INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE RENIN-ANGIOTENSIN system is a critical hormonal regulator of fetal cardiovascular function and organ development. In adult animals, renin secretion is tightly controlled by renal nerves, ion transport at the macula densa, the renal baroreceptor, and ANG II concentrations. During fetal development, the coupling of each of these regulators to renin secretion may mature at different gestational ages. Even so, end product feedback regulation of renin secretion by ANG II is known to occur relatively early in gestation in sheep and other mammals. For example, Robillard et al. (18), Lumbers and Lewes (11), and Sheikh et al. (21) showed that acute reduction of ANG II with a converting enzyme inhibitor increased plasma renin activity, tissue renin concentration, and mRNA for renin. Also, Lumbers and Lewes (11) showed that ANG II treatment reduced these variables. Despite these demonstrations of the regulation of renin secretion in the fetus, the development of this feedback does not seem to be a simple linear progression from fetal life to adulthood. Renin secretion is elevated in late gestation compared with both earlier gestational age or postnatal time periods (13, 16), which suggests that some component of the negative feedback regulation of renin secretion is less sensitive in late gestation.

Previous studies have primarily focused on acute, in vivo manipulations of renin secretion from fetal kidneys and examined only the secretion of active renin (11, 16). The purpose of our studies was to determine the age-related consequences of chronic, in vivo manipulation of ANG II concentration on the fetal kidney and whether enduring changes in renin secretion were measurable in vitro. Second, our studies examined the feedback control of the age-related recruitment of renin-secreting juxtaglomerular cells. Finally, by examining the content and secretion of both prorenin and active renin, we examined the role of ANG II feedback on the developmental relationship between renin processing and secretion.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Surgical preparation and in vivo treatment. All experimental protocols were approved by the Wake Forest University Animal Care and Use Committee and were performed on mixed breed sheep with known pregnancy dates. Sixteen immature fetuses (101-103 days of gestation, term = 145 days of gestation) and 18 mature fetuses (130-133 days of gestation) were studied. Under general anesthesia induced with ketamine and maintained with halothane, uterotomy was performed with sterile technique on the pregnant ewes and the fetal hindlimbs were exposed. Bilateral femoral arterial and venous catheters were placed in the fetal hindlegs and advanced to the aorta and vena cava. The fetus was returned to the uterine cavity, and an amniotic catheter was placed in the cavity before the uterine incision was closed. The fetal catheters, along with the maternal femoral artery and vein catheters, were exteriorized through the skin in the ewe's lateral abdominal wall and secured in an attached back pouch. Animals were given prophylactic gentamicin, allowed to recover for 4-5 days, and then studied.

The animals at each gestational age were divided into three groups. Control immature and mature fetuses (n = 5 each) were infused intravenously with 5% dextrose at 2 ml/h for 72 h. Angiotensin converting enzyme inhibitor-treated fetuses (immature = 7 and mature = 6) were given enalaprilat as an intravenous bolus (0.5 mg/kg), followed by a constant infusion rate of 2 ml/h (0.25 mg · kg estimated fetal wt¹ · h¹) for 72 h. The third group (immature = 4 and mature = 7) was given a constant infusion of ANG II (50 ng · kg estimated fetal body wt¹ · min¹) at 2 ml/h for 72 h. At the end of the infusion treatment, the ewe was anesthetized and the fetus was delivered by uterotomy and was euthanized by an umbilical venous injection of pentobarbital sodium.

In vitro studies of renin secretion. Renal cortical slices (300 µm thick) from the three groups of fetuses were prepared and incubated using a modified method described by Rawashdeh et al. (15). Slices were placed in cold Robinson's medium (pH 7.4 and 295 mosmol/kgH₂O) that contained ascorbic acid (6 mM) as a protectant of isoproterenol degradation and 3-isobutyl,1-methylxanthine (1 mM) as a phosphodiesterase inhibitor. Randomly selected cortical slices from each animal were frozen (80°C) for later measurement of renin content. The remainder of the slices were randomly allocated two per well in a 12-well culture plate. Each well contained 2 ml of Robinson's medium. Plates were preincubated for 20 min in a shaking water bath at 37°C with continuous suffusion of 95% O₂-5% CO₂. After the preincubation period, cortical slices were transferred to new culture plates containing fresh Robinson's medium for two consecutive 20-min incubation periods. The first of these periods was used as a control and the cortical slices were incubated in the medium only. During the second period of incubation, duplicate wells of renal cortical slices were incubated with either Robinson's medium alone (basal secretion) or with isoproterenol (10⁸-10⁴ M) added to the medium. At the end of the second period, 1.25 ml of incubation medium was saved in a sample tube containing 25 µl of a mixture of 137 mM disodium EDTA, 161 mM dimercaprol, and 168 mM -hydroxyquinoline to inhibit ANG II converting enzyme and angiotensinase activity and stored in a freezer at 80°C until assayed for renin activity. The tissue slices were removed, gently blotted, and weighed.

