WO1993012814A2 - Vaccination and methods against diseases resulting from pathogenic responses by specific t cell populations - Google Patents

Abstract

The present invention provides vaccines and a means of vaccinating a vertebrate so as to prevent or control specific T cell mediated pathologies, including autoimmune diseases and the unregulated replication of T cells. The vaccine is composed of a T cell receptor (TCR) or a fragment thereof corresponding to a TCR present on the surface of T cells mediating the pathology. The vaccine fragment can be a peptide corresponding to sequences of TCRs characteristic of the T cells mediating said pathology. Such a peptide can bind to conventional antigens complexed to MHC antigen presenting cells or to superantigens. Means of determining appropriate amino acid sequences for such vaccines are also provided. The vaccine is administered to the vertebrate in a manner that induces an immune response directed against the TCR of T cells mediating the pathology. This immune response down regulates or deletes the pathogenic T cells, thus ablating the disease pathogenesis. The invention additionally provides specific β-chain variable regions of T cell receptors, designated Vβ3, Vβ4, Vβ12, Vβ14 and V, β17, which are associated with the pathogenesis of autoimmune diseases, such as rheumatoid arthritis (RA) and multiple sclerosis (MS). Also provided are means to detect, prevent and treat RA and MS. Methods of administering DNA or RNA encoding the polypeptides useful as vaccines of the present invention into the tissue cells of an individual is also provided.

Description

VACCINATION AND METHODS AGAINST DISEASES RESULTING FROM PATHOGENIC RESPONSES BY SPECIFIC T CELL POPULATIONS

BACKGROUND OF THE INVENTION This invention relates to the immune system and, more specifically, to methods of modifying

pathological immune responses.

Higher organisms are characterized by an immune system which protects them against invasion by

potentially deleterious substances or microorganisms.

When a substance, termed an antigen, enters the body, and is recognized as foreign, the immune system mounts both an antibody-mediated response and a cell-mediated

response. Cells of the immune system termed B

lymphocytes, or B cells, produce antibodies that

specifically recognize and bind to the foreign substance. Other lymphocytes termed T lymphocytes, or T cells, both effect and regulate the cell-mediated response resulting eventually in the elimination of the antigen. A variety of T cells are involved in the cell- mediated response. Some induce particular B cell clones to proliferate and produce antibodies specific for the antigen. Others recognize and destroy cells presenting foreign antigens on their surfaces. Certain T cells regulate the response by either stimulating or

suppressing other cells.

While the normal immune system is closely regulated, aberrations in immune response are not

uncommon. In some instances, the immune system functions inappropriately and reacts to a component of the host as if it were, in fact, foreign. Such a response results in an autoimmune disease, in which the host's immune system attacks the host's own tissue. T cells, as the primary regulators of the immune system, directly or indirectly effect such autoimmune pathologies.

Numerous diseases are believed to result from autoimmune mechanisms. Prominent among these are rheumatoid arthritis, systemic lupus erythematosus, multiple sclerosis. Type I diabetes, myasthenia gravis and pemphigus vulgaris. Autoimmune diseases affect millions of individuals world-wide and the cost of these diseases, in terms of actual treatment expenditures and lost productivity, is measured in billions of dollars annually. At present, there are no known effective treatments for such autoimmune pathologies. Usually, only the symptoms can be treated, while the disease continues to progress, often resulting in severe

debilitation or death. In other instances, lymphocytes replicate inappropriately and without control. Such replication results in a cancerous condition known as a lymphoma. Where the unregulated lymphocytes are of the T cell type, the tumors are termed T cell lymphomas. As with other malignancies, T cell lymphomas are difficult to treat effectively.

Thus, a long-felt need exists for an effective means of curing or ameliorating T cell mediated

pathologies. Such a treatment should ideally control the inappropriate T cell response, rather than merely

reducing the symptoms. The present invention satisfies this need and provides related advantages as well.

Summary of the Invention

The present invention provides vaccines and a means of vaccinating a vertebrate so as to prevent or control specific T cell mediated pathologies. The vaccine is composed of a substantially pure T cell receptor (TCR) or an immunogenic fragment thereof

corresponding to a TCR present on the surface of T cells mediating the pathology. The vaccine fragment can be a peptide corresponding to sequences of TCRs characteristic of the T cells mediating said pathology.

The invention additionally provides specific β- chain variable regions and their immunogenic segments, and in particular three T cell receptors, designated Vβ3, Vβ14 and Vβ17, which are associated with the pathogenesis of autoimmune diseases, for example rheumatoid arthritis (RA) and multiple sclerosis (MS). Additional VDJ

junctional (CDR3) regions associated with other

autoimmune diseases are also provided. The present invention further relates to means for detecting,

preventing and treating RA, MS and other autoimmune diseases.

The invention further provides methods of preventing or treating T cell mediated pathologies, including RA and MS, by gene therapy. In these methods, pure DNA or RNA encoding for a TCR, an immunogenic fragment thereof or an anti-idiotype antibody having an internal image of a TCR or an immunogenic fragment is administered to an individual. Vectors containing the DNA or RNA and compositions containing such vectors are also provided for use in these methods.

Brief Description of the Figures

Figure 1 shows the variable region sequences of Vβ3, Vβ14 and Vβ17. The boxed segments depict the CDR1, CDR2 and CDR4 hypervariable regions of each Vβ chain.

The sequences between the CDR2 and CDR4 regions represent an overlap between these two hypervariable regions.

Figure 2(A) shows the location of primers used in polymerase chain reaction amplification of T cell receptor β-chain genes, and 2(B) shows primer sequences used in polymerase chain reaction.

Figure 3 shows the location and sequence of primers used in polymerase chain reaction amplification of HLA-DR B₁ genes. Also shown are HLA-DR allele specific oligonucleotides.

Detailed Description of the Invention

The invention generally relates to vaccines and their use for preventing, ameliorating or treating T cell-mediated pathologies, such as autoimmune diseases and T cell lymphomas. Vaccination provides a specific and sustained treatment which avoids problems associated with other potential avenues of therapy. As used herein, the term "T cell-mediated pathology" refers to any condition in which an

inappropriate T cell response is a component of the pathology. The term is intended to encompass both T cell mediated autoimmune diseases and diseases resulting from unregulated clonal T cell replication. In addition, the term is intended to include both diseases directly

mediated by T cells and those, such as myasthenia gravis, which are characterized primarily by damage resulting from antibody binding, and also diseases in which an inappropriate T cell response contributes to the

production of those antibodies.

As used herein, "substantially the amino acid sequence," or "substantially the sequence" when referring to an amino acid sequence, means the described sequence or other sequences having any additions, deletions or substitutions that do not substantially effect the

ability of the sequence to elicit an immune response against the desired T cell receptor sequence. Such sequences commonly have many other sequences adjacent to the described sequence. A portion or segment of the described immunizing sequence can be used so long as it is sufficiently characteristic of the desired T cell receptor or fragment thereof to cause an effective immune response against desired T cell receptors, but not against undesired T cell receptors. Such variations in the sequence can easily be made, for example by

synthesizing an alternative sequence. The alternate sequence can then be tested, for example by immunizing a vertebrate, to determine its effectiveness.

As used herein, the term "fragment" means an immunogenically effective subset of the amino acid sequence that comprises a TCR. The term is intended to include such fragments in conjunction with or combined with additional sequences or moieties, as for example where the peptide is coupled to other amino acid

sequences or to a carrier. The terms "fragment" and "peptide" can, therefore, be used interchangeably since a peptide will be the most common fragment of the T cell receptor. Each fragment of the invention can have an altered sequence, as described above for the term

"substantially the sequence." Reference herein to a "fragment," "portion" or

"segment" of a T cell receptor does not mean that the composition must be derived from intact T cell receptors. Such "fragments," portions" or "segments" can be produced by various means well-known to those skilled in the art, such as, for example, manual or automatic peptide

synthesis, various methods of cloning or enzymatic treatment of a whole TCR.

As used herein when referring to the relationship between peptide fragments of the invention and sequences of TCRs, "corresponding to" means that the peptide fragment has an amino acid sequence which is sufficiently homologous to a TCR sequence or fragment thereof to stimulate an effective regulatory response in the individual. The sequence, however, need not be identical to the TCR sequence as shown, for instance, in Examples II and III.

By "immunogenically effective" is meant an amount of the T cell receptor or fragment thereof which is effectively elicits an immune response to prevent or treat a T cell mediated pathology or an unregulated T cell clonal replication in an individual. Such amounts will vary between species and individuals depending on many factors for which one skilled in the art can

determine.

As used herein, "binding partner" means a compound which is reactive with a TCR. Generally, this compound will be a Major Histocompatibility Antigen (MHC) but can be any compound capable of directly or indirectly stimulating T cell activation or proliferation when bound to the TCR. Such compounds can also, for example, be a superantigen that binds to a superantigen binding site on the TCR.

As used herein, "individual" means any vertebrate, including humans, capable of having a T cell mediated pathology or unregulated clonal T cell

replication and is used interchangeably with

"vertebrate."

As used herein, "ligand" means any molecule that reacts with another molecule to form a complex.

As used herein, "selectively binds" means that a molecule binds to one type of molecule or related group of molecules, but not substantially to other types of molecules. In relation to Vβs, "selective binding" indicates binding to TCRs or fragments thereof containing a specific Vβ without substantial cross-reactivity with other TCRs that lack the specific Vβ.

The immune system is the primary biological defense of the host (self) against potentially pernicious agents (non-self). These pernicious agents may be pathogens, such as bacteria or viruses, as well as modified self cells, including virus-infected cells, tumor cells or other abnormal cells of the host.

Collectively, these targets of the immune system are referred to as antigens. The recognition of antigen by the immune system rapidly mobilizes immune mechanisms to destroy that antigen, thus preserving the sanctity of the host environment.

The principal manifestations of an antigen- specific immune response are humoral immunity (antibody mediated) and cellular immunity (cell mediated). Each of these immunological mechanisms are initiated through the activation of helper (CD4+) T Cells. These CD4+ T cells in turn stimulate B cells, primed for antibody synthesis by antigen binding, to proliferate and secrete antibody. This secreted antibody binds to the antigen and

facilitates its destruction by other immune mechanisms. Similarly, CD4+ T cells provide stimulatory signals to cytotoxic (CD8+) T cells that recognize and destroy cellular targets (for example, virus infected cells of the host). Thus, the activation of CD4+ T cells is the proximal event in the stimulation of an immune response. Therefore, elaboration of the mechanisms underlying antigen specific activation of CD4+ T cells is crucial in any attempt to selectively modify immunological function.

T cells owe their antigen specificity to the T cell receptor (TCR) which is expressed on the cell surface. The TCR is a heterodimeric glycoprotein, composed of two polypeptide chains, each with a molecular weight of approximately 45 kD. Two forms of the TCR have been identified. One is composed of an alpha chain and a beta chain, while the second consists of a gamma chain and a delta chain. Each of these four TCR polypeptide chains is encoded by a distinct genetic locus containing multiple discontinuous gene segments. These include variable (V) region gene segments, joining (J) region gene segments and constant (C) region gene segments.

