Reactive oxygen species (ROS), including hydrogen peroxide (H2O2), have recently been shown to be generated upon agonism of several members of the G protein-coupled receptor (GPCR) superfamily, including β2-adrenergic receptors (β2ARs). Previously, we have demonstrated that inhibition of intracellular ROS generation mitigates β2AR signaling, suggesting that β2AR-mediated ROS generation is capable of feeding back to regulate receptor function. Given that ROS, specifically H2O2, are able to post-translationally oxidize protein cysteine sulfhydryls to cysteine-sulfenic acids, the goal of the current study was to assess whether ROS are capable of S-sulfenating β2AR. Using a modified biotin-switch assay that is selective for cysteine-sulfenic acids, our results demonstrate for the first time that H2O2 treatment facilitates S-sulfenation of transiently overexpressed β2AR in human embryonic kidney 293 cells. It is noteworthy that stimulation of cells with the β-agonist isoproterenol produces both dose- and time-dependent S-sulfenation of β2AR, an effect that is receptor-dependent, and demonstrates that receptor-generated ROS are also capable of oxidizing the β2AR. Receptor-dependent S-sulfenation was inhibited by the chemoselective sulfenic acid alkylator dimedone and the cysteine antioxidant N-acetyl-l-cysteine. Moreover, our results reveal that receptor oxidation occurs in cells that endogenously express physiologically relevant levels of β2AR, because treatment of human alveolar epithelial A549 cells with either H2O2 or the β2-selective agonist formoterol promoted receptor S-sulfenation. These findings provide the first evidence, to our knowledge, that a mammalian GPCR can be oxidized by S-sulfenation and signify an important first step toward shedding light on the overlooked role of ROS in the regulation of β2AR function.
The β2-adrenergic receptor (β2AR) is the prototypical and best characterized member of the G protein-coupled receptor (GPCR) superfamily, mediating physiological responses upon binding of the catecholamines epinephrine and norepinephrine. Agonism of β2AR facilitates the generation of intracellular second messengers, most notably, cAMP, which activates protein kinase A, leading to cellular responses. Termination of agonist-occupied receptor signaling occurs upon phosphorylation of β2AR by members of the G protein-coupled receptor kinase family, facilitating receptor desensitization and initiating a cascade of G protein-independent signaling through the recruitment of β-arrestin partner proteins. Although cAMP and its effectors, as well as β-arrestins, have been viewed as chief mediators of cellular responses upon β2AR activation, our laboratory has been examining a novel aspect of β2AR signaling, specifically, the role of reactive oxygen species (ROS) in the regulation of β2AR function.
ROS, which include hydrogen peroxide (H2O2), superoxide (O2−), and hydroxyl radical (OH•), have traditionally been viewed as cytotoxic byproducts of cellular respiration and metabolism. However, a growing body of evidence has also demonstrated that ROS play central roles in transducing intracellular signal events because of their high reactivity within the cellular milieu. In this regard, it is well described that ROS can post-translationally modify proteins through modification of cysteine sulfhydryl (-SH) groups, yielding oxidized cysteine-sulfenic acids (-SOH). This reversible post-translational modification can lead to formation of higher-order redox states such as sulfinic (-SO2H) or sulfonic (-SO3H) acids, disulfides (S-S), or upon reaction with nitrogen species, S-nitrosothiols (-SNO), any of which can lead to altered protein function (Hess et al., 2005; Charles et al., 2007; Reddie and Carroll, 2008).
