receptors productive association with cardiac ryanodine ... - CiteSeerX

Role of calmodulin methionine residues in mediating productive association with cardiac ryanodine receptors Edward M. Balog, Laura E. Norton, David D. Thomas and Bradley R. Fruen AJP - Heart 290:794-799, 2006. First published Sep 30, 2005; doi:10.1152/ajpheart.00706.2005 You might find this additional information useful... This article cites 33 articles, 12 of which you can access free at: http://ajpheart.physiology.org/cgi/content/full/290/2/H794#BIBL Updated information and services including high-resolution figures, can be found at: http://ajpheart.physiology.org/cgi/content/full/290/2/H794 Additional material and information about AJP - Heart and Circulatory Physiology can be found at: http://www.the-aps.org/publications/ajpheart

AJP - Heart and Circulatory Physiology publishes original investigations on the physiology of the heart, blood vessels, and lymphatics, including experimental and theoretical studies of cardiovascular function at all levels of organization ranging from the intact animal to the cellular, subcellular, and molecular levels. It is published 12 times a year (monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 2005 by the American Physiological Society. ISSN: 0363-6135, ESSN: 1522-1539. Visit our website at http://www.the-aps.org/.

Downloaded from ajpheart.physiology.org on May 3, 2006

This information is current as of May 3, 2006 .

Am J Physiol Heart Circ Physiol 290: H794 –H799, 2006. First published September 30, 2005; doi:10.1152/ajpheart.00706.2005.

Role of calmodulin methionine residues in mediating productive association with cardiac ryanodine receptors Edward M. Balog, Laura E. Norton, David D. Thomas, and Bradley R. Fruen Department of Biochemistry, Molecular Biology and Biophysics, University of Minnesota, Minneapolis, Minnesota Submitted 28 June 2005; accepted in final form 23 September 2005

is triggered by Ca2⫹ influx through voltage-gated Ca channels, which in turn initiates a much larger Ca2⫹ release from the sarcoplasmic reticulum (SR) via ryanodine receptor Ca2⫹ release channels (RyR2). The ⬃2,000kDa RyR2 homotetramer consists of a COOH-terminal transmembrane assembly and a massive cytoplasmic NH2-terminal domain. Although RyR2 is activated by Ca2⫹, endogenous effectors and posttranslational modifications modulate Ca2⫹ activation of the channel. Indeed, the cytoplasmic domain of RyR2 acts as scaffolding to which accessory proteins, including the FK506-binding protein and calmodulin (CaM), bind (14). At micromolar Ca2⫹, CaM inhibits both RyR2 and the skeletal muscle ryanodine receptor isoform (RyR1). However, at submicromolar Ca2⫹, CaM activates RyR1 but inhibits RyR2 (3). This isoform-specific effect of CaM may be attributed in part to differential tuning of the Ca2⫹ affinity of CaM on CaM binding to the different RyR isoforms (16). Thus, at submicromolar Ca2⫹, RyR2-bound CaM may be Ca2⫹-CaM, whereas RyR1-bound CaM may be Ca2⫹-free CaM (apo-CaM). CaM is a 148-amino acid Ca2⫹-binding protein composed of NH2- and COOH-terminal globular domains connected by a flexible linker. High-affinity Ca2⫹ binding to two EF-hand Ca2⫹-binding motifs in each of the globular domains induces a

conformational change from the more compact apo-CaM structure to the more open Ca2⫹-CaM structure. This conformational change also exposes hydrophobic target-binding surfaces in each of the globular domains (9). These hydrophobic patches and the conformational flexibility of CaM allow it to bind to and regulate numerous, structurally diverse targets. Indeed, CaM interactions are quite promiscuous in that CaMbinding domains share little sequence homology and the mode of CaM interaction with targets varies greatly (30). CaM contains nine Met residues out of 148 amino acids. This is a much higher Met content than the statistical average for the occurrence of Met in other proteins (24). In mammalian CaM, four Met residues are clustered in each of the globular domains at residues 36, 51, 71, and 72 in the NH2 terminus and residues 109, 124, 144, and 145 in the COOH terminus. A ninth Met is located at position 76 in the linker region. As a result of this highly localized distribution, Met residues account for approximately half the surface area of the hydrophobic target-binding interface of Ca2⫹-CaM (24). The importance of the Met residues of CaM is indicated by their evolutionary conservation. For example, in CaMs from widely divergent organisms such as Tetrahymena and Dicytostelium, all nine Mets have been preserved (12). Two functions have been ascribed to the Met residues of CaM: stabilizing the open conformation of Ca2⫹-bound CaM (23) and providing a targetbinding interface (24, 31). Whereas the high Met content of CaM contributes to effective target binding, the Met residues in Ca2⫹-CaM are surface exposed (35) and susceptible to oxidation. Oxidation of nonpolar Met to polar Met sulfoxide (MetO) decreases the efficacy of target regulation by CaM (5, 28, 33). Site-specific replacement of CaM Met residues has been used to define the role of individual Met residues in the productive association of CaM with targets, i.e., both the binding of CaM to targets and the subsequent transduction of the functional effect of CaM to the targets. When compared with NH2-terminal mutations, site-specific replacement of COOH-terminal Met residues is more detrimental to the productive association of CaM with many (2, 8, 10, 32, 34) but not all targets (8, 19). However, even among the targets for which COOH-terminal CaM Mets are critical, the specific essential residue(s) vary (2, 10, 22). As such, the role of individual CaM Met residues in RyR2 regulation is undefined. This work used three approaches to study CaM Met function: Met oxidation, Met-to-Leu substitution, and Met-to-Gln substitution. Results presented here identify a critical Met residue within the COOH-terminal domain of CaM as key in the regulation of RyR2 channels.

