Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

FK506 is a potent immunosuppressant drug that has been used in the prophylaxis of allograft rejection in organ transplantation. The immunosuppressant actions of FK506 are mediated by its cytosolic receptors, FKBPs (FKBP12 or FKBP12.6) which are members of a family of the immunophilin proteins (Marks, 1996). The immunophilin proteins are highly conserved and widely distributed in virtually all cell types (Schreiber, 1991). The drug-immunophilin complex of FK506-FKBPs binds to calcineurin and inhibits its phosphatase activity resulting in the inhibition of T-lymphocyte activation.

Marks and colleagues (Jayaraman et al., 1992; Timerman et al., 1993; Brillantes et al., 1994; Marx et al., 1998) have demonstrated that FKBP12 is tightly associated with the ryanodine receptor (RyR) of skeletal muscle and modulates its functions by increasing RyR subunit cooperativity, decreasing open probability, and increasing RyR/Ca²⁺ release channel conductance. Another FK506 binding protein, FKBP12.6, is associated with the RyR of cardiac muscle sarcoplasmic reticulum (Sewell et al., 1994; Timerman et al., 1994; Lam et al., 1995; Kaftan et al., 1996). The actions of FKBP12.6 on cardiac RyR (RyR2) are similar to the actions of FKBP12 on skeletal RyR (RyR1). Whereas FK506 stimulates the binding of FKBPs to calcineurin, FK506 dissociates FKBPs from RyR. The dissociation of FKBPs from the RyR-FKBP complex by treatment with FK506 causes increased sensitivity of RyR to Ca²⁺, more frequent openings, and the appearance of subconductance states of the calcium release channel in isolated SR vesicles or the purified RyR reconstituted into planar lipid bilayers (Jayaraman et al., 1992; Brillantes et al., 1994; Lam et al., 1995; Kaftan et al., 1996). Thus, removal of FKBPs from the RyR-FKBP complex could cause the calcium release channels to become more sensitive to Ca²⁺, producing a positive inotropic effect; or "leaky", causing a negative inotropic effect by depleting the SR of Ca²⁺ (Marks, 1997). FK506 has been reported to increase the [Ca²⁺]_i transient in rat cardiac myocytes (McCall et al., 1996; duBell et al., 1997; Xiao et al., 1997), but effects of FK506 on [Ca²⁺]_i transients in other species have not been described. Since there are substantial species-dependent variations in Ca²⁺-handling components (Bassani et al., 1994; Sham et al., 1995; Su et al., 1999a,b), it is possible that the effects of FK506 on [Ca²⁺]_i transients are species-dependent. Therefore, we compared the effects of FK506 on [Ca²⁺]_i transients in adult rabbit and mouse ventricular myocytes.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Dissociation of Myocytes. Adult rabbit and mouse ventricular myocytes were isolated as described previously (Su et al., 1999a). Animals were anesthetized with pentobarbital sodium (65 mg/kg i.v. for rabbit, i.p. for mouse). Collagenase D (Roche Diagnostics, Indianapolis, IN) was used for mouse myocyte isolation and type 1 collagenase (Worthington Biochemicals, Freehold, NJ) for rabbit myocyte isolation. All experiments using single myocytes were performed at 25°C.

Measurement of [Ca²⁺]_i. The [Ca²⁺]_i was measured as previously described (Yao et al., 1998). Fluo-3-loaded myocytes were illuminated by a 485-nm excitation light through an epifluorescence attachment (510-nm dichroic mirror; Omega Opticals Inc., Brattleboro, VT) and a ×40 Fluor oil objective lens. The resulting fluorescence signals at 530 nm (DF30, Omega) were detected with a photomultiplier (SFX-2; Solamere Technology Group, Salt Lake City, UT). Fluo-3 fluorescence was transformed to [Ca²⁺]_i by a pseudoratio method (Cheng et al., 1993). [Ca²⁺]_i = K_d(F/F_o)/(K_d/[Ca²⁺]_irest + 1  (F/F_o)), where K_d is the dissociation constant for Fluo-3 (493 nM at 25°C), F the fluorescence intensity, F_o the intensity at rest, and [Ca²⁺]_irest the [Ca²⁺]_i at rest and assumed to be 100 and 150 nM for mouse and rabbit myocytes, respectively, based on previous calibrated measurements of [Ca²⁺]_i in these species in our laboratory (Yao et al., 1998a,b).

