FK-binding Protein Is Associated with the Ryanodine Receptor of Skeletal Muscle in Vertebrate Animals*

The ryanodine receptor/calcium release channel (RyR1) of sarcoplasmic reticulum from rabbit skeletal muscle terminal cisternae (TC) contains four tightly associated FK506-binding proteins (FKBP12). Dissociation and reconstitution studies have shown that RyR1 can be modulated by FKBP12, which helps to maintain the channel in the quiescent state. In this study, we found that the association of FKBP with RyR1 of skeletal muscle is common to each of the five classes of vertebrates. TC from skeletal muscle representing animals from different vertebrates, i.e. mammals (rabbit), birds (chick-en), reptiles (turtle), fish (salmon and rainbow trout), and amphibians (frog), were isolated. For each, we find the following: 1) FKBP12 is localized to the TC (there are four FKBP binding sites/ryanodine receptor); 2) soluble FKBP exchanges with the bound form on RyR1 of TC; 3) release of FKBP from terminal cisternae by drug (FK590) treatment leads to a significant reduction in the net calcium loading rate, consistent with channel activation (the calcium loading rate is restored to the control value by reconstitution with FKBP12); and 4) RyR1 of skeletal muscle TC can bind to and exchange with either FKBP12 or FKBP12.6 (FKBP12.6 is the novel FKBP isoform found selectively associated with RyR2 of dog cardiac sarcoplasmic reticulum). We conclude that FKBP is an integral part of the RyR1 of skeletal muscle in each of the classes of vertebrate animals. 35 S-labeled FKBP was estimated by quantitative densitometry of Coomassie Blue-stained SDS-PAGE gels using an image processing system (Technology Re- source Inc., Nashville, TN). Bovine serum albumin was used as the protein standard. Western Blot Analysis of TC Fractions— TC (20 m g of protein) was subjected to SDS-gel electrophoresis and transferred to 0.45 m m Immo- bilon-P transfer membranes (Millipore) using a semidry transfer appa-*

The ryanodine receptor/calcium release channel (RyR1) of sarcoplasmic reticulum from rabbit skeletal muscle terminal cisternae (TC) contains four tightly associated FK506-binding proteins (FKBP12). Dissociation and reconstitution studies have shown that RyR1 can be modulated by FKBP12, which helps to maintain the channel in the quiescent state. In this study, we found that the association of FKBP with RyR1 of skeletal muscle is common to each of the five classes of vertebrates. TC from skeletal muscle representing animals from different vertebrates, i.e. mammals (rabbit), birds (chicken), reptiles (turtle), fish (salmon and rainbow trout), and amphibians (frog), were isolated. For each, we find the following: 1) FKBP12 is localized to the TC (there are four FKBP binding sites/ryanodine receptor); 2) soluble FKBP exchanges with the bound form on RyR1 of TC; 3) release of FKBP from terminal cisternae by drug (FK590) treatment leads to a significant reduction in the net calcium loading rate, consistent with channel activation (the calcium loading rate is restored to the control value by reconstitution with FKBP12); and 4) RyR1 of skeletal muscle TC can bind to and exchange with either FKBP12 or FKBP12.6 (FKBP12.6 is the novel FKBP isoform found selectively associated with RyR2 of dog cardiac sarcoplasmic reticulum). We conclude that FKBP is an integral part of the RyR1 of skeletal muscle in each of the classes of vertebrate animals. The studies are consistent with a role for FKBP in skeletal muscle excitation-contraction coupling.
FK506 is a powerful immunosuppressive drug that prevents T-cell activation and is thereby used to prevent allograft rejection following transplant surgery. FK506-binding protein (FKBP12), 1 the cytosolic receptor for FK506, has a molecular mass of 11.8 kDa and is widely expressed in eukaryotic cells and tissues, predominantly in the cytosol. The sequence is highly conserved throughout eukaryotic phylogeny (1,2).
