the V~2 subfamily, TA39 (Vh2.1) and TAl9 20 (V~2.2) (93%) and 96% similar, respectively (Arden et al. (1985) Nature ~16, 783-787). The single V~4 gene sequenced is identical to a previously sequenced member of this subfamily, TA28 (V~4.2) (Arden et al. (1~85) supra). A subfamily is a set of V gene segments that 25 cross hybridize and generally are 75% or more similar to one anGther (Crews et al. (1981) Cell 25, 59-66).
The J~39 element utilized by both V~2.3 and V~4.2 genes as depicted in Figure 1 has been reported previously (TA39; Arden et al. (1985) supra).
2. V~ Genes The V~ genes from six TH hybridomas specific for peptide l-9NAc using the methods described for the V~
genes. The results of the DNA sequence analysis 35 demonstrated that five ~f the six Vgenes use the same W09l/0~23 PCT/-J~tO~699 ~ 74 ~ 20723~

V~8.2 and J~2.6 gene segments, whereas the sixth employs different gene segments (V~13 and J~2.2) (results not shown). At least four of the V~8 genes have arisen from unique rearrangements, since they 5 possess unique junctional nucleotide sequences. The fifth was derived from a distinct clone exhibiting a unique additional rearrangement.

Antibody Treatment A. Antibodies Specific for V~8 Chains Block the Recognition of the MBP Peptide 1 9NAc by V~8TH Cells In Vitro It was noted that the V~8-specific monoclonal antibody F23.1 is extremely effective in blocking the antigen recognition of V~8.2TH hybridomas specific for the N-15 terminal ME3P peptide 1-9NAc (results not shown). In agreement with these data, the addition of F23.1 antibody produced virtually complete suppression of interleukin-2 (IL-2) release by V~8.2-expressing hybridomas PL127.6 and PL414 in response to the MBP
20 peptides 1-9NAc and pM1-20 (>98% inhibition, Table III). The minimum concentration of F23.1 antibody required for complete suppression of antigen recognition was approximately 35 ~g/ml; at this concentration, the level of inhibition was roughly 25 equivalent to a 300-fold reduction in the amount of stimulating antigen ~data not shown). This inhibition was specific for hybridomas expressing V~8 chains, since no significant IL-2 suppression (<8%) was exhibited by hybridomas PL154 and PL413 which expressed V~13 rather than V~8 chains (Table III). In addition, no significant IL-2 suppression was exhibited by a V~8-negative MBP-specific SJL/J T-cell hybridoma, 5317.1.

Wo91/09623 PCl/US90/07699 _75- 20723~6 ¢
b --v ~ ^^ O ^ O ^ ~
& ~ ~^ ~^ ov ~ e , e ~ ~
C~

o o o o o o o o a~ ~D ~ O ~ I~ O O
V ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ U~ ~ O ~ ~ ~ ~ ~ O U~
-d ~d E ~o o ~n 1` o `D 1` -- ~1 ~ 0 ~ ~ ~ r~ ~ ~ o~> ~ r~
L~ ,~ V ~ Cl~ CO d' - r~ U~ ,~ o V~ U') o ~J 1` o~ co ov ~ tl:~ ~ ~O O 1--O 3: ~ U ~ O
(~l .~ t~ l U~ OJ ~ I` C~ ~ 0~ C`l ~ C~ ~ ~1 N ~ C~ ~ t~ _ U
_ ~ C U~
~ O ~ CO CO C~
v) ~ C r~ C ~ ~ C r~ ~ C r~ ~ C ~ ~ ~ C r C ~ C
C ~ o ~ ~ o ~ ~ o c~ d` O ~ ~ O ~ O ~ O ~ O C`~ O C~
C ~ ~ C ~ - C ~4 ~ C ~ ~ C ~ C ~ C ~4 C ~ C 14 ~ ~ u ~ z c ~ o o ~l o ~ o ~
~ C~ V o~
,, r :~
~C ~ o~
O D ~ ~ g ~ _~ ~ ;~ C
e u u c ~" aa ~a .~ a~

u ~ w 4~ ~
D. a) u C ~ ~ ~
v D

v¦ E
cl ~ ~ c ~ ~ ~
~ l ~ O
o v ~ :~ ~ 7 rn o I cq o u~ o u~ v~ o WO 91/09623 P~r/US90/07699 2nq~
tJV

o o C~ o _1 O ~ ~ O ~ O O
J ~ o ~1 ~ ~ O ~ ~ O
A ~o r_ ~ ~ ~ ~ O~ A V ~ ~ V

v ~ o~ r~ ~ ~ o r~
V C~ f~ _ O O `O C ~J ~ ~ ~ ~ O
. U ~ O ~ O ~ o~ co r O ~ u~ I_ O U~ ~n ~ ~ o ~1 u r~ o ~o o ~ n -- V ~ ~--~ ~ ~O _l _ `D ~ _ . I U~ U~ ~ V U~ U~ _ V ~ V

C
o e D V _ _ _ _ ~ _ _ _ _ _ _I _ ~
g 1 g 1~ g 1'~ C ~ C r~ cg ~ g ~ C 1' g r~ g r~ C 1~-C )~1 o c~ o c~ o ~1 o ~ o N O ~`J O C`J O ~ O ~`1 O ~ O ~ O ~ O
C ~ C ~4 C ~ C ~ C ~ C ~ C ~ C ~ C 14 C ~ ~ ~ C 14 C L

0 U U U U U O` U U C~`
bl) ~ ¢
V ~ ~ ~ ~ ~ ~ ~ ~ ~ X ~ ` W

~ U~
g ~ ~ ~ e~
~ ~ 3 ~ D U :1 ~
:1 m C _~

~C ~ ~0; ~ X
~:: G~ ~
m ~ btJ D ~ .
'S ~ ~ ~ ~ ~ O U~ O ~U~ o WO 91/09623 PClr/US90/07699 2072~6 0 ~ ~ ~ ~~ O
C.C 07 ~o~ ~ O~0 h D

c~, ~ ` 0 ~ C 0 C ~v ~ o v -- ~ ~ ~ ~ c 3 ~ ~ ~ ~ ~ 3, o u~ ~ ~
o ~ rd 3 v ~ o ~ v . 0 " "
C r~ ~ ~ v ~ U ~D ~ ~ ~ ~ C ~" r ~ j X O ~ ~
~, ~.. o v a CL> ~. ~ C P- ~ o :~ aJ X U) C ~ ~,C 0 o ~ C ~ X O ~a ~ ~ ~ a) U~ 1: ~ a ~ u E~
V 0~ ~ v Q~ 0 ~U ~U 0 c o ~ v v . c Ll ~ ~ O U
D~ C 5,~ ~ C ~ C ~ U o o Zo ~ 0 ~ V
. ~ 1 ' ~ v ~o u P4 o ~ c ~ e ~ o C ~ v a ~ o o ~ v Q~ ~ C u ~
~~ o ,~0 0 0 0 ~C~ Ul C ~ g ~ V 0 ~ V
V _ ~ U ~ V ~ U ~
U ~1 ~1 1~ C V~ C U7 0 ~ Dl VC 3 ~ ) (O C~ V ~ ~ O
U~ ~ ~ ~~ ~ ~ ~ ~ . C U~ 0 ~ ~ O ~ U ~ V
~ r _ ~0 ~ ~ --I o C ~ --I V ~ C n~ C
u ~ u v '\ ~ 0 0 0 U ~ ~ O ~ .C v ~
~ r C ~ ~ o e ~ ~ ~ u _ 3 ~ ~ ~ e ~ ~u u u, - _, u c: :~ 0 3 ~4, ,~ ~ ~ _ C ~ C r C ~ o ~C ~D Ul ~ ~ N
0~ U ~ . V 0 ov a~ v al ~ C ~ O ~~
V 1~ G~ V ot~ ~ ~ C ~` v v~ ~ C
u o ~ ~ o ~- IJ ~ 3 Ei ~~
C~ C ~ ~1 0 ~ o ~ V u ~ ~ o ~ e ~ ~ o 2 C ~ ~ C a~ ~ C ~ :> ~ U
C o ~ o o~ o)~ ~1 V )~ 0 ~ o Ei l 0 C u~ Ll V ~ V O
.. ~ ~ ~ Ll' ~ :~ ~ V oD-- ~ U ~ ~ V~ O -- P. U 13 ~ L~
~ ~ ~ D U ~D br) U U 0 ~ ~ o 3 C
GJ C ~ -v -d E O Vl V) 0 a C ~c N v u~ ~ ~1 ~ o v ~ O ~C - o Ul ~ ~ 0 V ~ 0 0 td v --~ -d ~ O
._~ r ~ u ~ 0 ~ U~ ~o C U 3 C v C~ ~ C )~ ~ 'O D Ll o V ~
e ~ o ~ . . ~ o u 8 b~ ~J ~ ~ ~ ~ v~
u ~ ~ 4J ~o c ~ ~ ~ ~ ~ _~ ,c c _~ ~
~ 3 ~ ~) 3 ~ ~ O L~ r nl ,_~ o o o ~ C o ~ ~ --~ o D o ~ o ~1 o ~ ~1l v C
V~ 0 C U V ~ o V ~ . 11~ Ul ~ U
C O ~LI 0 ~ ~ o ~ v ~ u C
_ ~ o a v ~ ~ ~
C - r ~ ,~1 u v ~ ~ --I e ~V 'v ~ ~ VC p, :~ ~ ,c --1 u ~_, > u o o v ~ e C ~ ~ D O ~ C g~ C
C _I ~ 3 ~ V C C ~ , ~ 0 O ~ _I O ~r~ ~ ~ O ~1 0 0 ~ 0 ~ o ~ V .C .~ E Dl v r~ ~ ~ ~ 0 u ~ v~ U e ra ~n ~ U ~ D U~ O V 0 t~ e c ~ ~ ~ ~ ~ c 5~ _ o O e,~ v c-~
e _ o 0 r~ E ~ ~ w 0 _~ u b~z e ,, ~ C~ J O ~ V
O ~ ~ e 4~ U c o~ ~ C ~ 0 C crr~ ~ oe ~ O 0 ~ ~ O ~
U ~ ~~; ~ D ~ ' U u~ C t~ I C
C ~ v) ~ ~ 0 r~ o 0 o ~ c 0~ O, V 0 vl C ~ D T~ r ~ ~ U ,C e Ul v v ~ ~ ~ u~
U O D ~ V e ~ ~~ O ~ ~ e ~ o,~
0 r~ 7 r C ~ ~ ,C ~ v ~ e c 0~ v _1 Ul ~ O ~ 0 1 --I .C V U U~ O ~ ~1 ~ C
o ~ ~ ¢ r~ V ~ J3 ~ 1 o C ~ ~ o vl ,C ~ ~J :~ Z '` 1/~ D ~ 0 ~ ~
_I v ~ 3 c~ p, e In ~ 5~ e ~ o 0 c t~ ~d V e ~ 0 ~ D ~ ,1 IJ v Q) e ~ 4~ U ,~ U ~ U ~ V ~ ~ 3 V ~ V O
Q~ ~ O ~ 4 D ~ bt) Vl ~a C bt~ L ~C IIJ ~ ~ C
X O ~\ 0 U ~I) V
C 3 ~ V) _~ V ~ Ll V U~ U ~ ~ U ~ V :~
e ~ v ~ w o ~ U ~J
vc ~ u ~ C~ e c u n c c ' ~ c ~ ~ o u u ~ e o ~ v ~ ~ u c ~ --l v ~ c c ~ ~q -O ~ O ~ ~ Cl. v~ V ~ O O e U ~ ~ O Ul Y C ~ O R
e ~ ~ U C ~ 3 a N~ ~U V ~1~0 -'~ e CL U
I.J Ul U V 0 ,D 0 o :~ ~q v u~ V q _1 0 Cl. V C 01 0 .--1 D ~ U al 0 ~ C u v~ 0 ~ e c ~ o l7 C D ,~ ~ c~ ~ z ~ ~ ~ o b~ ~C ~ r ~
.C ~ e --~ o E ~ bD u ~ v~ u ~ ~ v o U n~ V ~~' U ~i N 0 U u ,1 _ U - ~1 U v ~J ~ ~ v E~ ~ D --I C 17 ~J 0 ~ O ~ ~ r ~u r ~ r ~ o ~ O v~
1~ 3 ~ v~ S~ v C ~ ~ V ~ U ~ O ~ ~ J 3 ~ ql -I C C O U--I~ ~ D U~

