Autoimmunity in Insulin-Dependent Diabetes Mellitus j0RN NERUP, M.D. AKE LERNMARK, M.D. &ntofte,
From the Steno Memorial Hospital and Hagedorn Research Laboratory Gentofte, Denmark. Work in this laboratory was supported in part by Grant AM 26190 from the U.S. Public Health Service, and Vera and Carl Johan Michaelsens Legat. Requests for reprints should be addressed to Dr. )@rn Nerup, Steno Memorial Hospital, DK-2820. Copenhagen, Denmark.
Genetic susceptibility, environmental factors and immune mechanisms are implicated in the pathogenesis and etiology of insulindependent diabetes mellitus. The onset of the disease is associated with a major loss of beta cells, and inflammatory cells may be seen in and around the islets of Langerhans. Insulin-dependent diabetes is often associated with autoimmune endocrinopathies. Antipancreatic cell-mediated autoimmunity is found predominantly in young patients with insulin-dependent diabetes of recent onset as are antibodies reactive with cytoplasmic components or cell surface determinants of islet cells. Antibodies reactive with live islet cells may mediate complement-dependent cytotoxicity. The antigen(s) in the islet cells remain(s) to be identified and characterized. Further studies will be necessary to determine whether autoimmunity is the cause of beta-cell destruction or whether it is a consequence of damages caused by exogenous agents in susceptible persons. The possibility that autoimmunity is involved in the pathogenesis of insulin-dependent diabetes mellitus is only one of the interesting observations made in the 1878s linking genetics, environmental factors and immune mechanisms together in a new stimulating theory trying to explain the etiology and pathogenesis of insulin-dependent diabetes mellitus. Based on the findings in the early years of the past decade, we suggested the following hypothesis [l]: “HL-A8 and W15 (present terminology HLA-B8 and B15) seem to be genetic markers for insulin-dependent diabetes. This type of diabetes is further characterized by early age of onset, nonobesity, lymphocytic infiltration of the islets of Langerhans, reduction of the functioning beta-cell mass and antipancreatic, cell-mediated immunity. Thus, our findings lend strong support to the concept that insulin-dependent diabetes is a disease entity in itself and different from noninsulin-dependent diabetes in etiology and pathogenesis. The frequency of HL-A8 was also reported to be increased in Graves disease and in idiopathic Addison’s disease. This points to the possibility that HL-A8 or an HL-A&associated immune response gene may be the common denominator for the development of endocrine autoimmunity. We do not know how some HL-A factors confer an increased risk of insulin-dependent diabetes. Histocompatibility complexes like HL-A contain, in addition to genes controlling serologically detectable antigens, so-called immune-response genes which control the development of cell-mediated immunity to certain antigens. One or more immune-response genes associated with HL-A8 and/or Wl5 might be responsible for an altered T-lymphocyte response. The genetically determined host response could fail to eliminate an infecting virus
January 1981 The American Journal of Medicine
Thyroid Globulin Antibody % 400 400 133 126 250 250
Diabetic Control Diabetic Control Diabetic Control
13 10 23 16
10 6 9 7
Thyroid Microsomal Antibody No. %
56 34 27 4 43 21
64 40 21 7 45 20
21 10 16 5 16 6
9 7 see
Parieial Cell Antibody
14 9 20 3 17 6
Intrinsic Factor Antibody
Adrenal Antlhudy No. % 3 0
2 0 -
(Coxsackie B4 and others) which, in turn, might destroy the pancreatic beta-cells or trigger an autoimmune reaction against the infected organ.” In this review we shall discuss aspects of present knowledge of autoimmunity in insulin-dependent diabetes mellitus and relate this to ideas around etiology and pathogenesis of insulin-dependent diabetes mellitus as of today. Several endocrinopathies seem to be associated with a variety of alterations in immune reactivity apparently terminating the natural unresponsiveness to self. An imbalance or malfunction within the complicated system of cells and events controlling the immune response may result in a variety of measurable phenomena, all reflecting antiself reactivity or autoimmunity. The list of endocrine diseases with autoimmune character includes Graves’ disease, primary myxedema, Hashimoto’s disease, idiopathic Addison’s disease, idiopathic hypoparathyroidism, female hypergonadotropic hypogonadism. pernicious anemia as well as insulindependent diabetes mellitus. In several of these diseases, there seems to be a production of antibodies against relevant endocrine cell- or organ-specific selfantigens. The possible presence of specific effector lymphocytes and, in particular, lymphocytes with receptors for self-antigens remains to be identified and characterized. Immunopathology.
