Abstract
Hydrogen sulfide (H2S) is widely distributed throughout the nervous system with various antioxidant and anti-inflammatory properties. Hypertension involves an increase in reactive oxygen species (ROS) and inflammation in the hypothalamic paraventricular nucleus (PVN). However, it is unclear how H2S in PVN affects hypertension.
Our study used spontaneously hypertensive rats (SHR) and control Wistar Kyoto (WKY) rats, microinjected with adenovirus-associated virus (AAV)-CBS (cystathionine beta-synthase overexpression) or AAV-ZsGreen in bilateral PVN, or simultaneously injected with virus-carrying nuclear factor erythroid 2-related factor 2 (Nrf2)-shRNA for 4 weeks. Blood pressure (BP) and plasma noradrenaline level were detected, and the PVN was collected. Finally, levels of CBS, H2S, Nrf2, Fra-LI, ROS, gp91phox, p47phox, superoxide dismutase 1, interleukin (IL)-1β, IL-6, IL-10, tumor necrosis factor-α, tyrosine hydroxylase, and glutamate decarboxylase 67 were measured.
We found that AAV-CBS increased H2S in the PVN, and BP, neuronal activation, oxidative stress, and inflammation of PVN were substantially reduced. Furthermore, endogenous H2S in the PVN activated Nrf2 and corrected the PVN’s imbalance of excitatory and inhibitory neurotransmitters. However, Nrf2 knockdown in the PVN was similarly observed to abolish the beneficial effect of H2S on hypertension.
The findings imply that endogenous H2S in SHR PVN is reduced, and PVN endogenous H2S can alleviate hypertension via Nrf2-mediated antioxidant and anti-inflammatory effects.
The central nervous system’s (CNS) regulation is critical in the development and progression of hypertension. Hypertension is caused by abnormal autonomic mechanisms characterized by sympathetic nervous system hyperactivity and defective cardiac vagal control. The hypothalamic paraventricular nucleus (PVN) regulates respiration, blood pressure (BP), and cardiovascular activity as integrative elements of the neuroendocrine and autonomic nervous systems.1–3 Other research teams and we have previously demonstrated that reactive oxygen species (ROS) buildup and an unbalanced ratio of anti- to pro-inflammatory cytokines (PICs) in the PVN were associated with high BP in hypertensive rats.4–6 Treatment with antioxidants or anti-inflammatory drugs in PVN has been proven effective in improving BP. Therefore, reducing the amounts of PICs, ROS, and NADPH oxidase in the PVN may help lower BP. In addition, the level of neurotransmitter imbalance in the PVN substantially contributes to sympathetic output in various hypertension models.5
Hydrogen sulfide (H2S) is a gas transmitter created by 3 enzymes in the cytoplasm or mitochondria: homocysteine (Hcy), l-cysteine (Cys), and β-mercaptopyruvate. Cystathionine beta-synthase (CBS) is the primary enzyme that produces H2S in the CNS. H2S effectively reduces inflammation7,8 and acts as an antioxidant.9
Numerous studies have implicated decreased endogenous H2S production or an enzyme deficiency in the development of hypertension. Spontaneously hypertensive rats (SHR) showed a decrease in circulating H2S levels; early H2S, d-, or l-cysteine administration could lower BP.10 Most notably, H2S can influence the CNS. In SHR, NaHS microinjections into the rostroventrolateral medulla (RVLM) reduced ROS and lowered BP.11 Chronic NaHS intracerebroventricular infusion can lower BP, enhance autonomic activity, and impact PVN microglia.12 Our group discovered that chronic infusion of GYY4137, an H2S sustained-release donor, in bilateral PVN for 4 weeks, may lower sympathetic activity and the hypertensive response, in part because ROS and PICs levels inside the PVN are reduced in high salt-induced hypertension.13
Nuclear factor erythroid 2-related factor 2 (Nrf2) promotes the transcription of cell-related protective genes by combining with antioxidant response elements, and it is crucial for maintaining cellular antioxidants and preventing diseases associated with them. For example, studies have shown that curcumin can improve hypertension by inducing Nrf2 into the nucleus.14 Resveratrol can reduce the renal inflammatory response in SHR by activating the Nrf2 pathway.15 Recent research has demonstrated that a particular Nrf2 gene deletion in the RVLM can reduce the expression of antioxidant enzymes, boost sympathetic nerve activity (SNA), and raise BP.16 According to our group’s research, oxidative stress and inflammation in PVN can be reduced by activating the Nrf2 pathway, which lowers BP.17,18
This study aimed to elucidate the specific mechanism by which H2S activation of the Nrf2 pathway in the PVN reduces oxidative stress and inflammatory responses during hypertension and subsequently attenuates sympathetic activity and BP.
