Chronic hypoxia in pregnancy affected vascular tone of renal interlobar arteries in the offspring

Hypoxia during pregnancy could affect development of fetuses as well as cardiovascular systems in the offspring. This study was the first to demonstrate the influence and related mechanisms of prenatal hypoxia (PH) on renal interlobar arteries (RIA) in the 5-month-old male rat offspring. Following chronic hypoxia during pregnancy, phenylephrine induced significantly higher pressor responses and greater vasoconstrictions in the offspring. Nitric oxide mediated vessel relaxation was altered in the RIA. Phenylephrine-stimulated free intracellular calcium was significantly higher in the RIA of the PH group. The activity and expression of L-type calcium channel (Cav1.2), not T-type calcium channel (Cav3.2), was up-regulated. The whole-cell currents of calcium channels and the currents of Cav1.2 were increased compared with the control. In addition, the whole-cell K+ currents were decreased in the offspring exposed to prenatal hypoxia. Activity of large-conductance Ca2+-activated K+ channels and the expression of MaxiKα was decreased in the PH group. The results provide new information regarding the influence of prenatal hypoxia on the development of the renal vascular system, and possible underlying cellular and ion channel mechanisms involved.

R enal vascular systems play a vital role in the regulation of vascular volume. Recent studies demonstrated that adverse factors during pregnancy could cause poor renal development in the fetus and offspring 1,2 . Hypoxia in utero not only altered renal development in the fetuses 3 , but also induced programming of aberrant kidney development and accelerate the aging process of the kidney during postnatal development 4,5 . Our recent work showed that exposure to hypoxia during pregnancy caused Becn1 signaling-mediated renal autophagy in the rat fetuses 3 . Autophagy is type II programmed cell death, which is different from type I programmed cell death apoptosis. Both of them can cause structural changes in cells [6][7][8] . However, it is unknown whether autophagy or apoptosis may occur in renal vascular systems following chronic hypoxia during early developmental periods. Addressing such questions is very important to further understand the etiology and pathology of renal and vascular diseases in developmental origins. Thus, this study used the prenatal hypoxia (PH) model and determined key molecular elements in the pathways of autophagy and apoptosis in the renal vessels of the offspring exposed to chronic hypoxia during pregnancy.
A number of adverse environmental factors (e.g. nicotine, malnutrition, and cocaine, etc.) during prenatal stage could induce vascular problems and increase susceptibility of cardiovascular diseases [9][10][11][12][13] . In utero hypoxia could contribute to an increased risk of cardiovascular diseases 4 and altered vascular functions 14 . The present study hypothesized that prenatal hypoxia may induce functional and molecular changes in renal blood vessels.
Altered vascular functions could result from either enhanced vasoconstriction or reduced vessel relaxation. For example, the offspring from hypoxic pregnancy had markedly impaired nitric oxide-dependent relaxation in peripheral resistance arteries 15,16 . Voltage-gated Ca 21 channels play a central role in the regulation of intracellular calcium ([Ca 21 ]i) and vascular tone 17 . In the present study, we measured vascular relaxation and [Ca 21 ]i in the renal vessels of the offspring. In addition, functions and expression of L-type (Cav1.2) and T-type calcium channel (Cav3.2) were tested.
Potassium channels play critical roles in the regulation of smooth muscle membrane potential and thereby controlling smooth muscle tone 18 . Large-conductance Ca 21 -activated K 1 channels (BK), one of the important K 1 channels, are dominantly expressed in vascular smooth muscle cells 19,20 . Notably, there has been very limited information regarding influence and underlying mechanism of prenatal hypoxia-affected ion channel activity in renal vessels. The present study also determined functions and expression of BK. Together, the data gained may provide important information on whether and how chronic hypoxia during early developmental stages affects renal blood vessels.

Results
Kidney and body weight. Prenatal hypoxia resulted in a significant decrease in body weight, kidney weight, and the ratio of kidney-tobody weight in the fetuses at gestational day 21 (P , 0.05; Figure 1A). For 5-month-old offspring, there were no significant differences in body and kidney weight as well as the ratio of kidney-to-body weight between the control and PH group (P . 0.05; Figure 1B).
