Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Hexachlorocyclohexanes (HCHs) are highly lipophilic molecules possessing biological half-lives that can be measured in years (Jung et al., 1997). The  isoform (-HCH; lindane) is a potent neurostimulant and convulsant that has been extensively used worldwide both as an insecticide and ectopariticide. There is significant evidence that -HCH mediated neurotoxicity is primarily the result of blockade of Cl influx through ionotropic -aminobutyric acid_a receptors, resulting in depolarization and hyperexcitation of the postsynaptic neuronal membrane (Matsumura and Tanaka, 1984; Cristofol and Rodriguez-Farre, 1994). However, -HCH has also been shown to enhance both spontaneous and evoked release of neurotransmitters from nerve terminals (Publicover and Duncan, 1979; Joy and Burns, 1988; Cristofol and Rodriguez-Farre, 1994), effects that can contribute to central nervous system hyperexcitability, and these actions have been correlated with the ability of -HCH to elevate Ca²⁺ in brain synaptosomes (Narbonne and Lievremont, 1983; Bondy and Halsall, 1988; Hawkinson et al., 1989). Consistent with the hypothesis of Ca²⁺ deregulation, -HCH has been shown to alter Ca²⁺ homeostasis in a variety of excitable and nonexcitable cells (Joy and Burns, 1988; Gandhi and Venkatakrishna-Bhatt, 1989; Gautam et al., 1989; Seifert et al., 1989; Carrero et al., 1990; Forgue et al., 1990; Ervens and Seifert, 1991; Wenzel-Seifert et al., 1991). -HCH has also been shown to alter contractile parameters in skeletal myocytes (Koohmaraie, 1987), an action probably related to altered Ca²⁺ regulation.

In comparison to -HCH, -HCH has been recently shown to exhibit a spectrum of potent activities in several cell types that appears to stem primarily from deregulation of Ca²⁺ signaling. Examples include pronounced positive inotropy and contracture in isolated rat atrial strips (Pessah et al., 1992), enhancement of antigen-stimulated secretion of serotonin from rat basophilic leukemia cells (Mohr et al., 1995), and pronounced cytotoxicity in cultured cerebellar granule cells (Rosa et al., 1996, 1997b).

Importantly, the underlying molecular mechanisms by which -HCH and -HCH alter Ca²⁺ regulation appear to be quite different. With atrial strips, -HCH is ~30-fold more potent than -HCH in enhancing contractile force, a difference that is quantitatively mirrored by the ability of - and -HCH to 1) mobilize Ca²⁺ from sarcoplasmic reticulum (SR) and 2) alter the binding of [³H]ryanodine to ryanodine receptors (RyR) on SR/endoplasmic reticulum (ER) membranes (Pessah et al., 1992). Studies with rat basophilic leukemia (RBL) cells have indicated that -HCH mobilizes Ca²⁺ from a thapsigargin-sensitive ER store and concomitantly inhibits depletion-activated Ca²⁺ entry. Surprisingly, -HCH-stimulated Ca²⁺ efflux from ER appears to proceed by a mechanism independent of the inositol-1,4,5-trisphosphate (IP₃) receptor in the RBL cell because heparin is unable to block its actions (Mohr et al., 1995). More recently, -HCH has been shown to stereoselectively mobilize Ca²⁺ from intracellular stores in cultured neural cells and appears to be mediated, at least in part, by interaction with ryanodine receptors (Rosa et al., 1997a, 1997b). A partial involvement of ryanodine-sensitive Ca²⁺ stores has also been seen in RBL cells where caffeine diminishes the actions of -HCH, possibly by depleting Ca²⁺ in this store (Mohr et al., 1995). Consistent with the hypothesis of distinct mechanisms by which - and -HCH disrupt cellular Ca²⁺ signaling, Criswell et al. (1994) used myometrial smooth muscle cells and found that -HCH increased intracellular calcium through modulation of IP₃-sensitive, but not ryanodine-sensitive, stores.

