Putative Roles of Type 3 Ryanodine Receptor Isoforms (RyR3)

extracellular matrix proteins. Thromb 14:1792–1798.

Arterio

Vlodavsky I, Friedmann Y, Elkin M, et al.: 1999. Mammalian heparanase: gene cloning, expression and function in tumor progression and metastasis. Nat Med 5:793–802. Weiser MC, Belknap JK, Grieshaber SS, et al.: 1996. Developmental regulation of perlecan gene expression in aortic smooth muscle cells. Matrix Biol 15:331–340. Wight TN: 1995. The extracellular matrix and atherosclerosis. Curr Opin Lipid 6:326–334.

Woods A, Couchman J, Syndecans R: 1998. Synergistic activators of cell adhesion. Trends Cell Biol 8:189–192. Yokokawa, K, Tahara, H, Kohno, M, et al.: 1993. Heparin regulates endothelin production through endothelium derived nitric oxide in human endothelial cells. J Clin Invest 92:2080–2085.

PII S1050-1738(00)00048-7

TCM

Putative Roles of Type 3 Ryanodine Receptor Isoforms (RyR3) Yasuo Ogawa*, Nagomi Kurebayashi, and Takashi Murayama Ca21-release from the sarcoplasmic or endoplasmic reticulum, the intracellular Ca21 store, is mediated by the ryanodine receptor (RyR) and/or the inositol trisphosphate receptor (IP3R). While IP3R is a ligand(IP3)-operated channel, RyR can be gated by a ligand (Ca 21) and/or mechanical coupling with the voltage sensor. There are three genetically distinct isoforms among RyR in mammals: RyR1–3. RyR1, the primary isoform in the skeletal muscle, can be gated by direct or indirect coupling with the conformation change of the a1S subunit of dihydropyridine receptor (DHPR) on the T-tubules (transversely invaginated sarcolemma) upon depolarization of skeletal muscles or by the increased cytoplasmic Ca21 (Ca21-induced Ca21 release, CICR). RyR2, the primary isoform in the cardiac ventricular muscle (and, in a lesser amount, the brain), can be gated by Ca21 which flows in through DHPR, especially the a1C subunit on depolarization. RyR3 is distributed ubiquitously in various tissues and may be coexpressed with RyR1 and RyR2. RyR3 is considered to be similar to RyR2 in the respect that it can be activated by Ca21, in view of the lack of available evidence to show the activation by the a1S subunit. Therefore, it is anticipated that RyR3 might take part through CICR in Ca21 signaling in smooth muscle and other non-muscle cells. To address the possible involvement of the CICR mechanism in the Ca21 signal transduction, it is critical to assess the effect of Mg 21 on the CICR activity and the cytoplasmic concentration of Mg 21. In this brief review, our discussion focuses on the effects of Ca21 and Mg21 on the activity of RyR3. (Trends Cardiovasc Med 2000;10:65–70). © 2000 Elsevier Science Inc.

Yasuo Ogawa, Nagomi Kurebayashi, and Takashi Murayama are at the Department of Pharmacology, Juntendo University School of Medicine, Tokyo, Japan * Address correspondence to: Yasuo Ogawa, Department of Pharmacology, Juntendo University School of Medicine, Hongo, Bunkyo-ku, Tokyo 113-8421 Japan © 2000, Elsevier Science Inc. All rights reserved. 1050-1738/00/$-see front matter

