Peptides as antigens. Importance of orientation.

Factors known to be important in producing protein-reactive peptide antibodies include the accessibility of the region from which the peptide sequence is derived, the hydrophilic-phobic character of the sequence, and the length of the peptide. The data presented here indicate that the orientation of the peptide coupled to a carrier protein also influences the binding pattern of peptide antibodies. An octapeptide, representing a sequence from the alpha chain of the human acetylcholine receptor, was coupled either through an N- or C-terminal cysteine-glycine-glycine linker to a carrier protein and used to immunize rabbits. The resulting antisera reacted at comparable titers to the uncoupled immunizing peptides, but did not crossreact with the identical but opposite-linked peptide. Characterization of the binding to other homologous peptides showed that immunization with the N-terminal-linked peptide induced antibodies reactive specifically with the C-terminal amino acid(s). Immunization with the C-linked peptide resulted in antibodies reactive with a site of the peptide near the C-terminus.

Specific pancreatic beta-cell surface antigens recognized by a xenogenic antiserum.

An antiserum (R4) from a rabbit immunized with suspensions of C57BL/61 ob/ob mouse islet cells contains antibodies which in a 125I-protein A radioligand assay can be demonstrated to bind to single cell suspensions of normal Naval Medical Research Institute (NMRI) mouse islet cells. The binding of 125I-protein A to islet cells was about four times that of normal rabbit serum (NRS) after incubation at a 1/600 dilution of R4 antiserum quantitatively absorbed to mouse spleen lymphocytes (R4A antiserum) and hepatocytes. Subsequent absorption of the R4A antiserum to islet cells significantly reduced the binding of 125I-protein A to islet cells incubated with the doubly absorbed serum. Immunoprecipitation of radiolabeled islet cell lysates followed by SDS polyacrylamide gel electrophoresis and autoradiography suggested that the R4A antiserum recognized a Mr 40,000 glycoprotein. This glycoprotein was not detected in spleen lymphocytes. Electron microscope detection of gold-protein A complexes suggested that the binding of islet cell surface antibodies was cell specific. islet cell suspensions incubated with R4A antiserum and gold-protein A showed that 86 +/- 3 gold particles were bound per 100 beta-cells (mean +/- SE for six experiments). In contrast, the number of gold particles per 100 endocrine non-beta-cells was 8 +/- 1 which was similar to the number achieved with NRS (3 +/- 1) on all endocrine islet cells. Our observations suggest that the pancreatic islet cells, in particular the beta-cells, express a specific antigen.

Autoantibodies to a 64-kilodalton islet cell protein precede the onset of spontaneous diabetes in the BB rat.

Spontaneous insulin-dependent diabetes mellitus (IDDM) in the BB rat is associated with the presence of antibodies to a 64-kilodalton rat islet cell protein. These protein antibodies appeared in young animals and remained for as long as 8 weeks before the clinical onset of IDDM. Antibodies to a 64-kilodalton human islet cell protein were found to be associated with human IDDM. Detection of the antibodies may therefore be used to predict an early immune reaction against pancreatic B cells.

Inhibition of diabetes in BB rats by virus infection.

BB rats serve as a model for human insulin-dependent diabetes mellitus (IDDM), since without insulin treatment, most 60-140-d-old animals die within 1 to 2 wk of developing polyuria, polydypsia, hyperglycemia, and hypoinsulinemia. Lymphoid cells accumulate in the islets of Langerhans and beta cells undergo destruction. We report that inoculation of such BB rats with lymphocytic choriomeningitis virus (Armstrong strain, clone 13) reduces over a prolonged period the incidence of IDDM, normalizes the concentration of blood sugar and pancreatic insulin, prevents the mononuclear cell infiltration in the islets of Langerhans, and for a short time after inoculation alters T lymphocyte subsets. Thus, a virus might be programmed to carry out useful functions.

Oral insulin treatment suppresses virus-induced antigen-specific destruction of beta cells and prevents autoimmune diabetes in transgenic mice.

