QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.107.133538v1 324/2/876    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Webb, G. D. Articles by Moore, P. K. Search for Related Content PubMed PubMed Citation Articles by Webb, G. D. Articles by Moore, P. K.

CARDIOVASCULAR

Contractile and Vasorelaxant Effects of Hydrogen Sulfide and Its Biosynthesis in the Human Internal Mammary Artery

Department of Molecular Physiology and Biophysics, College of Medicine, University of Vermont, Burlington, Vermont (G.D.W.); Departments of Medicine (V.M.S.O., S.B.Y., E.A.T.), Pharmacology (L.H.L., Y.P.C., M.Y.A., M.B., P.K.M.), Surgery (M.G.C., R.E.O., C.N.L., P.S.W.), and Pathology (M.S.-T.), Office of Life Science Cardiovascular Biology Programme, National University of Singapore, Singapore; and Clinical Trials and Epidemiology Research Unit, Ministry of Health, Singapore (E.S.Y.C.)

Received October 24, 2007; accepted November 19, 2007.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
This study aimed to test these hypotheses: cystathionine -lyase (CSE) is expressed in a human artery, it generates hydrogen sulfide (H₂S), and H₂S relaxes a human artery. H₂S is produced endogenously in rat arteries from cysteine by CSE. Endogenously produced H₂S dilates rat resistance arteries. Although CSE is expressed in rat arteries, its presence in human blood vessels has not been described. In this study, we showed that both CSE mRNA, determined by reverse transcription-polymerase chain reaction, and CSE protein, determined by Western blotting, apparently occur in the human internal mammary artery (internal thoracic artery). Artery homogenates converted cysteine to H₂S, and the H₂S production was inhibited by DL-propargylglycine, an inhibitor of CSE. We also showed that H₂S relaxes phenylephrine-precontracted human internal mammary artery at higher concentrations but produces contraction at low concentrations. The latter contractions are stronger in acetylcholine-prerelaxed arteries, suggesting inhibition of nitric oxide action. The relaxation is partially blocked by glibenclamide, an inhibitor of K_ATP channels. The present results indicate that CSE protein is expressed in human arteries, that human arteries synthesize H₂S, and that higher concentrations of H₂S relax human arteries, in part by opening K_ATP channels. Low concentrations of H₂S contract the human internal mammary artery, possibly by reacting with nitric oxide to form an inactive nitrosothiol. The possibility that CSE, and the H₂S it generates, together play a physiological role in regulating the diameter of arteries in humans, as has been demonstrated in rats, should be considered.

In water or blood plasma, H₂S is a weak acid that dissociates as follows: H₂S HS^- + H⁺. The pK_a1 at 37°C is 6.76; therefore, when either NaHS or H₂S is dissolved in Krebs' solution that is buffered to remain at pH 7.4 at 37°C, it will form 18.5% H₂S and 81.5% HS^- (and a trace of S^2-;pK_a2 is 11.96), as predicted by the Henderson-Hasselbach equation (Dombkowski et al., 2004). Whiteman et al. (2006) point out that the term "hydrogen sulfide" or "H₂S" is often used (as we do) to encompass the total mixture of H₂S, HS^-, and S^2-; this is appropriate because we do not know which really is the active form of H₂S in vivo; it is likely to be HS^-. The techniques generally used to measure H₂S in blood plasma or serum measure total sulfides; either strong acid is added, converting HS^- and S^2- to H₂S, which is then measured; or strong base is added, converting H₂S and HS^- to S^2-, which is measured.

H₂S relaxes rat blood vessels both in vitro and in vivo; this relaxation is partly due to the opening of K⁺_ATP channels (Zhao et al., 2001; Cheng et al., 2004). A part of the effect on K⁺_ATP channels may be due to decreased intracellular ATP concentration, consequent to inhibition of cytochrome c oxidase, but patch-clamp experiments, in which ATP concentration was controlled, showed that H₂S also has a direct effect on K⁺_ATP channels (Zhao et al., 2001).

Cystathionine -lyase (CSE) is one of the enzymes responsible for the formation of H₂S from L-cysteine. CSE mRNA, as well as H₂S generation when exogenous L-cysteine is provided, has been reported in the rat aorta (Hosoki et al., 1997; Zhao et al., 2001), pulmonary and tail arteries (Zhao et al., 2001), and mesenteric artery (Zhao et al., 2001; Cheng et al., 2004). Perfusion of the rat mesenteric artery bed with physiological solution containing L-cysteine causes vasodilation that is abolished by inhibiting CSE (Cheng et al., 2004); thus, endogenously produced H₂S plays a physiological role in relaxing rat resistance arteries.

The physiological functions of CSE and H₂S have been reviewed previously (Kimura, 2002; Wang, 2002; Moore et al., 2003). Large amounts of CSE occur in rat liver and kidney (Ishii et al., 2004; Mok et al., 2004). CSE has been found in rat heart (Geng et al., 2004) and pancreas (Bhatia et al., 2005). The amounts of CSE in aorta and of H₂S in plasma are reduced in hypertensive rats (Yan et al., 2004). Administration of H₂S to normal (Zhao et al., 2001) or hypertensive (Zhong et al., 2003; Yan et al., 2004) rats decreases blood pressure, whereas the administration of a CSE inhibitor raises normal rat blood pressure (Zhao et al., 2003). In a recent study, it was found that low, physiological concentrations of H₂S contract rat aorta when endogenous nitric oxide is present, suggesting a role of H₂S in modifying the relaxing effects of nitric oxide (Ali et al., 2006).

