Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Quinine and its stereoisomer quinidine are two alkaloids derived from cinchona bark. Quinine is less potent and less toxic compared with quinidine and has been used as an antimalarial drug for more than a century, whereas quinidine is preferred for the treatment of cardiac arrhythmias (Salata and Wasserstrom, 1988). Both drugs have long been known to produce vasodilation, an increase in renal blood flow, and a decrease in blood pressure in dogs with experimental neurogenic hypertension (Hiatt, 1948). In addition, parenteral administration of therapeutic doses of quinidine and quinine caused forearm vasodilation and decreased mean arterial pressure in humans (Schmid et al., 1974; Mariano et al., 1992). These vasodilatory effects of quinine have been suggested to be due to actions on the adrenergic innervation of vascular smooth muscle as well as direct effects on the vascular smooth muscle cell. In rat cardiac membranes, human platelets, and rat kidney cells, quinidine is a competitive antagonist of ₁- and ₂-receptors (Motulsky et al., 1984). However, the observation that quinidine-induced vasodilation persisted even in denervated limbs suggested that a nonadrenergic effect may also be involved (Nelson et al., 1974; Schmid et al., 1974).

Quinidine and quinine inhibit KCl-induced contractions in rat and rabbit aorta (Cook et al., 1987) and rat rectum (Del Pozo et al., 1996) as well as angiotensin II-stimulated contractions in rabbit aorta (Cook et al., 1987). These compounds also inhibit KCl-induced ⁴⁵Ca uptake in A7r5 cells (Cook and Quast, 1990) and shift the CaCl₂-force response to the right in depolarized guinea pig taenia coli (Spedding and Berg, 1985). Taken together, these findings suggest that quinine inhibits calcium influx through voltage-dependent calcium channels in smooth muscle. Quinine's effects also seem to be at the level of intracellular calcium handling. Quinine has been shown to inhibit calcium release from internal stores of brain microsomes and macrophages (Lee and Go, 1996; Misra et al., 1997). More to the point of this present study, quinine reduced 5-hydroxytryptamine-induced contractions of rat aorta smooth muscle in calcium-free medium (Del Pozo et al., 1996).

In addition to effects on divalent cations, quinine inhibits monovalent conductances in membranes of a variety of different cell types (Walden and Speckmann, 1981; Salata and Wasserstrom, 1988). Using smooth muscle as the experimental model, quinine has been described as a potassium current blocker. For example, quinine has been shown to inhibit both the K_v1.5 (Overturf et al., 1994) and K_v2.2 (Schmalz et al., 1998) currents in vascular smooth muscle. It is difficult, however, to understand how inhibition of an outward potassium current could account for inhibition of a vascular response. A more likely response to inhibition of rectifying potassium currents would be an augmentation or at least prolongation of the contraction (Jackson, 2000).