MATERIALS AND METHODS
Animals
This experiment complied with the NIH Guide for the Care and Use of Laboratory Animals and was approved by the Xi’an Jiaotong University Ethics Committee of Laboratory Animals. Charles River Laboratories provided SHR and Wistar Kyoto (WKY) male rats weighing 220–250 g. Rats were kept in a room with a controlled temperature (23 ± 2 °C) and lighting (12/12-hour light–dark cycle) along with a regular diet and tap water.
Adenovirus-associated virus preparation
Experiment 1: Hanbio Biotechnology (Shanghai, China) provided the control vectors (HBAAV2/9-ZsGreen) and the plasmid vector (pHBAAV-CMV-MCS-3flag-T2A-ZsGreen) carrying the mRNA of Rattus norvegicus CBS targeting sequence (Transcript: NM 012522.2).
Experiment 2: Hanbio Biotechnology provided the plasmid vector (pHBAAV-U6-MCS-CMV-EGFP) containing the Nrf2-small hairpin RNA (AAV-Nrf2 shRNA) and control vectors (AAV-EGFP NC).
The adenovirus-associated virus (AAV) had a titer of 1 × 1012 µg/ml, and was subpackaged (20 µl/tube) and stored at −80 °C. Furthermore, the vectors needed to be mixed 1:1 before injection in experiment 2.
AAV injection
AAV was infused into the hypothalamic PVN, as previously mentioned.19 Briefly, rats were fixed with the head on the brain stereotaxic equipment after intraperitoneal ketamine (80 mg/kg) and xylazine (10 mg/kg) anesthesia. Next, AAV was contained in a 5-µl microinjector attached to an infusion pump. Following that, 1 µl (experiment 1) or 2 µl (experiment 2) of AAV were infused bilaterally into the PVN (the location referenced to Paxinos and Watson rat brain atlases the Paxinos and Watson rat brain atlas) within 10 minutes. After the injection, the microinjector was left for 10 minutes before removal.
General experimental protocol
Two weeks were spent acclimating the rats to their surroundings prior to the start of the experiment. Then, all groups of rats were randomly assigned (n = 10 for each group): experiment 1: (i) WKY + AAV-ZsGreen; (ii) WKY + AAV-CBS; (iii) SHR + AAV-ZsGreen; (iv) SHR + AAV-CBS; experiment 2: (i) SHR + AAV-ZsGreen + NC; (ii) SHR + AAV-ZsGreen + Nrf2 shRNA; (iii) SHR + AAV-CBS + Nrf2 shRNA; (iv) SHR + AAV-CBS + NC. Four weeks after the PVN injection, the trial came to an end. Plasma samples were obtained as described previously, and PVN was isolated from fresh rat brain tissue. Another portion of fresh rat brain tissue was stored in 4% paraformaldehyde for 3 days, dehydrated in 30% sucrose, and finally embedded in OCT. To prepare samples for subsequent analysis, they were all stored at −80 °C.
BP measurements
We measured BP every 4 days between 8:00 and 11:00 am by tail-cuff plethysmography as previously described.3 Additionally, anesthesia-induced rats’ left carotid arteries were intubated with polyethylene catheters after the experiment to assess mean arterial blood pressure (MAP). Within 30 minutes, MAP and heart rate (HR) data were gathered and averaged.
H2S level in the PVN
PVN was isolated from rat brains, and the H2S level was determined using an H2S concentration determining kit (Solarbio Science & Technology, Beijing, China) as per the instruction.