Address for reprint requests and other correspondence: E. M. Balog, School of Applied Physiology, Georgia Institute of Technology, 281 Ferst Dr., Atlanta, GA 30332-0356 (e-mail: [email protected]).

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

sarcoplasmic reticulum; excitation-contraction; oxidative stress; methionine sulfoxide; ryanodine receptor Ca2⫹ release channel CARDIAC CONTRACTION 2⫹

H794

0363-6135/06 $8.00 Copyright © 2006 the American Physiological Society

http://www.ajpheart.org


Balog, Edward M., Laura E. Norton, David D. Thomas, and Bradley R. Fruen. Role of calmodulin methionine residues in mediating productive association with cardiac ryanodine receptors. Am J Physiol Heart Circ Physiol 290: H794 –H799, 2006. First published September 30, 2005; doi:10.1152/ajpheart.00706.2005.—Calmodulin (CaM) binds to the cardiac ryanodine receptor Ca2⫹ release channel (RyR2) with high affinity and may act as a regulatory channel subunit. Here we determine the role of CaM Met residues in the productive association of CaM with RyR2, as assessed via determinations of [3H]ryanodine and [35S]CaM binding to cardiac muscle sarcoplasmic reticulum (SR) vesicles. Oxidation of all nine CaM Met residues abolished the productive association of CaM with RyR2. Substitution of the COOH-terminal Mets of CaM with Leu decreased the extent of CaM inhibition of cardiac SR (CSR) vesicle [3H]ryanodine binding. In contrast, replacing the NH2-terminal Met of CaM with Leu increased the concentration of CaM required to inhibit CSR [3H]ryanodine binding but did not alter the extent of inhibition. Site-specific substitution of individual CaM Met residues with Gln demonstrated that Met124 was required for both high-affinity CaM binding to RyR2 and for maximal CaM inhibition. These results thus identify a Met residue critical for the productive association of CaM with RyR2 channels.

CAM METHIONINES AND RYR2 REGULATION METHODS

AJP-Heart Circ Physiol • VOL

is not protonated at low pH, it can be selectively oxidized under acidic conditions (7). Site-directed mutagenesis of CaM. Two types of CaM mutants were generated. To determine the role of the Met residues clustered in the NH2 and COOH termini of CaM, Leu was substituted for all of the Met residues in either the NH2 terminus (residues 36, 51, 71, and 72) along with Met76 in the linker region (NH2-terminal Met to Leu) or the COOH terminus (residues 109, 124, 144, and 145; COOHterminal Met to Leu). To examine the role of individual Met residues, nine additional CaM mutants were produced in which individual Met residues were replaced with Gln. Mutants were constructed from the wild-type rat CaM cDNA by using primer-based site-directed mutagenesis following the protocol provided by the QuikChange SiteDirected Mutagenesis Kit (Stratagene, La Jolla, CA). DNA sequence analysis confirmed the correct generation of each mutant. CaM expression and purification. CaM was expressed in Escherichia coli [BL21(DE3)pLys5], purified via phenyl-Sepharose chromatography (18) and dialyzed overnight at 4°C against 2 mM HEPES, pH 7.0. CaM concentration was determined by using the published extinction coefficient ⑀277 nm–302 nm ⫽ 3,029 M/cm (28). [35S]methionine incorporation. Wild-type rat CaM subcloned into the pET-7 vector was metabolically labeled with [35S]methionine and purified by phenyl-Sepharose chromotagraphy (15). Briefly, bacterial growth was initiated in M9 media containing ampicillin. When the optical density value (measured at 600 nm) of the bacteria reached 0.5, the pelleted bacteria were resuspended in RPMI 1640 media containing ampicillin, one-fortieth of methionine, cysteine, and glucose concentration compared with the regular RPMI 1640 medium, and isopropyl-␤-D-thiogalactopyranoside was added to a final concentration of 1 mM. A 2.5-mCi aliquot of [35S]methionine and [35S]cysteine was then added to the medium, and the bacteria were cultured for 5– 6 h at the same conditions. Analysis. The CaM concentration dependence of CSR vesicle [3H]ryanodine binding and the inhibition of [35S]CaM binding by unlabeled CaM were fit with a four-parameter Hill equation. The Ca2⫹ dependence of ryanodine binding was fit with an equation that assumes a high-affinity Ca2⫹-binding site, which when bound will activate the RyR, and a lower-affinity Ca2⫹-binding site, which when bound will inhibit channel opening (1). All curve fitting was performed with SigmaPlot 6.0 (Systat Software, Point Richmond, CA). Statistics. Data are presented as means ⫾ SE. [3H]ryanodine and [3S]CaM CSR-binding curves in the presence and absence of CaM and CaM mutants were analyzed with a one-way ANOVA with Tukey’s multiple comparison test as a post hoc test or by Student’s paired or unpaired t-tests as appropriate. Statistical analysis was performed with SigmaStat 3.0 (Systat Software). The level of significance was P ⬍ 0.05. RESULTS