We have demonstrated that agonism of the β2AR facilitates generation of intracellular ROS, in particular superoxide and hydrogen peroxide (H2O2) (Moniri and Daaka, 2007). These results have been corroborated by other groups using cardiomyocyte models of cardiomyopathy and heart failure, suggesting that the ROS-β2AR link is physiologically significant and, if unregulated, may be involved in pathophysiological states (Li et al., 2010; Xu et al., 2011). It is noteworthy that the agonism of β2AR in human neutrophils facilitates a decrease in formyl-Met-Leu-Phe-mediated ROS generation (Opdahl et al., 1993; Anderson et al., 1996; Barnett et al., 1997), and others have shown that β2AR agonism specifically decreases only extracellular ROS from these cells (Kopprasch et al., 1997). Together, these results suggest that β2AR agonism may have cell type-specific differential effects on modulating ROS levels. Our previous results demonstrate that β2AR-mediated ROS generation was abrogated by the ROS scavenger and cysteine residue antioxidant N-acetyl-l-cysteine (NAC), the flavin-containing (NADPH) oxidase inhibitor diphenyliodonium chloride, and the Rac1 inhibitor NSC23766 [N6-[2-[[4-(diethylamino)-1-methylbutyl]amino]-6-methyl-4-pyrimidinyl]-2-methyl-4,6-quinolinediamine trihydrochloride], suggesting that β2 receptors can signal through one of the Rac1-dependent NADPH oxidase isoforms to generate intracellular ROS (Moniri and Daaka, 2007). Other results confirm this view, whereby β2AR agonism was shown to stimulate ROS generation via β-arrestin/Rac1-dependent activation of NADPH oxidases (Gong et al., 2008). Most importantly, our previous work demonstrates that inhibition of ROS markedly abrogates β2AR-mediated cAMP formation and protein kinase A activity, as well as receptor phosphorylation and internalization, but it does not affect ligand binding (Moniri and Daaka, 2007). These results suggested that oxidants are indispensible for receptor signaling, and ROS are likely able to regulate β2AR function. Based on these aggregate data, we hypothesized that ROS that are generated upon β2AR agonism can feed back to oxidize the receptor, and this modification allows β2AR to maintain functionally competent conformations that allow for proper signaling. Because ROS such as H2O2 are capable of oxidizing cysteine residues, initially forming sulfenic acids, we hypothesized that one or more β2AR cysteine residues could be oxidized to cysteine sulfenic acids. In this study, we have developed a modified biotin-switch assay to determine whether exogenous oxidant (i.e., H2O2) or receptor agonism could mediate direct oxidation of the β2AR. Our results demonstrate, for the first time, that ROS are capable of S-sulfenating the β2AR and that receptor agonism also facilitates this oxidative modification.
Materials and Methods
Chemicals and Reagents.
Isoproterenol hydrochloride and (S)-propranolol were obtained from MP Biomedicals (Solon, OH) and Sigma-Aldrich (St. Louis, MO), respectively. (±)-Formoterol hemifumarate was obtained from Tocris Bioscience (Ellisville, MO). Dimedone (5,5-dimethyl-1,3-cyclohexanedione), N-acetyl-l-cysteine, and 9.7 M H2O2 were obtained from Sigma-Aldrich. All other chemicals used were obtained at the highest available purity from Thermo Fisher Scientific (Waltham, MA) or Sigma-Aldrich. The FLAG-epitope tagged pcDNA3-FLAG-β2AR construct (Tang et al., 1999) was a kind gift from Dr. Robert J. Lefkowitz (Duke University Medical Center, Durham, NC).
Cell Culture, Transfection, and Treatment.
Cell growth media and supplements were obtained from Thermo Fisher Scientific. Human embryonic kidney (HEK) 293 cells from the American Type Culture Collection (Manassas, VA) were grown at 37°C in a humidified atmosphere of 5% CO2 in Dulbecco's modified Eagle medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin. At approximately 90% confluence, cells grown in 100-mm dishes were transfected with 5 μg of pcDNA3-FLAG-β2AR plasmid using LipoD293 according to the manufacturer's instructions (SignaGen, Rockville, MD). Cells were subcultured accordingly on collagenized (rat tail type I) six-well plates at a density of approximately 2 × 106 cells/well for all experiments and serum-starved overnight before assay. The human alveolar epithelial cell line A549, which endogenously expresses β2AR, was a kind gift from Dr. Philip Moos (University of Utah, Salt Lake City, UT). A549 cells were cultured exactly as described above except they were grown to 70% confluence in 100-mm dishes and serum-starved for 4 h for all experiments (Szentendrei et al., 1992).
Sulfenic Acid-Selective Biotin-Switch Assay.
