FI104179B - Method for Preparation of a Therapeutically Useful Polypeptide - Google Patents

Description

104179

A process for the preparation of a therapeutically useful polypeptide

Divided by FI application 905614.

5

FIELD OF THE INVENTION The present invention relates to a process for preparing a polypeptide comprising the sequence Val-Leu-Gly-Gly-Gly-Cys-Ala-Leu-Leu-Arg-Cys-Ile-Pro-Ala-Leu-10 Asp-Ser-Leu. Thr-Pro-Ala-Asn-Glu-Asp or any variant thereof that is immunologically cross-reactive with it but which is not a complete heat shock protein, provided that the polypeptide having the following amino acid sequence

1 MAKTIAYDEE ARRGLERGLN ALADAVKVTL GPKGRNWLE KKWGAPTITN DGVSIAK2IE 61 LEDPYEKIGA ELVKEVAKKT DDVAGDGTTT ATVLAQALVR EGLRNVAAGA NPLGLKRGIE 121 KAVEKVTETL LKGAKEVETK EQIAATAAIS AGDQSIGDLI AEAMDKVGNE GVITVEESNT 181 FGLQLELTEG MRFDKGYISG YFVTDPERQE AVLEDPYILL VSSKVSTVKD LLPLLEKVIG 241 AGKPLLIIAE DVEGEALSTL WNKIRGTFK SVAVKAPGFG DRRKAMLQDM AILTGGQVIS 2 0 301 EEVGLTLENA DLSLLGKARX VWTKDETTI VEGAGDTDAI AGRVAQIRQE IENSDSDYDR

361 EKLQERLAKL AGGVAVIKAG AATEVELKER KHRIEDAVRN AKAAVEEGIV AGGGVTLLQA 421 APTLDELKLE GDEATGANIV KVALEAPLKQ IAFNSGLEPG WAEKVRNLP AGHGLNAQTG 481 VYEDLLAAGV ADPVKVWGGG LQ

and polypeptides having amino acid sequences 234-540, 2561-540, 109-540, 171-540, 272-540 and 280-540, as well as human mitochondrial protein P1 are excluded. The polypeptide is useful for the prophylaxis of insulin dependent diabetes mellitus (IDDM) or for the initial treatment of IDDM.

BACKGROUND OF THE INVENTION * The incidence of insulin dependent diabetes mellitus (IDDM) has multiplied in recent years in many countries, and it is estimated that 1% of people currently living have IDDM before age 35. IDDM is caused by 2,104,179 autoimmune processes, which destroy insulin-producing beta cells. Diabetes only becomes clinically detectable after irreversible destruction of 9 very large (perhaps 90%) beta cells, and the life of this individual becomes dependent on an external source of insulin. In other words, at the time of making the clinical diagnosis, the autoimmune process has already caused irreversible damage, most of it without detectable symptoms.

Successful treatment of the autoimmune process that causes the said condition should be started before the patient has obvious symptoms of diabetes mellitus and needs insulin to replace the lost ability to produce insulin.

Stopping this autoimmune process would only lead to disease cure and obstruction of external insulin requirements if the disease process could be stopped at a stage when the patient still has sufficient beta cells to produce endogenous insulin. Therefore, any form of treatment would be more effective if individuals at risk could be identified when they had no visible symptoms of IDDM and before patients needed exogenous insulin.

Fortunately, there are various animal models of IDDM, including BB rats and NOD mice [see, e.g., 25 Rossini et al., Ann. Rev. Immunol. 3: 289-320 (1985)].

Many animals spontaneously develop autoimmune IDDM, which has many features of human IDDM.

Heat shock proteins (hsps) are a group of proteins produced by cells exposed to elevated temperatures or other stresses. hsp include proteins of different molecular weights, including 20 kD, 65-68 kD, 70 kD, 90 kD, 110 kD, etc. Heat shock proteins are ubiquitous in nature; they are produced by bacteria, yeasts, plants, insects and higher animals, including man, hsp proteins are highly evolved and remarkably homologous to all these various organisms. Because heat shock proteins are extremely well preserved throughout evolution, they are thought to perform vital functions. The synthesis of these is usually increased as the cells become exposed to stressful stimuli, including heat, certain metals, drugs or amino acid analogs. However, the specific functions of these proteins are still unknown.

For example, in patients with systemic lupus erythematosus (SLE), antibodies to the 90 kD heat shock protein have been detected [Minota et al., J. Clin. Invest. 81: 106-109 (1988). The function of these hsp90 antibodies is unknown.

15 hsp65 has been found to be associated with adjuvant arthritis in rats (see van Eden et al., Nature 331: 171-173 (1988)). Adjuvant arthritis is an autoimmune arthritis triggered by immunization of certain rat strains against Mycobacterium tuberculosis (MT). It has been found that a T-lymphocyte clone that responds to a 9 amino acid peptide sequence (180-188) in MT hsp65 can transmit this disease to immunologically undeveloped irradiated rats. Thus, adjuvant arthritis appeared to be an autoimmune disease caused by anti-hsp65 T lymphocytes. Autoimmune attacks on joints

it was caused by partial sequence homology between hsp65 peptide 180-185 and a segment of the cartilage proteoglycan linker protein (see Cohen, Scientific American 256 (988) 52-60). Also, T-lymphocytes * of the synovial fluid of patients with rheumatoid arthritis have been found to react with hsp65 of MT [see Res et al., Lancet II

(1988) 478-480].

Administration of hsp65 to rats prior to induction of adjuvant arthritis was found to prevent subsequent development of arthritis. Thus, the presence of an immune response to hsp65 was associated with r 4,104,179 in arthritis in both rats and humans, and administration of hsp65 was able to lead to resistance to arthritis.

EP-A-262 710 discloses polypeptides useful in the alleviation, treatment and diagnosis of autoimmune arthritis and related autoimmune diseases. The entire primary structure of the human P1 protein, including the nucleotide sequence and deduced amino acid sequence, has recently been published [Jindal, S. et al., Primary Structure of a Human Mito-10 Chondrial Protein Homologous to the Bacterial and Plant Chaperons and 65-Kilodalton Mycobacterial Antigen, Molecular and Cellular Biology 9: 5,279-2283 (1989)]. This protein, which is reported to have a molecular weight of about 63 kD, is a human thermal shock protein, referred to herein as the hHSP65 protein. All contents of this publication are incorporated herein by reference. The structure of this protein, reproduced in Figure 3 of this application, is intended to be identical to that of Jindal.

EP-A-261 648 discloses the use of activated T cells specific for autoimmune-20 disease in the treatment of such a disease. It is preferable to treat the T cells first with pressure and then with a chemical crosslinking agent and / or cell body disrupting agent to enhance their immunogenicity. All or 25 parts of the treated cell may be used as a vaccine against the autoimmune disease for which the T cell is specific.

In a known procedure for blocking autoimmune T cells, the target is immunized with attenuated or avirulent T cells having the autoimmune specificity or fragments or portions thereof. The subject responds by activating at least two types of regulatory T cells: antiergotype T cells, which recognize T cell activation markers, and anti-idiotype T cells, which appear to recognize self-antigen receptors on pathogenic endogenous autoimmune T cells. 5,104,179 and. T cell vaccination effectively induces permanent remission of the disease found in experimental animals as well as prevents the disease. Howell et al. [Scence 246: 668-670 (1989)] and Van denbark et al. [Nature 341: 541-544 (1989)] use the peptide sequences of the T-5 cell receptor β chain to vaccinate rats against experimental autoimmune encephalomyelitis and thus support the conclusion that the autoimmune T cell receptor itself may serve as a target epitope source for regulatory T cells. .

Although such use of T cells or fragments thereof in autoimmune diseases is well known, the antigen specific for IDDM has not been previously known and activated T cells for vaccination against IDDM have thus not been available.

