Hydrogen sulfide regulation of renal and mesenteric blood flow

Abstract

Hydrogen sulfide (H₂S) dilates isolated arteries, and knockout of the H₂S-synthesizing enzyme cystathionine γ-lyase (CSE) increases blood pressure. However, the contributions of endogenously produced H₂S to blood flow regulation in specific vascular beds are unknown. Published studies in isolated arteries show that CSE production of H₂S influences vascular tone more in small mesenteric arteries than in renal arteries or the aorta. Therefore, the goal of this study was to evaluate H₂S regulation of blood pressure, vascular resistance, and regional blood flows using chronically instrumented rats. We hypothesized that during whole animal CSE inhibition, vascular resistance would increase more in the mesenteric than the renal circulation. Under anesthesia, CSE inhibition [β-cyanoalanine (BCA), 30 mg/kg bolus + 5 mg·kg⁻¹·min⁻¹ for 20 min iv) rapidly increased mean arterial pressure (MAP) more than saline administration (%Δ: saline −1.4 ± 0.75 vs. BCA 7.1 ± 1.69, P < 0.05) but did not change resistance (MAP/flow) in either the mesenteric or renal circulation. In conscious rats, BCA infusion similarly increased MAP (%Δ: saline −0.8 ± 1.18 vs. BCA 8.2 ± 2.6, P < 0.05, n = 7) and significantly increased mesenteric resistance (saline 0.9 ± 3.1 vs. BCA 15.6 ± 6.5, P < 0.05, n = 12). The H₂S donor Na₂S (50 mg/kg) decreased blood pressure and mesenteric resistance ,but the fall in resistance was not significant. Inhibiting CSE for multiple days with dl-proparglycine (PAG, 50 mg·kg⁻¹·min⁻¹ iv bolus for 5 days) significantly increased vascular resistance in both mesenteric (ratio of day 1: saline 0.86 ± 0.033 vs. PAG 1.79 ± 0.38) and renal circulations (ratio of day 1: saline 1.26 ± 0.22 vs. 1.98 ± 0.14 PAG). These results support our hypothesis that CSE-derived H₂S is an important regulator of blood pressure and vascular resistance in both mesenteric and renal circulations. Furthermore, inhalation anesthesia diminishes the effect of CSE inhibition on vascular tone.

NEW & NOTEWORTHY These results suggest that CSE-derived H₂S has a prominent role in regulating blood pressure and blood flow under physiological conditions, which may have been underestimated in prior studies in anesthetized subjects. Therefore, enhancing substrate availability or enzyme activity or dosing with H₂S donors could be a novel therapeutic approach to treat cardiovascular diseases.

INTRODUCTION

Hydrogen sulfide (H₂S), a biological gasotransmitter in mammalian tissues, is enzymatically synthesized by the pyridoxal-5′-phosphate (PLP)-dependent enzymes cystathionine γ-lyase (CSE) and cystathionine β-synthase (CBS) and the PLP-independent enzyme 3-mercaptopyruvate sulfurtransferase (3-MST) (35, 39, 46). Endogenous H₂S exerts multiple effects in many organ systems, including the cardiovascular, respiratory, gastrointestinal, and the central and peripheral nervous systems (26, 39, 46, 49).

In the cardiovascular system (where CSE predominates) (31, 50), H₂S acts primarily as a vasodilator and to potentially lower blood pressure (24, 50). Furthermore, multiple cardiovascular diseases have been associated with a loss of H₂S production, suggesting that the loss of H₂S production might contribute to the development of cardiovascular disease (2, 5, 21, 45). In our previous investigations, inhibiting CSE to prevent H₂S synthesis augmented constriction of small mesenteric (21) and renal arteries. In addition, global knockout of CSE increases arterial pressure and decreases endothelium-dependent dilation in mice (9, 50). Our findings in isolated arteries also suggest CSE-derived H₂S suppresses myogenic tone (21), but the contribution of endogenous H₂S to the regulation of vascular resistance and blood flow in vivo remains largely unknown other than effects on blood pressure observed in knockout mice or in anesthetized rats after enzyme inhibition for many weeks (30, 34).