20723~6 V) rd 3~ --El ~ v rd o rD o U ~ ~ U ..
IJ u u ~ 3 o -- ~ o o ~ _l ~ ~ U
c ~:L o o C C
~ \J ~ O
s ~ ~ ~
X U rd rd o U~ ~ ~ V~ ~
a u o o ~ ~
o ~ _~ O ~ _/
o o - C ~ C
~J la p. U
e ~ u u o ~ U
o ~ _ o rd ~ U .~ 3 0 r~ eu ~,~
U ra ~
e ~ u ~ c ~ ~n C Uu rd 4~ rD U ~ o rd U
r C IJ 1 C) C) ~ Uu .C
bO ~J C V~ V
~ C
rd a~ - O~ 3 e rn rd C ~ ~
O ~ a~l P. U U
d S E
rd ~ ut ~-o O
L~ S ~ uJJ ~o U ~
~ . ~ bn U ~ C
o rn O ~~ ra IJ ~ U ~
~ C r~C '' v ~ o rd U o u U U
~n o u C ~d u d rd 3 L~ e C U C C~ U
,o r~ ~ o o _4 -~ m r .C
rD E CD m ~
o ,~ ~n v r~l o C rD o .
u ~ :~ ~ C ~o O ~
O S -,~ v o U W U C ~ C C
L V 1~ 0 d 1.~ VJ tD rd ~
~DO ~. 'O U U
rD(n~C ~ - ;tl .Q O
~1 o U S O ~ .~ r C -o u o ~ L u _1 ~0 ~ O C r~
U~ . U 0 ~ D _~ 3 ~:o ~ ~ u V ~
OD~ 3 J u e ~~ rn e uCD C~.
C O m E U Z U --d o rD o ~
x ~r~~ ~ U 0~ CD O
.1~ C ~ :~ U ~ r o ~ rd t~ r .

~ 4~

~7 ~ 2072356 Since approximately 90% f TH cells responding to the MBP peptide 1-9NAc express V~8 chains on their surface, F23.1 antibodies were used to inhibit the proliferation of lymph node cells from MBP peptide-injected mice.
Such lymph node cells are capable of adoptively transferring EAE, when injected back into irradiated (350 R) BlO.PL mice after a five day period of 5 restimulation with the peptide in vitro (data not shown). BlO.PL mice were injected with 200~g of MBP
peptide l-9NAc in complete Freund's adjuvant (CFA) containing purified protein derivative (PPD) and 10 days later, the draining lymph nodes were removed and 10 cultured in vitro in the presence or absence of F23.1 antibody. Table III shows that in four separate experiments. Proliferation by BlO.PL lymphocytes in response to the MBP peptide 1-9NAc was virtually eliminated by addition of F23.1 antibody (average 5 inhibition 88%) compared with only minimal inhibition to the PPD control antigen (average inhibition 8%).

To further test for the specificity of this suppression, F1 mice (BlO.PL X SJL/J) were injected 20 with the MBP peptides 1-9NAc and pM87-98. The latter peptide is specifically recognized by F1 and SJL/J (H-2') TH cells capable of causing EAE. Since the SJL/J
genome has deleted approximately 50% of its V~ gene segments, including all of the V~8 gene segments 25 (Behlke et al. (1986) Proc. Natl. Acad. Sci. USA 83, 767-771), all SJL/J T-cells are V~8-negative. In vitro, BlO.PL T-cells from mice immunized with MBP do not respond to peptide pM87-98 and conversely, the T-cells from SJL/J mice immunized with MBP do not 30 respond to peptide 1-9NAc (Zamvil et al. (19B6) Nature 324, 258-260, and data not shown). Table III shows that in three separate experiments, treatment of cultured Fl lymph node cells with F23.1 antibody resulted in nearly complete suppression of the response to the l-9NAc peptide (average inhibition >100%), but WOslJOs623 PC~/IJS90/07699 - 80 - 20723~6 had only a small effect on the response to the pM87-98 peptide (average inhibition 1%).

B. V~8-specific Antibodies Can Block EAE Disease Induction In Vivo in BlO.PL Mice Since the F23.1 antibody appears to completely block the proliferation of V~-positive encephalitogenic TH
cells in_vitro, the ability of this antibody to suppress the induction of EAE in vivo was studied.
10 Conditions under which specific deletion of the V~8-positive subset of T-cells could be achieved in BlO.PL
mice were established. It was found that administration of approximately 500~g of F23.1 antibody intraperitoneally resulted in the virtual elimination of V~8-positive lymphocytes from the spleen, lymph nodes, peripheral blood and thymus within 72 hours.

The lower half of Figure 9 presents representative data from a typical experiment. Spleen cells were removed 20 at various times following intraperitoneal injection of F23.1 antibody and subjected to quantitative fluorescence flow cytometry using the F23.1 antibody and fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse IgG. Within 24 hours the level of V~8 25 expression on individual cells had decreased by about 5-fold (see Figure 9D). By 72 hours, the number of detectable V~8-positive spleen cells was essentially zero (not significantly different from background staining due to the FITC antibody alone, Figure 9D), 30 and this depletion remained complete for up to 7 days post-injection (Figure 9E). In further experiments it was determined that this depletion remains virtually complete for up to four weeks following administration of antibody (data not shown). A similar depletion was 3~ noted for T-cells from peripheral blood, lymph nodes, WO91/09623 PC~/US~07699 - 81 - 20723~6 and thymus, and L3T4+ and Lyt-2+ V~8-positive T~cells were affected equally (data not shown) At present, it is not known whether these cells have been permanently eliminated or whether they have merely temporarily lost 5 surface expression of their V~8-positive T-cell receptors.

To determine whether such V~ T-cell receptor-depleted mice could develop EAE, B10.PL and (S3L/3 x B10.PL)Fl 10 mice were challenged with the encephalitogenic MBP 1-9NAc peptide and B. pertussis and then treated with F23.1 antibody 7 days later. Only 8 of 14 antibody-treated mice (57%) developed EAE, compared to 13 to 14 control mice (93%) given saline and 5 of 5 mice (100%) 15 given and isotypematched control immunoglobulin ascites (UPC10, Sigma Chemical Co.) (data not shown). This incidence was significantly different (p<0.05) when the F23.1-treated group was analyzed against the pooled control groups (18 of 19, 95%). The average maximal 20 severity of EAE for mice which developed the disease in both treatment and control groups did not differ.
Thus, once escape from antibody protection occurred, it was as severe in the treated as in the untreated animals. Analysis of spleen cells from mice treated 25 with the F23.1 antibody who developed EAE demonstrated that treatment had removed most V~8-positive spleen cells. Therefore, it appears that the mice which develop EAE despite treatment with anti-V~ antibody may expand V~13 or other non-V~8-positive TH cells specific 30 for the MBP 1-9NAc epitope. Antibody treatment for T-cells expressing V~13 TCAR is described in Example 3.

W09l/09623 PCT/US90/076~
- 82 - 2072~56 Prevention and Treatment of Murine EAE
with TCAR V~8.2 and V~13 Specific Antibodies Mice 5 BlO.PL mice (6~10 weeks old) were purchased from Jackson Laboratories (Bar Harbor, ME) or bred at the California Institute of Technology animal facility.

Mvelin Basic Protein 10 MBP was prepared from frozen rat or mouse brains purchased from Pel-Freeze Biologicals (Rogers, AR) according to a previously published protocol (Smith (1987) J. Neurochem 16, 83-92).

15 Immunization MBP was emulsified with an equal volume of complete Freund's adjuvant supplemented with 4 mg/ml H37Ra (Difco Laboratories, Inc., Detroit, MI). Mice were immunized with 150 ~g of M~P in each of the hind 20 footpads. For induction of EAE mice were also given an intravenous dose of 75 ng of purified pertussis toxin (List Biological Laboratories, Inc., Campbell, CA) at 24 and 72 hours after MBP immunization.

25 Proliferation Assays Cells were isolated from draining popliteal and inguinal lymph nodes 10 days after MBP immunization (except where noted). Lymph node cells were resuspended in Ventex HL-1 medium and plated at 4 X 105 30 cells per well in 96-well plates. MBP and Protein A-purified antibodies were added to the cultures as indicated in the text. All samples were run in triplicate. Four days later, 1 mCi [3H]thymidine was added to each well and cells were collected after an 35 additional 16 hours of culture using a PHD cell W091/09623 PCT/US~/076~
- 83 - 2~72~56 harvester (Cambridge Technology, Inc., Cambridge, MA).
counts per minute were determined by liquid scintillation counting.

5 Flow Cytometry ~nti-v~8.2 (F23.2) was a gift from Michael Bevan.
Anti-V~13 (MR12-4) was donated by Osami Kanagawa. The antibodies were purified by Protein A hromatography and then biotinylated with NHS-LC-Biotin (Pierce, 10 Rockford, IL) according to the manufacturers recommendations. 8iotinylated antibodies were diluted in phosphate-buffered saline (PBS) containing 5% fetal calf serum and 1 mM Hepes. 1 X 106 cells were stained with 20 ~g/ml of antibody at 37C for 20 minutes.
Cells were washed with PBS and then resuspended in a SO ~1 of a 1:~0 dilution of FITC-conjugated stretavidin (Pierce, Rockford, IL). After 20 minutes at 20~C, cells were washed twice with PBS and analyzed using an Ortho 50H Cytofluorograph (53S nm green band pass 20 filter with an argon laser excitation of 488 nm).

Antibodies ~eactive to V~8.2 and V~13 Inhibit MBP-specific Lymph Node Proliferation Responses in BlOP~ Mice The analysis of 33 independent MBP-specific BlOPL-25 derived T hybridomas in Example 2 detected the use ofonly V~8.2 and V~13 gene segments (see Figure 1). To measure the contribution of T-cells expressing these two V-region qenes in the response to MBP (Smith (1987) J. Neurochem 16, 83-92), specific antibodies were used to block in vitro proliferation responses after MBP
immunization. ThP monoclonal antibodies employed were F23.2 (V~8.2-specific; Staerz et al. (1985) Molecular Bioloqy of the Immune System, Streilein et al. eds., Cambridge University Press, 61-64) and MR12-4 (V~13 specific; gift from 0. Kanagawa~. Constant region WO9~/09623 PCr/US~/076~
- 84 - 2072~5~

isotype analysis of these monoclonal antibodies classified them both as IgG1/K. Lymph node cells from BlO.PL mice were primed in vivo with MBP and restimulated in vitro with various concentrations of 5 MBP in the presence of anti-V~ antibodies or an IgGl control antibody (MOPC-31C). Proli~erative respon es were measured by [3H]thymidine incorporation. Briefly, ten days after immunization of a group of 5 B10.PL mice with MBP, draining lymph node cells were removed, 10 pooled, and restimulated in vitro with various concentrations of MBP in the presence of the indicated antibodies. After four days of culture, cells were pulsed with [in3H]thymidine for 16 hours. The results are shown in Figure 10. Each point represents mean value of triplicate measurements with mean background value (no an~igen) subtracted. The antibodies were purified by Protein A affinity chromatography and added to the cultures to final concentrations of 10~g/ml.
Similar results were obtained at final antibody 20 concentrations of 1 ~g/ml. The experiment was repeated once with similar results. As can be seen, both anti-V~8.2 and anti-V~13 partially blocked the response to MBP. At the highest concentration of MBP used for stimulation (3mM), the percent inhibitions were 59% for 25 anti-V~8.2 and 34% for anti-V~13. Simultaneous addition of both antibodies resulted in an 85%
inhibition of the response. These results lead to the conclusion that the bulk of the response to MBP in B10.PL mice is composed of V~8.2 and V~13-expressing 30 T-cells.