Cells were treated with the appropriate agents for the indicated times. For propranolol experiments, cells were preincubated with 10 μM (S)-propranolol for 5 min before agonist stimulation. Micromolar concentrations were used to ensure that biased β-arrestin agonism, which has been demonstrated to occur by propranolol at the related β1AR, did not contribute to S-sulfenation. For experiments with NAC, cells were preincubated with 10 mM NAC in starvation media for 4 h, as described previously (Moniri and Daaka, 2007). After stimulation, cells were immediately placed on ice and washed twice with ice-cold phosphate-buffered saline. Whole cells were resuspended in ice-cold phosphate-buffered saline and collected by centrifugation at 3000 rpm for 8 min at 4°C. Cells were then lysed in HENS buffer (25 mM HEPES, pH 7.6, 1 mM EDTA, 1% SDS, and 0.01 mM neocuproine), homogenized, and cleared of cellular debris by centrifugation at 14,000g for 15 min. Protein concentrations were standardized by DC Protein Assay (Bio-Rad Laboratories, Hercules, CA), and the cysteine-reactive methylator methyl methanethiosulfonate (MMTS) (Thermo Fisher Scientific) was added (20 mM) to methylate free cysteine sulfhydryl groups, thereby protecting them from further chemical modification, as performed for the detection of S-nitrosylation (Fig. 1, step 1) (Jaffrey et al., 2001). SDS was also added to a final concentration of 2.5% (v/v) to ensure protein denaturation, and lysates were incubated at 50°C for 30 min with frequent agitation followed by acetone precipitation of the protein. Acetone-precipitated proteins were resuspended in HENS buffer, and 20 mM sodium arsenite (RICCA, Arlington, TX) was added to selectively reduce cysteine sulfenic acids (-SOH) groups back to the cysteine sulfhydryl (-SH), as demonstrated by Saurin et al. (2004) (Fig. 1, step 2). The ability of NaAsO2 to selectively reduce sulfenic acids, and not higher-order sulfinic acids (-SO2H), sulfonic acids (-SO3H), disulfides (S-S), or S-nitrosothiols (-SNO), provides sulfenic acid selectivity to this assay (Saurin et al., 2004). Reduction of sulfenic acids to free sulfhydryls, coupled with MMTS-mediated protection of initial sulfhydryls, allows for selective biotinylation of the formerly oxidized sulfenic acids with the sulfhydryl-reactive biotinylating agent biotin-HPDP (400 μM) (Thermo Fisher Scientific) (Fig. 1, step 3), as described for biotin-switch assays that detect protein S-nitrosothiols (Jaffrey et al., 2001). For experiments testing the specificity of S-sulfenation, 5 mM dimedone was included per reaction for 30 min before biotin switch, as described by others (Saurin et al., 2004). After biotin switch of sulfenic acids, lysates were acetone-precipitated and resuspended in HENS/10 (1:10 HENS/H2O), and an aliquot was reserved to determine reaction protein input. The remaining sample was neutralized with neutralization buffer (20 mM HEPES, pH 7.6, 1 mM EDTA, 100 mM NaCl, and 0.5% Triton X-100), and 30 μl of streptavidin-agarose (Thermo Fisher Scientific) was added to pull down biotinylated proteins (Fig. 1, step 4). After overnight tumbling at 4°C, beads were washed three times in wash buffer (20 mM HEPES pH 7.6, 1 mM EDTA, 600 mM NaCl, and 0.5% Triton X-100), and proteins were eluted in Laemmli buffer with 2.5% 2-mercaptoethanol. Samples, as well as the corresponding reserved inputs, were subjected to SDS-polyacrylamide gel electrophoresis, transferred to polyvinylidene difluoride membranes, and immunoblotted with anti-FLAG antibody (M2; Sigma-Aldrich) for transfected HEK293 cells or with anti-β2AR antibody (GeneTex Inc., Irvine, CA) for endogenous β2AR in A549 cells (Fig. 1, step 5). Immunoblotted reaction bands correspond to “biotin-switched” S-sulfenated FLAG-β2AR or β2AR and were detected in the appropriate range (∼50 kDa) similar to input lysates by using enhanced chemiluminescence.
Resulting data were quantified by densitometric analysis using Image J (National Institutes of Health, Bethesda, MD) and Prism 3.0 (GraphPad Software, Inc., San Diego, CA). Data are expressed as mean ± S.E.M for representative experiments repeated at least three independent times. Where not visible, error bars fall within the symbol size. Statistical analysis was performed, as appropriate, either by one-way analysis of variance and post hoc Tukey's test or by Student's t test using GraphPad Instat.
H2O2 and Isoproterenol Stimulate Sulfenic Acid Oxidation of β2AR.