Beta and delta chains contain an additional element termed the diversity (D) gene segment. Since D segments and elements are found in only some of the TCR genetic loci, and polypeptides, further references herein to D segments and elements will be in parentheses to indicate the inclusion of these regions only in the appropriate TCR chains. Thus, V(D)J refers either to VDJ sequences of chains which have a D region or refers to VJ sequences of chains lacking D regions.

With respect to the beta chain of the variable region referred to as a Vβ, the nomenclature used herein to identify specific Vβs follows that of Kimura et al., Eur. J. Immuno. 17:375-383 (1987), with the exception that the Vβ14 herein corresponds to Vβ3.3 of Kimura et al.

During lymphocyte maturation, single V, (D) and J gene segments are rearranged to form a functional gene that determines the amino acid sequence of the TCR expressed by that cell. Since the pool of V, (D) and J genes which may be rearranged is multi-membered and since individual members of these pools may be rearranged in virtually any combination, the complete TCR repertoire is highly diverse and capable of specifically recognizing and binding the vast array of binding partners to which an organism may be exposed. However, a particular T cell will have only one TCR molecule and that TCR molecule, to a large degree if not singly, determines the specificity of that T cell for its binding partner.

Animal models have contributed significantly to the understanding of the immunological mechanisms of autoimmune disease. One such animal model, experimental allergic encephalomyelitis (EAE), is an autoimmune disease of the central nervous system that can be induced in mice and rats by immunization with myelin basic protein (MBP). The disease is characterized clinically by paralysis and mild wasting and histologically by a perivascular mononuclear cell infiltration of the central nervous system parenchyma. The disease pathogenesis is mediated by T cells having specificity for MBP. Multiple clones of MBP-specific T cells have been isolated from animals suffering from EAE and have been propagated in continuous culture. After in vitro stimulation with MBP, these T cell clones rapidly induce EAE when adoptively transferred to healthy hosts. Importantly, these EAE- inducing T cells are specific not only for the same antigen (MBP), but usually also for a single epitope on that antigen. These observations indicate that discrete populations of autoaggressive T cells are responsible for the pathogenesis of EAE.

Analysis of the TCRs of EAE-inducing T cells has revealed restricted heterogeneity in the structure of these disease-associated receptors. In one analysis of 33 MBP-reactive T cells, only two alpha chain V region gene segments and a single alpha chain J region gene segment were found. Similar restriction of beta chain TCR gene usage was also observed in this T cell

population. Only two beta chain V region segments and two J region gene segments were found. More importantly, approximately eighty percent of the T cell clones had identical amino acid sequences across the region of beta chain V-D-J joining. These findings, confirm the notion of common TCR structure among T cells with similar antigen specificities and indicate that the TCR is an effective target for immunotherapeutic strategies aimed at eliminating the pathogenesis of EAE.

An alternative mechanism for T cell activation has been suggested in which endogenous and exogenous superantigens have been shown to mediate T-cell

stimulation as described, for example, in White et al.

Cell 56:27-35 (1989) and Janeway, Cell 63:659-661 (1990).

As used herein, "superantigens" means antigens or fragments thereof that bind preferentially to T cells at specific sites on the β chain of a TCR and stimulate T cells at very high frequency rate. Such superantigens can be endogenous or exogenous. "Frequency" refers to the proportion of T cells responding to antigens and ranges from about 1/5 to 1/100 in response to

superantigens. Thus, superantigens are distinguishable from conventional antigens, which have a much lower T cell response frequency rate ranging from about 1/10⁴ to 1/10⁶. Superantigens activate T cells by binding to specific Vβs. The superantigen binding sites of various TCRs have been distinguished from the conventional hypervariable regions (CDRs) of TCRs. These CDRs

represent the regions of TCRs thought to be responsible for binding conventional antigens that are complexed to MHC.

The present invention provides an effective method of immunotherapy for T cell mediated pathologies, including autoimmune diseases, which avoids many of the problems associated with previously suggested methods of treatment. By vaccinating, rather than passively

administering heterologous antibodies, the host's own immune system is mobilized to suppress the autoaggressive T cells. Thus, the suppression is persistent and may involve any or all immunological mechanisms in effecting that suppression. This multi-faceted response is more effective than the uni-dimensional suppression achieved by passive administration of monoclonal antibodies or ex vivo-derived regulatory T cell clones which requires a highly individualized therapeutic approach because of MHC non-identity among humans in order to avoid graft versus host reactions. The methods of the present invention are also more effective than vaccination with attenuated disease-inducing T cells that lack specificity for the protective antigen on the surface of a particular T cell as well as the variable induction of immunity to that antigen. In addition, vaccination with attenuated T cells is plagued by the same labor intensiveness and need for individualized therapies as noted above for ex vivo derived regulatory T cell clones.

As they relate to autoimmune disease, the vaccine peptides of the present invention comprise TCRs or immunogenic fragments thereof from specific T cells that mediate autoimmune diseases. The vaccines can be whole TCRs substantially purified from T cell clones, individual T cell receptor chains (for example, alpha, beta, etc.) or portions of such chains, either alone or in combination. The vaccine can be homogenous, for example, a single peptide, or can be composed of more than one type of peptide, each of which corresponds to a different portion of the TCR. Further, these peptides can be from different TCRs that contribute to the T cell mediated pathology. These vaccine peptides can be of variable length so long as they can elicit a regulatory response. Preferably, such peptides are between about 5 - 100 amino acids in length, and more preferably between about 6 - 25 amino acids in length. In a specific embodiment, the immunizing peptide can have the amino acid sequence of a β-chain VDJ region when the subject has MS or RA. Any immunogenic portion of these peptides can be effective, particularly a portion having substantially the sequence SGDQGGNE or CAIGSNTE of Vβ4 and Vβ12, respectively, for MS or

substantially the sequence ASSLGGAVSYN, ASSLGGEETQYF, ASSLGGFETQYF or ASSLGGTEAFF or RA. Thus, amino acid substitutions can be made which do not destroy the immunogenicity of the peptide. Additionally, this peptide can be linked to a carrier to further increase its immunogenicity. Alternatively, whole T cell

receptors or TCR fragments that include these sequences can be used to vaccinate directly. In a further specific embodiment, T cell receptors, whole T cells or fragments of TCRs that contain Vβ17, Vβ14 or Vβ3 can be used to immunize an individual having a T cell mediated pathology to treat or prevent the disease. In a specific embodiment,

rheumatoid arthritis can be so treated. The immune response generated in the individual can neutralize or kill T cells having Vβ17, Vβ14 or Vβ3 and, thus, prevent or treat the deleterious effects of such Vβ-bearing T cells. Moreover, to the extent that Vβ17, Vβ14 or Vβ3 is common to T cell receptors on pathogenic T cells

mediating other autoimmune diseases or autoimmune

diseases in general, such vaccines can also be effective in ameliorating such other autoimmune diseases.

As used herein, "Vβ17" refers to a specific human β-chain variable region of three T cell receptors. Vβ17 has the following amino acid sequence: MSNQVLCCWLC FLGANTVDGGITQSPKYLFRKEGQNVTLSCEQNLNHDAMYWYRQDPGQGLRLIYYSQ IVNDFQKGDIAEGYSVSREKKESFPLTVTSAQKNPTAFYLCASS.

"Vβ14" refers to a specific human β-chain variable regions of another TCR. Vβ14 has the following amino acid sequence: MGPQLLGYWLCLLGAGPLEAQVTQNPRYLITV TGKKLTVTCSQNMNHEYMSWYRQDPGLGLRQIYYSMNVEVTDKGDVPEGYKVSRKEK RNFPLILESPSPNQTSLYFCASS. "Vβ3" refers to a family of specific human β- chain variable region. Two members of the Vβ3 family have been identified as Vβ3.1 and Vβ3.2. Vβ3.1 has the following amino acid sequence: MGIRLLCRVAFCFLAVGLVDVKV TQSSRYLVKRTGEKVFLECVQDMDHENMFWYRQDPGLGLRLIYFSYDVKMKEKGDIP EGYSVSREKKERFSLILESASTNQTSMYLCASS. Vβ3.2 has the

following amino acid sequence: MGIRLLCRVAFC

FLAVGLVDVKVTQSSRYLVKRTGEKVFLECVDMDHENMFWYQRQDPGLGLRLIYFSY DVKMKEKGDIPEGYSVSREKKERFSLILESASTNQTSMYLCASS.

The hypervariable or junctional regions are useful for the vaccines of the present invention.

Hypervariable regions useful in the present invention include CDR1, CDR2, CDR3 and CDR4. The amino acid sequences of the CDR1, CDR2 and CDR4 hypervariable regions for Vβ3, Vβ14 and Vβ17 are shown in Figure 1. The CDR3, also known as the V(D)J region, is useful as a vaccine of the present invention since T cell immunity elicited by peptides corresponding to this region is expected to be highly specific for a particular antigen. Due to the recombination of the V, D and J region genes prior to maturation, the amino acid sequence across these regions is virtually unique to each T cell and its clones.

However, as a germ-line element, the CDR2 region is also useful in human diseases such as MS and in particular RA. In RA studies, the results indicate a limited number of Vβs among the activated T cells

infiltrating the synovial target tissue with only a few incidences of sequence homology in the βVDJ region. Thus, peptides corresponding to the CDR2 region are viable alternatives for use as vaccines of the present invention. For example, the CDR2 region of Vβ3,

DPGLGLRLIYFSYDVKMKEKG, of Vβ14, DPGLGLRQIYYSMNVEVTDKG, or of Vβ17, DPGQGLRLIYYSQIVNKFQKG, can be used.

Modifications in these sequences that do not affect the ability of the receptor or an immunogenic fragment thereof to act as an immunogen to stimulate the desired immune response are contemplated and are included in the definition of TCR fragment. The variable region can be joined with any D and J segment of the TCR.

Further, immunogenically representative fragments of Vβ3, Vβ14 and Vβ17 are also included in the definition of "Vβ3," "Vβ14" and "Vβ17," respectively. By "substantially pure," it is meant that the

TCR is substantially free of other biochemical moieties with which it is normally associated in nature. Such substantially pure TCRs or fragments thereof, for

instance, can be synthesized, produced recombinantly by means known to those skilled in the art. In addition, whole TCRs can be enzymatically treated to produce such fragments.

In another embodiment, vaccine peptides can correspond to the Vβ regions that contain sequences of high homology which are conserved among pathogenic TCRs. These regions of conserved homology include the

conventional CDRs, such as CDR1 and CDR2, which are common to T cells bearing the same Vβ, and also the superantigen binding site, which can be common to

pathogenic TCRs bearing different Vβs. The superantigen binding site is also known to be in or around the CDR4 hypervariable region.