It is clear that quinine and its isomer quinidine inhibit contraction of vascular smooth muscle by multiple pathways. The pathways proposed to date have suggested membrane receptor and calcium handling as sites of action. However, to our knowledge, no studies have shown whether quinine has effects directly at the level of activation/modulation of the contractile apparatus in vascular smooth muscle. After a stimulus-induced increase in calcium, contraction of vascular smooth muscle is initiated by phosphorylation of the 20-kDa myosin light chain (MLC). This phosphorylation step is catalyzed by the calcium- and calmodulin-dependent MLC kinase (Kamm and Stull, 1985). Relaxation follows dephosphorylation of the MLC by a MLC phosphatase. In addition to the direct regulation of the contractile proteins by the MLC kinase/phosphatase system, the sensitivity of the proteins to calcium can be modulated by a receptor and G protein-dependent pathway (Somlyo and Somlyo, 1994). Therefore, the goal of this study was to determine whether quinine inhibits vascular smooth muscle of the swine carotid artery and if so, is the inhibition solely due to effects on calcium or are steps in the activation or modulation of the contractile proteins also involved. The specific hypothesis tested was that quinine-induced inhibition is not mediated by changes in the level of activator calcium or events upstream from the contractile proteins. If this is correct then it will lend support to the secondary hypothesis that quinine-induced inhibition involves components downstream from the contractile proteins important in mechanotransduction such as paxillin or talin.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Intact and Permeabilized Tissue Preparation. Swine carotid arteries were obtained from a local slaughterhouse and transported to the laboratory in an ice-cold MOPS-buffered physiological salt solution. Arteries were cleaned of connective tissue and then dissected free of both intima and adventitia, leaving a thin medial layer for experimentation. Intact medial strips of swine carotid artery (7 × 0.7 mm) were suspended between a Grass FT.03 force transducer and a stationary clip in water-jacketed organ baths. The strips were equilibrated in physiological salt solution at 37°C, pH 7.4, and bubbled with 100% O₂ for 90 to 120 min. A passive force of 3 g was applied to all tissues. This passive force sets the muscle at a length that approximates L_o. During the equilibration period, tissues were maximally contracted with 30 µM histamine for 5 min at 45-min intervals. For experiments involving permeabilized tissues, medial strips (200 × 700 µm) were mounted in a Muscle Research Station (Scientific Instruments, Heidelberg, Germany) at room temperature and allowed to equilibrate for 90 min under a passive tension of 100 mg. The tissues were contracted with 30 µM histamine every 45 min until a reproducible contraction was attained, followed by incubation in relaxing solution composed of 100 mM K-acetate, 5 mM EGTA, 5 mM MgCl₂, 5 mM Na₂ATP, 20 mM creatine phosphate, 20 mM imidazole, pH 7.0, and 0.5 mM DTT, for 30 min. Tissues were then permeabilized by exposure to either 850 U/ml Staphylococcus aureus -toxin for 30 min or 0.2% Triton X-100 for 18 min. Solutions for the permeabilized tissue studies contained 20 mM imidazole, pH 7.0, 1 mM Mg²⁺, 5 mM ATP, 5 mM EGTA, sufficient K-acetate to maintain ionic strength at 120 mM, and levels of free Ca²⁺ appropriate for the particular experimental design. In addition, all solutions used with -toxin permeabilized tissues contained 1 µM ionomycin and 5 mM creatine phosphate, and all solutions used with Triton X-100 permeabilized fibers contained 0.5 µM calmodulin.

Determination of Myosin Light Chain Phosphorylation Levels. Muscle tissues that were used for determination of myosin light chain phosphorylation levels were treated identically to those used for force measurement. The tissues were rapidly frozen, at rest or after stimulation in an acetone/dry ice slurry containing 6% trichloroacetic acid and 10 mM DTT. The tissues were slowly brought to room temperature and rinsed in acetone containing 10 mM DTT. The tissues were then air dried, weighed, and homogenized on ice in a solution containing 6 M urea, 50 mM Tris pH 6.8, 10 mM DTT, 10 mM EGTA, 5 mM EDTA, and 5 mM NaF. Homogenized samples were assayed for total protein using the Bradford technique. Five micrograms of protein from each sample was then subjected to two-dimensional gel electrophoresis followed by transfer to nitrocellulose membrane as described previously (Moreland et al., 1992). Proteins were visualized using colloidal gold (Amersham Biosciences, Piscataway, NJ). MLC phosphorylation levels were quantified by densitometric analysis of optically scanned images. Phosphorylation levels were calculated by determining the volume of phosphorylated MLC as a percentage of the volume of both phosphorylated and unphosphorylated MLC.

Chemicals. Quinine, indomethacin, 1H-[1,2,4]oxadiazole[4,3-]quinoxalin-1-one (ODQ), ATP, and histamine were obtained from Sigma-Aldrich (St. Louis, MO). Ionomycin and GTPS were purchased from Calbiochem (La Jolla, CA). S. aureus -toxin was obtained from List Biological Laboratory (Campbell, CA).

