During activation of cardiac muscle, Ca influx across the cell membrane causes local Ca release from adjacent ryanodine receptors (e.g. Wier & Balke, 1999); the global Ca transient is the sum of these independent, localised releases. This requires that Ca influx occurs adjacent to the ryanodine receptors. Immunohistochemical studies of cardiac ventricular myocytes have shown ryanodine receptors and L-type Ca channels, which provide the main trigger for Ca release, concentrated and co-localised at the t-tubule (e.g. Scriven et al. 2000), but the localisation of other proteins has been more controversial. Furthermore, immunolocalisation of a protein does not allow localisation of its function, which may be modulated by other factors, and quantification is difficult. We have, therefore, used osmotic shock to detubulate isolated ventricular myocytes and thus investigate the distribution of membrane currents between the t-tubule and surface membranes.

To induce detubulation, isolated rat ventricular myocytes are exposed to 1.5 M formamide for 15 min, and then resuspended in control Tyrode solution (Kawai et al. 1999). This results in a 30 % decrease in cell capacitance and loss of the t-tubules, visualised by staining the cell membrane with di-8-ANEPPS. However, if the cells are stained with di-8-ANEPPS before treatment with formamide, irregular fluorescence can be observed in the cell following formamide treatment, and if FITC-labelled dextran is included in the bathing solutions during detubulation, localised FITC fluorescence can be observed within the cell following detubulation. These data suggest that the t-tubules uncouple from the surface membrane and reseal within the cell during formamide treatment.

Detubulation causes an 87 % decrease in the amplitude of the L-type Ca current (I_Ca), with no change in its time course or voltage dependence; this agrees with immunohistochemical studies showing concentration of this channel in the t-tubules. Detubulation also causes almost complete loss of Ni-sensitive Na-Ca exchange current elicited by ramp clamps. Because immunohistochemical studies of the distribution of the Na-Ca exchanger have given inconsistent results (see Blaustein & Lederer, 1999), this was investigated further using confocal microscopy to monitor Ca efflux from ventricular myocytes, by bathing cells in a solution containing the fluorescent Ca indicator fluo-3 during the application of caffeine. In control cells, application of caffeine is accompanied by a rapid rise of extracellular Ca, as Ca is extruded from the cell. This rise is markedly reduced following detubulation (the amplitude of the rise of intracellular [Ca] is unchanged, although its rate of decline is slowed), and the residual rise observed in detubulated cells is abolished by the Ca-ATPase inhibitor carboxyeosin. These data suggest that in rat ventricular myocytes 87 % of I_Ca and almost all Na-Ca exchange activity are concentrated in the t-tubules. This is consistent with the observation that in control cells, electrical stimulation results in a rapid, synchronous increase of intracellular [Ca], whereas in detubulated cells Ca initially rises at the cell periphery followed by sarcoplasmic reticulum-dependent propagation into the cell interior.

In contrast, the density of other currents – I_Na, I_K and I_K1 – does not change following formamide treatment of rat ventricular myocytes, suggesting that they are more homogeneously distributed between the t-tubule and surface membranes. Steady-state current (I_SS) density decreases slightly following detubulation, so it may be concentrated in the t-tubules, compatible with recent work showing that TASK-1, one of the channels that might underlie I_SS, is found at a high concentration in the t-tubules (Jones et al. 2002).

It is unlikely that direct effects of formamide can account for these data, because exposure of rat atrial myocytes, which lack t-tubules, to 1.5 M formamide for 15 min has no significant effect on cell capacitance, I_Ca, Ca efflux during application of caffeine, the configuration of the action potential or the Ca transient.

Thus it appears that the major trans-sarcolemmal Ca flux pathways (I_Ca and Na-Ca exchange) are concentrated within the cardiac t-tubules, which therefore play an important and specialised role in excitation-contraction coupling in the heart.

This work was supported by The Wellcome Trust and British Heart Foundation.