The vaccines of the present invention comprise peptides of varying lengths corresponding to the TCR or immunogenic fragments thereof. The vaccine peptides can correspond to regions of the TCR which distinguish that TCR from other nonpathogenic TCRs. Such specific regions can, for example, be located within the various region(s ) of the respective TCR polypeptide chains , for example , a short sequence spanning the V(D)J junction, thus

restricting the immune response solely to those T cells bearing this single determinant. The vaccines are administered to a host

exhibiting or at risk of exhibiting an autoimmune

response. Definite clinical diagnosis of a particular autoimmune disease warrants the administration of the relevant disease-specific TCR vaccines. Prophylactic applications are warranted in diseases where the

autoimmune mechanisms precede the onset of overt clinical disease (for example. Type I Diabetes). Thus,

individuals with familial history of disease and

predicted to be at risk by reliable prognostic indicators could be treated prophylactically to interdict autoimmune mechanisms prior to their onset.

TCR peptides can be administered in many possible formulations, including pharmaceutically

acceptable mediums. In the case of a short peptide, the peptide can be conjugated to a carrier, such as KLH, in order to increase its immunogenicity. The vaccine can include or be administered in conjunction with an

adjuvant, of which several are known to those skilled in the art. After initial immunization with the vaccine, further boosters can be provided. The vaccines are administered by conventional methods, in dosages which are sufficient to elicit an immunological response. Such dosages can be easily determined by those skilled in the art. Appropriate peptides to be used for

immunization can be determined as follows . Disease- inducing T cell clones reactive with the target antigens are isolated from affected individuals . Such T cells are obtained preferably from the site of active

autoaggressive activity such as a lesion in the case of pemphigus vulgaris , the central nervous system (CNS ) in the case of multiple sclerosis or the synovial fluid or tissue in the case of rheumatoid arthritis .

Alternatively, such T cells can be obtained from blood of affected individuals . The TCR genes from these

autoaggressive T cells are then sequenced. Polypeptides corresponding to TCRs or portions thereof that are selectively represented among disease inducing T cells (relative to non-pathogenic T cells ) can then be selected as vaccines and made and used as described above. An alternative method for isolating pathogenic T cells is provided by Albertini in PCT Publication No. WO88/10314 , published on December 29 , 1988. Alternatively, the vaccines can comprise anti- idiotypic antibodies which are internal images of the peptides described above. Methods of making, selecting and administering such anti-idiotype vaccines are well known in the art. See, for example, Eichmann, et al., CRC Critical Reviews in Immunology 7:193-227 (1987), which is incorporated herein by reference.

In a further aspect of the present invention, methods of preventing the proliferation of T cells associated with a T cell mediated pathology are also contemplated. Such methods include determining a T cell receptor binding partner according to the above methods and administering an effective amount of such binding partner in an appropriate form to prevent the

proliferation of the T cells. The methods can be used, for example, to build a tolerance to self antigens as in the case of an autoimmune disease.

The present invention also relates to other methods of preventing or treating a T cell pathology by inhibiting the binding of a T cell receptor to its TCR binding partner in order to prevent the proliferation of T cells associated with the T cell pathology. Ligands that are reactive with the T cell receptor or its binding partner at binding sites that inhibit the T cell receptor attachment to the binding partner can be used. Such ligands can be, for example, antibodies having

specificity for the T cell receptor or its binding partner.

The invention also provides a method of

preventing or treating a T cell mediated pathology in an individual comprising cytotoxically or cytostatically treating a Vβ-containing T-cells, particularly Vβ3, Vβ14 and Vβ17, in the individual. The Vβ-containing T cells are treated with a cytotoxic or cytostatic agent that selectively binds to the Vβ region of a T cell receptor that mediates a pathology, such as RA or MS for example. The agent can be an antibody attached to a radioactive or chemotherapeutic moiety. Such attachment and effective agents are well known in the art. See, for example, Harlow, E. and Lane, Antibodies, A Laboratory Manual, Cold Spring Harbor Laboratory (1988), which is

incorporated herein by reference.

Rheumatoid Arthritis

Rheumatoid arthritis (RA) is a T cell mediated autoimmune disease. The invention describes clonal infiltrates of activated Vβ3, Vβ14 and Vβ17 T cells in the synovium of rheumatoid arthritis patients. The presence of these T cells in the diseased tissue of most of patients examined, their clonality, and the cytotoxic activity of one such T cell for synovial adherent cells, demonstrate a central role for T cells bearing these Vβs in the pathogenesis of RA.

Activated T cell populations in the synovial tissue of RA patients have been examined by analyzing T cell receptor (TCR) mRNAs isolated from IL-2 receptor positive (IL-2R+) synovial T cells. As described in Example X(C), TCR mRNAs were amplified using a polymerase chain reaction (PCR) protocol designed to amplify human TCR β-chain genes containing virtually any desired Vβ gene element. In this analysis, clonal Vβ17

rearrangements were found to be enriched in the IL2-R+ population, indicating that Vβ17 T cells are likely involved in the pathogenesis of RA. A CD4+, Vβ17 bearing T cell clone has been isolated from one of the synovial tissue specimens and its in vitro cytotoxicity for synovial adherent cells supports the direct involvement of Vβ17 T cells in RA.

Additional studies were conducted to ensure that the prevalence of Vβ17 T cells in the initial studies did not result from an amplification bias for the Vβ consensus primer, and to examine the involvement of other TCR Vβ gene families in RA. As described in

Example XI, RNAs from activated (IL2-R+) synovial T cells were analyzed by PCR-amplification with a panel of Vβ- specific PCR primers. In this analysis, Vβ17 transcripts were found in four of the five patients tested,

confirming the association of Vβ17 with RA and validating the utility of the Vβ consensus primer. In addition, Vβ14 was found in four of the five RA patient samples and Vβ3 and Vβ9 were detectable in three of five patients.

The sequences of these various Vβ polypeptides were examined for homology to Vβ17. The results are reported in Table 1.

As shown in Table 1, mouse Vβ6 is most closely homologous, followed by hVβ3, hVβ12.1, hVβ14, and mVβ7. Three of the human Vβs, Vβ3, Vβ14 and Vβ17, were detected in the synovium of RA patients. Vβ12.1 was negligible in the synovium despite considerable overall homology with Vβ17. In contrast, Vβ9 was found in three of five synovial samples, yet is only weakly homologous to Vβ17.

A surprising discovery is the greater homology found among all of the Vβs detected in the synovia of RA patients, except Vβ9, in a contiguous stretch of 15 amino acids located carboxy to the CDR2 region. The 15 amino acid sequences of these Vβs as well as other human and mouse Vβs are shown in Table 2. Within the β-chain, this region of conservation corresponds positionally to that previously shown to contain superantigen binding sites.

The exogenous superantigen, SEC₂, stimulates human Vβ13.2 T cells as a result of binding to a site on Vβ13.2 as described in Choi et al., Nature 346:471-473 (1990), which is incorporated herein by reference. The sequence of this binding site is shown in Table 2 as the last 11 amino acids for Vβ13.2.

A binding site for Mls-1a, an endogenous superantigen, has also been mapped to this region as described in Pullen et al., Cell 61:1365-1374 (1990), which is incorporated herein by reference.

Identification of this region as the Mis binding site involved the study of Vβ8.2a, the Vβ8.2 isoform common to laboratory mice, and Vβ8.2c, a β-chain found in wild mice. These β-chain polypeptides are distinguished functionally by their differential reactivities with Mis- la and structurally by a difference of five amino acids. Of particular importance are the residues at position 70 and 71. The responder β-chain, Vβ8.2c, has lysine and glutamic acid, respectively, at these positions.

Specific mutagenesis of the non-responder gene to encode a lysine-glutamic acid pair at positions 70 and 71 rendered that non-responder β-chain Mls-1a-reactive.

This confirms the region as one of superantigen binding. Thus, both these exogenous and endogenous superantigens can bind in the vicinity of the 15 amino acid sequence homology identified in Table 2. In addition, the lysine- glutamic acid pair charge motif is implicated in Mls-1a reactivity. Involvement of this charge motif in Mls-1a reactivity is confirmed by its presence in mouse Vβ6 and Vβ7, two other Mls-1a reactive murine β-chains. The Vβ8.2c, Vβ6 and Vβ7 charge motif of a lysine or arginine followed by a glutamic acid residue is underlined in

Table 2. Thus, the superantigen binding site for Mls-1a is characterized by this charge motif contained within the region of local homology.

The present invention is directed to the unexpected discovery that human Vβ3, Vβ14 and Vβ17 have a region that corresponds to the Mls-1a binding site.

These three human Vβs display a significant degree of overall homology within the entire 94 amino acid sequence and of local homology within the 15 amino acid sequence with mVβ6 and mVβ7. Each of these Vβs possess a lysine or arginine/glutamic acid pair, which are underlined in Table 2 and represent what is meant by the term "charge motif." Vβ12.1, while displaying a high degree of overall and local homology with Vβ3, Vβ14 and Vβ17, lacks the charge motif, perhaps accounting for its presence in the synovium of only one of five RA patients. Vβ9 shows no overall or local homology to Vβs 3, 14 or 17 and lacks the charge motif.

The presence of Vβ3, Vβ14 and Vβ17-bearing T cells has been demonstrated among the activated synovial T cells in RA. These three β-chain polypeptides, in contrast to other known Vβs, possess overall and local sequence homolo-gy and an apparent superantigen binding charge motif. These results indicate that Vβ-specific T cell activation by superantigen plays a role in RA. Vβ3, Vβ14 and Vβ17 are the only known human Vβ chains to possess this apparent superantigen binding site characterized by this local sequence homology and the identified charge motif. However, it is possible that other Vβs may become known that contain such binding sites. Thus, a substantially pure Vβ3, Vβ14 or Vβ17 sequence containing the charge motif can be used as an immunogen in the vaccines of the present invention. For example, the sequences or fragments thereof shown in Table 2 for Vβ3, Vβ14 and Vβ17 can be used. Vaccines containing any combination of these three Vβ sequences, including all three sequences, can be used effectively to ameliorate T cell associated diseases.

In addition, other common V(D)J sequences of the β-chain observed in RA patients are listed in Table 3. The results taken from two different RA studies show sequence homologies in the βVDJs from four different clones, which indicates the usefulness of peptides

corresponding to the CDR3 region as appropriate vaccine candidates.

As noted, the invention provides the discovery that specific variable regions of the β-chains of three TCRs, designated Vβ3, Vβ14, and Vβ17, are closely

associated with T cell mediated pathologies, especially rheumatoid arthritis in human subjects. This discovery allows for the detection, prevention and treatment of rheumatoid arthritis using the methodology set out in this invention. Similar therapeutic approaches set out above for EAE can be applied to rheumatoid arthritis by those skilled in the art.

Specifically, the invention provides a method of diagnosing or predicting susceptibility to T cell mediated pathologies in an individual comprising

detecting T cells having the β-chain variable region from Vβ3, Vβ14 or Vβ17 in a sample from the individual, the presence of abnormal levels of such Vβ-containing T cells indicating the pathology or susceptibility to the

pathology. The Vβ-containing T cells can be

qualitatively or quantitatively compared to that of normal individuals. Such diagnosis can be performed, for example, by detecting a portion of the Vβs that does not occur on non-rheumatoid arthritis associated β-chain variable region T-cell receptors. The Vβs of the present invention can be detected, for example, by contacting the Vβs with a detectable ligand capable of specifically binding to the individual Vβs. Many such detectable ligands are known in the art, e.g. an enzyme linked antibody. Alternatively, nucleotide probes,

complementary to the individual Vβ of interest, encoding nucleic acid sequences can be utilized to detect such Vβ- containing T cells, as taught, for instance, in Examples X and XI.

