0021-972X/79/4906-0810$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1979 by The Endocrine Society
Vol. 49, No. 6 Printed in U.S.A.
Decreased Insulin Binding to Monocytes from Normal Pregnant Women* HENNING BECK-NIELSEN, CLAUS KUHL, OLUF PEDERSEN, CLAUS BJERRECHRISTENSEN, TORSTEN TOFTEGAARD NIELSEN, AND JOACHIM G. KLEBE Medical Department III and the Department of Clinical Chemistry Amtssygehuset, Aarhus; the Department of Obstetrics and Gynecology, Kommunehospitalet, Aarhus; the Medical Department, Bispebjerg Hospital, Copenhagen; and the Diabetes Center, the Department of Obstetrics and Gynecology YB, Rigshospitalet, Copenhagen, Denmark
ABSTRACT. To ascertain whether the decrease of glucose tolerance in pregnancy might be mediated by changes in insulin receptors, we have studied insulin binding to monocytes in 12 normal women during late pregnancy and 14 healthy, young, normal weight, nonpregnant female controls. The pregnant women had significantly higher fasting insulin concentrations in plasma than the controls (18 ± 3.5 vs. 8 ± 1.1 juU/ml; P < 0.01). Fasting concentrations of glucose and ketone bodies in plasma were not significantly different in the two groups. Insulin binding to monocytes from pregnant women was about 35% lower at
each insulin concentration tested compared to the nonpregnant controls (P < 0.01 at tracer insulin concentrations). Changes in cellular insulin binding were due to changes of the receptor number per cell, whereas the receptor affinity was unaffected. Insulin binding was not significantly correlated with the fasting plasma insulin in either of the two groups (P > 0.1). Our results suggest that the deterioration of glucose tolerance in normal late pregnancy might be explained by a decrease of insulin sensitivity caused by a reduction of the number of insulin receptors. (J Clin Endocrinol Metab 49: 810, 1979)
I
reported that cellular insulin binding is decreased in obese subjects and obese maturity-onset diabetics and that the decrease in insulin binding is positively correlated to the insulin insensitivity of these patients (7-9). The purpose of the present study was to examine whether the insulin receptor is defective in pregnancy.
N PREGNANCY, glucose tolerance gradually deteriorates in spite of steadily increasing levels of plasma insulin both in the fasting state and after oral intake of glucose or meals (1). Moreover, the fasting molar insulin to glucagon ratio (2) is enhanced (3), the suppression of glucagon during hyperglycemia is exaggerated and sustained (3), and the glucagon response to a meal is reduced (4) in pregnancy. Finally, the MCR of insulin and the insulin disappearance rate from circulation are unchanged in pregnancy (5, 6). Hence, the diabetogenicity of pregnancy is not caused by inappropriate changes in the secretory function of the endocrine pancreas or an increased degradation of insulin. Impaired glucose tolerance associated with high levels of insulin in plasma points to a state of insulin insensitivity as the most likely explanation for the diabetogenicity of pregnancy. Pregnant women have a striking resemblance to obese subjects and obese maturity-onset diabetics as regards their insulin insensitivity. It has been
Materials and Methods Subjects Twelve healthy pregnant women (average age, 26.1 yr) were studied between the 34th and 36th weeks of pregnancy. None of them received diuretics or any other drug during pregnancy. Before the actual pregnancy, all were within 10% of their ideal body weight (10). The women were without glycosuria and the weight gain during pregnancy never exceeded 11 kg. The control group comprised 14 healthy, young, normal weight, nonpregnant female volunteers (average age, 22.3 yr). None of them had a family history of diabetes. Investigative procedure
Received February 26,1979. Address all correspondence and requests for reprints to: Dr. Claus Kiihl, Hvid0re Hospital, DK-2930, Klampenborg, Denmark. * This work was supported by grants from the Danish Medical Research Council; the Danish Hospital Foundation for Medical Research Region of Copenhagen, the Faroe Islands, and Greenland; Aarhus Universitets forskningsfond; Landsforeningen for Sukkersyges Fond; Novo's fond; and Nordisk Insulinfond.
All subjects ate a normal diet containing at least 300 g carbohydrate for at least 3 days before each sampling. Blood (120 ml) was drawn from an antecubital vein at 0800 h after an overnight fast and abstinence from smoking. The blood was transferred to tubes containing EDTA (dipotassium salt) for the analytical procedures. 810
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 11 May 2015. at 02:47 For personal use only. No other uses without permission. . All rights reserved.
