Previous studies have shown that intrauterine growth restriction (IUGR) can impair nephrogenesis, but uncertainties remain about the importance of the gestational timing of the insult and the effects on the renal renin-angiotensin system (RAS). We therefore hypothesized that induction of IUGR during late gestation alters the RAS, and this is associated with a decrease in nephron endowment. Our aims were to determine the effects of IUGR induced during the later stages of nephrogenesis on 1) nephron number; 2) mRNA expression of angiotensin AT1 and AT2 receptors, angiotensinogen, and renin genes in the kidney; and 3) the size of maculae densae. IUGR was induced in fetal sheep (n = 7) by umbilical-placental embolization from 110 to 130 days of the ∼147-day gestation; saline-infused fetuses served as controls (n = 7). Samples of cortex from the left kidney were frozen, and the right kidney was perfusion fixed. Total kidney volume, nephron number, renal corpuscle volume, total maculae densae volume, and the volume of macula densa per glomerulus were stereologically estimated. mRNA expression of AT1 and AT2 receptors, angiotensinogen, and renin in the renal cortex was determined. In IUGR fetuses at 130 days, body and kidney weights were significantly reduced and nephron number was reduced by 24%. There was no difference in renin, angiotensinogen, or AT1 and AT2 receptor mRNA expression levels in the IUGR kidneys compared with controls. We conclude that fetal growth restriction late in nephrogenesis can lead to a marked reduction in nephron endowment but does not affect renal corpuscle or macula densa size, or renal RAS gene expression.
- nephron endowment
- macula densa
it is now apparent that intrauterine growth restriction (IUGR), which affects up to 10% of pregnancies, can alter fetal organ development, leading to changes that increase the risk of cardiovascular and renal disease in adulthood (2, 18, 35, 37). A major factor that may contribute to the increased risk of adult renal disease after exposure to IUGR is a reduction in nephron endowment (19, 24). Indeed, recent experimental studies suggest that a congenital nephron deficit renders the kidney more vulnerable to the effects of secondary postnatal insults such as high circulating levels of advanced glycation end products (as observed in diabetes) and induction of acute glomerulonephritis (33, 45). In humans, the number of nephrons has been related to birth weight, suggesting that IUGR inhibits nephrogenesis (15, 20, 21). A reduced nephron endowment has also been reported in the kidneys in many experimental models of IUGR (3, 22, 25, 26, 42, 45). However, there are still uncertainties as to critical periods during which nephrogenesis can be impaired by IUGR.
In humans, ∼60% of nephrons are formed in the last trimester of pregnancy, with nephrogenesis being completed by 36 wk (∼0.9) of gestation; no new nephrons are formed after this time (15). So, although there may be catch-up in growth of the IUGR infant after birth, the nephrons cannot increase in number, but they may enlarge. The sheep is an excellent model for studying the effects of IUGR on the fetal kidney; like humans, the sheep has a long gestation (term, 147–150 days) with much of renal ontogeny occurring before birth (10). In sheep, IUGR can be induced by umbilical-placental embolization (UPE), which replicates key aspects of the placental insufficiency associated with IUGR in humans during the later stages of gestation (23). In a recent study using sheep, we found that UPE from 120 to 140 days of gestation led to a marked reduction in birth weight and kidney size but had no effect on nephron endowment (26). This is not surprising, since the fetal growth restriction occurred after the completion of nephrogenesis; in sheep, nephrogenesis is complete by ∼120 days (∼0.8) of gestation (10). However, in twins, in which IUGR occurs during at least the latter one-half of gestation, there was a reduction in nephron number (26). This earlier study has highlighted the importance of the timing of the gestational insult when considering the effects of perturbations in utero on renal ontogeny. Therefore, in the present study, we have induced UPE from 110 days (∼0.75) of gestation to determine whether IUGR coinciding with the later stages of nephrogenesis can affect nephron number. This is important, as IUGR in humans occurs predominantly during the third trimester.