To assess the role of CaM Met residues in the productive association of CaM with RyR2, the Ca2⫹ dependence of CSR vesicle [3H]ryanodine binding in media containing either no CaM, 1 ␮M native CaM, or 1 ␮M oxidized CaM was compared (Fig. 1). The CaM oxidation protocol used here (incubation in 50 mM H2O2 for 24 h) was previously shown to selectively oxidize all nine Met residues of CaM to MetO (2). In the absence of CaM, CSR vesicle [3H]ryanodine binding exhibited a biphasic Ca2⫹ dependence (EC50 ⫽ 2.9 ⫾ 0.3 ␮M; IC50 ⫽ 2,003 ⫾ 503 ␮M). Native CaM (1 ␮M) depressed CSR [3H]ryanodine binding (EC50 ⫽ 10.1 ⫾ 6.1 ␮M; IC50 ⫽ 1,073 ⫾ 591 ␮M). In contrast, the Ca2⫹ dependence of CSR [3H]ryanodine binding determined in media containing 1 ␮M oxidized CaM virtually overlapped the Ca2⫹ dependence of CSR [3H]ryanodine binding determined in the absence of CaM (EC50 ⫽ 3.8 ⫾ 0.7 ␮M; IC50 ⫽ 2,012 ⫾ 398 ␮M).

290 • FEBRUARY 2006 •

www.ajpheart.org


Materials. The experimental protocols were approved by the Institutional Animal Care and Use Committee of the University of Minnesota. Pigs were obtained from the University of Minnesota Experimental Farm. Tran[35S]Met and Cys were obtained from ICN Radiochemicals (Costa Mesa, CA). [3H]ryanodine was purchased from PerkinElmer (Boston, MA). Unlabeled ryanodine was obtained from Calbiochem (La Jolla, CA). Myosin light chain kinase-derived CaMbinding peptide was obtained from Peptide Technologies (Gaithersburg, MD). Disodium ␤,␥-methyleneadenosine 5⬘-triphosphate (AMP-PCP) was from Sigma (St. Louis, MO). Isolation of cardiac SR vesicles. Cardiac SR (CSR) vesicles were prepared from porcine ventricular tissue as previously described (15). Briefly, ventricular tissue was homogenized in 10 mM NaHCO3 and centrifuged for 5 min at 4,000 g. The supernatant was filtered through gauze, centrifuged for 20 min at 4,000 g, filtered a second time, and centrifuged 30 min at 80,000 g. Pelleted membranes were extracted in 0.6 M KCl and 20 mM Tris (pH 6.8) on ice for 45 min and then centrifuged 30 min at 120,000 g. The pellets were resuspended in 10% sucrose, centrifuged 30 min at 120,000 g, and resuspended in a minimal volume of 10% sucrose. CSR was frozen in liquid N2 and stored at ⫺70°C. Endogenous CaM was removed from the SR by incubating SR in 120 mM potassium propionate, 10 mM PIPES, pH 7.0, 100 ␮M Ca2⫹, and 1 ␮M myosin light chain kinase-derived CaM-binding peptide for 30 min at room temperature (4). The SR was then centrifuged through 15% sucrose in a Beckman 70.1 Ti rotor at 40,000 rpm for 30 min at 4°C to remove the CaM and CaM-binding protein. All isolation buffers contained 1 ␮g/ml aprotinin, 1 ␮g/ml leupeptin, 1 ␮g/ml pepstatin A, 1 mM benzamidine, and 1 mM phenylmethylsulfonylfluoride. [3H]ryanodine binding. Ryanodine binds selectively to RyRs in the open state; thus CSR vesicle [3H]ryanodine-binding measurements are sensitive indicators of RyR2 channel activity (11, 21). [3H]ryanodine binding to CSR vesicles was performed as described by Balshaw et al. (4). CSR vesicles were incubated with 7 nM [3H]ryanodine in 150 mM KCl, 20 mM K-PIPES, pH 7.0, 0.1 mg/ml bovine serum albumin, 1.0 ␮g/ml leupeptin, 1.0 ␮g/ml aprotinin, and Ca-EGTA buffer to achieve the desired free Ca2⫹ concentration (6). Estimates of maximal [3H]ryanodine-binding capacity of each CSR vesicle preparation were determined in media that in addition contained 600 mM KCl, 10 mM ATP, 10 ␮M Ca2⫹, and 50 nM [3H]ryanodine. Nonspecific binding was measured in the presence of 10 mM MgCl2 and 20 ␮M nonradioactive ryanodine. After 20 h at room temperature (20°22°C), CSR vesicles were collected on Whatman GF/B filters and washed with 8 ml of ice-cold 100 mM KCl buffer. All assays were performed in duplicate. [35S]CaM binding to cardiac muscle SR vesicles. [35S]CaM binding to CSR vesicles was performed as previously described (4). CSR vesicles were incubated in 150 mM KCl, 20 mM PIPES, pH 7.0, 5 mM GSH, 1 ␮g/ml aprotinin, 1 ␮g/ml leupeptin, and 100 nM [35S]CaM. After 2 h incubation at 24°C, 0 –5,000 nM unlabeled CaM was added and incubated for 30 min. Pellets were collected after centrifugation at 40,000 rpm for 20 min in a Beckman TLA-55 rotor at 20°C. Pellets were solubilized by overnight incubation in 10% SDS. The pellets were then diluted in 200 ␮l double-distilled H2O, and bound [35S]CaM was determined by scintillation counting. Nonspecific binding was determined by using 100-fold excess unlabeled CaM. Oxidation of CaM. CaM (60 ␮M) was incubated in 50 mM homopiperazine-N,N⬘-bis-2-(ethanesulfonic acid), pH 5.0, 0.1 M KCl, 2.0 mM MgCl2, 50 mM H2O2 at room temperature for 24 h. The reaction was stopped by overnight dialysis (mol wt cutoff 3,500) at 4°C in distilled water (1 liter, for five times) buffered with 10 mM ammonium bicarbonate (pH 7.7). Although H2O2 can potentially oxidize a number of amino acids because the thioether group of Met