DECREASED INSULIN BINDING IN PREGNANCY Laboratory analyses Plasma insulin concentration was measured with the RIA of Heding (11). Glucose concentration in plasma was measured using an o-toluidine method (12). The concentrations of acetoacetate and 3-hydroxybutyrate in plasma were measured separately using an enzymatic micromethod (13). Cell binding studies. Mononuclear leucocytes were isolated from the EDTA-blood by gradient contrifugation (14). The cells were washed twice and incubated in dublicates of 250 /x\ of Hepes buffer (100 mM, pH 7.8, at 15 C, with 1% human serum albumin) at a concentration of about 50 X 106/ml for 100 min at 15 C with [I25I]insulin at a concentration of 34 pM (0.2 ng/ ml). The specific activity of the tracer was about 150 juCi//ig. For competition studies, native insulin in increasing concentrations was added to the incubation medium. At the end of the incubation period, 250 ju.1 cell suspension were added to a 1.5-ml microtube containing 1.0 ml ice-cold buffer and 100 /xl silicone oil (relative density, 1.04). Cell-bound and free insulin were separated by centrifugation. The specific cell-binding fraction is defined as the total cell-binding fraction minus the nonspecific cell-binding fraction, i.e. the radioactivity which remained bound in the presence of an excess of native insulin (10 JUM). This fraction averaged 10% of the total binding. The monocytes were identified in cytocentrifuged smears stained with a-naphthyl acetate esterase, and the specific cell-binding fraction was adjusted to a standard concentration of monocytes of 1.0 X 107/ ml using the formula previously described (14). There was no statistically significant difference between the numbers of monocytes in the suspensions of mononuclear leucocytes incubated from normal women and in those from pregnant women (8.7 ± 2.8 vs. 10.5 ± 4.5 X 106/ml; P > 0.1).
TABLE 1. Overnight fasting values of plasma glucose, insulin, and ketone body concentrations in the groups studied
Plasma glucose (mmol/ liter) Plasma insulin (/xU/ml) Plasma ketone bodies (mmol/liter)
Statistical methods. Wilcoxon's two-sample test was employed for comparison of values in the two groups studied, while Spearman's coefficient of rank (r) was applied in correlation studies.
Results As expected (1), fasting plasma insulin was significantly (P < 0.01) higher in the pregnant women than in the nonpregnant controls. In contrast, fasting plasma glucose and ketone bodies did not differ significantly in the two groups (Table 1). Thus, the pregnant women appeared to be insulin resistant. Correspondingly, insulin binding to monocytes from pregnant women was significantly lower than that observed in the controls (Fig. 1). In the pregnant subjects, insulin binding was significantly reduced at each insulin concentration tested (0.034-5 nM,
Pregnant
Control
P
4.6 ± 0.24
4.8 ± 0.34
NS
18.0 ± 12.0 0.097 ± 0.046
8.0 ± 4.0 0.134 ± 0.095
Vol. 49, No. 6 Printed in U.S.A.
Decreased Insulin Binding to Monocytes from Normal Pregnant Women* HENNING BECK-NIELSEN, CLAUS KUHL, OLUF PEDERSEN, CLAUS BJERRECHRISTENSEN, TORSTEN TOFTEGAARD NIELSEN, AND JOACHIM G. KLEBE Medical Department III and the Department of Clinical Chemistry Amtssygehuset, Aarhus; the Department of Obstetrics and Gynecology, Kommunehospitalet, Aarhus; the Medical Department, Bispebjerg Hospital, Copenhagen; and the Diabetes Center, the Department of Obstetrics and Gynecology YB, Rigshospitalet, Copenhagen, Denmark
ABSTRACT. To ascertain whether the decrease of glucose tolerance in pregnancy might be mediated by changes in insulin receptors, we have studied insulin binding to monocytes in 12 normal women during late pregnancy and 14 healthy, young, normal weight, nonpregnant female controls. The pregnant women had significantly higher fasting insulin concentrations in plasma than the controls (18 ± 3.5 vs. 8 ± 1.1 juU/ml; P < 0.01). Fasting concentrations of glucose and ketone bodies in plasma were not significantly different in the two groups. Insulin binding to monocytes from pregnant women was about 35% lower at
each insulin concentration tested compared to the nonpregnant controls (P < 0.01 at tracer insulin concentrations). Changes in cellular insulin binding were due to changes of the receptor number per cell, whereas the receptor affinity was unaffected. Insulin binding was not significantly correlated with the fasting plasma insulin in either of the two groups (P > 0.1). Our results suggest that the deterioration of glucose tolerance in normal late pregnancy might be explained by a decrease of insulin sensitivity caused by a reduction of the number of insulin receptors. (J Clin Endocrinol Metab 49: 810, 1979)
I
reported that cellular insulin binding is decreased in obese subjects and obese maturity-onset diabetics and that the decrease in insulin binding is positively correlated to the insulin insensitivity of these patients (7-9). The purpose of the present study was to examine whether the insulin receptor is defective in pregnancy.
