Troponin I phosphorylation enhances crossbridge kinetics during β-adrenergic stimulation in rat cardiac tissue


  • Lynne Turnbull,

    Corresponding author
    1. Biomedical Systems Research Group, Division of Information and Communication Sciences, Macquarie University, NSW 2109, Australia
    • Corresponding author L. Turnbull: Veterans Affairs Medical Center, Cardiology Division, 111C, 4150 Clement Street, San Francisco, CA 94121, USA. Email:

    Search for more papers by this author
  • Joseph F. Y. Hoh,

    1. Department of Physiology and The Sydney Institute for Biomedical Research, University of Sydney, NSW 2006, Australia
    Search for more papers by this author
  • Russell I. Ludowyke,

    1. Centre for Immunology, St Vincent's Hospital, Sydney, NSW 2010, Australia
    Search for more papers by this author
  • Gunther H. Rossmanith

    1. Biomedical Systems Research Group, Division of Information and Communication Sciences, Macquarie University, NSW 2109, Australia
    Search for more papers by this author

  • Author's present address
    L. Turnbull: University of California San Francisco, Veterans Affairs Medical Center, Cardiology Division, 111C, 4150 Clement Street, San Francisco, CA 94121, USA.


Inotropic agents that increase the intracellular levels of cAMP have been shown to increase crossbridge turnover kinetics in intact rat ventricular muscle, as measured by the parameter fmin (the frequency at which dynamic stiffness is minimum). These agents are also known to increase the level of phosphorylation of two candidate myofibrillar proteins: myosin binding protein C (MyBPC) and Troponin I (TnI), but have no effect on myosin light chain 2 phosphorylation (MyLC2). The aim of this study was to investigate whether the phosphorylation of TnI and/or MyBPC was responsible for the increase in crossbridge cycling kinetics (as captured by fmin) seen with the elevation of cAMP within cardiac tissue. Using barium-activated intact rat papillary muscle, we investigated the actions of isobutylmethylxanthine (IBMX), an inhibitor of cAMP-dependent phosphatase, which simulates the action of β-adrenergic agents, and the chemical phosphatase 2,3-butanedione monoxime (BDM), which has been shown to dephosphorylate a number of contractile proteins. The presence of 0.6 mm IBMX approximately doubled the fmin value of intact rat papillary muscle. This action was unaffected by the addition of BDM. In the presence of IBMX and BDM, the level of phosphorylation of MyBPC was unchanged, that of MyLC2 was reduced to 60 % of control, yet that of TnI was markedly increased (to 30 % above control levels). We conclude that TnI phosphorylation, mediated by cAMP-dependent protein kinase A, is the molecular basis for the enhanced crossbridge cycling seen during β-adrenergic stimulation of the heart.

The β-adrenergically induced positive inotropy of the myocardium is associated with an enhanced rate of cardiac relaxation. This feature is functionally very important in view of the associated tachycardia, which reduces the time available for ventricular filling before the next contraction. Several mechanisms mediated by the β-adrenergically induced elevation of intracellular cAMP have been proposed to effect this enhanced rate of relaxation: an enhanced rate of calcium uptake due to phosphorylation of phospholamban (Garvey et al. 1988), more rapid release of calcium from troponin C due to phosphorylation of troponin I (TnI) (Holroyde et al. 1979; Robertson et al. 1982; Gaponenko et al. 1999) and an enhanced rate of crossbridge cycling (Hoh et al. 1988, 1991; Layland et al. 1997).