To evaluate the immediate effects of CSE inhibition on blood flow and vascular resistance, Sprague-Dawley rats were instrumented with a Doppler flow probe on either the superior mesenteric or renal artery, a telemetry blood pressure device, and a femoral vein catheter for drug administration. We hypothesized that endogenous H₂S production regulates local control of blood flow and systemic blood pressure and that this effect would be seen after acute or daily administration of CSE inhibitors.

METHODS

Animals

Male Sprague-Dawley rats (275–330 g) were anesthetized with isofluorane (induction, 5%; maintenance, 2.0%). For treatment infusions, a microrenathane tubing (MRE-040; Braintree Scientific) was inserted into the femoral vein for intravenous injections, filled with saline (1 unit heparin/mL) then tunneled to the back of the head where it was secured beneath the skin until use. All animal protocols were reviewed and approved by the Institutional Animal Care and Use Committee of the University of New Mexico School of Medicine and abide to the National Institutes of Health guidelines for animal use.

For studies in conscious animals, the catheter of a telemetry device (Data Sciences International) was advanced into the abdominal aorta through the left femoral artery to record mean arterial pressure (MAP) and heart rate (HR). The core of the telemetry transmitter was secured subcutaneously in the left upper flank (42).

For studies in anesthetized animals, MAP and HR were measured using an arterial catheter connected to a pressure transducer (Hugo Saks Electronic T75). All variables were recorded and data analyzed using Laboratory Chart 7 software (Life Science Data). Pulsed-Doppler flow probes (20 MHz, 0.7 mm diameter; Iowa Doppler Products) were placed on either the superior mesenteric or left renal arteries, as previously described (13).

Briefly, in both acute (anesthetized or conscious) and conscious 5 day-infused experiments a midline laparotomy was performed and the left kidney was denervated by removing any visible renal nerves, stripping the renal artery adventitia, and painting the renal artery with 20% phenol in ethanol. The flow probes were positioned around the vessels, secured in place, and the leads tunneled subcutaneously to the head and placed in a protective cap along with the venous catheter. For studies in conscious rats, buprenorphine (0.03 mg/kg sc) was administered during the surgery to provide postsurgical analgesia, and rats recovered for 5 days before experiments were performed.

Protocols

Anesthetized rat studies.

After surgery, telemeter and Doppler signals were recorded for at least 60 min to allow stabilization. After 20 min of baseline data was captured, either saline, β-cyanoalanine (BCA, 30 mg/kg bolus + 5 mg·kg⁻¹·min⁻¹ for 20 min), or Na₂S (0.6 mg·kg⁻¹·min⁻¹ for 10 min) were administered through the venous line. Only one treatment was administered each day in random order of administration. Data were collected throughout the study and used to calculate percent change from the resting baseline for MAP, HR, relative blood flow and calculated vascular resistance (MAP/flow). Vehicle (saline) infusions did not alter responses to Na₂S or BCA. Rats were euthanized at the end of the study by exsanguination.

Conscious rat studies.

After recovery from surgery (at least 5 days), rats were placed in a clean cage and acclimated for 60 min. In acute studies, BCA or saline was administered as described above. To evaluate the effect of propargyl glycine (PAG, a reported selective inhibitor of CSE) on the same animal (before and after), another group of rats were administered daily for 5 days. For these studies, baseline recordings of MAP, HR, and Doppler signals were made before administration of the first dose of saline or PAG (50 mg·kg⁻¹·min⁻¹ iv). The dose of the inhibitor used was previously shown to lower endogenous H₂S production (41). Blood pressure and heart rate were recorded daily before the injections and blood flows were also recorded on days 1 and 5. Data were used to calculate percent change from day 1 in MAP, HR and calculated vascular resistance (MAP/flow). Rats were euthanized at the end of the study by exsanguination.

Immunofluorescence studies.

