Actions of cADP-Ribose and Its Antagonists on Contraction in Guinea Pig Isolated Ventricular Myocytes
Influence of Temperature
Abstract Although it is becoming widely accepted that cADP-ribose (cADPR) can regulate calcium release from the endoplasmic reticulum in sea urchin eggs and in a variety of mammalian cell types, it remains controversial whether this substance might influence calcium release during excitation-contraction coupling in cardiac muscle. We have investigated possible actions of cADPR in intact cells isolated from guinea pig ventricle, paying particular attention to the possible influence of temperature. At 36°C, myocyte contraction was influenced by cytosolic application of cADPR in a concentration-dependent manner (showing an ≈30% increase in contraction with 5 μmol/L cADPR applied via a patch pipette in myocytes stimulated to fire action potentials at 1 Hz). Calcium transients measured with fura 2 were also increased by 5 μmol/L cADPR. Antagonists of cADPR reduced contraction at 36°C (by ≈35% with either 50 μmol/L 8-Br-cADPR or 5 μmol/L 8-amino-cADPR applied via the patch pipette). At room temperature (≈20°C to 24°C), no significant effects on contraction were detected with either cADPR or its antagonists. At 36°C, treatment of the cells with a mixture of 2 μmol/L ryanodine and 1 μmol/L thapsigargin to suppress function of the sarcoplasmic reticulum stores of calcium prevented the action of 5 μmol/L cADPR applied via a patch pipette. These observations are consistent with an action of cytosolic cADPR to enhance calcium-induced calcium release from the sarcoplasmic reticulum in guinea pig ventricular myocytes at 36°C. The observed influence of temperature under the conditions of our experiments is one factor that might help to account for failure to detect actions of cADPR and its analogues in some previous studies.
Evidence continues to accumulate supporting a role for cADP-ribose (cADPR) in regulating calcium release from ryanodine-sensitive stores in a wide variety of tissues.1 2 Since release of calcium from ryanodine-sensitive stores is crucial for contraction of cardiac muscle, the question arises whether cADPR might also play an important regulatory role in the heart. Evidence to date has led to conflicting conclusions. In studies of microsomes and calcium-release channels prepared from cardiac sarcoplasmic reticulum membranes, an early study provided support for regulation of calcium-release channels by cADPR,3 whereas later studies showed either no effect4 or effects that would be expected to be antagonized by ATP at concentrations normally expected to be present in cardiac muscle cytoplasm.5 In intact cardiac muscle cells, evidence from our laboratory has shown that cytosolic application of the cADPR antagonist 8-amino-cADPR6 reduces cardiac muscle contraction, whereas other careful experiments have failed to detect effects of either cADPR or 8-amino-cADPR on calcium transients.7
In an attempt to provide an explanation for these apparently conflicting observations, we have investigated actions of cADPR and its antagonists in intact cells both at room temperature and at 36°C. It appears that actions of cADPR analogues may indeed be influenced by temperature, leading to a possible explanation of the failure to detect actions of these compounds in some previous experiments and providing further evidence that cADPR may play an important regulatory role in the control of cardiac muscle contraction.
Materials and Methods
Myocytes were isolated from guinea pig ventricle using methods described previously.8 9 10 Extracellular solution for superfusion of isolated cells contained (mmol/L) NaCl 118.5, NaHCO3 14.5, KCl 4.2, KH2PO4 1.2, MgSO4 1.2, CaCl2 2.5, and glucose 11.1 (pH 7.4, 36°C, 95% O2/5% CO2).
Action potentials evoked by 2-ms current pulses (2 to 10 nA in amplitude) at a frequency of 1 Hz were recorded from the cells using patch-clamp techniques (Axopatch 200 or Axoclamp 2B). Both conventional whole-cell (ruptured-patch) and permeabilized-patch (amphotericin B in the pipette solution at 200 μg/mL) techniques were used. When cADPR or its antagonists were to be applied to the cytosol, these were included in the pipette solution. In the case of conventional whole-cell (ruptured-patch) recording, control records were taken in the first minute, when it was thought that dialysis from the pipette to the cytosol was minimal, and compared with records 5 to 7 minutes after rupture of the membrane, when access of the compound to the cytosol was thought to be well established.11 In permeabilized-patch experiments, it appeared that cADPR and its analogues in the pipette solution could not permeate the amphotericin B channels, and contractions were well maintained over a period of many minutes. Rupture of the membrane under the pipette led to changes in contraction (eg, an increase with 5 μmol/L cADPR or a decrease with 5 μmol/L 8-amino-cADPR) over a period of ≈3 minutes as the cADPR analogue in the pipette gained access to the cytosol. The two methods had advantages and disadvantages, with the permeabilized patch allowing more stable control recordings before rupture of the patch, although it appeared that this method was less effective for long-term stability of recording (for periods of more than ≈5 minutes after rupture), probably as a consequence of amphotericin B entering the cell and making the cell membrane “leaky.” The findings concerning effects of cADPR were similar for the two methods (see “Results”).
