Norbert Freinkel

Banting Lecture 1980: of Pregnancy and Progeny

A profile characteristic of fuel economy in the mother during normal pregnancy has been delineated. The evidence indicates that pregnancy changes the metabolism of every class of foodstuff. The mechanisms by which the conceptus may be implicated are reviewed. The gestational interactions create a pattern of “accelerated starvation” whenever food is withheld, especially in late pregnancy, and they tend to “facilitate anabolism” when food is ingested. The consequent heightened metabolic oscillations during the shuttlings from fed to fasted state provide a basis for more aggressive therapy with exogenous insulin when endogenous insulin is lacking in pregnancy. It is emphasized that developing fetal structures may be exquisitely attuned to fine alterations in maternal fuel economy and that pregnancy complicated by diabetes may merely exaggerate these normal dependencies since maternal insulin affects all maternal fuels. The manifest changes in the offspring of mothers with. even the mildest limitations in insulin reserve, i.e., gestational diabetes, attest to the sensitivity of the relationships. It is suggested that concepts of teratogensis should be expanded to include alterations occurring subsequent to organogenesis during the differentiation and proliferation of fetal cells. Such changes could cause long-range effects upon behavioral, anthropometric, and metabolic functions. It is hypothesized that all of these could constitute expressions of fuel-mediated teratogenesis and that the potentialities should be incorporated into any evaluation of the outcome of pregnancy in gestations attended by disturbances in maternal fuel metabolism.

Simultaneous Estimation of Rates of Thyroxine Degradation and Thyroid Hormone Synthesis

Implicit in many studies of thyroidal function is the assumption that the rates of synthesis and degradation of thyroid hormone are equal (1-4). Yet this assumption has not been critically examined. Constancy of concentrations of serum precipitable iodine (SPI), particularly under conditions of environmental stress, need not indicate an equilibrium of this type, for interposed between the site of hormonal synthesis and the peripheral circulation is a large pool of stored hormone (1, 4-6). This pool is capable of maintaining, at least for a time, constancy of SPI, despite marked discrepancies between rates of hormonal production and degradation (7). A method has been devised for simultaneous estimation of the rate of degradation of thyroxine within the body and the rate of iodination of organic moieties within the thyroid gland.4 The equivalence of these two functions has been assessed in patients with diverse thyroidal states. In addition, the method has been applied to an evaluation of several factors which influence the degradation of thyroxine.

Method for Large-Scale Isolation of Pancreatic Islets by Tissue Culture of Fetal Rat Pancreas

Detailed studies of the maturation of stimulus-secretion coupling of the pancreatic B-cell requires a supply of isolated fetal islets, which has so far been difficult to obtain. To overcome this problem we have maintained minced and mildly collagenase-digested fetal rat pancreatic glands (21.5 days gestational age) in tissue culture to enable degeneration of the acinar part, leaving the endocrine cells in an isolated and surviving state. Indeed, after 1 wk in culture there was a complete separation between acinar and endocrine cells with the appearance of numerous discrete islets and the disappearance or dedifferentiation of the exocrine cells. Isolated islets were either free floating or attached on top of a monolayer of fibroblast-like cells. Their number after 1 wk in culture was estimated as about 90 per explanted fetal pancreas and a total yield of about 5000 isolated islets was easily achieved. Both light arid electron microscopic examinations showed an excellent structural preservation with a marked predominance of well-granulated B-cells. Numerous islets of the same weight as that measured in cultured islets of adult rats were regularly found after 1 wk in culture. The insulin concentration of the cultured fetal islets was related to the glucose concentration of the growth medium. A similar relationship was found with respect to the insulin release in response to glucose. Thus, fetal islets cultured for 8 days in growth media containing 11.1 or 22.2 mM glucose showed a marked and significant insulin response to glucose in batch-type incubations at the end of the culture period. By contrast, the glucose stimulation of insulin release was insignificant in islets cultured at 5.5 or 2.8 mM glucose. When the culture period was confined to 1 day, there were no effects of glucose on the insulin release irrespective of the glucose concentration of the growth medium. It is concluded that the present technique for tissue culture of fetal rat pancreas makes it possible to isolate substantial amounts of fetal islets predominantly composed of B-cells. The transition in vitro from a poor glucose sensitivity to an adult-type insulin response indicates that the technique can be used in further detailed studies of the molecular mechanisms involved in the growth and development of the pancreatic B-cell.

