Age-related changes in hepatic and splenic insulin receptors and serum insulin and glucose levels in inbred mice

Age-Related

Changes in Hepatic and Splenic Insulin Receptors Insulin and Glucose Levels in Inbred Mice

and Serum

T.M. Murray, K.S. Desai, and B. Cinader Inbred mice of strains A/J, DBAllJ, and SJL/J were housed and aged in our animal colony, and parameters of carbohydrate metabolism were assessed at various ages. The patterns of age-related change were both organ- and strain-specific. Age-related changes in two of the strains were associated with relative carbohydrate intolerance. Common to all three strains was a biphasic pattern of change in hepatic insulin receptor number, with a decrease in early life and a return to earlier levels late in life. In both A/J and DBA/lJ mice, there was a sharp increase in serum insulin level (twofold to 9.7-fold) that corresponded to the decrease in hepatic insulin receptors and was associated with hyperglycemia; no significant change in serum insulin or glucose levels was seen in SJL/J mice, despite a similar biphasic pattern in hepatic insulin receptor concentration. Age-related changes in splenic insulin receptors resembled changes in the liver in A/J and SJL/ J mice, ie, there were synchronous biphasic age-related patterns. This was not the case in the spleens of DBAI 1J mice, in which we did not observe age-related changes. There was no change in insulin receptor affinity with age, nor was there any difference in affinity between tissues or mouse strains. The pattern of change in hepatic insulin receptors and serum insulin levels was more complex than has been previously recognized. We do not know the mechanisms responsible for this complex pattern, but it must involve at least two discrete age-related events. Further experiments will be necessary to elucidate the mechanisms underlying the pattern. Future studies of age-related change in carbohydrate metabolism in inbred mice must include comparisons among multiple ages rather than comparisons between only two ages, ie, young and old mice, as has usually been done in the past. Only by following age-related changes throughout life can we determine the dynamics of age progression and reveal the complexity of changes in carbohydrate metabolism. Copyright 0 1993 by W. B. Saunders Company

T

HROUGHOUT LIFE there are age-related changes in the quantity of macromolecules, including hormones, receptors, and cytokines. These changes usually involve a decrease with age’-“; in some, the magnitude of the decrease is directly proportional to the magnitude of the output during youth and may be regulated by feedback controls In a few cases, an increase rather than a decrease is found and, in at least one of these instances, this can be attributed to changes in the relative proportions of two cell subpopulations that secrete the same product,6 although in different quantities, or a different overlapping spectrum of products7 We have found different types of age-related rate changes in the analysis of different receptors. The rate of changes in adrenoceptors appears to be compartmentalized, ie, the rate differs on B and T lymphocytes2 The rate of agerelated changes of some molecules such as dopamine receptors is directly proportional to the relative density of the receptor in early life.3 The interaction of receptors within cell membranes can undergo age-related changes, eg, the interaction of insulin receptors with class 1 major histocompatibility complex antigens decreases with age.s There have been many reports of age-related changes in insulin receptors in different species.9-‘s We wished to study these changes against the defined genetic background of inbred strains of mice in order to assess the polymorphism

of age-related changes in allelic genetic factors; genes regulating carbohydrate metabolism result in different phenotypes depending on various factors in the genetic background.19 Preliminary studies in our laboratoryzO revealed polymorphism, ie, strain differences, in insulin receptor numbers in liver and spleen. We therefore conducted more detailed assessments at a number of different ages and included measurements of plasma insulin and glucose. We found a more complex time-course of agerelated changes in murine liver and spleen insulin receptor numbers than has been previously demonstrated. In two of three strains studied, changes in hepatic insulin receptors correlated with age-related alterations in carbohydrate tolerance.

MATERIALS Materials

All chemicals were reagent grade. Rat insulin standard (28191 79) was supplied by Novo Biolabs, Danbury, CT. A 14-monoradioiodinated porcine 1251-iodoinsulin (receptor grade, 2,200 Ciimmol) was obtained from New England Nuclear (Boston, MA). Bacitratin (Zn salt), aprotinin, phenylmethylsulfonyl fluoride (PMSF), and bovine serum albumin (fraction V) were purchased from Sigma (St Louis, MO).

