INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

GRAM-NEGATIVE SEPSIS is a serious systemic disorder characterized by multiple hemodynamic derangements and cardiac failure associated with high mortality (23). Cardiac dysfunction at late or terminal stages of septicemia is compounded and exacerbated by systemic hypotension (6) and abnormal distribution of blood supply to vital organs (7, 8). However, in animal models of sepsis and its experimental counterpart, endotoxemia, cardiac dysfunction can occur well before systemic alterations in blood pressure or vascular blood supply (18, 22). Furthermore, ventricular myocytes isolated from guinea pigs (26) or rabbits (13) after in vivo treatment with endotoxin exhibit reduced cell shortening, as well as reduced rates of shortening and relengthening. Decreased cell shortening has been correlated with decreased action potential duration in endotoxemic rabbit myocytes (13) and decreased free intracellular Ca²⁺ concentration ([Ca²⁺]_i) in endotoxemic guinea pig myocytes (29). Although the cellular mechanisms responsible for myocardial dysfunction during sepsis/endotoxemia are unresolved, these data suggest that Ca²⁺ influx, specifically Ca²⁺ current through L-type Ca²⁺ channels of the sarcolemma, may be altered as part of the host response to gram-negative infection.

Ca²⁺ influx through L-type Ca²⁺ channels plays a crucial role in cardiac excitation-contraction coupling. Ca²⁺ influx during the action potential not only triggers Ca²⁺ release from the sarcoplasmic reticulum (SR) but also replenishes the SR Ca²⁺ stores for subsequent release (5). In the present study, we tested the hypothesis that L-type Ca²⁺ current (I_Ca,L) is reduced in ventricular myocytes isolated from an early, nonhypotensive model of endotoxemia in guinea pigs and that reduced I_Ca,L is responsible for the lipopolysaccharide (LPS)-induced depression in systolic [Ca²⁺]_i and cell contraction of these myocytes. Using whole cell patch-clamp and fura 2 microfluorescence techniques, we introduced different inotropic challenges and compared I_Ca,L and [Ca²⁺]_i in ventricular myocytes isolated from both control and endotoxin-injected guinea pigs.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Animal model. Male albino guinea pigs weighing 300-400 g (Sasco, Omaha, NE) were injected intraperitoneally with Escherichia coli endotoxin (4 mg/kg, LPS; Sigma, St. Louis, MO) or an equivalent volume of sterile saline (control). Four hours after LPS injection, animals were injected intraperitoneally with 1,000 U of heparin and were killed by decapitation 15 min later. Hearts were removed quickly by thoracotomy and placed immediately into ice-cold Ca²⁺-free isolation media. Previous studies with this endotoxemic model demonstrate that at 4 h animals are normotensive, and hypotension and shock develop between 8 and 12 h (21). All animal procedures were reviewed and approved by the Institutional Animal Care and Use Committee of the University of Missouri-Columbia.

Isolation of ventricular myocytes. Ventricular myocytes were isolated as previously described (26). Briefly, guinea pig hearts were perfused retrogradely through the aorta with Ca²⁺-free isolation media [Earle's balanced salt solution (GIBCO) supplemented with (g/l) 0.35 MgCl₂, 0.37 NaHCO₃, 0.2 KH₂PO₄, 0.3 glutamine, 1.1 glucose, 5.03 N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES), and 1 each of the essential amino acids and vitamins (GIBCO); pH 7.15-7.2; 280 mosM] at ~37°C. Hearts were then perfused with Ca²⁺-free isolation media containing 0.08% collagenase B (Boehringer Mannheim, Indianapolis, IN) for ~10 min at 37°C. Ventricles were then isolated, minced, and incubated in fresh isolation media containing 0.02% collagenase B and 50 µM Ca²⁺ for ~3 min at 37°C. Myocytes were mechanically dispersed with a large-bore, fire-polished pipette, filtered through sterile gauze, and centrifuged at low speed (15 g). After repeated rinses and centrifugations at gradually increasing Ca²⁺, cells were resuspended in HEPES-buffered Krebs-Henseleit (HKH) solution consisting of (in mM) 118 NaCl, 4.7 KCl, 1.2 MgSO₄, 1.2 KH₂PO₄, 2.0 CaCl₂, 13.5 NaHCO₃, 11 glucose, and 10 HEPES (pH 7.2-7.3) (26).