In insulin-dependent diabetes mellitus glucose-induced insulin release is dramatically reduced, and there is a progressive loss in beta-cell function during the course of the disease . Adequate secretory levels’of other islet hormones (glucagon, somatostatin, pancreatic polypeptide) seem to be preserved. Modern staining techniques have demonstrated morphologic changes corresponding to functional findings, i.e., a disappearance of beta cells and the preservation or even proliferation of the other endocrine islet cells . In patients with newly diagnosed insulin-dependent diabetes mellitus, particularly in the very young, the islets of Langerhans are found infiltrated with
Diabetic Control For details
Prevalence of Thyroid, Gastric and Adrenal Antibodies in Diabetic Patients
mononuclear cells , representing the lesion characteristic of insulin-dependent diabetes mellitus--“insulitis.” The exact nature of the infiltrating cells is unknown. By light and electronmicroscopy the cells seem to have the feature of small lymphocytes. Plasma cells are rarely seen. Cell surface markers specific for lymphocytic subpopulations remain to be applied in studies to identify the cell type responsible for these infiltrates. The recognition of beta cells by lymphocytes may require highly specific receptors for beta-cell membrane proteinIs) on the lymphocytes to allow the formation of a cellular infiltration. In experimental systems, cell-cell interactions, e.g., between a lymphocyte and a virusinfected target cell, seem to require the presence of a syngeneic transplantation antigen on the target cell surface [5,6]. To clarify whether the inflammatory lesion is of pathogenic importance, it will be of great interest to disclose the type of cell able to infiltrate a pancreatic islet. Until recently, insulitis was thought to be rare, occurring only in the early stages of clinically manifest insulin-dependent diabetes mellitus. It is interesting, however, to note that destruction of beta cells and acute and chronic inflammatory infiltrates were found in about half of the infants who died with Coxsackie B virus infections . Further studies to clarify whether, and how frequently, viral infections do produce transient or persisting insulitis-like changes in the islets of Langerhans are clearly warranted. Insulin-Dependent Diabetes Mellitus in Clinical AsIt has sociation with Autoimmune Endocrinopathies.
been recognized for several years that insulin-dependent diabetes mellitus often occurs in endocrine diseases of autoimmune character. Insulin-dependent diabetes mellitus was found among 15 percent of the patients with idiopathic Addison’s disease and among 7 to 10 percent of the patients with thyroid autoimmune disorders . The prevalence is about 30 to 50 times higher than the expected prevalence in the general population. There is, in addition, evidence to suggest
The American Journal of Medicine Volume
that different endocrinologic disorders of autoimmune character are four to five times more prevalent among patients with insulin-dependent diabetes mellitus than in the general population. In addition, the serum from such patients contains several organ-specific autoantibodies in an increased frequency (Table I). The prevalence of antithyroid, parietal cell, intrinsic factor or adrenal antibodies seemed higher than that in the control population. Thyroid, gastric and adrenal antibodies were most frequent in young, female patients. It is unclear whether the presence of different organspecific autoantibodies in diabetes should simply be taken as an expression of an autoimmune cause of insulin-dependent diabetes mellitus or whether they reflect some kind of heterogeneity within insulin-dependent diabetes mellitus. From HLA-typing and analysis of autoantibodies, it was suggested that insulin-dependent diabetes mellitus and thyrogastric autoimmunity have different genetic determinants [lo]. Antipancreatic Cell-Mediated Autoimmunity in InMany investigasulin-Dependent Diabetes Mellitus. tors have searched for the presence of cell-mediated reactions directed against the endocrine pancreas in patients with insulin-dependent diabetes mellitus. Nerup et al. [ll]were the first to demonstrate leukocyte migration inhibition specific for the endocrine pancreas in insulin-dependent diabetes mellitus. The antigen used was a homogenate prepared from pooled porcine pancreatic glands in which atrophy of the exocrine tissue had been induced by duct ligation. Twelve of 22 diabetic patients showed migration indices of less than 0.70 and were all found to be insulin-dependent. Half of the patients who showed low migration indices were untreated and had not received exogenous insulin, That the in vitro reaction was in fact a cell-mediated immune reaction was confirmed by intracutaneous testing with the same preparation in six patients. Leukocyte migration inhibition studies were subsequently confirmed and extended (Table II] [u]. Antipancreatic cell-mediated autoimmunity was found predominantly in young patients with insulindependent diabetes mellitus of recent onset but could be demonstrated also in some patients with late-onset diabetes mellitus. All of these either were typical patients with insulin-dependent diabetes mellitus at the time of onset or became insulin-dependent during the course of the disease. Studies of populations with insulin-dependent diabetes mellitus of different duration showed that migration inhibition tended to fade with increasing duration of disease, possibly reflecting the near-total destruction of the beta-cell mass after a year or more of insulin-dependent diabetes mellitus, coinciding with the disappearance of the lymphocytic insulitis. The endocrine pancreas antigen(s) able to induce migration inhibition in vitro is unknown. It seems to be organ-specific, to some extent species-nonspecific, but different from insulin.
