Fetal pancreatic glucagon responses in glucose-intolerant nonhuman primate pregnancy

Fetal pancreatic glucagon responses in glucose-intolerant nonhuman primate pregnancy MICHAEL EPSTEIN, M.D. RONALD A. CHEZ, :'vf.D. GARY K. OAKES, M.D. DANIEL H. MINTZ, M.D.

Miami, Florida

Rhesus monlu'y pancreatic alpha-all jimction in streptozotocin-induced glucose-intolerant pregnancy is similar to that in normal primate pregnancy. Specifically, basal maternal and fetal plasma glucagon levels equate, and the fetal alpha cell does not respond to thf' glucagonogenic stimulus of either intraz•enous alanine or insulin-induced hypogZwemia. This contrasts with the accelerated maturation of the feta./ beta cell in glucose-intolerant pregnancy, and does not support the concept rif jimctional coupling of the pancreatic isltt by a rommon glucose-based proces~. Fetal plasma glucagon levels do increase after L-dopa injection to thcfetus. These data indicate that alpha all unresponsiveness is a function of the glucagon-releasing mechanism rather than inadequate hormonal synthesis. (AM.j. 0BSTET. GYNECOL. 127:268, 1977.)

hyperglycemia accelerated the time of appearance of glucose recognition site(s) on the primate fetal beta cell. The current experiments were undertaken to examine the possibility of acceleration of alpha-cell function in STZ glucose-intolerant pregnancies. The response of fetal plasma glucagon levels in the normal fetus when an appropriate secretagogue, L-dopa, was administered were also examined.

WE HAvE previously examined fetal pancreatic alpha-cell responsivity in normal monkey pregnancy and found that induced fetal hypoglycemia, hyperglycemia, and hyperalaninemia were not associated with changes in fetal plasma glucagon levels. 1 In other studies, we also demonstrated that near term normal monkey fetal pancreatic beta cells were unresponsive to acute hyperglycemia. 2 In contrast, the normal fetal beta cell did respond to insulinogenic stimuli that do not depend upon the presence of a fully developed glucoreceptor mechanism. 3 • 4 Further, glucose administered to fetuses from streptozotocin (STZ)-induced maternal glucose-intolerant pregnancies was associated with acute increments in fetal plasma insulin concentrations. 5 These contrasting observations established that abnormal maternal-fetal

Methods

Monkeys (Macaca mulatta) with accurately known I day) were obtained from the gestational periods primate colony of the National Institutes of Health. Glucose intolerance was induced by the use of streptozotocin (gift of Upjohn Company. Dr. Dulin, Lot No. 10518-GGS-37), 48 mg. per kilogram of body weight intravenously between 50 to 7.1) days' gestation. 5 111travenous ~;lucose tolerance tests were performed at 64 to 90 davs' gestation to determine the extent of the induced maternal glucose intolerance. The animals were housed individually. Water was freely available; food consisted of Purina monkey chow and daily fruit. Supplemental daily feedings were presented to the STZ animals to maintain weight gain during gestation. Fetal experiments were performed between 141 and 152 days' gestation; this is 86 to 93 per cent of term (term= 164 days::!:: 3 days). The operative experimen-

From the Pregnancy Research Branch. National Institute of Child Health and Human Development, National institutes of Health and the Department of Medicine, Div~~ion of Endocrinology and Metabolism, University of Miami School of Medicine. Supported in part ~V NIH Grant No. AM14106 arul gifts jrom the juvenile Diabetes Foundation of Florida. Rn-i'lved for publication Februlll) 2, 19i6. Rer•ised August 23, I9i6. Atctpted September 10, 19i6. Reprint requests: Ronald A. Chez, AI .D., NIH Clinical Center, Room !3N266, Bethesda, Maryland 20014.

268

Fetal pancreatic glucagon responses

Volume 127 Number 3

269

Table I. Fasting plasma concentrations in the Macaca mulatta Glucose (mg. '7c)

Glucagon (pg.lml.) No.

Mean

::t

S.E.

I

Range

Mean± S.E.