Oral administration of self-antigens has been proposed as a therapy to prevent and treat autoimmune diseases. Here we report that oral treatment with insulin prevents virus-induced insulin-dependent diabetes mellitus (IDDM) in a transgenic (tg) mouse model. Such mice express the viral nucleoprotein (NP) of lymphocytic choriomeningitis virus (LCMV) under control of the rat insulin promoter in their pancreatic beta cells and < 2% spontaneously develop diabetes. However, 2 mo after challenge with LCMV, IDDM occurs in > 95% of tg mice but not in controls. Oral treatment with 1 mg of insulin twice per week for 2 mo starting either 1 wk before or 10 d after initiating LCMV infection prevents IDDM in > 50% of the tg mice (observation time 8 mo). Thus, insulin therapy is effective in preventing progression to overt IDDM in prediabetic tg mice with ongoing islet infiltration. Oral administration of insulin does not affect the generation of LCMV-NP-specific anti-self cytotoxic T lymphocytes nor the infiltration of lymphocytes into the pancreas. However, less beta cells are destroyed in insulin-treated mice, upregulation of MHC class I and II molecules does not occur, and antiviral (self) cytotoxic T lymphocytes are not found in the islets, events present in tg mice developing IDDM. The majority of lymphocytes in the islets of insulin-treated tg mice without IDDM produces IL-4, IL-10, and TGF-beta. In contrast, lymphocytes from islets of tg mice developing IDDM mainly make gamma-IFN.

Molecular mimicry and myasthenia gravis. An autoantigenic site of the acetylcholine receptor alpha-subunit that has biologic activity and reacts immunochemically with herpes simplex virus.

The large majority of patients with the autoimmune disease myasthenia gravis characteristically have detectable antibodies against the acetylcholine receptor (AChR). We used synthetic peptides to identify antibodies in sera of myasthenia gravis patients reactive with the human acetylcholine receptor (HuAChR) alpha-subunit, residues 160-167. Affinity purification of these antibodies, using the HuAChR alpha-subunit 157-170 peptide immobilized on thiopropyl-Sepharose, yielded IgG antibodies that bound to the native AChR and inhibited the binding of alpha-bungarotoxin to the receptor. The HuAChR alpha-subunit 160-167 peptide demonstrated specific immunological cross-reactivity with a shared homologous domain on herpes simplex virus glycoprotein D, residues 286-293, by both binding and inhibition studies. Thus, HuAChR alpha-subunit, residues 160-167, elicits antibodies in myasthenic patients that binds to the native AChR protein and is capable of eliciting a biologic effect. Immunologic cross-reactivity of this "self" epitope with herpes simplex virus suggest that this virus may be associated with the initiation of some cases of myasthenia.

Quantitative assay using recombinant human islet glutamic acid decarboxylase (GAD65) shows that 64K autoantibody positivity at onset predicts diabetes type.

At and before onset, most insulin-dependent diabetics (IDDM) have islet GAD65 autoantibodies (GAD65Ab). Since IDDM also occurs in older patients where non-insulin-dependent diabetes is common, we studied GAD65Ab at onset to classify diabetes type. Our quantitative immunoprecipitation assay uses recombinant human islet GAD65 stably expressed in hamster fibroblasts. Electrophoretic mobility was identical to native islet GAD65. Like native antigen, recombinant GAD65 migrated as two bands during electrophoresis, but converted to one under stronger reduction. Immunoprecipitation was linear with respect to antibody or antigen concentration. In 120 population-based diabetic patients of all ages grouped by treatment at onset and after 18 mo, GAD65Ab were present in 70% on insulin (n = 37), 10% on oral agent (n = 62, P < 0.0001), 69% changing from oral agent to insulin (n = 16, P < 0.001), and 1 of 33 controls. 65% with GAD65Ab, versus 8% without, changed from oral agent to insulin (P < 0.01). The GAD65Ab quantitative index was remarkably stable, and only 2 of 32 patients changed antibody status during follow-up. Concordance between GAD65Ab and islet cell antibodies was 93%. Quantitative correlation was approximate but significant. This highly sensitive, quantitative, high capacity assay for GAD65Ab reveals treatment requirements better than clinical criteria, perhaps guiding immunomodulatory therapy.

Prediction of insulin-dependent diabetes mellitus in siblings of children with diabetes. A population-based study. The Childhood Diabetes in Finland Study Group.