Results from our previous studies with isolated (SR) membranes isolated from rat cardiac ventricles have shown that -HCH, but not -HCH, has potent and selective actions on ryanodine-sensitive Ca²⁺ efflux and the binding of [³H]ryanodine (Pessah et al., 1992). To clarify the molecular mechanisms underlying HCH toxicity, we undertook the present study to investigate the direct action of - and -HCH on single cardiac RyR channels using planar lipid bilayer techniques and quantified changes in channel gating kinetics. The results presented here reveal two unique mechanisms by which -HCH, but not -HCH, alters cellular homeostasis and signaling. The first mechanism involves direct interaction with the RyR2-protein complex, which enhances channel gating kinetics and accounts for altered SR/ER Ca²⁺ transport and cellular Ca²⁺ homeostasis (current report). Our companion report reveals a unique property of -HCH to produce a Ca²⁺-dependent, K⁺-selective ionic current in biological membranes, which can account for altered membrane potential. These results provide insight into the complex pharmacology that has been seen with HCH isomers.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Preparation of SR Membrane Vesicles. Cardiac SR vesicles were isolated from Sprague-Dawley rats (Simonsen, Gilroy, CA) or male Hartley guinea pigs (Charles River, Wilmington, MA) according to the method of Feher (Feher and Davis, 1991) with a few modifications. Briefly, animals were sacrificed by cervical dislocation, and the ventricles were quickly removed, placed over ice, and trimmed of fat and connective tissue. Isolated ventricles (10-15 g) were washed in ice-cold homogenization buffer (1 M KCl, 10 mM imidazole, 10 µg/ml leupeptin, 100 µM phenylmethylsulfonyl fluoride, pH 7.0.), blotted to absorbent paper, weighed, minced, and homogenized with a Polytron homogenizer (Brinkmann Instruments, Westbury, NY) with three 30-s bursts at 20,000 rpm in 5 volumes of ice-cold homogenization buffer. The homogenate was centrifuged for 20 min at 10,000g_max at 4°C. The supernatant was discarded, and the pellets were rehomogenized in the same volume of ice-cold homogenization buffer (as described above) and centrifuged for 20 min at 6000g_max at 4°C. The supernatant was centrifuged for 25 min at 24,000g_max at 4°C, and the supernatant was collected and further centrifuged for 2 h at 41,000g_max at 4°C. The resulting pellet was collected and resuspended in homogenization buffer using a Dounce homogenizer and repelleted as described in the preceding step. The final pellets were resuspended in 10% sucrose and 20 mM Tris, pH 7.0, at approximately 3 mg/ml protein. Protein concentration was determined according to the method of Lowry et al. (1951) using BSA as a standard. Aliquots were snap frozen in liquid N₂ and stored at 80°C until used.

Measurement of Equilibrium [³H]Ryanodine Binding. Specific binding of [³H]ryanodine to the Ca²⁺-activated, high-affinity site on the channel complex from guinea pig cardiac SR (30 µg protein) was measured in a final volume of 500 µl in an assay buffer consisting of 250 mM KCl, 15 mM NaCl, 50 µM CaCl₂, and 20 mM HEPES, pH 7.1. To ensure consistent partitioning with the biophase, HCH isomers (10⁸ to 10⁴ M in DMSO) were first added to incubation mixtures containing the membrane vesicles; then, 1 nM [³H]ryanodine was added to initiate radioligand binding. DMSO never exceeded 1% and was present in all controls. Reactions were equilibrated for 3 h at 37°C and quenched by rapid filtration through GF/B glass-fiber filters using a cell harvester (Brandel, Gaitherburg, MD). The filters were washed with 2 × 5 ml of ice-cold Tris buffer, pH 7.1, containing 50 µM CaCl₂. Inhibition curves were repeated at least three times, each in duplicate. IC₅₀ values were calculated by linear regression analysis of logit plots or fit to a two-site model.