The renal renin-angiotensin system (RAS) is considered to play a key role in nephrogenesis and in the development of renal vascular and tubular structures (8, 11, 14). Angiotensin II (ANG II) is the main effector hormone in the RAS cascade. In humans and sheep, ANG II acts via two main receptors: the ANG II type 1 (AT1) and type 2 (AT2) receptors. However, little is presently known about the effects of IUGR on the renal RAS, including the number of AT1 and AT2 receptors and the level of angiotensinogen, the precursor for renal angiotensin. In the ovine fetal kidney, the expression of the renal AT1 receptor is normally low during early gestation and increases during the last one-third of gestation (36). In contrast, the AT2 receptors are abundantly expressed in the metanephric mesenchyme and downregulated late in gestation, when nephrogenesis is complete; the AT2 receptors are then predominantly expressed in the macula densa (10, 40). The maculae densae consist of specialized epithelial cells that lie within the cortical thick ascending limb of the juxtaglomerular apparatus, between the afferent and efferent arterioles (4). Their main role is to monitor NaCl concentration in the distal tubule and to regulate the release of renin (32). Since the macula densa cells are intimately associated with renin release, they play a key role in the renal RAS. We recently observed that maculae densae appeared enlarged in sections from the kidneys of fetuses that were experimentally or spontaneously growth restricted (26).
We hypothesized that induction of IUGR during late gestation leads to alteration of the RAS, and this is associated with a decrease in nephron endowment. Hence, our aims were to determine the effects of IUGR induced during the later stages of nephrogenesis on 1) nephron number; 2) mRNA expression of the AT1 and AT2 receptors, angiotensinogen, and renin genes in the kidney; and 3) the size of maculae densae.
MATERIALS AND METHODS
Animals and surgery.
All surgical and experimental procedures were approved by the Monash University Animal Welfare Committee. Details are provided in Bubb et al. (7), and only a brief account will be given here. Fourteen time-mated Border Leicester × Merino ewes were anesthetized at 105–106 days of gestation and prepared for surgery (term is ∼147 days). The uterus was incised, and the hind quarters of the fetus were exposed to allow insertion of catheters into a femoral artery and vein and the amniotic sac. The arterial catheter was positioned such that its tip lay in the fetal descending aorta below the renal arteries and above the umbilical arteries (9). After 4–5 days of postsurgical recovery, IUGR was induced in seven fetuses (3 female, 4 male); another seven fetuses (4 female, 3 male) were allowed to grow normally and served as controls. IUGR was induced by UPE from 110 to 130 days of gestation by daily injections of microspheres into the fetal aortic catheter. The microspheres become embedded into the fetal tissue of the placenta where they impair the exchange of oxygen and nutrients to the fetus (9). The control group received daily saline (2–5 ml) via the aortic catheter.
At 115 and 125 days of gestation (5 and 15 days after commencing embolization, respectively) a 2-ml sample of fetal blood was collected via the femoral artery catheter for measurement of fetal plasma renin activity and ANG II levels using radioimmunoassay (ProSearch International, Malvern, Australia).
At necropsy (130 days of gestation), the fetal kidneys were excised and weighed. From the left kidney, pieces of cortex were snap frozen in liquid nitrogen and stored at −75°C. The right kidney was perfusion fixed at 80 mmHg with 4% formaldehyde (pH = 7.4) and then stored in 10% buffered Formalin.
Kidney sampling and tissue processing.
The right kidney was cut through the hilus in the longitudinal plane followed by a transverse section through the hilus, resulting in four quarters. Two diagonally opposite quarters were randomly selected and cut into slices 2-mm thick; every second slice was embedded in glycolmethacrylate resin (Technovit 7100 resin; Heraeus Kulzer, Hanau, Germany). These blocks were sectioned at 20 μm using a Leica RM 2165 microtome (Leica Microsystems, Nussloch, Germany) fitted with glass blades. We collected the 10th and 11th sections, starting from a random number, and stained them with periodic acid Schiff's reagent.
In the remaining slices, small cubes of cortex (1 per kidney slice; ∼8 cubes per kidney) were cut from a randomly chosen site of cortex and then embedded in epon-araldite. These blocks were sectioned at 1 μm using a Leica OmU3 Ultramicrotome (C. Reichert, Vienna, Austria) and stained with toluidine blue to examine the size of the maculae densae (see below).
Measurement of kidney volume, nephron number, and renal corpuscular volume.
In the glycolmethacrylate sections, total kidney volume (Vkid), nephron number, and renal corpuscle volume were stereologically estimated. The sampled kidney sections (every 10th section) from each block were used for the estimation of kidney volume using the Cavalieri principle (13).
Nephron number (Nglom,kid) was estimated in the pairs (10th and 11th sections) of intact kidney sections using an unbiased physical disector/fractionator technique (5, 26). The total number of glomeruli (and thereby nephrons) in the kidney was determined by the formula where 4 represents the inverse of the slice sampling fraction, 10 is the inverse of the section sampling fraction, Ps/Pf and 1/2fa represent the fraction of the total section area used to count glomeruli, and Q− is the total number of glomeruli counted from all selected pairs.