Immunofluorescence and immunohistochemistry
In a frozen microtome, OCT-embedded brain tissue was cut with a thickness of 18 um from bregma −0.92 to −2.12 mm (Leica, CM1860). As previously mentioned, PVN immunohistochemical and immunofluorescence staining was carried out.3 The primary antibodies used in the study consisted of the following: rabbit anti-Nrf2 (Abcam, 1:400 dilution), mouse anti-CBS (Santa Cruz, 1:100 dilution), rabbit anti-Fra-LI (Santa Cruz, 1:50 dilution), rabbit anti-gp91phox (Santa Cruz, 1:100 dilution), mouse anti-p47phox (Santa Cruz, 1:20 dilution), mouse anti-IL-10 (Santa Cruz, 1:100 dilution), rabbit anti-GAD67 (Santa Cruz, 1:20 dilution), rabbit anti-TH (Millipore Sigma, 1:300 dilution), rabbit anti-IL-1β (Bioss, 1:200 dilution), and rabbit anti-HO-1 (Bioss, 1:100 dilution). In immunohistochemistry, slices were treated for 1 hour in a blocking solution with a secondary antibody from Abcam. Next, we detected the horseradish peroxidase response using the 3,3-diaminobenzidine (DAB) kit (Beyotime) to detect the horseradish peroxidase reaction. Finally, the microscope captured sections in photographs (Nikon Eclipse, 80i, Japan).
Dihydroethidium staining
In order to determine whether superoxide was forming in the PVN, dihydroethidium (DHE) staining was used, and the procedure was carried out as previously described.20 Briefly, dihydroethidium (0.05 mM) was incubated with the slices for 30 minutes at 37 °C. After 3 phosphate buffered saline (0.01 M) rinses, sections were imaged using a Nikon fluorescence microscope.
Western blotting
As previously mentioned, the western blotting methodology was followed.3 First, ultrasound was used to fracture the rat PVN, and the associated tissue proteins were retrieved. Next, the Pierce BCA protein assay kit (Waltham) measured the protein concentration in PVN tissues. Next, 20 µg of protein from each group was separated using sodium dodecyl sulfate polyacrylamide gel electrophoresis gels, and the associated proteins were subsequently transferred to a polyvinylidene fluoride (PVDF) membrane using the wet transfer method. Next, skimmed milk (5%) was used to seal the membranes for 2 hours at room temperature before they were treated with primary antibodies overnight at 4 °C. The primary antibodies used in the study consisted of the following: rabbit anti-Nrf2 (Abcam, 1:2,000 dilution), rabbit anti-SOD1 (Abcam, 1:1,000 dilution), mouse anti-IL-6 (Abcam, 1:1,000 dilution), rabbit anti-HO-1 (Abcam, 1:1,000 dilution), mouse anti-CBS (Santa Cruz, 1:500 dilution), rabbit anti-gp91phox (Santa Cruz, 1:500 dilution), mouse anti-p47phox (Santa Cruz, 1:200 dilution), mouse anti-tumor necrosis factor (TNF)-α (Santa Cruz, 1:500 dilution), mouse anti-IL-10 (Santa Cruz, 1:300 dilution), mouse anti-TH (Santa Cruz, 1:1,000 dilution), rabbit anti-GAD67 (Santa Cruz, 1:500 dilution), and rabbit anti-IL-1β (Bioss, 1:1,000 dilution). The next day, membranes were incubated at room temperature for 2 hours with goat anti-rabbit IgG Horse Radish Peroxidase (HRP)-linked antibody from Abcam or goat anti-mouse HRP-linked antibody from Abcam. The internal control used in this study was β-actin. After that, a chemiluminescence reagent was added, and protein content was detected.
ELISA
According to the manufacturer’s instructions, commercial ELISA kits (Abnova, Taiwan) were used to test the levels of norepinephrine (NE), interleukin (IL)-1β, and IL-6 in plasma.
Statistical analysis
Data from the group were presented as mean ± SEM, with a probability (P) value <0.05 denoting significance. The systolic blood pressure (SBP) values were analyzed using repeated-measures ANOVA. Two-way ANOVA analyzed other parameters in experiment 1 with Tukey’s post hoc test. One-way ANOVA analyzed other parameters in experiment 2. The data were analyzed and graphics were produced using GraphPad Prism (Version 7.0; La Jolla, CA).
RESULTS
AAV-CBS increased CBS expression and H2S levels in the PVN
Figure 1a shows that after AAV injection into the PVN, ZsGreen was strongly expressed in the PVN but not in the supraoptic nucleus or the subfornical organ. Additionally, SHR showed decreased CBS-positive neurons (Figure 1b, P < 0.0001), CBS protein level (Figure 1c, P < 0.0001), and H2S level (Figure 1d, P < 0.0001) in their PVN compared with control rats. Administration of AAV-CBS to PVN could significantly increase PVN expression of CBS (P < 0.0001) and H2S levels (P < 0.0001) in SHR.