Carbohydrate metabolism in pregnancy

Diurnal profiles have been constructed for glucose, free fatty acids (FFA), triglycerides, cholesterol, and ten neutral amino acids in subjects with normal carbohydrate metabolism during late pregnancy and in age- and weight-matched nongravid women. Samples of blood were secured during a 24 hour period while the subjects were receiving a liquid formula diet (containing 2,110 kcal with 275 gm carbohydrate and 75 gm protein) in three equal feedings at 0800, 1300, and 1800 hours. Postprandial excursions for most nutrients, as well as plasma concentrations after overnight fast and before each meal, were significantly different in the pregnant subjects. The studies indicate that criteria of normalcy based on observations in nongravid women cannot be invoked assess fuel homeostasis in late pregnancy, and that separate criteria are necessary to evaluate nutrient regulation at this time.

Frequency of diabetes mellitus in mothers of probands with gestational diabetes: Possible maternal influence on the predisposition to gestational diabetes

Interviews for genetic histories were conducted prior to delivery in 166 pregnant diabetic probands and 83 control gravidas with normal carbohydrate metabolism throughout gestation. A significant association was observed between parental diabetic phenotypes and type of diabetes in the probands (chi 2(12)) = 32.413; p less than 0.001). In particular a higher than expected number of mothers with diabetes was encountered in 91 probands with gestational diabetes mellitus. The findings are examined in relationship to the hypothesis that vulnerability to gestational diabetes may be increased by exposure to an abnormal environment during intrauterine development.

Gestational Diabetes Mellitus: Heterogeneity of Maternal Age, Weight, Insulin Secretion, HLA Antigens, and Islet Cell Antibodies and the Impact of Maternal Metabolism on Pancreatic B-Cell and Somatic Development in the Offspring

We have examined gravida with gestational diabetes mellitus (GDM), as defined by the National Diabetes Data Group (Diabetes 1979; 28:1039), for phenotypic and genotypic heterogeneity. Fasting plasma glucose (FPG) at diagnosis was used for further stratification of GDM according to putative metabolic severity into class A, (FPG < 105 mg/dl [N = 129]), class A2 (FPG 105–129 mg/dl [N = 47]), and class B1 (FPG ≥ 130 mg/dl [N = 23]). All GDM classes tended to be older and heavier than consecutive gravida with documented normal glucose tolerance (controls, N = 148). Subdivision into “lean” and “obese” indicated that plasma immunoreactive insulin (IRI) was greater after overnight fast in the obese of all groups except B1. However, absolute increases in IRI above fasting levels in response to glucose during OGTT were significantly enhanced by obesity only in class A2 gravida. Adjustment for the effects of age and weight by covariate analysis indicated that the IRI response to glycemic stimulation is usually attenuated in all forms of GDM. Mean values for increases in IRI above fasting values during the first 15 min and IRI increments relative to the increases in plasma glucose throughout the 180-min OGTT were below control values in all GDM groups and progressively so, i.e.,A1 < A2 < B1. The absolute insulinopenia was not invariable; a small number of gravida from all GDM groups displayed well-preserved IRI responses to oral glucose. Genotypic evaluation of the GDM population disclosed an increased occurrence of “markers” known to be associated with type I diabetes mellitus. HLA antigens DR3 and DR4 were more frequent in all GDM groups, and the incidence of cytoplasmic islet cell antibodies was enhanced significantly in class A2, and even more so in class B1. Thus, GDM appears to be a heterogeneous entity with substantial phenotypic and genotypic diversity in the mothers. Offspring from some class A1 and diet-treated class A2 gravida were examined to assess whether minimal abnormalities in maternal metabolism suffice to impact on intrauterine development independent of maternal diversity. Amniotic fluid insulin at 36 ± 0.1 wk of gestation and cord plasma C-peptide at birth were increased in offspring of mothers with class A1 GDM, thus indicating that even the mildest forms of GDM can cause accelerated maturation of fetal islet function. Birth weight and symmetry index in the newborn from class A1 arid diet-treated class A2 gravida were significantly increased above control values, even after adjustment for maternal age and weight, thus documenting for the first time that GDM per se can influence the anthropometric characteristics of the neonate. The findings underscore that GDM constitutes an independent risk factor with particular implications for islet and somatic development during fetal life. These unequivocal effects of maternal metabolism on cell development in the fetus may provide the most compelling reason for aggressive approaches to GDM, especially if prospective as well as retrospective studies continue to support their postulated association with increased obesity and diabetes in later life (i.e., “fuel-mediated teratogenesis”; Diabetes 1980; 29:1023).

