Angiotensin type 1a receptors in the median preoptic nucleus support intermittent hypoxia-induced hypertension

Abstract

Chronic intermittent hypoxia (CIH) is a model of the hypoxemia from sleep apnea that causes a sustained increase in blood pressure. Inhibition of the central renin-angiotensin system or FosB in the median preoptic nucleus (MnPO) prevents the sustained hypertensive response to CIH. We tested the hypothesis that angiotensin type 1a (AT1a) receptors in the MnPO, which are upregulated by CIH, contribute to this hypertension. In preliminary experiments, retrograde tract tracing studies showed AT1a receptor expression in MnPO neurons projecting to the paraventricular nucleus. Adult male rats were exposed to 7 days of intermittent hypoxia (cycling between 21% and 10% O₂ every 6 min, 8 h/day during light phase). Seven days of CIH was associated with a FosB-dependent increase in AT1a receptor mRNA without changes in the permeability of the blood-brain barrier in the MnPO. Separate groups of rats were injected in the MnPO with an adeno-associated virus containing short hairpin (sh)RNA against AT1a receptors to test their role in intermittent hypoxia hypertension. Injections of shRNA against AT1a in MnPO blocked the increase in mRNA associated with CIH, prevented the sustained component of the hypertension during normoxia, and reduced circulating advanced oxidation protein products, an indicator of oxidative stress. Rats injected with shRNA against AT1a and exposed to CIH had less FosB staining in MnPO and the rostral ventrolateral medulla after intermittent hypoxia than rats injected with the control vector that were exposed to CIH. Our results indicate AT1a receptors in the MnPO contribute to the sustained blood pressure increase to intermittent hypoxia.

INTRODUCTION

Sleep apnea (SA) is characterized by repeated interruptions in normal respiration during sleep that can lead to cardiovascular disease (16). Diurnal hypertension, which is a sustained increase in mean arterial pressure (MAP) throughout the waking hours, and elevated sympathetic nerve activity (SNA) are associated with SA (16, 39, 58). Chronic increases in sympathetic activity and elevated MAP produce a variety of cardiovascular sequelae (46). To address the sustained hypertension associated with SA, a deeper understanding of the pathophysiological mechanisms is necessary. Chronic intermittent hypoxia (CIH) is an experimental model of the episodic hypoxemia experienced by SA patients (22). This model effectively produces the increased MAP seen in SA patients after only a few days of exposure (16, 19, 47). A number of mechanisms contributing to the sympathoexcitation and the sustained increase in MAP associated with SA have been proposed, including changes in chemoreceptor function (47) and respiratory sympathetic coupling (37). Recent studies in rodents have focused on neural mechanisms that are essential for the sustained component of CIH hypertension that occurs during their active phase when they are breathing room air (12, 50, 54).

Neural mechanisms of CIH-induced hypertension involve the lamina terminalis, which demonstrates increased FosB/ΔFosB staining following 7 days of CIH, along with other regions involved in central autonomic regulation such as the paraventricular nucleus (PVN) of the hypothalamus, the nucleus tractus solitarius (NTS), and rostral ventrolateral medulla (RVLM) of the hindbrain (28). Lesions of the anteroventral region of the third ventricle, which includes the ventral lamina terminalis, block hypertension in a number of animal models (6) and prevent the sustained component of CIH hypertension that occurs during normoxia (12). This same antihypertensive effect can be produced by using a viral vector containing a dominant negative construct against FosB/ΔFosB in the median preoptic nucleus (MnPO) of the lamina terminalis (12). Subsequent studies indicate that the contribution of the lamina terminalis and FosB/ΔFosB to CIH hypertension is related to the renin-angiotensin system (RAS) (29, 50).

The link between the RAS and CIH hypertension was established by Fletcher’s laboratory (20, 23). Circulating angiotensin II (ANG II), a major peptide hormone generated by activation of the RAS, has several systemic effects that could contribute to CIH hypertension (2, 32, 33, 45). In addition, ANG II binds to circumventricular organs, such as the subfornical organ (SFO), and organum vasculosum lamina terminalis (OVLT) (18). Circumventricular organs are highly vascularized, and they lack a blood-brain barrier, allowing circulating peptides to access neurons in these regions (34, 36). The MnPO has an intact blood-brain barrier and is situated between the SFO and OVLT (35). All three of these areas have projections to regions that control autonomic outflow (7, 35) and contain angiotensin receptors (3).

Recent studies have linked the peripheral RAS to activation of the lamina terminalis and the brain RAS to CIH hypertension (29, 50, 54). Virally mediated knockdown of angiotensin type 1a (AT1a) receptors in the SFO reduces the sustained component of CIH hypertension and decreases FosB staining in autonomic regions located inside the blood-brain barrier (50). Chronic intracerebroventricular infusions of the angiotensin receptor antagonist losartan block both the sustained component of CIH hypertension that occurs during the normoxia and reduce the elevated FosB staining in the lamina terminalis (29). Direct PVN infusions of angiotensin antagonists produce similar effects on CIH hypertension (14). Virally mediated AT1a receptor knockdown in the SFO prevents the diurnal hypertension associated with CIH (50). Together, these results indicate that both the peripheral and the brain RAS may play an interactive role in the development of CIH hypertension. Our working hypothesis is that chemoreceptor stimulation during CIH leads to activation of the RAS (54). Circulating ANG II would activate neurons in the SFO, which in turn increases activity in the MnPO and induces FosB expression (50, 54). FosB, which is part of the AP-1 transcription factor family, increases the expression of MnPO AT1a receptors (30, 69) creating a feed-forward loop that sustains the increase in blood pressure during periods when room air is available. Blocking the increase in MnPO AT1a receptors could break this loop, allowing blood pressure to return to baseline during the dark phase of the CIH protocol when they are exposed to room air. We tested this hypothesis using different virally mediated approaches to either inhibit FosB with a dominant negative construct or knockdown AT1a receptor expression using short hairpin (sh)RNA.

MATERIALS AND METHODS

Animal Care

Experiments were performed according to the National Institutes of Health Guide for the Care and Use of Laboratory Animals guidelines and were approved by the University of North Texas Health Science Center Institutional Animal Care and Use Committee. These experiments used 6-wk-old (250–300 g) adult male Sprague-Dawley rats (Charles River Laboratory, Wilmington, MA). Animals were individually housed in temperature-controlled rooms on a 12:12-h light-dark cycle with the light phase lasting from 0700 to 1900. Standard laboratory rat chow and water were available ad libitum except where indicated for experimental protocols. Surgeries were performed using aseptic techniques, and postoperative infection was prevented by subcutaneous administration of procaine penicillin G (30,000 U). The nonsteroidal anti-inflammatory drug carprofen (Rimadyl, 2-mg tablet po in hydrogel) was given before and after surgery for pain management.

Stereotaxic Surgery

Rats were anesthetized with 2% isoflurane, and their scalps were shaved and disinfected with alcohol and iodine. Each rat was placed in a Kopf stereotaxic head frame (David Kopf Instruments, Tujunga, CA). To ensure accurate injections, skulls were leveled between two cranial suture landmarks, bregma and lambda. The injector was angled 8° from medial to lateral, and the injection coordinates used for the MnPO were 0.9 mm lateral, and 6.7 mm ventral from bregma as previously described (12). After a burr hole was drilled at the site of injection, a 30-gauge steel injector was lowered to the MnPO and 200–300 nl of AAV (described below in FosB inhibition) was delivered at a rate of 200 nl/min. The injector was connected to a Hamilton 5-µl syringe (cat. no. 84851; Hamilton Reno, NV) by calibrated polyethylene, which was used to determine the injection volume. The injector remained inserted for 5 min to allow for absorption, and then the injector was slowly withdrawn. Gel foam was packed into the opening in the cranium. Absorbable antibiotic suture was used to close the incision site and minimize postsurgical infection. Rats were allowed 7 days to recover before telemetry instrumentation.

