Renal Dysfunction Is Associated With a Reduced Contribution of Nitric Oxide and Enhanced Vasoconstriction After a Congenital Renal Mass Reduction in SheepCLINICAL PERSPECTIVE
Background—Children born with reduced congenital renal mass have an increased risk of hypertension and chronic kidney disease in adulthood, although the mechanisms are poorly understood. Similar sequelae occur after fetal uninephrectomy (uni-x) in sheep, leading to a 30% nephron deficit. We hypothesized that renal dysfunction is underpinned by a reduced contribution of nitric oxide (NO) and vascular dysfunction in uni-x sheep.
Methods and Results—In 5-year-old female uni-x and sham sheep, mean arterial pressure, glomerular filtration rate, and renal blood flow were measured before and during NO inhibition (Nω-nitro-l-arginine methyl ester [L-NAME]). Reactivity was assessed in resistance arteries, including renal lobar and arcuate arteries. Basal mean arterial pressure was 15 mm Hg higher and glomerular filtration rate and renal blood flow were ≈30% lower (P<0.001) in uni-x animals. L-NAME increased mean arterial pressure by ≈17 mm Hg in both groups, whereas glomerular filtration rate and renal blood flow were decreased less in uni-x sheep (PInteraction<0.01). Endothelial NO synthase and Ser-1177–phosphorylated endothelial NO synthase protein levels were upregulated in renal cortex of uni-x sheep (P<0.05). Lobar arteries of uni-x sheep had enhanced responsiveness to phenylephrine and nitrotyrosine staining and reduced sensitivity to endothelial stimulation. Vasodilator prostanoid contribution to endothelium-dependent relaxation was reduced in lobar arteries of uni-x sheep, accompanied by reduced cyclooxygenase-1 and -2 gene expression (P<0.05). Neurovascular constriction was enhanced ≈1.5-fold in renal arteries of uni-x sheep (P<0.05).
Conclusions—Renal dysfunction after congenital renal mass reduction is associated with impaired regulation of renal hemodynamics by NO. Reductions in renal blood flow and glomerular filtration rate are underpinned by impaired basal NO contribution, endothelial dysfunction, and enhanced vascular responsiveness to sympathetic nerve stimulation.
Epidemiological studies reveal that children born with a single functioning kidney have a predisposition to developing renal insufficiency and hypertension in adult life.1–4 A congenital reduction in renal mass is present in patients with unilateral renal agenesis (≈1 in 500 births)1 or unilateral multicystic kidney (≈1 in 4300 births).3 A reduced congenital renal mass results in impaired renal function, with 50% of patients developing hypertension before 18 years of age2 and a 50% probability of requiring dialysis by 30 years of age.4 Furthermore, individuals born with low birth weight have smaller kidneys and lower estimated glomerular filtration rate (GFR), with an increased risk of developing cardiovascular diseases.2 Many animal models of early life insults have a similar relationship between low birth weight and the development of hypertension in adulthood, with this link appearing to be mediated, at least in part, by an associated congenital nephron deficit.5 Substantial evidence suggests that developmental perturbations that alter kidney development play a critical role in the pathogenesis of hypertension.6 However, the mechanisms underlying the alterations in renal function are poorly understood, especially in more clinically relevant animal models of advanced age.
