Hydrogen sulfide inhibits Arabidopsis inward potassium channels via protein persulfidation

Hydrogen sulfide inhibits the inward-rectifying potassium ion current by inducing the persulfide modification on three cysteine residues of the inward potassium channel KAT1. This persulfidation inhibits the activity of KAT1 and KAT2 and suppresses the activity of heterologous channels formed by KAT1 and KAT2. Stomata play essential roles in sensing environmental stress and transmitting signals to plants. Stomatal opening is triggered by increased H+-ATPase activity, leading to membrane hyperpolarization (Liu and Xue, 2021), thus activating inward potassium (K+in) channels (KAT1 and KAT2) and resulting in K+ influxes (Qi et al., 2018). Conversely, during the stomatal closure, activation of both slow (S)-type and rapid (R)-type anion channels depolarizes the guard cell membrane (Wang et al., 2016). Hydrogen sulfide (H2S) is a vital signaling molecule in promoting stomatal closure (Papanatsiou et al., 2015; Wang et al., 2016; Liu et al., 2021). It participates in the post-translational modification of cysteine residues, leading to the formation of persulfide (R-SSH) groups (Filipovic et al., 2018). Abscisic acid (ABA) induces H2S accumulation by enhancing the self-activation of l-cysteine desulfhydrase 1 (DES1) via persulfidation, thereby amplifying the H2S signals in guard cells (Shen et al., 2020). Hydrogen sulfide was previously reported to regulate K+in channels in promoting stomatal closure (Papanatsiou et al., 2015). However, the mechanisms by which H2S regulates K+in channels in guard cells remain unclear. In this study, we provide evidence that H2S directly induces the persulfidation of the C-terminus of the KAT1 subunit, likely disrupting its oligomerization and inhibiting inward K+ translocation. Here, we found that H2S-fumigated leaves presented lower water loss than the control in Arabidopsis thaliana (Figure S1). We previously observed that 100 µM NaHS (H2S donor) activates slow-type anion currents in Arabidopsis (Wang et al., 2016). As the inward Shaker K+ channels also play a critical role in stomatal movement (Qi et al., 2018), we next investigated the effect of H2S on inward K+ currents in the guard cell via a patch-clamp technique. Whole-cell patch-clamp recordings revealed that H2S significantly inhibited the inward K+ current (Figure 1A). In Arabidopsis guard cells, at least four genes encode inward Shaker subunits, with KAT1 and KAT2 being the most prominent (Lebaudy et al., 2010). Therefore, we aimed to examine the impact of H2S on the K+ influxes of KAT1 and KAT2. The two-electrode voltage clamp (TEVC) technique was used to assess K+ channel activity. KAT1 and KAT2 were fused with YFP fluorescent tags to monitor protein expression and localization (Figure S2). As reported, large inward K+ currents are readily recorded upon expression of KAT1 protein in Xenopus oocytes (Papanatsiou et al., 2015). The current of KAT1 was significantly reduced after a brief incubation with 500 μM NaHS in bath solution (Figures 1B, C, S3A). Similarly, 50 μM sodium polysulfide (Na2S3), an H2S-releasing agent, also inhibited KAT1 current. Na2S3 was more effective than NaHS (Figures 1B, C, S3). KAT2, a homologous gene of KAT1, exhibited similar current behavior (Figures 1B, C, S3) (Lebaudy et al., 2010), and its K+ current was also suppressed by various H2S donor treatments (Figures 1C, S3). H2S inhibits K+ currents in conjunction with the persulfidation of KAT1 occurring in the KHA domain (A) Current–voltage relationships of whole-cell currents recorded in guard cell protoplasts without NaHS treatment (open circles, n = 8) and pre-exposed to NaHS (filled circles, n = 9). Whole-cell currents were recorded either in the control (top inset) or pre-exposed to NaHS (bottom