SUMMARY OF THE INVENTION The present invention relates to a process for preparing a polypeptide comprising the sequence Val-Leu-Gly-Gly-Gly-Cys-Ala-Leu-Leu-Arg-Cys-Ile-Pro-Ala-Leu-Asp-Ser-Leu. Thr-Pro-Ala-Asn-Glu-Asp, or any variant thereof that is immunologically cross-reactive with it but which is not a complete heat shock protein, provided that the polypeptide having the following amino acid sequence

1 MAKTIAYDEE ARRGLERGLN ALADAVKVTL GPKGRNWLE KKWGAPTITN DGVSIAKEIE

ώ Sd ·

61 LED ARVKEVAKKT DDVAGDGTTT ATVLAQAL VR EGLRNVAAGA NPLGLKRGIE

121 KAVEKVTETL LKGAKEVETK EQIAATAAIS AGDQSIGDLI AEAMDKVGNE GVITVEESNT

181 FGLQLELTEG MRFDKGYISG YFVTDPERQE AVLEDPYILL VSSKVSTVKD LLPLLEKVIG

241 AGKPLLIIAE DVEGEALSTL WNKIRGTFK SVAVKAPGFG DRRXAKLQDM AILTGGQVIS

301 EEVGLTLENA DLSLLGKARK VWTKDETTI VEGAGDTDAI AGRVAQIRQE IENSDSDYDR

30 361 EKLQERLAKL AGGVAVIKAG AATEVELKER KHR1EDAVRN AKAAVEEGIV AGGGVTLLQA

421 APTLDELKLE GDEATGANIV KVALEAPLKQ IAFNSGLEPG WAEKVRNLP AGHGLNAQTG

: 481 VYEDLLAAGV ADPVKVTRSA LQNAASIAGL FLTTEAWAD KPEKEKASVP GGGDMGGMDF

and polypeptides having the amino acid sequences 234-540, 61-540, 109-540, 171-540, 272-540 and 280-540,35, as well as human mitochondrial protein P1 are excluded, t 6,104,179 characterized in that the polypeptide is synthesized by peptide chemistry. by conventional methods.

The novel polypeptides are useful in the prevention and treatment of IDDM, even in the early stages.

According to the observations made in connection with the present invention, during the development of IDDM, the animals produce hsp65 molecules or molecules which are cross-reactive therewith and are transported into the blood and urine of the animals. They also produce antibodies and T cells that are specifically against such molecules. Thus, the presence of hsp65 or its cross-reactive molecules or its specific antibodies or T cells in the blood or urine indicates the IDDM process before the beta cells are completely destroyed and the individual is sentenced to 15 lifetime diabetes mellitus.

Detection of onset diabetes will then allow the patient to take measures aimed at stopping the autoimmune process. For example, administration of hsp65 or its active epitope or any other immunologically cross-reactive molecule (antigen) with it effectively induces resistance to the autoimmune process associated with IDDM. Administration of T cells specific for such antigens in attenuated or avirulent form or after antigenic treatment, or fragments or active portions thereof, also induces resistance to the autoimmune process associated with IDDM.

In the context of the invention, it has also been found that immunization with hsp65 in an appropriate adjuvant, or an active epitope thereof, or other molecule (antigen) immunologically cross-reactive therewith, can induce IDDM. However, vaccination with such antigen without effective adjuvant, and preferably in combination with a tolerogenic carrier, can provide specific tolerance to this antigen. This 7 104179 effectively induces resistance to the IDDM autoimmune process. The same applies to vaccination with such T antigens specific for such antigens in attenuated or virulent form or after antigenic enhancement treatment or fragments or active portions thereof. If a patient is already shown to be in the early preclinical phase of IDDM, injection of such antigen or T cells (or fractions) may generate tolerance to this antigen and thus stop the autoimmune process prior to the occurrence of significant, permanent damage.

Brief Description of the Drawings

The invention may be more readily understood from the following brief description of the drawings and the following detailed description of preferred embodiments.

Figure 1 shows the nucleotide sequence and deduced amino acid sequence of human P1 protein, which is hHSP65. The numbers to the left refer to the nucleotide sequence relative to coordinate 1 at the beginning of the putative 20 start codons. The amino acid sequence is numbered starting from the same point, starting with 1. The extension of the 5 'end of this reading frame is represented by one-letter codes. The location of the internal EcoRI site (nucleotide 712) indicating the start of the lambda22a sequence is indicated. The poly '25 adenylation signal 15 nucleotides from the 3' end A-tail is underlined. The putative mitochondrial alignment sequence at the N-terminus and the collin-like amino acid sequence at the C-terminus containing repetitive Gly-Gly-Met sequences are labeled with la-coli. The positively charged amino acids in the C-terminal start sequence are indicated by (+).

Figure 2 is a graph showing the spontaneous reactivity of T cells in NOD / Lt mice with human hsp65, MT-hsp65 and MT-hsp70 as a function of age.

8 104179

Figure 3 is a graph showing T cell proliferation response to p277 and p278 as a function of peptide concentration f.

Figure 4 is a bar graph showing the proliferation response of T-cell-5 clones C7 and C9, which have the ability to transfer acute diabetes to young, pre-diabetes NOD / Lt mice, to p277, MThsp65, and plasmid control.

Figure 5 shows the results of immunization against peptides p277 and p278 in the control of induced diabetes. The spots indicate blood glucose levels three weeks after the immunization of each mouse in the test groups.

Detailed Description of Preferred Embodiments

The following examples illustrate specific embodiments of the present invention and tests related to the present invention. They are intended to be illustrative only and are not intended to limit the invention.

Example 1 Production of MThsp65

The hsp molecule of Mycobacterium tuberculosis was transfected into E. coli by conventional procedures and purified by van Eden et al. [Nature 331: 171-173 (1988)]. Such genetically engineered E. coli cells produce substantial amounts of MThsp65. Due to the close homology of hsps from various sources, mammalian or human hsp65 is also effective when produced by genetic engineering or isolation from so-30 bones.

hHSP65, present in blood or urine in individuals developing IDDM, could come from a few sources. These sources may include hHSP65, which is produced by normally healthy beta cells and released as a precursor to immunological killing when beta cells are exposed to viral infection or toxicity, or may be released from beta cells due to the stress caused by immuno-killing. hHSP65 could also be secreted by cells of the immune system during its prolonged anti-beta cell-5 activity. Although the sources of hHSP65 in this system are not currently known definitively, it has been determined that when hHSP65 is released, the individual is stimulated to produce antibodies to the hHSP65 mole molecule.

Antibodies to an unspecified molecule of 64,000 molecular weight have been reported to occur in some newly diagnosed IDDM patients (Baekeskov et al., Nature 298: 167-168 (1982)). However, it is not known whether the 64 kDm antigen is hsp. In addition, this 64 kD antigen is not known to be present in blood or urine prior to the onset of IDDM. In contrast to this unspecified 64 kD beta-beta antigen, it is a hsp65 determined protein whose amino acid sequence is known (Thole et al., Infection and Immunity 55 (1987) 1446-1475). Similarly, the amino acid sequence of hHSP65 is known and shown in Figure 1.

Example 2 Production of antibodies to MThsp65

Conventional laboratory strains of rabbits (New Zealand White) were inoculated by subcutaneous injection of 100 µg of MThsp65 produced in Example 1 in physiological saline (0.5 ml) emulsified in mineral oil (0.1 ml) (incomplete adjuvant). After one month, rabbits were boosted with 100 μg MThsp65 in physiological saline (1.0 mL) and after two weeks, rabbits were blood sampled and serum collected. The rabbits were boosted with the same dose after two months and a blood sample again.

104179

Example 3

Many murine molecules are cross-reactive with mycobacterial hsp65

Rabbit anti-hsp65 antiserum as well as the monoclonal antibody designated TB78 were tested to identify the mammalian molecules recognized by mycobacterial hsp65. This antibody was developed and supplied by Dr. J. Ivanyi, MCR Tuberculosis Unit, Hammersmith Hospital, London. This antibody is specific for the hsp65 molecule of M. tuberculosis. Binding to these antibodies from five types of preparations was examined: cloned M. tuberculosishsp65; Sera from NOD mice developing IDDM and healthy comparators; and lysates and reference fibroblast lysates of heat-shocked rat fibroblasts.

hsp65 was prepared as described in Example 1. Rat embryofibroblasts were cultured by standard procedures. To induce heat shock proteins, fibroblast-20 cultures were incubated at 42.5 ° C for 2 hours and then at 37 ° C for 1 hour. Heat shock treated fibroblasts and control fibroblasts were cultured at 37 ° C and then lysed (about 5 x 10 6 cells per species) using a lysis buffer containing 0.1% SDS and 1% Triton in combination with protease inhibitors. . The protein content was adjusted to 2 mg / ml by the Bradford assay. The material was subjected to conventional electrophoretic treatment on a 10% polyacrylamide gel under reducing conditions (2% 2-mercaptoethanol); the number of samples was 100 µg / lane.