Our previous results suggested that agonism of β2AR with the nonselective β-receptor agonist isoproterenol (ISO) leads to marked generation of intracellular ROS. Because inhibition of ROS abrogated β2AR function, we hypothesized that ROS were capable of directly regulating the β2AR. Given that ROS are well known to modify protein cysteine residues by oxidizing them, at least initially, to sulfenic acids, we hypothesized that ROS could directly oxidize the β2AR, yielding sulfenic acids. To test this hypothesis, we used the biotin-switch assay shown in Fig. 1, which had been modified for selective detection of sulfenic acids. As shown in Fig. 2A, treatment of HEK293 cells transiently expressing FLAG epitope-tagged β2AR with exogenous ROS in the form of H2O2 (100 μM) yielded rapid and marked oxidation of β2AR. Full-length immunoblots (Supplemental Figures) demonstrate that the biotin-switch signal detected with the anti-FLAG M2 antibody is highly selective for a FLAG-immunoreactive band at approximately 50 kDa, which represents the expected size of FLAG-β2AR. Nonspecific bands in the 95- to 180-kDa range appeared in an exposure length-dependent manner and are likely to represent aggregated receptors, an effect that is highly typical upon heating and elution of GPCRs in SDS-sample buffer (Supplemental Figures). The H2O2-induced S-sulfenation signal peaked at 1 min after the addition of the oxidant and yielded a 350 ± 20% increase in oxidation compared with untreated control cells (p < 0.001 compared with control) (Fig. 2, A and B; Supplemental Fig. 1). Moreover, H2O2-mediated oxidation of β2AR was sustained through the time course of treatment (5–30 min) but began to decline to basal levels after 30 min (Fig. 2, A and B; Supplemental Fig. 1). β2AR S-sulfenation was also detected at lower H2O2 concentrations, and it is noteworthy that, although treatment with H2O2 for 1 to 30 min did not affect cell viability, treatment for 60 min often facilitated detachment of approximately 30% of HEK293 cells (data not shown), which is likely promoted by the relatively weak adherence of this cell line. Furthermore, control experiments in which biotin labeling was performed in the absence of NaAsO2 demonstrated no appreciable biotin-switch signal (data not shown).
Because our and others' previous results have shown that agonism of β2AR with ISO promotes robust generation of intracellular ROS, we examined whether agonism of β2AR with ISO could facilitate sulfenic acid oxidation of β2AR, similar to that seen upon treatment with exogenous H2O2. Results in Fig. 2, C and D, as well as Supplemental Fig. 2, demonstrate that agonism of β2AR with ISO (10 μM) facilitates rapid and significant sulfenic acid oxidation of β2AR. After 1 min of agonism, receptor sulfenic acid oxidation was 175 ± 22% of control, and this increased to 252 ± 24% of control after 5 min of agonism (p < 0.01 for both time points versus unstimulated control). Unlike that seen with H2O2, the ISO-mediated sulfenic acid oxidation of β2AR peaked 15 min after agonism, yielding a 306 ± 25% increase in oxidation compared with untreated control cells (p < 0.001 versus unstimulated control), and decreased toward basal levels thereafter (114 ± 5 and 128 ± 11% of control at 30 and 60 min, respectively) (Fig. 2, C and D; Supplemental Fig. 2). Significantly, this timeframe correlates precisely with our previous results that showed that the height of ISO-mediated generation of intracellular ROS occurs within 20 min of receptor agonism and declines thereafter (Moniri and Daaka, 2007) and suggests that agonist-mediated ROS generation within this timeframe modulates β2AR S-sulfenation. As seen in full-length immunoblots in Supplemental Figs. 1 and 2, the anti-FLAG immunoreactivity was similar in both H2O2- and ISO-treated conditions, with only the major β2AR-specific band (approximately 50 kDa) and the nonspecific aggregation toward the top of the blot (95–180 kDa) being recognized. To establish whether the ISO-mediated receptor oxidation is dose-dependent, we performed the biotin-switch assay with increasing concentrations of ISO. Initial dose responses were performed at the peak time point seen in Fig. 2, C and D (i.e., 15 min); while a lower oxidative signal was revealed at concentrations of 10 to 100 nM, a strong signal was revealed at micromolar concentrations, suggesting that the oxidative signal could be saturated in the presence of agonist during this duration (data not shown). Meanwhile, a potent and saturable dose-dependent sulfenic acid oxidation of β2AR was observed after 5 min of agonism with ISO (Fig. 2, E and F; Supplemental Fig. 3), demonstrating a dose-response relationship at earlier time points. Nonlinear regression analysis of the dose-response graph resulted in a pEC50 of −7.4 ± 0.3 (R2 = 0.90), reaching plateau at 1 μM (365 ± 10% of control; p < 0.001 versus unstimulated control). It is noteworthy that the ISO dose response produces a shallow curve with a calculated Hill slope of approximately 0.7, implying that agonist cooperativity may play a role in the S-sulfenation effect. Although our previous data demonstrated that inhibition of ROS does not affect agonist binding to the β2AR in membrane preparations (Moniri and Daaka, 2007), results presented here cannot discount the possibility that agonist-mediated receptor S-sulfenation could indeed affect binding of subsequent agonist molecules in situ. Moreover, S-sulfenation can be influenced by intracellular ROS pools, which are increased by β2AR agonism in HEK293 cells, as well as by favorable reducing conditions that constitutively exist inside cells. These factors could contribute a ROS-specific component to the extended ternary model of agonist binding and thereby possibly influence dose-response relationships in regard to any particular agonist's potential to induce receptor S-sulfenation.