Whole cell patch-clamp technique. Action potential and sarcolemmal Ca²⁺ current were recorded using whole cell single-electrode current-clamp and voltage-clamp modes, respectively, using an Axopatch-1D patch-clamp amplifier. Patch pipettes were pulled from borosilicate glass capillary tubes with a two-stage puller and then fire-polished. Tip resistance was typically 1-3 M. The liquid junctional potential between the pipette and the superfusate was corrected before seal formation. Myocytes were perfused with either normal Tyrode solution (current clamp) or K⁺-free Tyrode solution (voltage clamp) at room temperature (22-24°C). Normal Tyrode solution contained (in mM) 137 NaCl, 5.4 KCl, 1.8 CaCl₂, 1.0 MgCl₂, 10 HEPES, and 10 glucose (pH adjusted to 7.4 with KOH). For K⁺-free Tyrode solution, the KCl was replaced by 5.4 mM CsCl, and the pH was adjusted to 7.4 with 1 N CsOH. After a 1-G seal was obtained by gentle suction near the center of the myocyte, the membrane was ruptured by increased suction, and the cell was voltage-clamped at a holding potential of 40 mV. Cell membrane capacitance and series resistance were determined using a 20-mV hyperpolarizing pulse and were compensated by the capacitance and series resistance compensation circuit of the patch-clamp amplifier. Current and voltage signals were filtered at 5 kHz, collected using the XOP Pulse program, and stored in a Macintosh computer for later analysis using Igor-pro software.

Cell shortening and [Ca²⁺]_i measurements. Ventricular myocytes freshly isolated from either LPS or control guinea pigs were incubated with the cell membrane-permeant form of fura 2 [fura 2-acetoxymethyl ester (AM), 2.5 µM, Molecular Probes] for 10 min at room temperature and were washed twice with HKH solution (without fura 2-AM). Cells were then resuspended in HKH solution for 1 h before subsequent measurement of [Ca²⁺]_i and cell contraction. Fura 2-AM was diluted from a 1 mM stock solution into HKH solution to a final concentration of 2.5 µM. The stock fura 2-AM solution was made in 100% dimethyl sulfoxide (DMSO). DMSO at 0.25% concentration had no effect on contractile function of either control or LPS myocytes.

Data collection and analysis. Fura 2 ratio and myocyte length changes were recorded continuously throughout the experiment. Values from the last six contractions after 3-min exposure to either HKH perfusion, Iso, or BAY K 8644 were averaged and considered representative of that myocyte. Three myocytes from each animal were used for every protocol. Membrane potential and I_Ca,L recordings were collected from two to three myocytes from each animal, and six animals are represented in each protocol. Data are presented as means ± SE, and n refers to the number of myocytes. Differences between groups were compared using the Student's t-test. Two-way analysis of variance was performed for multiple comparisons between two groups. P < 0.05 was considered significantly different.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Membrane potential of LPS and control myocytes. The LPS-induced reductions in both peak systolic [Ca²⁺]_i and cell contraction were correlated with a significant reduction in action potential duration, as shown with typical action potentials from control (Fig. 1A) and LPS (Fig. 1B) myocytes after establishment of whole cell patch-clamp configuration. Characteristics of action potentials for both groups of myocytes are summarized in Table 1. Neither the resting membrane potential nor the action potential amplitude differed between control and LPS myocytes. However, action potential duration (APD) of LPS myocytes was significantly shorter than that of control myocytes. For example, time to 50% repolarization (APD₅₀) and 90% repolarization (APD₉₀) were decreased 40 and 35%, respectively, for action potentials of LPS myocytes relative to corresponding control values (Table 1). Because the plateau phase of the action potential is determined in part by transmembrane Ca²⁺ influx through L-type Ca²⁺ channels, present data suggest that Ca²⁺ influx through L-type Ca²⁺ channels is reduced in LPS myocytes.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Representative traces of action potentials from control (A) and lipopolysaccharide (LPS) myocytes (B) 5 min after establishing whole cell configuration. Myocytes were superfused with normal Tyrode solution (1.8 mM Ca2+) and maintained in current-clamp mode. Action potentials were elicited by applying a 12.5-ms current pulse 30% above threshold.