Sera from healthy control mice and NOD-developing IDDM mice were diluted to 2 mg / ml. Each of these serum preparations was subjected to polyacrylamide gel electrophoresis separation as described above. The separated proteins were then transferred overnight to? 104179 nitrocellulose paper by standard procedures. The papers were then incubated for one hour at room temperature with hemoglobin (1%, for protection) and then with either normal rabbit serum or anti-hsp65 at a dilution of 1: 100 or a dilution of TB78 or monoclonal reference antibody. at 1: 100 for two hours at room temperature. Binding of antibodies to any of the distinguished bands was detected by incubation with goat 125 I-labeled anti-rabbit Ig or anti-mouse Ig, washed and developed by autoradiography. Molecular weight standards were included.

It was observed that few bands were recognized by anti-hsp65 antibodies. Mycobacterial hsp65 was recognized by both rabbit antiserum and TB68 monoclonal. The antibodies also recognized a single 65 kD band of murine fibrioblasts that was aided after heat shock. The heated fibroblast lysates also had positive bands at 30 kD and 47 kD moles. One additional band at approximately 25 kD was detected in the sera of NOD mice developing IDDM. Thus, mammalian molecules with molecular weights of 65, 47, 30 and 25 kD are cross-reactive with mycobacterial hsp65.

Example 25 hsp65 is formed in pancreatic islets

Because the development of IDDM is accompanied by an increased presence of hsp65 in blood and urine, it was thought that islet beta cells could be a source of hsp65. To test these 30 theories, rabbit anti-hsp65 was examined to see if it binds to islet cells.

A conventional method for the preparation of frozen rat pancreatic slices of 6-8 μτη: η was used. Sections were covered with normal rabbit serum 35 or anti-hsp65 antiserum (absorption-treated liver 12 104179 powder to remove non-specific antibodies) diluted 1:50, incubated for 30 min at room temperature, washed extensively with PBS and then incubated for 5 min at 5%: with normal goat serum before incubation with goat fluorescein-labeled anti-rabbit Ig for 30 min at room temperature. The islands were strongly stained by anti-hsp65 serum but not by rabbit control serum. Therefore, the islets produce hsp65.

Example 5

Immunization to hsp Induces IDDM Since it was observed that islet cells produce hsp65, it was assumed that the immune response against hsp would damage the beta cells and therefore induce IDDM.

Male C57BL / Ksj mice, 8 weeks old or NOD female mice, 4.5 weeks of age, were immunized by intraperitoneal injection of 50 μg hsp65 and tested for development of IDDM with evidence of blood glucose above 250 mg / 100 ml .

In NOD mice 4.5 weeks of age, spontaneous IDDM development continues for at least another three months. C57BL / Ksj mice do not develop spontaneous IDDM. hsp65 was administered emulsified in oil or PBS. Oil-Emulsified Bovine Serum Albumin (BSA) was used as a reference. The results are shown in Table 1. It was found that hsp65 inhibited IDDM in oil but not in PBS. Therefore, the immune response to hsp65 may trigger IDDM, probably because beta cells produce antigen cross-reactive with hsp65.

30 13 104179

Table 1 in hsp65 adjuvant induces IDDM occurrence of IDDM 3 µc Antigen Adluvant after administration NOD hsp65 oil 7/10 hsp65 PBS 0/10 BSA oil 0/20 C57BL / ksj hsp65 oil 6/7 hsp65 PBS 0/9 no anti- no adjuvant 0/15 gene

In an additional experiment, normal mouse strains that do not spontaneously develop IDDM, such as NOD mice, or even after a small dose of streptozotocin, such as C57BL / ksj mice, were inoculated intraperitoneally with 50 μg of antigen, either hsp65 or bovine serum antigen ( incomplete Freund's adjuvant (oil). Blood samples were taken from mice in the morning after 19 days and blood glucose was measured. Blood glucose levels above 200 mg / 100 ml were considered IDDM diagnosis. The results are shown in Table 2. It can be seen that immunization with hsp65 can induce IDDM even in apparently normal mouse strains, especially at the appropriate dose of an-. 25 as shown. This supports the conclusion that hsp65 or molecules cross-reactive with it are target antigens in the context of IDDM.

* r 104179 14

Table 2

Immunization with hsp65 induces IDDM in diabetic non-diabetic mouse strains.

Blood glucose mistlheys_

Antigens Antigens

- Mouse strain hSP65 BSA | hsp65 BSA

10 I

C3H.eB / Fej 270 + 41 96 + 32 I 5/5 0/5 C57BL / 6j 298 + 52 122 + 26 I 5/5 0/5 DBA / 2 14 6 + 33 126 + 21 1 0/5 0 / 5 15 SJL / j 162 ± 27 139 + 26 | 0/5 0/5

Example 6 hsp65 Can Induce Resistance to the Development of IDDM It is well known that antigen administered without effective adjuvant or in combination with a tolerogenic antigen can induce immunological inertia, i.e. specific tolerance to said antigen. Therefore, mice injected with hsp65 in PBS were tested to determine if the mice had acquired resistance to IDDM induced by hsp65 in oil. One month after the C57BL / Ksj mice received hsp65 in PBS, they were given hsp65 in oil and none of these mice developed IDDM whose mark-30 strain was maintained at> 260 mg / 100 ml after three weeks. . In contrast, 8 control mice out of 10 who had not received hsp65 in PBS developed IDDM when they received hsp65 in oil.

104179 15

In another experiment, hsp65 in PBS was administered intraperitoneally to 30-day-old female NOD mice 15 days prior to the administration of hsp65 (50 µg) in oil to induce IDDM. The presence of IDDM was measured by 5 blood glucose concentrations above 200 mg / 100 ml 35 days after the induction dose. The presence of IDDM was re-measured at 5 months of age. It is known that at this age 50% of all untreated NOD mice have detectable IDDM.

10 The results are shown in Table 3.

Table 3 Use of hsp65 as a vaccine against XDDM Frequency of 25 IDDMs hsp65 PBSrssM 35 days after immunization at 5 months of age (ua). _, 0 7/8 1 0/8 0/8 20 5 0/8 0/8 50 0/8 0/8

Thus, it is apparent that hsp65 can be used to induce tolerance to the diabetes process leading to diabetes. This tolerance is not only effective against immunogenic hsp65 attack but remains an effective treatment against the natural development of spontaneous IDDM in NOD mice.

Example 7 ·, Treatment of onset IDDM using hHSP65

As shown in Example 6, hsp65 can be used to induce resistance to the IDDM autoimmune process. This appears to be caused by the immunologic tolerance mechanism of beta cells upon exposure to exogenous 104179 16 hsp65. Thus, hsp65 may be useful in the treatment of IDDM before the disease becomes clinically detectable, and the autoimmune process may be stopped before significant permanent damage occurs. The experimental results of size 3, compiled in Table 5, that the natural development of spontaneous IDDM can be inhibited in NOD mice provide significant evidence that hsp65, particularly hHSP65, can be used therapeutically. This autoimmune process begins very early in NOD mice. Insulitis is detectable at 10 months of age. IDDM becomes clinically detectable at 5 months of age in 50% of female mice in this strain. Administration of hsp65 to 30-day-old mice stops this natural development. This demonstrates that treatment can be effective even after birth to the 15 islands of autoimmunity.

Example 8 T-cell response to hHSP65 is associated with an emerging diabetes mellitus

The human hsp65 gene shown in Figure 3 was cloned for expression in a conventional manner and resulted in approximately pure human recombinant hsp65.

