Introduction

The osmolarity of body fluids is normally very tightly regulated so that most cells do not experience changes in osmotic pressure under physiological conditions1, but osmotic changes can occur in pathological states. Extensive evidence indicates that cardiac cell swelling occurs under abnormal conditions such as ischemia and reperfusion2, 3, 4, 5, 6. It has been shown that cell swelling can modulate the function of a number of membrane channels and ion transporters. These effects will lead to changes in the cardiac electrical activity and may contribute to arrhythmogenesis under pathological conditions7, 8. Swelling can also reduce the efficacy of some antiarrhythmic drugs and render them less effective during ischemia/reperfusion9, 10, 11.

The cardiac L-type Ca2+ channel (ICa,L) plays a critical role in cardiac excitability and in excitation-contraction coupling. However, the effect of cell swelling on cardiac ICa,L remains controversial. For example, some studies have shown clear increases in ICa,L12, 13, 14, while others have shown a decrease in ICa,L15, 16 or no effect10, 17. Some authors have even demonstrated biphasic effects on ICa,L, an increase followed by a more sustained decrease18. Previous reports indicated that osmotic swelling increases ICa,L in neonatal rat ventricular myocytes14, but decreases ICa,L in adult rat ventricular myocytes16. But these reports did not explore the mechanism underlying the changes of ICa,L. Thus it is important to unveil the details of osmotic regulation on ICa,L and the underlying mechanisms, which may provide clues to develop novel therapeutic approaches useful for pathological conditions that induce cell swelling. Therefore in this study we have investigated the consequences and mechanisms by which cell swelling modulates ICa,L.

Materials and methods

Solutions and drugs

The Tyrode solution contained (in mmol/L): 135 NaCl, 5.4 KCl, 1.8 CaCl2, 1 MgCl2, 0.33 NaH2PO4, 10 glucose, and 10 HEPES, pH 7.4 with NaOH. The KB solution contained (in mmol/L): 70 KOH, 40 KCl, 1 MgCl2, 20 KH2PO4, 20 taurine, 50 glutamic acid, 0.5 EGTA, 10 glucose, and 10 HEPES, pH 7.4 with KOH. The pipette solution used to record ICa,L contained (in mmol/L): 115 CsCl, 20 TEA-Cl, 5 MgATP, 0.1 Na2-GTP, 10 phosphocreatine, 1 CaCl2, 10 EGTA, and 10 HEPES, pH 7.2 with CsOH. The standard isosmotic bath solution (∼300 mosmol/L) used to record ICa,L was composed of (in mmol/L) 88 choline chloride, 5.4 CsCl, 1 MgCl2, 1.8 CaCl2, 10 glucose, 10 HEPES, and 96 D-mannitol, pH 7.4 with CsOH19. D-mannitol concentrations were modified to be 15 mmol/L (∼220 mosmol/L)19, 20 and 177 mmol/L (∼380 mosmol/L) for the standard hyposmotic solution and hyperosmotic solution, respectively. The osmolarity of the solutions was calculated by the formula: mosmol/L=∑φiniCi, where Φ is the osmotic coefficient; n is the number of particles into which a molecule dissociates; C is the molar concentration of the solute; the index i represents the identity of a particular solute21.

Collagenase type I was obtained from Gibico (GIBCO TM, Invitrogen, Paisley, UK). Isoproterenol hydrochlordie, forskolin, H-89, phorbol 12-myristate 13-acetate (PMA), Bisindolylmaleimide IV (BIM), S-nitroso-N-acetylpenicillamine (SNAP), Rp-8-Br-PET-cGMPS, TEA-Cl, CsCl, CsOH, MgATP, and Na2-GTP were purchased from Sigma Chemical (Saint Louis, MO, USA). Forskolin, H-89, PMA, and BIM were dissolved in DMSO and diluted into the test solution to obtain the required concentration immediately before use. The final concentration of DMSO in the test solution did not exceed 0.1%. We confirmed that DMSO at this concentration had no effect on ICa,L. H-89 and Rp-8-Br-PET-cGMPS were dissolved into the pipette solution at the final concentration indicated in the text.