Clinical Perspective on p 288
We have established an ovine model of fetal uninephrectomy (uni-x) that replicates the consequences of a congenital nephron deficit in programming hypertension7,8 without the confounding effects of low birth weight often linked with other models.9 In this model, renal dysfunction precedes the onset of hypertension in female uni-x sheep, with a decline in GFR observed at 1 year of age, before the development of hypertension, that is associated with elevations in blood volume and cardiac output at 2 years of age.10 Furthermore in this model, a greater reduction in renal blood flow (RBF), associated with increases in renal and total peripheral resistance, was observed between 2 and 5 years of age, indicative of disease progression with age.10 Similarly, uni-x in rats on postnatal day 1 (a period of active nephrogenesis in rodents) results in age-dependent reductions in GFR and hypertension.11 Nitric oxide (NO) produced within the kidney plays an important role in the regulation of renal hemodynamics and sodium excretion, thus maintaining systemic vascular volume and blood pressure.12 Vascular endothelial and smooth muscle cells and the extracellular matrix are also important determinants of vascular homeostasis and blood pressure regulation.13,14 Deficiency in NO production/bioavailability has been reported in different animal models of chronic kidney disease associated with renal mass reduction and aging, contributing to a progressive loss of renal function.15 Moreover, vascular dysfunction in adulthood is a common legacy of exposure to early life insults and may contribute to the development of hypertension.16
The aim of the present study was to investigate the cardiovascular and renal responses to in vivo NO synthase (NOS) inhibition via the administration of Nω-nitro-l-arginine methyl ester (L-NAME) in conscious female uni-x and sham sheep at 5 years of age. The effects of fetal uni-x on smooth muscle and endothelial function, passive wall stiffness, nerve-mediated vasoconstriction, and oxidative stress were assessed in isolated renal arteries. To examine the effects of uni-x over the wider circulation, resistance arteries from 3 other vascular beds (mesenteric, femoral, and coronary) were also studied. We hypothesized that NO-mediated control of renal and vascular function is impaired in animals born with a congenital reduction in renal mass.
Please refer to the online-only Data Supplement for an expanded Methods section.
All experiments were approved by the Monash University, School of Biomedical Sciences Animal Ethics Committee and were in agreement with the guidelines of the National Health and Medical Research Council of Australia. Merino ewes carrying female fetuses of known gestational age underwent surgery at 100 days after conception, as described previously.17 The left renal artery, vein, and ureter were ligated, and the kidney was removed in the uni-x group. The kidney was cleared of surrounding fat but left intact in the sham group. After natural birth, lambs were kept with their mothers until weaned at 4 months of age and had carotid arterial loops prepared at 5 months of age.18 At 5 years of age, uni-x and sham animals were placed in individual metabolic cages, and fitted with a carotid catheter for measurement of arterial pressure, and the left and right jugular veins were catheterized for drug infusions. Animals were fitted with a removable Foley bladder catheter for urine collection and measurement of renal function.
Blood Pressure and Renal Function
Mean arterial pressure (MAP) and heart rate were recorded continuously via the carotid arterial loop. GFR was determined via the clearance of 51chromium-ethylenediamine-tetra-acetic acid (51Cr EDTA); RBF was determined via the clearance of para-aminohippuric acid; and filtration fraction was calculated. Para-aminohippuric acid concentrations were assessed with a rapid microplate assay19; 51Cr EDTA levels were assessed with a gamma counter (PerkinElmer Wizard 1470); and urinary sodium concentrations were measured with a RapidChem 744 Electrolyte analyzer.
Cardiovascular and Renal Responses to NO Inhibition
Cardiovascular function and renal function were monitored during a 60-minute basal period, after which animals were infused with L-NAME (40 mg/kg bolus plus 20 mg·kg−1·h−1 IV; Sapphire Biosciences) for 60 minutes. Urine samples were collected at 30-minute intervals, with an arterial blood sample (5 mL) collected at the midpoint of each urine sample collection during the first basal hour and 1 hour after L-NAME infusion. Seven days after the in vivo experiments, sheep were euthanized (overdose of pentobarbitone [Lethabarb]; 325 mg/mL), and renal lobar and arcuate arteries and mesenteric, femoral, and coronary arteries were isolated for assessment of vascular function and passive mechanical wall properties.
