COMPOSITIONS WITH ANTI-PROLIFERATIVE ACTIVITIES AND METHODS FOR USE THEREOF Background of the Invention Diabetes mellitus is the most commonly occurring endocrine disease. Although an accurate frequency of the disease is not available because of differing standards for diagnosis, millions of Americans suffer from some form of the disease. There are several distinct forms or classes of diabetes, including both primary (no associated disease is present) and secondary (some other identifiable cause allows for development of diabetes) forms. Primary diabetes is classified into Type I and Type II, while secondary forms are classified by their proximal cause such as pancreatic disease- induced, hormonal abnormalities-induced, drug-induced, etc.

Human Type I diabetes mellitus (IDDM), or insulin- dependent diabetes, is a disorder of immunity characterized by the selective destruction of insulin-producing pancreatic beta cells (-cells). These P-cells are the targets of autoreactive T lymphocytes which are the primary mediators of the disorder. Typical IDDM is characterized by destruction of more than 90% of the (3-cells. However, the precise cellular mechanisms and immunological mediators involved in IDDM have not been elucidated.

Research into the causes and treatments for IDDM has often employed an animal model known as the non-obese diabetic mouse (NOD), an animal model which exhibits an immunological disorder similar to that seen in humans (Hanafusa et al. 1994.

Diabetes Res. Clin. Pract. 24: S307-S311). This model is currently accepted as the principal animal model and the optimal tool for examining mechanisms underlying the disease, as well as for preclinical screening of potential treatments for IDDM. Using this model, therapeutic strategies that target T cells have been successful in preventing IDDM (Makino

et al. 1980. Exp. Anim. 29: 1-13). These strategies have included neonatal thymectomy, cyclosporin, and infusion of anti-pan T cell, anti-CD4, or anti-CD25 monoclonal antibodies (Ogawa et al. 1985. Biomed. Res. 6: 103; Mori et al. 1986.

Diabetologia 29: 244-247 ; Tarui et al. 1986. Insulitis and Type I Diabetes, Lessons from the NOD Mouse. Academic Press: Tokyo, p. 143; Shizuru et al. 1988. Science 240: 659-662 ; Kelley et al. 1988. J. Immunol. 140: 59-61). In addition, individual T cell clones have been isolated from NOD mice that proliferate in response to syngeneic islets and induce insulitis but not IDDM, whereas other T cell clones can severely damage mouse islets when co-mixed with islet cell transplants derived from NOD mice (Reich et al. 1989. Nature 341: 326-328; Haskins et al. 1989. Proc. Natl. Acad. Sci. USA 86: 8000-8004).

Studies have been performed examining the role of a particular CD8+ T cell clone, IS-2.15, in the promotion or prevention of diabetes (Pankewycz et al. 1991. Eur. J.

Immunol. 21: 873-879). The T cell clone IS-2.15 was isolated directly from the islets of disease-resistant NOD male mice.

Although both male and female NOD mice develop a strong autoimmune reaction against pancreatic (3-cells, in males the process aborts by an unknown mechanism and does not lead to total destruction of insulin-producing ß-cells. in vitro studies, IS-2.15 cells exhibited an anergic phenotype as they failed to kill any tested target or produce any known cytokine when stimulated with anti-CD3 monoclonal antibodies. In in vivo studies, IS-2. 15 cells prevented the expression of overt diabetes. Spleen cells harvested from older diabetic mice were transferred into young, female, irradiated NOD mice which led to diabetes within 28 days. In contrast to a 100% incidence of diabetes in control mice, only 60% of those that received IS-2.15 cells became diabetic. Young male NOD were also tested by administering cyclophosphamide, which leads to diabetes within 10 days. In control animals, 75% developed

diabetes while only 18% of mice that received IS-2.15 cells became diabetic. These data indicated that IS-2.15 cells regulated the immune response and were able to prevent the autoimmune process.

Later studies with IS-2.15 cells demonstrated that these cells produced a soluble immunoregulatory factor that was able to block T cell proliferation in vitro (Pankewycz et al. 1992.

Eur. J. Immunol. 22: 2017-2023). This unidentified T cell- derived substance was not cytotoxic and did not prevent T cell activation.

It has now been found that a T cell clone of NOD mice produces novel lymphocyte-derived peptides with anti- proliferative activity. Thus, it is believed that these peptides or active fragments or variants thereof and molecular mimics of these peptides can be used to treat or prevent autoimmune disorders, transplant rejection and immunoproliferative disorders. Further, it is believed that compounds which inhibit the activity of these peptides can be used in the treatment of immunodeficient conditions.