Leukocyte Migration Inhibition Studies in Diabetes Yellitus
Porcine duct-ligated pancreas Migration inhibition in 65% of 22 (homogenate prepared by patients with juvenile-onset differential centrifugation) insulin-dependent diabetes mellitus Migration inhibition in 28 % of 112 Fetal calf pancreas diabetic patients (all types) (homogenate) Migration inhibition in 29% of 101 Human pancreas diabetic patients (all types) (homogenate) Human insulinoma (extract) Migration inhibition in 50% of 28 patients with juvenile-onset insulin-dependent diabetes mellitus Migration inhibition in 54% of 24 Human pancreas (homogenate) patients with juvenile-onset insulin-dependent diabetes mellitus For details see
[ 1 l].
The scarcity of information about cell-mediated autoimmunity in insulin-dependent diabetes mellitus relates to the lack of suitable methods. Obviously, renewed interest in this area is important. Antibody-Mediated Autoimmunity in Insulin-Dependent Diabetes Mellitus. Further evidence to support the role for autoimmune mechanisms in the pathogenesis of insulin-dependent diabetes mellitus came from the detection of circulating antibodies reacting with islet cells in the serum of patients with insulin-dependent diabetes mellitus. Two such antibodies have been demonstrated. Islet Cell Cytoplasmic Antibody was described in 1974 by Bottazzo et al. [l3]. Islet cell cytoplasmic antibody is detected in an indirect immunofluorescence assay using sections of frozen human pancreas from persons with blood group 0. The antibody-reaction towards the islet cells is weak, and the fluorescent reaction is seen over the cytoplasm of all the different endocrine islet cells. Antibodies bound to the frozen sections, briefly fixed in acetone, are detected by fluoresceinisothiocyanateconjugated antihuman immunoglobulin G (IgG) antiserum. Based on analogous anti-immunoglobulin M (IgM) or anti-immunoglobulin A (IgA) preparations, it is claimed that islet cell cytoplasmic antibody mainly represents IgG antibodies. Using fluorescein-conjugated sheep antihumanC3 [third component of complement) or anti-Clq reagents, a large proportion of serum from patients with insulindependent diabetes mellitus of short duration was found to induce islet cell immunofluorescence. These experiments suggest that some, but not all, islet cell antibodies reactive with cytoplasmic components are complement fixing . Summarized in Table III are some recent reports on the prevalence of islet cell cytoplasmic antibody. In the
The American Journal of Medicine
Prevalence ( %) of Islet Cell Cytoplasmlc Antlbodles Results from 1171
Total 0.6 (51811)
(01iO0, -(49EO5, (&6, (24:337, (20/L)
(3& First degree relatives to insulin-dependent diabetic patients &6,
Insulin-dependent diabetes without autoimmune disease (22438574)
Insulin-dependent diabetes with autoimmune disease (68Y76,
Noninsulin dependent diabetes Autoimmune disease without diabetes
NOTE: Figures in parentheses indicate the number of patients antibodies and the total number of patients studied.
