Inhibitory Action of 4-Aminopyridine on Ca2'-ATPase of the Mammalian Sarcoplasmic Reticulum*

In the isolated guinea pig diaphragm muscle, 4-ami-nopyridine (4-AP) elicited a marked potentiation of twitch contraction evoked by direct electrical stimuli. Although tetraethylammonium (TEA) and charybdotoxin only slightly potentiated twitch contraction, 4- AP, but not TEA, also augmented a contractile response to caffeine. These effects of 4-AP on muscle contraction could not be interpreted by a simple inhibition of potassium channels on the plasma membrane. In the fragmented sarcoplasmic reticulum (SR) prepared from the guinea pig psoas muscle, 4-AP inhibited the ATP-driven Ca2+ uptake from the extravesicular medium. Furthermore, 4-AP at concentrations less than 10 mM elicited a selective inhibition of Ca2+-activated SR ATPase in a competitive manner against the Ca2+ concentration of the medium and 10 mM 4-AP showed the unsurmountable inhibition. 4-AP at 30 mM apparently inhibited activities of other ATPases such as Na+,K+- and myosin ATPases. In contrast, other potassium channel blockers such as TEA, apamin, charybdotoxin, and glibenclamide did not inhibit the SR function. These results suggest that, although the specific concentration range is rather small, 4-AP elicits an inhibition of SR Ca2+-pumping activity, leading to the marked potentiation of muscle contractile responses to electrical stimuli and caffeine. The on

The inhibitory action of 4-aminopyridine (4-AP)l on potassium channels of plasma membrane has been revealed after the finding of its stimulatory effects on neuromuscular transmission through experimental and clinical studies in Bulgaria (1). Although the 4-AP-induced increase in the quantity of acetylcholine liberated from nerve endings has been established (1,2), 4-AP, especially a t higher concentrations around the millimolar range, could increase contractility of rat and frog skeletal muscle preparations through the direct action on muscle cells (3)(4)(5). Since, in the presence of 4-AP, changes in membrane potential were not always associated with contractile response (6), and the relaxation time was prolonged (3), it has been suggested that 4-AP may have a direct action on the resequestration of Ca2+ in the muscle cell. Also, in cardiac muscle of the dog ventricle, the prolongation of the action potential was much smaller than the prolongation of the * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. relaxation time, suggesting that the primary action of 4-AP may be to alter the intracellular Ca2+ metabolism (7). Our preliminary experiments using diaphragm muscle of the guinea pig revealed that, in contrast to tetraethylammonium (TEA), another potassium-channel blocker (8), 4-AP elicited a marked augmentation of contractile responses to caffeine which is known to release Ca2+ from the SR (9). Thus, we made SR preparations and investigated the effects of 4-AP on the SR Ca2+-pumping activity and Ca2+-activated ATPase activity. We report here that 4-AP has a direct action on the mammalian sarcoplasmic reticulum (SR) which may associate with increase in contractility.
Ca2+-pumping Activity of the Fragmented SR-The Ca2+-pumping activity of the fragmented SR was measured using a Ca2+ electrode as described elsewhere (11,12). The heavy fractions of the fragmented SR were prepared from the guinea pig psoas muscle by the modified method of Kim et al. (13). The heavy fraction of the fragmented SR was stored a t 0 "C and used within 3 days. The fragmented SR (1.5 mg of protein/ml) was incubated in the medium composed of (in mM) KC1, 100; MgC12, 0.5; and MOPS, 50 at pH 7.0 (30 "C). Then, CaCI2, 50 /IM, fragmented SR, 0.5-1 mg of protein/ml, phosphocreatine, 5 mM; and ATP, 1 mM plus creatine kinase, 0.1 mg/ml, were successively added to the medium. After the medium Ca2+ was taken up by SR, various drugs were administered. A Ca2+ electrode was made using a Ca2+ mixture (Fluka, Switzerland). ATPase Actiuities-Myosin and deoxycholic acid-treated SR were prepared from the guinea pig psoas muscle according to the methods of Perry (14) and Yamamoto and Tonomura (15), respectively. Pig brain Na+,K+-ATPase preparation was purchased from Sigma. ATPase activities were determined by measuring the liberated Pi according to Martin and Doty (16). Compositions of the reaction medium were as follows. SR preparation, 2 fig protein/ml; CaCl,, a t desired concentration, MgC12, 5 mM; ATP, 2 mM; KC1, 90 mM; EGTA, 1 mM; and HEPES, 30 mM (pH 7.1, 30 "C) for SR ATPase; myosin, 0.1 mg/ml; actin, 0.1 mg/ml; KCI, 50 mM; MgC12, 2 mM; ATP, 2 mM; EGTA, 1 mM; and Tris, 20 mM (pH 6.8,30 "C) for myosin ATPase; and enzyme preparation, 0.01 unit/ml; NaC1, 100 mM; KC1, 20 mM; MgC12, 5 mM; ATP, 2 mM; and Tris, 50 mM (pH 7.4, 37 "C) for Na+,K+-ATPase. The SR Ca2+-ATPase activity was estimated by the difference between ATPase activities in the presence and absence of Ca". M $ + -ATPase activity of the SR was approximately 2% of the total activity. Free Ca2+ concentration was calculated on a computer as described by Fabiato and Fabiato (17).