Experimental From the Division of Endocrinology and Metabolism, St. Michael’s Hospital, Toronto; and the Departments of Medicine and Immunology and the Centre for the Study of Aging, University of Toronto, Ontario, Canada. Submitted July 22,199O; accepted May 5, 1992. Address reprint requests to T.M. Murray, MD, St. Michael’s Hospital, 38 Shuter St, #212, Toronto, Ontario, Canada MSB lA6. Copyright 0 1993 by W B. Saunders Company 0026-0495/93/4202-0002$03.00/0 140

AND METHODS

Animals

Female 5week-old mice of strains A/J, DBAIlJ, and SJL/J were purchased from Jackson Laboratory (Bar Harbor, ME) and allowed to age in our colony. All animals were housed in groups of five in plastic cages (30 x 18 x 12 cm) at 23°C; a contact bedding, BetaChips (Northeastern Products, Warrensburg, NY), was used and changed twice a week, and the mice were kept on a 12-hour light/dark cycle. Purina Laboratory Rodent Chow no. 5001 (P.M.I. Seeds, Richmond, VA) and water were provided ad libitum. All animals were checked regularly for health problems; only animals without tumors were used in this study. Metabolism,

Vol

42, No 2

(February).

1993:

pp 140-144

141

INSULIN RECEPTORS AND AGING

Biochemical Tests Insulin level was determined by radioimmunoassay in the Core Laboratory of the University of Toronto Banting and Best Diabetes Centre, as described by Livesey et al,21 except that rat insulin was used as a standard. Glucose level was determined by an autoanalyzer method. Blood was taken from the tail vein at various ages for determination of glucose and insulin levels without anesthesia. Assays were performed on several pooled samples from four mice each. Food was withheld for 12 hours before sampling of tissues or blood.

Tissue Preparation Preparation of liver membranes was achieved by a modification of the method of Cuatrecasas??: all work was performed on ice. Freshly obtained livers from five to 10 mice were weighed and brought up to 10 times their volume level with 0.25 mol/L sucrose containing PMSF 2.0 mmol/L, pH 7.4. Livers were minced with scissors and then homogenized in a Polytron homogenizer (Kinematica. Lucerne, Switzerland) for 20 seconds. The homogenates were left to stand at 4°C for 10 minutes, filtered through a nylon sieve, and then centrifuged at 600 x g for 10 minutes. The supernatant was then centrifuged at 12,000 x g for 30 minutes. The resultant supernatant was adjusted to 100 mmol/L with NaCl and to 200 kmol/L with MgS04 and then centrifuged at 40.000 x g for 40 minutes. The membrane pellet was suspended in one half the original volume of 50 mmol/L Tris, pH 7.4, containing 2.0 mol/L PMSF and then recentrifuged at 40,000 X g for 40 minutes to obtain the final pellet, which was suspended in Tris/PMSF and frozen at -70°C. Protein levels were measured in membrane preparations by the method of Lowry et a1.Z3 Suspensions of spleen cells were prepared by mincing spleen tissue gently with scissors in cold CMRL 1066 medium (supplied by the Medical Preparation Unit of the Princess Margaret Hospital, Toronto, Canada) and passing it through a no. 200 mesh stainless steel screen. Thereafter, suspensions were centrifuged at 300 x g for 10 minutes. The cell pellets were resuspended in 17 mmol/L Tris hydrochloride, pH 7.2. containing 0.144 mol/L ammonium chloride for erythrocyte lysis,‘” and then washed in cold 1066 medium containing 10% fetal calf serum (FCS) twice at 300 x g for 10 minutes. The cell pellets were washed in 5 mmol/L Tris hydrochloride with 150 mmol/L NaCl (pH 7.5) twice at 300 x g for 10 minutes, and then the resulting cell pellets were resuspended. as above. and filtered through a no. 200 mesh stainless steel screen. The resulting cell concentration was determined by a Coulter Counter (Coulter Electronics, Hialeah, FL); the cell suspension was adjusted to 4 x lo7 cells/ml. The viability of the cells as determined by dye-exclusion using trypan blue was always greater than 9.5%. Cells were always assayed on the day of preparation.