In vivo Injection of Anti-V~ Antibodies Leads to a Long Term Depletion of Corresponding V~-expressing T-cells Anti-TCAR V-region antibodies can be used to deplete specific T-cells. In order to optimize the conditions WO9l/os623 PCT/US~/076~
- 85 - 20723~

necessary for T-cell depletion, various concentrations of anti-V~8.2 and anti-V~13 were injected into the peritoneal cavity of BlO.PL mice. Antibody concentrations in unpurified ascites were determined by 5 comparison to a IgGl standard in an ELISA. The antibodies were then diluted and injected without further purification in a total volume of 0.1 ml p~r mouse. Seventy-two hours later, peripheral T-cells were purified over nylon wool and analyzed by 10 fluorescence flow cytometry. Cells were stained with biotinylated F23.2 followed by fluorescein isothiocynate (FITC)-conjugated streptavidin. 2x104 cells were counted for each histogram. A
representative histogram is shown in Figure 11 and the 15 data is summarized in Table IV. At all concentrations tested, injec~ion of both anti-V~8.2 and anti-V~13 resulted in a virtually complete elimination of corresponding T-cells expressing high levels of surface receptors.
TABLE IV
Down Modulation of V~8.2 and V~13 T-cells Followinq In Vivo Antibody Treatment Percent Percent Percent Percent 25 Antibody Amount _ V~8.2~ V~8.2 V~13 ~ 13h~h IgG1 control 0.500 mg 1.6 8.5 0.1 2.0 anti-V~8.2 0.500 mg 2.1 0.2 ND ND
30 anti-V~8.2 0.250 mg 2.1 0.1 0.3 1.9 anti-V~8.2 0.125 mg 2.7 0.0 ND ND
anti-V~13 0.250 mg 1.5 8.8 0.5 o.o anti-V~13 0.125 mg ND ND 0.9 0.1 --.. ..

WO 91/09623 PCr/~lS90/07699 - 86 - 2~723~6 Depletion of V~B.2 and V~13 T-cells following ~_y~ antibody treatment.
Various concentrations of either F23.2 (V~8.2-specific) or MR12-4 (V~13-specific) were injected into groups of 3 BlO.PL
mice. Three days later, lymph node cells were purified over nylon wool and analyzed by quantitative fluorescence flow cytometry. Cells were stained with either biotinylated F23.2 or biotinylated MR12-4 followed by FITC-conjugated streptavidin.
. ~
A small population of T-cells expressing lower levels 10 of surface TCARs persisted following in vivo antibody injection. In fact, the percentage of T-cells present in this dull staininq population appeared to increase after antibody treatment. This increase was especially noticeable when the lowest concentrations of antibody 15 (0.125 mg) were injected. No increase over background was seen after cell surface staining with a fluorescein-conjugated anti-mouse immunoglobulin reagent (data ~ot shown). This indicates that after 72 hours, all of the injected antibodies were cleared from 20 the T-cell surface. Therefore the residual dull staining population of T-cells did not result from subsaturated binding of the biotinylated anti-V-region antibodies used in the fluorescence analysis due to the competitive binding of any remaining injected 25 antibodies. The observed elevation in the percentage of dull staining T-cells following antibody treatment could be due to either the expansion of pre-existing dull staining populations or the antibody-induced down modulation of TCAR levels on formerly bright-staining 30 T-cells.

Mice examined at 4, 8 and 12 weeks after antibody injections showed identical staining patterns to the mice examined after 72 hours (data not shown). Sixteen 35 weeks after antibody injection the depleted T-cells ~ WO~1/09623 PCT/US~/076~
- 87 2~7~356 began to reemerge. Thus, the elimination of specific T-cells following a single anti-V-region antibody treatment appears to be a relatively long term effect.

5 Responsiveness to MBP is Diminished Following In Vivo Administration of Anti-VB Antibodies Since in vivo depletion of V~8.2 and V~13 T-cells was possible, the effect of this depletion was evaluated on the response to MBP. BlO.PL mice were treated with 250 ~g anti-V~8.2 and anti-V~13, alone and in combination, or with an IgGl control antibody, and then primed in vivo with MBP. Draining lymph node cells were isolated 10 days later and restimulated in yitro with MBP (3mM MBP) in the presence of various blocking 15 antibodies (final concentration lOmg/ml). Cells were cultured for four days and then pulsed with [31t]thymidine. The results from triplicate cultures with mean background subtracted and shown in Figure 12.

20 Injection of the control antibody had no effect on reactivity to MBP. There was a vigorous proliferation response that was inhibitable in the in vitro addition of both anti-V~8.2 and anti-V~13.

25 ~nti-V~8.2 injection reduced, but did not eliminate, reactivity to MBP. The remaining response was inhibited by anti-V~13 but not significantly hindered by addition of anti-V~8.2.

30 Anti-V~13 injection, by itself, did not lead to a significant reduction in MBP reactivity. However, the response was no longer significantly inhibited by anti-V~13, while it remained susceptible to inhibition by anti-V~8.2.

WO91/09623 ~ 6 Simultaneous injection of both anti-V~8.2 and anti-V~13 resulted in the most extreme reduction in MBP
reactivity. However, a minor proliferative response to MBP remained, that was no~ significantly inhibited by 5 either anti-v~8.2 or anti V~13 . This residual response could be the consequence of subdominant MBP~specific T-cells that might expand after the elimination of the dominant V~8.2 and V~13 MBP-specific T-cells.

10 Anti-V~ Antibody Treatment is an Effective Therapy for the Prevention of EAE in BlO.PL Mice The considerable reduction in responsiveness to MBP
brought on by injection of anti-V~ antibodies suggested that this treatment might be effective in preventing 15 the induction of EAE. To assess the efficacy of this treatment, BLO.PL mice were either left untreated or injected with various monoclonal antibodies, and then immunized with MBP in attempts to induce EAE. Results from these experiments are summarized in Table V.
20 Fourteen of 17 mice (82.4%) that were either untreated or had been injected with an IgG1 control antibody developed EAE. The average disease severity index of the animals in this group was 2.2 (see legend to Table V for denotation of the severity index scale).

W O 91/~9623 PCT/US90/07699 2~7235~
TABI.E V
Antl-TCAR M~noclonal Antibodv Tre~tment Protects A~Rlnst EAE
~vg, Severltv EAE Percent Avg, I)~y I'rentm~nt Incl~enc~ Incld~nce nll dlsensed of Onset nnlmals nn~mals none ll/13 leGl l0c~ntrol 3/4 82.llX 2.2~.4 2.7~1.19.4~0.6 nntL-Vp8. 2 5/2~a 25.0X 0.7:~:1.2C 2 6~:0.5 11.4i:0.5 antl-Val3 9/10 90.0X 2 . 7il . 4 3 . 0+1. 2 9 . 3~0 . 7 ~n~i- v~n . 2 1 1.20b 5.0% 0.1~.2d 1 19 15~ntl-Vpl3 Croups of B10.PL mice were treated wlth the lndlcated antibodies and tl~en immunized in the hlnd footpads with lS0 ~g of MBP. 75 ng of purifled pertussls toxln was ln~ected ~.v. at 24 and 48 20 hours after lmmunlzatlon. Mice were then observed daily for si~ns of EAE. Dlsease severlty was graded on a 5 polnt scale: O
normal; 1 - loss of tall tone; 2 - hlnd limb weskness, difficulty walkln~; 3 - hlnd limb paralysis, difflculty turning over; 4 - severe whole body paralysls; 5 - death,. The average 25 severity for each group was calculated ln two different ways:
(1) by averaging the maxlmum severity of all of the nnimals in the group, and (2) by averaglng the maxi~um severlty of only the diseased anlmals ln the group. The arithmetlc means and standard deviatlons are lndlcseed ln the table.
a p - 0.000085 when compared to the control group.
b p _ 0.000014 when compared to the control group and p - ~.091 when com~ared to the ~ntl-V~8.2 treated group.
c p < 0.005 when compar~d to the control ~roup.
d p < 0.001 when compared to the control group and p < 0.025 when com~ared to the nnti-Vp8.2 trested group.
p values for comparisons of EAE incldence were calcul~ted using the e~act probablllty test; p values for compnrisons of average severlty were calculated uslng the Student's t test.

Nine of 10 mice (90%) pretreated with anti-V~13 developed E~E. Disease symptoms in the afflicted mice were just as severe as those ~een in the control mice. Therefore, anti-V~13 pretreatment, by itself, W091/Os623 PCT/U~90/07699 go 20 7~3~ ~

Nine of 10 mice (90%) pretreated with anti-V~13 5 developed EAE. Disease symptoms in the afflicted mice were just as severe as those seen in the control mice.
Therefore, anti-V~13 pretreatment, by itself, i5 not an effective therapy for EAE. This is consistent with the ln vitro proliferation data presented in Figure 15 which d~monstrated that anti-V~13 when injected by itself did not reduce the overall responsiveness to MBP.

Five of 20 mice treated with anti-V~8.2 developed EAE.
15 The average severity index of the animals in this group was O.7. Therefore, anti-V~8.2 treatment alone can lead to a very significant protection against EAE
(p<0.005). This is consistent with previous results that demonstrated A reduction in EAE incidence in H-2U
20 mice pretreated with monoclonal antibodies that recoynize members of the V~8 family (Acha-Orbea et al.
(1988) Cell 54, 263-273; Urban et al. (1988) Cell 54, 577-592 and Example 2). The symptoms in the five diseased animals in this group were just as severe as 25 those seen in the control animals but were slightly delayed in onset.

Pretreatment with a combination of anti-V~8.2 and V~13 proved to be extremely effective in protecting BlO.PL
30 mice from EAE. Only one of twenty mice (5%) that had been injected with this combination of antibodies developed signs of paralysis. The average severity index of the animals in this group was 0.1. This is a very significant protection against EAE when compared 35 to control mice (p<O.OOl~ and a significant protection when compared to anti-V~8.2-treated mice (p<0.025).

W091/09623 PCT/US90/076~
2~72356 The paralysis in the one mouse that exhihited disease symptoms despite double antibody treatment was mild (a loss of tail tone) and extremely delayed in onset compared to control mice (19 days versus an average o f 5 9.4 days after MBP immunization).

The T-cells responsible for the development of EAE in this single exception could be either Y~8.2 or V~13 T-cells that were not eliminated by the treatment or subdominant T-cells that express as yet uncharacterized V~ genes. To address this issue, MBP-induced proliferative responses of lymph node cells isolated from this animal were compared to that of control animals with EAE and double antibody treated mice without EAE. Draining lymph node cells used in this experiment we~e isolated 21 days after MBP
immunization. A proliferative response was measured in the control animals with EAE that was inhibitable by both anti-V~8.2 and anti-V~13 (see Figure 13).
20 Briefly, BlO.PL mice were either untreated or treated with injections of both anti-V~8.2 and anti-V~13. Mice were then immunized with MBP and followed for signs of EAE. Three weeks after immunization, animals with and without EAE were sacrificed. Draining lymph node cells 25 were collected and assayed for MBP reactivity as previously described.