Taken together, these various observations in rats suggest that H₂S (generated by the activity of CSE) may also play a role in controlling human blood vessel diameter in health and disease. A PubMed search found no previous reports of the presence of CSE, or the biosynthesis of H₂S, in human blood vessels, although CSE protein has been described in human liver (Levonen et al., 2000) and leukocytes (Glode et al., 1981). We undertook the present study to test the hypotheses that CSE protein is expressed in human arteries and that homogenates of human vessels synthesize H₂S in vitro. We also tested whether various concentrations of H₂S relax or contract human artery, whether the relaxation is blocked by a K_ATP channel inhibitor, and whether the contraction is augmented by acetylcholine.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Arteries. The Research Ethics Committee of the National University of Singapore approved our research protocol. After documenting informed patient consent, remainder segments, usually approximately 10 mm long, of human internal mammary artery (also known as the internal thoracic artery) were collected during coronary artery bypass grafting. Arteries from 71 patients were studied. Of the patients, 94% were being treated for hypertension, 56% had type 2 diabetes mellitus, 13% had type 1 diabetes, and 31% were nondiabetic. The group was 77% male. Immediately after obtaining the remainder piece of artery from the surgeon, it was often cut in two, and one-half was submerged in RNAlater (QIAGEN, purchased from Research BioLabs, Singapore). The other half was submerged in phosphate-buffered saline (8 g/l NaCl, 0.2 g/l KCl, 1.44 g/l Na₂HPO₄, 0.24 g/l KH₂PO₄). The adhering connective tissue was then carefully removed from both halves; this required approximately 5 min for each segment. The segment that was submerged in RNAlater (Qiagen) was used for total RNA extraction, followed by reverse transcription (RT) and polymerase chain reaction (PCR) with CSE primers.

The segment that was submerged in phosphate-buffered saline was divided into a 3-mm segment that was snap-frozen in liquid nitrogen and stored at -80°C before protein extraction for Western blotting of CSE and a 2-mm segment that was placed in buffered 10% (v/v) formalin before paraffin embedding and immunohistochemistry. All of these procedures were carried out in the operating theater. Sometimes the entire cleaned artery segment was snap-frozen in liquid nitrogen for assay of H₂S biosynthesis, and on other days, the uncleaned artery was placed into cold, sterile RPMI 1640 tissue culture medium (Invitrogen, Carlsbad, CA) and carried to the laboratory for organ bath studies.

Measurement of CSE mRNA. Total RNA from artery segments was extracted using TRIzol obtained from Invitrogen Life Technologies (Singapore) and reverse transcribed to cDNA using SuperScript II Reverse Transcriptase from Invitrogen. Real-time PCR on the cDNA samples was carried out using the ABI PRISM 7000 Sequence Detection System (Applied Biosystems, Foster City, CA). A human CSE variant 1-specific primer pair (635–656 and 895–875 in GenBank accession number BC015807) was used, in which one of the primers was in the exon that is deleted from variant 2. The CSE variant 1-specific primer sequences were as follows: forward, 5'-AGA AGG TGA TTG ACA TTG AAG G-3'; and reverse, 5'-CAA TAG GAG ATG GAA CTG CTC-3'.

A CSE variant 2-specific primer pair (591–609 and 763–743 in NM_153742) was also used, in which one of the primers straddled the deleted exon. The CSE variant 2-specific primer sequences were as follows: forward, 5'-AGA AAC CAA GCG CCC TTT G-3'; and reverse, 5'-CAA TAG GAG ATG GAA CTG CTC-3'. For human β actin, we used 440 to 460 and 639 to 617 in GenBank accession number NM_001101 [GenBank] : forward, 5'-ATG TTT GAG ACC TTC CAA CAC C-3'; and reverse, 5'-TCC ATC AGT CAG TCC AGG GCC GG-3'.

The primers were custom made for us by Research Biolabs. Two cultures of Escherichia coli transformed with plasmids containing the human CSE sequences, one for variant 1 and one for variant 2, were purchased from the American Type Culture Collection (Manassas, VA). We grew the two E. coli cultures and extracted the CSE-containing plasmids. DNA sequencing confirmed that the plasmids contained the CSE variant 1 or 2 sequences. These plasmids were used as CSE standards. The β actin standards were a gift from Dr. Wai W. Cheung (Department of Pediatrics, National University of Singapore).

Measurement of CSE Protein. Total protein was extracted from frozen human internal mammary arteries by pulverizing the artery with a mortar and pestle under liquid nitrogen, together with modified radioimmunoprecipitation assay buffer containing freshly added protease inhibitor cocktail (Sigma-Aldrich, St. Louis, MO). Protein extracts were stored at -80°C in small aliquots to avoid repeated freeze/thaw, which degrades the CSE protein. Standard SDS-polyacrylamide gel electrophoresis separation of the extracted proteins was carried out using 7.5% polyacrylamide precast gels from Bio-Rad (Hercules, CA). Gels were Western blotted onto polyvinylidene difluoride membranes. After blocking nonspecific sites with 5% powdered skim milk in Tris-buffered saline, the blots were incubated overnight at 4°C with 2 µg/ml polyclonal anti-CSE antibody. US Biological (Swampscott, MA) custom made the anti-human-CSE antibody for the study. They immunized rabbits to the epitope peptide, kndrdvlgisdtlir (361–375, GenBank accession number P32929 [GenBank] ), and purified IgG from the rabbits' serum.