hsuiitl Receptor Assays The method used was essentially that previously described by Desai et al.? Liver membranes were incubated with lZsIiodoinsulin in phosphate-buffered saline containing 50 mmol/L Tris, pH 7.4. 1% bovine serum albumin, 1 mmol/L PMSF, 0.11 TIUimL aprotinin, and 1.000 mU/mL bacitracin. Binding incubations were performed for 2 hours at 21”C, by which time steadystate binding was achieved. The membrane concentration was 3 mg/mL. At the end of the incubation, membrane-bound lZsIiodoinsulin was separated from unbound radioactivity by pipetting 200 PL incubation mixture into 400-FL plastic microfuge tubes containing 200 )LL chilled buffer and rapidly centrifuging the tubes at 12,000 x g in a microfuge (Beckman Instruments, Fullerton, CA). The supernatant was aspirated and the pellet was washed once with another 200 PL cold buffer. The tube tip containing the

pellet was then sliced off with a scalpel blade and counted in an automatic gamma counter (Searle Instruments, Des Plaines, IL). Insulin degradation was monitored by assessing the precipitation of 1151-iodoinsulin in the incubation mixture when trichloroacetic acid was added to a final concentration of 5%. In all experiments the degradation of radioligand was less than 9% of the total. Specific binding was calculated as the difference between total binding and that measured in the presence of 50 kg/tube excess unlabeled insulin. Insulin receptor concentrations were estimated from binding isotherms obtained by determining binding at nine concentrations of insulin ranging from 10 to 1,250 ng/mL. Results were corrected for between-assay variations in levels of membrane protein. Insulin binding to splenocytes was determined in a similar fashion, but with the following differences. Binding incubations were performed at 4°C with the cell number optimized in order to minimize the receptor loss that occurred at higher incubation temperatures. Binding incubations were performed for 120 minutes at 4°C in a shaker bath. The optimal cell concentration was 4 to 7 x 10’ cells/ml. Binding data were normalized to 2 x IO” cells/tube.

Data AnaJysis Receptor number and affinity were determined from binding isotherms by nonlinear regression analysis using the LIGAND computer program of Munson and Rodbard.lh Scatchard analysis of insulin binding to liver membranes resulted in curves that seldom appeared perfectly linear, but which. on analysis by the LIGAND program, could only be fitted to a single-site model even when data from large numbers of experiments were pooled; significance for a fit to two sites or more was not seen in a single experiment. On the other hand, Scatchard analysis of splenocyte binding consistently gave rise to a straight line plot. Because of the lack of significance for a two-site or more fit in liver membranes and because we wanted to compare our data between tissues using the same methodology, we elected to use data derived from the LIGAND program’s one-site fit throughout our analysis. Tests for statistical significance of differences in receptor number and affinity were performed using a standard two-tailed Student’s t test for unpaired data. RESULTS

When of age

insulin in the

receptor three

binding

strains

was analyzed

of mice, A/J.

as a function DBA/l/J,

and

in receptor number, but not in receptor affinity, and receptor affinity did not differ significantly between organs or strains. In the three strains, the mean affinity of liver insulin receptors was 3.6 nmol/L (range, 1.1 to 12.5 nmol/L) in 17 different independent fits. Figure 1 shows age-related changes in hepatic insulin receptors and serum insulin and glucose levels in the three mouse strains. Common to all strains was a significant decrease in hepatic insulin receptors during early life, and a return at an older age to values similar to those of the earliest age. To explore whether the changes were primarily at the receptor or secondary to changes in circulating insulin, plasma insulin and glucose levels were also determined at various ages; different patterns were seen in different strains. In strain A/J, a significant elevation of serum insulin level was noted at the age at which the density of hepatic insulin SJL/J,

there

were

significant

age-related

changes

142

MURRAY,DESAl,AND

CINADER

DISCUSSION

1.0

0.5

a

b

I&ff do

4b

do do do Age,weeks

do

lb

Fig 1. Hepatic insulin receptors (0). serum insulin levels (0). and serum glucose levels (A) in groups of mice from three strains, A, DBA/ 1, and SJL, at different ages. Data are means * 1 SEM; at points where no error bars are placed, assays were performed on single pools of serum from a group of four animals. The different strains were sampled at different ages because of known strain differences in the rate of aging; in particular, the SJL strain is known to have a short life span.