inset). (B) The current–voltage relationship in oocytes expressing KAT1, and the effects of treatment with 500 μM NaHS or 50 μM Na2S3. ddH2O in control (n = 11), NaHS (n = 10), Na2S3 (n = 10), and KAT1 in control (n = 9), NaHS (n = 9), Na2S3 (n = 9). (C) The amplitude of K+ current record at −180 mV. (D) In vitro NaHS-induced persulfidation of KAT1-Flag was detected using the biotin-switch assay. (E) The C-terminus of KAT1 is persulfidated in the presence of H2S with a dosage-dependent. The addition of DTT reverses the persulfidation. (F) Persulfidation analysis of the C-terminus of KAT1 wild-type and mutant derivatives in vitro. (G) The current–voltage relationship in oocytes expressing KAT1 and its mutants (C569A, C606A, C647A, and 3 M) bathing with 100 mM K (control) and 100 mM K + 500 μM NaHS. ddH2O in control (n = 5) and NaHS (n = 5), KAT1 in control (n = 10) and NaHS (n = 13), KAT1-C596A in control (n = 13) and NaHS (n = 12), KAT1-C606A in control (n = 15) and NaHS (n = 15), and KAT1-C647A in control (n = 9) and NaHS (n = 8), and KAT1-3M in control (n = 21) and NaHS (n = 19). (H, L) The amplitude of K+ current recorded at −180 mV. (I) Percentage inhibition of KAT1 current by H2S. (J) Typical recordings of KAT1 + KAT2 and KAT1-C647A + KAT2, bathing with 100 mM K, 100 mM K + NaHS (500 μM). (K) The current–voltage relationship in oocytes expressing KAT1 + KAT2 and KAT1-C647A + KAT2, and the effects of treatment with 500 μM NaHS. ddH2O in control (n = 4) and NaHS (n = 4), KAT1 + KAT2 in control (n = 14) and NaHS (n = 15), and KAT1-C647A + KAT2 in control (n = 15), NaHS (n = 15). (M) Percentage inhibition of KAT1 + 2 by H2S. All the data are means ± SD, and statistical significance was determined using one-way ANOVA (Tukey's multiple comparisons test) or multiple t-tests (Holm–Sidak method). H2S-induced protein persulfidation (S-sulfhydration) is essential for exerting its biological function (Liu et al., 2021). To investigate whether H2S inhibited the K+ current through persulfidation of KAT1, we expressed the KAT1-Flag protein in expi293F cells (Figure S4) and used a tag-switch assay in which persulfidated cysteine was labeled with HPDP-biotin and detected by anti-biotin. The results showed that NaHS dosage induced persulfidation of KAT1 protein, and this modification was reversed by dithiothreitol (DTT) (Figure 1D). Amino acid sequence analysis displayed KAT1 containing 10 cysteine residues (Figure S6A). To screen out persulfidated cysteine sites induced by H2S, we performed direct mass spectrometry (MS) analysis. KAT1 was digested with a protease before MS analysis, but concerns arose because the enzyme cleavage site was close to a cysteine residue, potentially affecting the results. Switching to a different protease did not improve the outcome; only the non-persulfidated Cys338 was detected (Figure S5). Alternatively, we analyzed the structure of KAT1 and observed a short intracellular domain at the N-terminus and a long intracellular structural domain at the C-terminus (Figure S6). The latter is a key region involved in spontaneous regulation in Shaker-type channels. Both N- and C-terminal regions contain cysteine sites (Figure S6), prompting us to investigate whether these segments are targets for H2S regulation of KAT1. We expressed the C-terminal fragment of KAT1 (KAT1-CT) in the Escherichia coli system (Figure S7) and analyzed the level of KAT1-CT persulfidation after treatment with different concentrations of NaHS. The results demonstrated that H2S induced a dose-dependent persulfidation of the KAT1-CT protein (Figure 1F). Subsequently, we constructed the KAT1-CT single mutant protein and assessed its persulfidation level (Figure S7). The results showed that mutating a single cysteine in KAT1-CT to alanine decreased the level of persulfidation, with