Isolated Myocyte Experiments. Male Hartley guinea pigs (Charles River, Wilmington, MA) were injected i.p. with heparin (3 units/g) 30 to 60 min before sacrifice. Hearts were rapidly removed, and the aorta was cannulated and perfused retrogradely with oxygenated (95% O₂/5% CO₂) Ca²⁺-free HEPES Tyrode's buffer (121 mM NaCl, 3.82 mM KCl, 1.18 mM KH₂PO₄, 11.1 mM glucose, 10 mM HEPES, and 30 mM taurine, pH 7.4) for 4 min. The heart was then perfused with the same buffer supplemented with 0.01% type II collagenase (Worthington Biochemical, Freehold, NJ), 0.01% type XIV protease (Sigma Chemical, St. Louis, MO), and 0.1% fatty acid-free BSA for 8 to 10 min. The ventricles were removed, diced, and triturated using a pipette with a 4-mm orifice. The cell suspension was filtered through a 250-mm mesh, gently centrifuged twice, and recovered in a buffer containing 140.8 mM KCl, 2 mM EGTA, 2 mM MgCl₂, 1 mM K₂ATP, and 5 mM HEPES, pH 7.4.

Single-Channel Measurements. Reconstitution of rat cardiac ryanodine receptor (RyR2) into the planar lipid bilayer was performed by forming bilayer membranes from a 5:3:2 mixture of phosphatidylethanolamine/phosphatidylserine/phosphatidylcholine suspended (50 mg/ml) in decane across a 200-µm hole in the side of a polystyrene cup that separates two chambers of 0.7 ml each. SR membranes containing cardiac ryanodine receptor (5 µg total protein) were added to the cis chamber, which contained either 250 or 500 mM CsCl, 200 µM CaCl₂, and 20 mM HEPES, pH 7.2, whereas the trans chamber contained either 50 or 100 mM CsCl and 20 mM HEPES, pH 7.2. After fusion of a single vesicle, more than 300 µM EGTA, pH 7.0, was added to halt the reaction, and the cis chamber was perfused with an identical buffer with no added Ca²⁺ or EGTA. Single-channel activity was measured at various holding potentials with respect to the trans (ground) side in the presence or absence of HCH. Data were amplified (3900A; Dagan Corporation, Minneapolis, MN), digitized (DigiData 1200; Axon Instruments, Burlingame, CA), and stored on a 486 PC computer using the Axotape program (version 2.0; Axon Instruments, Burlingame, CA). Subsequent data analysis was performed using pClamp (version 6.0; Axon Instruments). Control channel activity was typically acquired for at least 2 min followed by the addition of the test compound from a 100× stock. Modified channel fluctuations were acquired for at least 5 min for subsequent analysis.

(1)

Materials. [³H]Ryanodine (63.5 Ci/mmol) was obtained from New England Nuclear (Wilmington, DE). Hexachlorocyclohexane, albumin, and protease were purchased from Sigma Chemical Co. (St. Louis, MO). All other chemicals were of the highest quality available commercially.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

-HCH Stereoselectively Alters RyR2 Function. The binding of [³H]ryanodine to receptors found in guinea pig ventricular muscle SR was assessed in the presence of 50 µM Ca²⁺, a condition favoring channel opening. As shown in Fig. 1, -HCH inhibited Ca²⁺-activated binding of [³H]ryanodine to its high-affinity site on the cardiac calcium release channel complex and tended to be biphasic. Inhibition constants were calculated using a two-site model giving IC₅₀ values of 2.5 ± 0.7 and 18 ± 3 µM for the high- and low-affinity interactions, respectively. The dose-response relationship for -HCH inhibition of [³H]ryanodine binding was steep, with a logit slope of 1.7 ± 0.1. In contrast to -HCH, inhibition of [³H]ryanodine binding by -HCH (lindane) was no more than 17 ± 8% at 100 µM.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Stereoselective inhibition of the binding of [3H]ryanodine (1-2 nM) to the Ca2+ release channel complex of guinea pig cardiac muscle by two HCH isomers. Receptor binding was assessed in the presence of optimal Ca2+ (50 µM) as described in Experimental Procedures. Data shown are the mean ± S.E.M. of at least four determinations performed in duplicate. Dose-response profiles consistently suggested the presence of two components in the inhibition curves. Inhibitory constants are 2.5 ± 0.7 and 18 ± 3 µM for the high- and low-affinity interactions, respectively, and the logit slope was 1.7 ± 0.1 for -HCH. Control binding of [3H]ryanodine in the presence of 0.5% DMSO was 0.30 ± 0.01 pmol/mg protein.