The secondary antibody was anti-rabbit IgG conjugated with horseradish peroxidase, and the bands on the Western blots were visualized by chemiluminescence using ECL Western blotting detection reagents. After imaging the CSE bands on Kodak Biomax Light film (Eastman Kodak, Rochester, NY), the blots were washed and incubated with antibody against heat shock protein (HSP) 90β for a same-lane loading control. The anti-HSP90β antibody was purchased from Chemicon (Temecula, CA). We could not use β-actin as a same-lane loading control because its 42-kDa band sometimes overlapped the CSE band.

The 44.5-kDa bands for variant 1 of CSE were detected in all of the 43 arteries tested with Western blotting. CSE variant 2 was also detected at 39.7 kDa in a few arteries, but the band was not as intense and was not detected in most arteries. Several protein bands of other sizes were also detected, but they did not interfere with the CSE or HSP90β bands. This lack of specificity precluded reliable immunohistochemistry. We plan to develop a specific anti-CSE monoclonal antibody for future experiments, including immunohistochemistry. Human liver lysate, purchased from Chemicon, served as a positive control because liver is known to express CSE in large quantities (Levonen et al., 2000).

Measurement of Tissue H₂S Synthesis. Tissue H₂S production rate was measured as described previously (Stipanuk and Beck, 1982) with modifications. In brief, a snap-frozen 10-mm segment of human internal mammary artery was thoroughly pulverized with 1 ml of 100 mM potassium phosphate buffer, pH 7.4, under liquid nitrogen, using a mortar and pestle. The assay mixtures (500 µl) contained 430 µl of tissue homogenate, 10 mM L-cysteine, and 2 mM pyridoxal 5'-phosphate, plus or minus 1 mM DL-propargylglycine (PAG) (an inhibitor of CSE). Incubation was carried out in tightly sealed Eppendorf vials.

After incubation (37°C, 30 min), zinc acetate (1% w/v, 250 µl) was injected to trap generated H₂S followed by trichloroacetic acid (10% w/v, 250 µl) to precipitate protein and stop the reaction. Next, we added N,N-dimethyl-p-phenylenediamine sulfate (20 mM; 133 µl) in 7.2 M HCl, then FeCl₃ (30 mM; 133 µl) in 1.2 M HCl, and finally filtered (0.22-µm Millipore filter; Millipore Corporation, Billerica, MA) the resulting solution. After 5 min, absorbance at 670 nm of 300-µl aliquots of the filtrates was determined. The H₂S concentration of each sample was calculated using a calibration curve of NaHS (3.125–200 µM), and the results were expressed as nanomoles of H₂S formed per milligram of soluble protein per 30 min. Protein was determined with the Bradford assay, using a kit from Bio-Rad.

Organ Bath Studies. The uncleaned artery was rapidly transported from the operating theater to the laboratory in cold RPMI 1640 tissue culture medium. In the laboratory, it was pinned down under RPMI 1640 solution bubbled with air and cleaned using a dissecting microscope. The artery was cut into 2.5- to 3.0-mm rings that were mounted on wire hooks in organ baths containing Krebs' solution of the following composition: 118 mM NaCl, 5.4 mM KCl, 2.5 mM CaCl₂, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, 25 mM NaHCO₃, and 11.1 mM glucose; pH was 7.4 when equilibrated with 5% CO₂/95% O₂ and maintained at 37°C. Up to four rings were pre-equilibrated for 2 h under2gof tension. After the 2-h equilibration, the artery rings were tested for contractility with 10 µM phenylephrine (PE); rings that gave a poor response were rejected. In later experiments, in the continued presence of PE, they were tested for relaxation with 1 µM acetylcholine; if the relaxation was less than 60% in a ring, that ring was rejected due to endothelial damage.

For the first four arteries we used, a concentration-contraction response curve for PE was obtained. From the average curve obtained for those four arteries, the EC_70% concentration of PE was found to be 3.2 µM. This concentration of 3.2 µM PE was used in all subsequent experiments to precontract the artery rings, before the application of progressively increased concentrations of the H₂Sdonor, NaHS (in the continued presence of PE), to obtain a concentration-relaxation response curve. After washing with fresh Krebs' solution, the concentration-relaxation response to NaHS experiment was repeated three more times when possible (sometimes the artery ring did not remain viable that long). On some days, glibenclamide was applied to half of the rings after the PE precontraction and before the NaHS concentration-relaxation response series; the other half of the rings served as controls. Results are shown as percent reduction of the PE-induced contraction. In another series of experiments, 3.2 µM PE was applied to contract the artery rings, then 1 µM acetylcholine (Ach) was applied to relax the artery rings; then increasing concentrations of NaHS were applied, and the results were expressed as the percentage of contraction toward the previous PE contraction tension. At pH 7.4 and 37°C, 18.5% of the HS^- anion combines with H⁺ to form H₂S (Dombkowski et al., 2004); this reaction did not significantly change the pH of our well buffered Krebs' solution, even at the highest NaHS concentration used.