the mutation of Cys647 having the most significant effect, followed by Cys596 and Cys606 (Figure 1F). To determine whether the persulfidation affects H2S-mediated inhibition of KAT1 transport, we expressed KAT1-YFP single mutant proteins in Xenopus oocytes with YFP tag (Figure S2), and assessed their K+ transport activity via TEVC. Among the three cysteine sites, each KAT1-YFP single mutant attenuated the inhibition of KAT1-mediated K+ current by H2S (Figure 1G, H). KAT1-3M exhibited reduced K+ transport (Figure 1G, H), and the currents mediated by KAT1-3M were unresponsive to H2S (Figure 1G, H). The mutation of Cys647 to alanine had a lesser effect on inhibition, followed by Cys596 and Cys606; with KAT1-3M showing minimal response (Figure 1I). These results suggest that cysteines at the C-terminus of KAT1 play a critical role in modulating KAT1 by H2S. Additionally, the KAT1-NT Cys14Ala mutation did not affect the K+ transport and the inhibitory effects of H2S on KAT1 in Xenopus oocytes (Figure S8). It was reported that KAT1 and KAT2 formed heteromeric K+in channels in guard cells (Lebaudy et al., 2010). We also detected their K+in current in Xenopus oocytes via TEVC. Compared with homomeric KAT1 or KAT2 (Figures 1B, C, S2, S3), H2S exerts a more significant inhibitory effect on heteromeric KAT1 + KAT2 (Figure 1J–L). The KAT1-C647A + KAT2 complex exhibited a significantly smaller K+ current compared with the KAT1 + KAT2 complex and exhibited reduced sensitivity to H2S inhibition (Figure 1J–M). Furthermore, bimolecular fluorescence complementation (BiFC) assays in tobacco demonstrated that H2S significantly inhibited the interaction between KAT1 and KAT2 (Figure S9). These results suggest that H2S inhibits the heteromeric assembly of KAT1 and KAT2, and that this inhibition is closely related to the persulfidation of Cys647 of KAT1. DES1 is the primary enzyme-producing H2S in guard cells (Liu and Xue, 2021), and overexpression of DES1 increases intracellular H2S levels (Chen et al., 2020). The stomatal apertures of DES1-OE plants were significantly smaller compared with Col-0 (Figure S10B, C). DES1-OEs displayed more substantial drought tolerance than Col-0 (Figure S10A), consistent with the phenotype of H2S-induced inhibition of K+ currents and promoted stomatal closure (Papanatsiou et al., 2015). To further analyze the structural basis of KAT1 inhibition by H2S through persulfide modification, we modeled the structure of the KAT1 subunit using AlphaFold 2.0 (https://alphafold.com/) and performed family and domain analysis in UniProt (https://www.uniprot.org). The analysis revealed that Cys647 was located in the KHA domain of KAT1, Cys606 is situated in the initial region of the KHA domain, and Cys596 was located in the polar residue region in the disordered region (Figure S11). The KHA domain includes two conserved blocks, enriched with hydrophobic and acidic residues, respectively, which are unique to plant K+in channels and vital for interactions among plant K+in channels (KAT1, KAT2, AKT2/3, AKT1) (Figure S12A). The KHA domain mediates tetramerization and/or stabilization of the heteromers (Zimmermann et al., 2001). The above analyses indicate that the inhibition of KAT1 current by H2S is closely related to the KHA domain. Interestingly, compared with the KHA domain sequences of KAT1 and KAT2 revealed that Cys647, a cysteine unique to KAT1, resides in a contiguous hydrophilic amino acid (–GGCN–) region, while the corresponding region of KAT2 contains three contiguous hydrophilic amino acids (–GYS–) (Figure S12A). Structure simulations further revealed that Cys647 in KAT1 forms a hydrogen bond with the carbonyl group of Phe644, likely playing a role in stabilizing the turn structure behind the α-helix where Cys647 resides (Figure S12B). Persulfidation