-HCH Alters Ca²⁺ Transient in Ventricular Myocytes. Ventricular myocytes isolated from guinea pig were electrically stimulated at 1.0 Hz in the absence or presence of HCH, and the resulting intracellular Ca²⁺ transients were measured fluorimetrically using the Ca²⁺ indicator dye indo-1. After exposure to 10 µM -HCH, the amplitude of the indo-1 transient was decreased with a corresponding increase in the recovery time the transient needed to return to baseline (Fig. 2). Analysis of transients recorded in the presence of HCH reveal that 10 µM -HCH decreased the peak amplitude by 29 ± 4%, whereas the same concentration of -HCH elicited only a small and statistically insignificant decrease (Fig. 3A). -HCH (10 µM) also significantly prolonged the time required for the Ca²⁺ transient to decrease to within 80% of the peak (Fig. 3B, 20% prolongation compared with 0.1% DMSO control, from 200 ± 10 to 244 ± 13 ms, respectively). In comparison, 10 µM -HCH did not induce any significant change in recovery time compared with 0.1% DMSO control (Fig. 3B; 203 ± 11 versus 214 ± 17 ms, respectively). Interestingly, in the time frame of these experiments, -HCH (10 µM) had no effect on resting indo-1 fluorescence in quiescent cells (data not shown), suggesting its actions were use dependent.

View larger version (7K): [in this window] [in a new window]   Fig. 2.   -HCH alters the responses of ventricular myocytes to electrical stimulation. Freshly isolated ventricular myocytes from guinea pig were loaded with indo-1 as described in Experimental Procedures. Ca2+ transients produced by electrical stimulation at 1.0 Hz were measured by ratio fluorometry. Shown is a representative indo-1 fluorescence transient before and 5 min after exposure to 10 µM -HCH. This experiments was repeated four times with similar results.

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Analysis of indo-1 Ca2+ transients in the absence and presence of HCH isomers. A, effect of HCH isomers on amplitude of indo-1 fluorescence transient in guinea pig ventricular myocytes electrically stimulated at 1.0 Hz. Data shown are mean ± S.E.M. (n = 4) except for 0.1% DMSO (n = 6). Results are expressed as percent of control. B, effect of the HCH isomers on the time required for indo-1 fluorescence transient to decline to 80% of maximum in guinea pig ventricular myocytes electrically stimulated at 1.0 Hz. Data shown are mean ± S.E.M. (n = 4) except for 0.1% DMSO (n = 5). *P  .05 HCH treatment versus control.

Effect of HCH on Cell Shortening. Ventricular myocytes isolated from guinea pig heart were electrically stimulated at 1.0 Hz, and contraction was monitored using edge motion detection as described in Experimental Procedures. Although exposure of myocytes to HCH induced a concentration-dependent decrease in cell shortening (Fig. 4A), -HCH was significantly more potent than -HCH in inhibiting electrically stimulated cell shortening. As shown in Fig. 4B, cardiac myocytes responded to 10 µM -HCH by contracting to 26 ± 6% of control (a 74% reduction in cell shortening), whereas exposure to the same concentration of -HCH generated a 54 ± 10% decrease in cell shortening. The same effects were obtained when myocytes were exposed to a bolus addition of 10 µM HCH (data not shown).