Ca²⁺ currents of control and LPS myocytes. The magnitude of the Ca²⁺ current diminishes as a function of time during patch-clamp recording, a phenomenon termed "run down" (4). In the experiments presented here, the rate of I_Ca,L run down was similar between control and LPS myocytes and resulted in ~30% reduction in peak current by the end of a 20-min recording period (data not shown). To properly compare peak I_Ca,L in two different cell populations (control and LPS), we measured peak I_Ca,L at the same time period, and all measurements were completed within 15 min after establishment of the whole cell configuration.

View larger version (16K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Fig. 2.   Representative traces of L-type Ca2+ currents (ICa,L) measured from control (A) and LPS myocytes (B) 3 min after cell membrane rupture. Myocytes were superfused with K+-free Tyrode solution containing 1.8 mM CaCl2. ICa,L was elicited by a voltage pulse (200 ms) to +10 mV from a holding potential of 40 mV. C: averaged peak ICa,L density for control (CTL) and LPS myocytes. D: time constant of ICa,L decay for both control and LPS myocytes. Data are means ± SE for control and LPS; n = no. of myocytes. * Significantly different from control value under same conditions (P < 0.05).

Effect of Ba²⁺ substitution on I_Ca,L. L-type Ca²⁺ channels are known to be modulated by intracellular Ca²⁺ (3). Although the intracellular Ca²⁺ was buffered by 10 mM EGTA in the present experiments, a transient increase in local subsarcolemmal Ca²⁺ concentration during Ca²⁺ influx could affect Ca²⁺ channel activity (16). To eliminate the potential effect of subsarcolemmal Ca²⁺ on Ca²⁺ channel function, we measured currents with Ba²⁺ rather than Ca²⁺ in the bathing solution. For these experiments, currents were measured from myocytes that were first superfused with K⁺-free Tyrode solution containing 1.8 mM Ca²⁺ and then superfused with K⁺-free Tyrode solution containing 5 mM Ba²⁺ and no Ca²⁺. As shown in Fig. 3, Ba²⁺ substitution nearly doubled the peak current amplitude for both control and LPS myocytes. However, Ba²⁺ substitution did not reverse the LPS-induced reduction in peak current. Both peak I_Ca,L density and peak Ba²⁺ current (I_Ba) density of LPS myocytes were significantly less than the corresponding values of control myocytes (Fig. 3C). These data indicate that the reduction of peak I_Ca,L density in LPS myocytes was not Ca²⁺ dependent.

View larger version (22K): [in this window] [in a new window]   View larger version (17K): [in this window] [in a new window]   Fig. 3.   Representative traces of ICa,L measured from control (A) and LPS myocytes (B) using either Ca2+ or Ba2+ as the permeant ion. A and B show ICa,L before and after replacement of Ca2+ (1.8 mM) with Ba2+ (5 mM) in K+-free Tyrode solution. ICa,L and Ba2+ current (IBa) were elicited by a voltage pulse (200 ms) from a holding potential of 40 mV to +10 mV. C: averaged peak ICa,L and IBa densities for control and LPS myocytes. D: time constant of ICa,L and IBa decay for both control and LPS myocytes. Data are means ± SE for control and LPS myocytes; n = no. of myocytes. * Significantly different from control value under same conditions (P < 0.05).

I-V relationship. To assess the voltage dependence of L-type Ca²⁺ channels in LPS myocytes, we measured peak I_Ca,L at different voltages using either Ca²⁺ or Ba²⁺ as the charge carrier. Currents of both control and LPS myocytes had similar voltage dependence regardless of whether Ca²⁺ or Ba²⁺ was the charge carrier (Fig. 4). Threshold potential, the potential eliciting maximum peak I_Ca,L density, and the reversal potential of I_Ca,L were similar between control and LPS myocytes. Despite the similar I-V relationship, averaged peak I_Ca,L density of LPS myocytes was significantly decreased at pulse potentials between 10 and +40 mV compared with control values (Fig. 4, A and B).