Isoproterenol-Mediated β2AR S-Sulfenation Is Inhibited by β2AR Antagonism.
Our previous results demonstrated that β2AR-mediated ROS generation directly depended on receptor agonism, because it was abolished in the presence of the β-receptor antagonist propranolol (Moniri and Daaka, 2007). To determine whether β2AR S-sulfenation was receptor-dependent, we performed the biotin-switch assay in cells pretreated for 5 min with (S)-propranolol in the absence or presence of ISO. (S)-propranolol alone had no significant effect on stimulating β2AR S-sulfenation throughout a 60-min time course, as demonstrated in Fig. 3A and Supplemental Fig. 4. In the presence of ISO, (S)-propranolol significantly abrogated the ISO-mediated oxidation of β2AR at 1, 5, and 15 min to 110 ± 20, 89 ± 7, and 86 ± 4% of control, respectively (Fig. 3, B and C; Supplemental Fig. 5) (p <0.05, <0.001, and <0.001 versus the ISO-stimulated conditions, respectively). These results demonstrate that receptor agonism facilitates the oxidation of the β2AR and suggest that β2AR-mediated ROS generation, which peaks within 20 min of agonism, is capable of facilitating S-sulfenation of the receptor.
β2AR S-Sulfenation Is Inhibited by Dimedone and N-Acetyl-l-Cysteine.
To confirm the nature of the β2AR oxidative product, we used the cyclic diketone nucleophile 5,5-dimethyl-1,3-cyclohexanedione (dimedone), which selectively and irreversibly reacts with S-sulfenic acids, forming a covalent adduct that prevents reduction of the sulfenic acid group by sodium arsenite (Benitez and Allison, 1974; Ellis and Poole, 1997; Saurin et al., 2004). In the presence of dimedone, the biotin-switch signal would be expected to be attenuated because of nucleophillic attack and “trapping” of the S-sulfenic acid by dimedone. As shown in Fig. 4, ISO-mediated S-sulfenation (262 ± 10% at 10 μM for 15 min) was significantly abrogated by dimedone treatment (105 ± 13.5%) (p < 0.01 versus ISO) (Fig. 4, A and B; Supplemental Fig. 6). Likewise, H2O2-mediated S-sulfenation (376 ± 9%) was also attenuated in the presence of dimedone (142 ± 35%) (p < 0.05 versus H2O2) (Fig. 4, A and B; Supplemental Fig. 6); however, complete abolishment of this signal was not seen, likely because of overoxidation, which becomes insensitive to dimedone upon treatment with H2O2, as described by others (Gutscher et al., 2009). These results demonstrate that the β2AR oxidant species induced by receptor agonism and exogenous oxidant treatment is indeed a sulfenic acid, at least for an initial stable period. We also used the membrane-permeable thiol antioxidant NAC, which is extensively used in redox studies to assess the effect of thiol excess in oxidation. The effects of NAC on H2O2-mediated oxidation are well described, such that NAC impedes oxidative processes by acting as a ROS scavenger. Specifically, the free thiol of NAC behaves as a martyr, taking on the brunt of oxidation by ROS to become oxidized itself, thereby reducing ROS' ability to modify other thiols. As such, NAC has been described as a cysteine-residue oxidation protectant, and as shown in Fig. 4 C and D and Supplemental Fig. 7, ISO-mediated S-sulfenation of β2AR (270 ± 10% at 10 μM for 15 min) was markedly attenuated in cells pretreated for 4 h with NAC (138 ± 14%) (p < 0.01 versus ISO). These results are consistent with our previously published results demonstrating that NAC decreases β2AR-mediated ROS generation and inhibits β2AR function (Moniri and Daaka, 2007) and demonstrate that the ISO-mediated S-sulfenation is specific to ROS.