Mesenteric tissue was fixed overnight with 4% paraformaldehyde (PFA) at 27°C and then processed and paraffin embedded with an automated tissue processor (ASP6025 Leica Microsystems) (6). Sections (5 µm) were deparaffinized, and heat-induced epitope retrieval (HIER) was performed at 90°C in sodium citrate buffer (pH 6, anti-3MST) and Tris-EDTA buffer (pH 9, anti-CSE, anti-CBS). Primary antibodies were applied at optimized concentrations determined on control tissues. Sections were incubated overnight with primary antibodies: anti-CSE (12217-1-AP, ProteinTech, 1:200), anti-CBS (GTX113400, GeneTex, 1:1000), anti-3MST (HPA001240, Sigma, 1:1000), anti-PECAM (SC-13537, Santa Cruz, 1:50), and anti-sm22 (ab10135, Abcam, 1:100), followed by 60-min incubation with secondary antibodies, (Dylight 549 or 649, 1:200, Jackson). Slides were counterstained with Sytox Green (1:5000), mounted with ProLong Gold (Cell Signaling Technology) and stored at −20°C until images were recorded. Slide images were collected using a Leica SP5 confocal microscope and imaging system (Leica MicroImaging). The anti-CSE antibody was previously validated by our group using rat aorta endothelial cells overexpressing CSE (17) but also by Ahmad et al. using CSE knockout mice (1). The same group validated the anti-3MST antibody using a 3MST knockout mouse. There is 100% homology between the human peptide used to generate the anti-3MST antibody and the rat 3-MST protein. The anti-CBS antibody was validated by GeneTex using a KO, shRNA, and protein-overexpression approaches. Figure 8B includes representative images of a rat kidney section stained with the anti-3MST antibody following the same protocol used to label mesenteric tissue showing positive 3MST staining in renal tubules but negative staining in a renal artery. Similarly, a liver section was used as positive control for the anti-CBS antibody.

Fig. 8.

Mesenteric tissue sections showing representative staining for cystathionine γ-lyase (CSE), cystathionine β-synthase (CBS), or 3-mercaptopyruvate sulfurtransferase (3-MST). A: sections from at least 5 different animals were imaged for each of the 3 antibodies, and representative images illustrate that CSE but not CBS or 3-MST is expressed in the vascular wall. B: antibodies against CBS and 3-MST were validated using liver and kidney sections, respectively, as positive controls.

Experimental Design and Data Analysis

Two different study designs were performed, using the rat as the experimental unit. All data are expressed as means ± SD, and elimination criteria were applied for parameters (lack of pulsatile blood pressure or blood flow recordings).

For acute studies (Figs. 1–3), the experimental design was transversal-completely randomized for one-factor treatment with two levels: saline (as control) and drug. The drugs assessed were BCA and Na₂S. For PAG studies, the design structure was longitudinal-completely randomized with a factorial treatment (two factors) and time and treatment (with two levels: saline or PAG).

Fig. 1.

β-Cyanoalanine (BCA, 50 mg/kg iv) or saline was administered to rats anesthetized with isoflurane. CSE, cystathionine γ-lyase. A: percent change in blood pressure from baseline. B: percent change in mesenteric blood flow (raw Doppler signal). C: percent change in calculated resistance from baseline [mean arterial pressure (MAP)/flow]. *P < 0.05, analyzed by unpaired t-test.

Fig. 3.

Na₂S (0.6 mg·kg⁻¹·min⁻¹ for 20 min iv) or saline was administered to rats instrumented 5 days previously with arterial and venous catheters. Data are shown for values at the end of the 20-min infusion. A: percent change in blood pressure from baseline. B: percent change in heart rate from baseline. C: percent change in mesenteric blood flow (raw Doppler signal). D: percent change in calculated resistance from baseline [mean arterial pressure (MAP)/flow]. *P < 0.05, significantly different from saline administration, analyzed by unpaired t-test.

For the acute studies, an unpaired t-test was performed to compare saline- and drug-treated groups with the data expressed as percent change from the base line. For PAG study, MAP and HR time courses were analyzed by two-way ANOVA with time as repeated-measure factor. An unpaired t-test was used to assess the effect of PAG on mesenteric and renal-Doppler signals and calculated resistance. Statistical analysis was performed with GraphPad Prism 8 or SAS JMP14 as needed.