No response to MBP was detected in mice that had received double antibody treatment and were symptom-30 free. This is in contrast to the slight proliferativeresponse which was detected in lymph node cells removed from double antibody treated animals 10 days after MBP
immunization (Figure 12). This difference in responsiveness might be due to the delay in removing 35 lymph node cells for analysis. Indeed, proliferative WO9l/09623 PCT/US90/076 - 92 ~
2072~5~

responses were lower in c~lls from ~ontrol animals isolated at day 21 compared to those of cells isolated at day lo.

5 A response to MsP was seen in lymph node cells isolated from the one double antibody treated mouse that showed signs of EAE. This response was not inhibitable by either anti-V~8.2 or anti-V~13. This implies that EAE
in ~his mouse may have been caused by MBP-specific 10 T-cells that do not express either V~8.2 or V~13.

Anti-V~ Antibody Treatment Can Reverse Paralysis in B10 ! PL Mice with EAE
Groups of five B10.PL mice matched for severity of EAE
symptoms were treated with IgGl control antibody or with a combination of anti-V~8.2 and anti-V~13 three days after the first signs of NBP-induced paralysis.
The animals were then observed on daily basis and graded for the severity of paralysis. The results are 20 shown in Figure 14 where the lines correspond to the overage disease severity of survivi~g mice in each group on that day. The solid line represents the control antibody treated ~roup; the broken line represents the anti-V~ treated group.
25 Disease symptoms remained the ~ame or worsened in all five of the control antibody treated animals during the two weeks of observation. In contrast, dramatic improvements were observed in anti-V~ treated animals within two to seven days after antibody injection.
30 Reversal from severe hind leg paralysis (grade 3) to a completely normal phenotype was seen in three of the five anti-V~ treated mice. In another anti-V~ treated mouse, a reversal from a near complete whole body paralysis (grade 4) to a limited tail paralysis 35 (qrade 1) was observed. One anti-V~ treated mouse that began with whole body paralysis died three days after W~91/09623 PCT/I~S90/07~99 2072~6 treatment. These results led to the conclusion that treatment with the combination of anti-V~8.2 and anti-V~13 is an effective therapy Eor the reversal of paralysis in BlO.PL mice.
s Discussion MBP-induced encephalomyelitis i8 considered to be a paradigm for T-cell mediated disease. Striking similarities in the pathology between EAE an~ MS in 10 humans makes EAE an especially important model system to study. The dominant T-cell response to MBP in H-2U
mice is directed towards a single N-terminal determinant and involves T-cells which express a limited number of TCAR V-regions. In this Example, we 15 have demonstrated that in vivo depleti~n of T-cells with a combin~tion of monoclonal antibodies specific for the TCAR V~-regions used in the response to MBP is an effective therapy for both the prevention and treatment of EAE.

The Majority of MBP-specific T-cells in BlO.PL Mice~Express Either V~8.2 or V~13 Two approaches were used in this study to characterize the T-cells involved in the response to MBP. In the 25 first approach, BlO.PL mice were immunized with MBP and the response was recalled in vitro in the presence of TCAR V~-specific blocking antibodies (Figure 10). The virtually complete loss of responsiveness in the presence of anti-V~8.2 and anti-V~13 indicate that the 30 major T-cell response to MBP is composed of cells expressing these V-regions.

In the second approach, specific subsets of T-cells were deleted in VlVO by the injection of V~-specific 35 antibodies. Animals were then immunized with MBP and the response was once again recalled in vitro in the WO91/09623 PCT/US~/076 20723~6 presence of blocking antibodies (Figure 12). Depletion of V~8 2-expressing T-cells prior to immunization led to a reduction in MBP responsiveness, while depletion of V~13-expressing T-cells did not lead to a 5 significant reduction. This implies that the presence of V~8.2-positive T-cells is necessary and sufficient for efficient in vivo priming to MBP. Simultaneous depletion of both V~8.2 and V~13-expressing T-cells led to the most dramatic loss of MBP responsiveness, 10 however, a small proliferative response was still observed. This suggests that in the absence of V~8.2 and V~13-expressing T-cells, subdominant T-cells that utilize alternative V-regions might expand in response to MBP immunization.

V~-Specific M~noclonal Antibodies Can be Used for the ~ YiY~ Çli~ination of_~ecific T-cells A single intraperitoneal (i.p.) injection of anti-V~
antibodies led to a depletion of T-cells expressing 20 corresponding V~ gene products that lasted for up to 12 weeks. T-cell elimination could have been the result of antibody-mediated cytotoxicity, as the monoclonal antibodies used were of the IgGl isotype. However, the emergence of a small population of dull staininq 25 T-cells following antibody treatment (Figure 11) suggests that down modulation of the level of TCARs on the cell surface may have also played a role. The persistence of T-cell depletion long after the injected V~-specific antibodies would be expected to have been 30 cleared from circulation is puzzling. Presumably, continued TCAR gene rearrangements should have led to a more rapid replenishment of the depleted T-cells.
Perhaps small amounts of the injected antibodies remained sequestered in locations where they interfered 35 with T-cell development. A previous study demonstrated that the maturation of T-cells expressing particular WO9lJ09623 PCT/US~/076 2~72~56 V-regions could be blocked by ne~natal injections of a monoclonal antibody specific for these V-regions (McDuffie et al., ~1986) proc. Natl. Acad. Sci. USA 83, 8728-8732). Alternatively, injection of V~-specific 5 antibodies may have set up some type of regulatory networ~ in which the development of T-cells expressing particular V~-regions was suppressed.

V~-specific Monoclonal Antibodies 0 Can be Used to Prevent.and Treat EAE
The ability to suppress the response to MBP with injections of V~-specific antibodies allowed for the strategy of using this treatment to block EAE. In one series of experiments, antibody treatment was administered prior to MBP i~munization in order to test the efficacy Df this treatment for the prevention of EAE (Table V). Injection of anti-V~8.2 (F23.2) by itself resulted in a significant reduction in EAE
incidence (82.4% to 25%). This finding is consistent 20 with previous reports of partial protection against EAE
after treatment with a V~8-specific antibody (F23.1) or V~.3-specific antibody (KJ16) (Acha-Orbea et al. (1988) supra; Urban et al. (1988) sup~a, and Example 2 herein). The use of F23.2 for therapy is preferential 25 to the use of either F23.1 or KJ16, since the later reagents have broader specificities which would lead to the elimination of larger percentages of non-MBP-specific ~-cells. The ideal therapeutic antibody would be reactive to clonotypic determinants, and indeed such 30 an antibody raised against an MBP-specific rat T-cell hybridoma has been shown to partially protect rats from EAE (Owashi et al. (1988) J. Exp. Med. 168, 2153-2164).

Injection of anti-V~13 along with anti-V~8.2 resulted 35 in an enhanced redurtion in EAE incidence (25% to 5%).
This result implies that the failure of anti-V~8.2 to WO91~09623 PCTt~JS~0/076~9 20723~

completely protect animals from EAE was due, at least in part, to the continued presence of MBP-reactive V~13-positive T-cells. One animal in our study that had received double antibody treatment developed a 5 delayed outbreak of tail paralysis. It is possible that rare V~8.2-negative, V~13-negative, MBP-specific T-cells can occasionally expand and lead an attack.
This explanation is consistent with the observation that draining lymph node cells isolated from this mouse 10 could respond in vitro to MBP, but the response could not be blocked by addition of either anti-V~8.2 or anti-V~13 (Figure 13).

It has also been demonstrated that anti-V~ treatment 15 can rapidly reverse paralytic symptoms of EAE
(Figure 14). ~This result suggests that neurological damage associated with EAE can be readily repaired once autoagqressive T--cells are removed. In addition, the maintenance of an EAE disease state must require the 20 constant presence of MBP-reactive T-cells. Relapses have not been seen in any of the cured animals during the first four weeks of post-treatment observations.

In summary, we have demonstrated that a single passive 25 immunization with a combination of V~-specific antibodies is a very effective therapy for the prevention and treatment of EAE. It is believed that the future characterization of TCAR V-gene usage in huma diseases will lead to similar strategies of immune 30 intervention.

The Germline Repertoire of T-cell Receptor ~-Chain Genes in Patients with Chronic ~roqressive Multi~le Sclerosis WO91/09~23 PCT/US~/076~
` ~ g7 ~ 2~72356 Subiects DNA samples from 4 0 MS patients who fulfilled criteria for clinically definite MS (Poser et al. (1983) Ann.
Neurol. 13, 227-231) of the chronic progressive subtype 5 were studied. The initial objectives of this study were to compare the V~ gene seg~ent repertoire size and extent of polymorphism in MS patients with normals.
For this phase of the study, restriction enzyme digests of DNA from 28 patients were compared to those from a 10 panel of 100 unrelated normal individuals previously described in Example 1 and Concannon et al (1987) J.
Exp. Med. 165, 1130-1140. The HLA haplotypes of these randomly selected controls, designated control population #1, were unknown and during the course of these studies it became apparent that HLA haplotypes may be impor~ant in the analysis of TCAR haplotypes (Kappler et al. (1987) Cell 49, 263-27I).
Consequently, an additional control population was used, consisting of 43 DR2+ unrelated normal 20 individuals (control population #2). To assess the distribution of TCAR haplotypes, an additional 12 patients were added to the study and non-Caucasian patients were excluded for a total population of 38 Caucasian MS patients. The distribution of TCAR
25 haplotypes of the 38 Caucasian MS patients, 84% of whom were DR2+, was compared to the Caucasian individuals comprising control populations #1 and #2.

DNA PreParation 30 Genomic DNA was prepared from Epstein-Barr virus-transformed lymphoblastoid B cell lines or the buffy coat from venous blood from the patient and co~trol populations, as previously described (Beall et al.
(1987) J. Immunol. 139, 1320-1325; Conconnan et al.
35 (1987) suPra).

20723~6 Probes cDNA prohes used in the analysis corresponded to members of human V~ gene segment subfamilies V~ through V~14 isolated from subclones described by Concannon et 5 al. (Concannon et al. (1986) Proc. Natl. Acad. Sci. USA
83, 6598-6602) as set forth in Figures 2 and 3, and a c~ specific probe isolated as de~cribed by Beall et al.
(Beall et al. (1987) supra). Gel-purified inserts containing the V~, and C~ specific sequences were 10 labeled by either random priming ~Feinberg et al.
(1984) Anal. Biochem. 137, 266-267) or nick translation (Maniatis et al. (1982) Molecular Cloninq: A
Laboratorv Manual Cold Spring Harbor Lab., Cold Spring Harbor, NY.) with a-[32P3triphosphates to specific 15 activities ranging from 5 x 108 to 2 x 109 cpm/mg.

Southern Blot AnalYsis DNA from each subject was digested with Bam HI, Bgl II, Eco RI, or Hind III using conditions specified by the 20 enzyme manufacturers. Ten ~g of DNA was loaded per lane on gels, transferred to Zeta-Probe (BioRad) nylon membranes (Reed et al. (1985) Nucleic A~ids Res. 13, 7207-7221). Hybridizations were carried out for 15-24 hours at 37-42C in 50% formamide, 5xSSC, 0.02 M sodium 25 phosphate, pH 6.7, 100 mg/ml sheared denatured salmon sperm DNA, 0.5% nonfat powdered dry milk, 10% dextran sulfate, 1% SDS and 1-2 ng/ml of probe. The filters were washed in 2xSSC and 0.1~ SDS at 60C and exposed to Kodak XAR-5 film for 18-76 hours. Probes were 30 removed from blots by washing twice in boiling 0.01 x SSC for 15 min/wash on a rocker platform.