Agonist- and H2O2-Mediated β2AR S-Sulfenation Occurs in Human Alveolar Cells.
Because our initial characterization of β2AR S-sulfenation here was performed in a clonal cell model of forced β2AR overexpression and depends on an epitope-tagged receptor, we wanted to examine whether β2AR S-sulfenation occurred in a physiologically relevant system. Given that β2 receptors are physiologically critical for pulmonary function, and β2 receptor agonists are foundational therapeutic agents used to dilate the bronchial and alveolar epithelium in pulmonary disorders such as asthma and chronic obstructive pulmonary disease, we examined whether H2O2 or β2 agonism would stimulate β2AR S-sulfenation in human lung alveolar epithelial cells. Here, we used the well characterized immortalized A549 human lung alveolar epithelial cell line, which endogenously expresses both β2 and β1 receptors (Szentendrei et al., 1992). Using the biotin-switch assay described in Fig. 1 in conjunction with a β2AR-specific antibody raised against the C-terminal tail of the receptor, our results demonstrate that stimulation of A549 cells with H2O2 facilitated significant S-sulfenation of β2AR compared with unstimulated control (Fig. 5A; Supplemental Fig. 8). Because A549 cells also express β1AR and ISO is a nonselective β-receptor agonist, we used the highly β2-selective agonist formoterol, which exhibits 300-fold greater potency for β2AR versus β1AR (Anderson 1993), to avoid concurrent stimulation of β1AR in these cells. Results of these experiments demonstrate that despite its relatively low intrinsic efficacy (Hanania et al., 2010) and the low expression levels of β2AR on A549 cells (Szentendrei et al., 1992), formoterol (1 μM, 15 min) stimulates S-sulfenation of β2AR compared with unstimulated control (Fig. 5B; Supplemental Fig. 8). Taken together, these results suggest that β2AR S-sulfenation by ROS or receptor agonism occurs in physiologically relevant tissues such as the alveolar epithelium, where β2AR agonism represents the primary therapy to achieve bronchodilation in airway disorders such as asthma.
Over the last 5 years it has been demonstrated and is now well accepted that agonism of the β2AR leads to generation of ROS. These results have been obtained from a variety of tissue and cell types including cardiomyocytes (Li et al., 2010; Xu et al., 2011), microglia (Qian et al., 2009), and embryonic kidney cells (Moniri and Daaka, 2007; Gong et al., 2008). In each case, ROS were not only shown to be generated upon β2AR agonism, but were also shown to serve purposeful roles in regulating β2AR signaling. Our previous results demonstrated that β2AR-mediated ROS generation was accompanied by a definitive requirement of ROS for proper β2AR signaling, suggesting that some degree of ROS is indispensible for receptor function. Xu et al. (2011) recently showed that mice transgenically overexpressing β2AR exhibited a greater degree of cardiac ROS production, which in of itself was not surprising, but this effect markedly contributed to cardiac inflammation and failure, suggesting that overexertion of the β2AR-ROS link may have pathophysiological consequences. These aggregate studies imply that low degrees of ROS are required for β2AR signaling, whereas higher levels of ROS lead to detrimental effects, similar to the current paradigm that suggests micromolar H2O2 levels may regulate signaling while higher levels facilitate unfavorable oxidative stress responses (Valko et al., 2007). One of the primary ROS species reported to be generated after β2AR agonism is superoxide, which occurs via the action of NADPH-oxidase enzymes and is subsequently dismutated, yielding H2O2 (Gong et al., 2008; Qian et al., 2009; Li at al., 2010; Xu et al., 2011). This two-electron oxidant species' chief biological activity comes from its ability to readily oxidize thiol groups of protein cysteine residues, and the initial product of this reaction is an S-sulfenic acid (S-OH). The overall purpose of this study was to determine whether ROS are capable of oxidizing the β2AR to form S-sulfenic acids. Using a biotin-switch assay that had been modified to assess S-sulfenation by way of the well characterized sulfenic acid-reducing agent sodium arsenite, the present study demonstrates for the first time that ROS are capable of S-sulfenating the β2AR. We also reveal that β2AR S-sulfenation occurs upon receptor agonism with the classic β-receptor catecholamine agonist isoproterenol as well as the β2-selective agonist formoterol, and importantly, the observation that this effect is receptor-dependent (i.e., blocked by propranolol) suggests that β2AR-generated ROS contribute to S-sulfenation.