ELISA analysis. The expression of Becn1 was significantly reduced in renal interlobar arteries of the PH offspring at 5 months old, where Caspase3 was increased (P , 0.05; Figure 2B-C). Map1lc3a, Map1lc3b, Sqstm1, Bax, Bcl2, and Caspase8 were not significantly different in renal arteries between the control and PH group (P . 0.05; Figure 2A, C, D).
Transmission electron microscope analysis. Transmission electron microscope analysis showed that significant mitochondrial injury in the hypoxia offspring characterized by mitochondrial swelling and disappearance of mitochondrial cristae in vascular smooth muscle cells of renal interlobar arteries ( Figure 2E).
Blood pressure. Following administration of phenylephrine, mean arterial pressure (MAP) in the PH offspring was significantly higher than that in the control (P , 0.05; Figure 3).
Vascular tension in renal interlobar arteries. There was an increase in phenylephrine-mediated vascular tension in the renal interlobar arteries. Phenylephrine-induced maximum contractions were significantly higher in the PH offspring than that in the control (P , 0.05; Figure 4A). There was a significant difference in response to PE between the control and PH offspring in the absence or presence of NG-Nitrol-arginine (L-Name, nitric oxide-synthase inhibitor) ( Figure 4B). Acetylcholine (ACh)-mediated relaxation was weaker in the PH offspring ( Figure 4C). Vessel integrity was confirmed immediately after the experiments using histologic analysis.
Mibefradil (a blocker for Cav1.2 and Cav3.2 channels) and nifedipine (a specific Cav1.2 channel blocker) depressed PE-induced contractile responses, and the decreased levels in the PH offspring were significantly greater than that in the control ( Figure 5A). Figure 5A showed that PE-mediated vessel tension was reduced to the same levels between the two groups by either mibefradil or nifedipine. Bay K8644 (an agonist for voltage-dependent calcium channel) caused stronger contractions in the renal interlobar arteries in the PH offspring than that in the control ( Figure 5B). In both groups, CTX (a BK channel blocker) potentiated PE-induced contractions; the increased level was greater in the control offspring ( Figure 5C).
[Ca 21 ]i in renal interlobar arteries. Phenylephrine and Bay k8644 mediated-increase of [Ca 21 ]i was significantly higher in renal interlobar arteries of the PH group (P , 0.05; Figure 6).
Whole-cell Ca 21 currents and K 1 currents on vascular smooth muscle cells. The calcium current amplitude was significantly higher in the PH vascular smooth muscle cells (VSMCs) from renal interlobar arteries than that in the control (P , 0.05), and the maximum current density calculated at 120 mV in the PH VSMCs was remarkably increased compared with the control. Incubation with nifedipine depressed Ca 21 current density in both groups, while the decreased level was greater in the PH group ( Figure 7A). Whole-cell K 1 current density tested from 210 to 160 mV in the PH VSMCs was significantly depressed compared with that in the control, the maximum current density calculated at 160 mV was 24.1638 6 1.3818 pA/pF in the control and 18.3982 6 2.6347 pA/pF in the PH (P , 0.05; Figure 7B). Exposure to 100 nmoll 21 iberiotoxin (IBTX, a selective and reversible inhibitor of BK channel) significantly depressed K 1 currents. I-V relationship displayed that whole-cell K 1 current density at 160 mV inhibited by IBTX in the VSMCs was 39.54% in the control, and 23.89% in the PH, respectively ( Figure 7B).
Western blot analysis. Cav1.2a1C was significantly higher in renal interlobar arteries of the PH (P , 0.05; Figure 8A). There was no difference in Cav3.2a1H between the control and PH offspring (P . 0.05; Figure 8B). Western blot analysis showed a significant reduction in the expression of MaxiKa protein in the PH offspring (P , 0.05; Figure 8C).

