Interaction between hydrogen sulfide-induced sulfhydration and tyrosine nitration in the K_ATP channel complex

Abstract

Hydrogen sulfide (H₂S) is an endogenous gaseous mediator affecting many physiological and pathophysiological conditions. Enhanced expression of H₂S and reactive nitrogen/oxygen species (RNS/ROS) during inflammation alters cellular excitability via modulation of ion channel function. Sulfhydration of cysteine residues and tyrosine nitration are the posttranslational modifications induced by H₂S and RNS, respectively. The objective of this study was to define the interaction between tyrosine nitration and cysteine sulfhydration within the ATP-sensitive K⁺ (K_ATP) channel complex, a significant target in experimental colitis. A modified biotin switch assay was performed to determine sulfhydration of the K_ATP channel subunits, Kir6.1, sulphonylurea 2B (SUR2B), and nitrotyrosine measured by immunoblot. NaHS (a donor of H₂S) significantly enhanced sulfhydration of SUR2B but not Kir6.1 subunit. 3-Morpholinosydnonimine (SIN-1) (a donor of peroxynitrite) induced nitration of Kir6.1 subunit but not SUR2B. Pretreatment with NaHS reduced the nitration of Kir6.1 by SIN-1 in Chinese hamster ovary cells cotransfected with the two subunits, as well as in enteric glia. Two specific mutations within SUR2B, C24S, and C1455S prevented sulfhydration by NaHS, and these mutations prevented NaHS-induced reduction in tyrosine nitration of Kir6.1. NaHS also reversed peroxynitrite-induced inhibition of smooth muscle contraction. These studies suggest that posttranslational modifications of the two subunits of the K_ATP channel interact to alter channel function. The studies described herein demonstrate a unique mechanism by which sulfhydration of one subunit modifies tyrosine nitration of another subunit within the same channel complex. This interaction provides a mechanistic insight on the protective effects of H₂S in inflammation.

protein function can be fundamentally altered during inflammation attributable to multiple oxidative reactions. The sulfhydryl groups on cysteine residues can undergo several modifications including S-nitrosylation, S-palmitoylation, S-glutathionylation, and/or S-sulfhydration (28). In addition to modifications of cysteine residues, nitration of tyrosine by peroxynitrite (ONOO⁻) can also alter functional properties of proteins (13, 14, 19). During colonic inflammation, S-sulfhydration and tyrosine nitration occur within smooth muscle cells. Our studies in colonic smooth muscle demonstrate that S-sulfhydration of the ATP-sensitive K⁺ (K_ATP) channel during colonic inflammation enhances K_ATP currents, whereas tyrosine nitration of the L-type calcium channel results in reduced calcium currents (6, 11, 12, 20).

Hydrogen sulfide (H₂S) is an endogenous gas involved in several biological functions, including neuromodulation, smooth muscle relaxation, inflammation, insulin release, and metabolic demand. H₂S is predominantly produced in mammalian tissues from l-cysteine by the enzymes cystathionine β-synthetase and cystathionine γ-lyase (16, 25). H₂S converts cysteine residues to thiocysteines (i.e., modifies the -SH group to -SSH; S-sulfhydration) in many proteins, including actin, tubulin, and GAPDH, and the K_ATP channel (17).

K_ATP channels play an important role in cellular function observed in a variety of tissues, including smooth muscles, pancreatic β-cells, myocardium, and neurons (18, 29). K_ATP channel in the smooth muscle colon is a heterooctamer consisting of two main subunits, a pore-forming subunit (Kir6.1) and a sulfonylurea receptor subunit (SUR2B) (8). H₂S S-sulfhydrates the sulfonylurea subunit, SUR2B, of the K_ATP channel in colonic smooth muscle. The S-sulfhydration of the SUR2B by H₂S provides an allosteric modulation of the channel, resulting in enhanced potency for the K_ATP channel opener, levcromakalim (6).

Since the discovery of endogenous H₂S, many studies have explored the role of this gas in animal models. In an experimental model of colonic inflammation, enhanced H₂S synthesis is thought to result in faster resolution of the inflammation (22, 23). Furthermore, inhibition of K_ATP channel in colitis rats also induces significant mortality (24), indicating that H₂S and K_ATP channels are important components in ulcerative colitis. The mechanism by which K_ATP channel is affected may be through sulfhydration of specific cysteine residues (17). However, the molecular mechanisms involved in such modulation are not fully elucidated. Previously, we showed in mouse models of colonic inflammation that oxidative stress is also associated with increased tyrosine nitration in smooth muscle (1). The increased expression of tyrosine-nitrated calcium channels in inflamed colon prevents the binding and regulation of the calcium channel by c-Src kinase (11, 21).