receptors approached their minimum. This was accompanied by hyperglycemia (206 mg/dL), despite the elevation of insulin levels, suggesting that relative glucose intolerance and insulin resistance had developed by 32 weeks. A similar pattern of age-related change was observed in DBA/lJ mice, with a 9.7-fold increase in serum insulin level occurring at the same age as the nadir in hepatic insulin receptors, again accompanied by hyperglycemia (215 mg/ dL). In SJL/J mice on the other hand, there was no large or consistent age-related pattern of change in either insulin or glucose concentrations accompanying the change in hepatic insulin receptors. Strain differences were also found in the number of splenic insulin receptors, as can be seen in Table 1. In A/J mice, splenic insulin receptor number showed an agerelated decrease and increase similar to the changes in liver, and correlating with changes in glucose tolerance and circulating insulin levels. In DBA/lJ mice, there was no significant change in the density of splenic insulin receptors despite the changes in insulin receptors in the liver, further evidence for compartmentalization of age-related changes.1,2,27 In SJL/J mice, there was an age-related decrease followed by an increase in the density of insulin receptors in both liver and spleen, but there were no significant changes in serum glucose and insulin levels.

It is well established that changes in glucose tolerance occur with age.11~1s,*8~29 A s pectrum of age-related changes in insulin action has been reported, including receptor number, receptor affinity, and post-receptor changes, as well as changes in the capacity of islet cells for insulin secretion; in addition, age-related changes in insulin degradation30 and clearance3i have been reported. While several studies have noted a decrease in human insulin receptors with age,9~llI,l5-17 others have not.14,3’ Pacini et a13” suggest that after correction for variables such as obesity and physical inactivity, there is no significant effect of age on glucose tolerance. On the other hand, Lonnroth and Smith observed an age-related change in insulin resistance in obese subjects, thus distinguishing the aging effect from the the greatest factor in effects of obesity. I7 In humans, age-related glucose intolerance appears to be in insulin target tissues at post-receptor sites.29.34 In aging rats, insulin resistance has also been demonstrated,35 and in addition to reduced receptor density 9.13there is evidence for reductions in receptor kinase activity36 and receptor recycling.37 Furthermore, changes in insulin binding in response to dietary variables are a function of age.12 Decreased insulin responsiveness in fat cells of older rats has been ascribed by Craig et al and Lawrence et al39 to fat cell hypertrophy rather than to a direct effect of aging. Finally, Reaven et a14’r have shown an age-related decrease in insulin secretion in rats, when corrected for cell number. Our analysis of murine aging has revealed a complex pattern of age-related change in hepatic insulin receptor number, plasma insulin level, and plasma glucose level, with interesting similarities and differences between strains. This pattern was more complex than has been previously reported for insulin receptors and more complex than that observed in dopamine receptors in mice of the same strains.” In most previous reports dealing with age-related changes, the changes have been assessed by comparing young and old animals at two arbitrarily chosen ages. In Table 1. Number of Insulin Receptors in Liver and Spleen at Various Ages in Three Strains of Mice Insulin Receptors (nlOl/L Strain

A

DBAII

SJL

x 10-y

Liver

Spleen

5

7.5 (1.3)

0.46 (.04)

24

5.8 (0.6)

33

2.1 (0.4)

Age (4

54

1.7(0.3)

88

5.8 (0.6)

92

6.2 (0.2)

0.26 (.02) 0.003 (.0002) 0.23 1.03)

5

6.9 (0.8)

0.24(.017)

29

1.4(0.3)

0.18 (.024)

64

1.9(0.1)

0.25 (.023)

85

4.0 (0.4)

5

12.0(1.2)

0.72 (.04)

19

1.4(0.2)

24

2.0 (0.5)

0.18 (.06) 0.15 (.Ol)

28

1.3(0.1)

0.22 (.03)

32

7.5 (1.6)

0.95 (.04)

44

10.0(1.0)

NOTE.Valuesin parenthesesarel SEM

0.20 (.015)