The enhancement of crossbridge mechanics during β-adrenergic stimulation is likely to be due to cAMP-mediated phosphorylation of one or more of the myofibrillar proteins. Phosphorylation of the molecular motor itself at the myosin light chain 2 (MyLC2) is unlikely to play any role, since β-adrenergic stimulation does not increase MyLC2 phosphorylation (England et al. 1984) and enhanced MyLC2 phosphorylation in response to endothelin stimulation is not associated with an increase in the frequency at which cardiac muscle dynamic stiffness is minimum (fmin; Rossmanith et al. 1997). It is well known that two myofibrillar proteins, TnI and myosin binding protein C (MyBPC), are phosphorylated in response to β-adrenergic stimulation (England, 1975; Jeacocke & England, 1980). During the course of the study presented here, another group (Kentish et al. 2001) published a study in which it was concluded that TnI phosphorylation was the molecular mechanism responsible for an increase in fmin and for enhancing relaxation during β-adrenergic stimulation. This was achieved using cardiac preparations from genetically modified mice in which the cardiac isoform of TnI was replaced by the slow skeletal TnI isoform. The slow skeletal isoform of TnI lacks the amino terminal peptide of cardiac TnI, which contains protein kinase (PK) A-specific phosphorylation sites, and PKA cannot phosphorylate it (Kentish et al. 2001). Our investigations, using intact rat cardiac muscle preparation without genetic modification, agree with these findings. Our work is based on the investigations of Venema and associates (Venema et al. 1993) who observed that in rat cardiac myocytes the chemical phosphatase 2,3-butanedione monoxime (BDM) generally dephosphorylated β-adrenergically stimulated myofibrillar proteins, but left TnI phosphorylation relatively intact. In the paper presented here, we confirm the observation of Venema et al. in intact rat papillary muscle using isobutylmethylxanthine (IBMX) to simulate β-adrenergic stimulation. We correlated changes in contracture tension, high-frequency stiffness, and the crossbridge cycling parameter fmin, with the phosphorylation status of TnI and MyBPC, and show that the enhanced crossbridge cycling rate seen during β-adrenergic stimulation in cardiac muscle is associated with TnI phosphorylation. To our knowledge, this is the first time that this has been achieved in an intact, wild-type muscle. The ability to change the phosphorylation status of contractile proteins within the one preparation bypasses the need for genetic modification and provides internal controls for the comparison between the native and altered states.


A detailed description of the mechanical apparatus, temperature control, electrical stimulation, preparation and mounting of the papillary muscle is given in previous papers (Rossmanith et al. 1986, 1997; Hoh et al. 1988). The same procedures were used here, but with the following variations. The rats were killed by decapitation and each heart was immediately removed by sternotomy and placed in standard solution (SS; see solutions section) containing 30 mm BDM at room temperature. All dissection procedures were performed in this solution. BDM has been shown to preserve the condition of in vitro preparations and its effects are reversible (Mulieri et al. 1989; Blanchard et al. 1999). Control experiments were performed to ensure that there was no effect of this procedure on the mechanical and biochemical studies. All procedures were carried out in accordance with The Animal Research Act and Regulation of Australia, 1990 (including the Code of Practice for the Care and Use of Animals for Scientific Purposes), and were monitored by the Animal Care and Ethics Committee of Macquarie University.


The SS used was a modified Krebs-Henseleit solution, with the following composition (mm): NaCl 120, KCl 4.69, CaCl2 1.5, MgCl2 0.54, KH2PO4 1.02, NaHCO3 25 and dextrose 10. This was modified further to obtain the following solutions: (1) for calcium-free solution (CFS), CaCl2 was omitted; (2) for Ba2+ contracture solution (BCS), CaCl2 was replaced by 0.5 mm BaCl2 and (3) for phosphate-free solution (PFS), KH2PO4 was omitted. All solutions were maintained at pH 7.4 by infusion with 95 % O2-5 % CO2.

For phosphorylation studies, the composition of the homogenising buffer (SSHB) was: 100 mm sodium pyrophosphate, 5 mm EGTA, 50 mm sodium fluoride, 10 % glycerol, 15 mmβ-mercaptoethanol, 100 μg ml−1 aprotinin, 10 μg ml−1 leupeptin, 2 μg ml−1 pepstatin and 2 μg ml−1 Pefabloc (Silver & Stull, 1982).

The drugs used were BDM and IBMX (Sigma Chemical, St Louis, MO, USA). All drugs and chemicals used were of analytical grade. Stock solutions of each drug were made daily by dissolving the drug in CFS.

Computer control and data analysis

A detailed description of the computer control and data analysis procedures has been reported in earlier publications (Rossmanith, 1986; Rossmanith et al. 1986). The signal used to perturb muscle length was pseudorandom binary noise (PRBN). As a time-domain signal, PRBN is akin to step-changes in muscle length, giving rise to the characteristic tension transients. Envisaged in the frequency domain, the effect of PRBN is to expose the muscle simultaneously to length changes embracing a whole set of frequencies, and is therefore more time efficient than the sinusoidal method that exposes the muscle to one frequency at a time. PRBN provides a convenient link between time and frequency characterisations of crossbridge turnover kinetics. The PRBN was software-generated and was introduced to the muscle via digital-to-analog conversion and the length driver. The required duration of perturbation of the papillary muscle to yield dynamic-stiffness values, at 300 frequencies, in the range 0.1-19 Hz, was 32.8 s. The amplitude of the signal was tailored to the length of the particular muscle preparation, and not to exceed 0.1 % of the operating muscle length (L0). The crossbridge cycling parameter, fmin, is the frequency at which dynamic stiffness attains a local minimum and corresponds to the point of inflexion in the rising part of the phase response. High-frequency stiffness values were obtained by measuring the tension response to small-amplitude (0.05 % L0) sinusoidal length changes at 200 Hz. The tension-to-stiffness ratio was obtained by dividing the steady-state contracture tension by the corresponding high-frequency stiffness.

