Characterization
of Purified Autoantibodies to the Insulin Receptor From Six Patients With Type B Insulin Resistance
Orlando
Rodriguez, Elaine Collier, Richard Arakaki,
and Phillip Gorden
Anti-insulin-receptor autoantibodies are present in the serum of patients with the type 6 syndrome of extreme insulin resistance. Sera from six patients with this syndrome were purified over protein-A agarose to remove insulin and other serum factors and obtain an immunoglobulin fraction. These purified fractions were used to quantitatively determine the antibodies’ activity in three separate assays. The ability to inhibit insulin binding was determined in an assay using fibroblasts that overexpress the human insulin receptor; the ability to immunoprecipitate the receptor was determined in an assay using biosynthetically labeled insulin receptors rather than insulin cross-linked receptors; and the ability to stimulate glucose oxidation was determined in isolated adipocytes. We show that the ability of these antibodies to inhibit insulin binding is tightly coupled to their ability to immunoprecipitate the biosynthetically labeled receptor, but neither assay predicts the bioactivity of these immunoglobulins. We suggest that the inability to show this tight coupling in the past may be due to methodological differences. We find no evidence that these antibodies are anti-idiotypic antibodies. Copyright 0 1992 by W. B. Saunders Company
A
UTOANTIBODIES to the insulin receptor are found in the sera of patients with the type B syndrome of extreme insulin resistance.‘.’ The clinical manifestations observed in these patients consist most commonly of severe fasting hyperglycemia associated with extremely high insulin levels.‘-” Occasionally, spontaneous fasting hypoglycemia follows periods of hyperglycemia: and, rarely, hypoglycemia is the only manifestation of these autoantibodiesP In vitro, these antibodies have been shown to inhibit [“‘I]insulin binding,‘,’ to immunoprecipitate the insulin receptor,6 to display insulin-like activity in fat’ and muscle cells,8 and with prolonged incubation, to desensitize target tissues to the effects of insulin.” In vivo, in the fasted rat, acute injections of these autoantibodies produce hypoglycemia, whereas chronic infusions into the fed rat produce hyperglycemia.” The precise nature of these antibodies in each patient is unknown. Since these sera are polyclonal, there can be multiple antibodies, each with its individual epitope, leading to multiple effects. Further, the presence of other components in serum may affect the behavior of these autoantibodies. In an attempt to further understand the nature of these pathologic autoantibodies, we have systematically studied six patients’ sera by partially purifying the immunoglobulins and measuring in a dose-dependent fashion the ability of the extract to inhibit [“‘I]insulin binding, immunoprecipitate the insulin receptor, and stimulate glucose oxidation in fat cells. Several features emerge that are helpful in characterizing the properties of these autoantibodies. MATERIALS
Purification
AND METHODS
of Immunoglobulins
concentrated in 0.2 mol/L phosphate buffer, pH 7.0. The Ig concentration was determined by BCA Protein Assay (Pierce, Rockford, IL) and spectrophotometry, using human IgG as the standard. Materials and Cell Culture
Protein-A agarose was obtained from Bethesda Research Laboratories (Bethesda, MD). [‘Hlmannose (60 Ci/mmol), D-[l“Clglucose (5 to 10 mCi/mmol), Econofluor, Protosol, and EN’HANCE were purchased from New England Nuclear (Boston, MA). Amersham provided the [‘ZSI]insulin(2,000 Cilmmol). Purified pork insulin was purchased from Elanco (Indianapolis, IN). Guinea pig anti-insulin antiserum (G-619) was obtained from Indiana University (Indianapolis, IN). All other chemicals used were reagent grade. National Institutes of Health (NIH) 3T3 cells transfected with human insulin receptor (hIR) cDNA” were grown in Dulbecco’s modified Eagle’s medium (DMEM) (Biofluids, Rockville, MD) supplemented with 10% fetal calf serum, 2 mmol/L glutamine, and 600 pg/mL G-418 (Gibco, Grand Island, NY). For insulin-binding studies, the cells were grown to confluence in 30-mm wells (-3 x 10’ cells). For biosynthetic labeling of insulin receptors, confluent cells (- lo9 cells) were incubated in complete media and continuously labeled with the addition of [SH]mannose (0.1 mCi/ mL) for 18 hours at 37°C. The cells were then solubilized in 1% (vol/vol) Triton X-100 (RPI, Elk Grove, IL) in 50 mmol/L HEPES, 150 mmol/L NaCl buffer, pH 7.4, with 1 mmol/L phenylmethylsulfonyl fluoride (PMSF) and aprotinin (1.5 trypsin-inhibitor units/ mL). This solubilized insulin-receptor preparation was aliquoted and stored at -70°C for use in the immunoprecipitation studies described below. [‘25Z]Znsulin-Binding-Inhibition