This experiment demonstrates that mice whose beta cells are spontaneously killed are T cells responsive to recombinant human hsp65. Spleen cell suspensions obtained from groups of 5 to 25 NOD / Lt female mice of different ages were assayed for T cell proliferation in a substantial manner as described for T cell thyroglobulin responses [Maron, R. et al., J. Immunol. 131: 2,316-2322 (1983)]. Briefly, cells (1 x 10 6 cells / ml) were incubated for 72 hours in triplicate in microplate wells in culture medium (0.2 ml) in the presence or absence of the following antigens (5 μg / ml): human hsp65, MThsp65 and MThsp70. Proliferation was measured by incorporating [3 H] thymidine into DNA for the last 12 hours of incubation. Results 35 are expressed as Δ cpm, which is the mean cpm of the wells containing the antigen si 104179 17 minus the mean cpm (SE) of the control wells cultured without antigen added. Reference cpm values ranged from 9,000 to 10,500. Onset of IDDM in about 50% of the mice occurred at 4-5 months of age, as indicated by "IDDM" in Figure 4.

The results of this experiment, which show the spontaneous reactivity of T cells in NOD / Lt mice to various antigens as a function of age, are plotted in Figure 10, section 4. At 1 month of age, with little or no insulin in NOD / Lt mice Reactivity of T cells to any antigen. However, at 2 and 3 months of age, along with increasing insulitis, a strong and increasing reactivity to human 15 hsp65, a relatively weaker reactivity to MThsp65, and no reactivity to MThsp70 were observed. Reactivity to human hsp65 and MThsp65 decreased with the onset of IDDM at 4.5 months and further decreased at 6 months after clinical onset of IDDM. Thus, T-cell response to human hsp65 appeared to be associated with an increase in beta cell damage prior to visible IDDM. The decrease in T-cell reactivity upon the onset of IDDM can be explained by reduced immune stimulation after loss of beta cells and their antigens.

Example 9

Human hsp65-responsive T cells cause diabetes

Spleen cell suspensions were made from groups containing si-30 containing five 3.5 months of strain NOD / L, • 5. C57NL / 6 or C57BL / KS female mice. One group of NOD / Lt mice (Group 2) had been primed 9 days earlier by intraperitoneal immunization with MThsp65 (50 μg / ml) in oil. Some spleen cells (groups 4, 6, and 8) were activated by incubation with Con A (1.25 μg / ml) for 48 hours.

104179 18

Con A-treated cells were transplanted for an additional 5 days in media lacking Con A. Spleen cells were tested for proliferation responses to human recombinant hsp65, recombinant MThsp65, or E. coli blood or 5 luantigen after activation as described in Example 12. way. E. coli control cells were transfected with plasmid pEX2, which did not contain hsp65 genes. The antigens were used at a concentration of 5 µg. Reproductive response results from three replicates are presented as the Stimulus Stability Index (SI), which is the ratio of [3 H] -thymidine incorporated into spleen cells incubated with the antigen to the [3 H] thymide incorporated into control antigen-free control cultures. Comparative cultures had a cpm of 5,000 -15,7,000 and standard deviations from the mean of cpm were always less than 10% of the mean. The ability of cells to induce diabetes was tested by activating some of them by culturing with Con A for 48 hours. Groups of pre-diabetes recipient 1 month-old recipient nasal-20 mice were inoculated intraperitoneally with intact spleen cells or Con A-activated spleen cells from MThsp65-primed or untreated mice. After 21 days, recipient mice were evaluated for the development of acute diabetes mellitus as evidenced by hyperglycemia (blood glucose> 200 mg / 100 ml) and histological signs of insulitis. Blood glucose was measured by taking a blood sample from the tail vein of each mouse at approximately 9 am and glucose concentration was measured using a Diascan Glucose Meter and 30 test strips (Behringwerke, Federal Republic of Germany). In- ·. the sulitis was determined by histological detection (hematoxylinosin and light green staining by the Histology Laboratory of The Weizmann Institute). The Insu connector was judged by an individual who did not know the 35 identities of the test discs.

r 104179 19

The results are shown in Table 4. The differences between NOD / Lt spleen cell responses to human hsp65 and MThsp65 (groups 1-4) were significantly greater (P <0.01) than NOD-Lt spleen cells in E. coli. -ver-5 responses to tail antigen or other mouse strains (groups 5-7) to hsp65 antigens using Student's T test; ND = not specified.

t 20 1 04 1 79

C/O

>> 0)

O rC

U «H

iJ m O O O O · · · ·

in ”H H H σ \ QQ QO

«N s v. \ 'V v. · · ·> Ioeoocozzzz

C * J

a> e

CO-H

A! -B

<u tn u h e »

X

Λ

•B

Ό

C-H

• H 4J

4J C

3 -H MJ MJ

3 O H H

t-l «H ····

<3 · <-H -H -H >> QQ QQ

• y «<0 JJ. . .

COM SS Z 2 2

01 U I

Q> m vo e a -h o)

CO X -M

J3 -H >> E- * 0); c0 · «· ·

2 S Hon HO Q Q QQ

M <U i-ti-l · - Λ · * ''

t-j 00ιΗ · ΗγΠγΗ γΗγΗ 2h Z Z

-H O CO

C JJ ϋ

0) C M

0) CO * 0) ·· ω> in / -s

vo M

O. CO

co w m K m IL in r- in r- cmh o σ> X to ~ r- - - ~ ^ • u jj .c ro v n «a · hh i-io

0 Tl H

01 "S

4J (0 10> M o

5 I S s ^ ^ Q. OH

JJ 4-1 2 -H vo · "λ« Λ · -. OH O H

3 m Ssarviio o co * '

Ή M hi JO -W H {N H {N HH H H

3 μ ® M 2 CO 0) ^ H m -h u

• H Ή c C: aJ C

iHO 4J AJ -H 4-1 -H S · Μ

OM: cd: M C: M C: 0 C

M CU g SC SC WC

• e: o: o -h: o -h 4j -h

4J 4-io 4-io: co O

C + J 4J> 4J> B>

0): c0: c0 4J: c0 -H: c0 H 4) -H

n 4J B r-t 3 S uj E -M H · υ 3 3 <U Γ-t 4J <u - * i 0) <U A!

t — 1 i — I i- “t ·· 41 i — t CO i — t CO 4-1 CO

0 o ai iou oi i e 4i i e 4J i e CO CO 4-1 vO CO 4J4I 4-1 41 -H 41 1 CO 4J 0.3 4J <3 4) 4J <3 41 CO <3 0).

H C -H C01- | -i-l, ώ -H -Ui MJ

M JOJO CO CH to CiH JJ CH

e: c0 H O: c0 0: ctJ: co Otrt O mj {M J <! 2 P. -ϋ O-n JJ Oi-ι w O-i-i O «V -P>>

* J »J

-H CO>> ® ® m 4J Q Q f «£ • h e o O m in

• h m; g 2 O U

SS JJ **: c0

S

£. . · · ^ H n n - <r m vo r »eo 2i 104179

The results, shown in Table 4, indicate that T-cell responses of 3.5-month-old NOD / Lt mice to human hsp65 could be enhanced by immunizing the mice with MThsp65 (group 2) or by activating the T cell population at T cell mitogen. concanavalin A (Con A; Group 4). Con A has been shown to stimulate primarily activated in vivo autoimmune T cells in animals. Thus, NOD / Lt T cells raised in vivo by immunization with MThsp65 or activated in vitro with Con 10 A naturally react to human hsp65.

Human hsp65 itself is not a mitogen; adult mice belonging to strains that do not spontaneously develop IDDM, such as strain C57BL / 6 (groups 5 and 6) or C57BL / KS (groups 7 and 8), show no T-cell reactivity after Con A stimulation.

Finally, Table 4 illustrates that in vitro activation of the human anti-hsp65 T cell population by Con A made T cells capable of transferring acute diabetes to NOD / Lt mice in the pre-diabetes 20 months (groups 2 and 4).

