Insulin binding to myotonic dystrophy fibroblasts

Journal ~! the Neurological Sciences, 1983~ 58:289 295

289

Elsevier Biomedical Press

INSULIN BINDING TO MYOTONIC DYSTROPHY FIBROBLASTS

LISTER LAM t, ARTHUR J. HUDSON 2, KENNETH P. STRICKLAND ~ and GERALD J.M. TEVAARWERK 3

IDep artment 0/ Biochemisto', Universi O' ~2[' Western Ontario; 2Department ~[' Clinical Neurological Sciences, University Ho.spital and Universi O, o[' Western Ontario," and 3Department c~[ Medicine, St. Joseph's Ho,spital and Universi O, of Western Ontario, London, Ontario (Canada) (Received 8 July, 1982) (Accepted 12 August, 1982)

SUMMARY

Insulin receptor binding was examined in cultured skin fibroblasts from 10 myotonic dystrophy patients and 10 age- and sex-matched control subjects. The conditions for insulin binding to fibroblasts were optimal and employed HEPES binding buffer, pH 8.0 at 15°C for 5 h. These conditions correspond to those previously employed with monocytes from MyD subjects. The normalized initial insulin binding capacity showed a decrease of 62'j'Jo from 5.04 ___0.28°,0 of the total labeled insulin added/rag protein in the control to 1.93 +_ 0.13°o in the myotonic dystrophy group (P < 0.01) due mainly to a marked reduction in high affinity receptors or in receptor affinity. The addition of 1.0 ng/ml of unlabeled insulin produced significant decreases to 3.80 + 0.25°/,iin the control group and 1.24 _+ 0.09°;i in the MyD group. The results are similar to previously reported findings with monocytes from myotonic dystrophy patients and suggest that a surface membrane defect exists in this disease. However, the conditions that have been employed in the binding procedures in all of the studies, while optimal, are performed at a high pH and low temperature and could have an important bearing on the interpretation of a membrane disorder.

This work has been supported by the Medical Research Council of Canada, Grant No. MA-5702. Lister Lam was supported by the Muscular Dystrophy Association of Canada, Grant No. PR924. Reprint requests should be addressed to: Arthur J. Hudson, M.D., Department of Clinical Neurological Sciences, University Hospital, London, Ontario, N6A 5A5, Canada. 0022-510X/82/0000-0000/$03.00 © 1983 Elsevier Biomedical Press

290 INTRODUCTION

The reports of insulin binding to monocytes in myotonic muscular dystrophy (MyD) have, with one exception, maintained that insulin binding is decreased (Kobayashi et al. 1977; Festoff and Moore 1979; Tevaarwerk et al. 1979; Moxley et al. 1981 ; Perurena et al. 1981). In the study in which no abnormality was found (Kobayashi et al. 1977) all of 7 patients had glucose intolerance and 5 demonstrated hyperinsulinemia and the possibility was raised that insulin resistance, if it exists, arises from an abnormality in insulin action that is located distal to the insulin binding step. It is difficult to correlate plasma insulin and glucose levels to insulin binding unless particular care is taken to determine these at the time of recovery of the cells from the patient. Recognizing this problem Perurena et al. (1981) were able to correlate insulin binding to the blood glucose and insulin levels by euglycemically "clamping" the blood glucose with a constant intravenous infusion of insulin. They found in confirmation of their earlier studies (Festoff and Moore 1979) both a reduced insulin binding capacity and affinity of MyD monocytes. The possibility remains, however, of altered insulin binding that is due to other in vivo effects when cells are studied immediately following their removal from the subject and to the in vitro conditions of the binding procedure. The present study was initiated to investigate insulin binding to cultured fibroblast monolayers from MyD and control subjects since it is possible to maintain a cellular milieu that is constant and unaffected by the in vivo conditions. MATERIALS AND METHODS

Subjects Eight male and 2 female patients ranging in age from 16 to 65 years (mean + SD 36.8 +_ 14.16) with the typical clinical and electromyographic features of MyD were chosen for the study. For each MyD patient, a sex- and age- (mean + SD 37.3 _+ 12.90) matched control subject with no known metabolic disease and who was undergoing elective surgery was selected.