RESULTS

To determine the impact of the H₂S synthesizing enzyme CSE on blood pressure, anesthetized rats were acutely infused with the inhibitor BCA (<20 min). MAP increased significantly (saline = −2.8 ± 1.1 vs. BCA = 7.2 ± 1.2%, Fig. 1A) without a change in HR (saline: −0.5 ± 1.4, n = 3 vs. BCA: −2.5 ± 5, n = 6). However, there was not a change in either blood flow (Fig. 1B) or in calculated resistance (Fig. 1C) in either the mesenteric or the renal vascular beds. Because resting MAP was lower in anesthetized compared with conscious rats [100.56 ± 2.5 (n = 9) versus 118.07 ± 2.25 mmHg (n = 35), P < 0.0001], it is possible that the suppression of sympathetic tone by the anesthetic as reported previously (6) lowered resting vascular tone, which masked vasoconstrictor effects of CSE inhibition. Since we previously observed that a H₂S donor causes greater dilation in arteries at 70 mmHg than in arteries at 60 mmHg (12, 29), diminished basal vascular tone would limit H₂S-dependent dilation.

To avoid the effect of anesthesia, studies were repeated in conscious rats. Acute administration of BCA to conscious rats increased blood pressure slightly more than in the anesthetized rats (saline = −0.8 ± 1.2 vs. BCA 7.9 ± 2.6%, Fig. 2A) without changing HR (Fig. 2B). Although, there was a significant fall in blood flow and calculated resistance in the mesentery but not in the kidney (Fig. 2C and D). Infusion of Na₂S caused a small but significant decrease in MAP (Fig. 3A) but did not affect HR (Fig. 3B) nor did it significantly affect blood flow or resistance (Fig. 3, C and D). Representative traces of the acute effects of BCA and Na₂S on MAP, HR, and mesenteric blood flow are shown in Fig. 4.

Fig. 2.

β-Cyanoalanine (BCA, 50 mg/kg iv) or saline was administered to conscious rats instrumented 5 days previously with arterial and venous catheters. Data are shown for values at the end of the 20 min infusion. CSE, cystathionine γ-lyase. A: percent change in blood pressure from baseline. B: percent change in heart rate from baseline. C: percent change in mesenteric blood flow (raw Doppler signal). D: percent change in calculated resistance from baseline [mean arterial pressure (MAP)/flow]. Data were analyzed by unpaired t-tests. *P < 0.05 and ****P < 0.001, significantly different from saline.

Fig. 4.

Representative traces of acute effect of β-cyanoalanine (BCA; left) and Na₂S (right) on mean arterial pressure (MAP), heart rate (HR), and mesenteric blood flow (MBF). CSE, cystathionine γ-lyase; bpm, beats/min.

To better evaluate the contribution of endogenous H₂S production, PAG (3), the irreversible inhibitor of CSE, was daily administered to rats instrumented with either a mesenteric or a renal vascular flow probe. Over the 5 days of PAG administration, MAP was significantly increased from baseline by day 4 and was higher than MAP of rats that received saline (Fig. 5A) accompanied by a parallel decrease in HR (Fig. 5B). Unlike acute inhibition of CSE, chronic inhibition of CSE significantly decreased blood flow in both the mesenteric (Fig. 6A) and renal (Fig. 6C) vascular beds while increasing vascular resistance (Fig. 6, B and D). In addition, PAG has no acute effects on MAP, HR, or mesenteric blood flow (Fig. 7). These results suggest that CSE-derived H₂S plays a significant role in the regulation of blood pressure by controlling mesenteric and renal vascular resistance and that acute inhibition using a reversible short-acting inhibitor may not have completely inhibited CSE.

Fig. 5.

Propargyl glycine (PAG, 50 mg·kg⁻¹·min⁻¹ iv) or saline were administered to conscious rats instrumented with an arterial telemeter, a venous catheter, and a pulsed-Doppler flow probe on the superior mesenteric artery or on the main renal artery at least 7 days before day 1 measurements. CSE, cystathionine γ-lyase. A: mean arterial pressure (MAP) for each day before administration of the daily dose of the inhibitor or saline. B: heart rate (in bpm, beats/min) recorded simultaneously with the blood pressure. Data were analyzed by repeated-measures, two-way ANOVA with treatment and time as the two factors with multiple comparisons using Sidak’s test. **P < 0.01, ***P < 0.001, and ****P < 0.0001, statistical significance vs. saline administration; n = 5 rats.

Fig. 6.

