Microtubule Disruption Modulates Ca2+ Signaling in Rat Cardiac Myocytes
Abstract—Microtubules have been shown to alter contraction in cardiac myocytes through changes in cellular stiffness. However, an effect on excitation-contraction coupling has not been examined. Here we analyze the effects of microtubule disruption by 1 μmol/L colchicine on calcium currents (ICa) and [Ca2+]i transients in rat ventricular myocytes. ICa was studied using the whole-cell patch-clamp technique. Colchicine treatment increased ICa density (peak values, −4.6±0.4 and −9.1±1.3 pA/pF in 11 control and 12 colchicine-treated myocytes, respectively; P<0.05). ICa inactivation was well fitted by a biexponential function. The slow component of inactivation was unchanged, whereas the fast component was accelerated after colchicine treatment (at −10 mV, 11.8±1.0 versus 6.7±1.0 ms in control versus colchicine-treated cells; P<0.005). [Ca2+]i transients were analyzed by fluo-3 epifluorescence simultaneously with ICa. Peak [Ca2+]i transients were significantly increased in cardiac myocytes treated with colchicine. The values of F/F0 at 0 mV were 1.1±0.02 in 9 control cells and 1.4±0.1 in 11 colchicine-treated cells (P<0.05). β-Adrenergic stimulation with 1 μmol/L isoproterenol increased both ICa and [Ca2+]i transient in control cells. However, no significant change was induced by isoproterenol on colchicine-treated cells. Colchicine and isoproterenol effects were similar and not additive. Inhibition of adenylyl cyclase by 200 μmol/L 2′-deoxyadenosine 3′-monophosphate blunted the colchicine effect. We suggest that β-adrenergic stimulation and microtubule disruption share a common pathway to enhance ICa and [Ca2+]i transient.
Microtubules constitute one of the main cytoskeletal components, together with actin and intermediate filaments. The microtubule network is dynamic, composed by the self-association of α,β-tubulin dimers. Thus, by polymerization and depolymerization, the cell can change the amount of microtubules at constant tubulin amount. The presence of microtubules in the cardiac myocytes is well known, but its role in physiology and pathology is thought to be purely mechanical. In this regard, it has been shown that in pressure-overload cardiac hypertrophy, there is an increase in the microtubule network, which would be responsible for the contractile dysfunction in hypertrophied cells.1 In this elegant work, Tsutsui et al1 studied right ventricular cardiac myocytes isolated from cats subjected to pulmonary artery constriction. Under these experimental conditions, hypertrophied cells presented an increased number of microtubules and contracted weakly. When treated with the depolymerizing agent colchicine, hypertrophied myocytes contracted normally. Tsutsui et al1 concluded that the contractile defect of hypertrophied cells is due to an increase in stiffness and viscosity on the cell imposed by the increased microtubule network triggered by the pressure overload.2 However, it is also possible that microtubule polymerization and depolymerization play other roles in addition to the mechanical one. In this regard, we have recently shown that heart failure after pressure-overload cardiac hypertrophy induces a dysfunction of the excitation-contraction (EC) coupling.3 This alteration can account for the decreased contractile function found in this animal model. Because microtubules are increased in the weakly contracting myocytes after pressure-overload cardiac hypertrophy,4 5 this cytoskeletal abnormality might be in part responsible for the contractile dysfunction observed in pressure overload–induced heart failure.
Electrical excitation during an action potential activates sarcolemmal Ca2+ channels. Ca2+ influx that follows opening of these channels induces a local elevation of [Ca2+]i around the sarcoplasmic reticulum (SR) Ca2+ channels, or ryanodine receptors (RyRs). Activation of RyRs by Ca2+ triggers the SR Ca2+ release that would be able to activate contractile fibrils and contraction.
In this study, we analyzed the effects of microtubule depolymerization on the 2 main components (Ca2+ current and SR Ca2+ release) of EC coupling, and we found that microtubule depolymerization increases them both. Moreover, this effect is blocked by inhibition of adenylyl cyclase. Thus, besides a mechanical role, microtubules seem to be important modulators of calcium signaling and, hence, cardiac function.
