Overexpression of FK-506–Binding Protein 12.0 Modulates Excitation–Contraction Coupling in Adult Rabbit Ventricular Cardiomyocytes
The effect of the 12-kDa isoform of FK-506–binding protein (FKBP)12.0 on cardiac excitation–contraction coupling was studied in adult rabbit ventricular myocytes after transfection with a recombinant adenovirus coding for human FKBP12.0 (Ad-FKBP12.0). Western blots confirmed overexpression (by 2.6±0.4 fold, n=5). FKBP12.0 association with rabbit cardiac ryanodine receptor (RyR2) was not detected by immunoprecipitation. However, glutathione S-transferase pull-down experiments indicated FKBP12.0–RyR2 binding to proteins isolated from human and rabbit but not dog myocardium. Voltage-clamp experiments indicated no effects of FKBP12.0 overexpression on L-type Ca2+ current (ICa,L) or Ca2+ efflux rates via the Na+/Ca2+ exchanger. Ca2+ transient amplitude was also not significantly different. However, sarcoplasmic reticulum Ca2+ load was ≈25% higher in myocytes in the Ad-FKBP12.0 group. The reduced ability of ICa,L to initiate sarcoplasmic reticulum Ca2+ release was observed over a range of values of sarcoplasmic reticulum Ca2+ content, indicating that overexpression of FKBP12.0 reduces the sensitivity of RyR2 to Ca2+. Ca2+ spark morphology was measured in β-escin–permeabilized cardiomyocytes. Ca2+ spark amplitude and duration were significantly increased, whereas frequency was decreased in cells overexpressing FKBP12.0. These changes were accompanied by an increased sarcoplasmic reticulum Ca2+ content. In summary, the effects of FKBP12.0 overexpression on intact and permeabilized cells were similar to those of tetracaine, a drug known to reduce RyR2 Ca2+ sensitivity and distinctly different from the effects of overexpression of the FKBP12.6 isomer. In conclusion, FKBP12.0-RyR2 interaction can regulate the gain of excitation–contraction coupling.
Ca2+ release via the sarcoplasmic reticulum (SR) Ca2+ release channel (ryanodine receptor [RyR2]) is modulated by regulatory proteins interacting with both the cytoplasmic and luminal aspects of RyR2. One such cytosolic protein is FK-506–binding protein (FKBP), which binds to the RyR2 with a maximum stoichiometry of 4 FKBP:1 RyR2.1 Two members of the FKBP family are expressed within mammalian ventricular cardiomyocytes: FKBP12.0 (12.0 kDa) and FKBP12.6 (12.6 kDa). The cytosolic concentration of FKBP12.0 is almost 10-fold higher than FKBP12.6.1,2 In some species (eg, canine), the affinity of FKBP12.0 for RyR2 is ≈500 times lower than for FKBP12.6, hence binding is negligible under physiological conditions.1 In others (including rabbit and human), the affinity of FKBP12.0 for RyR2 is only ≈7 times lower than for FKBP12.61; therefore, it is conceivable that FKBP12.0–RyR2 interaction occurs in these species under physiological conditions.1 The consequences of FKBP12.0 binding on RyR2 function are unclear; currently, no pharmacological intervention can differentiate between the effects of FKBP12.0 and -12.6. FKBP12.0-null mice showed RyR2 dysfunction3 either resulting from the absence of RyR2-FKBP12.0 binding or as an indirect consequence of the accompanying developmental defects. FKBP12.6-null mice lack developmental defects, supporting the view that the FKBP12.0 isoform is indispensable for normal cardiac development. In the current study, adenoviral-mediated transfection of the human FKBP12.0 gene was used to overexpress FKBP12.0 in isolated rabbit ventricular cardiomyocytes over 48 hours. The choice of rabbit heart tissue was based on the similarities to human heart in terms of relative affinities of FKBP12.0 and FKBP12.6 for RyR2.
Materials and Methods
Recombinant Adenovirus Vector Construction
Full-length cDNA of the human FKBP12.0 gene was cloned by polymerase chain reaction (PCR) from human heart muscle–specific cDNA samples by the use of PCR primers that span the entire coding region of FKBP12.0 cDNA. This sequence was inserted downstream from a cytomegalovirus promoter into vector pACCMV · pLpA, and recombination with vector pJM17 was performed in HEK293 cells. The production, purification, and titration of adenovirus containing the FKBP12.0 gene (Ad-FKBP12.0) were performed according to standard procedures.4 Previous studies have used an adenovirus containing the human FKBP12.6 gene.2,5 At 100 multiplicities of infection, this Ad-FKBP12.6 vector caused an overexpression of FKBP12.6 of ≈6-fold normal values after 48 hours of incubation.
Ventricular Cardiomyocyte Isolation and Transfection
New Zealand White rabbits (2 to 2.5 kg) were euthanized by administration of an intravenous injection of 500 IU heparin, together with an overdose of sodium pentobarbitone (100 mg · kg−1). Hearts were removed in accordance with the United Kingdom Animals (Scientific Procedures) Act 1986 and conformed to the Guide for the Care and use of Laboratory Animals (NIH publication no. 85-23, revised 1985). Ventricular cardiomyocytes were then isolated as described previously.5 Adenoviral infection with a multiplicity of infection of 100 was performed to produce 2 populations of adenovirus-transfected cardiomyocytes: (1) overexpressing FKBP12.0 (Ad-FKBP12) and (2) expressing β-galactosidase as control (Ad-LacZ). Infected cardiomyocytes were subsequently cultured in supplemented medium M199 (Sigma) for 48 hours. Western blot analysis indicated the level of FKBP12.0 overexpression to be 2.6±0.4-fold (n=5, P<0.05) higher compared with nontransfected control cells.
Glutathione S-Transferase Pull-Down Assay
Prepared cell lysates and cardiac homogenates were incubated with 30 μg of glutathione S-transferase (GST)-FKBP12.0 or GST-FKBP12.6 immobilized on glutathione–sepharose for 5 hours at 4°C. Human FKBP12-pGEX-3X or FKBP12.6-pGEX-3X constructs (generously provided by Dr Wayne Chen, Department of Physiology and Biophysics, University of Calgary, Canada) were used to express and purify the GST-FKBP12.0 and GST-FKBP12.6 according to the instructions of the manufacturer (Amersham). After washing the glutathione–sepharose precipitates 3 times in solubilization medium, the GST-tagged protein and any associated proteins were competitively eluted with 10 mmol/L reduced glutathione in 50 mmol/L Tris–HCl, pH 8.0, for 15 minutes at room temperature. Incubation of samples with immobilized GST (Sigma) and elution with 60 μmol/L FK-506 or vehicle alone served as controls.
FKBP12.0 from cell lysates was analyzed by 15% SDS-PAGE. Eluates obtained during GST pull-down assay were size fractionated on 4% to 20% SDS-PAGE. The immunoblotting procedure was performed as described previously.2
Electrophysiological Measurements in Rabbit Cardiomyocytes
The isolated cardiomyocytes were superfused with a Krebs–Henseleit solution with additional 5 μmol/L tetrodotoxin, 0.1 mmol/L niflumic acid, and 5 mmol/L 4-aminopyridine at 19 to 20°C in a chamber mounted on the stage of an inverted microscope (see the online supplement at http://circres.ahajournals.org for details about solutions, including pipette solutions). Cytosolic loading of Fura-2 was achieved by incubating cardiomyocytes with 5 μmol/L Fura-2/acetoxymethyl ester at room temperature for 12 minutes.
Excitation–Contraction Coupling Protocol
Using an Axon Instruments Switch-clamp (2B), cardiomyocytes were held at −80 mV and the voltage was stepped to −40 mV (50 ms) to inactivate the remaining inward Na+ current, stepping to 0mV (150 ms) before returning to -80 mV. This protocol was repeated every 2s for 80s to achieve steady-state Ca2+ transients. SR Ca2+ content and Na+/Ca2+ exchanger (NCX) activity were then estimated by rapidly switching to 10 mmol/L caffeine to cause SR Ca2+ release. In the continued presence of caffeine, the SR is unable to reaccumulate Ca2+ and elimination of Ca2+ is mainly attributable to NCX. The time course of the decay of [Ca2+] and INCX represent rates of extrusion of Ca2+ from the cell predominately by NCX.6 These signals were fitted to exponential decays >80% of their amplitude.
The relationship between SR Ca2+ content and Ca2+ transient amplitude at low SR Ca2+ loads was investigated by manipulating SR Ca2+ content. Reduction of SR Ca2+ content was achieved by superfusing cardiomyocytes for set periods of time with thapsigargin (5 μmol/L).7 Increased SR Ca2+ content was achieved with a holding potential of −50 mV between voltage-clamp pulses.
Ca2+ Spark Measurements in Permeabilized Cardiomyocytes
Isolated rabbit cardiomyocytes were superfused with a mock intracellular solution and permeabilized using β-escin (Sigma) as detailed previously.5 Fluo-3 (10 μmol/L) in the perfusing solution was excited using a confocal microscope. Confocal line scan images were recorded using a Bio-Rad Radiance 2000 confocal system, further details of which have been published previously.5 All Ca2+ spark measurements were made within 2 to 3 minutes of cell permeabilization.
Data were expressed as means±SEM. For ionic currents, intracellular [Ca2+], and Ca2+ spark parameters, comparisons were performed by using the unpaired Student’s t test; otherwise, the paired Student’s t test was used. Differences were considered significant when P<0.05. ANOVA statistics with a Tukey post test were used in cases of multiple comparisons.
Overexpression of FKBP12.0 Within Isolated Adult Rabbit Ventricular Cardiomyocytes
Transgene expression at mRNA and protein levels were verified by means of RT-PCR and immunoblotting, respectively, both of which revealed a specific dose-dependent increase of exogenous FKBP12.0 (Figure 1A through 1C). To ensure effective overexpression of the transgene, cells were transfected at a multiplicity of infection of 100, which provides a moderate increase of the exogenous FKBP12.0 to 2.6±0.4 (n=5) fold above normal.
[Ca2+]i Measurements in Voltage-Clamped Rabbit Cardiomyocytes
No significant change in Ca2+ transient amplitude (231±15 versus 238±19 nmol/L; Ad-LacZ, n=33cells versus Ad-FKBP12, n=20cells) was observed in voltage-clamped rabbit cardiomyocytes (Figure 2A and 2C, i). Similarly, no significant difference between the Ca2+ decay of the Ca2+ transient was noted between the 2 populations of cells, suggesting no effect of Ad-FKBP12.0 overexpression on SERCA-mediated SR Ca2+ uptake (Figure 2C, ii). This was supported by measurements of SERCA-mediated Ca2+ uptake in oxalate equilibrated permeabilized cardiac myocytes after FKBP12.0 overexpression (data not shown). As illustrated in Figure 2A, ICa,L amplitude was monitored by incorporating a prepulse to −40 mV for 50 ms to inactivate the small amount of remaining inward INa (in the presence of 5 μmol/L tetrodotoxin). In addition to the measurements of ICa,L amplitude (Figure 2B and 2D, i), the time integral of ICa was calculated and converted to a Ca2+ influx (normalized to cell capacitance; Figure 2D, ii). The amplitude of ICa,L and the integral of the current were not different between the 2 experimental groups. This confirms the absence of an effect of FKBP12.0 overexpression on ICa in these experiments.
SR Ca2+ Content as Assessed by Rapid Application of Caffeine
Application of caffeine caused a rapid increase of [Ca2+]i as a result of SR Ca2+ release, the subsequent reduction of which results from extrusion of Ca2+ across the sarcolemma mainly via the NCX. This extrusion of Ca2+ via NCX generates a transient inward current, the amplitude and time course of which was monitored together with the Ca2+ transient (Figure 3A and 3B). As shown in Figure 3A (mean values shown in Figure 3B, i), the amplitude of the caffeine-induced Ca2+ release was significantly larger in Ad-FKBP12.0–transfected cardiomyocytes, suggesting an increased SR Ca2+ content (762±42 versus 910±74 nmol/L; Ad-LacZ [n=26 cells] versus Ad-FKBP12 [n=24 cells]). As described previously,6 the time integral of the current can be used as a measure of the amount of Ca2+ extruded by NCX during a caffeine application (an indicator of the SR Ca2+ content). As shown in Figure 3B, iii, the mean integral of the NCX-mediated inward current (INCX) in the Ad-FKBP12.0–transfected group was ≈25% higher than the control (Ad-LacZ) group (normalized to cell capacitance), supporting the conclusion that SR Ca2+ content was significantly higher in the FKBP12.0-overexpressing cardiomyocytes.
Sarcolemmal Ca2+ Efflux Rates in Rabbit Cardiomyocytes
Sarcolemmal flux rates can be estimated from the time course of the inward current decay and the corresponding decrease in [Ca2+] after rapid application of 10 mmol/L caffeine. As shown in Figure 3C, i and ii, both INCX and [Ca2+] decayed with a similar time course in Ad-FKBP12.0-transfected cardiomyocytes. These decays were fitted to a single exponential function, and mean rate constants were calculated. The rate constants for both parameters were not significantly different in the 2 experimental groups, suggesting that the rate of extrusion of Ca2+ via the NCX was not affected by FKBP12.0 overexpression.
Relationship Between SR Ca2+ Content and Ca2+ Transient Amplitude
To determine the relationship between the SR Ca2+ content and Ca2+ transient amplitude, measurements were made using thapsigargin to progressively decrease SR Ca2+ content. A holding potential of −50 mV was used to increase SR Ca2+ content. As shown in Figure 4, the amplitude of ICa,L was not altered by this range of experimental conditions. The plot of the INCX integral and Ca2+ transient amplitude for the Ad-LacZ group generated an approximately exponential relationship. The data from the Ad-FKBP12.0 group was shifted to the right and described an exponential curve that was significantly different from that required to fit the Ad-LacZ data (Figure 4). Analysis of ICa,L indicated that there were no significant changes in the amplitude or time course of this current in any of the datasets. Therefore, the exponential relationship described by the Ad-LacZ data represents the relationship between SR Ca2+ content and the ability of ICa,L to trigger Ca2+ release from the SR, ie, excitation–contraction (E–C) coupling “gain.”8 Overexpression of FKBP12.0 moved this curve to the right, indicating a reduced gain of E–C coupling. A similar but larger effect was observed in a separate group of Ad-LacZ cells superfused with 100 μmol/L tetracaine (a drug known to decrease the Ca2+ sensitivity of RyR2). As can be seen in Figure 4, this caused an increase in SR Ca2+ content (2.2±0.14 versus 3.78±0.33 coulombs/farad; Ad-LacZ [n=6 cells] versus Ad-LacZ±tetracaine [n=6 cells]) but no increase in Ca2+ transient amplitude (231±14 versus 232±53 nmol/L; n=6 cells). Data obtained from parallel studies using a previously characterized Ad-FKBP12.6 virus5 are shown in gray. As reported previously, FKBP12.6 overexpression significantly increased the Ca2+ amplitude and SR Ca2+ content,5 with no obvious shift from the normal relationship between SR Ca2+ content and Ca2+ transient amplitude. This suggests that, in contrast with FKBP12.0 overexpression, FKBP12.6 does not alter RyR2 Ca2+ sensitivity.
Interaction of FKBP12.0 and FKBP12.6 With RyR2
Previously published work based on immunoprecipitation experiments suggests interaction of RyR2 with FKBP12.6.9 The present study used similar immunoprecipitation techniques to confirm FKBP12.6 binding but could not detect specific interaction of FKBP12.0 with RyR2 (Figure I in the online data supplement). To further examine the interaction between FKBP12.0 and RyR2, GST pull-down assays were performed. GST fusion FKBP12.0 and FKBP12.6, immobilized on glutathione–sepharose beads, were allowed to interact with solubilized proteins of cell lysates and cardiac homogenates. The resulting complexes were then visualized by immunoblotting and silver staining. Figure 5A shows the extraction of rabbit RyR2 using GST-FKBP12.0 and GST-FKBP12.6 immobilized on glutathione–sepharose. RyR2, shown in lanes 3 and 4, was precipitated on the basis of its association with GST-FKBP12.0 and GST-FKBP12.6, respectively. The retained RyR2 was specifically bound to GST-FKBP12.0 because GST alone did not bind any RyR2 (Figure 5A, i and iii, lane 5). Furthermore, from 1 of the precipitates formed after interaction with GST-FKBP12.0, RyR2 was eluted with 60 μmol/L FK-506 (instead of 10 mmol/L glutathione) as a specific competitor (Figure 5A, i and iii, lane 7). To clarify whether FK-506 removed all of RyR2 from the immobilized FKBP12.0 (Figure 5C, i and ii, lane 1), GST-FKBP12.0, still bound to the glutathione–sepharose after application of FK-506, was eluted with reduced glutathione (Figure 5C, lane 2). The absence of RyR2 in this fraction, as revealed by silver staining (Figure 5C, i, lane 2) and immunoblotting (Figure 5C, ii, lane 2), confirmed that in FKBP12.0-immobilized complex, there was no RyR2 bound in a nonspecific manner, ie, independent of FK-506 treatment. Similarly, GST-FKBP12.0 displayed a specific FK-506–displaceable interaction with human RyR2, with efficiency similar to GST-FKBP12.6 (Figure 5B).
To exclude the possibility of detecting RyR1 instead of RyR2 when GST-FKBP12.0 is used as the ligand, the high-molecular-weight proteins from each species recognized by anti-RyR antibody and precipitated with GST-FKBP12.0 were subjected to MALDI-TOF (matrix-assisted laser desorption ionization–time of flight) mass spectrometry. Samples from both species were identified as the cardiac isoform of RyR (see supplemental Figures II through IV).
Canine RyR2 Interacts Exclusively With the FKBP12.6 Isoform
When subjected to the GST pull down, canine RyR2 was able to interact with GST-FKBP12.6 (Figure 5D, i and iii, lane 6) but not with GST-FKBP12.0 (Figure 5D, i and iii, lane 3), whereas rabbit RyR2, probed in parallel, interacted with both GST-FKBP12.0 (Figure 5C, i and iii, lane 8) and GST-FKBP12.6 (Figure 5D, i and iii, lane 9). Analysis of the same precipitates by Western blot revealed that all canine RyR2 molecules subjected to interaction with GST-FKBP12.0 remained unabsorbed, because it was completely recovered in the flow-through fraction (Figure 5D, iii, lane 1). In contrast, GST-FKBP12.6 provided efficient extraction of RyR2 from the crude homogenate (Figure 5D, iii, lanes 4 and 6).
Ca2+ Sparks and Caffeine-Induced Ca2+ Release in Permeabilized Cardiomyocytes Overexpressing FKBP12.0
An investigation of the direct effects of FKBP12.0 overexpression on Ca2+ sparks in intact rabbit cardiomyocytes is complicated by the rapid sarcolemmal extrusion of intracellular Ca2+ and reduction of resting [Ca2+] and SR Ca2+ content during the quiescent periods required for Ca2+ spark recording. For this reason, sarcolemmal fluxes were functionally bypassed by the acute permeabilization of the sarcolemma with β-escin and the cytosolic [Ca2+] maintained at 155 to 165 nmol/L. Ca2+ spark activity and SR Ca2+ content was monitored by the inclusion of 10 μmol/L Fluo-3 in the perfusing solution. SR Ca2+ content was assessed at the end of spark recording by rapid application of 10 mmol/L caffeine; the mean values are shown in Figure 6B. SR Ca2+ content was significantly higher in cells transfected with Ad-FKBP12. (627±94 versus 1901±338 nmol/L; Ad-LacZ [n=8 cells] versus Ad-FKBP12 [n=6 cells]). To quantify Ca2+ spark activity, the line scan images (Figure 6A) were analyzed using an automated spark detection program.10 The results collated from a number of cardiomyocytes are shown in Figure 6C. Increases in mean values of peak (1.96±0.03 versus 2.14±0.09 F/F0) and duration (40.4±1.5 versus 47.3±2.1 ms) were observed (Ad-LacZ: n=18 cells, 1742 events; versus Ad-FKBP12: n=10 cells, 626 events). Spark frequency was significantly reduced (0.064±0.005 versus 0.047±0.005 events · μm−1 · sec−1). Incubation of cells with rapamycin (3 μmol/L) during the 48 hours of quiescent culture inhibited the ability of Ad-FKBP12.0 to alter Ca2+ spark parameters (Figure 6D). To support the conclusion that these Ca2+ spark characteristics were attributable a decreased Ca2+ sensitivity of RyR2, Ad-LacZ transfected cells were equilibrated with 50 μmol/L tetracaine. Under these conditions the SR Ca2+ content in these cells increased to similar levels observed in FKBP12.0 overexpressing cells (720±116 versus 1640±320 nmol/L, Ad-LacZ, n=10cells versus Ad-LacZ+Tetracaine, n=10cells) as shown in Figure 7B. Ca2+ spark parameters were also affected in a similar manner with increases in Ca2+ spark peak (1.95±0.03 versus 2.05±0.04 F/F0), duration (36.7±0.9 versus 40.58±1.8 ms), Ad-LacZ n=25 cells 2971 events; Ad-LacZ versus Ad-LacZ±tetracaine n=11cells 1646 events) and a decrease in Ca2+ spark frequency (0.062±0.004 versus 0.047±0.004 events · μm−1 · sec−1). To compare the effects FKBP12.0 and tetracaine with previously published data on FKBP12.6 overexpression,5 the relative effects of these interventions on peak, duration, width, and frequency of Ca2+ sparks is shown in Figure 7D. The data are expressed relative to the corresponding Ad-LacZ (control) data. FKBP12.0 and tetracaine both caused an increase in peak and duration to ≈115% of control. As published previously, FKBP12.6 overexpression caused a decrease in peak and duration to ≈90% of control. All interventions decreased Ca2+ spark frequency.
Relationship Between Ca2+ Transient Amplitude and SR Ca2+ Content
This study is the first to show that overexpression of FKBP12.0 increases SR Ca2+ content (by ≈25%) but does not alter Ca2+ transient amplitude (Figures 2 and 3⇑). Previous studies have shown an approximately exponential relationship between SR Ca2+ content and Ca2+ transient amplitude.11 In the present study, the Ad-LacZ group demonstrated a similar exponential relationship (Figure 4), indicating that an ≈25% increase in SR Ca2+ content would be expected to increase Ca2+ transient amplitude by ≈100%. The absence of any significant effect on the Ca2+ transient amplitude suggests that the sensitivity of the Ca2+-induced Ca2+ release process is reduced by FKBP12.0 overexpression.11 In support of this conclusion, tetracaine, a drug known to reduce the Ca2+ sensitivity of RyR2, produced a similar effect on E–C coupling.
A previous study has shown that overexpressing FKBP12.6 increases Ca2+ transient amplitude and SR Ca2+ content.5 Similar effects shown in the present study indicated that FKBP12.6 overexpression does not significantly affect E–C coupling gain (Figure 4). In contrast, overexpression of FKBP12.0 reduced E–C coupling gain, suggesting that the 2 isoforms of FKBP may modulate RyR2 activity in different ways. This is reinforced by the contrasting effects of FKBP12.0 and 12.6 on Ca2+ sparks. A more detailed study comparing the 2 isoforms is required to clarify this point further.
Verification of RyR2-FKBP12.0 Interaction
Previously, Jeyakumar et al used an exchange binding assay with soluble [35S]FKBP12.0 and [35S]FKBP12.6 to demonstrate the ability of both FKBP isoforms to bind to RyR2.1 This work provided the first evidence that RyR2-FKBP12.0 interaction may occur in cardiac muscle in some vertebrates. The current study used immunoprecipitation techniques to demonstrate RyR2/FKBP12.6 association in rabbit heart cell lysates as previously reported in myocardium from rat and mice.9 However, this technique failed to show RyR2/FKBP12.0 association. A GST pull-down assay involving purified proteins subsequently showed that both GST-FKBP isoforms were capable of binding with rabbit and human RyR2 (Figure 5A and 5B). In the immunoblot analysis that followed GST pull down, the anti-RyR antibody used could not discriminate between the 2 RyR isoforms. Recent data have demonstrated the coexpression of 3 different RyR isoforms in both healthy and failing human hearts.12 In addition, possible contaminations of starting material with RyR1-containing cells/tissue may also occur. MALDI-TOF mass spectrometry confirmed that the high-molecular-weight proteins precipitated with GST-FKBP12.0 was the cardiac isoform (RyR2). The reason for the discrepancy between immunoprecipitation and GST pull-down assays with regard to FKBP12.0/12.6-RyR2 interactions is unclear. The dichotomy would indicate that the nature of the interaction of the 2 FKBP isoforms with RyR2 is different. The association of FKBP12.6 appears to be more stable and capable of surviving the preparation of the tissue for biochemical studies. Although the functional data strongly indicate that FKBP12.0 affects RyR2 function, and the GST pull down indicates that direct FKBP12.0-RyR2 interaction is possible, the nature of the association may not be sufficiently strong to survive tissue preparation for immunoprecipitation studies.
Specificity and Reliability of GST Biochemical Data
To validate the specificity of the GST pull-down assay, canine RyR2 was used as a negative control. When subjected to the GST pull-down assay, canine RyR2 was able to interact with GST-FKBP12.6 but not with GST-FKBP12.0 (Figure 5D). This result supplements previously published data suggesting that canine RyR2 has a higher specificity for FKBP12.6 over FKBP12.0 compared with other mammals.1
Effect of FKBP12.0 Overexpression on Ca2+ Sparks
Steady-state Ca2+ spark activity and SR Ca2+ load was measured in permeabilized cardiomyocytes at a standardized bathing [Ca2+] (155 to 165 nmol/L).13,14 Previous work has established that Ca2+ spark characteristics (amplitude, time course, and frequency) in permeabilized cells are indistinguishable from those observed in intact cells14 and regulated by known modulators of RyR2 activity (Ca2+–calmodulin and cyclic ADPribose).14 It is unlikely that there was any loss of FKBP from permeabilized preparations because previous work on isolated SR vesicles indicates minimal FKBP loss even after vigorous homogenization procedures.15 FKBP12.0 overexpression resulted in a significantly increased steady-state SR Ca2+ load in permeabilized cells. This confirmed the conclusion that FKBP12.0 overexpression acted directly on the SR and not via sarcolemmal-based processes. The relative effect of FKBP12.0 overexpression on SR content was greater than that seen in intact cells. This may reflect the higher cytoplasmic [Ca2+] and higher SR loads observed in the permeabilized cell experiments.
Ca2+ sparks were significantly increased in amplitude (≈110%) and duration (≈114%), but frequency was reduced (≈73%) compared with control. These effects were prevented by preincubation with rapamycin. The changes in Ca2+ spark characteristics are unlikely to be a simple consequence of the increased SR Ca2+ content because manipulations that increase SR Ca2+ content by stimulating the SR Ca2+ pump increase both Ca2+ spark size and frequency.13 In the present study, SR Ca2+ content was increased by inhibition of RyR2 using tetracaine (50 μmol/L); this increased Ca2+ spark amplitude (≈106%) and duration (≈111%) but reduced Ca2+ spark frequency (≈76%). The similarity of the changes in Ca2+ spark parameters produced by tetracaine to those of the Ad-FKBP12.0 group supports the conclusions from the E–C coupling studies (Figure 4), namely that FKBP12.0 overexpression acts to reduce the overall Ca2+ sensitivity of RyR2. This effect is distinct from the effect of FKBP12.6 overexpression reported earlier5 and summarized in Figure 7D, ie, a decrease in spark peak and duration to ≈90% of control. This reinforces the point that the two forms of FKBP have differing effects on RyR2.
Does FKBP12.0 Influence E–C Coupling Under Physiological Conditions?
In rabbit cardiomyocytes, FKBP12.0 protein expression is ≈10 times higher than that of FKBP12.6.2 The affinity of FKBP12.0 for rabbit RyR2 was estimated as ≈7 times lower than FKBP12.61; therefore, under physiological conditions, FKBP12.0 bound to RyR2 may be comparable to FKBP12.6. Under resting conditions, FKBP12.6 knockout mice appear to have a functional Ca2+-induced Ca2+ release mechanism.16 However, work on isolated RyR2 in lipid bilayers has shown that complete absence of FKBP12.6 and 12.0 from RyR2 causes dramatic increases in open probability and uncoupling of channels,17 behavior that is not compatible with normal E–C coupling. This paradox may be explained by the retention of the influence of FKBP12.0 on RyR2 in animals without the FKBP12.6 isoform.
This study provides the first evidence that FKBP12.0 influences RyR2 function in a distinct fashion from that observed with FKBP12.6. The differences are summarized as follows. (1) Unlike FKBP12.6, the association of FKBP12.0 to RyR2 could not be detected by immunoprecipitation but was evident in GST pull-down assays. (2) FKBP12.0 overexpression reduces the gain of E–C coupling, an effect not observed with FKBP12.6 overexpression. (3) FKBP12.0 overexpression increases Ca2+ spark amplitude, the opposite from that observed with FKBP12.6. The molecular actions of FKBP on RyR are not clear; changes in isolated RyR2 channel kinetics have been observed as a result of FKBP12.6 binding.18 FKBP12.6 is also thought to mediate the coupled gating observed in RyR2 channel clusters.19 These data suggest that the net effect of FKBP12.6 on a cluster of RyR2s in the dyad is therefore a combination of at least 2 effects. One possible explanation for the distinct effects of FKBP12.0 observed in this study is a different balance of effects on the single-channel kinetics and interchannel cooperatively. But there is, as yet, no evidence to suggest that FKBP12.0 and FKBP12.6 act on the same site on RyR2. Further studies of FKBP12.0– and FKBP12.6–RyR2 interactions would benefit from a conditional FKBP12.6/FKBP12.0 knockout to circumvent the developmental cardiac defects observed in FKBP12.0 knockout mice and allow the effects of FKBP12.6 and FKBP12.0 to be studied in isolation.
We thank Michael Kothe, Sandra Ott-Gebauer, Aileen Rankin, Jessica Spitalieri, and Anne Ward for expert technical assistance.
Sources of Funding
This study was supported by Deutsche Forschungsgemeinschaft grant HA 1233/7 (to T.S. and G.H.), Deutsche Forschungsgemeinschaft Graduiertenkolleg 335 (to G.H., H.K., D.Z.), the British Heart Foundation (C.M.L.), and the Wellcome Trust (S.K.).
↵*The first 3 authors contributed equally to this work.
Original received June 20, 2005; first resubmission received January 11, 2007; second resubmission received April 20, 2007; revised second resubmission received July 31, 2007; accepted August 31, 2007.
Prestle J, Janssen PML, Janseen AP, Zeitz O, Lehnart SE, Bruce L, Smith G, Hasenfuss G. Overexpression of FK-506 binding protein FKBP12.6 in cardiomyocytes reduces ryanodine receptor-mediated Ca2+ leak from sarcoplasmic reticulum and increases contractility. Circ Res. 2001; 88: 188–194.
Diaz ME, Trafford AW, O’Neill SC, Eisner DA. Measurement of sarcoplasmic reticulum Ca2+ content and sarcolemmal Ca2+ fluxes in isolated rat ventricular myocytes during spontaneous Ca2+ release. J Physiol. 1997; 501.1: 3–16.
Seidler T, Miller SL, Loughrey CM, Kania A, Burow A, Kettlewell S, Teucher N, Wagner S, Kogler H, Meyers MB, Hasenfuss G, Smith GL. Effects of adenovirus-mediated sorcin overexpression on excitation-contraction coupling in isolated rabbit cardiomyocytes. Circ Res. 2003; 93: 132–139.
Reiken S, Gaburjakova M, Guatimosim S, Gomez AM, D’Armiento J, Burkhoff D, Wang J, Vassort G, Lederer WJ, Marks AR. Protein kinase A phosphorylation of the cardiac calcium release channel (ryanodine receptor) in normal and failing hearts. Role of phosphatases and response to isoproterenol. J Biol Chem. 2003; 278: 444–453.
Eisner DA, Trafford AW, Diaz ME, Overend CL, O’Neill SC. The control of Ca2+ release from the cardiac sarcoplasmic reticulum regulation versus autoregulation. Cardiovasc Res. 1998; 38: 589–604.
Lukyanenko V, Gyorke S. Ca2+ sparks and Ca2+ waves in saponin-permeabilized rat ventricular myocytes. J Physiol. 1999; 521.3: 575–585.
Timerman AP, Onoue H, Xin HB, Barg S, Copello J, Wiederrecht G, Fleischer S. Selective binding of FKBP12.6 by the cardiac ryanodine receptor. J Biol Chem. 1996; 271: 20385–20391.
Wehrens XH, Lehnart SE, Huang F, Vest JA, Reiken SR, Mohler PJ, Sun J, Guatimosim S, Song LS, Rosemblit N, D’Armiento JM, Napolitano C, Memmi M, Priori SG, Lederer WJ, Marks AR. FKBP12.6 deficiency and defective calcium release channel (ryanodine receptor) function linked to exercise-induced sudden cardiac death. Cell. 2003; 113: 829–840.
Gomez AM, Valdivia HH, Cheng H, Lederer MR, Santana LF, Cannell MB, McCune SA, Altschuld RA, Lederer WJ. Defective excitation-contraction coupling in experimental cardiac hypertrophy and heart failure. Science. 1997; 276: 800–806.
Marx SO, Gaburjakova J, Gaburjakova M, Henrikson C, Ondrias K, Marks AR. Coupled gating between cardiac calcium release channels (ryanodine receptors). Circ Res. 2001; 88: 1151–1158.