Redox Sensitivity of the Ryanodine Receptor Interaction with FK506-binding Protein*

The ryanodine receptor (RyR) calcium release channel functions as a redox sensor that is sensitive to channel modulators. The FK506-binding protein (FKBP) is an important regulator of channel activity, and disruption of the RyR2-FKBP12.6 association has been implicated in cardiac disease. In the present study, we investigated whether the RyR-FKBP association is redox-regulated. Using co-immunoprecipitation assays of solubilized native RyR2 from cardiac muscle sarcoplasmic reticulum (SR) with recombinant [35S]FKBP12.6, we found that the sulfydryl-oxidizing agents, H2O2 and diamide, result in diminished RyR2-FKBP12.6 binding. Co-sedimentation experiments of cardiac SR vesicles with [35S]FKBP12.6 also demonstrated that oxidizing reagents decreased FKBP binding. Matching results were obtained with skeletal muscle SR. Notably, H2O2 and diamide differentially affected the RyR2-FKBP12.6 interaction, decreasing binding to ∼75 and ∼50% of control, respectively. In addition, the effect of H2O2 was negligible when the channel was in its closed state or when applied after FKBP binding had occurred, whereas diamide was always effective. A cysteine-null mutant FKBP12.6 retained redox-sensitive interaction with RyR2, suggesting that the effect of the redox reagents is exclusively via sites on the ryanodine receptor. K201 (or JTV519), a drug that has been proposed to prevent FKBP12.6 dissociation from the RyR2 channel complex, did not restore normal FKBP binding under oxidizing conditions. Our results indicate that the redox state of the RyR is intimately connected with FKBP binding affinity.

The ryanodine receptor (RyR) calcium release channel functions as a redox sensor that is sensitive to channel modulators. The FK506-binding protein (FKBP) is an important regulator of channel activity, and disruption of the RyR2-FKBP12.6 association has been implicated in cardiac disease. In the present study, we investigated whether the RyR-FKBP association is redoxregulated. Using co-immunoprecipitation assays of solubilized native RyR2 from cardiac muscle sarcoplasmic reticulum (SR) with recombinant [ 35 S]FKBP12.6, we found that the sulfydryloxidizing agents, H 2 O 2 and diamide, result in diminished RyR2-FKBP12.6 binding. Co-sedimentation experiments of cardiac SR vesicles with [ 35 S]FKBP12.6 also demonstrated that oxidizing reagents decreased FKBP binding. Matching results were obtained with skeletal muscle SR. Notably, H 2 O 2 and diamide differentially affected the RyR2-FKBP12.6 interaction, decreasing binding to ϳ75 and ϳ50% of control, respectively. In addition, the effect of H 2 O 2 was negligible when the channel was in its closed state or when applied after FKBP binding had occurred, whereas diamide was always effective. A cysteine-null mutant FKBP12.6 retained redox-sensitive interaction with RyR2, suggesting that the effect of the redox reagents is exclusively via sites on the ryanodine receptor. K201 (or JTV519), a drug that has been proposed to prevent FKBP12.6 dissociation from the RyR2 channel complex, did not restore normal FKBP binding under oxidizing conditions. Our results indicate that the redox state of the RyR is intimately connected with FKBP binding affinity.
Ryanodine receptors (RyRs) 2 are tetrameric intracellular Ca 2ϩ channels that mediate the release of Ca 2ϩ from the sarco/ endoplasmic reticulum in muscle and nonmuscle cells (1). Three genes coding for mammalian RyRs have been identified: RyR1 in skeletal muscle, RyR2 in heart and brain, and RyR3 in a number of tissues. The deduced primary structure of all RyRs suggests a hydrophobic C terminus forming the channel pore, with the remaining ϳ80% being cytoplasmic. RyR channel activity is regulated by Ca 2ϩ , Mg 2ϩ , ATP, phosphorylation and redox status, and a number of accessory proteins.