In rabbit skeletal muscle, FK-binding protein is bound to the ryanodine receptor of terminal cisternae of SR (3)(4)(5)(6)(7)(8)(9)(10), in a stoichiometry of four FKBP/ryanodine receptor (4,8); i.e. for mammalian skeletal muscle, the ryanodine receptor is a hetero-oligomer with a structural formula of (RyR1 protomer) 4 (FKBP12) 4 . Although FKBP is tightly bound to the RyR of terminal cisternae of SR, soluble FKBP readily exchanges with FKBP12 on the RyR in TC (6). The ryanodine receptor of heart SR is also associated with FKBP (7) in a stoichiometry of 4 (8), albeit with a novel isoform, i.e. FKBP12.6. The latter differs from FKBP12 by 18 of 108 amino acids (8). Recently, we found that the ryanodine receptor from skeletal muscle binds to and exchanges with both FKBP12 and FKBP12.6, whereas the RyR2 of heart binds to and exchanges only with FKBP12.6 (9). In vitro reconstitution studies indicate that FKBP12 modulates the channel function of the ryanodine receptor of skeletal muscle (RyR1) (4,5,10). In this regard, FKBP stabilizes the closed conformation of the skeletal muscle ryanodine receptor. Image enhancement analysis of cryoelectron micrographs of RyR1 shows that four FKBPs bind to RyR1 in 4-fold symmetry and pinpoint the binding sites (11). FKBP binds near the transverse tubule face of RyR1. This observation is consistent with FKBP being involved in the coupling of the dihydropyridine receptor to the ryanodine receptor in excitation-contraction coupling in skeletal muscle. If FKBP is essential for excitation-contraction coupling in skeletal muscle, it would be expected to be generally associated with and modulate the function of ryanodine receptors from diverse animals. In this study, we find that FKBP12 is associated with and modulates the skeletal muscle RyR1 calcium release channel in each of the five classes of vertebrates. H]ryanodine were purchased from NEN Life Science Products. Goat anti-rabbit alkaline phosphatase conjugate and nitro blue tetrazolium/5-bromo-4-chloro-3indolyl phosphate color development reagents for immunoblotting were obtained from Promega. Immobilon-P membrane was purchased from Millipore Corp. CHAPS, antipyrylazo III, and ruthenium red were obtained from Sigma. Sodium dodecyl sulfate, acrylamide, methylene bisacrylamide, and SDS-PAGE molecular weight standards were obtained from Bio-Rad. TSK GEL G3000SW column was purchased from TosoHaas.

Materials-Both
General Methods-The protein concentration of the TC membrane fractions was determined by the Folin reaction (13) using bovine serum albumin as a standard. SDS-PAGE was performed with a mini-slab gel apparatus (Hoeffer Scientific) using the buffer system described by Laemmli (14). The protein content of recombinant 35 S-labeled FKBP was estimated by quantitative densitometry of Coomassie Blue-stained SDS-PAGE gels using an image processing system (Technology Resource Inc., Nashville, TN). Bovine serum albumin was used as the protein standard.
Western Blot Analysis of TC Fractions-TC (20 g of protein) was subjected to SDS-gel electrophoresis and transferred to 0.45 m Immobilon-P transfer membranes (Millipore) using a semidry transfer appa-ratus. The transfer time was 15 min for FKBP and 1 h for ryanodine receptor (4). The membrane was blocked and washed as described previously (4). Finally, the membrane was developed with substrates for alkaline phosphatase.
Isolation of Terminal Cisternae from Skeletal Muscle-Terminal cisternae of sarcoplasmic reticulum fractions were isolated from predominantly white muscle of chicken, turtle, frog, and fish (rainbow trout and salmon) as described previously for rabbit fast twitch skeletal muscle (12,15,16). Microsomes obtained from the first and second homogenate (I and R series, respectively) were loaded onto a sucrose step gradient consisting of four steps, i.e. 27% (0.9 M), 32% (1.1 M), 38% (1.3 M), and 45% (1.6 M), and centrifuged overnight to equilibrium. The I4 and R4 fractions were obtained from the interfaces of the sucrose gradient between 1.3 and 1.6 M, and the I3 and R3 fractions were obtained from the 1.1 and 1.3 M interfaces. These fractions were analyzed for stoichiometry of FKBP/ryanodine receptor (4), drug binding activity (4,18), and calcium loading rate (with/without ruthenium red) in loading medium containing 1 mM Mg 2ϩ (16,20).