Chemicals and Reagents. The chemicals and reagents were obtained from Sigma-Aldrich, unless otherwise noted.

Patient Data. In the operating theater, during surgery, the following information was copied from the patient's case notes: gender, race, birth date, height, weight, blood pressure the previous day, history of hypertension, diabetes (presence and type), hyperlipidemia, smoker (present, former, or never), serum glucose concentration, serum K⁺ concentration, and medication history.

Data Analysis. Results are expressed as the mean ± S.E. The Student's two-tailed t test for paired samples was used to compare H₂S production of arteries in the presence and absence of CSE inhibitor and to compare the contraction or relaxation responses of the artery rings in the presence and absence of glibenclamide at each of the six NaHS concentrations used. A one-sample Student's t test was used to compare the contraction responses with zero response. Repeated-measures analysis of variance was used to test for the presence of an effect modification of glibenclamide on the response to NaHS and also to test for a trend in response to varying NaHS concentrations.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Expression of CSE mRNA in a Human Artery. Representative agarose gels, showing the products produced from real-time RT-PCR using primers for human CSE variant 1, are shown in Fig. 1, A and B. The gels show that PCR of the cDNA from some of the artery samples produced a 261-bp product, the size expected from our CSE variant 1 primers. Further amplification and sequencing of the cDNA from one of these artery PCR product bands confirmed that it was the CSE variant 1 amplicon. The real-time data could not be used for quantifying the amount of CSE mRNA in the artery samples, due to dimer formation, as may be seen at approximately 100 bp in many of the lanes. Lanes 2 in Fig. 1, A and B, are from tubes containing 10 copies of the CSE plasmid standard. Only dimer is seen in lane 2 of gel A, whereas the expected band at 261 bp is seen in lane 2 of gel B. It is possible that there were fewer than 10 starting copies of CSE standard in gel A, due to random molecular distribution. The absence of visible CSE bands for two arteries in both the A and B gels and the absence of β actin bands in the same arteries in C and D is probably due to degradation of the RNA before transcription. The gels (data not shown) after PCR with variant 2 primers usually showed either weak or no visible bands at 173 bp, the size of the CSE variant 2 product. We also carried out RT-PCR using primers for human cystathionine β synthase, which, like CSE, can release H₂S from cysteine. We did not detect any cystathionine β synthase mRNA in our artery samples.

View larger version (34K): [in this window] [in a new window]   Fig. 1. Products from real-time RT-PCR for CSE variant 1. Agarose gels A and B show real-time PCR products resulting from use of the human CSE variant 1-specific primers, which produce a 261-bp amplicon. The real-time data could not be used for quantifying the amount of CSE mRNA in the artery samples, due to dimer formation, as may be seen at about 100 bp. Lane 1, 100-bp size markers; the brighter band near the top of the lanes is 500 bp. Lane 2 in A and B are from tubes containing 10 copies of the CSE plasmid standard; note that only dimer is seen in gel A (because of random molecular distribution, there may have been less than 10 starting copies). Lanes 3 to 6 are from tubes starting with 100, 1000, 10,000, and 100,000 copies of the CSE variant 1 standard. Lanes 7 are no-transcript controls, and the rest are cDNA transcripts (2.5 µg/tube) of total RNA extracted from artery samples. Agarose gels C and D are products from primers for β actin; they are lined up with gels A and B just above, so that the same artery samples match. The standards were for β actin and were 10 to 100,000 copies in C and 100 to 100,000 copies in D.

Western Blotting to Detect CSE Protein. Figure 2 shows typical Western blots. The anti-CSE antibody used was raised against a 15-amino acid epitope common to both variants 1 and 2 of human CSE. A band at 39.7 kDa, the molecular mass of CSE variant 2, was rarely seen, in contrast to variant 1, which was always found at 44.5 kDa, the molecular mass of CSE V1. Thus, the shorter splice variant 2, which is enzymatically inactive (Levonen et al., 2000), is not consistently expressed in adult human internal mammary artery, although it is expressed in neonatal human liver (Levonen et al., 2000).

View larger version (74K): [in this window] [in a new window]   Fig. 2. Western blots showing CSE protein in the human internal mammary artery. CSE variant 1 and HSP90β bands (the loading control) from three representative blots are shown, each with four different artery samples (Art), plus a liver sample (Liv; the positive control).

Hydrogen Sulfide Production by Artery Homogenates. The CSE protein expressed in the internal mammary artery produces H₂S when incubated (30 min) with cysteine (Fig. 3). H₂S production by homogenates of arteries from five patients was evaluated. Biosynthesis of H₂S, from added L-cysteine (10 mM) in the presence of pyridoxal phosphate (2 mM), was detected in all five arteries tested: the mean ± S.E. was 7.7 ± 1.9 nmol H₂S/mg protein/30 min. The positive control, rat liver, generated 39.5 ± 5.2 nmol H₂S/mg protein/30 min, n = 6. This higher rate is consistent with the fact that rat and mouse liver express 50 times or more CSE protein than do most other tissues (Ishii et al., 2004). PAG (1 mM), a known inhibitor of CSE, inhibited H₂S formation in human arteries by 39 ± 9%, p = 0.020, n = 5. PAG inhibited rat liver H₂S formation by 96 ± 18%, p = 0.0004, n = 6. Thus CSE is probably the primary source of H₂S generation in liver, whereas about half of the H₂S generated by human arteries seems to come from other sources.