In the present study, we explore the interaction of S-sulfhydration and tyrosine nitration within the K_ATP channel complex. Here, we demonstrate that sulfhydration of the SUR2B subunit prevents nitration of the pore-forming Kir6.1 subunit of the channel. This type of interaction provides mechanistic insight into the protective effects of H₂S in colonic inflammation.

MATERIALS AND METHODS

All animal protocols conformed to the guidelines and approval by the Virginia Commonwealth University Institutional Animal Care and Use Committee.

Cell culture and transient transfection.

Chinese hamster ovary (CHO) cells were purchased from the American Type Cell Collection (ATCC). CHO cells were cultured in Dulbecco's modified Eagle's medium/F12 medium supplemented with 0.2 mg/ml geneticin and 10% fetal bovine serum (Invitrogen). Cells were grown in six-well plates overnight and transfected with Kir6.1, SUR2B, enhanced green fluorescent protein, a combination of plasmid DNAs, or with Cav1.2b and β2-subunit by Genejammer (Stratagene), as described by the manufacturer. Untransfected CHO cells were maintained in the same media. After ∼24–48 h of incubation, protein samples were isolated and were quantified using a protein assay kit (Pierce). Cells were maintained under standard cell culture conditions at 37°C and 5% CO₂ in a humid environment.

Immunoblots.

For measuring the extent of tyrosine nitration on cells, samples were treated with 100 μM 3-morpholinosydnonimine (SIN-1) chloride (Cayman Chemical) for 20 min or 150 μM ONOO⁻ (Cayman Chemicals) twice at 5-min intervals. Protein samples were prepared as previously described (9) and were examined for nitration using anti-nitrotyrosine (NY) (cat. no. 487923-50UG RRID:AB_212231, EMD Chemicals) antibody after immunoprecipitation with anti-Kir6.1 (cat. no. sc-11225 RRID:AB_2130774, Santa Cruz Biotechnology) or anti-SUR2B antibody (cat. no. sc-5793 RRID:AB_2219773, Santa Cruz Biotechnology). Standard protocols for Western blots were performed using site-specific antibodies. Protein samples were run on 5% and 10% SDS-PAGE and transferred on nitrocellulose membranes (Bio-Rad). Membranes were blocked with 3% nonfat milk blocking buffer with 0.1% Tween 20. Specific antibodies were used for blotting at concentrations of 1:200 to 1:1,000. The membranes were then incubated with secondary antibodies at concentrations of 1:1,000 to 1:5,000 and visualized using the Odyssey Infrared Imaging System (Licor Biosciences).

In-cell Western blot assay.

Concentration-response measurements were performed by in-cell Western (ICW) blot assay. Enteric glial cells (EGCs) were obtained from the ATCC (CRL 2690) and cultured and maintained in Dulbecco's modified Eagle's medium/F12 medium supplemented with 0.2 mg/ml geneticin and 10% fetal bovine serum (Invitrogen). EGCs are derived from rat jejunum myenteric plexus. EGCs were stimulated by adding ONOO⁻ at the indicated doses for 15 min at 37°C. For effect of NaHS on the tyrosine nitration, the cells were pretreated with 1 mM NaHS for 15 min at 37°C. ONOO⁻ was removed, and cells were fixed (4% formaldehyde in 1× PBS, 20 min at 25°C), permeabilized (0.1% Triton X-100 in 1× PBS, 4 washes for 5 min each at 25°C with rotation), and blocked [0.5× Odyssey Blocking Buffer (OBB) (Licor Biosciences), 1 h at 25°C with rotation]. Primary antibodies, anti-NY, or anti-glial fibrillary acidic protein (GFAP) (cat. no. sc-9065 RRID:AB_641022, Santa Cruz Biotechnology) were diluted 1:100 in 0.5× OBB (50 μl per well) and allowed to incubate overnight at 4°C. Cells were then washed three times with 1× PBS-Tween (0.1%) for 5 min at 25°C with rotation, followed by the addition of secondary antibody (diluted 1:500 in 0.5× OBB) for 1 h at 25°C. Following three additional wash cycles, wells were aspirated dry and scanned using the Licor Odyssey Scanner (Licor Biosciences). Data were normalized and presented as means ± SE of IR700 responses with reference to controls (100% standard).

Modified biotin switch assay.