survey of an unselected population, 0.01 percent (nine of 3,766people) had the antibody .The prevalence of islet cell cytoplasmic antibody in nonautoimmune hospital populations and normal subjects was 0 to 2 percent and was slightly increased among first-degree relatives of patients with insulin-dependent diabetes mellitus (Table III]. Disregarding the duration of the disease, 20 to 40 percent of the patients with insulindependent diabetes mellitus had islet cell cytoplasmic antibody. However, the presence of antibodies seemed to be dependent on the duration of the disease. Thus, 60 to 70 percent of the patients with insulin-dependent diabetes mellitus had islet cell cytoplasmic antibody at the time of diagnosis or a few weeks thereafter. The proportion of patients with the antibody declined rapidly, being only about 5 percent after 10 to 20 years, preferably in those who had HLA-BB. The evanescent character of islet cell cytoplasmic antibody remains to be explained but could be due to loss of the antigenic stimulus. However, in patients with a concomitant organ-specific autoimmune disease , the islet cell antibodies are more likely to persist. Only rarely did islet cell cytoplasmic antibody seem to appear if it was not present at the time of diagnosis [v’]. However, positive serum from patients without insulin-dependent diabetes mellitus was found among patients with autoimmune polyendocrinopathies . Islet cell cytoplasmic antibody is uncommon among noninsulin-dependent diabetic patients; when present, it may signify future development of severe beta-cell dysfunction . In patients with insulin-dependent diabetes mellitus, the prevalence of ‘islet cell cytoplasmic antibody was independent of residual beta-cell function as determined by serum C-peptide . In contrast, in some noninsulin-dependent diabetic patients with islet cell antibody, blood glucose levels were higher and serum insulin levels lower in an oral glucose tolerance test than in patients without islet cell antibodies . Nondiabetic persons with islet cell antibodies were reported to release less insulin and glucagon
The American Journal of Medicine
in response trols .
to an arginine
Thus, islet cell cytoplasmic antibody might be considered a marker of developing insulin-dependent diabetes mellitus in nondiabetic subjects and in patients with noninsulin-dependent diabetes mellitus. In this context it is worth noting that islet cell cytoplasmic antibody was recently reported to occur in children with mumps infection . Furthermore, the observation that rodenticide induced diabetes mellitus in some patients was accompanied by the occurrence of islet cell surface antibody  points to the possibility that environmental agents might be able to elicit antipancreatic autoimmunity in man. It is surprising, however, that knowledge about islet cell cytoplasmic antibody is still rather unsatisfactory. The antigen in the beta cells with which islet cell cytoplasmic antibody reacts is species-nonspecific, but its exact nature remains to be elucidated through absorption studies, including subcellular fractions of beta cells. Furthermore, systematic studies on antibody titers relative to clinical characteristics remain to be made. The qualitative approach to the assay is also reflected by the fact that there is little or no information about reproducibility, assay sensitivity or intra- and interassay variability in and between different laboratories. The establishment of an international standard and the development of a quantitative islet cell cytoplasmic antibody assay would be of great interest. Islet Cell Surface Antibodies. Living cells are in general impermeable to antibodies. Antibodies directed against a hypothetic self-antigen expressed on the surface of a cell may remain accessible for either assay reagents or hypothetically for in vivo reactions with either complement (complement-mediated antibody cytotoxicity) or cells [antibody-dependent cellular cytotoxicity). By indirect immunofluorescence on viable, insulin-producing cells, freshly prepared from collagenase-isolated rat or mouse pancreatic islets, islet cell surface antibodies were found in serum from many
I ( ICSAI
Figure 1. Islet cell antibodies in children and young adults with insulin-dependent diabetes mellitus. Cytoplasmic (ICA) and cell surface antibodies (ICSA) were determined in 17 patients with diabetes followed prospectively for 18 months. Each point represents the analysis of serum samples from six to 1i patients.
patients with insulin-dependent diabetes mellitus . The results of a prospective analysis of islet cell surface antibody in children with insulin-dependent diabetes mellitus are shown in Figure 1. Islet cell surface antibodies were present in 67 percent of the newly diagnosed children. The antibodies were present among 3 to 6 percent of healthy, age- and sex-matched controls [24,25]. Experiments with absorption of serums from human subjects with insulin-dependent diabetes mellitus to various rat tissue preparations suggested that the islet cell surface antibodies are organ-specific, but nonspecies-specific. The technique of demonstrating islet cell surface antibodies including more quantitative assays  will, of course, be applicable to suspensions of human islet cells once such material becomes more readily available. Whether the assays for islet cell cytoplasmic antibody and islet cell surface antibody detect antibodies against different antigens is not known. Using both assays to determine islet cell antibodies, it appeared that islet cell
cytoplasmic antibody and islet cell surface antibody may occur independently of one another in patients with insulin-dependent diabetes mellitus  and that the proportion of patients who have islet cell cytoplasmic antibody and/or islet cell surface antibody was higher than the proportion of patients who have only one or the other antibody . The existing knowledge about islet cell surface antibody is incomplete and unsatisfactory in the same way as described for islet cell cytoplasmic antibody. The Significance of Autoimmunity in Insulin-Dependent Diabetes Mellitus. The demonstration of the existence of autoimmune phenomena such as antipancreatic cell-mediated autoimmunity, islet cell cytoplasmic antibody and islet cell surface antibody in insulin-dependent diabetes mellitus is only indirect evidence for a role of autoimmune mechanisms in this disease. A second generation type of studies which aim to show directly that specifically reactive lymphocytes and
The American Journal of Medicine
AUTOIMMUNITY IN INSULIN-DEPENDENT DIABETES MELLITUS-NERUP.
