JP2928176B2 - Insulin-dependent diabetes mellitus treatment composition - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION Diabetes is a major health problem for people. As reported in the 1987 report of the National Diabetes Advisory Committee's Long-Term Diabetes Eradication Plan, 6 million people in the United States are known to have diabetes, with another 5 million diagnosed More than 500,000 new diabetics are found each year with this disease not being treated. In 1984, diabetes was a direct cause of death in 35,000 Americans, and contributed to a further 95,000 deaths.

[0002] Diabetic eye complications are a major cause of new, blind, law-defined patients in the United States between the ages of 20 and 74. The risk of lower limb amputation is 15 times greater in individuals with diabetes than in individuals without diabetes. This assessment is performed annually in the United States by people with diabetes, approximately 12
This suggests that more than half of 5,000 foot amputations have experienced amputations.

[0003] Kidney disease is a frequent complication of severe diabetes. Approximately 30% of all new patients in the United States who have been treated for end-stage renal disease have diabetes. This percentage is increasing at a rate of approximately 2% each year, so within the next decade, diabetes-related kidney disease will account for more than half of all registrants in end-stage renal disease programs It will be.

[0004] Individuals with diabetes are also at increased risk for periodontal disease. Periodontal infection progresses rapidly, leading to compromised metabolic inhibition as well as tooth loss.
Women with diabetes are at risk for serious complications of pregnancy. Current statistics suggest that fetuses of mothers with diabetes have a mortality rate of approximately 7%.

[0005] The financial burden on diabetes is also great. Each year, patients with diabetes or co-morbidities exhaust 24 million patient days in the hospital. A conservative estimate of the total annual cost attributable to diabetes is at least $ 24 billion (American Diabetes Association assessment, 198
8), and the overall economic impact of the disease is greater because the additional healthcare expenditure is often associated with certain complications of diabetes rather than diabetes itself. Diabetes is a chronic, complex metabolic disease that prevents the body from properly maintaining and using carbohydrates, fats and proteins. Diabetes arises from the interaction of various genetic and environmental factors and is characterized by high blood sugar levels resulting from the inability to produce insulin or its impaired utilization. Many cases of diabetes fall into two clinical types. Insulin-dependent diabetes mellitus (IDDM or IDD) and non-insulin-dependent diabetes mellitus (NIDDM or NIDD). Each type has a different prognosis, treatment and cause.

[0006] Approximately 5-10% of diabetics have IDD, which was earlier referred to as juvenile diabetes because it appeared early in life, usually in childhood or puberty. IDD is characterized by the inability to make insulin partially or completely. Patients with IDD die without daily insulin injections to control their illness.

[0007]

2. Description of the Related Art Little progress has been made in elucidating the etiology of diabetes until the mid-1970s.
Evidence has started accumulating that D has an autoimmune etiology. It is now generally accepted that IDD results from the chronic autoimmune destruction of pancreatic β-cells that make insulin. Lymphocytes and other inflammatory cells are observed in the islets of Langerhans in newly diagnosed patients with IDD, and are regenerating (β) cells more composed of β cells than those of other cell types (Langerhans)
More commonly found throughout the island. This active immune process is associated with various autoantibodies to the cytoplasmic and membrane antigens of β-cells, insulin and insulin receptors.

[0008] IDD is therefore a disease filled with autoantibodies. These include cytoplasmic islet cell autoantibodies (ICA) (Bottazo, GF, A. florin-christensen (Flo)
rin-Cristensen) and D. Doniac [1974]
Lance. 2: 1279-1283), islet cell surface autoantibodies (ICSA) (Maclaren, N., SW Huang, and J. Fog (1975) Lancet.i: 997-1000) , Insulin autoantibody (IAA) (Palmer, JP, CM
lin), and P. Cleamons [1983] Science 222:
1337-1339) and possible anti-genotype insulin receptor autoantibodies (Ins.RA) (Maron, R., D. Elias, M. de J. Bartelt, GJ Bruining, JJ Van Rood, Y.
Shechter, and IR Cohen [198
3] Nature 303: 817-818). 6 in 1982
Antibodies to 4,000 m _r islet cell antigen has been reported to have been detected in IDD patients (Bekkesukofu (Bekkeskov), S. Et al. [1
982] Nature 298: 167-169). In a later publication, Beckeskov et al. Reported that the 64,000 M _r antibody was associated with IDD in a BB rat model (Bkkeskov, S. et al.
[1984] Science 224: 1348-1350). Follow-up tests examined patients associated with individuals who had IDD and were therefore known to be at risk of developing the disease
(Bekkeskov, S. et al. [1987] J. Clin. Inves
t.79: 926-934). The results of this subsequent study suggested that the 64,000 M _r antibody might be present in some IDD patients before the appearance of clinical symptoms. However, in the same study, Beckeskov and colleagues did not know the function of the 64,000 M _r protein, and could not conclude that this data could be concluded whether the 64,000 M _r antibody was present before the onset of beta cell function began to decrease. Reporting.

It has been reported that other antibodies to various β-cell non-specific molecules are present in increasing numbers in IDD patients. These include tubulin, single and double stranded DNA, gastric parietal cells, intrinsic factor adrenocortical cells, thyroid peroxidase enzyme and antibodies to thyroglobulin. IDD patients have signs of polyclonal activation of B-lymphocytes, and increased antibody titers against various antigens may be a consequence.

[0010]

Further studies over the past decade have shown that patients with IDD have a genetic marker called a histocompatibility (histo-compatible) antigen that is associated with a greater susceptibility to IDD. It indicates that you are doing. Because these genetically susceptible markers are necessary, but not sufficient, for the development of IDD, some additional environmental factors, as yet unknown, may be involved in initiating beta cell destruction and in the development of diabetes. Seems necessary. Environmental factors, whether viral or chemical, initiate an immune response against beta cells, allowing the genetically susceptible person to immunodestruct beta cells (Atkinson, MA, N., et al.
K. Maclaren. [1990] Scientific Amereica
n 7: 62-67). Therefore, finding diabetic patients in a pre-diabetic condition is essential in efforts to prevent the occurrence of this disease. Perhaps the single most important advance in diabetes research over the past two decades is the recognition that autoimmune destruction of beta cells can take months and years to complete. Currently, clinical diagnosis of diabetes is rarely made until the destruction process is almost complete and insulin injections are needed to prevent death,
Interference before insulin-producing cells are irreversibly destroyed may provide a measure to prevent the progression of diabetes and its complications.

Therefore, before the disease progresses to the clinical stage, IDD
It is essential to find a way to accurately predict the onset of

Although many biological markers are associated with IDD, to date, none of these markers have been uniformly present before the onset of clinical signs of the disease. Not shown (Maclaren, NK [1988] Diabetes 3
7: 1591-1594). Such an early presence is essential for a compound to qualify as a useful predictive test of the disease. However, the initial presence alone is not sufficient for an accurate initial detection method. Predictive tests based on the presence or absence of a particular marker are valuable only if the predictive marker is present only in those who will have IDD and not in those who do not. . Since the treatment of diabetes and the psychological impact of the disease can have a significant effect on the health of the person being diagnosed, it is important to develop a predictive test that is specific and therefore very unlikely to be falsely positive. is important.

It is also important to find a predictive test that can recognize the onset of the disease many years before clinical symptoms appear. Early detection provides an opportunity to take action that can proactively or prevent serious health problems associated with clinical stage IDD. For example, cyclosporin A,
Non-specific immunosuppression tests with drugs such as azathioprine and steroids have been performed in newly diagnosed patients. Although a limited number of disease remissions were gained over the first year of their use, these patients never returned to normal because beta cells had limited regenerative capacity and were Treatment that begins at times is often almost too late.

To date, no suitable predictive test has been found for IDDs that meet these requirements. Although many autoantibodies are associated with the disease, studies that identify evenly predictive autoantibodies have not been successful. For example, 1976
Many years ago, autoantibodies (ACA) that react with the cytoplasm of glucagon-secreting (α) cells of islet cells have been described (Bottazo (Botta
zo), GF, and R. Lendrum [1976] Lancet
2: 873-876). ACA may be a predictive marker for deficiency of the islet cell hormone glucagon, as autoantibodies to endocrine glands have been found to occur before or accompany the clinical onset of the disease. So I thought. Unfortunately, later studies have revealed that ACA is not associated with defective α-cell function, not with IDD, and does not appear to be associated with any identifiable pancreatic pathology (Winter) , WE, NK Maclaren, WJ
Riley, RH Unger, M. Neufold
(Neufeld), and PT Ozand [1984] Diabetes
33 (5): 435-437).

[0015] Studies on the predictive value of the pancreas and other antibodies associated with the clinical stage of IDD have created a means of early detection of the disease. For example, both insulin autoantibodies (IAA) and islet cell autoantibodies (ICA) have been found to be present in many newly diagnosed patients. ICA has been shown to exist before the clinical stage of IDD, and has been found especially in nondiabetic relatives at risk for IDD. But,
Neither ICA nor IAA provide predictive tests that confer almost absolute specificity and sensitivity to IDD.

Thus, before clinical symptoms began and before insulin therapy was required, a more accurate means of detecting and treating IDD in the early stages had been felt substantially and for a long time.

[0017]

Means for Solving the Problems The invention described in the present specification is:
The present invention relates to a method for accurately detecting insulin-dependent diabetes mellitus (IDD) at an early stage, and more particularly, to detect IDD before significant loss of β-cell function, and to treat IDD at an early stage.

More specifically, it was found that autoantibodies to the 64,000 M _r protein of islet cells existed several years before the clinical appearance of IDD was observed. This antibody is 64
Named KA. Early detection and treatment of this potentially spoiling disease facilitates ineffective interfering treatment in later stages of the disease where irreversible destruction is widespread.

64KA can be used alone to detect and predict the onset of IDD, or in combination with other autoantibodies such as ICA and IAA that are known to be associated with IDD. 64KA described herein is highly disease-specific for IDD. 64KA has been found to be present uniformly in children and young adults tested months and years prior to the commencement of IDD. Despite the lack of other antibodies associated with the disease, the 64KA test is
It was found to predict the onset of IDD.

It has also been found that protein antigens reactive with 64KA are useful for treating IDD. As part of the present invention, ID
Novel means for treating D or delaying the onset of IDD are described herein. A novel means of treatment is to construct a novel hybrid β-cell 64K protein, that is, to construct a toxin product that can deprive immunological mediators involved in the pathogenesis of IDD .