View larger version (29K): [in this window] [in a new window]   Fig. 4.   -HCH diminishes contractile force of electrically paced isolated cardiomyocytes. A, representative trace displaying cell shortening in an isolated ventricular myocyte from guinea pig electrically stimulated at 1.0 Hz. B, effect of HCH isomers on steady-state cell shortening in guinea pig ventricular myocytes electrically stimulated at 1.0 Hz. Data are expressed as percentage of cell shortening in the presence of 0.1% DMSO, which had no effect on myocytes. Data shown are mean ± S.E.M. from at least seven and five observations for -HCH and -HCH, respectively. *P  .05 relative to respective control.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   The influence of pacing frequency on the action of -HCH on cell shortening in isolated ventricular myocytes from guinea pig. Cell shortening is expressed as percentage of control shortening at 1.0 Hz in the presence of 0.1% DMSO. Data shown are mean ± S.E.M. from at least seven observations. Depression of cell shortening was found to be significantly different from control (P  .05) at all pacing frequencies except in the presence of 3 µM -HCH at 0.5 Hz.

-HCH Stimulates Single-Channel Kinetics. To further elucidate the molecular mechanism underlying the action of -HCH on cardiac function, the Ca²⁺ release channel from cardiac SR was incorporated into the planar lipid bilayer, and channel gating activity was measured in the presence of -HCH and dantrolene, singly or in combination. -HCH was found to significantly alter RyR2 channel gating kinetics at concentrations as low as 1 to 10 µM, and the threshold for its actions was dependent on the degree of channel activity determined by the concentration of Ca²⁺ in the cis chamber (cytoplasmic face of the channel) at the start of the experiment. -HCH added to the cis chamber significantly increased P_o in a dose-dependent manner in the concentration range of 5 to 50 µM. This range of concentrations was selected to span those used in the cellular studies while accounting for the well-characterized differences in ligand sensitivity of proteins studied in the planar lipid bilayer. For example, the addition of 50 µM -HCH to the cis chamber in the presence of 50 µM Ca²⁺ increased channel P_o by 2.3-fold from 0.27 ± 0.19 (Fig. 6A, control) to 0.63 ± 0.23 (Fig. 6A, -HCH). The subsequent addition of 50 µM dantrolene to the cis chamber had no significant effect on channel activity (P_o = 0.68 ± 0.26; Fig. 6A, dantrolene). A final addition of 10 µM ryanodine induced the well characterized half-conductance state (Fig. 6A, ryanodine). If Ca²⁺ in the cis chamber was adjusted to give low open probability (P_o = 0.11 ± 0.13), the addition of 50 µM -HCH more dramatically increased channel activity 4.5-fold (P_o = 0.49 ± 0.27), an effect that could be fully blocked by 10 µM ruthenium red (Fig. 6B). The influence of -HCH on single-channel P_o was also observed when dantrolene was first introduced. Figure 6C shows representative plots of a membrane containing three cardiac Ca²⁺ release channels. Under these conditions, the average open probability, P_oavg, for the control trace was determined to be 0.33 ± 0.18 when calculated as described in Experimental Procedures. Dantrolene (50 µM), in the absence of -HCH, did not significantly alter channel P_o from that of control (P_oavg = 0.33 ± 0.18 and 0.41 ± 0.19 for control and dantrolene treatment, respectively). Nevertheless, the subsequent addition of 50 µM -HCH to these channels increased P_oavg by 2.2-fold to 0.73 ± 0.25. As expected, the further addition of 10 µM ruthenium red completely blocked all channels in the membrane (data not shown). These results indicate that -HCH enhanced cardiac SR Ca²⁺ transport and altered myocyte Ca²⁺ signaling by direct interaction with the RyR2-channel complex and that the modified channels remain responsive to known pharmacological reagents of RyR2.