View larger version (13K): [in this window] [in a new window]   Fig. 3. H2S production by the human internal mammary artery. H2S production, in the presence of 10 mM L-cysteine, by homogenates of human arteries and of rat liver are shown in the bar graph. T = 0 is at the start of the experiment. The next four bars are after 30 min. Each of five artery and six liver homogenates was run with and without 1 mM PAG, an inhibitor of CSE. Bars, average of results from five human patients and from six rats ± S.E. The H2S production was significantly reduced by PAG (p = 0.020 for human artery and p = 0.0004 for rat liver).

View larger version (16K): [in this window] [in a new window]   Fig. 4. Inhibition of the H2S relaxation by a K+ATP channel blocker, glibenclamide, and contraction produced by low concentrations of NaHS. The artery rings were precontracted with 3.2 µM phenylephrine, which we previously found to be the EC70% concentration. Points are the average of four arteries from four patients. Three rings from one patient, and two from each of the other three patients were studied. On each ring, two to four replicate experiments were carried out. Half of the rings served as no glibenclamide controls, and for the other half, 50 µM glibenclamide was added 5 min before beginning the NaHS concentration-response series. A Student's t test was used to compare the with- and without-glibenclamide average values for each of the six concentrations of NaHS. A significant difference was found between the with- and without-glibenclamide responses to 50 and 200 µM NaHS (p < 0.01). A one-sample Student's t test found a significant difference between the contraction response to 100 µM NaHS and zero (p < 0.05). A weighted average of the responses to 800 and 1600 µM NaHS found a significant difference between the with- and without-glibenclamide relaxation responses (p < 0.01). There was overall a significant effect of varying NaHS concentrations on the response of the artery rings (p = 0.01) and a significant effect modification by glibenclamide on this response to NaHS (p < 0.05).

View larger version (33K): [in this window] [in a new window]   Fig. 5. Tension traces showing responses to NaHS and a no-NaHS control after precontraction of human artery rings with 3.2 µM PE (to produce 70% of the maximal contraction), followed by prerelaxation with 1.0 µM Ach. The top trace shows the effects of progressively increased concentrations of NaHS on tension. The bottom trace is a control experiment run simultaneously on another ring from the same piece of artery. Similar results, showing contraction with low NaHS concentrations and relaxation with high concentrations after a successful relaxation with Ach, were obtained with arteries from four different patients.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
The present study found CSE mRNA and protein in the human internal mammary artery. Homogenates of this human artery generate H₂S when the CSE substrate, L-cysteine, is present, and the CSE inhibitor DL-propargylglycine blocks H₂S production by the artery homogenates. Finally, higher concentrations of H₂S relax the artery, an effect that is partially blocked by glibenclamide, a K⁺_ATP channel blocker. Lower physiological concentrations of H₂S contract the artery; this effect is especially prominent when the artery is prerelaxed with ACh, which releases NO from the endothelium. Although all the above results have been reported for rat arteries (Hosoki et al., 1997; Zhao et al., 2001; Cheng et al., 2004; Yan et al., 2004; Ali et al., 2006), to the best of our knowledge, they have not been previously reported for human arteries.

Normal Plasma Sulfide Concentration. Jiang et al. (2005) found that in serum from normal human subjects, total sulfides averaged 52 µM, and Chen et al. (2005) found that normal serum sulfide concentration averaged 35 µM. Three studies of normal rat plasma found averages of 46 (Zhao et al., 2001), 39 (Zhong et al., 2003), and 48 (Yan et al., 2004) µMH₂S (total sulfide).

Discussion of Our Organ Bath Findings. In experiments (Fig. 4, without glibenclamide), 50 µM NaHS produced significant contraction of an artery ring. This concentration is almost identical to the normal human serum sulfide concentration reported by Jiang et al. (2005). Evidence was recently published suggesting that low concentrations of NaHS add further contraction to the precontracted rat aorta because the phenylephrine contraction induces the endothelium to release NO, which then attenuates the PE contraction, but H₂S reacts with NO to form a nitrosothiol, which is not vasoactive. Therefore, H₂S inactivates the relaxing influence of NO (Ali et al., 2006; Whiteman et al., 2006). This hypothesis is consistent with our human artery results; low concentrations of NaHS produced contraction of the PE precontracted human internal mammary artery, as seen in Fig. 4.