N PREGNANCY, glucose tolerance gradually deteriorates in spite of steadily increasing levels of plasma insulin both in the fasting state and after oral intake of glucose or meals (1). Moreover, the fasting molar insulin to glucagon ratio (2) is enhanced (3), the suppression of glucagon during hyperglycemia is exaggerated and sustained (3), and the glucagon response to a meal is reduced (4) in pregnancy. Finally, the MCR of insulin and the insulin disappearance rate from circulation are unchanged in pregnancy (5, 6). Hence, the diabetogenicity of pregnancy is not caused by inappropriate changes in the secretory function of the endocrine pancreas or an increased degradation of insulin. Impaired glucose tolerance associated with high levels of insulin in plasma points to a state of insulin insensitivity as the most likely explanation for the diabetogenicity of pregnancy. Pregnant women have a striking resemblance to obese subjects and obese maturity-onset diabetics as regards their insulin insensitivity. It has been
Materials and Methods Subjects Twelve healthy pregnant women (average age, 26.1 yr) were studied between the 34th and 36th weeks of pregnancy. None of them received diuretics or any other drug during pregnancy. Before the actual pregnancy, all were within 10% of their ideal body weight (10). The women were without glycosuria and the weight gain during pregnancy never exceeded 11 kg. The control group comprised 14 healthy, young, normal weight, nonpregnant female volunteers (average age, 22.3 yr). None of them had a family history of diabetes. Investigative procedure
Received February 26,1979. Address all correspondence and requests for reprints to: Dr. Claus Kiihl, Hvid0re Hospital, DK-2930, Klampenborg, Denmark. * This work was supported by grants from the Danish Medical Research Council; the Danish Hospital Foundation for Medical Research Region of Copenhagen, the Faroe Islands, and Greenland; Aarhus Universitets forskningsfond; Landsforeningen for Sukkersyges Fond; Novo's fond; and Nordisk Insulinfond.
All subjects ate a normal diet containing at least 300 g carbohydrate for at least 3 days before each sampling. Blood (120 ml) was drawn from an antecubital vein at 0800 h after an overnight fast and abstinence from smoking. The blood was transferred to tubes containing EDTA (dipotassium salt) for the analytical procedures. 810
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 11 May 2015. at 02:47 For personal use only. No other uses without permission. . All rights reserved.
DECREASED INSULIN BINDING IN PREGNANCY Laboratory analyses Plasma insulin concentration was measured with the RIA of Heding (11). Glucose concentration in plasma was measured using an o-toluidine method (12). The concentrations of acetoacetate and 3-hydroxybutyrate in plasma were measured separately using an enzymatic micromethod (13). Cell binding studies. Mononuclear leucocytes were isolated from the EDTA-blood by gradient contrifugation (14). The cells were washed twice and incubated in dublicates of 250 /x\ of Hepes buffer (100 mM, pH 7.8, at 15 C, with 1% human serum albumin) at a concentration of about 50 X 106/ml for 100 min at 15 C with [I25I]insulin at a concentration of 34 pM (0.2 ng/ ml). The specific activity of the tracer was about 150 juCi//ig. For competition studies, native insulin in increasing concentrations was added to the incubation medium. At the end of the incubation period, 250 ju.1 cell suspension were added to a 1.5-ml microtube containing 1.0 ml ice-cold buffer and 100 /xl silicone oil (relative density, 1.04). Cell-bound and free insulin were separated by centrifugation. The specific cell-binding fraction is defined as the total cell-binding fraction minus the nonspecific cell-binding fraction, i.e. the radioactivity which remained bound in the presence of an excess of native insulin (10 JUM). This fraction averaged 10% of the total binding. The monocytes were identified in cytocentrifuged smears stained with a-naphthyl acetate esterase, and the specific cell-binding fraction was adjusted to a standard concentration of monocytes of 1.0 X 107/ ml using the formula previously described (14). There was no statistically significant difference between the numbers of monocytes in the suspensions of mononuclear leucocytes incubated from normal women and in those from pregnant women (8.7 ± 2.8 vs. 10.5 ± 4.5 X 106/ml; P > 0.1).
TABLE 1. Overnight fasting values of plasma glucose, insulin, and ketone body concentrations in the groups studied
Plasma glucose (mmol/ liter) Plasma insulin (/xU/ml) Plasma ketone bodies (mmol/liter)
Statistical methods. Wilcoxon's two-sample test was employed for comparison of values in the two groups studied, while Spearman's coefficient of rank (r) was applied in correlation studies.
Results As expected (1), fasting plasma insulin was significantly (P < 0.01) higher in the pregnant women than in the nonpregnant controls. In contrast, fasting plasma glucose and ketone bodies did not differ significantly in the two groups (Table 1). Thus, the pregnant women appeared to be insulin resistant. Correspondingly, insulin binding to monocytes from pregnant women was significantly lower than that observed in the controls (Fig. 1). In the pregnant subjects, insulin binding was significantly reduced at each insulin concentration tested (0.034-5 nM,
Pregnant
Control
P
4.6 ± 0.24
4.8 ± 0.34
NS
18.0 ± 12.0 0.097 ± 0.046
8.0 ± 4.0 0.134 ± 0.095