Statistics. All results are expressed as the means ± S.E.M. with n representing the number of observations. Data were compared for statistical significance using the Student's t test (paired and unpaired). Probability level <0.05 was taken as statistically significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Quinine Inhibition of Histamine-Induced Contraction. The first set of experiments was designed to determine whether quinine inhibited intact strips of swine carotid artery in a concentration-dependent manner. The strips were subjected to the cumulative addition of histamine (0.1-100 µM) after which they were rinsed and allowed to fully relax. The strips were then incubated in varying concentrations of quinine (100-400 µM) and then subjected again to the cumulative addition of histamine. Data were normalized to the force generated in response to a single contraction of 10 µM histamine before the initiation of the cumulative concentration-response curves. As shown in Fig. 1, incubation of the vascular strips with different concentrations of quinine induced a concentration-dependent inhibition of the contractile response to histamine. Increasing quinine concentration produced concentration-dependent depression of the maximal histamine-stimulated force generation. Quinine also decreased the sensitivity of the strips to histamine as evidenced by a significant increase in the EC₅₀ value (control, 2.8 ± 0.3 µM; 100 µM quinine, 3.0 ± 0.6 µM; 200 µM quinine, 4.2 ± 0.7 µM; and 400 µM quinine, 6.6 ± 0.7 µM). Time controls were performed to ensure consistency of histamine concentration-response curves over the same time frame as the quinine inhibition experiments (data not shown).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Effect of quinine on cumulative histamine concentration-response curves in intact swine carotid arterial strips. Histamine (0.1-100 µM) was added to the muscle bath chambers in a cumulative manner. Fifteen minutes before the addition of histamine, the tissues were exposed to either 0 (), 100 (), 200 (), or 400 () µM quinine. Quinine inhibited histamine-induced contractions in a concentration-dependent manner. Values shown are the means ± S.E. for five to six determinations. Asterisk (*) denotes significantly different from 0 quinine; P < 0.05.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Effect of indomethacin and ODQ on quinine-induced inhibition of swine carotid arterial contraction. A, smooth muscle strips were incubated in vehicle (), 1 µM indomethacin (), 1 µM indomethacin, and 300 µM quinine (), or 300 µM quinine alone () then subjected to the cumulative addition of histamine (0.1-100 µM). B, smooth muscle strips were incubated in vehicle (), 10 µM ODQ (), 10 µM ODQ, and 300 µM quinine (), or 300 µM quinine alone () then subjected to the cumulative addition of histamine (0.1-100 µM). Neither indomethacin nor ODQ pretreatment significantly affected the ability of quinine to inhibit the contraction. Values shown are the means ± S.E. for four to five determinations.

Effect of Quinine on Myosin Light Chain Phosphorylation. The primary step involved in activation of smooth muscle contraction is the calcium- and calmodulin-dependent phosphorylation of the 20-kDa myosin light chain (Kamm and Stull, 1985). Therefore, inhibition of myosin light chain phosphorylation is a logical site of action for the quinine-induced reduction in contraction. Moreover, although a precise relationship does not exist between calcium and myosin light chain phosphorylation, in most cases myosin light chain phosphorylation levels follow a directional change in calcium concentration. Thus, we measured histamine-induced levels of myosin light chain phosphorylation in the presence and absence of quinine, and the results are shown in Fig. 3. The data in Fig. 3 demonstrate that neither 300 µM quinine (open symbols) nor 600 µM quinine (inserted column) have any effect on basal or peak (1 min) histamine-induced increases in myosin light chain phosphorylation. The time-course data shown in Fig. 3 indicate that 300 µM quinine does produce a small but significant depression in myosin light chain phosphorylation levels after 10 min of histamine stimulation, suggesting that steady-state calcium levels may be decreased or MLC phosphatase activity may be increased. However, even taking into account the potential for cooperativity (Butler and Siegman, 1998), this small albeit significant decrease in myosin light chain phosphorylation is not consistent with the nearly 60% decrease in force shown in Fig. 1.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Effect of quinine on histamine-induced changes in myosin light chain phosphorylation levels. Vascular smooth muscle strips were incubated in the absence () or presence of 300 µM () or 600 µM (inserted column) quinine for 15 min and then stimulated with 30 µM histamine. MLC phosphorylation levels were quantified at 0, 0.5, 1, 3, and 10 min of histamine stimulation. Quinine did not significantly affect the level of histamine-induced increase in phosphorylation during stimulation durations of up to 3 min. Values shown are the means ± S.E. for four to five determinations.

Effect of Quinine on Ca²⁺-Induced Contraction of Permeabilized Vascular Tissue. To directly test the hypothesis that quinine inhibits vascular contractile activation directly rather than indirectly through actions in calcium metabolism, we used the S. aureus -toxin permeabilized preparation. This preparation allows control of the intracellular environment while maintaining physiologically relevant signaling pathways intact. The permeabilized strips were subjected to the cumulative addition of Ca²⁺ (0.3-3 µM) in the absence and presence of 300 µM quinine; the results are shown in Fig. 4. Even in conditions of constant free calcium (5 mM EGTA and 10 µM ionomycin), quinine produced significant decreases in force development at every [Ca²⁺]. Smooth muscle is known to be modulated by receptor-dependent pathways upstream from the contractile filaments. To determine whether quinine is acting at the level of receptor-mediated, G protein-dependent pathways, we contracted the -toxin permeabilized strips with Ca²⁺ plus GTPS in the absence and presence of 300 µM quinine (Fig. 5). Quinine had no effect on the GTPS-dependent increase in force development, suggesting receptor and G protein-mediated effects are not involved.