TCM Vol. 10, No. 2, 2000

Before discussion on this matter advances further, it will be helpful to review the ryanodine receptor (RyR); for details, please refer to recently published reviews or books (Sutko and Airey 1996, Sitsapesan and Williams 1998, Ogawa and Murayama 1998, Sorrentino and Reggiani 1999, Ogawa et al. 1999a and 1999b) that supply many invaluable references which were omitted here because of limited space. Table 1 shows a comparison of general properties of RyR1–3 (Sutko and Airey 1996, Sorrentino and Reggiani 1999, Ogawa et al. 1999a). They are functionally very similar in Ca21induced Ca21 release (CICR) activity in spite of some differences in the amino acid sequences: a low concentration of Ca21 (submicromolar to tens of micromolar) activates RyR, and ATP and caffeine are stimulatory. Although there is a slight difference in the channel-gating behavior among isoforms, their unit conductances are similar. A high concentration of Ca21 (millimolar), in contrast, inhibits RyR. Extent of the inhibitory effect of Mg21 is dependent on the coexisting Ca21 concentration, while that of procaine, in contrast, is independent of it (Kurebayashi and Ogawa 1998). These actions of Ca21 and Mg21 are understood as follows: the RyR has highaffinity activating Ca21-sites (A-sites) and low-affinity inactivating Ca21-sites (Isites). Mg21 exerts competitive antagonism against Ca21 at the A-sites and agonistic action with Ca21 at the I-sites. Ruthenium red is inhibitory. A FK506binding protein, FKBP12, and calmodulin can be reversible modulators of RyR activity as shown in Table 1, and they are considered to be accessory proteins to RyR molecules in the native Ca21-release channel. Ryanodine, on the other hand, irreversibly modulates the channel activity in a particular way. Diaphragm and soleus muscle in a mammalian adult have a little RyR3 (roughly less than 1% of RyR1) in addition to RyR1, whereas the other skeletal muscles express RyR1 alone (Ogawa and Murayama 1998, Sorrentino and Reggiani 1999). Many non-mammalian vertebrate skeletal muscles, however, express both a- and b-RyR in nearly equal amounts. a- and b-RyR in non-mammalian vertebrates are determined to be homologous to RyR1 and RyR3, respectively (Ogawa and Murayama 1998). Their genetic loci, however, are not yet identified, whereas

65

Table 1. Comparison among RyR isoforms RyR1

RyR2

RyR3

Gene locus in human Amino acid residues Calculated MW (kDa)

19q13.1 5037 565

1q42.1–43 4976 565

15q14–15 4872 552

Different regions D1 D2 Mobility on SDS-PAGE Active form Appearance

4249–4626 1302–1408 Smallest Homotetramer Quatrefoil, foot in situ

4210–4562 1315–1408 Intermediate Homotetramer Quatrefoil, foot in situ

4100–4400 Deleted Largest Homotetramer Quatrefoil

[3H]ryanodine binding Bmax (mol/mol tetramer ) KD (nM) EC50 / Ca21 (mM) IC50 / Ca21 (mM)

1 4–7 1–4 2

1 3–6 2 7

1 2 10–14 3

Single channel (500 mM KCl) Conductance (pS) Mean open time (ms) Stimulatory Inhibitory FKBP Removal of FKBP

630 2.58 Adenine nucleotides, caffeine Procaine, Mg21, ruthenium red FKBP12 Activation

550

743 4.85 Adenine nucleotides, caffeine Adenine nucleotides, caffeine Procaine, Mg21, ruthenium red Procaine, Mg21, ruthenium red FKBP12.6 FKBP12 Controversiala No activationb

Inhibition Activation ( 0.1 2 10mM Ca21 )

Inhibition Activation

a1S b, a2/d, g 1 alternate apposition of tetrads to every other foot

a1C b, a2/d 2 in close proximity, but no particular relationship

CaM

Ca21 <0.1mM or >3mM Ca21 : 0.1–1mM

DHPR (voltage sensor) Main subunit Modulatory subunits Tetrad formation of DHPR DHPR-Foot

Inhibition Activation

Partly modified from Ogawa, Y., et al. (1999a) a Some reported “No activation” [Timerman et al. (1996)] whereas others, “Activation” [Marx et al. (2000)]. b Copello et al. (1999).

those for RyR1-3 are determined in some mammals. The mRNA for RyR3 can be detectable not only in skeletal muscles but also in the brain, vascular smooth muscles and various other tissues (Ogawa and Murayama 1998, Sorrentino and Reggiani 1999). Cardiac ventricles in vertebrates have a distinct isoform of RyR, RyR2 or cardiac isoform (Sutko and Airey 1996). In various vertebrate tissues, mRNAs for multiple RyR isoforms were detected. Therefore, vascular smooth muscles as well as skeletal muscles and the brain may coexpress distinct isoforms of RyR in the same cells. It is characteristic of RyR that the homotetramer is formed as a functional unit in contrast to the case with IP3R, which may form a heterotetramer as well as a homotetramer.