Exchange Isotherms of 35 S-Labeled FKBP with Terminal Cisternae-Assays for the exchange of bound FKBP on TC with soluble 35 S-labeled FKBP were performed essentially as described previously (6). Exchange isotherms of 35 S-labeled FKBP to TC (2 mg of protein/ml) were performed over a concentration of 0.05-1.2 M FKBP for 30 min at different temperatures depending on the species (37°C for rabbit and chicken, 0°C for fish and room temperature for turtle and frog), as necessary to retain loading activity. Nonspecific binding was determined in the presence of a 50-fold excess of unlabeled FKBP12 and FKBP12.6 and was subtracted from the total binding to give specific binding. Nonspecific binding did not change significantly with increased incubation time and temperature. Free FKBP was separated from bound FKBP isoform by diluting a 50-l sample (100 g of protein) into 150 l of ice-cold IHM buffer (10 mM imidazole, 0.3 M sucrose, 3 mM NaN 3 ) and immediately sedimentating the TC in a Beckman TL100 ultracentrifuge (TL100.1 rotor) at 35,000 rpm for 15 min at 4°C. The pellet was quickly rinsed twice with 200 l of distilled water, resuspended in 200 l of distilled water, and counted in 5 ml of CytoScint liquid scintillation mixture.
Expression and Purification of 35 S-Labeled FKBP12 and FKBP12.6 -Both 35 S-labeled FKBP isoforms were produced as described previously (17) with some modification. Briefly, an overnight culture of FKBP12-or FKBP12.6-producing Escherichia coli strain was inoculated with 200 ml of M9 medium containing 50 g/ml carbenicillin until the A 600 reached 0.4 OD units. The pellet obtained by centrifugation at 3000 rpm for 5 min was resuspended in 200 ml of RPMI 1640 medium containing 50 g/ml carbenicillin, 1:40 methionine, 1:40 cysteine, and 1:40 glucose compared with the regular medium. Isopropyl-1-thio-␤-D-galactopyranoside (final concentration 0.1 mM) was added and incubated 15 min at 37°C. Then 2.5 mCi of [ 35 S]methionine and [ 35 S]cysteine mixture was added, and the culture was incubated 5 h with 250 rpm rotary shaking at 37°C and further cultured overnight with 150 rpm shaking at 30°C. The production of the bacterial lysate from the FKBP-producing bacteria and purification of 35 S-labeled FKBP12 or 12.6 was as described for unlabeled FKBP (4,8). The specific activity of the [ 35 S]FKBP12 and [ 35 S]FKBP12.6 in the time span of these experiments changed from 1773 to 1700 cpm/pmol and from 2580 to 2200 cpm/pmol, respectively.
Preparation of FK590 Drug-treated and FKBP-depleted Terminal Cisternae-The endogenous FKBP was dissociated from TC (1.5 mg/ml) by incubation with either 5 or 10 M FK590 at different temperatures and times (30 min at 37°C for rabbit, 30 min at room temperature for chicken and turtle, 15 min at room temperature for frog and fish). TC resulting from this treatment are referred to as "drug-treated TC"; subsequently, sedimentation of the mixture in a Beckman TL 100.1 rotor at 35,000 rpm for 15 min at 2°C removes the soluble drug-FKBP complex, which remains in the supernatant (4). The pellet referred to as FKBP-depleted TC was resuspended in IHM buffer for drug binding assay (6).
Drug Binding Assays-Ryanodine binding of TC vesicles (25 g of rabbit skeletal muscle TC or 50 g of TC of other species) was measured essentially as described previously (4,18).