Statistical Analysis The chi-square test was used for comparisons of allelic 35 frequencies and haplotype frequency distributions _ 9~ _ 20~3~

between patients and the control populations. In the haplotype frequency distribution comparisons, data for the four rarest haplotypes were combined so that small expected numbers would not degrade the accuracy of the 5 chi-square test. The Bonferroni correction (Ingelfinger et al. (1983) Biostatistics in Clinical Medicine, MacMillan, NY, 160-176j was applied to p values of tests comparing frequencies of individual haplotypes. All chi-square calculations were performed 10 using the MINITAB computer program by the Biometry and Field Studies Branch, NINCDS, NIH.

TCAR VB Gene S~ent Repertoire_Size Germline DNA from 28 MS patients was digested with four different restriction enzymes and probed with cDNAs representing ~each of the previously characterized human T-cell receptor V~ gene subfamilies V~1 through V~14 and the C~ genes (C~1 and C~2). The autoradiograms obtained were compared to those of the control 20 population #1 to determine if either duplication or deletion of V~ segment hybridizing bands were detected in the MS patient population. No deletions of the 53 gene seqments defined by the V or C region gene families were detected (data not shown).
TCh~V~ Allelic Fre~uencies The RFLP's associated with 12 of the 14 human V~ gene segment subfamilies are described in Example 1. For most V~ probes, one predominant hybridization pattern 30 was observed, and only a few individuals displayed a second pattern. The patterns observed in our MS
patients were similar in form and frequency to those observed in control population #l (data not shown).

WO~]/09623 PCT/US~/076~
- lOO - 2072356 Two variable gene probes (V~8 and V~11) were hybridi~ed to DNA digested with Bam HI and a C~ probe was ~Iybridized to DNA digested with Bgl II. These probe/enzyme combinations detect polymorphisms whose 5 alleles occur at approximately equal frequency in the normal control population #1 (Concannon et al. (1987 supra; Robinson et al. (1985) Proc. Natl. Acad. Sci.
USA 82, 3804-3808). The frequencies of these alleles in the MS patient population were analyzed under the 10 same conditions and were not significantly different from control population #l (p values - 0.089, 0.076, and 0.204, r~spectively, Figure 15). ~enotype frequencies for each of these loci (V~8, V~ll, and C~) in the MS population were not significantly different 15 from those predicted by Hardy-Weinberg equilibrium.

TCAR~ Haplotype Frequencies The polymorphisms generated by the V~8, V~11, and C~
alleles were used to define informative haplotypes.
20 Haplotype frequencies in 38 Caucasian MS patients were compared to those of two normal populations.
Haplotypes could be assigned in individuals who were doubly or triply homozygous for the three loci. These individuals were assigned haplotypes consisted of 25 eighteen of the MS patients and 65 of the Caucasian individuals comprising control populations #1 and #2.
The distribution of haplotype frequencies seen in the MS patients differed significantly from that seen in Caucasian individuals of the control population 30 (p=0.012, Table VI). In the MS population, there was a slight over representation of the haplotype 23/20/9, as defined by the V~8 23~0 kb allele, the V~11 20.0 kb allele, and the C~9.0 kb allele (Bonferroni corrected p=0.055, Relative Risk (RR)=2.96).

WO9l/09623 PCT/~S90/076~

~0723~6 TA LE VI
Haplotype Frequencies Defined by the Vp8, V~ll and Ca Probes in MS Patients and Normal Sublects l~aplotypes Observed ~requencies 5 (V~8/V~11/C~) MSNormala MSNormal 23/25/10 o 2 o.oo 0.02 2/25/10 4 35 0.11 0.27 23/20~10 9 30 0.25 0.23 2/20/10 1 4 0.03 0~03 23/25/9 2 o 0.~6 o.oo 2/25/9 4 33 0.11 0.25 23/20/9 14 23 0.39 ~.18 ~/20/9 2 3 0.06 0.02 llaplotypes (N) 36 130 Defined X2 = 12.93,p=0.012(4df)b a caucasian individuals from control populations ~1 and #2 b Chi-square analysis of the haplotype frequency distributions between the normal population and the MS patient population. Data for the four rarest haplotypes (23/25/10, 2/20/10, 23/25/9, and 2/20/9) were combined so that small expected cell numbers would not degrade the accuracy of the chi-square test.
Thirty-two of the 38 Caucasian MS patients used ln the analysis of allelic and haplotype frequencies were DR2+ (84% DR2+). In order to determine if HLA
contributed to differences in haplotype frequencies, three comparisons were performed with control population #2 (100%DR2~). First the two control groups were compared. The allelic frequencies for each of the three different probes were not significantly different in the populations (p values-0.24, 0.78, and 0.60 for the V~8, V~ll, and C~ loci, respectively). In contrast to the findings in the MS
patients, the distribution of haplotype frequencies between these control populations was not significantly different (p=0.48).

WO9l/09623 PCT/US~/076~
~ 102 - 2~723~

Secondly, when the haplotype frequencies defined by the two Y~
and one C~ gene segment probes in the DR2+ subset of 5 Caucasian MS patients (a total of 34 haplotypes) were compared to the distribution in the Caucasian individuals of the two control populations (a total of 130 haplotypes), a significant difference was found (p=0.0028, Table VII, comparison 1). In the MS
10 population, haplotype 23/20/9 was over represented (corrected p=0.028, RR=3.26), and the frequency of the haplotype 2/25/10 was decreased (corrected p-0.072, RR=0.17). Thirdly, comparison of DR2+ Caucasian MS
patients to the DR2+ Caucasian individuals in control population #2 showed a significant difference in the haplotype fre~uency distribution (p=0.015) (Table VII, comparison 2) and a slight under representation of the haplotype 2/25/10 (corrected p=0.056).

WO91/09623PCT/US~/076~
-103- 2072~56 TABI.E VII
comparisons of the Haplotype Frequencies in DR2~ M5 Patients and Normal Sub~ects llaplotypes DR2+ MSa Normal DR2+ Normal 5 (V~8/Vp1l/C~) Subjectsa~b SubjectSalc 23/25/loo(o.oo) 2~0.02) 1(0.04) 2/25/10 2(0.06) 3~(0.27) 9(0.32~
23/20/109(0.26) 30(0.23) 6(0.21) ]o 2/20/10 1(0.03) 4(0.03) O(O.Oo) 23/25/9 2(0.06) 0(0.00) 0(0.00) 2/25/9 4(0.12) 33(0.25 7(0.25) 23/20/914(0.41) 23(0.18) 5(0.18) 2/20/9 2(0.06) 3(0.02) 0(0.00) 15l~aplotypes (N) 34130 28 Defined Comparison 1. D~2+ MS vs. Nordmal Subjectsb:X2=16.14, P = 0.0028(4df) Comparison 2. DR2~ MS vs. DR2+ Normal SudjectsC:
x2 = 12.34, P = 0.015(4~f) a Frequencies are given in parentheses.
b Caucasian individuals f~-om control populations #1 and #2.
25 c Caucasian individuals from control populations #2.
d Data for the four rarest haplotypes (23/25/10, 2/20/10, 23/25/9, and 2/20/9) were combined so that small expected cell numbers would not degrade the accuracy of the chi-square test.

Ilaplotypes based on single or double homozygoslty at the V~8 and V~11 loci were also assigned. Analysis of 32 Caucasian, DR2+ MS patients indicated that 20 of these subjects were singly or doubly homozygous for the V~8 and V~11 loci which allowed assignment of 40 haplotypes. A comparison of these haplotypes in thP MS population with l6 Caucasian, DR2+ normal individuals (a total of 32 haplotypes from control population #2) showed a significant dlfference (p=o.oos) (Table VIII) with an over representation of the haplotype 23.20 (RR-2.14) and an under W091tO9623 PCT/~S90/076~
-104- ~07235~

representation of the haplotype 2/25 (corrected p=O.013, RR=0.21).

ABLE VIII
Comparison of the Haplotype Frequencies Defined by V~B and V~ll alleles in DR2+ MS Patients and DR2~ Normal Subiects llaplotypes Observed Erequencies (V~8/V~ll)MS Normala MS Normal 23/25 2 2 0.05 0.06 2/25 7 16 0.18 0.50*
23/20 25 14 0.63 0.44 2/20 6 0 0.15 0.00 x2 = 9.4s,p = 0.009(2df)b _ a Caucasian individuals from control population #2 b Chi-square analysis of the haplotype frequency distributions between the normal population and t he MS patient population. Data for the two rarest haplotypes (23/25 and 2/20) were combined so that small expected cell numbers would not degrade the accuracy of the chi-square test.
* Corrected p value = 0.013 It is believed that genetic hackground and an environmental agent, possibly a virus, are involved in the etiology of MS (Spielman et al. (1982) E~idemoil Rev. 4, 45-65; McFarland et al (1984) Ann.
NY. Acad. Sci. 436, 118-124: Ebers et al. (1986) N.
~gLl J. Med. 315 1638-1642). ~owever, the disease is not entirely genetically controlled because the concordance in monozygotic twins is considerably less than 100~. Three is increased representation of D~2 in the WO91/09623 PCT/US90/076~
- 105 ~ 2Q72356 Caucasian MS population which indicates that one genetic element may reside in the genes encoding the HLA complex (Tiwari et al (1980) Histocompatibilitv Testina 1980, P. Terasaki (~d.), UCLA Tissue Typing 5 Laboratory, Los Angeles, CA, 687-692; Batchelor (1978) Br. Med._ Bull. 34, 279-284).

Additional MS susceptibility genes have been postulated to exist, and recently, it has been suggested (Goodman 10 et al. (1987) Current Neuroloqy, S.H. Appel (Ed.), Year Book Medical Publishers, Inc., Chicago, IL, 91-127) that genetic variations in the TCAR genes may play a role in the pathogenesis of MS. Support for this has been obtained from animal models for demyelination.
15 Both EAE (Mokhtarian et al. (1984) Nature 309, 356-358) and TMEV-induced demyellnating disease (Lipton (1975) Infect. Immun. 11, 1147-1155; Melvold et al. (1987) J.
Immunol. 138, 1429-1433) are produced in SJL mice, a strain which has deleted approximately 50% of the V~
20 germline gene repertoire (Behlke et al. (1985) Science (Wash. DC) 229, 566-570; Behlke et al. (1986) Proc.
Natl. Acad. Sci. USA 83, 767-771; Lai et al. (1987) Proc. Natl. Acad. Sci. 84, 3846-3850). The present study failed to find deletions or duplications of any 25 of the 51 gene segments defined by 14 of the 20 known subfamilies of the V~ locus or of the two gene segments of the C~ locus in any of the MS patients. Thus, although the SJL mouse may have "holes" in its T-cell repertoire that could make it more susceptible to 30 demyelinating disease, the failure to find gross deletions of V~ gene segment members of the MS
population suggests that a similar abnormality does not exist in MS.

