Abstract
The most common cardiac pathologies, such as myocardial infarction and heart failure, are associated with oxidative stress. Oxidation of the cardiac ryanodine receptor (RyR2) Ca2+ channel causes spontaneous oscillations of intracellular Ca2+, resulting in contractile dysfunction and arrhythmias. RyR2 oxidation promotes the formation of disulfide bonds between two cysteines on neighboring RyR2 subunits, known as intersubunit cross-linking. However, the large number of cysteines in RyR2 has been a major hurdle in identifying the specific cysteines involved in this pathology-linked post-translational modification of the channel. Through mutagenesis of human RyR2 and in-cell Ca2+ imaging, we identify that only two cysteines (out of 89) in each RyR2 subunit are responsible for half of the channel’s functional response to oxidative stress. Our results identify cysteines 1078 and 2991 as a redox-sensitive pair that forms an intersubunit disulfide bond between neighboring RyR2 subunits during oxidative stress, resulting in a pathological “leaky” RyR2 Ca2+ channel.
Similar content being viewed by others
Introduction
Calcium ion (Ca2+) is a universal second messenger that regulates a myriad of cellular processes, including contraction, gene regulation, synaptic transmission, hormone secretion, migration, and cell growth1. In cardiac myocytes, Ca2+ released from the sarcoplasmic reticulum (SR) through the type 2 ryanodine receptor (RyR2) is essential for initiating a robust myocardial contraction2. Consequently, defects in RyR2 regulation cause contractile dysfunction and decreased cardiac output. The most common cardiac pathologies with fatal outcomes, such as myocardial infarction and heart failure, are associated with abnormal Ca2+ regulation3. Pathological alterations during those conditions commonly involve substantial structural and functional remodeling of RyR2 by oxidative stress4,5,6,7. It has been shown that reactive oxygen species (ROS) can directly activate RyR2-mediated Ca2+ release in the form of diastolic spontaneous Ca2+ sparks8,9 and Ca2+ waves4,10, causing arrhythmias and depletion of SR Ca2+ load. Oxidative stress also causes the dissociation of an important regulator of RyR2 function, calmodulin (CaM)11,12,13. As a result, RyR2 oxidation has a significant impact on Ca2+ signaling and contraction in infarcted and failing hearts4,6,14,15,16.
Because RyR2 comprises up to 89 cysteines per subunit17, it is no surprise that RyR2 plays a key role in the cellular response to oxidative stress. However, a large number of cysteines in RyR2 has been a major limitation in identifying the specific sites of ROS action. As a result, the current understanding of the molecular mechanisms of redox regulation and redox-sensitive sites within RyR2 remains incomplete. Hence, the prospect of targeted therapy against RyR2 oxidation is absent. In contrast, functionally important phosphorylation sites on RyR2 have been identified and characterized18,19,20,21, and pre-clinical studies targeting these sites are currently underway. Thus, identifying the functionally important cysteines in RyR2 is essential for advancing our understanding of molecular mechanisms of RyR2 dysfunction during pathological conditions associated with oxidative stress.
In our previous studies, we discovered a novel posttranslational modification (PTM) of RyR2 induced by oxidative stress. As a common mechanism of channel assembly, the RyR2 complex is formed by four identical subunits. We have shown that oxidative stress can promote the formation of disulfide bonds between cysteines on neighboring RyR2 subunits, also known as intersubunit cross-linking13,22. There is structural evidence that a transition of RyR between the closed and the opened state (i.e., channel gating) is associated with an intersubunit dynamic23. Thus, we hypothesized that intersubunit cross-linking should significantly impact RyR2 function and Ca2+ signaling. In agreement with our hypothesis, we found that in ventricular myocytes from healthy hearts, the degree of SR Ca2+ leak induced by oxidative stress positively correlated with the level of RyR2 cross-linking13,22. Moreover, a significant level of RyR2 cross-linking has been detected in ventricular myocytes from failing hearts7.
The effect of oxidative stress on the channel’s structure and function is better described for the skeletal muscle type 1 RyR isoform (RyR1). Historically, the oxidation of thiol groups in RyR1 by reactive sulfhydryl agents has been linked to increased SR Ca2+ release in skeletal muscle fibers24. It has been suggested that cysteine 3635 in RyR1 is the most important residue responsible for redox regulation of the channel25,26. It has been shown that cysteine 3635 can be oxidized during oxidative stress to form a disulfide bond27,28. We re-evaluated the functional significance of the corresponding cysteine 3602 in RyR2 and showed that this residue is not involved in channel cross-linking nor Ca2+ leak activation by oxidative stress13. It appears that despite ~70% homology between RyR1 and RyR2, these channels exhibit important structural and functional differences, particularly in redox regulation. In the end, the specific mechanisms of redox regulation and the redox-sensitive residues within RyR2 remain largely unknown.
In this study, we identify cysteines 1078 and 2991 as a critical redox-sensitive pair that can form a disulfide bond between neighboring RyR2 subunits during oxidative stress, resulting in “leaky” RyR2.
Results
Cell system to study the functional role of RyR2 intersubunit cross-linking in Ca2+ signaling
To define the functional role of cysteines involved in RyR2 intersubunit cross-linking, we took advantage of our assay to study RyR2 function in living cells29. Changes in the ER luminal [Ca2+] ([Ca2+]ER) were measured with the genetically encoded ER Ca2+ sensor R-CEPIA1er in HEK293 cells expressing human RyR2 (hRyR2; Fig. 1a). R-CEPIA1er has an appropriate affinity to measure [Ca]ER dynamics in non-muscle30 and muscle cells31. The cells were also transfected with SERCA2a Ca2+ pump to maintain physiological ER Ca2+ load. Co-expression of hRyR2 and SERCA2a produced regular [Ca2+]ER depletions (i.e., Ca2+ waves) due to spontaneous hRyR2 activation followed by SERCA-mediated Ca2+ reuptake (Fig. 1b). Thus, the expression of both hRyR2 and SERCA2a creates a heterologous cell system that reproduces key aspects of cardiac Ca2+ signaling. As expected, these Ca2+ waves were sensitive to the RyR agonist caffeine. By activating hRyR2, caffeine (10 mM) caused a global ER Ca2+ release and [Ca2+]ER depletion. The completeness of [Ca2+]ER depletion by caffeine was validated by the SERCA inhibitor thapsigargin (10 µM). Since the application of thapsigargin did not cause additional [Ca2+]ER depletion, we used the caffeine output as Fmin during signal normalization (Supplementary Fig. 1). Since HEK293 cells lack endogenous RyR (verified by Western blot analysis and the lack of response to caffeine; Supplementary Fig. 2), this cell system allows studying only recombinant RyR2. Thus, we consider this minimalistic Ca2+ handling system to be ideal for examining the protective properties of different cysteine mutations against hRyR2 intersubunit cross-linking.