Effects of AAV-CBS in the PVN on CBS expression, H2S level, blood pressure, heart rate, PVN Fra-LI expression and plasma NE. (a) Green fluorescent protein ZsGreen fluorescence images of the PVN, supraoptic nucleus (SON), and subfornical organ (SFO) as examples. (b) Representative immunofluorescence images at ×40 magnification showing double staining of CBS (red) and nucleus (blue) in the PVN; and the summary data for the CBS-positive neurons in 4 groups. (c) The representative immunoblot images for CBS in the PVN; and quantification of western blotting images for CBS in 4 groups. (d) Measurement of H2S level in the PVN in 4 groups. (e) Time course of systolic blood pressure (SBP) in 4 groups. (f) Mean arterial blood pressure (MAP) was assessed at the end of the experiment. (g) Measurement of heart rate in 4 groups. (h) Representative immunohistochemistry images at ×10 magnification showing staining of Fra-LI (brown) in the PVN; and the summary data for Fra-LI in the PVN in 4 groups. (i) Measurement of NE level in plasma in 4 groups. 3 V, the third ventricle. n = 6–8/group. Values are mean ± SEM. In SBP data: #P< 0.05 vs. WKY groups (WKY + AAV-ZsGreen or WKY + AAV-CBS); εP < 0.05 vs. SHR + AAV-ZsGreen. In other parameters: ****P < 0.0001. Abbreviations: AAV, adenovirus-associated virus; CBS, cystathionine beta-synthase; H2S, hydrogen sulfide; NE, norepinephrine; PVN, paraventricular nucleus; SHR, spontaneously hypertensive rats; WKY, Wistar Kyoto.
Endogenous H2S in the PVN lowered BP of SHR
In our investigation, SBP was measured using a tail-cuff plethysmograph. As shown in Figure 1e, SHR significantly increased BP compared with WKY rats (SHR + AAV-ZsGreen: 186 ± 3 mm Hg vs. WKY + AAV-ZsGreen: 125 ± 4 mm Hg; P < 0.05). However, from day 12, the BP in the SHR + AAV-CBS rats gradually dropped and stayed lower than in the SHR + AAV-ZsGreen group by the end of our study (SHR + AAV-CBS: 155 ± 5 mm Hg vs. SHR + AAV-ZsGreen: 186 ± 3 mm Hg; P < 0.05). These SBP data were supported by MAP records (Figure 1f, P < 0.0001). Even though endogenous H2S in PVN decreased SHR’s HR (Figure 1g), the impact was insignificant (P = 0.11). Additionally, we discovered that H2S had no discernible impact on WKY rats’ BP and HR.
Endogenous H2S in the PVN decreased the expression of Fra-LI and plasma levels of NE of SHR
Plasma NE levels were an indirect predictor of SNA, and additionally, we examined Fra-like immunoreactivity (Fra-LI) in the PVN to evaluate the impact of endogenous H2S on neuronal activity. We found that both Fra-LI expression in the PVN (Figure 1h, P < 0.0001) and plasma NE levels (Figure 1i, P < 0.0001) were much higher in SHR than those in WKY rats. Furthermore, increased endogenous H2S in the PVN significantly reduced PVN Fra-LI expression (P < 0.0001) and plasma NE (P < 0.0001) concentration of SHR. Additionally, we discovered that in WKY rats, H2S had no discernible impact on Fra-LI expression or plasma NE levels.
Endogenous H2S in the PVN increased Nrf2 expression of SHR
According to immunofluorescence and western blotting analyses, compared with the WKY rats, Nrf2-positive neurons (Figure 2a, P < 0.0001) and protein level (Figure 2b, P < 0.0001) in the PVN of SHR were lower, and Nrf2 expression was increased by an increase in endogenous H2S in SHR but not in WKY rats (Figure 2a, P < 0.0001; Figure 2b, P < 0.01).