The relationship between T cell responsiveness to human hsp65 and diabetes is confirmed in Table 5. In the experiment which yielded the results of Table 5, T cell lines were obtained by: repeated stimulation of spleen cells with MThsp65 (5 µg / ml) as described by Maroni (supra). . Clones were isolated from cell lines by the limit dilution method. T-souclones were tested for their ability to substantially transport acute diabetes as described in Table 4, except that '5-106 Con A-activated cells were transplanted intraperitoneally. Clone reproductive responses to antigens were measured as SI as described in Table 4. The control cultures had a cpm value of 4,500-6,500 and standard deviations from the mean cpm were always 35 less than 10%.

104179 22

Table 5

Diabetogenic T-Cell Clones Recognize Human Hsp65 T-Cell Acute Diabetic Proliferation Responses (SI) Transfer of Anti-Clone Text to Genes__ _ (Occurrence-Human E. coli ____ Density) _hsp65 MT-hsn65_ Reference Sample 27 9/11 16.9 7, 1 0.9 C7 10/12 23.8 6.7 1.1 10 1 C9 10/15 38.5 5.8 1.2 21 0/13 6.3 2.8 1.0 15 Clones were found to react more strongly human hsp65 than MThsp65 or E. coli reference antigen. In addition, clones 27, C7 and C9, which responded strongly to human hsp65, were diabetogenic, whereas clone 21, which reacted relatively poorly to human 20 hsp65, was unable to transfer from diabetes. Because cloned T cells have only one type of antigen receptor specificity, it can be concluded that acute diabetes can be transferred to pre-diabetes NOD / Lt mice by T cells recognizing. 25 human hsp65 has an epitope that is somewhat cross-reactive. MThsp65 epitope.

Example 10

Virulent T-Cells Vaccine Against Spontaneous IDDM Groups of 5-7 pre-diabetic two-month-old female NOD / Lt mice were primed with antigen (50 μg) in oil or left un primed. Mouse spleen cells were activated by incubation with antigen (5 µg / ml) or Con A (1.25 µg / ml) as described in Example 231041979 in Table 5, or not activated. Cells (25-106) were then transferred intraperitoneally into groups of one-month-old, pre-diabetic mice. 5 mice were examined for acute diabetes (hyperglycemia; blood glucose> 200 mg / 100 ml) after 3 weeks as described in Example 9. Spontaneous IDDM was determined at 8 months of age by examining hyperglycemia and insulin. Differences with the control group (group 1) were significant in the indicated households (* p <0.01). The numbers represent the cumulative results of 2-3 experiments.

«·% T 24 104179

U J ~ S

C * «Η τ-τ C -rl • rl M O) e Ο>, *

te «rl in * o ι-l -K

rt C -C 00 1-1 OJ Γ ^ Γ «o 4J ·· -rl e a <->

O O M

rt P * Q * rl

«M H B

rt | X 1

01 II

4-> ^> ä * rl 0) 0) 4-1 XI U 10 m e οι ti • h rt x> a o rt rt

4-1 1-1 4J

e to -ö e JCÖ CÖ Ή / —s >> e -H N! «« *

0) Ή t »^ O O O

J J §3

SS Q <03 SS

o a

M

e

• H rt I

4-1 -rl 1-4

rt C -H

rt rt s: M rt: rt

rt 4J c M

C rt: rt

U O l-ι KO

rt o. rt S

f »to to 4-1 rt to in rt τι Λ

Slrt S HH N O If »O 1-1 O

o aa · ηο η- <νηη <νγν

She Ό 6 E Si P! M C rt rH O rt.

3 i-i ss rt e -h;

1— l: o> M M

3 4J >> I

rt rt E C

H rH “Hl

O CO «U« H

CO p CO u H rt ”-h m o m j_) so vo r- vo u u si> Q, Dj Cl ·

rt O O-Hrt -H tnC WW

tn c 4J c <u ϊί ^ rs n m ^ rtCrtf. 5ir, C_.

4-i rH E ·· O w H H

tou O S U Λ £ X.

rt rt tn,>> l to rt E H 2 O in .. -h to Γ «vo

m o D · Η »H * H 04 CU

vO «U t-jOJ -ΦΟ) WW

(O * rt 'oi * $ * t * m s s: rt

E

x! H m n TT in vo *> y cd 104179 25

As expected, untreated control mice (Group 1) were not hyperglycemic at 7 weeks of age, and spontaneous IDDM was observed in 81% at 8 months of age. In contrast, acute diabetes (blood glucose above 200 mg / 100 ml) was transferred using spleen cells from pre-diabetes mice, provided that the spleen cells were activated in vitro by MThsp65 (group 2) or Con A (group 3).

Spleen cells obtained from mice primed in vivo with MThsp65, BSA or MThsp70 were administered in vitro activated with the corresponding antigen. Anti-MThsp65 cells (group 6) induced acute diabetes, but anti-BSA or anti-MThsp70 cells (groups 4 and 5) did not. Unlike Con A and MThsp65-activated cells, which responded strongly to human hsp65 (Tables 3 and 4), no proliferation assays for MThsp65 or human were observed in proliferation assays for BSA or MThsp70. hsp65 (results not shown). Thus, acute diabetes transmission was specific for cell populations containing human hsp65-responsive T cells.

Table 6 also shows that the acute diabetes attack was followed by a significant decrease in the incidence of spontaneous IDDM developing at 8 months of age. Mice ·. Inoculation with spleen cells against BSA or MThsp70 did not induce acute diabetes or prevent spontaneous IDDM (groups 4 and 5).

The ability of acute exogenous diabetes to interrupt the spontaneous IDDM process was also observed in experiments performed with the anti-hsp65 T cell clones described above (Table 4). 31 mice were injected with virulent clone 27, C7 or C9, and all developed acute diabetes within 1-2 weeks. All recovered spontaneously from hyperglycemia within 2 weeks and only 35 out of 31 developed spontaneous IDDM at 8 months of age. In contrast, 13 mice <104179 2,6 injected with avirulent clone 21 did not develop acute diabetes, and 6 of these mice developed spontaneous diabetes at 6 months of age (P <0.01).

Thus, diabetes caused by the direct translocation of ul-5 to human anti-hsp65 is transient. However, the mice not only recovered from acute diabetes but also acquired resistance to subsequent spontaneous development of IDDM. Thus, instead of accelerating or increasing the severity of the natural onset of IDDM, exposing mice in the pre-diabetes stage to virulent T cells actually resulted in resistance to the diabetes process.

Example 11

Weakened T cells act as a vaccine against autoimmunity against hsp65 and block IDDM

T cell vaccines were prepared from spleen cells of 3-month-old pre-diabetic NOD / Lt mice by activating the cells by incubation with Con 20 A or MThsp65 as described in Example 8 Table 3, and then attenuating the activated cells by gamma irradiation (1500 R) or glutarid (0.3%, 15 min) Lider et al. as described in Proc. Natl. Acad. Sci. U.S. 84: 4,577 (1987). Groups, ·. 25, consisting of 15 or 25 five-week-old NOD-Lt female mice in pre-diabetes, were then unvaccinated or inoculated intraperitoneally with 107 treated spleen cells. At six months of age, 30 of the 5 mice in each group were tested for spleen T cell proliferation responses to human hsp65, which is presented as a Stimulation Index (SI). The reference cpm values without antigen added were 2,465 ± 235 and 2,246 ± 185 in unvaccinated and vaccinated mice, respectively. Blood-35 samples were taken from the remaining mice to determine human anti-hsp65 antibodies by the standard solid phase radioimmunoassay [Schechter et al., J. Biol. Chem. 259, 6, 411-6,419 (1984)]. To detect serum antibodies, micromalayers were coated by incubation with hsp65 (to determine hsp65), insulin (to determine anti-insulin), or guinea pig DM idiotype positive anti-insulin (to determine anti-idiotype antibody) ( 50 μg / ml each). The presence of antibodies to these antigens was detected by incubating the coated wells with test mouse sera (diluted 1:50) and developing with goat 125 I-labeled anti-mouse Ig (Amersham, England; 105 cpm / well). For detection of hsp65 antigen (or cross-reactive antigen), the wells were incubated with 1.5 μ1: η of 15 test sera diluted 1: 5 and then covered with rabbit anti-hsp65-Ig (diluted 1: 100). . Binding was measured using goat 125 I-labeled anti-rabbit Ig (Amersham). Relative titer represents the binding of 20: 1 diluted serum to human hsp65 in cpm. The mean cpm of human anti-hsp65 serum antibody from non-diabetic 1-month-old NOD / Lt mice was 1450 ± 194.

