Localization of the 12.6-kDa FK506-binding Protein (FKBP12.6) Binding Site to the NH2-terminal Domain of the Cardiac Ca2+ Release Channel (Ryanodine Receptor)*

The 12.6-kDa FK506-binding protein (FKBP12.6) interacts with the cardiac ryanodine receptor (RyR2) and modulates its channel function. However, the molecular basis of FKBP12.6-RyR2 interaction is poorly understood. To investigate the significance of the isoleucine-proline (residues 2427–2428) dipeptide epitope, which is thought to form an essential part of the FKBP12.6 binding site in RyR2, we generated single and double mutants, P2428Q, I2427E/P2428A, and P2428A/L2429E, expressed them in HEK293 cells, and assessed their ability to bind GST-FKBP12.6. None of these mutations abolished GST-FKBP12.6 binding, indicating that this isoleucine-proline motif is unlikely to form the core of the FKBP12.6 binding site in RyR2. To systematically define the molecular determinants of FKBP12.6 binding, we constructed a series of internal and NH2- and COOH-terminal deletion mutants of RyR2 and examined the effect of these deletions on GST-FKBP12.6 binding. These deletion analyses revealed that the first 305 NH2-terminal residues and COOH-terminal residues 1937–4967 are not essential for GST-FKBP12.6 binding, whereas multiple sequences within a large region between residues 305 and 1937 are required for GST-FKBP12.6 interaction. Furthermore, an NH2-terminal fragment containing the first 1937 residues is sufficient for GST-FKBP12.6 binding. Co-expression of overlapping NH2 and COOH-terminal fragments covering the entire sequence of RyR2 produced functional channels but did not restore GST-FKBP12.6 binding. These data suggest that FKBP12.6 binding is likely to be conformationdependent. Binding of FKBP12.6 to the NH2-terminal domain may play a role in stabilizing the conformation of this region.

Ryanodine receptors (RyRs) 1 are intracellular Ca 2ϩ channels located in the sarco(endo)plasmic reticulum of muscle and non-muscle cells. They govern the release of Ca 2ϩ from intracellular stores and play an essential role in various cellular processes including muscle contraction, fertilization, secretion, and apoptosis (1). These channels are regulated by a number of protein modulators, such as the 12-and 12.6-kDa FK506-binding proteins (FKBP12 and FKBP12.6) (2)(3)(4)(5)(6). FKBP12 is tightly associated with the type 1 ryanodine receptor (RyR1) predominantly expressed in skeletal muscle (7), whereas FKBP12.6 is selectively associated with the type 2 ryanodine receptor (RyR2) mainly expressed in cardiac muscle and in the brain (8,9). The type 3 ryanodine receptor (RyR3), which is expressed at relatively low levels in a variety of tissues, has also been shown to be capable of interacting with both FKBP12 and FKBP12.6 (10). The interactions between FKBP and RyR are believed to be involved in the stabilization of the full conductance state (11), channel gating (12), and modulating the sensitivity to Ca 2ϩ activation of RyR (13,14). Alterations in these interactions have been implicated in cardiomyopathy (15), cardiac hypertrophy (16), and heart failure (17,18).
Given the important roles of FKBPs in RyR regulation, a number of studies have focused on the structural basis of FKBP-RyR interactions, and some insights into the molecular determinants of FKBP binding have recently been revealed (10, 12, 14, 19 -21). Using the yeast two-hybrid technique, it has been shown that a 114-amino acid region containing residues 2497-2520 of RyR1 interacts with FKBP12. An analogous region in the inositol 1,4,5-trisphosphate receptor (IP3R) has also been shown to bind FKBP12 (19). This 114-amino acid fragment contains a valine-proline (residues 2461 and 2462) dipeptide epitope thought to be the FKBP12 binding motif in RyR1 (19), and its significance in FKBP12 binding to the intact full-length RyR1 protein has been confirmed. Mutations of valine 2461 to glycine, glutamate, or isoleucine eliminate FKBP12 binding to RyR1 (12). This valine-proline dipeptide motif is conserved in RyR3, where mutation of the corresponding valine to aspartate also diminishes FKBP12 binding (10). These observations demonstrate that the valine-proline motif is essential for FKBP12 interaction with RyR1 and RyR3.