The invention also provides a method of

preventing or treating a T cell mediated pathology comprising preventing the attachment of a Vβ3-, Vβ14- or Vβ17-containing T-cell receptor to its binding partner. In one embodiment, attachment is prevented by binding a ligand to Vβ3, Vβ14 or Vβ17. In an alternative

embodiment, attachment is prevented by binding a ligand to the Vβ3, Vβ14 or Vβ17 binding partner. Attachment can be prevented by known methods, e.g. binding an antibody to the individual Vβs or to its binding partner in order to physically block attachment. Multiple Sclerosis

T cells causative of multiple sclerosis (MS) have not previously been identified, though MBP-reactive T cells have been proposed to play a role due to the clinical and histologic similarities between MS and EAE. In rat and mouse models of EAE, MBP-reactive,

encephalitogenic T cells show striking conservation of β- chain V(D)J amino acid sequence, despite known

differences in MHC restriction and MBP-peptide antigen specificity. One embodiment of the invention is premised on the observation that a human myelin basic protein

(MBP)-reactive T cell line, derived from an MS patient, has a TCR β-chain with a V(D)J amino acid sequence of Vβ4 homologous with that of β-chains from MBP-reactive T cells mediating pathogenesis in experimental allergic encephalomyelitis (EAE), an animal model of MS, as shown in Table 4. This finding demonstrates the involvement of MBP-reactive T cells in the pathogenesis of MS and demonstrates that TCR peptides similar to those described herein for the prevention of EAE can be appropriate in treating MS. As shown in Table 4, a VDJ sequence of Vβ12, CAIGSNTE and one of Vβ4, SGDQGGNE, have been observed in MS patients.

The activated cells of one MS patient from the CSF were analyzed and found to be predominantly Vβ14 and Vβ3. In both cases, there was a predominate clone of Vβ14 and a predominate clone of Vβ3. These findings indicate that the results from the RA studies relating to the same Vβs can be extended to other autoimmune

pathologies, including MS.

In this regard, the invention is directed to the discovery that β-chain VDJ fragments homologous to VDJ sequences found in rodent EAE, such as SGDQGGNE and CAIGSNTE, are closely associated with multiple sclerosis in human subjects. This discovery allows for the

detection, prevention and treatment of multiple sclerosis using the methodology set out in this invention. Similar therapeutic approaches set out herein for EAE can be applied to multiple sclerosis by those skilled in the art.

Specifically, the invention provides a method of diagnosing or predicting susceptibility to multiple sclerosis in an individual comprising detecting T cells having various Vβs, such as Vβ4 or Vβ12, and particularly having substantially the sequence SGDQGGNE or CAIGSNTE, in a sample from the individual, the presence of the sequence indicating multiple sclerosis or susceptibility to multiple sclerosis. The sequences can be detected, for example, by contacting T cells or TCRs with a

detectable ligand. Many such ligands are known in the art, for example, an enzyme linked or otherwise labeled antibody specific for the sequence. Alternatively, nucleotide probes complementary to the nucleic acid encoding the sequence can be utilized as taught, for instance, in Example IX.

The invention also provides a method of

preventing or treating multiple sclerosis comprising preventing the attachment of a T-cell receptor containing various Vβs, including Vβ4, Vβ12 or fragments thereof, such as those having substantially the SGDQGGNE or

CAIGSNTE sequence to its binding partner. In one

embodiment, attachment is prevented by binding a ligand to the sequence. In an alternative embodiment,

attachment is prevented by binding a ligand to the binding partner. Attachment can be prevented by known methods, such as binding an antibody to these Vβs, and in particular to the SGDQGGNE or CAIGSNTE sequences, to physically block attachment. The invention also provides a method of preventing or treating multiple sclerosis in an

individual comprising cytotoxically or cytostatically treating T cells containing various Vβs, including Vβ4, Vβ12 and fragments thereof, particularly those having substantially the SGDQGGNE or CAIGSNTE sequence in the individual. In one embodiment, T-cells are treated with a cytotoxic or cytostatic agent which selectively binds to these Vβs or their immunogenic fragments. The agent can be, for example, an antibody attached to a

radioactive or chemotherapeutic moiety.

T Cell Pathologies of Malignant Etiology

To illustrate the utility of TCR vaccination, autoimmune disease has been discussed.

However, T cell lymphoma is another T cell pathology which would be amenable to this type of treatment.

Application of this technology in the treatment of T lymphoma would be conducted in virtually identical fashion. In one respect, however, this technology is more readily applied to T cell proliferative disease since the isolation of the pathogenic T cells is more easily accomplished. Once the clones are isolated, the technology is applied in the manner described herein.

Specifically, the TCR genes of the T lymphomas are seguenced, appropriate regions of those TCRs are

identified and used as vaccines. The vaccines can comprise single or multiple peptides, and can be

administered in pharmaceutically acceptable formulations, with or without adjuvants, by conventional means. Gene Therapy

The present invention further relates to an alternative method of treating or preventing a T cell mediated pathology by gene therapy. In this method, a nucleic acid encoding for a TCR or an immunogenic

fragment thereof is first inserted into an appropriate delivery system, for example a plasmid. The nucleic acid can be DNA or RNA encoding for TCRs, immunogenic

fragments thereof or anti-idiotype antibodies that can be used as vaccines in the present invention. Such DNA or RNA can be isolated by standard methods known in the art. The isolated nucleic acid can then be inserted into a suitable vector by known methods. Such methods are described, for example, in Maniatis et al., Molecular Cloning: A Laboratory Manual (Cold Spring Harbor

Laboratory 1982), which is incorporated herein by

reference.

The vector is subsequently administered

directly into a tissue of an individual. Preferably, the DNA or RNA-containing vector is injected into the

skeletal muscle of the individual. For example, a 1.5 cm incision can be made to expose the quadricep muscles of the subject. A 0.1 ml solution containing from 10-100 μg of a DNA or RNA plasmid and 5-20% sucrose is injected over 1 minute into the exposed quadricep muscles about 0.2 cm deep. The skin is thereafter closed. The amount of DNA or RNA plasmid can range from 10 to 100 μl of hypotonic, isotonic or hypertonic sucrose solutions or sucrose solutions containing 2 mM CaCl₃. The plasmid containing solutions can also be administered over a longer period of time, for example, 20 minutes, by

infusion. The in vivo expression of the desired gene can be tested by determining an increased production of the encoded polypeptide by the subject according to methods known in the art or as described, for example, in Wolff et al., Science 247:1465-1468 (1990).

It is believed that the treated cells will respond to the direct injection of DNA or RNA by

expressing the encoded polypeptide for at least about 60 days. Thus, the desired TCR, immunogenic fragment or anti-idiotype antibody can be effectively expressed by the cells of the individual as an alternative to

vaccinating with such polypeptides. The present invention also relates to vectors useful in the gene therapy methods and can be prepared by methods known in the art. Compositions containing such vectors and a pharmaceutically acceptable medium are also provided. The pharmaceutically acceptable medium should not contain elements that would degrade the desired nucleic acids.

The following examples are intended to illustrate but not limit the invention. EXAMPLE I

RAT MODEL OF EAE

Female Lewis rats, (Charles River Laboratories, Raleigh-Durham, NC) were immunized in each hind foot pad with 50 μg of guinea pig myelin basic protein emulsified in complete Freund's adjuvant. The first signs of disease were typically observed 9-11 days post- immunization. Disease severity is scored on a three point scale as follows: 1=limp tail; 2=hind leg

weakness; 3=hind leg paralysis. Following a disease course of approximately four to six days, most rats spontaneously recovered and were refractory to subsequent EAE induction.

EXAMPLE II SELECTION AND PREPARATION OF VACCINES Vaccinations were conducted with a T cell receptor peptide whose sequence was deduced from the DNA sequence of a T cell receptor beta gene predominating among EAE-inducing T cells of BIO.PL mice. The DNA sequence was that reported by Urban, et al., supra, which is incorporated herein by reference. A nine amino acid peptide, having the sequence of the VDJ junction of the TCR beta chain of the mouse, was synthesized by methods known to those skilled in the art. The sequence of this peptide is: SGDAGGGYE. (Amino. acids are represented by the conventional single letter codes.) The equivalent sequence in the rat has been reported to be: SSDSSNTE (Burns et al., J. Exp. Med. 169:27-39 (1989)). The peptide was desalted by Sephadex G-25 (Pharmacia Fine Chemicals, Piscataway, NJ) column chromatography in 0.1 M acetic acid and the solvent was subsequently removed by two cycles of lyophilization. A portion of the peptide was conjugated to keyhole limpet hemocyanin (KLH) with glutaraldehyde at a ratio of 7.5 mgs of peptide per mg of KLH. The resulting conjugate was dialyzed against phosphate buffered saline (PBS).

EXAMPLE III VACCINATION AGAINST EAE

Vaccines used in these studies consisted of free VDJ peptide and also of VDJ peptide conjugated to KLH. These were dissolved in PBS and were emulsified with equal volumes of either (1) incomplete Freund's adjuvant (IFA) or (2) complete Freund's adjuvant (CFA) made by suspending 10 mg/ml heat killed desiccated

Mycobacterium tuberculosis H37ra (Difco Laboratories, Detroit, MI) in IFA. Emulsions were administered to 8-12 week old female Lewis rats in a final volume of 100 microliters per animal (50 μl in each of the hind

footpads). 5 μg of unconjugated VDJ peptide were

administered per rat. KLH-VDJ conjugate was administered at a dose equivalent to 10 μg of KLH per rat. Twenty- nine days later each rat was challenged with 50 μg of guinea pig myelin basic protein in complete Freund's adjuvant in the front footpads. Animals were monitored daily beginning at day 9 for clinical signs of EAE and were scored as described above. The results are

presented in Table 5. As can be seen, not only was there a reduced incidence of the disease in the vaccinated individuals, but in those which did contract the disease, the severity of the disease was reduced and/or the onset was delayed. The extent of protection varied with the vaccine formulation, those including CFA as the adjuvant demonstrating the greatest degree of protection.

EXAMPLE IV

Vaccination against EAE with Lewis Rat VDJ peptides

The VDJ peptide used in the previous examples was synthesized according to the sequence of TCR β chain molecules found on EAE-inducing T cells in B10.PL mice. In addition, peptides were synthesized and tested which correspond to sequences found on encephalitogenic T cells in Lewis rats. These VDJ sequences are homologous with that of B10.PL mice, but not identical. The rat peptides were synthesized according to the DNA sequences reported by Burns, et al. and Chluba, et al., Eur. J. Immunol.

