Neuroprotective use-dependent blockers of Na+ and Ca2+ channels controlling presynaptic release of glutamate.

We have originated a family of N,N'-disubstituted guanidines that block the voltage-activated Ca2+ and Na+ channels governing glutamate release. These compounds, CNS 1237 (N-acenaphthyl-N'-methoxynaphthyl guanidine) and its analogues, are "use dependent" in their ability to attenaute neurotransmitter release: they block glutamate release with greater efficacy under conditions of persistent or repetitive depolarization, as would be encountered under pathophysiological circumstances, relative to their ability to block glutamate release elicited by brief, transient depolarizations more characteristic of normal physiological release events in nonischemic brain. Using electrophysiological and rapid kinetic methods, we have differentiated the use-dependent block of the relevant Na+ and Ca2+ channels governing neurotransmitter release from the mechanism of channel antagonism exhibited by, respectively, the substituted guanidine Na+ channel blocker tetrodotoxin (TTX) and venom peptide Ca2+ antagonists. To characterize use-dependent Na+ channel block by CNS 1237, we have employed whole-cell voltage-clamp recordings from a Chinese hamster ovary (CHO) cell line expressing cloned mammalian type II Na+ channels. These experiments demonstrated that, in contrast to the actions of TTX under the same conditions, the potency of Na+ channel block by CNS 1237 is greatly enhanced by depolarizing stimuli in a frequency-dependent manner. Ca2+ channel-activated glutamate release from brain nerve terminal preparations was measured with approximately 300 msec time resolution over a 5-second period of high K(+)-depolarization, using a rapid superfusion technique. CNS 1237 and analogues, at 1-3 microM, accelerated the decay of glutamate release by 40-70%, reflecting depolarization-induced enhancement of block. In contrast, blockade of glutamate release by the Ca2+ channel antagonist peptide toxins omega-aga IV-A (from spider venom) and omega-conotoxin M-VII-C (from cone snail venom) exhibited "reverse-use-dependence:" at concentrations of 0.3 microM, which blocked the initial amplitude of glutamate release by 40-60%, the decay time constant for glutamate release was significantly increased, indicating depolarization-induced relief of block. These findings establish that CNS 1237 and other members of this compound series are use-dependent blockers of the voltage-activated ion channels governing glutamate release. Studies of CNS 1237 in the rat middle cerebral artery occlusion (MCAO) focal stroke model have indicated infarct size reduction comparable to that observed by the same investigators for the glutamate release blocker (BW 619C89 (Burroughs-Wellcome, now in clinical development). Maximal infarct size reduction is achieved with a 3-mg/kg bolus followed by a 4-hour infusion of 0.75 mg/kg/hr.(ABSTRACT TRUNCATED AT 400 WORDS)

New CNS-specific calcium antagonists.

Ischemic insults to the brain in stroke or traumatic brain injury produce excessive release of glutamate from depolarized nerve terminals. This excessive glutamate release in turn stimulates massive calcium entry into nerve cells, activating a biochemical cascade that results in cell death. A major pathway of calcium entry into depolarized nerve cells is through voltage-sensitive, high threshold calcium channels. A large fraction of this calcium entry is mediated through "R-type" calcium channels, channels resistant to blockage by dihydropyridine calcium antagonists such as nimodipine. A newly discovered compound derived from spider venom, CNS 2103, antagonizes both R-type channels and dihydropyridine-sensitive ("L-type") calcium channels. This broad spectrum of action, coupled with selectivity for calcium channels over other classes of voltage-sensitive and ligand-gated ion channels, makes CNS 2103 an interesting lead for development of drugs to treat ischemic brain injury. Activation of presynaptic ("N-type") calcium channels in nerve terminals is a primary cause of excessive neurotransmitter release in brain ischemia. Prevention of glutamate release by blockade of N-type channels in glutamatergic nerve terminals may, at an early stage in the pathophysiological cascade, abort the process leading to nerve cell death. Cambridge NeuroScience has developed a novel rapid kinetic approach for monitoring glutamate release from brain nerve terminals in vitro, and this has led to CNS 1145, a substituted guanidine that selectively blocks a kinetic component of calcium-dependent glutamate release mediated by persistent depolarization. Additional evidence suggests that CNS 1145 antagonizes presynaptic N-type calcium channels, and this may account at least in part for its ability to block glutamate release.