Renin concentration in cortical tissue and medium was measured using a modified method that we have previously described (14, 15). Renal tissue was thawed, weighed, and added to 1 ml of 4°C saline solution. The tissue was homogenized with a Teflon pestle and centrifuged at 1,000 g for 10 min at 4°C. The supernatant was diluted with buffer containing 10 mM Na₂EDTA and 5 mM 8-hydroxyquinoline (pH 7.4). The final concentration of tissue ranged from 0.1 to 0.5 mg/ml depending on gestational age.

Renin concentration was assessed by incubating the samples with excess substrate obtained from sheep plasma collected 72 h after nephrectomy. In a portion of each sample, total renal renin concentration in the supernatant was measured after trypsin activation of prorenin. Briefly, 100 µl of trypsin was added to 300 µl of sample (final concentration 0.75 mg/ml) and incubated for 30 min at 4°C. The reaction was stopped by adding 100 µl of soybean trypsin inhibitor at the same final concentration and incubating for 15 min at room temperature. For the assay, 100 µl of sample for active renin or 100 µl of the trypsin-treated sample for total renin was added to 500 µl of renin substrate along with 25 µl of a mixture of inhibitors, which was supplied with the assay kit, and incubated for 2 h at 37°C at pH 7.4. The same methodology was used to generate ANG I from incubation medium samples; however, dilutions and changes in time of incubation were made depending on gestational age and treatments, i.e., in samples where renin was expected to be low, incubation times were increased and dilutions of the unknowns were less. The generation of ANG I was measured with a radioimmunoassay kit (Isotex Laboratories, Friendswood, TX) and expressed in nanograms ANG I per milligram tissue per hour with an intra-assay coefficient of variation of 7% and interassay coefficient of variation of 9%. Prorenin activity was determined by subtracting active renin from total renin activity.

The concentrations of prorenin and active renin from duplicate samples were averaged, and the data from each group were expressed as means ± SE. Homogeneity of variance between groups was accomplished by ln transformation. These data were analyzed by a three-way ANOVA with Crunch (Crunch Software, Oakland, CA), which examined the effects of gestational age (immature or mature), in vivo pretreatment (vehicle, enalaprilat, or ANG II), and dose of isoproterenol (10⁸-10⁴ M). Subsequently, a two-way ANOVA was used to determine the effect of enalaprilat or ANG II treatment on basal and isoproterenol-stimulated release of renin from kidney slices from immature and mature fetuses. Differences in tissue concentrations of prorenin and active renin were analyzed using a two-way ANOVA. If the differences were statistically significant between the groups (P < 0.05), comparisons among means were performed using Newman-Keuls test.

Immunohistochemical localization of renin. In additional experiments, kidneys from 12 immature (vehicle = 4, enalaprilat = 5, and ANG II = 3) and 16 mature (vehicle = 5, enalaprilat = 5, and ANG II = 6) fetal sheep were bisected longitudinally. Methods used were modified from our previous studies (17). Small, full-thickness slices of cortex were made at the level of the hilum and placed in fresh Bouin's fixative containing 20% sucrose as a cryoprotectant for 48 h at 4°C. After fixation, nonconsecutive 20-µm sections were made on a cryostat and placed in individual wells of a culture plate filled with 0.1 M PBS at pH 7.4. Slices were rinsed twice with PBS for 10 min each time with constant gentle agitation in a shaking water bath.

Sections were then incubated for 10 min in 3% hydrogen peroxide as a quencher, rinsed twice with 0.01 M PBS for 10 min, and then incubated with normal goat serum for 20 min at room temperature. Normal goat serum was used as a blocking antibody solution. The excess normal goat serum was removed carefully from each well containing the sections.