These NaHS-induced contractions are apparently due either to inactivation of NO or inhibition of NO production because prerelaxation with Ach, which releases large amounts of NO, greatly increases the amount of contraction produced by low concentrations of H₂S, as seen in Fig. 5. This supports the nitrosothiol formation hypothesis but also leaves open the possibility that part of the contraction may be due to inhibition of nitric oxide synthase (NOS). Kubo et al. (2007) found that NaHS inhibits endothelial NOS in a concentration-dependent manner. However, 100 µM NaHS inhibited endothelial NOS by only 23%; thus, we think the pronounced contraction we observed with 10 or 20 µM NaHS (Fig. 5) is predominantly due to nitrosothiol formation. Because we observed the contraction effect of H₂S at or below the concentrations reported in normal human blood plasma, attenuation of the effect of NO may be an important physiological role of H₂S in human arteries. It should be noted that the human internal mammary artery is a conduit artery, as is the rat aorta studied by Ali et al. (2006). It remains to be tested whether this contractile response occurs in rat or human resistance arteries. To date, only relaxant effects of H₂S have been observed in rat resistance arteries (Wang, 2002; Cheng et al., 2004). This includes relaxation of rat mesenteric resistance arteries in response to endogenously released H₂S (Cheng et al., 2004).

The Role of K⁺_ATP Channels in H₂S Relaxation. Glibenclamide, a K⁺_ATP channel blocker, partially blocked the relaxation caused by higher concentrations of NaHS (Fig. 4). This result suggests that part of the relaxation of the human internal mammary artery produced by moderate to high concentrations of H₂S is due to increased open time of the K⁺_ATP channels. Our result confirms in a human artery what has already been demonstrated in rat arterial smooth muscle (Zhao et al., 2001; Cheng et al., 2004; Ishii et al., 2004). In patch-clamp studies of isolated rat arterial smooth muscle cells, whole cells produced sufficient endogenous H₂Stoincrease the K⁺_ATP channel current, whereas the inhibition of CSE with PAG decreased the K⁺_ATP channel current (Tang et al., 2005). Another patch-clamp study, in which the ATP concentration was controlled, showed that a reduction in the ATP concentration increases the K⁺_ATP channel current, as expected, but H₂S further increases the current at all ATP concentrations tested, thus suggesting a direct action of H₂S on the K⁺_ATP channels (Zhao et al., 2001). The Fig. 4 results show that K⁺_ATP channels play a role in the H₂S-induced relaxation in the human internal mammary artery. Part of the H₂S relaxation in human artery may be due to a direct effect of H₂SonK⁺_ATP channels. A reduction in intracellular ATP concentration, however, due to the inhibition of metabolism by H₂S, which inhibits cytochrome c oxidase (Beauchamp et al., 1984), might be the dominant cause of the apparent increase in K⁺_ATP channel current produced by H₂S. This is supported by the finding that metabolic inhibitors produce a glibenclamide-sensitive vasodilation in guinea pig coronary arteries (Daut et al., 1990). The glibenclamide-insensitive portion of the H₂S relaxation may be due to reduced availability of ATP for actin-myosin cross-bridge cycling.

A Role of H₂S in Hypertension? In a study of patients with coronary artery disease, those with normal blood pressure had an average serum H₂S concentration of 34 µM, whereas the patients with hypertension had a significantly lower average H₂S concentration of 20 µM (Jiang et al., 2005). Recall from Fig. 5 that 20 µM NaHS contracted the human artery; thus, we can speculate that sulfide-induced contraction might play a role in human hypertension. Rat experiments also suggest a role for H₂S in hypertension. Spontaneously hypertensive rats have significantly lower plasma concentrations of H₂S compared with normotensive control rats, and the administration of H₂S to spontaneously hypertensive rats decreases their blood pressure (Yan et al., 2004). Likewise, H₂S administration lowered the blood pressure in rats made hypertensive by long-term treatment with the nitric oxide synthase inhibitor, L-N-nitro-arginine methyl ester (Zhong et al., 2003). In contrast to the hypotensive effect of H₂S, the administration of the CSE inhibitor, PAG, to normal rats increased their blood pressure (Zhao et al., 2003).

Other Possible Pathological Roles. Hypertension and other pathological conditions can lead to vascular smooth muscle cell proliferation and deleterious remodeling of arterial walls. When cultured human aorta smooth muscle cells are exposed to exogenous H₂S or are transfected with CSE containing adenovirus, apoptosis is induced (Yang et al., 2006). This suggests a possible therapeutic treatment for diseases associated with vascular remodeling. In experimental hemorrhagic shock in rats, CSE mRNA and H₂S synthesis are increased in the liver, leading to generalized vasodilation and decreased blood pressure that can be reversed by inhibiting CSE with PAG (Mok et al., 2004). In experimental pancreatitis in rats, CSE mRNA and H₂S production are increased in the pancreas, and inhibition of CSE with PAG reduces the resulting damage to the pancreas and lungs (Bhatia et al., 2005). Excessive production of H₂S may be involved in human chronic obstructive pulmonary disease, as Chen et al. (2005) found that serum sulfide concentration was significantly higher in patients with stable chronic obstructive pulmonary disease than in age matched control subjects.

H₂S may play a role in septic shock. Li et al. (2005) found that human patients with septic shock had an average plasma concentration of 151 µMH₂S compared with 44 µM in similar aged healthy control subjects. In our organ bath experiments with human artery, 150 µMH₂S caused relaxation (Fig. 5). Li et al. also found that lipopolysaccharide-induced inflammation in mice was associated with increased plasma H₂S concentrations. Finally, the contractile effect of H₂S on human arteries prerelaxed with Ach, which we saw with concentrations of H₂S that are similar to those found in normal human blood plasma, suggest that H₂S may be a physiological mediator for moderating the relaxant effect of NO.