Mouse sera were assayed for the presence of hsp65 or cross-reactive antigen using the solid phase radioimmunoassay described above. The results are shown in Table 7. Vaccinated mice differed significantly from unvaccinated control mice (1p <0.01).

(28 104179

/-OF

: ce w to s *. * «« "Ϊ o o o o

ci ^^ C

G 10 VO H O ΓΜ

·· 'jj rH

m e * ph

4J

3 rt f «G

0) Ή iH

M B S -H

• HO) O CO

ε 4) PO) JZ 60 m ^

• rt «H

4-1

Iti C

4J π) I

4- * Ή Ή VO r-v

DB -U Q

3D to -H to 03 M «rl C + l • h tu tn turn

> <u <u I

rt to rl MO

* H Ifl «H H

4-1 Ct) 4-14- ··. * 4-1 CB * 3 to m et) Q. CO o CU * H vO O «r- U> C tee) CX C ^ ~ • H ai to to tu. »» E * rl C CO DC ICO +1 +1 tu 4-i -h: <o ii to ot · O ·

-2-3 5> «n a n O

• H cd 4-1 Ή tI, d B. (Tl · dl tU to e B -H -rl ® 2 cu «·· D4J n a 4-1¾

H rl>. * U ^ C

td 4J, * oi rt ·· i to e tu oi, * cd iH ..VO CO J-ι Ai uum to VO -rl rt rt ex 4J / -> -> tn to 4j Ω m tn X. <u \ n cd e dl io i-ι cd e 4J <u -H j.

. *. * tu -rt q m ψ. * e tn e -h C i P * "0 · D <UC iH D cd O 1 <h

• H! -) ·· B g I ,, o O

D Dm jz S rt fr · i -,. + | cd rito ΗΉ «n Ϊ

H OCX en -H q. IT) Q Q O

en w cö · η υ ** * · * ·.

ij => -m w n Z S r> j

B

:c/o

• ri: co O 4J

rH e AJ

rH <U 3 p £ ^ n * * *

Xc .rt O r-v irv r-, rv 4-1> tn H TT UI Γ ~ ΓΊ e | to H o- n

o »{—i '—-« —1 IN {N

·:. * Q dl to you

D 4-1 4-1 XI O D

Z4 tn o u e D ·· C -H -rl i —t m ··· n -n 0 X> S Ph Ph

CO ex Q JZ XI

1 tn o di cu H Xl M TJ T3

tn r-t rH

to η rt

• fr, 4-1 -H «H

• e 1-t Ph 1-1 M

C dl rH rt Cd tu iH rt rt

M dl 4J 4J

• H 4-1 D D

<D ICÖ rH r— | 'X en bo night • H m

'4-1 VO

e a o <<j2

3 ® O O H

m new <29,104,179

The results presented in Table 7 show that otherwise virulent T cells activated by Con A or MThsp65 can be attenuated by treatment with gamma radiation (1500 R) or glutaraldehyde (0.3%) and can be used for vaccination against spontaneous IDDM. . As shown in other experimental diseases, irradiated or glutaraldehyde-treated autoimmune T cells were not virulent and did not cause acute diabetes (data not shown). Inhibition of IDDM was associated with a significant reduction in spontaneous T cell and antibody activity in vaccinated mice with respect to human hsp65.

As described in the previous examples, islet damage in NOD / Lt mice is significant up to the time of the appearance in the serum of a protein recognized by anti-hsp65 antibodies. Because this serum tigene is possibly derived from damaged beta cells, it was interesting to see if T cell vaccination had any effect on its amount. It can be seen (Table 7) that administration of a T cell vaccine significantly reduced the amount of hsp65 or cross-reactive antigen appearing in the serum at 6 months of age. This decrease was related to the lack of detectable insulin in the histological examination (data not shown).

Thus, MThsp65 or Con A activated Viru lentils against human hsp65 could be attenuated and used to vaccinate NOD / Lt mice against the development of spontaneous IDDM. The vaccinated condition was marked by a decrease in the immunological markers of the IDDM process, anti-human hsp65 T cells and antibodies. The amount of serum tigene cross-reacting with hsp65 or hsp65 was reduced, which can be explained by insulitis. stagnation.

30, 104179

Example 12 T cell vaccination confers resistance to acute induced diabetes

Example 5 showed that acute diabetes 5, characterized by insulitis and hyperglycemia, could be induced in pre-diabetes NOD / Lt mice by immunization with MThsp65 in oil. I also found that this form of acute diabetes was transient and in fact led to resistance to subsequent 10 spontaneous IDDM. This experiment investigates the effects of human hsp65 immunization and the effect of T cell vaccination on acute induced diabetes.

Groups consisting of 4-week-old NOD / Lt female mice were vaccinated with 107 Con A-activated 15 glutaraldehyde-treated spleen cells as described in Example 15 in Table 12 or left unvaccinated. After two weeks, mice were given MThsp65 or human hsp65 (50 μg in oil) to induce acute diabetes. The mice were then examined for spon-20 Danish IDDM development at 8 months of age by measuring hyperglycemia and insulin as described in Example 9. The results are shown in Table 8. The differences were significant at * p <0.01.

Table 8 25 T cell vaccination against acute hsp65-induced diabetes and spontaneous diabetes

Induced Diabetes Spontaneous IDDM Blood Occurrence - 3Q T-Cell Cell - Occurrence Glucose Density

gene vaccination Induced by: i Hhpyg. (niq / dl) IDDM

no no 0/10 151 + 19 8/10 no Human hsp65 12/15 258 ± 22 * 2/15 * no MT hsp65 8/10 339 + 116 * 0/10 * yes MT hsp65 0/20 14 0 ± 20 1 / 20 * 35 104179 31

It is seen that unvaccinated mice responded to human hsp65 or MThsp65 and developed acute transient diabetes. This was followed by resistance to spontaneous IDDM. In contrast, T cell-treated mice were resistant to antigen-induced acute diabetes. They were also protected from spontaneous IDDM. Thus, T cell vaccination effectively confers resistance to both acute diabetes induced by artificial immunization and subsequent spontaneous 10 IDDM.

Example 13

Peptide Synthesis

It was assumed that the key epitope of the human hsp65 molecule was an amino acid sequence that was partially, but not completely, homologous to the MThsp65 sequence. Because the latter functions less well, the sequence of the corresponding epitope is presumably slightly different.

A series of peptides were synthesized starting from the carboxyl terminus of the human hsp65 sequence and selecting sequences with slight differences between mycobacterial and human sequences. One such peptide has the amino acid sequence 437-460 of the human hsp65 shown in Figure 1, i.e., H-Val-Leu-Gly-Gly-Gly-Cys-Ala-Leu-Leu-Arg-Cys-25 Ile-Pro-Ala-Leu -Asp-Ser-Leu-Thr-Pro-Asn-Ala-Glu-Asp-OH.

This peptide is designated p277.

Reference peptide p278 overlaps the carboxyl terminus of p277 by 3 amino acids and has the following sequence: H-Asn-Glu-Asp-Gln-Lys-Ile-Gly-Ile-Glu-Ile-Ile-Lys- Thr-Arg-Leu-Lys-Ile-OH. This corresponds to the amino acid sequence 458-474 of Figure 1.

t 32 1 04 1 79

Example 14

Immune response to p277 in the diagnosis of development of IDDM

The spleen cell suspensions of 3-month-old NOD / Lt female mice in the pre-diabetes phase were tested for their proliferation responses to p277 as described in Example 8, and the results are shown in Figure 3. Although these mice did not develop visible IDDM for another 1-3 months, T cells respond strongly to 10 p277 but not to p278. The optimal concentration of peptide in this and other in vitro assays is 5 µg / ml.

Example 15

Pathogenic T cell clones respond to p277 and hsp65 variants Figure 4 shows that T cell clones C7 and C9, which have the ability to transfer acute diabetes to young, pre-diabetes NOD / Lt mice (see Table 5 above), react to p277. Clones are found to react with p277 as well as the whole mycobacterium 20 ri-hsp65 but not the control plasmid preparation which does not contain the hsp65 gene. Thus, it has been concluded that peptide p277 contains a pathogenic epitope because it is recognized by pathogenic T cells.