Yet another aspect of the present invention is the imminence of pancreatic islet cell transplantation by administering 64K hybrid toxin molecules to minimize immune-related destruction and rejection of the new pancreas. The use of 64KA to diagnose and prevent destruction. The invention further relates to the use of anti-genotype antibodies against 64KA as a means of interference therapy.

The present invention is further directed to the discovery that the pancreatic 64K protein has sequence homology to the glutamate decarboxylase (GAD) protein. Furthermore, 64K protein is GAD
It has immunological reactivity with antibodies to the enzyme. Thus, yet another element of the present invention relates to the use of GAD proteins or immunologically active fragments thereof for the diagnosis, prevention and treatment of IDD.

DETAILED DESCRIPTION OF THE INVENTION The invention described herein relates to an agent for treating IDD discovered using 64K autoantibodies as an accurate and specific early indication of the onset of IDD. The finding of 64KA as a useful predictive marker for IDD stems from an exhaustive biochemical study focused on a number of compounds related to IDD. The high specificity of 64KA for IDD, as well as the frequent presence of 64KA in humans many years before the onset of IDD, is both unexpected. these
The nature of 64KA makes 64KA useful as a means of early detection of IDD.

To ascertain whether certain substances are valid predictors of the onset of IDD, it is necessary to determine whether these substances are present before clinical symptoms of diabetes occur. During this prediabetic period, a number of high-risk relatives of the proband with the disease must be screened for circulating autoantibodies to identify susceptible individuals, resulting in diabetes. It is difficult to study because long-term observations are needed to document the natural history of β-cell failure. Therefore, animal models were used to increase and further confirm the data obtained for human testing.

A non-obese diabetic (NOD) mouse is a human IDD
Is a useful animal model for The study of NOD mice gives important insights into the sequence of etiological events and leads to an understanding of the nature of target islet cell autoantigens involved in the autoimmune process. Another important model for IDD is the biobreeding (BB) rat. The present invention is a person,
Studies and tests involving BB rats and NOD mice were found.

In human and BB rat models, it has been discovered that islet cell autoimmunity is associated with autoantibodies to β-cell protein with a relative molecular weight (M _r ) of 64,000 (64K). Sera from newly diagnosed NOD mice were also determined to contain autoantibodies that immunoprecipitate 64K antigen from a detergent lysate of ³⁵ S methionine-labeled rat islet cells as well (Atkins ( Atokinson, M., and NK McLaren (Mac
laren) [1988] Diabetes 38: 1587-1590). This autoantibody is detectable in the NOD mice by the time of weaning, which disappears within weeks of the onset of diabetes, with older non-diabetic mice as well as all three non-diabetic constitutional strains tested. Was not present in In humans, 64KA was present prior to clinical diagnosis in 96% of those whose later onset of IDD was below the age of 35.
64KA was found to exist predictably well before the clinical appearance of IDD. The presence of 64KA was also found to be highly specific for IDD. 64 different from IAA or ICA
KA was not observed in individuals without a high or increased risk of developing IDD.

Materials and Methods Islet cell preparation Human pancreatic islets were isolated from cadaver pancreas as previously reported (Diabetes 37: 413-420, 1988). All batches of islets are in vitro (in vitro) in supplemented medium CMRL1066
Held in. 84,946 ± 11,061 average per batch (x ± SEM, n =
12) obtained an isolated human islet with an average island purity of 91.2 ±
The insulin content was 2.3% and the insulin content per island was 0.4 ± 0.1 mU insulin. 4.68 ± 0.6 by KREBS perfusion test at alternating glucose concentrations of 60 mg / dl-300 mg / dl
An average insulin stimulation index of 6 (G300 / G60) was given. All batches were labeled within 7 days.

Metabolic labeling of islet cells Following in vitro culture for a minimum of 36 hours (6000 islets / 50 ml
RPMI 1640 medium in 16 mM glucose, 20 mM HEPES, 100 u
U / ml penicillin, 100 μg / ml streptomycin, 2% (v /
v) Supplemented with normal human serum, 37 ° C, 95% air / 5% C
O ₂ ), these islets were replanted and cultured for 15 minutes in supplemented RPMI 1640 medium lacking methionine (1000 islets / ml). The cells were then plated at 37 ° C. (95% air / 5 ml) in methionine-free medium (1000 islets / ml) supplemented with ³⁵ S methionine (> 500 Ci / mmol) at a concentration of 0.5 mCi / 1 × 10 ³ islet cells. % CO ₂ ) for 5 hours. Methionine-containing (0.5 mM) RPM supplemented with cells
Incubate with I1640 medium (1000 islets / ml) for 30 minutes, twice (200 xg, 5 minutes) 20 mM Tris (pH 7.4), 150 mM Nacl, 100 KIE / ml aprotinin and 2 mM phenylmethylsulfonyl fluoride (PM
Washed with buffer containing SF) and then snap frozen until extraction of their detergent
(-80 ° C).

[0029] Immunoprecipitated islet cells of the 64K autoantigen were lysed in the above buffer containing 2% TRITON ^™ X-114. Unnecessary matter was removed by ultracentrifugation (1
00,000xg, 30 minutes, 4 ° C). The lysate was then separated into an aqueous, sucrose and detergent phase prior to immunoprecipitation (Bordia (Bord-
ier), C. [1981] J. Biol. Chem 256: 1064-1067). Lysate was added to normal control serum (10 μl per 100 μl supernatant, 1 h, 4
C), followed by protein A Sepharose C
L-4B was adsorbed in excess (100 μl of swollen Protein A Sepharose was used for 25 μl of serum). Aliquots of unbound (precleared) lysate (100 μl containing 5-10 × 10 ⁶ cpm) were incubated with 25 μl of test serum (either IDD or control) for 18 hours (4 ° C.). Detergent extracts from 1000 islets were used for each serum sample tested. To each assay tube, pre-swelled protein A Sepharose CL-4B was added and the reaction mixture was incubated for 45 minutes (4
° C). Protein A cephalometric - scan CL-4B was washed 5 times by centrifugation at then 0.5% Triton ^TM X-114 buffer (200 × g, 10 seconds), ice cold double distilled water (200 × g, 30 seconds) in In 1
Washed once. The bound protein was eluted and 80 mM Tris (p
H6.8), 3.0% (w / v) sodium dodecyl sulfate (SDS), 15% (v
/ v) boil for 3 minutes in a sample buffer containing sucrose, 0.001% (v / v) bromophenyl blue, and 5.0% (v / v) β-mercaptoethanol. Denatured. Sepharose beads were removed by centrifugation (200 ×
g, 1 minute) and the supernatant was electrophoresed through a discontinuous SDS 10% polyacrylamide separation gel, followed by coomassie
Brilliant Blue-Staining and Enhance (New England New Clear, Mass., Mass.,
(Boston) was performed.

ICA analysis (Neufeld, MN Maclaren,
W. Riley, D. Lezotte, H. McLaughlin, J. Silverstein, and A. Rosenbroom [1980] Diabetes 2
9: 589-592). ICA was assayed by indirect immunofluorescence on blood group O cryocut pancreatic sections as described in (9: 589-592). All results were read on coded samples and control negative and positive sera were used in each batch. Serum was considered positive when fluorescent antibodies of the tissue were observed in the islets of the pancreas. Positive results are from the Immunology of Diabetes Workshops
(IDW) and defined by Noel K. McLaren (M
aclaren), standardized by the Global ICA Proficiency Testing Program managed by the
Positive if autoantibodies exceed the concentration (or titer) of DF).

The insulin autoantibody binding procedure The RIA method used to detect IAA was modified to maximize sensitivity without losing specificity for IDD. Mean serum insulin binding has no family history of IDD9
Calculated from sera obtained from 4 non-diabetic Caucasian children and laboratory personnel (mean age 13.7 ± 7.7 years, 51 men / 43 women). None of these control individuals were positive for ICA. At the position of A14 at a concentration of 0.15 ng / ml
Optimum results were found using monoiodinated human insulin radiolabeled with ¹²⁵ I. More recently, improved specificity of the assay could be achieved by measuring specific binding, which can be displaced through the addition of excess unlabeled insulin.

IVGTT Procedure To examine the response of insulin to glucose stimulation, an intravenous glucose tolerance test (IVGTT) was performed.
(Srikanta, S., OP Ganda, GS
Eisenbarth and JS Soeldonah (S
oeldner) [1983] N. Engl. J. Med. 308: 322-325), the amount of glucose administered is 0.5 g / Kg given exactly intravenously over 2-4 minutes. Serum sample after glucose injection-1
Collected at 0, 1, 3, 5, 10, 15, 30, and 60 minutes. Insulin deficiency is less than the control population of non-IDD individuals with both ICA negative insulin levels at 1 and 3 minutes.
It is defined as when it is the 3rd percentile (n = 15
0, age 20.3 ± 10.2 years, range 2 to 59 years). 70 for this analysis
Insulin values less than μIU / ml (<3rd percentile for 150 controls)
c).

HLA DR Blood Grouping Method The HLA DR blood grouping method is described in Van Rood and Van Leuwen (Van Rood), JJ, A.
Van Leuwen, and JSPloem [1976] Nat
ure 262: 795-797). Standard control antisera was used for blood typing.

Construction of a Human Islet cDNA Library A cDNA library for human islet cells was constructed in a lambda gt11 vector (Young, RA, and Davis (Dav).
is), RW [1983] Proc.Natl.Acad.Sci. USA 80: 1194-11
98). Messenger RNA was isolated and purified by Chilgwin.
(win) et al. were isolated and purified from approximately 200,000 purified human islets according to a modification of the method of (Chirgwin,
JM, AE Przybyla, RJ McDonald (M
acDonald), WJ Rutter (1979) Biochemistry
18: 5294-5299). Place the island in 1M 2-mercaptoethanol and 0.1%.
Dissolved in 4M guanidine thiocyanate containing 1M Tris-HCL (pH 7.4). The lysate was layered on a 5.7M cushion of CsCl and 3 hours in a Beckman SW50.1 rotor for 17 hours.
Centrifuged at 7,000 rpm. The RNA pellet was removed and resuspended in sterile water, followed by precipitation with 3 volumes of absolute ethanol. This last step was repeated three times to ensure that a clean preparation of RNA was obtained. Using human islet total RNA as a template for the cDNA against the library, and enzymatically in the presence of human placental ribonuclease inhibitor, oligo (dT) primer, and four deoxyribonucleotide triphosphates Reverse transcription
Synthesized using ze (reverse transcriptase). Mix the mixture at 40 degrees
The cells were cultured for 2 hours. A second strand of cDNA was synthesized using the reaction product of the first strand and the enzymes E. coli DNA polymerase I and E. coli ribonuclease H. The mixture was incubated sequentially at 12 ° C. for 60 minutes, then at 22 ° C. for 60 minutes, followed by heat inactivation at 65 ° C. for 10 minutes. Mix the mixture to ensure that the ends of the cDNA are flush ends.
Incubated with T4 DNA polymerase at 10 ° C for 10 minutes. This mixture of double-stranded cDNA (ds-cDNA) was then purified using extraction with an equal volume of buffer-saturated phenol / chloroform, followed by extraction with an equal volume of chloroform.
Unincorporated nucleotides were removed by the addition of an equal volume of 4M ammonium acetate, followed by the addition of 2 volumes of anhydrous ethanol. After cooling on dry ice for 15 minutes, the mixture was centrifuged at 12,000 rpm in a microcentrifuge for 10 minutes. This precipitation step was repeated twice to ensure complete removal of unincorporated nucleotides.