In the corresponding location of RyR2, the valine-proline motif is replaced with isoleucine-proline. The role of this motif in FKBP12.6 binding to RyR2 has not yet been defined. Using the yeast two-hybrid method, Marx et al. (14) reported that a 136-amino acid fragment of the human RyR2 containing residues 2361-2496 interacted with FKBP12. 6 and that this interaction was inhibited by rapamycin, a drug known to dissociate FKBP12.6 from RyR2. Based on these observations, Marx et al. (14) proposed that the FKBP12.6 binding site is defined by isoleucine 2427 and proline 2428, analogous to the FKBP12 binding motif in IP3Rs, RyR1, and RyR3. However, contradictory to this proposal and the observations by Marx et al. (14), recent studies by Zissimopoulos and Lai (21) using the yeast two-hybrid and immunoprecipitation assays demonstrated that of the 10 overlapping fragments covering the entire sequence of human RyR2, none interacted with FKBP12.6. Using the same assay, they were, however, able to observe interactions between the cytoplasmic domain of the type 1 tumor growth factor-␤ receptor and FKBP12.6 and FKBP12 (21). Zissimopoulos and Lai (21) also found that a large COOHterminal fragment of human RyR2 encompassing all 10 transmembrane helices interacted with FKBP12.6, leading to the suggestion that the FKBP12.6 binding site is located at the COOH terminus of RyR2 (21). Therefore, the location of the FKBP12.6 binding site in RyR2 remains controversial and has yet to be defined.
In the present study, we investigated the significance of the isoleucine-proline motif in FKBP12.6 interaction with RyR2. Using a GST-FKBP12.6 pull-down assay, we found that mutations of this motif did not prevent GST-FKBP12.6 from binding to RyR2, in contrast to observations with RyR1 and RyR3. This prompted us to conduct systematic studies to define the FKBP12.6 binding site in RyR2. Through internal, NH 2 -terminal, and COOH-terminal deletion analysis, we showed that the first 1937 NH 2 -terminal residues are sufficient for GST-FKBP12.6 binding, whereas the last ϳ3000 COOH-terminal residues encompassing the proposed FKBP12.6 binding motif are neither required nor sufficient for GST-FKBP12.6 binding. In addition, we demonstrated that multiple regions within the NH 2 -terminal domain are required for GST-FKBP12.6 interaction. Furthermore, co-expression of overlapping NH 2 and COOH-terminal fragments led to the formation of functional RyR channels but did not reconstitute the GST-FKBP12.6 binding site. These observations suggest that FKBP12.6 binding may be conformation-dependent.

EXPERIMENTAL PROCEDURES
Materials-Restriction endonucleases and DNA-modifying enzymes were purchased from New England Biolabs Inc. The anti-c-Myc antibody was kindly provided by the Immunology Core Facilities at the Wadsworth Center of the New York State Department of Health. Soybean phosphatidylcholine was obtained from Avanti Polar Lipid. CHAPS and other reagents were purchased from Sigma.
Cell Culture and DNA Transfection-HEK293 cells were maintained in Dulbecco's modified Eagle's medium as described previously (22). HEK293 cells grown on 100-mm tissue culture dishes for 18 -20 h after subculture were transfected with 6 -12 g of wild type or mutant RyR cDNAs using Ca 2ϩ phosphate precipitation (23).
