Glucose intolerance in pregnancy: consequences for offspring kidney development and adult renal health

The prevalence of glucose intolerance in pregnancy is increasing worldwide. Both pre-existing Type 2 diabetes and gestational diabetes have known health implications for offspring yet few studies have assessed offspring renal development and function. In rodents, offspring exposed to persistent hyperglycaemia have a nephron deficit. However, this model does not reflect the typical clinical condition and is associated with fetal growth restriction that may confound the effect on nephron endowment. In humans, estimation of total nephron number is only feasible in autopsy tissue. Studies assessing fetal kidney volume, as a proxy of renal sufficiency, in women with glucose intolerance in pregnancy are sparse and report equivocal findings. The studies presented in this thesis investigated offspring kidney development in mouse models of both pre-gestational Type 2 diabetes and gestational diabetes. It was hypothesised that offspring of pregnancies complicated by glucose intolerance would be macrosomic, yet exhibit deficits in nephron endowment. It was further hypothesised that offspring would be overweight and glucose intolerant in adulthood with perturbed renal function and morphology. Additionally, this thesis aimed to further our understanding of how gestational diabetes affects human fetal kidney size. It was hypothesised that fetuses of women with gestational diabetes would have smaller kidney volumes. Three major studies were undertaken. In the first experimental study, C57BL6 mice were fed a high fat diet (HFD, 21% fat) prior to and throughout pregnancy to model obesity-associated pre-gestational Type 2 diabetes. Compared with mice fed a normal fat diet (NFD, 6% fat), mice fed a HFD were overweight and glucose intolerant prior to pregnancy, but unfortunately not in late gestation. Offspring of HFD dams were macrosomic at embryonic day (E) 18.5 and had a tendency to be heavier at postnatal day (PN) 21. Using unbiased stereology, the kidneys of HFD offspring were found to contain 20-25% more nephrons than kidneys of NFD offspring. Nephron number was increased in HFD offspring regardless of maternal weight and glucose profiles prior to pregnancy. Offspring of HFD dams displayed mild fasting hyperglycaemia at 9 months of age, but surprisingly had normal body weight, glucose tolerance and body composition. Renal function and morphology, measured by transcutaneous clearance of FITC-sinistrin and stereology respectively, were also normal at 9 months of age. In the second experimental study, C57BLKS/J leptin receptor deficient mice (Leprdb/+) were characterised to assess their suitability as a model of gestational diabetes. Leptin is a hormone produced primarily by adipocytes that acts on receptors in the hypothalamus to regulate energy balance by decreasing food intake and increasing energy expenditure. Compared with wildtype (+/+) mice, leptin resistant Leprdb/+ mice were overweight and glucose intolerant prior to pregnancy, with glucose profiles exacerbated in late gestation. Leprdb/+ mice were therefore a model of pre-gestational Type 2 diabetes rather than gestational diabetes as anticipated. Offspring of Leprdb/+ dams were of normal body weight at E18.5 and PN21 and exhibited a 15% nephron deficit which was evident before birth. The opposing effects on nephron endowment in offspring of HFD and Leprdb/+ dams are likely reflective of differences in maternal phenotypes during the period of kidney development. The increased nephron endowment in HFD offspring was not attributable to maternal obesity or glucose intolerance and is possibly be due to constituents of the HFD promoting nephrogenesis. Conversely, the deficit in nephron endowment in Leprdb/+ offspring was best predicted by maternal glucose intolerance in late gestation. Additional studies are required to identify the molecular mechanisms that lead to increased and decreased nephron endowment in these two models. The third study compared fetal biometry and fetal kidney size in a prospective cohort study of women with and without gestational diabetes. Fetal biometry and kidney volume (calculated using the approximation of a prolate ellipsoid) were measured by obstetric ultrasound at 32-34 weeks gestation in women identified with (n=36) and without (n=42) gestational diabetes by routine testing at 24-28 weeks. In this preliminary study, estimated fetal weight and birth weight were lower in women with gestational diabetes. Amniotic fluid index and fetal cerebral vascular resistance were also lower, although values were within the normal clinical range. Despite these outcomes, fetal kidney volume in offspring of women with gestational diabetes was similar to those of normal control pregnancies. As maternal diabetes is typically associated with neonatal macrosomia, it is unclear why lower fetal weight was observed in the human gestational diabetic cohort. However, a possible explanation is low-level placental dysfunction. The preliminary findings from the prospective cohort study suggest that hyperglycaemia does not affect fetal kidney volume. Analysis of fetal kidney volume in a larger population incorporating maternal ethnicity and degree of glucose intolerance, as well as in women with longer periods of hyperglycaemia (such as in women with early gestational diabetes and undiagnosed pre-gestational Type 2 diabetes) is required