The sections were incubated with rabbit antimouse renin antibody (gift from Dr. Chris Deschepper) at a dilution of 1:32,000 for 48 h at 4°C. The ability of this antibody to cross-react with sheep renin was previously demonstrated in our laboratory. Subsequently, the primary antibody was removed and the sections were rinsed twice with PBS for 10 min each time. Sections were then incubated with biotinylated antirabbit antibody (Vectastin ABC kits, Vector Laboratory, Burlingame, CA) at a dilution of 1:1,000 for 45 min and rinsed twice for 10 min each time. Next, the sections were incubated in an avidin-biotin horseradish peroxidase mixture diluted 1:500 in PBS and rinsed twice in sodium acetate buffer. Finally, they were exposed to 0.033% 3',3'diaminobenzidine tetrahydrochloride (DAB) diluted in nickel sulfate buffer and 0.03% hydrogen peroxide for 8 min and rinsed twice to remove excess DAB. Renin-containing cells were stained purple on a blue background. Negative controls were assessed by omitting the primary antibody.

Sections were mounted on gelatin-coated slides and dried overnight. They were counterstained with cresyl violet and dehydrated in ethanol dilutions and xylene. The slides were then placed under a coverslip.

Light microscopy was used to determine the extent and localization of renin-containing cells. The total number and number of positively stained glomeruli in a microscopic field (2.5 mm²) were counted in a blind coded manner in three different sections, and an average value was calculated for each animal. The percent of positive glomeruli was calculated by dividing the number of positively stained glomeruli by the total number of glomeruli. The length of the renin staining along the afferent vasculature was scored depending on the location. A value of 0 was assigned to glomeruli with no visible staining for renin in the polar region, 1 indicated that staining for renin was confined to the polar portion of the glomerulus, a value of 2 meant the polar location plus spreading along the first half of the afferent arteriole, a value of 3 indicated all of the afferent arteriole stained positively, and a value of 4 represented spreading beyond the afferent arteriole to the interlobular arteriole.

The number of glomeruli per 2.5 mm² and the percent of positive glomeruli in each treatment group of young and mature fetuses were expressed as means ± SE. Three sections from each kidney were examined, and the values were averaged to give a single value for number of glomeruli for that kidney. The change in number of glomeruli and the percentage of positive glomeruli were analyzed by two-way ANOVA comparing gestational age and in vivo pretreatment. The vascular distribution scores were first ranked as a group and then analyzed by two-way ANOVA. If differences were statistically significant between the groups (P < 0.05), comparisons among means were performed using Newman-Keuls test.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The kidney tissue concentrations of active renin and prorenin increased with advancing gestational age (active renin: F = 6.5, P < 0.05; prorenin: F = 10.1, P < 0.05). Renal cortical slices from both the immature and mature groups of fetuses pretreated in vivo with vehicle contained a similar percentage of active renin (50%; Fig. 1). The overall statistical analysis of all data also showed that in vivo pretreatments with enalaprilat or ANG II had highly significant effects across ages on the basal and isoproterenol-stimulated secretion of active renin and prorenin (P < 0.001 for each). Thus there were age, pretreatment, and isoproterenol dose effects. These data are summarized in Tables 1 and 2.

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Effects of vehicle, enalaprilat (Enal), or ANG II (A-II) pretreatments in vivo on renal cortical prorenin and active renin concentrations. Bars represent means ± SE. * Significant difference from vehicle-treated fetuses within an age group; ** significant difference from immature fetuses within treatments; and + only prorenin is significantly different from vehicle treatment. All symbols represent P < 0.05.

Renal renin concentration. Kidneys from vehicle-treated fetuses contained equivalent proportions of active renin and prorenin at both gestational ages. However, when fetuses were pretreated with the converting enzyme inhibitor enalaprilat there were dramatic increases in tissue concentration of both active renin and prorenin (Fig. 1) in both age groups, without altering the proportion of active renin and prorenin. Conversely, ANG II pretreatment produced substantially different results on prorenin compared with active renin. At both ages, the kidney tissue concentration of prorenin was decreased compared with vehicle-pretreatment groups (F = 6.2, P < 0.05; Fig. 1). However, in sharp contrast, the kidney tissue concentrations of active renin from both groups of fetuses were not altered significantly so that the percentage of active renin in the kidney was increased (F = 13.4, P < 0.05). The mature fetuses had higher renal cortical concentrations of both prorenin (P < 0.05) and active renin (P < 0.05) compared with immature animals.