Effects of endogenous H2S in the PVN on Nrf2 expression and oxidative stress. (a) Representative immunofluorescence images at ×20 magnification showing double staining of Nrf2 (red) and nucleus (blue) in the PVN; and the summary data for the Nrf2 in the PVN in 4 groups. (b) The representative immunoblot images for Nrf2; and quantification of western blotting images for Nrf2 in the PVN in 4 groups. (c) An illustration of DHE staining in the PVN at ×10 magnification; and the data of DHE staining in different groups. (d) Representative immunofluorescence images at ×20 magnification showing double staining of gp91phox-positive neurons (red) and nucleus (blue) in the PVN; and the summary data for the gp91phox in the PVN in 4 groups. (e) Representative immunofluorescence images at ×20 magnification showing double staining of p47phox-positive neurons (red) and nucleus (blue) in the PVN; and the summary data for the p47phox in the PVN in 4 groups. (f) The representative immunoblot images for gp91phox, p47phox, and SOD1; and quantification of western blotting images for gp91phox, p47phox, and SOD1 in the PVN in 4 groups. 3 V, the third ventricle. n = 6–8/group. Values are mean ± SEM. **P < 0.01; ***P < 0.001; ****P < 0.0001. Abbreviations: DHE, dihydroethidium; H2S, hydrogen sulfide; PVN, paraventricular nucleus.
Endogenous H2S in the PVN attenuated oxidative stress of SHR
We first detected ROS changes in the PVN using dihydroethidium staining. Figure 2c shows that endogenous H2S in the PVN significantly reduced the increase in ROS in SHR rats but not in WKY rats (P < 0.0001).
The level of gp91phox and p47phox has been measured in numerous earlier investigations to represent NADPH oxidase (NOX) activity.18 Results revealed that compared with the control rats, the number of gp91phox (Figure 2d, P < 0.0001) and p47phox-positive neurons (Figure 2e, P < 0.0001), as well as their protein levels (Figure 2f, P < 0.0001) in the PVN of SHR rats, were significantly higher. However, after the microinjection of AAV-CBS into PVN bilaterally, the number of gp91phox- and p47phox-positive neurons and their protein levels was decreased.
Superoxide dismutase (SOD) is an enzyme that can catalyze the disproportionation of superoxide anion radicals O2 and H2O2 in living things. SOD is essential for maintaining the proper balance of oxidation and antioxidants in the body.19 SOD1 mainly exists in the cytoplasm of eukaryotic cells and is considered the most widely distributed among primitive biological groups. SHR rats had lower SOD1 expression in the PVN than WKY rats, as shown in Figure 2f (P < 0.001). However, endogenous H2S in the PVN increased SOD1 expression in SHR.
Endogenous H2S in the PVN reduced PICs of SHR
Increased neuroinflammation in autonomic brain regions is strongly associated with hypertension. Compared with the WKY + AAV-ZsGreen group, significantly increased numbers of IL-1β-positive neurons (Figure 3a, P < 0.0001) and decreased numbers of IL-10-positive neurons (Figure 3b, P < 0.0001) were observed in the PVN of the SHR + AAV-ZsGreen group. Endogenous H2S in the PVN significantly reduced the number of IL-1β-positive neurons in SHR but not WKY rats (P < 0.0001) and increased the number of IL-10-positive neurons (P < 0.001). In addition, western blotting results revealed that the protein levels of TNF-α (P < 0.0001), IL-1β (P < 0.0001), and IL-6 (P < 0.0001) in the PVN of SHR were significantly higher than those of WKY rats, while the protein level of IL-10 (P < 0.0001) was significantly lower than that of WKY rats (Figure 3c). Endogenous H2S in the PVN significantly increased the level of IL-10 in SHR (P < 0.05) and lowered the level of TNF-α (P < 0.0001), IL-1β (P < 0.05), and IL-6 (P < 0.0001), but not in WKY rats. Additionally, we used an ELISA technique to measure IL-1β and IL-6 expression to ascertain the impact of endogenous H2S on plasma levels of PICs. Plasma levels of IL-1β (Figure 3d, P < 0.0001) and IL-6 (Figure 3e, P < 0.0001) were higher in the SHR + AAV-ZsGreen group than in the WKY + AAV-ZsGreen group, and these changes were significantly reduced in the group that received AAV-CBS PVN microinjections (P < 0.0001). Furthermore, we discovered that H2S had no discernible effect on the plasma level of PICs in WKY rats.