Statistics and Drugs-Values are expressed as mean f S.E., and the number of experiments was a t least three unless otherwise stated. The statistical analysis was performed by Student's t test.
Watanabe of the Peptide Institute Inc., Minoh, Osaka, Japan. Cromakalim and nicorandil were generous gifts from Beecham-Japan, Tokyo and Chugai Pharmaceutical Co., Tokyo, respectively.

RESULTS
Isolated diaphragm muscle of the guinea pig caused twitch contractions in response to strong electrical stimuli which directly activated muscle cells. When 4-AP (0.01-10 mM) was administered, it potentiated twitch contractions dose-dependently ( Fig. 1). Low concentrations of 4-AP (e.g. 1 mM or less) elicited a sustained potentiation of twitch contraction. The high concentration such as 10 mM elicited a transient potentiation and an increase in basal tension. In the presence of 3 and 10 mM 4-AP, the twitch amplitude was more than twice of control. Such a strong potentiation of contraction with a similar 4-AP concentration range was also reported in the rat diaphragm (19). The presence of d-tubocurarine (0.1 mM) did not inhibit the potentiation induced by 1 mM 4-AP, suggesting that the 4-AP-induced potentiation is not mediated through the acetylcholine liberation from nerves, as already reported (4). On the other hand, administration of TEA (1-30 mM) also potentiated twitch contraction, but the potentiation was much smaller than that induced by 4-AP. Charybdotoxin (1 FM) but not apamin (1 FM), peptide potassium channel blockers (20, 21), also potentiated twitch contraction by 30% of control, a similar magnitude to that induced by 30 mM TEA (data not shown).
When external Ca2+ was removed, twitch contraction gradually decreased (Fig. IA, c and d ) . During the falling phase of twitch contraction, an administration of 10 mM 4-AP elicited a similar increase in twitch contraction to that in the normal medium (Fig. lA, c ) , being consistent to the reported results in frog muscle (4). 10 mM 4-AP also raised the basal tension in the absence of external Ca"; although 30 mM TEA elicited a very small potentiation in the absence of external Ca2+ (Fig.  lA, d).
Next, we examined the effects of 4-AP on the contractile response of the diaphragm muscle to caffeine (Fig. 2). An  administration of 3 mM caffeine elicited a small contraction with the magnitude of 12.9 f 1.61% ( n = 26) compared with the twitch amplitude evoked by supramaximal electrical stimuli. The presence of 4-AP (1-10 mM) also elicited a small contraction in a dose-dependent manner (Fig. 2). In the presence of 4-AP at 1 mM, but not 0.3 mM, the caffeineinduced contraction was augmented approximately five times as much as control. Raising 4-AP concentrations to 3 and 10 mM attenuated the degree of augmentation gradually, but at 10 mM 4-AP, the contractile response to caffeine was still twice greater than control, whereas the presence of 30 mM TEA did not elicit contraction or augment the caffeine-induced contraction (Fig. 2).
To understand the biochemical mode of action, we made SR preparations from the guinea pig psoas muscle (a white muscle). Using a Ca2+ electrode, Ca2+-pumping activity of fragmented SR and caffeine-induced Ca2+ release from SR were measured (Fig. 3). An addition of ATP to the SRincubated medium gradually lowered the medium Ca2+ concentration, indicating the active Ca2+ uptake by SR. After the Ca2+ level was lowered, administration of caffeine (0.5 mM) elicited a transient increase in the medium Ca2+ concentration, as reported previously (11,12). The presence of 4-AP (10 mM) delayed the active Ca2+ uptake and prolonged the duration of Ca2+ increase induced by caffeine. In the presence of 10 mM 4-AP, time to reach the nominally Ca2+-free level after the ATP addition prolonged, 1.97 f 0.05 ( n = 4), the times control significantly ( p < 0.05), suggesting that the presence of 4-AP inhibits the SR Ca2+-pumping activity, whereas 3 mM 4-AP did not show significant effects on the Ca2+ uptake process and caffeine-induced response. On the other hand, 30 mM TEA did not affect the active Ca2+ uptake or the caffeine-induced Ca2+ release. Potassium channel blockers of 1 PM charybdotoxin, 1 PM apamin, and 10 PM glibenclamide.(22) or potassium channel activators (23) of 10 PM cromakalim and 10 p M nicorandil did not affect the active Ca2+ uptake or the caffeine-induced Ca2+ release (data not shown). Effects of 4-AP on the various ATPase activities were investigated (Fig. 4). When the Ca2+-ATPase of the deoxycholate-treated SR prepared from the guinea pig psoas was T h e level before addition of Ca2+ represents the nominally Ca2+-free level, about 1 p~. Lowercase letters represent the following. a, addition of CaC1,; b, addition of the fragmented SR c, addition of phosphocreatine; and d, addition of ATP and creatine kinase to the medium. A prompt reduction and the following gradual decrease in Ca2+ concentration after addition of ATP are due to the formation of the ATP. Ca2+ complex and the active Ca2+ uptake by SR, respectively. nearly maximally activated in the presence of 1 pM Ca2+, 4-A P (0.3-30 mM) dose-dependently inhibited the ATPase activity. Against the basal ATPase (Me-activated ATPase) of the deoxycholate-treated SR and the Na+,K+-ATPase of the hog brain, 0.1-3 mM 4-AP did not elicit inhibition and 10 mM 4-AP inhibited these ATPase activities by 7-10% of control (Fig. 4). The actin-activated myosin ATPase prepared from the guinea pig psoas was not affected by 3 and 10 mM 4-AP (Fig. 4). 