19:279-284 (1989). The sequences of these peptides designated IR1, 2, 3 and 9b are shown below, aligned with the BIO.PL mouse sequence used in Examples I through III (VDJ):

VDJ S G DAG G Y E IR1 CA S S D - S S N T E VF F GK

IR2 CAS S D - S G N T E VF F G K

IR3 CAS S D - S GN - VL YF GE G S R

IR9b A S S D - S S N T E

The preparation, administration and evaluation of these vaccines were conducted as described in Examples I through III with the following exceptions: 50 μg of the individual VDJ peptides were incorporated into vaccine formulations containing CFA; neither vaccinations in IFA nor vaccinations with peptides conjugated to KLH were conducted. Control animals were untreated prior to MBP challenge as in Example III or were vaccinated with emulsions of PBS and CFA to assess the protective effect of adjuvant alone. The results are shown in Table 6 below.

As shown in Table 6, disease in unvaccinated control animals was observed as early as day 10. Disease was characterized by severe paralysis and wasting, persisted for 4 to 6 days and spontaneously remitted. PBS-CFA vaccinated rats displayed disease courses

virtually indistinguishable from those of unvaccinated controls. In contrast, delays in onset were observed in some of the IR1, 2 or 3 vaccinated animals and others showed both delayed onset as well as decreased severity and/or duration of disease. Overall, however,

vaccinations with the rat VDJ peptides (IR1-3) were slightly less effective than those with the mouse VDJ peptide (Example III). Vaccination with IR9b, however, afforded complete protection in all four animals in which it was tested. Importantly, no histologic lesions characteristic of disease were found in any of the four animals vaccinated with IR9b indicating that sub-clinical signs of disease were also abrogated.

EXAMPLE V

Vaccination with V region specific peptides A peptide specific for the Vβ8 gene family was tested as a vaccine against EAE. Vβ8 is the most common β chain gene family used by encephalitogenic T cells in both rats and mice. A peptide was synthesized based on a unique DNA sequence found in the Vβ8 gene, and which is not found among other rat Vβ genes whose sequences were reported by Morris, et al., Immunogenetics 27:174-179 (1988). The sequence of this Vβ8 peptide, designated IR7, is:

IR7 DMGHGLRLIHYSYDVNSTEK The efficacy of this Vβ8 peptide was tested in the Lewis rat model of EAE (Example I) as described in Examples II and III. 50 μg of peptide were tested in CFA. Vaccinations in IFA or with peptide-KLH conjugates were not conducted. The results of these studies are shown in Table 7.

EXAMPLE VI

Comparison of Vβ8.2 Peptide Lengths

The results of vaccinations conducted with the rat Vβ8 peptide are similar to those observed with the mouse and rat IR1, 2 and 3 peptides. Delayed onset as well as decreased severity and duration of disease was observed in one animal. One animal was completely protected.

It has been found that a corresponding 21 amino acid sequence of Vβ8.2 (residues 39-59) provided less protection than IR7 as shown in Table 8. The 21 amino acid sequence of Vβ8.2 is DMGHGLRLIHYSYDVNSTEKG.

Each peptide was used at 100 μg doses and dissolved in saline prior to being emulsified in an equal volume of complete Freund's adjuvant (CFA). Animals were challenged after 42 days with 50 μg of guinea pig myelin basic protein in CFA. The CFA contained 10 mg/ml of mycobacteria tuberculosis. Injections and evaluation of clinical signs and histology were performed as previously described. Other animals (five per group) were immunized with peptides in the same way and their splenocytes were removed after 14 days to test for lymphocyte

proliferation as described in Olee et al., J.

Neuroimmunol. 21:235-240 (1989). The sequences of the 20 amino acid peptide and the 21 amino acid peptide are designated in Table 8 as Vβ8.2₂₀ and Vβ8.2₂₁ respectively. EXAMPLE VII

Vaccination with J region peptides

A peptide was synthesized which corresponds to the J α gene segment, TA39, found among both rat and mouse encephalitogenic T cell receptors. The sequence of this peptide, designated IR5, is:

IR5 RFGAGTRLTVK

The efficacy of the JαTA39 peptide was tested in the Lewis rat model of EAE (Example I) as described in Examples II and III. 50 μg of peptide were tested in CFA. Vaccinations in IFA or with peptide-KLH conjugates were not conducted. The results of these studies are shown in Table 9.

The results of vaccinations conducted with the rat J a TA39 peptide are more effective than those observed with the mouse VDJ peptide or the Vβ8 peptide. Two of three animals were totally protected and, in the third, disease onset was markedly delayed. Severity was also reduced in this animal though disease persisted for a normal course of 5 days. Importantly, the two animals which were completely protected showed no histologic evidence of T cell infiltration of the CNS. This result indicates that vaccinating with the J_αTA39 very

efficiently induces a regulatory response directed at encephalitogenic T cells. Even sub-clinical signs of disease were abrogated.

EXAMPLE VIII

Vaccination with mixtures of TCR peptides

Vaccinations were conducted with a mixture of TCR peptides. This mixture contained 50 μg of each of the peptides IR1, 2, 3 and 5 (the three rat VDJ peptides and the rat JαTA39 peptide).

The efficacy of this peptide mixture was tested in the Lewis rat model (Example I) as described in Examples II and III. Peptides were tested in CFA.

Vaccinations in IFA or with peptide-KLH conjugates were not conducted. The results of these studies are shown in Table 10.

The results of vaccinations conducted with the rat JαTA39 and three VDJ peptides were as effective as those described for IR9b in Table 6. All three animals were totally protected. In addition to the absence of any clinical signs of EAE, two of these three animals were completely free of histological evidence of T cell infiltration into the CNS while the third showed only two small foci of lymphocytic infiltration at the base of the spinal cord.

EXAMPLE IX

Multiple Sclerosis Vaccine

A. Human MBP-reactive T cells

MBP-reactive T cell lines were established from peripheral blood mononuclear cells (PBMC) of nine chronic progressive MS patients and two healthy controls. Cells were maintained in culture by regular stimulation with purified human MBP and irradiated-autologous PBMC for three days followed by four days in IL-2 containing medium.

B. PCR Amplification of TCR β-chain genes from MBP- reactive T cell lines T cells were harvested from log phase cultures and RNA was prepared, amplified with the Vβ16mer primer and nested Cβ primers for 55 cycles as described in Example X.

C. TCR β-chain sequences of human MBP-reactive T cells Vβ16mer amplified TCR β-chain genes from human

MBP-reactive T cell lines were sequenced using the Cβseq primer. Amplification products were gel purified, base denatured and sequenced from the Cβseq primer. Readable DNA sequence was obtained from 5 of these lines,

indicating that predominant T cell clones had been selected by long term in vitro passage. One of these sequences, from the MS-Re cell line (Table 11), possessed a β-chain VDJ amino acid sequence that shared five of the first six and six of nine total residues with the β-chain VDJ amino acid sequence conserved among MBP reactive, encephalitogenic T cells in the B10.PL mouse model of EAE. This sequence was not present among the predominant TCR rearrangements found in the remaining four human MBP reactive T cell lines. To determine if similar sequences were present in the β-chain repertoire of the MBP-reactive T cell lines from other MS patients, PCR amplification was conducted with a degenerate (n=1024) 21-nucleotide primer (VβRe) corresponding to seven amino acids of this

sequence. RNAs were reversed transcribed and amplified in 20 cycle stage I reactions with the Vβ16mer and Cβext primers. One μl aliquots of these stage I reactions were reamplified for 35 cycles with the VβRe and Cβ int primers. One μl aliquots of these reactions were

analyzed by Southern blot hybridization with a ³²P-labeled human Cβ probe. This analysis revealed the 300 bp amplified product in the Re cell line and in one of the other MS patient lines, but not in MBP-reactive T cells from control subjects or in non-MBP reactive human T cell lines and clones. The presence of this sequence in two of the nine MS patient lines tested is compelling. Since this sequence is known to be conserved among

encephalitogenic T cells in EAE, its detection among MBP- reactive T cells from MS patients demonstrates a role for T cells bearing this determinant in the pathogenesis of MS. Immunogenic peptides having the sequence

SGDQGGNE can be synthesized as shown in Example II and used to immunize human subjects by methods demonstrated in Example III. Such immunizations can result in an effective immune response.

EXAMPLE X

Detection of Clonal Infiltrates of

ActivatedVβ17 T Cells in the Synovium of

Rheumatoid Arthritis Patients A. T cell preparations from synovial tissue

Synovial tissue specimens were obtained from radiographically proven rheumatoid arthritis patients undergoing joint replacement therapy. Activated T cells were selected using magnetic beads and antibodies

reactive with the human IL2-R (αIL2-R) as follows.

Synovial tissue was digested for 4 hrs at 37°C in RPMI + 10% Fetal Bovine Serum (FBS) containing 4 mg/ml

collagenase (Worthington Biochemical, Freehold, NJ) and 0.15 mg/ml DNAse (Sigma, St. Louis, MO.). Digests were passed through an 80-mesh screen and single cells were collected by Ficoll density gradient centrifugation.

Cells at the interface were washed and were incubated at lOVml for 30 min at 0°C with 5 μg/ml control mouse IgG (Coulter Immunology, Hialeah, FL) in PBS containing 2% FBS (PBS-FBS). Cells were washed three times and

incubated for 30 min at 0°C with magnetic beads conjugated to goat anti-mouse IgG (Advanced Magnetics, Cambridge, MA). Beads were magnetically separated and washed three times with PBS-FBS. This preselection with mouse IgG (mlgG) and magnetic beads was used to control for nonspecific adsorption of T cells. The cells remaining in the initial suspension were further incubated 30 minutes at 0°C with 5 μg/ml monoclonal mouse IgG reactive with the human T cell IL2-R (Coulter Immunology, Hialeah, FL).

Cells were washed and selected with magnetic beads as above. Beads from the IgG preadsorption and the IL2-R antibody selection were immediately resuspended in acidified-guanidinium-phenol-chloroform and RNA prepared as described in Chonezynski and Sacchi, Anal. Biochem. 162:156 (1987), which is incorporated herein by reference. Since RNAS were prepared without in vitro culture of the cells and the accompanying bias that may be induced, they are expected to accurately reflect T cell distributions in synovial tissue at the time of surgical removal. Only half of the mlgG and αIL2-R beads from patient 1012 were immediately processed for RNA.

The remainder were cultured for 5 days in RPMI 1640, 5% FBS, 20% HL-1 (Ventrex Laboratories Inc., Portland, ME), 25mM HEPES, glutamine, antibiotics and 20% LAK

supernatant (Allegretta et al., Science, 247:718 (1990)), which is incorporated by reference herein, as a source of IL-2. RNA was extracted from cultures of the αIL2-R beads (1012lL2.d5), but not from the 1012mIgG sample as no viable cells were present at the end of the 5 day culture.

A T cell clone was derived from the Ficoll pellet of patient 1008. The cells in the pellet were cultured at 2 x 10⁶/ml in media without IL-2 for two weeks. Non-adherent cells from this culture were cloned by limiting dilution onto autologous synovial cell monolayers. A CD4+ T cell clone 1008.8 was obtained and adapted to culture by regular stimulation with autologous synovial monolayers for 3 days in media without IL-2 followed by a 4 day culture in medium with LAK

supernatant.