Effects of endogenous H2S in the PVN on inflammation and neurotransmitter. (a) Representative immunofluorescence images at ×20 magnification showing double staining of IL-1β (red) and nucleus (blue) in the PVN; and the summary data for the IL-1β in the PVN in 4 groups. (b) Representative immunofluorescence images at ×40 magnification showing double staining of IL-10 (red) and nucleus (blue) in the PVN; and the summary data for the IL-10 in the PVN in 4 groups. (c) The representative immunoblot images for TNF-α, IL-1β, IL-6, and IL-10; and quantification of western blotting images for TNF-α, IL-1β, IL-6, and IL-10 in the PVN in different groups. (d) Measurement of IL-1β levels in plasma in different groups. (e) Measurement of IL-6 levels in plasma in different groups. (f) Representative immunohistochemistry images at ×40 magnification showing staining of GAD67 (brown) in the PVN; and the summary data for the GAD67 in the PVN in 4 groups. (g) Representative immunofluorescence images at ×20 magnification showing double staining of TH (red) and nucleus (blue) in the PVN; and the summary data for the TH in the PVN in 4 groups. (h) The representative immunoblots images for GAD67 and TH; and quantification of western blotting images for GAD67, and TH in the PVN in different groups. 3 V, the third ventricle. n = 6–8/group. Values are mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Abbreviations: H2S, hydrogen sulfide; PVN, paraventricular nucleus; TH, tyrosine hydroxylase.
Endogenous H2S in the PVN restored the neurotransmitter imbalance of SHR
In the PVN, gamma amino butyric acid (GABA) is a significant inhibitory neurotransmitter, and NE is a significant excitatory neurotransmitter. They are closely related to how SNA is controlled in hypertension. Tyrosine hydroxylase (TH) is a crucial enzyme in the production of catecholamines. The enzyme glutamate decarboxylase (GAD) 67 is essential for the synthesis of GABA. Therefore, the levels of NE and GABA may be indirectly reflected in the expression of TH and GAD67.
Immunofluorescence and immunohistochemistry results indicated that compared with the WKY group, SHR had significantly more TH-positive neurons (Figure 3g, P < 0.0001) and significantly fewer GAD67-positive neurons (Figure 3f, P < 0.0001) in the PVN. Furthermore, endogenous H2S in the PVN markedly enhanced GAD67 expression (P < 0.01) while decreasing TH expression (P < 0.0001). These data were supported by the results of western blotting results (Figure 3h). Additionally, we discovered that in WKY rats, H2S had no discernible impact on TH or GAD67.
Microinjection of Nrf2 shRNA into PVN decreased the expression of Nrf2
Next, we measured Nrf2 expression levels after injecting AAV-Nrf2 shRNA into PVN to assess the consequences of Nrf2 silencing. Western blotting results revealed that SHR treated with AAV-Nrf2 shRNA had considerably lower levels of Nrf2 protein in their PVN than SHR treated with AAV-EGFP NC (Figure 4a, P < 0.0001).
Effects of AAV-Nrf2 shRNA in the PVN on Nrf2 expression, blood pressure, HO-1 expression, and oxidative stress. (a) The representative immunoblot image for Nrf2; and quantification of western blotting images for Nrf2 in the PVN in 4 groups. (b) Time course of SBP in 4 groups. (c) Representative immunofluorescence images at ×20 magnification showing double staining of HO-1 (red) and nucleus (blue) in the PVN; and the summary data for the HO-1 in the PVN in 4 groups. (d) The representative immunoblot images for HO-1; and quantification of western blotting images for HO-1 in the PVN in different groups. (e) An illustration of DHE staining in the PVN at ×10 magnification; and the data of DHE staining in different groups. (f) Representative immunofluorescence images at ×40 magnification showing staining of gp91phox (red) in the PVN; and the summary data for the gp91phox in the PVN in 4 groups. (g) The representative immunoblot images for gp91phox, p47phox, and SOD1; and quantification of western blotting images for gp91phox, p47phox, and SOD1 in the PVN in 4 groups. 3 V, the third ventricle. n = 6–8/group. Values are mean ± SEM. In SBP data: #P < 0.05 vs. SHR +AAV-CBS + AAV-Nrf2 shRNA. In other parameters: ***P < 0.001; ****P < 0.0001. Abbreviations: AAV, adenovirus-associated virus; CBS, cystathionine beta-synthase; DHE, dihydroethidium; HO-1, heme oxygenase-1; PVN, paraventricular nucleus; SBP, systolic blood pressure; SHR, spontaneously hypertensive rats; SOD, superoxide dismutase.