The biotin switch assay was carried out as described (6). In brief, cells treated with or without 1 mM NaHS (Cayman Chemical) were homogenized in 250 mM Hepes-NaOH (pH 7.7), 1 mM EDTA, 2.5% SDS, and 0.1 mM neocuproine (HEN buffer) supplemented with 100 μM deferoxamine. Protein samples (250 μg) were added to blocking buffer (HEN buffer adjusted to 2.5% SDS and 20 mM S-methyl methanethiosulfonate) at 50°C for 20 min with frequent vortexing. After acetone precipitation, the proteins were resuspended in HEN buffer adjusted to 1% SDS (HENS buffer). Biotin-N-[6-(biotinamido)hexyl]-3-pyridyldithio)propionamide (4 μM) in dimethyl sulfoxide was added to the suspension. After 3 h of incubation, biotinylated proteins were precipitated by streptavidin-agarose beads, which were then washed with HENS buffer. The biotinylated proteins were eluted by SDS-PAGE sample buffer and subjected to Western blot analysis with specific antibodies to each protein. For quantitation of protein sulfhydration, samples were run on blots alongside total lysates. Anti-SUR2B and anti-Kir6.1 were used at 1:1,000 dilutions.

Site-directed mutagenesis.

The mutants were constructed using the QuikChange Site-Directed Mutagenesis System (Stratagene) according to the manufacturer's procedures. Murine Kir6.1-cDNA (Open Biosystems) inserted into pcDNA3 (Invitrogen) was used as the template. Primers were designed by Vector NTI program (Invitrogen), containing the desired mutation in the middle region (Table 1). Sequences that are underlined indicate site mutated and change in the amino acid [cysteine (C) to serine (S)]. The presence of the desired mutation was confirmed by DNA sequencing of the relevant DNA region.

Table 1.

Primer sequences containing the desired mutations

Target	Forward Primer	Reverse Primer
C6S	GAAATGAGCCTTTCTTTTTCTGGGAACAACATCTCCTCC	GGAGGAGATGTTGTTCCCAGAAAAAGAAAGGCTCATTTC
C24S	GTTCTCCAAAACCCCTCCTTTGTGGACGCACTCAACCTGG	CCAGGTTGAGTGCGTCCACAAAGGAGGGGTTTTGGAGAAC
C263S	GGGCGGTGACCAATTATGTCTCCCTAAAGGAAGCCTATGAAG	CTTCATAGGCTTCCTTTAGGGAGACATAATTGGTCACCGCCC
C706S	GTAGGCCAAGTGGGTTCTGGAAAATCATCTCTTCTCC	GGAGAAGAGATGATTTTCCAGAACCCACTTGGCCTAC
C785S	GGCTGTCACCGATGCCTCCTCTCTTCAGCCAGACATTG	CAATGTCTGGCTGAAGAGAGGAGGCATCGGTGACAGCC
C818S	GTCAGAGGCAGAGAATCTCCGTGGCACGGGCACTCTACC	GGTAGAGTGCCCGTGCCACGGAGATTCTCTGCCTCTGAC
C1410S	CCTGGATCCTGAATGCAAGTCCACAGATGACAGGCTCTGG	CCAGAGCCTGTCATCTGTGGACTTGCATTCAGGATCCAGG
C1455S	GGACAGAGACAGCTGTTCTCCCTGGCCAGGGCCTTTG	CAAAGGCCCTGGCCAGGGAGAACAGCTGTCTCTGTCC

Underlined sequences indicate site mutated and change in the amino acid, with cysteine (C) to serine (S).

Immunohistochemistry.

Immunohistochemical staining was performed by standard protocols. CHO cells were transfected with SUR2B or C1455S SUR2B mutant, with Kir6.1, and treated with 100 μM SIN-1 chloride for 20 min, 1 mM NaHS for 15 min, or a combination of both. After being washed in PBS, cells were fixed in 4% formaldehyde in PBS for 15 min at room temperature and then washed in PBS three times for 5 min each. After formaldehyde fixation, cells were incubated for 10 min in PBS/0.3% Triton X-100 (PBST) following by three washes in PBS. Cells were then blocked for 60 min with 10% nonimmuno goat serum. Subsequently, samples were incubated with anti-NY antibody (1:200) in 1% nonimmuno goat serum overnight at 4°C. Cells were washed in PBS three times and then were treated with the Alexa Fluor 568-conjugated goat polyclonal secondary antibody at 1:1,000 dilutions in PBST for 1 h at room temperature in the dark. Images were collected using an Olympus FV300 confocal microscope (Olympus America) at the appropriate excitation wavelength.