66

While RyR1- or RyR2-gene knockout is lethal in mice (Takeshima et al. 1994, 1995, 1998), RyR3-knockout mice survive to produce offspring (Takeshima et al. 1996). No abnormality except for hyperactivity in locomotion was observed. A crooked-neck dwarf mutant chick which failed to express a-RyR was also lethal, as true of RyR1-knockout mouse (Sutko and Airey 1996); however, there is no report about b-RyR knockout animals. We would like to point out that RyR1 and RyR2 appear at an early stage of development and that their lack results in the impaired formation of the other intracellular components including myofibrils. RyR3, on the other hand, appears at a late stage of development, just prior to birth, and increases tran-

siently in amount after the birth up to about two weeks, thereafter fading out to almost nothing in skeletal muscles (Sorrentino and Reggiani 1999). In the brain, however, this transient change in amount of RyR3 was not observed (Balschun et al. 1999). The lack of RyR3 causes much less impairment of myofibril formation. • Methodology-Dependent Ca21-Sensitivity of RyR3 Figure 1A shows Ca21-dependent single channel activity of RyR3 together with that of RyR1 for comparison as incorporated into the lipid bilayer (Murayama et al. 1999). The threshold of Ca21 concentration for activation of RyR3 is around 0.1 mM (pCa 7), and its channel opening TCM Vol. 10, No. 2, 2000

was definitely detected at pCa 6.7 (0.2 mM). The channel was activated steeply in a nearly all-or-none manner by Ca21 between pCa 7 and 6, and was almost in the fully activated state (Po ~ 1) in the wide range of pCa 6-3. The inactivation by Ca21 was weak, and the activity in the presence of 3 mM Ca21 was maintained as high as Po ~ 0.6. RyR1, in contrast, showed biphasic Ca21 dependence: this dependence is less steep than for RyR3 in the range of stimulatory Ca21 concen-

trations, whereas it is more marked in the range of inhibitory Ca21 concentrations. Observation of two states of activity, high-Po and low-Po, both of which show analogous Ca21-dependence is another characteristic of RyR1. The EC50 values for Ca21 in the channel opening were 1 mM with RyR1 and 0.3 mM with RyR3: RyR3 was more sensitive to Ca21 in its activation than RyR1. RyR3 was more resistant or immune to Ca21 inactivation in the presence of high concen-

Figure 1. Ca21-dependence of RyR1 and RyR3 isoforms. (A) Single channel activity on lipid bilayer. (B) [3H]ryanodine binding. The values 100% stand for 132 and 156 pmol/mg protein for RyR1 and RyR3, respectively. For details, refer to Murayama et al. (1999). Courtesy of the American Society for Biochemistry and Molecular Biology.