Materials and Methods
Cardiac ventricular myocytes were isolated from adult male Wistar rats (275 to 325 g) as previously described.6
Cardiomyocytes were fixed and immunolabeled with anti–β-tubulin and a fluorescence secondary antibody (C. Frederick and W.J. Lederer, unpublished data, 1996; see online-only supplementary information, http://www.circresaha.org). Cells were viewed using a confocal microscope Zeiss LSM 510 fitted with an argon laser (488-nm wavelength). Emission was collected through a low-pass filter at 505 nm. Parameters were first adjusted with a control cell and maintained constant to examine all cells.
The whole-cell mode of the patch-clamp technique7 was used to study L-type Ca2+ current (ICa). Myocytes were perfused with HEPES solution containing, in mmol/L, NaCl 140, MgCl2 0.5, CsCl 5, glucose 5.5, HEPES 5, and CaCl2 1.8 (pH set to 7.4 with NaOH). Myocytes were voltage-clamped (Axopatch 200A, Axon instruments) with a suction pipette filled with a solution containing, in mmol/L, CsCl 130, MgCl2 1, NaH2PO4 1, Na2 phosphocreatine 3.6, MgATP 5, HEPES 10, and fluo-3 (pentapotassium salt) 0.1 (pH fixed at 7.2 with CsOH). In some experiments, 200 μmol/L 2′-deoxyadenosine 3′-monophosphate (2′d3′AMP) was added to the pipette solution. Pipettes had tip resistances of 0.9 to 1.2 MΩ. Capacitance and series resistance were electronically compensated to ≈60%. ICa was elicited as previously explained.3
Cells were loaded with the fluorescence-Ca2+ dye fluo-3 (Molecular Probes) either by diffusion of its salt form through the patch pipette or by using its acetoxymethyl ester derivative as previously described.6 This second method was used in the experiments conducted to estimate the SR Ca2+ load.
Fluo-3–loaded cells were excited with a xenon lamp at 460- to 490-nm wavelength through an epifluorescence attachment. Emission fluorescence (520 nm) was detected with a photomultiplier tube. Microscope and fluorescence equipment were from Nikon France. The signal was then amplified and low-pass filtered at 100 kHz (Fern Development).
ICa and fluorescence signals were simultaneously digitized (Digidata 1200, Axon instruments) and acquired at sampling rate of 100 μs using pClamp 7.
Data are presented as mean±SEM. An unpaired Student t test was performed to compare control and colchicine-treated cells or control and paclitaxel (Taxol)–treated cells, whereas a paired Student t test was used to test the isoproterenol (ISO) effect. P<0.05 was considered significant.
An expanded Materials and Methods section is available online at http://www.circresaha.org.
The role of microtubules in EC coupling was investigated by analyzing the effect of colchicine, a substance known to induce microtubule depolymerization, on Ca2+ current and SR Ca2+ release.
To be sure that colchicine treatment was effectively disrupting microtubules in our hands, some cells were fixed for immunofluorescence assay. Control myocytes and cells exposed to 1 μmol/L colchicine at different times were marked with anti–β-tubulin and fluorescein-conjugated secondary antibody as described in Materials and Methods. To allow comparison between different cell groups, we used the same parameters in the microscope configuration to visualize all cells. After 30 minutes and for up to 4 hours, microtubules were effectively disrupted. Because we started doing patch-clamp experiments after 1.5 hours of colchicine treatment, we chose to present images and data of immunofluorescence after 2 hours colchicine treatment. Figure 1A⇓ shows examples of 1 cardiac myocyte not exposed to colchicine (left) and after 2 hours of colchicine treatment (right). The observed filamentous structures in control myocytes were absent in colchicine-treated cells. Because all cells were marked in the same way and images were taken under the same conditions, we measured the averaged fluorescence in each cell, which is correlated with the number of microtubules. Fluorescence values were 61.4±3.4 versus 38.3±3.3 in 14 control versus 19 colchicine-treated myocytes, P<0.0001.