Assay of Calcium Loading Rate-The calcium loading rate of TC vesicles was measured spectrophotometrically with the calcium indicator antipyrylazo III at room temperature as described previously (4,20). TC vesicles (75 g of proteins) were added to 1 ml of loading medium (100 mM KPO 4 , pH 7.0, 0.5 mM antipyrylazo III, 1 mM MgCl 2 , and 1 mM ATP). Ca 2ϩ loading was initiated with the addition of 1 mM ATP. Following the uptake of calcium from the medium, 10 nmol of CaCl 2 was pulsed into the cuvette. The calcium uptake was followed by dual wavelength spectrophotometry (710 -790 nm). The calcium loading rate was measured following Ca 2ϩ uptake after pulsing three times with 10 nmol of Ca 2ϩ . The net Ca 2ϩ loading rate was calculated by averaging the rates from the first two pulses of Ca 2ϩ .
To observe the effect of FKBP on the calcium loading rate in FKBPdepleted TC, FKBP12 or FKBP12.6 (3.5 M) was added to the medium and preincubated several minutes, and then the calcium loading assay was performed as described above. In order to measure the enhancement of the calcium loading rate in TC vesicles, ruthenium red (7 M) was added to the assay medium following the addition of TC and before adding ATP and Ca 2ϩ .

TABLE I
Characteristics of TC from animals of different vertebrate classes Ca 2ϩ loading rates were measured in a medium containing 1 mM Mg 2ϩ in response to a single pulse of 50 mol of Ca 2ϩ with 50 g of protein in the presence (ϩRR) and absence (ϪRR) of ruthenium red. Analysis of fish and frog TC was carried out on two different preparations, and each was analyzed twice with quantitatively similar results (Ϯ10% for frog and Ϯ20% for fish). The chicken and turtle TC fractions were prepared once and analyzed twice with similar results (Ϯ10%).

Isolation and Characterization of Skeletal Muscle SR Fractions from Different
Vertebrate Species-Terminal cisternae enriched fractions were prepared from skeletal muscle from the different vertebrate animals by the procedure developed for rabbit skeletal muscle (12). Muscle microsomes were prepared and subfractionated on a sucrose density gradient. As found previously for rabbit, the highest density fractions I3, I4, R3, and R4 were most enriched in terminal cisternae of SR as indicated by ryanodine binding (Table I). This was confirmed by SDS-PAGE. In other words, the terminal cisternae fractions of rabbit, chicken, turtle, fish, and frog, representing the five classes of vertebrates, were enriched in bands referable to ryanodine receptor, calcium pump protein, and calsequestrin. The calcium pump protein and calsequestrin are quantitatively the major proteins of terminal cisternae as observed by Coomassie Blue staining (Fig. 1A) and staining with Stains-all for calsequestrin (Fig. 1B). The ryanodine receptor band was confirmed by Western blot analysis (Fig. 2) probed with our general RyR antipeptide antibody (derived from the highly conserved region, sequence 4681-4700 in rabbit skeletal muscle RyR1). The calsequestrin band gave a characteristic dark blue or purple color with Stains-all ( Fig. 1B) (22). The calsequestrin from different species migrates with somewhat different mobility; chicken is the fastest, whereas fish and frog have the slowest mobility.