In vitro renin secretion. Gestational age significantly increased the basal and isoproterenol-stimulated secretion of renin in vehicle-pretreated fetuses. Kidney slices from mature fetuses secreted two- to threefold more active renin and prorenin under basal conditions compared with immature fetuses (Tables 1 and 2). Whereas renal tissue contained equivalent proportions of active renin and prorenin at both ages, basal renin secretion was ~75% active and this proportion was unchanged by isoproterenol stimulation. There was also a gestational effect on basal prorenin secretion from kidney cortical slices. It increased from 0.3 ± 0.01 to 1.0 ± 0.1 ng ANG I · mg tissue¹ · h¹ (P < 0.05). The -adrenoceptor agonist significantly increased secretion of active renin (P < 0.002) when all doses were considered, with the peak effect at 10⁶ M. Although there was a tendency for prorenin secretion to be increased by isoproterenol, the changes were not significant at either age in vehicle-treated animals (Tables 1 and 2, Fig. 2).

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Effect of in vivo pretreatments on isoproterenol (106 M)-stimulated change in secretion of prorenin and active renin from renal cortical slices in vitro. * Significant difference from vehicle-treated fetuses within an age group; ** significant difference from immature fetuses within treatments. All symbols represent P < 0.05.

Pretreatment of fetuses with the converting enzyme inhibitor enalaprilat resulted in highly significant increases (P < 0.001) in the basal secretion of active renin and prorenin at both ages. Isoproterenol significantly increased the secretion of both renin forms at both ages (P < 0.001) in a dose-related manner (Tables 1 and 2). The slope of the dose-response curve was steeper compared with the vehicle-treated animals, although peak effect still occurred at ~10⁶ M. Figure 2 shows that this concentration of isoproterenol increased active renin and prorenin secretion significantly more (P < 0.05) in slices from enalaprilat-treated animals than from vehicle-treated controls at both gestational ages. The percentage of active renin in the incubation media was lower from kidney slices from enalaprilat-treated immature fetuses than from vehicle-treated controls (P < 0.05, Table 1).

ANG II pretreatment only decreased basal secretion of active renin by kidney slices from immature fetuses (P < 0.05; Tables 1 and 2). Although ANG II pretreatment decreased renin content at both gestational ages, kidney slices from mature fetuses still secreted more active renin than did slices from immature fetuses (P < 0.05).

In vivo pretreatment of immature fetuses with ANG II abolished the net secretion of both prorenin and active renin in response to 10⁶ M isoproterenol (P < 0.01; Fig. 2). In contrast, in kidney slices from mature fetuses ANG II pretreatment only reduced the net secretion of prorenin in response to isoproterenol; net secretion of active renin was unchanged. Thus isoproterenol failed to stimulate the secretion of prorenin at either gestational age after ANG pretreatment. The dose-response curve to isoproterenol in slices from these immature fetuses was essentially flat for active renin and not different from basal secretion, whereas responses to isoproterenol in slices from mature animals pretreated with ANG II were not different from the vehicle-treated group. Because ANG pretreatment reduced the content and secretion of prorenin and not active renin in these kidneys, the percentage of active renin secreted into the media increased in both age groups (P < 0.002).

Intrarenal renin distribution. The intrarenal distribution of renin increased with development in vehicle-treated animals. In the immature fetuses immunoreactive staining for renin surrounded the distal afferent arterioles next to the glomeruli. The majority of renin containing glomeruli was located in the inner portion of the renal cortex consistent with normal centrifugal maturation. Gestational development increased the percent of renin immunoreactive glomeruli from 24.3 ± 3.0% in the immature fetus to 39.0 ± 2.1% in the mature fetus (P < 0.05, Fig. 3). Furthermore, in mature fetal sheep, the renin immunoreactive staining shifted from a juxtaglomerular localization to a more extensive distribution throughout the length of the afferent arteriole and a portion of interlobular arteries (P < 0.05 from control; Fig. 4). The positively staining glomeruli were dispersed throughout the inner and outer cortices. The number of glomeruli per 2.5 mm² (glomerular density) did not significantly change with gestational age, although there was a tendency to increase in mature fetuses (Fig. 5).

View larger version (31K): [in this window] [in a new window]   Fig. 3.   Percentage of juxtaglomerular apparati staining positively for renin. * Significant difference from vehicle-treated fetuses within an age group; ** significant difference from immature fetuses within treatments. All symbols represent P < 0.05.