Smooth Muscle, Endothelial, and Nerve-Mediated Responses in the Vasculature
Rings of renal, mesenteric, femoral, and coronary arteries 1 to 2 mm in length were mounted on a 4-channel myograph (model 610M, Danish Myo Technology, Aarhus, Denmark) for measurement of isometric tension, as previously described.20 Contractile properties were assessed by cumulative additions of either the α1-adrenoreceptor agonist phenylephrine (10−9–10−4 mol/L, renal, mesenteric and femoral arteries) or the thromboxane analog U46619 (U4; 10−9–3×10−6 mol/L, coronary artery). Contractions were expressed as a percentage of contraction evoked by high K+ physiological saline solution (isotonic replacement of Na+ with 100 mmol/L K+). In arteries that were submaximally constricted, endothelium-dependent relaxation was tested with either bradykinin (10−10–10−6 mol/L, renal, mesenteric, and coronary arteries) or acetylcholine (10–9–10–5 mol/L, femoral artery). Responses were obtained before and after sequential blockade of NOS with L-NAME (2×10−4 mol/L) and cyclooxygenase with indomethacin (10−6 mol/L). Endothelium-independent relaxation was tested with the NO donor sodium nitroprusside (10−9–10−5 mol/L).21 For perivascular nerve stimulation, segments of renal lobar arteries were mounted on a single-channel wire myograph, and the smaller renal arcuate arteries were mounted on a pressure myograph (Living Systems Instrumentation, Burlington, VT). Platinum electrodes positioned on either side of the artery were used to stimulate the perivascular nerves along the artery segment, as previously described.22,23 Passive mechanical wall properties were determined in leak-free segments of arteries mounted on a pressure myograph with no luminal flow and bathed in zero-Ca2+ physiological saline solution containing 2 mmol/L EGTA at ≈36°C, as previously described.20
Markers of Endothelial Function and Oxidative Stress
At postmortem, a 0.5-cm slice in the transverse plane was taken from the right kidney, the cortex, and medulla divided; and snap-frozen for molecular studies. Renal lobar arteries were isolated and snap-frozen for gene and protein expression and 3-nitrotyrosine immunohistochemistry.
Data are reported as mean±SEM, and n represents the number of animals studied. Renal function variables were corrected for body weight. Data were analyzed with repeated-measures ANOVA or an unpaired Student t test. For endothelium-dependent relaxation, the contribution of each vasodilator was determined with an area under the curve (AUC) analysis. The response evoked by NO was determined by subtracting the AUC in the presence of L-NAME from that obtained in normal physiological saline solution. The relaxation attributed to the vasodilator prostanoids was calculated by subtracting the AUC in the presence of L-NAME+indomethacin from that obtained in L-NAME alone. Responses remaining in the presence of both blockers were attributed to endothelium-derived hyperpolarizing factor (EDHF). Statistical analysis was performed with GraphPad PRISM 5.03.
Reduced Role for NO in the Modulation of Renal Hemodynamic Function in Uni-x Sheep
Body weight and kidney weight were not significantly different between the uni-x and sham groups at 5 years of age, as previously reported for this cohort.10,24 Basal MAP (≈15 mm Hg) and renal vascular resistance (RVR; ≈55%) were greater and RBF (≈42%), GFR (≈38%), and urinary sodium excretion (≈46%) were lower in the uni-x compared with the sham group (all P<0.01; Table I in the online-only Data Supplement). In response to L-NAME infusion, MAP increased by 16±3 mm Hg in uni-x and 18±3 mm Hg in sham sheep, with the response not significantly different between groups (PInteraction=0.3; Figure 1A). This increase in MAP was associated with a similar reduction in heart rate in both groups (PInteraction=0.2; Figure 1B). The reduction in RBF in response to L-NAME was blunted in the uni-x compared with the sham sheep (13±5% versus 60±8%, respectively; PInteraction<0.001; Figure 1C). In response to L-NAME infusion, the increase in RVR was modest in uni-x compared with sham sheep (20±10% versus 160±31%, respectively; PInteraction=0.04; Figure 1D).
L-NAME infusion caused a reduction in GFR, with the change attenuated in uni-x compared with sham animals (27±5% versus 48±6%, respectively; PInteraction=0.002; Figure 1E). Filtration fraction did not significantly change in response to L-NAME infusion in either group (PTreatment=0.9; Figure 1G). In response to L-NAME infusion, there was no significant difference in urine flow between the 2 groups (PInteraction=0.6; Figure 1F); however, urinary sodium excretion was markedly decreased in sham but not in uni-x animals (PInteraction=0.002; Figure 1H).