Ryanodine binding to the terminal cisternae enriched fractions is the highest in I4 and R4 fraction and next highest in I3 and R3 (see Table I). The lowest density fractions (I1, I2, R1, and R2) had little ryanodine binding, as observed for the rabbit skeletal muscle terminal cisternae preparations (data not shown) (12). Another diagnostic of terminal cisternae is the enhanced net Ca 2ϩ loading rate in the presence of ruthenium red (20). The net calcium loading rate of terminal cisternae is determined predominantly by the Ca 2ϩ loading rate referable to the Ca 2ϩ pump protein, minus the Ca 2ϩ leak rate via the calcium release channel. A decrease in net Ca 2ϩ loading rate reflects channel activation. In the presence of ruthenium red, which closes the RyR channel, this enhanced Ca 2ϩ loading rate in terminal cisternae from rabbit skeletal muscle is approxi-mately 5-10-fold. The enhanced Ca 2ϩ loading rate in the terminal cisternae from other species varied from approximately 3 to 6. The longitudinal tubule fractions (R2 and I2), which were essentially devoid of ryanodine receptor, had a higher calcium loading rate compared with R4 and I4, which was not enhanced with ruthenium red (data was shown, see Ref. 20 for rabbit). FIG. 1. SDS-PAGE of skeletal muscle in terminal cisternae vesicles from diverse species. Electrophoresis was carried out on a 5-15% gradient of polyacrylamide in SDS. TC fraction (30 g) was stained with Coomassie Blue (Fig. 1A) and by Stains-all for calsequestrin (22) (Fig. 1B). Protein were loaded as follows. were analyzed by Western blotting using the FKBP antibody (3) that recognizes the N-terminal sequences of both FKBP12 and FKBP12.6 (Fig. 3). The immunoreactive bands in the terminal cisternae fractions from the different species of vertebrates were observed with mobility in the range of ϳ12 and 12.6 kDa. The mobility of the FKBP bands from the different animals varied somewhat. Rabbit has the same mobility as human recombinant FKBP12, and chicken is only slightly slower. The fish had the slowest mobility, just slightly faster than human recombinant FKBP12.6. The turtle and frog were between chicken and fish.
Quantitation of FKBP Binding Sites in the RyR of Terminal Cisternae-The stoichiometry of FKBP/ryanodine receptor in the terminal cisternae fractions from mammal, bird, reptile, and fish was measured by drug binding. The stoichiometry was 4 for rabbit and frog, but was Ͻ4 for fish (1.7), reptile (1.5), and bird (2.5) (Table II). This lower stoichiometry reflects unfilled FKBP binding sites on the ryanodine receptor (Table II), which were measured using FKBP exchange methodology (6). We find approximately four FKBP binding sites on the RyR1 from skeletal muscle in each of the classes of vertebrates. The binding sites could be filled by either FKBP12 or FKBP12.6 ( Table III).
The FKBP exchange methodology (6) previously developed for rabbit terminal cisternae had to be modified for some of the vertebrate classes, since the skeletal muscle terminal cisternae of turtle, fish, and frog were found to be sensitive to temperature inactivation. This was manifest by the inability to obtain good FKBP exchange as well as the loss of ryanodine binding. The terminal cisternae from fish skeletal muscle lost nearly all ryanodine binding activity at 25°C incubation for 30 min. Terminal cisternae from turtle and frog lost the ryanodine binding activity when incubated at 37°C for 30 min. For this reason, the exchange was performed at different temperatures.
The conditions for exchange are summarized in Table III. The stoichiometry of FKBP12 or FKBP12.6 binding sites/ryanodine receptor for the different animals is approximately 4 as determined from the B max for FKBP, obtained from exchange isotherms (Fig. 4), and the measurement of high affinity ryanodine binding.
The combined studies in Tables II and III indicate that there are four FKBP binding sites/ryanodine receptor in RyR1 from skeletal muscle in each of the classes of vertebrates, although the terminal cisternae of fish, reptile, and bird were isolated with less than a full quota of FKBP (stoichiometry of less than 4).
RyR1 Contains Four FKBP Binding Sites, Which Can Be Filled by FKBP12 or FKBP12.6 -We previously reported that the ryanodine receptor of rabbit skeletal muscle TC can bind approximately four equivalents of FKBP12 or FKBP12.6. This could mean that there are a total of eight FKBP binding sites, four each for FKBP12 and FKBP12.6; alternatively, there may be four sites that can bind either FKBP12 and/or FKBP12.6. To answer this question, exchange was carried out with a combination of both FKBP12 and 12.6, each at nearly saturating concentration (6,9). We find that there are only four FKBP binding sites, which can be occupied by either isoform or the combination of FKBP isoforms (Table IV).