The enzyme E in the presence of s-adenosyl-methionine
Starting with a 1 hour incubation at 37 ° C. with coRI methylase, ds-cDNA was prepared for cloning into the vector by a series of modification steps. This methylation c
Synthesize DNA overnight at 15 ° C in the presence of T4 DNA ligase and ATP
It was linked to EcoRI linker. Cohesive EcoRI ends were generated on linker-linked cDNA by digestion with EcoRI endonuclease at 37 ° C for 5 hours. Following this step, the excess linker was removed from the mixture by column chromatography. The material eluted in the void volume of this column represents the ds-cDNA that has been methylated and linked to the EcoRI linker and selected in size.

This material was ligated to a lambda gt11 vector that had been digested with EcoRI and treated with alkaline fospase. Ligation was performed in the presence of T4-DNA ligase and ATP.
Progressed overnight at ° C. Each ligation mixture was packaged into phage lambda coat protein using a commercially available extract (Strategene). The level of packaged phage was determined by titration on E. coli host strain Y1088. The complexity of the library was measured to be approximately 2.0 × 10 ⁶ pfu and bromo-chloro-indolyl-galactoside (BCIG)
Of these, 81% contained the inserted cDNA as detected by the absence of color on the indicator. Screening of the recombinant library with oligonucleotide probes
Amplification prior to ning.

The following describes the procedure including the best mode for carrying out the present invention. These examples should not be construed as limiting. Unless stated otherwise, all percentages are by weight and all solvent mixture proportions are by volume.

Statistical analysis includes both chi-square analysis and Fisher's exact test after the Yate correction factor has been applied (Snedecor).
(Snedecor), TN and WG Cochran [1967] S
tatistical Methods. Iowa State University Press, Ames, Iowa).

Example 1 Association of 64KA with IDD Five groups of persons were selected for the 64KA study. There were 31 newly diagnosed patients with IDD defined according to established National Diabetes Data Group (NDDG) standards (National Diabetes Data Group [1979], Diabetes Classification and Diagnosis and -Other categories of Sintolerance-, Diabetes 28: 1039
-1057). 26 non-diabetic controls who also lack ICA and have no known family history of autoimmune disease;
Nondiabetic ICA and IAA in said newly diagnosed patient
Forty relatives lacking thyroid; 36 nondiabetic but IC
A positive individuals, 28 of whom are first-degree relatives of the proband with IDD, 4 are apparently healthy normal controls, and 4 have autoimmune Addison's disease or thyroiditis Patients; 5 unaffected relatives with IDD who are ICA-negative but have high titers of IAA-positive; and sera before their documented onset of IDD Twenty-eight additional people gathered in the country were also tested.

The above patients were two of the ongoing prospective ICA screening population studies of more than 5,000 first-degree relatives of IDD probands, and 8200 normal control individuals Approximately 50 of normal control
00 were school students. All sera were stored at 20 ° C. for 1 week to 7 years prior to autoantibody analysis.

64KA was extremely susceptible to this disease in that it was present in 84% of newly diagnosed IDD patients and completely absent from the population of subjects tested (p <0.0001). It turned out to be specific.

Example 2-64KA as an Indicator of IDD Serum samples from 28 individuals were analyzed for 64KA to analyze the ability of 64KA to actually predict the onset of IDD.
Serum samples were collected prior to the onset of any clinical appearance of IDD. At later points in their lives, each of the 28 individuals developed clinical symptoms of IDD. These 28
Twenty-five of the people developed clinical symptoms before age 35.

It was found that 64KA (96%) was present in 24 of 25 people under the age of 35 at the onset of clinical IDD. Tested serum samples from these individuals have been collected somewhere between 3 months and 7 years prior to the clinical onset of IDD (mean 27 months). Seven years before the onset of clinical symptoms, the presence of 64KA occurred.

The same serum sample was tested for the presence of ICA and IAA. ICA was present in 22 of the 28 samples (78%) and IAA compared to 64KA which was present in 82% of all cases.
Was present in only 13 (48%) of the 27 samples. These results are shown in Tables 1-4. Obviously ID
The predicted value of 64KA for D is the best of all three autoantibodies.

In another analysis of 9 IDD patients at the onset of the disease, retention of responding SUSTACAL ^™ versus loss of islet cell insulin conservation correlated better with 64KA than ICA. Thus, 64KA also provides an indication of the amount of β cells remaining.

[0046] Table 1 64,000 M _r frequency group n 64 kA ICA control 26 0 (0) autoantibodies 0 (0) IDD 31 26 newly diagnosed (84) 25 (81) ICA- / IAA- relatives 40 1 (2 ) 0 (0) ICA + relatives 28 23 (82) 28 (100) ICA + other 8 7 (87) 8 (100) ICA- / IAA + <age 5 5 4 (80) 0 (0) (ICA Islet cell autoantibodies, IAA ...- insulin autoantibodies)

Table 2 Identification of non-diabetic patients at risk of IDD due to ICA positive Summary IDD-related autoantibody Plasma insulin IVGTT Age of HLA-DR confirmed ICA ^* 64K IAA (1 + 3 min ( percentile )) M 6 3,-IDD family survey 80 +-144 (40) M 9 3,1 IDD family survey 40 +-57 (<1) F 11 3,4 IDD family survey 80 + + ND M 11 3,4 IDD family survey 80 + + 105 (20) F 12 4,3 IDD family survey 40 +-194 (60) F 13 2,4 IDD family survey 40 + + 62 (<3) M 17 3,4 IDD family survey 20 +-67 ( <3) M 20 2,4 IDD family survey 40 + + 58 (<1) F 29 3,4 IDD family survey 40 +-57 (<1) F 30 3,-IDD family survey 160 +-82 (10) F 33 3,8 IDD family survey 20 +-125 (30) F 34 3,4 IDD family survey 40 +-167 (50) M 44 4,1 IDD family survey 20--ND F 10 4,8 IDD family survey 360 +-98 (20) F 38 4,-IDD family survey 20 + + 71 (5) M 7 4,8 IDD family survey 40 + + 114 (25) M 7 3,4 IDD family survey 20 + + 116 ( 25) M 10 4,-IDD family survey 160 + + 52 (<1) F 10 5.9 IDD family survey 20 + + 128 (30) F 43 3,1 IDD family survey 80--293 (70) F 13 1, 3 IDD Family survey 80--100 (20) M 5 2,4 IDD family survey 40 +-125 (30) F 39 4,-IDD family survey 80 + + 112 (25) M 11 3,4 IDD family survey 40 + + 56 (<1) F 35 4,7 IDD family survey 40 +-420 (90) M 43 3,7 IDD family survey 160--ND F 38 3,4 IDD family survey 20--184 (60) F 26 ND IDD Family Survey 40 + + ND M 8 3,4 School Survey 160 + + 67 (<3) M 10 3,4 School Survey 160--268 (75) F 13 4,6 School Survey 40 +-67 (<3 ) M 11 3,4 School survey 320 + + 124 (30) M 16 3,4 Addison's disease / thyroiditis 160 +-ND M 32 3,4 Addison's disease / thyroiditis 80 +-197 (60) F 41 3, 4 Addison's disease / thyroiditis 320 + + 69 (<3) M 12 3,4 Addison's disease / thyroiditis 160 +-149 (40) ^* JDF units

[0048] Table 3 IDD occurs before the test has been Mato Me IDD-related that was seen in the first serum sample that can be obtained from a patient self-antibody HLA-DR start age verification diabetes before period (month) ICA ^* 64K IAA M 3,3 5 IDD family survey 9-+-M 4,6 10 IDD family survey 34 20 + + F 4,7 11 IDD family survey 34 40 + + M 3,3 12 IDD family survey 8 40 +-F 4, 6 15 IDD Family Survey 3 80 +-M 3,4 15 IDD Family Survey 19 80 + + M 3,4 17 IDD Family Survey 14-+-M 3,4 19 IDD Family Survey 3 40 +-M ND 21 IDD Family Survey 32 80 +-F 4,-33 IDD Family Survey 14 10 + + F 4,5 38 IDD Family Survey 2 80--M 3,4 42 IDD Family Survey 42 80 + + M 4,6 5 IDD Family Survey 30 - ^** - ^** + M 3,4 15 IDD family survey 6-+ + M 3,4 30 IDD family survey 13 160 + ND M ND 4 IDD family survey 3 20--M 6,1 33 IDD family survey 14 - ^** +-F 4,6 37 IDD Family Survey 3 80--F 3,4 7 Peptide intolerance 1 + (NT) + + M 4,4 15 Plot sugar intolerance 2 80 + + F 3,6 17 Plot sugar Intolerance 2 20 +-M 2,3 18 Dextrose intolerance 12 40 +-M ND 27 Dextrose intolerance 6 40 + + F 2,3 40 Dextrose intolerance 19 + (NT)--F 1,4 9 School survey 3 40 + + F 3,4 12 School survey 44 320 + + F 3,3 24 Quercus disease 72- ^** +-F 1,4 14 Thyroiditis 75 + (NT) + + 22/28 23/28 13/27 (78%) (82 %) (48%) NT = JDF titer not measured ND = Not measured ^* = JDF unit ^** = Converted to autoantibody positive in later sample data

[0049] Table 4 IAA non-diabetic patients found at risk of IDD for positive Summary IDD associated autoantibodies plasma insulin IVGTT of age HLA-DR confirmation ICA ^* 64K IAA (1 + 3 minutes (percentile)) F 2 4,5 IDD family survey-+ + 129 (40) M 3 2,1 IDD family survey-+ + 27 (<1) F 5 ND IDD family survey-+ + ND M 0.5 1,6 IDD family survey-+ + 342 F 3 ND IDD Family Survey--+ ND

Example 3-Relationship between 64KA and IDD in NOD mice
Related Recently two types of animals (BB rats and NOD mice) have been developed that can serve as models for the study of human IDD. The use of animal models to evaluate as an indicator of IDD indicates that the number of high-risk relatives of a proband with a disease who must screen circulating autoantibodies to find susceptible individuals Due to the large number, the pre-diabetes period is invaluable because of the difficulty of study and for the long-term observations that need to document the natural history of β-cell failure in the disease. NOD mouse and B
B rats give important insights into the etiological event sequence involved in human IDD. Analysis of these animals also leads to an understanding of the nature of target islet cell autoantigens involved in the autoimmune process. We therefore IDD NOD mice and BB rats
64K autoantibodies that may be present both before and after the start of the test.