View larger version (11K): [in this window] [in a new window]   Fig. 4.   Effect of quinine on Ca2+-induced contraction of -toxin permeabilized vascular smooth muscle strips. Permeabilized strips were incubated in the absence () or presence () of 300 µM quinine and then subjected to increasing [Ca2+]. Quinine significantly depressed Ca2+-dependent force development in the -toxin permeabilized arterial strips. Values shown are the means ± S.E. for four to six determinations.

View larger version (31K): [in this window] [in a new window]   Fig. 5.   Effect of quinine on calcium plus GTPS-induced contraction of -toxin permeabilized strips. -Toxin permeabilized strips were incubated in the absence or presence of 300 µM quinine and then stimulated with 0.5 µM Ca2+ plus 100 µM GTPS. Stimulation with 0.5 µM Ca2+ alone is shown for comparison. Values are the means ± S.E. for four to six determinations.

View larger version (27K): [in this window] [in a new window]   Fig. 6.   Differential effect of quinine on Ca2+-dependent contractions of -toxin permeabilized and Triton X-100 skinned vascular strips. -Toxin permeabilized and Triton X-100 skinned strips were incubated in the absence or presence of 300 µM quinine and then stimulated with 0.7 µM Ca2+. Quinine significantly depressed Ca2+-dependent contractions of -toxin permeabilized preparations but had no effect on Ca2+-dependent contractions of the Triton X-100 skinned tissues. Values shown are the means ± S.E. for four to six determinations.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

We have demonstrated in the present study that quinine inhibits histamine-induced contraction of isolated strips of swine carotid artery smooth muscle. This finding is consistent with previous reports describing the actions of quinine on rabbit and rat aorta (Cook et al., 1987; Del Pozo et al., 1996). Quinine was earlier identified as a sympatholytic drug (Hiatt, 1950) exhibiting a higher degree of inhibition of the ₁-adrenoceptor compared with the ₂-adrenoceptor (Motulsky et al., 1984). The results of our study do not rule out the possibility that quinine produces receptor blockade, but instead have extended the list of inhibitory effects of quinine on smooth muscle contractility to include intracellular sites.

Vascular smooth muscle can synthesize and release prostanoids that are capable of modulating the contractile response (Mombouli and Vanhoutte, 1999). Bioactive prostanoids are derived from enzymatic activity downstream of cyclooxygenase. Inhibition of cyclooxygenase activity by indomethacin did not abolish the inhibitory effect of quinine on vascular smooth muscle. Thus, we were able to rule out the possibility that quinine-induced inhibition is mediated by either the activation of relaxing prostanoids or the inhibition of contractile prostanoids. A second relaxation pathway that could account for the effect of quinine is the NO/cGMP cascade. Endogenous NO and related nitrovasodilators regulate vascular smooth muscle contraction by activation of soluble guanylate cyclase, elevation of cGMP, and activation of cGMP-dependent protein kinase (Ignarro et al., 1999). Cyclic GMP-mediated vascular smooth muscle cell relaxation is characterized by a reduction in intracellular calcium concentration (Lincoln et al., 2001) and by activation of the MLC phosphatase, which dephosphorylates phosphorylated MLC (Lee et al., 1997; Surks et al., 1999). In addition, NO directly stimulates Ca²⁺ activated K⁺ channels in smooth muscle cells (Bolotina et al., 1994; Koh et al., 1995), resulting in hyperpolarization, a decrease in Ca²⁺ influx, and consequently muscle relaxation (Lincoln and Cornwell, 1991; Robertson et al., 1993; Mazzuco et al., 2000). Thus, we examined the role played by cGMP-dependent mechanisms in quinine-induced relaxation of the artery. ODQ had no effect on quinine-induced relaxation of the arterial strips. This suggests that the quinine-induced relaxation in porcine carotid artery is not mediated through a cGMP-dependent pathway.