B. Lysis of Synovial Adherent Cells by 1008.8

Lysis of synovial adherent cells by 1008.8 was demonstrated as follows. Synovial cell monolayers were labeled as described in Stedman and Campbell, J. Immunol. Meth. 119:291 (1989), which is incorporated herein by reference, with ³⁵S for use as targets in CTL assays.

Cells were typsinized, washed and plated at 2000 cells per well of a 96-well round bottom microtiter plate.

1008.8 cells, cultured for 3 days prior to the assay with synovial adherent cells and medium containing LAK supernatant, were added to the targets at the indicated effector:target ratios. Cultures were incubated

overnight at 37°C, centrifuged at 300xg for 2 minutes and radioactivity in 50 μl of the supernatant quantified. Per cent specific lysis was calculated relative to detergent-lysed targets by standard formulas. This clone is cytotoxic for synovial adherent cell targets in CTL assays (Table 12).

C. PCR Amplification of TCR β-chain genes

TCR β-chain genes were amplified with several combinations of the primers shown in Figure 2. The vβl6mer primer is a degenerate Vβ primer (n=256) which is predicted to bind 85% of human TCR β-chain genes at all 16 residues and 95% at 15 residues. This primer has been used to amplify TCR β-chains from more than 25 different human T cell clones, lines or primary tissue

preparations. A spectrum of Vβ genes has been sequenced from these amplified DNAs, arguing against a significant bias of the primer for certain Vβ families. Thus, PCR amplification with the Vβ16mer primer facilitates

analysis of T cell populations for which a priori

knowledge of Vβ gene usage is unavailable. T cell receptor β-chain genes were amplified in two-stage amplification reactions with nested pairs of the primers shown in Figure 2. The primer sequences used in the polymerase chain reactions are listed in Table 13.

RNAs were reverse transcribed for 1 hour at 42°C with 40pmol of the Cβext primer in a 12 μl reaction using conditions described by Hart et al., The Lancet, p. 596 (1988), included by reference herein. Reactions were diluted with a master mix containing 40 pmols of the Vβ16mer primer, nucleotides and reaction buffer as above but without MgCl₂ to give a final Mg⁺² concentration of 3.6 mM. Samples were denatured for 15 minutes at 95°C, 1 unit of heat stable recombinant DNA polymerase (Cetus Corporation, Emeryville, CA, Ampli-taq™) was added and 20 cycles of PCR conducted. Each cycle consisted of a 1 min denaturation at 95°C, a two minute annealing step and a two minute extension at 72°C. The first two cycles were annealed at 37°C and 45°C, respectively, and the remainder at 50°C. One microliter aliquots of these stage I

reactions were added to 100 μl stage II amplification reactions (Cetus, Gene-Amp Kit™) containing 100 pmols of the Cβint primer and 100 pmols of the Vβ8, Vβ17 or 5'Cβ primers or 700 pmols of the Vβ16mer primer. Stage II amplifications were conducted as above with a 50°C annealing temperature and without the 37°C and 45°C ramping.

RNA samples from 1012lL2.d5 and 1008.8 cultures were amplified with the Vβ16mer and Cβext primers in stage I reactions and with the Vβ16mer and the Cβint primer in 35 cycle stage II reactions. Reaction

products, purified from low melting agarose gel slices with Gene Clean glass beads (Bio 1D1, San Diego, CA), were base denatured and sequenced from the Cβseq primer with T7 polymerase (Sequenase, United States Biochem,

Cleveland, OH). A predominate Vβ sequence, corresponding to a single Vβ17 rearrangement (Table 14), was clearly readable in the 1012IL2.d5 sample. Other, less frequent rearrangements were detected as faint, uninterpretable background bands in the sequencing gels. Culture of these 1012.IL2 beads in IL2-containing medium without added accessory cells or antigen is not expected to induce de novo activation of T cells. Thus, the

predominance of a single Vβ17 rearrangement in this sample reflects in vivo clonal expansion of Vβ17+ T cells in this patient. DNA sequence determination of TCR β- chain DNA amplified from the cytotoxic T cell clone, 1008.8, also revealed a Vβ17 rearrangement (Table 14). The presence of Vβ17 rearrangements in these two

different types of synovial T cell samples, derived from two separate RA patients, implicates Vβ17 bearing T cells in the pathogenesis of RA.

To determine whether or not Vβ17 rearrangements were present in the other magnetic bead RNA preparations, TCR β-chain genes were amplified with a Vβ17-specific primer in the second stage amplification after an initial amplification with the Vβ16mer. Vβ17 TCR DNA could be amplified from magnetic bead samples derived from the 4 patients examined. Ethidium bromide staining of

electrophoresed reaction products revealed greater Vβ17 amplification in some of the IL-2R+ samples than in the corresponding controls. Accordingly, the relative amounts of Vβ17 TCRs in each control and IL-2R+ sample were quantified by slot blot hybridization analysis as follows. RNAs from magnetic bead preps were amplified in the first stage with the Vβ16mer and Cβext primers and then reamplified for twenty cycles with the Cβint primer and each of the Vβ17, BB8 and 5'Cβ primers.

Amplification reactions were serially diluted in 20X SSC, denatured by boiling and chilled in an ice slurry.

Samples were loaded onto nitrocellulose membranes, hybridized to a human TCR β-chain constant region probe and washed with 0.1X SSC, 0.1% SDS at 56°C. Bound radioactivity was quantified by liquid scintillation spectroscopy and endpoint dilutions were those samples with fewer than 200 cpm bound. The amounts of product produced by forty total cycles with each of the

respective primer combinations falls in the linear portion of a product versus cycle number quantification curve. Amplifications with the 5'Cβ and Cβint primer pair were used to estimate the total β-chain amplified from each sample, providing benchmarks for normalizing the results of Vβ17 and Vβ8 quantification in the

respective IL-2R+ and control sample pairs (Table 15). The quantity of Vβ17 DNA amplified was increased in the IL-2R+ samples, relative to the control samples, in 3 of the 4 patients. The magnitude of the increase ranged from 5-fold in patient 1015 to 40-fold in patient 1014 (Table 15). This enrichment was not a product of the isolation procedure, since the quantity of Vβ8 DNA

amplified was increased in the IL-2R+ fraction only in patient 1015.

*In a follow-up study in which Vβ8 was sequenced, it was found that the enrichment of this Vβ was an artifact and that the value is ≤ 1. Vβ17 rearrangements from the IL-2R+ RNAs of the three patients showing enrichment were amplified with the Vβ17 and Cβint primer pair and the reaction products sequenced with the Cβseq primer. As was shown for sample 1012 IL-2.d5, 1014 and 1015 contained single sequences (Table 14), indicative of clonal expansion of Vβ17 T cells in vivo. In contrast, direct sequencing of the rearrangements amplified with the Vβ8 specific primer was not possible due to significant heterogeneity in the β- chain product. D. HLA-DR Analysis in Rheumatoid Arthritis Patients

HLA-DR analysis in rheumatoid arthritis

patients was performed as follows. DNA from each patient was prepared by boiling 10⁵ synovial cells in 200 μl dH₂O. Ten μl were amplified for 35 cycles in a 100 μl reaction (Cetus, Gene Amp Kit™) containing 100 pmols of each of the DRβ PCR primers shown in Table 16 as DRβl and DRβ3. One-tenth μl of this reaction was reamplified in 10 μls containing only the DRB2 primer and 17 pmol of α32P-dCTP as the sole source of dCTP for 10 cycles. Reactions were spiked with 200 μM dCTP and chased for 2 cycles. The resulting negative strand probes were hybridized to slot blots containing 10 pmol of the HLA-DR allele specific oligos (positive strands) using conditions previously described by Amar et al., J. Immunol. 138:1947 (1987), which is incorporated herein by reference. The slots were washed twice for 20 minutes with

tetramethylammoniumchloride (Wood et al., Proc. Natl.

Acad. Sci. USA 82:1585 (1985)) which is incorporated herein by reference) at 65-68°C and exposed to X-ray film.

Each of the patients in this study possessed at least one allele of the HLA-DR genes, DR4w4, DR1, DR4w14 or DR4w15, that are known to predispose for RA (Table 16). Also shown in Table 16 are HLA-DR allele specific oligonucleotides.

* The dashes refer to amino acids in common with DR1, DR4w14, DR4w15

T cell receptors containing Vβ17 or fragments thereof which are immunogenic or can be made immunogenic can be used to immunize human subjects by methods

demonstrated by Example VIII. Such immunizations can

result in an effective immune response. EXAMPLE XI

Synovial tissue specimens were obtained from proven RA patients undergoing joint replacement surgery. HLA-DR analysis was conducted as described in Example IX(D).

A. Polymerase chain reaction (PCR) amplification of T cell receptor β-chain genes

T cell receptor β-chain genes were amplified in two-stage amplification reactions with nested pairs of HPLC-purified oligonucleotide primers (Midland Certified Reagents, Midland, TX.) shown in Figure 3. RNAs were reverse transcribed (1 hour, 42°C) with the Cβext primer (40 pmol) in 12 μl of reaction buffer (Hart et al.,

Lancet ii:596-599 (1988)). Reactions were diluted with a master mix (8 μl) containing the Vβcons primer (40 pmol), nucleotides and reaction buffer minus MgCl₂ (final Mg⁺² concentration= 3.6 mM). Samples were denatured (15 minutes, 95°C) and 20 cycles of PCR were conducted using Tag polymerase (1 unit, Cetus Ampli-taq). Each cycle consisted of a 1 minute denaturation at 95°C, a two minute annealing step and a two minute extension at 72°C. The first two cycles were annealed at 37°C and 45°C and the remainder at 50°C. One microliter aliquots of these Stage I reactions were added to 100 μl Stage II amplification reactions (Cetus Gene-Amp Kit) containing the Cβint primer (100 pmol) and either the Vβ8, Vβ12, Vβ17, or 5'Cβ primers (100 pmol) or the Vβcons primer (100-700 pmol). Stage II amplifications were conducted for the indicated number of cycles with a 50°C annealing temperature and without the 37°C and 45ºC ramping.

B. Vβ17 T cells are cytotoxic for synovial adherent cells

Two parallel cultures were established from single cell suspensions, derived by enzymatic digestion as described below from the synovial tissue of patient 1008. The first, a bulk culture of total synovial cells, was plated at 2 x 10⁶/ml in RPMI 1640, 5% FBS, 20% HL-1 (Ventrex Laboratories, Portland, ME), 25mM HEPES,

glutamine and antibiotics. The culture was grown

undisturbed for two weeks in the absence of exogenous antigen or growth factors. The second, a synovial cell monolayer culture, was initiated as above and regularly trypsinized and passaged during this two-week period.