The RyR functions as a redox sensor that is sensitive to channel modulators (22), and channel activity has been correlated with the number of free sulfydryl groups (23). In general, oxidizing reagents activate, whereas reducing reagents inhibit, channel activity (24,25). Pharmacological sulfydryl-reactive reagents, including thimerosal, dithiodipyridines, N-ethylmaleimide, and diamide, have been shown to activate skeletal and cardiac muscle RyRs (26 -28). This effect was reversed by reducing reagents, such as dithiothreitol (DTT), and reduced glutathione. Glutathione, which constitutes the major redox buffer system in eukaryotic cells, inhibits the channel in its reduced form (GSH), whereas its oxidized form (GSSG) is stimulatory (29). Reactive oxygen species, such as hydrogen peroxide (H 2 O 2 ) and superoxide anion radical, activate the channel (30 -33). A sarcoplasmic reticulum (SR)-associated NAD(P)H oxidase coupled to superoxide anion radical production has been shown to modulate RyR-mediated Ca 2ϩ release in skeletal and cardiac muscle (33)(34)(35)(36). Oxidation-induced RyR activation is due to increased sensitivity of the channel to Ca 2ϩ activation as well as a decrease in Mg 2ϩ inhibition (24,25).

EXPERIMENTAL PROCEDURES
Cell-free Protein Expression-In vitro cell-free protein expression was carried out using the TNT T7 quick coupled transcription and translation system (Promega). Reactions were carried out in 10-l volumes by adding the TNT mix with 1 g of plasmid DNA together with 1 l (0.53 MBq or 14 Ci) of [ 35 S]methionine (Pro-Mix; Amersham Biosciences). Reaction samples were incubated in a 30°C water bath for 90 min and terminated by placing on ice. Hemoglobin was removed from the TNT reactions using one bed volume of Ni 2ϩ -nitriloacetate resin (Qiagen).
Co-immunoprecipitation Assays-Cardiac heavy SR vesicles (1 mg), prepared as described previously (12), were solubilized in 200 l of IP buffer (20 mM Tris, 150 mM NaCl, 0.4% CHAPS, pH 7.4, and Complete protease inhibitors (Roche Applied Science)) by overnight incubation at 4°C with continuous mixing. The insoluble material was pelleted at 20,000 ϫ g for 10 min at 4°C, and the supernatant was withdrawn and treated with an appropriate redox reagent (2 mM DTT, 1 mM H 2 O 2 , 200 M diamide) for 30 min at room temperature. In vitro synthesized, hemoglobin-free, radiolabeled FKBP12.6 was added and incubated for 1 h at room temperature. The RyR2-specific Ab 1093 was added at a 1:40 dilution, and the sample was incubated for 2 h at 4°C, followed by the addition of 20 l of protein G Dynabeads (Dynal) and incubation for a further 2 h with continuous mixing. Protein immunocomplexes were isolated with the use of a magnetic particle concentrator (MPC-S; Dynal), and beads were washed three times with IP buffer. Immunoprecipitated proteins were resuspended in SDS-PAGE loading buffer, heated at 85°C for 5 min, and analyzed by SDS-PAGE and autoradiography.
For the experiments to test the effect of K201, the drug (10 M) was added to the IP samples before mixing with [ 35 S]FKBP12.6. For the experiments to test whether oxidizing conditions induce FKBP12.6 dissociation, redox reagents were applied for 30 min at room temperature or for 6 h at 4°C at the end of the IP protocol, once RyR2-FKBP12.6 immunocomplexes had formed following overnight incubation.
Co-sedimentation Assays-Cardiac (200 g) or skeletal muscle (100 g) heavy SR vesicles were resuspended in 200 l of buffer (10 mM Na 2 -Pipes, 120 mM KCl, pH 7.4, and Complete protease inhibitors (Roche Applied Science)) and treated with an appropriate redox reagent (2 mM DTT, 5 mM GSH, 2 mM GSSG, 1 mM H 2 O 2 , 200 M diamide) for 30 min at room temperature. In vitro synthesized, hemoglobin-free, radiolabeled FKBP was added and incubated for 6 h at 4°C with continuous mixing. SR vesicles were recovered at 20,000 ϫ g for 10 min at 4°C, washed once, resuspended in SDS-PAGE loading buffer, heated at 85°C for 5 min, and analyzed by SDS-PAGE and autoradiography.