Patch-clamp recording of K_ATP current.

Whole cell K⁺ currents were recorded in transfected cells using standard methods as previously described (6–8). After a seal was obtained, cells were voltage clamped, and K⁺ currents were generated by progressive 10-mV depolarizing steps (500-ms duration, 5-s intervals) from a holding potential of −60 mV. K_ATP currents were measured in a high K⁺ bathing solution (140 mM) and dialyzed with low ATP (0.1 mM) in the internal solution. This protocol allows for inward currents because of K_ATP channel activation by levcromakalim (Tocris Bioscience).

Isometric tension recording.

All animal protocols confirmed to the guidelines and approval by the Virginia Commonwealth University Institutional Animal Care and Use Committee. Tissue preparation was performed as described previously (21). Approximately 1.5-cm strips of mouse ileum were suspended in the longitudinal direction in an organ bath containing calcium-free Krebs solution (118 mM NaCl, 4.6 mM KCl, 1.3 mM NaH₂PO₄, 1.2 mM MgSO₄, 25 mM NaHCO₃, and 11 mM glucose), bubbled continuously with carbogen (95% O₂ and 5% CO₂) at 37°C under a resting tension of 1 g, and equilibrated for a period of 1 h. Isometric contractions were recorded by a force transducer (model no. GR-FT03, Radnoti) connected to a personal computer using Acqknowledge 382 software program (BIOPAC Systems). After equilibration in Krebs solution, tissues were incubated for 45 min in calcium-free high-potassium solution (80 mM) in which equimolar NaCl was replaced by KCl and changed every 15 min. Tissues were preincubated for 20 min either in the presence or absence of NaHS (100 μM). Tissues were washed with Krebs solution after the 20-min preincubation, and cumulative concentration-dependent contraction responses to CaCl₂ (1 μM to 3 mM) were performed. The results were expressed as percentages of the maximal response for CaCl₂ alone, and the curves were statistically compared.

Statistical analysis.

All experiments were performed and repeated at least three times. Results are presented as the means ± SE for the number of cells (n). Sigmaplot 12.5 software was used for detection of differences in paired or unpaired t-tests as appropriate, and EC₅₀ values were calculated using four-parameter logistic nonlinear regression model in Sigmaplot. A P value of <0.05 was considered significant.

RESULTS

Dithiothreitol reduces sulfhydration of SUR2B.

Recently, we reported that, in cells cotransfected with Kir6.1 and SUR2B and in freshly isolated mouse colonic smooth muscle cells, NaH₂S induces sulfhydration of SUR2B as cysteine thiols (-SH) react with H₂S to form hydropersulfides (-SSH) (6). Figure 1 shows the effect of the reducing agent, dithiothreitol (DTT), on K_ATP channel sulfhydration in CHO cells transfected with Kir6.1 and SUR2B. Cell lysates were isolated and treated with 1 mM NaHS, 1 mM DTT, or the combination and subjected to the modified biotin switch assay with antibody against SUR2B to detect K_ATP channel sulfhydration. There was basal sulfhydration of SUR2B (0.20 ± 0.025), and this sulfhydration was significantly enhanced with NaHS treatment (0.35 ± 0.07) (Fig. 1A). There was no significant effect of DTT on the basal sulfhydration of SUR2B subunit. However, treatment with DTT significantly reversed H₂S-mediated sulfhydration (0.21 ± 0.05). H₂S did not induce sulfhydration of Kir6.1 subunit of the channels (6) (not shown). Thus SUR2B sulfhydration is a covalent modification involving sulfhydryl groups. The presence of functional K_ATP channels upon cotransfection of Kir6.1 and SUR2B was confirmed by whole cell voltage clamp in CHO cells. Levcromakalim-induced currents were measured at a holding potential of −60 mV in the presence of high K⁺ solution with low internal ATP to isolate the K_ATP currents as described previously (6). Levcromakalim-induced currents were significantly inhibited by the channel blocker, glibenclamide (Fig. 1B).

Fig. 1.