143

INSULIN RECEPTORS AND AGING

examining parameters at a number of different ages, we have detected a biphasic pattern of change. Age-related changes in liver insulin receptors were similar in the three strains studied; there was a decrease in the density of insulin receptors during early life and a return to youthful levels with aging. In two of the three strains, A/J and DBA/lJ, the decrease in insulin receptors was synchronized with a significant concomitant increase in plasma insulin level, associated with mild hyperglycemia. This age-related change in carbohydrate metabolism appeared to differ in SJL/J mice; changes in hepatic and splenic insulin receptors were similar to those in the other two strains, but there was no significant change in either glucose or insulin levels. It seems reasonable to conclude that, at least in some strains, significant changes in hepatic insulin receptors may occur without affecting carbohydrate metabolism. For this to occur without a change in circulating insulin levels would require a compensatory change in post-receptor mechanisms. However, it is conceivable that our sampling schedule was not sufficiently frequent to detect a brief transient change in insulin and glucose concentrations in SJL/J mice. We are unable to identify the mechanism for the previously unrecognized age-related change in glucose tolerance of A/J and DBA/lJ mice, but we will make some speculations. The biphasic pattern of change in hepatic insulin receptor numbers suggests two age-related processes, one responsible for the early decrease and the other related to the later increase. The density of hepatic insulin receptors decreases from youth to middle age at a rate proportional to the magnitude of the youthful level in a manner that we have observed in a number of other systems, eg, dopamine receptors3 and have referred to as economic correction.5 The early decrease occurs well past sexual maturity and is therefore not likely to be triggered by changes in secretion of sex hormones related to puberty. The observation can perhaps be interpreted most simply as a primary decrease in the density of insulin receptors, with insulin resistance and secondary hyperinsulinemia. Primary hyperinsulinemia with a secondary decrease in receptor numbers and insulin resistance would be unlikely to cause a significant elevation in plasma glucose level, particularly such a sharp increase. However. other explanations are possible. An elevation in glucose level, with secondary hyperinsulinemia and a resultant decrease in receptor number, might have been induced by a sharp increase in food intake. This seems unlikely to us, since weights of mice in our colony from these strains do not change significantly during this period or indeed during old age (data not shown). The unique feature of the hepatic insulin receptor system is an increase in receptor number following the decrease, ie, a reversal of the decrease in receptor density that occurs

between middle and old age. It is possible that the increase in insulin receptors occurring in old age is due to an age-related decrease in caloric intake; again, this seems unlikely since we observed no significant weight loss in these aging mouse strains. Alternatively, the increase in liver insulin receptors in old age could be due to an age-related change in the balance between subpopulations of liver cells, resulting in a relative increase in cells bearing insulin receptors and/or a relative decrease in cells without insulin receptors. Although age-related metabolic changes have been traditionally ascribed to intracellular metabolic changes, it has become apparent in recent years that such changes may also reflect age-related changes in the relative proportion of cellular subpopulations. We have observed age-related increases in other systems and have attributed them to changes in the relative proportion of two different subpopulations of cells. This has been observed in the case of different types of cardiac myocytes” and. with respect to xanthine oxidoreductase, in rat hepatocytes4’ We have also observed this phenomenon in relation to different types of T helper cells and age-related changes in interleukin-3 and interleukin4.‘.“‘Changes in cell subpopulations must therefore be regarded as an important potential mechanism for age-related changes in the output of a product. Pursuing this line of thought, it is possible to postulate that we can account for the observed biphasic pattern of change on the basis of changes in relative proportions of three subpopulations of cells. Other explanations could involve (1) changes in receptor synthesis at the transcriptional and/or posttranscriptional level, as has been reported recently in genetically obese and diabetic mice,43 or (2) changes in receptor recycling or degradation. Whatever the explanation, there must be at least two separate age-related changes. The fact that the complex pattern of age-related changes in insulin receptors and carbohydrate tolerance reported here has not been observed in earlier studies may be due to the fact that, in the past, experiments on the effects of aging have tended to compare changes at two ages, ie, young and old. Our results indicate that several ages should be compared over the entire life span to get a true picture of age-related changes. Simply comparing liver insulin receptors between 5 weeks of age and the oldest age studied in each strain would have failed to show the significant age-related changes detected in our study; we would have missed the remarkable changes in middle age. More detailed studies of age-related changes may reveal new complexities in the regulation of the insulin receptor throughout life.