Isolation of myocytes

Adult female Sprague-Dawley rats (220–260 g) were anesthetized with pentobarbital sodium (30 mg/kg, ip) 20 min after an intraperitoneal injection of 2000 U heparin. Hearts were excised rapidly and perfused retrogradely on a langendorff apparatus with Ca2+-free Tyrode solution for 5 min; subsequently, the perfusate was switched to an enzyme-containing solution [0.3 g/L collagenase type I, 0.5 g/L bovine serum albumin (BSA) in the same solution] for 25–30 min. The perfusate was finally changed to a KB solution for 5 min. These perfusates were bubbled with 100% O2 and maintained at 37 °C. The ventricles were cut into small chunks and gently agitated in KB solution. The cells were filtered through nylon mesh and stored in KB solution at 4 °C until use. All procedures met the Guide for the Care and Use of Laboratory Animals of Hubei Province.

Current recordings

Before patch clamp recording, cells were transferred onto a coverslip in a recording chamber that was mounted on the stage of an inverted microscope (Zeiss, Germany). The recording chamber was continuously perfused with bath solution. Conventional whole-cell patch clamp was performed at room temperature (20–22 °C) using an Axopatch 200A amplifier and Digidata 1200 A/D converter (Axon instruments, Union city, CA, USA). The patch electrodes had resistances of 1.5–2.5 MΩ when filled with the pipette solutions. Capacitance and series resistances were adjusted to obtain minimal contribution of the capacitive transients. A 50%–80% compensation of the series resistance was usually achieved without ringing. Currents were filtered at 2 kHz, digitized at 10 kHz and stored on a computer hard disk for further analysis.

To measure ICa,L, cells were voltage clamped at a holding potential (HP) of -80 mV. A 50-ms prepulse to -40 mV was used to inactivate ICa,TTX22, and ICa,L was then elicited by a 200-ms depolarizing pulse from -40 to 0 mV. Stimulation frequency was 0.2 Hz. The amplitude of ICa,L was measured as the difference between peak inward current and the current remaining at the end of the 200-ms voltage-clamp pulse. For the experiments testing drug effects, cells were exposed to drugs and the effects were monitored by recording ICa,L every 5 s. For the steady-state activation protocol, HP was −80 mV. Following a 50-ms prepulse to −40 mV, ICa,L was elicited by 200-ms pulses to different test potentials (between −40 and +50 mV in 10 mV increments). For the steady-state inactivation protocol, HP was −40 mV. 2000-ms conditioning prepulses (between −50 and 0 mV in 10 mV increments) were applied followed by a 200-ms test pulse to 0 mV. Stimulation frequency was 0.2 Hz. These measurements were made after a minimal time of 20 min after rupture of the patch to minimize the contribution of time-dependent shifts of steady-state gating parameter measurements.

Data analysis

Data were analyzed using pClamp 9.0 (Axon instruments) and Sigmaplot 7.0 (SPSS Inc, Chicago, IL, USA). Current density was calculated by dividing the current amplitude by the cell capacitance. The average cell capacitance was 183±48 pF (n=82 cells). Data from steady-state activation and inactivation relationship of ICa,L were fitted to the Boltzmann equation: Y=1/{1+exp[(Vm–V1/2)/k]}, where Vm is the membrane potential, V1/2 is the half-activation or half-inactivation potential, and k is the slope factor. For steady-state activation curves, Y stands for the relative conductance. The chord conductance was calculated using the ratio of the current to their electromotive for potential in individual current-voltage relationships. Then these conductances were normalized to their individual maximal conductance. For steady-state inactivation curves, Y represents the relative current (ICa,L/ICa,L max).

An unpaired Student's t-test was perfomed for the comparison between two groups. Significance was tested by one-way analysis of variance (ANOVA) followed by Tukey test if multiple comparisons were made. All values were expressed as mean±SD, and the number of cells (n) in each group was given. P<0.05 was considered to be statistically significant.

Results

Biphasic effects of hyposmotic challenge on ICa,L

Cells were continuously perfused with an isosmotic solution. Experiments were performed 10 min after establishment of the whole-cell configuration. After the peak inward current had reached a steady state value, the isosmotic external solution was replaced by a hyposmotic external solution.