Male and female NODs and subjects of various ages (BALB
/ c, C57BL / 6, C2H / Hej) mice were obtained. NOD
The diagnosis of IDD in mice was characterized by their throat and weight loss and persistent hyperglycemia above 240 mg / dl. All diabetic animals per day to maintain health
Received daily insulin doses of 0.5-2.0 units of protamine-zinc insulin (Eli Lilly, Indianapolis, IN).

[0052] Brunstedt (Bru
ndstedt), J., Nielsen, JH, Lahnmark
(Lernmark), A. and Hagedorn study group, Methods in Diabetes Research, J. Lerne (Larne)
r), SL poll (Pohl) [John Willie & Sons (John
Willey & Sons), New York, 1987]) and isolated BALB / c islands and labeled with ³⁵ S methionine. RPMI1640 (GIBCO, New York) medium supplemented with islet cells
Wash islet cells twice (4 ° C.) in (2.0% fetal calf serum, 100 units / ml penicillin, 100 μg / ml streptomycin), followed by 20 mM Tris (pH 7.4), 150 mM NaCl, 1000 KIE / ml trayilose. Washes once in a buffer solution containing 1 mM and 2 mM PMSF. Cells were then incubated with 1% Nonidet P-40 (NP-4
0) (v / v) was dissolved in the above buffer. Unwanted substances were removed by ultracentrifugation (100,000 × g, 30 minutes). The lysate was incubated with normal mouse serum (Balb / c) (10 μl per 100 μl supernatant, 1 hour, 4 ° C.) and subsequently adsorbed against excess protein A Sepharose CL-4B. This culture / adsorption procedure was repeated three times. Aliquots (100 μl) of unbound (pre-cleared) lysate were incubated with 25 μl of either NOD serum or control serum for 18 hours (4 ° C.). For each test tube
100 μl of pre-swollen protein A Sepharose CL-4B was added and the reaction mixture was incubated for 45 minutes (4 ° C.). Protein A Sepharose CL-4B was then washed five times by centrifugation in 0.5% NP-40 buffer (200 × g, 10 seconds) and once by centrifugation in cold water (200 × g, 30 seconds). The bound immune complex was treated with 80 mM Tris (pH 6.8), 3.0% (w / v) sodium dodecyl sulfate (SDS), 15% sucrose, 0.001 bromophenyl blue and
Denatured by boiling for 5 minutes in sample buffer containing (5.0% v / v) β-mercaptoethanol. The beads of Sepharose were removed by centrifugation (200 xg, 1
Minutes), and the supernatant was electrophoresed through a SDS 12% polyacrylamide separation gel, followed by Coomassie staining and electrophoresis.
Tradography was performed.

The first test was to determine if 64KA had developed in NOD mice. Serum newly diagnosed
Eight female NOD mice with IDD (mean age at onset 104.5 ± 13.5 days, range 93-134 days) and BALB / c (n = 5), C
11 healthy control mice of the 57BL / 6 (n = 4), C3H / HeJ (n = 2) strain (2 males and 9 females, average age 97.7 ± 30.4 days, range 50-120 days) From serum.

At the onset of their diabetes, 64KA is NO
Although detected in 87% of D, 64KA was not detected in any of the non-diabetic control lines (P <0.001). NOD mice with newly diagnosed IDD have a high percentage of 64K
The presence of autoantibodies against, in combination with the absence of 64KA in age-matched, non-diabetic control mice, confirm that 64KA is specific for IDD and for its presymptomatic state. means. The presence of 64KA is also one of the first events to be detectable, and is generally present at or even before weaning, the earliest age at which insulitis is found.

Example 4 As Index of IDD in NOD Mice
To know about the natural history of 64K autoantibodies for 64KA IDDs , 5 mice (2 males and 3 females,
Serum samples obtained from ages 23-25) were analyzed and their pancreas was histologically examined for insulin. Insulin was detected as a distinct intraislet infiltration of lymphocytes on hematoxylone and eosin stained pancreatic sections containing 5 or more islets. 64KA was already present in 80% (4/5) of these mice. At the same time, early insulin was also present in three of the four mice whose pancreatic tissue was analyzed. Two animals tested at the age of 190 days or older (both males), which did not produce any IDD symmetrically, did not have 64KA, although both had insulinity.

Example 5-64KA in BB Rats In a study of IDD found in BB rats (n = 8), autoantibodies to the 64K protein appeared early in the natural history of the disease,
Significant specificity is shown as an IDD marker (absent in the control of 6 Lewis rats). People, NOD
Mouse, and the discovery of 64KA in BB rats, M _r 64,000
Autoimmunity to islet cell proteins is of considerable etiological significance, suggesting that 64K proteins may have important physiological functions in normal beta cells.

Example 6 Biochemical Characterization of 64K Antigen For biochemical characterization, the higher resolution of the gel of the 64K protein means that it has two components (63,500 and 64,000 M _r ).
Suggests that they can consist of doublets that differ by about 0.5 kilodaltons between. The 64K component shifts to the M _r 64,000 range under reducing conditions and can be found as a dimer (approximately 130K) under non-reducing conditions. 64
The K protein shows charge heterogeneity in that it has an isoelectric point of 6.4 and 6.7. The 64K component most likely represents an integral islet cell membrane protein due to the hydrophobicity observed in the immunoprecipitation of the detergent phase extract. "64K antigen" used in this patent application
The term refers to a compound, compounds or complexes that behave on a gel in the manner described herein. The 64K antigen can also be identified by immunoreactivity with the anti-GAD antibodies described below.

We have also found that immunoprecipitation (immunoprecipitation), detergent-extracted from human islet cell proteins with serum from persons with IDD, often results in the identification of the M _r 67,000 protein. We are U.S. school students and
Antibodies to this 67,000 M _r antigen were analyzed in a population of relatives of a proband with IDD. Autoantibodies to the 67,000 M _r protein are present in 60% (12/2
0) and only one of the 18 controls (5% P <0.001). 67,000 in this IDD population
Autoantibodies to M _r were often observed in people positive for (73% 11/15) 64KA, but only 67,000 M _r autoantibodies were found in only one patient lacking 64KA Did not. Autoantibodies to 67,000M _r protein are 64KA,
Only 1 in 34 non-diabetic first-degree relatives lacking both islet cell cytoplasmic and insulin autoantibodies
Only present in people (3%). Coomassie-stained SDS-PAGE
Analysis of the gel, the presence of autoantibodies against 67,000M _r protein was often associated with the protein of the strong staining of 70,000M _r. Gas phase amino acid sequencing of this 70,000 M _r protein
This reveals that the immunoglobulin V-III region chain is thereby directed to the IgM class of antibodies. Thus, immunity to the 67,000 M _r antigen may be increased with, but not limited to, IgM class antibodies. We also clearly showed that almost all 67,000 M _r autoantibody positive patients also contained 64 KA,
00M _r Autoantibody positive but 64KA negative is rarely seen. Therefore 64,000 and 67,000M be antigenically cross-reactivity between the _r islet cell antigens may be present, and epitope autoantibodies to these proteins these immunological cross-reactivity - that can be identified flop Believe in

As used in this application, "immunologically equivalent to the 64K antigen" refers to peptides that react with 64KA. The term 64KA or 64K autoantibody refers to a compound or compounds that react with the 64K antigen.

Since the 64,000 M _r protein is amphiphilic, we will observe whether the 64K protein is essentially restricted to the plasma membrane of the cell, and if so, the normal islet. Surface expression of this protein in cells (surface e
An experiment was performed to see if there was evidence for xpression).

Our biochemical characterization tests showed no evidence of glycosylation of the 64K protein, a common feature for surface expressed proteins. A series of broad enzyme spectra was used to detect common forms of protein carbohydrate glycosylation. This analysis includes the enzymes N-GLYCANASE ^™ (for N-linked carbohydrates), treatment for O-linked carbohydrates, neurominidase (for sialic acid groups) and β-endogalactosidase (for galactose groups). ) Was used. If one of the carbohydrate groups is present on the 64K protein, the enzyme will cleave the carbohydrate, the mobility of the 64K protein will increase upon subsequent SDS-PAGE analysis, and the reduced molecular weight Protein. After treatment of the protein with this particular set of enzymes, no difference in the mobility of the 64K protein in SDS-PAGE gels was detected. We also used a series of 19 broad and narrow spectrum carbohydrate binding lectins to thoroughly analyze the 64K protein for the presence of glycosylation. However, none of these lectins bound to the 64K protein. Thus, there appears to be no glycosylation in the 64K protein.

Finally, using standard techniques for radiolabeled cell surface proteins at ¹²⁵ I, none of the 64K proteins were purified from surface-labeled islet cells in serum from diabetics.
It has previously been shown that immunoprecipitation was not possible (Colmanda [1987] Diabetes 36,1432-1440). The study suggested that there is no evidence for surface representations. However, the general technique is labeling tyrosine (a relatively rare amino acid that may not be present in the extracellular portion of the protein), and the conditions were harsh in nature. Thus, it may be severe enough to destroy the antigen binding site of the 64K protein, and thus the antibody may no longer bind to the protein. However,
We used a newly described technique (Thompson J.
[1987] Biochemistry 26,743-750). It is much less harsh in chemistry and radiolabels lysine (which is a much more common amino acid than tyrosine) or the NH ₂ -terminal amino acid, and is less capable than the method used in Colman et al. It is shown to label cell surface proteins with much higher specific activity. Cell surface amino groups ¹²⁵ I sulfo succinate stains diyl (hydroxyphenyl) Purobione - were derivatized with bets by ¹²⁵ I (hydroxyphenyl) propionyl groups. However, even with these assay modifications, the 64K protein was not immunoprecipitated from intact cell surface labeled human islet cells.