View larger version (17K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Fig. 4.   Averaged peak current density-voltage relationships (A and B) and voltage-dependent activation of ICa,L (C). Peak ICa,L density-voltage relationship was obtained using Ca2+ (A) or Ba2+ (B) as the charge carrier. C: for voltage-dependent activation, relative peak conductance (GCa/GCa max) was plotted as a function of test pulse potential (see text). Step changes in voltage were elicited at 6-s intervals. Data points are means ± SE; n = no. of myocytes. * Significantly different from control value under same conditions (P < 0.05).

Steady-state I_Ca,L activation and inactivation and recovery from inactivation. The voltage dependence of I_Ca,L activation was determined as the ratio of peak conductance (G_Ca) to the maximal peak conductance (G_{Ca max}) and was expressed as dV = G_Ca/G_{Ca max} and G_Ca = I_Ca,L/(V_m  V_rev), where V_rev is the apparent reversal potential of I_Ca,L, and d is the steady-state activation parameter (29). When d was depicted as the function of test potentials, the activation curve of LPS myocytes shifted slightly toward more positive potentials but was not significantly different from that of control myocytes (Fig. 4C). When the activation curves of individual myocytes were fit to the Boltzmann equation, d = {1 + exp[(V_1/2  V)/K]}¹, where V_1/2 is the membrane potential producing half-maximal activation, V is voltage, and K is the slope of the activation curve, neither V_1/2 (1.8 ± 0.3 mV) nor K (5.54 ± 0.8 mV) of LPS myocytes was different from those values of control myocytes (V_1/2: 1.6 ± 0.2 mV; K: 5.52 ± 0.7 mV).

View larger version (11K): [in this window] [in a new window]   View larger version (11K): [in this window] [in a new window]   View larger version (17K): [in this window] [in a new window]   Fig. 5.   Steady-state inactivation of ICa,L obtained from control and LPS myocytes using a double-pulse protocol (see MATERIALS AND METHODS). A: representative current record obtained in response to a 2-pulse protocol with a prepulse of 0 mV (300 ms) and a test pulse of +10 mV (200 ms) separated by a 5-ms resting interval at 40 mV. B: normalized ICa,L (ICa /ICa max) of the test pulse plotted as a function of the prepulse potential (see MATERIALS AND METHODS). C: normalized ICa,L of the test pulse plotted as a function of the prepulse potential with either Ca2+ (5 mM) or Ba2+ (5 mM) as the charge carrier. The 2-pulse protocol was slightly modified with a prepulse duration of 150 ms, a test pulse duration of 100 ms, and an interpulse interval of 10 ms (see text). Data are means ± SE; n = no. of myocytes.

View larger version (13K): [in this window] [in a new window]   View larger version (11K): [in this window] [in a new window]   Fig. 6.   Recovery of peak ICa,L from inactivation using a double-pulse protocol (see MATERIALS AND METHODS). A: representative current record in response to identical 200-ms prepulse and test pulse to +10 mV from a holding potential of 40 mV, separated by a resting interval of 100 ms. B: normalized ICa,L of the test pulse plotted as a function of the interpulse resting interval (see MATERIALS AND METHODS). Data are means ± SE; n = no. of myocytes.