Propargyl glycine (PAG, 50 mg·kg⁻¹·min⁻¹ iv) or saline was administered to conscious rats instrumented with an arterial telemeter, a venous catheter, and a pulsed-Doppler flow probe on the superior mesenteric artery or main renal artery at least 7 days before day 1 measurements. CSE, cystathionine γ-lyase. A: ratio of the final vs. start values for mesenteric flow. B: resistance. C: flow. D: resistance for the renal circulation in rats given 5 days of saline or 5 days of PAG. Data were analyzed by unpaired t-tests. *P < 0.05, statistical significance vs. saline administration.

Fig. 7.

Representative traces of lack of acute effect of propargyl glycine (PAG) on mean arterial pressure (MAP; top), heart rate (HR; in bpm, beats/min; middle), and mesenteric blood flow (MBF; bottom). CSE, cystathionine γ-lyase.

To determine which H₂S-producing enzymes were present in rat mesenteric arteries, immunofluorescence was used to detect CSE, CBS, and 3-MST in rat mesenteric arterial sections. The staining demonstrates that CSE is expressed in the vascular wall of mesenteric arteries, whereas CBS and 3-MST are not (Fig. 8A). Validation of primary antibodies is shown in Fig. 8B. This supports a greater role of CSE in the generation of vasoactive H₂S.

DISCUSSION

Our previous studies show that CSE-derived H₂S regulates myogenic tone in isolated small mesenteric arteries (<100 μm internal diameter) by dampening the constrictor response to increases in intravascular pressure (21). The same size arteries show endothelium-dependent vasodilation to exogenous H₂S, and in these arteries CSE-derived endogenous H₂S contributes to acetylcholine (ACh)-induced vasodilation (17).

CSE global KO mice are hypertensive and showed impaired endothelium-dependent dilation (50). However, whether CSE activity acutely and differentially controls regional vascular blood flow is unknown. Several previous studies using systemic CSE inhibition in vivo did not report increased arterial pressure (10, 32, 34). These prior inhibitor studies were performed in either anesthetized animals or using a tail-cuff method to record blood pressure. Our study shows that only in conscious rats does acute inhibition of CSE increase vascular resistance both in the mesenteric and to a lesser extent in the renal circulation. This is in agreement with our prior studies showing that acute inhibition of CSE in isolated mesenteric arteries increases basal or myogenic tone (21). In addition, increases in vascular resistance and MAP are more modest in anesthetized rats. Thus in conscious rats with chronic inhibition of CSE, resistance increased in both vascular beds examined and caused a greater increase in MAP than acute inhibition. These data suggest that CSE activity is an important regulator of blood flow through direct production of H₂S in the vasculature and that CSE activity contributes to chronic blood pressure regulation.

Many studies have demonstrated that endogenous H₂S produced by CSE plays important, protective roles in the vasculature. Apoliproprotein E^−/− (ApoE^−/−) mice fed a high-fat diet for 16 wk developed significant endothelial disease marked by extensive atherosclerotic lesions and impaired endothelium-dependent dilation. However, pairing the diet with daily intraperitoneal injections of NaHS (10 µmol·kg⁻¹·min⁻¹) significantly reduced lesion size and improved vasodilator responses (15). In a model of cardiac ischemia-reperfusion, administration of the CSE inhibitor PAG increased the myocardial infarct size (36), whereas the administration of exogenous H₂S reduced necrotic damage and improved cardiac function and structural outcomes (36, 37). Furthermore, the endogenous production of H₂S appears to contribute importantly to improved tissue perfusion and vascular remodeling in cases of chronic ischemia (8, 33, 51). Thus, endogenous H₂S is protective in many vascular beds including the kidney (18, 30), heart (36), brain (19, 28), and liver (44). However, the contribution of endogenous H₂S synthesized by CSE to regulation of blood flow in each of these tissues is not clear.

Multiple mechanisms have been suggested to mediate protective effects of H₂S including activation of vascular smooth muscle cell relaxation leading to increased tissue blood flow (25, 34, 41) and reduced tissue oxidative stress (9, 16, 18). We have reported that inhibiting CSE in resistance-size arteries isolated from the mesenteric circulation elevates myogenic tone providing indirect evidence that endogenous H₂S can regulate blood flow in the mesenteric bed (21). In addition, isolated mouse mesenteric vascular arteries relaxed in response to exogenous cysteine (a CSE substrate) but not after pretreatment with PAG (11). The current studies, however, provide the first direct measure of CSE control of blood flow in vivo.