Ca2+ Current and [Ca2+]i Transient
The effect of disrupting microtubules on L-type calcium current, ICa, was analyzed in rat ventricular cells under whole-cell patch clamp. ICa was elicited by applying 100-ms depolarizing pulses from −50 to +60 mV every 10 seconds from a holding potential of −80 mV. To allow steady-state Ca2+ load of the SR, 4 steps to 0 mV were applied at 1 Hz between test pulses. Sodium current was inactivated by predepolarization to −50 mV (achieved by a 500-ms ramp followed by maintaining at this potential for 100 ms) before each test pulse.3
Colchicine was first dissolved in DMSO and then added to an aliquot of cells. DMSO concentration in the cell suspension was 0.01%. This concentration of DMSO did not induce any significant change in ICa (data not shown). Experiments were made 1.5 to 4 hours after colchicine addition.
ICa recorded in cells treated with 1 μmol/L colchicine was markedly increased when compared with ICa recorded in cells in control conditions (Figure 1B⇑, bottom). To avoid error in pooling data from different-sized myocytes, we normalized the ICa amplitude by the cell capacitance, to get ICa density. Cell capacitance was of similar magnitude in control cells and in cells that were incubated with colchicine (172.0±15.0 [n=13] versus 156.9±23.4 pF [n=12] in control versus colchicine). The current density/voltage relationship of ICa is shown in Figure 1C⇑ (bottom). Microtubule disruption induced an increase in ICa that is statistically significant from −10 to +20 mV. To check whether or not the observed effect was due to a direct effect of colchicine, some control cells were patch-clamped and perfused with 1 μmol/L colchicine. Direct perfusion of myocytes up to 5 minutes with colchicine was without effect on ICa (data not shown).
Cardiac myocytes contraction arises when the increase in [Ca2+]i triggered by Ca2+ influx through Ca2+ channel activates neighboring RyRs. The opening of RyRs provokes a large Ca2+ release that activates contractile fibrils. Microtubule disassembly increased ICa; thus, we addressed the issue of whether microtubules could modulate [Ca2+]i transient. Figure 1B⇑, top, shows representative fluorescence traces recorded simultaneously with ICa in a control and a colchicine-treated cell. Colchicine treatment markedly increased [Ca2+]i transient. Comparison of average data are shown in Figure 1C⇑ (top). This result was expected, because ICa, which triggers Ca2+ release, is increased under these conditions (Figure 1C⇑, bottom).
ICa and [Ca2+]i Transient Kinetics
This is the first time to our knowledge that an effect of microtubules on ICa at the whole-cell level is shown. There is, however, a study in single Ca2+ current.8 In this study, performed in embryonic chick ventricle cells, colchicine increased the inactivation of single-channel current. We further analyzed the effect of colchicine on ICa kinetics. As observed in the current records (see Figures 1B⇑ and 5A⇓), ICa inactivation seems faster in colchicine-treated myocytes. To quantify this observation, current decay was fitted by the following biexponential equation: y=C+Afast×e−t/τfast+Aslow×e−t/τslow
where C is a constant, Afast and Aslow are the maximal amplitude of the fast and slow components respectively, and τfast and τslow are the time constant of the fast and slow components, respectively. The slow time constant of ICa inactivation was unaltered by microtubule disruption (Figure 2A⇓, right). The fast component of ICa inactivation, τfast, was shorter after microtubule disruption (at −10 mV, 11.8±1.0 versus 6.7±1.0 ms in control [n=11] versus colchicine-treated cells [n=12], P<0.005) (Figure 2A⇓, left).
ICa steady-state inactivation was analyzed by applying voltage steps at 0 mV preceded by a 1-second predepolarization to different potentials ranging from −50 to +50 mV. The stimulation frequency was 0.1 Hz. The current elicited during each test potential was normalized to the current obtained during the test potential that was preceded by predepolarization at −50 mV. Normalized currents are plotted against predepolarization voltages in Figure 2B⇑. By fitting to a Boltzmann function, we obtained the following voltages of half-inactivation, which were similar in both myocyte groups: –36.2±1.4 mV for control cells (n=8) and –32.4±1.6 mV after colchicine treatment (n=4).
[Ca2+]i transient kinetics were then examined. The time from start of the depolarizing pulse to peak of the [Ca2+]i transient was not statistically different in both cell groups (Figure 3A⇓). However, decay time of the [Ca2+]i transient was accelerated. The [Ca2+]i transient decay could be well fitted to a single exponential function. The [Ca2+]i transient decay was significantly faster in cells treated with colchicine than in control cells (Figure 3B⇓).
