Previously, we and others have determined that restricted nutrition in both humans and animal studies reduces the number of offspring nephrons.31–36
These findings suggest that maternal FR impacts nephrogenesis during fetal development, which may lead to the higher incidence of hypertension seen in the maternal FR rat model.35,37
Our study evaluated transcriptional and growth factor expression at e20, a time during which ureteric bud branching and mesenchymal to epithelial transformation are ensuing. Maternal FR inhibition of UB branching was evidenced by a reduction in GDNF gene expression. The importance of GDNF signal transduction to renal development is demonstrated in knock-out mice, as well as in mice lacking its receptor, c-Ret, or its co-receptor, GDNFR-α1; these animals have no kidneys as a result of a failure of UB outgrowth, while mice heterozygote for GDNF have small kidneys with 30% fewer nephrons.38
This, together with the ability of ectopically expressed or applied GDNF to induce the formation of supernumerary UBs from the Wolffian Duct in explants of e10–11 urogenital regions39
or in vivo40
confirms the critical role of GDNF in promoting UB outgrowth.
FGF7, a TGFβ family member, is strongly expressed in the developing rat kidney from e17 until birth,41
and also is important in branching morphogenesis.42
The RNA and protein expression of FGF7 were reduced by maternal FR and also may contribute to a reduction in ureteric bud branching and nephron number.
E20 kidney WT1 mRNA and protein are upregulated significantly while Pax2 mRNA and protein are down-regulated, suggesting that the Pax2 cascade is suppressed by maternal FR. WT1 inhibits transcription of Pax2,43;44
and may be, in part, responsible for its downregulation. Others have reported a decrease in Pax2 mRNA expression by 1.4 fold in the rat growth restricted fetus,45
and some Pax2 alleles in humans are associated with renal hypoplasia in adults,46
which support our findings. WT1 inhibits and Pax2 upregulates transcription of WNT4,47
and the observed dysregulation of these factors is associated with reduction in WNT4 RNA and protein, in this model.. The Wnt superfamily, are secreted glycoproteins that have important regulatory functions during vertebrate embryonic development.48
WNT4, an autocrine epithelializing signal, is expressed in several compartments in the developing nephron and is necessary for nephron formation.48;49
Wnt signaling is considered sufficient to trigger tubule development experimentally in the separated, uninduced kidney mesenchyme.48
WNT4 binds to the frizzled family of transmembrane receptors, activates the canonical WNT signaling pathway50,51
and activates mTOR, an important nutrient sensing pathway.52
The importance of this transcription factor is evidenced by WNT4 null mice, which fail to epithelialize the mesenchyme.53
Therefore, the observed significant decrease in WNT4 mRNA and protein expression in the offspring of food restricted dams may play a key role resulting in the reduced number of nephrons.
BMP4 is localized to the metanephric mesenchyme and functions to inhibit branching morphogenesis possibly acting through a GDNF reduction mechanism at a specific point in kidney nephrogenesis to contribute to lowered nephron number.54
BMP4 also prevents apoptosis and promotes growth of the mesenchyme depending on the specific localization.55,56
Mice heterozygote for BMP4 with reduced protein levels do not show abnormal ureteric bud branching.55
In contrast, upregulation of BMP4 mRNA has been identified in offspring of glucocorticoid treated spiny mouse dams in association with reduced nephron number.57
RNA and protein levels were significantly downregulated in the FR model, and the significance of this reduction in BMP4 is uncertain; it may result in increased apoptosis and inhibit mesenchymal growth although further studies in this area are needed.
We identified several factors in maternal FR offspring with dysregulated mRNA levels for which associated protein expression was not significantly different. The lack of change in protein expression or discrepant mRNA and protein level findings for these factors may be due to compensatory mechanisms present at e20 versus earlier points in fetal kidney development. It is possible that protein translation, post-translational modification, and degradation, may influence the level of a protein present within the kidney. Additionally, gene expression and protein concentrations may vary within the different compartments of the developing kidney and during the different stages of development. Changes in nuclear proteins may not be fully reflected in analysis of whole tissue homogenates. Further studies localizing these factors during renal development may help elucidate the significance of these changes. Finally, a confounding factor in this study is the lack of stratification of the e20 kidneys by sex. There is evidence that male offspring are more affected by maternal FR than females.36
Separate assessment of embryonic kidneys by gender and with larger cohorts may uncover additional significant aberrations in growth and signaling factor mRNA levels and proteins during renal development.
In conclusion, in this study we have identified maternal FR-induced dysregulation of GDNF, WT1, Pax2, WNT4, BMP4 and FGF7, factors impacting ureteric bud branching and mesenchymal to epithelial transformation in the developing fetal kidney at embryonic day 20. Our previous work has identified dysregulated TGF-β receptor 1 in the kidney earlier in nephrogenesis at embryonic day 16.58
These data suggest that dysregulation of critical renal genes at different stages of development is a key mechanism of the nephropenia induced by maternal FR. Further studies elucidating this process will provide additional insights into our understanding of reduced nephron number associated with fetal growth restriction. This is necessary to develop appropriately timed interventions to correct this maladaptive programming.
We would like to thank Guang Han, Bindu Cherian, Linda Day, and Stacy Behare for their technical assistance. This study was supported in part by grants from the March of Dimes Birth Defects Foundation #6-FY04-72 (MGR) and NIH 5P20MD000545 (TRM).