Summary of the Invention An object of the present invention is to provide peptides derived from a T cell clone which are useful in preventing or treating autoimmune disorders including diabetes and transplant rejection. In a preferred embodiment, these peptides comprise P1, MCACVCPSACASVSLKNNLLCDFLWSFCSGYSAAPQ (SEQ ID NO : 1) or P2, MTSLIKAFLDNRCFLSAPVSYSLVGFFSDCLPSYSPGGRGA (SEQ ID NO : 2) or active fragments or variants thereof.

Another object of the present invention is to provide methods for inhibiting T cell proliferation and for treating or preventing autoimmune disorders via these peptides or active fragments or variants thereof or via molecular mimics of these peptides.

Yet another object of the present invention is to provide methods for identifying small molecule inhibitors of

these peptides for use in treatment of immunodeficient disorders.

Detailed Description of the Invention With the identification that a T cell clone, IS-2.15, from NOD mice produces a substance or substances that can prevent diabetes in this preclinical model of human IDDM, studies were done to isolate and characterize the novel substance or substances. The novel anti-diabetic compounds have been identified in terms of their amino acid sequence and their biological activity as two distinct peptides, referred to herein as P1 (SEQ ID NO : 1) and P2 (SEQ ID NO : 2). The nucleotide sequence of the DNA segment 722415 (SEQ ID NO : 3) encodes for PI (SEQ ID NO : 1) and P2 (SEQ ID NO : 2). P1 (SEQ ID NO : 1) is encoded by an open reading frame at positions 172 through 295 of SEQ ID NO : 3. P2 (SEQ ID NO : 2) is encoded by an open reading frame at positions 404 through 511 of SEQ ID NO : 3. A variant of the P1 peptide, namely Plb (SEQ ID NO : 5), has also been identified, and is believed to share the immunological properties of P1 due to their significant amino acid homology. The nucleotide sequence encoding for Plb (SEQ ID NO : 5) is depicted in SEQ ID NO : 4. Plb (SEQ ID NO : 5) is encoded by an open reading frame at positions 343 through 520 of SEQ ID NO : 4. The isolation of these immunosuppressive peptides opens new areas of research for the design of human therapeutics for treatment of IDDM.

Initial studies into the identity of the novel immunoregulatory substance or substances produced by IS-2.15 cells were performed to eliminate factors already known to mediate T cell function. One potential known factor was transforming growth factor beta (TGFß). Using the mink lung epithelial cell line bioassay (CCL-64) for active TGFß, however, it was found that IS-2.15 T cells failed to produce active TGFß when cultured alone or following stimulation with anti-CD3 monoclonal antibodies, syngeneic spleen cells, or

allogenic spleen cells. Therefore, the soluble immunoregulatory factor was not related to TGFß. In similar experiments, it was found that IS-2.15 T cells also failed to make detectable quantities of interleukins 2,4 and 10 as well as gamma interferon.

Accordingly, IS-2.15 cells were cultured alone in either serum-free medium or in the presence of 10% fetal calf serum at a concentration of 105 cells/200 ml. A soluble factor was isolated using siliconized tubes and non-tissue culture treated well plates. This soluble factor was added to mixed lymphocyte cells in culture and found to inhibit T cell proliferation by more than 50%.

Experiments were then performed to examine conditions which neutralized the IS-2. 15-derived soluble factor. The inhibitory activity of the soluble factor on T cell proliferation in mixed lymphocyte cell cultures was not affected by the addition of neutralizing concentrations of monoclonal antibodies against known mediators such as TGF, IL4 and IL10. Therefore, the IS-2. 15-derived soluble factor was not one of the known cytokines that may affect T cell proliferation in mixed lymphocyte cell cultures. In addition, IS-2.15-derived soluble factor that was boiled before addition to the mixed lymphocyte cell cultures still inhibited T cell proliferation by 72%, indicating that the factor resists neutralization by boiling and was a small molecular weight substance.

The IS-2.15 cell supernatant was then fractionated into >30 kDa and <30 kDa fractions using size selective centrifugation techniques. The entire immunoregulatory activity was found in the <30 kDa fraction, with no activity at all noted in the >30 kDa fraction. This activity was confirmed in the mixed lymphocyte cell culture assay. Further size selection fractionation of the <30 kDa fraction and testing in vitro demonstrated that activity was in the <10 kDa fraction. Passage of the supernatant through a 3 kDa size

exclusion membrane led to a decrease in activity. These results confirmed that the IS-2. 15-derived factor was a small molecule with a molecular weight close to, but greater than 3 kDa, and less than 10 kDa.