View larger version (33K): [in this window] [in a new window]   Fig. 4.   Effect of in vivo pretreatments on intrarenal distribution of renin during fetal development. Nonparametric distribution values were ranked and expressed as means ± SE. * P < 0.05, significantly different from vehicle-treated fetuses; ** P < 0.05, significantly different from immature fetuses.

View larger version (37K): [in this window] [in a new window]   Fig. 5.   Number of glomeruli per 2.5 mm2 of cortex. * Significant difference from vehicle-treated fetuses within an age group; ** significant difference from immature fetuses within treatments. All symbols represent P < 0.05.

Enalaprilat treatment significantly changed the immunoreactive staining for renin along the vasculature in the mature fetal sheep (Fig. 4). It was distributed further along the afferent arterioles (P < 0.05 from control) and interlobular arteries (Fig. 6). Positively staining juxtaglomerular regions were located in the inner cortex and increasingly in the outer cortex. In the immature fetus the immunoreactive renin distribution was not significantly different (Fig. 4) from vehicle-treated animals; however, enalaprilat tended to increase renin distribution along the afferent arteriole from the previous periglomerular localization. Positively stained arterioles were mainly associated with inner cortical nephrons. Enalaprilat pretreatment increased the percentage of glomeruli that stained for renin in immature fetuses to 45.0 ± 2.0% (P < 0.05 compared with vehicle) and similarly increased it in mature fetuses to 65.6 ± 2.0% (P < 0.05, Fig. 3). Enalaprilat pretreatment significantly increased (P < 0.05) the cortical density of glomeruli in mature but not in immature fetuses (Fig. 5).

View larger version (95K): [in this window] [in a new window]   Fig. 6.   Immunostaining of renin in renal cortex of a mature fetal sheep treated in vivo with vehicle (A), enalaprilat (B), or ANG II (C; ×250).

In contrast, ANG II did not produce significant changes in renin distribution along the vasculature in either mature or immature fetuses (Figs. 4 and 6). Renin was found in the juxtaglomerular region and in the afferent arteriole with occasional distribution to the interlobular arteries as in vehicle-pretreated fetuses. Positively staining glomerular regions were similarly distributed in the cortex as in vehicle-pretreated fetuses. However, ANG II pretreatment did significantly decrease the fraction of positively staining glomerular regions in immature fetuses to 14.5 ± 3% (P < 0.05 compared with vehicle) and in mature fetuses to 32.4 ± 1.9% (P < 0.05, Fig. 3). ANG II pretreatment did not change the cortical density of glomeruli at either gestational age (Fig. 5).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The results of the studies in this paper suggest that ANG II feedback modulates not only the synthesis and content of renin but the sensitivity of the coupling between stimulus and secretion in the fetus. Thus, although the stimulation of renin secretion through -adrenoceptors does not develop until relatively late in the third trimester in sheep (13, 15, 16), we found that enalaprilat pretreatment not only increased the content and basal secretion of renin, but augmented isoproterenol-stimulated secretion even in immature animals. Interestingly, this was apparent with prorenin secretion as well, in that enalaprilat pretreatment permitted isoproterenol-stimulated prorenin secretion, whereas this was not seen in vehicle-pretreated animals. Opposite changes occurred with ANG II pretreatment in vivo. Consequently, our studies show the end product feedback of ANG II on active renin and prorenin secretion even in early development. This has been shown previously in our laboratory by Sheikh et al. (21), who reported that chronic in vivo enalaprilat treatment in fetal sheep increased renin mRNA and plasma and tissue concentrations of prorenin and active renin. Our current work extends these findings to show that alterations in ANG II adjust the recruitment and location of renin-secreting cells along the afferent vasculature and produce a change in the capacity of the renal cortex to secrete renin that persists when the kidney is studied in vitro in the absence of hemodynamic influences.

