INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

THE RENIN-ANGIOTENSIN SYSTEM (RAS) functionally matures during fetal development to assume a significant role in physiological homeostasis. Activators of the RAS, such as the intrarenal baroreceptor, the macula densa mechanism, and the renal sympathetic nerves have been shown to be important regulators of renin secretion in the near-term fetus, newborn, and adult sheep (3, 12, 14, 18, 19, 21). Whether there exists a modulating interaction between the renal sympathetic nerves and the intrarenal mechanisms that ultimately affect renin gene expression and renin secretion in the fetus is less well understood.

During the transition from fetal to newborn life, stimulation of the renal sympathetic nerves is thought to evoke a rapid rise of plasma renin activity (PRA²). Surgical denervation in utero results in suppression of the rise in renin mRNA associated with delivery and decreased PRA in the neonatal lamb (14, 21). This ability of the renal nerves to modulate the expression of renin mRNA at birth suggests a regulatory role of the renal sympathetic nerves on renin gene expression during development. Furthermore, these findings suggest that the ability of the RAS to respond to challenge may be attenuated by the absence of renal nerves in the fetus.

Studies exploring the interaction of the renal nerves in modifying the response of the RAS to various stimuli have produced unclear results. El-Dahr et al. (2) have shown that renal denervation in neonatal rats blunts the renin mRNA response to chronic ureteral obstruction, inferring intrarenal RAS response impairment. In contrast, the findings of Wagner et al. (23) and Holmer et al. (5) in adult rats, show renin mRNA increases of similar levels in both unilaterally denervated and contralaterally innervated kidney in response to ANG II receptor blockade or hemorrhage, suggesting that in response to hypotensive challenges, the intrarenal RAS is not functionally impaired by renal denervation. This interaction has not been explored during fetal life. To further investigate the developmental aspects of renal sympathetic innervation on renal renin gene expression, we examined the affect of unilateral denervation on renin gene expression under basal conditions and in response to chronically reduced renal perfusion pressure (RPP) in the near-term ovine fetus. Specifically, the present study was designed to 1) determine if unilateral renal denervation alters renin gene expression in either the unstimulated state or after a chronic reduction in RPP, 2) determine if renal denervation reduces the number of glomeruli staining positive for renin under basal conditions or reduces the response after a chronic reduction in renal perfusion pressure, and 3) explore the effects of denervation on renal active and prorenin concentrations in the basal state and after 24 h of reduced RPP.

	MATERIAL AND METHODS

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

Animal Preparation

On the day of surgery, the ewes were sedated with Ketamine intramuscularly (20 mg/kg) and orally intubated and mechanically ventilated using 1-2% halothane in 100% oxygen. These anesthetics also provided anesthesia to the fetus. A midline incision was made in the ewe's abdomen, exposing the gravid uterus. A uterine incision was made over the fetal hindlimbs, allowing exposure to the fetal lower extremities and adjacent lower abdomen. Polyvinyl catheters (1.0 mm ID, 1.7 mm OD) were advanced into the fetal femoral arteries bilaterally to measure descending aortic arterial blood pressure. The catheters were then passed through the uterine incision, and the hysterotomy was closed. Next, the uterus over the fetal lower thoracic spine was opened, and a longitudinal skin incision was made overlying the fetal spine. Further left flank dissection exposed the fetal left kidney and adjacent descending abdominal aorta. Careful dissection along the left renal artery, vein, and ureter disclosed the renal nerve. With a microscopic technique, the nervous tissue was striped away from the vessels mechanically. Application of 10% phenol in absolute alcohol was then performed twice. Caution was taken to avoid disruption of the sympathetic tissue on the abdominal aorta. Proximal to both the left and right renal arteries, a 6-mm inflatable Silastic occluder was placed around the aorta for the occluder group, whereas similar dissection or placement of a noninflatable occluder around the aorta was preformed in the control group. The inflation line was passed through a separate fetal skin incision, and both incisions were closed. An amniotic fluid pressure catheter was secured to the fetal back by a single suture, and all lines were passed through the hysterotomy incision. The uterus was closed. The fetal catheters were then brought through a separate maternal left flank incision to allow access in the recovery and study periods. The maternal abdomen was closed. Finally, a small incision in the ewe's left groin allowed placement of both femoral arterial and venous lines (1.3 mm ID, 2.3 mm OD), which were tunneled subcutaneously to exit the maternal flank incision with the fetal lines. All lines were kept in a clean pouch within elastic netting that surrounded the ewe's abdomen. The ewes were examined daily throughout this protocol. For the first 3 days, 1 g ampicillin and 80 mg gentamycin, both intravenous, were administered to the ewes. All lines were flushed with saline and filled with heparin 1,000 U/ml solution every 24 h. A recovery period of 4-5 days in each group was completed before instituting further study.