ACKNOWLEDGMENT We are grateful to

Dr Errol1

Marks

for helpful discussions.

REFERENCES

senescence: Individual variation and multi-centricity, in de Week A (ed): Lymphoid Cell Functions in Aging, Topics in Aging Research in Europe. vol 3. Rijswijk, The Netherlands, Eurage, 1984, pp 3-16 2. Kohno A, Cinader B. Seeman P: Age-related changes in beta-adrenoceptors of lymphocytes. lmmunol Lett 13:79-82, 1986 1. Cinader

B: Immune

3. Leprohon-Greenwood CE, Cinader B: Variations in agerelated decline in striatal Dz-dopamine receptors in a variety of mouse strains. Mech Ageing Dev 38:199-206. 1987 4. Noth RH. Mazzaferri EL: Age and the endocrine system. Clin Geriatr Med 1:223-250, 1985 5. Cinader B. Dubiski S, Greenwood C, et al: Economic correc-

144

bon: Quantity and age-related rate of change. Mech Ageing Dev 48:111-116, 1989 6. Schoutsen B, De Jong JW: Age-dependent increase in xanthine oxidoreductase differs in various heart cell types. Circ Res 61:604-607, 1987 7. Kubo M, Cinader B: Polymorphism of age-related changes in interleukin production-Differential changes of the subpopulations synthesizing IL-2, IL-3, and IL-4. Eur J Immunol 20:12891296,199O 8. Yip CC, Moule ML, Hirstwood C, et al: Class I histocompatibility antigens and insulin receptors: Evidence for age dependent interactions, in Jaworski MA, Molnar GD, Rajotte RV, et al (eds): The Immunology of Diabetes Mellitus. Amsterdam, The Netherlands, Elsevier, 1986, pp 215-222 9. Olefsky JM, Reaven GM: The effects of age and obesity on ‘251-insulin binding to isolated rat adipocytes. Endocrinology 96: 1486-1495.1975 10. Pagan0 C, Cassader M, Diana A, et al: Insulin resistance in the aged: The role of the peripheral insulin receptors. Metabolism 30:46-49, 1981 11. DeFronzo RA: Glucose intolerance and ageing. Diabetes Care 4:493-501,198l 12. Iritani N, Fukuda H, Tashiro S, et al: Age-dependent alterations of insulin binding to hepatic plasma membranes, erythrocytes and adipocytes in rat. J Biochem 92:569-574,1982 13. Clark JB, Clark CM: Age-related changes in insulin receptor regulation in liver membranes from Zucker fatty rats. Endocrinology 111:964-969, 1982 14. Rowe JW, Minaker KL, Pallotta JA, et al: Characterization of the insulin resistance of aging. J Clin Invest 71:1581-1587, 1983 15. Lonnroth P, DiGirolamo M, Krotiewski M, et al: Insulin binding and responsiveness in fat cells from patients with reduced glucose tolerance and type II diabetes. Diabetes 32748-754. 1983 16. Bolinder J, Osterman J, Amer P: Influence of aging on insulin receptor binding and metabolic effects of insulin in human adipose tissue. Diabetes 32:959-964, 1983 17. Lonnroth P, Smith U: Aging enhances the insulin resistance in obesity through both receptor and post-receptor alterations. J Clin Endocrinol Metab 62:433-437, 1986 18. Fulop T Jr, Nagy JT, Worum I, et al: Glucose tolerance and insulin resistance with aging-Studies on insulin receptors and post-receptor events. Arch Gerontol Geriatr 6:107-115,1987 19. Coleman DL: Obese and diabetes: Two mutant genes causing diabetes-obesity syndromes in mice. Diabetologia 14:141148,1978 20. Desai KS, Ruth J, Murray TM, et al: Age-related changes in insulin receptors in mouse spleen and liver. Clin Invest Med 8:A83. 1985 (abstr) 21. Livesey JH, Hodgkinson SC, Roud HR, et al: Effect of time, temperature and freezing on the stability of immunoreactive LH, SH, TSH, growth hormone, prolactin, and insulin in plasma. Clin Biochem 13:151-155.1980 22. Cuatrecasas P: Affinity chromatography and purification of the insulin receptor of liver cell membranes. Proc Nat1 Acad Sci USA 69:1277-1281, 1972 23. Lowry OH, Rosebrough NJ, Farr AL, et al: Protein measurement with the Folin phenol reagent. J Biol Chem 193:265-275,195l 24. Mishell RI, Shiigi JM, Mishell BB, et al: Prevention of the