The Honeybee Syndrome — Implications of the Teratogenicity of Mannose in Rat-Embryo Culture

Abstract Lethal effects of D-mannose in the honeybee have been recognized for more than half a century. We observed another toxic effect of D-mannose during culture of rat embryos from the early head-fold stage to the 26–to-29-somite stage (Days 9Vz through 111/2 of gestation). The addition to culture mediums of 1.5 mg of D-mannose per milliliter caused growth retardation and faulty neural-tube closure in approximately two thirds of the embryos. Mannose effects occurred during the first 24 hours of culture and were attended by modest inhibition of the glycolysis that constitutes the principal energy pathway at this stage of development. Adding more glucose to preserve glycolytic flux or increasing atmospheric oxygen to promote oxidative metabolism offset the mannose teratogenesis. Our findings highlight the metabolic vulnerabilities that exist during early organogenesis, before oxidative flexibility is established. They may serve as a model to explain the teratogenicity of many other seemingly unrelated a...

THYROXINE-BINDING BY SERA OF PREGNANT WOMEN, NEW-BORN INFANTS, AND WOMEN WITH SPONTANEOUS ABORTION

Among the many physiological alterations which may occur during pregnancy are thyromegaly (1), augmented thyroidal avidity for iodine (2, 3), and increase in the concentration of circulating thyroid hormone (4, 5). This triad of anatomical and functional alterations is ordinarily considered to be diagnostic of thyrotoxicosis. Paradoxically, however, these findings are neither accompanied by symptomatic stigmata of hyperthyroidism, nor are they associated, during the first half of pregnancy, with increase in the basal metabolic rate. Although the basal metabolic rate increases during the latter half of pregnancy, this change has been ascribed by some to fetal needs, and not to alterations in maternal energy requirements (6). The origins and consequences of this unique functional dissociation are unknown. Recently there has been considerable interest in the physiological role and physicochemical properties of the specific thyroxine-binding protein or proteins of plasma (TBP). Although the physiological role of TBP has not been elucidated, demonstration of this alpha globulin moiety in 1952 provided another parameter of thyroidal economy which is susceptible to measurement (7-11). Even earlier it had been suggested that the increased SPI of pregnancy might result from an alteration in the manner in which thyroxine in plasma is protein-bound (12, 13). Therefore, it appeared possible that study of the thyroxine-

Effects of gestational diabetes on diurnal profiles of plasma glucose, lipids, and individual amino acids.