FosB inhibition.

Rats were injected with an adeno-associated virus (AAV) vector containing ΔJunD to test the role of FosB in the regulation of AT1a receptors during CIH. The construct ΔJunD is a truncated form of JunD that functions as a dominant negative to FosB (65, 66). Rats were anesthetized with isoflurane (2%) and injected in the MnPO with an AAV vector (serotype 2 with a CMV promoter; 2.0 × 10⁷ genomic particles/ml) containing ΔJunD and green fluorescent protein (GFP) or a control vector containing only GFP (both provided by the laboratory of E. J. Nestler). After a 2-wk recovery period, the rats were exposed to CIH for 7 days. On the morning of the eighth day, the rats were anesthetized with inactin (100 mg/kg ip; SigmaAldrich, St. Louis, MO) and euthanized as previously described (12). Punches (23-gauge) containing the MnPO were harvested from each brain. A PARIS kit (Ambion, ThermoFisher Scientific, Grand Island, NY) was used to extract RNA from the samples according to the manufacturer’s instructions and used for qRT-PCR analysis.

AT1a receptor knockdown.

Viral vectors were used to locally decrease the expression of AT1a receptors as previously described (50, 61). The recombinant viruses AAV1/2 (with a CMV promoter) that contained either a small hairpin (sh)RNA sequence to match the AT1a receptor (AAV-shAT1a) or a scrambled (AAV-SCR) sequence were obtained from GeneDetect (GeneDetect.com, Auckland, NZ). Viruses were used undiluted at a titer of 1.1 × 10¹² genomic particles/ml. Both viruses expressed GFP to verify accuracy of the injection location.

Drinking Tests with Central ANG II

A separate group of rats was used to test the efficacy of the AT1a receptor knockdown in the MnPO on drinking responses to centrally administered ANG II. For these studies, the rats were injected with AAV-AT1a or AAV-SCR and also received a chronic intracerebroventricular cannula during the same surgery. Fourteen and 18 days after the surgery, they were tested for their drinking responses to ANG II (2 ng/µl icv). Twenty-one days after surgery, the rats were injected with the same dose of ANG II but were not given access to water. Ninety minutes after the ANG II injections, they were anesthetized with inactin (100 mg/kg ip), and their brains were harvested for c-Fos immunohistochemistry.

Intracerebroventricular cannula implantation.

Animals were induced and anesthetized with 2% isoflurane, and their scalps were shaved and disinfected with alcohol and iodine. The rats were placed in a stereotaxic frame (David Kopf Instruments, Tujunga, CA). Each rat received a 200- to 300-nl injection of either AAV-AT1a or AAV-SCR as previously described above. Each rat also was implanted with a chronic intracerebroventricular cannula. After the MnPO injection, a craniotomy was performed at −1.0 posterior; +1.5 lateral to bregma (43). A cannula (Plastics One, Roanoke VA), which extended 4.5 mm below the pedestal, was implanted and secured via Loctite (Henkel, Westlake, OH) and dental cement (CO-ORAL-ITE; Dental Mfg., Diamond Springs, CA) adhering the cannula to three screws mounted in the skull. The cannula was capped with an obturator to obstruct the opening and prevent infection. After cannula implantation, animals were given carprofen and allowed 14 days of recovery before drinking tests.

Drinking tests.

Rats were tested for drinking responses stimulated by centrally administered ANG II (2 ng/µl icv). All tests were conducted during the light phase between 1300 and 1500 h. For each test, the food and water were taken away 90 min before the intracerebroventricular injections. After 60 min, water was returned for 30 min, after which the bottles were weighed to measure water intake. Next, each rat was administered ANG II intracerebroventricularly, and the water bottles were returned for 2 h. At the end of the 2-h drinking test, all of the water bottles were weighed to measure water intake, after which the water and food were returned. All animals received drinking tests 14 and 18 days after injections of the AAV vectors. On day 21 after the stereotaxic surgery, each rat was injected with 2 ng/µl ANG II icv, but they were not given water to drink. They were anesthetized with inactin (100 mg/kg ip) and euthanized 90 min later for c-Fos immunohistochemistry, as described in the following section.

c-Fos staining.

Rats from the drinking behavior experiments were anesthetized with inactin (100 mg/kg ip) and euthanized via transcardiac perfusion with 4% paraformaldehyde (PFA). Brains were divided between hindbrain and forebrain and then placed in PFA for 24 h. After the fixation, the brains were dehydrated in 30% sucrose until they were equalized with sucrose gradient. Brains were serially sectioned on a Leica CM 1950 cryostat at 40 µm. Serial sections were stored in cryprotectant (64) at −20°C. One set of free-floating serial sections was utilized for c-Fos immunohistochemistry. Sections were exposed to peroxide and incubated with a goat primary antibody against c-Fos (sc-52-G, 1:1,000; Santa Cruz Biotechnology, Santa Cruz, CA). After 3 days of incubation at 4°C, the tissue was incubated with a biotinylated anti-goat secondary antibody (Vector Laboratories, Burlingame, CA), which was visualized using 3,3′-diaminobenzidine hydrochloride, as previously described (50). After the c-Fos staining was completed on sections containing the PVN, they were processed with a guinea-pig vasopressin antibody and anti-guinea pig CY3 (Jackson ImmunoResearch Laboratories, West Grove, PA). Stained tissue was coverslipped using Permount and imaged with an epifluorescence microscope (Olympus BX41; Olympus, Center Valley, PA) with a digital camera (Olympus DP70). Cell counting was performed utilizing NIH ImageJ software (v.1.49; National Institutes of Health, Bethesda, MD). Another set of serial sections, containing the MnPO, were directly mounted onto slides and coverslipped using Vectashield medium (Vector Laboratories).

CIH and AT1a Receptor Knockdown in MnPO

In these experiments, all of the rats were injected in the MnPO with either AAV-SCR or AAV-AT1a. One week later, they received a second surgery to implant radio telemetry transmitters. After a 7-day recovery period, the rats were moved to the CIH apparatus for 5 days of baseline recording followed by 7 days of CIH or control. On the morning following the last day of CIH (day 8), the rats were anesthetized and euthanized as described below.

Radio Telemetry Implantation

All rats were implanted with a Data Sciences International (DSI St. Paul, MN) TA11PA-C40 telemetry unit. This telemetry transmitter was used in conjunction with the Dataquest A.R.T. Four receiver and data acquisition system. Before implantation of the telemetry unit, rats were anesthetized with 2% isoflurane, and their abdomens were shaved, cleaned, and sterilized with 100% ethanol and betadine. A midline incision was made in the abdomen from the xyphoid process to the start of the pelvis. The abdominal aorta was located, isolated, and punctured with a 23-gauge needle so that the tip of the telemetry could be inserted. Vetbond (3M, St. Paul, MN) and cellulose were used to close the blood vessel and adhere the telemetry unit to the abdominal aorta. The battery of the telemetry unit was secured to the abdominal muscle via prolene suture, and the incision site was closed with absorbable Vicryl antibiotic suture. Animals were allowed 7 days to recover after telemetry implantation. During the experiments, heart rate (HR), respiration rate (RR), activity (ACT), and mean arterial pressure (MAP) were measured for 10 s every 10 min for 24 h/day and averaged as previously described (28).