Effect of Iso on I_Ca,L. L-type Ca²⁺ channels are modulated by -adrenergic receptor activation, and channel activity is stimulated through phosphorylation of channel subunits by adenosine 3',5'-cyclic monophosphate (cAMP)-dependent protein kinase (28). The -adrenergic receptor agonist Iso has been shown to reverse the contractile dysfunction of myocytes isolated from this guinea pig model of endotoxemia (26). Therefore, we evaluated whether Iso also would reverse the reduction in I_Ca,L of LPS myocytes. In the presence of Iso, I_Ca,L of both control and LPS myocytes increased compared with I_Ca,L for the same cell in the absence of Iso (Fig. 7). In this set of myocytes, peak I_Ca,L density of LPS myocytes (2.7 ± 1.7 pA/pF) was significantly lower than the value of control myocytes (4.6 ± 0.4 pA/pF) under basal conditions. In the presence of Iso, peak current density increased 174% in control myocytes and 283% in LPS myocytes (Fig. 7C). I_Ca,L density was no longer statistically different between these two groups. The enhancement of I_Ca,L by Iso occurred at pulse potentials between 30 and +40 mV such that in the presence of Iso there was no significant difference in the I-V relationship between control and LPS myocytes (Fig. 7D). Iso had no effect on threshold potential or the reversal potential but shifted the peak potential for I_Ca,L from +10 to 0 mV for both control and LPS myocytes.

View larger version (22K): [in this window] [in a new window]   View larger version (20K): [in this window] [in a new window]   Fig. 7.   Representative traces of ICa,L from control (A) and LPS myocytes (B) before (basal) and after exposure to isoproterenol (Iso). Myocytes were superfused with K+-free Tyrode solution containing 1.8 mM Ca2+ and then switched to K+-free Tyrode solution containing 1.8 mM Ca2+ and 0.1 µM Iso. ICa,L was elicited by a test pulse (100 ms) to 0 mV from a holding potential of 40 mV. C: averaged peak ICa,L density before (basal) and after Iso exposure. D: peak ICa,L density-voltage relationship in presence of Iso. Data are means ± SE; n = no. of myocytes. * Significantly different from control value under same conditions (P < 0.05).

Effect of BAY K 8644 on I_Ca,L. BAY K 8644, a dihydropyridine receptor agonist, has been used widely to stimulate L-type Ca²⁺ channels of various cell types. Enhancement of I_Ca,L by BAY K 8644 is independent of cAMP-dependent phosphorylation of Ca²⁺ channels (27, 28). We determined whether BAY K 8644 could overcome the endotoxin-induced I_Ca,L deficiency in LPS myocytes. Representative tracings of I_Ca,L from control and LPS myocytes before and after BAY K 8644 demonstrate that BAY K 8644 increased I_Ca,L in both populations of cells (Fig. 8, A and B). Averaged peak I_Ca,L density increased 87 and 110% in control and LPS myocytes, respectively (Fig. 8C). Stimulation of I_Ca,L by BAY K 8644 occurred over test potentials from 20 to +20 mV for both control and LPS myocytes (Fig. 8D). However, BAY K 8644 failed to reverse the endotoxin-induced reduction in I_Ca,L. Peak I_Ca,L density of LPS myocytes remained significantly less than control values even in the presence of BAY K 8644 (Fig. 8, C and D). Similar to the effect of Iso, BAY K 8644 shifted the I-V relationship such that the pulse potential required for peak current changed from +10 to 0 mV. Neither the threshold potential nor the reversal potential changed in either group of myocytes after exposure to BAY K 8644.

View larger version (20K): [in this window] [in a new window]   View larger version (20K): [in this window] [in a new window]   Fig. 8.   Representative traces of ICa,L from control (A) and LPS myocytes (B) before (basal) and after exposure to BAY K 8644 (BAY K). Myocytes were superfused with K+-free Tyrode solution containing 1.8 mM Ca2+ and then switched to K+-free Tyrode solution containing 1.8 Ca2+ and 0.1 µM BAY K 8644. ICa,L was elicited by a test pulse (200 ms) to +10 mV from a holding potential of 40 mV. C: averaged peak ICa,L density before (basal) and after BAY K 8644 exposure. D: peak ICa,L density-voltage relationship after BAY K 8644 exposure. Data are means ± SE; n = no. of myocytes. * Significantly different from control value under same conditions (P < 0.05).