The antihypertensive impact of endogenous H2S in SHR is eliminated by PVN knockdown Nrf2
Next, we assessed whether endogenous H2S in the PVN ameliorates hypertension through the Nrf2 pathway. As shown in Figure 4b, PVN endogenous H2S reduced BP in the SHR + AAV-EGFP NC group compared with the AAV-Nrf2 shRNA group (P < 0.05). The findings showed that the hypotensive impact of endogenous H2S might be inhibited by decreasing Nrf2 expression in PVN.
The Nrf2-dependent antioxidant response of endogenous H2S in the PVN of SHR is eliminated by PVN knockdown Nrf2
Heme oxygenase-1 (HO-1) is a powerful antioxidant, and Nrf2 can directly regulate the activity of the HO-1 promoter. The results showed that endogenous H2S in the PVN increased the HO-1-positive neuron number (Figure 4c, P < 0.0001) and the expression level of HO-1 protein (Figure 4d, P < 0.0001) in SHR + AAV-EGFP NC group, but the effect of H2S was eliminated in the AAV-Nrf2 shRNA group, the effect of H2S was eliminated. These findings suggest that PVN endogenous H2S acts via the Nrf2/HO-1 pathway.
The effect of PVN endogenous H2S on oxidative stress of SHR is abrogated by PVN knockdown Nrf2
We discovered that endogenous H2S in the PVN reduced ROS formation (Figure 4e, P < 0.0001) and the expression of gp91phox (Figure 4f,g, P < 0.0001) and p47phox (Figure 4g, P < 0.0001), increased SOD1 (Figure 4g, P < 0.0001) protein expression level in SHR + AAV EGFP NC group. However, these changes were not present in the SHR + Nrf2 shRNA group. These findings imply that Nrf2 knockdown can completely reverse the effects of PVN endogenous H2S on oxidative stress and antioxidant capacity.
DISCUSSION
This research mainly looked into how oxidative stress, inflammation, and hypertensive response are affected by endogenous H2S in PVN. Here are the main conclusions: (i) PVN administration of AAV-CBS can reduce hypertension by increasing endogenous H2S content and decreasing SNA in SHR; (ii) Endogenous H2S in the PVN of SHR rats activates the Nrf2 signaling pathway, increases the production of HO-1 and SOD1, reduces oxidative stress and PICs, and balances the levels of excitatory and inhibitory neurotransmitters; (iii) PVN Nrf2 knockdown eliminated the ameliorative effects of endogenous H2S on hypertension in SHR.
The gas signaling molecule H2S is widely distributed in the cardiovascular system and nervous systems. Its functions include inhibiting inflammatory response, antioxidant stress, and regulating BP.13 There is much evidence that in numerous hypertension models H2S donors and precursors can lower BP,21,22 and reduced peripheral H2S synthesis promotes hypertension development.23 However, it is unclear how endogenous H2S affects hypertension in the CNS, particularly in the PVN. CBS is an enzyme that produces endogenous H2S in the CNS.24,25 According to the findings, CBS is expressed in the PVN, and microinjecting AAV-CBS into the PVN can increase the expression of endogenous H2S while decreasing SHR BP. Previous studies by our group also showed that PVN microinjection of GYY4137 resulted in a hypotensive response in high salt-induced hypertensive rats, whereas PVN microinjection of hydroxylamine hydrochloride (HA, CBS inhibitor) showed an increased BP response.13 These findings imply that the concentration of H2S in PVN is inversely related to BP and HR, and the relationship between them is negative. However, our results differ from the evidence in 2 other acute experiments.26,27 One study found no significant differences in BP, HR, or lumbar SNA after bilateral microinjections into the RVLM or PVN of NaHS (0.2–2000 pmol/side), CBS inhibitor amino-oxyacetate, or HA (0.2–2.0 nmol/side) of WKY rats. The other study found that CBS activity and H2S levels were lower in the PVN in rats with chronic heart failure, and PVN microinjection of low doses of GYY4137 (0.01 and 0.1 nmol) had no significant effect on MAP, renal SNA, or cardiac sympathetic afferent reflex, while high doses of GYY4137 (1, 2, and 4 nmol) increased baseline renal SNA. This may be related to the fact that they were both acute experiments and ours were chronic. This may be related to synthesizing different proteins or substances in chronic experiments. Notably, some reports also support the hypotensive response of H2S in the brain. Duan et al. found that chronic intracerebroventricular infusion of NaHS relieved Ang II-induced hypertension and impacted PVN microglia.28 Guo et al. found a dose-dependent decrease in MAP, renal SNA, and HR after microinjection of NaHS (4, 8, and 16 mM, 50 nl) in bilateral RVLM.29
Antioxidant enzyme activity and content are significantly reduced in hypertension individuals. An increase in ROS levels causes increased oxidative stress. Our earlier research has demonstrated that the development of hypertension and PVN sympathetic excitation are significantly influenced by oxidative stress. H2S has a powerful antioxidant function. It inhibits ROS-mediated CHOP apoptotic signaling in endoplasmic reticulum stress.30 In this investigation, we discovered that endogenous H2S in the PVN decreased ROS level, gp91phox and p47phox, and increased expression of SOD1. This suggests that elevated H2S levels in PVN reduce BP, presumably due to NOX and ROS being downregulated in PVN.