~V09]/09623 PC~/US~/0769g ~U7~356 The distribution of haplotypes defined by V~8, V~11, and C~ RFLP alleles in a group of chronic progressive MS patients in this study was different from that found ; in normal individuals (Tables VI, VII and VIII). There 5 was a predominance of the haplotype 23/20t9, and an under representation of the haplotype 2t25. Although multiple restriction enzymes and probes were used to assess V gene segment repertoire size, only those combinations previously show to define TCAR ~
10 haplotypes were used to determine allelic and haplotype frequencies. The prevalence of MS varies among ethnic populations (Matthews et al. (1987) McAlpine's Multiple Sclerosis W.B. Matthews (Ed.), Churchill Livingstone, Edinburgh, London, Melbourne, and NY, 3-26) and 15 recently, it has been shown that TCAR allotypes (Posnett et al. (1986) Proc. Natl. Acad. Sci. USA 83, 7888-7892) can vary in ethnic groups. For these reasons, the TCAR haplotypes in Caucasian individuals in the MS and control populations were compared.
The association of HL~-B7 and DR2 with susceptibility to MS in Caucasians has been calculated to have a Relative Risk (Svejgaard et al. (1983) Immunol. Rev.
70, 193-218) of 3.66 and 3.64, respectively (Tiwari et 25 al. (1980) sup~a). To determine if the presence of the TCAR b haplotype 23/20/9 contributed to an increase in the susceptibility to MS in DR2+ individuals, the frequencies of the haplotype 23/20/9 were compared between the DR2+ Caucasian MS patients and DR2+
30 Caucasian normal individuals (Table VII). A Relative Risk of 3.26 was obtained, indicating that possession of both DR2 and the 23/20/9 TCAR b haplotype provides an increased risk for developing MS over just having DR2 alone. This suggests a coordinated action of these 35 two susceptibility genes and may reflect the fact that WO91/09623 PCT/U~tO76 one encodes a receptor molecule (TC~R1 and the other (DR2) a molecule that presents antigen to the receptor molecule. Because haplotypes could only be assigned to a subset of the MS patients, this increased Relative 5 Risk may also reflect the small sample size in this study.

It is possible that the differences in haplotype frequencies are representative of linkage to an MS
10 susceptibility gene that maps between V~8 and V~ll. It will be important to determine if particular V~
haplotypes segregate with susceptibility to MS in families, and if additional genetic differences exist in the Vl locus. Such differences may produce 15 structural changes in the antigen binding specificity of the available TCAR repertoire. In addition, changes in a regulatory element could occur that would lead to a lack of expression of some V~ genes in MS patients (with a consequent hole in the TCAR repertoire).
20 Conversely, aberrant expression of certain V~ genes could produce T-cells with autoreactive potential.
Each of these alternate consequences of V~ region polymorphism have implications for immune reactivity to either autoantigens or viruses that have been 25 postulated to be involved in the pathogenesis of MS.

Restriction Fragment Length Polymorphism Correlated with Susceptibility Gene fo~ MS in V~able Reqion of ~-chain Locus 30 The locus encoding the variable region gene segments for the ~-chain of the TCAR was also investigated in a manner analogous to that presented in Example 4. In this study, 120 control samples from unrelated Caucasians of Northern European descent and 79 Caucasians of Northern European descent afflicted with WOgl/09623 P~T/US90/076 MS were analyzed for detection of a restriction fragment length polymorphism associated with a multiple ; sclerosis susceptibility gene in the ~-chain locus. of this latter MS disease ~roup, approximately 30 of the 5 individuals were the same as those used in Example 4.

This study revealed a RFLP associated with MS as detected by a V~16 probe (Klein M. et al. tl987) Proc.
Natl. Acad. Sci. 84, 6884-6888) and the restriction 10 enzy~e Mspl. Two restriction fragments were detected corresponding to 1.0 kb band and a 0.7 kb band. The results of this study are presented in Table IX.

TABLE IX
Vn 16/Mspl Genotype and Allele F're~uencies in MS Patients ~nd Controls Genotypea ~ controls 1.0/1.046 51 l.o/0.731 52 0.7/0.7 2 17 Tot~l genotypes 79 120 x2 = 9.36**(df=2 Allelea 1.0 123 156 Total alleles158 242 .
x2 = 8.12***(df=1) ---a Numbers indicate fragment size in Kb ** p<O. 01 *** pc0.005 W091/09623 PCT/US9~/076~

20723~6 As can be seen, a significant correlation exists between this restriction fragment length polymorphism and an MS. This result indicates that an MS
susceptibility qene is located in the V~ locus.

Since there is an overlap between the MS population studied herein and that of Example 4 relating to the detection of a V~ RFLP associated with an MS
susceptibility gene in the V~ locus, there is a 10 substantial likelihood that the inheritance of an MS
susceptibility gene in the V~ locus and an MS
susceptibility gene in the V~ locus i5 required for predisposition to this disease.

WQ9l/09623 PCT/US90/076~
- llO - 2072~6 Additional Sequence Polymorphisms in the T cell Receptor ~ and ~ Loci In order to determine whether certain T cell receptor haplotypes correlate with particular ~isease predispositîons, it is desirable to identify about 10 polymorphic markers that evenly span the human V~ and 0 V~ loci. To date, we have identified approximately 8 V~ locus markers and 2 V~ locus markers in addition to those already described (see examples 1 & 5; Charmley et al., l9g0, Proc. Natl. Acd. Sci. USA 87: 4~23-4827;
Concannon et al., 1990, Am. J. Hum. Genet., 47: 45-52).
These are being converted into OLA assays (Nickerson et al., 1990, Proc. Natl. Acad. Sci. USA 87:892~-8927).
These markers will also be employed to determine whether there are hot spots of recombination in the V~
or V~ loci. Once the polymorphic markers for the V~
20 and V~ loci of humans are determined, we will be able to screen populations that have particular diseases for haplotype associations. Such associations can then be used to screen populations for their risk in developing the disease in question, e.g. their predisposition to 25 disease.
Although many different assays can be done to determine the existence of specific polymorphisms and haplotypes in an individual, the preferred assay is OLA. 0LA
30 assays can currently be performed at the rate of 1200 ligation assays per day by one technician. Multiple analyses can be done on the cells obtained from a sample of saliva. Saliva contains more than sufficient cells such that, with PCR techniques, 5 to 10 analyses 35 can easily be done.

WO91/09623 PCT/US90/076~
2~723~6 Highly Informative Pol ymorphic Markers Biallelic markers are not intrinsically informative.
However, we have made the startling discovery that 5 clusters of biallelic markers can be highly informative. For example, we have identified four biallelic markers in the second intrsn of the human C~
gene, for example, see Table XI. These markers are within 400 base pairs of one another, yet strikingly, 10 they appear to be in partial linkage disequilibria with one another. This implies either hot spots of recombination have separated these markers or that gene conversion has occurred. In either case, the important point is, a vast array of different C~ haplotypes has 15 been created in the human population. For just three of these markers, the heterozygosity index exceeds 70% - an extremely informative marker. The C~ locus results are even more striking, for example, See Table XI. Here we have identified 7 biallelic markers, most 20 of which are in partial linkage disequilbria with one another. If we put combinations of three of these markers together, we can, together with any one V
polymorphism, obtain heterozygosities which range between 90 and 95%. This is a heterozygosity that is 25 as good as those in the HLA locus. The importance of these two observations is that we can, in human populations, uniquely mark T cell receptor haplotypes for family studies. A fascinating question is whether these phenomena are unique to the T cell receptor loci, 30 or are in fact present throughout the human genome.

OLAs for MHC polvmorphisms We are now attempting to automate the analysis of MHC
polymorphisms by converting these polymorphisms into 35 OLA assays. Currently, we have finished the task for W09l~09623 PCT/US~/076 - 112 ~
207~3~6 the 8 alleles of the DQ~ locus and are now on the 11 alleles of the DQ~ locus. We plan to create OLA assays for all the major Class II human polymorphisms. These will be extremely important in our family and 5 population studies in humans to make correlations between particular immune receptor loci (~ and ~ T cell receptor loci and MHC loci) and particular huma diseases.

10 Sequence Polymorphisms in the T cell receptor ~ and ~ qene com~lexes These polymorphisms were found as part of a larger effort to derive the DNA sequence information for DNA
polymorphisms in the T cell receptor ~ and ~ gene 15 complexes- Specificially, DNA primers were synthesized in order to a~plify, using PCR, genes from every known human TCR ~ and ~ subfamily (see Table X). The PCR
primers were selected from the V or C gene sequence by analyzing the percentage of G/C content and calculating 20 the melting temperatures. A primer that was fairly A/T
rich was generally longer (about 24 nucleotides) than one that was quite G/C rich (18 nucleotides). Each primer was selected to have a melting point within the range of 58 - 62C. The melting point was determined 25 by assigning a value of 4c for each G or c residue in the primer sequence and a value of 2C for each A or T
residue and then summing up the assigned values. After PCR a~plif ication using the specific primers, the PCR
products from several unrelated individuals were then 30 electrophoresed through a denaturing gradient gel.
This gel makes possible the discrimination of DNA
sequence differences on the basis of altered melting properties of allelic forms of the same gene (Meyers et al., 1987, Meth. Enzymol. 155: 501 527). In this way, 35 several TCR genes were found to be polymorphic. These WO91/09623 PCT~VS~/076~
- 113 - 2~72356 include the genes for V~2, V~4.1, V~10 and V~15. The exact sequence pol~morphism in these genes is now being determined by standard sequencing techniques.
Polymorphisms can also be detected by the Southern 5 blot/RFLP procedure described in example 1 as well as by sequencing the same gene from multiple individuals to determine the existance of sequence polymorphisms.
Once a polymorphism has been detected, many individuals from a population are screened to determine the 10 frequency of the polymorphism. In general, very rare polymorphisms that do not occur very often in the population are less valuable for defining haplotypes containing multiple polymorphic markers, since they do not occur often enough to be useful.
After identi~ying pol~morphic genes, each allelic form is then sequenced in order to determine the specific DNA base(s) which differ between the allelic forms (see Table XI). With such sequence information, different 20 strategies can be employed which allow the discrimination of the allelic forms in any individual.
We have used restriction enzyme digestion (if any DNA
change alters a restriction site, e.g. RFLP analysis, see example 1 and Charmley et al., 1990, Proc. Natl.
25 Acad. Sci. UsA 87: 4823-4827; concannon et al., 1990, Am. J. Hum. Genet., 47: 45-52), oligonucleotide ligase assay (OLA, Nickerson et al., 1990, Proc. Natl. Acad.
Sci. USA 87:8923-8927), and also the denaturing gradient gel as techniques for determining the 30 genotypes of individuals. Other strategies will also be known to one skilled in the art given the disclosure of the specific polymorphisms and teachings of this application. These include the strategy of Allele Specific Oligonucleotide (ASO) analysis and others (see W~ 9623 P~/U~ 76~

references 7 through 21 ~ited in Nickerson et al., 1990, Proc. Natl. Acad. sci. USA 87: Bs23-8927).

The DNA sequence identified by the use of the DRS18 5 primers (Table X and XI) was derived from a non-coding region contained in a cosmid which had ~een initially isolated by screening a total human cosmid library with various V region gene sey~ents~ This polymorphism is an example of sequence polymorphisms which can be 10 detected in "anonymous" regions in and around the TCR
gene complexes. From sequence information of the cosmid, PCR primers were synthesized and product amplified from multiple unrelated individuals. The amplified products were then scanned for polymorphisms l5 by direct sequencing.

The V~18 polymorphism was initially discovered by sequencing the V~18 gene from multiple individuals.
This is an extremely interesting polymorphism, because 20 the polymorphic sequence results from an altered sequence that generates an amino acid stop codon.
Thus, in Caucasians, the 10-20S of individuals who are homozygous for this stop codon are unable to express the V~18 protein product. This is an example of a 25 polymorphism that can alter function.