Earlier studies in developing rats (6, 7) have shown renin distribution is not confined to the glomerular pole but is widespread in the renal vasculature of the fetus close to term (0.9 gestation and above). In other species the pattern of distribution is somewhat different during early development (3, 8, 10, 12) but is similar later with renin present along the renal vascular tree in the perinatal period (8, 10, 17). These differences in the histochemical localization of renin may be related to the relative level of maturation at the time of birth. It may be that some of the changes seen postnatally in the rat are found prenatally in other species. In the present work, manipulation of ANG II concentrations in vivo significantly altered the distribution of renin-containing cells along the afferent vasculature in the fetus. This had been reported earlier in adult rats (9), but the interaction of gestational development with differing levels of ANG II had not been documented. Our histochemical studies clearly show the recruitment of more renin-containing cells with converting enzyme inhibition, but they cannot distinguish between tissue distribution of prorenin and active renin because the antibody recognizes both forms. In addition to increasing renin distribution, enalaprilat pretreatment significantly increased (P < 0.05) the cortical density of glomeruli in mature fetuses. This may be consistent with its reported action to decrease renal tubulogenesis as a consequence of preventing the mitogenic effect of ANG at a time when glomeruli have already been established. In immature fetuses, a change in glomerular density was not seen, perhaps because both glomerular and tubular maturation were retarded. Conversely, ANG II pretreatment did not change the cortical density of glomeruli at either age, which likely indicates that the basal concentrations of ANG were adequate for this effect.

A critical observation of our study is that the higher renal tissue concentrations of prorenin and active renin in mature fetuses are, at least in part, a consequence of reduced sensitivity to ANG II feedback. This is supported by the findings that ANG II pretreatment reduced the concentration and secretion of prorenin at both gestational ages, yet the secretion of active renin by renal slices from mature fetal kidneys was not changed (Fig. 2). This decrease in the sensitivity of ANG feedback on renin secretion is consistent with the increased tissue and plasma concentrations of renin found in late gestation (13, 14, 16, 23). Our findings in fetal sheep extend previous observations in neonatal rats (27) showing that ANG feedback is functional but reset at a different baseline. The reasons for the apparent reduced sensitivity of the mature fetal kidney to feedback inhibition of renin secretion by ANG II remain elusive. However, our studies do show that in late gestation, in vivo treatment with ANG II decreases prorenin content and secretion without an accompanying decrease in active renin. These findings are consistent with an inhibition of renin gene transcription and/or translation of the renin mRNA. This is supported by the studies of Schunkert et al. (22) in which a 3-day infusion of ANG II into adult rats suppressed renal renin concentration and renal renin mRNA and of Robillard et al. (19), who noted a decrease in renin mRNA in fetuses infused with ANG II. Furthermore, Stanley et al. (24) reported that chronic in vivo treatment with ANG II in mature fetal sheep reduced prorenin and active renin plasma concentrations, decreased renal prorenin content, and reduced renal renin mRNA. Our studies extend these reports by comparing the age-related effect of manipulating ANG II levels in vivo on renin secretion in vitro. All of the findings are consistent with an inhibition of the genetic control of prorenin production. Alternatively, or concomitantly, a direct inhibitory effect of ANG II on active renin secretion may increase the time prorenin remains in the intracellular granules, thereby increasing its processing to active renin and maintaining content and secretion.

Yet to be determined is the role of the ANG AT₁ receptor in this process. In adult animals and humans, the AT₁ receptor has been shown to be involved in the coupling between ANG II and the renin gene (9, 22). The fetal kidney contains AT₂ receptors expressed strongly in the macula densa during development with a transition to AT₁ receptor predominance at term (2, 19, 28). Because ANG II feedback regulation of renin synthesis was clearly demonstrable even in early gestation, the developmental role of AT receptor subtypes needs examination.

A developmental coupling between nephrogenesis and the regulatory transducers of renin synthesis and secretion may also be involved. Zhang and Morgan (29) concluded that in adult mice, the macula densa plays a critical role in regulating the synthesis and processing of prorenin to active renin and may less directly regulate renin secretion. Several potential mediators of the sensitivity of renin synthesis and secretion are known to change during fetal development. Among these are regulators of tubular sodium transport such as hydroxyeicosatetraenoic acids in the loop of Henle (1, 25), macula densa nitric oxide, which mediates renin secretion (4, 5), and oxytocin receptors in the macula densa, which decrease toward parturition (20, 26).

In summary, our studies demonstrate in fetal lambs that chronic in vivo alteration of ANG II exposure changes renin content and secretion in vitro. We found that the end product feedback of ANG II was present even in immature fetuses and that enalaprilat causes an age-related recruitment of renin-containing juxtaglomerular cells. The decreased feedback sensitivity of ANG treatment on active renin secretion from mature fetal kidneys may partially explain the elevated levels of plasma renin found in near-term mammals.

Perspectives