H2S has anti-inflammatory properties at physiological concentrations. According to 1 study, H2S inhibits the activation of the NLRP3 inflammasome, slowing the progression of diabetes-accelerated atherosclerosis and protecting endothelial cells.31 We investigated whether H2S significantly affects the PICs expression in the PVN. Given bilateral injections of AAV-CBS in the PVN, we discovered that the expression levels of TNF-α, IL-6, and IL-1β were decreased while the expression levels of IL-10 were increased in SHR. Additionally, the SHR + AAV-CBS group’s plasma NE levels were considerably lower, suggesting that PVN endogenous H2S may lower BP by lowering sympathetic activation. Additionally, in hypertensive rats, the balance of excitatory and inhibitory neurotransmitters is off. Our findings align with earlier research showing that TH expression is upregulated. In contrast, glutamate decarboxylase isoform GAD67, a hallmark of GABAergic neurons, is downregulated in the PVN of SHR. Endogenous H2S in the PVN increased the GAD67 expression and decreased the TH expression. These findings imply that endogenous H2S’s ability to lower BP may be due to its ability to restore neurotransmitter balance.
The antioxidant transcription factor Nrf2 can bind to antioxidant response elements and stimulate the development of antioxidant enzymes such as HO-1 and SOD, increasing the body’s ability to scavenge ROS.32 Nrf2 is extensively expressed in the CNS, and HO-1 is a crucial Nrf2 target gene. Our research team has previously shown that activating the Nrf2 signaling pathway can shield PVN against oxidative stress by enhancing mitochondrial function.17 The current study found that endogenous H2S in the PVN activated the Nrf2 signaling pathway in SHR, reduced levels of oxidative stress, induced antioxidant enzyme expression, and restored the balance of anti-inflammatory cytokines and PICs. Furthermore, Nrf2-shRNA in the PVN abrogated the hypotensive and antioxidant stress effect of endogenous H2S.
Our findings suggest that endogenous H2S in the PVN protects against SHR hypertension, which can be partially explained by activating the Nrf2/HO-1 pathway, which reduces ROS generation and PICs expression while restoring the appropriate neurotransmitter balance in the PVN.
FUNDING
This study was supported by the National Natural Science Foundation of China (Nos. 82070439 and 82070440), Natural Science Research Program of Shaanxi Province (No. 2020GCZX-13), Fundamental Research Funds for the Central University (No. xzy012022045), and Xi’an Children’s Hospital Research Grant (No. 2020A02).
AUTHORS’ CONTRIBUTIONS
Conceptualization: Yu-Ming Kang and Xiao-Jing Yu; methodology: Nianping Zhang and Tingting Meng; experiments conducted and data gathered: Wen-Jie Xia, Xiao-Min Wang, and Yu Yang; data examined and interpreted: Wen-Jie Xia, Kai-Li Liu, and Jin-An Qiao; writing—original draft preparation: Wen-Jie Xia and Xiao-Jing Yu; writing—review and editing: Yu-Ming Kang. The finished manuscript was examined by each author.
DISCLOSURE
The authors declared no conflict of interest.
DATA AVAILABILITY
The data underlying this article will be shared on reasonable request to the corresponding author.