Figure 1 shows typical traces of ICa,L before and after perfusion with hyposmotic solution. Application of hyposmotic solution always led to a biphasic changes of ICa,L. The amplitude of ICa,L increased transiently, then decreased gradually and reached a new steady state value (Figure 1), where the current was able to last for at least 15 min at this level. Furthermore, the decrease of ICa,L induced by long hyposmotic challenge was accompanied with slowing of the inactivation of ICa,L (Figure 1B inset). However, the time course was variable. ICa,L reached a maximum within 20 s–3 min and a new stable value within 5–10 min after perfusion with hyposmotic solution. After 10 min of hyposmotic challenge, 8 cells were perfused with isosmotic solution. But the effects of hyposmotic solution on ICa,L could not be reversed (Figure 1). On average, current density of ICa,L was increased to 119.1%±6.1% of control after short exposure (within 3 min) (isosmotic: −4.71±0.71 pA/pF; short hyposmotic: −5.58±0.68 pA/pF, n=15, P<0.01), decreased to 78.1%±11.0% of control after long exposure (10 min) to hyposmotic solution (long hyposmotic: −3.65±0.64 pA/pF, n=15, P<0.01 vs both groups), and recovered to 82.3%±9.6% of control after hyposmotic solution was washed out (washout: −3.89±0.74 pA/pF, n=8, P>0.05 vs long hyposmotic) (Figure 1C).

Figure 1
figure 1

Effects of hyposmotic swelling on ICa,L in rat ventricular myocytes. (A) Time course of peak current measurements from a cell under isosmotic (Iso), hyposmotic (Hypo), and washout conditions. Currents were recorded every 5 s. (B) Representative current traces recorded at different time under different conditions. Inset, currents were normalized to the peak amplitudes and traces were superimposed. Long exposure to hyposmotic solution (3) causes reduction of peak ICa,L and slowing of ICa,L inactivation. These changes of ICa,L could not be reversed by washout (4). (C) The mean current densities of ICa,L under different conditions. Values are Mean±SD. n=8–15. bP<0.05, cP<0.01.

Full size image

To confirm that the observed changes of ICa,L were due to the osmolarity change, we also studied the effects of hyperosmotic solution on ICa,L in 7 cells. Hyperosmotic challenge increased the amplitude of ICa,L and accelerated the inactivation of ICa,L (Figure 2). Current density of ICa,L was increased to 126.8.1%±11.1% of control after application of hyperosmotic solution (isosmotic: -4.93±0.60 pA/pF; hyperosmotic: −6.25±0.96 pA/pF, n=7, P<0.05) and recovered to 105.5%±11.8% of control after hyperosmotic solution was washed out (washout: −5.19±0.61 pA/pF, n=6, P>0.05 vs both groups) (Figure 2C).

Figure 2
figure 2

Effects of hyperosmotic challenge on ICa,L in rat ventricular myocytes. (A) Time course of peak current measurements from a cell under isosmotic (Iso), hyperosmotic (Hyper), and washout conditions. Currents were recorded every 5 s. (B) Representative current traces recorded at different time under different conditions. Inset, currents were normalized to the peak amplitudes and traces were superimposed. Hyperosmotic challenge (2) causes enhancement of peak ICa,L and acceleration of ICa,L inactivation. These changes of ICa,L was largely reversible (3). (C) The mean current densities of ICa,L under different conditions. Values are Mean±SD. n=6–7. bP<0.05.

Full size image

We further examined the effects of hyposmotic swelling on the steady-state activation and inactivation curves of ICa,L. As shown in Figure 1A, ICa,L was increased gradually and could not reach a stable state during short hyposmotic challenge. Thus we only studied the effects of long hyposmotic challenge (10 min) on both curves. Figure 3A shows typical examples of ICa,L recordings obtained with incremental test pulses of the steady-state activation protocol. We again observed that hyposmotic swelling causes reduction of peak ICa,L and slowing of ICa,L inactivation (Figure 3A). The voltage dependence of activation of ICa,L was unaltered (isosmotic: V1/2=−16.51±1.97 mV, k=6.38±1.65 mV, n=21; hyposmotic: V1/2=−16.85±2.13 mV, k=7.24±1.79 mV, n=6, both P>0.05 vs isosmotic) (Figure 3B). However, the voltage dependence of inactivation of ICa,L was slightly shifted in the depolarizing direction (isosmotic: V1/2=−36.78±2.25 mV, k=−6.31±1.45 mV, n=25; hyposmotic: V1/2=−33.91±2.23 mV, P<0.05 vs isosmotic, k=−7.26±1.44 mV, P>0.05 vs isosmotic, n=6) (Figure 3D).