An unselected population of 755 siblings of children with insulin-dependent diabetes mellitus (IDDM) was studied to evaluate the predictive characteristics of islet cell antibodies (ICA), antibodies to the IA-2 protein (IA-2A), antibodies to the 65-kD isoform of glutamic acid decarboxylase (GADA), insulin autoantibodies (IAA), and combinations of these markers. We also evaluated whether the histochemical ICA test could be replaced by the combined detection of other markers. 32 siblings progressed to IDDM within 7.7 yr of the initial sample taken at or close to the diagnosis of the index case (median follow-up, 9.1 yr). The positive predictive values of ICA, IA-2A, GADA, and IAA were 43, 55, 42, and 29%, and their sensitivities 81, 69, 69, and 25%, respectively. In contrast to the other three antibody specificities, GADA levels were not related to the risk for IDDM. The risk for IDDM in siblings with four, three, two, one, or no antibodies was 40, 70, 25, 2, and 0.8%, respectively. Combined screening for IA-2A and GADA identified 70% of all ICA-positive siblings, and all of the ICA-positive progressors were also positive for at least one of the three other markers. The sensitivity of the combined analysis of IA-2A and GADA was 81%, and the positive predictive value was 41%. In conclusion, combined screening for IA-2A and GADA may replace the ICA assay, giving comparable sensitivity, specificity, and positive predictive value. Accurate assessment of the risk for IDDM in siblings is complicated, as not even all those with four antibody specificities contract the disease, and some with only one or no antibodies initially will progress to IDDM.

Differential responsiveness to interferon-alpha in beta-cells and non-beta cells.

Interferon-alpha (IFN-alpha) is important in the innate immune defense, particularly in viral infections. IFN-alpha induces 2',5'A synthetase, the products of which, 2',5'-oligoadenine nucleotides, activate mRNA degrading enzymes. IFN-alpha is the first detectable cytokine in the insulitis lesion seen in recent-onset IDDM, and insulin promoter directed expression of IFN-alpha in transgenic mice leads to development of IDDM. Here, we demonstrate that IFN-alpha induces 2',5'A synthetase activity only in insulin-producing betaTC3 cells and in isolated single rat beta-cells but not in alphaTC3 cells or in isolated rat non-beta-cells. The increased responsiveness of beta-cells but not non-beta-cells to IFN-alpha with the ensuing activation of the mRNA-degrading 2',5'A synthetase system suggests why only the beta-cells are destroyed in the diabetogenic process.

Spontaneously diabetic BB rats have age-dependent islet beta-cell-specific surface antibodies at clinical onset.

Diabetes-prone BB rats were examined before, during, and after clinical onset of diabetes for the occurrence of circulating islet cell surface antibodies (ICSAs) with a specific binding to islet beta-cells. The presence of ICSA was assessed by incubating serum immunoglobulin fractions with normal Wistar rat islet cells and identifying cell-bound immunoglobulins by indirect immunofluorescence and by a complement-induced cell toxicity reaction. Under the selected experimental conditions, none of the diabetes-resistant rats were ICSA positive over the entire study period (45-120 days of age). In two diabetes-prone BB rat strains, 19 animals developed diabetes with onset between 50 and 120 days. At day 45, none of these animals was positive for ICSA. In rats developing diabetes between 50 and 85 days of age, 7 of 9 animals presented ICSA at clinical onset, determined by either the immunofluorescence or cytotoxicity test. The antibodies bound to insulin-containing beta-cells but not to other islet cell types, and binding was not eliminated by absorption with liver powder. In animals developing diabetes between 85 and 120 days, only 1 of 10 was positive for beta-cell-specific surface antibodies at onset of the disease. After 30 days of insulin treatment, beta-cell-specific antibodies were detectable in 3 of 4 animals of age 50-85 days, whereas only 3 of 12 older rats presented ICSAs that were, in addition, of low titer or affinity. Our data confirm that ICSAs develop in diabetic BB rats and indicate that these antibodies can bind specifically to islet beta-cells compared with other islet cell types.(ABSTRACT TRUNCATED AT 250 WORDS)

Restriction fragment polymorphisms in the major histocompatibility complex of diabetic BB rats.

DNA isolated from diabetic BB (BB/Hagedorn) rats was examined for restriction fragment length differences within the major histocompatibility complex (MHC) as compared with nondiabetic (W-subline) BB rats. Polymorphisms were detected using a mouse class I MHC gene as probe. Specifically, a 2-kb BamHI fragment was present in all the nondiabetic rats examined, but absent in the diabetic rats. Similar polymorphisms were observed with various other restriction enzymes, particularly XbaI, HindII, and SacI. There were no polymorphisms detected using either a human DR-alpha (class II antigen heavy chain) or a human DC-beta (class II antigen light chain) gene as probes. These results indicate that the BB rat diabetic syndrome may be linked to differences in class I MHC genes.