When ryanodine and thapsigargin were applied in the solution superfusing the cells, a tap close to the inflow of the bath was used to switch to the drug-containing solution. In choosing this drug combination, account was taken of the possibility that ryanodine alone reduces calcium release from stores but not calcium uptake and that the mixture with thapsigargin also inhibits the function of the Ca2+-ATPase of the sarcoplasmic reticulum.12
In one series of experiments, the pipette solution contained (mmol/L) KCl 150, MgCl2 5, K2ATP 1, and HEPES 3 (pH 7.2). In a second series of experiments, the pipette solution contained (mmol/L) KCl 140, NaCl 5, MgCl2 2, K2ATP 1, and HEPES 5 (pH 7.2). Contractions with the two solutions appeared approximately similar under the conditions of our experiments.
The solution was heated by passage through stainless steel tubes mounted in a heating block on its way to the recording chamber, where the cells were mounted on a coverslip. The temperature was monitored by a thermocouple positioned close to the cells. In the majority of experiments, this temperature was maintained at 36°C; in others, the power supply to the heating block was turned off, allowing the solution to cool to room temperature (≈20°C to 24°C). During this procedure, the pH of the extracellular solution remained approximately stable, although a slight acidification from pH 7.4 to 7.3 was observed. If there were any slight acidification of the cytosol under these conditions, contraction would be expected to be decreased, although this is likely to be outweighed by other factors leading to increased contraction at reduced temperature, and indeed, an increase in contraction was observed at room temperature (see “Results”). As an additional check of whether a reduction of pH, though small, might influence our observations, the effects of cADPR were investigated for one series of experiments with the extracellular pH adjusted to pH 7.2 (mean control shortening under these conditions before application of cADPR was 7.1±1.3% of resting length for contractions accompanying action potentials at 1 Hz; see below); the permeabilized-patch technique followed by membrane rupture as described above was used to apply 5 μmol/L cADPR from the pipette solution. This cytosolic application of cADPR increased cell shortening at 36°C by 31±9% (n=6 cells, P<.05), a change that was similar to that observed at an extracellular pH of 7.4 and 36°C (see “Results”); thus, the lack of effect of cADPR on contraction at room temperature reported below cannot be attributed to the small reduction in pH alone.
The possibility was also considered that temperature might influence diffusion of substances applied from the patch pipette into the cytosol. This was thought not to be a serious source of error since (1) the Q10 for diffusion-limited processes is no more than 1.3 and therefore any slowing of access would be expected to be small, and (2) when the measurement period for possible changes in contraction was extended for observations at room temperature from 5 to 8 minutes (using the conventional-patch method described above), no further changes were detected.
Measurement of Cell Contractions
Cell shortening accompanying action potentials stimulated at 1 Hz was measured from the video image of the cells viewed microscopically (Cohu 4710 or Pulnix 765E CCD camera) as described previously.6 13 Spatial resolution was 1 in 1024 pixels, and time resolution was 20 ms/point (acquisition board and software from Brian Reece Scientific). Cell shortening is quoted in the text as percent resting length.
Measurement of Fluorescence
Cells were loaded with fura 2-AM (5 μmol/L for 20 minutes) and excited by light from a Xenon arc lamp (PTI) via a quartz fiberoptic positioned close to the cell at a wavelength of either 340±7.5 or 380±7.5 nm, with collection of the emitted fluorescence at 510±20 nm (interference filters from Molecular Probes), as described previously.6 14 Fluorescent light was detected with a Photonic Science Isis III intensified CCD camera (time resolution of transients was 20 ms/point; acquisition and software were from Brian Reece Scientific). Fluorescence signals were very consistent in response to successive stimuli in the steady state (before and during application of drug), and switching of excitation wavelength was therefore carried out between stimuli (averaging eight transients at each excitation wavelength, collected in the following sequence: four at 340 nm, eight at 380 nm, and four at 340 nm). Transients in the absence of drug applied to the cytosol were stable over the period of the experiments (see Fig 6⇓).