Mean renal corpuscle volume (Vcorp) was determined stereologically by dividing the volume density of renal corpuscles in the kidney (VVcorp, kid) by the numerical density of glomeruli in the kidney (NVglom, kid) (5, 26).
Estimation of the total volume of maculae densae and volume of macula densa per glomerulus.
The epon-araldite sections (1 per block; 8 per kidney) were used for the stereological determination of the total volume of maculae densae. The sections were sampled using a uniform step length of 250 μm in the horizontal and vertical direction. This allowed for uniform systematic sampling throughout the kidney sections, commencing at a random start point. An unbiased counting frame was superimposed over the sections at a magnification of ×1,250. In each section, grid points that fell on maculae densae (PMD) and on the renal corpuscle (Pcorp) were counted. Maculae densae were identified as closely packed columnar epithelial cells in the wall of the distal tubule, which is in direct contact with the glomerulus at the vascular pole. The total volume of maculae densae in the kidney (total VMD) was calculated using the formula where total VMD was determined from multiplying VMD/Vcorp by total volume of renal corpuscle (total Vcorp) (17).
The volume of macula densa per glomerulus (VMD) was calculated by division of the total volume of maculae densae in the kidney (total VMD) by the total number of glomeruli in the kidney (Nglom, kid) (17)
Gene expression studies: real-time PCR.
Total RNA was extracted from the cortex of the left kidneys (control n = 7, IUGR n = 6) using an RNeasy Midi Kit (Qiagen), and 1 μg of RNA was then reverse transcribed into cDNA. Negative reactions were prepared to check for genomic DNA contamination. Reverse transcription reactions were run in a PCR machine (Bio-Rad i-cycler) at 25°C for 10 min, 48°C for 30 min, and 95°C for 5 min and then held at 4°C. Relative mRNA expression levels of ANG II receptors AT1 and AT2, angiotensinogen, and renin were determined using a 25-μl TaqMan multiplex real-time PCR assay in the ABI PRISM 7700 Sequence Detection System (Perkin-Elmer Applied Biosystems, Foster City, CA). Table 1 shows the primer and probe sequences used in this study. The sequences used for the primer and probes for the AT1 and AT2 receptors, angiotensinogen, and renin were optimized previously (28).
A comparative cycle of threshold (Ct) method was used for mRNA quantitation of genes of interest as described previously (28), where 18S was used as a housekeeping gene (endogenous/internal control). Each sample was run and analyzed in triplicate. To calculate relative gene expression, the Ct value for 18S was subtracted from the Ct value of the gene of interest to give a ΔCt for each sample. The ΔCt value of a “calibrator” (in this case, the mean ΔCt value of the control group) was then subtracted from the ΔCt of each sample to give a ΔΔCt. This was put into the equation 2−ΔΔCt to give relative gene expression. The values obtained for each gene (AT1 and AT2 receptors, angiotensinogen, and renin) were then compared between the treatment groups. The intra-assay coefficients of variation (34) were calculated for all genes of interest: they were 5, 12, 8, and 11% for AT1, AT2, angiotensinogen, and renin, respectively.
Data are presented as means ± SE. Statistical analyses were performed using Graphpad Prism for Windows (v.3.00; Graphpad Software, San Diego, CA). For all data, gender was included in an initial ANOVA to identify gender differences. Because no significant gender differences were detected, data from males and females were pooled and effects of treatment (IUGR vs. control) compared using an unpaired two-tailed Student's t-test. If data were not normally distributed, a Mann-Whitney correction was applied. Within each group, linear regression analyses of fetal body weight vs. kidney weight, kidney volume, and nephron number were determined as well as kidney weight vs. nephron number. Statistical significance was accepted where P ≤ 0.05.
Fetal body weight, kidney weight, and kidney volume.
There was a 25% reduction in body weight (P < 0.05) and right kidney weight (P < 0.05) in IUGR fetuses compared with controls (Table 2). Kidney volume was also significantly reduced (by 31%) in the IUGR group (Table 2) compared with controls. In both the control and IUGR groups, linear regression analyses showed a strong linear correlation between fetal body weight and kidney weight (R2 = 0.765, P = 0.011 in controls; R2 = 0.676, P = 0.023 in the IUGR group; Fig. 1, A and B, respectively).
Plasma renin activity and ANG II levels.
