Neural tube defects are among the most common of the malformations associated with diabetic embryopathy. To study the molecular mechanisms by which neural tube defects occur during diabetic pregnancy, we have developed a new experimental system using pregnant diabetic mice. In this system, the rate of neural tube defects is about three times higher in embryos of diabetic mice than in embryos of nondiabetic mice. Most of the defects affected presumptive midbrain and hindbrain structures and included open defects (i.e., exencephaly) and gross maldevelopment. By semiquantitative reverse transcription–polymerase chain reaction and in situ hybridization, we found that expression of Pax-3, a gene required for neural tube closure in the area of the midbrain and hindbrain, is significantly reduced in the embryos of diabetic mice. The same regions of the neural tube where Pax-3 had been underexpressed were found subsequently to contain high concentrations of cells undergoing apoptosis. Reduced expression of Pax-3 appears to be responsible for this apoptosis because apoptotic cells were also found at sites of neural tube defects in embryos carrying null mutation of the Pax-3 gene. Finally, mouse strains that carry null mutations in Pax-3 develop neural tube defects that resemble the malformations that occur in embryos of diabetic mice. These results suggest that Pax-3 is an important developmental control gene, expression of which is disturbed in embryos of diabetic mice, and that as a consequence, apoptosis of the neural tube occurs. This pathway may be responsible for many of the neural tube defects resulting from diabetic pregnancy.

Pax-3 is a transcription factor that is expressed in the neural tube, neural crest, and dermomyotome. We previously showed that apoptosis is associated with neural tube defects (NTDs) in Pax-3-deficient Splotch (Sp/Sp) embryos. Here we show that p53 deficiency, caused by germ-line mutation or by pifithrin-α, an inhibitor of p53-dependent apoptosis, rescues not only apoptosis, but also NTDs, in Sp/Sp embryos. Pax-3 deficiency had no effect on p53 mRNA, but increased p53 protein levels. These results suggest that Pax-3 regulates neural tube closure by inhibiting p53-dependent apoptosis, rather than by inducing neural tube-specific gene expression.

Congenital malformations, including neural tube defects (NTDs), are significantly increased in the offspring of diabetic mothers. We previously reported that in the embryos of a mouse model of diabetic pregnancy, NTDs are associated with reduced expression of the gene Pax-3, which encodes a transcription factor that regulates neural tube development, and that reduced expression of Pax-3 leads to neuroepithelial apoptosis. In this study, we used three approaches to test whether glucose alone could be responsible for these adverse effects of diabetes on embryonic development. First, primary culture of embryo tissue in medium containing 15 mmol/l glucose inhibited Pax-3 expression compared with culture in medium containing 5 mmol/l glucose. Second, inducing hyperglycemia in pregnant mice by subcutaneous glucose administration significantly inhibited Pax-3 expression (P < 0.05), as demonstrated by quantitative reverse transcription-polymerase chain reaction assay of Pax-3 mRNA, and also increased neural tube apoptosis (P < 0.05). NTDs were significantly increased in glucose-injected pregnancies when blood glucose levels were >250 mg/dl (P < 0.002) but not in moderately hyperglycemic pregnancies (150-250 mg/dl, P = 0.37). Third, phlorizin administration to pregnant diabetic mice reduced blood glucose levels and the rate of NTDs. As seen with glucose-injected pregnancies, the rate of NTDs in phlorizin-treated diabetic pregnancies was related to the severity of hyperglycemia, since NTDs were significantly increased in severely hyperglycemic (>250 mg/dl) diabetic pregnancies (P < 0.001) but not in moderately hyperglycemic pregnancies (150-250 mg/dl, P = 0.35). These two findings, that elevated glucose alone can cause the changes in Pax-3 expression observed during diabetic pregnancy and that the NTD rate rises with significant increases in blood glucose levels, suggest that congenital malformations associated with diabetic pregnancy are caused by disruption of regulatory gene expression in the embryo in response to elevated glucose.

To evaluate whether there were differences in acquisition of research grant support between male and female faculty at eight Harvard Medical School-affiliated institutions. Data were obtained from the participating institutions on all research grant applications submitted by full-time faculty from 2001 through 2003. Data were analyzed by gender and faculty rank of applicant, source of support (federal or nonfederal), funding outcome, amount of funding requested, and amount of funding awarded. Data on 6319 grant applications submitted by 2480 faculty applicants were analyzed. Women represented 29% of investigators and submitted 26% of all grant requests. There were significant gender differences in the mean number of submissions per applicant (women 2.3, men 2.7), success rate (women 41%, men 45%), number of years requested (women 3.1, men 3.4), median annual amount requested (women $115,325, men $150,000), mean number of years awarded (women 2.9, men 3.2), and median annual amount awarded (women $98,094, men $125,000). After controlling for academic rank, grant success rates were not significantly different between women and men, although submission rates by women were significantly lower at the lowest faculty rank. Although there was no difference in the proportion of money awarded to money requested, women were awarded significantly less money than men at the ranks of instructor and associate professor. More men than women applied to the National Institutes of Health, which awarded higher dollar amounts than other funding sources. Gender disparity in grant funding is largely explained by gender disparities in academic rank. Controlling for rank, women and men were equally successful in acquiring grants. However, gender differences in grant application behavior at lower academic ranks also contribute to gender disparity in grant funding for medical science.