Student’s paired t test was used for the majority of experiments in which measurements were made on single cells; unpaired t tests were used when observations on two groups of cells were to be made. A value of P<.05 was used to indicate statistical significance. Data are quoted in the text as mean±SEM.
Sources of Drugs
8-Amino-cADPR and fura 2-AM were obtained from Molecular Probes. cADPR and 8-Br-cADPR were obtained from Sigma, and thapsigargin and ryanodine were from Calbiochem.
Experiments at 36°C
In myocytes stimulated to fire action potentials at a frequency of 1 Hz, the accompanying contraction was well maintained over the time scale of our experiments: with conventional whole-cell patch recording, the mean change at 5 minutes after rupture of the cell membrane in 17 cells was a reduction of 3±4% (expressed as a percentage of cell shortening measured within 1 minute of rupture of the patch membrane, when dialysis was minimal; cells were maintained at 36°C; P>.05; mean cell shortening in these experiments was 8.1±0.8% of resting length). In contrast, inclusion of 5 μmol/L cADPR in the patch pipette solution caused a marked increase in contraction, as shown in Fig 1A⇓. Under these conditions, the effects appeared to develop over the period expected for dialysis of the cytosol (≈2 to 3 minutes11 ). In 13 cells exposed to cytosolic application of 5 μmol/L cADPR, contraction was increased by 25±7% (measured at 5 minutes, again expressed as a percentage of contraction measured within 1 minute; cells were maintained at 36°C; P<.05).
One potential disadvantage of this procedure, particularly with high concentrations of cADPR in the pipette, is that cADPR may enter the cell and exert effects on contraction even during the first minute, when control measurements were made. In another series of experiments, control records were obtained with the permeabilized-patch technique (amphotericin B in the pipette), and then the membrane was ruptured by applying reduced pressure to allow access to the cytosol of the cADPR, which had been included in the pipette solution (the cADPR being considered too large to permeate the amphotericin-induced pores before physical rupture of the membrane). Under these conditions, 5 μmol/L cADPR increased contraction by 30±5% (compared with contractions before rupture of the patch; n=9 cells, 36°C, P<.05; Fig 1B⇑).
The effects of cytosolically applied cADPR on myocyte contraction were dependent on the concentration in the patch pipette. Fig 2⇓ shows a log(concentration)-response curve in which concentrations of cADPR between 10 pmol/L and 10 μmol/L were applied to the cytosol by rupturing the patch membrane as described for the first method described above (open symbols). The second method using the permeabilized patch followed by membrane rupture was used for 10 nmol/L and for 5, 50, and 100 μmol/L cADPR (solid symbols). It can be seen that the effects were broadly similar with the two approaches and that the effects were maximal under these experimental conditions at a concentration of ≈5 μmol/L. The EC50 under these conditions appeared to be between 2 and 3 μmol/L. A curve was not fitted to the data, since a simple model relating drug binding and effect seemed not to be appropriate in view of evidence that the effect of cytosolically applied cADPR appeared to be biphasic: at 10 nmol/L cADPR there was a small but statistically significant reduction in contraction found with both the conventional whole-cell patch and the permeabilized-patch techniques followed by rupture (reduction of 10±2% [n=6 for conventional method] and of 11±3% [n=7 for permeabilized method], P<.05 in both cases). When the pipette concentration was further reduced to 10 pmol/L, there were no significant changes in contraction.
If the actions of cADPR were associated with an increased sensitivity of calcium-induced calcium release (CICR) from the sarcoplasmic reticulum, the actions would be expected to be suppressed when the function of sarcoplasmic reticulum was inhibited. This possibility was tested using a mixture of 2 μmol/L ryanodine (to interfere with release channels) and 1 μmol/L thapsigargin (to inhibit the Ca2+-ATPase of the sarcoplasmic reticulum). Under these conditions, mean cell shortening was 6.4±0.5% of resting length. Fig 3⇓ shows that when 5 μmol/L cADPR was applied via the patch pipette to cells pretreated with the ryanodine/thapsigargin mixture, there was no significant increase in contraction in contrast with the increase reported above in the absence of ryanodine and thapsigargin (permeabilized-patch method). The observations are therefore consistent with an enhancing effect of cADPR on CICR from the sarcoplasmic reticulum.