Experimental Design

After the experiment was completed, the ewes were given 10 ml Ketamine (100 mg/ml iv), and the fetuses were delivered by cesarean section. The ewe was immediately killed on delivery of the fetus by rapid KCl intravenous injection. The fetuses were killed with an overdose of pentobarbital sodium intravenously. The fetal kidneys were immediately harvested, decapsulated, and weighed, and the cortex was dissected. Cortex samples were immediately placed in cryoprotective vials, frozen in liquid nitrogen, and subsequently stored at 80°C until extraction of mRNA and determination of tissue renin content. Cortex samples were also placed in Bouin's solution for later fixation and immunocytochemical determination of renin staining glomeruli.

Blood Chemistries and Renal Renin Determinations

To determine renal tissue total and active renin concentrations, 200 mg of previously (80°C) stored renal cortical tissue was submerged into 3 ml of cold saline. This solution was then homogenized (Tri-R Stir-R model K43) on ice for ~1 min using a Teflon pestle. The homogenate was centrifuged at 3,000 g for 10 min, and the supernatant was collected and diluted to 0.25 mg/ml with saline containing 5.2 mM 2,3-dimercapto-1-propanol, 0.59 mM 8-hydroxy-quinolone, and 10 mM disodium EDTA. At the time of assay, this solution was further diluted to a concentration of 0.125 mg/ml with saline. Tissue total and active renin concentrations were determined by incubating aliquots of this mixture, at 37°C with nephrectomized sheep plasma, with and without previous activation with trypsin. Concentrations of ANG I generated were then measured using an RIA developed in this lab. Iodinated ANG I (catalog JJR1087) and rabbit anti-A1 serum (catalog JU1087) were purchased from Advance ChemTech (Louisville, KY). The antiserum cross-reacts 100% with human ANG I and 0% with human ANG II, ANG III, endothelin-1, neurotensin, substance P, [Arg⁸]vasopressin, and -atrial natriuretic peptide-(1-28). Both tracers and antiserum were diluted with assay buffer (0.25% BSA, 3.0 mM EDTA, 0.15 mM 8-hydroxyquinoline, 0.02% neomycin sulfate, and 0.1% Triton X-100 in 0.05% phosphate buffer, pH 7.4) to a final concentration of 10,000 cpm per tube of tracer and 1:35,000 antiserum dilution. After addition of 0.05 ml of sample or standard, the final assay volume was 0.55 ml/tube. ANG I (Sigma) concentrations ranging from 0.078 to 5.0 ng/ml were used for standard curve preparation. Samples and standards were incubated in 12 × 75-mm polystyrene tubes overnight at 4°C, and the complex was precipitated by incubating for 1 h with anti-rabbit gammaglobulin (Antibodies, Davis, CA) at 1:40 dilution in 0.05 M phosphate buffer containing 0.25% BSA, 5% polyethylene glycol (Carbowax PEG 8000; Fisher) and normal rabbit serum at 1:800 dilution, pH 7.5. After centrifugation, the supernatant was aspirated and the pellet was counted in a gammacounter. ANG I concentrations (ng/ml) of the samples were read from a standard curve generated by Graphpad Prism using the four-parameter logistic equation. Samples were measured in duplicate within the same assay (EC₅₀ = 0.99 ng/ml). Results are reported in nanongrams ANG I per milligrams tissue per hour.

Determination of renal renin mRNA. For a detailed description of the procedure used to extract and measure renin mRNA reference is made to our previous paper (18).

Determination of renal catecholamines. Norepinephrine concentrations were determined in renal tissue specimens obtained at the time of kidney harvesting. Samples of renal cortex (100-200 mg) were homogenized in 300 µl of chilled 0.1 M PCA with dihydrobenzylamine (DHBA) at 125 pg/50 µl concentration as internal standard by using an ultrasonic homogenizer. The homogenates were centrifuged at 16,000 rpm for 15 min at 4°C, and the catecholamines in the supernatant were adsorbed onto alumina. After elution with DHBA, the catecholamines were separated by HPLC on a C₁₈ column using a mobile phase of 0.1 M sodium phosphate, 600 mg/1.1 octanesulfonic acid, 0.1 mM EDTA, and 8% methanol with a pH of 7.3. Quantitation was accomplished with electrochemical detection (EG & G 400, Princeton, NJ). The sensitivity of the assay was 1 pg/ml with variance <2%.

Analysis of Glomeruli

Tissue preparation. A slice (~1.0 × 0.5 × 0.5 cm) of kidney cortex was fixed (8 h) in Bouin's solution, dehydrated, and embedded in paraffin. Five-micrometer, serial, sagittal sections were obtained throughout the tissue; every fifth section was collected onto a microscope slide. Each slide contained five sections that represented ~100 µm of tissue and was the estimated thickness of a typical glomerulus. In this manner, each tissue block yielded ~20 microscope slides and every fourth slide was used for immunocytochemistry.