To identify the levels of oxidative stress sufficient to induce hRyR2 cross-linking in our cell model system, we treated cells expressing the wild-type hRyR2 (hRyR2WT) with increasing concentrations of the thiol oxidant diamide. Western blot analysis revealed that increasing concentrations of diamide caused the disappearance of the hRyR2WT 560 kDa monomeric band and promoted the formation of cross-linked oligomers. This effect could be reversed by treating cell samples with the reducing agent DTT (Fig. 1c). Next, we tested the effect of diamide on [Ca2+]ER dynamics in our cell model system. For this set of experiments, we took advantage of stable inducible Flp-In T-Rex-293 cells expressing SERCA2a29. We used normalized and averaged values of Ca2+ wave peaks (marked by dashed lines; Fig. 1b) to estimate the relative change in [Ca2+]ER loading upon the application of diamide. The concentration of diamide (50 µM) that caused the maximal cross-linking (91.8 ± 1.7%; n = 13) also resulted in 50.1 ± 1.4% (n = 72) depletion of [Ca2+]ER (or ER Ca2+ load) in cells that express hRyR2WT tagged with GFP (Fig. 1a, b). Diamide (50 μM) did not affect [Ca2+]ER in cells that do not express hRyR2WT (Supplementary Fig. 2), suggesting that this level of oxidative stress does not affect other endogenous ion channels/pumps in our cell system. Because SERCA function is unaffected by these diamide concentrations29, the decline in [Ca2+]ER was mainly mediated by an increase in ER Ca2+ leak via hRyR2WT.
N-terminal truncation of hRyR2 domains A and B prevents hRyR2 intersubunit cross-linking and activation of hRyR2 by diamide
To identify hRyR2 cysteine oxidation which promotes cross-linking and ER Ca2+ leak, we analyzed the channel’s N-terminal region (NTR) structure (residues 1-547), which includes three domains (A, B, and C)32. The NTR of RyR2 has been shown to play an important role in the channel’s tetramer formation and regulation of the channel’s function33,34,35,36. It has also been characterized as a “hotspot” for the hRyR2 mutations associated with catecholaminergic polymorphic ventricular tachycardia (CPVT)37,38, suggesting its important role in channel gating. Based on the analysis of the NTR domains orientation in the available RyR2 cryo-EM models23,39,40, we decided to focus on domains A and B (residues 1-409), which harbor 9 of 10 NTR cysteines (Fig. 2a). We did not evaluate the role of cysteine 501 from domain C because of its distant location from the intersubunit interface of RyR2.
To study the role of NTR cysteines in hRyR2 cross-linking and hRyR2 redox regulation, we removed domains A and B, generating the hRyR2ΔAB construct (Fig. 2b). We compared cross-linking levels in hRyR2WT, and hRyR2ΔAB expressed in HEK293 cells after a treatment of cells with diamide. The mutant channel was expressed at a similar level as hRyR2WT (estimated from an intensity of “Diamide 0” bands; Fig. 3a). The truncation of hRyR2 domains A and B significantly prevented cross-linking induced by 25 and 50 µM diamide (Fig. 3a). Overall, the mutation preserved 45.6 ± 7.9% (n = 4, p = 5.2 × 10−4) and 43.5 ± 10.7% (n = 4, p = 0.014) of hRyR2ΔAB monomers from cross-linking at 25 and 50 µM diamide, correspondingly (Fig. 3b). In these experiments and hereafter, the protective effect of hRyR2 mutations was estimated based on the amount of the preserved monomer, which was calculated as a difference between normalized amounts of the mutant and hRyR2WT monomeric forms.
Next, we evaluated the functional response of hRyR2ΔAB to diamide in our cell system. It has been previously shown that the truncation of the domains A and B does not cause the channel loss-of-function34. While hRyR2ΔAB mutant in our model responded to caffeine activation (Fig. 3c), this mutation decreased Ca2+ wave frequency and increased ER Ca2+ load, suggesting an inhibition of hRyR2 activity. The application of 50 µM diamide resulted in 52.7 ± 2.6% (n = 22) decline in ER Ca2+ load caused by increased ER Ca2+ leak through hRyR2WT. However, in cells expressing hRyR2ΔAB, the same diamide concentration caused only 11.1 ± 1.3% depletion of [Ca2+]ER, yielding a relative protective effect of 79% (n = 20, p = 5.2 × 10−17, Fig. 3d). Thus, the removal of domains A and B was sufficient to eliminate the most of excessive ER Ca2+ leak mediated by hRyR2 oxidation, suggesting a critical role of hRyR2 NTR domains in the channel’s activation by oxidative stress.