The acceleration of the [Ca2+]i transient decay could be due to an acceleration of the SR Ca2+-ATPase activity. If this were true, SR Ca2+ content could be elevated. The increased amount of Ca2+ entry though Ca2+ channels observed after colchicine treatment (Figure 1⇑) could also account for differences in the SR Ca2+ load. After steady-state field stimulation at 1 Hz for 2 minutes, the amplitude of the fluorescence transient obtained by a rapid application of 10 mmol/L caffeine was used to assess SR Ca2+ content.6 Figure 4A⇓ shows representative traces of caffeine-induced fluorescence transients in a control (open circles) and a colchicine-treated (closed circles) cell. After colchicine treatment, SR Ca2+ content was increased. On average, the amplitude of the caffeine-induced [Ca2+]i transient (F/F0) was 2.4±0.3 in 10 control myocytes versus 4.0±0.6 in 9 colchicine-treated myocytes (P<0.05). To check whether or not the observed increase in [Ca2+]i transient (Figure 2⇑) was due to an increase in the activating ICa, we analyzed the “gain” function estimated by normalizing the rate of Ca2+ release to the integral of ICa, which we called calcium-induced calcium release (CICR) gain. If colchicine treatment alters EC coupling, CICR gain should be modified. Figure 4B⇓ shows that CICR gain is voltage-dependent and not significantly different after colchicine treatment. It appears that microtubule disruption does not affect coupling but rather induces changes in other aspects of EC coupling such as ICa and [Ca2+]i transient.
Effect of β-Adrenergic Stimulation
In cardiac tissues, β-adrenergic stimulation is known to increase cAMP that will activate phosphorylation by protein kinase A (PKA). It has been suggested that microtubules modulate the β-adrenergic response in rat cardiac hypertrophy.9 Forskolin, a direct activator of PKA, is ineffective after microtubule disruption on colonic epithelia.10 These observations led us to test the effects of PKA activation by a β-agonist on control cells and cells after microtubule disruption. After measuring ICa and [Ca2+]i transient, we added to the bath solution 1 μmol/L ISO. ISO induced a marked increase in both ICa and [Ca2+]i transient in control myocytes. However, only a weak increase in ICa and [Ca2+]i transient could be registered after ISO application in the cells treated with colchicine (Figure 5A⇓). Moreover, despite the difference in ICa in both cell groups, after ISO treatment ICa became similar in control and colchicine-treated myocytes.
Figure 5B⇑ summarizes the effect of ISO application on ICa and [Ca2+]i transient at 0 mV in control cells and in cells treated with colchicine. For each cell, the values obtained after ISO application are divided by the values before ISO application (unit=no ISO effect). In control cells, there is a ≈76.5% increase in ICa and ≈83.5% in [Ca2+]i transient (F/F0). However, the β-adrenergic agonist had no more significant effects in cardiac cells after microtubule disruption by colchicine.
Effect of Taxol on ICa
As a result of microtubule disruption, the amount of free tubulin increases within the cell. It is possible that the observed effect on Ca2+ signaling after colchicine treatment is due to the diminution of polymerized tubulin or to an increase on free tubulin. To check this possibility, the microtubule-stabilizing agent Taxol was used. Taxol was first dissolved in DMSO as colchicine. The final DMSO concentration was 0.01%.
Because the changes in Ca2+ signaling are mainly due to the increase on ICa, we analyzed ICa density in cells incubated with 10 μmol/L Taxol for at least 1.5 hours, which stabilized microtubules. Figure 6A⇓ shows confocal images of a control (left) and Taxol-treated (right) myocyte labeled with anti–β-tubulin antibody, as explained in Materials and Methods. Averaged fluorescence levels in Taxol-treated cells were higher than in control cells: 61.4±3.4 (control, n=14) versus 93.2±6.0 (Taxol, n=31); P<0.01. Figure 6B⇓ shows voltage dependence of ICa density recorded in control cells (left, open circles) and in Taxol-treated myocytes (right, open triangles). Both curves were not statistically different, as recently reported.11 After 1 μmol/L ISO addition, both groups of cells responded in a similar manner (control, left, closed circles; and Taxol-treated cells, right, closed triangles).