TCM Vol. 10, No. 2, 2000

trations of Ca21 than RyR1. Similar results were also reported by Jeyakumar et al. (1998). This Ca21- dependence is similar to that of RyR2 (Meissner 1994, Sutko and Airey 1996). This is also the case with b-RyRs from chicken (Percival et al. 1994) and fish skeletal muscle (O’Brien et al. 1995), and heterologously expressed RyR3 (Chen et al. 1997). When the activity of each isoform was monitored by [3H]ryanodine binding which was brought about only by the open channels, however, the conclusions were different (Figure 1B). RyR1 showed a single state of the activity. It was more sensitive to Ca21 activation than RyR3 (EC50 values for Ca21 were 3.5 mM with RyR1 and 14 mM with RyR3), whereas the two isoforms showed indistinguishable inactivation with high concentrations of Ca21 or Mg21 (Murayama et al. 1999). These results are consistent with those of the CICR activity which were determined in skinned skeletal muscle fibers from wildtype (mainly RyR1) and RyR1-knockout mice (RyR3) (Takeshima et al. 1995). It should be noted that a- and b-RyR from frog skeletal muscle showed only minor difference at most in the Ca21-dependent [3H]ryanodine binding activity and were very similar to that of RyR3. Therefore, the higher sensitivity to Ca21 of RyR1 than RyR3 or a-RyR is due to speciesspecificity, and not to the isoform specificity. In other words, we may conclude that RyR3 and its homologue b-RyR show very similar properties irrespective of their origins, whereas RyR1 and its homologue, a-RyR, are more variable (Ogawa and Murayama 1998). The reason the distinct Ca21-dependence was obtained by the methods adopted to determine the activity has not yet been completely elucidated. A possible explanation for this finding, however, is that the channel activity on the lipid bilayer may be vulnerable to oxidation. Marengo et al. (1998) clearly showed that transition from the low Po state to the high Po state (more oxidized state) occurred spontaneously. RyR1 showed two states, high-Po and low-Po as shown in Figure 1A, consistent with the results already reported (Copello et al. 1997, Marengo et al. 1998). RyR3 may be more sensitively affected by oxidation. While the Po value of RyR3 at pCa 6.7 was intermediate between high-Po and low-Po of RyR1, this value steeply increased to ~1 with a slight increase in the Ca21 con-

67

centration. This result may suggest that some Ca21-dependent process(es) could also be involved in the abrupt transition of the channel activity in RyR3, and may be the reason a single state of activity was detected with RyR3. The activity attained to Po ~ 1 implies that ATP or caffeine would hardly show their stimulatory effect at a saturating Ca21 concentration, say, 10 mM Ca21 [see also the results with b-RyR reported by Percival et al. (1994)]. [3H]ryanodine binding and CICR activity showed a stimulatory effect of ATP and caffeine on RyR3 in the presence of saturating Ca21 concentrations (Takeshima et al. 1995, Ogawa et al. 1999a and 1999b). [3H]ryanodine binding activity, therefore, will hereafter be adopted to make a fair evaluation of Ca21 activation of RyR molecules. The CICR activity with skinned fibers may measure the activity in situ. • Affinities for Ca21 and Mg21 of A-sites and I-sites Mg21 and ATP occur in the sarcoplasm. The intracellular Mg21 concentration is estimated to be 1 mM in striated muscle (Westerblad and Allen 1992, Konishi et al. 1993). In smooth muscles and nonmuscle cells, however, it is reported to be lower and estimated to be about 0.5 mM (Nakayama and Tomita 1991, Raju et al. 1989). The intracellular ATP concentration is estimated to be 3–9 mM (Godt and Maughan 1988), most of which occurs as MgATP. It is well known that free ATP stimulates the RyR activity: CICR, Ca21-dependent [3H]ryanodine binding and single channel activity on the lipid bilayer. However, it remains to be determined whether MgATP is as potent as free ATP. To assess the functional role of CICR mediated by RyR in cells, the effects of Mg21 and MgATP on the RyR must be evaluated. Another concern is that purified RyR isoforms might be different in their properties from the native Ca21 release channels in the SR which are composed of RyR, FKBP, calmodulin and other components. The Ca21 dependence of [3H]ryanodine binding was indeed affected by the addition of CHAPS and phospholipids (Ogawa et al. 1999b). To address these problems, Murayama et al. (1998, 2000) determined the affinities for Ca21 and Mg21 of A- and I-sites using a- and b-RyR purified from bullfrog skeletal muscle, designing an effective procedure for analysis of determi-