Individual cysteines of the NTR of hRyR2 are not directly involved in hRyR2 intersubunit cross-linking
To narrow the search of the cross-linking sites within the NTR of hRyR2, we produced a construct lacking only the A domain (hRyR2ΔA). hRyR2ΔA showed substantial resistance to cross-linking similar to the one observed for the hRyR2ΔAB construct (Fig. 4a), preserving 43.8 ± 7.5% (n = 5, p = 2.4 × 10−4) and 21.8 ± 8.3% (n = 5, p = 0.0035) of the hRyR2ΔA monomeric form after the corresponding treatment with 25 and 50 µM diamide (Fig. 4b). Thus, removing the NTR domain A was sufficient to significantly prevent cross-linking of the channel. Next, we investigated which cysteine residues in domain A are involved in forming intersubunit disulfide bonds during channel oxidation. It has been shown that domain A contains 7 cysteines, and some of them can be partially exposed to solvent and avato oxidation (e.g., cysteines 36 and 158)17. However, when we generated a series of hRyR2 constructs with the cysteine to serine mutations in domain A (hRyR2C24S hRyR2C36S, hRyR2C47S/C65S, hRyR2C131S/C132S/C158S), none of them prevented the channel’s cross-linking (Supplementary Fig. 3). These results concluded that the individual cysteines in the A domain were not directly involved in hRyR2 cross-linking. Based on the available cryo-EM structural models23,39,40,41, cysteines from domain B of hRyR2 could only be cross-linked with cysteines from domain A of the neighboring subunit. Thus, these results also excluded the possibility of the involvement of cysteines in the hRyR2 domain B. It appears that the domains A and B allosterically modulate intersubunit cross-linking of hRyR2 instead of directly providing thiol groups to form intersubunit disulfide bonds.
Cysteine 1078 is directly involved in hRyR2 intersubunit cross-linking and activation of hRyR2 by diamide
Since the NTR cysteines are not directly involved in intersubunit cross-linking, we sought to explore other possible hRyR2 regions capable of forming intersubunit disulfide bonds. The recently solved Cryo-EM models of hRyR240,41 made it possible to highlight a potential interface for the disulfide bond formation between cysteine 2991 on the helical domain 2 (HD2) loop on one RyR2 subunit and cysteine 1078 of the P1-SPRY2 linker on another subunit (Fig. 5 and Supplementary Fig. 4). To test whether these cysteines are involved in cross-linking, we first designed the construct with cysteine 1078 mutated to serine (hRyR2C1078S). Western blot analysis revealed that the hRyR2C1078S mutation substantially prevented cross-linking caused by diamide (Fig. 6a, b). Overall, the mutation protected 58.7 ± 9.4% (n = 5, p = 1.5 × 10−4) of hRyR2C1078S monomers from cross-linking at 25 µM and 53.4 ± 7.4% (n = 5, p = 3.2 × 10−5) at 50 µM diamide (Fig. 6b).
Our recent data demonstrated that RyR2 intersubunit cross-linking positively correlated with RyR2 activation by oxidative stress7,13. We tested whether this residue is essential for the channel’s functional response to oxidative stress. The hRyR2C1078S mutant generated Ca2+ waves with similar properties as in cells expressing hRyR2WT (Fig. 6c and Supplementary Fig. 5). The application of 25 µM diamide caused the 42.1 ± 2.1% (n = 43) decline of ER Ca2+ load in cells expressing hRyR2WT and 24.3 ± 2.0% (n = 40) in cells expressing hRyR2C1078S (Fig. 6d). Since we observed some decline in hRyR2C1078S monomer levels with an increase in diamide concentration from 25 to 50 µM, we studied the effect of 50 µM on [Ca2+]ER. Diamide (50 µM) caused the 46.9 ± 2.0% decline (n = 27) of ER Ca2+ load for hRyR2WT and 28.5 ± 2.0% (n = 36) for hRyR2C1078S (Fig. 6d). Thus, the relative protective effect of the C1078S mutation was ~40% for both diamide concentrations (corresponding p = 2.7 × 10−8 and p = 2.8 × 10−8). Since the higher diamide concentration only slightly increased the ER Ca2+ load depletion, these results suggest that the most functionally significant disulfide cross-linking occurs at 25 µM diamide (by oxidizing cysteine 1078), while higher diamide concentrations oxidize other hRyR2 cysteines without a substantial increase in ER Ca2+ leak.
Cysteines 1078 and C2991 form the functionally important intersubunit disulfide bond within the hRyR2 complex
To explore whether cysteine 1078 forms the intersubunit disulfide bond with cysteine 2991, we mutated cysteine 2991 to serine (hRyR2C2991S). Western blot analysis showed that the C2991S mutation significantly prevented hRyR2 cross-linking, protecting 48.7 ± 7.3% (n = 3, p = 0.0037) of its monomeric form at 25 µM diamide and 72.3 ± 2.3% (n = 3, p = 2.0 × 10−4) at 50 µM diamide (Fig. 7a, b). We next tested whether hRyR2C2991S would have a similar degree of functional protection as seen for hRyR2C1078S. Cells expressing the hRyR2C2991S mutant generated Ca2+ waves similar to those observed in cells expressing hRyR2WT or hRyR2C1078S (Fig. 7c and Supplementary Fig. 5). The treatment with 25 µM diamide produced 46.1 ± 2.5% (n = 29) depletion of ER Ca2+ load for hRyR2WT and 25.0 ± 2.8% (n = 28) for hRyR2C2991S. The treatment with 50 µM diamide caused 47.0 ± 1.9% (n = 40) depletion of ER Ca2+ load for hRyR2WT and 28.9 ± 1.9% (n = 48) for hRyR2C2991S (Fig. 7d). Overall, the relative protective effect of the C2991S mutation was ~40% for both diamide concentrations (corresponding p = 5.5 × 10−7 and p = 2.7 × 10−9). The level of protection was similar to those seen for the hRyR2C1078S mutant, suggesting a common mechanism for the redox resistance in hRyR2C1078S and hRyR2C2991S mutants. Thus, cysteines 1078 and 2991 are likely involved in the hRyR2 intersubunit disulfide formation.