The p277 epitope is also present in the mycobacterial hsp65-25 molecule because C7 and C9 also react to that molecule. Thus, the mycobacterial hsp65 sequence homologous to p277 is immunologically active. The following shows a mycobacterial sequence with underlined human p277 sequence substitution: H-Asp-Ala-Glv-Gly-Gly-Val-Thr-Leu-Leu-Gln-Ala-Ala-Pro-Thr-Leu-Asp-Glu -: Leu-Lvs ----- Leu-Glu-Glv-Asp-OH. It can be seen that 13 of the 24 amino acids have been replaced. Therefore, we can conclude that the immunological properties of peptide p277 are able to withstand approximately 60% sequence change. Although the 35 smallest possible epitope would be 10 amino acids out of 24 of 104179, there is no 10 amino acid sequence of p277 that does not differ by at least 4-6 amino acids from the mycobacterial sequence.

Example 16 Peptide p277 can be used to treat diabetes

In Example 6, it has been shown that the whole hsp65 molecule in non-immunogenic form can induce resistance to acute diabetes in NOD / Lt mice induced by immunization with immunogenic hsp65 in 10 adjuvants. Figure 5 shows that also peptide p277, but not peptide p278, could be used to confer resistance to induced diabetes. Groups consisting of 7 NOD / Lt female mice at pre-stage 5 weeks of diabetes were treated with p277 or p278 (50 μg) in incomplete Freund's adjuvant. After two weeks, mice were immunized with immunogenic hsp65 (50 μ%) in incomplete Freund's adjuvant to induce acute diabetes. After three weeks, blood glucose was measured. It can be seen that none of the mice treated with p277 became hyperglycemic (blood glucose concentration of at least 200 mg / 100 ml). Instead, 5 of the 7 mice treated with peptide p278 developed diabetes.

Treatment with p277 also prevented the development of spontaneous diabetes in all 7 of these mice, whereas 80% of control mice treated with various antigens such as bovine serum albumin or hsp70 developed diabetes by 7 months of age. Thus, treatment with p277 resulted in resistance to both induced and spon-30 Danish diabetes. Thus, a particular peptide may *: provide a therapeutic effect of the entire hsp65 molecule administered in a non-immunogenic form.

Although the above results obtained with p277 relate to diabetes in NOD mice, it is apparent that this 35 peptide, as well as the entire hsp65 molecule or any other molecule immunologically cross-reactive therewith, can be used for treatment in humans. This is because diabetes in NOD mice is recognized as a reliable model for human IDDM. In addition, Toddin et al. mu-5 [Nature 329: 599 (1987)] NOD mice have a major histocompatibility complex group (MHC) II molecule (Ia) that is similar to human DQ6 associated with human IDDM. Thus, it can be expected that both human and NOD diabetogenic T cells should recognize the same peptide sequence present in MHC Group II lacking aspartic acid at position 57 of the DQs chain. If humans and NOD mice developing diabetes recognize a similar peptide antigen such as p277 can be used in humans as well as in NOD mice to treat IDDM.

The particular protein produced by the human body during the development of IDDM is a human heat shock protein of about 65 kD (or cross-reactive antigen with it). The 20 nucleotide sequence and deduced amino acid sequence of the human 65 kD heat shock protein is shown in Figure 1. Hereafter referred to as hHSP65. Other proteins that cross-react with antibodies binding to 65 kD protein may also be generated in vivo. For example, mice and rats have found molecules of 47, 30 and 25 kD that also cross-react with a monoclonal antibody specific for M. tuberculosis hsp65. A 47 kD molecule that is cross-reactive with such an antibody has also been found in rat fibroblasts. Given the persistence of heat-shock protein among species, it is fully expected that they will also occur in humans. In addition, during IDDM, the protein released into the blood and urine may be a molecule other than hHSP65, which however is cross-reactive with it. It may be a 104179 35 surface protein present on beta cells or even a protein fragment that retains the epitope present on hHSP65.

Thus, a molecule that acts as a diagnostic marker for the presence of an onset IDDM is a molecule that immunologically reacts with polyclonal antibodies raised against hHSP65, and preferably monoclonal or polyclonal antibodies raised against the p277 sequence of hHSP65. agents. Throughout this specification and claims, the term "hHSP65" is intended to mean, in addition to the human 65kD heat shock protein, any other related molecule in human serum that cross-reacts with polyclonal antibodies raised against the 65kD heat shock protein of any species. In particular, the 65, 30, 25 and 47 kD proteins already found and discussed herein are intended to be included within the scope of this definition.

Although in the experiments detailed above, p277 has been shown to correspond to a pathogenic epitope in NOD / Lt mice, T cells do not respond to p277 in another mouse strain, C57BL / 6, in which diabetes can be induced by human hsp65 and whose T cells respond to human hsp65. Thus, it is apparent that this is not the only pathogenic epitope of hHSP65. In fact, it has been determined that the molecular weight of the marker protein in the blood and urine of pre-diabetes patients is approximately 62 kD. Thus, it would be expected that the epitope (human leukocyte: * antigen, HLA) present in the major histocompatibility complex 30 (MHC) may vary from individual to individual. Thus, although p277 is currently preferred, those skilled in the art will appreciate that other antigenic sequences having the same or similar activity to p277 have been found in hHSP65. The purpose of the present invention is to encompass all such sequences.

Significantly, spontaneous IDDM is relieved when desired for an acute diabetic attack either mediated by virulent anti-hsp65 cells or induced by active immunization with hsp65. This suggests that the kinetics and potency of the autoimmune response may lead to self-regulation. Apparently, the insidious, chronic process can "slip through" a natural defense against autoimmune disease, while an open autoimmune stimulus may potentiate regulation. Resistance to autoimmune disease after an acute attack of the disease itself is regularly detected in an experimental autoimmune encephalomyelitis (EAE) rat model. When spontaneously recovered from acute EAE induced by either active immunization to myelin basic protein or passive transfer of activated T cells, rats are resistant to further attempts to induce EAE. It has been found that the EAE attack activates me-20 canisms, possibly anti-idiotype suppression, which have the ability to control virulent T cells responsive to myelin basic protein.

T cell vaccination using attenuated auto-immune T cells seems to be one way to activate regulatory mechanisms without the cost of acute illness. Anti-isotype and antiergotype T cells, such as those disclosed in the context of T cell vaccination against EAE, suppress autoimmune T cells against hsp65 and thus prevent the clinical appearance of IDDM.

These arguments help to explain how induction of tolerance or suppression of the immune response to hHSP65 prevents or ameliorates the diabetes process even after its onset. Thus, hsp65, low molecular weight molecules (25, 30, or 35 47 kD) or fragments cross-reacting with hsp65, modified peptide sequences, synthetic peptides, or even organic molecules based on copy of the fusion protein and thus designed that meet the physiochemical requirements for hsp65 can be used to inhibit or treat the 5 IDDM process as long as they are cross-reactive with polyclonal antibodies raised against hHSP65 or induce the generation of antibodies that are cross-reactive with hHSP65: with. The hHSP65 sequence p277 is a preferred agent for this purpose. In addition, attenuated T cells specific for 10 hHSP65 or a related gene thereof, as well as fragments or active portions thereof, may be used to induce such immunity or to suppress such immune response.

The hsp65 molecule has been shown to be useful as a therapeutic composition effective against continued development of IDDM by inducing tolerance to hHSP65 and thereby stopping beta cell self-destruction. The active ingredient for use in such treatment of ald-20 formula IDDM may be any material that is immunologically cross-reactive with hHSP65, i.e., either cross-reacting with hHSP65-raised polyclonal antibodies or inducing hHSP65-cross-reactive antibodies. Formation of substances. Such a material, be it a peptide, protein, carbohydrate or other substance, when tolerogenically administered, induces a tolerance to hHSP65 due to its cross-reactivity. If this substance is a hsp65 protein, it may be of any species. The agent does not need to be a complete protein in order to be immunologically cross reactive with hHSP65. It could be a protein fragment that retains the antigen activity of the protein itself, such as the sequence p277. Whether a particular agent is cross-reactive with hHSP65 is determined by routine-35 assays. If the agent cross-reacts with a polyclonal antibody formed against hHSP65, or if it induces the formation of cross-reactive antibodies with hHSP65, it is useful for the treatment of onset IDDM. Further evidence of the ability of such an agent to act on human 5 would be obtained by testing tolerance induction in the mouse assay described in Example 10. Performing such experiments would be routine and would not involve unnecessary experimental work.