Effect of Blocking Adenylyl Cyclase
Our findings using colchicine are quite similar to the effects of β-adrenergic stimulation, with increased ICa and [Ca2+]i transient, as well as acceleration of [Ca2+]i transient decay time. In this regard, it has been reported in rat cerebral cortex that free tubulin stimulates adenylyl cyclase.12 To analyze the involvement of adenylyl cyclase in the signal pathway under our experimental conditions, we tested the effect of the adenylyl cyclase inhibitor 2′d3′AMP on the whole-cell ICa enhancement after colchicine treatment. Cardiac myocytes were incubated with colchicine as earlier, and ICa was analyzed by the patch-clamp technique. Figure 7A⇓ shows ICa density at 0 mV over time after whole-cell configuration achievement in the presence of 200 μmol/L 2′d3′AMP in the internal solution, on a control myocyte (open circles) and a colchicine-treated myocyte (closed circles). Shortly after breaking into the cell, ICa was markedly higher on the colchicine-treated cell (Figure 7A⇓a) but became similar to the control myocyte with adenylyl cyclase dialysis (Figure 7A⇓b). Data are summarized in Figure 7B⇓; whereas 2,d3′AMP had no statistically significant effect on the control myocytes, it did block the ICa increase after colchicine treatment.
The present data show that microtubules are able to modulate EC coupling by modulating Ca2+ current and consequently [Ca2+]i transient. Particularly, microtubule disassembly induces an increase in both ICa and [Ca2+]i transient and accelerates both ICa inactivation and [Ca2+]i transient decay. Thus, the effects of disrupting microtubules appear rather similar to β-adrenergic stimulation. Moreover, after microtubule disruption, the effects of ISO on ICa and [Ca2+]i transient are blunted. Furthermore, by blocking adenylyl cyclase activity we can reverse colchicine effects. These results show that colchicine and ISO actions share an inotropic effect involving activation of adenylyl cyclase.
Since 1993, Tsutsui et al1 13 and Tagawa et al4 have demonstrated that colchicine treatment can increase contraction in hypertrophied cardiac myocytes in which contraction was decreased. The suggested mechanism was a decrease in stiffness and viscosity, consequent to the decrease in microtubules.2 A passive effect of the microtubule network is a plausible mechanism; however, the effect that these authors observed with colchicine treatment in hypertrophied cells might also be explained by the increase in ICa and SR Ca2+ release that we report in this study, although they fail to see an effect on normal cells.1 We do not discard, however, a concomitant action on the cellular viscous load. In fact, these authors also observed a decrease in contraction after microtubule stabilization with Taxol1 that so far can only be explained by an increase in cell viscosity and stiffness, because we failed to observe significant modification on ICa by Taxol (Figure 6⇑). Returning to the cardiac effect of microtubule disruption, it has been shown that colchicine treatment accelerates the beating frequency in neonatal cardiac cells.14 15 This effect could also be explained by the increase in ICa that we report. However, after shorter periods of colchicine treatment than ours, some authors did not find an effect on contraction in either control or hypertrophied myocytes.16 17
Cytoskeleton, and in particular microtubules, can bind several proteins, probably including ionic channels.18 Galli and DeFelice8 have found that colchicine modifies Ca2+ channel inactivation, but a direct effect of colchicine was discarded, because they failed to see an effect on excised patches. In a similar way, we observed that the fast component of ICa inactivation is accelerated in colchicine-treated cells. This first inactivation phase is dependent on Ca2+. Because sarcolemmal Ca2+channel and RyR are close to each other in a restricted space,19 a bigger Ca2+release by the RyR could increase the Ca2+-induced inactivation of the sarcolemmal Ca2+channel.20 After colchicine treatment, as a result of a bigger triggering ICa, we obtained a larger [Ca2+]i transient. The faster inactivation of ICa that we observed (see Figure 2A⇑) can be the result of the increase in [Ca2+]i transient induced after colchicine treatment (Figure 1⇑).