Sulfhydration is dependent on cysteine residues. A: representative biotin switch assay of sulphonylurea 2B (SUR2B) sulfhydration in Chinese hamster ovary (CHO) cells transfected with Kir6.1 and SUR2B. The bar graph represents the ratio of sulfhydration of SUR2B and total SUR2B and is presented as means ± SE from 3 independent experiments. *Represents significant difference between untreated (left bar) vs. NaHS (1 mM) (left middle bar). Dithiothreitol (DTT) treatment also reduced the sulfhydration by NaHS (1 mM) (right bar). WB, Western blot. B: representative levcromakalim (LevK)-induced whole cell K⁺ currents recording in a single cell. The currents were recorded from CHO-transfected cells. LevK currents were recorded in the presence of high external K⁺ and low intracellular ATP at −60 mV holding potential. Glib, glibenclamide.

Cysteine residues affect sulfhydration.

To determine the specific cysteine residues within SUR2B that are sulfhydrated, a preliminary analysis of the possible cysteine residues was carried out using the program PSIPRED, and the mutants were constructed using the QuikChange Site-Directed Mutagenesis System. The following cysteine residues were identified as most likely to be involved, based on the preliminary analysis: C6, C24, C263, C706, C818, C1410, and C1455. CHO cells were transfected with Kir6.1 and SUR2B or mutants of SUR2B and subjected to the modified biotin switch assay with antibody against SUR2B to detect K_ATP channel sulfhydration following NaHS. Figure 2 shows that mutation of cysteine 24 (C24S) and 1455 (C1455S) reduced sulfhydration, C24S (0.12 ± 0.023) and C1455S (0.12 ± 0.035) compared with control (0.31 ± 0.033) (Fig. 2). The other mutants did not affect the extent of sulfhydration of the channel.

Fig. 2.

Cysteine residues on SUR2B affect sulfhydration. Biotin switch assay of SUR2B sulfhydration in CHO cells transfected with Kir6.1 and SUR2B mutants. Top: bar graph represents the ratio of sulfhydrated SUR2B and total SUR2B and is presented as means ± SE from 3 independent experiments. *Presents significant difference between control (CTL) vs. mutants. Both C24S and C1455S were significantly different from control. Bottom: representative blot for wild-type and all mutants.

Sulfhydration prevents tyrosine nitration.

An interaction between tyrosine nitration and H₂S has been suggested in neuronal cells (26), in interstitial cells of Cajal, and in processes involving apoptosis/necrosis and poly(ADP-ribose)polymerase 1 activation (4, 27, 30). We examined the interaction of H₂S-induced sulfhydration with tyrosine nitration in the K_ATP channel complex. In CHO cells cotransfected with the two subunits, treatment with SIN-1 and ONOO⁻ resulted in nitration of Kir6.1 but not SUR2B (Fig. 3, A and B). Immunoprecipitation with anti-Kir6.1 followed by immunoblot with anti-NY revealed that the extent of nitration of Kir6.1 was significantly reduced by pretreatment with NaHS (1 mM). NaHS treatment did not affect the basal level of nitration of Kir6.1. The reduction in nitration by NaHS was also evident when examined by immunohistochemistry. Cells cotransfected with Kir6.1 and SUR2B or the C1455S were treated with SIN-1 or NaHS or the combination. Tyrosine nitration was readily observed in cells treated with SIN-1 but markedly reduced in the presence of NaHS (Fig. 3C). To further confirm that SUR2B subunit is the potential target for sulfhydration, the C1455S mutant was cotransfected with Kir6.1 and examined for tyrosine nitration following sulfhydration. As shown in Fig. 3C, bottom, mutation of C1455 resulted in robust expression of tyrosine nitration, indicative of the requirement of sulfhydration of C1455 for the reduction in tyrosine nitration. The effect of NaHS was further evaluated in EGCs by ICW blot assay. Tyrosine nitration was evaluated over a 1,000-fold concentration range. We also examined the level of GFAP, marker for EGCs, to establish the validity of cells in each well. Significant increase in total cellular tyrosine nitration in EGCs was observed at the concentration of 1 and 3 mM ONOO⁻ stimulation (1,054 ± 101-fold and 516 ± 49-fold, respectively, n = 4) (Fig. 3D). Pretreatment with NaHS reduced the level of nitration in the EGCs to 181 ± 19-fold (1 mM ONOO⁻) and 84 ± 23-fold (3 mM ONOO⁻), respectively (n = 4, ***P ≤ 0.001). The expression of GFAP remained constant across the concentration range. These data suggest that NaHS may have the potential to inhibit tyrosine nitration formation.

Fig. 3.