Monolayer cells were plated at 5 x 10⁴ cells/well in flat- bottomed, 96-well microtiter plates and cultured

overnight. Non-adherent cells from the total synovial cell culture were then plated at 10 cells/well onto the monolayers. Wells positive for T cell growth were expanded and adapted to culture by regular stimulation with autologous synovial monolayers for 3 days in media without IL-2 followed by a 4 day culture in medium

containing 20% supernatant from lymphokine activated killer (LAK) cells, a source of IL-2 (Allegretta et al., Science 247:718-721 (1990)). Later passages were adapted to weekly stimulation with allogeneic PBLs and anti-CD3 antibody (Coulter Immunology, Hialeah, FL) in place of synovial cell monolayers. Cytotoxicity assays were conducted as described

Stedman et al., Immunol. Meth. 119:291-294 (1989), incorporated herein by reference, using ³⁵S-labeled synovial monolayer cells or EBV-transformed B cells as targets. Target cells were labeled overnight,

trypsinized (adherent cells only), washed and plated at 2000 cells per well in 96-well round bottom microtiter plates. T cells were activated with allogenic PBLs and anti-CD3 antibody in medium containing LAK supernatant for 7 days prior to the assay and added to the targets at the indicated effector:target ratios. Cultures were incubated overnight at 37°C, centrifuged at 300xg for 2 minutes and radioactivity in 50 μl of the supernatant quantified.

The per cent specific lysis was calculated relative to detergent-lysed targets as described in Townsend et al., Cell 44:959-968 (1986), incorporated herein by reference. At effector:target ratios of 1.0, 2.5 and 5.0, the percent lysis of synovial adherent cells by 1008.8 T cells was approximately 4%, 14% and 33%, respectively. By comparison, at the same effector:target ratios, 1008.8 T cells had no demonstrated effect on EBV- transformed B cell targets. Similarly, at the same effector:target ratios, the percent lysis of synovial adherent cells by MS3 cells was approximately 0%, 0% and 3%, respectively. T cells were isolated from the synovial tissue of patient 1008 by co-cultivation with synovial cell monolayers. Since the antigens recognized by pathogenic T cells in RA are unknown, synovial cell monolayers were used as stimulators of in vitro synovial T cell growth. The relevant target cells were assumed to be present in a bulk adherent cell culture from diseased synovium. Of 192 microwell co-cultures plated, 7 were positive for T cell growth within 10-14 days. T cells were expanded and maintained in vitro by alternately stimulating with synovial cell monolayers and medium containing LAK supernatant. Flow cytometry revealed that each of these seven cultures was 100% CD4+. One culture designated 1008.8, grew especially vigorously and, as assessed microscopically promoted the destruction of the monolayer cells. CTL assays confirmed the cytotoxicity of 1008.8 for these monolayer targets as discussed above.

Cytolysis was specific for synovial cell targets, as no lysis of autologous Epstein-Barr virus-transformed B cells was demonstrable. Neither of the targets was lysed by a CD4+, myelin basic protein-reactive human CTL clone. MS3, excluding the possibility that the monolayer cells were susceptible to lysis by any activated T cell. In repeated assays with 1008.8, specific lysis never

exceeded 30-35%, suggesting that the relevant synovial target cell comprises approximately that proportion of the total monolayer culture, which based on morphology is a mixture of multiple cell types.

The T cell receptor (TCR) β-chain gene of

1008.8 was amplified by the polymerase chain reaction (PCR) as described in Mullis et al., Meth. Enzym.

155:335-350 (1987), incorporated herein by reference, and its DNA sequence was determined. Amplification was accomplished using a consensus Vβ primer (Vβσons), a degenerate 16 nucleotide primer (n=256) which is

homologous at all 16 residues with 78%, and at 15 of 16 residues with 98%, of known human TCR Vβ genes, which have been compiled by Kimura et al., J. Immol. 17:375-383 (1987). This primer was designed to amplify β-chain rearrangements containing any of the known Vβ genes, thus allowing analysis of T cell clones for which a priori knowledge of Vβ gene usage is unavailable. Sequencing of the Vβcons-amplified β-chain gene of 1008.8 revealed a single Vβ17-Jβl.l rearrangement as shown in Table 16.

Subsequently, the remaining six cultures derived from patient 1008 were analyzed by PCR amplification using a Vβ17-specific primer shown in Table 13. TCR genes from three of those six were amplifiable, indicating that, of the 7 T cell cultures obtained from the synovium of this patient, 4 had rearranged and expressed the Vβ17 gene. C. Vβl7 T cells are enriched among activated T cells in RA synovium

Synovial tissue was digested with agitation for 4 hrs at 37°C in RPMI-1640 and 10% fetal bovine serum (FBS) containing 4 mg/ml collagenase (Worthington

Biochemicals, Freehold, NJ) and 0.15 mg/ml DNAse (Sigma Chemical, St. Louis, MO). Digests were passed through an 80-mesh screen and single cells were collected from the interface of Ficoll density gradients, washed and

incubated at 10⁶/ml for 30 min at 0°C with 5 μg/ml control mouse IgG (Coulter) in PBS containing 2% BSA. Cells were washed 3X and incubated for 30 min at 0°C with magnetic beads conjugated to goat anti-mcuse IgG (Advanced

Magnetics, Cambridge, MA). After magnetic removal of the beads, the remaining cells were incubated 30 minutes at 0°C with 5 μg/ml mouse anti-human IL-2R (Coulter), washed and selected with magnetic beads as above. Cell-coated beads from the mlgG preadsorption and the IL-2R antibody selection were washed 3X, immediately resuspended in acidified-guanidinium-phenol-chloroform and RNA was prepared as described in Chomczynski et al., Anal.

Biochem. 162:156-159 (1987).

RNAs from magnetic bead preparations were reverse transcribed and amplified for 20 cycles in stage I reactions with the Vβcons and Cβext primers. One μl of each reaction was reamplified for 20 cycles in individual stage II reactions containing the Cβint primer in

conjunction with the Vβ17, Vβ8, Vβ12 and 5'Cβ primers. Aliquots of each reaction were diluted in 20X SSC, denatured by boiling and chilled in an ice slurry.

Samples were loaded onto nitrocellulose membranes, hybridized to a human TCR β-chain constant region probe and washed with 0.1X SSC, 0.1% SDS at 56°C. Bound

radioactivity was quantified by liquid scintillation spectroscopy. The amounts of product produced by 40 total cycles with each of the respective primer

combinations falls in the linear portion of a product versus cycle number quantification curve. Values shown in Table 17 below reflect the relative increase or

decrease for the specific Vβs in the IL2-R+ versus mlgG controls calculated according to the formula:

Next, the presence of Vβ17 T cells in the synovial tissue of other RA patients was determined. Since the rheumatoid synovium contains a mixture of activated and non-activated T cells, the activated T cells were identified as the most relevant for the initiation and perpetuation of the disease pathogenesis. Thus, activated T cells from single-cell suspensions of synovial tissue were selected using magnetic beads and antibodies reactive with the human interleukin-2 receptor (IL-2R). Cell suspensions from each patient were

pretreated with an isotype-matched mouse IgG (mlgG) and magnetic beads to control for non-specific adsorption. RNAs were directly extracted from cells in the IL-2R+ and control samples without in vitro culture and, therefore, are expected to accurately reflect T cell distributions in synovial tissue at the time of surgical removal.

The initial PCR amplifications of these

magnetic bead RNA preparations revealed greater amounts of Vβ17 PCR product in the IL2-R+ samples than in the corresponding mlgG controls. The presence of TCR mRNA in the mlgG samples indicates that T cells non-specifically adhered to the magnetic beads. The apparent increase in the IL2-R+ sample suggests that the activated T cell compartment contained more Vβ17 T cells than the

unselected synovial T cell compartment. Thus, a

quantitative PCR analysis was used to formally examine the relative proportions of Vβ17 T cells in the IL-2R+ and mlgG control samples (Table 17). Magnetic bead RNAs were reverse transcribed, preamplified with Vβcon and Cβext and reamplified in separate reactions with a constant region primer (Cβint) and each of the Vβ- specific primers, Vβ17, Vβ8 and Vβ12. The second stage amplification was also performed with two Cβ primers (5'Cβ and Cβint) in order to estimate the total β-chain present in each sample and to provide benchmarks for normalizing the results of specific Vβ quantification in the respective IL-2R+ and control sample pairs. The proportion of Vβ17 DNA, relative to total Cβ, was

increased in the IL-2R+ samples over that found in the mlgG control samples for each of the 5 patients examined. This increase was observed in multiple analyses and the means are shown in Table 17. Enrichment was not a product of the isolation procedure, since the quantity of Vβ8 or Vβ12 DNA amplified was not significantly increased in the IL-2R+ fraction of any of the patients. Thus, activated T cells in the rheumatoid synovium do not represent a cross section of all possible Vβ families, but preferentially contain Vβ17 T cells, and possibly other Vβ families which were not quantified in this analysis. D. Synovial Vβ17 T cells display limited heterogeneity

RNAs were reverse transcribed and amplified with the Vβcons and Cβext primers in 20 cycle stage I reactions and with the Cβint and Vβcons (1008.8) or Vβ8, VT312, and Vβ17 primers (magnetic bead RNA preparations) in 35 cycle stage II reactions. Double stranded reaction products were electrophoresed in 2% Nu-Sieve agarose gels. After purification from gel slices with Gene Clean (Bio 101, San Diego, CA), samples were base denatured and either directly sequenced or cloned into plasmid for sequencing of multiple independent rearrangements. In all cases, samples were sequenced from the Cβseq primer (Figure 2) with T7 polymerase (Sequenase, United States Biochemicals, Cleveland, OH). Vβ17 rearrangements present in the IL-2R+ RNAs were amplified from the Vβcons and Cβext preamplification with the Vβ17 and Cβint primer pair and the reaction products sequenced. PCR products from patients 1014 and 1015 were directly sequenced and the results obtained were consistent with the presence of a single Vβ17 rearrangement in each of these amplified samples (Table 17). The PCR product of patient 1013 was cloned into plasmid and the thirteen isolates that were sequenced contained identical Vβ17 rearrangements. Sequencing of plasmid clones from patient 1012 revealed the presence of two dominant rearrangements (5 isolates of each) and a single isolate of another. Thus, the Vβ17 repertoire in the RA synovium is of limited heterogeneity, indicative of clonal or oligoclonal expansion of Vβ17 T cells in vivo. This is in contrast to the Vβ8 and Vβ12

repertoires, from these same synovial preparations, which showed significant heterogeneity. None of the PCR- amplified Vβ8 and Vβ12 samples analyzed were directly sequenceable and plasmid cloning of Vβ8 rearrangements from patient 1012 revealed 4 different sequences in 5 clones analyzed. Vβ17 rearrangements in PBLs from patient 1012 also revealed greater diversity than that seen in the synovial IL-2R+ sample. RNA from a 3 day culture of PHA/PMA stimulated 1012 PBLs was amplified with the Vβ17 primer, as for the synovial sample, and the products cloned into plasmid. Nine different rearrangements, none of which corresponded to those present in the 1012 synovium, were found in the 10 clones that were

sequenced. Thus, the restricted heterogeneity of Vβ17 rearrangements in the activated synovial T cell

population of patient 1012, as well as the other patients examined, likely results, not from random T cell

trafficking, but from the selective expansion of those Vβ17-bearing T cells in the diseased tissue. EXAMPLE XII

A. Detection of TCR β-chain transcripts in synovial T cells using Vβ-specific PCR-amplification

T cell receptor β-chain genes were amplified in two-stage reactions with individual Vβ-specific primers as described in Wucherfpennig et al., Science, 248:1016- 1019 (1990), incorporated herein by reference, and nested Cβ primers. RNAs were reverse transcribed (1 hour, 42°C) with the Cβext primer (40 pmol) in 12 μl of reaction buffer (Hart et al., supra). Reactions were diluted with a master mix (8 μl) containing nucleotides, reaction buffer minus MgCl₂ (final Mg⁺² concentration = 3.6 mM) and Taq polymerase, apportioned among 19 tubes containing the individual Vβ primers. Samples were denatured (15 minutes, 95°C) and 20 cycles of PCR were conducted. Each cycle consisted of a one minute denaturation at 95°C, a two minute annealing step and a two minute extension at 72°C. The first two cycles were annealed at 37°C and 45°C and the remainder at 50°C. One microliter aliquots of these Stage I reactions were added to 100 μl Stage II amplification reactions (Cetus Gene-Amp Kit) containing the Cβint primer (100 pmol) and 100 pmol of the Vβ primers used in the corresponding preamplifications.