In summary, we have shown for the first time that CSE is expressed in the human internal mammary artery. We also demonstrated that homogenates of human internal mammary artery convert cysteine to H₂S. Furthermore, we have shown that H₂S can contract this human artery, when NO is present, at concentrations similar to the normal concentration of H₂S in human blood plasma. At somewhat higher concentrations, it relaxes the artery. Our results suggest that CSE and H₂S may play a role in the control of vascular function and of blood pressure in humans.

	Acknowledgements

 
We thank Lim Seng Gee for sharing the ABI-Prism 7000 Sequence Detection System, Shanthi Wasser for technical advice, and Joni Lo, Neil Kad, and William Barnes for help with the figures. We are especially grateful to the registrars (resident surgeons) who informed the patients about the study and obtained signed consent from those who generously agreed to let us use the piece of their artery that is left over during coronary artery bypass grafting. The registrars were W. Maung Maung Aye, Atasha Asmat, Ali Zamir, Hardip Singh, Ooi Oon Cheong, Hon Ian Chong, Hua Gwee Nang, Lim Kai Toh, and Vitaly Sorokin. We also thank Bok Lan Goh, the head perfusionist, who facilitated the collection of the remainder segment of artery and its rapid transfer to us.

	Footnotes

 
The Agency for Science, Technology and Research (A^*STAR) awarded a Graduate Scholarship to M.Y.A., and the Office of Life Science of the National University of Singapore (Grant R-184-000-074-712) provided the financial support for this project (principal investigator, P.K.M.) Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.

doi:10.1124/jpet.107.133538.

ABBREVIATIONS: H₂S, hydrogen sulfide; NaHS, sodium hydrogen sulfide, but often called sodium hydrosulfide; CSE, cystathionine -lyase; RT, reverse transcription; PCR, polymerase chain reaction; HSP, heat shock protein; PAG, DL-propargylglycine; PE, phenylephrine; Ach, acetylcholine; PE, phenylephrine; NOS, nitric-oxide synthase.

Address correspondence to: George Webb, Department of Molecular Physiology and Biophysics, University of Vermont, College of Medicine, Health Science Research Facility Building, Burlington, VT 05405. E-mail: george.webb{at}uvm.edu

	References

Top Abstract Materials and Methods Results Discussion References

 

Ali MY, Ping CY, Mok Y-YP, Ling L, Whiteman M, Bhatia M, and Moore PK (2006) Regulation of vascular nitric oxide in vitro and in vivo; a new role for endogenous hydrogen sulphide? Br J Pharmacol 149: 625-634.[CrossRef][Medline]

Beauchamp RO Jr, Bus JS, Popp JA, Boreiko CJ, and Andjelkovich DA (1984) A critical review of the literature on hydrogen sulfide toxicity. Crit Rev Toxicol 13: 25-97.[Medline]

Bhatia M, Wong FL, Fu D, Lau HY, Moochhala SM, and Moore PK (2005) Role of hydrogen sulfide in acute pancreatitis and associated lung injury. FASEB J 19: 623-625.[Abstract/Free Full Text]

Blackstone E, Morrison M, and Roth MB (2005) H₂S induces a suspended animation-like state in mice. Science 308: 518.[Abstract/Free Full Text]

Chen Y-H, Yao W-Z, Geng B, Ding Y-L, Lu M, Zhao M-W, and Tang C-S (2005) Endogenous hydrogen sulfide in patients with COPD. Chest 128: 3205-3211.[CrossRef][Medline]

Cheng Y, Ndisang JF, Tang G, Cao K, and Wang R (2004) Hydrogen sulfide-induced relaxation of resistance mesenteric artery beds of rats. Am J Physiol Heart Circ Physiol 287: H2316-H2323.[Abstract/Free Full Text]

Daut J, Maier-Rüdolph W, von Beckerath N, Mehrke G, Günther K, and Goedel-Meinen L (1990) Hypoxic dilation of coronary arteries is mediated by ATP-sensitive potassium channels. Science 247: 1341-1344.[Abstract/Free Full Text]

Dombkowski RA, Russell MJ, and Olson KR (2004) Hydrogen sulfide as an endogenous regulator of vascular smooth muscle tone in trout. Am J Physiol Regul Integr Comp Physiol 286: R678-R685.[Abstract/Free Full Text]

Geng B, Yang JH, Qi YF, Zhao J, Pang YZ, Du JB, and Tang CS (2004) H₂S generated by heart in rat and its effects on cardiac function. Biochem Biophys Res Commun 313: 362-368.[CrossRef][Medline]

Glode LM, Epstein A, and Smith CG (1981) Reduced -cystathionase protein content in human malignant leukemia cell lines as measured by immunoassay with monoclonal antibody. Cancer Res 41: 2249-2254.[Abstract/Free Full Text]

Hosoki R, Matsuki N, and Kimura H (1997) The possible role of hydrogen sulfide as an endogenous smooth muscle relaxant in synergy with nitric oxide. Biochem Biophys Res Commun 237: 527-531.[CrossRef][Medline]

Ishii I, Akahoshi N, Yu XN, Kobayashi Y, Namekata K, and Komak G (2004) Murine cystathionine -lyase: complete cDNA and genomic sequences, promoter activity, tissue distribution and developmental expression. Biochem J 381: 113-123.[CrossRef][Medline]