Tyrosine nitration is attenuated by H₂S. A and B: representative immunoblot analyses of tyrosine nitration proteins in CHO-transfected cells. Cells were lysed in immunoprecipitation assay buffer (RIPA), and Kir6.1 (A) or SUR2B (B) was immunoprecipitated (IP) using anti-Kir6.1 or anti-SUR2B and immunoblotted using anti-nitrotyrosine antibody (NY). The bar graph represents the ratio of nitration of Kir6.1 (A) and SUR2B (B) against total Kir and SUR2B, respectively (means ± SE from 3 independent experiments). *Significant difference between 3-morpholinosydnonimine (SIN-1)-induced nitration of Kir6.1 and in the presence of NaHS. C: representative immunostaining for tyrosine nitration in CHO-transfected cells. CHO cells were transfected with Kir6.1 and SUR2B or C1455S and treated with 100 μM SIN-1 for 20 min, 1 mM NaHS for 10 min, or a combination of both. Cells were visualized by fluorescence confocal microscopy. Scale bar = 20 μm. D: in-cell Western of enteric glial cells treated with the indicated doses of peroxynitrite (ONOO⁻) for 15 min in the absence or presence of NaHS (1 mM, 15 min) and processed for imaging (bottom), as described in materials and methods. The image shows whole cell fluorescence from actual wells of a 96-well plate, highlighting the increase in tyrosine nitration signal. The signal is background subtracted and normalized to control. Bar graph (top) shows the fold induction of tyrosine nitration measured as intensity over the control (absence of ONOO⁻). Values represent the means ± SE. ***Significant difference between ONOO⁻-treated vs. NaHS + ONOO⁻-treated cells. GFAP, glial fibrillary acidic protein.

Mutation of cysteine residues prevents nitration.

The above data suggest that Kir6.1 is the potential target for tyrosine nitration, whereas cysteine residues in SUR2B subunit are sulfhydrated, and that sulfhydration prevents tyrosine nitration. To test this, the cysteine residues C24 and C1455 that were identified as potential sulfhydration sites were mutated to serine residues. In CHO cells cotransfected with SUR2B, wild-type, C263S (control), C24S, and C1455S with Kir6.1 were examined for prevention of nitration by NaHS. Figure 4, A and B, shows that SIN-1-induced nitration of Kir6.1 is reduced by NaHS in wild-type and in C263S SUR2B mutant but not in C24S or C1455S mutants. None of the mutants of SUR2B were nitrated by SIN-1 (Fig. 4C). We next examined whether tyrosine nitration of Kir6.1 could be reversed by NaHS. Enhanced tyrosine nitration of Kir6.1 was detected with SIN-1 treatment (2.024 ± 0.58), but, interestingly, NaHS did not reduce postnitration of Kir6.1 (2.11 ± 0.26) (Fig. 4D). These data suggest that sulfhydration of SUR2B is required for reducing nitration.

Fig. 4.

Tyrosine nitration of Kir6.1 and the effect of cysteine mutations of SUR2B. A and B: representative immunostaining for tyrosine nitration in CHO-transfected cells. Wild-type or mutants of SUR2B were transiently expressed, immunoprecipitated using Kir6.1 antibody, and then immunoblotted for tyrosine nitration of Kir6.1 using anti-NY antibody. The bar graph represents normalized data for tyrosine nitration and is presented as means ± SE from 3 independent experiments. *Significant differences between CTL vs. SIN-1 and SIN-1 vs. NaHS + SIN-1. Cells were treated with NaHS (150 μM, 15 min), SIN-1 (100 μM, 20 min), or both. B: representative Western blots of immunoprecipitated Kir6.1 and immunoblotted with anti-NY. C: representative immunoblot analyses of tyrosine nitration of SUR2B proteins in CHO-transfected cells. Wild-type or mutants of SUR2B were transiently expressed, treated with SIN-1 (100 μM, 20 min), immunoprecipitated using anti-SUR2B, and immunoblotted using antibody for tyrosine nitration proteins (NY). D: representative SIN-1- induced tyrosine nitration of Kir6.1 is not reversed by NaHS after tyrosine nitration. Kir6.1 and wild-type SUR2B were transiently expressed and treated with SIN-1 (100 μM, 20 min) followed by NaHS (150 μM, 15 min). The cell lysates were immunoprecipitated using Kir6.1 antibody and then immunoblotted for tyrosine nitration. The bar graph represents normalized data for tyrosine nitration and is presented as means ± SE from 3 independent experiments. *Significant difference between wild-type and SIN-1.

Sulfhydration prevents Ca²⁺ channel nitration.