To confirm that the 64K protein is indeed present in the plasma membrane, we made a crude membrane preparation of islet cells and used the reagent Triton X-114 to obtain from patients with IDD. Serum was used to immunoprecipitate the 64K protein. Next we developed an improved method for the detection of autoantibodies to the 64K protein.
This method did not use detergents and produced results that were easier to interpret and faster to perform. This method involves making a crude membrane preparation of ³⁵ S methionine labeled islet cells followed by trypsinization at 4 ° C. for 60 minutes (2 mg / ml 50 mM Tris buffer pH 7.
Four). This material is then centrifuged, followed by our standard immunoprecipitation technique. Following gel electrophoresis and autoradiography, the 40K band is observed in immunoprecipitates from sera of diabetics, but not in controls. Isolated 64K treated with tyrosine
We have already shown that the protein also displays 40K protein, so this 40K immunoprecipitated directly from trypsinized islets
The bands are very likely to represent the same protein. Assays for the 40K protein are not only improvements for easier detection of 64KA, but also immobilon (immolone).
gas phase sequence analysis using bilon) membranes yields products that are much more familiar. This improvement of the reason for 40K fragment Applied Biosystems gas phase Sekuensa - no major hydrophobic region caused difficulties in our determination of using the Inobiron (IMMOBILON) N-terminal amino acid sequence of 64K protein bound to ^TM This is because the possibility is very strong. Gas phase sequencer with protein having strong hydrophobic region
It has been established that the protein is solubilized in the solvent used, and thus has become a major technical handle for direct sequencing of this protein. The 64,000 M _r than the range, the 40,000 _r range is lower background staining than 64K protein in silver staining of the SDS page gels, much easier to be able to see the 40K protein It is.

Gas-phase amino acid sequencing of either the intact 64K protein or the 40K tyrosine-cleaved fragment revealed NH ₂ -terminal blockade. This information, along with the solubility of the 40K fragment, suggests that the 40K protein is attached to the β-cell plasma membrane near its COOH terminus. Specifically, we would predict that a string of hydrophobic amino acids occurs in the nearest 230 amino acids at the COOH terminus, followed by a series of 3-4 highly charged amino acids. .

Example 7-Phosphorylation of the 64K Protein Many key regulatory proteins exist in phosphorylated or dephosphorylated form in cells, and their steady state levels of phosphorylation are interconverted. This is reflected in the relative activities of proteinase and protein phosphatase, which catalyze the process. Phosphorylation triggers small conformational changes in these proteins that alter biochemical and biological properties. Hormones and other extracellular signals convey information via this mechanism.

Although the full role of phosphorylation within human islets has not yet been identified, serum and islet cells have been used as a method to identify all phosphorylated proteins that have a major role in autoimmune destruction caused by IDD. An in vitro (in vitro) kinase assay using the protein was applied.

Two groups of individuals (individuals) were selected for the in vitro phosphorylation assay. Diagnosed according to established National Diabetes Data Group (NDDG) criteria (National Diabetes Data Group Diabetes. 1979; 28: 1039-5).
7) There were 7 newly started IDD patients referenced at the University of Florida Diabetes Clinic. We also tested eight non-diabetic controls without any known family history of autoimmune disease. Human pancreatic islets isolated from cadaver pancreas were 91.2 ± 2.3% pure and had an insulin content of 0.4 ± 0.1 mU per islet. Following 48 hours of in vitro culture, including the use of normal human serum alone, islet cells were immunoprecipitated as follows. Immune complexes were prepared with serum from either patients with IDD or controls, but the serum was cultured with islet cell detergent extract,
Subsequently gamma - ³² cultured with P-ATP, and the products phosphorylated in vitro were analyzed on SDS-PAGE gels. Islet cells (1,000 / test serum) were treated with 50 mM Tris buffer (2% Triton X-114, aprotinin, PMSF, NaF)
(3 hours, 4 ° C.). The detergent lysate was cultured with normal human serum (100 μl / 1,000 islets) and subsequently adsorbed to excess Protein A Sepharose CL-4B (Pharmacia). Aliquots containing unbound (precleared) lysate were incubated with either IDD or control serum (25 μl, 18 hours, 4 ° C.). Following incubation of the immunoglobulin with protein A Sepharose CL-4B (2 hours, 4 ° C.), the complex was then incubated with 0.1% SDS, 1.
50% with 0% Triton X-114, and 2 mM EDTA
Tris-HCl (pH 7.4) three times, and 50 mM Tris-HCl
(PH 7.4) twice. Kinase reaction is 10 mM MnC
50 mM H containing l ₂ , 10 mM MgCl ₂ , 1% Triton X-114 and 20 μCi of adenosine (gamma- ³² P) 5 ′ triphosphate (Amersham,> 5,000 Ci / mmol).
Started by adding 10 μl of EPES buffer (pH 7.4). The precipitate was suspended and incubated at 30 ° for 10 minutes. The reaction was stopped by the addition of electrophoresis sample buffer and heated at 95 ° C. for 4 minutes. Discontinuous SDS for ³² P labeled products
Separation was performed on a 15% polyacrylamide separation gel, followed by autoradiography on an intensifying screen (Kodak, X-omat AR5) for 6-72 hours. Serum of IDD patients specifically M _r 64,000
The protein was immunoprecipitated, which was not observed in the control serum.

To analyze whether the identification of M _r 64,000 protein detected by the peptidase reaction is the same as that observed in the immunoprecipitation of metabolically labeled islet cells, - [0068] Finally vitro quinic A tryptic peptide mapping experiment was performed. IDD serum treatment ³⁵ S methionine labeled islet cell M _r 64,000 protein immunoprecipitated showed original similar tryptic peptide mapping of the in vitro phosphate warp Le of M _r 64,000 protein.

[0069] In terms specific to, 64,000M _r and 125,000M _r
Cut protein from SDS-PAGE gel and remove 1 mg protein
/ Ml of trypsin (3-30 minutes, 4 ° C). The resulting protein was run again on an SDS-PAGE gel. 125,000M _r
The protein was not affected by this treatment. But 64,0
00M _r protein increased motility to the 40,000 M _r range. Therefore the 64,000 ³² P-labeled islet cell protein shows the M _r 64,000 ³⁵ S-methionine labeled island cellular protein the same M _r as we have observed (40,000) when treated with trypsin.

Example 8 Alternative biochemical 64K autoantigen
Unique identification of the 64K autoantigen is glutamate decarboxylase (GA
We have established that it shows homology to D) and is immunoreactive with anti-GAD antibodies. GAD enzymes are localized in the brain and in the β cells of the pancreas. The biochemical characterization of GAD in the literature has been limited to its role in the formation of gamma-aminobutyric acid (GABA), a major inhibitory neurotransmitter (neurotransmitter) in the mammalian brain. The GAD sequence disclosed by Kobayashi et al. (Kobayashi Y., DL Kaufmann and AJ Tobin [1987] "Glutamate decarboxylase cDN
A: Nucleotide sequence encoding an enzymatically active fusion protein "J. Neuroscience 7 (9): 2768-2772) is reproduced in Table 6 herein.

Antigenic cross-reactivity between GAD and the 64K protein was confirmed. A series of immunoprecipitations using autoantibodies to the 64K protein and monoclonal antibodies directed against GAD confirmed competition for binding to the 64K protein. Monoclonal antibodies reactive with the GAD protein are readily available to those skilled in the art, for example, from American Type Culture Collection (ATCC) 12301 Rockville, Midland, Midland, USA.
The monoclonal antiserum GAD-1 (ATCC HB184) is specific for binding to GAD.

For these studies, human islet cells were metabolically labeled with ³⁵ S methionine, and the detergent phase was separated as described above. The islet cell extract was then split into two samples of equal volume. Each sample was precleared twice with either control (no 64K autoantibody) or IDD (64K autoantibody containing) serum. Preclear with IDD serum
The step removes or significantly reduces the amount of 64K protein from the islet cell protein preparation, while pre-clearing in the control serum does not significantly affect the amount of 64K protein.
Both precleared samples (control and IDD) were individually immunoprecipitated with monoclonal antibodies against GAD. The results showed that pre-clearing with control (64K autoantibody negative) serum did not affect the ability of the monoclonal anti-GAD antibody to immunoprecipitate the 64K protein of islet cells. However, IDD serum containing 64K autoantibody as a preclearing agent significantly reduced the ability of the monoclonal anti-GAD antibody to immunoprecipitate the 64K protein of islet cells. This means that 64K autoantigen and GAD
Demonstrating antigenic cross-reactivity between

Serum from individuals with 15 new onset IDD patients and samples obtained 2-3 years after IDD onset, 5
Serum from 65 individuals was tested for anti-GAD activity, including 10 individuals with IDD for a duration of 1010 years, 20 control individuals and 5 individuals at high risk of IDD. Immunization in the form of autoantibodies against GAD was clearly associated with IDD.

[0074] As described in Example 6 on, to immunoprecipitate the islet cells in IDD sera sometimes not only results in the identification of 64K protein, several people or is immunoprecipitated a protein of 67,000M _r Was. This observation was particularly noted when the detergent phase preparation did not produce a sharp separation between aqueous (primarily cytosolic) and detergent phase (membrane associated) proteins. Thus, the 64K protein can be found in islet cells in two forms. That is, 67,000 M _r soluble (cytosolic) form and 6
4,000M _r hydrophobic (membrane) type. The major autoantigen in IDD is of type 64,000M _r , but there is also autoimmunity to type 67,000.

The gas phase amino acid sequencing of the 64K autoantigen isolated from ³⁵ S-labeled human islet cells is shown in Table 6 below.
It shows homology between the K protein and the GAD sequence. This homology was discovered as follows.