Calcium as a coagonist of inositol 1,4,5-trisphosphate-induced calcium release.

Inositol 1,4,5-trisphosphate (IP3)-induced calcium release from intracellular stores is a regulator of cytosolic-free calcium levels. The subsecond kinetics and regulation of IP3-induced calcium-45 release from synaptosome-derived microsomal vesicles were resolved by rapid superfusion. Extravesicular calcium acted as a coagonist, potentiating the transient IP3-induced release of calcium-45. Thus, rapid elevation of cytosolic calcium levels may trigger IP3-induced calcium release in vivo. Extravesicular calcium also produced a more slowly developing, reversible inhibition of IP3-induced calcium-45 release. Sequential positive and negative feedback regulation by calcium of IP3-induced calcium release may contribute to transients and oscillations of cytosolic-free calcium in vivo.

A superfusion system designed to measure release of radiolabeled neurotransmitters on a subsecond time scale.

A new method for subsecond measurement of release of neurotransmitters from nerve terminal preparations (e.g., synaptosomes) in vitro is described. Synaptosomes were prelabeled with [3H]GABA via a Na-dependent GABA uptake system. The prelabeled nerve terminals are retained on small glass fiber filters in a superfusion chamber accessed by three high speed, solenoid-driven valves. Microcomputer-programmed circuitry controls the timing of valve operation. Each valve controls the delivery of a separate solution to the chamber, permitting rapid and independent control of membrane potential, [Ca2+]e, and drug delivery. The minimal dead volume of the chamber and the relatively high solution flow rate afford time resolution for release of at least 60 ms. This time resolution was necessary to observe the most rapid of at least three components of GABA release.

Multiple components of synaptosomal [3H]-gamma-aminobutyric acid release resolved by a rapid superfusion system.

Release of [3H]-gamma-aminobutyric acid ([3H]GABA) from rat brain synaptosomes was studied with 60-ms time resolution, using a novel rapid superfusion method. Synaptosomes were prelabeled with [3H]GABA via an associated GABA uptake system. KCl depolarization stimulated at least three distinct components of GABA release: (1) a phasic Ca-dependent component, which develops rapidly and decays with a time constant of at most 60 ms; (2) a tonic Ca-dependent component that persists after KCl depolarization is ended; (3) a Ca-independent component. The three components of GABA release are pharmacologically distinct. The phasic component was selectively blocked by 50 microM Cd2+, while the tonic component was selectively blocked by 100 microM Ni2+. The Ca-independent component was selectively blocked by nipecotic acid (IC50 = 21 microM), a known inhibitor of Na+-dependent GABA uptake. The time course and amplitude of Ca-dependent GABA release evoked by the Ca2+ ionophore A23187 were nearly identical with Ca-dependent release evoked by depolarization. This result indicates that Ca-dependent GABA release depends primarily on Ca2+ entry into the nerve terminal, and not depolarization, per se. The properties of the phasic component suggest that it is normally initiated by a voltage-sensitive Ca2+ channel that is functionally and pharmacologically distinct from those previously described. The Ca-independent component of GABA release is probably mediated by reversal of the Na-dependent, electrogenic GABA uptake system. The ability to identify multiple components of GABA release on a physiologically relevant time scale may afford a more precise definition of the mechanism of action of drugs thought to affect neurotransmission in the brain.

Biochemical and immunological evidence for a calcium pump in chromaffin granules.

ATP stimulated the accumulation of 45Ca2+ by chromaffin granule ghosts which contained 5 mM oxalate to trap transported calcium within the lumen. Inasmuch as the ATP-dependent 45Ca2+ transport was resistant to 25 mM ammonium acetate as well as the proton ionophore, carbonylcyanide-m-chlorophenylhydrazone, the chromaffin granule proton translocating ATPase does not provide the energy for this process. Instead, we suggest that chromaffin granules contain a calcium translocating ATPase which catalyzes the 45Ca2+ uptake directly. The observation that chromaffin granules bind to a monoclonal antibody raised against a calcium pump from bovine brain supports this hypothesis.