I

Range

Normal pregnane)•:

Mother Fetus

6 6

59± II 73 ±II

23-88 37-116

40 ± 6 32 ± 3

32-4R 27-35

8

71 ± 3 53± 5

45-126 23-70

60 ± 2 48 ± 2

53-75 40-53

STZ pregnancy:

Mother Fetus

8

tal design had been detailed previously . 1 Brief1y, sequential maternal and fetal blood samples were obtained at hysterotomy under nitrous oxide-oxygen-halothane anesthesia. Maternal blood was sampled from the inferior vena cava and fetal blood from an interplacental vessel via a silicone rubber t-tube cannula. Fetal and maternal blood pH, Po2 and Pcoz as well as vital signs were monitored throughout the experiment; they remained within normal limits. STZ animals. Sustained fetal and maternal hypoglycemia was obtained by infusion of increasing concentrations of insulin into maternal upper arm veins (N=5). Glucagon-free insulin (courtesy of M.A. Root, Lilly Research Laboratories) was diluted in saline and infused over 30 minutes at increasing concentrations, 0.3 Jk per minute for 10 minutes, 0.6 Jk per minute for 10 minutes. and 0.9 Jk per minute for lO minutes. Thirty minutes following the termination of the insulin infusion, 0.75 Gm. per kilogram body weight of dextrose was infused into the maternal circulation, and the study was continued for an additional 30 minutes. Blood samples were obtained simultaneously from the maternal and fetal circulations before and every 15 minutes for 90 minutes following the initiation of the insulin infusion. This protocol was selected from a previous study 3 • 5 in which we demonstrated that it was an efficient method of obtaining sequential hypoglycemia and hyperglycemia in both the maternal and fetal circulations. In three separate experiments, L-alanine (Sigma Chemical Co.) was dissolved in normal saline and infused (75 mg. per 1.5 mi. over 3 minutes) directly into the fetal circulation. Blood samples were obtained simultaneously from the maternal and fetal circulations prior to and every 15 minutes for 60 minutes following the alanine infusions. Normal animals. L-Dopa, 10 mg. per I ml. of normal saline, was injected into the circulation of three fetuses of normal pregnancies. Blood was obtained simultaneously from the maternal and fetal circulations, prior to and 5, 15, 30, and 45 minutes following the L-dopa injection.

Blood samples of 1 ml. each, collected in heparinized syringes, were immediately transferred to tubes containing ethylenediamine tetra-acetate (EDT A) and 500 KI.U. per milliliter ofTrasylol (FBA Pharmaceuticals). After cold centrifugation, a plasma aliquot was immediately analyzed for glucose content (Beckman Glucose analyzer), and the remainder was stored at -20° C. for the determination of glucagon, insulin. and growth hormone levels by immunoassay. The sensitivity and specificity of the assay system in simian plasma was previously recorded. I. >

Results Baseline. Eight animals with STZ-induced glucose intolerance were examined. The mean plasma glucose disappearance half-times during intravenous glucose tolerance tests performed 12 to 25 days following the streptozotocin administration were significantly prolonged compared to those of normal controls (18.2 0.6 minutes vs. 12.6 = 0.8 minutes, p < 0.00 I). Moreover, the other stigmas of STZ pregnancy. including increased fetal weights and placental weights, and amniotic fluid volumes exceeding the normal range were also observed in these pregnancies. Table I compares baseline maternal and fetal plasma glucagon and glucose concentrations in the late third trimester of the eight STZ pregnancies with six normal pregnancies examined concurrently. Although the mean STZ maternal and fetal plasma glucagon levels were higher and lower. respectively, than in plasma from normal pregnancies. the differences did not reach statistical significance. In normal gestation, fetal plasma glucagon was greater than the simultaneous maternal plasma glucagon concentrations in four of the six animals. The mean fetal/maternal ratio in normal pregnancy 0.93 (range 0.62 to 1.20) was no different from the mean fetal/maternal ratio of O.tn (range 0.52 to 1.13) in the STZ gestations. Both the mean maternal and fetal plasma glucose concentrations in the glucose intolerant pregnancies were significantly greater than the respective values from normal pregnancy (p < 0.001).

270

Epstein et al.

Februarv l, 1\177 Am . .J. Ohstet..Gvnecol.

10 mg I.V.

• FETUS

225'r----~~------------~ 150 " I

I

l~-1---,,, f '

-

.......

75 ~

1

............

CHO

~ 0

::r:-

(.) e

150

en-.... :s• a. e

75

cg :::& ...

0

o~~~--L--L~--~~--L-~~

0

10

20

30

40

50

l

MINUTES Fig. 1. Normal pregnancy. Near term fetal plasma glucagon responses to the intravascular injection of L-dopa (N 3, mean and range.)