View larger version (37K): [in this window] [in a new window]   Fig. 6.   -HCH, but not -HCH, dramatically alters the gating kinetics of single cardiac Ca2+ channels (RyR2). Cardiac Ca2+ release channels were incorporated into the planar lipid bilayer as described in Experimental Procedures, and current records acquired in the presence of a 500:100 mM CsCl ratio (cis/trans). A, traces show a representative section of channel transitions in the presence of 50 µM Ca2+ (Control); the same channel after exposure to 50 µM -HCH (-HCH); the -HCH-modified channel after the addition of 50 µM dantrolene (Dantrolene), which did not significantly change Po; and the same channel after the addition of 10 µM ryanodine (Ryanodine). This experiment is representative of four trials. Po values for A, B, C, and D are stated in the text. B, channel Po was initially set to a low value by reducing cis Ca2+. The addition of 50 µM -HCH dramatically increased channel activity 4.5-fold. Trace shows channel after the addition of 10 µM ruthenium red (RR) (a known blocker RyR2 function). This experiment was repeated three times with similar results. C, the orders of the addition of -HCH and dantrolene were reversed. The membrane shown contains three channels. The Po for a single channel is calculated as described in Experimental Procedures. Traces are representative of channel transitions in the presence of 50 µM Ca2+ (Control), show the same channel after exposure to 50 µM dantrolene (Dantrolene), and show that the addition of 50 µM -HCH to the dantrolene-treated channel dramatically increased (-HCH). A final addition of 10 µM ruthenium red completely blocked all channels (not shown). This experiment is representative of three trials. D, control channel fluctuations in the presence of 10 µM Ca2+. The addition of 50 µM -HCH had no significant effect on channel Po. The subsequent addition of ruthenium red (10 µM) completely blocked the -HCH-exposed Ca2+ channel. This experiment is representative of three trials. In all traces, solid lines represent the closed state of the channel. All of the traces shown are at a holding potential of +20 mV.

View larger version (12K): [in this window] [in a new window]   Fig. 7.   Analysis of -HCH-exposed channel kinetics. In the presence of a 500:100 mM CsCl gradient (cis/trans) and 25 µM Ca2+, -HCH dose dependently increased Po of a cardiac RyR over a concentration range of 0 to 50 µM. Lifetime analysis of the data reveals that the increase in Po is due to a decrease in channel mean closed time () with a relatively small corresponding increase in channel mean open time (). A, results of analysis of channel gating kinetics for the first time constant, tau1. B, results of analysis of channel gating kinetics for the second time constant, tau2. The data shown are the average ± range of two independent determinations. In B, the range falls within the symbol.

View larger version (17K): [in this window] [in a new window]   Fig. 8.   Exposure of cardiac RyR to -HCH does not alter the intrinsic unitary channel conductance. Cardiac RyR was incorporated into a lipid bilayer as described in Experimental Procedures and channel conductance for Cs+ monitored in the presence of symmetric 100 mM CsCl and 25 µM Ca2+ cis. Channel conductance in the absence of 50 µM -HCH () was 455 pS and was not different from that in the presence of -HCH (). The data shown are representative of three independent I/V determinations before and after the addition of -HCH.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Chlorinated hydrocarbons have been shown to alter intracellular Ca²⁺ homeostasis. In particular, the mechanisms by which HCHs alter cellular Ca²⁺ regulation in excitable and nonexcitable cells have received attention in recent years. However, it is not clear whether these compounds alter Ca²⁺ homeostasis by a selective interaction with critical macromolecules or less specifically by disruption of membrane lipids. Pessah et al. (1992) reported -HCH-induced positive inotropy and contracture in rat atrial strip preparations that was correlated with activation of ryanodine-sensitive Ca²⁺ release from SR vesicles and inhibition of [³H]ryanodine equilibrium binding. Criswell et al. (1994) found that -HCH (lindane) increases intracellular calcium in rat myometrial smooth muscle cells through the modulation of IP₃-sensitive stores because carbachol, but not caffeine and ryanodine, eliminated the lindane-induced increase in intracellular Ca²⁺. In contrast, Mohr et al. (1995) showed that -HCH potently mobilized Ca²⁺ from an intracellular store in cultured RBL cells and concomitantly eliminated depletion-activated Ca²⁺ entry. The mechanism was not blocked by the nonselective IP₃ channel inhibitor heparin, suggesting a mechanism either downstream of the IP₃-binding site (an allosteric site) or independent of the IP₃ receptor. A recent series of reports by Rosa et al. (1996, 1997b) examined the effects of short- and long-term HCH exposure on cytotoxicity and cellular Ca²⁺ homeostasis in cultured cerebellar granule cells. Their work indicates that - and -HCH differentially modulate Ca²⁺ release from multiple intracellular Ca²⁺ stores, including those sensitive to ryanodine and dantrolene. Thus, the molecular mechanisms by which these HCH isomers alter intracellular Ca²⁺ signaling involve mobilization of Ca²⁺ from intracellular stores, although the mechanisms appear to be fundamentally different.