Neonatal tolerization with glutamic acid decarboxylase but not with bovine serum albumin delays the onset of diabetes in NOD mice.

To test the role of glutamic acid decarboxylase (GAD65) or bovine serum albumin (BSA) autoimmunity in the pathogenesis of diabetes, GAD65 or BSA was injected intraperitoneally into neonatal female NOD mice (100 micrograms/mouse of each protein). Treatment with GAD65, but not with BSA, significantly delayed the onset of diabetes compared with control mice (P < 0.05). At 18 weeks, 6 of 10 control mice compared with 0 of 10 GAD65-treated mice (P = 0.005) and 7 of 14 BSA-treated mice had developed diabetes. However, after 79 weeks, 6 of 10 of the GAD65-treated mice were diabetic compared with 9 of 10 of the control mice and 12 of 14 of the BSA-treated mice. In GAD65-treated mice without diabetes, insulitis was markedly reduced compared with control or BSA-treated mice (P < 10(-4)). To further elucidate why GAD becomes an autoantigen, the expression in NOD mice islets was studied. Quantitative immunohistochemistry revealed that islet cell expression of GAD was increased in 5-week-old NOD mice compared with BALB/c mice (P = 0.02). With the occurrence of insulitis (9-15 weeks), the GAD expression was further increased relative to 5-week-old NOD mice (P < 0.02). In conclusion, GAD, but not BSA, autoimmunity is important for the development of diabetes in NOD mice. Furthermore, concordant with the appearance of insulitis, the GAD expression increased in NOD mouse islets, which could possibly potentiate the beta-cell-directed autoimmunity.

Glutamic acid decarboxylase (GAD65) autoantibodies in prediction of beta-cell function and remission in recent-onset IDDM after cyclosporin treatment. The Canadian-European Randomized Control Trial Group.

We have investigated whether glutamic acid decarboxylase (GAD) autoantibodies (GAD65 Ab) were affected by cyclosporin therapy and were related to subsequent non-insulin-requiring remission and loss of glucagon-stimulated C-peptide response in 132 recent-onset insulin-dependent diabetes mellitus (IDDM) patients treated with cyclosporin or placebo for 12 months. GAD65 Ab were detected in a quantitative radioligand assay using as tracer recombinant, in vitro translated, human islet [35S]methionine-labeled GAD65. GAD65 Ab were found at onset in 66% (87 of 132) of IDDM patients and in 1% (1 of 100) of healthy control subjects. The prevalence of GAD65 Ab and median GAD65 Ab levels did not change in serum samples taken 3, 6, 9, and 12 months after study entry in either the cyclosporin- or the placebo-treated groups. The presence or absence of GAD65 Ab at study entry did not predict non-insulin-requiring remission in either cyclosporin- or placebo-treated patients. However, the relative (compared with 0 months) glucagon-stimulated C-peptide response was more than 30% lower in GAD65 Ab+ patients receiving placebo at 9 and 12 months compared with the GAD65 Ab- placebo patients (P < 0.035). Islet cell cytoplasmic antibody (ICA) and GAD65 Ab+ placebo-treated patients showed no significant differences in stimulated C-peptide levels compared with those who were ICA- and GAD65 Ab+, suggesting that ICA was not independently associated with loss of beta-cell function.(ABSTRACT TRUNCATED AT 250 WORDS)

Islet cell surface antibodies and lymphocyte antibodies in the spontaneously diabetic BB Wistar rat.

Plasma from 14 diabetic and 6 nondiabetic BB Wistar rats along with plasma from 6 non-BB Wistar rats was evaluated for the presence of islet cells surface antibodies (ICSA) and antibodies to spleen lymphocytes by the protein-A radioligand assay. Dispersed Wistar rat islet cells incubated with plasma from diabetic rats bound 4255 +/- 2208 cpm 125I-protein A/5 x 10(4) islet cells (mean +/- SD) compared with 984 +/- 454 cpm/5 x 10(4) islet cells in islet cells incubated with plasma from nondiabetic BB rats (P less than 0.005). Twelve of the 14 diabetic rats with a duration of diabetes for 3-11 days bound radioactivity above the mean and 2 x SC of controls. The binding of 125I-protein A did not differ between nondiabetic BB rats and non-BB Wistar rats. Wistar rat spleen lymphocytes incubated in diabetic plasma bound 20,249 +/- 10,783 cpm/2 x 10(6) spleen lymphocytes compared with 3460 +/- 1809 in the controls (P less than 0.005). Animals positive for ICSA correlated with those positive for spleen lymphocyte antibodies. It is concluded that ICSA and splenic lymphocyte antibodies are present in diabetic BB rats.