To directly test that these residues provide thiol groups to form an intersubunit disulfide bond, we created a double mutant with cysteines C1078 and C2991 mutated to serines (hRyR2C1078S/C2991S). When expressed in HEK293 cells, the hRyR2C1078S/C2991S mutant demonstrated significant protection against hRyR2 intersubunit cross-link induced by diamide. The protection of hRyR2C1078S/C2991S monomers from cross-linking was in line with the one seen for hRyR2C1078S and hRyR2C2991S mutants: 53.7 ± 7.6% (n = 5, p = 2.9 × 10−5) at 25 µM diamide and 53.1 ± 3.3% (n = 5, p = 2.4 × 10−6) at 50 µM diamide (Fig. 8a, b). Because there was no further protection against hRyR2 cross-linking in the double cysteine mutant, we concluded that cysteines 1078 and 2991 partner with one another to create a disulfide bond between neighboring hRyR2 subunits.
Next, we studied the effect of the simultaneous C1078S and C2991S mutations on the magnitude of [Ca2+]ER depletion caused by diamide. Cells expressing the hRyR2C1078S/C2991S mutant generated Ca2+ waves similar to those seen in cells expressing hRyR2WT (Fig. 8c and Supplementary Fig. 5). The treatment with 25 µM diamide caused the 48.6 ± 2.0% (n = 39) decline in ER Ca2+ load for hRyR2WT and 24.3 ± 1.6% (n = 44) for hRyR2C1078S/C2991S (Fig. 8d). Corresponding values for 50 µM diamide were 60.8 ± 2.1% (n = 24) for hRyR2WT and 28.3 ± 1.9% (n = 27) for hRyR2C1078S/C2991S (Fig. 8d). Overall, the simultaneous cysteine 1078 and 2991 mutations prevented the ER Ca2+ load depletion by ~50% for both diamide concentrations (p = 3.5 × 10−15 and p = 2.7 × 10−15). This level of protection against the cross-linking was similar to those seen for the single cysteine mutants (Figs. 6, 7), confirming that cysteines 1078 and 2991 form the functionally important intersubunit disulfide bond within the hRyR2 complex.
This study reveals that out of 89 cysteines in hRyR2, two cysteines (C1078 and C2991) play a particularly important role in the functional response of the channel to oxidative stress. Mutation of one of those cysteines substantially protected against hRyR2 cross-linking and ER Ca2+ leak activation caused by diamide. Notably, we observed a similar protective effect of hRyR2 C1078S and C2991S mutations in cells treated with increasing hydrogen peroxide (H2O2) concentrations (Supplementary Fig. 6). While diamide is an effective exogenous agent that specifically oxidizes thiol groups, H2O2 is an endogenous oxidant produced in vivo during oxidative stress42,43. Therefore, the redox response caused by cysteines 1078 and 2991 oxidation represents a universal reaction of RyR2 to oxidative stress.
Discussion
In the mammalian heart, SR Ca2+ released through RyR2 plays a central role in initiating a robust myocardial contraction. Consequently, defects in RyR2 function can lead to contractile dysfunction in a variety of cardiac pathologies. It has been shown that redox-dependent PTMs of RyR2 contribute to abnormal Ca2+ regulation during myocardial infarction and heart failure4,5,6,7. In the canine model of chronic heart failure, disulfide formation within the RyR2 complex has been characterized as the major redox modification responsible for “leaky” RyR2, abnormal Ca2+ regulation, and arrhythmias5. We previously showed that oxidative stress could promote the formation of disulfide bonds between two neighboring subunits of RyR2 through a PTM known as intersubunit cross-linking22. We found a strong positive correlation between the level of RyR2 cross-linking and an elevated SR Ca2+ leak7,13,22. Moreover, a substantial portion of RyR2 in cardiomyocytes from failing hearts exists in the cross-linked form7. Despite its clinical significance, the redox-sensitive sites involved in RyR2 cross-linking have not yet been identified.
By analyzing the recently published cryo-EM structures of hRyR241, we identified two candidates for the channel’s intersubunit disulfide formation at neighboring subunits of RyR2: cysteines 1078 and 2991 (Fig. 5 and Supplementary Fig 4a). We found that C2991 on the HD2 loop is solvent accessible in both closed and open conformations (Fig. 5 and Supplementary Fig 4b, Supplementary Table 1). Although a limited resolution in the area of the P1-SPRY2 linker does not provide structural information about the precise localization of C1078, this cysteine might be also accessible to oxidation (Fig. 5 and Supplementary Fig 4c, Supplementary Table 1). Therefore, based on structural localization and solvent accessibility, we decided to study the role of cysteines 1078 and 2991 in RyR2 disulfide cross-linking and functional response to oxidative stress. We discovered that mutations of either one of those cysteines or the mutation of both substantially protected against hRyR2 cross-linking caused by diamide or H2O2. To get further insights into the mechanism of the C1078-C2991 disulfide formation, we analyzed orientation and solvent accessibility of these cysteines at different RyR2 conformations, including phosphorylation by protein kinase A (PKA), CPVT mutation, and the presence of CaM or the Rycal drug ARM21041 (for the detailed analysis see Supplementary Table 1 and Supplementary Figs. 4, 7–9). The analysis revealed that the C2991 thiol group is accessible to a solvent, with higher accessibility in open and phosphorylated conformations. The binding of CaM or ARM210 also increases the solvent accessibility. However, an introduction of the CPVT R2474S mutation decreases the accessibility, especially in the open conformation. Due to its limited resolution, we could not reliably evaluate the conformation of the P1-SPRY2 linker and C1078 orientation in different RyR2 models. However, this residue could be accessible for oxidation due to the high levels of the linker’s flexibility. Even though available cryo-EM models41 do not indicate the favorable mutual orientation of C1078 and C2991 for the disulfide bond formation, we believe this bond is sterically possible in some P1-SPRY2 linker conformations (Supplementary Fig. 10)40,41. We propose a model where an oxidant (diamide, oxidized glutathione) would bind to C1078 and stabilize the linker in the favorable conformation for further interaction with C2991. In the next step, C1078 would create a disulfide bond with C2991, releasing the reduced form of oxidant (hydrazine, glutathione). The disulfide bond formation between two RyR2 subunits highlights the essential functional role of the intersubunit interface comprised of the P1-SPRY2 linker from one subunit and the HD2 loop from another subunit. In the previous structural study of the RyR2R2474S CPVT mutant41, ARM210 has been shown to bind to the P1 domain, stabilizing the intersubunit interface between the beginning of the P1-SPRY2 linker (preceding C1078) from one subunit and the HD2 loop (including C2991) from another subunit, reverting the primed state of the mutant channel to the closed state. In the proposed model, the stabilization of the intersubunit interface occurs through the interaction between D1070 on one RyR2 subunit and H2995 on another, with the possible involvement of H1071 and R298841. The intersubunit disulfide formation between C1078 and C2991 would likely destabilize this interface, shifting the channel’s structure to a “leaky” conformation.