FKBP Modulates the Net Calcium Loading Rate of TC in Different Vertebrate Animals-We previously found that removal of FKBP from rabbit terminal cisternae activated the RyR by making the RyR more sensitive to Ca 2ϩ (5,21) as studied by single channel measurements in planar lipid bilayers. This was supported by the macroscopic assay of net Ca 2ϩ loading in the terminal cisternae (4,20). The Ca 2ϩ loading rates of terminal cisternae from different species were differentially sensitive to inactivation. The time and temperature of  drug treatment had to be optimized as a compromise for efficient extraction of FKBP and the ability to restore Ca 2ϩ loading activity with FKBP reconstitution. The percentage of FKBP remaining in the pellet after extraction of FKBP with FK590 was 30% (chicken), 29% (turtle), 18% (frog), and 6% (fish), respectively. Release of FKBP from the ryanodine receptor of terminal cisternae by drug treatment reduced the net Ca 2ϩ loading rate, consistent with the increased Ca 2ϩ leak from the RyR referable to the release of FKBP (Fig. 5); i.e. FKBP-depleted terminal cisternae have a reduced net Ca 2ϩ loading rate compared with the control. The Ca 2ϩ loading rate was restored to control levels by the addition of either FKBP12 or FKBP12.6 to the assay, reflecting that the channel again becomes quiescent. Thus, FKBP12 and FKBP12.6 modulate the calcium release channel in the skeletal muscle terminal cisternae from the different species. DISCUSSION We find that FK-binding protein is associated with the ryanodine receptor of skeletal muscle terminal cisternae in animals representing each of the five classes of vertebrates. Until now, studies of FKBP association with RyR1 were limited to mammalian skeletal muscle. In this study, we find that each of the five classes of vertebrates shares a similarity of key char-  Fig 3) (6). [ 3 H]Ryanodine binding was performed at 60 nM [ 3 H]ryanodine. The binding data for TC fractions are expressed as the mean Ϯ S.E. for two preparations. The exchange studies on rabbit, fish, and frog were carried out on two different preparations with quantitatively similar results. The studies on chicken and turtle were carried out twice on the same preparation with quantitatively similar results. Fraction 3 of chicken skeletal muscle TC, which has lower ryanodine binding, more comparable with fractions 4 of frog and fish, was used for the exchange isotherms. The exchange isotherms were performed for 30 min at different temperatures (37°C for rabbit and chicken, room temperature for turtle and frog, and 0°C for fish). .0 for 30 min at 37°C (for rabbit () and chicken (q)), 0°C (for fish (f)), and room temperature (for turtle (ࡗ) and frog (OE)). Nonspecific binding was obtained by adding a 50-fold excess of unlabeled FKBP at each concentration of [ 35 S]FKBP. Specific binding is calculated from (total Ϫ nonspecific) binding. K d and B max obtained from this data are summarized in Table III. Exchange studies of chicken SkM TC were carried out with fraction R3 of chicken TC, which has lower ryanodine binding, more comparable with fraction 4 of fish and frog.  (20). Thus, FKBP modulates the RyR. The generality of FKBP association, exchange with RyR-bound FKBP, and modulation of RyR channel activity was found to be common for each of the vertebrate classes of animals.
Terminal cisternae fractions were prepared from animals representing each of the five classes of vertebrates by the procedure previously developed for rabbit skeletal muscle (12) ( Table I). The terminal cisternae share characteristic features: 1) higher isopycnic density by equilibrium sedimentation; 2) the enrichment in SR proteins (i.e. calcium pump protein, calsequestrin, and ryanodine receptor); 3) net Ca 2ϩ loading enhancement by the addition of ruthenium red, which closes the Ca 2ϩ leak referable to the ryanodine receptor (Table I); 4) association of FKBP with the ryanodine receptor and FKBP modulation of channel function (Fig. 5).