Sulfenic acid, unlike higher-order sulfinic and sulfonic acids, is the simplest of the organosulfur oxyacids, and as a consequence these species are predicted to be unstable and reactive. Despite these characteristics, previous work has demonstrated that limited solvent access and a lipophilic environment, along with a lack of other reactive sulfhydryl groups in the vicinity of the site of S-sulfenation, allows for marked stabilization of protein S-sulfenic acids (Allison, 1976; Liu, 1977). Although our results demonstrate that an S-sulfenated β2AR species can be stabilized upon treatment of cells with β2 agonist as well as physiologically relevant concentrations of H2O2, we cannot rule out the possibility that this modification could be further oxidized to higher-order species, either concurrent with or after S-sulfenic formation. Nevertheless, to date, more than 200 proteins accounting for vast biochemical functions that include metabolism, transcription, DNA repair, nuclear transport, intracellular trafficking, immune regulation, and cell signaling, among others have been identified as at least being initially S-sulfenated (Leonard et al., 2009). To our knowledge, this study represents the first time that a mammalian G protein-coupled receptor has been recognized as being oxidized to an S-sulfenic acid. Although our results suggest a role for S-sulfenation in β2AR regulation, an absolute physiological role of protein S-sulfenation remains elusive, making it difficult to gauge what role, if any, S-sulfenation has on β2AR physiology. However, our observation that β2AR S-sulfenation occurs in human lung alveolar epithelial cells suggests that this modification occurs endogenously upon agonism of physiologically relevant levels of receptor.
A notable feature of the ISO-mediated oxidation is its apparent transient nature, which saturates at 15 min after agonism. This observation is of pharmacological interest because of the temporal dependence of ISO-stimulated ROS generation, which also peaks approximately 15 to 20 min after agonist addition in these cells (Moniri and Daaka, 2007). Although the decrease in receptor S-sulfenation seen upon treatment with ISO may be attributed to reducing favorable intracellular conditions, we cannot rule out the possibility that further modification of the receptor by phosphorylation, ubiquitination, or signal-dependent events such as sequestration and internalization may play a role in desensitizing the S-sulfenation signal. Meanwhile, the kinetics of ISO-mediated S-sulfenation is in accordance with previously reported G protein-dependent β2AR signaling events such as cAMP formation and correlates precisely to our previous data on the ROS dependence of β2AR signaling (Moniri and Daaka, 2007). It is noteworthy that because β-arrestin recruitment and β2AR-mediated G protein-independent signaling are known to be temporally regulated by G protein-coupled receptor kinase-mediated phosphorylation, specifically occurring after 10 to 15 min of agonism (Shenoy et al., 2006), our kinetic data with ISO may also suggest a role for β-arrestin in the desensitization of the S-sulfenation signal.