68

nations of [3H]ryanodine binding in the presence of various concentrations of Ca21 and Mg21. The analysis was further extended to CICR activity in skinned fibers. The RyR activity ([3H]ryanodine binding or CICR activity), A, will be expressed in the presence of specified concentrations of Ca21 and Mg21 as follows: A 5 A *fA*(1 2 fI) where fA and (1 2 fI) represent the fractions of A-site occupied by Ca21 and I-site which is free of Ca21 or Mg21, respectively; A stands for the peak activity which could be attained when fA 5 1 and fI 5 0. A full description of fA and (1 2 fI) is found in the legend to Figure 2. Whereas fA and fI are functions of Ca21 and Mg21, A is independent of these divalent cations. Alternatively, A could be an entirely different function of Ca21. Representative values of parameters are as follows: a-RyR has the A-site (KCa 5 11 mM, KMg 5 320 mM) and the I-site

(KCa 5 2.4 mM, KMg 5 2.8 mM) (Murayama et al. 1998, 2000), and b-RyR shows the A-site (KCa 5 18 mM, KMg 5 330 mM) and the I-site (KCa 5 2.3 mM, KMg 5 3.1 mM) (Murayama et al. 2000). These results clearly show that a- and bRyR are affected by Ca21 and Mg21 in a way so similar as to be indistinguishable between the two isoforms. The A-site favors Ca21 over Mg21 about 30 times, whereas the I-site shows similar affinity for Ca21 and Mg21. From the determinations of CICR activity in skinned frog skeletal muscle fibers, furthermore, the following conclusions were obtained (Murayama et al. 1998, 2000). (1) The affinity values for Ca21 and Mg21 of the A-site and I-site of the SR in skinned fibers were greater by 3–5 times than the counterparts of purified RyR with unchanged selectivity between Ca21 and Mg21. The representative values with skinned fibers are: for the A-site, KCa 5 2.5 mM, KMg 5 75 mM; and for the I-site, KCa 5 0.4 mM, KMg 5 0.3 mM. (2) There

Figure 2. Simulated CICR activity of RyR3 and the effect of 0.5 mM Mg21. The CICR activity, A, is expressed as follows: A 5 A *fA*(1 2 fI) fA 5 [Ca21]/{[Ca21] 1 KA,Ca*(1 1 [Mg21]/KA,Mg)} (1 2 fI) 5 1/(1 1 [Ca21]/KI,Ca 1 [Mg21]/KI,Mg). The parameter values which bring about curves shown in the figure are as follows: A 5 45 min21, KA,Ca 5 2.5 mM, KA,Mg 5 75 mM, KI,Ca 5 0.4 mM, KI,Mg 5 0.3 mM. Although Hill coefficient for each site was not always equal to 1, Murayama et al. (2000) showed that it could be assumed to be 1 for CICR in the presence of 4 mM ATP. Here the equation is expressed under the assumption of Hill coefficient 5 1 for simplicity. The CICR activity of the SR in skinned frog skeletal muscle is assumed to represent that of RyR3 in the mammalian skeletal muscles. Continuous lines correspond to CICR activity in the presence of 4 mM ATP with or without 0.5 mM Mg21; interrupted curves, fA with or without 0.5 mM Mg21; dotted curves, (1 2 fI) with or without 0.5 mM Mg21. Please note that the cytoplasmic Ca21 concentration changes from about 0.1 mM at rest to around 1 mM upon smooth muscle activation in the presence of 0.5 mM Mg21.