Supplies and media Monocomponent porcine insulin (26.6 U/mg) was kindly donated by Dr. R. Dolman, Eli Lilly Co., Scarborough, Ont. Bio-Rad Protein Assay Kit was purchased from BIO-RAD Laboratories, Mississauga, Ont. POPOP [1,4-bis-2(5-phenyloxazolyl)benzine] was purchased from Amersham Searle, Oakville, Ont. PPO (2,5-diphenyloxazole), scintillation glass vials, tissue culture and other chemicals were obtained from Fisher Scientific Co., Toronto, Ont. Unless otherwise noted culture and incubation media supplies were purchased from Grand Island Biological Co., Grand Island, NY. The culture media used included (i) modified Eagle's minimal essential medium (MEM) supplemented with Earle's salts and non-essential amino acids, including t-glutamine (Flow Laboratories Inc., Mississauga, Ont.) 10~ (v/v) fetal calf serum and gentamycin (50/~g/ml),

291 (ii) Dulbecco's phosphate-buffered saline (DPBS) and (iii) N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES), added to (i) for buffering, pH 7.4. The incubation medium (IM) consisted of modified Eagle's basal medium (BME) containing NaCI (0.14 M), KC1 (2.7 mM), CaCL_ (0.49 raM), L-glutamine (2 mM), BME amino acids, BME vitamins and bovine plasma albumin (1 mg/ml). Presence of the latter (bovine plasma albumin, fraction V) which was obtained from Armour Pharmaceuticals, Phoenix, AZ would reduce to a minimum any action of extracellular proteases on the fibroblasts.

Cell culture technique All skin biopsies were obtained at the University Hospital, London, Ont. with the informed consent of the patient and the approval of the University of Western Ontario Committee on Human Experimentation. For MyD patients, skin biopsies were obtained from the anterior forearm and abdomen and for the control subjects skin was obtained from the neck, breast or abdomen. Fibroblast cultures were established from the skin of M y D and control subjects and grown in monolayers using methods described by Howard et al. (1976, 1979). At confluence the cultures were subcultivated using a 0.05-0.02'); trypsin EDTA solution in calciumand magnesium-free DPBS. Cultures were screened for mycoplasma. All experiments were performed on cultures between the 6th and the 15th passage with each passage a I :2 split. Cell counts in duplicate were done using a particle counter (Coulter Electronics, Inc., New York, NY). Protein was determined by the method of Bio-Rad Protein Assay. hlsulin binding assay Monoiodinated porcine monocomponent insulin was prepared by the method described by Schneider et al. (1976) yielding [~25I]insulin with a specific activity of 50/~Ci//~g. The binding assay was done in triplicate according to the procedure of Schilling et al. (1979). Fibroblasts were inoculated at high density (5 x 105 cells per well) into Linbro Multiwell (FB-6-TC) tissue culture plates containing six 35-ram wells (Fisher Scientific Co., Toronto, Ont.) and submitted to assay 7 8 days later. Prior to assay, the cell monolayer was washed twice with DPBS containing Ca ~+ and Mg e'. It was then incubated in HEPES binding buffer (pH 8.0; 0.1 M HEPES, 0.12 M NaCI, 1.2 mM MgSO4, 2.5 mM KC1, 10 mM glucose and 10 mg bovine serum albumin/ml) containing ~251-1abeled insulin (100 pg/ml) and varying concentrations of unlabeled insulin (0, 1, 10, 100, 1000 ng/ml) in a final volume of 1.0 ml. Non-specific binding was determined in the presence of a large excess of unlabeled insulin (10 #g/ml). After 5 h of incubation at 15 °C, the incubation medium was aspirated and the cell monolayer was washed quickly 3 times with cold (4°C) HEPES binding buffer and solubilized with 1.0 ml of a solution of NaOH (0.1 M) and Na2CO, (21~;;) for 2 h at room temperature. For assay of radioactivity, 0.4 ml of the lysate of each well was transferred to counting tubes in duplicate and the radioactivity was determined in a ;,-counter (Micromedic Systems 4/200 Automatic Gamma Counter, CanLab, Toronto, Ont.). For protein determination, 0.04 ml of

292

the lysate of each well was diluted to 0.4 ml in duplicate and the amount of protein was determined as described above. Results were expressed as the percent of total radioactivity added to each well/mg protein.

Statistical analysis The Wilcoxon rank sum test for independent samples was used to determine differences between control and MyD groups and differences between insulinstimulated and basal conditions (Colton 1974). RESULTS The results of the insulin binding studies on the 10 M y D and 10 control subjects were normalized to 1 mg protein to allow comparisons of the data. Figure 1 shows the initial binding capacity (expressed as the mean _+ SEM percent of the total labeled insulin added/mg of protein) of 5.04 +_.0.28% in the control group to be significantly decreased (P < 0.01) by 62% to 1.93 + 0.13~ in the MyD group: The addition of 1.0 ng/ml unlabeled insulin produced a significant decrease (P < 0101) in the control group to 3:80 _+ 0.25% and in the MyD group to 1.24 + 0.09~o. The ratio of cell number and protein concentration were not different in the MyD and control fibroblasts. No significant differences were found in insulin binding to fibroblasts from different biopsy sites. Any effects of internalization and degradation of insulin were minimal in view of the low temperature (15 °C) of the binding assay (Soil et al. 1975; Gliemann and Sonne 1978). Scatchard analysis (Fig. 2) showed that a significant decrease in affinity occurred in both the control and MyD groups as receptor occupancy increased (a > b ; b > c ; c > d ; P <0.01 for the control group and a' > b ' ; b' > c ' ; c' > d ' ; P < 0.01 for the MyD group). The mean (+ SEM) affinities of the normal and MyD groups for the initial part of each curve (i.e. a and a', respectively) were not 6£)O