To comprehend our results, one might consider the peculiar feature of microtubules. Microtubules are formed by the self-assembly of α,β-tubulin dimers that polymerize and depolymerize dynamically.18 α,β-Tubulin dimer is a GTP-binding protein with amino acid homologies and significant functional similarities to the G proteins.21 Moreover, in the neuronal system, relatively high-affinity binding between dimeric tubulin and the α subunits of Gs, Gi1, and Gq have been reported,22 23 whereas assembled microtubules bind G protein quite weakly. It has also been observed that the tubulin dimer, also called free tubulin, causes stimulation of adenylyl cyclase in rat cerebral cortex membranes.12 This effect results from a direct transfer of nucleotide from the exchangeable GTP-binding site of tubulin to the Gs protein.12 24 These findings strongly suggest that the increase in ICa and [Ca2+]i transient presently reported in cardiac myocytes might be due to the activation of adenylyl cyclase by tubulin dimers. As a matter of fact, in cardiac tissues, cAMP-dependent activation of PKA has various effects on EC coupling. Among them, a phosphorylation of the L-type Ca2+ channel induces an increase in ICa (see Figure 1⇑). There is also a phosphorylation of phospholamban that will result in an acceleration of the SR Ca2+ pump (see Figure 3B⇑) and an increase in the SR Ca2+ load (Figure 4A⇑). Moreover, RyRs can also be phosphorylated, modulating in this way their sensitivity to Ca2+.25 We thus suspected that microtubule disruption increases both ICa and SR Ca2+ release by increasing free tubulin, which leads to adenylyl cyclase activation.
This hypothesis was supported by the observation that, after microtubule disruption, the β-adrenergic stimulatory effect of ICa and [Ca2+]i transient are blunted. The lack of additivity suggests that both the β-adrenergic and the colchicine effects occur through a similar pathway. Two previous experimental reports are in line with this hypothesis. First, after microtubule disruption by colchicine, the cAMP-dependent Cl− secretion is no longer sensitive to forskolin, whereas the Cl−- secretory response of colonic epithelia is still Ca2+-dependent.10 Second, the forskolin-induced relocation of CFTR on T84 cells is blocked by the microtubule-disrupting agent nocodazole.26 Furthermore, Limas and Limas27 have suggested that microtubules can fix β-adrenergic receptors in the membranes and that after colchicine treatment, the fraction of β-receptors in internal vesicles compared with sarcolemma was increased. This possibility could explain the decrease in ISO effect that we observed; however, it cannot account for the increase in ICa and Ca2+ transient after colchicine treatment. In fact, their observation could rather be interpreted as β-adrenergic receptor endocytosis, a secondary phase of agonist-independent phosphorylation and receptor desensitization mediated by PKA.28 In our experiments, if free tubulin activates adenylyl cyclase and consequently PKA, as well as ISO application, it would be possible that once PKA is activated by increase in tubulin dimer and phosphorylation is induced, further activation of PKA by ISO would seem ineffective. We thus repeated the experiments in the presence of 2′d3′AMP, an inhibitor of the adenylyl cyclase that interacts with the purine site of the cyclase.29 In line with our hypothesis, blocking the adenylyl cyclase reversed the microtubule disruption effect on ICa (Figure 7⇑).
In conclusion, we show that microtubules can modulate calcium signaling in cardiac cells. We suggest that the microtubule disruption–increased level of soluble tubulin dimers activates Gs protein and leads to activation of the adenylyl cyclase. This effect triggers the cascade that leads to an increase in ICa and [Ca2+]i transient and, in the end, to an increase in contraction. Moreover, this mechanism could help to explain, at least in part, the alterations in heart contraction observed in several pathologies in which changes in microtubules are reported.1 30
This work has been supported by a grant from Association Française contre les Myopathies (MNM 6086, 1997). We are indebted to Jean-Pierre Bénitah for help with the manuscript. We thank W. Jonathan Lederer and Paco Lorente for comments on the paper and Eva Le Charpentier for technical assistance. We also thank Cecilia Frederick for advice on the immunolabeling experiments. A.M. Gómez is supported by the Centre National de la Recherche Scientifique.
- Received September 1, 1999.