The 64K autoantigen was isolated from ³⁵ S-labeled human islet cells as described above. After SDS-PAGE, the proteins from the gel were transferred to IMOBILON-P ^™ membranes and made visible by autoradiography. The 64K band was localized, excised, and treated with cyanogen bromide, which cleaves the peptide chain at methionine residues. The resulting peptide fragments were eluted from the PBDF membrane and re-run on SDS-PAGE before being transferred to IMOBILON-P ^™ . Oh the band - autoradiography - and it so that can be seen in, and cut out the fragments in the M _r 33,000, and subjected to gas phase peptide sequencing. For example, the following sequence profile was obtained: MET-ILE-PRO-GLU-VAL-LYS

We have found a high degree of similarity at two levels between the 64K amino acid sequence and GAD. As previously mentioned in the first, when our 64K protein treated with cyanogen bromide, it produces peptide fragments of 33,000M _r. One would expect this fragment to consist of approximately 300 codons (900 nucleotides). Examination of the GAD sequence shown in Table 6 shows an open reading frame of 625 codons and 1875 nucleotides in the total protein. Nucleotide positions 916-933 of the sequence code for the next amino acid. PHE-PHE-PRO-GLU-VAL-LYS

Thus, the 64K peptide sequence shows a high similarity in amino acid sequence to the GAD sequence. In addition, this GAD peptide sequence, when a PHE or MET amino acid substitution is made in that part, when cleaved with cyanogen bromide,
It is located in the region of the protein where 33,000 M _r peptide is expected to be observed. Second, the nucleotides have differences between the elucidated amino acid sequences (MET-PHE-PRO-GLU-VAL-LYS), and nucleotide sequences 916-933 involve only minor nucleotide substitutions. In fact, methionine rather than phenylalanine appears to be in the first position of the 64K sequence of this protein, since the peptide is cleaved with cyanogen bromide, which is specific for methionine. So we have our DNA
The above sequence information was used to construct a probe to identify the gene coding for the human pancreatic cell 64K protein from the library.

A mixed oligonucleotide having the following sequence was synthesized: The synthetic oligonucleotide, T4 polynucleotide Kina - peptidase and gamma - ³² using P-ATP is radiolabeled, and human island cDNA library - Pro - was used as a drive. Approximately 120,000 clones were screened and three positive plaques were detected. Thus, DNA coding for the 64K protein was identified and isolated.

The DNA coding for the 64K protein can be introduced into an appropriate prokaryotic expression vector for producing a large amount of the recombinant protein. This product can be used in assays (eg, ELISA, RIA) to determine levels of anti-64K antibodies in pre-diabetic patients or to diagnose pre-diabetic individuals.
Etc.) can be used for development. Suitable eukaryotic expression vectors can also be used for production in native 64K eukaryotic cells. This overcomes the possibility that bacteria will not be able to properly process the 64K precursor and will be recognized by autoantibodies.

Recently, considerable interest has been focused on proteins recognized by effectors of the immune system (Benjamin D., J. Benzowski, I. East, F. Gard.
C. Hannam, S. Reach [1984] "Protein antigen structure recognized by T cells: antigen structure of protein; re-evaluation"
Ann. Rev. Immunol. 2: 67-101). Based on algorithms that predict T and β cell epitopes, a great deal of knowledge is available about T cell and antibody binding epitopes (Delici, S. and JA Berzowski [1985] “T
Cell antigen sites tend to be amphiphilic structures. ''
Natl. Acad. Sci. 82: 7048-7052; Hopp, TP and K.
R. Woods [1981] "Prediction of protein antigenic data from amino acid sequence" Proc. Natl. Acad. Sci. 78: 3824.
-3828). In addition, experimental studies on model protein antigens such as cytochrome C have allowed for the strict localization of T and B cell locations and the characterization of critical amino acid residues important for recognition ( Militi, DR [1989] "Synthetic T and B cell recognition sites; suggestions for vaccine development," Dickson, FJ, Advances in Immunology. 45 Academic Press, London 1989; 195-282). But,
In autoimmune diseases specific to human organs, the lack of knowledge of the nature and availability of self antigens has limited the study of epitopes (McKay, IR and ME Gershwin [1989] Molecular Basis of Mitochondrial Self-Reactivity during Cirrhosis of Sexual Bladder ", Immunol. Today 10:
315-318). In organ-specific autoimmune diseases, the determination of the nature of the self-antigen region recognized by T cells and autoantibodies allows the design of antagonists and immunomodulatory compounds to drive tissue damage.

One aspect of the present invention is the identification of immunoreactive peptides consisting of the 64K protein epitope. For these studies, the GAD cloned cDNA template can be used in conjunction with DNA amplification techniques to represent a selected segment of the 64K antigen as a recombinant protein in E. coli. On average, 80 amino acid residues
Multiple overlapping different recombinant fragments, and larger fragments encompassing the entire extracellular region of the molecule, can be produced, and the autoantibody binding site has been labeled with a radioisotope. It is analyzed by immunoprecipitation of the product of the expressed bacteria growing in the presence of the precursor. Using these procedures allows for epitopic mapping of the correct antigenic amino acid sequence of the GAD molecule. 64K
Synthetic peptides consisting of an antigenic epitope of a protein can be used for both diagnostic and therapeutic purposes, as described herein.

[0083] One such peptide of the invention consists of the PRO-GLU-VAL-LYS sequence described above. Thus, one peptide that can be used as described herein for diagnostic or therapeutic purposes can include from about 8 to about 30 amino acids including the PRO-GLU-VAL-LYS sequence. Most advantageously, such a peptide may be from about 10 to about 20 amino acids. The peptide may have one sequence of the GAD protein or the 64K islet cell protein itself. It is well known that there are several forms of GAD that may differ in the exact amino acid sequence. The isolation of GAD from rats, pigs, cattle, cats and humans has all been reported. See, e.g., Clam, DS, LD Burnett, JL Joseph, LC Harrison (1991) "Human glutamate decarboxylase cD from brain and pancreatic islets.
NA Cloning and Partial Nucleotide Sequences "
m. Biophys. Res. Comm. 176 (3): 1239-1244; Spink
DC, TG porter, S. C, DL Martin (1987)
"Reactive Kinetically Different Types of Glutamate Decarboxylase in the Rat Brain Brain Res. 421: 235-244; Denna LA, SC Way, HS Phosphorus, CT Phosphorus, JY U (1987)" Brain L-glutamate decarboxylase, purification and subunit structure "Proc. Natl. Acad,
Sci. USA84: 668-672; Chang YC, DI Gottlie
(1988) "Characterization of Proteins Purified with Monoclonal Antibodies to Glutamate Decarboxylase", J. Biol.
Neurosci. 88: 2123-2130; Spink DC, TG Porter, SJ C, DL Martin (1985) "Three kinetic distinctions of glutamate decarboxylase from pig brain. Characterization of different forms "Biochem. J. 231: 695-
703; Blinda-Man JM, M. Maittle, L. Osora, P. Mandel (1978) "L-Glutamase from the Human Brain"
Purification and Some Properties of Todecarboxylase "Eur.
Biochem. 86: 143-152; and Fan WM, L. Lead Hokwet, E. U., JY U. (1990) Proc. Natl. Acad.
Sci. USA 87: 8491-8495. Each of these GAD sequences contains 6
Epitopes that are immunoreactive with 4K autoantibodies may be included.
Thus, in accordance with the present invention, peptides from these various forms of GAD can be used for therapeutic or diagnostic purposes.
Also, these various GAD sequences can be used to provide information about the types of amino acid substitutions that can be made in the 64K sequence while maintaining critical immunoreactivity. The synthetic peptide of the present invention can be represented as follows. A-Pro-Glu-Val-Lys-B

In the formula, both A and B both have a length of 0 to
Consisting of a sequence of about 20 amino acids, wherein A and B are such that the synthetic peptide is totally immunoreactive with the 64K autoantibody or is recognized by T cells or β cells.
It has sufficient homology with 64K. This homology can be, for example, about 90% with the 64K or GAD protein. The synthetic peptides described herein are typically less than about 50 amino acids, and in any event less than the full length of a GAD or 64K protein.

The 64K protein we describe here exists in at least two forms. That is, a form or forms having a size of about 63-65K, and a form having a size of about 67K. These forms are clearly distinguished. The lower molecular weight form was observed as a doublet as described above. As noted above, the presence of multiple forms of 64K may indicate that a person at risk for diabetes may have some form of
64K has been made advantageous for diagnostic purposes in determining whether it has antibodies to the 64K antigen. Although the various forms of 64K share homology, they also have non-homologous regions that can give rise to different immunoreactivity. Therefore, the most complete diagnosis or treatment plan is 64
Multiple peptides (or hybrid peptides) consisting of sequences that are immunoreactive with each form of autoantibody to K
Would include the use of In making peptides for diagnostic or therapeutic use in accordance with the present invention, one of ordinary skill in the art can use sequences from various forms of GAD or 64K, or even sequences modified with various amino acid substitutions. Can be used. To be within the scope of the present invention, these peptides
Immunoreactive with antibodies to the 64K protein.

The term analog used in this application is defined as
A compound that is substantially the same as another compound, but that has been modified, for example, by adding additional amino acids or side chains. As used herein, the mutant (mut
The terms ants and variants refer to an amino acid sequence that is substantially identical to another sequence, but has an amino acid substitution at a certain position in the amino acid sequence. A fragment refers to a portion of a longer amino acid sequence.

The present invention encompasses the use of specific GAD and 64K amino acid sequences. The invention further encompasses analogs and mutants of these sequences, as well as fragments and analogs and mutants of the sequences. These analogs, mutants, and fragments are included within the scope of the present invention, as long as the analog, fragment, or mutant retains substantially the same immunoreactivity as an antibody to 64K. For example, making conservative amino acid substitutions is within the skill of the art. For example, amino acids can be divided into the following classes: basic, hydrophobic, acidic, polar,
And amides. Substitutions in which one class of amino acid is replaced by another amino acid of the same type are included within the scope of the invention, as long as the substitution does not substantially alter the biological activity of the compound. The following table lists examples of amino acids belonging to each class.

Table 5 Examples of Class Amino Acids of Amino Acids Nonpolar Ala, Val, Leu, Ile, Pro, Met, Phe, Trp Polarity Gly, Ser, Thr, Cys, Tyr, Asn, Gln Negatively charged Asp , Glu plus charged Lys, Arg, His

In some cases, non-conservative substitutions can also be made. Construction of various mutants can occur by cassette mutagenesis of synthetic genes. This step is known to any person skilled in the art. To the extent that these substitutions do not substantially alter the immunological reactivity with the antibody to 64K, the resulting compounds are included within the scope of the invention. Conservative amino acid substitutions are merely an example of one type of modification that is within the scope of the present subject matter.