2.ooi!. . . .

o.,

,.o

,o

,ooo

INSULIN (ng/m[)

Fig. I. Dose-response curves of insulin binding in 10 M y D patients ( O - - - O ) and 10 matched normal control subjects ( O - - O ) . Data are corrected For non-specific binding and are represented as the mean +_ SEM (where no bars are shown the SEM is less than the diameter of the point) of the percent

of insulin bound/mg protein/well. The cells were incubated for 5 h at 15°C in the presenceof 0.1 ng/ml [125I]insulin and increasing doses of unlabeled insulin.

293 0 061

--0-CONTROL ~ 0 ~ " MyD

OD

0.04C Z 0.03C

0.020( (23 Z 0"0101

0

0100

0200

0.500

0.400

0.500

0.900

1.000

INSULIN BOUND ( n M )

Fig. 2. Scatchard plot [bound/free, (B/F) plotted against the concentration of bound insulin (B)] or the data shown in Fig. 1. The slope is the negative of the association or equilibrium binding constant. K (i.e. slope K), while the intercepts on the abscissa represent the insulin bound (receptor concentration). The results plotted are the means of the individual values of 10 MyD patients (© O) and 10 matched normal subjects (Q O). The letters a, b, c and d, a', b', c' and d' identify parts of Ihe curve and are compared in the text. significantly different giving values of 2.50 + 0.15 nM ~ and 3,18 _+ 0.34 nM respectively. However, the intercept of the a and a' slopes on the abscissa is less in the MyD than in the control group which suggests an actual loss of receptors in the high affinity range. Due to the relatively low insulin binding evident in fibroblasts as occupancy increases, it was not possible to extrapolate exact total receptor concentrations. DISCUSSION Fibroblast culture provides a useful preparation for the study of cellular defects, such as in MyD, since the influence of the in vivo milieu is removed from the experimental conditions. However, this is replaced by conditions that could produce other effects. In the present study of insulin binding to cultured fibroblasts from M y D subjects the findings of reduced insulin binding in the high affinity range were similar to those that have been described earlier using monocytes from MyD subjects ( F e s t o f f a n d Moore 1979; Tevaarwerk et al. 1979; Perurena et al. 1981). It appears, therefore, that the decrease in high affinity binding in MyD does not result from the in vivo milieu and is in support to the earlier views that a cell surface membrane defect is present in MyD. Such a defect is likely an important factor in the altered glucose metabolism that is observed clinically and in the reduced 2-deoxyglucose uptake in adipose tissue from M y D subjects reported by Mably et al. (1981). Insulin binding studies are usually performed under conditions that provide optimal binding but they are also unphysiological. The optimal insulin binding conditions for fibroblasts, as employed generally and in the present study, consists of cells that are exposed to [t25I]insulin and variable concentrations of cold insulin