By analyzing [Ca2+]ER dynamics in HEK293 cells expressing hRyR2 and SERCA2a Ca2+ pump, we showed that the mutations of cross-linking cysteines (C1078S or/and C2991S) substantially ( ~ 50 %) protect against a depletion of [Ca2+]ER induced by oxidative stress. Since the level of oxidative stress that induced almost complete hRyR2WT cross-linking (at 50 μM diamide) did not affect the function of SERCA2a29 and endogenous ion channels/pumps in our cell model system, the depletion of ER Ca2+ load was chiefly mediated by increased ER Ca2+ leak via hRyR2. It is well known that oxidation of RyR2 is commonly associated with increased channel activity and ER/SR Ca2+ leak (reviewed in refs. 44,45). In the present study, we discovered that oxidation of only two cross-linking cysteines was responsible for half of the channel’s functional response to oxidative stress. It appears that defective intersubunit dynamics due to RyR2 intersubunit cross-linking plays a key role in RyR2 dysfunction during oxidative stress.
Discovered redox-sensitive cysteines 1078 and 2991 are conserved in RyR2 but are missing in RyR1, indicating a unique redox-sensitive pair for the cardiac isoform of the channel (Supplementary Fig. 11). They contribute to the ~50% of RyR2 functional activation during oxidative stress, which raises a question about the additional RyR2 sites involved in redox signaling of the channel. It has been shown that the RyR2 molecule comprises more than 30 cysteines accessible to redox modifications46,47. In addition to disulfide formation, the mechanisms of direct RyR2 regulation by oxidative stress may include a sulfonic acid formation, S-nitrosylation and S-glutathionylation46,48,49. Alternatively, the indirect mechanism of redox regulation might can be mediated by CaM dissociation from the RyR2 complex11,12,13. Recently, it has been shown that RyR2 redox regulation also occurs in the SR lumen of cardiomyocytes, where oxidative stress disrupts the redox-dependent association between RyR2 and ERp44, through the upregulation of Ero1α reductase50.
Although the NTR of hRyR2 was shown to be essential for the structural integrity of the channel33,38, our experiments revealed that the individual cysteines from this region were not directly involved in RyR2 intersubunit cross-linking. At the same time, we believe that NTR domains A and B of RyR2 have an allosteric effect on hRyR2 structural re-arrangements during oxidative stress. The complete removal of domains A and B substantially reduced the levels of hRyR2 cross-linking and prevented ER Ca2+ leak caused by oxidative stress. It appears the intact structure of the NTR is required for two cross-linking cysteines in neighboring RyR2 subunits to be in close proximity for a formation of disulfide bond. It has been shown that recombinant RyR2 can self-associate in an organized manner51, suggesting a possibility for channel-channel cross-linking. However, both the P1-SPRY2 linker harboring C1078 and the loop that includes C2991 are parts of the RyR2 inner intersubunit area and do not belong to the proposed channel-channel self-organization interfaces41.
In conclusion, the results of this study revealed two critical redox-sensitive residues in human RyR2. We found that, out of 89 cysteines in each hRyR2 subunit, cysteines 1078 and 2991 play a critical role in the channel’s functional response to oxidative stress. Given the structural orientation on the neighboring RyR2 subunits, we suggest that cysteines 1078 and 2991 form a redox sensor pair capable of forming an intersubunit disulfide bond during oxidative stress, resulting in structural remodeling of RyR2 and the formation of pathological “leaky” channel.
Methods
Vector production
The production of vectors encoding the RyR2, R-CEPIA1er, and SERCA2a has been previously described in ref. 29. The vector encoding human RyR2 cDNA fused to a green fluorescent protein (GFP) at the N-terminus was a gift from Dr. Christopher George (University of Cardiff). pCMV R-CEPIA1er was kindly provided by Dr. Masamitsu Iino (Addgene plasmid 58216; http://n2t.net/addgene:58216; RRID: Addgene_58216). The vector encoding human SERCA2a cDNA was a gift from Dr. David Thomas (University of Minnesota). The SERCA2a cDNA was cloned into the mCerulean-M1 modified plasmid (Addgene, plasmid # 15214), yielding SERCA2a recombinant protein fused to a modified cerulean fluorescent protein (mCer) at the NH2-terminus. The mutations in hRyR2 (hRyR2ΔAB, hRyR2ΔA hRyR2C24S, hRyR2C36S, hRyR2C47S/C65S, hRyR2C131S/C132S/C158S, hRyR2C1078S, hRyR2C2991S, hRyR2C1078S/C2991S) were introduced by PCR using specific primers containing the mutations (IDT, USA. Sequences of these primers are listed in the Supplementary Table 2), and the Q5 mutagenesis kit (Nee England Biolabs, USA). After verification of the mutagenesis by single-pass analysis, the plasmids were amplified and used for experimentation. To generate the C131S/C132S RyR2 mutant we used primers simultaneously containing both point mutations. However, all the other multiple cysteine RyR2 mutants were generated by multi mutagenesis steps. Every new mutation was introduced using RyR2 already containing mutations at other residues as a backbone.