Figure 3
figure 3

Effects of long hyposmotic challenge on steady-state activation and inactivation curves for ICa,L. (A) Examples of ICa,L recordings elicited according to the activation protocol. (B) Comparison of mean activation curves under iso-smotic (open circle) and hyposmotic condition (filled circle). Lines represent the fit of data to a Boltzmann distribution function. Hyposmotic swelling does not change the activation curve of ICa,L. (C) Examples of ICa,L recordings elicited according to the inactivation protocol. (D) Comparison of mean inactivation curves under isosmotic (open circle) and hyposmotic condition (filled circle). Lines represent the fit of data to a Boltzmann distribution function. Hyposmotic swelling shifts slightly the inactivation curve of ICa,L in depolarizing direction.

Full size image

Protein kinases, such as PKA, PKC, and PKG, may modulate function of ICa,L channel23, 24. Some studies have reported that hyposmotic swelling modulates function of ion channels by changing activities of protein kinases, such as PKC, tyrosine protein kinase8, 17, 25, 26. To examine mechanism underlying the hyposmotic swelling-induced biphasic changes of ICa,L, we tested whether PKA, PKC, or PKG were involved in the signal transduction pathway, respectively.

PKA signaling pathway

As shown in Figure 4A, application of PKA agonist foskolin (1 μmol/L) under isosmotic conditions significantly increased ICa,L by 41.3%±21.6% (control: −5.02±0.66 pA/pF; foskolin: −7.05±1.26 pA/pF, n=7, P<0.05). Switching the perfusate from isosmotic to hyposmotic solution in the presence of foskolin reverted the prestimulated ICa,L to 90.7%±30.9% of control (hyposmotic+foskolin: -4.65±1.98 pA/pF, n=7, P<0.05 vs foskolin) (Figure 4A). In another set of experiments, cells were treated with foskolin after long hyposmotic exposure (10 min). Again, peak ICa,L was significantly decreased to 79.3%±7.1% of control when cells were exposed to hyposmotic solution (isosmotic: -5.00±0.32 pA/pF; hyposmotic: −3.97±0.52 pA/pF, n=6, P<0.05). Application of foskolin under hyposmotic conditions significantly reversed the reduced ICa,L to 142.1±15.4% of control (hyposmotic+foskolin: −7.11±1.00 pA/pF, n=6, P<0.01 vs both groups) (Figure 4B). Similar results were observed in those experiments using β-adrenergic receptor agonist isoproterenol (1 μmol/L) (data not shown).

Figure 4
figure 4

Interactive effects of protein kinase A activator foskolin and hyposmotic swelling on ICa,L. (A) Hyposmotic swelling decreases the stimulated ICa,L by foskolin. Left panel shows the representative example of currents recordings in isomostic (Iso), isomostic plus 1 μmol/L foskolin (Iso+FSK), and hyposmotic plus 1 μmol/L foskolin (Hypo+FSK) solution. Right panel shows the mean current density of ICa,L under different conditions. Values are Mean±SD. n=7. bP<0.05. (B) The decreased ICa,L induced by hyposmotic swelling is reversed by foskolin. Left panel shows the representative example of currents recordings in isomostic (Iso), hypomostic (Hypo), and hyposmotic plus 1 μmol/L foskolin (Hypo+FSK) solution. Right panel shows the mean current density of ICa,L under different conditions. Mean±SD. n=6. bP<0.05, cP<0.01. Inset, currents were normalized to the peak amplitudes and traces were superimposed.

Full size image

To further study the role of PKA in the swelling-induced biphasic changes of ICa,L, selective PKA inhibitor H-89 (10 μmol/L) was applied. Cells were dialyzed with a pipette solution containing H-89 for 20 min before the solution change. Treatment with H-89 did not change the amplitude of basal ICa,L under isosmotic conditions (control: −6.48±1.76 pA/pF, n=44; H-89: −7.51±1.37 pA/pF, n=10, P>0.05). In the presence of H-89, hyposmotic swelling still significantly decreased ICa,L to 79.4±15.2% of control (isosmotic: −7.51±1.37 pA/pF; hyposmotic: −5.82±0.79 pA/pF, n=10, P<0.01). However, H-89 abolished the transient increase of ICa,L induced by hyposmotic swelling in 9 of 10 cells. The biphasic changes of ICa,L was only observed in 1 of 10 cells.