A preferred compound for treating human IDDM is 10 hHSP65. The amino acid sequence of the human heat shock protein is shown in Figure 1. This protein can be used for the purpose mentioned. Its sequence p277 is another particularly preferred molecule for this purpose.

In addition to the hsp65 protein discussed herein, its salts, functional derivatives, precursors, and active fractions having the ability to immunologically cross-react with hHSP65 may also be used. Sequences such as those of Figure 1 or Van Eden et al. (supra) sequences in which one or more amino acids are deleted, added or replaced by other amino acids are intended to be included within the scope of this invention as long as they have the ability to cross-react with hHSP65 immunologically.

As used herein, the term "salts" refers to salts of both carboxylic groups and acid addition salts of amino groups of a protein molecule. Salts of the carboxyl group may be formed by methods known in the art, and include inorganic salts, for example, sodium, calcium, ammonium, iron (III) or zinc salts, and the like, and salts with organic bases such as those formed by, for example, amines such as triethanolamine, arginine or lysine, piperidine, procaine and the like. Acid addition salts include, for example, salts formed with mineral acids such as hydrochloric acid or sulfuric acid, and organic acids such as acetic acid or oxalic acid.

The term "functional derivatives" as used herein includes derivatives which may be prepared from functional groups present as side chains or N- or C-terminal groups by methods known in the art and as long as they remain pharmaceutically acceptable, i.e.. they do not destroy the activity of the protein, impart toxic properties to the compositions containing them, and do not adversely affect their antigenic properties.

These derivatives may include, for example, aliphatic esters of carboxyl groups, amides of carboxyl groups formed by reaction with ammonia or primary or secondary amines, N-acyl derivatives of free amino groups on carboxyl groups or aryl groups (e.g. or O-acyl derivatives of a free hydroxyl group (e.g., a hydroxyl group of a seryl or threonyl group) formed with acyl groups.

"Precursors" are compounds that are formed before hsp65 and which are converted to hsp65 in the animal or human body.

The "active" fractions of the substantially purified protein include any fragment or precursor of the polypeptide chain of the hsp65 protein molecule, alone or in combination with related molecules or linked groups, such as sugar or phosphate groups, or protein molecule aggregates themselves. provided that the fraction has the ability to cross-react immunologically with hHSP65. One example of such an active fraction is the sequence p277 of hHSP65.

104179 40

It is crucial that the active ingredient described above is administered in a manner that induces tolerance rather than eliciting an immunogenic response. Thus, it should not be administered in oil or in any other immunogenic adjuvant. One preferred way of administering the active ingredient so as to induce tolerance is to administer it in association with a carrier which favors the induction of antigen tolerance upon administration of the antigen-carrier conjugate. Such carriers are known as tolerogenic carriers. Examples of known tolerogenic carriers are polymers of D-amino acids, polyethylene glycol, polymers of sugar molecules, native IgG molecules, native spleen cells and fatty acid molecules. The antigen may also be administered in monomeric highly soluble form to induce tolerance. One known method for inducing tolerance to an antigen is to administer it orally, even without a carrier selected for its tolerogenic properties. Methods for administering antigen in a tolerance-inducing manner will be known to those skilled in the art, and any of such methods may be used in accordance with the present invention. Such methods are not themselves part of the present invention.

Such a tolerogenic composition may be administered as a vaccine to prevent the development of IDDM, for example, in family members of IDDM patients who may have a hereditary risk of developing IDDM. However, the composition is preferably used to stop the continued development of IDDM in subjects having detectable hHSP65 in their blood or urine, but preferably before they have developed an immune response to hHSP65. Induction of tolerance prevents damage (IDDM) caused by said immune response and thus uncontrolled response to hHSP65. However, it is not too late to use the composition, even after anti-hHSP65 antibodies have been recovered. The experiments, the results of which are shown in Tables 8 and 13 of "104179 41", show that the present invention can stop the immune response even after the onset of autoimmunity to the islets. Because the autoimmune process can take years in humans, even regulation that diminishes response would be favorable.

hsp65 or a related molecule (discussed above) can be used as an immunogen in pharmaceutical compositions, particularly vaccines, for the alleviation or treatment of IDDM.

One way to improve the efficacy of hsp65 as a vaccine or therapeutic agent is to form, by known genetic engineering methods, microorganisms that produce hsp65 either as such or as part of a fusion protein or multimer. These microorganisms themselves can be used as a live vaccine for production, which not only triggers the production of antibodies against that microorganism, but is also useful in the alleviation and treatment of IDDM. Examples of suitable genetically engineered microorganisms are Vaccinia and Salmo-20 nella strains.

The compound of this invention may be administered orally or parenterally, such as subcutaneously, intramuscularly, intravenously, intranasally, or rectally. Pharmaceutical tolerance gene compositions may be prepared in a manner known in the art.

The foregoing description of specific embodiments fully illustrates the general nature of the invention that others may readily apply such specific embodiments and / or applications to various uses without departing from the spirit of the invention, and therefore such applications and modifications are intended to be within the scope of circuit. It is to be understood that the phraseology or terminology used herein is for the purpose of illustration and not limitation.

r

Claims (3)

1 MAKTIAYDEE ARRGLERGLN ALADAVKVTL GPKGRHWLE KKWGAPTITN DGVSIAKEIE 61 LEDPYEKICA ELVKEVAKKT DDVAGDGTTT ATVLAQALVR EGLRNVAAGA NPLGLKRGIE 121 KAVEKVTETL LKGAKEVETK EQIAATAAIS AGDQSIGDLI AEAMDKVGNE GVITVEESNT 181 FGLQLELTEG MRFDKGYISG YFVTDPERQE AVLEDPYILL VSSKVSTVKD LLPLLEKVIG 241 AGKPLLIIAE DVEGEALSTL WNKIRGTFK SVAVKAPGFG DRRKAMLQDM AILTGGQVIS 15301 EEVGLTLENA DLSLLGKARX WVTKDETTI VEGAGDTDAI AGRVAQIRQE IENSDSDYDR 361 EKLQERLAKL AGGVAVIKAG AATEVELKER KHRIEDAVRN AKAAVEEGIV AGGGVTLLQA 421 APTLDELKLE GDEATGANIV KVALEAPLKQ IAFNSGLEPG WAEKVRNLP AGHGLNAQTG 481 VYEDLLAAGV ADPVKVTRSA LQNAASIAGL FLTTEAWAD KPEKEKASVP GGGDHGGMDF and the polypeptide with the amino acid sequences 234 to 540, 61 -20 540, 109-540, 171 -540, 272-540 and 280-540 and human mitochondrial protein P1 is excluded, characterized by the polypeptide being synthesized by peptide chemistry conventional methods. 1. A method for producing a polypeptide containing the sequence Val-Leu-Gly-Gly-Gly-Cys-Ala-Leu Leu-Arg-Cys-Ile-Pro-Ala-Leu-Asp-Ser-Leu-Thr-Pro-Ala-Asn-Glu-Asp or any modification thereof, which modification is immunologically cross-reactive therewith, but not is a complete heat shock protein, provided that the polypeptide having the amino acid sequence 10 2. A method according to claim 1, characterized in that a polypeptide containing the sequence Val-Leu-Gly-Gly-Gly-Cys-Ala-Leu-Leu-Arg-Cys-Ile-Pro Ala-Leu-Asp-Ser-Leu-Thr-Pro-Ala-Asn-Glu-Asp or any modification thereof, the sequence of which at least 40% is homologous to it. 30 3. A method according to claim 1, characterized in that a polypeptide of 24 amino acids is produced.