In our studies, increased vascular resistance was apparent in both mesenteric and renal circulations of conscious rats after 5 days of PAG administration, an irreversible CSE inhibitor (3). This is in contrast to the effect of acute administration of BCA, a competitive inhibitor (3) where the increase in resistance was smaller and only observed in conscious rats. A previous study in acutely instrumented anesthetized rats also reported that administration of the CSE inhibitor PAG did not increase renal vascular resistance unless it was combined with inhibition of cystathionine β synthase (CBS) and maintained for 4 mo (34). Although the dose of PAG used in that study was lower than in our study, both suggest CSE participates more in blood flow control in conscious than in unconscious rats, perhaps because of higher activity of the sympathetic nervous system in regulating blood flow. This supposition is based on our observations that dilation to the H₂S donor NaHS is ~40% when arteries are pressurized to 60 mmHg (12), but 100% when arteries are pressurized to 75 mmHg (20). Thus decreased vascular tone has a very profound impact on the response to H₂S.

Similar to the study by Roy et al. (34), PAG required repeated administration to increase blood pressure. Interestingly, the faster-acting competitive inhibitor BCA at the same dose increased blood pressure and mesenteric vascular resistance acutely. This may be why PAG administered acutely does not show effects of CSE inhibition on blood pressure and flow (Fig. 7). The structures of the two inhibitors suggest that the less lipid-soluble PAG may not efficiently cross membranes (40) and is also an irreversible inhibitor explaining why it requires repeated administration to cause sustained inhibition of CSE.

Inhibiting CSE is generally deleterious in the cardiovascular system as discussed above, whereas the administration of exogenous H₂S donors is generally protective. Oral administration of NaHS has been shown to protect renal function in a rat model of renal insufficiency (4) and to attenuate atherosclerosis (27, 48). Plasma and tissue levels of H₂S have been reported as lower in patients with hypertension (38, 45), including preeclampsia-induced hypertension (45), in patients with atherosclerosis (14, 48) and in those with chronic kidney disease (43). Thus, loss of H₂S production may be an important contributor to multiple cardiovascular diseases. However, it is unclear whether circulating H₂S is the primary source for vascular control by H₂S and whether acute or chronic production is the more important role for this gasotransmitter. Indeed, acute administration of an H₂S donor at levels needed to decrease blood pressure has a very narrow therapeutic index so that even moderate doses are lethal to healthy rats (23). In our studies, administering doses sufficient to decrease blood pressure (50 mg/kg iv) led to respiratory arrest and death while a nonlethal dose of 0.6 mg·kg⁻¹·min⁻¹ elicited only a transient fall in blood pressure and apparent discomfort to the conscious animals. Other studies have also reported negative effects of exogenous intravenous H₂S administration (7, 23).

In contrast to intravenous administration, studies in which donors (4) or precursors (38, 47) are administered in the diet appear to effectively lower blood pressure, perhaps due to slower absorption, suggesting substrate supplementation to activate local generation may be more efficacious. Thus, while our studies strongly suggest endogenous CSE-derived H₂S importantly regulates blood flow and blood pressure, much remains unknown about the synthesis and mechanism of action of this vasoactive molecule.

GRANTS

This work was funded by National Heart, Lung, and Blood Institute Grants HL-123301 and HL-07736.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

A.B., J.S.N., L.V.G.-B., and N.L.K. conceived and designed research; H.M.-L., A.B., J.G., and C.E.P. performed experiments; H.M.-L., A.B., J.G., C.E.P., L.V.G.-B., and N.L.K. analyzed data; A.B., J.S.N., L.V.G.-B., and N.L.K. interpreted results of experiments; A.B., J.G., L.V.G.-B., and N.L.K. prepared figures; A.B. drafted manuscript; H.M.-L., A.B., J.G., J.S.N., L.V.G.-B., and N.L.K. edited and revised manuscript; and H.M.-L., A.B., J.G., C.E.P., J.S.N., L.V.G.-B., and N.L.K. approved final version of manuscript.