High prevalence of autoantibodies to glutamic acid decarboxylase in long-standing IDDM is not a marker of symptomatic autonomic neuropathy.

Immune reactivity to the enzyme glutamic acid decarboxylase (GAD), a pancreatic islet autoantigen, is present at the diagnosis of insulin-dependent diabetes mellitus (IDDM). Because GAD is also highly expressed in the nervous system, we investigated the presence of autoantibodies to the isoform GAD65 in patients with diabetic neuropathy, which is a debilitating complication of the disease. We studied 39 patients with autonomic and somatic neuropathy, 28 patients matched for age and IDDM duration, and 13 patients with a shorter duration of IDDM, all with no diabetic complications, as well as 50 recently diagnosed diabetic patients, 23 neurologic patients with idiopathic autonomic failure unrelated to IDDM, and 72 healthy subjects. An immunoprecipitation radioligand assay was used to detect anti-GAD65 autoantibodies with in vitro transcribed and translated human islet GAD65 as antigen. Autoantibodies to GAD65 were present in 56% of the diabetic patients with neuropathy, 57% of the long-duration and 69% of the short-duration diabetic control subjects, 78% of the recently diagnosed patients, and 13% of the nondiabetic neuropathic patients. Among the diabetic patients with neuropathy, there was no correlation between the presence of anti-GAD65 antibodies and the presence of autoantibodies to sympathetic ganglia, vagus nerve, or adrenal medulla structures identified by immunofluorescence. Our study shows that anti-GAD65 antibodies are present in a high proportion of patients with diabetic neuropathy but are not exclusively associated with it, rendering it unlikely that they have a role as a disease marker or that they are pathogenetic.(ABSTRACT TRUNCATED AT 250 WORDS)

Recombinant glutamic acid decarboxylase (representing the single isoform expressed in human islets) detects IDDM-associated 64,000-M(r) autoantibodies.

GAD is an autoantigen in IDDM. Molecular cloning and specific antibodies allowed us to demonstrate that only the lower M(r) GAD64 isoform is expressed in human islets, in contrast to human brain, rat islets, and rat brain, all of which express both GAD64 and GAD67. Expression of the human islet GAD64 isoform in COS-7 and BHK cells resulted in an enzymatically active rGAD64, which is immunoreactive with diabetic sera comparable with that of the islet 64,000-M(r) autoantigen. Immunoprecipitation analyses showed that 21/28 (75%) IDDM sera had rGA D64 antibodies compared with only 1/59 (1.7%) of the healthy control sera. In immunoblot analyses, an SMS serum--but only 1/10 randomly selected IDDM sera--recognized the blotted rGAD64 without relation to immunoprecipitation titers. In conclusion, only the GA D64 isoform is expressed in human islets, in contrast to rat islets, which also express the GAD67 isoform. The immunological properties of human rGAD64 are comparable with the native 64,000-M(r) islet autoantigen, allowing further studies of the immunopathogenesis of IDDM.

Detection of GAD65 antibodies in diabetes and other autoimmune diseases using a simple radioligand assay.

Autoantibodies to glutamic acid decarboxylase (GAD) are frequent at or before the onset of insulin-dependent diabetes mellitus (IDDM). We have developed a simple, reproducible, and quantitative immunoprecipitation radioligand assay using as antigen in vitro transcribed and translated [35S]methionine-labeled human islet GAD65. By using this assay, 77% (77 of 100) of serum samples from recent-onset IDDM patients were positive for GAD65 antibodies compared with 4% (4 of 100) of serum samples from healthy control subjects. In competition analysis with unlabeled purified recombinant human islet GAD65, binding to tracer was inhibited in 74% (74 of 100) of the GAD65-positive IDDM serum samples compared with 2% of the control samples. The levels of GAD antibodies expressed as an index value relative to a standard serum, analyzed with or without competition, were almost identical (r = 0.991). The intra- and interassay variations of a positive control serum sample were 2.9 and 7.6%, respectively (n = 4). The frequency of GAD antibodies was significantly higher with IDDM onset before the age of 30 (80%, 59 of 74) than after the age of 30 (48%, 10 of 21) (P < 0.01). The prevalence of islet cell antibodies showed a similar pattern relative to age at onset. Because simultaneous occurrences of multiple autoimmune phenomena are common, we analyzed sera from patients with other autoimmune diseases. The frequency of GAD antibodies in sera positive for DNA autoantibodies (8% [2 of 25] and 4% [1 of 25] in competition analysis) or rheuma factor autoantibodies [12% (4 of 35) and 3% (1 of 35) in competition analysis] was not different from that in control samples. In contrast, in sera positive for ribonucleoprotein antibodies the frequency of GAD antibodies was significantly increased (73% [51 of 70] and 10% [7 of 70] in competition analysis [P < 0.025]). In conclusion, even large numbers of serum samples can now be tested for GAD65 antibodies in a relatively short time, allowing screening of individuals without a family history of IDDM for the presence of this marker.