Autoradiography of [ 35 S]FKBP12.6-Radiolabeled proteins were separated by SDS-PAGE, and the gel was fixed (40% methanol, 10% acetic acid) for 30 min. The fixing solution was removed, and the gel was incubated with a fluorographic enhancer (Amplify; Amersham Biosciences) for a further 30 min and subsequently dried. The dried gel was exposed to an x-ray film (Hyperfilm; Amersham Biosciences) for variable periods of time. Densitometric analysis was performed using a GS-700 scanner (Bio-Rad) and Quantity One software (Bio-Rad).

Glutathione-induced Aggregation of RyR and FKBP-
Initially, we tested the effect of redox reagents (i.e. 2 mM DTT, 5 mM GSH, 2 mM GSSG, 1 mM H 2 O 2 , 200 M diamide) on the solubilization of the RyR. Heavy SR vesicles were treated with redox reagents for 1 h at room temperature, pelleted to remove reagents, and then incubated overnight in IP buffer followed by a centrifugation step to remove the insoluble material. We estimate that under our conditions, ϳ50% of the RyR is solubilized in control or DTT-treated samples. As shown in Fig. 1A, the sulfydryl-oxidizing reagents H 2 O 2 and diamide reduces RyR solubilization from both skeletal and cardiac SR. This could be due to increased protein aggregation, since such oxidants have been shown to induce intra-and intermolecular disulfide bonds within RyR (37,48) and with the SR integral protein triadin (38,39). Surprisingly, RyR could not be solubilized from SR samples that had been treated with glutathione. No soluble RyR1 was detected from skeletal muscle SR treated with either reduced or oxidized glutathione or soluble RyR2 treated with GSH, whereas there was modest solubilization of RyR2 from cardiac SR treated with GSSG (Fig. 1A). We subsequently examined the effect of redox reagents on RyR, postsolubilization. Solubilized RyR remained soluble following treatment with DTT, H 2 O 2 , or diamide but became insoluble with GSH or GSSG (Fig. 1B). Some soluble RyR2 remained after treatment with GSSG.
We also tested for any glutathione effects on FKBP12/12.6. In vitro synthesized [ 35 S]FKBP12.6 was incubated with 5 mM GSH or 2 mM GSSG in the presence or absence of 10 M rapamycin, followed by a centrifugation step. As shown in Fig. 1C, FKBP12.6 became insoluble following treatment with GSH. Interestingly, this effect was prevented by rapamycin, suggesting that the glutathione effect involves the FK506-and rapamycin-binding pocket. Treatment with GSSG occasionally resulted in partial insolubility that was again prevented by rapamycin, whereas DTT, H 2 O 2 , and diamide had no effect (data not shown). Identical results were obtained for FKBP12 (data not shown).
The above results suggest that solubilized RyR as well as FKBP form protein aggregates in the presence of glutathione, followed by centrifugation at 20,000 ϫ g for 10 min at 4°C to remove insoluble material. The soluble supernatant was withdrawn, the pellet was resuspended in 200 l of IP buffer, and 40 l of each was loaded on 4% SDS-polyacrylamide gels for Western blot analysis. B, skeletal (200 g) or cardiac heavy SR (1 mg) was solubilized in the absence of any redox reagent as described above, and the insoluble material was removed. Solubilized SR was treated with redox reagents (2 mM DTT, 5 mM GSH, 2 mM GSSG, 1 mM H 2 O 2 , 200 M diamide) for 6 h at 4°C followed by centrifugation at 20,000 ϫ g for 10 min at 4°C. The soluble supernatant was withdrawn, the pellet was resuspended in 200 l of IP buffer, and 40 l of each was loaded on 4% SDS-polyacrylamide gels for Western blot analysis. C, in vitro synthesized, hemoglobin-free [ 35 S]FKBP12.6 in IP buffer was incubated for 6 h at 4°C with 5 mM GSH or 2 mM GSSG, in the presence or absence of 10 M rapamycin, followed by centrifugation at 14,000 ϫ g for 10 min at 4°C. The supernatant was discarded, and the pellet was resuspended in SDS-PAGE loading buffer and analyzed by SDS-PAGE (15% gel) and autoradiography. An aliquot of the TNT reaction, equivalent to the volume processed as described above, was included in the first lane of the autoradiogram. and therefore the effect of the latter on RyR-FKBP association cannot be investigated. Also, in subsequent IP experiments, DTT, H 2 O 2 , and diamide were added once SR vesicles had been solubilized, in order to ensure that identical amounts of soluble RyR are processed in all IP samples examined.