Measurements of Na⁺/Ca²⁺ Exchange and L-type Ca²⁺ Currents and Membrane Potential. The exchange current (I_Na/Ca) was measured by means of a whole-cell voltage clamp technique and a rapid solution switching technique (Su et al., 1999a). The voltage-clamped (40 mV) cell was initially superfused in a switcher microstream containing 140 mM NaCl. The outward I_Na/Ca was activated when the cell was abruptly exposed to an adjacent microstream of solution in which Li⁺ replaced Na⁺, using the solution-switching device. L-type Ca²⁺ current (I_Ca,L) and membrane potential were measured by whole-cell voltage clamp and current clamp techniques as described previously (Yao et al., 1998), except that current-clamp measurements of membrane potential were carried out with 10 mM EGTA in the pipette to buffer intracellular [Ca²⁺]_i.

Measurement of SR Calcium Content. SR Ca²⁺ content was estimated by measuring the integral (pA · s, pC) of the caffeine-induced inward I_Na/Ca normalized by cell capacitance (pF) (Yao et al., 1998). Following a train of steady-state conditioning pulses (twelve 200-ms pulses to 0 mV, 0.5 Hz), the voltage-clamped (80 mV) myocyte was abruptly immersed for 6 s in an adjacent switcher microstream of solution in which 10 mM caffeine was added to release SR Ca²⁺.

Preparation of FK506 Solutions. A stock solution (20 mM) of FK506 was prepared in ethanol and stored at 20°C. Aliquots of this stock solution were added to the perfusate immediately before use. The final ethanol concentration is equal to or less than 0.05%. In preliminary experiments, ethanol at 0.05% had no effects on calcium transients in either species. In the measurement of calcium transients, all myocytes were treated with FK506 for at least 10 min, a time required to reach the steady-state effect of FK506. In the measurements of action potentials, L-type calcium channel currents, and SR calcium content, myocytes were pre-exposed to FK506 for 10 to 14 min. In the measurement of Na/Ca exchange current, myocytes were treated for up to 15 min.

Data Analysis. All recordings were digitized online with a DigiData 1200 interface (Axon Instruments, Inc., Foster City, CA) and stored on disk. The digitized data were analyzed with pCLAMP6 (Axon Instrument, Inc.) and ORIGIN (Origin LabCorp., Northampton, MA). Results were expressed as means ± S.E.M., and statistical differences were determined by unpaired or paired t tests, as appropriate. Differences were considered statistically significant at p < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effects of FK506 on [Ca²⁺]_i Transients. In this study, we found that FK506 increased the diastolic and peak systolic [Ca²⁺]_i in mouse ventricular myocytes (Fig. 1A). The alteration of the peak of the [Ca²⁺]_i transient is similar to that previously observed in rat myocytes (McCall et al., 1996; Xiao et al., 1997). However, FK506 decreased the diastolic and peak systolic [Ca²⁺]_i in rabbit ventricular myocytes (Fig. 1B). The effects of FK506 on [Ca²⁺]_i transients in ventricular myocytes of both species were noted within 3 to 4 min, were stable by 10 min of exposure, were completely reversible upon washout of FK506, and were reproducible after a second application of FK506 in the same myocyte (data not shown). As shown in Table 1, there was no effect of FK506 on resting membrane potential or on action potential characteristics under these experimental conditions. Although duBell et al. (2000) have reported that FK506 reduces action potential duration (APD) in mouse myocytes at 32°C with 5 Hz stimulation frequency, electrophysiologic effects of FK506 do not appear to account for the alterations in [Ca²⁺]_i we have observed under our experimental conditions.

View larger version (28K): [in this window] [in a new window]   Fig. 1.   A, effects of FK506 on [Ca2+]i transients in mouse ventricular myocytes loaded with Fluo-3. Left panel: representative traces of the steady-state [Ca2+]i transients in a field-stimulated (0.5 Hz) single mouse ventricular myocyte in the absence (control) or the presence of 10 µM FK506. Fluo-3 fluorescence was transformed to [Ca2+]i by a pseudoratio method, as described under Materials and Methods. Myocytes were treated with 10 µM FK506 for 10 min. Right panel: average values of diastolic and peak systolic [Ca2+]i before and after exposure to FK506 (means ± S.E.M., n = 8; , p < 0.05, , p < 0.01 compared with control). B, effects of FK506 on [Ca2+]i transients in rabbit ventricular myocytes loaded with Fluo-3. Left panel: representative traces of the steady-state [Ca2+]i transients in a field-stimulated (0.5 Hz) single rabbit ventricular myocyte in the absence (control) or the presence of 10 µM FK506. The myocytes were treated with 10 µM FK506 for 10 min. Right panel: average values of diastolic and peak systolic [Ca2+]i before and after FK506 for 10 min (means ± S.E.M., n = 12; , p < 0.05, , p < 0.01 compared with control).

Effects of Cyclosporin A on [Ca²⁺]_i Transients. Calcineurin is also one of the major protein phosphatases in cardiac muscle. To examine whether calcineurin inhibition is responsible for the different effects of FK506 on [Ca²⁺]_i transients in rabbit and mouse ventricular myocytes, we observed the effects of cyclosporin A (CsA) on [Ca²⁺]_i transients in both species. Although CsA and FK506 are structurally unrelated and bind to distinct immunophilins, both drugs are potent inhibitors of calcineurin (Liu et al., 1991). Figure 2 shows that CsA had no effects on [Ca²⁺]_i transients in either rabbit or mouse ventricular myocytes. These results indicate that calcineurin inhibition is not responsible for the different effects of FK506 on [Ca²⁺]_i transients in rabbit and mouse ventricular myocytes.

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Effects of cyclosporin A on [Ca2+]i transients in rabbit and mouse ventricular myocytes. Fluo-3-loaded myocytes were field-stimulated at 0.5 Hz and treated with cyclosporin A (10 µM) for 10 min. Cyclosporin A had no effect on [Ca2+]i transients (shown by Fluo-3 fluorescence) in either rabbit (panel A) or mouse (panel B) ventricular myocytes. Similar results were observed in six rabbit myocytes and seven mouse myocytes.

Effects of Rapamycin on [Ca²⁺]_i Transients. Rapamycin is a new immunosuppressive drug that is a structural analog of FK506 and binds to the same immunophilins (FKBPs) (Schreiber, 1991; Marks, 1996). The complexes of rapamycin and FKBPs bind to calcineurin but do not change its phosphatase activity (Marks, 1996). Therefore, we also observed the effects of rapamycin on [Ca²⁺]_i transients in mouse and rabbit ventricular myocytes. As shown in Fig. 3, A and B, rapamycin enhanced the amplitude of [Ca²⁺]_i transients in mouse myocytes but reduced the amplitude of [Ca²⁺]_i transients in rabbit myocytes, effects that are similar to those of FK506.

View larger version (30K): [in this window] [in a new window]   Fig. 3.   A, effects of rapamycin on [Ca2+]i transients in mouse ventricular myocytes loaded with Fluo-3. Left panel: representative traces of the steady-state [Ca2+]i transients in a field-stimulated (0.5 Hz) single mouse ventricular myocyte in the absence (control) or the presence of 10 µM rapamycin. Myocytes were treated with 10 µM rapamycin for 10 min. Right panel: average values of diastolic and peak systolic [Ca2+]i before and after exposure to rapamycin (means ± S.E.M., n = 12, , p < 0.05, , p < 0.01 compared with control). B, effects of rapamycin on [Ca2+]i transients in rabbit ventricular myocytes loaded with Fluo-3. Left panel: representative traces of the steady-state [Ca2+]i transients in a field-stimulated (0.5 Hz) single rabbit ventricular myocyte in the absence (control) or presence of 10 µM rapamycin. The myocytes were treated with 10 µM rapamycin for 10 min. Right panel: average values of diastolic and peak systolic [Ca2+]i before and after exposure to rapamycin (means ± S.E.M., n = 5 cells, , p < 0.01 compared with control).

Effects of FK506 on L-type Calcium Currents and Na⁺/Ca²⁺ Exchange Currents. To examine the possible mechanisms for the different effects of FK506 on [Ca²⁺]_i transients in mouse and rabbit ventricular myocytes, we investigated the influence of FK506 on L-type calcium currents and Na⁺/Ca²⁺ exchange currents in both mouse and rabbit ventricular myocytes. As shown in Fig. 4, A and B, FK506 did not alter the I_Ca,L densities and current-voltage relationships in either rabbit or mouse ventricular myocytes. This is consistent with the findings of McCall et al. (1996) and Xiao et al. (1997) that FK506 did not affect the amplitude and kinetics of the L-type Ca²⁺ channel currents in rat ventricular myocytes. Figure 5, A and B, shows a similar lack of effect of FK506 on outward Na⁺/Ca²⁺ exchange currents in both rabbit and mouse ventricular myocytes. Note that the I_Na/Ca densities are greater in mouse myocytes than in rabbit myocytes, consistent with our previous observations (Su et al., 1999a).

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Effects of FK506 on L-type calcium current (ICa,L) in rabbit (A) and mouse (B) ventricular myocytes. ICa,L density was measured as described in Methods. FK506 (10 µM) did not alter the ICa,L densities and current-voltage relationships in either rabbit (n = 11-15, Panel A) or mouse (n = 8-13, Panel B) ventricular myocytes. Examples of current traces are shown in the inserts.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Effects of FK506 on Na/Ca exchange current (INa/Ca) in rabbit and mouse ventricular myocytes. A, example of effects on outward INa/Ca density (in rabbit myocyte) measured as described under Materials and Methods. The pipette [Na+] was 10 mM. B, average results expressed as means ± S.E.M. FK506 (10 µM) did not alter the INa/Ca densities in either rabbit (n = 11-15) or mouse (n = 8-13) ventricular myocytes. #, p < 0.05 compared with control mouse values.

SR Ca²⁺ Load. Because of the lack of effects of FK506 on L-type calcium currents and Na⁺/Ca²⁺ exchanger currents, we concluded that the different effects of FK506 on [Ca²⁺]_i transients in rabbit and mouse ventricular myocytes are not due to the variable effects of FK506 on calcium channel currents and Na⁺/Ca²⁺ exchanger currents. In addition to I_Ca,L and I_Na/Ca, the amplitude of [Ca²⁺]_i transients is also very sensitive to the Ca²⁺ loading status of SR. We therefore examined the SR Ca²⁺ content in rabbit and mouse ventricular myocytes when they were exposed to FK506. Figure 6, A and B, shows that FK506 treatment did not alter the SR Ca²⁺ content in mouse ventricular myocytes but reduced the SR Ca²⁺ content in rabbit ventricular myocytes. It is also interesting to note that rabbit ventricular myocytes have a significantly smaller SR Ca²⁺ content than mouse myocytes.

View larger version (19K): [in this window] [in a new window]   Fig. 6.   SR Ca2+ content in rabbit and mouse ventricular myocytes. A, example of currents elicited by rapid application of caffeine in a mouse myocyte treated with FK506 (10 µM) for 10 to 14 min. B, average SR Ca2+ content assessed by integration of currents produced by rapid application of caffeine (see Materials and Methods; pipette [Na+] was 15 mM for mouse and 5 mM for rabbit myocytes). Results are expressed as means ± S.E.M. Note that FK506 did not alter the SR Ca2+ content in mouse myocytes (n = 5-7) but reduced SR Ca2+ content in rabbit myocytes (n = 5-6). , p < 0.01 compared with control. ##, p < 0.01 compared with control mouse values.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In the present study, we have demonstrated that FK506 exerts different effects on [Ca²⁺]_i in rabbit and mouse ventricular myocytes. FK506 increased diastolic [Ca²⁺]_i and the amplitude of [Ca²⁺]_i transients in mouse myocytes but decreased diastolic [Ca²⁺]_i and the amplitude of [Ca²⁺]_i transients in rabbit myocytes (Fig. 1). Our findings of FK506 effects in mouse are similar to those reported by McCall et al. (1996), Xiao et al. (1997), and duBell et al. (1997), in rat myocytes.

duBell et al. (1997, 1998) have shown that FK506 can inhibit repolarizing K⁺ currents and have suggested that the resulting prolongation of APD is the cause of the increased [Ca²⁺]_i transient in rat myocytes. However, our results in rabbit ventricular myocytes demonstrate that the effects of FK506 on [Ca²⁺]_i transients are species-dependent and, under our experimental conditions, are not associated with marked alterations of APD. Furthermore, since FK506 reduces binding of FKBP12.6 to RyR, our finding of an FK506-induced decrease in the [Ca²⁺]_i transient in rabbit myocytes is consistent with the recent report by Prestle et al. (2001) that overexpression of FKBP12.6 (and presumably increased FKBP12.6-RyR interaction) increases fractional shortening and SR Ca²⁺ content in rabbit myocytes. CsA and FK506 are structurally unrelated and bind to distinct immunophilins. Both drugs are potent inhibitors of calcineurin (Liu et al., 1991). Rapamycin is a new immunosuppressive drug that is a structural analog of FK506 and binds to the same immunophilins (FKBPs), but does not inhibit calcineurin (Marks, 1996). Our results show that a concentration of CsA (10 mM) that significantly inhibits calcineurin (duBell et al., 1998) had no effects on [Ca²⁺]_i transients in either mouse or rabbit ventricular myocytes (Fig. 2), and that the effects of rapamycin on [Ca²⁺]_i transients in mouse and rabbit (Fig. 3) were similar to those of FK506. Thus, dissociation of FKBP from RyR, not calcineurin inhibition or inhibition of K⁺ currents, appears to be responsible for the effects of FK506 on [Ca²⁺]_i transients in mouse and rabbit myocytes.

To examine factors responsible for the species-dependent difference in effects of FK506, we investigated whether this species-dependent difference is due to possible different effects of FK506 on L-type Ca²⁺ channel, Na/Ca exchanger, membrane potential, or SR Ca²⁺ load. FK506 did not alter the I_Ca,L densities and current-voltage relationships in either rabbit or mouse ventricular myocytes (Fig. 4). FK506 also had no effects on Na⁺/Ca²⁺ exchange currents and membrane potential in either rabbit or mouse ventricular myocytes (Fig. 5; Table 1). From these results, we can exclude the possibility that the different effects of FK506 on [Ca²⁺]_i transients in mouse and rabbit myocytes are due to the variable direct effects of FK506 on sarcolemmal membrane currents.

As discussed, FK506-induced dissociation of FKBP12.6 from RyR2 in ventricular myocytes could produce an increased sensitivity of RyR to Ca²⁺, which could cause an increase in the [Ca²⁺]_i transient, or decreased RyR cooperativity and increased open probability. Both these effects could cause a decrease in the [Ca²⁺]_i transient, the latter by causing a "leak" of Ca²⁺ from the SR, which could deplete SR Ca²⁺ stores (Marks, 1997). Indeed Yano et al. (2000) have shown that defective interaction of FKBP12.6 with RyR in a canine model of pacing-induced heart failure leads to an abnormal SR Ca²⁺ leak, which may contribute to impaired function of the myocardium.

Previous work in our laboratory has shown that the [Na⁺]_i in rabbit myocytes is 4 to 5 mM (Yao et al., 1998b), whereas in mouse and rat myocytes, [Na⁺]_i is much higher, in the range of 12 to 15 mM (Yao et al., 1998). Since [Na⁺]_i is a major determinant of the rate at which [Ca²⁺]_i can be extruded by forward Na⁺/Ca²⁺ exchange, we postulate that in mouse (and rat) myocytes, an increased leak of Ca²⁺ from SR induced by FK506 does not cause Ca²⁺ depletion. Since Ca²⁺ is not extruded from the cell by Na⁺/Ca²⁺ exchange, the diastolic [Ca²⁺]_i is high and the Ca²⁺ is taken back up into the SR by the Ca²⁺-ATPase. In this situation, SR Ca²⁺ loading is maintained, and the effect of FK506 to increase sensitivity of RyR to Ca²⁺ could predominate, resulting in an increase in the [Ca²⁺]_i transient. On the other hand, in rabbit myocytes with a low [Na⁺]_i, Ca²⁺ leaking from SR could be extruded by Na⁺/Ca²⁺ exchange, resulting in a lower diastolic [Ca²⁺]_i and sufficient SR Ca²⁺ depletion to cause a decrease in the [Ca²⁺]_i transient despite an increase in sensitivity of RyR to Ca²⁺. The SR Ca²⁺ content measurements are consistent with this hypothesis, as is the effect of exposure to zero [Na⁺]_o on the change in resting [Ca²⁺]_i in rabbit myocytes induced by FK506. This idea is also consistent with the observations of Janiak et al. (1996). These investigators showed that exposure of guinea pig ventricular myocytes to 1.0 µM ryanodine, which induces an SR Ca²⁺ leak, caused a decrease in resting [Ca²⁺]_i that was dependent on function of the Na⁺/Ca²⁺ exchanger.

These findings may be relevant to the observation of Atkison et al. (1995) who reported that hypertrophy and heart failure were induced in five pediatric patients undergoing immunosuppression with FK506 after organ transplantation. Reduced SR Ca²⁺ content and release induced by FK506 in young myocardium could potentially induce hypertrophy and failure (Meyer et al., 1998). Although the [Na⁺]_i in human myocardium is not defined, rabbit myocardium is more similar to human myocardium than are rat and mouse. Thus, the effects of FK506 we note in rabbit myocytes may better predict effects in human myocardium. It is also clear that dissociation of FKBP from RyR induced by hyperphosphorylation of RyR would be predicted to have a negative inotropic effect in rabbits, and perhaps also in larger mammals. This would be consistent with the proposal by Marx et al. (2000) that protein kinase A-induced hyperphosphorylation of RyR may contribute to the progression of heart failure.