Immunocytochemistry for renin-containing cells. All slides from a single kidney cortex were processed and immunostained simultaneously. Sections were rehydrated and then incubated overnight with renin antiserum (1:1,000) in Tris-NaCl buffer containing 1.0% BSA. After this step, the slides were washed and incubated 2 h with secondary antiserum (1:250, goat anti-rabbit IgG-horseradish peroxidase conjugates). Reaction product was visualized with diaminobenzidine-H₂O₂; sections were counterstained with hematoxylin and eosin and mounted. The antiserum made against mouse renin was a generous gift from Dr. C. Deschepper.

Tissue sampling and analysis. A single photomicrograph of a selected region within a paraffin section was obtained by using a ×10 objective. All glomeruli showing both parietal and visceral layers were included in the survey and classified as either immunopositive (i.e., staining of juxtaglomerular cells) or immunonegative. Data were expressed as a percentage of immunopositive glomeruli in the total population, which was typically 200-300 glomeruli.

Data Analysis

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

Mean Femoral Arterial Blood Pressure Change Over Time

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Change in mean arterial pressure (MAP) in fetuses during continuous suprarenal aortic occlusion and time-matched control fetuses from 0 to 24 h. *P < 0.05 between groups.

Renal Renin mRNA Levels

View larger version (11K): [in this window] [in a new window]   Fig. 2.   Renal renin mRNA level from occluder fetuses and control fetuses comparing denervated (Denerv) and innervated (Innerv) kidney. No significant effect of denervation is noted.

Plasma Active Renin Over Time

1

1

1

1

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Plasma levels of active renin in fetuses during continuous suprarenal aortic occlusion and in control fetuses from 0 to 24 h. A1, ANG I. *P < 0.05 between groups.

Plasma Prorenin Over Time

1

1

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Plasma levels of prorenin in fetuses during continuous suprarenal aortic occlusion and in control fetuses, from 0 to 24 h. *P < 0.05 between groups.

Renal Active Renin and Prorenin Concentration

1

1

1

1

Comparison of the renal prorenin concentrations in the denervated and innervated renal cortex samples of the control group revealed no difference (43.3 ± 26.6 vs. 46.1 ± 21.2 ng ANG I · mg¹ · h¹). Overall, the renal prorenin concentration response to decreased aortic perfusion pressure was unaffected by renal denervation (54.9 ± 18.2 vs. 47.7 ± 16.9 ng ANG I · mg¹ · h¹, P = 0.78).

Glomerular Immunocytochemistry

View larger version (143K): [in this window] [in a new window]   Fig. 5.   Sagittal sections of fetal kidney immunocytochemically stained for renin. Left, denervated kidney; right, innervated kidney. A: control fetus. B: occluder fetus. Note that the percentage of positive-staining glomeruli increased in the occluder group compared with the control group irrespective of innervation.

Renal Norepinephrine Levels, Animal Data, and Blood Gas Data

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

The present study demonstrates that renal denervation in the near-term ovine fetus does not alter the renin mRNA response to chronic reductions in RPP. After 24 h of reduced descending aortic perfusion pressure, renin mRNA level increased similarly in both the innervated and denervated renal cortex. Although increases in renin mRNA are not associated with measurable differences in active renin or prorenin content when comparing innervated and denervated samples in the basal state or in response to aortic occlusion, the increase in renin mRNA is associated with a significant elevation in pARC and pPRC. Quantitative immunocytochemical analysis for renin-containing glomeruli confirms an increase in the percentage of positive-staining glomeruli in response to decreased descending aortic perfusion pressure in both intact and denervated kidney. Together, these observations suggest that in the near-term ovine fetus, the absence of renal innervation does not alter the renal baroreceptor-mediated renin gene expression and secretory responses to chronic reductions in descending aortic perfusion pressure.

Our findings of unaltered basal renin mRNA levels in the denervated renal cortex compared with the contralateral innervated cortex within the same animal suggest that renal sympathetic nerve tone in the fetus is not required to maintain basal renin gene expression. Although it has been proposed that a tonic stimulation of renin-expressing cells may be necessary to maintain the RAS in a state of preparedness both at the gene expression and secretory levels (5, 22), our findings are consistent with previous work in near-term fetal sheep showing no significant effect of renal denervation on basal renal functioning (14, 16, 17, 20, 21). Our data agree with the observations of El Dahr et al. (2), who showed that chemical sympathectomy does not change renin gene expression in maturing neonatal rats. They also agree with previous work in ovine fetuses showing no reduction in renin mRNA levels after bilateral renal denervation (1). However, our observations contrast with most reports on the effect of denervation in adult animals that indicate a marked reduction in renin mRNA level after renal denervation (5, 14, 22). This suggests that the effect of denervation on basal renin mRNA level is minimal in earlier developmental ages.