Jiang HL, Wu HC, Li ZL, Geng B, and Tang CS (2005) Changes of the new gaseous transmitter H₂S in patients with coronary heart disease. Di Yi Jun Yi Da Xue Xue Bao 25: 951-954.[Medline]

Kimura H (2002) Hydrogen sulfide as a neuromodulator. Mol Neurobiol 26: 13-19.[CrossRef][Medline]

Kubo S, Doe I, Kurokawa Y, Nishikawa H, and Kawabata A (2007) Direct inhibition of endothelial nitric oxide synthase by hydrogen sulfide: contribution to dual modulation of vascular tension. Toxicology 232: 138-146.[CrossRef][Medline]

Levonen AL, Lapatto R, Saksela M, and Raivio KO (2000) Human cystathionine -lyase: developmental and in vitro expression of two isoforms. Biochem J 347: 291-295.[CrossRef][Medline]

Li L, Bhatia M, Zhu YZ, Zhu YC, Ramnath RD, Wang ZJ, Anuar FBM, Whiteman M, Salto-Tellez M, and Moore PK (2005) Hydrogen sulfide is a novel mediator of lipopolysaccharide-induced inflammation in the mouse. FASEB J 19: 1196-1198.[Abstract/Free Full Text]

Mok YYP, Atan MSBM, Cheong YP, Wang ZJ, Bhatia M, Moochhala S, and Moore PK (2004) Role of hydrogen sulfide in haemorrhagic shock in the rat: protective effect of inhibitors of hydrogen sulfide biosynthesis. Br J Pharmacol 143: 881-889.[CrossRef][Medline]

Moore PK, Bhatia M, and Moochhala S (2003) Hydrogen sulfide: from the smell of the past to the mediator of the future? TRENDS in Pharmacol Sci 24: 609-611.[CrossRef]

Stipanuk MH and Beck PW (1982) Characterisation of the enzymic capacity for cysteine desulphhydration in liver and kidney of the rat. Biochem J 206: 267-277.[Medline]

Tang G, Wu L, Liang W, and Wang R (2005) Direct stimulation of K_ATP channels by exogenous and endogenous hydrogen sulfide in vascular smooth muscle cells. Mol Pharmacol 66: 1757-1764.

Wang R (2002) Two's company, three's a crowd: can H₂S be the third endogenous gaseous transmitter? FASEB J 16: 1792-1798.[Abstract/Free Full Text]

Whiteman M, Li L, Kostetski I, Chu SH, Siau JL, Bhatia M, and Moore PK (2006) Evidence for the formation of a novel nitrosothiol from the gaseous mediators nitric oxide and hydrogen sulphide. Biochem Biophys Res Commun 343: 303-310.[CrossRef][Medline]

Yan H, Du J, and Tang C (2004) The possible role of hydrogen sulfide on the pathogenesis of spontaneous hypertension in rats. Biochem Biophys Res Commun 313: 22-27.[CrossRef][Medline]

Yang G, Wu L, and Wang R (2006) Pro-apoptotic effect of endogenous H₂S on human aorta smooth muscle cells. FASEB J 20: 553-555.[Abstract/Free Full Text]

Zhao W, Zhang J, Lu Y, and Wang R (2001) The vasorelaxant effect of H₂S as a novel endogenous gaseous K_ATP channel opener. EMBO J 20: 6008-6016.[CrossRef][Medline]

Zhao WM, Ndisang JF, and Wang R (2003) Modulation of endogenous production of H₂S in rat tissues. Can J Physiol Pharmacol 81: 848-853.[CrossRef][Medline]

Zhong GZ, Chen FR, Cheng YQ, Tang CS, and Du JB (2003) The role of hydrogen sulfide generation in the pathogenesis of hypertension in rats induced by inhibition of nitric oxide synthase. J Hypertens 21: 1879-1885.[CrossRef][Medline]

This article has been cited by other articles:

				M. Xia, L. Chen, R. W. Muh, P.-L. Li, and N. Li Production and Actions of Hydrogen Sulfide, a Novel Gaseous Bioactive Substance, in the Kidneys J. Pharmacol. Exp. Ther., June 1, 2009; 329(3): 1056 - 1062. [Abstract] [Full Text] [PDF]

				Q. C. Yong, S. W. Lee, C. S. Foo, K. L. Neo, X. Chen, and J.-S. Bian Endogenous hydrogen sulphide mediates the cardioprotection induced by ischemic postconditioning Am J Physiol Heart Circ Physiol, September 1, 2008; 295(3): H1330 - H1340. [Abstract] [Full Text] [PDF]

				L. Li, M. Whiteman, Y. Y. Guan, K. L. Neo, Y. Cheng, S. W. Lee, Y. Zhao, R. Baskar, C.-H. Tan, and P. K. Moore Characterization of a Novel, Water-Soluble Hydrogen Sulfide-Releasing Molecule (GYY4137): New Insights Into the Biology of Hydrogen Sulfide Circulation, May 6, 2008; 117(18): 2351 - 2360. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.107.133538v1 324/2/876    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Webb, G. D. Articles by Moore, P. K. Search for Related Content PubMed PubMed Citation Articles by Webb, G. D. Articles by Moore, P. K.