TCM Vol. 10, No. 2, 2000

are no differences in the stimulatory effect of an adenine nucleotide among its free form, Mg- and Ca-bound form. The effect of AMPPCP was to increase the peak value ( A ) without change in the affinity for Ca21 or Mg21 of the A- or I-site. As MgATP concentrations increase, in other words, only the peak value ( A ) increases, but the Ca21-dependences of CICR are unchanged. The presence of 4 mM ATP would cause over 200-fold increase in A , without significant change in fA or fI. This conclusion is at variance with Meissner et al. (1986, 1997), who claimed that the decreased affinity values of the I-sites were responsible for the stimulatory effect of the reagent. (3) The effect of caffeine is both an increase in the peak value ( A ) and an increase in the affinity for Ca21 alone of the A-site with unchanged affinity for Mg21 (resulting in an increase of fA). The affinity for Ca21 and Mg21 of the I-site was slightly decreased; however, this decrease was too small to explain the Ca21-releasing action of caffeine. The extent of the increase in affinity for Ca21 of the A-site (i.e., fA by caffeine), was similar between mammalian and frog skeletal muscle SR vesicles. However, the increase in A was much greater with frog SR vesicles than that with mammalian SR vesicles. This is consistent with the observation that caffeine is more effective in inducing muscle contracture with frog skeletal muscle than with mammalian (Ogawa et al. 1999b). These findings indicate that the effect of the increase in A is important for the induction of Ca21 release in vivo. ATP causes remarkable increase in A without change in fA or fI. Its stimulatory effect is not only Ca21dependent but also Ca21-independent (Kermode et al. 1998, Murayama et al. 2000). In the presence of Ca21, the increase of Po by 0.1–0.2 mM ATP in single channel activity was reported to be due primarily to the increase in the frequency of channel openings, whereas large increases in the duration of the open state was claimed to be the main cause of the further elevations in Po by mM ATP. No change in the channel conductance was reported. Therefore the effects of ATP on RyR may be multiple and heterogeneous. The increase in A for CICR and [3H]ryanodine binding, however, appears to be homogeneous. The Bmax for [3H]ryanodine binding (the maximum binding in the presence of an TCM Vol. 10, No. 2, 2000

infinite amount of the ligand) was increased to 1 mol/tetramer of RyR (see Table 1) together with the enhanced affinity for [3H]ryanodine as the ATP concentration was increased. Whereas the single channel activity reports the gating property of active channels alone and nothing about silent channels, CICR and [3H]ryanodine binding inform us of the averaged activity of the whole channel. Taking these findings into consideration, a simple, but curious, interpretation might be that the occupation by Ca21 of the A-site may be a necessary but not sufficient condition for open channels and that ATP and caffeine which increase A modulate RYR conformations on Ca21 activation, resulting in an increased affinity for the ligand in [3H]ryanodine binding and enhanced P0 in CICR. This means that the gating properties of the channels activated by Ca21 are further affected in favor of open channels by ATP and caffeine. The Ca21 and Mg21 dependence of the native RyR3 on the SR membrane within cells is not yet known. Because purified RyR3 is very similar to a- and b-RyR from frog skeletal muscle in Ca21-dependent [3H]ryanodine binding, the CICR activity in skinned frog skeletal muscle fibers may represent that for native RyR3 within cells. The cytoplasmic Mg21 concentration is estimated to be about 0.5 mM in smooth muscle and non-muscle cells. Figure 2 shows the CICR activity curves at various Ca21 concentrations in the presence and absence of 0.5 mM Mg21 which are calculated from the parameters mentioned above. These curves may give the expected CICR activity of RyR3 within mammalian cells. As shown, Mg21 shifts the pCa-CICR activity relationship toward a higher Ca21 concentration range with a reduced peak value. The resting myoplasmic Ca21 concentration is around 0.1 mM in smooth muscle cells, and will increase to near 1 mM during activation which produces the development of the full tension (Bárány 1996). Figure 2 shows that the CICR activity within cells may be very low at pCa 6, only a few percent of the maximum value. This may be unfavorable for involvement of RyR3 in the Ca21 signal transduction. The RyR3-knockout mice grew up normally and produced offspring, having only slightly increased locomotive activity (Takeshima et al. 1996, Balschun et al. 1999). This suggests that