Previously, we have shown that L-type calcium channels in colonic smooth muscle cells are tyrosine nitrated during colonic inflammation, resulting in reduced currents and inhibition of contractions (21). Cav1.2 channels in rat cardiomyocytes have been suggested to be sulfhydrated (31). We therefore examined whether tyrosine nitration of Cav1.2b is reduced by NaHS. Figure 5 shows that pretreatment with NaHS reduced the nitration of Cav1.2.

Fig. 5.

Sulfhydration prevents Ca²⁺ channel nitration. Bar graph represents nitration of calcium channel in CHO-transfected cells. Ca²⁺ channel (Cav1.2b) was transiently expressed and treated with NaHS (150 μM, 15 min), SIN-1 (100 μM, 20 min), or NaHS followed by SIN-1. Protein was immunoprecipitated using Cav1.2b antibody and then immunoblotted for tyrosine nitration. The bar graph represents normalized data for tyrosine nitration and is presented as means ± SE from 3 independent experiments. *Significant differences between NaHS and SIN-1.

H₂S prevents ONOO⁻-induced inhibition of CaCl₂-induced contraction.

To further evaluate the physiological role of H₂S on tyrosine nitration, CaCl₂-induced contractions of the isolated mouse ileum were measured in the presence and absence of ONOO⁻. Mouse ileum tissues were incubated in a depolarizing calcium-free, high-potassium (80 mM) physiological saline solution. A cumulative concentration response to CaCl₂ was established in control tissues. Preincubation with ONOO⁻ (150 μM) for 10 min significantly reduced the calcium-induced contraction and shifted the dose-response curve to the right (Fig. 6). Tissues preincubated with NaHS (100 μM) for 15 min followed by ONOO⁻ resulted in a significant recovery of CaCl₂-induced contractions. The EC₅₀ values for CTL, H₂S + ONOO⁻, and ONOO⁻ were 143.3 ± 15.5 μM, 197.3 ± 23.5 μM, and 270.2 ± 23.7 μM, respectively. These data demonstrate that the inhibition of the calcium influx under depolarizing conditions by ONOO⁻ is reduced by NaHS.

Fig. 6.

Effect of ONOO⁻ or H₂S + ONOO⁻ on the cumulative CaCl₂-induced concentration-dependent contractions in isolated mouse ileum. Maximal contractions of control group (CTL) were considered as 100%. The CaCl₂ contractions were elicited in tissues depolarized by high-potassium (80 mM) physiological saline solution. A: representative graph showing the cumulative CaCl₂ concentration-response curves by ONOO⁻ and H₂S + ONOO⁻ in mouse ileum. B: representative tracing of CaCl₂-induced concentration-dependent contractions in isolated ileum from mouse. Points represent means ± SE. Data were analyzed by repeated-measures ANOVA followed by Bonferroni's posttest. *P ≤ 0.05 compared with ONOO⁻. #P ≤ 0.05, ##P ≤ 0.01 vs. ONOO⁻ compared with CTL, n = 3–4.

DISCUSSION

H₂S is emerging as a highly significant endogenous gaseous mediator and potential target for pharmacological manipulation (5, 15). It is known to have several intracellular targets, of which K_ATP channels represent a significant target for H₂S, particularly in colonic inflammation. Previously, we have shown that the K_ATP channel opener levcromakalim and NaHS-induced currents are enhanced in a 4,6-trinitrobenzene sulfonic acid model of colonic inflammation. An enhancement in the S-sulfhydration of the K_ATP channel subunit SUR2B is also seen after colonic inflammation, which may allow for the allosteric modulation of the channel. In the present study, we examined the potential sites within the K_ATP channel complex that are sulfhydrated and determined the interaction with nitration of tyrosine residues. We have found that 1) NaHS significantly enhanced sulfhydration of SUR2B but not Kir6.1 subunit; 2) SIN-1 (a donor of ONOO⁻) induced nitration of Kir6.1 subunit but not SUR2B; 3) two specific mutations within SUR2B, C24S and C1455S, prevented sulfhydration by NaHS, and these mutations prevented NaHS-induced reduction in tyrosine nitration of Kir6.1; 4) NaHS also reversed ONOO⁻-induced inhibition of smooth muscle contraction.