Measurements of RyR2 cross-linking
HEK293 cells (AAV pro 293 T Cell Line, Takara, #632273) were grown on 100 mm plastic dishes for 24 h and then transfected with the wild-type or mutant GFP-hRyR2 plasmid using polyethylenimine (PEI, 1 µM/ml). Cells were further grown for another 48 h. Cells were harvested and resuspended in the solution containing (in mM): 150 K-aspartate, 0.25 MgCl2, 0.1 EGTA, 10 HEPES, and pH 7.2. The cells were treated with corresponding diamide concentrations, harvested, and resuspended in lysis buffer (150 mM NaCl, 25 mM Hepes pH 7.4, 1% Triton X-100, phosphatases inhibitors) with 5 mM N-ethylmaleimide (NEM) to block free sulfhydryl groups. Lysate samples were incubated with non-reducing Laemmli buffer for 10 min, ran on 2–10% gradient SDS-PAGE, and blotted overnight onto the nitrocellulose membrane. Cross-linking was detected using F1 anti-RyR2 primary antibody (1:1000; sc-376507, Santa Cruz, U.S.A.) and HRP-conjugated goat anti-mouse secondary antibody (1:5000; 1430, Thermo Fisher Scientific, U.S.A.). For N-truncated hRyR2 mutants rabbit polyclonal anti-GFP antibody (1:1000; A6455, Thermo Fisher Scientific, U.S.A.) was used as primary, and HRP-conjugated goat anti-rabbit antibody (1:5000, A0545, Sigma-Aldrich Co, U.S.A.) as secondary. The amount of hRyR2 cross-linking was analyzed in ImageJ software (NIH, USA) based on the disappearance of the monomeric 560 kDa band using the densitometry analysis.
Generation of the inducible T-Rex SERCA2a stable cell line
The generation of stable inducible Flp-In T-Rex-293 cell line expressing SERCA2a using the Flp-In T-REx Core Kit (Invitrogen) has been previously described in ref. 29. Briefly, SERCA2a cDNA was cloned into the inducible expression vector pcDNA5/FRT/TO. Obtained vector, together with the pOG44 vector encoding the Flp recombinase, was used to co-transfect Flp-In-T-Rex293 cells (R780-07, Thermo Fisher Scientific, U.S.A.). Forty-eight hours after transfection, the growth medium was replaced with a selection medium containing 100 µg/ml hygromycin (Invitrogen). The selection medium was changed every 2 days until the stable line cell foci could be isolated. The hygromycin-resistant cell foci were selected and expanded. Stable line cells were cultured in high-glucose DMEM supplemented with 100 U/ml penicillin, 100 mg/ml streptomycin, and 10% fetal bovine serum at 5% CO2 and 37 °C.
Measurements of the ER luminal [Ca2+] ([Ca2+]ER)
For functional Ca2+ measurements we used HEK293 or T-Rex SERCA2a stable cells. Stable cells were co-transfected with plasmids containing GFP-hRyr2 and R-CEPIA1er genes. HEK293 cells were also transfected with the plasmid containing the mCer-SERCA2a gene to ensure the physiological ER Ca2+ concentrations. Experiments were carried out 48 h after transfection when expression of the exogenous genes was optimal. All experiments were conducted using laser scanning confocal microscopy (Radiance 2000 MP, BioRad, UK) equipped with a 40x oil objective lens (N.A. = 1.3). The cells prepared for an experiment were bathed in Tyrode solution containing (in mM): 135 NaCl, 4 KCl, 2 CaCl2, 1 MgCl2, 10 D-glucose, and 10 HEPES. To record changes in [Ca2+]ER, R-CEPIA1er was excited with the 543 nm line of a He-Ne laser, and fluorescence was measured at > 580 nm. For measurements of [Ca2+]ER load-leak balance, 2D images (512 × 512 pixels) were collected every 5 s at a scanning speed of 6 ms/line. The cells were treated with corresponding diamide concentration for at least 15 min until the complete equilibration of [Ca2+]ER levels. At the end of each experiment, caffeine (10 mM) was applied to induce complete depletion of [Ca2+]ER. Then, Fmax was estimated by applying 2 µM ionomycin, escin 0.005%, and 10 mM Ca2+. The changes in [Ca2+]ER were presented normalized according to [Ca2+]ER = (F-Fmin)/(Fmax-Fmin). To obtain the change in [Ca2+]ER caused by oxidative stress, Ca2+ wave peak values were averaged before and after diamide application, and the relative change was calculated.
Statistics
All data are presented as means ± SEM of n measurements. Obtained 2D images were analyzed using ImageJ software (NIH, USA). Groups were compared using the two-sided Student’s t-test for unpaired data sets or one-way ANOVA followed by Tukey’s post-hoc test where appropriate. Differences were considered statistically significant at P < 0.05. Peak analysis, statistical analysis, and graphical representation of averaged data were carried out on Origin 2021b SR2 software (OriginLab, U.S.A.).
Structural analysis and model visualization
The cryo-EM models of RyR2 were obtained from RSCB protein data bank. The structural analysis was performed using PyMOL 1.7.0.3, WinCoot 0.9.8.1 and AREAIMOL program from CCP4i package (v. 8.0.007). PyMOL 1.7.0.3 was used to create graphical representations of the RyR2 structural models.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
The data supporting this article and other findings are available within the manuscript, figures, supplementary data, and from the corresponding authors upon request. The cryo-EM models of RyR2 used in this study 7UA5, 7UA9, 7U9T, 7U9Q, 7U9R, 7UA3, 7UA1, 7U9X, 7UA4, 7U9Z, 6WOU, 6WOV were obtained from Protein Data Bank. Source data are provided with this paper.
References
Berridge, M. J., Bootman, M. D. & Roderick, H. L. Calcium signalling: dynamics, homeostasis and remodelling. Nat. Rev. Mol. Cell Biol. 4, 517–529 (2003).
Bers, D. M. Cardiac excitation-contraction coupling. Nature 415, 198–205 (2002).