Immature transformed rat islet beta-cells differentially express C-peptides derived from the genes coding for insulin I and II as well as a transfected human insulin gene.

Synthetic peptides representing unique sequences in rat proinsulin C-peptide I and II were used to generate highly specific antisera, which, when applied on sections of normal rat pancreas, confirm a homogeneous coexpression of the two C-peptides in all islet beta-cells. Insulin gene expression is induced in the transformed heterogeneous rat islet cell clone, NHI-6F, by transient in vivo passage. During this process a transfected human insulin gene is coactivated with the endogenous nonallelic rat insulin I and II genes. Newly established cultures from NHI-6F insulinomas having a high frequency of insulin-producing cells showed highly differential expression at the cellular level of the three proinsulin C-peptide immunoreactivities, as follows: C-peptide I greater than human C-peptide greater than C-peptide II. The fractions of cells expressing human C-peptide and C-peptide II decreased in time and were absent after more than 50 successive passages, while a C-peptide I-producing population was still present. Double-labeling experiments revealed a heterogeneous distribution of the three different C-peptides. Surprisingly, in the early passages a large fraction of cells would express only a single species of proinsulin-C-peptide immunoreactivity but still at high levels. However, rat C-peptide II and human C-peptide were often colocalized, even in later passages. In situ hybridization studies combined with the immunocytochemical data suggest that the differential expression occurs at the level of transcription.(ABSTRACT TRUNCATED AT 250 WORDS)

Cortisone fails to affect levels of islet cell surface antibodies and incidence of diabetes in the BB rat.

Cortisone acetate (250 micrograms/kg X day) was given by im injections to 40 21-day-old diabetes-prone BB rats. The animals were followed longitudinally to determine islet cell surface antibodies (ICSA), as an expression of an abnormal immune reaction against the pancreatic islet cells and plasma glucose to estimate the degree of metabolic control. ICSA were detected 10-150 days before the diagnosis of diabetes. In the cortisone-treated group the diabetic rats showed significantly higher ICSA values compared to the nondiabetic ones, both in frequency of positive tests (P less than 0.05) and in mean binding values (P less than 0.02). In the control group, no difference in ICSA levels were seen between diabetic and nondiabetic rats. The cortisone regimen also failed to influence the degree of insulitis, commonly associated with diabetes in these rats. These experiments in well defined animals which spontaneously develop diabetes do not support the use of low dose cortisone treatment in attempts to improve or prevent insulin-dependent diabetes in human subjects.

Inhibition of insulin release after passive transfer of immunoglobulin from insulin-dependent diabetic children to mice.

We used the mouse passive transfer model to test whether islet cell antibodies affect beta-cell function. The immunoglobulin (Ig) fraction of plasma from 5 islet cell surface antibody-positive, newly diagnosed insulin-dependent diabetic children or of a pool of plasma from 12 normal subjects was injected daily (7-16 mg IgG/day) for 14 days into normal immunosuppressed BALB/c mice. Insulin secretory responses in the Ig-injected mice were then examined by perfusing the rodent pancreata in vitro. Insulin release induced by 20 mmol/liter D-glucose during 30 min of stimulation decreased from 900 ng insulin (median; range, 814-1138) from pancreata of mice injected with control Ig to 511 ng (range, 130-786) from pancreata of mice injected with diabetic Ig (P less than 0.003). Both the initial peak and the sustained second phase of glucose-stimulated insulin release were depressed in 4 of the 5 pancreata from mice injected with diabetic Ig. These results indicate that circulating antibodies in diabetic children may alter beta-cell function and possibly contribute to the pathogenesis of insulin-dependent diabetes.