Glucagon responses to L-dopa injections, normal. L-Dopa, 17.0 to 18.6 mg. per kilogram of body weight, was injected intravascularly to the fetus in three normal gestations between 148 to 153 days of term, and sequential blood samples were obtained for 45 minutes. Fig. 1 depicts the mean and the range of fetal plasma glucagon levels in association with the L-dopa injections. Fetal plasma glucagon concentrations were elevated at 5 minutes after the injection, reached mean peak levels of 132 per cent from baseline at 15 minutes, and declined to or below basal levels by 45 minutes. Maternal plasma glucagon levels did not change during the experimental period, nor did maternal and fetal plasma insulin or fetal plasma growth hormone levels. Induced hypoglycemia/hyperglycemia, STZ. The mean maternal and fetal plasma glucagon and glucose responses to induced maternal/fetal hypoglycemia are depicted in Fig. 2. A significant and sustained hypoglycemic stimulus was achieved in both maternal (nadir mean== 18 mg. per 100 mi.) and fetal (nadir mean 13 mg. per 100 mi.) blood pools. In response to maternal hypoglycemia, the mean maternal plasma glucagon rose to reach a value 280 per cent above baseline (range == 156 to 400 per cent). By contrast, the mean concentrations of plasma glucagon and growth hormone in the fetus were not significantly altered by the induced fetal hypoglycemia. Glucose (0.75 Gm. per kilogram of body weight) was injected intravenously to the mother at 60 minutes. This resulted in an 11- and 10-foid increase, respectiveiy, in maternai and fetai blood glucose levels within 15 minutes after injection. Plasma glucagon levels in the fetus remained unchanged, whereas the mean maternal glucagon con-

15

30

45

60

75

90

MINUTES

Fig. 2. Streptozotocin pregnancy. Near term fetal and maternal plasma glucose and glucagon responses to the intravenous infusion of insulin to the mother. The mother also received an intravenous glucose injection at I hour. (:--/ "' 5, mean and standard error.) • FETUS MOTHER

75 mg I.V.

o

~

N=3

lOOr---------------------------~

~

:Ci

~~

100

I

~,

Cl.

0

0

;o

==~

10

20

~

30

~

?

0

j

! 40

50

eo

MINUTES

Fig. 3. Streptozotocin pregnancy. Near term fetal and maternal plasma glucose and glucagon responses to the intravascular infusion of alanine to the fetus. (N 3, mean and standard error.)

Volume 127 Number 3

centrations continued to decline from their 60 minute value to levels at 90 minutes that were significantly lower than the initial baseline determinations (87 ± 15 pg. per milliliter vs. 50 ± 11 pg. per milliliter p < 0.001\. During these experiments, fetal plasma growth hormone levels remained stable or did not change in a consistent manner. Specifically, there was no relationship between the plasma growth hormone values and the presence of fetal hyperglycemia or hypoglycemia. Alanine infusion to the STZ fetus. L-Aianine. 75 mg .. was infused into three fetuses o\'er 3 minutes and sequential hlood samples were obtained for I hour as depicted in Fig. 3. There were no significant changes in fetal and maternal plasma levels of glucagon and glucose. or fetal levels of growth hormone. Comment

In STZ pregnancy, the primate fetal pancreatic beta cell undergoes accelerated maturation in its response to an intravenous glycemic stimulus. 5 The present study was directed at determining if the fetal pancreatic alpha cell was similarly affected by maternal glucose intolerance. In normal monkey pregnancy. induced fetal hypoglycemia. hyperglycemia, and hyperalaninemia were not associated with changes in fetal plasma glucagon levels.' The experiments depicted in Fig. 2 demonstrate that acutely induced fetal hypoglycemia also is not accompanied bv fetal hyperglucagonemia in STZ pregnancy. This is in contrast to the change in glucagon levels found in STZ maternal plasma. Similarly, acutely induced fetal hyperglycemia failed to suppress the basal len·ls of fetal plasma glucagon in STZ pregnancy. Again, this is the same unresponsiveness that was found in the normal fetus. 1 Thus, these experiments provide no support for the notion that the fetal pancreatic alpha cell in glucose-intolerant gestation shares in the accelerated maturation of the metabolic pron:sst:s necessary for pancreatic glucose recognition in the fetal beta n~ll. The functional unresponsiveness of the STZ fe-

REFERENCES !. Chez, R. A., Mintz, D. H., Epstein, M. F., eta!.: Glucagon metabolism in nonhuman primate pregnancy, AM. J. 0BSTET. GYNECOL 120: 690, 1974. 2. Mintz. D. H., Chez, R. A., and Horger, E. 0., III.: Fetal insulin and growth hormone metabolism in the subhuman primate, J. Clin. Invest. 48: 176, 1969. 3. Chez, R. A., Mintz, D. H., Horger. E. 0., Ill, eta!.: Factors affecting the response to insulin in the normal subhuman pregnant primate,]. Clin. Invest. 49: 1517, !970.