Sulfydryl-oxidizing Reagents Weaken the RyR-FKBP Interaction-The redox sensitivity of the RyR2-FKBP12.6 interaction was tested by co-immunoprecipitation assays. Solubilized cardiac heavy SR was treated with an appropriate redox reagent (2 mM DTT, 1 mM H 2 O 2 , 200 M diamide) for 30 min at room temperature followed by incubation with exogenous [ 35 S]FKBP12.6. RyR2 was immunoprecipitated with the isoform-specific Ab 1093 and the presence of co-precipitated FKBP12.6 was analyzed by SDS-PAGE and autoradiography. As shown in Fig. 2A, there was less FKBP12.6 recovered in the RyR2 immunoprecipitate when cardiac SR was treated with the sulfydryl-oxidizing reagents H 2 O 2 and diamide compared with control or DTT-treated SR. On average, H 2 O 2 decreased FKBP12.6 binding by ϳ25% and diamide by ϳ50%, whereas DTT had no effect (Table 1). This result suggests that diamide reacts with different or additional RyR2 cysteine(s) compared with H 2 O 2 .
The effects of H 2 O 2 and diamide on the binding interaction could be due to cysteine(s) modification of the RyR2 and/or FKBP12.6. In order to test whether the two FKBP12.6 cysteine residues are involved, we generated the double mutant protein, FKBP12.6 C23S/C77I . Cys 77 was replaced by Ile, the corresponding residue present in FKBP12, whereas Ser was substituted for Cys 23 to preserve the size and polarity of the side chain at that position. Both cysteines are located on the outer surface of the protein, away from the FK506binding pocket, and they have not been previously impli-cated in the interaction with the RyR. Using co-IP assays, we found that the C23S/C77I double mutant FKBP12.6 retains binding to the RyR2 and that binding is weakened by H 2 O 2 and diamide (Fig.  2B). FKBP12.6 C23S/C77I binding to RyR2 was 102.2 Ϯ 7.8% for DTT, 74.7 Ϯ 6.3% for H 2 O 2 , and 47.5 Ϯ 7.1% for diamide compared with control (n ϭ 5). These data with FKBP12.6 C23S/C77I are identical to data obtained for the wild-type protein ( Table 1), suggesting that FKBP12.6 cysteine modification is not directly involved in the interaction with RyR.
The redox sensitivity of the RyR2 interaction with wild-type FKBP12.6 was also tested by co-sedimentation assays. The centrifugation-based binding assay is conducted in a detergentfree environment with the RyR resident in the SR membrane. Thus, the effect of redox reagents on RyR2 in its native conformation can be studied, as opposed to with solubilized RyR2, where normally inaccessible cysteine residue(s) may be exposed to redox reagents. Intact cardiac heavy SR was treated with an appropriate redox reagent (2 mM DTT, 1 mM H 2 O 2 , 200 M diamide) for 30 min at room temperature followed by incubation with exogenous [ 35 S]FKBP12.6. SR vesicles were pelleted, and the presence of co-sedimenting FKBP12.6 was analyzed by SDS-PAGE and autoradiography. We observed less FKBP12.6 co-sedimenting with cardiac SR treated with H 2 O 2 and diamide compared with control or DTT-treated SR (Fig. 3A). Cumulative data demonstrate that H 2 O 2 decreased FKBP12.6 binding by ϳ10% and diamide by ϳ35% (Table 2). These values are lower than those obtained by co-IP assays, which could be due to increased accessibility of cysteine(s) or modification of additional residue(s) in detergent-solubilized RyR2.