Discussion
Multiple adverse factors during pregnancy may cause poor renal development in fetuses and influence health of the offspring 1-3 . Hypoxia in utero during the critical developmental periods has been demonstrated to cause programming of aberrant renal development and accelerate aging process of the kidney 4,21 . In the present study, prenatal hypoxia resulted in a decrease in fetal kidney and body weight as well as the ratio of kidney/body weight, indicating that renal development was affected by the hypoxia. However, those weight differences disappeared in the adult offspring, probably due to ''catch-up'' growth as previously reported 22 .
A recent study demonstrated that similar prenatal exposure to hypoxia altered renal development in the rat fetuses and induced Becn1 signaling-mediated renal autophagy in the renal tissue by using multiple approaches, including measuring of protein, mRNA, and immunostaining 3 . In the present study, Becn1 expression was significantly decreased, while caspase3 was increased in the renal blood vessels of the PH offspring, although Map1lc3a, Map1lc3b, Sqstm1, Bax, Bcl2, and caspase8 were not changed. The changes in the expression of Becn1 and caspase3 in the RIA suggest that both autophagy (type II cell death) and apoptosis (type I cell death) may occur in the renal blood vessels following prenatal hypoxia, adding new information on prenatal hypoxia-affected development in kidney vascular systems. Further histological analysis showed significant mitochondrial swelling in vascular smooth muscle cells of the renal arteries in the hypoxia offspring. This histological alteration was associated with molecular changes of Becn1 and caspase3 in the RIA, indicating a possible link between mitochondrial injury and cellular death, and certain molecular and mitochondrial changes in the renal blood vessels following prenatal hypoxia.
Whether renal blood vessels can be functionally influenced by prenatal factors is an important question and the focus of the present     study, since accumulated evidence have suggested a relationship between environmental insults during prenatal periods and increased cardiovascular risks in postnatal life 2,9,11,13 . The present study found PE-mediated pressor responses and vascular tension of RIA were significantly higher in the PH group, as new finding of functional changes in blood pressure and renal vascular responses following prenatal exposure to hypoxia. Increased vascular tension by PE could result from either enhanced vessel contraction or reduced relaxation. Previous studies showed a markedly impaired nitric oxide-dependent relaxation in mesenteric arteries and femoral arteries in the offspring exposed to prenatal hypoxia 15,16 . Nitric oxide mediation of endotheliumdependent relaxation was evaluated using L-Name and acetylcholine (ACh) in the present study. L-Name significantly potentiated PE-mediated vasoconstrictions in both groups; however, the potentiated effect by L-Name was significantly weaker in the PH than that in the control group. Acetylcholine acts on cholinergic receptors in blood vessels, by stimulating the endothelium to produce NO that causes vascular relaxation. The present study demonstrated that ACh-induced relaxation was significantly decreased in the PH offspring. These results indicated that NO functions or pathways in the endothelium of renal interlobar arteries might be injured in the PH offspring, which could cause reduced vascular relaxation. To best of our knowledge, this was the first in demonstration of renal vascular dysfunctions following chronic exposure to adverse environments during pregnancy. Small arteries in the kidney, especially afferent arterioles, play important roles in renal autoregulation 23 . The present  study found functional and molecular changes in the interlobar arteries, which suggested that afferent arterioles in the same kidney could be affected too. This is worth further investigation.
Phenylephrine is a widely accepted vasoconstrictor. One of underlying mechanisms for PE-mediated vasoconstrictions is linked to free intracellular calcium. A rapid increase in [Ca 21 ]i is a major determinant of vascular contraction. Ca 21 binds to calmodulin causing contraction in smooth muscle cells 24 28 . One previous study demonstrated that chronic hypoxia selectively enhanced both L-and T-type voltage-dependent Ca 21 channel activity in the pulmonary artery by up-regulating Cav1.2 and Cav3.2 27 . In the present study, we tested both Cav1.2 and Cav3.2 in renal interlobar arteries of the offspring. The data showed that the pretreatment with either mibefradil or nifedipine could significantly reduce PEincreased vascular tension, and the inhibition by either mibefradil or nifedipine was greater in the PH than that in the control offspring. In addition, the inhibition by mibefradil was stronger than that by nifedipine in both groups. These results suggested that L-type voltage-dependent Ca 21 channels, not T-type, played critical roles in the PE-increased vasoconstrictions via [Ca 21 ]i in the renal blood vessels of the PH offspring. To further determine the possible influence of Ltype or T-type channels, we did patch clamp and protein analysis. The results showed that Cav1.2 currents were significantly increased in the smooth muscle cells, in association with significantly higher expression of Cav1.2a1C in renal arteries of the PH offspring, which may contribute to the increased [Ca 21 ]i and vascular tension by PE in the PH offspring.