Stage II amplifications were conducted for 20 cycles with a 50°C annealing temperature and without the 37ºC and 45°C ramping. Five μl of each reaction was

electrophoresed in 2% agarose gels, transferred to nitrocellulose and hybridized to a ³²P-labeled β-chain constant region probe. Blots were exposed to X-ray film and scored for the presence or absence of Vβ-specific amplification.

B. Vβ17 , Vβ14 , VB9 and Vβ3 transcripts are common among activated synovial T cells

To ensure that the prevalence of Vβ17 rearrangements in the earlier studies did not result from an amplification bias of the Vβ16mer primer, and to assess the presence of other Vβ genes in the synovium, TCR transcripts in the IL2-R+ samples were analyzed with a panel of 19 PCR primers , specific for known human Vβ gene families (Table 18). The Vβcons primer of Example XI is equivalent to the Vβ1όmer primer. The number of Vβ genes detectable in these samples was variable, ranging from two to twelve. Vβ17 was found in four of the five patients, confirming the previous analyses using the Vβ16mer primer. Vβ14 transcripts also were found in four of the five patients and Vβ3 and Vβ9 transcripts were detectable in three of the five samples. Thus, T cells bearing these 3 Vβ polypeptides may also contribute to synovial inflammation.

EXAMPLE XIII

Vaccination Against EAE with a CDR4 Peptide of Vβ8.2

Rats were immunized with 100 μg of a synthetic peptide containing the sequence VPNGYKVSRPSQGDFFLTL found in the fourth hypervariable or CDR4 region of the rat TCR β chain identified as Vβ8.2. The peptide, dissolved in saline, was emulsified in an equal volume of complete Freund's adjuvant (CFA) containing 10 mg/ul of

mycobacterial tuberculosis. The animals were challenged 30 days later with 50 μg of guinea pig myelin basic protein in CFA. The CDR4 peptide vaccination resulted in a reduced incidence of disease as well as a reduced severity as measured both clinically and histologically as shown in Table 19. The histology was performed essentially as described in Hughes & Powell, J.

Neuropath. Exp. Neurol. 43:154 (1984), which is

incorporated herein by reference. TABLE 19

Protection from EAE in the Rat by Vaccination

with a TCR-CDR4 Peptide¹

Vaccination Challenge #Clinical Signs Mean Max Mean Mean #With Histology Mean

#Tested Severity² Onset Duration #Tested Histology

Score³

¹ rat CDR4 peptide sequence: VPNGYKVSRPSQGDFFLTL

² graded on a 3-point scale as described in Example I

3 graded on a 4-point scale

EXAMPLE XIV

Cross-Species Efficacy of Peptides

Synthetic peptides representing amino acid sequences from the CDR4 region of both rat Bβ8.2 T cells and human Vβ17 T cells can protect rats from the

development of EAE. The rat Vβ8.2 CDR4 peptide has the sequence VPNGYKVSRPSQGDFFLTL (residues 61-79) and the human Vβ17 CDR4 peptide has the sequence EKKESFPLTVT (residue 68-79). Lewis rats were immunized with emulsions containing equal volumes of peptide in saline and of IFA intradermally in 0.1 ml doses divided equally between the two hind footpads (0.05 ml/footpad). Each animal of Groups A and B received 100μg of peptide. Group C, 50 μg of peptide. Group D received IFA alone and Group E, no treatment. After 30 days, each animal was challenged with 50 μg guinea pig myelin basic protein (GP-MBP) in saline emulsified with an equal volume of complete

Freund's adjuvant containing mycobacteria tuberculosis (10 mg/ml). Each animal received 0.1 ml in two doses divided equally between the front footpads (0.05

ml/footpad) and observed for clinical signs of disease graded on a 3 point scale consisting of 1) loss of motor control of the tail, 2) hind leg weakness and 3) hind leg paralysis.

Table 20 shows that animals receiving either human CDR4 (Group A) or rat CDR4 (Group B) had mild or nonclinical signs of disease after challenge whereas animals receiving control peptide (C), IFA (D) or no treatment (E) had a normal disease course. The results establish that CDR4 peptides can prevent EAE and that a peptide with the sequence EKKESFPLTVT from human Vβ17 can protect from autoimmune disease in a different species (rat).

Table 21 shows that animals receiving human CDR4 peptides (Group A) from Vβ17 or Vβ14 had mild or no disease and 3 of 5 animals receiving a similar Vβ3 peptide had no disease. In contrast, animals receiving IFA (Group D) or no pretreatment (Group E) had a normal disease course. These results establish that the human Vβ14 CDR4 peptide having the sequence KEKRNFPLILE can also protect from autoimmune disease in a different species (rat) and that a human Vβ3 CDR4 peptide having the sequence REKKERFSLILE also has a protective effect.

o

Although the invention has been described with reference to the presently-preferred embodiment, it should be understood that various modifications can be made without departing from the spirit of the invention. Accordingly, the invention is limited only by the following claims.

Claims

We Claim:

1. A vaccine for preventing or treating a T cell mediated pathology in a vertebrate, comprising an immunogenically effective amount of a substantially pure T cell receptor or an immunogenic fragment thereof corresponding to a T cell receptor present on the surface of T cells mediating said pathology, and a

pharmaceutically acceptable medium.

2. The vaccine of claim 1, wherein said immunogenic fragment comprises substantially the amino acid sequence of a β-chain variable region.

3. The vaccine of claim 2, wherein said immunogenic fragment comprises substantially the amino acid sequence of the CDR2 region of said β-chain variable region.

4. The vaccine of claim 3, wherein said CDR2 region comprises substantially the sequence

DPGLGLRLIYFSYDVKMKEKG or an immunogenic portion thereof.

5. The vaccine of claim 3, wherein said CDR2 region comprises substantially the sequence

DPGLGLRQIYYSMNVEVTDKG or an immunogenic portion thereof.

6. The vaccine of claim 3, wherein said CDR2 region comprises substantially the sequence

DPGQGLRLIYYSQIVNKFQKG or an immunogenic portion thereof.

7. The vaccine of claim 3, wherein said pathology is rheumatoid arthritis.

8. The vaccine of claim 7, wherein said amino acid sequence comprises substantially the sequence of Vβ3, Vβ14 or Vβ17.

9. The vaccine of claim 3, wherein said pathology is multiple sclerosis.

10. The vaccine of claim 1, wherein said fragment comprises substantially the amino acid sequence of a β- chain VDJ region of said T cell receptor.

11. The vaccine of claim 10, wherein said pathology is rheumatoid arthritis or multiple sclerosis.

12. The vaccine of claim 11, wherein said β-chain VDJ region comprises substantially the sequence ASSLGGAVSYN.

13. The vaccine of claim 11, wherein said β-chain VDJ region comprises substantially the sequence ASSLGGEETQYF, ASSLGGFETQYF or ASSLGGTEAFF.

14. The vaccine of claim 11, wherein said β-chain VDJ region comprises substantially the sequence CAIGSNTE.

15. The vaccine of claim 11, wherein said β-chain VDJ region comprises substantially the sequence SDGQGGNE.

16. A method of diagnosing or predicting

susceptibility to a T cell mediated pathology in an individual comprising detecting T cells having a T cell receptor associated with said pathology in a sample from the individual, the presence of abnormal expression of T cells containing said T cell receptor indicating said pathology or susceptibility to said pathology.

17. A method of preventing or treating an individual exhibiting or at risk of exhibiting a T cell-mediated pathology, comprising administering to the individual the vaccine of claim 1.

18. A method of preventing or treating a T cell mediated pathology in an individual, comprising

contacting Vβ-containing T cells in the individual with an effective amount of a cytotoxic or cytostatic agent specifically reactive with said Vβ-containing T cells to inhibit the activity of said Vβ-containing T cells.

19. The method of claim 18, wherein said agent is an antibody attached to a moiety selected from the group consisting of radioactive moieties, chemotherapeutic moieties and chemotoxic moieties.

20. The method of claim 19, wherein said antibody is specifically reactive with the variable region or V(D)J region of said Vβ-containing T cells.

21. The method of claim 18, wherein said pathology is rheumatoid arthritis or multiple sclerosis.

22. A method of preventing or treating a T cell- mediated pathology in an individual, comprising

administering to the individual a nucleic acid encoding a T cell receptor or an immunogenic fragment thereof in a form capable of being expressed in said individual.

23. The method of claim 22, wherein said nucleic acid is DNA.

24. The method of claim 22, wherein said nucleic acid is RNA.

25. The method of claim 22, wherein said nucleic acid is administered into a muscle of the individual.

26. The method of claim 22, wherein said immunogenic fragment comprises the amino acid sequence of a β-chain V(D)J region of said T cell receptor.

27. The method of claim 26, wherein an amino acid sequence is substantially the amino acid sequence of Vβ3, Vβ4, Vβ12, Vβ14 or Vβ17.

28. The method of claim 27, wherein said pathology is arthritis or multiple sclerosis.

29. The method of claim 22, wherein said immunogenic fragment comprises the amino acid sequence of a variable region of said T cell receptor.

30. The method of claim 29, wherein said variable region is a β-chain variable region.

31. The method of claim 30, wherein said amino acid sequence comprises substantially the sequence of the CDR2 region of said β-chain variable region.

32. The method of claim 30, wherein said amino acid sequence comprises substantially the sequence of the superantigen binding region of said β-chain variable region.

33. The method of claim 30, wherein said β-chain variable region is Vβ3, Vβ4, Vβ12, Vβ14 or Vβ17.

34. The method of claim 29, wherein said pathology is rheumatoid arthritis or multiple sclerosis.

35. A vector containing the nucleic acid of claim 22.

36. A composition comprising the vector of claim 22 and a pharmaceutically acceptable medium.