MURRAY, DESAI, AND CINADER

immunosuppressive effects of glucocorticosteroids by cell-free factors from adjuvant-activated accessory cells. Immunopharmacology 2:233-245,198O 25. Desai KS, Zinman B. Steiner G, et al: Effect of calcium on ‘151-insulin binding to rat adipocytes. Can J Biochem 56:843-848. 1978 26. Munson PJ, Rodbard D: LIGAND: A versatile computer approach for characterization of ligand-binding systems. Anal Biochem 107:220-239, 1980 27. Meneilly GS, Minaker KL, Elahi D, et al: Insulin action in aging man: Evidence for tissue-specific differences at low physiologic insulin levels. J Gerontol42:196-201, 1987 28. Morrow LA, Halter JB: Carbohydrate metabolism in the elderly, in Sowers JR, Felicelta JU (eds): The Endocrinology of Aging. New York, NY. Raven, 1988, pp 151-173 29. Elahi D, Clark B. Andres R: Glucose tolerance, insulin sensitivity, and age, in Armbrecht HJ, Coe RM, Wonsurawat N, (eds): Endocrine Function and Aging. New York, NY, SpringerVerlag, 1990, pp 48-63 30. Runyan K, Duckworth WC, Kitabchik AE, et al: The effect of age on insulin-degrading activity in rat tissue. Diabetes 28:324325, 1979 31. Minaker KL, Rowe JW, Tonino R, et al: Influence of age on clearance of insulin in man. Diabetes 31851-855, 1982 32. Fink RI, Kolterman OG, Griffin J. et al: Mechanisms of insulin resistance in aging. J Clin Invest 71:1523-1535, 1983 33. Pacini G, Valerio A, Beccaro F, et al: Insulin sensitivity and beta-cell responsivity are not decreased in elderly subjects with normal OG’IT. J Am Geriatr Sot 36:317-323, 1988 34. Fink RI, Kolterman OG, Kao M, et al: The role of the glucose transport system in the post-receptor defect in insulin action associated with human aging. J Clin Endocrinol Metab 58:721-725, 1984 35. Narimiya S, Dolkas CB, Mondon CE, et al: Insulin resistance in older rats. Am J Physiol246:E397-E404, 1984 36. Carrasocsa JM, Ruiz P, Martinez C, et al: Insulin receptor kinase activity in rat adipocytes is decreased during aging. Biochem Biophys Res Commun 160:303-309,1989 37. Trischitta V. Reaven GM: Evidence of a defect in insulinreceptor recycling in adipocytes from older rats. Am J Physiol 254:E39-E44, 1988 38. Craig BW, Garthwaite SM, Holloszy JO: Adipocyte insulin resistance: Effects of aging, obesity, exercise, and food restriction. J Appl Physiol62:95-100. 1987 39. Lawrence JC Jr. Colvin J, Cartee GD, et al: Effects of aging and exercise on insulin action in rat adipocytes are correlated with changes in fat cell volume. J Gerontol44:B88-B92,1989 40. Reaven E, Wright D, Mondon CE, et al: Effect of age and diet on insulin secretion and insulin action in the rat. Diabetes 32:175-180, 1983 41. Levinson DJ, Chalker D: Rat hepatic xanthine oxidase activity. Age and sex specific differences. Arthritis Rheum 23:7782, 1980 42. Kubo M, Cinader B: The distinction between age- and disease-related change. Immunol Lett 24:133-136,199O 43. Ludwig S, Muller-Wieland D, Goldstein BJ, et al: The insulin receptor gene and its expression in insulin-resistant mice. Endocrinology 123:594-600.1988