Two functionally distinct forms of guanosine cyclic 3',5'-phosphate stimulated cation channels in a bovine rod photoreceptor disk preparation.

Cyclic nucleotide stimulated efflux of 22Na+ and 45Ca2+ from a purified bovine rod outer segment disk preparation was measured on the 25-100-ms time scale by a novel rapid superfusion method. Activation of cation efflux by 8-bromoguanosine cyclic 3',5'-phosphate (8-Br-cGMP) was maximal within 25 ms. Over a wide range of concentrations of 8-Br-cGMP, the kinetics of termination of efflux precisely conformed to the sum of two exponential decay processes: a rapid phase (decay constant of 200 ms) and a slower phase (decay constant of 1.6 s). The kinetics of the biphasic decay of efflux cannot be explained by depletion of a pool of releasable 22Na but appear to reflect an intrinsic process for inactivation of the channels. 8-Br-cGMP-stimulated release of actively accumulated 45Ca exhibited identical biphasic decay kinetics. The maximum rate of Ca release [5 nmol.(mg of disk protein)-1.min-1] may be sufficient to produce a 1 microM change in local cytoplasmic [Ca] within 20 ms. The Ca:Na selectivity ratio is approximately 0.5:1 for both decay phases. 8-Br-cGMP demonstrated a lower potency (EC50 of 8.4 microM vs 2.8 microM) but a higher degree of cooperativity in its activation of the rapid vs the slower decay phase of 22Na efflux. The slower phase of decay was selectively inhibited by 25 microM l-cis-diltiazem, a relatively weak inhibitor of the rapid decay phase. Sodium ion (5-10 mM) selectively inhibited the rapid decay phase of 8-Br-cGMP-stimulated 45Ca release. These two kinetically and pharmacologically distinct phases of decay are hypothesized to represent two functionally distinct forms of cGMP-stimulated cation channels.

Guanosine 3',5'-cyclic monophosphate stimulates release of actively accumulated calcium in purified disks from rod outer segments of bovine retina.

Parallel lines of evidence have suggested that light initiates changes in both cGMP metabolism and calcium levels in rod outer segments (ROS). We report that cGMP stimulates release of a pool of Ca2+ actively accumulated within purified ROS disks. Disks were purified and actively loaded with 45Ca2+ by an associated ATP-dependent calcium uptake activity as previously described [Puckett, K.L., Aronson, E.T., & Goldin, S.M. (1985) Biochemistry 24, 390-400]. Spikes of 45Ca2+ released from disks were observed in a rapid superfusion system. The Ca2+ release was specifically stimulated by physiological levels of cGMP (Kapp approximately 20 microM; Hill coefficient = 1.7). 8-Bromo-cGMP could also activate the release mechanism, but cAMP was ineffective. At cGMP levels of greater than or equal to 100 microM, approximately 20% of the loaded Ca2+ was released. The Ca2+ release rate at saturating cGMP levels reached a maximum within the 10-s time resolution of the assay system. In contrast to other recent reports of cGMP activation of ROS ion conductances, the majority of the release activity terminated in a spontaneous manner, suggestive of an intrinsic inactivation process. The amount of Ca2+ released and the release kinetics were similar to the presence or absence of an unbleached pool of rhodopsin. Cyclic nucleotides did not stimulate release from disks passively equilibrated with 45Ca2+, i.e., in the absence of ATP but otherwise under identical conditions. Preincubation of the disks with cGMP also reduced the level of ATP-dependent Ca2+ uptake (approximately 30%); this apparent inhibition may be due to activation of the release mechanism, rather than direct modulation of the uptake activity.(ABSTRACT TRUNCATED AT 250 WORDS)

Calcium channels in rat brain synaptosomes: identification and pharmacological characterization. High affinity blockade by organic Ca2+ channel blockers.