Data are presented as means ±s.e.m. One-way analysis of variance and Fisher's protected least-significant difference (PLSD) test were used to determine differences between means. Values of P < 0.05 were considered to be significant. All data presented were analysed using StatView SE and Graphics software (Abacus Concepts, Berkeley, CA, USA, 1987).

Protocols used for muscle activation and mechanics

After mounting, the muscle was stretched by 30 % of the length corresponding to the onset of passive tension, to arrive at L0 (Rossmanith et al. 1986; Saeki et al. 1987). The muscle was stimulated electrically at 25 °C at a frequency of 12 pulses min−1. The isometric twitch was displayed on a digitising oscilloscope (model 5223, Tektronix, Beaverton, Ore, USA) and on a chart recorder (model WX 4422, Graphtec, Tokyo, Japan). The muscle was washed in several changes of fresh SS to ensure that any residual BDM was removed from the bath and muscle.

Isometric twitch

When the isometric twitch response had stabilised, it was recorded by means of a Polaroid oscilloscope camera (model C59A, Tektronix) and was taken to be the control response. BDM or IBMX was then added to the muscle bath and the time course of the effect on the twitch profile was recorded. Cumulative and single-addition dose-response effects were recorded.

Ba2+ contracture

After the twitch had stabilised in SS, the muscle was washed in several changes of CFS until the twitch force had become negligibly small. A steady Ba2+ contracture was then induced, without electrical stimulation, by adding 0.5 mm BaCl2 to the CFS bathing solution. Contracture tension was typically 50–100 % of twitch force.

Mechanics during steady Ba2+ contracture

The mechanics measured, once contracture tension had attained steady state, were dynamic stiffness and fmin, steady contracture tension (T), high-frequency stiffness (S), and tension-to-stiffness ratio (T/S). After these control mechanical measurements had been obtained, the inotropic agents were added to the muscle bath. The concentration and sampling time were chosen from dose-response curves in isometric twitch. BDM (5 mm) was shown to cause a > 50 % reduction in peak isometric twitch tension within 5 min of addition to the bath. The suite of mechanical measurements was made 5 min after the addition of 5 mm BDM to the bath. IBMX was added to the muscle bath at a concentration known to affect the crossbridge cycling rate (0.6 mm), and again the mechanical parameters were recorded at the maximally effective time (Hoh et al. 1991). BDM (5 mm) was added to the muscle bath either before or after the addition of IBMX. At the end of each experiment the muscle was relaxed in SS and isometric twitch re-established to verify that the mechanical functionality of the muscle had not been compromised by the experimental procedure. Muscles in which twitch could not be re-established were excluded from the data set.

Protocols for the determination of the phosphorylation state of contractile proteins

The protocol used for this analysis has been described in detail previously (Rossmanith et al. 1997). Briefly, papillary muscles were dissected and placed in PFS and continuously infused with 95 % O2-5 % CO2. The muscles were washed and incubated in 333 μCi ml−1 of [32P]-orthophosphoric acid (Amersham, UK) for 4 h. After incubation, samples were washed four times in fresh PFS for 10 min. BDM and/or IBMX was then added to the experimental vessels. Control muscles were incubated in PFS in the absence of inotropic agents. After stimulation, muscles were freeze-clamped with large forceps cooled with liquid nitrogen. Frozen samples were homogenised immediately in 200 μl SSHB. Samples were placed on ice for 15 min before centrifugation at 13000 g for 10 min at 4 °C. An aliquot of supernatant was removed for protein assay. The remaining supernatant was then added to an equal volume of Laemmli sample buffer and heated at 100 °C for 6 min. Equivalent amounts of total protein for each sample were electrophoresed in a 10 % SDS-PAGE gel, which was used for analysis of protein phosphorylation. Equivalence of protein loading was confirmed in each gel by laser densitometry. The phosphoprotein bands were analysed from autoradiograms by laser scanning densitometry or from phosphorimages (Molecular Dynamics, CA, USA).