It is interesting to note that the small but statistically significant reduction of contraction with 10 nmol/L cADPR was also prevented by exposure of the cells to 2 μmol/L ryanodine and 1 μmol/L thapsigargin (increase of 7±8%, n=7 cells, permeabilized-patch method, P>.05).
8-Br-cADPR has been reported to be a competitive antagonist of cADPR in the sea urchin egg preparation.15 The effects of this compound were tested in a series of experiments at two concentrations of 8-Br-cADPR in the pipette solution (conventional whole-cell patch method). When 5 μmol/L 8-Br-cADPR was applied to the cytosol, contraction was reduced by 23±4% (n=7 cells, P<.05). A larger reduction of contraction of 37±8% (n=7 cells, P<.05) was observed when 50 μmol/L 8-Br-cADPR was applied to the cytosol (P<.05 for a comparison of the two sample means) as shown in Fig 4⇓. The observations are therefore similar to those previously reported for another cADPR antagonist, 8-amino-cADPR (applied to the cytosol via a theta-glass microelectrode6 ). To test for the effects of 8-amino-cADPR under the conditions of the present experiments, a further series of observations was made with 5 μmol/L 8-amino-cADPR applied via the patch pipette; again, contraction was found to be reduced by 35±7% (n=8 cells, P<.05) in the presence of this antagonist. The observations made when using these cADPR analogues are therefore consistent with a suppression by the two antagonists of an enhancing influence of endogenous cADPR on CICR.
If 8-Br-cADPR can inhibit the influence of endogenous cADPR, the question arises whether it might prevent the actions of exogenous cADPR applied via the patch pipette. This was tested in experiments in which 5 μmol/L 8-Br-cADPR and 5 μmol/L cADPR were both added to the pipette solution. Under these conditions, there was no significant change in the amplitude of contraction (decrease of 5±4%, n=6 cells, conventional whole-cell method, 36°C). This antagonistic effect of 5 μmol/L 8-Br-cADPR could be overcome by increasing the concentration of cADPR in the patch pipette to 50 μmol/L in the continued presence of 5 μmol/L 8-Br-cADPR (Fig 4⇑). This was the case both when a conventional patch method was used, as described above (an increase of 20±5%, n=9 cells, P<.05), and in experiments in which the second method was used to allow control measurements with permeabilized patches before rupture of the membrane (an increase of 28±11%, n=6 cells, P<.05).
Observations at Room Temperature
The possible influence of temperature on the actions of cADPR and its analogues was investigated in another series of experiments in which the heating system normally used to warm the solution before it entered the recording chamber was switched off. Under these conditions, contraction accompanying action potentials at 1 Hz was again well maintained (reduction of 3±5%, n=10 cells, P>.05; changes were measured at 5 minutes and expressed as a percentage of contraction measured within 1 minute of rupture of the patch membrane). Observations with 5 μmol/L cADPR, 50 μmol/L 8-Br-cADPR, or 5 μmol/L 8-amino-cADPR in the patch pipette are shown in Fig 5⇓. It can be seen that none of the cADPR analogues caused significant changes in contraction under these conditions (for 5 μmol/L cADPR, there was a reduction of 13±6% [n=13 cells, P>.05]; for 50 μmol/L 8-Br-cADPR, there was a decrease of 5±6% [n=7 cells, P>.05]; and for 5 μmol/L 8-amino-cADPR, there was an increase of 11±9% [n=5 cells, P>.05]). In contrast, the changes seen at 36°C are shown in the same bar graph (mean values for cADPR and 8-Br-cADPR are reported above; for 5 μmol/L 8-amino-cADPR, there was a reduction of 35±7% [n=8 cells, P<.05]). It is clear that temperature exerted a profound influence on the actions of cADPR and its analogues.