Chronic Intermittent Hypoxia

Rats were exposed to 7 days of CIH as previously described (28, 50). All rats were transferred to the room containing the CIH chambers a week after telemetry instrumentations and 2 wk after virus microinjections. Animals that were exposed to CIH were housed in an 8 × 9” cage that was placed inside of custom Plexiglas chambers and were housed in these chambers for a 5-day baseline period before the start of the CIH protocol. CIH occurred in 6-min cycles: 3 min of 21% oxygen room air pumped in, 90 s of nitrogen pumped in to lower the chamber O₂ to 10% oxygen, and then 90 s of maintenance at 10% oxygen. This cycle was repeated 10 times per hour, 8 h a day (0800–1600) for 80 total cycles. Animals were exposed to room air for the remainder of the day. Normoxic controls were housed in the same room under similar conditions, only exposed to room air. At the end of the 7-day protocol, the rats were euthanized on the morning after the last day of CIH after they were anesthetized with inactin (100 mg/kg ip). Blood was drawn from some of the rats after they were anesthetized for measuring advanced oxidation protein products (AOPP) as an index of circulating oxidative stress, as previously described (13, 57).

Immunohistochemistry

FosB immunohistochemistry.

Rats from the CIH experiments were prepared for immunohistochemistry and their brains were processed for FosB staining as previously described (28). After 7 days of CIH or normoxia, rats were anesthetized with inactin (100 mg/kg ip) and perfused with PBS followed by 4% PFA in PBS. One set of forebrain sections containing the MnPO was directly mounted on slides and used to verify the placement of the stereotaxic injections. Radio telemetry data from rats exposed to CIH and injected with AAV-shAT1a outside of the MnPO were used to form a separate control groups, whereas rats from other treatment groups with misplaced injections were not included in the study. Separate sets of forebrain and brain-stem sections were bleached with peroxide and stained for FosB/ΔFosB, using a goat primary antibody (1:1,000; sc-48-G, Santa Cruz Biotechnology). Sections were incubated in the primary antibody for 3 days at 4°C (28). This primary antibody was reacted with a biotinylated anti-goat secondary (Vector Laboratories) and finally visualized using 3,3′-diaminobenzidine hydrochloride. Hindbrain sections were colabeled with a mouse dopamine-β-hydroxylase antibody (1:1,000; Millipore, Billerica, MA), and a CY3-labeled anti-mouse secondary antibody (Jackson ImmunoResearch). Stained sections were imaged using a microscope equipped for epifluorescence (Olympus BX41, Olympus) with a digital camera (Olympus DP70) and counted utilizing NIH ImageJ software (v.1.49). Counts were averaged between sections for each nucleus and subregions of the PVN.

IgG staining.

Forebrains from a separate set of rats exposed to CIH or normoxia were stained for rat IgG to test the integrity of the blood-brain barrier, as previously described (1, 48, 51). Free-floating sections were incubated for 48 h at 4°C in goat anti-rat IgG conjugated with Alexa-Fluor 488 (1:500, ab150157, Abcam). After the sections were rinsed, they were mounted on gelatin-coated slides and coverslipped using antifade mounting medium with DAPI (Vectashield, Vector Laboratories). Sections containing the SFO, MnPO, and OVLT were imaged as described above, using ×100 magnification and a CY2 filter. All images were captured with the same exposure time and converted to an 8-bit grayscale in ImageJ. To calculate integrated density, background signals were eliminated by adjusting the threshold of negative controls for each respective region to be as close to zero as possible in ImageJ. The calculated minimum threshold for each respective region of negative controls was then applied to calculate the integrated density for comparable regions in the CIH-exposed and normoxic control groups. The integrated density calculated for negative controls was then subtracted from that calculated for the CIH-exposed and normoxic control groups to obtain the delta integrated density (Fig. 4). Negative controls were stained according to the immunohistochemistry protocol described above with a GFP anti-mouse primary antibody (1:500, DSHB-GFP3-1F5; Developmental Studies Hybridoma Bank, University of Iowa) and 48 h later stained with the secondary donkey anti-mouse IgG conjugated with Alexa Fluor 488 (1:500, 715-745-151, Jackson ImmunoResearch). Analysis of the integrated density of the Alexa Fluor 488 signal in sections from negative controls, CIH, and normoxia were performed in ImageJ as previously described (63). The same threshold parameter was used for each region of interest and all images.

Fig. 4.

A: virally mediated expression of the ΔJunD dominant negative construct prevents elevation in angiotensin type 1a (AT1a) receptor message vs. controls and chronic intermittent hypoxia (CIH) animals (*P < 0.05). B: representative image of virally induced green fluorescent protein (GFP) expression in dorsal and ventral median preoptic nucleus (MnPO) Hit. ac, anterior commissure. C: the same section from B shown in bright field after removal of dorsal and ventral MnPO by laser capture microdissection. D: injections of shAT1a prevent the increase of AT1a expression after 7 days of CIH. qRT-PCR from laser-captured MnPOs demonstrate ability of shAT1a to prevent an increase in AT1a receptor message after 7 days of CIH vs. CIH scramble (SCR) animals (*P < 0.05). Expression of AT1a in CIH shAT1a was comparable to that in normoxic (N) counterparts (P > 0.05).

In Situ Hybridization

A separate group of rats that were not exposed to CIH were used for RNAscope in situ hybridization for AT1a mRNA. One set of these rats was anesthetized with isoflurane and injected with retrograde tract tracer FluoroGold (100 nl/side; Fluorochrome, Denver, CO) bilaterally in the PVN, using the following coordinates: 1.9 mm caudal, 0.3 mm lateral from bregma, and 7.5 mm ventral from the dura (59).

Reagents used for RNAscope in situ hybridization were purchased from Advanced Cell Diagnostics (Newark, CA) unless otherwise specified. The procedures for in situ hybridization have been previously described (15, 62, 63). Briefly, rats were transcardially perfused with 4% PFA. Brains were quickly extracted, submerged in 4% PFA at 4°C for 24 h, and then transferred into RNase-free 30% sucrose at 4°C. After being equalized with the sucrose gradient, brains were stored at −80°C. Brains were sectioned coronally at 20 µm using a Leica CM 1950 cryostat. Brain sections were rinsed twice in RNase-free PBS and mounted onto Superfrost Plus Gold slides (Thermo Fisher Scientific). The sections were air-dried for ~1 h at room temperature and were then stored at −80°C until they were processed as described below.

In situ hybridization and immunohistochemistry.

To study whether astrocytes in the MnPO express AT1a, brain sections were incubated in hydrogen peroxide for 10 min at room temperature, rinsed twice with RNase-free water, and incubated in Protease Plus for 30 min at room temperature. After two additional rinses with RNase-free water, brain sections were hybridized with AT1a probe (cat. no. 422661), followed by amplification procedures and staining with 3,3′-diaminobenzidine (DAB) as described in the RNAscope 2.5 HD Detection Reagent - BROWN User Manual PART2 (document no. 322310-USM, Advanced Cell Diagnostics).

In situ hybridization and retrograde tract tracing.