Effects of BDM and IBMX on isometric twitch contractions

The main thrust of this work was to investigate T, S, fmin and the phosphorylation status of myofibrillar proteins when papillary muscles in Ba2+ contracture were treated with BDM and IBMX. The effects of these agents on twitch parameters, at the doses used later during contracture, were recorded in order to assess the actions of these agents at these doses on the more physiological twitch contractions. Our results are given in Fig. 1. The addition of 0.6 mm IBMX to the muscle bath caused an approximately two-fold increase in peak isometric force and a reduction in the twitch-time parameters, time to peak and time to half-relaxation, signifying a faster isometric twitch. The addition of 5 mm BDM also resulted in a faster isometric twitch but caused a 60 % reduction in peak twitch force. These results confirmed previous observations (Coulombe et al. 1990; Hoh et al. 1991; Suetake et al. 1992; Backx et al. 1994; Ebus & Stienen, 1996). The decrease in twitch-time parameters caused by the addition of IBMX was greater than the reduction caused by BDM. When the two agents were added consecutively to the muscle bath, the relative effect of the second agent was almost the same as the action of the agent on control tissue: IBMX caused a 100 % increase in the BDM-induced peak twitch force, while the addition of BDM resulted in a 60 % reduction in the IBMX-induced peak twitch force. Therefore, within the time frames explored, the actions of BDM and IBMX on peak twitch force were the same, irrespective of whether the agent acted on control muscle or on muscles already exposed to the other agent. Similar results followed for the summed effects of BDM and IBMX on the twitch time parameters (Fig. 1).

Figure 1.

The effect of phosphatase 2,3-butanedione monoxime (BDM) and isobutylmethylxanthine (IBMX) on isometric twitch parameters of intact rat papillary muscle

Twitch parameters (peak force, time to peak (TTP) and half-relaxation time (RT50))are normalised to the control value (C, ▪) and presented as mean ±s.e.m. All muscles act as their own control. BDM (B, inline image, n/ 10) sampled at 5 min after addition to muscle bath, IBMX (I, □, n= 8) sampled at 3.5 min, BDM + IBMX (B + I, □, n= 20) sampled at plateau, typically 5 min after addition of the second agent. All timing parameters are significantly different from control and between treatments. For peak isometric tension: a significantly different when compared to control value; b significantly different when compared to BDM value; c significantly different when compared to IBMX value.

Effects of BDM and IBMX on contracture T and S

Values of S and T were obtained from papillary muscles in steady-state Ba2+ contracture. Each papillary muscle acted as its own control, and values obtained after the addition of BDM and/or IBMX were normalised to this control value. Figure 2 shows the time course of the effects of BDM and IBMX on the T/S ratio. The addition of BDM to the muscle bath depressed both T and S, whereas IBMX increased these parameters. The relative effects of both agents on T and S can be seen in the respective T/S ratios. BDM depressed the T/S ratio, while IBMX left this ratio unchanged. In the presence of 0.6 mm IBMX, the T/S ratio did not alter significantly from control values over a 5 min period, while in the presence of 5 mm BDM, the T/S ratio decreased steadily over a 5 min period to 80 % of the control value. This means that whereas the effect of BDM was to depress tension more than stiffness, IBMX caused the same proportional increase in both stiffness and tension.

Figure 2.

The effect of BDM and IBMX on the relationship between tension (T) and stiffness (S) in intact rat papillary muscle

Simultaneous measurements of T and S and the effect of the addition of BDM then IBMX (□, n/ 7) or IBMX then BDM (•, n= 5) were recorded. The arrows indicated the addition of the first and then second agent to the muscle bath. The T/S ratio is normalised to the control ratio (Time = 0) for each papillary muscle and is presented as mean ±s.e.m.

Consecutive addition of BDM and IBMX produced results that were order dependent. The addition of IBMX to a BDM-conditioned muscle caused a slowing in the BDM-induced rate of decrease in the T/S ratio. At 10 min after the addition of IBMX, the T/S ratio had decreased from 80 % to 73 % of control values (see Fig. 2). The addition of 5 mm BDM to an IBMX-conditioned muscle resulted in a 42 % decrease in T/S ratio within a 2 min period. This decrease was sustained for the period of observation (see Fig. 2). So, whereas IBMX caused a gradual slowing of the BDM-induced decline in the T/S ratio, BDM caused a ‘sudden’ and maintained decrease in the IBMX-induced T/S ratio.