One possibility that deserves consideration is that since reduction of temperature from 36°C to room temperature leads to an increase in contraction, the observed lack of effect of cADPR at room temperature might simply reflect a marked increase in basal inotropic state and magnitude of the calcium transient under these conditions. To investigate this possibility, we carried out a series of experiments with extracellular calcium concentration reduced from 2.5 to 1.8 mmol/L. Under these conditions, myocyte contraction accompanying action potentials at 1 Hz was reduced from 9.7±0.7% at 2.5 mmol/L to 8.7±0.4% at 1.8 mmol/L calcium; the magnitude of this reduced contraction at room temperature and 1.8 mmol/L calcium was therefore similar to that at 36°C and 2.5 mmol/L calcium (8.1±0.8%). In myocytes maintained at room temperature and 1.8 mmol/L calcium, no significant increase in contraction was observed when 5 μmol/L cADPR was applied to the cytosol via patch electrodes (the mean change was a reduction of 2±5%; n=11 cells, P>.05). However, at 36°C and 1.8 mmol/L calcium, application of 5 μmol/L cADPR via the patch electrode did cause an increase in contraction of 29±10% (n=6 cells, P<.05), similar to that reported above for cells at 36°C and 2.5 mmol/L calcium. It seems unlikely therefore that an increased inotropic state at room temperature can be the sole reason for the lack of significant increase in contraction by cytosolically applied cADPR under these conditions.
Effects of cADPR on Calcium Transients
The reduction in contraction caused by cytosolically applied 8-amino-cADPR has previously been shown to be associated with a corresponding decrease in the calcium transient accompanying the action potential in cardiac ventricular cells.6 In another series of experiments, we have investigated whether the increase in contraction caused by 5 μmol/L cADPR is also associated with changes in calcium transients measured with fura 2. The experiments were carried out at 36°C. Fig 6A⇓ shows that the calcium transient was well maintained 5 minutes after rupture of the patch membrane in the absence of cADPR in the pipette solution, whereas when 5 μmol/L cADPR was applied to the cytosol, there was a substantial increase in the recorded calcium transient (Fig 6B⇓). Mean changes are shown in the bar graph in Fig 6C⇓.
The main findings reported in the present study are that the contraction of cardiac ventricular myocytes was significantly increased by cADPR and decreased by its antagonists at 36°C but that possible effects of these compounds were not detected at room temperature under the conditions of our experiments. The temperature sensitivity of the actions of cADPR analogues may be one important factor leading to reports of the lack of effects of these analogues when studied at room temperature, although the possible influence of other additional factors cannot be ruled out.
The observations at 36°C with the antagonists 8-Br-cADPR and 8-amino-cADPR applied via a patch pipette are similar to the reductions in myocyte contraction reported for 8-amino-cADPR applied from a theta-glass microelectrode.6 However, lower concentrations of 8-amino-cADPR were required in the present experiments when this antagonist was applied via a patch pipette rather than from a “sharp” theta-glass electrode, as might be expected from the improved access of drugs to the cytoplasm under these conditions (since the diameter of the open end of a patch pipette is expected to be much larger than the tip of a theta-glass electrode).
In the case of 8-amino-cADPR, we have suggested that the reduction in myocyte contraction may arise from competitive inhibition of endogenous cADPR, which normally acts to sensitize CICR to calcium.6 We have said that a direct effect of 8-amino-cADPR on the CICR ryanodine receptor could not be excluded, although we have pointed out that such an effect could not result from simple channel block, since the calcium-releasing actions of caffeine were not prevented by this compound. The observation that 8-Br-cADPR, which is another substance found to be a competitive antagonist of cADPR in the sea urchin preparation,15 also reduces myocyte contraction provides further support for our hypothesis of antagonism of endogenous cADPR in myocytes. The ability to overcome the actions of 8-Br-cADPR by increasing the concentration of cADPR is also consistent with competitive antagonism rather than block of release channels by the antagonist. Furthermore, there was a close parallel between the observations that higher concentrations of 8-Br-cADPR than of 8-amino-cADPR are required to competitively antagonize the actions of cADPR on calcium release from sea urchin egg microsomes15 and that reduction of myocyte contraction shows a similar difference in sensitivity to these two compounds.
The observations reported in the present study are the first demonstration that cADPR, which is known to act as an agonist in the sea urchin egg preparation, can increase contraction in intact cardiac muscle cells. As with the reductions of contraction observed with the antagonist 8-amino-cADPR,6 the increase in contraction with 5 μmol/L of the agonist cADPR was shown to be prevented by exposure to drugs that interfere with the function of the sarcoplasmic reticulum. The observations therefore support a role of cADPR to increase calcium release from the sarcoplasmic reticulum, perhaps by increasing the calcium sensitivity of CICR. The finding that 5 μmol/L cADPR applied to the cytosol also increased calcium transients measured with fura 2 provides further support for this hypothesis.