Experiments were performed to examine the biochemical nature of the IS-2. 15-derived factor. The isolated factor supernatant was digested with trypsin. Following digestion, the supernatant was size fractionated into the <30 kDa portion which removed the trypsin. The digested <30 kDa fraction contained an equal or even greater immunoregulatory activity in terms of T cell proliferation when compared to the original supernatant. When digested with Protease K, all immunoregulatory activity was lost. These data indicated that the IS-2. 15-derived factor was a small protein, a peptide or peptides.

Genes in IS-2.15 encoding the immunoregulatory peptide or peptides were then isolated. The mRNA content from IS-2.15 cells was first compared with that of control CD4+ and CD8+ T cell clones derived from NOD mice. Unlike IS-2.15 T cells, control T cell clones were not anergic and proliferated in response to syngeneic or allogenic stimulator cells. The CD8+ T cell clone showed potent cytotoxic activity. Differential Display was carried out using a commercial kit (Display Systems, Vista, CA). DNA segments unique to IS-2.15 cells were sequenced and compared to the established DNA database, GenBank. Initial experiments yielded over 80 IS-2.15-specific gene fragments, 40 of which were not found in the known GenBank database. These 40 fragments were used to obtain longer cloned genetic sequences from the established database.

The 40 genes were then cloned into CMV-driven mammalian expression vectors and transfected into a human lung adenocarcinoma cell line. Supernatants obtained from the transfected cells and from control cells transfected with vector alone were tested for in vitro immunoregulatory activity in the mixed lymphocyte cell culture assay.

Only one DNA sequence, 722415, cloned into the CMV expression vector repeatedly conferred immunosuppressive activity into supernatant of transfected adenocarcinoma cells.

The 722415-derived supernatant consistently blocked T cell proliferation in the mixed lymphocyte assay. Such activity was not observed when vector alone or other DNA clones were tested.

Sequencing of the 722415 DNA (SEQ ID NO : 3) segment revealed two potential open reading frames that encoded for peptide products. Neither the genetic sequences nor the peptides encoded had homology with known genetic or peptide databases. The two peptides identified are referred to herein as P1 and P2. P1 comprises the amino acid sequence MCACVCPSACASVSLKNNLLCDFLWSFCSGYSAAPQ (SEQ ID NO: 1). P2 comprises the amino acid sequence MTSLIKAFLDNRCFLSAPVSYSLVGFFSDCLPSYSPGGRGA (SEQ ID NO: 2).

PI is an unusual peptide in that it contains 6 cysteines out of a total of 36 amino acids. Using computer-based 3D- protein modeling, the protein structure of P1 was shown to be similar to the structure of beta-defensin proteins. The spacing of the six cysteine amino acids and the intervening residues are markedly different from the known canonical sequence of beta-defensin molecules. Thus, while PI shares no genetic or protein sequence similarity to beta-defensins, when this peptide assumes its normal protein folding structure, it strongly resembles this group of strong biologically active proteins. Since beta-defensins are potent mediators of anti-proliferative activity for immune cells, the striking similarity of P1 with these molecules is indicative of Pl also being a mediator of the immunoregulatory activity of IS-2.15 T cells.

Three separate experiments were performed to confirm the immunoregulatory activity of P1. Further, the gene segment encoding PI exclusively was isolated using PCR techniques and was cloned into a mammalian CMV driven protein expression

vector. Thus, all extraneous sequences contained within the EST fragment were eliminated. This CMV driven expression vector was transfected into a human lung adenocarcinoma cell line. At a dilution of 1: 6, supernatants from P1 expressing cells inhibited mixed lymphocyte cultures by greater than 50%.

In contrast, supernatants obtained from control vector transfected cells were inactive.

The specific P1 sequence was also cloned into an appropriate vector for in vitro translation using the rabbit reticulocyte lysate method. The presence of P1 in vector containing lysates was confirmed by SDS-PAGE gel analysis.

P1 containing lysate was added to mixed lymphocyte cultures.

In contrast to control lysates obtained when vector alone was used for in vitro translation, in two separate experiments P1 containing lysates were inhibitory to lymphocyte proliferation in mixed lymphocyte reticulocytes.

Recombinant P1 was also synthesized using a bacterial expression system. The specific sequence coding for P1 was cloned into a GST tagged bacterial expression vector. The GST-P1 fusion protein was isolated using sepharose beads. The peptide was cleaved from the GST protein with thrombin. The purified P1 containing solution was added to mixed lymphocyte reticulocytes at varying dilutions. The inhibitory activity of the Pl containing solutions was compared to a control solution containing an equal concentration of GST and thrombin alone. The recombinant peptide containing solution was strongly anti-proliferative for T cells at all dilutions tested.