H795

H796

CAM METHIONINES AND RYR2 REGULATION

Fig. 1. Ca2⫹ dependence of cardiac sarcoplasmic reticulum (CSR) vesicle [3H]ryanodine binding. [3H]ryanodine binding to CSR vesicles as a function of free Ca2⫹ was performed as described in METHODS in media containing no calmodulin (No CaM), 1 ␮M wild-type CaM, or 1 ␮M oxidized CaM. Data are means ⫾ SE from 3 experiments. Maximum binding capacity (Bmax) ⫽ 1.1 ⫾ 0.1 pmol ryanodine/mg SR protein. EC50: No CaM ⫽ 2.9 ⫾ 0.3 ␮M; native CaM ⫽ 10.1 ⫾ 6.1 ␮M; oxidized CaM ⫽ 3.8 ⫾ 0.7 ␮M. IC50: No CaM ⫽ 2,003 ⫾ 502 ␮M; native CaM ⫽ 1,073 ⫾ 591 ␮M; oxidized CaM ⫽ 2,012 ⫾ 398 ␮M. *Significantly different from No CaM (P ⬍ 0.05).

Fig. 2. CaM oxidation abolished functional interaction between CaM and ryanodine receptor Ca2⫹ release channel (RyR2). [3H]ryanodine and [35S]CaM binding to CSR vesicles was performed as described in METHODS in media containing 700 ␮M free Ca2⫹. A: native CaM inhibited [3H]ryanodine binding to CSR (IC50 ⫽ 27 ⫾ 1.8 nM), but oxidized CaM has no effect on CSR vesicle [3H]ryanodine binding. B: unlabeled native CaM completely displaced (IC50 ⫽ 386 ⫾ 68 nM) prebound native [35S]CaM (100 nM). Unlabeled oxidized CaM, up to 10 ␮M, did not displace prebound native [35S]CaM. Data are means ⫾ SE of 3 experiments and are expressed as B/Bo, where B is [3H]ryanodine or [35S]CaM bound in presence of unlabeled CaM and Bo is [3H]ryanodine or [35S]CaM bound in absence of unlabeled CaM. AJP-Heart Circ Physiol • VOL

Fig. 3. Inhibition of CSR vesicle [3H]ryanodine binding by wild-type CaM and NH2- and COOH-terminal Met-to-Leu CaM mutants. [3H]ryanodine binding to CSR vesicles was performed as described in METHODS in media containing 5 mM GSH and 3 ␮M Ca2⫹. A: comparison of CSR vesicle [3H]ryanodine binding in media containing either No CaM, 3 ␮M wild-type CaM, 3 ␮M NH2-terminal Met-to-Leu CaM mutant, or 3 ␮M COOH-terminal Met-to-Leu CaM mutant. *Significantly different from No CaM; P ⬍ 0.05. Bmax ⫽ 1.4 ⫾ 0.2 pmol ryanodine/mg SR protein; n ⫽ 18. B: CaM concentration dependence of CSR vesicle [3H]ryanodine binding in media containing either indicated concentration of wild-type CaM or NH2-terminal Met-to-Leu CaM mutant. Bmax ⫽ 1.1 ⫾ 0.1 pmol ryanodine/mg SR protein; n ⫽ 4. C: CaM concentration dependence of CSR vesicle [3H]ryanodine binding in media containing either indicated concentration of wild-type CaM or COOH-terminal Met-to-Leu CaM mutant. Bmax ⫽ 1.9 ⫾ 0.3 pmol ryanodine/mg SR protein; n ⫽ 4.

290 • FEBRUARY 2006 •

www.ajpheart.org


To determine the effects of CaM oxidation on the affinity of CaM for RyR2, the dependence of CSR vesicle [3H]ryanodine binding on the concentration of native and oxidized CaM was compared. Native CaM inhibited CSR vesicle [3H]ryanodine binding with an IC50 of 27 ⫾ 1.8 nM (Fig. 2A). CSR vesicle ryanodine binding in media containing up to 3 ␮M oxidized CaM did not differ from CSR vesicle ryanodine binding in the absence of CaM. Thus oxidation abolished CaM inhibition of RyR2. In competitive binding experiments (Fig. 2B), unlabeled native CaM completely displaced [35S]CaM previously bound to CSR. In contrast, unlabeled oxidized CaM (up to 10 ␮M) was unable to compete with [35S]CaM for CSR binding. Thus the loss of CaM regulation of RyR2 was the result of the