Fetal pancreatic glucagon responses

271

tal pancreatic alpha cell also applied to ::l!lother glucagonogenic stimulus, alanine, a substrate which has been demonstrated previously to elicit a rise in glucagon levels in the normal monkey neonate but not the normal monkey fetus." That an immutable constraint exists for fetal glucagon serretion was excluded in the L-dopa experiments. This secretagogue administered directly to the normal fetus was associated with similar increments in fetal plasma glucagon similar to those that occur in nonpregnant adult monkeys similarly challengecl. 6 We have previously suggested that the chronic presence of fetal hyperglycemia accelerates the maturation of glucose recognition site(s) on the fetal pancreatic beta cell. It is now recognized that the pancreatic beta and alpha cells are both structurally 7 and functionally coupled, imparting to the pancreatic islets as a whole a coordinated system for the opposing actions of glucose on insulin and glucagon release. The hypothesis has been advanced that functional coupling of the alphaand beta-cell responses to glucose may be accomplished by a common glucose recognition process either through molecular events during the metabolism of glucose/· 9 or through a glucoreceptor located on the plasma membrane of both alpha and beta cells. 10 If this hypothesis is correct, our experiences \\ ith the animal model detailed here and in previous \\ork, 2 provide evidence for the lack of coupling of the fe· tal pancreatic alpha and beta cells in STZ-induced glucose-intolerant pregnancies. ·whether this ontogenic disparity implies that gluwreceptor mt"chanisms are not primarily required for the induction of alpha-cell responsivity will require continued study. The data from this animal model of human pregnancy continue to provide unique insights int.o fetal carbohydrate metabolism in both normal and pathologic pregnancy. Interpretation and extrapolation of the data to human subjects requires the recognition that the disease diabetes mellitus and the druginduced glucose-intolerant state secondary to streptozotocin are not synonymous pathologic states.

4. Chez. R. A., Mintz, D. H., and Hutchinson, D. L: Effect of theophylline on glucagon and glucose-mediated plasma insulin responses in subhuman primate fetus and neonate, Metabolism 20: 805, 197 I. 5. Mintz, D. H .. Chez. R. A., and Hutchinson, D. L: Subhuman primate pregnancy complicated bv strep· tozotocin-induced diabetes mellitus. ]. Clin. Im·est 51: 837, 1972. 6. George, D. T., and Rayfield, E. J.: L-dopa induced plasma glucagon release,]. Clin. Endocrinol. 39: 618, 1974.

272

Epstein et al. Am

7. Orci, L.. Unger, R. H., and Renold, E. A.: Structural coupling between pancreatic islet cells, Experimentia 29: 1015, 197:l. 8. Ashcroft, S. ].. Hedescov, C. V., and Randle, P. J.: Glucose metabolism in mouse pancreatic islets, Biochem. ]. 118: 143, 1970. 9. Edwards, J. C., and Taylor, K. W.: Fatty acids and the

J.

Februarv I, 1\)7/ Obstet. 'Gvnerol.

release of glucagon from isolated Guinea pig islets of Langerhans incubated in Yitro, Biochem. Biophvs. Acta 215: 310. 1970. I 0. Matschinsky, F. 1\1., Pagliara. A. S., Ho\'er, B. A .. et al.: Differential effects of alpha-and beca-D-glucose on insulin and glucagon secretion from th{· isolated perfused rat pancreas, Diabetes 24: 369, 1975.

The Changing Seasons of Life: The Physician's Role The Continuing Medical Education Institute, "The Changing Seasons of Life: The Physician's Role," scheduled for the S.S. Rotterdam, April 23-30, I 977, is sponsored by: Sea and Shore Seminars, Washington, D. C.; The Charles R. Drew Postgraduate Medical School, The Office of Continuing Education; and The University of Cincinnati, College of Medicine, The Office of Continuing Medical Education. Medical problems of interest to both specialists and generalists will be presented in a continuum, prenatal through geriatric years. This course is acceptable for 25 hours in Category I for the Physician's Recognition Award, American Medical Association. Additional professional areas for which accreditation will be available will be sent on request. Registration: $175.00 (after February 23, 1977, $200.00). For further information, contact: Sea and Shore Seminars, 2100 19th St. N. W., No. 601, Washington, D. C. 20009, Att: N. Chamberlain.