Marks, A. R. Calcium cycling proteins and heart failure: mechanisms and therapeutics Find the latest version: Review series Calcium cycling proteins and heart failure: mechanisms and therapeutics. J. Clin. Invest. 123, 46–52 (2013).
Bovo, E., Mazurek, S. R. & Zima, A. V. Oxidation of ryanodine receptor after ischemia-reperfusion increases propensity of Ca 2+ waves during β-adrenergic receptor stimulation. Am. J. Physiol. Circ. Physiol. 315, H1032–H1040 (2018).
Terentyev, D. et al. Redox Modification of Ryanodine Receptors Contributes to Sarcoplasmic Reticulum Ca 2+ Leak in Chronic Heart Failure. Circ. Res. 103, 1466–1472 (2008).
Belevych, A. E. et al. Redox modification of ryanodine receptors underlies calcium alternans in a canine model of sudden cardiac death. Cardiovasc. Res. 84, 387–395 (2009).
Bovo, E., Mazurek, S. R. & Zima, A. V. The role of RyR2 oxidation in the blunted frequency-dependent facilitation of Ca2+transient amplitude in rabbit failing myocytes. Pflugers Arch. Eur. J. Physiol. 470, 959–968 (2018).
Prosser, B. L., Ward, C. W. & Lederer, W. J. X-ROS signaling: Rapid mechano-chemo transduction in heart. Science 333, 1440–1445 (2011).
Yan, Y. et al. Bidirectional regulation of Ca2+ sparks by mitochondria-derived reactive oxygen species in cardiac myocytes. Cardiovasc. Res. 77, 432–441 (2008).
Bovo, E., Lipsius, S. L. & Zima, A. V. Reactive oxygen species contribute to the development of arrhythmogenic Ca2+ waves during β-adrenergic receptor stimulation in rabbit cardiomyocytes. J. Physiol. 590, 3291–3304 (2012).
Oda, T. et al. Oxidation of ryanodine receptor (RyR) and calmodulin enhance Ca release and pathologically alter, RyR structure and calmodulin affinity. J. Mol. Cell. Cardiol. 85, 240–248 (2015).
Ono, M. et al. Dissociation of calmodulin from cardiac ryanodine receptor causes aberrant Ca2+ release in heart failure. Cardiovasc. Res. 87, 609–617 (2010).
Nikolaienko, R. et al. The functional significance of redox-mediated intersubunit cross-linking in regulation of human type 2 ryanodine receptor. Redox Biol. 37, 101729 (2020).
Belevych, A. et al. Enhanced ryanodine receptor-mediated calcium leak determines reduced sarcoplasmic reticulum calcium content in chronic canine heart failure. Biophys. J. 93, 4083–4092 (2007).
Domeier, T. L., Blatter, L. A. & Zima, A. V. Alteration of sarcoplasmic reticulum Ca2+ release termination by ryanodine receptor sensitization and in heart failure. J. Physiol. 587, 5197–5209 (2009).
Mochizuki, M. et al. Scavenging Free Radicals by Low-Dose Carvedilol Prevents Redox-Dependent Ca2+ Leak Via Stabilization of Ryanodine Receptor in Heart Failure. J. Am. Coll. Cardiol. 49, 1722–1732 (2007).
Lobo, P. A. & Van Petegem, F. Crystal Structures of the N-Terminal Domains of Cardiac and Skeletal Muscle Ryanodine Receptors: Insights into Disease Mutations. Structure 17, 1505–1514 (2009).
Marx, S. O. et al. PKA Phosphorylation Dissociates FKBP12.6 from the Calcium Release Channel (Ryanodine Receptor). Cell 101, 365–376 (2000).
Wehrens, X. H. T. Ca2+/Calmodulin-Dependent Protein Kinase II Phosphorylation Regulates the Cardiac Ryanodine Receptor. Circ. Res. 94, e61–e70 (2004).
Xiao, B. et al. Ser-2030, but not Ser-2808, is the major phosphorylation site in cardiac ryanodine receptors responding to protein kinase A activation upon β-adrenergic stimulation in normal and failing hearts. Biochem. J. 396, 7–16 (2006).
Marx, S. O. & Marks, A. R. Dysfunctional ryanodine receptors in the heart: New insights into complex cardiovascular diseases. J. Mol. Cell. Cardiol. 58, 225–231 (2013).
Mazurek, S. R., Bovo, E. & Zima, A. V. Regulation of sarcoplasmic reticulum Ca(2+) release by cytosolic glutathione in rabbit ventricular myocytes. Free Radic. Biol. Med. 68, 159–167 (2014).
Peng, W. et al. Structural basis for the gating mechanism of the type 2 ryanodine receptor RyR2. Science 354, aah5324 (2016).
Zaidi, N. F., Lagenaur, C. F., Abramson, J. J., Pessah, I. & Salama, G. Reactive disulfides trigger Ca2+ release from sarcoplasmic reticulum via an oxidation reaction. J. Biol. Chem. 264, 21725–21736 (1989).
Sun, J., Xin, C., Eu, J. P., Stamler, J. S. & Meissner, G. Cysteine-3635 is responsible for skeletal muscle ryanodine receptor modulation by NO. Proc. Natl. Acad. Sci. USA. 98, 11158–11162 (2001).
Aracena-Parks, P. et al. Identification of cysteines involved in S-nitrosylation, S-glutathionylation, and oxidation to disulfides in ryanodine receptor type 1. J. Biol. Chem. 281, 40354–40368 (2006).
Moore, C. P., Zhang, J. Z. & Hamilton, S. L. A role for cysteine 3635 of RYR1 in redox modulation and calmodulin binding. J. Biol. Chem. 274, 36831–36834 (1999).
Zhang, H., Zhang, J. Z., Danila, C. I. & Hamilton, S. L. A noncontiguous, intersubunit binding site for calmodulin on the skeletal muscle Ca2+ release channel. J. Biol. Chem. 278, 8348–8355 (2003).
Bovo, E. et al. Novel approach for quantification of endoplasmic reticulum Ca 2+ transport. Am. J. Physiol. Circ. Physiol. 316, H1323–H1331 (2019).
Suzuki, J. et al. Imaging intraorganellar Ca2+ at subcellular resolution using CEPIA. Nat. Commun. 5, 4153 (2014).
Bovo, E., Martin, J. L., Tyryfter, J., De Tombe, P. P. & Zima, A. V. R-CEPIA1er as a new tool to directly measure sarcoplasmic reticulum [Ca] in ventricular myocytes. Am. J. Physiol. - Hear. Circ. Physiol. 311, H268–H275 (2016).
Zhang, J. et al. Cardiac ryanodine receptor N-terminal region biosensors identify novel inhibitors via FRET-based high-throughput screening. J. Biol. Chem. 298, 101412 (2022).
Zissimopoulos, S. et al. N-terminus oligomerization regulates the function of cardiac ryanodine receptors. J. Cell Sci. 126, 5042–5051 (2013).
Liu, Y. et al. Roles of the NH2-terminal domains of cardiac ryanodine receptor in Ca2+ release activation and termination. J. Biol. Chem. 290, 7736–7746 (2015).
Faltinova, A., Tomaskova, N., Antalik, M., Sevcik, J. & Zahradnikova, A. The N-terminal region of the ryanodine receptor affects channel activation. Front. Physiol. 8, 443 (2017).
Zhang, Y., Rabesahala de Meritens, C., Beckmann, A., Lai, F. A. & Zissimopoulos, S. Defective ryanodine receptor N-terminus inter-subunit interaction is a common mechanism in neuromuscular and cardiac disorders. Front. Physiol. 13, 1–7 (2022).
Tung, C. C., Lobo, P. A., Kimlicka, L. & Van Petegem, F. The amino-terminal disease hotspot of ryanodine receptors forms a cytoplasmic vestibule. Nature 468, 585–588 (2010).
Zhong, X. et al. Conformational dynamics inside amino-terminal disease hotspot of ryanodine receptor. Structure 21, 2051–2060 (2013).
Gong, D. et al. Modulation of cardiac ryanodine receptor 2 by calmodulin. Nature 572, 347–351 (2019).
Iyer, K. A. et al. Structural mechanism of two gain-of-function cardiac and skeletal RyR mutations at an equivalent site by cryo-EM. Sci. Adv. 6, 1–13 (2020).
Miotto, M. C. et al. Structural analyses of human ryanodine receptor type 2 channels reveal the mechanisms for sudden cardiac death and treatment. Sci. Adv. 8, eabo1272 (2022).
Cumming, R. C. et al. Protein disulfide bond formation in the cytoplasm during oxidative stress. J. Biol. Chem. 279, 21749–21758 (2004).
Han, H. M., Wei, R. S., Lai, A. F. & Yin, C. C. Molecular nature of sulfhydryl modification by hydrogen peroxide on type 1 ryanodine receptor. Acta Pharmacol. Sin. 27, 888–894 (2006).
Nikolaienko, R., Bovo, E. & Zima, A. V. Redox Dependent Modifications of Ryanodine Receptor: Basic Mechanisms and Implications in Heart Diseases. Front. Physiol. 9, 1775 (2018).
Zima, A. V. & Mazurek, S. R. Functional impact of ryanodine receptor oxidation on intracellular calcium regulation in the heart. Rev. Physiol. Biochem. Pharmacol. 171, 39–62 (2016).
Xu, L. Activation of the Cardiac Calcium Release Channel (Ryanodine Receptor) by Poly-S-Nitrosylation. Science 279, 234–237 (1998).
Donoso, P., Sanchez, G., Bull, R. & Hidalgo, C. Modulation of cardiac ryanodine receptor activity by ROS and RNS. Front. Biosci. 16, 553–567 (2011).
Sánchez, G., Pedrozo, Z., Domenech, R. J., Hidalgo, C. & Donoso, P. Tachycardia increases NADPH oxidase activity and RyR2 S-glutathionylation in ventricular muscle. J. Mol. Cell. Cardiol. 39, 982–991 (2005).
Becerra, R. et al. Reversible redox modifications of ryanodine receptor ameliorate ventricular arrhythmias in the ischemic-reperfused heart. Am. J. Physiol. - Hear. Circ. Physiol. 311, H713–H724 (2016).
Hamilton, S. et al. Ero1α -Dependent ERp44 Dissociation from RyR2 Contributes to Cardiac Arrhythmia. Circ. Res. 130, 711–724 (2022).
Cabra, V., Murayama, T. & Samsó, M. Ultrastructural Analysis of Self-Associated RyR2s. Biophys. J. 110, 2651–2662 (2016).
Acknowledgements
This work was supported by the National Institutes of Health Grants R01HL151990 (to A.V.Z.). The authors would like to thank Dr. Christopher George (University of Cardiff, UK) for providing the vector encoding the human RyR2. The authors also would like to thank Dr. Masamitsu Iino for donating the R-CEPIA1er vector.
Author information
Authors and Affiliations
Contributions
R.N. and A.V.Z. conceived and supervised the study. R.N., E.B., R.G., T.J., and A.V.Z designed and performed Ca2+ imaging and biochemistry experiments. E.B. and D.K. performed DNA cloning experiments. R.N., E.B., R.G., and T.J. performed data analysis. R.N. and A.V.Z. wrote the manuscript. All the authors read and approved the manuscript version to be published.
Corresponding author
Ethics declarations
Competing interests
The authors declare on competing interests.
Peer review
Peer review information
Nature Communications thanks Isaac Pessah and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Source data
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Nikolaienko, R., Bovo, E., Kahn, D. et al. Cysteines 1078 and 2991 cross-linking plays a critical role in redox regulation of cardiac ryanodine receptor (RyR). Nat Commun 14, 4498 (2023). https://doi.org/10.1038/s41467-023-40268-z
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-023-40268-z
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.