Endothelial Dysfunction Localized Predominantly to the Renal Arteries of Uni-x Sheep
Stimulation of the endothelium with bradykinin or acetylcholine evoked concentration-dependent relaxation in all arteries (Figure 1 in online-only Data Supplement). Renal arteries from uni-x animals were less sensitive to bradykinin compared with those from sham sheep (pD2 [negative logarithm of the concentration of agonist producing half maximal response], P<0.01; Table II in the online-only Data Supplement), and the AUC for endothelium-dependent relaxation was significantly reduced in uni-x sheep (P=0.04; Figure 2A). In contrast, the AUC for endothelium-dependent relaxation was not different between treatment groups for mesenteric, coronary, or femoral arteries (Figure 2B–2D). In the presence of L-NAME, the renal arteries of uni-x sheep developed markedly less spontaneous tone compared with those from sham sheep (P=0.04; Figure 2E). In the presence of L-NAME, the concentration-relaxation curve to bradykinin was shifted to the right to a significantly greater extent in renal lobar arteries of uni-x (4-fold) compared with sham sheep (2.5-fold; Figure IA and IB in the online-only Data Supplement). In the presence of L-NAME and indomethacin, the rightward shift of the concentration-relaxation curve to bradykinin was markedly greater in renal arteries of sham (16-fold) compared with uni-x (3-fold) sheep (Figure IA and IB in the online-only Data Supplement). AUC analysis revealed that endothelium-dependent relaxation attributed to prostanoids was significantly (≈46%) reduced in renal arteries of uni-x compared with sham sheep (P<0.05; Figure 2A). The contribution of NO and EDHF to bradykinin-induced endothelium-dependent relaxation was unaltered between the sham and uni-x groups.
Sensitivity to bradykinin or acetylcholine was not different in mesenteric, coronary, and femoral arteries between the sham and uni-x sheep (Figure IC–IH in the online-only Data Supplement). The AUC for endothelium-dependent relaxation in these arteries was also not different between the sham and uni-x sheep (Figure 2B–2D). However, the contribution of prostanoids was reduced (≈37%) in coronary arteries from uni-x compared with sham sheep (P<0.05; Figure 2C). The contribution of NO and EDHF to total agonist-evoked endothelium-dependent relaxation was not different between the experimental groups across all vascular beds (Figure 2).
Enhanced Vasoconstriction in Uni-x Sheep
Phenylephrine evoked concentration-dependent contraction, with renal lobar, mesenteric, and femoral arteries from uni-x sheep all demonstrating enhanced sensitivity to the α1-adrenoreceptor agonist compared with arteries from sham sheep (pD2; P<0.05 for all; Figure 3A, 3B, and 3D and Table II in the online-only Data Supplement). Maximum contraction to phenylephrine in renal lobar, mesenteric, and femoral arteries was significantly greater in uni-x sheep (P<0.05 for all; Figure 3A, 3B, and 3D and Table II in the online-only Data Supplement). Contraction evoked by U46619 in coronary arteries was not different between uni-x and sham sheep. Absolute contraction evoked by high K+ physiological saline solution was not different between uni-x and sham animals for any of the arteries tested (Table II in the online-only Data Supplement). Sodium nitroprusside produced concentration-dependent relaxation, and the sensitivity and maximal response were not different between the uni-x and sham groups across all vascular beds (Figure 3E–3H and Table II in the online-only Data Supplement).
Enhanced Renal Vasoconstriction to Sympathetic Nerve Activation
Stimulation of the renal lobar and arcuate arteries with pulses of increasing frequency evoked contractions of increasing amplitude (PFrequency<0.001; Figure 4A and 4C). Neurovascular constriction was enhanced in both the renal lobar and arcuate arteries of uni-x compared with sham sheep (PGroup=0.04 [lobar] and PGroup=0.02 [arcuate]; Figure 4A and 4C). These contractions were markedly attenuated by the α1- adrenoceptor blocker prazosin and all but abolished in the presence of tetrodotoxin (Figure III in the online-only Data Supplement). For any given stimulus frequency, increasing stimulus voltage increased contraction amplitude, and to a greater extent in renal lobar arteries from uni-x compared with sham sheep (PGroup <0.05 for all; Figure IV in the online-only Data Supplement). The enhanced responses to nerve stimulation in renal lobar arteries of uni-x sheep were associated with an augmented sensitivity of the smooth muscle to exogenous α1- adrenoceptor stimulation (PGroup=0.001; Figure 4B). In contrast, enhanced neurovascular constriction in renal arcuate arteries from uni-x sheep was not accompanied by changes in smooth muscle sensitivity to α1- adrenoceptor stimulation (Figure 4D).