The rats that received PVN injections were perfused after 1 wk. The brain sections were incubated in Protease IV for 20 min at room temperature. After two rinses with RNase-free water, brain sections were hybridized with AT1a probe, followed by amplification procedures as described in the RNAscope Fluorescent Multiplex Kit User Manual PART 2 (document no. 320293, Advanced Cell Diagnostics). After staining was completed, slides were coverslipped using ProLong Diamond Antifade Mountant (Life technologies, Carlsbad, CA). Images containing the MnPO were captured with bright-field illumination or epifluorescence with a CY3 filter and Olympus BX41 microscope (Olympus, Center Valley, PA) with a digital camera (Olympus DP70). The densities of the AT1a mRNA signals were measured using ImageJ. Specifically, images were converted to 8-bit grayscale. Background signals were eliminated by adjusting the threshold; threshold was determined using images hybridized with DapB (negative control, cat. no. 320871), and the same threshold parameter was used on all images. The integrated density of reaction product in the MnPO was measured to represent AT1a mRNA expression.

Laser Capture Microdissection and qRT-PCR

On the morning of the eighth day after CIH, animals were anesthetized with inactin (100 mg/kg ip) and euthanized, and their brains were removed. Each brain was immediately frozen in a beaker containing ice-cold isopentane on dry ice. Frozen brains were sectioned at 10 µm using a Leica CM 1950 cryostat, and sections containing the MnPO were transferred to polyethylene naphthalate membrane-coated slides (Arcturus Biosciences, Mountain View, CA). An Arcturus Veritas laser capture microdissection system (13553-00, version-c) was used to identify and laser capture dorsal and ventral MnPO regions that were GFP positive. Captured regions were placed on an Arcturus CapSure Macro LCM Caps (LCM0211). Samples collected from rats used for LCM and PCR with injections that did not include the MnPO were not included in subsequent PCR analysis due to low numbers, although the radio telemetry data from rats exposed to CIH with injections that did not include the MnPO were included in a separate control group. RNA was purified from samples by use of the epicenter ArrayPure Nano-scale RNA Purification Kit, as previously described (41). The purified RNA underwent amplification using the Epicenter TargetAmp 2-Round AminoallylaRNA Amplification Kit 1.0 and was purified using the RNeasy MinElute Cleanup Kit (Madison, WI). Quality and purity of the RNA were verified spectrophotometrically on a Nanodrop 2000c (Thermo Scientific). We considered RNA pure when the 260/280-nm wavelength absorbance ratio was above 1.8. The purified RNA (40 ng) was used for RT-PCR via the Sensiscript Reverse Transcription Kit described in our previous publications (8, 40, 41, 42, 50).

The cDNA produced from reverse transcription was used for quantitative (q)PCR. Individual 15-µl qPCR reactions consisting of 1.8 µl of cDNA, 1.2 µl of a primer mix, 7.5 µl of iQ SYBR Green Supermix, and 4.5 µl of RNase-free water were performed in a 96-well plate. A Bio-Rad iQTM5 iCycler 191 system (Bio-Rad Laboratories, Hercules, CA) and thermocycler were used to perform qPCR. Each reaction was performed in duplicate, and RFU data from the iCycler were analyzed according to the 2^−∆∆CT method in accordance with our previous publications (8, 40, 41, 42). Primers used were the following: S18, Forward 5′-CAGAAGGACGTGAAGGATGG-3′, Reverse 5′-CAGTGGTCTTGGTGTGCTGA-3′; AT1a, Forward 5′-ACTCACAGCAACCCTCCAAG-3′, Reverse 5′-ATCACCACCAAGCTGTTTCC-3′.

Advanced Oxidative Protein Products Assay

Plasma oxidative stress was measured using Cell Biolabs OxiSelect Advanced Oxidative Protein Products (AOPP) assay kit (STA-318) and performed using the manufacturer’s recommended protocol, as previously described (55, 57). The AOPP assay kit provides the ability to measure the concentration of total oxidized protein in a sample (μM) by reacting with chloramine to initiate a color change. Samples were read using a wavelength of 340 nm, and AOPP concentration was calculated using a chloramine standard curve (55).

Statistical Analysis

Baseline data for telemetry recordings that were obtained during the 3–5 days before the start of CIH were analyzed for between-group differences with a two-way mixed-effects ANOVA. The baseline measurements were averaged for each rat, and within-group differences for each condition were analyzed using separate repeated-measures ANOVA followed by a Holm-Sidak post hoc test to compare against baseline conditions. Differences between groups were tested by calculating changes in MAP, HR, RR, and ACT and analyzing the changes by separate two-way mixed-effects ANOVA, as previously described (50). Post hoc analysis on between-group differences was performed using the Student-Newman-Keuls (SNK) test. Data from the mRNA quantification generated from qRT-PCR were analyzed as 2^−ΔΔCT values using a one-way ANOVA with SNK post hoc tests as previously described (41). Cell counts from the FosB studies were analyzed for between-group effects using separate one-way ANOVA followed by SNK post hoc tests. All analyses were performed using SigmaPlot software (SigmaPlot v.12, Systat Software, San Jose, CA). Statistical significance was set at P < 0.05. All data are reported as means ± SE.

RESULTS

AT1a Receptors in the MnPO

Brain sections containing the MnPO from rats (n = 4) with bilateral injections of FluoroGold in the PVN were processed for AT1a in situ hybridization. Forebrain sections from a separate group of rats (n = 5) were for processed for glia fibrillary acid protein (GFAP) immunohistochemistry and AT1a receptor in situ hybridization. In the MnPO of rats injected with FluoroGold, we observed that an average of 20 ± 3% of the neurons retrogradely labeled from the PVN were positive for AT1a receptors. Examples of retrograde labeling from the PVN and AT1a receptor staining are shown in Fig. 1, A–C. In contrast, we observed little to no colocalization of the AT1a signal and GFAP-positive cells (Fig. 1, C and D). These results indicate that AT1a receptors in the MnPO are present on neurons that project to the PVN but not highly expressed in GFAP-positive astrocytes.

Fig. 1.

Expression of angiotensin type 1a (AT1a) receptor in median preoptic nucleus (MnPO) neurons that project to the paraventricular nucleus (PVN). A: digital image of neurons on the dorsal MnPO retrogradely labeled with FluoroGold injected in PVN. B: the same section showing in situ hybridization for the AT1a receptor in the dorsal MnPO. C: digitally merged image of A and B, with high-magnification inset showing retrograde labeling of a neuron that is positive for AT1a receptor. D: representative image showing expression of AT1a mRNA (black punctate dots) and the presence of astrocytes [yellow, glia fibrillary acid protein-positive cells] in MnPO. Digital image of dorsal MnPO with high-magnification inset showing that hybridization signal is not present in astrocytes. E: digital image of colocalization in ventral MnPO. Magenta arrow, astrocyte positive for AT1a mRNA; red arrows, astrocytes with adjacent AT1a mRNA signals; blue arrows, astrocytes that lack AT1a mRNA.

MnPO AT1a Knockdown Reduces Responses to Central ANG II

Intracerebroventricular ANG II.

The injection sites were verified on the basis of GFP staining in the MnPO (Fig. 2A). Rats that did not have GFP staining in the MnPO (Fig. 2B) were used to form a separate control group. Central injections of 2 ng/µl ANG II into the lateral ventricle significantly increased water intake in the scramble injected and shAT1a Miss groups (Fig. 2C). The AT1a Hit group drank significantly less than either two groups on both day 14 and 18 [F(2, 20) = 9.63, P < 0.05; Fig. 2C].

Fig. 2.