Effect of H 2 O 2 Depends on Activation State of the Channel-We next examined whether the open or closed state of the RyR2 channel affects the redox sensitivity of the interaction with FKBP12.6. The RyR is known to undergo conformational changes in its transition from the open to the closed state (50,51). It is plausible that such conformational changes may expose cysteine residue(s), thereby providing additional targets for sulfydryl-oxidizing agents. Alternatively, conformational  Co-immunoprecipitation experiments to determine the redox sensitivity of the RyR-FKBP interaction were carried out as described in the legend to Fig. 2, followed by quantitation of the ͓ 35 S͔FKBP12.6 band by densitometric analysis of autoradiograms and normalization against control. changes may bury cysteine(s), thereby blunting the effect of oxidants.

Control
In order to distinguish between these two possibilities, we carried out co-IP and co-sedimentation assays under conditions that should promote a predominantly open (100 M CaCl 2 ) or closed RyR2 channel (2 mM EGTA or 2 mM MgCl 2 ). These results, summarized in Tables 1 and 2, show that DTT treatment is largely without any effect, whereas diamide reduced FKBP12.6 binding to the same extent (by ϳ50 or ϳ35%, depending on the assay), irrespective of the RyR2 acti-vation state. On the other hand, H 2 O 2 treatment reduced FKBP12.6 binding (by ϳ25 or ϳ10%, depending on the assay) when the channel was activated by 100 M CaCl 2 but had almost no effect when the channel was inhibited by 2 mM EGTA or 2 mM MgCl 2 . Under ambient conditions, H 2 O 2 reduced FKBP12.6 binding to the same extent as that obtained with 100 M CaCl 2 , most probably because RyR2 is in a partially open configuration due to contaminating calcium in the buffers. These data indicate that, whereas diamide has access to its cysteine target(s) irrespective of RyR2 channel state, H 2 O 2 has access to its target(s) in only the open configuration.
Diamide, but Not H 2 O 2 , Induces FKBP12.6 Dissociation-The above results demonstrate that oxidized RyR2 has a diminished binding affinity for FKBP12.6. The modified RyR2 cysteine(s) could be directly involved in the interaction with FKBP12.6 or indirectly through an allosteric mechanism. It is also possible that FKBP12.6 binding provides a direct steric hindrance effect, denying free access to the oxidants and thereby protecting RyR2 cysteine(s) from oxidation. Thus, it was of interest to determine whether oxidizing agents induce FKBP12.6 dissociation when the protein is already bound to the RyR2.
To test this hypothesis, we carried out co-IP assays after allowing FKBP12.6 to bind first, followed by treatment with redox reagents. Solubilized cardiac heavy SR was incubated with [ 35 S]FKBP12.6, and then the RyR2 was immunoprecipitated specifically with Ab 1093 . Redox reagents (2 mM DTT, 1 mM H 2 O 2 , 200 M diamide) were then applied for 30 min or for 6 h. Protein immunocomplexes were isolated, and the presence of co-precipitated FKBP12.6 was analyzed by SDS-PAGE and autoradiography. Representative autoradiograms are shown in Fig. 4, and cumulative data are given in Table 3. We found that diamide results in loss of FKBP12.6 in a time-dependent manner (ϳ65% of control after 6 h), whereas H 2 O 2 is ineffective. These results suggest that FKBP12.6 protects RyR2 sulfydryl(s) targeted by H 2 O 2 but not the one(s) reacting with diamide. DTT also induced partial FKBP12.6 dissociation (ϳ80% of control after 6 h); this finding was somewhat unexpected, since DTT was without effect in the previous experiments. One plausible explanation is that FKBP12.6 binding to RyR2 produces a significant protein conformational change, thus exposing cysteine(s) that are then endogenously modified. RyR2 is known to be endogenously S-glutathionylated (36) and S-nitrosylated (52), and sulfydryl reduction of such modified cysteine(s) may result in FKBP12.6 removal.  SR vesicle co-sedimentation experiments to determine the redox sensitivity of the RyR-FKBP interaction were carried out as described in the legend to Fig. 3, followed by quantitation of the ͓ 35 S͔FKBP12.6 band by densitometric analysis of autoradiograms and normalisation against control.  6 Association under Oxidative Conditions-K201, also known as JTV519, is a drug that has recently been proposed to stabilize RyR2 function by promoting FKBP12.6 association (53,54). We tested whether K201 can restore FKBP12.6 binding to diamide-oxidized RyR2 using co-IP assays. Solubilized cardiac heavy SR was treated with 2 mM DTT or 200 M diamide for 30 min at room temperature in the presence or absence of 10 M K201, followed by incubation with exogenous [ 35 S]FKBP12.6. RyR2 was immunoprecipitated with Ab 1093 , and the presence of co-precipitated FKBP12.6 was analyzed by SDS-PAGE and autoradiography. As shown in Fig. 5, K201 had no effect on the RyR2-FKBP12.6 interaction in control, DTT-treated, or diamide-treated SR. K201 also failed to enhance FKBP12.6 binding to diamidetreated RyR2 in the predominantly open (100 M CaCl 2 ) or closed configuration (2 mM EGTA or 2 mM MgCl 2 ) (not shown).

DISCUSSION
The RyR and its accessory protein FKBP12.6 have attracted much attention in recent years because of their involvement in the pathogenesis of heart failure. It has been proposed that heart failure is caused by a chronic hyperadrenergic state that results in "hyperphosphorylation" of RyR2 mediated by cAMPdependent protein kinase at a unique site (Ser 2809 ), which in turn promotes the dissociation of FKBP12.6 from the RyR2 channel complex (7). Although specific details of this mechanism remain highly controversial (55)(56)(57)(58)(59), the loss of the stabilizing FKBP12.6 subunit is expected to result in a constitutively increased Ca 2ϩ leak from the SR, leading to decreased SR Ca 2ϩ content and increased diastolic Ca 2ϩ release. However, additional or alternative mechanisms may account for FKBP12.6 dissociation from the RyR2 Ca 2ϩ release channel complex. In the present study, we demonstrate that sulfydryl-oxidizing agents may provide a novel form of regulation due to their ability to disrupt the RyR-FKBP interaction.
Using co-immunoprecipitation assays with detergent-solubilized RyR, we found that FKBP12.6 binding was decreased by ϳ25 or ϳ50% when RyR2 was treated with H 2 O 2 or diamide, respectively (Fig. 1, Table 1). Identical results were obtained with a cysteine-null mutant version of FKBP12.6 ( Fig. 2), indicating that the effect of the oxidizing reagents is via modification of one or more RyR2 cysteine residue(s). SR vesicle co-sedimentation assays (Fig. 3) confirmed that H 2 O 2 and diamide reduce FKBP12.6 binding, although to a lesser extent (by ϳ10 and ϳ35%, respectively) ( Table 2). Nevertheless, the two sets of data are compatible with each other and demonstrate that oxidized RyR2 has reduced binding affinity for FKBP12.6. Treatment of RyR2 with DTT did not affect FKBP12.6 binding compared with control, despite DTT reducing any sulfydryls that could have been oxidized by air exposure. However, this is not surprising, since oxidation of up to ϳ10 thiols from the physiological redox state (muscle O 2 tension 10 mm Hg and 5 mM GSH) was shown to have little effect on channel activity (23). H 2 O 2 and diamide target disparate sulfydryl(s) and/or a distinct number of sulfydryl(s), since they decreased FKBP12.6 binding to a different extent. This conclusion is supported by the observation that H 2 O 2 was almost ineffective under conditions that result in a closed RyR2 channel, whereas diamide was equally effective under open or closed conditions (Fig. 4). Differences in the action of H 2 O 2 and diamide should be attributable to differences in redox chemistry, redox potential, and/or hydrophobic nature. Indeed, it has been shown that both agents induce RyR activation mediated by formation of intersubunit disulfide bonds, which increases with the more potent oxidizing agents (e.g. diamide), and the latter notably also induces additional intrasubunit cross-linking (27,37,48). A recent report also suggested that the type of redox modification differentially affects FKBP12 binding to RyR1, since RyR1 S-nitrosylation reduced binding affinity, whereas S-glutathionylation (induced by a combination of H 2 O 2 and GSH) was without any effect (45). In the present study, H 2 O 2 and diamide are expected to induce disulfide bond formation within the RyR rather than mixed sulfydryls with glutathione, whereas the effect of NO donors was not investigated.
Interestingly, H 2 O 2 could not induce FKBP12.6 dissociation when this protein was already bound to RyR2 (Table 3). Thus, FKBP12.6 binding to the RyR2 protects the cysteine residue(s) reacting with H 2 O 2 , either directly, if such residue(s) are within the primary FKBP-binding site, or indirectly, by stabilizing a RyR2 conformation that occludes such residue(s). A similar observation has been made for calmodulin, with the bound cal-  Co-immunoprecipitation experiments to determine the redox sensitivity of the RyR-FKBP interaction were carried out as described in the legend to Fig. 4, followed by quantitation of the ͓ 35 S͔FKBP12.6 band by densitometric analysis of autoradiograms and normalization against control.  (61)). We have previously presented evidence that FKBP12.6 binds at the C terminus of the RyR2, and we suggested that a Val-Pro-Leu-Val motif (amino acids 4594 -4597 for human RyR2) within transmembrane segment M6 could constitute part of the FKBP-binding core (12). Therefore, it is plausible that the two cysteines within M6, which are the only predicted intramembranous sulfydryls present in RyR, are involved in the interaction with FKBP12.6. In contrast to H 2 O 2 , diamide did result in partial FKBP12.6 loss from RyR2 (Table 3). Oxidation-induced FKBP removal is consistent with oxidation-induced RyR activation, since interaction with this accessory protein promotes channel closure (5,6,15). This raises the possibility that FKBP dissociation from the RyR channel complex may be a causative mechanism in oxidative stress-related disease and also in heart failure. Oxidative stress, a condition where the production of reactive oxygen species overrides the scavenging effects of the antioxidant defense system and the associated intracellular Ca 2ϩ overload, has been implicated in the genesis of various cardiac diseases, including ischemia-reperfusion injury, diabetic and catecholamine-and doxorubicin-induced cardiomyopathies, and the transition from cardiac hypertrophy to heart failure (62)(63)(64). A role for FKBP12.6 has been proposed in hypoxia-induced, RyR2-mediated Ca 2ϩ signaling in vascular smooth muscle, since FKBP12.6 deficiency (gene knock-out or FK506 exposure) enhanced hypoxia-induced Ca 2ϩ release (65). Heart failure is a condition characterized by increased activity of the sympathetic nervous system and increased catecholamine levels. Elevated catecholamine levels produce reactive oxygen species due to autoxidation, leading to cardiomyopathy as well as age-related neurodegeneration (62,64,66). It is noteworthy that a recent study reported on reduced RyR2-FKBP12.6 asso-ciation in an animal model of heart failure that was corrected with the administration of an antioxidant (67).

Control
In these studies, we used redox reagent concentrations believed to mimic the oxidative conditions caused by localized, highly reactive oxygen free radicals produced in certain pathophysiological responses (68). The intracellular level of H 2 O 2 has been estimated to reach 100 M in pathological cases, whereas diamide is a nonphysiological redox reagent. Under oxidative stress conditions, H 2 O 2 is not the primary reactive oxygen species produced, since the extremely reactive superoxide anion radical ( ⅐ O 2 Ϫ ) is the most prominent and more toxic (68). The ⅐ O 2 Ϫ is converted to H 2 O 2 through the action of superoxide dismutase, and the H 2 O 2 can in turn give rise to the highly reactive hydroxyl radical ( ⅐ OH) in the presence of heavy metal ions (e.g. Fe 2ϩ ). In addition, an SR-associated NAD(P)H oxidase coupled to ⅐ O 2 Ϫ production has been shown to regulate RyR channel activity (34 -36, 69). It will be of interest to assess the RyR-FKBP association following stimulation of the NAD(P)H oxidase; however, these experiments are beyond the scope of the present work.
In conclusion, we have presented direct evidence that the fundamental RyR-FKBP regulatory protein interaction is redox-sensitive. Our results suggest that oxidative conditions may result in FKBP12.6 dissociation from the RyR2 channel complex, leading to altered Ca 2ϩ release, and this phenomenon may contribute to aberrant Ca 2ϩ signaling-mediated cardiac disease.