Studies
The hIR-transfected, NIH 3T3-cell monolayers were washed with 1 mL of phosphate-buffered saline (PBS), pH 7.4, and
From Patients With Insulin
Resistance
Sera from six patients with type B extreme insulin resistance were studied. Four patients, B-d, B-7, B-10, and B-g, have been previously reported,‘“~‘3 and two new patients, B-15 and B-b, are included in this study. Sera from these six patients and a normal individual were applied to a lo-mL column of protein-A agarose and allowed to equilibrate for 15 hours at 4°C. The column was washed with 200 mLof 0.2 mol/L phosphate buffer, pH 7.0, and the bound immunoglobulin (Ig) was eluted with 100 mL of 0.2 mol/L glycine-HCI buffer, pH 2.8.14 The Ig was dialyzed and then Metabolism, Vol41, No 3 (March), 1992: pp 325-331
From the Diabetes Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD. Address reprint requests to Elaine Collier, MD, Diabetes Branch, National Institute of Diabetes and Digestive and Kidney Diseases, Bldg IO, Room S-S-243, National Institutes of Health, Bethesda, MD 20892. Copyright 0 1992 by W.B. Saunders Company 0026-0495/92/4103-0017$03.00/O 325
RODRIGUEZ ET AL
incubated in duplicate with various concentrations of Ig in binding buffer (100 mmol/LHEPES, 120 mmol/L NaCI. 1 mmol/L MgSO,, 1 mmol/L EDTA, 10 mmol/L glucose, 1.5 mmol/L sodium acetate, and 0.1% bovine serum albumin [BSA], pH 8.0) for 2 hours at 22°C. The binding buffer was then aspirated, and the cells were incubated with 1 mL of [“51]insulin (- 15,000 cpm) for 1 hour at 15°C. The specific binding was determined by incubating with binding buffer alone followed by [‘ZI]insulin (45% to 55% of total counts), and the nonspecific binding (
of Purified Autoantibodies to the Insulin Receptor From Six Patients With Type B Insulin Resistance
Orlando
Rodriguez, Elaine Collier, Richard Arakaki,
and Phillip Gorden
Anti-insulin-receptor autoantibodies are present in the serum of patients with the type 6 syndrome of extreme insulin resistance. Sera from six patients with this syndrome were purified over protein-A agarose to remove insulin and other serum factors and obtain an immunoglobulin fraction. These purified fractions were used to quantitatively determine the antibodies’ activity in three separate assays. The ability to inhibit insulin binding was determined in an assay using fibroblasts that overexpress the human insulin receptor; the ability to immunoprecipitate the receptor was determined in an assay using biosynthetically labeled insulin receptors rather than insulin cross-linked receptors; and the ability to stimulate glucose oxidation was determined in isolated adipocytes. We show that the ability of these antibodies to inhibit insulin binding is tightly coupled to their ability to immunoprecipitate the biosynthetically labeled receptor, but neither assay predicts the bioactivity of these immunoglobulins. We suggest that the inability to show this tight coupling in the past may be due to methodological differences. We find no evidence that these antibodies are anti-idiotypic antibodies. Copyright 0 1992 by W. B. Saunders Company
A
UTOANTIBODIES to the insulin receptor are found in the sera of patients with the type B syndrome of extreme insulin resistance.‘.’ The clinical manifestations observed in these patients consist most commonly of severe fasting hyperglycemia associated with extremely high insulin levels.