4-AP at 30 mM abolished the SR Ca2+-ATPase activity and inhibited other tested ATPase activities by 30-55% of control (Fig. 4). In addition, 10 mM 4-AP did not inhibit the myosin ATPase activities in the presence of Ca2+ or in the presence of EDTA and K+ (data not shown). Other potassium channel blockers or activators of 1 p~ charybdotoxin, 1 p~ apamin, 1 p~ glibenclamide, 10 p M cromakalim, and 10 PM nicorandil did not inhibit any kind of ATPase activities tested (data not shown). The effects of TEA on ATPase activities could not be tested, since it interfered with the formation of the Pi-molybdate complex for the determination of liberated Pi through ATPase reaction. tions of more than 0.01 mM, which is consistent with those reported previously (24,25). The presence of 4-AP (2, 4, and 6 mM) seemed to dose-dependently shift the Ca2+ concentra-tion-ATPase activity relationship to the right in a parallel manner (Fig. 5A). 4-AP at 10 mM apparently showed an unsurmountable inhibition of the SR Ca2+-ATPase activity.
Using values of ATPase activity in the presence of various tested concentrations of 4-AP atpCa 6,6.2, and 6.5, the Dixon plot (26) for the determination of enzyme inhibitor constants ( Fig. 5B) reveals that 4-AP inhibits the SR Ca2+-ATPase in a competitive manner against Ca2+ activation with the apparent dissociation constant of 2.4 mM, although the 4-AP concentration for the competitive inhibition is limited less than 10 mM.

DISCUSSION
Among tested potassium channel modulators, only 4-AP elicited the inhibition of Ca2+-pumping activity of the fragmented SR and Ca2+-activated SR ATPase. Therefore, this inhibitory action may not be related to the known mode of action, the potassium channel inhibition, but rather may be defined as a novel action for 4-AP. 4-AP at concentrations less than 10 mM elicited a significant inhibition of Ca2+-ATPase activity of the SR but had little or no effect on other tested ATPase activities, whereas 30 mM 4-AP more or less inhibited all tested ATPase activities. These results suggest that 4-AP elicits a specific inhibition of Ca2+-activated SR ATPase activity, although the specific concentration range is narrow.
4-AP at concentrations of 2-6 mM seemed to inhibit the Ca2+-activated SR ATPase activity in a competitive manner against the medium Ca2+ concentration. 4-AP a t concentrations more than 10 mM elicited an unsurmountable inhibition of the Ca2+-activated SR ATPase activity. The mechanism of action for such a high concentration of 4-AP could not be well defined at present but seems to be nonspecific, since 4-AP at 30 mM apparently elicited the inhibitory effects on activities of Me-activated SR ATPase as well as other tested ATPases. Presumably, these competitive and nonspecific actions of 4-AP at low and high concentrations represent the fact that the Ca2+-pumping activity of the fragmented SR was inhibited only by the high concentration (10 mM) of 4-AP in the presence of a high concentration (50 PM) of extravesicular Ca2+.
Considering that 4-AP could penetrate the cell across the plasma membrane (l), 4-AP may be capable of interacting with the SR membrane, although the real 4-AP concentration at the cytoplasm could not be determined. Thus, 4-AP elicited a n augmentation of contractile response to caffeine, presumably due to the inhibition of resequestration of Ca2+ to the S R through the inhibitory effects on Ca2+-pumping activity as described above. The 4-AP-induced potentiation of twitch contraction evoked by electrical stimuli may be explained by the combined mechanism of interference with SR Ca2+-pumping activity and inhibition of potassium channels on the plasma membrane; presumably the former is a major factor for the potentiation than the latter, since other potassium channel blockers of TEA and charybdotoxin elicited much smaller potentiation compared to 4-AP. Present experiments clearly demonstrate that, although the specific concentration range for 4-AP is small, 4-AP inhibits Ca2+-ATPase activity and active Ca2+ uptake of the SR, presumably resulting in the marked potentiation of skeletal and cardiac muscle contractile responses. This mode of action may also account for the smooth muscle response, since, in the absence of external ea2+ plus EGTA, the guinea pig isolated aorta produced a big contractile response 'to 10 mM 4-AP, but not to 30 mM TEA.' A similar intervention of 4-AP to intracellular Ca2+ metabolism was suggested in nerve cells (27)(28)(29). Thus, this newly defined mode of action for 4-AP may be adopted in the cellular reponses through excitation-contraction coupling as well as excitation-secretion coupling of nerve cells.