PKG signaling pathway

Application of NO donor SNAP (1 μmol/L) under isosmotic conditions did not have significant effects on ICa,L (control: -3.50±0.24 pA/pF; SNAP: -3.28±0.22 pA/pF, n=7, P>0.05). Switching the perfusate from isosmotic to hyposmotic solution in the presence of SNAP significantly decreased ICa,L to 78.1%±13.8% of control (hyposmotic+SNAP: -2.75±0.30 pA/pF, n=7, P<0.01 vs both groups) (Figure 5A). In another set of experiments, cells were treated with SNAP after long hyposmotic exposure (10 min). Peak ICa,L was significantly decreased after long exposure to hyposmotic solution (isosmotic: -3.53±0.47 pA/pF; hyposmotic: -2.46±0.22 pA/pF, n=4, P<0.05). Application of SNAP under hyposmotic conditions did not have further effects on ICa,L (hyposmotic+SNAP: -2.46±0.33 pA/pF, n=4, P>0.05 vs SNAP) (Figure 5A).

Figure 5
figure 5

Protein kinase G is not involved in swelling-dependent modulation of ICa,L. (A) NO donor SNAP does not interference with swelling-dependent modulation of ICa,L. The mean current density of ICa,L under different conditions is shown. Data in left panel and right panel are from 7 and 4 cells, respectively. Mean±SD. bP<0.05, cP<0.01. (B) PKG inhibitor Rp-8-Br-PET-cGMPS do not block swelling-induced biphasic changes of ICa,L. 1 μmol/L Rp-8-Br-PET-cGMPS is applied via pipette solution. The mean current density of ICa,L under different conditions is shown. Mean±SD. n=5. bP<0.05, cP<0.01.

Full size image

In another 5 cells, selective PKG inhibitor Rp-8-Br-PET-cGMPS (1 μmol/L) was applied via the pipette solution. Cells were dialyzed with the pipette solution for 15 min before the solution change. Treatment with Rp-8-Br-PET-cGMPS did not change the amplitude of basal ICa,L under isosmotic conditions (control: -6.48±1.76 pA/pF, n=44; Rp-8-Br-PET-cGMPS: -5.28±0.59 pA/pF, n=5, P>0.05 vs control). Under such conditions, hyposmotic swelling still caused the biphasic changes of ICa,L (Figure 5B). These data suggest that PKG system is not involved in swelling-dependent modulation of ICa,L.

PKC signaling pathway

Involvement of PKC in the hyposmotic swelling-induced biphasic changes of ICa,L was evaluated with the PKC activator PMA (100 nmol/L). Application of PMA under isosmotic conditions significantly decreased ICa,L to 80.8%±13.8% of control (control: -4.05±0.40 pA/pF; PMA: -3.28±0.70 pA/pF, n=7, P<0.05). Interesting, switching the perfusate from isosmotic to hyposmotic solution in the presence of PMA reversed the decreased ICa,L to 114.3%±8.7% of control (hyposmotic+PMA: -4.70±0.40 pA/pF, n=5, P<0.01 vs PMA) (Figure 6). In another set of experiments, cells were treated with PMA after long hyposmotic exposure (10 min). Hyposmotic swelling significantly decreased ICa,L to 64.8%±14.8% of control (isosmotic: -5.95±0.22 pA/pF; hyposmotic: -3.66±0.54 pA/pF, n=3, P<0.01). Subsequent application of PMA under hyposmotic conditions further decreased ICa,L to 41.8%±11.4% of control (hyposmotic+PMA: -2.45±0.57 pA/pF, n=3, P<0.05 vs hyposmotic) (Figure 6C).

Figure 6
figure 6

Interactive effects of protein kinase C activator PMA and hyposmotic swelling on ICa,L. (A) Time course of peak current measurements from a cell after application of 100 nmol/L PMA under isosmotic and hyposmotic conditions. Currents were recorded every 5 s. (B) Representative current traces recorded at different time under different conditions. Inset, currents were normalized to the peak amplitudes and traces were superimposed. PMA decreases ICa,L and slows the inactivation of ICa,L. Subsequent application of hyposmotic solution plus PMA reverses PMA-induced decrease of ICa,L. (C) The mean current density of ICa,L under different conditions. Values are Mean±SD. n=3–7. bP<0.05, cP<0.01.