Preservation of Arterial Passive Mechanical Wall Properties
Arterial wall stiffness, as assessed from stress-strain relationships, was not different between uni-x and sham animals for any of the arteries tested (Figure VA–VD in the online-only Data Supplement). The ratio of media thickness to lumen diameter at 100 mm Hg was not different between the uni-x and sham groups for any of the arteries tested (Table II in the online-only Data Supplement).
Enhanced Endothelial NOS Protein Expression and Oxidative Stress in Renal Arteries of Uni-x Sheep
At 5 years of age, endothelial NOS (eNOS) mRNA expression was significantly upregulated in the renal cortex of uni-x compared with sham sheep (P<0.001; Figure VIA in the online-only Data Supplement). Although expression of eNOS was greater within the medulla compared with the cortex (P=0.002; Figure VIA in the online-only Data Supplement), there was no difference in eNOS expression in the medulla between sham and uni-x sheep. Similar to renal cortex, eNOS mRNA expression was significantly greater in isolated renal lobar arteries of uni-x sheep (P=0.03; Figure VIB in the online-only Data Supplement). Renal lobar arteries of uni-x sheep had significantly lower expression of both cyclooxygenase (COX)-1 (P=0.03; Figure VIC in the online-only Data Supplement) and COX-2 (P=0.026; Figure VID in the online-only Data Supplement) mRNA compared with those from sham animals. Consistent with the increase in renal eNOS gene expression in the uni-x sheep, eNOS protein expression was also increased in the renal cortex of uni-x compared with the sham sheep (P=0.02; Figure 5A). Furthermore, the Ser-1177–phosphorylated eNOS protein level in the renal cortex was also increased in the uni-x compared with the sham sheep (P=0.02; Figure 5B). However, the ratio of eNOS to phosphorylated eNOS protein was not significantly different between groups (Figure 5C). The intensity of 3-nitrotyrosine staining was stronger in isolated renal lobar arteries of uni-x compared with sham sheep (Figure 6A and 6B). Quantification of 3-nitrotyrosine revealed a greater fluorescence intensity in arteries of uni-x compared with sham sheep (P=0.02; Figure 6D).
The present study provides compelling in vivo and in vitro evidence of the deleterious effects of being born with a reduction in congenital renal mass on the regulation of renal and vascular function. Our study provides the first report that renal dysfunction in adult sheep after fetal uni-x is associated with impaired modulation of renal hemodynamics and sodium excretory function by NO. We also report renal vascular dysfunction with impaired contribution of basal NO, endothelial dysfunction, increased oxidative stress, and enhanced smooth muscle responsiveness to vasoconstrictors and sympathetic nerve stimulation in sheep born with a reduction in congenital renal mass. Dysfunction was also observed in blood vessels from other regions of the body, but this was less extensive than in the renal vasculature.
In the present study, sham sheep had significant increases in MAP and RVR within 60 minutes of L-NAME infusion. This finding is consistent with studies in several species, including sheep, and confirms the important contribution of NO to the cardiovascular system.12 The striking increase in RVR (≈160%) in sham sheep was associated with reductions in both RBF (≈60%) and GFR (≈48%), suggesting that NO tonically generated within the kidney lowers resting RVR and directly modulates glomerular hemodynamics. Systemic NO blockade in rodents produces increases in both afferent and efferent arteriolar resistances, leading to reductions in renal plasma flow.25,26 There is also evidence for a role for NO in facilitating sodium excretion by blocking sodium transporters in proximal tubules and collecting ducts and in mediating pressure natriuresis via increases in renal medullary blood flow and interstitial hydrostatic pressure.12 Sham sheep exhibited significant reductions in urinary sodium excretion (≈50%) after L-NAME infusion, consistent with studies in anesthetized dogs in which significant reductions in natriuresis occurred.27 These results highlight the importance of endogenous NO within the kidney in the regulation of renal hemodynamics and sodium excretory function under normal physiological conditions.