Studies addressing the importance of the renal nerves in the renin mRNA response to stimulation in adults have not been uniform. Renal denervation does not affect the increase in renin mRNA produced by hemorrhage (5), hypoxia (11), or the administration of the ANG II antagonist losartan (23). These manipulations may influence renin mRNA quantitatively by more than one mechanism. For example, hemorrhage probably stimulates renin mRNA by increasing renal sympathetic nerve activity and plasma catecholamines while triggering the renal baroreceptor mechanism through reductions in RPP. The multiplicity of the different renin mRNA stimulatory events could therefore conceal any effect resulting from loss of the renal nerves. On the other hand, unilateral renal artery stenosis increases renin mRNA level in the hypoperfused kidney, and this response is blunted by renal nerve ablation (23). Thus the application of a stimulus that operates through one or possibly two mechanisms demonstrates that the renal nerves are required for the stimulation of renin mRNA produced by a reduction in RPP in the adult animal. This contrasts with our observation that denervation of the fetal kidney does not attenuate the rise in renin mRNA produced by a reduction in RPP and suggests that the effect of denervation on renin mRNA level is influenced by developmental age. Although, alternatively, the findings or reduced basal renin mRNA and mRNA response may be related to the species studied, studies in fetal and adult sheep in our laboratory support our findings that the renal nerve is not essential for maintaining basal renin mRNA levels and response (6). In addition, our data support the premise that the sympathetic nervous system in adult animals is more influential in maintaining renin mRNA expression compared with the fetus.

Another means by which renin gene expression may be activated by the reductions in RPP imposed in this study is by the triggering of the macula densa mechanism. However, this experiment remains unable to ascertain those changes in renin gene expression that may have occurred as a result of activation of the macula densa. In studies of induced reductions in renal blood flow of up to 26% in the presence of renal denervation and -adrenergic antagonism, Nakamura and Johns (13) suggested that renin gene expression as a consequence of change in renal blood flow and the activation of the macula densa mechanism appears to be relatively small.

Other variables considered important in modulating renin mRNA levels include the reno-renal reflex, in which ipsilateral renal denervation may increase the contralateral efferent renal nerve activity (7, 8) and the denervation induced reduction of specific neurotransmitters (9) or peptides (10) important for gene expression. We believe these changes would tend to accentuate rather then minimize any measurable difference in renin gene expression occurring between the ipsilateral denervated and contralateral innervated kidneys. Conversely, one may suggest that renal denervation results in denervation hypersensitivity and an exaggerated response to circulating catecholamines that could potentially increase renin gene expression, making observable differences between the denervated and innervated kidneys less obvious.

The chronic reduction in descending aortic perfusion pressure elevated both pARC and pPRC similar to that pattern observed in an earlier study (18). This increase in secretory activity was not accompanied initially by similar changes in plasma PRC. The proportion of the ARC or PRC that originated from the left, denervated kidney as opposed to the right intact kidney is unknown from the current experimental protocol.

Analysis of basal state renal renin content after denervation shows no significant change in renal ARC or PRC. In response to aortic occlusion, renal ARC elevations in both the denervated and innervated kidney resulted in a 26 and 19% increase in active renin content, respectively. However, this represented a tendency, as these were not significant increases. The change in renal PRC content was minimal in both the basal state and after aortic occlusion when comparing denervated and innervated cortex. The increase in plasma levels of active and prorenin coupled with the lack of change in tissue concentrations of renin suggest either large pools of active and prorenin exist in the kidney, or there is an increase in renin synthesis coupled with increased secretion caused by the reduction in RPP (18).

Immunocytochemical analysis for renin-containing glomeruli show a similar increase in renin-expressing juxtaglomerular cells in both the denervated and innervated kidney in response to aortic occlusion. The immunoreactivity was not altered by renal innervation status in the basal state. These findings contrast to that of Holmer et al. (5), in which adult rats with left renal denervation and a reduction in renin mRNA were found to have fewer renin-positive immunoreactive areas per vascular pole compared with the adjacent kidney. In our study, increases in renin-containing glomeruli with aortic occlusion were associated with similar directional changes in renal renin mRNA values in the fetus, confirming the relative relationship between renin gene expression and renin immunoreactivity (4).

In summary, our findings in the near-term ovine fetus indicate that renal denervation does not alter the basal renin mRNA level or renin expression response to reductions in descending aortic perfusion pressure. Although no significant change in renal active renin or prorenin content is observed after 24 h of reduced RPP, significant increases in plasma active renin and prorenin persist. Increases in renin-staining glomeruli in response to reduced RPP are quantitatively similar in denervated and innervated kidney. More importantly, our findings suggest that the degree of influence of the renal nerves in maintaining the renin mRNA expression is developmentally regulated.

Perspectives