If the hypothesis of antagonism between either 8-Br-cADPR or 8-amino-cADPR and endogenous cADPR is correct, temperature sensitivity might arise from an influence of temperature on the enzymes (for example ADP-ribosyl cyclase and cADPR hydrolase) that regulate the endogenous levels of cADPR and that are known to be present in cardiac muscle.16 17 Thus, at room temperature there may be reduced endogenous levels of cADPR in the cytoplasm, so that this mechanism for increasing the calcium sensitivity of CICR might be suppressed, leading to reduced effects of the antagonists. Another possibility to account for the temperature sensitivity of the antagonists is that the binding of antagonists to their postulated inhibitory sites might be temperature sensitive.
In this context, it is interesting that the actions of the cADPR agonist also show a high temperature sensitivity. In the sea urchin egg system, it has been suggested that cADPR may act via a binding protein to influence the ryanodine receptor and that the actions may be modulated by calmodulin.18 There is also intriguing evidence that the FK-506 binding protein (FKBP), which has been associated with the actions of cADPR,19 may also influence cardiac muscle contraction.20 Recent experiments have shown that the exchange reaction between soluble FKBP-12.6 and the FKBP-ryanodine receptor complex is temperature sensitive.21 Temperature sensitivity either of the binding of cADPR to the postulated binding proteins or of the interaction between the binding protein complex with the CICR channel (ryanodine receptor) might account for the temperature sensitivity of the agonist cADPR. If such mechanisms were to occur, they might also contribute to the temperature sensitivity of the antagonists, 8-amino-cADPR and 8-Br-cADPR.
In summary, the observations reported in the present study provide further support for the hypothesis that cADPR might function as an important cytosolic regulator of cardiac muscle contraction. The temperature sensitivity of these effects of cADPR and its antagonists may provide important clues concerning mechanism of action and may help to explain previous negative reports concerning cADPR actions.
This study was supported by The Wellcome Trust and by The Japan Heart Foundation. We thank Prof H.C. Lee, University of Minnesota, for helpful discussions.
- Received June 4, 1997.
- Accepted September 17, 1997.
- © 1997 American Heart Association, Inc.
Galione A, Summerhill RS; Sorrentino V, ed. Ryanodine Receptors. Boca Raton, Fla: CRC Press; 1996:52-70.
Sitsapesan R, McGarry SJ, Williams AJ. Cyclic ADP-ribose competes with ATP for the adenine nucleotide binding site on the cardiac ryanodine receptor Ca2+-release channel. Circ Res.. 1994;75:596-600.
Guo X, Laflamme MA, Becker PL. Cyclic ADP-ribose does not regulate sarcoplasmic reticulum Ca2+ release in intact cardiac myocytes. Circ Res.. 1996;79:147-151.
Eilers J, Schneggenburger R, Konnerth A; Sakmann B, Neher E, eds. Single-Channel Recording. Second ed. New York, NY: Plenum Publishing Corp; 1995:213-229.
Rakovic S, Annetts C, Flaxman P, Cui Y, Terrar DA. Measurement of cytosolic calcium transients in myocytes isolated from guinea-pig ventricle. J Physiol (Lond).. 1995;487:7P–8P.
Rusinko N, Lee HC. Widespread occurrence in animal tissues of an enzyme catalyzing the conversion of NAD+ into a cyclic metabolite with intracellular Ca2+-mobilizing activity. J Biol Chem.. 1989;264:11725-11731.
Noguchi N, Takasawa S, Nata N, Tohgo A, Kato I, Ikehata F, Yonekura H, Okamoto H. Cyclic ADP-ribose binds to FK506-binding protein 12.6 to release Ca2+ from islet microsomes. J Biol Chem.. 1997;272:3133-3136.
McCall E, Li L, Satoh H, Shannon TR, Blatter LA, Bers DM. Effects of FK-506 on contraction and Ca2+ transients in rat cardiac myocytes. Circ Res.. 1996;79:1110-1121.
Timerman AP, Wiederrecht G, Marcy A, Fleischer S. Characterization of an exchange reaction between soluble FKBP-12 and the FKBP ryanodine receptor complex. J Biol Chem.. 1995;270:2451-2459.