Effect of Iso on systolic [Ca²⁺]_i transients and cell shortening. To correlate the ability of Iso to reverse endotoxin-induced depression in I_Ca,L with myocyte contractile functions, we measured both systolic [Ca²⁺]_i and myocyte shortening before and after exposure to Iso. Myocytes were loaded with fura 2-AM and field stimulated at 0.5 Hz. Both fura 2 ratios and cell length were recorded simultaneously. LPS myocytes exhibited reduced peak systolic [Ca²⁺]_i as well as reduced maximal rates of Ca²⁺ rise and fall (±dCa²⁺/dt_max) compared with control myocytes (Table 2). Correlating with reduced systolic [Ca²⁺]_i, cell shortening of LPS myocytes also was decreased (Table 2). Exposure of myocytes to Iso increased peak systolic [Ca²⁺]_i and cell shortening of both control and LPS myocytes. Iso increased systolic [Ca²⁺]_i 26% in control and 111% in LPS myocytes over basal values. The relatively greater increase in the size of Ca²⁺ transients of LPS myocytes caused by Iso essentially eliminated the difference in systolic [Ca²⁺]_i as well as ±dCa²⁺/dt_max between control and LPS myocytes. Furthermore, as predicted by the improved systolic [Ca²⁺]_i, Iso had relatively greater effect on cell shortening of LPS myocytes (26% in control and 156% in LPS) such that there was no significant difference in cell shortening between control and LPS myocytes.

Effect of BAY K 8644 on systolic [Ca²⁺]_i transients and cell shortening. In contrast to the effects of Iso, BAY K 8644 was ineffective in reversing either contractile dysfunction or reduced systolic [Ca²⁺]_i of LPS myocytes. Enhancement of systolic [Ca²⁺]_i and cell shortening by BAY K 8644 was similar between control and LPS myocytes: systolic [Ca²⁺]_i increased 53% in control and 60% in LPS myocytes (Table 3) and cell shortening increased 50% in control and 66% in LPS myocytes (Table 3). Thus both systolic [Ca²⁺]_i and cell shortening of LPS myocytes were still significantly less than respective control values even in the presence of BAY K 8644. 

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

E. coli endotoxemia consistently produces a deleterious depression in cardiodynamic function, including diminished contractile reserves of left ventricular myocardium and decreased end-diastolic compliance of the left ventricular chamber (23, 32). Our previous work indicated that endotoxin-induced loss of inotropic power can be correlated with reduced systolic [Ca²⁺]_i and corresponding reduction in cell shortening of individual myocytes (31). In the present study, we determined that myocytes isolated from a guinea pig model of E. coli endotoxemia have a shortened action potential duration and decreased peak Ca²⁺ and Ba²⁺ currents through L-type Ca²⁺ channels. Reduction in I_Ca,L of LPS myocytes could not be attributed to alterations in the I-V relationship, steady-state inactivation, or recovery from inactivation of the Ca²⁺ channel. Thus intrinsic voltage-dependent properties of Ca²⁺ channels appeared normal in LPS myocytes. Activation of Ca²⁺ channels by the direct Ca²⁺ channel agonist BAY K 8644 or by the -adrenergic receptor agonist Iso increased I_Ca,L. However, only Iso reversed the LPS-induced depression in I_Ca,L, systolic [Ca²⁺]_i, and cell contraction of ventricular myocytes. These data demonstrate that Ca²⁺ influx through L-type channels is reduced in ventricular myocytes isolated from LPS-injected animals and that reduced Ca²⁺ influx may underlie the early myocardial dysfunction associated with gram-negative endotoxicosis.

Reduced peak I_Ca,L density of LPS myocytes is unlikely to be due to experimental conditions associated with myocyte isolation or current recording. We previously established that resting cell length of myocytes from this LPS model is similar to that of control cells and that yields of Ca²⁺-tolerant myocytes are similar between control and LPS populations (26). Recording conditions for I_Ca,L measurement were carefully managed to avoid contamination by other currents and were confirmed by complete current blockade with 0.5 mM Cd²⁺. Although I_Ca,L of cardiac myocytes is characterized by time-dependent run down during patch-clamp measurements (4), differences in run down could not account for reduced I_Ca,L of LPS myocytes. First, the rate of I_Ca,L run down was similar for the two groups, and second, all recordings were made at the same time after membrane rupture. Thus reduction of I_Ca,L density in LPS myocytes was not an artifact of recording conditions.