antibodies might be cytotoxic to beta cells will probably appear in the literature soon. The first results from this type of approach have been published. Immunization of experimental animals with pancreatic islets or islet cell suspensions results in the production of cell surface antibodies [25,27]. These antibodies mediate complement-dependent cytotoxicity [28,29] as well as antibody-dependent cellular cytotoxicity (Soderstrum, personal communication). Using intact hamster islets in a tissue culture system cross-species complement-mediated cytotoxic effects of serums from patients with insulin-dependent diabetes mellitus were reported [3O]. An effect of lymphocytes and/or antibodies from patients with insulin-dependent diabetes mellitus on human islet cells still remains to be demonstrated. If direct beta-cell cytotoxicity mediated through autoimmune mechanisms in patients with insulin-dependent diabetes mellitus should be demonstrated, the crucial question is still unanswered: Does autoimmunity cause beta-cell destruction per se? Or is it rather the consequence of initial beta-cell damage (minor changes in the cell surface recognized by the immune system as not self) caused by exogenous agents, e.g., beta-cell cytotropic virus, beta-cell cytotoxic chemicals?
Furthermore, to fully understand the role of autoimmunity in insulin-dependent diabetes mellitus, results must be related to recent progress in the genetical analysis of insulin-dependent diabetes mellitus. HLAD/DR genotyping on carefully ascertained populations of insulin-dependent diabetes mellitus suggests that heterogeneity exists within insulin-dependent diabetes mellitus [31,32]. Thus, our understanding of insulin-dependent diabetes mellitus is far from complete. In our opinion, however, the following updated version of our original hypothesis is worth pursuing:
“The genetic susceptibility to insulin-dependent diabetes mellitus is conferred by two genes on chromosome 6-one associated with HLA-D/DR3, one with HLA-D/DR4. The two susceptibility genes act through different mechanisms. The reaction of susceptible persons to certain exogenous stimuli [beta-cell cytotropic virus, beta-cell cytotoxic chemicals] is abnormal, leading to beta-cell destruction, directly by the exogenous agents, through autoimmune mechanisms and/or because of lack of beta-cell regeneration.”
REFERENCES 00. et al.: HL-A antigens and diabetes and diabetes mellitus. Lancet 1974; 2: 864-866. Rubenstein AH, Steiner DF, Horwitz DL, et al.: Clinical significance of circulating proinsulin and c-peptide. Ret Prig Horm Res 1977: 33: 435-475. Geots W. Le Comote I? The oancreatic islets in diabetes. Am
Gepts W, De Mey J: Islet-cell survival determined by morphology. Diabetes 1978: 27 (suppl 11:251-256. Zinkernagel RM, Doherty PC: H-2 compatibility requirement for T-cell mediated lysis of targets infected with lymphocytic choriomeningitis virus. J Exp Med 1975; 141: 14271436. Zinkernagel RM: H-2 restriction of cell-mediated virusspecific immunity and immunopathology: self recognition, altered self and autoaggression. In: Talal N, ed. Autoimmunity. New York: Academic Press, 1977; 363-384. Jenson AB, Rosenberg HS, Notkins AL: Pancreatic islet cell damage in children with fatal viral infections. Lancet 1980; 2: 354-358. Nerup J: The clinical and immunological association of diabetes mellitus and Addisons disease. In: Bastenie PA, Gepts W. eds. Immunity and autoimmunity in diabetes mellitus. Amsterdam; Excerpta Medica 1974; 149-152. MacCuish AC, Irvine WJ: Autoimmunological aspects of diabetes mellitus. Clin Endocrinol Metab 1975: 4: 435471. Gorsuch AN, Dean BM, Bottazzo GF, et al.: Evidence that type I diabetes and thyrogastric autoimmunity have different genetic determinants. Br Med J 1980; 1: 145-147. Nerup J. Andersen 00, Bendixen G. et al.: Antipancreatic cellular hypersensitivity in diabetes mellitus. Diabetes 1971: 20: 424-427. Nerup’j, Christy M, Deckert T. et al.: Diabetes mellitus. In: Samter M. ed. Immunological diseases. Boston: Little, Brown & Company, 1978; 1330-1347.