Full size image

To further evaluate the involvement of PKC in the biphasic changes of ICa,L induced by hyposmotic swelling, we treated cells with specific PKC inhibitors BIM (2 μmol/L) before and during exposure to hyposmotic solution. Application of the PKC inhibitor BIM did not have any detectable effects on ICa,L under isosmotic conditions. However, BIM prevented the hyposmotic swelling-induced biphasic changes of ICa,L. The transient increase of ICa,L was abolished, and ICa,L was decreased slightly by hyposmotic swelling (Figure 7).

Figure 7
figure 7

Protein kinase C inhibitor BIM prevents the swelling-induced biphasic changes of ICa,L. (A) Time course of peak current measurements from a cell after application of 2 μmol/L BIM under isosmotic and hyposmotic conditions. Currents were recorded every 5 s. (B) Representative current traces recorded at different time under different conditions. Inset, currents were normalized to the peak amplitudes and traces were superimposed. (C) The mean current density of ICa,L under different conditions. Values are Mean±SD. n=7–10. aP>0.05 vs isosmotic.

Full size image

Discussion

Extensive evidence shows that cardiac cells undergo significant swelling during ischaemia and reperfusion2, 3, 4, 5, 6. During myocardial ischemia, cell metabolites such as inorganic phosphates and lactate accumulate intracellularly and to a certain degree extracellularly. Cardiac myocytes are exposed to a hyposmotic challenge and cell swelling occurs7. This swelling may be more marked on reperfusion when the hyperosmotic extracellular milieu is replaced by isosmotic blood4, 6, 27. It has been shown that cell swelling can modulate the function of a number of membrane channels and ion transporters. These effects will lead to changes in the cardiac electrical activity and may contribute to arrhythmogenesis under pathological conditions7, 8. Many studies have reported that reperfusing hearts with a hyperosmotic perfusate reduces the incidence of reperfusion-induced arrhythmias and results in better recovery of cardiac function2, 3, 27, 28. Moreover, swelling can also reduce the efficacy of some antiarrhythmic drugs and render them less effective during ischemia/reperfusion9, 10, 11. A full understanding of effects and the mechanisms by which cell swelling modulates function of ion channels and transporters will provide important insights into the electrical effects of myocardial ischemia. In this study we investigated the consequences and mechanisms by which hyposmotic swelling modulates ICa,L in rat ventricular myocytes.

There are controversies in the literature regarding how cell swelling can modulate cardiac ICa,L channel. For example, ICa,L increases in neonatal rat ventricular myocytes14 and rabbit sinoatrial node and atrial cells12, 13, remains unaltered in canine ventricular myocytes17, decreases or remains unaltered in guinea-pig ventricular myocytes10, 15. Brette et al had reported that in rat ventricular myocytes osmotic swelling decreases ICa,L16. On the other hand, Li et al had found that in rabbit ventricular myocytes hyposmotic swelling induces biphasic changes, an increase followed by decrease, of ICa,L under perforated whole-cell patch-clamp conditions18. In the present study, we found that osmotic swelling also induced biphasic changes of ICa,L, a transient increase followed by sustained decrease, in rat ventricular myocytes (Figure 1). Our results are similar to the report of Li et al and different from the report of Brette et al. These contradictory data may be caused by the wide differences in experimental conditions, including different species, tissues, preparations, and the method employed to measure ICa,L.

We also compared the effects of hyposmotic swelling and hyperosmotic shrinkage on ICa,L. Hyperosmotic challenge increased amplitude of ICa,L and accelerated inactivation of ICa,L (Figure 2), which was different from that of hyposmotic swelling (Figure 1). When cells are exposed to hyperosmotic solution, cell shrinkage will occur. Then hyperosmotic solution is switched to isomotic solution, now cell swelling will occur. Thus we observed that the increased ICa,L induced by hyperosmotic challenge was reverted after the hyperosmotic solution was washed out (Figure 2), possibly by the same mechanism underlying cell swelling.