294 in HEPES binding buffer, pH 8.0, at 15 °C for 5 h. These conditions are similar to those that have been used in the earlier studies for insulin binding to monocytes (pH 7.6, 16 °C for 90 min). Thus, it may be justifiably argued that the variable in vivo conditions that prompted the present study, and were a concern of Perurena et al. (1981), are of little significance when compared with those of a somewhat extreme nature used in the binding procedure. Insulin binding to fibroblasts can vary considerably with changes in p H and temperature. Podskalny and Kahn (1982) found that a reduction from pH 8.0 to pH 7.2 produced a decrease in insulin binding by more than 50~o in normal cells and Hidaka et al. (1981) also observed a striking decrease in insulin binding, but only at low insulin concentrations, to fibroblasts with a pH change from 7.4 to 6.9. Reduced insulin binding at the lower p H is due to a decrease in the high affinity receptors (De Meyts et al. 1976). Receptor binding studies with fibroblasts, as in other reported studies, are performed at low temperatures (15°C in the present study) because receptor degradation is a factor at higher temperatures. Soll et al. (1975) found that receptor degradation in liver plasma membrane was 15% of the initial receptor population/h at 37 °C. In cultured lymphocytes Taylor et al. (1981) found that insulin binding is decreased by approximately 250~/o as the temperature is increased from 12 °C to 37 °C (pH 7.8). At 15 °C receptor degradation is only 5 ~ or less in 5 h and while there might be difference in degradation in MyD fibroblasts compared with control cells we believe that it is essential to assess this under physiological conditions of temperature and pH. The effects of temperature and pH may be important in the study of membrane functions in MyD. Vickers et al. (1979) found an altered temperature response of membrane phosphorylation in erythrocyte membranes. They compared protein kinase activity at 30 °C and 37 °C and observed that the increase in activity occurring at the higher temperature was less in MyD. It is concluded that insulin receptor binding studies in cultured MyD fibroblasts confirm earlier studies of insulin binding to monocytes in this disease. However, similar conditions were also employed in the binding procedure in these studies and could have an important bearing on the findings. Accordingly, insulin binding studies under physiological conditions are required in order to assess more appropriately insulin binding in relation to the apparent cell surface membrane defect in MyD. REFERENCES Colton, T. (1974) Statistics in Medicine, Little, Brown & Co., Boston, MA. De Meyts, P., A. R. Bianco and J. Roth (1976) Site-site interactions among insulin receptors, J. biol. Chem., 251:1877 1888. Festol/". B.W. and W.V. Moore (1979) Evaluation of the insulin receptor in myotonic dystrophy, Ann. Neurol., 6: 60--65. Gliemann, J. and O. Sonne (1978) Binding and receptor-mediateddegradation of insulin in adipocytes, J. biol. Chem., 253: 7857-7863. Hidaka, H., B.V. Howard, F. lshibashi, F.C. Kosmakos, J.W. Craig, P.H. Bennett and J. Larner (1981) Effect of pH and 3-hydroxybutyrateon insulin binding and action in cultured human fibroblasts, Diabetes, 30: 402406.

295 Howard, B. V., M. de la Llera and W. J. Howard (1976) A new method for the establishment ot" diploid fibroblast cell cultures from human foreskins, Proc. Soc. exp. Biol. Med., 153: 28(~283. Howard, B.V., D. M. Mort, R. M. Fields and P.H. Bennett (1979) Insulin stimulation of glucose entry in cultured human fibroblasts, J. Cell Physiol., 101:129 138. Kobayashi, M., J.C. Meek and E. Streib (1977) The insulin receptor in myotonic dystrophy, J. elm. Endocrinol. Metab., 45: 821--824. MaNy, E., K.P. Strickland, G . J . M . Tevaarwerk and A.J. Hudson (1981) Glucose transport and oxidation in adipose tissue of patients with myotonic dystrophy, J. neurol. Sci., 52:11-23. Moxley, III, R.T., J.N. Livingston, D.H. Lockwood, R.C. Griggs and R.L. Hill (1981) Abnormal regulation of monocyte insulin-binding affinity after glucose ingestion in patients with myotonic dystrophy, Proc. Nat. Aead. Sci., 78:2567 2571. Perurena, O. H., W.V. Moore and B.W. Festoff (1981) Correlation of receptor capacity, affinity and in vivo tissue insulin sensitivity in myotonic dystrophy (MyD), Abstract Issue, Amer. Academy o! Neurology, p. 116. Podskalny, J. M. and R. Kahn (1982) Cell culture studies on patients with extreme insulin resistance, Part 1 (Receptor defects on cultured fibroblasts), J. din. Endocrinol. Metab., 54:261 268. Schilling, E.E., M.M. Rechler, C. Greenfield and A . M . Rosenberg (1979) Primary defect of insulin receptors in skin fibroblasts cultured from an infant with leprechaunism and insulin resistance, Proc. Nat. Acad. Sci., 76:5877 5881. Schneider, B.S., E. Straus and R.S. Yalow (1976) Some considerations in the preparation of radioiodoinsulin for radioimmunoassay and receptor assay, Diabetes, 25: 26(3~267. Soil, A. H., C. R. Kahn and D. M. Neville, Jr. (1975) Insulin binding to liver plasma membranes in the obese hyperglycemic (ob/ob) mouse Demonstration of a decreased number of functionally normal receptors, J. biol. Chem., 250:4702 4707. Taylor, S. 1.. J. Roth, R.M. Blizzard and M.J. Elders (1981) Qualitative abnormalities in insulin binding in a patient with extreme insulin resistance - Decreased sensitivity to alterations in temperature and pH, Proe. Nat. Aead. Sci., 78:7157 7161. Tevaarwerk, G. J. M., K. P. Strickland, C. H. Lin and A.J. Hudson (1979) Studies on insulin resistance and insulin receptor binding in myotonia dystrophica, J. clin. Endocrinol. Metab., 49:216 222. Vickers, J. D., A~ J. McComas and M. P. Rathbone (1979) Myotonic muscular dystrophy - - Abnormal temperature response of membrane phosphorylation in erythrocyte membranes, Neurology (Mimwap.), 29:791 796.