1. Neruo. 1. ,. Platz P. Andersen 2.
3. 4. 5.
9. 10. 11. 12.
17. 18. 19. 20.
22. 23. 24.
Bottazzo GF, Florin-Christensen A, Doniach D: Islet cell antibodies in diabetes mellitus with autoimmune oolvendocrine deficiencies. Lancet 1974; 2: 1279-1283. A ” Bottazzo GF, Dean BM. Gorsuch AN, et al.: Complement-fixing islet-cell antibodies in type-1 diabetes: possible monitors of active B-cell damage. Lancet 1980; 1: 668-672. Rodger B. Whittingham S, Martin FIR. et al.: A population survey of pancreatic islet cell antibodies. Clin Exp Immunoll980; 89: 125-129. Irvine WI.,. McCallum Cl.,. Grav_ RS. et al.: Pancreatic islet cell antibodies in diabetes mellitus correlated with the duration and tvoe of diabetes. coexistent autoimmune disease, and HLA:iype. Diabetes 1977; 26: 138-147. Lendrum R. Walker G, Cudworth AG, et al.: Islet-cell antibodies in diabetes mellitus. Lancet 1976: 2: 1273-1276. Irvine WJ, Gray RS, McCallum CJ: Pancreatic islet cell antibodv as a marker for asvmotomatic and latent diabetes and prediabetes. Lancet 1976: 2: 1097-1102. Madsbad S. Bottazzo GF. Cudworth AG. et al.: Islet-cell antibodies and beta-cell function in insulin-dependent diabetics. Diabetologia 1980; 18: 45-47. Grav RS. Irvine WI. Cameron EHD. et al.: Glucose andinsulin responses to oral glucose in overt non-insulin-deoendent diabetics with and without the islet cell antibody. Diabetes 1980: 29: 312-316. Tiengo A, Del Prete GF, Nosadini R, et al.: Insulin and glucagon secretion in diabetic and non-diabetic patients with circulating islet cell antibodies. Diabetologia 1977: 13: 451-458. Helmke K. Otten A. Williams W: Islet cell antibodies in children’with mumps infection. Lancet 1980; 2: 211-212. Karam lH, Prosser PR. Lewitt PA: Islet cell surface antibodies in a -patient with diabetes mellitus after rodenticide ingestion. N Engl J Med 1978: 299: 1191. Lernmark A, Freedman ZR, Hofmann C, et al.: Islet cell
January 1981 The American Journal of Medicine Volume 70
AUTOIMMUNITY IN INSULIN-DEPENDENT DIABETES MELLITUS-NERUP.
25. 26. 27.
surface antibodies in juvenile diabetes mellitus N Engl ] Med 1978; 299: 375-380. Freedman ZR, Feek CM, Irvine WJ, et al.: Islet cell cytoplasmic and cell surface antibodies in diabetes mellitus. Trans Assoc Am Physicians 1979; 92: 64-76. Lernmark A. Kanatsuna T, Patzelt C. et al.: Antibodies directed against the pancreatic islet cell plasma membrane: detection and specificjty. Diabetologia (in press). Kromann H. Lernmark A, Vestergaard BF, et al.: The influence of the maJor histocompatibility complex (H-2) on experimental diabetes in mice. Diabetologia 1979: 16:
Soderstrum WK, Freedman ZR, Lernmark A: Complementdependent cytotoxic islet cell surface antibodies in insulin-dependent diabetes. Diabetes 1979; 28: 397. Kanatsuna T, Lernmark A, Rubenstein AH, et al.: Block in
insulin release from column-perifused pancreatic B-cells induced by cell surface antibodies and complement. Diabetes [in press). Rittenhouse HG, Oxender DL, Pek S, et al.: Complementmediated cytotoxic effects on pancreatic islets with sera from diabetic patients. Diabetes 1980; 29: 317-322. Ryder LP, Christy M, Nerup J, et al.: HLA studies in diabetics, In: Camerini-Davalos RA, Hanover B, eds. Treatment of early diabetes. New York: Plenum Publishing Corp. 1979; 41-48. Platz P, Jakobsen BK. Marling N. et al.: HLA-D and -DR antigens in genetical analysis of insulin-dependent diabetes mellitus. Submitted to Diabetologia. Del Prete GF, Betterle C, Padovan D, et al.: Incidence and significance of islet cell autoantibodies in different types of diabetes mellitus. Diabetes 1977; 26: 909-915.
The American Journal of Medicine