Effects of BDM and IBMX on fmin

Analysis of the small-amplitude PRBN length changes and the resulting force changes gave rise to the characteristic dynamic stiffness and phase data (Rossmanith, 1986, Rossmanith et al. 1986). This data displays a local minimum in dynamic stiffness and a mini-max feature in the phase-frequency plot. The frequency, fmin, at which this minimum in stiffness occurs, corresponds to the point of inflexion in the phase plot (see Fig. 3). It has been established in previous studies that intact ventricular muscle from juvenile rats has an fmin of 2 Hz at 25 °C (Rossmanith et al. 1986, 1997; Hoh et al. 1988, 1991). This was confirmed in the present study (Fig. 3, control curves). IBMX has also been shown to effect an increase in fmin of approximately 80 % (Hoh et al. 1991). This was again confirmed in the present investigation (Fig. 3A and D).

Figure 3.

Dynamic stiffness and phase plots from intact rat papillary muscle

Representative dynamic stiffness (dashed lines) and phase plots (continuous lines) from rat papillary muscle under steady-state 0.5 mm Ba2+ contracture at 25 ° C are shown in A, B and C. The pseudorandom binary noise length signal and corresponding dynamic force response gave rise to dynamic stiffness and phase data at 340 equally spaced frequencies in the range 0.1-19 Hz. A stiffness value of 1 is equivalent to 11.1 N m−1. fmin corresponds to the frequency at which the dynamic stiffness minimum occurs. This value corresponds to the point of inflexion of the phase plot. A, stiffness and phase plots before (thin line) and 3.5 min after (thick line) the addition of 0.6 mm IBMX to the muscle bath. The effect of IBMX was to shift the stiffness and phase plots to the right, indicating an increase in the crossbridge cycling kinetics. The value of fmin for this experiment increased from 2.0 to 3.5 Hz with the addition of IBMX. B, stiffness and phase plots before (thin line) and 5 min after (thick line) the addition of 5 mm BDM to the muscle bath. BDM did not alter fmin, indicating that the crossbridge cycling kinetics were unchanged. C, stiffness and phase plots before (thin line) and 8 min after (thick line) the addition of 0.6 mm IBMX and 5 mm BDM to the muscle bath. The effect of the combination of IBMX and BDM was to shift the stiffness and phase plots to the right, indicating an increase in the crossbridge cycling kinetics. The value of fmin for this experiment increased from 2.1 to 3.5 Hz. D, pooled results for all experiments. Steady-state tension and fmin values are given for rat papillary muscle in response to BDM (B, inline image, n/ 12), IBMX (I, □, n= 8) and a combination of BDM and IBMX (B + I, □, n= 15). All values are normalised to the control (C, ▪). The control fmin value was 2.1 ± 0.7 Hz (n= 20). aSignificantly different when compared to the control value; bsignificantly different from BDM value; csignificantly different from IBMX value.

In contrast to IBMX, BDM did not cause a shift in fmin in intact ventricular muscles (Fig. 3B). The presence of both BDM and IBMX caused fmin to increase, regardless of the order of addition of the agents (Fig. 3C). The increase in fmin effected by the presence of both BDM and IBMX is comparable to the effect of IBMX alone on control muscle (see Fig. 3C and D). The same concentrations of inotropic agents were used in both the isometric twitch and perturbation analysis protocols. Whereas both BDM and IBMX had an effect on the speed of the isometric twitch, only IBMX increased kinetics (as captured by fmin) during steady activation.

Effects of BDM and IBMX on phosphorylation state of contractile proteins

SDS-PAGE and autoradiographic analysis indicated that a basal level of 32Pi radiolabelling occurred in a number of proteins in rat cardiac tissue, identified as MyBPC, TnT, tropomyosin, TnI and MyLC2. The levels of phosphorylation of these proteins varied slightly between the different experiments, but after careful densitometric analyses of at least four separate hearts for each experimental protocol, the only proteins in which the phosphorylation state was consistently altered (compared with the control) were MyBPC, TnI and MyLC2. Individual results for the addition of IBMX, BDM and BDM plus IBMX are shown in Fig. 4A, B and C, respectively. Pooled results for the effects of the agents on the phosphorylation state of these proteins are shown in Fig. 4D. The addition of BDM (5 mm) caused a marked decrease in the radiolabelling of MyBPC and MyLC2, but had no effect on the basal level of phosphorylation of TnI (see Fig. 4D), confirming the observation of Venema et al. (1993) on cardiac myocytes. IBMX at 0.6 mm markedly and consistently increased the radiolabelling of two of these proteins, MyBPC and TnI, but had little effect on MyLC2 phospholabelling (Fig. 4D). The addition of the two agents together caused an increase in the radiolabelling of TnI, no change in the radiolabelling of MyBPC, yet a marked decrease in the radiolabelling of MyLC2 (Fig. 4D). The increase in the phospholabelling of TnI seen when the two agents where combined was about 30 %, whereas that seen when IBMX was used alone was about 60 % (see Fig. 4D). The action of IBMX and BDM on both the mechanics and on the level of phosphorylation of the contractile proteins is given in Table 1.