Median preoptic nucleus (MnPO) knockdown of angiotensin type 1a (AT1a) receptors and responses to intracerebroventricular ANG II. AAV, adeno-associated virus; sh, short hairpin RNA. A: digital image of green fluorescent protein (GFP) staining in MnPO from AAV-shAT1a injection. B: image from a different subject showing the same vector injected lateral to the MnPO. ac, anterior commissure. C: drinking responses to ANG II (2 ng/1 µl icv) in rats injected with AAV-shAT1a that included the MnPO (open bars, n = 9), AAV-shAT1a injections that missed (shAT1a Miss) MnPO (gray bars, n = 7), and rats injected in MnPO with AAV-shRNA-scrambled (SCR; black bars, n = 7). *Significantly different from all other groups. D: summary of effects of AT1a knockdown in MnPO on cFos staining in the organum vasculosum lamina terminalis (OVLT; n = 5–8), MnPO (n = 7–8), supraoptic nucleus (SON; n = 7–8), paraventricular nucleus (PVN; n = 7–8), and subregions of the PVN (n = 7–8). *Significantly different from all other groups [Student-Newman-Keuls (SNK), P < 0.05; †significantly different from shAT1a Miss group (SNK, P < 0.05)]. E–J: digital images of cFos staining in MnPO (right column) and OVLT (left column) in rats from the AAV-shAT1a group (E and F), AAV-shAT1a Miss group (G and H), and rats injected in MnPO with AAV-shSCR (I and J).

c-Fos.

The effects of AT1a receptor knockdown in the MnPO on c-Fos staining in the OVLT, MnPO, supraoptic nucleus (SON), and PVN in rats with intracerebroventricular injections of ANG II were also examined (Fig. 2D). The AT1a receptor knockdown animals had significantly reduced staining in the MnPO, SON, and PVN (Fig. 2D; all P < 0.05). In contrast, AT1a receptor knockdown did not influence c-Fos staining associated with intracerebroventricular injections of ANG II in the OVLT (Fig. 2D). Examples of c-Fos staining in the MnPO and OVLT are shown in Fig. 2, E–J.

MnPO AT1a Knockdown and CIH Hypertension

The effects of CIH on the blood-brain barrier in the MnPO were tested by staining coronal sections containing the lamina terminalis with anti-rat IgG (48, 51). In normoxic control rats, both the SFO and the OVLT (not shown) demonstrated abundant IgG fluorescence that was two orders of magnitude greater than IgG staining in the MnPO (Fig. 3A). Exposing the rats to 7 days of CIH did not change the calculation of the density of IgG staining in any of the three regions (Fig. 3B; ANOVA P > 0.05). This lack of staining suggests that CIH, or the hypertension associated with it, did not compromise the integrity of the blood-brain barrier in the MnPO.

Fig. 3.

A: digital images of IgG staining in subfornical organ (SFO; left) and median preoptic nucleus (MnPO; right) from unstained controls (bottom), normoxic control (middle), and a rat exposed to chronic intermittent hypoxia (CIH; top). B: calculated integrated density of IgG staining in organum vasculosum lamina terminalis (OVLT), SFO, and MnPO from normoxic control rats and rats exposed to CIH. There were no differences in the density of IgG staining between normoxia (n = 6) and CIH (n = 6) for any of the 3 regions (all unpaired t tests).

To determine the role of MnPO AT1a receptors in CIH hypertension, we first tested the effects of virally mediated dominant negative inhibition of FosB on changes in MnPO AT1a mRNA associated with CIH. Seven days of CIH was associated with a significant increase of AT1a receptor message in the MnPO, and this increase was blocked by virally mediated inhibition of FosB with the dominant negative construct ΔJunD [Fig. 4A; F(2, 11) = 33.72, P < 0.001; SNK, P < 0.001].

Next, we investigated whether this CIH-induced increase in AT1a message could be blocked using an AAV vector approach. Injection sites were identified by GFP fluorescence and collected using laser capture microdissection. Examples of GFP labeling in the MnPO and the same section after LCM can be seen in Fig. 4, B and C. CIH was associated with a significant increase of AT1a receptor mRNA in rats injected with shSCR, and this increase was blocked by MnPO injections of shAT1a [Fig. 4D; F(3,19) = 4.427, P = 0.05; SNK test, P < 0.05]. These results confirm that shAT1a injection in the MnPO blocked the increase in AT1a receptor shRNA associated with CIH.

Both MAP and HR were recorded for 5 days before the start of the CIH protocol (Table 1). There were no significant differences among the treatment groups for either MAP or HR during the light phase or the dark phase of the rats’ diurnal cycle (Table 1). This suggests that AT1a receptors in the MnPO do not significantly contribute to basal MAP or HR.

Table 1.

Average MAPs, HRs, RRs, and ACTs recorded during light phase (0800–1600) and dark phase (1800–0600) during a 5-day baseline period before the 7-day CIH protocol

	shSCR Norm	shSCR CIH	shAT1a Norm	shAT1a CIH Hit	shAT1a CIH Miss
n	8	7	8	7	13
MAP, mmHg
Light baseline	90.8 ± 2.3	91.7 ± 2.3	92.5 ± 2.3	96.9 ± 2.6	91.7 ± 1.8
Dark baseline	99.1 ± 2.4	96 ± 2.4	100.4 ± 2.3	102.1 ± 2.7	97.4 ± 1.9
HR, beats/min
Light baseline	313.8 ± 8	328.5 ± 8.5	326.1 ± 8	344.1 ± 9.5	323.6 ± 7
Dark baseline	386.7 ± 8	391.5 ± 8	386.2 ± 7	391.7 ± 9	382.1 ± 6
RR, beats/min
Light baseline	101.7 ± 2	102.8 ± 2	95.2 ± 3	104.1 ± 3	99.6 ± 2
Dark baseline	98.0 ± 1	102.3 ± 1	99 ± 2	101.5 ±	100.8 ± 1
ACT, counts/min
Light baseline	1.1 ± 0.1	0.95 ± 0.1	1.1 ± 0.2	0.8 ± 0.2	0.9 ± 0.1
Dark baseline	5 ± 0.6	3.6 ± 0.6	4.5 ± 0.7	3.5 ± 0.7	4 ± 0.4

Values are means ± SE; n, number of rats. ACT, activity; HR, heart rate; MAP, mean arterial pressure; RR, respiratory rate; SCR, scramble; sh, short hairpin. Groups are the following: rats injected in median preoptic nucleus with control virus and exposed to normoxia (shSCR + Norm), control virus and chronic intermittent hypoxia (CIH; shSCR + CIH), virus against AT1a and normoxia (shAT1a + Norm), angiotensin type 1a knockdown virus and CIH (shAT1a + CIH Hit), and injections of AT1a knockdown virus not including median preoptic nucleus (shAT1a + CIH Miss). No significant differences exist among any of the groups.

During CIH exposure from 0800 to 1600, the average daily changes in MAP varied significantly among the treatment groups [Fig. 5A; F(4, 44) = 7.684, P < 0.001]. During the normoxic dark phase (1800–0600), significant differences among the treatment groups also were detected for changes in MAP [Fig. 5B, F(4, 44) = 7.594, P < 0.001].

Fig. 5.