These synthetic peptides will be used in the development of diagnostic reagents to be used in similar assays developed with the recombinant proteins described above. Synthetic peptides can also be used to make vaccines for the prevention of IDD, similar to its purpose for vaccine development in general (Militi [1989], supra).

Example 9 Insulin-dependent diabetes mellitus
Blood against glutamate decarboxylase
Identification of T Lymphocyte Response Insulin-dependent diabetes mellitus (IDD) appears to result from autoimmune destruction of the insulin-producing pancreatic β-cells. The early appearance and high frequency of antibodies immunoreactive with glutamate decarboxylase (GAD)
This suggests that this protein may be a target antigen in diabetes. However, since the disease is believed to be dependent on T lymphocytes, we have peripheral blood for children with newly diagnosed IDD and recombinant GAD from people at high and low risk of developing IDD. The mononuclear cell response was analyzed.

The mononuclear cell infiltration of pancreatic islet cells (Insulitis) at the clinical onset of IDD is a pathological proof of the disease. The lesions have been proposed primarily for T-lymphocytes, and it is these cells that are thought to play a pivotal role in the pathogenesis of IDD disease.
Evidence supporting this idea includes the observation that ICA titers do not rise, but that the disease recurs in long-term diabetic patients who have received pancreatic transplants. Immunosuppressants have also been shown to be effective in improving the metabolic potential of individuals with IDD. In a spontaneous animal model of IDD (NOD mice), diabetes can be introduced into non-diabetic mice through conversion of CD4 and CD8 T lymphocytes. Disease in NOD mice can also be prevented through the use of anti-CD4 therapy. However, the nature of the target antigen playing a role in this autoimmune phenomenon remains unclear.

[0093] We here at the beginning of the IDD,
We also examined whether T-lymphocyte reactivity was characteristic in populations at increased risk of disease.

Recombinant GAD antigen was replaced with rat brain GAD65 PET
Manufactured from expression. Peripheral blood mononuclear cells
(Hypaque) isolated by density centrifugation
(Sigma, St. Louis, Missouri). 1 × 10 ⁶ mononuclear cells were cultured for 7 days in the presence of 10 μg / ml antigen in a 24-well tissue culture tray. Six days after culture, 1.0 μCi ³ H thymidine (NEN, Wilmington, Del.) Was added to each well. Media was harvested semi-automatically after 18 hours and thymidine incorporation was assessed by liquid scintillation counting. Cell proliferation was expressed as stimulation index (SI) minus cpm taken up in the presence of antigen divided by cpm taken up in the absence of antigen (medium only). An SI of 3 or more (≧ 3) was determined to be a statistically positive response.

Our studies have shown that T-lymphocyte responsiveness to GAD occurs at the onset of IDD and occurs in persons with an increased risk of IDD. This T cell responsiveness to GAD is absent in healthy people.
These findings support the hypothesis that GAD contains antigenic epitopes on T-lymphocytes and may therefore be involved in the pathogenesis of IDD disease. T- in patients with newly initiated IDD
There is no correlation between the degree of lymphocyte response and ICA titer, age at onset of disease, gender or HLA type. Thus, this T-lymphocyte reactivity to GAD contributes to providing additional information and provides another way to diagnose the disease.

Example 10-Organisms for the Detection of 64K Autoantibodies
Collection of body fluids More than 500 microliters of whole blood was collected from humans for testing for 64KA autoantibodies.
Blood was drawn directly into glass evacuated vessels, or drawn for injection, and subsequently transferred to glass evacuated vessels. In order to obtain serum (blood without clotting factors), the common vacuum vessel tube used is called a red top tube (no sodium heparin) or a serum separator (STS) tube. When a common red top tube was used, the tube was allowed to clot (for a period longer than 10 minutes) and the clot was removed.
During this time, any sample tube can be centrifuged at 1000 rpm for 5 minutes at room temperature. The serum in the sample was removed, placed in a plastic storage bottle, and sealed tightly.
Samples can be frozen at -20 ° C until 64K autoantibody analysis.

Example 11-Used to test biological fluids
Preparation of 64K Antigen for Preparation Before human islet cell preparations can be tested for 64K autoantibodies, they must be isolated and metabolically labeled to provide a source of 64K antigen. The human islet cell preparation was prepared according to the method of Ricodi et al.
Lacy, PE, Finke, EH, Olac
k), BJ, and Scharp, DW [1988] Automated Methods for Isolation of Human Pancreatic Islets, Diabetes 37: 413-420). Can be isolated from pancreatic donations. Islet cells were 6,000 in 50 ml CMRL medium.
Can be stored in T-150 plastic tissue culture flask at the concentration of islets / T-150 flask.

Before testing, remove and discard the CMRL medium. Add 50 ml of RPMI 1640 medium (16 mM glucose, 100 μU / m
l penicillin, 100 μg / ml streptomycin and 2% v / v
(Supplemented with normal human serum) and cultured overnight at 37 ° C. in culture medium in an atmosphere of 95% air / 5% CO ₂ . Islets were collected in centrifuge tubes at a concentration of 50 ml of medium per tube.

The test tube was capped and 3 rpm at room temperature at 1000 rpm.
Centrifuge for minutes. The supernatant was removed and discarded. Small amount (approximately
4 ml) RPMI-labeled medium (16 mM glucose, 20 mM HEPES, 2 mM)
(Supplemented with% normal human serum and no amino acid methionine) was added to each tube for approximately 30 seconds per tube with gentle mixing. Next 50 ml of all islet cells
Collected in a polypropylene centrifuge tube at
Centrifuge for minutes. The supernatant was removed and discarded. Islet cells next
Resuspended in RPMI labeled medium (1000 islets / ml medium) and transferred to 100 mm plastic tissue culture media Petri dishes. Island 15 minutes 37
C. (95% air / 5% CO ₂ ) followed by ³⁵ S methionine
It was added to the medium at a concentration of 0.5 mCi / 1 × 10 ⁴ cells. Islet cells again
The cells were cultured at 37 ° C. for 5 hours (95% air / 5% CO ₂ ). The medium in the Petri dish was aspirated without agitating (or collecting) the islet cells and the medium was discarded.

The cells were then added with an equal volume of RPMI 1640 medium at a concentration of 1000 islets / ml and incubated at 37 ° C. (95% air / 5% CO ₂ )
For 30 minutes. The islet cells and their medium were collected in 15 ml conical polypropylene centrifuge tubes. The test tube is filled with 20 mM Tris (pH 7.4), 150 mM NaCl, 1000K to a mark of 15 ml.
Fill with IE / ml aprotin and 2 mM phenylmethylsulfonyl fluoride, cap, centrifuge (1000 rpm, 5 minutes),
And the supernatant was discarded. The tube containing the pellet was then frozen (-80 ° C).

Example 12-Detergent extraction of 64K antigen and 64K autologous
Testing of Biological Fluids for Antibodies Islet cells were slowly thawed in tubes on ice to perform immunoprecipitation of 64K antigen using test serum. Once thawed, add detergent extraction buffer (20 mM Tris (pH 7.4), 150 mM NaCl, 1000 KIE / ml aprotin and 2 mM phenylmethylsulfonyl fluoride and 2% Triton ^™ X-114) to the test tube, and place the islets on ice. Dissolved for 3 hours. For a total of 60 seconds (4 times with a 1 minute break at 15 second intervals on ice between each sonication)
Using an ultrasonic probe inserted into the detergent extract,
The islet cells were sonicated in vitro. At 15 minute intervals, the island was also destroyed mechanically.

An equal volume of extract was pipetted into a centrifuge tube, the balanced tube was placed in a Beckman 50 rotor, and the tube was ultracentrifuged (100,000 × g, 30 minutes, 4 minutes).
C) to remove insoluble material from the detergent extracted islets. The supernatant was removed and placed in a new polypropylene test tube. The pellet can be discarded. The islet cell lysate (supernatant) can then be separated into an aqueous phase, a sucrose phase and a detergent phase before the immunoprecipitation test.

A method for the extraction of the detergent phase is described in Bodille (Bordier, C. [1981], Triton ^TM X-11.
4 Phase separation of integral membrane proteins in solution '' J. Biol. Chem. 256: 1
604-07).

After extraction, the detergent phase of the islet cell lysate was then placed in a capped 1.5 ml Epindorf tube with normal human serum (containing 10 μl test serum per 100 μl detergent phase supernatant, 1 h, 4 ° C.)
[Ice bath]) and subsequently adsorbed to pre-expanded Protein A Sepharose CL-4B (available from Pharmacia, Sweden). 100 μl of swollen Protein A Sepharose CL-4B was covered with 15 ml
Was added to each 400 μl test serum / detergent phase mixture and kept on a shaking platform at 4 ° C. (300 rpm for 2 hours) in a conical centrifuge tube. At that time the test tube is 1000r
Centrifuge for 1 minute at pm and remove the supernatant. The Protein A Sepharose pellet can be discarded. Aliquots of unbound (pre-cleared) lysate (supernatant) (100 μl containing 5-10 × 10 ⁶ cpm)
l) into 1.5 ml Epindorf centrifuge tubes and add 64K
Cultured with 25 μl of serum to be tested for the presence of autoantibodies. The tubes were then capped and incubated at 4 ° C. for 18 hours (in an ice bath). To each assay tube was added 100 μl of preswollen Protein A Sepharose CL-4B and the reaction mixture was incubated at 4 ° C. on a shaking platform at 300 rpm for 2 hours. To the tubes, 900 μl of 0.5% Triton ^™ X-114 buffer was added to wash protein A Sepharose CL-4B. Swirl the tubes for 10 seconds for each, followed by 1000 seconds for 10 seconds
Centrifuged at rpm. The supernatant was discarded by suction, and the pellet was left in the test tube. Another 900 μl 0.5% Triton
^TM wash buffer was added to the test tube and the wash procedure was repeated five more times. One final wash step was performed by replacing the 0.5% Triton ^™ buffer with an ice-cold water wash. At this point, the tubes should contain only protein A Sepharose CL-4B and any immune complexes formed between the antibodies and / or islet proteins. Bound immune complexes can be removed, 80 mM Tris (pH 6.8), 3.0% (w / v) sodium dodecyl sulfate (SDS), 15% sucrose, 0.001 bromophenyl blue and
(5.0% v / v) β-mercaptoethanol with the same stopper 1.5
Denaturation can be achieved by boiling for 3 minutes in 100 μl of sample buffer contained in ml Epindorf tubes. The bead of sepharose is then removed by centrifugation of the tube (1000 rpm, 1 minute) and the supernatant is carefully removed. A 50 μl sample is then electrophoresed through a discontinuous SDS 10% polyacrylamide separation gel, followed by the method of Laemli [Laemmli, UK 1970; Cleavage of structural proteins during assembly of bacteriophage T4 heads Nature.
227: 680-85], followed by Coomassie brilliant blue staining, followed by fluorography using the methods described in detail in the Enhance of commercial products (New England New York). Clear-).