We found that NO donor SNAP had no significant effects on the basal ICa,L in rat ventricular myocytes, which was consistent with previous report29. SNAP also had no any effects on the decreased ICa,L induced by hyposmotic swelling (Figure 5A). Furthermore, PKG inhibitor could not prevent hyposmotic swelling induced biphasic changes of ICa,L (Figure 5B). These results suggest that PKG system is not involved in swelling-dependent modulation of ICa,L.

The previous reports about the effects of PKC on ICa,L is controversial. Experiments utilizing direct activators of PKC have also demonstrated a range of effects on ICa,L23, 24. In this research, we observed that PKC activator (PMA) decreased the basal ICa,L (Figure 6) and PKC inhibitor (BIM) prevented the biphasic effects of hyposmotic swelling on ICa,L (Figure 7). These results show that PKC system is involved in swelling-dependent modulation of ICa,L and hyposmotic swelling may increase PKC activity, which can modulate function of ICa,L. In this experiments, we observed an interesting result that the reduced ICa,L induced by PMA was reversed by hyposmotic solution containing PMA (Figure 6). A possible explanation is that, under full activation of PKC by PMA conditions, hyposmotic swelling could act on a signaling system, which can also modulate function of ICa,L channel, and increase ICa,L. In this study, we found that PMA could further decrease ICa,L which had been inhibited by hyposmotic swelling (Figure 6C). A PMA concentration of 100 nmol/L is known to elicit maximum activation of PKC. On the other hand, the degree of activation of PKC by hyposmotic swelling may well be less than that of 100 nmol/L PMA. This could account for the discrepancy in the quantitative responses to hyposmotic swelling and PMA.

In addition, we found that PKC inhibitor prevented the biphasic effects of hyposmotic swelling on ICa,L and the swelling-induced transient increase of ICa,L also was abolished (Figure 7). On the contrary, in absence of PKC inhibitor, short hyposmotic challenge could result in the transient increase of ICa,L (Figure 1). These results suggest that activation of PKC could be necessary for the swelling-induced transient increase of ICa,L. Hyposmotic swelling caused reduction of peak ICa,L and slowing of ICa,L inactivation (Figure 1B, Figure 3A), and shifted rightwardly the steady-state inactivation curve of ICa,L (Figure 3D). PKC activator PMA also had similar effects on ICa,L (Figure 6B). Thus, these effects of hyposmotic swelling on ICa,L could be associated with swelling-induced activation of PKC.

Extensive literature show that PKA can increase cardiac ICa,L by phosphorylation of ICa,L channels. In the present study, PKA inhibitor (H-89) abolished the transient increase of ICa,L during short exposure to hyposmotic solution in 9 of 10 cells. It seems that hyposmotic swelling can cause PKA activation, which participates in the transient increase of ICa,L induced by hyposmotic swelling in most cells.

According to our results and above discussion, hyposmotic swelling can cause activation of both PKA and PKC, which participate in the biphasic changes of ICa,L induced by hyposmotic swelling. PKA and PKC have contrary effects on ICa,L. Hyposmotic swelling induced monophasic decrease of ICa,L after blocking PKA activity using inhibitor, indicating that swelling-induced PKC activation became the dominant factor in regulation of ICa,L in this situation. We found that hyposmotic swelling could revert PKA enhanced ICa,L (Figure 4A). The activation of PKC inhibits PKA stimulated cardiac ICa,L30, 31, thus we supposed that, under full activation of PKA conditions, PKC could again dominate in swelling-induced changes of ICa,L and decrease PKA stimulated ICa,L. In this experiment, activation of PKA could increase the decreased ICa,L induced by long hyposmotic challenge (Figure 4B), implying that hyposmotic swelling-induced inhibition of ICa,L is reversible in some conditions.

In summary, our present results indicated that hyposmotic challenge induced biphasic changes of ICa,L, a transient increase followed by a sustained decrease, in rat ventricular myocytes through PKC pathway, but not PKG pathway. PKA system could be responsible for the transient increase of ICa,L during short exposure to hyposmotic solution. One limitation of our study is the fact that the current density of ICa,L under basal conditions is quite different among different experiment groups.

Author contribution

Ming TANG, Jürgen HESCHELER, and An-tao LUO designed the research; An-tao LUO, Hong-yan LUO, Xin-wu HU, Lin-lin GAO, and Hua-min LIANG performed the research; An-tao LUO, Ming TANG, and Jürgen HESCHELER analyzed the data; An-tao LUO and Ming TANG wrote the paper.