Knockdown of angiotensin type 1a (AT1a) receptors in median preoptic nucleus (MnPO) prevents the sustained component of hypertension from chronic intermittent hypoxia (CIH). A: changes in mean arterial pressure (MAP) recording from 0800 to 1600 h during light phase when rats were exposed to CIH or room air. AAV, adeno-associated virus; sh, small hairpin RNA. CIH scramble (CIH SCR) and AAV-AT1a not including MnPO (CIH shAT1a Miss) animals had significantly elevated blood pressure during CIH vs. normoxic (Norm) controls (*P < 0.05). AAV-shAT1a rats exposed to CIH were not significantly different from the other CIH groups. BL, baseline; IH, intermittent hypoxia. B: changes in MAP from the same rats recorded during dark phase from 1800 to 0600. During dark phase, increases in MAP of AAV-shSCR-injected and AAV-shAT1a Miss groups exposed to CIH were significantly greater than the 2 normoxic groups or (*P < 0.05) the CIH AAV-AT1a group (†P < 0.05). Symbols represent the following groups: ●, rats injected in MnPO with AAV-shAT1a and exposed to CIH (CIH shAT1a, n = 7); ○, CIH-treated rats with AAV-shAT1a injections that did not include MnPO (CIH AT1a Miss, n = 13); ▲, CIH-treated rats injected in MnPO with AAV-SCR (CIH SCR, n = 7), △, rats exposed to normoxia injected in MnPO with AAV-AT1a (Norm shAT1a, n = 8); ■, rats exposed to normoxia and injected in MnPO with AAV-SCR (Norm SCR, n = 8). C: average daily increases in MAP in CIH-treated AAV-shSCR injected group (CIH SCR), and CIH treated shAT1a Miss group (CIH AT1a Miss) were significantly greater than in normoxic groups [Student-Newman-Keuls (SNK), †P < 0.05] but not different from CIH-treated rats injected in MnPO with AAV-AT1a (CHI shAT1a). D: changes in MAP during dark phase from 1600 to 0800 h, when all rats were exposed to room air. Increased MAP from CIH SCR and CIH AT1a Miss groups were significantly greater that the other 3 groups (SNK, *P < 0.05). Numbers of rats are the same as in A and B.

The rats exposed to CIH that were injected with the control vector (CIH SCR) and rats with AAV-AT1a that did not include the MnPO (CIH shAT1a Miss) had a significantly greater increase in MAP compared with the two normoxic control groups (Fig. 5C; SNK, P < 0.05). Rats exposed to CIH with AAV-shAT1a injections in the MnPO were not statistically different from any of the other groups (Fig. 5C; SNK P > 0.05).

While similar results were observed for MAP recorded during the dark phase (1800–0600), rats treated with CIH and injected with AAV-shAT1a in the MnPO (CIH shAT1a) had average daily changes in BP that were not significantly different from those of the normoxic controls (Fig. 5D; SNK, P > 0.05). The responses of the CIH shAT1a-treated rats were significantly lower than those of the other two CIH-treated groups (Fig. 5D; P < 0.05). These results indicate that CIH produced sustained diurnal increases in MAP in rats injected with the control vector or the AAV-AT1a injections that missed the MnPO. In contrast, rats injected in the MnPO with AAV-shAT1a did not have a sustained increase in MAP, and their MAP was significantly lower than that of the other CIH groups during the normoxic dark phase (01800–0600).

No significant differences were found overall among any of the groups for changes in HR during CIH [Fig. 6A; F(4, 44) = 1.957, P = 0.12]. A significant treatment by day interactions was observed for changes in HR that occurred during dark phase [Fig. 6B; F(24, 222) = 2.89, P < 0.001]. This was due to a decrease in HR in AAV-shAT1a Miss group that was significantly different from those of the other two groups treated with AAV-shAT1a (Fig. 6B; SNK, P < 0.05). There was also a trend for rats exposed to CIH to have increased RR during the intermittent hypoxia [Fig. 6C; F(4, 44) = 2.3, P = 0.073; SNK]. There were no significant effects of CIH or viral treatments on RR during the dark phase [Fig. 6D; F(4, 44) = 0.360, P = 0.835]. No significant differences were found in activity among the groups during either CIH or the dark phase (P > 0.05; data not shown).

Fig. 6.

Daily changes from baseline in heart rate (HR) and respiratory rate (RR). A: changes in HR recorded from 0800 to 1600 h when the rats were exposed to intermittent hypoxia (IH) during the light phase. There were no differences in HR among the groups. B: changes in HR recording during the normoxic (Norm) dark phase from 1800 to 0600 h. There was a significant difference between chronic intermittent hypoxia (CIH) shAT1a animals and their normoxic counterparts during normoxic dark phase (‡P > 0.05). C and D: changes in RR recorded during 0800–1600 when the rats were exposed to IH (C) and from 1800–0600 during the normoxic dark phase (D). There were no differences among the groups. AT1a, angiotensin type 1a; sh, short hairpin RNA; Miss, not including ventral median preoptic nucleus (MnPO); SCR, scramble. Norm shAT1a, n = 8; CIH shAT1a, n = 7; CIH shAT1a Miss, n = 13; Norm SCR, n = 8; CIH SCR, n = 7. Symbols and abbreviations represent the same groups as described in Fig. 5.

Differential Effects of MnPO AT1a Knockdown on CIH-Induced FosB Staining

Injections of AAV-shAT1a in the MnPO blocked the increase in FosB/ΔFosB staining associated with CIH exposure in a region-specific manner. In the MnPO, CIH AAV-SCR-injected rats had significantly more FosB/ΔFosB than any of the other groups, including rats exposed to CIH after AAV-shAT1a injections [Fig. 7, A–E; F(3,22) = 4.215, P = 0.017; SNK P < 0.05]. In the OVLT, CIH significantly increased FosB/ΔFosB staining, and this effect was not influenced by AT1a knockdown in the MnPO [Fig. 7E; F(3, 19) = 14.810, P < 0.001; SNK P < 0.05]. There were no significant effects on FosB/ΔFosB staining in the SON [Fig. 7E; F(3, 22) = 2.01, P = 0.142]. FosB/ΔFosB staining in the PVN was significantly increased by CIH, and this increase was not affected by AAV-AT1a injections in the MnPO [Fig. 7, F–J; F(3, 22) 15.492, P < 0.001; SNK P < 0.05]. Similar patterns were observed in the PVN subregions (Fig. 7J).

Fig. 7.

A–D: representative examples of FosB/ΔFosB staining in ventral median preoptic nucleus (MnPO). AT1a, angiotensin type 1a; norm, normoxic; sh, short hairpin RNA; CIH, chronic intermittent hypoxia; AAV, adeno-associated virus; SCR, scramble. A: normoxic control injected with AAV-shAT1a. B: CIH and AAV-shAT1a injected. C: normoxic control injected with AAV-SCR. D: CIH and injected with AAV-SCR. E: summary data showing average FosB staining in the organum vasculosum of the lamina terminalis (OVLT), MnPO, and supraoptic nucleus (SON) in normoxic control injected with AAV-shAT1a (Norm shAT1a, open bars; n = 4–6), CIH and AAV-shAT1a injected (CIH shAT1a, light gray bars; n = 4–8), normoxic control injected with AAV-SCR (Norm SCR, dark gray bars; n = 6–7), and CIH and injected with AAV-SCR (CIH SCR, black bars; n = 6–7). No significant effects were observed for the SON. F–I: representative examples of FosB/ΔFosB staining in paraventricular nucleus (PVN). A: normoxic control injected with AAV-shAT1a. G: CIH and AAV-shAT1a injected. H: normoxic control injected with AAV-SCR. I: CIH and injected with AAV-SCR. J: summary of FosB/ΔFosB staining in total PVN and specific subregions in normoxic control injected with AAV-shAT1a (Norm shAT1a, open bars; n = 6), CIH and AAV-shAT1a injected (CIH shAT1a, light gray bars; n = 8), normoxic control injected with AAV-SCR (Norm SCR, dark gray bars; n = 7), and CIH and injected with AAV-SCR (CIH SCR, black bars; n = 7). *Significantly different from both normoxic exposed groups (P < 0.05); †significantly different from all other groups for that area (P < 0.05). Subregions of PVN: PVN-DP, dorsal parvocellular; PVN-MP, medial parvocellular; PVN-VLP, ventrolateral parvocellular; PVN-PM, posterior magnocellular.