Rat brain synaptosomes are shown to contain functional voltage-sensitive Ca2+ channels that are inhibited by organic Ca2+ channel blockers. Depolarization of synaptosomes with high K+ stimulates uptake of 45Ca2+ which is biphasic in its time course. Replacement of external Na+ with choline eliminates the slower phase of depolarization-stimulated Ca2+ uptake, leaving only a rapid uptake process which terminates within 1 sec. This rapid, tetrodotoxin-insensitive Ca2+ uptake can be inactivated by prior depolarization of the synaptosomes. Depolarization has no effect on the rate of synatptosomal 22Na+ efflux. These results are interpreted as ruling out Na+/Ca2+ exchange as a mediator of the rapid phase of depolarization-stimulated Ca2+ uptake. A portion (30 to 50%) of the rapid phase of depolarization-stimulated Ca2+ uptake is inhibited by nitrendipine, as is depolarization-stimulated [3H]norepinephrine release from synaptosomes. In external Na+, the inhibition constant (Kapp) for nitrendipine inhibition of Ca2+ uptake is 56 nM. The potency of nitrendipine is increased in the absence of external Na+ (Kapp = 1.7 nM), such that inhibition correlates more closely with the equilibrium dissociation constant for [3H] nitrendipine binding to synaptosomes (Kd = 0.35 nM). Other organic channel blockers (nifedipine, verapamil, D600, and dilitiazem) inhibit the rapid Ca2+ uptake. The potencies of all Ca2+ channel blockers tested by us are in reasonable agreement with their potencies, observed in other laboratories, as blockers of Ca2+ channels in smooth and cardiac muscle. These data demonstrate the existence of active voltage-sensitive Ca2+ channels in synaptosomes.(ABSTRACT TRUNCATED AT 250 WORDS)

ATP-dependent calcium uptake activity associated with a disk membrane fraction isolated from bovine retinal rod outer segments.

Ca2+ sequestration and release from disks of rod outer segments may play a critical role in visual transduction. An ATP-dependent Ca2+ uptake activity has been identified in association with purified disks of bovine rod outer segments. A crude preparation of osmotically active disks was obtained from rod outer segments by hypoosmotic shock and subsequent flotation on a 5% Ficoll 400 solution. These "crude" disks were further purified by separation into two distinct components by centrifugation in a linear Ficoll gradient. Disks comprised the major component; at least 60% of the protein was rhodopsin. This fraction also contained a Ca2+ uptake activity stimulated approximately 4-fold by ATP. The initial rate was approximately 0.21 nmol of Ca2+ (mg of protein)-1 min-1. Most of the ATP-dependent accumulation of 45Ca2+ was released by the calcium ionophore A23187. The uptake activity was sensitive to vanadate (Ki approximately 20 microM) and insensitive to the mitochondrial Ca2+ uptake inhibitor ruthenium red (10 microM). The ATP-dependent Ca2+ uptake exhibited Michaelis-Menten activation kinetics with respect to [Ca2+] (Km approximately 6 microM). The osmotic properties of the sealed disk membranes were exploited to determine whether the association of Ca2+ transport activity with the disks was merely coincidental. The sedimentation properties of these disks, upon centrifugation on a second Ficoll linear density gradient, varied with the osmolarity of the gradient solution. In several separate gradient solutions of differing osmotic and ionic strengths, the Ca2+ uptake activity always comigrated with the disks. These results indicate that the ATP-dependent Ca2+ uptake activity was physically associated with sealed native disk membranes. The characteristics of the Ca2+ uptake activity suggest that it may play a major role in the regulation of cytosolic Ca2+ levels in rod cells in vivo.

Partial purification and functional identification of a calmodulin-activated, adenosine 5'-triphosphate-dependent calcium pump from synaptic plasma membranes.