Figure 4.

The effect of BDM and IBMX on the phospholabelling of contractile proteins in rat cardiac tissue

A, B and C, representative experiments of the phospholabelling of contractile proteins. Papillary muscles were phospholabelled with 32P for 4 h and then incubated without (control) or with 0.6 mm IBMX for 5 or 10 min (A), 5 mm BDM for 5 min (B), or a combination of BDM and IBMX for 10 min (C). Samples were freeze-clamped, homogenised and separated on 10 % SDS-PAGE gels. The gels were dried and exposed to autoradiographic film. Proteins were identified by Coomassie blue staining. D, pooled data from all of the experiments. Autoradiographs were scanned and analysed by laser scanning densitometry. Values were normalised to their relative control and are presented as mean ±s.e.m. The bars represent control (C, ▪), BDM (B, o, n/ 4), IBMX (I, □, n= 4) and BDM + IBMX (B + I, □, n= 5). MyBPC, Myosin binding protein C; TnI, troponin I; MLC2, myosin light chain 2. a Significantly different when compared to control value; b significantly different from BDM value; c significantly different from IBMX value; d significantly different from BDM + IBMX value.

Table 1. The observed actions of phosphatase 2,3-butanedione monoxime (BDM) and isobutylmethylxanthine (IBMX) on intact rat myocardium
 STT/S f min TnI-PMyBPC-PMyLC2-P
  1. The table shows the relative actions of BDM and IBMX, both alone and in combination (B + I) when compared to control values. The upward- and downward-pointing arrows indicate an increase or decrease in the parameter, respectively. The lack of effect of the agents on a parameters is indicated by ‘–’. S, stiffness; T, isometric contracture tension; T/S tension-to-stiffness ratio; fmin, minimum stiffness frequency; TnI-P; level of phosphorylation of Troponin I; MyBPC-P, level of phosphorylation of myosin binding protein C; MyLC2-P, level of phosphorylation of myosin light chain 2.

B + I   


The aim of this investigation was to elucidate in intact muscle, the molecular mechanism underlying the β-adrenergically induced increase in crossbridge cycling kinetics, in particular to resolve whether this action was mediated by phosphorylation of TnI or MyBPC. The prospect of resolving this problem was greatly enhanced by the observation of Venema et al. (1993) that the chemical phosphatase BDM was able to dephosphorylate MyBPC without significantly affecting the phosphorylation status of TnI. Using an intact cardiac muscle preparation, we confirmed these observations. With the muscle in steady contracture induced by Ba2+, we showed that BDM did not affect fmin. Furthermore, in the presence of BDM, IBMX, a phosphodiesterase inhibitor that raises the intracellular cAMP level, was still able to induce an enhancement of fmin to a level equal to that induced by IBMX alone in control tissue. This enhancement of fmin was associated with a significant increase in TnI phospholabelling, but left the phosphorylation status of MyBPC at the control level. Thus, we conclude that the enhancement of fmin that is observed during β-adrenergic stimulation in intact, wild-type cardiac tissue is due to the phosphorylation of TnI.

It is widely accepted that during a crossbridge cycle, crossbridges can occupy a number of attached and detached states (for example, Huxley & Simmons, 1971; Smith, 1990; Rossmanith & Tjokorda, 1998). In such models, a detached crossbridge is thought to initially attach to actin in a low-tension state and then, while still attached, undergo a power stroke(s) to generate a high-tension state(s). Finally it detaches, completing the crossbridge cycle. The parameter S is a correlate of the number of crossbridges in all the attached states, while T also depends upon the distribution of the crossbridges between these attached states. The frequency fmin, at which dynamic stiffness assumes a minimum, has been shown to be correlated with the rate of detachment of crossbridges from the high-tension state and with the rate at which crossbridges perform the power stroke. fmin is insensitive to changes in the rate at which crossbridges attach, while such a change would modulate both S and T (Rossmanith & Tjokorda, 1998). In terms of such generic crossbridge models, the mechanical parameters T, S and fmin give insight into how BDM, IBMX and the phosphorylation of contractile proteins alter the number of crossbridges that are attached, their distribution between the attached states, and their rate of detachment.