inability of oxidized CaM to interact with the CaM binding site on the channel. The functional role of the NH2- and COOH-terminal Metrich patches of CaM in the regulation of RyR2 was examined by using two Met-to-Leu mutants. In the NH2-terminal Metto-Leu mutant, the four NH2-terminal Mets (residues 36, 51, 71, and 72) and Met76 in the linker region were changed to Leu. In the COOH-terminal Met-to-Leu mutant, the four COOH-terminal Mets (residues 109, 124, 144, 145) were changed to Leu. Wild-type CaM (3 ␮M) and the NH2-terminal Met-to-Leu CaM mutant (3 ␮M ) significantly (P ⬍ 0.05) decreased CSR vesicle [3H]ryanodine binding compared with the absence of CaM (Fig. 3A). In contrast, CSR vesicle [3H]ryanodine binding in media containing 3 ␮M COOH-terminal Met-to-Leu CaM mutant was similar to CSR ryanodine binding in the absence of CaM. This suggests that the COOH terminus of CaM is critical for inhibition of RyR2. Thus the dependence of CSR vesicle [3H]ryanodine binding on the concentration Met-to-Leu mutants was compared with that of wild-type CaM (Fig. 3, B and C). Figure 3B shows that 3 ␮M NH2-terminal Met-to-Leu CaM mutant indeed inhibited CSR vesicle [3H]ryanodine binding to the same extent as wild-type CaM. However, a significantly higher concentration of the NH2-terminal Met-to-Leu mutant was required to half-inhibit CSR vesicle [3H]ryanodine binding (IC50: wild type, 9 ⫾ 1 nM; NH2terminal Met-to-Leu mutant, 111 ⫾ 40 nM; P ⬍ 0.05). In contrast, the COOH-terminal Met-to-Leu CaM mutant inhibited CSR vesicle [3H]ryanodine binding to a significantly lesser extent than wild-type CaM (percent inhibition: wild type, 57 ⫾ 3; COOH-terminal Met-to-Leu mutant, 27 ⫾ 2; P ⬍ 0.01).


H797

Fig. 5. Inhibition of CSR vesicle [35S]CaM binding by wild-type CaM and Met124Gln CaM. Binding was performed as described in METHODS in media containing 700 ␮M free Ca2⫹. Data are means ⫾ SE of 5 experiments. Wild-type: IC50 ⫽ 598 ⫾ 198; Met124Gln: IC50 ⫽ 1,617 ⫾ 441; P ⬍ 0.05.

Thus the COOH-terminal Met residues of CaM are critical for CaM inhibition of RyR2. To define the role of specific Met residues in the productive association of CaM with RyR2, individual Met residues were substituted with Gln (Fig. 4 and Table 1). Wild-type CaM depressed CRS vesicle [3H]ryanodine binding 50% with an IC50 of 37 ⫾ 3 nM. Three Met-to-Gln mutations significantly altered CaM inhibition of RyR2. In the NH2 terminus of CaM, the Met72Gln mutant lessened the extent of CaM inhibition to 68% of wild-type CaM but had a similar IC50 as wild-type CaM. Two COOH-terminal mutants altered the productive Table 1. Calmodulin concentration dependence of inhibition of CSR vesicle [3H]ryanodine binding

WT M36Q M51Q M71Q M72Q M76Q M109Q M124Q M144Q M145Q

IC50, nM

nH

%Inhibition

37⫾3 50⫾6 23⫾7 36⫾11 38⫾9 36⫾4 42⫾8 106⫾21* 47⫾5 35⫾4

1.1⫾0.9 0.8⫾0.1 1.4⫾0.5 0.9⫾0.2 2.6⫾1.2 2.4⫾0.6 1.7⫾0.5 0.9⫾0.1 2.1⫾0.6 1.7⫾0.3

50⫾2 42⫾2 44⫾6 47⫾5 34⫾5* 48⫾3 40⫾4 29⫾2* 32⫾2* 41⫾2

Values are means ⫾ SE. Half-inhibitory concentration (IC50), Hill coefficient (nH), and percent inhibition were derived from fits to Hill equation as described in METHODS and shown in Fig. 4. CSR, cardiac sarcoplasmic reticulum; WT, wild type. *Significantly different from wild-type calmodulin (P ⬍ 0.05). M, methionine; Q, glutamine. AJP-Heart Circ Physiol • VOL

Fig. 6. Dependence of CSR vesicle [3H]ryanodine binding on concentration of wild-type and Met109Gln. Binding was performed as described in METHODS in media containing 10 mM caffeine, 1 mM EGTA, 0.023 mM CaCl2 (calculated free Ca2⫹ ⫽ 10 nM), 5 mM GSH, and indicated concentration of either wild-type CaM or Met109Gln CaM. Bmax ⫽ 1.4 ⫾ 05 pmol ryanodine/mg SR protein. Data are means ⫾ SE of 7 experiments. *P ⬍ 0.05 vs. No CaM.