W091/09623 PCT/US~tO76~

TABLE X
PCR Primers Used to Detect Polymorphisms by Denaturing Gradient Gel Electrophoresis and/or DNA Sequence Analysis V gene Primer Primer Sequence (5' to 3')*
. .
V~2 VA2-1 ATGATGAAATCCTTGAGA

V~4.1 VA4-1 TGAAGTTGGTGACAAGCATTACT

V~7 VA7-1 GGGGAGCTTTCCTTCTCTATG

V~8.1 VA8-1 ACATCCATTCGAGCTGTATTTATAT

V~10 VBlO-l CAGAAAGCAAAGATGGATTGTGT

V~12 V812-l GCCCTTTGTCTCCTGTGGACA

20 V~15 VB15-1 ATGGCCTCCCTGCTCTTCTT

V~18 VB18-l GAGTGCACTGTGGAGGGAA

C~ PCB45-l CTGTTTCCCTGAAGATTGAG

C~ CAP2-l GAGCTAAGAGAGCCGTACTGG

~locus DRSl8-l CTTCTGATACTAGGAGGTCAG

30 * The primers designated "-1" are from the sense ~
strand sequence; those designated "-2" or "-3" are from the anti-sense strand sequence WO 91/Og623 PCI`/U~;90/07699 -116- 2~723~6 .C
o U
o o~
~ U ~
e ~ ~ u :

~ ~ E ~ .~ ~ c :~ E~ ~ h S O ~ U~ v ~ U ~ a) ~
15~ e E~ 8 ~ U ~ o ~
o .~ ~ ~ ~, ~ ~ ~ ~, ~ t, O~ V ~ U ~ Q~ O
x ~ o ~

20 g U~ ~ ~ U~ ~o~

a c u E~ ~ ¢ ~ u ~ = R~R
R ~ h ~ ~: ~7 u h ~-~
25~cl c ~S ~ U ~ U ~ c ,~ ~ o o OcO

3 ~ ~ = ' ~ ~' x ~ e U

35 ~ a) u~ u--I C s~
:~ ~ '~ ~ O
a~ ~ ~ > ~ > ~ 3 ~ u~ x E~ ~ ~

WO91/Og623 PCT/US90~076 20723~
Having described the preferred embodiments of the invention, it will appear to those of ordinary skill in the art that various modifications may be made to the disclosed embodiments and that such modifications are : 5 intended to be within the scope of the invention.

Claims (100)

WHAT IS CLAIMED IS:

1. A method for diagnosing predisposition to a disease comprising, obtaining a DNA test sample from an animal and detecting in said test sample a first predetermined target DNA sequence correlated with a first susceptibility gene for said disease, wherein said first target DNA sequence comprises a DNA sequence contained within or in close proximity to a genomic DNA
sequence encoding a variable region of a chain of a T-cell antigen receptor.

2. The method of Claim 1 wherein said chain of said T-cell antigen receptor is an .alpha.-chain or .beta.-chain.

3. The method of Claim 2 wherein the presence of said first target sequence defines an .alpha. or .beta.-chain haplotype indicative of the predisposition of said animal to said disease.

4. The method of Claim 1 wherein said genomic DNA
sequence encodes a gene segment of the variable region of a .beta.-chain of a T-cell antigen receptor selected from the group consisting of V.beta., D.beta. and J.beta. gene segments.

5. The method of Claim 1 wherein said genomic DNA
sequence encodes a V.beta. gene segment of the variable region of a .beta.-chain of a T-cell antigen receptor.

6. The method of Claim 1 wherein said genomic DNA
sequence encodes a gene segment of the variable region of an .alpha.-chain of a T-cell antigen receptor selected from the group consisting of V.alpha. and J.alpha. segments.

7. The method of Claim 1 wherein said genomic DNA
sequence encodes a V.alpha. gene segment of the variable region of an .alpha.-chain of a T-cell antigen receptor.

8. The method of Claim 1 wherein said first susceptibility gene comprises a V.beta. gene segment.

9. The method of Claim 1 wherein said first susceptibility gene comprises a V.alpha. gene segment.

10. The method of Claim 1 wherein said DNA test sample is derived from genomic DNA.

11. The method of Claim 10 wherein said genomic DNA
test sample is substantially free of rearranged T-cell receptor genes.

12. The method of Claim 11 wherein said genomic test DNA is obtained from a sample which is substantially free of T-cells.

13. The method of Claim 10 wherein said first target DNA sequence encodes a restriction fragment length polymorphism.

14. The method of Claim 13 wherein said animal is a human, said disease is multiple sclerosis and said restriction fragment length polymorphism is linked to a V gene segment.

15. The method of Claim 14 wherein said V gene segment is a V.beta. gene segment selected from the group consisting of gene segments between V.beta.8 and V.beta.11.

16. The method of Claim 15 wherein said V.beta. gene segment is V.beta.8 and said detecting of said restriction fragment length polymorphism comprises digesting said genomic DNA with BamHI and determining whether a BamHI
restriction site linked to said V.beta.8 gene segment is present in said sample as an indication of whether said first susceptibility gene is present in said human.

17. The method of Claim 15 wherein said V.beta. gene segment is V.beta.11 and said detecting of said restriction fragment length polymorphism comprises digesting said genomic DNA with BamHI and determining whether a BamHI
restriction site linked to said V.beta.11 gene segment is present in said sample as an indication of whether said first susceptibility gene is present in said human.

18. The method of Claim 14 wherein said V gene segment is a V.alpha. gene segment comprising V.alpha.16.

19. The method of Claim 18 wherein said detecting of said restriction fragment length polymorphism comprises digesting said genomic DNA with Msp1 and determining whether a Msp1 restriction site linked to said V.alpha.16 gene segment is present in said sample as an indication of whether said first susceptibility gene is present in said human.

20. The method of Claim 1 wherein said animal is a human and said disease is an autoimmune disease selected from the group consisting of rheumatoid arthritis, type I diabetes, juvenile diabetes, multiple sclerosis, thyroiditis, myasthenia gravis, systemic lupus erythematosus, Sjogren's syndrome, Grave's disease, Addison's disease, Goodpasture's syndrome, scleroderma, dermatomyositis, myxedema, pernicious anemia, atrophic gastritis, and hemolytic anemia.

21. The method of Claim 20 wherein said autoimmune disease is multiple sclerosis.

22. The method of Claim 21 wherein said first susceptibility gene comprises a V.beta. gene segment.

23. The method of Claim 22 wherein said first susceptibility gene is between gene segments V.beta.8 and V.beta.11.

24. The method of Claim 3 further comprising determining whether said animal has an MHC haplotype correlated with said disease as a further indication of predisposition to said disease.

25. The method of Claim 24 wherein said animal is a human, said disease is multiple sclerosis, said first target DNA sequence encodes a restriction fragment length polymorphism linked to a V.alpha. or V.beta. gene segment and said MHC haplotype is DR2+.

26. The method of Claim 3 wherein said first target DNA sequence is contained within or in close proximity to a genomic DNA sequence encoding a variable region of a T-cell antigen receptor .beta.-chain and said method further comprises detecting in said test sample a second target DNA sequence correlated with a second susceptibility gene for said disease wherein said second target DNA sequence comprises a DNA sequence contained within or in close proximity to a genomic DNA
sequence encoding a variable region of a T-cell receptor .alpha.-chain.

27. The method of Claim 26 wherein the presence of said first and said second target nucleic acid sequences define .beta.-chain and .alpha.-chain haplotypes indicative of the predisposition of said animal to said disease.

28. The method of Claim 27 wherein said method further comprises determining whether said animal has an MHC haplotype also correlated with said disease as a further indication of predisposition to said disease.

29. The method of Claim 26 wherein each of said first and said second target DNA sequences encode first and second restriction fragment length polymorphisms respectively.

30. The method of Claim 29 wherein said animal is a human, said disease is multiple sclerosis, said first restriction fragment length polymorphism is linked to a V.beta. segment and said second restriction fragment length polymorphism is linked to a V.alpha. segment.

31. The method of Claim 26 wherein said animal is a human, said disease is multiple sclerosis, said first susceptibility gene comprises a V.beta. segment and said second susceptibility gene comprises a V.alpha. segment.

32. A method for diagnosing the onset or monitoring the course of a disease comprising, obtaining a T-cell nucleic acid test sample from an animal and detecting in said test sample a first target nucleic acid sequence correlated with said disease wherein said first target nucleic acid sequence is formed of DNA or RNA corresponding to a rearranged genomic DNA sequence encoding the variable region of a chain of a T-cell antigen receptor.

33. The method of Claim 32 wherein said chain of said T-cell antigen receptor is an .alpha.-chain or .beta.-chain.

34. The method of Claim 33 wherein said rearranged genomic DNA sequence a gene segment of the variable region of a .beta.-chain of a T-cell antigen receptor selected from the group consisting of V.beta., D.beta. and J.beta.
segments.

35. The method of Claim 33 wherein said rearranged genomic DNA sequence comprises a V.beta. gene segment of the variable region of a .beta.-chain of a T-cell antigen receptor.

36. The method of Claim 33 wherein said genomic DNA
sequence comprises a gene segment of the variable region of an .alpha.-chain of a T-cell antigen receptor selected from the group consisting of V.alpha. and J.alpha.
segments.

37. The method of Claim 32 wherein said genomic DNA
sequence comprises a V.alpha. gene segment of the variable region of an .alpha.-chain of a T-cell antigen receptor.

38. The method of Claim 32 wherein said nucleic acid test sample comprises DNA or RNA from T-cells.

39. The method of Claim 32 wherein said first target nucleic acid sequence comprises a V.beta. gene segment.

40. The method of Claim 32 wherein said first target nucleic acid sequence comprises a V.alpha. gene segment.

41. The method of Claim 32 wherein said animal is a human and said disease is an autoimmune disease selected from the group consisting of rheumatoid arthritis, type I diabetes, juvenile diabetes, multiple sclerosis, thyroiditis, myasthenia gravis, systemic lupus erythematosus, Sjogren's syndrome, Grave's disease, Addison's disease, Goodpasture's syndrome, scleroderma, dermatomyositis, myxedema, pernicious anemia, atrophic gastritis, and hemolytic anemia.

42. The method of Claim 41 wherein said autoimmune disease is multiple sclerosis.

43. The method of Claim 42 wherein said first target nucleic acid sequence comprises a V.beta. gene segment.

44. The method of Claim 43 wherein said V.beta. gene segment is between gene segments V.beta.8 and V.beta.11.

45. The method of Claim 42 wherein said first target nucleic acid sequences comprises a V.alpha. gene segment.

46. The method of Claim 33 wherein said first target nucleic acid sequence comprises a DNA sequence encoding a variable gene segment of a T-cell antigen receptor .beta.-chain and said method further comprises detecting in said T-cell nucleic acid test sample a second target nucleic acid sequence correlated with said disease wherein said second target nucleic acid sequence is formed of DNA or RNA corresponding to a rearranged genomic DNA sequence comprising a variable gene segment of a T-cell receptor .alpha.-chain.

47. The method of Claim 46 wherein said animal is a human, said disease is multiple sclerosis, and said first and said second target nucleic acid sequences comprise DNA sequences encoding .beta.-chain and .alpha.-chain variable regions respectively.

48. The method of Claim 46 wherein said animal is a human, said disease is multiple sclerosis and said first and said second target nucleic acid sequences comprise DNA sequences encoding V.beta. and V.alpha. gene segments respectively.

49. The method of Claim 32 further comprising quantitating the detection of said first target nucleic acid sequence and comparing said value to that measured for the same target sequence in an appropriate baseline T-cell nucleic acid test sample.

50. A method for diagnosing the onset or course of a disease comprising obtaining a T-cell receptor test sample from an animal and detecting in said test sample a first target polypeptide sequence correlated with said disease comprising a variable region of a chain of a T-cell antigen receptor.

51. The method of Claim 50 wherein said chain of said T-cell antigen receptor is an .alpha.-chain or .beta.-chain.

52. The method of Claim 51 wherein said first target polypeptide sequence comprises a peptide segment of the variable region of a .beta.-chain of a T-cell antigen receptor selected from the group consisting of V.beta., D.beta.
and J.beta. peptide segments.

53. The method of Claim 51 wherein said first target polypeptide sequence comprises a V.beta. peptide segment of a .beta.-chain of a T-cell antigen receptor.

54. The method of Claim 51 wherein said first target polypeptide sequence comprises a peptide segment of an .alpha.-chain of a T-cell antigen receptor selected from the group consisting of V.alpha. and J.alpha. peptide segments.