In contrast to sham animals, the reductions in GFR (27%) and RBF (13%) in uni-x sheep in response to L-NAME infusion were modest and occurred despite a similar rise in MAP (≈17%) in both groups. The minimal renal hemodynamic responses to NO inhibition suggest a deficit in basal NO generation/bioavailability within the vasculature of the uni-x kidney in female sheep at 5 years of age. A role of endothelial NO deficiency has been implicated in mediating hypertension in several animal models of fetal programming in which the maternal environment has been adversely affected.28,29 Moreover, there have been reports of reductions in neuronal NOS expression and activity within the renal cortex and medulla in a chronic kidney disease model produced by renal mass reduction after 5/6 nephrectomy in adult rats.30 The increase in RVR in response to L-NAME was far greater in sham (≈160%) compared with uni-x (≈20%) sheep, and this was associated with a reduction in sodium excretion in sham but not uni-x animals. A reduction in basal NO generation within the kidneys has been shown to decrease renal sodium excretory function directly by inhibiting tubular reabsorption or indirectly by increasing basal RVR or enhancing renal vascular responsiveness to vasoconstrictors such as angiotensin II and renal sympathetic nerve activity.12 Previously, we have shown that these 5-year-old uni-x sheep have a reduced ability to excrete a saline load; thus, reduced NO production may be a contributing factor.24 In the present study, the eNOS gene and eNOS and phosphorylated eNOS protein expression were upregulated in the renal cortex of uni-x sheep, although the ratio of phosphorylated eNOS to eNOS protein was unchanged. Furthermore, 3-nitrotyrosine staining, indicative of oxidative stress, was increased in the renal vasculature. Under conditions of increased oxidative stress, for example, in diseases such as hypertension and diabetes mellitus, an upregulation of eNOS gene and protein expression is generally observed with eNOS uncoupling, resulting in decreased NO bioavailability, increased superoxide formation, and disrupted eNOS dimer formation.31,32 Collectively, our in vivo data suggest that the contribution of NO to the regulation of renal hemodynamics and sodium excretion is markedly blunted in uni-x sheep (See the online-only Data Supplement for further discussion of the timing of these changes).
Endothelium-dependent vasodilation in vivo is elicited by shear stress and paracrine and circulating agents.33 The reduction in basal RBF and increased RVR in uni-x animals, together with the blunted renal hemodynamic responses to NO inhibition, may point to dysfunction in NO release or bioavailability in these animals. Indeed, a decrease in flow/shear stress–induced dilation has commonly been reported in humans with chronic hypertension, and this is underpinned by reductions in NO-mediated vasodilation.34 Studies in spontaneously hypertensive rats have demonstrated impairments in flow-dependent dilation of small arteries in which endothelial NO activity in response to shear stress was impaired35 or preserved,36 depending on the region of the vascular bed under investigation. In the present study, basal tone generation after NOS inhibition in isolated renal lobar arteries was markedly blunted in uni-x animals, suggesting impairments in basal NO production. However, this was not observed in arteries from the 3 other vascular beds examined, highlighting region-dependent alterations in vascular function in response to fetal uni-x. Interestingly, the minimal renal hemodynamic responses to systemic NO blockade and the lower vascular tone generated under NOS inhibition occurred in the presence of an upregulation of eNOS gene and protein expression in the renal cortex and eNOS gene expression in lobar arteries. Furthermore, renal artery smooth muscle responsiveness to sodium nitroprusside was unaltered, indicating that the guanylate cyclase–cGMP pathway is intact in arteries from uni-x sheep. Collectively, our results indicate the possibility of a reduction in NO production or bioavailability. The observation that renal arteries had significantly greater 3-nitrotyrosine staining in the aged uni-x sheep suggests that a reduction in NO bioavailability may be occurring.