In the RVLM, AT1a knockdown in the MnPO attenuated the FosB/ΔFosB staining associated with CIH (Fig. 8, A–D). The numbers of FosB/ΔFosB-positive cells in the RVLM of AAV-SCR CIH animals were significantly greater than in CIH AAV-shAT1a or normoxic animals [Fig. 8I; F(3, 19) = 8.048, P = 0.001; SNK P < 0.05]. Overall, CIH significantly increased FosB/ΔFosB staining in the NTS of rats injected with either AAV-shAT1a or AAV-SCR, and there was no significant difference between the two CIH-treated groups [Fig. 8, E–H; F(3, 19) = 12.321, P < 0.001]. Whereas similar results were observed in the commissural and precommissural NTS, AAV-shAT1a injections in the MnPO were associated with a significant reduction in FosB staining in the subpostremal NTS compared with CIH-treated rats that had been injected with AAV-SCR in MnPO [Fig. 8I; F(3, 19) = 4.579, P = 0.014; SNK P < 0.05]. FosB staining in the CVLM was not different among the groups [Fig. 8I; F(3, 19) = 1.181, P = 0.343].

Fig. 8.

Effects of angiotensin type 1a (AT1a) knockdown in median preoptic nucleus (MnPO) on FosB/ΔFosB staining in hindbrain. Norm, normoxic; sh, short hairpin RNA; CIH, chronic intermittent hypoxia; AAV, adeno-associated virus; SCR, scramble. Representative images of FosB staining in the MnPO (A–D) and subcommissural nucleus tractus solitarius (NTS; E–H). Images are from A and E: normoxic control injected with AAV-shAT1a; B and F: CIH and AAV-shAT1a injected. C and G: normoxic control injected with AAV-SCR. D and H: CIH and injected with AAV-SCR. I: summary data of FosB/ΔFosB staining in NTS: precommissural NTS (NTS-PC), commissural NTS (NTS-C), subpostremal NTS (NTS-SP), caudal ventrolateral medulla (CVLM), and ventrolateral medulla (RVLM) in normoxic control injected with AAV-shAT1a (Norm shAT1a, open bars; n = 6), CIH and AAV-shAT1a injected (CIH shAT1a, light gray bars; n = 8), normoxic control injected with AAV-SCR (Norm SCR, dark gray bars; n = 7), and CIH and injected with AAV-SCR (CIH SCR, black bars; n = 7). *Significantly different from both normoxic groups (P < 0.05); †significantly different from all other groups (P < 0.05).

As previously reported (57), 7 days of CIH was associated with increased circulating AOPP in rats that were not injected with an AAV and only received telemetry surgery and rats that were injected with the control vector in the MnPO [Fig. 9; F(1, 33) = 6.3, P < 0.017]. This increase in AOPP was not observed in rats that were exposed to CIH but were injected with AAV-shAT1a in the MnPO (Fig. 9; AT1a Norm vs. CIH SNK P = 0.949). This indicates that AT1a knockdown in the MnPO significantly reduced the increase in AOPP associated with CIH, although there appeared to be an increase in circulating AOPP related to AAV treatment in the central nervous system (CNS). The data also indicated that stereotaxic injection of either AAV-SCR or AAV-shAT1a significantly increased circulating AOPP independently of CIH (Fig. 9).

Fig. 9.

Effects of angiotensin type 1a (AT1a) knockdown in median preoptic nucleus (MnPO) on circulating advanced oxidation protein products (AOPP) in rats exposed to normoxia (Norm) or chronic intermittent hypoxia (CIH) after radio telemetry surgery alone (Uninj), or radio telemetry surgery followed by MnPO injections with AAV-SCR (shSCR) or adeno-associated virus (AAV)-shAT1a (shAT1a). *Significantly different from normoxic control group within the same surgical condition [Student-Newman-Keuls (SNK), P < 0.05]; +significantly lower than the other 2 normoxic groups (SNK, P < 0.05). Norm Uninjected, n = 5; CIH Uninjected, n = 6; Norm shSCR, n = 7; CIH shSCR, n = 9; Norm shAT1a, n = 6; CIH shAT1a, n = 6.

DISCUSSION

In the MnPO, AT1a receptor mRNA was significantly increased following 7 days of CIH. This same 7-day CIH protocol had previously been shown to increase MnPO staining for the transcription factor FosB (28). The results of the tract-tracing studies showed that MnPO neurons that project to the PVN express AT1a receptor and that in MnPO these receptors are found mostly on neurons. Blocking the transcriptional effects of FosB with dominant negative inhibition in the MnPO prevented the increase in AT1a receptor mRNA associated with CIH. It was previously shown that dominant negative inhibition of FosB in the MnPO prevented the sustained increase in blood pressure associated with CIH that is observed during normoxia without affecting the blood pressure increases that occur during the interment hypoxia (12). In the current study, shRNA knockdown of AT1a receptors in the MnPO was used to test the role of AT1a in CIH hypertension. The results show that the shRNA injection in the MnPO blocked the increase in AT1a receptor messenger RNA expression associated with CIH. Functionally, AT1a receptor knockdown in the MnPO affected CIH hypertension in the same manner as FosB inhibition in our previous study (12). Only the sustained component of the hypertension that occurs during normoxia was blocked by AT1a receptor knockdown in the MnPO. In contrast, the increase in blood pressure that was recorded during the intermittent hypoxia exposures from 0800 to 1600 h was not different from those of rats exposed to CIH and injected with a control vector or shRNA injections that missed the MnPO.

CIH hypertension is inhibited by either intracerebroventricular (29) or PVN (14) infusions of losartan as well as by inhibition of the peripheral RAS (20). In a genetic model of ANG II-dependent hypertension, increased permeability of the blood-brain barrier allows circulating ANG II to access the hypothalamus (5). It is possible that the increase in blood pressure associated with CIH is associated with similar changes in the blood-brain barrier. The results of the IgG staining experiments indicate that this 7-day CIH protocol does not appear to compromise the blood-brain barrier at the level of the MnPO. The model used in the previous study had higher levels of blood pressure and a much longer time course (5). It could be that a longer exposure to CIH or a greater sustained blood pressure increase would result in increased blood-brain barrier permeability. Additional research will be required to address this issue.

Taken together, our results suggest that circulating ANG II acts through a circumventricular organ to contribute to the increase in blood pressure associated with 7 days of CIH. This is supported by a previous study demonstrating that virally mediated knockdown of AT1a receptors in the SFO blocks the sustained component of CIH hypertension (50). Activation of the SFO could result in the activation of the AT1a receptors in the MnPO and the PVN, which may involve angiotensin peptides generated in the CNS. The existence and organization of a brain RAS remains controversial (25, 60), and additional research will be required to address the brain RAS and its relationship to CIH.