290 • FEBRUARY 2006 •

www.ajpheart.org


Fig. 4. Inhibition of CSR vesicle [3H]ryanodine binding by wild-type CaM and Met-to-Gln CaM mutants. [3H]ryanodine binding to CSR vesicles was performed as described in METHODS in media containing 700 ␮M free Ca2⫹. A: CaM NH2-terminal Met-to-Gln substitutions. B: CaM COOH-terminal Met-to-Gln substitutions. Here Bo is CSR vesicle [3H]ryanodine binding in absence of CaM and B is CSR vesicle [3H]ryanodine binding in presence of indicated concentration of CaM. Data are means ⫾ SE of 29 wild-type CaM experiments and 4 – 8 Met-to-Gln CaM experiments. Bmax ⫽ 1.0 ⫾ 0.1 pmol ryanodine/mg SR protein. Best fit parameters for each curve are presented in Table 1.

association of CaM with RyR2. Like Met72Gln, Met144Gln inhibited CSR vesicle [3H]ryanodine binding to a lesser extent than wild-type CaM (64% of wild type) but with a similar IC50 as wild-type CaM. Met124Gln inhibited CSR vesicle [3H]ryanodine binding to 57% of wild-type CaM and required a greater concentration to do so (IC50 ⫽ 106 ⫾ 21 nM). The Hill coefficients of SR vesicle [3H]ryanodine did not differ between wild-type CaM and any of the Met-to-Gln CaM mutants (Table 1). To determine whether the higher concentration of Met124Gln CaM required to inhibit CSR vesicle [3H]ryanodine binding was due to a lower affinity of the mutant for RyR2, displacement of prebound wild-type [35S]CaM by unlabeled wild-type and Met124Gln CaM was compared (Fig. 5). In a medium containing 700 ␮M Ca2⫹, wild-type CaM displaced CSR vesicle-bound [35S]CaM with an IC50 of 598 ⫾ 198 nM. In comparison, 2.7-fold more Met124Gln CaM was required to displace CSR-bound [35S]CaM (IC50 ⫽ 1,617 ⫾ 441 nM). Thus the Met124Gln mutation decreased the affinity of CaM for RyR2. Previous work (2) demonstrated that substitution of Met109 with Gln abolished CaM activation of RyR1 at submicromolar Ca2⫹ without affecting CaM inhibition of the channel at micromolar Ca2⫹. Thus experiments examined the effects of the Met109Gln mutation on CaM regulation of RyR2 at submicromolar Ca2⫹ (Fig. 6). To enhance cardiac SR vesicle [3H]ryanodine binding at low Ca2⫹ (1 mM EGTA, 0.023 mM CaCl2; calculated free Ca2⫹ ⫽ 10 nM), 10 mM caffeine was included in the assay media. Under these conditions, high concentrations of wild-type CaM (⬎10 ␮M) enhanced CSR vesicle [3H]ryanodine binding. In contrast, Met109Gln CaM, up to 30 ␮M, did not enhance CSR vesicle [3H]ryanodine

H798


binding. Thus the Met109Gln CaM mutation similarly impairs CaM activation of both RyR1 and RyR2 at submicromolar Ca2⫹. DISCUSSION


ACKNOWLEDGMENTS Present address of E. M. Balog: School of Applied Physiology, Georgia Institute of Technology, 281 Ferst Dr., Atlanta, GA 30332-0356. GRANTS This work was supported by American Heart Association Greater Midwest Affiliate (to E. M. Balog) and National Institutes of Health Grants K02 AR-050144 (to B. R. Fruen) and R01 AG-026160 (to D. D. Thomas). REFERENCES 1. Balog EM, Fruen BR, Shomer NH, and Louis CF. Divergent effects of the malignant hyperthermia-susceptible Arg6153 Cys mutation on the Ca2⫹ and Mg2⫹ dependence of the RyR1. Biophys J 81: 2050 –2058, 2001. 2. Balog EM, Norton LE, Bloomquist RA, Cornea RL, Black DJ, Louis CF, Thomas DD, and Fruen BR. Calmodulin oxidation and methionine to glutamine substitutions reveal methionine residues critical for functional interaction with ryanodine receptor-1. J Biol Chem 278: 15615– 15621, 2003. 3. Balshaw DM, Yamaguchi N, and Meissner G. Modulation of intracellular calcium-release channels by calmodulin. J Membr Biol 185: 1– 8, 2002. 4. Balshaw DM, Xu L, Yanaguchi N, Pasek DA, and Meissner G. Calmodulin binding and inhibition of cardiac muscle calcium release channel (ryanodine receptor). J Biol Chem 276: 20144 –20153, 2001.