Using our protocol for inducing a Ba2+ contracture, T and S are very stable over a prolonged period of time. This implies that the intracellular concentration of Ba2+ is at equilibrium. Studies on skeletal muscle have shown that Ba2+ is not taken up by the sarcoplasmic reticulum (SR), nor can it release Ca2+ from the SR by calcium-induced Ca2+ release (Blaineau et al. 1993). Although BDM has been shown to decrease the opening probability of L-type Ca2+ channels, and IBMX to increase it (Allen & Chapman, 1995; Sakai et al. 1999), these agents are unlikely to influence the intracellular Ba2+ concentration, as the preparation was not stimulated during contracture. Notwithstanding the possibility that BDM and IBMX might influence T and S by affecting the intracellular Ba2+ concentration in some other way, a change in the intracellular level of Ba2+ cannot explain the changes in the T/S ratio. The effect of BDM and IBMX on the T/S ratio can, however, be explained by the action of these agents on contractile proteins.

BDM at 5 mm decreased both the T and S of cardiac muscle during Ba2+ contracture, consistent with a net decrease in the number of attached crossbridges. This can be envisaged by BDM causing a decrease in the rate of attachment of crossbridges. Since T decreased more than S, this suggests further a redistribution of attached crossbridges that favours occupation of low-tension states. The observed decrease in T is consistent with the effects of BDM on tension observed by others (Li et al. 1985; West & Stephenson, 1989; Gwathmey et al. 1991). The fact that BDM did not alter fmin suggests that the kinetics of the power stroke and of the rate of detachment are unchanged.

BDM significantly reduced the level of phosphorylation of MyBPC and MyLC2 without significantly changing the level of phosphorylation of TnI. This resistance to BDM is presumably due to a favourable chemical environment at the PKA-specific phosphorylation sites of cardiac TnI. Phosphorylation of MyBPC increases the effective concentration of myosin heads available for interaction with the thin filament without changing crossbridge kinetics (Weisberg & Winegrad, 1998; Kunst et al. 2000; Levine et al. 2001). These effects are likely to increase the rate of crossbridge attachment, thus increasing T and S. MyLC2 phosphorylation is also thought to cause crossbridges to swing out towards the thin filaments, and is associated with the enhanced force development, as shown in a recent study on endothelin (Rossmanith et al. 1997). The dephosphorylation of MyBPC and MyLC2 may reduce the effective number of crossbridges available for binding and cause these crossbridges to lie closer to the thick filament backbone. Thus, the dephosphorylation of these proteins can in part explain the action of BDM in reducing T and S. The fact that an endothelin-induced increase in the level of phosphorylation of MyLC2 also did not alter fmin (Rossmanith et al. 1997) is consistent with the non-effect of BDM on fmin. The BDM-induced dephosphorylation of these proteins is unlikely to explain the redistribution of crossbridges in favour of the low-tension states; for this, the direct action of BDM on actin and myosin may be the cause.

The action of IBMX on cardiac muscle during Ba2+ contracture was to increase both S and T, pointing to a net increase in the number of attached crossbridges, and suggests an increase in the rate of attachment of crossbridges. We note that in contrast to the isometric twitch, where IBMX caused an approximate two-fold increase in maximal force, the effect of IBMX on Ba2+ contracture tension was to increase it by only 16 % (Fig. 3D). This is because the use of Ba2+ contracture bypassed the effects of IBMX on the L-type Ca2+ channel, phospholamban and intracellular Ca2+ cycling. The observation that the T/S ratio was not changed by IBMX during Ba2+ contracture indicates that the distribution of crossbridge numbers between the attached states was not altered. The fact that IBMX also caused fmin to increase suggests an increase in the rate of detachment of crossbridges from the high-tension state. The increase in S and T would require that this increase in the rate of detachment be more than compensated by an increase in the rate of attachment of crossbridges.

We confirm in the present study that IBMX, a phosphodiesterase inhibitor that elevates intracellular cAMP, has the ability to phosphorylate both TnI and MyBPC (England, 1975, 1976; Kranias & Solaro, 1982; Miyakoda et al. 1987; Garvey et al. 1988). We found no change in MyLC2 phosphorylation in response to this IBMX-induced increase in intracellular cAMP. This is consistent with the results of others (High & Stull, 1980; Kranias& Solaro, 1982; England et al. 1984). It is likely that the increase in the level of phosphorylation in MyBPC is the mechanism underlying the IBMX-induced compensatory increase in the rate of crossbridge attachment postulated herein. This suggestion offers an insight into the functional significance of the tight coupling between the phosphorylation of TnI and MyBPC. Without a concomitant increase in MyBPC phosphorylation, high values of T and S cannot be maintained at a given level of activation in the presence of accelerated crossbridge detachment due to increased TnI phosphorylation.