As stated above, an increase in intracellular Ca²⁺ initiates smooth muscle cell contraction by activation of the calmodulin-dependent MLC kinase that catalyzes phosphorylation of the 20-kDa MLC. MLC phosphorylation activates the myosin molecule, allowing interaction with actin and the resultant active crossbridge cycling and development of force. In view of this, we determined whether inhibition of the contraction by quinine is related to the phosphorylation level of MLC. Our results demonstrate that quinine inhibits contraction without significantly affecting MLC phosphorylation levels, at the initial force development stage of contraction. This supports the hypothesis that quinine-induced decreases in contraction are not associated with changes at the level of the Ca²⁺/calmodulin complex or MLC kinase/phosphatase. This possibility is also supported by the finding that 600 µM quinine completely abolished a histamine-induced contraction without significantly affecting MLC phosphorylation levels at the typical peak value at 1 min of stimulation. If MLC phosphatase activity was increased by quinine, MLC phosphorylation levels would be expected to be decreased. Our suggestion is again supported by the fact that quinine failed to inhibit Ca²⁺-induced contraction in Triton X-100 permeabilized strips, a preparation typically devoid of modulatory signaling pathways. Intact smooth muscle treated with quinine had lower steady-state levels of force and the time-dependent decrease in MLC phosphorylation levels was more rapid in the presence of quinine. Although the decrease in maintained force was significantly greater than the decrease in MLC phosphorylation, the fact that MLC phosphorylation levels were lower suggests that in addition to any calcium-independent pathways affected by quinine, the Ca²⁺-dependent pathway may also be impaired or MLC phosphatase activity may be enhanced.

It is well known that the calcium sensitivity of smooth muscle contractile filaments can be modulated (Morgan and Morgan, 1984). One of the primary models used to study this phenomenon has been the -toxin permeabilized fiber. Smooth muscles permeabilized with -toxin retain their responsiveness to agonist activation and maintain intracellular signaling pathways, while allowing control of the intracellular environment (Nishimura et al., 1988; Gong et al., 1996). Quinine significantly inhibited Ca²⁺-dependent contractions of -toxin permeabilized strips, whereas Ca²⁺-dependent contractions of the Triton X-100 permeabilized strips were unaffected. The primary difference between these two permeabilized preparations is the presence of modulatory pathways in the -toxin preparation compared with the lack of these pathways in the Triton X-100 preparation. This information, coupled with the finding that quinine did not depress initial MLC phosphorylation levels supports the speculation that quinine does not inhibit contraction by alterations in the G protein-dependent change in myofilament calcium sensitivity but instead inhibits contraction downstream from the contractile apparatus.

Our results do not rule out the thin filaments as a site for quinine inhibition. It has been, in our opinion, clearly shown that thin filament proteins can regulate or at least modulate smooth muscle contraction (Earley et al., 1998; Wang 2001). The Triton X-100 permeabilized fibers contain physiological levels of thin filament proteins. However, they may or may not contain the components required for activation/inhibition of the thin filament regulatory proteins. This could account for the lack of effect of quinine in the detergent skinned fiber compared with the pronounced inhibition at constant calcium in the -toxin permeabilized fiber.

One growing area of study on smooth muscle modulation is at the connection of the actin- and myosin-containing contractile lattice structure with the cell membrane. Gunst and collaborators have provided, over the years, compelling evidence suggesting that the mechanical transduction of the work performed by the crossbridges to the cell membrane may be modulated by proteins at the level of the focal adhesions (Pavalko et al., 1995; Wang et al., 1996). Most notably, phosphorylation levels of the proteins talin and paxillin have been shown to change with mode of contractile stimulation. It is interesting to speculate that quinine may alter the pathways involved in talin and/or paxillin phosphorylation and thus change the coupling efficiency between the crossbridges and the focal adhesions on the cell membrane.

In summary, we have shown that quinine inhibits contraction of the swine carotid arterial smooth muscle. Our results suggest that the mechanism by which quinine inhibits contraction is not exclusively through a decrease in activator calcium concentrations. We have also shown that quinine does not inhibit contraction by stimulation of pathways involved in active relaxation, at least not through the prostaglandin or nitric oxide pathways. Instead, our results suggest that quinine-dependent inhibition of contraction involves a modulatory pathway present in the smooth muscle cell. Based in part on the fact that 1) quinine inhibited Ca²⁺-dependent contractions of the -toxin permeabilized preparation, which retains modulatory pathways both upstream and downstream from the contractile proteins; 2) quinine did not inhibit the GTPS-induced contraction of the -toxin permeabilized preparation important in upstream modulation of contraction; and 3) quinine did not inhibit MLC phosphorylation levels and force in the Triton X-100 permeabilized strips, which are devoid of all modulatory pathways; we suggest that a pathway(s) downstream from the contractile proteins is the site of quinine-dependent inhibition. Therefore, we propose the hypothesis that quinine decreases the coupling of the contractile filaments with the focal adhesions on the sarcolemma.