Besides calcium channels, K 1 channels also are critical for vascular functions. Potassium channels play roles in regulating VSMC membrane potentials, and thereby controlling vascular tone 18 . In the present study, charybotoxin potentiated-vasoconstriction by PE was altered in the PH offspring compared with the control, indicating K 1 channels in renal interlobar arteries were affected, which could contribute to the increase of the stimulated vascular tension in the kidney of the PH offspring. Furthermore, BK current density in smooth muscle cells from renal arteries was decreased by IBTX in the PH offspring. One of important K 1 channels is BK channels that exist in vascular muscle cell 18,19 . BK channels appear to play a negative feedback role in limiting active vasoconstrictions 29 , as well as in development of hypertension 30 . The expression of BK channels in vascular smooth muscle membranes was increased in hypertension 19 . In the present study, BK functions and BK current density in smooth muscle cells from renal arteries were decreased in the PH offspring. The expression of MaxiKa protein in the renal vessels was also reduced in the PH offspring. This was the first to demonstrate that the prenatal factor could affect BK channels in renal arteries of the offspring. Those data provide new information that prenatal insults such as chronic hypoxia could induce molecular changes in expression of BK protein and alter functions of BK channels in the renal arteries.

Conclusions
The present study demonstrated that prenatal hypoxia had significant influence on PE-mediated blood pressure, histological structures, molecular expression, vascular tissue functions, and ion channel activity in smooth muscle cells and in renal arteries of the offspring. Importantly, PE-stimulated pressor responses and renal vascular tension were increased by prenatal hypoxia, which could be critical contribution to vascular diseases in life later. To determine possible underlying mechanisms, the present study found both vascular abilities in relaxation and constriction were changed following exposure to chronic prenatal hypoxia. Those findings include: NO functions in renal arteries were disabled, PE-increased free intracellular calcium was higher, and the whole-cell K 1 currents, especially BK, were relatively weaker in the renal vascular tissue. Subsequent testing found the probable causes for the observed phenomenon in altered L-type voltage-dependent Ca 21 channel activity, which included the changes in Cav1.2 currents and protein expression, altered BK channel currents and reduced MaxiKa expression in the renal vessels and smooth muscle cells. Together, the data gained provide important information to further understand the long-term influence and mechanisms of chronic hypoxia during early developmental periods on the renal vascular system, which may offer new insight for developing new preventive medicines or approaches targeting at vascular and cellular levels against renal and vascular diseases in fetal origins.

Methods
Animals. Pregnant Sprague-Dawley rats from the Animal Center of Soochow University were randomly divided into the control (21% oxygen) and hypoxia group (10.5% oxygen) from gestational day 4 to 21. Hypoxia was induced by a mixture of nitrogen gas and air in an individual chamber. The control group was housed with room air flowing through chambers. All animals were fed with standard food and tap water. Some of the pregnant rats were sacrificed on day 21 of gestation for fetal studies. Others were allowed to give birth naturally. Pups were kept with their mothers until weaning. The male offspring were tested at 5 months old. Kidneys were collected immediately after sacrifice of animals. Body and kidney weight was determined. All experimental procedures were approved by the Institutional Animal Care Committee and in accordance with the Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85-23, 1996).
ELISA and transmission electron microscope analysis. The contents of Map1lc3a, Map1lc3b, Sqstm1, Becn1, Caspase3, Caspase8, Bax, and Bcl2 in RIA tissue of the offspring were measured with enzyme-linked immunosorbent assay (ELISA) using commercially available kits by Y-Y Biochemical Corporation (Shanghai, China). All data were processed in a blind manner.