the functional contribution of RyR3 is minor. The miniscule amount of RyR3 molecules may also be the cause. The conclusion, however, must be reserved for the case of some neurons, where the Ca21 increase may occur in a very strictly localized region (Eilers et al. 1995). Recently, Balschun et al. (1999) reported modulated synaptic plasticity of the hippocampal region of RyR3-knockout mouse. Further investigation is required to clarify whether this modulation is directly caused by the change of Ca21 which is brought about by the lack of RyR3. • Conclusions On the basis of single channel activity on the lipid bilayer, RyR3 was believed to be similar to RyR2; high sensitivity to the Ca21 activation and resistance to the Ca21 inactivation. [3H]ryanodine binding activity and CICR activity, however, showed that RyR3 was as sensitive to Ca21 inactivation and to Mg21 inhibition as RyR1. Therefore, the CICR activity of the RyR3 in the sarcoplasmic or endoplasmic reticulum may be markedly reduced by the cytoplasmic Mg21, which is estimated to be 0.5–1 mM. Although mRNA for RyR3 was detected, the content of the RyR3 protein was not determined except in skeletal muscle or brain, where the amount was miniscule. Careful consideration must therefore be given to the functional contribution of CICR mediated by RyR3 under physiological conditions.

References Balschun D, Wolfer DP, Bertocchini F, et al.: 1999. Deletion of the ryanodine receptor type 3 (RyR3) impairs forms of synaptic plasticity and spatial learning. EMBO J 18:5264– 5273. Bárány M, ed.: 1996. Biochemistry of Smooth Muscle Contraction. San Diego, CA, Academic Press. Chen SRW, Li XL, Ebisawa K, et al.: 1997. Functional characterization of the recombinant type 3 Ca21 release channel (ryanodine receptor) expressed in HEK293 cells. J Biol Chem 272:24,234–24,246. Copello JA, Barg S, Onoue H, et al.: 1997. Heterogeneity of Ca21 gating of skeletal muscle and cardiac ryanodine receptors. Biophys J 73:141–156. Copello JA, Jeyakumar L, Fleischer S: 1999. FK binding proteins (FKBP12 and FKBP 12.6) do not modulate channel activity of

69

ryanodine receptor isoform 3 (RyR3) from bovine diaphragm muscle (BDM). Biophys J 76:A469. Eilers J, Augustine GJ, Konnerth A: 1995. Subthreshold synaptic Ca21 signaling in fine dendrites and spines of cerebellar Purkinje neurons. Nature 373:155–158. Godt RE, Maughan DW: 1988. On the composition of the cytosol of relaxed skeletal muscle of the frog. Am J Physiol 254: C591–C604. Jeyakumar LH, Copello JA, O’Malley AM, et al.: 1998. Purification and characterization of ryanodine receptor 3 from mammalian tissue. J Biol Chem 273:16,011– 16,020. Kermode H, Williams AJ, Sitsapesan R: 1998. The interactions of ATP, ADP, and inorganic phosphate with the sheep cardiac ryanodine receptor. Biophys J 74:1296– 1304. Konishi M, Suda N, Kurihara S: 1993. Fluorescence signals from the Mg21/Ca21 indicator furaptra in frog skeletal muscle fibers. Biophys J 64:223–239. Kurebayashi N, Ogawa Y: 1998. Effect of luminal calcium on Ca21 release channel activity of sarcoplasmic reticulum in situ. Biophys J 74:1795–1807. Marengo JJ, Hidalgo C, Bull R: 1998. Sulfhydryl oxidation modifies the calcium dependence of ryanodine-sensitive calcium channels of excitable cells. Biophysical J 74:1263–1277. Marx SO, Reiken S, Hisamastu Y, et al.: 2000. PKA phosphorylation dissociates FKBP12.6 from the calcium release channel (ryanodine receptor): defective regulation in failing hearts. Cell 101:365–376.

Murayama T, Oba T, Katayama E, et al.: 1999. Further characterization of the type 3 ryanodine receptor (RyR3) purified from rabbit diaphragm. J Biol Chem 274:17,297– 17,308.

Sorrentino V, Reggiani C: 1999. Expression of the ryanodine receptor type 3 in skeletal muscle: a new partner in excitationcontraction coupling? Trends Cardiovasc Med 9:54–61.

Nakayama S, Tomita T: 1991. Regulation of intracellular free magnesium concentraion in the taenia of guinea-pig caecum. J Physiol (London) 435:559–572.

Sutko JL, Airey JA : 1996. Ryanodine receptor Ca21 release channels: does diversity in form equal diversity in function? Physiol Rev 76:1027–1071.