Ion channel activity can be significantly altered by the intracellular redox state maintained by reactive oxygen species (ROS) and reactive nitrogen species (RNS). Physiological generation of these compounds is highly regulated; however, an imbalance occurs in pathophysiological conditions, such as in colonic inflammation. Generated ROS/RNS can directly induce posttranslational modifications at specific residues and alter the ion channel function. These include redox modification of cysteine residues by ROS, S-nitrosylation by nitric oxide, and/or nitration of tyrosine residues by ONOO⁻. Our studies demonstrate that, within a single protein complex involving two subunits, there is significant specificity, as sulfhydration occurred in the SUR2B but not Kir6.1, whereas nitration occurred in Kir6.1 but not SUR2B. Since the discovery of endogenous H₂S, many studies have explored the roles of this gas in animals and humans and have shown H₂S to be involved in diverse physiological and pathophysiological processes. However, our understanding of the pharmacology and the pathophysiological roles of H₂S in the gastrointestinal (GI) tract is still quite rudimentary. There is growing evidence that, during colonic inflammation, including ulcerative colitis, there is an increased production of H₂S in the GI tract (23) that may have protective effects during colitis. In the present study, we demonstrate that sulfhydration of one subunit (SUR2B) of the K_ATP channel prevents the tyrosine nitration in the other subunit (Kir6.1). We also show that H₂S sulfhydrates SUR2B at C24 and C1455 residues and causes the reversal of tyrosine nitration of the Kir6.1 in the K_ATP channel complex. This suggests that sulfhydration of cysteine residues of SUR2B on either side of the membrane is required to alter the tyrosine nitration of Kir6.1, as C24 is localized on the extracellular side, whereas C1455 is localized toward the intracellular side. The specific mechanism by which these cysteine residues affect modulation of the K_ATP currents requires further work, but it is noteworthy that K_ATP channel activity is markedly enhanced in a model of experimental colitis (8). In addition to the allosteric effects on channel function, the present studies show that sulfhydration also appears to reduce tyrosine nitration of the pore-forming subunit Kir6.1. Because postnitration of Kir6.1 is not affected by NaHS, it would appear that the order of exposure to the free radicals may be critical in the regulation of the ion channel activity.

During colonic inflammation, reduced muscle contractility occurs partly as a result of decreased calcium influx through L-type calcium channels and enhanced K_ATP currents (2). Calcium channels along with the K_ATP play an important role in regulating the excitability of the smooth muscle cells. Previous work has shown that nitration of the tyrosine residues within the COOH terminus of Cav1.2 prevents Src-kinase-mediated tyrosine phosphorylation, leading to decreased calcium currents (9–12, 20, 21). Cav1.2 channels can also be sulfhydrated (31). The inhibition by tyrosine nitration of calcium-induced contractions that occurs as a result of calcium influx under depolarizing conditions is reduced by NaHS. Thus, similar to the findings for the K⁺ channel, sulfhydration may also reduce tyrosine nitration in Cav1.2. However, further works needs to done to identify the cysteine residues involved in the sulfhydration of the Cav1.2. Occurrence of this interaction in the K_ATP channel complex in transfected CHO cells and EGCs and in Cav1.2 in transfected CHO cells demonstrates the ubiquitous nature of this phenomenon. This unique interaction within a protein complex may provide a mechanistic basis by which H₂S affords a protective role in colonic dysmotility during colitis. Enhanced production of endogenous H₂S and that via resident luminal bacteria may compete with the RNS produced during colonic inflammation. ONOO⁻ is enhanced during colonic inflammation as a result of the uncoupling of nitric oxide synthase (NOS) and enhanced expression of inducible NOS (3).

Ion channels play a central role in the regulation of cell excitability; therefore, altered regulation of ion channels as a consequence of protein modification is of significance, not only for colonic inflammation, but also in other disease states and may help provide novel approaches in channel-targeted therapeutics. Our findings that sulfhydration of the K_ATP channel prevents the detrimental effects of tyrosine nitration within a single ion channel complex provide a novel basis to examine the antioxidant properties of H₂S. (Fig. 7).

Fig. 7.

A schematic of the proposed protective effects of H₂S in colonic inflammation. Exogenous H₂S enhances sulfhydration of 2 cysteine residues (24 and 1455) within the SUR2B subunit of ATP-sensitive K⁺ (K_ATP) channels, which prevents tyrosine nitration within the Kir6.1 subunit of the K_ATP channel complex.

GRANTS

This work was supported by NIH DK046367.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

Author contributions: M.K. and H.I.A. conception and design of research; M.K., A.H., and A.R.G. performed experiments; M.K., A.H., and A.R.G. analyzed data; M.K., A.H., A.R.G., and H.I.A. interpreted results of experiments; M.K. and A.R.G. prepared figures; M.K. and H.I.A. drafted manuscript; M.K. and H.I.A. edited and revised manuscript; M.K., A.H., and H.I.A. approved final version of manuscript.