The summed effect of BDM and IBMX was to increase fmin, and to decrease T, S and the T/S ratio. The increase observed in fmin was the same as that induced by IBMX alone on control muscle. The combined use of BDM with IBMX in an intact cardiac muscle resulted in a selective increase in the phosphorylation of TnI whilst maintaining the phosphorylation state of MyBPC at the control level and depressing that of MyLC2 to below its basal level (Fig. 4D). To our knowledge, this is the first time that uncoupling of the phosphorylation of TnI and MyBPC has been achieved in an intact cardiac muscle, enabling us to implicate TnI phosphorylation as the mechanism for increased fmin. It is of interest that the crossbridge model discussed herein presents the increase in the rate of detachment of crossbridges as the mechanism underlying the shift in fmin. A corollary of this increase in the rate of crossbridge detachment is an increase in the rate of tension relaxation upon withdrawal of the activator, consistent with the tension-relaxation results presented by Kentish et al. (2001). Further work is required to determine how phosphorylation of cardiac TnI increases the rate of detachment of crossbridges.

The order of addition of the agents BDM and IBMX does produce qualitatively the same, but quantitatively different results with regard to the T/S ratio (see Fig. 2). Whereas the addition of BDM caused a modest decrease in the T/S ratio, which is little changed by IBMX, the addition of BDM to the IBMX-equilibrated muscle caused a ‘rapid’ and large decrease in the T/S ratio, and this decrease was maintained over the next 10 min. The order-of-addition dependence of the rate of decrease in the T/S ratio suggests that the ease of redistributing attached crossbridges to favour the low-tension state is dependent upon the initial condition. The dephosphorylating action of BDM outweighs the phosphorylating action of IBMX on both MyBPC and MyLC2, but whereas the level of phosphorylation of MyBPC was returned to control levels, BDM caused the level of phosphorylation of MyLC2 to be reduced to below control levels (Fig. 4D). The reduction in MyLC2 phosphorylation below control levels can in part account for the BDM-with-IBMX-induced decrease in T and S.

It is of interest that whilst BDM does not affect the basal level of phosphorylation of TnI, it does decrease the level of phospholabelling of TnI in an IBMX-affected muscle (Fig. 4D). From the pooled data from all the experiments (Fig. 4D), in response to IBMX alone, the phosphorylation of TnI increased to 60 % above control levels, but when a muscle was exposed to a combination of BDM and IBMX, the phosphorylation of TnI was increased only 30 % above control levels. This suggests that while BDM does not dephosphorylate PKA-specific sites in the amino terminal peptide of cardiac TnI, it may dephosphorylate other sites of phosphorylation of TnI, such as those acted on by PKC (Jideama et al. 1996). As the magnitude of the change in fmin induced by IBMX with BDM is the same as the change in fmin induced by IBMX alone, the BDM-resistant sites are sufficient to support the observed fmin shift (Fig. 3D).

Our results and those of others indicate the occurrence of significant TnI phosphorylation in control muscle. It has not been determined whether the basal phosphorylation of TnI is located at the PKA-specific sites or elsewhere. Since the level of phosphorylation of TnI is not altered by BDM, in the light of the above analysis this is likely to be located at the PKA-specific sites. This suggests that in the absence of basal TnI phosphorylation, fmin would be lower than in normal control muscle.

Clear evidence for the contribution of TnI phosphorylation to positive lusitropy has come from transgenic mice, where the cardiac TnI isoform has been replaced by the slow skeletal isoform that cannot be phosphorylated (Fentzke et al. 1999). Isoproterenol increased fmin in intact muscles from wild-type mice but not in transgenic mice. In skinned preparations from these animals, the rate of relaxation was unchanged after incubation with PKA, even though MyBPC was phosphorylated (Kentish et al. 2001). These findings are in agreement with the conclusions from the present study, in which decoupling of TnI and MyBPC was achieved by the use of a pharmacological agent. The advantage of this approach lies in its simplicity and the expected applicability to other species. Furthermore, our analysis provides a rational explanation for the tight coupling between TnI phosphorylation and MyBPC phosphorylation during β-adrenergic stimulation.


The study received support from the National Health and Medical Research Council of Australia (G.H.R., J.F.Y.H.), the Australian Postgraduate Award Scheme (L.T.) and The Macquarie University Postgraduate Research Grant Scheme (L.T.).