O’Brien J, Valdivia HH, Block BA: 1995. Physiological differences between the alpha and beta ryanodine receptors of fish skeletal muscle. Biophys J 68:471–482.

Takeshima H, Iino M, Takekura H, et al.: 1994. Excitation-contraction uncoupling and muscular degeneration in mice lacking functional skeletal muscle ryanodinereceptor gene. Nature 369:556–559.

Ogawa Y, Kurebayashi N, Murayama T: 1999a. Ryanodine receptor isoform in excitationcontraction coupling. Adv Biophys 36:27–64. Ogawa Y, Murayama T: 1998. Characteristics of ryanodine receptor 3 isoform (RyR3) and its homologues. In Sitsapesan R, Williams AJ, ed. The Structure and Function of Ryanodine Receptors. London, Imperial College Press, pp. 5–22.

Takeshima H, Ikemoto T, Nishi M, et al.: 1996. Generation and characterization of mutant mice lacking ryanodine receptor type 3. J Biol Chem 271:19,649–19,652. Takeshima H, Komazaki S, Hirose K, et al.: 1998. Embryonic lethality and abnormal cardiac myocytes in mice lacking ryanodine receptor type 2. EMBO J 17:3309–3316.

Ogawa Y, Murayama T, Kurebayashi N: 1999b. Comparison of properties of Ca 21 release channels between rabbit and frog skeletal muscles. Mol Cell Biochem 190: 191–201.

Takeshima H, Yamazawa T, Ikemoto T, et al.: 1995. Ca21-induced Ca21 release in myocytes from dyspedic mice lacking the type-1 ryanodine receptor. EMBO J 14:2999– 3006.

Percival AL, Williams AJ, Kenyon JL, et al.: 1994. Chicken skeletal muscle ryanodine receptor isoforms: ion channel properties. Biophys J 67:1834–1850.

Timerman AP, Onoue H, Xin H-B, et al.: 2000. Selective binding of FKBP12.6 by the cardiac ryanodine receptor. J Biol Chem 271:20,385–20,391.

Raju B, Murphy E, Levy LA, et al.: 1989. A fluorescent indicator for measuring cytosolic free magnesium. Am J Physiol 256:C540– C548.

Westerblad H, Allen DG: 1992. Myoplasmic free Mg21 concentration during repetitive stimulation of single fibres from mouse skeletal muscle. J Physiol (London) 453: 413–434.

Sitsapesan R, Williams AJ, eds: 1998. The Structure and Function of Ryanodine Receptors. London, Imperial College Press.

PII S1050-1738(00)00050-5

TCM

Meissner G: 1994. Ryanodine receptor/Ca21 release channels and their regulation by endogenous effectors. Annu Rev Physiol 56:485–508. Meissner G, Darling E, Eveleth J: 1986. Kinetics of rapid Ca21 release by sarcoplasmic reticulum: effects of Ca 21, Mg21, and adenine nucleotides. Biochemistry 25:236–244. Meissner G, Rios E, Tripathy A, et al.: 1997. Regulation of skeletal muscle Ca21 release channel (ryanodine receptor) by Ca21 and monovalent cations and anions. J Biol Chem 272:1628–1638. Murayama T, Kurebayashi N, Ogawa Y: 1998. Effects of caffeine and Mg 21 on the Ca21 activation and inactivation sites of frog ryanodine receptor. Biophys J 74: A163. Murayama T, Kurebayashi N, Ogawa Y: 2000. Role of Mg21 in Ca21 release through ryanodine receptors of frog skeletal muscle: modulations by adenine nucleotides and caffeine. Biophys J 78:1810–1824.

70

REPRINTS Reprints of articles in TCM are available (minimum order: 100). Please contact: Dan Cronin, Advertising Sales Elsevier Science Inc. 655 Avenue of the Americas New York, NY 10010 Phone: (212) 633-3813

Fax: (212) 633-3820

TCM Vol. 10, No. 2, 2000