The mean plasma renin activity and ANG II levels at 115 days of gestation and 125 days of gestation are shown in Table 3. There was a wide range in plasma renin activity and ANG II concentrations within both the control and IUGR fetuses at the gestational ages studied. There was no significant change within either the control or the IUGR groups in the levels of renin activity or ANG II from 115 days of gestation to 125 days of gestation; in some fetuses, there was a rise in renin activity and ANG II from 115 days of gestation to 125 days of gestation, whereas in others, the levels decreased. There was wide variability among individual measurements within groups, which is reflected in the absence of a significant difference (P = 0.07) between groups in renin activity and ANG II levels at 115 days of gestation. At 125 days of gestation, there was no difference in either renin activity or ANG II levels between groups.
The number of glomeruli (and thereby nephrons) was significantly reduced in IUGR fetuses (Fig. 2); the IUGR kidneys had 24% fewer nephrons compared with controls. In the control group, linear regression analysis demonstrated a strong linear relationship between nephron number and birth weight (R2 = 0.791, P = 0.007); interestingly, there was no such association in the IUGR group (R2 = 0.169, P = 0.359). In the control group, nephron number and kidney weight were strongly associated (R2 = 0.689, P = 0.021; Fig. 3A), whereas no such association was found in the IUGR group (R2 = 0.002, P = 0.929; Fig. 3B). The number of glomeruli per unit of kidney volume (numerical density) was not different between groups (Table 2).
Renal corpuscle size.
The mean volume of renal corpuscles was not significantly different between groups (Table 2). There was no correlation between nephron number and renal corpuscle volume in either the control group (R2 = 0.453, P = 0.098) or the IUGR group (R2 = 0.344, P = 0.220).
Maculae densae volume.
In general, the maculae densae in the IUGR kidneys when examined microscopically appeared to be longer than those in controls (Fig. 4, B and A, respectively). However, stereological analysis demonstrated that the total volume of maculae densae was not different between groups (Table 2). Likewise, there was no difference in the volume of the macula densa per glomerulus between groups (Table 2).
Relative expression of AT1 and AT2 receptors, angiotensinogen, and renin genes.
In the renal cortex, the mRNA levels for the AT1 and AT2 receptor genes were not different between groups (Fig. 5, A and B, respectively). There was also no difference in angiotensinogen or renin mRNA expression in the renal cortex between groups (Fig. 5, C and D, respectively).
In this study, IUGR induced during late gestation led to significant reductions in both fetal body weight and kidney weight, and this was accompanied by a substantial decrease in nephron endowment. Our findings show that a brief period of placental insufficiency, coinciding with the later stages of nephrogenesis, can markedly reduce renal growth and nephron endowment. However, we found no evidence that the renal RAS was altered by late gestational IUGR and found no difference in the total volume of the maculae densae or in the volume of the macula densa per nephron.
Nephrogenesis in the sheep closely resembles that in humans in that it commences early in gestation, when the ureteric bud first invades the metanephric mesenchyme, and is complete before birth (10). Nephrogenesis occurs as a result of branching of the ureteric bud and induction of nephron formation within the metanephric mesenchyme at the branch tips (39). The first branching of the ureteric bud has been observed at day 27 in the ovine fetal kidney (27), and nephrogenesis is completed at ∼120 days, after which no further nephron formation has been observed (10). The nephrogenic zone often remains visible up to ∼130 days of gestation in sheep as many of the recently formed nephrons undergo maturation. Thus, in the present study in which IUGR began at day 110 of gestation, there was a developmental window of only ∼10 days in which nephrogenesis could be affected by the IUGR. Importantly, our results demonstrate that nephron endowment can be reduced substantially by fetal growth restriction in a relatively short developmental window during late gestation. This is likely linked to the extensive level of branching of the ureteric bud in the kidney during the latter stages of nephrogenesis (39).
There is mounting evidence that a primary congenital nephron deficit can render the kidney susceptible to secondary postnatal insults (30, 33, 45). Furthermore, the majority of IUGR infants show catch-up growth within the first 2 yr of life (1), resulting in a disproportionately low number of nephrons in relation to adult body weight. Hence, our findings are clinically important, as our IUGR model mimics the growth restriction often observed in human pregnancies that occurs as a result of late gestational placental insufficiency. Our findings suggest that IUGR babies may be at risk of reduced nephron endowment at birth and subsequent renal disease, especially if there is catch-up in body growth. Interestingly, in a previous study in our laboratory, in which IUGR was induced in the same manner from 120 to 140 days of gestation, there was no difference in nephron endowment between the treatment groups; we concluded that this was because nephrogenesis was already complete in these fetuses (26). The results from both of these studies demonstrate that the timing of the insult leading to IUGR during fetal development is crucial to the effects on nephron endowment.