To assess the effects of gestational diabetes mellitus (GDM) on intermediary metabolism in late pregnancy, circulating levels of glucose, FFA, triglycerides, cholesterol, and individual amino acids were monitored for 24 h while subjects received a liquid formula diet (containing 2110 cal and 275 g carbohydrate) in three equal feedings at 0800, 1300, and 1800. Attempts were made to distinguish between varying degrees of severity of gestational diabetes by subdividing the population into those with fasting plasma glucose within the normal range for pregnancy, i.e., below 105 mg/dl (GDM < 105), and those with fasting plasma glucose of 105 mg/dl or greater (GDM ≥ 105). Both groups were compared with pregnant women with normal carbohydrate metabolism (NM). The diurnal profiles indicated that premeal, postprandial averages, and integrated 24-h values for plasma glucose were consistently higher in GDM ≥ 105 than in GDM < 105; both GDM groups uniformly exceeded the values in NM. Plasma FFA tended to be higher in all GDM, with maximal increments occurring in the early hours before breakfast and assuming greatest significance in GDM ≥ 105. The elevations in circulating cholesterol that occur in pregnancy were not significantly different at any time point in NM, GDM < 105, and GDM ≥ 105. However, the increases in plasma triglycerides were greater in GDM < 105 and GDM ≥ 105 than in NM and most marked in GDM ≥ 105. Diurnal profiles for a number of individual amino acids (phenylalanine, tyrosine, alanine, serine, proline) were not affected by gestational diabetes. However, certain other amino acids, particularly the branched chain (leucine, isoleucine, valine), tended to be elevated in subjects with GDM and to the most significant extent in GDM ≥ 105. Although the GDM subjects tended to be heavier than the NM, the progressively more pronounced metabolic abnormalities in GDM ≥ 105 than in GDM < 105 would suggest that relative insulinopenia rather than obesity contributed to the differences. Our findings indicate that gestational diabetes is attended by disturbances of varying degrees in all major classes of insulin-dependent foodstuffs and must be viewed as a disorder of multiple fuels.

Gestational Diabetes Mellitus: Correlations Between the Phenotypic and Genotypic Characteristics of the Mother and Abnormal Glucose Tolerance During the First Year Postpartum

We evaluated glucose tolerance during the first year postpartum in 113 women with gestational diabetes mellitus (GDM) diagnosed according to the criteria of the First International Workshop-Conference on GDM and the National Diabetes Data Group. The high incidence of abnormal postpartum glucose tolerance (38% “diabetes mellitus” plus 19% “impaired glucose tolerance”) was correlated with certain of the heterogeneous characteristics of the population at the time of antepartum diagnosis. Virtually all women with antepartum fasting plasma glucose (FPG) >130 mg/dl (GDM class BO remained abnormal postpartum (21/22 [95%]), which suggests that this group may include women with preexisting glucose intolerance unrecognized before pregnancy. In the remainder, those with FPG > 105–129 mg/dl (GDM class A2) were more likely to be abnormal postpartum than those with FPG <105 mg/dl (GDM class A,). Within the A, and A2 groups, increasing maternal age, relative insulinopenia, and hyperglycemia at 2 h during antepartum OGTT were also associated with a greater likelihood of abnormal glucose tolerance postpartum. The presence of HLA-DR3 and/or -DR4 antigens was not predictive of the status of glucose tolerance during the first year postpartum, although the increased frequency of cytoplasmic islet cell antibodies in A2 and B1 subjects was associated with a high incidence of abnormal postpartum glucoregulation. The high incidence of abnormal postpartum glucose tolerance in all GDM classes makes a compelling case for careful, early, and continuing follow-up of all women with a diagnosis of GDM. The risks of abnormal glucose tolerance in the first year postpartum appear to be correlated with some of the heterogeneous characteristics of women with GDM. Differences in such genotypic and phenotypic features as we have analyzed may account for differences in the incidence of postpartum abnormality in different populations. They must be considered in all studies attempting to elucidate the pathophysiology of GDM or to evaluate therapeutic strategies designed to alter the prognosis for glucoregulation under nongravid conditions.