In the present report, we examined the relationship between the ability of - and -HCH to alter functional parameters of isolated ventricular myocytes and their ability to directly interact with cardiac ryanodine-sensitive Ca²⁺ release channels reconstituted in bilayer lipid membranes. Ryanodine-sensitive Ca²⁺ channels are widely expressed and trigger a variety of cell functions (Berridge, 1990). In cardiac muscle, the ryanodine receptor complex is localized at the junctional and corbular regions of SR (Jorgensen et al., 1993) and is responsible for the efflux of calcium during excitation-contraction coupling. Previous findings have shown that ligands known to activate the ryanodine receptor and stimulate SR Ca²⁺ release also stimulate high-affinity [³H]ryanodine binding, whereas ligands which close the channel and inhibit Ca²⁺ release inhibit [³H]ryanodine binding (Pessah et al., 1987; Chu et al., 1990; Zimanyi and Pessah, 1991). Thus, [³H]ryanodine binding has been used as an indicator of the functional state of the channel. Notable exceptions to this paradigm is the disparity between [³H]ryanodine binding and channel function observed with HCHs (Pessah et al., 1992) or Ag⁺ ions (Pessah et al., 1987). In agreement with a previous study performed in rat atria (Pessah et al., 1992), we found in the present study that -HCH inhibits equilibrium binding of [³H]ryanodine to its site on the calcium channel complex from SR preparations isolated from guinea pig ventricles even though its prominent actions on SR Ca²⁺ transport and single-channel activity indicate that it is an effective activator. Although results from equilibrium receptor binding with [³H]ryanodine suggested that -HCH induced Ca²⁺ channel closure, results showing -HCH stimulated Ca²⁺ release from SR membrane vesicles and enhanced single-channel P_o values (Figs. 6 and 7) indicated that -HCH may in fact elicit a time-dependent biphasic action on ryanodine receptor function by first stimulating, then inhibiting RyR channel activity over several hours. Consistent with this interpretation, -HCH was shown to significantly enhance the apparent rate of [³H]ryanodine association with cardiac RyR (Pessah et al., 1992). Therefore, functional measurements of SR Ca²⁺ transport and single-channel activity that examine the initial (seconds to minutes) consequence of -HCH interaction with the channel complex are consistent with its initial effects on the rate of [³H]ryanodine-binding kinetics while equilibrium measurements of [³H]ryanodine-binding that take 3 h suggest a subsequent inhibition of channel activity. The work of Rosa and colleagues (Rosa et al., 1996) using cultured cerebellar granule cells further supports this assertion. They reported that upon exposure to -HCH at concentrations as low as 25 µM resulted in a dramatic increase in intracellular Ca²⁺ within 2-4 min of exposure. However, exposure of the cells to either -HCH or -HCH for 18 h resulted in similar levels of cytotoxicity (as measured by propidium iodide entry) for both isomers. These findings suggest that -HCH can act to modify cellular functions in a heterogeneous, time-dependent manner.

The present study reveals that -HCH is a negative inotropic agent in guinea pig ventricular myocytes which exhibits significant selectivity over -HCH. Moreover, the negative inotropy observed is correlated with a decrease in the peak indo-1 transient. In addition, electrically stimulated cell shortening was significantly reduced at all pacing frequencies in the presence of -HCH. The negative inotropic effects, the prolongation of the indo-1 transient decline, and the decrease in cell shortening observed with -HCH presented here are all consistent with a decreased ability of SR to sequester Ca²⁺. This action of -HCH on SR Ca²⁺ regulation does not appear to be due to nonspecific effects such as membrane permeabilization. Even though -HCH is highly lipophilic, the present results reveal a direct stimulatory action of -HCH on the RyR2 channel complex and our companion paper proves that -HCH is incapable of transporting Ca²⁺ across a phospholipid membrane, and is therefore not acting as Ca²⁺ selective ionophore. Furthermore, it is unlikely that the loss of contractile function and depletion of intracellular Ca²⁺ stores seen with -HCH is the result of Ca²⁺ (Mg²⁺) ATPase (SR pump) inhibition because we have shown -HCH at the concentrations used in the present study only minimally affect SR pump activity (Pessah et al., 1992). The current findings support the hypothesis that the -HCH induced negative inotropy seen in ventricular myocytes is primarily a result of prolonged activation of RyR2 and depletion of intracellular Ca²⁺ stores since reduction of the peak amplitude of the Ca²⁺ transient and the prolongation of the return of intracellular Ca²⁺ to baseline can both be accounted for by the observation that -HCH directly stimulated RyR2.