at Cys647 probably disrupted this hydrogen bond, thereby affecting KAT1 tetramerization via the KHA domain. Conversely, simulations of the KAT2 structure showed no Cys–Phe hydrogen bond in the corresponding region. Instead, a stronger C–N peptide bond is formed between Gly663 and Gly659 (Figure S12C). The presence of Cys647 in KAT1 may explain why H2S exerts a more significant inhibitory effect on KAT1 than on KAT2 (Figures 1B, C, S3A). Our results also displayed that H2S inhibited the KAT2 currents (Figure S3B, C), similar to the effect on the KAT1 C647A mutant protein (Figure 1E-I). We speculate that Cys596/Cys592 on KAT1 and Cys606/Cys611 on KAT2 were partially involved in H2S inhibition of the KAT1/KAT2 currents. The position of cysteine sites from the KHA domain may determine the extent of inhibition of KAT1 channels by persulfidation (Figure 1E–I). In summary, our study provides new insights into the interplay between persulfidation and ion channel activity, advancing our understanding of the complex physiological processes underlying stomatal function and plant responses to environmental cues. We gratefully acknowledge the support from the Stress biology scientific research team at the National Key Laboratory of Crop Genetics Improvement, Huazhong Agricultural University (HZAU). We also thank research associates at the Center for Protein Research (CPR) at HZAU for MS technical support, the Arabidopsis Research Center at HZAU for the resources, and Dr. Ting Peng from HZAU for editing the language. This work was funded by the National Natural Science Foundation of China (32070214, 31670267), the Postdoctoral Fellowship Program of CPSF (GZC20230908), the China Postdoctoral Science Foundation (CN: 2023M741289), the Postdoctor Project of Hubei Province under Grant No. 2004HBBHCXA042. The authors declare that they have no conflict of interest. H.L. and S.X. designed the experiment. H.L. performed electrophysiology, biochemical, and physiological experiments with input from X.L., R.L., C.L., S.L., and Z.Z. X.L. purified proteins of KAT1-CT and the three types of point mutants and performed in vitro transcription of cRNA. R.L. constructed the plasmid of the KAT1 protein and expressed and purified KAT1 in HEK293 cells. C.L. constructed the plasmid of the KAT1 point mutants and mass analyses on the persulfidation of KAT1. H.L., X.L., and S.X. analyzed data. Z.L. and S.X. supervised the experiments; H.L., X.L., and S.X. wrote and modified the manuscript. All authors read and approved of the final manuscript. Additional Supporting Information may be found online in the supporting information tab for this article: http://onlinelibrary.wiley.com/doi/10.1111/jipb.13851/suppinfo Figure S1. Hydrogen sulfide (H2S) treatment improves drought tolerance in Arabidopsis leaves Figure S2. Confocal fluorescence images of oocytes expressing KAT1-YFP (C596A, C606A, C647A, and 3M) and KAT2-YFP Figure S3. Hydrogen sulfide (H2S) inhibits KAT1 and KAT2 currents Figure S4. Expression and purification of KAT1 protein using expi293F cells Figure S5. Identification of S-persulfidated Cys residues of KAT1 using mass spectrometry Figure S6. Analysis of KAT1 transmembrane structural domains Figure S7. Expression and purification of KAT1-CT protein using BL21 cells Figure S8. Hydrogen sulfide (H2S) does not impede K+ transport by persufidating Cys14 at the N-terminus of KAT1 Figure S9. Hydrogen sulfide (H2S) inhibits interaction between KAT1 and KAT2 Figure S10. Hydrogen sulfide (H2S)-producing enzyme DES1 promotes stomatal closure and plant drought resistance Figure S11. Structural prediction and analysis of KAT1 Figure S12. Cys647 is site specific to the inhibition of KAT1 by persulfidation Table S1. List of primers used in this study Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.