Whereas ISO-mediated ROS generation has been well described, ROS generation induced by other β2AR agonists has not been characterized to a great extent. Incidentally, our laboratory has also detected S-sulfenation induced by other β-agonists including epinephrine, albuterol, and formoterol. However, interpretation of these S-sulfenating effects must be made with caution because the ROS-generating effects of these agents remain virtually unstudied. Our results here should also be interpreted under the context of several other important factors that may influence β-agonist-mediated S-sulfenation. These include the well described rapid auto-oxidation of the catecholamine-like β-agonists, a consequence that leads to loss of agonist function (Misra and Fridovich, 1972). This effect may be significant given that many cell-based investigations use the addition of antioxidants (e.g., ascorbic acid) to catecholamine-agonist drug stocks to prevent catechol oxidation. However, by sequestering ROS pools, this approach could conceivably also hinder receptor S-sulfenation and possibly influence receptor function. In addition, many β2-agonists, including albuterol, formoterol, and salmeterol, have atypical binding kinetics that induce slow receptor dissociation (Teschemacher and Lemoine, 1999), and it is possible that this unique pharmacological property could affect ROS generation and receptor S-sulfenation. Furthermore, the well described variable intrinsic efficacies of β-receptor agonists could be expected to contribute to differing abilities of these agents to promote ROS generation and receptor S-sulfenation (Hanania et al., 2002, 2010). In this regard, ISO and epinephrine are reported to have full agonist efficacies, whereas formoterol and albuterol exhibit one-fifth and one-twentieth of the respective responses. Further studies underway in our laboratory will seek to address these issues and aid in assessing the contribution of S-sulfenation on β2AR conformations, signaling, and physiological function.
Our future studies will also address the localization of the exact sites of β2AR S-sulfenation. Whereas many studies that aim to decipher sites of thiol modification use expression of mutant protein constructs in which cysteines are mutated to nonreactive residues, these studies often lead to biological responses that are divergent from wild type. In this regard, mutation of several cytosolic-facing or helix-associated β2AR cysteines has been shown to lead to loss of expression or function, which would confound interpretation of such results. In addition, although the biotin-switch assay shown here effectively demonstrates β2AR S-sulfenation, the method is limited because of its dependence on SDS-denaturing conditions, which may destabilize modified thiols, decreasing the sensitivity when used to localize sites of modification based on site-directed mutagenesis (Charles et al., 2007). Based on these issues, we are using a proteomic approach to identify the sites of oxidation by using mass spectrometry methodology coupled with fluorescent dimedone congeners that allow for quantification of S-sulfenation and localization of the sites of oxidation (Poole et al., 2005). The human β2AR contains 13 cysteine residues, and several investigations have examined the importance of various cysteine residues in β2AR structure and function. Our efforts to identify the sites of oxidation will focus on the intracellular-exposed cysteines with particular emphasis on Cys341, which creates a fourth intracellular pseudo-loop and is required for creation of a high-affinity agonist-coupling state (O'Dowd et al., 1989; Moffett et al., 1993), as well as residues within transmembrane helices VI and VII and the C-terminal tail, which have been shown to be involved in rotation and conformational changes upon agonism (Ghanouni et al., 2001; Shi et al., 2002; Granier et al., 2007).
In summary, the current study demonstrates for the first time that the human β2AR can be oxidized by ROS to form receptor-S-sulfenic acids, and importantly S-sulfenation occurs upon receptor agonism, which has previously been shown to induce intracellular ROS generation. Because β2AR is the prototype of the class A GPCRs and is often used as a model for the study of other rhodopsin-like receptors, some of which have been linked to ROS generation, our results constitute an important first step toward characterization of the ROS/GPCR relationship and may have broader implications in the overall study of the role of ROS in the regulation of GPCRs.
Participated in research design: Burns and Moniri.
Conducted experiments: Burns and Moniri.
Contributed new reagents or analytic tools: Burns and Moniri.
Performed data analysis: Burns and Moniri.
Wrote or contributed to the writing of the manuscript: Burns and Moniri.
We thank Vivienne Oder for excellent secretarial assistance and Dr. Robert J. Lefkowitz (Duke University Medical Center) and Dr. Philip Moos (University of Utah) for reagents.
This study was funded, in part, by a New Investigator Program grant (to N.H.M.) from the American Association of Colleges of Pharmacy. R.N.B. is supported by a predoctoral fellowship from the American Foundation for Pharmaceutical Education.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- β2-adrenergic receptor
- reactive oxygen species
- G protein-coupled receptor
- human embryonic kidney
- hydrogen peroxide
- methyl methanethiosulfonate
- sulfenic acid
- N6-[2-[[4-(diethylamino)-1-methylbutyl]amino]-6-methyl-4-pyrimidinyl]-2-methyl-4,6-quinolinediamine trihydrochloride
- 25 mM HEPES, pH 7.6, 1 mM EDTA, 1% SDS, and 0.01 mM neocuproine
- Received July 11, 2011.
- Accepted September 12, 2011.
- Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics