Low N-Ethylmaleimide Concentrations Activate Ryanodine Receptors by a Reversible Interaction, Not an Alkylation of Critical Thiols*

Previous studies proposed thatN-ethylmaleimide (NEM) alkylates 3 classes of thiols on skeletal muscle ryanodine receptors (RyRs) producing 3 phases of channel modification, as function of time and concentration. NEM (5 mm) decreased, increased, and then decreased the open probability (P o) of the channel by thiol alkylation, a reaction not reversed by reducing agents. We now show that low NEM concentrations (20–200 μm) elicit Ca2+ release from sarcoplasmic reticulum (SR) vesicles, but contrary to expectations, the effect was fully reversed by reducing agents or by washing SR vesicles. In bilayers, NEM (0.2 mm) increased P o of RyRs within seconds when added to the cis (not trans) side, and dithiothreitol (DTT; 1 mm) decreased P o in seconds. High (5 mm) NEM concentrations elicited SR Ca2+ release that was not reversed by DTT, as expected for an alkylation reaction. A non-sulfhydryl reagent structurally related to NEM, N-ethylsuccinimide (0.1–0.5 mm), also elicited SR Ca2+ release that was not reversed by DTT (1 mm). Other alkylating agents elicited SR Ca2+ release, which was fully (N-methylmaleimide) or partially (iodoacetic acid) reversed by DTT and inhibited by ruthenium red. Nitric oxide (NO) donors at concentrations that did not activate RyRs inhibited NEM-induced Ca2+ release, most likely by an interaction of NO with NEM rather than an inactivation of RyRs by NO. Thus, at low concentrations, NEM does not act as a selective thiol reagent and activates RyRs without alkylating critical thiols indicating that the multiple phases of ryanodine binding are unrelated to RyR activity or to NEM alkylation of RyRs.


Previous studies proposed that N-ethylmaleimide (NEM) alkylates 3 classes of thiols on skeletal muscle ryanodine receptors (RyRs) producing 3 phases of channel modification, as function of time and concentration. NEM (5 mM) decreased, increased, and then decreased the open probability (P o ) of the channel by thiol alkylation, a reaction not reversed by reducing agents. We now show that low NEM concentrations (20 -200 M) elicit
Ca 2؉ release from sarcoplasmic reticulum (SR) vesicles, but contrary to expectations, the effect was fully reversed by reducing agents or by washing SR vesicles. In bilayers, NEM (0.2 mM) increased P o of RyRs within seconds when added to the cis (not trans) side, and dithiothreitol (DTT; 1 mM) decreased P o in seconds. High (5 mM) NEM concentrations elicited SR Ca 2؉ release that was not reversed by DTT, as expected for an alkylation reaction. A non-sulfhydryl reagent structurally related to NEM, N-ethylsuccinimide (0.1-0.5 mM), also elicited SR Ca 2؉ release that was not reversed by DTT (1 mM). Other alkylating agents elicited SR Ca 2؉ release, which was fully (N-methylmaleimide) or partially (iodoacetic acid) reversed by DTT and inhibited by ruthenium red. Nitric oxide (NO) donors at concentrations that did not activate RyRs inhibited NEM-induced Ca 2؉ release, most likely by an interaction of NO with NEM rather than an inactivation of RyRs by NO. Thus, at low concentrations, NEM does not act as a selective thiol reagent and activates RyRs without alkylating critical thiols indicating that the multiple phases of ryanodine binding are unrelated to RyR activity or to NEM alkylation of RyRs.
In striated muscle, the sarcoplasmic reticulum (SR) 1 is the major storage compartment of intracellular Ca 2ϩ that controls cytosolic free Ca 2ϩ and developed force by sequestering and releasing Ca 2ϩ during each contraction. Studies of Ca 2ϩ flux across the SR membrane and single channel recordings in planar lipid bilayers have suggested that Ca 2ϩ release from skeletal muscle SR is mediated by a high conductance Ca 2ϩ channel (ryanodine receptor; RyR) (1,2). A wide range of intracellular signaling molecules and pharmacological agents modulated RyRs (3). In addition the following oxidants have been shown to activate the channel: heavy metals, reactive disulfides, hydrogen peroxide, Fe 2ϩ /ascorbate, and Thimerosal (4 -10). Sulfhydryl-reducing agents have been shown to reverse the effect of some oxidants to close the channel. Such studies demonstrated that skeletal and cardiac RyRs contain hyperreactive or critical thiols of low pK a values that could be oxidized and reduced to reversibly open and close the Ca 2ϩ -release channel (7,9). Heavy metals and sulfhydryl oxidants, much like Ca 2ϩ -induced Ca 2ϩ release, exhibit biphasic effects on RyRs, where low concentrations activate and high concentrations inhibit the open probability of the channel (10 -12).
Stoyanovsky et al. (13) first showed that NO donors oxidize free thiols on RyRs causing an increase in the open probability of the channel (P o ) and eliciting Ca 2ϩ release from skeletal and cardiac SR vesicles. NO gas and NO donors interact with RyRs to promote channel opening by a nitrosylation of critical sulfhydryl sites involved in the modulation of channel gating, and the effect is reversed by sulfhydryl-reducing agents (e.g. reduced glutathione, GSH; dithiothreitol, DTT; and L-cysteine) (10,13). NO gas elicited SR Ca 2ϩ but required high concentrations both in ambient oxygen levels (200 -400 M) and deoxygenated solutions (60 -200 M). Because SR Ca 2ϩ transport is not altered by oxygen levels this indicated that direct nitrosylation of RyRs by NO gas is not biologically significant (13). Unlike authentic NO and non-thiol NO donors, nitrosothiols such as S-nitrosocysteine (cys-SNO) and S-nitroso-N-acetyl-D-L-penicillamine (SNAP) were potent at activating RyRs even in the presence of ambient oxygen levels because of a robust transnitrosylation of NO from the donor to critical thiols on RyRs (13). As with other sulfhydryl oxidants, the binding of radiolabeled [ 3 H]ryanodine to its receptor was inhibited by the NO donor cys-SNO, whereas the activity of Ca 2ϩ -ATPases was not altered by NO (13). In contrast, Mészà ros et al. (14) reported that SNAP inhibited SR Ca 2ϩ release by partially reversing caffeine-induced Ca 2ϩ release and decreased P o of car-diac RyRs in planar bilayers (15). Subsequent studies confirmed that sulfhydryl NO donors activate cardiac RyRs (16) and showed that NO-like sulfhydryl oxidants and Ca 2ϩinduced Ca 2ϩ release elicits a biphasic response, first an activation at low NO concentrations and then an inhibition at higher concentrations (17,18).
The divergent findings of the effects of NO on Ca 2ϩ -release channel could arise from differences in the reaction conditions and/or NO concentration. Aghdasi et al. (19) proposed that NO acts at different classes of thiols depending on the NO concentration. Low NO concentrations oxidize thiols that inhibit channel activity, and high NO levels oxidize thiols that activate RyRs (19). Their conclusions were based on the multiple classes of sulfhydryls with which the sulfhydryl-alkylating agent Nethylmaleimide (NEM) could interact (19). In RyRs reconstituted in planar bilayers, NEM (5 mM) produced 3 phases of channel modification as a function of time. First, there was a decrease in P o after 2 min (phase 1), then there was an increase after 10 min (phase 2), and finally there was a decrease after 12 min (phase 3) (19). Low NEM (0.2 mM) concentrations decreased P o of the channel, and a subsequent addition of NEM (5 mM) increased P o (19). The three phases of channel activity elicited by NEM (5 mM) seemed to represent three phases of alkylation to three classes of sulfhydryl sites on RyRs, with decreasing affinity for NEM. This interpretation was supported by parallel changes of [ 3 H]ryanodine binding (i.e. a decrease, increase, and then decrease in B max ). In addition, the sulfhydryl cross-linking reagent, diamide, was shown to activate RyRs in bilayers and to cross-link RyR monomers to form dimers. RyRs activated by diamide could be activated further by NEM (5 mM), and alkylation with NEM (0.5 mM) blocked the cross-linking of monomers by diamide. This implied that alkylation of phase 1 sulfhydryls by NEM prevents the formation of dimers elicited by diamide (19). Low NEM concentrations (Յ5 mM) alkylate high affinity sulfhydryl sites (phase 1) reducing RyR activity (P o ) and do not to proceed to phase 2 and 3 alkylation according to ryanodine binding and changes in P o . Most intriguing was that NO donors (0.1 to 0.5 mM) like SNAP, NOC-9, and NOC-15 blocked phase 1 alkylation of RyRs, measured as an increase in ryanodine binding during phase 1 of NEM alkylation (20). NO donors also blocked intersubunit cross-linking induced by diamide. In contrast, high concentrations of NOC-9 and NOC-15 (5 mM) activated RyRs by increasing P o and induced intersubunit cross-linking (20). Aghdasi et al. (20) concluded that NO interacts with at least two classes of sulfhydryls on RyRs sites, a high affinity site where NO protects against oxidants (site of phase 1 alkylation) and a low affinity site where NO acts as an oxidant to directly activate RyRs and promote intersubunit cross-linking.
These results are opposite to the biphasic effects of NO, Ca 2ϩ , and other sulfhydryl reagents where low concentrations first activate and higher concentrations inhibit channel activity (10,18). The dual role of NO as an antagonist to other oxidants and as an oxidant of free thiols was previously demonstrated with SR, but the protective effects against oxidative stress were attributed to direct interaction between NO and the oxidant (21,22).
The present study re-examines the triphasic effect of NEM on channel activity for the following three reasons: 1) The selectivity of NEM as a thiol-alkylating reagent cannot be relied upon.
2) The level of ryanodine binding is not equivalent to a measurement of channel activity, particularly for activation of the channel by sulfhydryl oxidants and nitrosylation of the channel. C) Evidence that NEM alkylates free sulfhydryl on RyRs was lacking, leaving the possibility that NEM interacted with RyRs by other mechanisms that could be reversed by sulfhydryl-reducing agents, particularly at low NEM concentrations.

MATERIALS AND METHODS
Preparation of SR Vesicles-Skeletal SR vesicles were isolated from rabbit white muscle from the hind leg, as described by Salama and Scarpa (23). Vesicles were suspended in sucrose (0.29 M) medium with histidine (20 mM) buffer at pH 7.0 and kept in liquid nitrogen until use. Protein concentrations were determined with the Bio-Rad protein assay kit with bovine serum albumin as a standard. Heavy SR vesicles were prepared to reconstitute RyRs by fusing heavy SR vesicles with planar lipid bilayers for single channel recordings (24). Briefly, SR vesicles were resuspended in a medium containing 0.6 M KCl, 100 M EGTA, 75 M CaCl 2 , 10 mM K-Pipes, pH 7.4, and were then loaded on top of a 15-ml tube in which a sucrose gradient was layered (5 ml of 20%, 5 ml of 35%, and 8 ml of 40%). The tube was centrifuged in a swinging bucket rotor (SW 28 Beckman rotor) at 26,000 rpm for 16 h. Heavy SR vesicles were collected from the 35-40% sucrose interface, and 2 volumes of 0.4 M KCl solution was added to the heavy SR fraction and centrifuged at 18,500 rpm for 90 min. The heavy SR pellet was resuspended in a medium containing 0.29 M sucrose, 5 mM K-Pipes, pH 7.4, and stored in liquid nitrogen until use.
Measurement of SR Ca 2ϩ Transport-Ca 2ϩ uptake and efflux from SR vesicles was measured spectrophotometrically through the differential absorption changes of antipyrylazo III (AP III) at 720 -790 nm (23). Measurements were performed in a temperature-controlled cuvette under continuous stirring. Ca 2ϩ uptake by the SR vesicles was measured in a reaction medium containing the following (in mM): 100 KCl, 0.2 AP III, 1 MgCl 2 , 0.5 ATP, 4 phosphocreatine, 20 HEPES, and 2.5 units/ml creatine kinase, pH 7.0, at 37°C. Creatine kinase and phosphocreatine provided an ATP-regenerating system to maintain a constant concentration of ATP and free Mg 2ϩ during Ca 2ϩ transport. Calibration of AP III responses was obtained for each experiment by adding 12 M Ca 2ϩ to the reaction medium, in the presence or absence of SR vesicles.
Single Channel Recordings-Ca 2ϩ -release channels were reconstituted through the fusion of heavy SR vesicles. SR 0 (.5-1 g protein/ml) was added to the cis side under an osmotic gradient to drive vesicle fusion with a planar lipid bilayer composed of phosphatidylethanolamine:phosphatidylserine:phosphatidylcholine at a ratio of 5:3:2, as described previously (9,13,24). The lipids were dissolved in decane at a total concentration of 50 mg/ml, and planar bilayers were formed across a 200-m diameter hole in a Kynar cup. Single channel activities were recorded in the following asymmetrical solutions: cis side, 250 mM cesium gluconate; trans side, 50 mM cesium gluconate. The cis and trans solutions were kept at pH 7.4 with 10 mM Tris-HEPES and pCa of 6.5 buffered with 1 mM EGTA. Channel activity was recorded at holding potentials varying from Ϯ 40 mV, and P o values were calculated from a minimum of 3 min of continuous recordings under each condition. An Axopatch 1D (Axon Instruments, Foster City, CA) amplifier with a CV-3B head stage was used to measure picoampere current fluctuations. Data were digitized (model VR-10; Instrutech Corp.) and stored on video tape for subsequent analysis of channel activity. Analog data output from the video recorder was digitized with an analog to digital converter (Labmaster TM-40; Scientific Solutions, Solon, OH), transferred to computer, and analyzed using pClamp software (Axon Instruments).
Chemicals-cys-SNO was prepared by mixing equal volumes of Lcysteine (0.6 M) and NaNO 2 (4.0 M) as described previously (13). Excess NaNO 2 was used to ensure that all the cysteine was converted to cys-SNO, because residual cysteine would act as a sulfhydryl-reducing agent to reverse the oxidation of RyRs by cys-SNO. All NO solutions were freshly prepared before an experiment. antipyrylazo III was purchased from ICN Biochemicals (Cleveland, OH). Other chemicals were obtained from Sigma.

RESULTS
Effects of NEM on Skeletal SR Vesicles-NEM is an alkylating agent that has been extensively used to label or block thiol residues on proteins with the presumption that the interaction is specific for thiols and is irreversible. As stated in a review article by Brocklehurst (25), a general criticism of such alkylating reagents (like NEM) is that they often require long reaction times, and the presence of excess reagent and their specificity for thiols cannot be relied upon. The specificity problem appears to be particularly serious in the case of maleimide derivatives that react with other nucleophiles such as amino and imidazole groups. Fig. 1A shows the effect of different concentrations of NEM on isolated skeletal SR vesicles. Once the SR vesicles were loaded with Ca 2ϩ , NEM (20 -200 M) was added to the reaction medium, which elicited Ca 2ϩ release. DTT was then added to test whether NEM was alkylating thiols on the SR. DTT reversed the effects of NEM (20 -200 M) resulting in Ca 2ϩ reuptake via the SR Ca 2ϩ pumps. An addition of the Ca 2ϩ ionophore, A23187, released the Ca 2ϩ stored in the vesicles indicating that in DTT, the vesicles reaccumulated the Ca 2ϩ released (Fig. 1A). On the other hand, DTT did not reverse NEM-induced Ca 2ϩ release when high concentrations of NEM (1-5 mM) were used, consistent with an alkylation of RyRs, which elicited a fast release of nearly 100% of the stored Ca 2ϩ . At an intermediate concentration of NEM (0.5 mM), DTT partially reversed NEM-induced Ca 2ϩ release, indicating that alkylation requires a higher concentration of the substrate compared with the non-covalent interaction between NEM and RyRs.
Reversibility of NEM-induced Ca 2ϩ Release-To determine whether the effect of NEM on SR Ca 2ϩ release can be reversed by washing out the vesicles, skeletal SR vesicles were incubated in low concentration of NEM for 5 min and then washed of NEM by rapid centrifugation and resuspended in NEM-free reaction medium. SR vesicles pretreated with NEM were then tested for their capacity to actively load Ca 2ϩ compared with control vesicles, which were treated in an identical manner but were resuspended with NEM in the reaction medium. Fig. 1B (dotted line, trace 1) illustrates that SR vesicles resuspended with NEM in the medium sequester Ca 2ϩ upon the addition of ATP during an early phase of rapid Ca 2ϩ transport and then release Ca 2ϩ because of NEM-induced Ca 2ϩ release. The rapid initial phase of Ca 2ϩ uptake indicated that NEM acts primarily by activating Ca 2ϩ -release channels rather than inhibiting Ca 2ϩ uptake by Ca 2ϩ -ATPase pumps. The addition of DTT reverses the effect of NEM resulting in complete reuptake of Ca 2ϩ by the vesicles (trace 1). In contrast, SR vesicles that were exposed to NEM and then resuspended in NEM-free medium sequestered the same amount of Ca 2ϩ as control vesicles, which were never exposed to NEM, and DTT had no further effect on enhancing SR Ca 2ϩ uptake (solid line, trace 2). Hence, the pretreatment of vesicles with NEM (100 M) was fully reversed by washing the vesicles. Fig. 2A shows that SR Ca 2ϩ release induced by low NEM concentrations was reversed by all of the sulfhydryl-reducing agents that were tested. The order of relative potency was DTT Ͼ GSH Ͼ L-cysteine Ͼ mercaptoethanol. Ruthenium red was effective but did not fully inhibit NEM-induced Ca 2ϩ release, supporting the view that release occurred mostly through an activation of RyRs ( Fig. 2A).
Effects of NO on NEM-induced Ca 2ϩ Release-The inhibition of NEM-induced Ca 2ϩ release by NO was examined by testing the effect of several NO donors. We tested several NO donors at FIG. 1. NEM-induced Ca 2؉ release from skeletal SR vesicles is reversible. Ca 2ϩ transport across SR vesicles was determined by measuring extravesicular free [Ca 2ϩ ] through the differential absorption changes of AP III at 720 -790 nm. After the SR vesicles (0.2 mg/ml) were actively loaded with Ca 2ϩ using an ATP-regenerating system, NEM was added to the reaction mixture to elicit Ca 2ϩ release. After release was completed, DTT was added to reverse the effects of the NEM. A23187 was added to measure the total intravesicular Ca 2ϩ . A, NEM (20 -5000 M) elicited Ca 2ϩ release from the SR vesicles, and DTT (1 mM) induced the reuptake of Ca 2ϩ by SR, except for 5 mM NEM. B, SR vesicles were pretreated with 100 M NEM at 4°C for 5 min. The suspension of SR vesicles was diluted to 0.5 ml, centrifuged for 20 min at 45,000 ϫ g, and resuspended in 1 ml of medium with (dotted line, trace 1) or without (solid line, trace 2) NEM (0.1 mM). SR vesicles were actively loaded with Ca 2ϩ and then DTT (1 mM) was added to reverse the effects of the NEM. SR with NEM exhibited a normal initial phase of Ca 2ϩ uptake followed by a rapid release of Ca 2ϩ (dotted line, trace 1). The phase of release was due to the presence of NEM, because DTT neutralized NEM resulting in rapid Ca 2ϩ uptake. SR vesicles resuspended in NEM-free medium accumulated and retained Ca 2ϩ , and DTT had no effect (solid line, trace 2).

FIG. 2. Sulfhydryl reductants, ruthenium red, and NO inhibit NEM-induced Ca 2؉ release.
Ca 2ϩ transport across SR vesicles was measured with AP III at 720 -790 nm. SR vesicles (0.2 mg/ml) were actively loaded with Ca 2ϩ using an ATP-regenerating system and NEM (50 M)-elicited Ca 2ϩ release. A, after release was completed, various sulfhydryl-reducing agents were added to reverse the effects of the NEM and then A23187 was added to measure the total intravesicular Ca 2ϩ . Traces were as follows: 1) 1 mM DTT; 2) 2 mM L-cysteine; 3) 1 mM GSH; 4) 1 mM mercaptoethanol. A prior addition of ruthenium red (5 M) inhibited NEM-induced Ca 2ϩ release (dotted line). B, NO donors inhibit Ca 2ϩ release induced by low NEM concentrations. SR vesicles were actively loaded with Ca 2ϩ using an ATP-regenerating system and then exposed to various NO donors for 5 min followed by NEM (25 M) to test for the inhibition of NEM-induced Ca 2ϩ release by NO donors. a wide range of concentrations and found that the inhibition of NEM-induced Ca 2ϩ release by NO donors was dependent on the concentrations of the NO donor relative to that of NEM. The inhibition by NO was best studied at the lowest possible NEM (25 M) concentration that caused release such that equally low concentrations of NO donors could be tested. This is important, because the higher concentrations of NO donor themselves elicit SR Ca 2ϩ release. Fig. 2B illustrates that NEM (25 M) induces Ca 2ϩ release from SR vesicles (control, trace 4) and was inhibited by the prior addition of NO donors. The order of potency of NO donors as inhibitors of NEM was SNAP (trace 1) Ͼ NOC-15 (PAPA-NONOate) (trace 2) Ͼ cys-SNO (trace 3). At the concentration used here, these NO donors had no detectable effect on SR Ca 2ϩ transport.
Effect of Low NEM Concentrations on Single Channel Properties of RyRs-The effect of low NEM concentrations on single channel fluctuations of skeletal muscle RyRs was investigated by reconstituting Ca 2ϩ -release channels through the fusion of heavy SR vesicles with planar bilayers. Single channel fluctuations were measured in asymmetrical solutions across the bilayer with cesium as the conducting ion (Fig. 3A). As illustrated in Fig. 3B, NEM (200 M) added to the cis side (but not trans) produced a marked increase in P o within 30 s. There was no detectable effect on single channel conductance at potentials ranging from Ϯ 40 mV. In four separate channels, mean P o increased from 0.28 Ϯ 0.05 to 0.75 Ϯ 0.1 in the presence of NEM. Continuous recordings showed that the presence of NEM (200 M) on the cis side for up to 20 min did not cause further changes in P o (n ϭ 3), and an initial inhibition of the channel was not observed. Lower concentrations (10, 25, and 50 M) of NEM had no effect on single channel properties for ϳ15 min, and 100 M NEM also increased P o but with a slower onset of activation (ϳ2 min). In three bilayers, the effect of NEM was tested on the trans side before testing it on the cis side. Single channel activation by NEM (200 M) was effectively reversed by 1 mM DTT, resulting in a decrease in P o to 0.05 in ϳ30 s (Fig.  3C).
Effects of Other Alkylating Agents and the Non-thiol Analog of NEM-N-methylmaleimide (NMM) had similar effects as NEM on SR Ca 2ϩ release. As shown in Fig. 4A, NMM at 50 -200 M caused SR Ca 2ϩ release, which was fully reversed by DTT. Higher NMM concentrations (1-5 mM) elicited release that was not reversed by DTT. Other alkylating agents like iodoacetic acid (IAA) (Fig. 4B) and iodoacetimide (not shown) also elicited SR Ca 2ϩ release but with slower kinetics, and they were partially reversed by DTT. As for NEM, ruthenium red inhibited release by these alkylating reagents but did not fully block release (Fig. 4, A and B) consistent with a possible inhibition of SR Ca 2ϩ -ATPase (26,27). Interestingly, control experiments with a non-thiol-reactive compound, which is structurally related to NEM, N-ethyl succinimide (NES) also elicited SR Ca 2ϩ release at 100 -500 M, and as shown for IAA, the rate of release was slower compared with NEM and NMM (Fig. 4C). NES-induced Ca 2ϩ release was not reversed by DTT as expected for a non-sulfhydryl reagent and was effectively blocked by ruthenium red, consistent with an activation of RyRs. DISCUSSION Several studies have shown that sulfhydryl oxidation-reduction plays an important role in the pathophysiology of skeletal and cardiac muscle (10). It has been demonstrated that skeletal and cardiac RyRs contain critical thiols of low pK a that can be oxidized and reduced to reversibly open and close the Ca 2ϩrelease channel. Free thiols on RyR can also be oxidized by NO donors resulting in an increase in the open probability of the channel and Ca 2ϩ release from skeletal and cardiac SR vesicles (13, 16 -18). Channel activation most likely occurs through the transnitrosylation of critical sulfhydryl sites on RyRs resulting in the formation of stable S-nitrosothiol residues or transient S-nitrosothiol bonds followed by disulfide bond formation with vicinal thiols on RyR. The complex effects of NO and other sulfhydryl oxidants highlight the multiple interactions that can occur and the need to identify the critical cysteine residues on RyRs to better understand the role of NO in the regulation of force in striated muscle (10).
Low NEM Concentrations Do Not Act at Sulfhydryl Sites on Skeletal RyRs-The main findings of this study are that low and high NEM concentrations act by different mechanisms. Low concentrations of NEM (20 -200 M) activate RyRs but do not alkylate the protein, because the effect is fully reversed by washing out NEM or by adding sulfhydryl-reducing agents. High concentrations of NEM (5 mM) also activate the channel but are not reversible with sulfhydryl-reducing agents. It is important to note that low NEM concentrations activate RyRs and may partially inhibit Ca 2ϩ , Mg 2ϩ -ATPases, because NEM effects were not fully reversed by ruthenium red. The Ca 2ϩ -ATPase enzyme is known to be inhibited by sulfhydryl reagents and up to three sulfhydryl groups out of the 24 cysteines of the enzyme are considered essential for full enzymatic activity (26 -29). High concentrations of alkylating reagents are typically required and used to inhibit Ca 2ϩ -ATPase activity. Low concentrations of alkylating agents are ineffective, which is consistent with the fast initial phase of Ca 2ϩ uptake seen after a brief exposure of the SR vesicles to NEM (Fig. 1B, trace 1). On the other hand, prolonged exposure of SR vesicles with low NEM concentrations may partially inhibit Ca 2ϩ -ATPase activity, as shown in Fig. 1B.
The reversal of NEM-induced Ca 2ϩ release by DTT indicates that the exogenously added DTT neutralizes the effect of NEM by the alkylation of DTT by NEM and that NEM did not significantly inhibit Ca 2ϩ -ATPase activity, because the vesicles take up the release Ca 2ϩ . These findings indicate that at low concentrations, NEM activates RyRs by a non-covalent interaction and at high concentrations by alkylating free thiols on RyRs. The NO donors SNAP, NOC-15 (PAPA-NONOate), and cys-SNO were tested at concentrations that had no effect on channel activity and were found to inhibit NEM-induced Ca 2ϩ release from SR vesicles. NO donors most likely inhibit the effect of low NEM by interacting with NEM, because the inhibition is only seen at low NEM concentrations compared with NO donor concentrations and at NEM levels that do not alkylate the RyR. Thus, inhibition of NEM-induced Ca 2ϩ release by NO cannot be explained by a poly-s-nitrosylation of essential cysteine residues on RyRs, because the action of NEM does not involve a sulfhydryl-mediated reaction. These findings are supported by equivalent measurements with other alkylating agents and with a non-thiol-reactive compound, structurally related to NEM. In all cases, these reagents elicited SR Ca 2ϩ release, and the effects of the thiol reagents were reversed by DTT, and as expected, the non-thiol reagent was not reversed by DTT. The finding that NES, a non-thiol reagent structurally related to NEM, elicits SR Ca 2ϩ release demonstrates that such release occurs without the possibility of a sulfhydryl interaction with RyRs. The block of NES-induced Ca 2ϩ release by ruthenium red (Fig. 4C) confirms that NES acts at RyRs.
Aghdasi et al. (19) reported three distinct time-dependent phases of channel modification primarily from measurements of ryanodine binding with the tacit assumption that ryanodine binding is a direct measurement of single channel activity. The greater the binding, the more active the channel. In general, it is known that agents that increase and decrease channel activity typically increase and decrease ryanodine binding (B max ), respectively. However, that is not always the case, because sulfhydryl reagents that activate RyRs have the opposite effect on ryanodine binding. The heavy metal Ag ϩ is a robust activator of RyRs (4) and dissociates ryanodine from its high affinity binding site in seconds instead of the normally very slow rate of dissociation (Ͼ45 min) (30). Similarly, reactive disulfide compounds and NO activate RyRs and inhibit ryanodine binding (7,13). Aghdasi et al. (19) found that low NEM concentrations alkylate thiols involved in phase 1 channel inhibition without progressing to phases 2 and 3. This was shown as an inhibition of RyR activity with 0.2 mM NEM in bilayers and a decrease in ryanodine binding with 0.5 and 1.0 mM NEM (19). However, no evidence was given to indicate that low NEM concentrations alkylate thiols or that 0.5 or 1 mM NEM inhibited activity in single channel recordings before progressing to phases 2 and 3 of channel modification.
In contrast, we found that low concentrations of NEM (0.2 mM) activated RyRs in bilayer experiments, and we found no signs of channel inhibition in either bilayer or SR vesicle measurements. In bilayer experiments, the activation of the channel by low concentrations of NEM was reversed by DTT (Fig. 4C) consistent with the reversal of NEM-induced Ca 2ϩ release in vesicle experiments by washing out NEM or adding DTT. The most likely explanation is that DTT interacts directly with NEM to neutralize NEM and NEM-induced Ca 2ϩ release. At high NEM concentrations (5 mM), Ca 2ϩ release from SR was not reversed by DTT as expected for an alkylation of thiol groups.
Several studies have shown that NO donors activated (13,16,17) or inhibited (14,15) skeletal and cardiac RyR. Aghdasi et al. (19) proposed that low levels of NO inactivate and high levels activate the receptor. Their conclusions were based on the findings that NEM could interact with multiple classes of sulfhydryls resulting in inactivation or activation of RyR as a function of time and NEM concentration (19). NO and H 2 O 2 blocked the phase 1 inhibitory effect of 5 mM NEM. NO donors, at low concentrations that have no detectable effect on channel activity, blocked intersubunit cross-linking, whereas higher NO levels activate the channel (17). The authors suggested that the two effects of NO resulted from interaction with distinct sulfhydryls. Our data show that NO donors can partially reverse the effects of low concentration of NEM, but the reversal of this reaction can also be accomplished by washing the SR vesicles or by adding a sulfhydryl-reducing agent. When SR vesicles were preincubated with NO donors at subactivating concentrations of NO donors, the subsequent NEM-induced Ca 2ϩ release was inhibited but not fully blocked. In this case, NO appears to interact directly with NEM, decreasing its effective concentration rather than causing an indirect inhibition of NEM-induced Ca 2ϩ release through an inactivation of RyRs.
The present findings indicate that 5 mM NEM alkylates thiols on RyRs, but low concentrations (20 -200 M) act by a reversible non-covalent interaction indicating that phase 1 effects measured as a decrease in ryanodine binding do not correspond to the inhibition of the channel with low NEM concentrations. The latter interpretation of the data rests heavily on ryanodine binding measurements and the questionable assumption that ryanodine binding is a measure of channel activity under all conditions.