Synaptic plasma membranes isolated from rat brain contain a calmodulin-activated Ca2+ pump. It has been purified 80- to 160-fold by solubilization with Triton X-100 and affinity chromatography on a calmodulin-Sepharose 4B column. After reconstitution into phospholipid vesicles, the affinity-purified pump efficiently catalyzed ATP dependent Ca2+ transport, which was activated 7- to 9-fold by calmodulin. The major protein component of the affinity-purified preparation had a Mr = 140,000; it was virtually the only band visualized on a Coomassie blue-stained SDS polyacrylamide gel. It has been identified as the Ca2+ pump by two functional criteria. First, it was phosphorylated by [gamma-32P]ATP in a Ca2+-dependent manner; the phosphorylated protein had the chemical reactivity of an acyl phosphate, characteristic of the phosphorylated intermediates of ion-transporting ATPases. Second, the protein was enriched by transport-specific fractionation, a density gradient procedure which uses the transport properties of the reconstituted Ca2+ pump as a physical tool for its purification. By analogy with calmodulin-activated (Ca2+ + Mg2+) ATPases of other cell types, and because of its presence in a synaptic plasma membrane fraction, we hypothesize that the calmodulin-activated Ca2+ pump functions in vivo to extrude Ca2+ from nerve terminals.

Purification and immunological characterization of a calcium pump from bovine brain synaptosomal vesicles.

In previous work ( Papazian , D., H. Rahamimoff , and S. M. Goldin (1979) Proc. Natl. Acad. Sci. U. S. A. 76: 3708-3712), an ATP-dependent calcium transport activity derived from rat brain synaptosomes was reconstituted into artificial lipid vesicles and substantially purified by transport-specific fractionation. When this procedure was applied to bovine brain synaptosomes, the approximately 70-fold purified, reconstituted Ca2+ uptake system contained two major polypeptides of Mr = 230,000 (" C230 ") and 94,000 (" C94 ") as observed on sodium dodecyl sulfate (SDS) gels. Evidence is presented here that these polypeptides are immunologically related to one another and that the synaptosomal Ca2+ pump is immunologically distinct from Ca2+ pumps in non-neuronal cells. Antisera and monoclonal antibodies to the purified, reconstituted protein did not significantly cross-react with the Ca2+ pumps or any other components of bovine sarcoplasmic reticulum or erythrocytes. However, these antibodies did cross-react with a component of bovine brain axolemma-enriched membranes. A monoclonal antibody was produced that immunoprecipitated the Ca2+ transport activity, both in native, synaptosomal vesicles and in liposomes containing the reconstituted transport system. This antibody bound C230 more prominently than C94 on Western blots of SDS gels. An antiserum raised against C94 alone, obtained by elution from SDS gels, was also found to bind most prominently to C230 on Western blots. These results suggest that this synaptosomal Ca2+ pump is specific to nerve tissue and that C94 and C230 are structurally homologous components of this transport activity.

Molecular characterization, reconstitution, and "transport-specific fractionation" of the saxitoxin binding protein/Na+ gate of mammalian brain.

The saxitoxin (STX) binding protein has been solubilized by sodium cholate, both from axolemma and from synaptosomal membranes of mammalian brain. On the basis of agarose gel filtration and sedimentation properties in H2O and 2H2O, the solubilized particle has the following molecular properties: Stokes radius, 120 A; partial specific volume, 0.85 cm3/g; mass, 1,020,000 daltons; frictional ratio f/fo, 1.6. The solubilized STX binding protein was incorporated into unilamellar (approximately 550-A) artificial phosphatidylcholine vesicles. Based on the expectation that the STX binding protein contains functional monovalent cation gating activity ("action potential Na+ gate") that can be activated by veratridine and inhibited by tetrodotoxin, a strategy was devised for partial purification of the reconstituted sodium gate/STX binding protein by "transport-specific fractionation." When the entire vesicle population was preloaded with 0.4 M cesium ion, addition of veratridine allowed Cs+ efflux from specifically those vesicles containing the ion gate; the concomitant reduction in intravesicular density permitted the ion gate/STX binding protein to be fractionated on density gradients. These observations demonstrate functional reconstitution and partial (30- to 50-fold) purification of the STX binding protein/Na+ gate of mammalian brain.