Renal arteries of 5 month-old offspring were isolated and fixed in 4% paraformaldehyde at 4uC, and then were prepared and examined using transmission electron microscope (JEOL-1010, Japan) in a blind manner at Nanjing Medical University.
Measurement of blood pressure. Rats were anesthetized with a mixture of ketamine (75 mg/kg) and xylazine (10 mg/kg) (Hengrui Medicine, Jiangsu, China) intraperitoneally. Polyethylene catheters were placed in the femoral artery and threaded to the abdominal aorta 8,11 . After three days of recovery from vascular catheterization, blood pressure was recorded via the artery catheter in conscious and freely moving offspring rats. At testing, phenylephrine (10 mg/kg) was injected via the catheter, blood pressure was recorded for at least 60 minutes using Power Lab system and software (AD Instruments, Australia).
Measurement of intracellular Ca 21 in renal interlobar arteries. In the RIA, [Ca 21 ]i concentrations in vascular tissue were measured using fluorescent Ca 21 indicator, the acetoxymethyl ester of fura-2 (fura2-AM; Calbiochem, San Diego, CA, USA). In brief, RIA ring was loaded in HEPES-PSS solution with fura2-AM (10 mmoll 21 ) for 4 hours at room temperature, gassed continuously with 95% O2-5% CO2. Then rings were incubated in HEPES-PSS solution for 30 minutes at 37uC. [Ca 21 ]i levels were calculated qualitatively by fluorescence ratio of fura-2AM at 340 and 380 nm wave length (Rf340/380). PE (10 24 moll 21 ) and Bay K8644 (10 25 moll 21 ) was added to the chamber respectively, and vessel responses were monitored and recorded continuously.
Electrophysiological measurement. Vascular smooth muscle cells were isolated enzymatically from dissected RIA as reported 11,31 . Whole-cell Ca 21 channel currents and whole cell K 1 currents were recorded in conventional whole-cell configuration, voltage-clamp mode using an Axon Multiclamp 700B with Clampex10.1 and normalized to cell capacitance as picoampere per picofarad (pA/pF). Voltagedependent Ca 21 channel current densities were assessed using standard pulse protocols and a patch-clamp station described previously 11,32 . Negative inward current through Cav1.2 was evaluated using Ba 21 as the charge carrier. T-type Ca 21 current was eliminated by using a holding potential of 260 mV. IBa was evoked using several voltage protocols. For activation of the inward current, Ba 21 current was elicited by 250 ms voltage steps from holding potential of 260 mV to test potentials in the range from 260 to 50 mV with 10 mV increments. Barium chloride (10 mmoll 21 ) as the charge carrier was used to limit current rundown. In a subset of cells, 1 mmoll 21 nifedipine was used to verify the identity of Cav1.2 currents. Only cells with an input resistance .2 GV without substantial run-down were analyzed. Whole cell K 1 currents were elicited by 500 ms voltage steps from holding potential of 270 mV to test potentials in the range from 260 to 60 mV with 10 mV increments. BK currents were tested with IBTX (100 nmoll 21 ).
Western blot. Vascular tissue collected was homogenized. The homogenate containing 100 mg protein was quantified and loaded in each lane for electrophoresis, and then was transferred to immobilon polyvinyldifluoride membranes (Millipore, MA). Membranes were incubated overnight at 4uC with primary antibodies against Cav1.2 (15200), Cav3.2 (15200), or MaxiKa (BK a-subunits, 15500)(Santa Cruz Biotech, Santa Cruz, CA), then incubated with donkey anti-goat antibody or goat anti-rabbit antibody (154,000), respectively, and visualized using chemiluminescence detection (Amersham Biosci., Piscataway, NJ) and UVP imaging system (EC3-Imaging-System, Upland, CA). Imaging signals were digitized and analyzed; the ratio of band intensity to b-actin was obtained for analysis.
Data analysis. Concentration-response curves for vascular responses were analyzed by computer-assisted nonlinear regression (Graph Pad Prism software) to fit the data. Data was expressed as mean 6 SEM. Statistical significance (P , 0.05) was determined by two-way ANOVA or t-test.