We have recently shown in the primate kidney that kidney size is directly correlated with nephron number during normal gestation (12). In accordance with this concept, in the present study, there was a strong linear relationship between nephron number and kidney weights and nephron number and kidney volumes in the control fetuses. Importantly, however, this regulation of nephron number was not evident in the kidneys of the IUGR fetuses. In these IUGR fetuses, a high kidney weight did not necessarily imply a greater nephron complement, and similarly a low kidney weight did not imply a reduced nephron endowment. Apparently the number of nephrons is not directly linked to kidney size in the IUGR fetuses, suggesting that other factors that have been altered in the intrauterine environment are affecting nephrogenesis. Potential factors that lead to the dysregulation of nephrogenesis in the IUGR fetus are the fetal hypoxia and hypoglycemia that are seen in association with IUGR (7). However, the mechanisms leading to the deregulation of nephron endowment within the kidneys of the IUGR fetuses cannot be determined from the present study, but it is an important area for future research.
Interestingly, there was no difference between groups in the mean corpuscular volume, although there were fewer nephrons in the IUGR group. Under normal circumstances, the postnatal kidney compensates for the loss of nephrons through an increase in the size of glomeruli (6, 29, 31), but this relationship is not always observed in the fetal or early postnatal kidney, implying that filtration is sufficient for body size at this age. It is expected that if the IUGR fetuses were born and allowed to grow to adulthood, especially if catch-up growth occurred, there would be induction of glomerular hypertrophy to maintain renal filtration surface area, which may in turn lead to adverse effects in the kidney as a result of chronic glomerular hyperfiltration (16).
Before the commencement of this study, we had made the observation in the kidneys of two other cohorts of growth-restricted ovine fetuses that the maculae densae appeared enlarged (26); preliminary analyses demonstrated an increase in the length of the macula densa per nephron. Since the maculae densae, and in particular the juxtaglomerular cells, play a key role in renin production in the kidney, in the present study we compared the total volume of the maculae densae and the volume of the macula densa per nephron in the IUGR and control kidneys. Using this stereological approach, we were able to avoid the difficulties of anisotropy associated with sampling of the anisotropic maculae densae. Interestingly, in the present study, as before, we observed that, microscopically, the maculae densae appeared longer in the IUGR kidney sections compared with controls, yet our stereological measurements demonstrated no difference in total volume of the maculae densae or in the volume of the macula densa per nephron.
Previous to this study, there have been a number of studies in other species on the effects of IUGR on the fetal renal RAS (38, 41, 43). In the case of the IUGR rat, there are reports of downregulation of the renal RAS at birth (41, 43), whereas in newborn spontaneously growth-restricted piglets, it has been recently reported that there is upregulation of the renal RAS (38). The differences in findings between species may relate to the relative developmental maturity of the kidneys at birth; in the rat, the kidneys are still undergoing nephrogenesis at birth, whereas in the piglet, it is complete. In this regard, our ovine model resembles the human and piglet models, with nephrogenesis being complete late in gestation (∼day 120 in the fetal sheep). However, unlike the findings in piglets (38), we found no difference in the mRNA expression of either the AT1 or the AT2 receptor or angiotensinogen in the IUGR kidneys at 130 days of gestation within the IUGR group compared with controls. Of particular relevance to the present study, the effects of IUGR induced by placental restriction in sheep on the intrarenal RAS have also been previously described (44). In that study, there was downregulation of renal angiotensinogen and renin expression in the IUGR fetus at ∼140 days of gestation; the authors therefore suggested that reduced activity of the intrarenal RAS may lead to impaired renal growth and development in the IUGR fetal sheep. In the present study, although there was obvious impairment of nephrogenesis as a result of late gestational IUGR, there was no clearly defined suppression of the intrarenal RAS. However, it must be noted that the mRNA expression of the components of the intrarenal RAS and plasma concentration of ANG II and renin activity varied considerably between animals within groups, which makes detection of subtle changes difficult.
We conclude that fetal growth restriction that coincides with the later stages of nephrogenesis can markedly reduce renal growth and nephron endowment. This may render the kidney increasingly vulnerable to renal insults later in postnatal life, especially if catch-up in body growth occurs, potentially contributing to the excess of renal disease in individuals born with low birth weight.
This study was supported by the National Health and Medical Research Council of Australia.
We acknowledge the technical assistance of Alex Satragno in fetal surgery, Sue Connell and Julie Hickey in tissue embedding and processing, Debbie Arena in real-time PCR, and Rebecca Douglas-Denton in the stereology.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society