Using the molecular technique of 5 prime and 3 prime RACE, the naturally occurring mRNA species encoding for PI and P2 proteins was isolated from both 2.15 T cells and mouse thymus. A variant version of the 722415 est species was isolated in which a GG sequence is replaced with a TT nucleotide pair. This variant nucleic acid sequence is depicted in SEQ ID NO : 4. This naturally occurring

substitution yields a variant of Pl. This novel 67 amino acid peptide variant of PI is referred to herein as Plb. P1b comprises the amino acid sequence <BR> <BR> MCACVCPSACASVSLKNNLLCDFLWSFCSFTVLLHNSTDPCPPTPGIKCIIWSQVGFLG TVSFCAVY (SEQ ID NO: 5). Since PI and Plb share significant amino acid homology including critical Cysteine bonds, Plb is expected to have immunological properties similar to that of P1.

The second peptide, P2, contained within the suppressive cDNA clone was also found to have in vitro biological activity. Unlike P1, P2 has 46 amino acids with only two cysteines forming one potential disulfide bridge. Thus, P2 can be synthesized in pure form for further testing and investigation of the functions of these unique peptides.

Based upon the ability of PI and P2 to inhibit T cell proliferation, it is believed that peptides comprising PI or P2 or an active fragment or variant thereof such as P1b will be useful in preventing and treating autoimmune disorders, transplant rejection and abnormal immunoproliferative disorders including cancer. By"active fragments or variants thereof"it is meant amino acid sequences shorter or longer than Pl or P2 with similar activity or amino acid sequences with conservative substitutions as compared to Pl or P2, such as Plb, which have similar activity to Pl or P2.

Recombinant PI was also tested for anti-proliferative activity in human mixed lymphocyte reactions. Human T cells were incubated with genetically disparate irradiated antigen presenting cells for 3 days. Cultures were then pulsed with tritiated thymidine and harvested according to standard methods. Similar to murine cultures, PI strongly inhibits the proliferation of human T cells in response to a strong antigenic stimulus.

Examples of autoimmune disorders linked to T cell function which can be treated with these peptides include, but are not limited to, IDDM, rheumatoid arthritis, asthma, and

colitis. It is believed that these peptides, fragments or variants thereof would also be useful in prevention of rejection in human solid organ transplantation and aid in engraftment of the transplants as well as in the prevention or treatment of immunoproliferative disorders such as cancer.

The peptides, fragments or variants thereof can be prepared recombinantly via cloning of gene fragments encoding amino acid sequences of P1, P2, or active fragments or variants thereof into mammalian expression vectors to establish the immunosuppressive properties in transfected human cell lines.

Compositions comprising isolated peptides or active fragments or variants thereof can then be administered to animals, including humans, to prevent or treat an autoimmune disorder, transplant rejection or an immunoproliferative disorder.

Alternatively, peptides may be prepared synthetically and incorporated into compositions for such treatment or prevention. In addition, based upon the amino acid sequence and structure of the Pl, Plb and P2 peptides, one of skill in the art could design chemicals which mimic these peptides, referred to herein as molecular mimics, for use in treating and preventing these disorders.

Further, based upon the structure and sequence of P1, Plb and P2, small molecular inhibitors of P1 or P2 activity can be designed and synthesized for use in reversing T cell mediated immunosuppression. Such chemicals would be useful in the treatment of patients with immunodeficiencies such as those caused by HIV.

The peptides, molecular mimics and small molecular inhibitors of the present invention are preferably administered in a pharmaceutically acceptable vehicle.

Examples of such vehicles include, but are not limited to, aqueous solutions of salts or buffers, topical creams or lotions, and solid dosage forms such as pills, gelatin capsules, or liquid-filled gelatin tablets. Peptides are also often administered in the form of nontoxic salts such as

hydrochloride, hydrobromide, sulphate, phosphate, maleate, ascorbate, acetate, citrate, benzoate, succinate, and tartrate salts. Design of therapeutic regimens for the peptides, peptide mimics and small molecular inhibitors can be determined routinely by one of skill based upon the determined activity of the peptide, peptide mimic or small molecular inhibitor.

Example 1 The peptide P1 encoding cDNA was cloned into a GST fusion protein plasmid containing a trypsin enzyme cut site between GST and P1. After bacterial expression, the fusion protein was purified using sepharose beads and cut using trypsin. The P1 containing solution was added to a human mixed lymphocyte reaction using disparate donors for T cells and stimulating antigen presenting cells. Stimulating cells were irradiated (3, OOOR) prior to incubation with purified T cells in 1% AB human serum media for 3 days, pulsed with 3H- thymidine and harvested according to standard methods. Assays were done in triplicate. Experimental wells contained a 1/100 dilution of GST fusion protein alone. Peptide 1 was found to strongly inhibit human mixed lymphocyte reaction (MLR).