However, the possibility that -HCH can disrupt the lipid-RyR2 protein interface in a unique manner remains plausible since it has been demonstrated using Raman analyses that the effect of HCH isomers on thermal transition properties of lipids is stereospecific (Verma and Rastogi, 1990). Indeed, -HCH was more disruptive of lipid-protein interactions while -HCH is more disruptive of lipid-lipid interactions. Thus if -HCH has some unique effects on the ion permeability of plasmalemma that alters membrane potential, then this could contribute significantly to the altered excitability observed with isolated cardiomyocytes. In this respect, our companion paper characterizes a Ca²⁺-dependent, K⁺ selective current induced by -HCH but not -HCH.

An interesting difference identified in the present study with isolated ventricular myocytes is the observation that -HCH induced a dose-dependent negative inotropy (Figs. 2-4), whereas it produced a marked positive inotropy in rat atrial strip preparations (Pessah et al., 1992). Several possible explanations can account for these functional differences. First is the probable influence from extrinsic cells capable of secreting neurohumoral agents that are present in atrial strip preparations, but absent in the isolated ventricular cardiomyocytes. Since -HCH has been shown to influence Ca²⁺ transport in neurogenic cells and to alter secretion from nonexcitable cells in a manner consistent with its actions on cardiomyocytes, it is likely that -HCH induced secretion of neurohumoral factors contributes in enhancing contractile force in atrial strip preparations. Second, differences in inotropic responses may stem from significant differences in SR structure between atrial and ventricular tissues (Lytton and MacLennan, 1992). In any case, the observed difference in responses is most likely unrelated to species differences since positive inotropy with -HCH has also been observed in guinea pig (Atrakchi and West, 1985).

The drug dantrolene has been reported to inhibit SR Ca²⁺ release and to either stimulate or inhibit single skeletal RyR depending on concentration (Nelson et al., 1996; Fruen et al., 1997). However the exact mechanism by which dantrolene produces muscle relaxation has remained elusive and it has remained unclear whether the drug directly interacts with RyR or acts through an as yet unidentified accessory protein (Parness and Palnitkar, 1995). Attempts in our lab to inhibit -HCH-, Ca²⁺-, or ryanodine-induced Ca²⁺ release from cardiac SR with up to 50 µM dantrolene failed (unpublished data), and no pronounced effect of dantrolene was observed using direct measurements of single Ca²⁺ channels incorporated into the planar lipid bilayer (Fig. 6). Single-channel P_o values were unaffected by exposure to 50 µM dantrolene either before or after stimulation by -HCH. These findings are in consonance with results using skinned cardiomyocytes which showed that dantrolene induced only a mild negative inotropy (Meissner et al., 1996; Fratea et al., 1997). Taken together, these results suggest that -HCH exerts its effects on ventricular myocytes by direct interaction with an allosteric site on RyR2 and that dantrolene may exert its muscle relaxant properties at a physiological location upstream of this site.

The hexachlorocyclohexanes are highly lipophilic molecules possessing long biological half-lives and much evidence has accumulated indicating that these compounds can be highly toxic to mammals. In spite of this toxicity, HCH continues to be used worldwide, often with high human exposures. The data presented here indicate that the hexachlorocyclohexanes, and in particular -HCH, can mediate their significant biological effects through disruption of intracellular Ca²⁺ regulation by direct interaction with RyR2.