In isolated arteries, agonist stimulation of the endothelium evoked the release of NO, vasodilator prostanoids, and EDHF. Whereas the contribution of basal NO to renal artery tone suppression was reduced in arteries of uni-x sheep, the role of NO in agonist-induced endothelium-dependent relaxation was unaltered. However, endothelium-dependent relaxation was impaired in renal lobar arteries of uni-x animals and attributable to the reduced contribution of vasodilator prostanoids. The expression and location of COX-1 and COX-2 enzymes appear to be the major rate-limiting factors in the synthesis of prostanoids.37 Both COX-1 and COX-2 are expressed in the vascular endothelium,37 with COX-2 proposed to be the major mediator of vasodilator prostanoid synthesis, especially during the evolution of compensatory renal functional changes after renal ablation.38 In the present study, the reduced contribution of vasodilator prostanoids to endothelium-dependent relaxation in renal lobar arteries of uni-x sheep was associated with reductions in both COX-1 and COX-2 gene expression. Endothelial dysfunction in programming models has been attributed to reductions in NO-mediated5,28,29 or EDHF-mediated 9,20,39,40 relaxation, whereas the involvement of vasodilator prostanoids in the regulation of vascular tone in programming models is less widely reported. Reduced prostanoid production underpins impaired coronary artery vasodilation in adolescent lambs exposed to dexamethasone in utero.41 Interestingly, the contribution of vasodilator prostanoids was also reduced in coronary arteries of uni-x animals, but not in the mesenteric and femoral beds. Within the renal vasculature, vasodilator prostanoids have myriad functions such as modulating RBF and GFR and promoting natriuresis via their vasodilator actions.37 Thus, the reduced vasodilator prostanoid contribution to endothelial vasodilation within the renal vasculature of uni-x animals may also contribute to reductions in RBF and GFR and subsequently to the lower basal urinary sodium excretion. Endothelial vasodilator dysfunction has been reported in patients with chronic renal failure, attributed mainly to a reduction in NO activity42,43; however, alterations in other vasodilators (eg, prostanoids and EDHF) are also implicated.43 Despite the hypertension associated with fetal uni-x, endothelial dysfunction was not generalized across all vascular beds. Our findings highlight the heterogeneity across the vasculature in response to fetal uni-x, rendering certain vascular beds more vulnerable than others. Similarly, when arteries from different vascular beds were tested in other fetal programming models, there appeared to be regional variations in the extent and nature of dysfunction.20,40,44
Enhanced vasoconstrictor responses have been reported for some fetal programming models.9 Enhanced responsiveness to α1-adrenoreceptor agonists occurs in femoral and renal arteries of adult rats exposed to nutrition deprivation in utero.44,45 We demonstrated enhanced smooth muscle responsiveness to α1-adrenoreceptor agonists in the renal lobar, mesenteric, and femoral arteries in uni-x sheep. The female uni-x sheep used in the present study had higher total peripheral resistance at 2 years of age.10 Furthermore, there was an exacerbation in total peripheral resistance with aging between 2 and 5 years, which was greater in uni-x compared with sham sheep.10 Thus, generalized increased responsiveness to vasoconstrictors may contribute to the physiological maintenance and progression of the higher baseline peripheral resistance in uni-x sheep.
Neurovascular constriction was upregulated in renal vessels of uni-x sheep. In the renal lobar arteries, this increase in responsiveness to neurovascular constriction could be explained by the enhanced sensitivity and maximal response of the smooth muscle to α1-adrenoreceptor stimulation. In contrast, in the smaller arcuate arteries, enhanced neurovascular constriction occurred in the absence of any changes in α1-adrenoceptor responsiveness of the smooth muscle. Thus, in the arcuate arteries of uni-x sheep, there may be augmentation in the density of the perivascular nerve network or alteration in the amount and regulation of transmitter release.46 Blockade of α1-adrenoreceptors virtually abolished the constrictor responses to nerve stimulation, indicating that noradrenaline is the main neurotransmitter responsible for vasoconstriction in renal arteries of both sham and uni-x sheep. These findings strongly support a role of an enhanced renal sympathetic innervation in the programming of hypertension in uni-x sheep. This is consistent with other fetal programming models of hypertension,6,47 and in them, renal denervation is effective in abolishing hypertension.