Construction of COOH-terminal Deletion Mutants of RyR2-The BsiWI (8864)-NotI (vector) fragment in the pcDNA3 plasmid was digested with HpaI and ligated with a linker containing a stop codon, 5Ј-CTAGCTAG-3Ј. The BsiWI (8864)-NotI (vector) fragment containing the inserted stop linker was then used to replace the corresponding fragment in the full-length RyR2 cDNA to yield the 1-4770 COOHterminal deletion mutant. Similarly, a stop linker, 5Ј-GTAGCTAC-3Ј was inserted into the Bsu36I (position 13,237) site after being cut and blunted by Klenow treatment to form the 1-4114 deletion mutant. A stop linker, 5Ј-TAGTGATCACTA-3Ј, was inserted into the SalI site (introduced at 11,821) after being cut and blunted by Klenow treatment to form the 1-3940 COOH-terminal deletion mutant. The 1-2958 COOH-terminal deletion mutant was constructed by inserting a stop linker, 5Ј-TAGTGATCACTA-3Ј, into the BsiwI (8864) site after being cut and blunted by Klenow treatment in the full-length RyR2 cDNA. An NheI (vector)-BsiWI (8864) fragment in pcDNA3 was digested with HpaI and ligated with a stop linker, 5Ј-CTAGCTAG-3Ј, yielding the 1-2531 COOH-terminal deletion mutant. Similarly, a stop linker, 5Ј-TAGTGATCACTA-3Ј, was inserted into the EcoRV (6439) site, forming the 1-2150 COOH-terminal deletion mutant. PCR was used to introduce a stop codon after residue 1937. The forward primer used is 5Ј-GATCCTCTGCAGTTCATGTCCCTC-3Ј. The reverse primer containing a stop codon followed by an EcoRV site is 5Ј-GATATCCTAGTCTT-GGAGTTTAGCTACAA-3Ј (1937-stop). The Bsu36I (4900)-EcoRV (introduced after the stop codon) PCR fragment was used to replace the Bsu36I (4900)-EcoRV (13,873) fragment in the full-length RyR2 cDNA to yield the 1-1937 COOH-terminal deletion mutant. In order to produce the 1-1636 COOH-terminal deletion mutant, a stop linker, 5Ј-GTAGCTAC-3Ј, was inserted into the Bsu36I (4900) site after being cut and blunted by Klenow treatment. To construct the 1-1072 and 1-531 COOH-terminal deletion mutants, a stop linker, 5Ј-AGTGATCACT-3Ј, was inserted into the XhoI (3216) and EcoRI (1590) sites, respectively, after being cut and blunted by Klenow treatment. All point mutations and deletions were confirmed by DNA sequencing.
GST-FKBP12. 6 Pull-down, Immunoprecipitation, and Immunoblotting Analyses-Cell lysates from 5-10 dishes (100 mm in diameter) of transfected HEK293 cells, prepared as described previously (26), were incubated with glutathione-Sepharose (30 l) that was prewashed with PBS (137 mM NaCl, 8 mM Na 2 HPO 4 , 1.5 mM KH 2 PO 4 , and 2.7 mM KCl) and prebound with 100 g of GST-FKBP12.6 or protein G-Sepharose (30 l) that was prewashed with PBS and prebound with 10 g of anti-c-Myc antibody at 4°C for 17-19 h. The glutathione-Sepharose and the protein G-Sepharose beads were washed with ice-cold lysis buffer (25 mM Tris, 50 mM Hepes (pH 7.4), 137 mM NaCl, 1% CHAPS, 0.5% soybean phosphatidylcholine, 2.5 mM dithiothreitol, 1 mM benzamidine, 2 g/ml leupeptin, 2 g/ml pepstatin A, 2 g/ml aprotinin, and 0.5 mM phenylmethylsulfonyl fluoride) three times, each time for 10 min. The proteins bound to the Sepharose beads were solubilized by the addition of 30 l of 2ϫ Laemmli's sample buffer (27) plus 5% ␤-mercaptoethanol and boiled at 100°C for 5 min. The solubilized proteins (20 l) were separated by 6% SDS-PAGE. The SDS-PAGE-resolved proteins were either stained with Coomassie Brilliant Blue or transferred to nitrocellulose membranes at 45 mV for 18 -20 h at 4°C in the presence of 0.01% SDS according to Towbin et al. (28). The nitrocellulose membrane was blocked for 30 min with PBS containing 0.5% Tween 20 and 5% skim milk powder. The blocked membrane was incubated with the anti-c-Myc antibody and washed three times each time for 15 min with PBS containing 0.5% Tween 20. The membrane was then incubated with the secondary anti-mouse IgG (H & L) antibodies conjugated with alkaline phosphatase for 30 -40 min. After washing three times each time for 15 min, the bound antibodies were visualized by the alkaline phosphatasemediated color reaction using nitro blue tetrazolium and 5-bromo-4chloro-3-indolyl phosphate as the substrates.
Ca 2ϩ Release Measurements in Transfected HEK293 Cells-Free cytosolic Ca 2ϩ concentration in transfected HEK293 cells was measured with the fluorescence Ca 2ϩ indicator dye fluo-3-AM as described previously (29).

Mutations in the Proposed FKBP12.6 Binding Motif Do Not
Abolish GST-FKBP12.6-RyR2 Interaction-It has been shown that mutating the proline at position 1401 in IP3R into glutamine or alanine abolishes FKBP12 binding (19). To determine the significance of the equivalent proline in RyR2 to FKBP12.6 binding, we mutated proline 2428 to glutamine (P2428Q). As shown in Fig. 1, both the wild type RyR2 (RyR2 (wt)) and the P2428Q mutant were pulled down by GST-FKBP12.6 (lanes 1 and 2) but not by GST (not shown), indicating that the P2428Q mutation does not abolish FKBP12.6 binding to RyR2, in contrast to observations with IP3R. It has also been shown that mutations of valine 2461 in RyR1 or valine 2327 in RyR3 abolish FKBP12 binding and that leucine 195 after the leucineproline motif in the tumor growth factor-␤ receptor is critical for FKBP12 binding (10,12,30). To assess the role of the corresponding RyR2 residues in FKBP12.6 interaction, we generated two double mutations, I2427E/P2428A and P2428A/ L2429E. Both mutants were pulled down by GST-FKBP12.6 ( Fig. 1, lanes 3 and 4) but not by GST (not shown). Like the WT RyR2-transfected cells, HEK293 cells transfected with mutants P2428Q, I2427E/P2428A, and P2428A/L2429E exhibited caffeine-induced Ca 2ϩ release (Fig. 1C). These data demonstrate that the isoleucine-proline motif is not essential for the interaction of RyR2 with FKBP12.6, although the corresponding motif is critical for FKBP12 binding to RyR1, RyR3, IP3Rs, and the tumor growth factor-␤ receptor.
GST-FKBP12. 6 Binding-deficient Mutants Are Capable of Forming Functional Heteromeric Channels-The effects of these internal deletions on RyR2 channel function were assessed by measuring the caffeine response of HEK293 cells transfected with each deletion mutant. All deletions completely abolished caffeine-induced Ca 2ϩ release in transfected HEK293 cells (Fig. 3, A, C, E, G, I, K, M, O, Q, S, and U), indicating that a large portion of the RyR2 sequence is required for activation by caffeine.
To assess their ability to form heteromeric channels, we co-expressed deletion mutants with an RyR2 mutant, RyR2 (I4827T), in HEK293 cells and determined their caffeine responses. We have shown previously that HEK293 cells trans-fected with the RyR2 (I4827T) mutant exhibited no caffeineinduced Ca 2ϩ release (24). Reasoning that if a deletion mutant is capable of forming a heteromeric channel with RyR2 (I4827T), co-transfection of HEK293 cells with a deletion mutant and RyR2 (I4827T) may rescue their caffeine response, since the mutants may complement each other's defects through the formation of heteromeric channel complexes. As shown in Fig. 3, HEK293 cells co-transfected with RyR2 (I4827T) and all internal deletion mutants except for mutant D2530 -4770 (Fig. 3B) displayed caffeine-induced Ca 2ϩ release. Hence, most deletion mutants retain the ability to form functional heteromeric channels, suggesting that these deletions did not grossly alter the channel structure.
The First 305 NH 2 -terminal Amino Acid Residues Are Not Essential for GST-FKBP12. 6 Binding and Caffeine Activation-To investigate the role of the NH 2 terminus in FKBP12.6 binding, we constructed a series of NH 2 -terminal deletion mutants and examined the effect of these deletions on GST-FKBP12.6 binding. As shown in Fig. 4, deletion of the first 305 NH 2 -terminal residues (D305) did not eliminate GST-FKBP12.6 binding (Fig. 4B), indicating that the first 305 NH 2terminal residues are not essential for FKBP12.6 binding. On the other hand, deletion of the first 784 (D784), 1072 (D1072), 1636 (D1636), or 2150 (D2150) NH 2 -terminal residues abolished GST-FKBP12.6 binding (Fig. 4, A and B). This indicates that a NH 2 -terminal region after residue 305 is critical for FKBP12.6 binding, which is consistent with the results of internal deletion studies (Fig. 2). These data also indicate that a Ͼ4000-amino acid COOH-terminal fragment of RyR2 encompassing the proposed FKBP12.6 binding motif is insufficient for GST-FKBP12.6 binding.
The effect of NH 2 -terminal deletions on channel function is shown in Fig. 5. HEK293 cells transfected with deletion mutant D305 remained sensitive to caffeine (Fig. 5A), indicating that the first 305 NH 2 -terminal amino acid residues are not essential for caffeine activation of RyR2. On the other hand, deletion of the first 784, 1072, 1636, or 2150 NH 2 -terminal amino acid residues abolished caffeine-induced Ca 2ϩ release in transfected HEK293 cells (Fig. 5, B, D, F, and H). Co-expression of these deletion mutants with RyR2 (I4827T) restored their response to caffeine activation (Fig. 5, C, E, G, I), indicating that these GST-FKBP12.6 binding-deficient NH 2 -terminal deletion mutants remain capable of forming functional heteromeric channels.

The NH 2 -terminal Fragment Including the First 1937 Amino Acid Residues Is Sufficient for GST-FKBP12.6 Binding and Is Capable of Restoring the Function of NH 2 -terminal Deletion
Mutants-The significance of the COOH-terminal region of RyR2 in FKBP12.6 interaction was assessed by constructing a series of COOH-terminal deletions and determining the effect of these deletions on GST-FKBP12.6 binding. Deletion of up to ϳ3000 COOH-terminal amino acid residues (Fig. 6, constructs 2-8) did not abolish GST-FKBP12.6 binding, whereas further deletion of the COOH terminus to residue 1636 led to a complete loss of GST-FKBP12.6 binding (Fig. 6B). These observations demonstrate that the COOH-terminal region starting at residue 1937 is not required for GST-FKBP12.6 binding and that the first 1937 NH 2 -terminal residues are sufficient for GST-FKBP12.6 binding.
These findings together with the observation that multiple deletions within the NH 2 -terminal region abolish GST-FKBP12.6 binding and channel function (Figs. 2 and 3) raise the possibility that the NH 2 -terminal region may constitute a unique domain structure. To test this possibility, we co-expressed this NH 2 -terminal fragment (amino acids  with the NH 2terminal deletion mutants, D1072, D1636, and D2150, in HEK293 cells (Fig. 7A) and examined the caffeine response of the transfected cells. As shown in Fig. 7B, caffeine-induced Ca 2ϩ release was observed in HEK293 cells co-transfected with the 1-1937 and D1072 fragments (Fig. 7Ba) and with the 1-1937 and D1636 fragments (Fig. 7Bb). On the other hand, cells cotransfected with 1-1937 and D2150 fragments did not exhibit caffeine-induced Ca 2ϩ release (Fig. 7Bc), which is consistent with the results of internal deletion studies in which residues between 1937-2150 are essential for caffeine activation (Fig. 3K). These results indicate that the NH 2 -terminal fragment, amino acids 1-1937, is capable of interacting functionally with the COOHterminal fragments, aa 1072-4967 and 1636 -4967.
Co-expression of Overlapping NH 2 and COOH-terminal Fragments Produces Functional RyR2 Channels but Does Not Restore GST-FKBP12. 6 Interaction-We next examined whether a GST-FKBP12.6 binding-deficient NH 2 -terminal fragment was able to form functional channels with an overlapping COOH-terminal fragment. We co-expressed the NH 2terminal fragment containing residues 1-1636 with the COOHterminal fragments D1072, D1636, and D2150 in HEK293 cells and examined caffeine-induced Ca 2ϩ release in transfected cells (Fig. 8, A and B). Co-expression of fragments 1-1636 and D1072, which exhibit no GST-FKBP12.6 binding, produced caffeine-sensitive Ca 2ϩ release channels in HEK293 cells (Fig.  8Ba). On the other hand, no significant caffeine-induced Ca 2ϩ release was observed in HEK293 cells co-transfected with fragments 1-1636 and D1636 and with fragments 1-1636 and D2150, which have little or no overlap, indicating that residues around 1636 are critical for channel function. We then reasoned that if co-expression of fragments 1-1636 and D1072 can form functional channels, it might also be able to restore the FKBP12.6 binding site. To this end, we carried out immunoprecipitation and pull-down assays using cell lysate from HEK293 cells co-transfected with fragments 1-1636 and D1072, D1636, or D2150. As shown in Fig. 8C, the anti-c-Myc antibody was able to precipitate all of the expressed fragments, whereas none of these fragments were pulled down by GST-FKBP12.6. These data indicate that overlapping fragments covering the entire sequence of RyR2 are not sufficient to form a stable FKBP12.6 binding site. DISCUSSION It is widely believed that the isoleucine-proline (residues 2427 and 2428) motif constitutes an essential part of the FKBP12.6 binding site in RyR2 (14,18,(31)(32)(33). This belief stems mainly from the observation that a small fragment of RyR2 encompassing this motif interacts with FKBP12.6 in the yeast two-hybrid assay and that the corresponding motif in RyR1, RyR3, and IP3R is critical for FKBP12 and FKBP12.6 binding (10,12,14,19,20). However, until now, the significance of the isoleucine-proline motif or regions encompassing this motif in FKBP12.6-RyR2 interaction has not been tested biochemically, and it has recently been questioned (21). In the present study, we directly examined the role of this isoleucineproline motif and regions containing this motif in the interaction of FKBP12.6 with the full-length RyR2. Site-directed mutagenesis and deletion analysis revealed that mutations of this motif do not abolish GST-FKBP12.6 binding to RyR2. In addition, the removal of large fragments containing the motif does not eliminate the ability of the mutant RyR2 to bind GST-FKBP12.6. These results provide the first biochemical evidence that the isoleucine-proline motif and regions flanking the motif are unlikely to form the core of the FKBP12.6 binding site in RyR2.
These findings differ from those observed with RyR1 and RyR3, in which mutations in the FKBP12 binding motif abolish FKBP12 and FKBP12.6 interactions (10,12), raising a question as to whether the FKBP12.6 binding site in RyR2 is structurally different from the FKBP12/12.6 binding site in RyR1 and RyR3. It has been shown that a 114-amino acid fragment containing residues 2497-2520 of RyR1 is sufficient for FKBP12 binding (19). However, we found that RyR2 fragments containing the corresponding region do not bind FKBP12.6 ( Fig. 4), whereas an NH 2 -terminal fragment of RyR2 contain- ing the first 1937 residues is sufficient for FKBP12.6 binding (Fig. 6). These observations suggest that the FKBP12.6 binding site in RyR2 and the FKBP12 binding site in RyR1 are located in different regions of RyR. However, this view is apparently inconsistent with the results of a recent study, which showed that the three-dimensional location of FKBP12.6 in RyR2 is similar to that of FKBP12 in RyR1 (34). Alternatively, it is possible that both the NH 2 -terminal and central regions are involved in binding with FKBP12 and FKBP12.6, but to different extents. FKBP12.6 binding may largely depend on the NH 2 -terminal region, whereas FKBP12 binding may mainly rely on the central region that encompasses the valine-proline motif. In addition, FKBP12.6 may bind more tightly to RyR2 than to RyR1 and RyR3. As a result, the central region may have less influence on FKBP12.6 binding to RyR2 than to RyR1 and RyR3. This may explain why mutations in the central region affect FKBP12.6 binding to RyR1 and RyR3, but not to RyR2. To further delineate the roles of these regions in FKBP12 and FKBP12.6 interactions, it will be of interest to determine whether the equivalent NH 2 -terminal fragment of RyR1 and RyR3 is capable of interacting with FKBP12.6 and FKBP12. It will also be important to determine biochemically whether truncated RyR1 and RyR3 proteins that lack the NH 2terminal region but contain the valine-proline motif are sufficient to bind FKBP12 and FKBP12.6.
Our observation that a truncated RyR2 lacking only the first 784 NH 2 -terminal amino acid residues fails to interact with GST-FKBP12.6 ( Fig. 4) also appears to be inconsistent with the results of early studies using the yeast two-hybrid method. In one study, a 136-amino acid fragment including residues 2361-2496 of RyR2 was found to be sufficient to interact with FKBP12.6 (14), whereas, in another study, FKBP12.6 was found to bind to a different region in RyR2, most likely at the COOH terminus (21). The reasons for these discrepancies are not clear. It is possible that there are multiple FKBP12.6 binding sites in RyR2 with different binding properties. The GST-FKBP12.6 pull-down assay used in the present study may not be sensitive enough to detect (probably low affinity or transient) FKBP12.6 binding to the central and COOH-terminal regions, as was detected by the yeast two-hybrid method. On the other hand, the yeast two-hybrid assay may fail to detect FKBP12.6 binding to the NH 2 -terminal region, because the RyR2 fragments used in this assay may not contain all of the necessary sequences to form a stable FKBP12.6 binding site as discussed below. In any event, based on our biochemical studies, it is likely that the NH 2 -terminal region of RyR2 represents the major binding site for FKBP12.6 if multiple binding sites exist.
An important finding of our present study is that multiple regions between residues 305 and 1937 of RyR2 are required for GST-FKBP12.6 binding. It is unlikely that all of these regions are involved in direct contact with FKBP12.6. Some of these regions are likely to be involved in the proper folding of the FKBP12.6 binding site. In other words, the binding of FKBP12.6 to RyR2 is likely to be conformation-dependent. We have found that co-expression of two large overlapping fragments covering the full sequence of RyR2 restores channel function but not GST-FKBP12.6 binding (Fig. 8). Similarly, Zissimopoulos and Lai (21) have shown that 10 overlapping fragments that cover the entire RyR2 sequence fail to bind to FKBP12.6 either individually or in multiple combinations. Thus, overlapping fragments, although they contain all of the linear sequence necessary for binding, are insufficient to reconstitute a stable FKBP12.6 binding site, supporting the idea that FKBP12.6 binding is dependent on the proper folding of the binding site.
This observation would in turn imply that binding of FKBP12.6 to the binding site might stabilize its conformation.
We have shown that an NH 2 -terminal fragment is able to form functional channels with overlapping COOH-terminal fragments (Fig. 7), suggesting that the NH 2 -terminal region contains one or more functional domains capable of interacting with the COOH-terminal portion of the channel protein. It is possible that binding of FKBP12.6 to the NH 2 -terminal region may stabilize intradomain interactions either within the NH 2terminal region or between the NH 2 -terminal and COOH-terminal regions. Further identification of the residues involved in direct interaction with FKBP12.6 should provide important insight into its roles in RyR function and regulation.