55. The method of Claim 51 wherein said first target polypeptide sequence comprise a V.alpha. peptide segment of an .alpha.-chain of a T-cell antigen receptor.

56. The method of Claim 50 wherein said T-cell polypeptide test sample comprises T-cells from said animal.

57. The method of Claim 50 wherein said animal is a human and said disease is an autoimmune disease selected from the group consisting of rheumatoid arthritis, type I diabetes, juvenile diabetes, multiple sclerosis, thyroiditis, myasthenia gravis, systemic lupus erythematosus, Sjogren's syndrome, Grave's disease, Addison's disease, Goodpasture's syndrome, scleroderma, dermatomyositis, myxedema, pernicious anemia, atrophic gastritis, and hemolytic anemia.

58. The method of Claim 57 wherein said disease is multiple sclerosis.

59. The method of Claim 58 wherein said first target polypeptide sequence comprises a V.beta. peptide segment.

60. The method of Claim 59 wherein said V.beta. peptide segment is encoded by a DNA sequence between gene segments V.beta.8 and V.beta.11.

61. The method of Claim 58 wherein said first target polypeptide sequence comprises a V.alpha. peptide segment.

62. The method of Claim 57 wherein said first target polypeptide sequence comprises a variable peptide segment of a T-cell antigen receptor .beta.-chain and said method further comprises detecting in said T-cell polypeptide test sample a second target polypeptide sequence correlated with said disease comprising a variable peptide segment of a T-cell receptor a-chain.

63. The method of Claim 62 wherein said animal is a human, said disease is multiple sclerosis and said first and said second target polypeptides comprise V.alpha.
and V.beta. segments respectively.

64. The method of Claim 50 further comprising quantitating the detection of said first target polypeptide sequence and comparing said value to that measured for the same target sequence in an appropriate baseline T-cell antigen receptor test sample.

65. A method for treating an animal afflicted with a disease correlated with at least one T-cell clone containing a specific variable region in the T-cell antigen receptor of said T-cell clone, said method comprising treating said animal with at least one antibody which is reactive with said specific variable region.

66. The method of Claim 65 wherein said disease is correlated with a T-cell clone containing a specific variable segment and said method comprises treating said animal with at least one antibody which is reactive with said specific variable segment.

67. The method of Claim 66 wherein said variable segment comprises a V.beta. segment.

68. The method of Claim 66 wherein said variable segment comprises a V.alpha. segment.

69. The method of Claim 65 wherein said disease is correlated with at least two T-cell clones, each with T-cell receptors having different variable regions and said method comprises treating said animal with at least two antibodies, each specific for each of said different variable regions.

70. The method of Claim 69 wherein said T-cell receptors have different variable segments and each of said antibodies are specific for each of said different variable segments.

71. The method of Claim 65 wherein said animal is a human and said disease is an autoimmune disease selected from the group consisting of rheumatoid arthritis, type I diabetes, juvenile diabetes, multiple sclerosis, thyroiditis, myasthenia gravis, systemic lupus erythematosus, Sjogren's syndrome, Grave's disease, Addison's disease, Goodpasture's syndrome, scleroderma, dermatomyositis, myxedema, pernicious anemia, atrophic gastritis, and hemolytic anemia.

72. The method of Claim 71 wherein said disease is multiple sclerosis.

73. The method of Claim 72 wherein said antibody is specific for a V.beta. segment encoded by a V.beta. gene segment between V.beta.8 and V.beta.11.

74. An antibody for detecting or treating a disease correlated with a T-cell clone containing a specific variable region in the T-cell antigen receptor of said T-cell clone, said antibody being specific for said variable region.

75. The antibody of Claim 74 wherein said antibody is specific for a peptide comprising a V.beta. region of said T-cell antigen receptor.

76. The antibody of Claim 74 wherein said antibody is specific for a peptide comprising a J.beta. segment of said T-cell antigen receptor.

77. The antibody of Claim 74 wherein said antibody is specific for a peptide comprising a V.alpha. region of said T-cell antigen receptor.

78. The antibody of Claim 74 wherein said antibody is specific for a peptide comprising a J.alpha. segment of said T-cell antigen receptor.

79. The antibody of Claim 74 wherein said antibody is a monoclonal antibody.

80. The method of claim 70 wherein a first antibody is reactive with a variable segment that comprises a V.beta.
segment and a second antibody is reactive with a variable segment that comprises a V.alpha. segment.

81. An isolated nucleic acid that comprises a nucleotide sequence selected from the group consisting of:
(a) , and subfragments thereof wherein said subfragment comprises at least about 10 nucleotides and comprises the nucelotide X;

(b) , and subfragments thereof wherein said subfragment comprises at least about 10 nucleotides; and (c) nucleic acids complementary to (a) or (b);
wherein X is C and Y is A or G.

82. An isolated nucleic acid that comprises a nucleotide sequence selected from the group consisting of:
(a) , and subfragments thereof wherein said subfragment comprises at least about 10 nucelotides and comprises the nucleotide X;
(b) , and subfragments thereof wherein said subfragment comprises at least about 10 nucleotides;
(c) , and subfragments thereof wherein said subfragment comprises at least about 10 nucleotides; and (d) nucleic acids complementary to (a), (b) or (c);
wherein X is G, Y is C or T and Z is G or C.

83. An isolated nucleic acid that comprises a nucleotide sequence selected from the group consisting of:
(a) , and subfragments thereof wherein said subfragment comprises at least about 10 nucleotides and comprises the nucleotide X;
(b) , and subfragments thereof wherein said subfragment comprises at least about 10 nucleotides and comprises the nucleotide Y; and (c) nucleic acids complementary to (a) or (b);

wherein X is G and Y is A.

84. An isolated nucleic acid that comprises a nucleotide sequence selected from the group consisting of:
(a) , and subfragments thereof wherein said subfragment comprises at least about 10 nucleotides and comprises the nucleotide X; and (b) nucleotide sequences complementary to (a);
wherein X is T.

85. An isolated nucleic acid that comprises a nucleotide sequence selected from the group consisting of:
(a) , and subfragments thereof wherein said subfragment comprises at least about 10 nucleotides; and (b) nucleic acids complementary to (a); wherein X is T or G.

86. An isolated nucleic acid that comprises a nucleotide sequence selected from the group consisting of:
(a) , and subfragments thereof wherein said subfragment comprises at least about 10 nucleotides; and (b) nucleic acids complementary to (a); wherein X
is A or C.

87. An isolated nucleic acid that comprises a nucleotide sequence selected from the group consisting of:
(a) ;

(b) ;
(c) a nucleic acid resulting from the PCR product that utilizes as primers the nucleic acids of (a) and (b);
(d) a subfragment of (a), (b) or (c) wherein said subfragment comprises at least about 10 nucleotides;
and (e) nucleic acids complementary to (a), (b), (c) or (d).

88. The nucleic acid of claim 96 wherein said nucleotide sequence is selected from the group consisting of:
(a) ;
(b) a subfragment of (a) wherein said subfragment comprises at least about 10 nucleotides; and (c) nucleotide sequences complementary to (a) or (b);
wherein X is T or C.

89. An isolated nucleic acid that comprises a nucleotide sequence selected from the group consisting of:
(a) a nucleic acid resulting from the PCR
product that utilizes the primers VA2-1 and VA2-2 wherein said nucleic acid resulting from said PCR
product contains the V.alpha.2 sequence:

, with the proviso that said PCR product contains a polymorphic nucleotide substitution that is not contained in said V.alpha.2 sequence;
(b) a subfragment of (a) wherein said subfragment comprises at least about 10 nucelotides and said subfragment contains the polymorphic nucleotide substitution; and (c) nucleotide sequences complementary to (a) or (b).

90. An isolated nucleic acid that comprises a nucleotide sequence selected from the group consisting of:
(a) a nucleic acid resulting from the PCR product that utilizes the primers VA4-1 and VA4-2 wherein said nucleic acid resulting from said PCR product contains the V.alpha.4.1 sequence:

, with the proviso that said PCR product contains a polymorphic nucleotide substitution that is not contained in said V.alpha.4.1 sequence;
(b) a subfragment of (a) wherein said subfragment comprises at least about 10 nucelotides and said subfragment contains the polymorphic nucleotide substitution; and (c) nucleotide sequences complementary to (a) or (b).

91. An isolated nucleic acid that comprises a nucleotide sequence selected from the group consisting of:
(a) a nucleic acid resulting from the PCR product that utilizes the primers VB10-1 and VB10-2 wherein said nucleic acid resulting from said PCR product contains the V.beta.10 sequence:

, with the proviso that said PCR product contains a polymorphic nucleotide substitution that is not contained in said V.beta.10 sequence;
(b) a subfragment of (a) wherein said subfragment comprises at least about 10 nucelotides and said subfragment contains the polymorphic nucleotide substitution; and (c) nucleotide sequences complementary to (a) or (b).

92. An isolated nucleic acid that comprises a nucleotide sequence selected from the group consisting of:
(a) a nucleic acid resulting from the PCR product that utilizes the primers VB15-1 and VB15-2 wherein said nucleic acid resulting from said PCR product contains the V.beta.15 sequence:

, with the proviso that said PCR product contains a polymorphic nucleotide substitution that is not contained in said V.beta.15 sequence;
(b) a subfragment of (a) wherein said subfragment comprises at least about 10 nucelotides and said subfragment contains the polymorphic nucleotide substitution; and (c) nucleotide sequences complementary to (a) or (b).

93. A method for determining the presence of a specific polymorphic allele in a nucleic acid sample from an animal which comprises detecting a nucleic acid sequence selected from the group consisting of those nucleic acids of claims 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91 and 92 that comprise a sequence polymorphism.

94. The method of claim 93 wherein said method further comprises comparing said nucleic acid sequence to a polymorphic allele correlated to a disease, thereby diagnosing predisposition to said disease, diagnosing the onset of said disease or monitoring the treatment of said disease.

95. The method of claim 94 which further comprises detecting in said nucleic acid sample a nucleic acid sequence comprising a polymorphic MHC locus.

96. The method of claim 65 wherein the specific variable region in the T-cell antigen receptor comprises a variable region encoded by a nucleic acid selected from the group consisting of those nucleic acids of claims 81, 82, 83, 84, 89, 90, 91 and 92 that are located in an exon.

97. The method of claims 1, 32, 50 or 65 wherein said disease is selected from the group consisting of neoplastic diseases, infectious diseases, hypersensitivity, transplantation, graft-versus-host disease and degenerative nervous system diseases.

98. The method of claim 57 wherein said disease is selected from the group consisting of leukemias, lymphomas, Non-Hodgkin's lymphoma, and Hodgkin's lymphoma, and breast cancer, colon cancer, lung cancer, liver cancer, pancreatic cancer, viral infections caused by the viruses HIV, HSV, EBV, CMV, Influenza, Hepatitis A, B or C; fungal infections caused by the yeast genus Candida; parasitic infections caused by schistosomes, filaria, nematodes trichinosis, protozoa causing sleeping sickness, plasmodium causing malaria or leishmania causing leischmaniasis; and bacterial infections caused by mycobacterium, corynebacterium, or staphylococcus, Type I hypersensitivities that lead to allergies, Type II hypersensitivities such as those present in Goodpasture's syndrome, myasthenia gravis, and autoimmune hemolytic anemia, and Type IV
hypersensitivities such as those manifested in leprosy, tuberculosis, sarcoidosis and schistosomiasis, and Alzheimer's disease.

99. The method of claims 32 or 50 further comprising detecting a nucleic acid sequence comprising a polymorphic MHC nucleotide sequence correlated with said disease.

100. The method of claim 65 further comprising treating said animal with at least one antibody which is reactive with an MHC haplotype correlated with said disease.