The present study demonstrates that renal dysfunction and hypertension after congenital nephron deficit are underpinned by a reduced contribution of NO to the modulation of renal hemodynamics and function in association with enhanced oxidative stress. This is further exacerbated by a proconstrictor profile in the renal vasculature characterized by endothelial vasodilator dysfunction, enhanced sensitivity to vasoconstrictors, and sympathetic nerve stimulation. Identification of these pivotal mechanisms may have potential implications for improving prognosis and treatment for children born with 1 kidney. For instance, recent studies have suggested that catheter ablation of the renal nerves may be effective in the treatment of hypertension and chronic renal disease. Our data suggest that such a strategy might slow the progression of renal dysfunction in patients with renal agenesis because it would remove the enhanced response to sympathetic activation and improve endothelial function. This warrants further investigation.
Sources of Funding
This study was funded by a National Heart Foundation grant (G 05M 2110), National Health and Medical Research council (SRF 1041844, GNT 0546087), Australia and New Zealand Trustees, The University of Queensland Research Fund, and the Monash University Research Fund.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/lookup/suppl/doi:10.1161/CIRCULATIONAHA.114.013930/-/DC1.
- Received June 28, 2014.
- Accepted October 24, 2014.
- © 2014 American Heart Association, Inc.
- Schreuder MF,
- Westland R,
- van Wijk JA
- Sanna-Cherchi S,
- Ravani P,
- Corbani V,
- Parodi S,
- Haupt R,
- Piaggio G,
- Innocenti ML,
- Somenzi D,
- Trivelli A,
- Caridi G,
- Izzi C,
- Scolari F,
- Mattioli G,
- Allegri L,
- Ghiggeri GM
- Alexander BT
- Kett MM,
- Denton KM
- Lankadeva YR,
- Singh RR,
- Tare M,
- Moritz KM,
- Denton KM
- Nuyt AM
- Kone BC,
- Baylis C
- Baylis C
- Agarwal R
- Deng A,
- Baylis C
- Baylis C,
- Harton P,
- Engels K
- Majid DS,
- Williams A,
- Navar LG
- Franco Mdo C,
- Arruda RM,
- Dantas AP,
- Kawamoto EM,
- Fortes ZB,
- Scavone C,
- Carvalho MH,
- Tostes RC,
- Nigro D
- Thum T,
- Fraccarollo D,
- Schultheiss M,
- Froese S,
- Galuppo P,
- Widder JD,
- Tsikas D,
- Ertl G,
- Bauersachs J
- Paniagua OA,
- Bryant MB,
- Panza JA
- Koller A,
- Huang A
- Roghair RD,
- Volk KA,
- Lamb FS,
- Segar JL
- Annuk M,
- Lind L,
- Linde T,
- Fellström B
- Morris ST,
- McMurray JJ,
- Rodger RS,
- Jardine AG
- Tripovic D,
- Pianova S,
- McLachlan EM,
- Brock JA
Children born with a reduced renal mass have an increased risk of chronic kidney disease and hypertension in adulthood. To understand the mechanisms underlying the early development of renal insufficiency in children born with a solitary functioning kidney, we established a model of fetal uninephrectomy (uni-x) in sheep, which, similar to humans, completes nephrogenesis before birth. This model results in a 30% reduction in nephron number rather than 50% as a result of compensatory nephrogenesis in the remaining kidney. Similar to children with a congenital solitary functioning kidney, uni-x sheep demonstrate a progressive increase in arterial pressure and loss of renal function with aging. The present study, using a combination of in vivo studies in conscious sheep and in vitro functional and molecular studies in isolated arteries, defines the mechanisms responsible for impairments in renal function in uni-x sheep at an advanced age. We report that renal dysfunction in conscious 5-year-old female sheep after fetal uni-x is associated with impaired modulation of renal hemodynamics and sodium excretion by nitric oxide. Furthermore, marked vascular dysfunction, predominantly in the renal vasculature, was observed in uni-x sheep that was underpinned by endothelial vasodilator dysfunction involving nitric oxide and prostanoid deficiencies, enhanced oxidative stress, and augmented smooth muscle responsiveness to vasoconstrictors and sympathetic nerve stimulation. Identification of these pivotal mechanisms may have potential implications for improving the prognosis and treatment of children born with a solitary functioning kidney.