290 • FEBRUARY 2006 •

www.ajpheart.org


In this study, oxidation and site-specific mutagenesis were used to identify CaM Met residues critical for inhibition of RyR2. CaM contains no cysteine residues, and under acidic conditions, oxidation can be specific for Met residues (7). The extensive in vitro oxidation protocol used here converted all nine Met residues of CaM to MetO (2). As a result, Met was changed from a nonpolar to a polar residue (17), the ␣-helical content of CaM was decreased (20, 33), the cooperativeness between the Ca2⫹ binding sites of CaM was reduced (20), and the productive association of CaM with RyR2 was abolished (Fig. 2). Because less extensive oxidation produced CaM samples containing multiple oxiforms, (2) site-directed mutagenesis was used to selectively modify specific Met residues. CaM function can be bifurcated, with the two lobes subserving different functions. A classic example is the two types of aberrant swimming behavior of Paramecium (26). One swimming behavior arises from a disregulation of a Ca2⫹-activated K⫹ current as a result of mutations in the COOH-terminal lobe of CaM. The other behavior is the consequence of an altered Ca2⫹-activated Na⫹ current due to mutations in the NH2terminal lobe of CaM. More recently, conversion of CaM from a RyR1 activator to an inhibitor has been attributed to Ca2⫹ binding to the COOH-terminal EF hands of CaM (16, 25). Thus the functional role of the NH2- and COOH-terminal Met clusters of CaM was examined by replacing all the Met residues in each cluster with Leu. In comparison with Met, Leu has a similar volume, is slightly more hydrophobic, and has a reduced flexibility but a similar propensity to form ␣-helices (17, 27). Through evolution, Leu is the most common substitute for Met (13). For instance, Sarccharomyces cercvisiae CaM has ⬃60% identity with mammalian CaM, with only three of the nine Met residues having been retained. The other six Met residues have been replaced by Leu (12). Thus Leu is a relatively conservative substitution for Met. As such, substitution of the NH2-terminal Met residues of CaM with Leu preserved CaM inhibition of RyR2. In contrast, replacement of COOH-terminal Mets significantly decreased the extent of CaM inhibition of the channel (Fig. 3). This suggests that not only is Ca2⫹ binding to the COOH terminus of CaM a determinant of RyR inhibition (25) but that the resulting exposure of COOH-terminal Mets is critical for RyR inhibition. We next endeavored to determine the relative importance of specific Met residues in the productive association of CaM with RyR2. For this we made use of CaM Met-to-Gln mutants. A Gln substitution may be considered similar to the oxidation of Met to MetO in that an oxygen is located in the same position in the amino acid side chain. Although of similar size, this substitution replaces a hydrophobic residue with a polar one. Of the nine Met-to-Gln mutations examined, replacing Met124 with Gln was the most deleterious to CaM regulation of RyR2. This substitution decreased the extent of inhibition and increased by more than twofold the CaM concentration required for half-maximal inhibition. The substitution, however, had no detectable effect on the secondary structure of CaM as assessed by circular dichroism nor did it prevent the

Ca2⫹-induced mobility shift on SDS-PAGE or alter the Ca2⫹ affinity of CaM (2). Like its effect on RyR2 regulation, the Gln substitution for Met124 was the only mutation that altered Ca2⫹-CaM inhibition of RyR1, increasing the CaM concentration required for half-maximal inhibition. The mutation also increased the apo-CaM concentration required for RyR1 activation (2). Thus Met124 is required for high-affinity binding of CaM to both RyR1 and RyR2. COOH-terminal Met124 has also been shown, along with Met109, to be critical for the high-affinity association of Ca2⫹-CaM with a number of enzyme targets, including CaM kinases II and IV, smooth muscle myosin light-chain kinase (8), and Bordetella pertussis adenylate cyclase (32). Although the Met109Gln mutation did not alter Ca2⫹-CaM inhibition of either RyR1 or RyR2, this mutation abolished apo-CaM activation of RyR1. Therefore, an attempt was made to determine whether RyR1 and RyR2 share a requirement for Met109 in CaM activation at nanomolar Ca2⫹. With Ca2⫹ as the sole RyR2 activator, CaM was inhibitory (Fig. 1). However, increasing RyR2 activity at low Ca2⫹ via the addition of caffeine revealed apo-CaM activation of this isoform (Fig. 6). This is consistent with the previous reports using caffeine (15) and Ca2⫹-insensitive CaM mutants (16). Thus the enhanced CSR vesicle [3H]ryanodine binding by wild-type CaM but the inability of the Met109Gln mutant to similarly enhance CSR vesicle [3H]ryanodine binding suggests that the two RyR isoforms do indeed share a requirement for a Met residue in CaM position 109 for CaM activation of the channels. In summary, the present study demonstrated that extensive in vitro oxidation of CaM abolished the productive association of CaM with cardiac RyR2 channels. Met-to-Leu substitutions revealed that the COOH-terminal Met residues of CaM are required for CaM inhibition of RyR2. Furthermore, replacement of Met124 in the COOH terminus with Gln lowered the affinity of CaM for RyR2 and decreased the extent of channel inhibition. Thus Met124 is required for high-affinity productive association of CaM with both RyR1 (2) and RyR2.