- Accepted October 19, 1999.
- © 2000 American Heart Association, Inc.
Tsutsui H, Ishihara K, Cooper G 4th. Cytoskeletal role in the contractile dysfunction of hypertrophied myocardium. Science. 1993;260:682–687.
Tagawa H, Wang N, Narishige T, Ingber DE, Zile MR, Cooper G 4th. Cytoskeletal mechanics in pressure-overload cardiac hypertrophy. Circ Res. 1997;80:281–289.
Gómez 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.
Tagawa H, Rozich JD, Tsutsui H, Narishige T, Kuppuswamy D, Sato H, McDermott PJ, Koide M, Cooper G 4th. Basis for increased microtubules in pressure-hypertrophied cardiocytes. Circulation. 1996;93:1230–1243.
Sato H, Nagai T, Kuppuswamy D, Narishige T, Koide M, Menick DR, Cooper G 4th. Microtubule stabilization in pressure overload cardiac hypertrophy. J Cell Biol. 1997;139:963–973.
Palmer BM, Valent S, Holder EL, Weinberger HD, Bies RD. Microtubules modulate cardiomyocyte beta-adrenergic response in cardiac hypertrophy. Am J Physiol. 1998;275:H1707–H1716.
Fuller CM, Bridges RJ, Benos DJ. Forskolin- but not ionomycin-evoked Cl− secretion in colonic epithelia depends on intact microtubules. Am J Physiol. 1994;266:C661–C668.
Tsutsui H, Tagawa H, Kent RL, McCollam PL, Ishihara K, Nagatsu M, Cooper G 4th. Role of microtubules in contractile dysfunction of hypertrophied cardiocytes. Circulation. 1994;90:533–555.
Lampidis TJ, Kolonias D, Savaraj N, Rubin RW. Cardiostimulatory and antiarrhythmic activity of tubulin-binding agents. Proc Natl Acad Sci U S A. 1992;89:1256–1260.
Janmey PA. The cytoskeleton and cell signaling: component localization and mechanical coupling. Physiol Rev. 1998;78:763–781.
Lederer WJ, Niggli E, Hadley RW. Sodium-calcium exchange in excitable cells: fuzzy space. Science. 1990;248:283.
Sham JS, Cleemann L, Morad M. Functional coupling of Ca2+ channels and ryanodine receptors in cardiac myocytes. Proc Natl Acad Sci U S A. 1995;92:121–125.
Wang N, Yan K, Rasenick MM. Tubulin binds specifically to the signal-transducing proteins, Gs alpha and Gi alpha 1. J Biol Chem. 1990;265:1239–1242.
Popova JS, Garrison JC, Rhee SG, Rasenick MM. Tubulin, Gq, and phosphatidylinositol 4,5-bisphosphate interact to regulate phospholipase Cbeta1 signaling. J Biol Chem. 1997;272:6760–6765.
Valdivia HH, Kaplan JH, Ellis-Davies GC, Lederer WJ. Rapid adaptation of cardiac ryanodine receptors: modulation by Mg2+ and phosphorylation. Science. 1995;267:1997–2000.
Tousson A, Fuller CM, Benos DJ. Apical recruitment of CFTR in T-84 cells is dependent on cAMP and microtubules but not Ca2+ or microfilaments. J Cell Sci. 1996;109:1325–1334.
Limas C, Limas CJ. Disparate effects of colchicine on thyroxine-induced cardiac hypertrophy and adrenoceptor changes. Circ Res. 1991;68:309–313.
Bohm SK, Grady EF, Bunnett NW. Regulatory mechanisms that modulate signalling by G-protein-coupled receptors. Biochem J. 1997;322:1–18.
Babenko A, Vassort G. Enhancement of the ATP-sensitive K+ current by extracellular ATP in rat ventricular myocytes: involvement of adenylyl cyclase-induced subsarcolemmal ATP depletion. Circ Res. 1997;80:589–600.
Watkins SC, Samuel JL, Marotte F, Bertier-Savalle B, Rappaport L. Microtubules and desmin filaments during onset of heart hypertrophy in rat: a double immunoelectron microscope study. Circ Res. 1987;60:327–336.