‘-” Occasionally, spontaneous fasting hypoglycemia follows periods of hyperglycemia: and, rarely, hypoglycemia is the only manifestation of these autoantibodiesP In vitro, these antibodies have been shown to inhibit [“‘I]insulin binding,‘,’ to immunoprecipitate the insulin receptor,6 to display insulin-like activity in fat’ and muscle cells,8 and with prolonged incubation, to desensitize target tissues to the effects of insulin.” In vivo, in the fasted rat, acute injections of these autoantibodies produce hypoglycemia, whereas chronic infusions into the fed rat produce hyperglycemia.” The precise nature of these antibodies in each patient is unknown. Since these sera are polyclonal, there can be multiple antibodies, each with its individual epitope, leading to multiple effects. Further, the presence of other components in serum may affect the behavior of these autoantibodies. In an attempt to further understand the nature of these pathologic autoantibodies, we have systematically studied six patients’ sera by partially purifying the immunoglobulins and measuring in a dose-dependent fashion the ability of the extract to inhibit [“‘I]insulin binding, immunoprecipitate the insulin receptor, and stimulate glucose oxidation in fat cells. Several features emerge that are helpful in characterizing the properties of these autoantibodies. MATERIALS
Purification
AND METHODS
of Immunoglobulins
concentrated in 0.2 mol/L phosphate buffer, pH 7.0. The Ig concentration was determined by BCA Protein Assay (Pierce, Rockford, IL) and spectrophotometry, using human IgG as the standard. Materials and Cell Culture
Protein-A agarose was obtained from Bethesda Research Laboratories (Bethesda, MD). [‘Hlmannose (60 Ci/mmol), D-[l“Clglucose (5 to 10 mCi/mmol), Econofluor, Protosol, and EN’HANCE were purchased from New England Nuclear (Boston, MA). Amersham provided the [‘ZSI]insulin(2,000 Cilmmol). Purified pork insulin was purchased from Elanco (Indianapolis, IN). Guinea pig anti-insulin antiserum (G-619) was obtained from Indiana University (Indianapolis, IN). All other chemicals used were reagent grade. National Institutes of Health (NIH) 3T3 cells transfected with human insulin receptor (hIR) cDNA” were grown in Dulbecco’s modified Eagle’s medium (DMEM) (Biofluids, Rockville, MD) supplemented with 10% fetal calf serum, 2 mmol/L glutamine, and 600 pg/mL G-418 (Gibco, Grand Island, NY). For insulin-binding studies, the cells were grown to confluence in 30-mm wells (-3 x 10’ cells). For biosynthetic labeling of insulin receptors, confluent cells (- lo9 cells) were incubated in complete media and continuously labeled with the addition of [SH]mannose (0.1 mCi/ mL) for 18 hours at 37°C. The cells were then solubilized in 1% (vol/vol) Triton X-100 (RPI, Elk Grove, IL) in 50 mmol/L HEPES, 150 mmol/L NaCl buffer, pH 7.4, with 1 mmol/L phenylmethylsulfonyl fluoride (PMSF) and aprotinin (1.5 trypsin-inhibitor units/ mL). This solubilized insulin-receptor preparation was aliquoted and stored at -70°C for use in the immunoprecipitation studies described below. [‘25Z]Znsulin-Binding-Inhibition
Studies
The hIR-transfected, NIH 3T3-cell monolayers were washed with 1 mL of phosphate-buffered saline (PBS), pH 7.4, and
From Patients With Insulin
Resistance
Sera from six patients with type B extreme insulin resistance were studied. Four patients, B-d, B-7, B-10, and B-g, have been previously reported,‘“~‘3 and two new patients, B-15 and B-b, are included in this study. Sera from these six patients and a normal individual were applied to a lo-mL column of protein-A agarose and allowed to equilibrate for 15 hours at 4°C. The column was washed with 200 mLof 0.2 mol/L phosphate buffer, pH 7.0, and the bound immunoglobulin (Ig) was eluted with 100 mL of 0.2 mol/L glycine-HCI buffer, pH 2.8.14 The Ig was dialyzed and then Metabolism, Vol41, No 3 (March), 1992: pp 325-331
From the Diabetes Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD. Address reprint requests to Elaine Collier, MD, Diabetes Branch, National Institute of Diabetes and Digestive and Kidney Diseases, Bldg IO, Room S-S-243, National Institutes of Health, Bethesda, MD 20892. Copyright 0 1992 by W.B. Saunders Company 0026-0495/92/4103-0017$03.00/O 325
RODRIGUEZ ET AL
incubated in duplicate with various concentrations of Ig in binding buffer (100 mmol/LHEPES, 120 mmol/L NaCI. 1 mmol/L MgSO,, 1 mmol/L EDTA, 10 mmol/L glucose, 1.5 mmol/L sodium acetate, and 0.1% bovine serum albumin [BSA], pH 8.0) for 2 hours at 22°C. The binding buffer was then aspirated, and the cells were incubated with 1 mL of [“51]insulin (- 15,000 cpm) for 1 hour at 15°C. The specific binding was determined by incubating with binding buffer alone followed by [‘ZI]insulin (45% to 55% of total counts), and the nonspecific binding (