20. Lafitte D, Tsvetkov PO, Devred F, Toci R, Barras F, Briand C, Makarov AA, and Haiech J. Cation binding mode of fully oxidized calmodulin explained by the unfolding of the apostate. Biochim Biophys Acta 1600: 105–110, 2002. 21. Meissner G and el-Hashem A. Ryanodine as a functional probe of the skeletal muscle sarcoplasmic reticulum Ca2⫹ release channel. Mol Cell Biochem 114: 119 –123, 1992. 22. Montgomery HJ, Bartlett R, Perdickis B, Jervis E, Squier TC, and Guillemette JG. Activation of constitutive nitric oxide synthases by oxidized calmodulin mutants. Biochemistry 42: 7759 –7768, 2003. 23. Nelson MR and Chazin WJ. An interaction-based analysis of calciuminduced conformational changes in Ca2⫹ sensor proteins. Protein Sci 7: 270 –282, 1998. 24. O’Neil KT and DeGrado WF. How calmodulin binds its targets: sequence independent recognition of amphiphilic ␣-helices. Trends Biochem Sci 15: 59 – 64, 1990. 25. Rodney GG, Krol J, Williams B, Beckingham K, and Hamilton SL. The carboxy-terminal calcium binding sites of calmodulin control calmodulin’s switch from an activator to an inhibitor of RyR1. Biochemistry 40: 12430 –12435, 2001. 26. Saimi Y and Kung C. Calmodulin as an ion channel subunit. Annu Rev Physiol 64: 289 –311, 2002. 27. Sharp KA, Nicholls A, Friedman R, and Honig B. Extracting hydrophobic free energies from experimental data: relationship to protein folding and theoretical models. Biochemistry 30: 9686 –9697, 1991. 28. Squier TC and Bigelow DJ. Protein oxidation and age-dependent alterations in calcium homeostasis. Front Biosci 504 –526, 2000. 29. Strasburg GM, Hogan M, Birmachu W, Thomas DD, and Louis CF. Site-specific derivatives of wheat germ calmodulin. Interactions with troponin and sarcoplasmic reticulum. J Biol Chem 263: 542–548, 1988. 30. Vetter SW and Leclerc E. Novel aspects of calmodulin target recognition and activation. Eur J Biochem 270: 404 – 414, 2003. 31. Vogel HJ and Zhang M. Protein engineering and NMR studies of calmodulin. Mol Cell Biochem 149/150: 3–15, 1995. 32. Vougier S, Mary J, Dautin N, Vinh J, Friguer B, and Ladant D. Essential role of methionine residues in calmodulin binding to Bordetella pertussis adenylate cyclase, as probed by selective oxidation and repair by peptide methionine sulfoxide reductases. J Biol Chem 279: 30210 –30218, 2004. 33. Walsh M and Stevens FC. Chemical modification studies on the Ca2⫹dependent protein modulator: the role of methionine residues in the activation of cyclic nucleotide phosphodiesterase. Biochemistry 17: 3924 – 3930, 1978. 34. Yin D, Sun H, Weaver RF, and Squier TC. Nonessential role for methionines in the productive association between calmodulin and the plasma membrane Ca-ATPase. Biochemistry 38: 13654 –13660, 1999. 35. Yuan T, Ouyang H, and Vogel HJ. Surface exposure of methionine side chains of calmodulin in solution. J Biol Chem 274: 8411– 8420, 1999.

290 • FEBRUARY 2006 •

www.ajpheart.org


5. Bigelow DJ and Squier TC. Redox modulation of cellular signaling and metabolism through reversible oxidation of methionine sensors in calcium regulatory proteins. Biochim Biophys Acta 1703: 121–134, 2005. 6. Brooks SJP and Storey KB. Bound and determined: a computer program for making buffers of defined ion concentrations. Anal Biochem 223: 271–281, 1983. 7. Brot N and Weissbach H. Biochemistry and physiological role of methionine sulfoxide residues in proteins. Arch Biochem Biophys 233: 271–281, 1983. 8. Chin D and Means AR. Methionine to glutamine substitutions in the C-terminal domain of calmodulin impair the activation of three protein kinases. J Biol Chem 271: 30465–30471, 1996. 9. Chin D and Means AR. Calmodulin: a prototypical calcium sensor. Trends Cell Biol 10: 322–328, 2000. 10. Chin D, Winkler KE, and Means AR. Characterization of substrate phosphorylation and use of calmodulin mutants to address implications form the enzyme crystal structure of calmodulin-dependent protein kinase I. J Biol Chem 272: 31235–31240, 1997. 11. Chu A, Diaz-Munoz M, Hawkes MJ, Brush K, and Hamilton SL. Ryanodine as a probe for the functional state of the skeletal muscle sarcoplasmic reticulum calcium release channel. Mol Pharmacol 37: 735–741, 1990. 12. Davis TN, Urdea MS, Masiarz FR, and Thorner J. Isolation of the yeast calmodulin gene: calmodulin is an essential protein. Cell 47: 423– 431, 1986. 13. Dayhoff MO. Atlas of Protein Sequence and Structure, Washington D. C.: Natl. Biomed. Res. Found., 1978. 14. Fill M and Copello JA. Ryanodine receptor calcium release channels. Physiol Rev 82: 893–922, 2002. 15. Fruen BR, Bardy JM, Byrem TM, Strasburg GM, and Louis CF. Differential Ca2⫹ sensitivity of skeletal and cardiac muscle ryanodine receptors in the presence of calmodulin. Am J Physiol Cell Physiol 279: C724 –C733, 2000. 16. Fruen BR, Black DJ, Bloomquist RA, Bardy JM, Johnson JD, Louis CF, and Balog EM. Regulation of the RyR1 and RyR2 Ca2⫹ release channel isoforms by Ca2⫹-insensitive mutants of calmodulin. Biochemistry 42: 2740 –2747, 2003. 17. Gellman SH. On the role of methionine residues in the sequenceindependent recognition of nonpolar protein surfaces. Biochemistry 30: 6633– 6636, 1991. 18. Gopalakrishna R and Anderson WB. Ca2⫹-induced hydrophobic site on calmodulin: application for purification of calmodulin by phenyl-Sepharose affinity chromatography. Biochem Biophys Res Commun 29: 830 – 836, 1982. 19. Hu¨hmer AFR, Gerber NC, de Montellano PRO, and Scho¨neich C. Peroxynitrate reduction of calmodulin stimulation of neuronal nitric oxide synthase. Chem Res Toxicol 9: 484 – 491, 1996.

H799