Reduced I_Ca,L density in LPS myocytes could result from alterations in the voltage-dependent properties of the L-type Ca²⁺ channel. On the basis of work in dogs, myocytes isolated from infarcted hearts exhibit a significant reduction in I_Ca,L that is associated with 1) reduced numbers of L-type Ca²⁺ channels, 2) altered voltage-dependent inactivation, and 3) increased rate of I_Ca,L decay (1). However, endotoxemia does not appear to affect the voltage-dependent properties of the L-type Ca²⁺ channels. The I-V relationships were similar between control and LPS myocytes when either Ca²⁺ or Ba²⁺ was used as the charge carrier. In addition, Ca²⁺ channels of both control and LPS myocytes exhibited similar voltage-dependent steady-state activation and inactivation, as well as recovery from inactivation. Although the rate of I_Ca,L decay was reduced in LPS myocytes, this reduction appears to be correlated with the reduced peak I_Ca,L in LPS myocytes because substitution with Ba²⁺ as the charge carrier eliminated the difference in the rate of current decay between control and LPS myocytes. These data indicate that endotoxemia does not change the intrinsic voltage-dependent properties of the L-type Ca²⁺ channel.

Elevation in [Ca²⁺]_i decreases the amplitude of I_Ca,L and accelerates I_Ca,L decay through Ca²⁺-dependent inactivation of I_Ca,L in mammalian cardiac myocytes (10, 24, 30). Thus reduced I_Ca,L in LPS myocytes could be explained by elevated [Ca²⁺]_i concentrations in LPS myocytes. In the present study, the pipette solution contained 10 mM EGTA with no added Ca²⁺, which will buffer myoplasmic Ca²⁺ to subnanomolar levels. Thus bulk myoplasmic Ca²⁺ concentration was not different between control and LPS myocytes. More significantly, peak current of LPS myocytes was less than control even when the superfusate Ca²⁺ was replaced with Ba²⁺ as the charge carrier. Ba²⁺ substitution eliminates potential Ca²⁺-dependent inactivation of I_Ca,L, which may persist even in the presence of EGTA (10, 16). Furthermore, the rate of I_Ca,L decay in the presence of Ca²⁺ was slower in LPS myocytes compared with control, inconsistent with Ca²⁺-dependent inactivation of I_Ca,L. In addition, resting [Ca²⁺]_i is not different between control and LPS myocytes (31). For example, resting [Ca²⁺]_i of fura 2-loaded LPS myocytes used for data in Table 2 (156 ± 11 µM) was similar to that of control myocytes (159 ± 27 µM, P > 0.05) although systolic [Ca²⁺]_i was significantly lower in LPS myocytes (Table 2). Thus it is unlikely that [Ca²⁺]_i-dependent inactivation is responsible for the reduction of I_Ca,L in LPS myocytes.

Reduced peak I_Ca,L density of myocytes from a cardiac hypertrophy model appears to result from a decrease in sarcolemmal Ca²⁺ channel density possibly due to an increase in myocyte size without concomitant increase in L-type Ca²⁺ channel number (19). However, there is no evidence of myocyte hypertrophy in the endotoxemic guinea pig model used in the current study. Resting length of ventricular myocytes from endotoxemic guinea pigs was not different from that of control myocytes (26), and membrane capacitance was similar for both control and LPS myocytes in the present study. On the other hand, decreases in L-type Ca²⁺ channel numbers independent of myocyte hypertrophy appear to underlie the reduced peak I_Ca,L density in human cardiac myocytes dissociated from failing hearts (20) and myocytes isolated from 5-day infarcted canine hearts (1). In addition, Ca²⁺ channel numbers measured by dihydropyridine binding assays are reduced in cardiac sarcolemmal membranes from endotoxemic rabbits (17). Although present data do not rule out reduced number of membrane channels as causative in the decreased I_Ca,L of LPS myocytes, the ability of Iso to reverse the endotoxin-induced reduction in I_Ca,L density suggests strongly that the absolute number of Ca²⁺ channels of LPS myocytes is similar to that of control cells.

L-type Ca²⁺ channel function can be modulated by -adrenergic receptor activation (15) and by the direct channel agonist, BAY K 8644 (27, 28). Direct binding of dihydropyridine agonists such as BAY K 8644 to Ca²⁺ channels enhances Ca²⁺ current by increasing the open time and shortening the close time of single channels (28). -Adrenergic receptor activation also increases channel activity by prolonging the open time and shortening the close time of Ca²⁺ channels. In addition, -adrenoceptor activation also increases the probability that a channel will open, as reflected by an increase in the number of channel openings per unit time during single-channel recording (12, 28). The -adrenoceptor-dependent increase in the probability that a channel will open appears to be dependent on cAMP-dependent protein kinase A (PKA)-mediated phosphorylation of Ca²⁺ channel subunits (12, 15, 28). Thus the primary difference between these agents is that -adrenergic receptor agonists increase the probability that a Ca²⁺ channel will open (28), although both agonists increase macroscopic Ca²⁺ current without increasing either the number of Ca²⁺ channels or the single-channel conductance (28).

In the present study, both Iso and BAY K 8644 increased I_Ca,L of control and LPS myocytes. However, only Iso reversed the endotoxin-induced reduction of I_Ca,L as reflected by the relatively greater increase in I_Ca,L of LPS myocytes compared with controls. In contrast, BAY K 8644 produced similar or parallel increases in I_Ca,L in both control and LPS myocytes as evidenced by reduced peak I_Ca,L density of LPS myocytes in the presence of BAY K 8644. These data suggest that under basal conditions, LPS myocytes may have fewer Ca²⁺ channels that are available to open. Whether a channel is available to open appears to depend on the phosphorylated state of the channel protein. Phosphorylation-dephosphorylation controls the cycling of individual cardiac L-type Ca²⁺ channels between two gating modes (11, 12). In the phosphorylation mode, the channel is believed to be available to open, but in the dephosphorylated mode the channel remains less available or silent (11, 12). Thus increased channel phosphorylation via activation of PKA increases the number of channels available to open without necessarily altering the total number of channels resident in the sarcolemma. Data from the present study suggest that endotoxemia somehow decreases the basal phosphorylation state of the L-type Ca²⁺ channel of cardiac myocytes, resulting in reduced peak I_Ca,L density, depressed systolic [Ca²⁺]_i, and impaired contractility of these cells.

The ability of Iso to reverse endotoxin-induced reductions in both peak I_Ca,L density and peak systolic [Ca²⁺]_i is an important finding, indicating that cardiac responses to -adrenergic receptor stimulation are conserved in the guinea pig model during early stages of endotoxemia (26). Endotoxemia and septicemia commonly evoke increased concentrations of circulating catecholamines as part of the sympathetic compensatory attempt to maintain cardiac output and circulation to vital tissues (14). Sympathetic support of cardiac work effort will be of limited duration because prolonged exposure of the myocardium to catecholamines typically leads to desensitization or downregulation of the cardiac -adrenoceptor population (14, 25). Present studies with Iso indicate clearly that cardiac -adrenoceptors controlling myocardial inotropy are functionally operative for at least the first 4 h of endotoxemia, but these studies do not address putative loss of such receptors during the later hypotensive and decompensatory stages of this form of endotoxicosis (21, 22). In any case, an endotoxin-induced diminution of the basal phosphorylation state of L-type Ca²⁺ channels, as we now propose, would modulate a mechanistic pathway used by sympathetic compensatory attempts to provide increased inotropic support of myocardial performance (14).

In conclusion, ventricular myocytes isolated from endotoxemic guinea pigs exhibited reduced peak I_Ca,L density that correlated with decreased systolic [Ca²⁺]_i and decreased cell shortening. Reduced peak I_Ca,L density, decreased systolic [Ca²⁺]_i, and diminished inotropic capability of LPS myocytes were selectively reversed by -adrenoceptor stimulation with Iso, but not by direct Ca²⁺ channel activation with the dihydropyridine agonist BAY K 8644. These data indicate that reduced Ca²⁺ influx through L-type Ca²⁺ channels plays a central role in myocardial contractile dysfunction during endotoxemia.