To assess the role of FKBP in the modulation of the ryanodine receptor in terminal cisternae from the different classes of vertebrates, we had to optimize conditions for drug treatment in order to be able to restore function by reconstitution studies. The conditions developed are summarized in Table V and Fig.  5. Likewise, the conditions for exchange of soluble FKBP with RyR-bound FKBP in TC of SR (Table III) had to be optimized to retain ligand binding. The exchange condition of 37°C previously worked out for mammalian TC was satisfactory also for bird. However, the treatment had to be modified for amphibian and reptile (to room temperature) and for fish (to 0°C).
The skeletal muscle terminal cisternae from mammals contain mainly RyR1 with a small amount of RyR3 (Ͻ1%). However, for the other vertebrate classes, approximately equal amounts of RyR1 (␣-isoform) and RyR3 (␤-isoform) are present as major constituents (24 -27). Even so, we measured four FKBP binding sites/ryanodine receptor in animals from each of the classes of vertebrates. The inference from these results is that RyR1 and RyR3 are similar with regard to the binding of FKBP; i.e. there are four FKBP binding sites/RyR3, and both RyR1 and RyR3 can bind either FKBP12 or FKBP12.6, unlike the cardiac RyR2, which selectively binds and exchanges only with FKBP12.6. Recent studies in our laboratory with the purified mammalian RyR3 receptor (28) confirm this inference. 2 The presence of FKBP in the skeletal muscle fraction was confirmed by Western blot analysis (Fig. 3). The mobility of the FKBP from the different animals varied somewhat depending FIG. 5. FKBP modulates channel function of skeletal muscle RyR from diverse species of vertebrate animals. The terminal cisternae fraction (75 g) was treated with FK590 (drug-treated (f)) or without FK590 (control (Ⅺ)) at different temperatures and times as described in Table IV. The calcium loading rate for control versus drug-treated terminal cisternae and reconstituted (treated plus FKBP12 (^) and treated plus FKBP12.6 (s)) are shown. For reconstitution, the FKBP12.0 or FKBP12.6 (3.5 M) were preincubated for 8 min with drug-treated TC, and then calcium loading was measured. The control in the presence of FKBP12 or FKBP12.6 is essentially unchanged (Ϯ10%) (not shown). The calcium loading rates from a single Ca 2ϩ pulse (10 M) are expressed as the mean Ϯ S.E. for two or more preparations. The loading rates of the drug-treated samples were significantly lower than the control and reconstituted samples: **, p Ͻ 0.01 for rabbit, chicken, and fish; *, p Ͻ 0.05 for frog and turtle. on the species. Mobility alone is not sufficient to distinguish FKBP12 from FKBP12.6 or perhaps yet another isoform, since the difference in mobility of FKBP12 and FKBP12.6 is small and can be accounted for in part by a change in one or two amino acids. 3 RyR1 and RyR3 can bind both FKBP12 and FKBP12.6. FKBP12 is the predominant isoform in the cytosol of rabbit skeletal muscle and dog heart muscle. Since FKBP12 and FKBP12.6 bind comparably to RyR1 (9) and RyR3, we can infer that the skeletal muscle RyRs (RyR1 and RyR3) bind the FKBP isoform that predominates in the cytosol, i.e. FKBP12. The heart RyR2 isolates with FKBP12.6 and not FKBP12 due to its specificity for binding FKBP12.6 (9).
We had previously reported that RyR1 can bind four equivalents of FKBP12 or FKBP12.6. This could mean that RyR1 has eight binding sites (four for FKBP12 and four for FKBP12.6), or alternatively the RyR receptor could have four binding sites that can bind either FKBP isoform. This issue has been resolved. There are only four FKBP binding sites, which can bind either isoform (Table IV).
In summary, we find that the ryanodine receptors from skeletal muscle in animals from each of the vertebrate classes are similar with regard to their association with FK-binding protein: 1) there are four binding sites/ryanodine receptor; 2) the general properties of binding and exchange are similar; and 3) the ryanodine receptor channel is modulated by the presence of FKBP. The association of FKBP with the RyR is conserved throughout vertebrate evolution and is consistent with a role for FKBP in EC coupling in skeletal muscle.