Studies using this model of CIH have shown that it is associated with increased circulating markers of oxidative stress and region-specific changes in the expression of pro- and anti-inflammatory cytokines in the CNS (56, 57). Our results suggest that the central injections of the AAV vectors alone were sufficient to significantly increase circulating AOPP compared with rats that had only radio telemetry implantation surgery. Although there did appear to be an effect of the AAV injections, CIH significantly increased AOPP in both uninjected and AAV-shSCR-treated rats compared with their respective normoxic control groups. This CIH effect on AOPP was not seen in the rats injected in the MnPO with AAV-shAT1a. AT1a receptor knockdown in the MnPO reduced circulating AOPP levels indicating a decrease in circulating oxidative stress. These rats were still exposed to intermittent hypoxia, and their blood pressure was significantly increased during this part of the day, although the blood pressure increase was not sustained for the full diurnal cycle. Thus, AT1a receptor knockdown may have reduced circulating AOPP by preventing the sustained increase in blood pressure or reducing sympathetic outflow, which was not directly measured in the current study. Since AT1a knockdown in the MnPO reduced both the sustained increase in blood pressure and circulating AOPP, it can be speculated that the longer term negative effects of both of these factors, i.e., hypertension and oxidative stress, may be initiated by activation of the brain RAS.

The results of the present study suggest that the AT1a receptor is a downstream target of FosB in the MnPO that contributes to CIH hypertension. This is consistent with previous studies that indicated that AP-1 transcription factors can increase AT1a receptor expression (30, 69). In our initial study that used dominant negative inhibition of FosB (12), a PCR array analysis identified other components of the RAS, such as angiotensin-converting enzymes ACE1 and ACE2, as possible downstream targets of FosB but the AT1a receptor was not included in the analysis.

Although the effects of AT1a receptor knockdown and dominant negative inhibition of FosB in the MnPO produced similar effects on CIH hypertension, there were differences in how these manipulations influenced CIH-induced FosB staining. The PVN exhibited a marked increase in FosB expression in response to CIH in both the knockdown animals and their scramble counterparts. The increase in FosB expression has previously been seen in numerous CIH studies (12, 29, 50), but the AT1a receptor knockdown failed to prevent this increase in FosB expression despite blocking of the sustained component of the CIH hypertension. In previous studies of CIH hypertension from our group, CNS treatments that have diminished CIH hypertension have also significantly reduced FosB staining in parvocellular PVN (4, 12, 29, 50). Chronic infusions of AT1 antagonists into the PVN are also capable of preventing hypertension from CIH (14). The PVN has been shown to contribute directly to increase sympathetic nerve activity associated with CIH (53). Given the role of the PVN in CIH hypertension, it is possible that this FosB staining may indicate activation of a subset of PVN neurons not responsible for sympathetic activity. CIH has been shown to sensitize the hypothalamic-pituitary-adrenal axis (31), and this aspect of PVN function was not assessed in the present study. Alternatively, the differences could be related to other functions of the AT1a receptor in this region. Sodium-sensitive MnPO neurons require AT1a receptors for their inhibitory response to γ-aminobutyric acid (GABA), and removing AT1a receptors may remove this brake, allowing for increased stimulation of nonsympathetic neurons in the PVN (26).

There was a significant increase in NTS FosB staining between the rats exposed to CIH and those that were not. Subsection analysis revealed that in the subpostremal NTS the numbers of FosB-positive neurons associated with CIH was significantly decreased in rats injected with AT1a shRNA compared with rats injected with scrambled RNA. This more rostral segment of the NTS is responsible for sympathetic respiratory control. After CIH, numerous laboratories have found respiratory coupled sympathetic outflow is increased at the end of expiration (17, 38, 68). This increase in expiratory sympathetic drive has been postulated to be partially responsible for the sustained hypertension. The ability of the MnPO AT1a knockdown to prevent an increase in subpostremal NTS FosB may indicate some role in this reported change in sympathetic respiratory coupling.

The RVLM is a major source of sympathetic premotor neurons that contribute to blood pressure regulation. AT1a knockdown in the MnPO prevented the increase in FosB staining in the RVLM that was seen in CIH-treated rats injected with the control vector. This result is consistent with the effects of AT1a knockdown in the MnPO on CIH hypertension and suggests that decreased SNA contributed to the reduction in blood pressure through changes in vascular tone and cardiac output.

In response to AT1a knockdown, there was a decrease in FosB expression locally in the MnPO, which indicates AT1a’s role in stimulating FosB expression. AT1 receptors are known to increase in expression or sensitivity in response to stimulation, and Fos proteins play a key role in this response, at least in neurons (9, 11). By eliminating the AT1a receptor, there could be a reduction in both the activation of NADPH oxidase and an activation of cAMP response element-binding protein, which may work independently, or in concert, leading to AP-1 dependent upregulation of the AT1a receptor (10, 69). A breakdown in either the angiotensin signaling or the transcriptional activation is sufficient to prevent the blood pressure increase from CIH. The nature of the cellular mechanisms that link FosB/ΔFosB to the AT1a receptor in the MnPO requires further investigation.

This study highlights the critical role that AT1a receptors in the MnPO play in the sustained hypertension from CIH. Angiotensin receptors are located in several regions that also contribute to hypertension related to CIH or SA in humans. For example, a recent study has demonstrated that angiotensin receptors contribute to sustained increases in muscle sympathetic nerve activity produced by acute intermittent hypoxia caused by voluntary apnea/breath holds in humans (27). This helps to solidify the role of central RAS in the pathogenesis of hypertension from CIH and could provide future avenues of treatment that are more efficacious and less palliative.

Perspectives and Significance

The pathophysiology of SA is poorly understood, in part due to the fact that sufferers cannot self-identify. Whereas the disease progresses silently, the sequelae become indistinguishable from the inciting pathophysiology. The 7-day model of CIH allows for study of the initiating factors in the pathogenesis of neurogenic hypertension. RAS activation both peripherally and centrally has been shown to be necessary for CIH hypertension (12, 19, 20, 24, 50). Our study has demonstrated the role MnPO AT1a receptors play in the onset of neurogenic hypertension from CIH. Treatment with AT1 receptor blockers that can cross the blood-brain barrier may provide relief from hypertension as well as stymie the progression of the neuropathology in the forebrain. There are numerous other pathophysiological processes occurring, such as sensory long-term facilitation in the chemoreceptors and weakened ATP-sensitive K⁺ channel responses to hypoxia that also likely contribute to CIH hypertension (44, 67). To reliably treat hypertension from advanced sleep apnea, it may be necessary to target multiple mechanisms. More studies are needed to determine whether a multitiered approach is necessary to target dysfunction in the forebrain separately from the chemoreceptors and hindbrain.

GRANTS

This work was supported by National Institutes of Health Grants P01 HL-088052 and T32 AG-020494.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

B.S., G.E.F., T.P.N., L.A.W., R.L.C., and J.T.C. conceived and designed research; B.S., T.P.N., L.A.W., A.B.M., and B.D.S. performed experiments; B.S., G.E.F., T.P.N., L.A.W., A.B.M., B.D.S., R.L.C., and J.T.C. analyzed data; B.S., G.E.F., T.P.N., L.A.W., A.B.M., B.D.S., R.L.C., and J.T.C. interpreted results of experiments; B.S., L.A.W., A.B.M., B.D.S., and J.T.C. prepared figures; B.S. drafted manuscript; B.S., G.E.F., T.P.N., L.A.W., A.B.M., B.D.S., R.L.C., and J.T.C. edited and revised manuscript; B.S., G.E.F., T.P.N., L.A.W., A.B.M., B.D.S., R.L.C., and J.T.C. approved final version of manuscript.