Following electrophoresis, the gel was placed on filter paper and
Dry in an electrophoresis gel dryer at 60 ° C for a time. Once dried, remove the gel, place on X-ray film (Kodak XROMAT) in a film cassette with enhanced screen, and place at -80 ° C for 3 weeks. The film (referred to as the fluororadiogram) is processed using standard radiographic processing. Test samples are scored as positive or negative after comparison to their respective positive and negative controls.

Example 13 Methods for Detecting 64KA In addition to using the immunoprecipitation technique outlined in Example 11, the present invention can be practiced using any procedure that facilitates the detection of the presence of 64KA. For example, other immunological methods that can be used include enzyme-linked immunosorbent assays (ELIS
A) and radioimmunoassay (RIA). The principles and experimental methods of these procedures are well known to those skilled in the art. This assay can be performed quickly and efficiently by using natural or recombinant proteins that bind to 64KA.
Both whole cell and cell lysate procedures are well known to those conducting research in the art and can be readily used to detect 64KA of the present invention.

For example, as described in Example 8, the amino acid sequences shown in Table 6 can be analyzed to identify immunologically reactive epitopes. These episodes are 64
Amino acid sequence that will react immunologically with KA.
Synthetic peptides that can be used according to these procedures
It is described in detail in FIG. These sequences can then be produced by recombination. For recombinant production, DNA encoding the epitope is inserted into a vector, which is then used to transform a suitable host cell, which in turn encodes the desired amino acid sequence. Expressed. Bacterial, insect, yeast and mammalian cells can all serve as suitable hosts, but eukaryotic cells are the preferred hosts if protein folding is an important factor in epitopic reactivity.

The purified protein or a lysate of the cell producing the protein can be used for the assay.

Another method of using the 64K antigen to detect 64KA is to use an antibody raised against 64KA otherwise known as an anti-64KA antibody. This antibody is immunoprecipitated with 64KA, and detection can be performed as described above.

Example 14-64K autoantigen in immune complex-autologous
Patients who have the assay IDD to detect their own antibodies or who are in the process of developing this disease can be expected to have immune complexes containing the 64K antigen in their serum. These immune complexes typically consist of a 64K autoantigen surrounded by several immunoglobulins. Since individuals who do not have this disease or who are not going to develop this disease will not have these immune complexes, detecting these complexes is the basis of assays for detecting this disease. I will provide a.

[0111] Blood is initially drawn from the patient or control undergoing the IDD test. Serum is obtained by spinning out (centrifuging) red blood cells from a blood sample. ID
For patients with D or those who are about to develop this disease, this serum can be expected to contain an immune complex containing the 64K antigen. The serum sample can then be incubated with Protein A Sepharose. The volume of the sample cultured with the serum is small, for example 25 μl. Following culture of protein A Sepharose CL-4B and immunoglobulin (2 hours, 4 ° C.), the complex was then lysed in 50 mM Tris-HCl (0.1% SDS, 1.0% Triton X-114, and 2 mM EDTA). pH 7.
It can be washed three times in 4) and twice with 50 mM Tris-HCl (pH 7.4). By washing the sepharose as described above, the outer serum proteins are washed away, leaving only the immunoglobulin bound to protein A sepharose. The radioactive phosphonylation of the 64K autoantigen can then be achieved using a kinase reaction. Kinase reaction is 10 mM MnCl ₂ , 10 mM MgCl ₂ ,
1% Triton X-114, and 20 μCi of adenosine (γ- ³² P) 5 ′
50m containing triphosphoric acid (Amersham> 5000Ci / mmol)
It can be started by adding 10 μl of M HEPES buffer (pH 7.4). The precipitate is suspended and can be incubated at 30 ° C. for 10 minutes. The reaction can then be stopped by the addition of electrophoresis sample buffer and heated at 95 ° C. for 4 minutes. ^{The 32} P-labeled product can be separated on a discontinuous SDS 15% polyacrylamide separation gel, followed by autoradiography on an enhanced screen for 6-72 hours (Kodak, X-omat AR5). Serum of IDD patients specifically immunoprecipitated an M _r 64,000 protein, which is not observed in the control of serum.

Example 15-Hybrid protein vaccine
The specific events or reagents used to trigger the treatment of IDD to initiate diabetes are unknown. Viruses having antigens similar to the 64K antigen can produce both a normal immune response to the virus and an abnormal autoimmune response to the 64K antigen through molecular mimicry with the virus. Thus, genetic susceptibility is expressed by exaggerated or prolonged immune responses to environmental factors that initiate the disease process.
It is also possible that the 64K protein has a delayed expression in the development of islet cells during ontogeny, and that it may be possible to use it as an antigen since resistance to it has not developed in the early stages of life .

Regardless of the mechanism of disease initiation, beta cell destruction can proceed in at least two ways. 64KA binds to foreign antigens initiated and stimulates other compounds of the immune system to destroy antibody-bound β cells. Or T-lymphocytes
Recognizes 64K antigen and kills cells directly. 64K in each case
Interfere with the destruction process using proteins, thereby causing IDD
Can be delayed or prevented.

The novel therapeutic method of the present invention involves injection of a toxin conjugated to a purified form of the 64K antigen into the bloodstream. The antigen-toxin complex quickly reaches the lymph nodes, where the complex is usually picked up by immune cells that produce 64KA. Also, the antigen-toxin complex will be bound by T-lymphocytes that recognize the 64K antigen on β cells. Therefore, β
Certain immune cells involved in cell destruction are poisoned and inactivated, leaving non-destructive immune cells harmless.

The hybrid protein may consist of, for example, diphtheria toxin together with the 64K antigen. The construction of such hybrid toxins is described, for example, in U.S. Pat.
5,382 (Muffy) can be proceeded according to the disclosure of the hybrid protein.

Another method involves creating a modified virus (from the disease initiation class of virus) that lacks the expression of the 64K-like protein but retains infectivity. This attenuated virus can be used to vaccinate individuals who are at high genetic risk for IDD, ie, siblings homologous to HLA or who are in the early stages of IDD. Candidate viruses that mimic the 64K protein, and thus initiate the path of the disease, include coxaki virus, rubella virus and EMC virus. Vaccine compositions can be made up of an attenuated virus such that a non-attenuated virus elicits an immune response to the epitope common to the 64K protein. In an attenuated form, the vaccine virus produces an immune response against the virus but not against the 64K protein.

Example 16-Use of 64KA in connection with pancreatic transplantation A preferred method of treating patients with IDD is to transplant normal islets as a replacement for damaged or destroyed beta cells. Partial and whole pancreas transplants have been successfully performed in many patients with diabetes. However, permanent immunosuppressive therapy is always needed to maintain the transplant and prevent rejection. Partial or total pancreas transplantation under continuous immunosuppressive therapy has produced normal levels of blood glucose in some patients with diabetes. Pancreas transplantation occurs late in the course of diabetes and probably does not reverse complications such as kidney disease, in fact exacerbating retinopathy.

Importantly, the success of pancreas transplantation between identical twins has been maintained without immunosuppression, but autoimmune islet cell destruction has occurred and diabetes has recurred. Therefore, even when transplantation is not rejected, immunotherapy is essential to prevent disease recurrence. Destruction (rejection) of the transplanted islets is due at least in part to the re-emergence of self-antigens responsible for autoimmune destruction. No specific immunotherapy currently exists to prevent autoimmune destruction (rejection of transplanted islets / pancreas). Specific immunotherapy can be used to prevent islet cell destruction, using a hybrid toxin as described in detail in Example 11 to prevent autoimmune destruction of the transplanted islet cells or pancreas. The combined use of immunotherapy makes islet cell / pancreas transplantation a therapeutic tool for the treatment of IDD.

Examples 17-64 Kits for Assaying KA Using the novel procedure used herein, reagent kits can be provided that facilitate convenient methods of analyzing serum samples. Kits can be made to use intact, recombinantly or synthetically produced 64K proteins or immunoreactive peptides that serve as antigens for 64KA detection. Alternatively, antibodies specifically developed to detect 64KA are also useful. The principles and methods of ELISA and RIA techniques for detecting antibodies are well established.

As an example, for an ELISA assay, such a kit consists of the following components: 1. 64K protein, peptide or anti-64KA antibody 2. Enzyme (eg, peroxidase) 3. Conjugate (complexed) animal anti-human immunoglobulin, and 4. Positive and negative controls 96 well plastic press
, A colorimetric reagent, an ELISA reader, a blocking reagent and a wash buffer.

Also, by way of example, for an RIA, such a kit may consist of the following components: 1. Radiolabeled 64K protein, peptide or anti-64KA antibody 2. Washing buffer 3. Polyethylene glycol 4. Anti-human precipitation (second) antibody of goat or sheep, and 5. Positive and negative controls Any of the above kits The can be modified to include any of the appropriate laboratory equipment.

The embodiments and specific examples described herein are for illustrative purposes only, and various changes or changes will be suggested to those skilled in the art in light of the above and are included within the spirit and scope of the present application. And should be included in the appended claims.

Table 6. Cat glutamate decarboxylase
mRNA nucleotides and derived sequences

[Table 1]

[Table 1]

[Table 1]

[Table 1]

Continued on the front page (72) Inventor William Castern United States 32608 Gainesville, Florida 47th Road S.S. Brew. 8123 (58) Field surveyed (Int. Cl. ⁶ , DB name) A61K 38/46 A61K 39/00 BIOTECHABS (STN) CA (STN) MEDLINE (STN)

Claims (2)

(57) [Claims] 1. A medicament comprising a glutamate decarboxylase protein or peptide for use in the treatment or prevention of insulin-dependent diabetes or for the manufacture of a medicament thereof. 2. The method according to claim 1, wherein the drug comprises an immune system activating fragment.
The agent according to claim 1, wherein the immune system activating fragment contains the amino acid sequence Pro-Glu-Val-Lys.