Faculty Bibliography

This paper provides a brief overview of the diversity of voltage-gated Ca2+ channels and our recent work on neuronal Ca2+ channels with novel pharmacological and biophysical properties that distinguish them from L, N, P or T-type channels. The Ca2+ channel alpha 1 subunit known as alpha 1A or BI [Mori Y., Friedrich T., Kim M.-S., Mikami A., Nakai J., Ruth P., Bosse E., Hofmann F., Flockerzi V., Furuichi T., Mikoshiba K., Imoto K., Tanabe T. and Numa S. (1991) Nature 350, 398-402] is generally assumed to encode the P-type Ca2+ channel. However, we find that alpha 1A expressed in Xenopus oocytes differs from P-type channels in its kinetics of inactivation and its degree of sensitivity to block by the peptide toxins omega-Aga-IVA and omega-CTx-MVIIC [Sather W. A., Tanabe T., Zhang J.-F., Mori Y., Adams M. E. and Tsien R. W. (1993) Neuron 11, 291-303]. Thus, alpha 1A is capable of generating a Ca2+ channel with characteristics quite distinct from P-type channels. Doe-1, recently cloned from the forebrain of a marine ray, is another alpha 1 subunit which exemplifies a different branch of the Ca2+ channel family tree [Horne W. A., Ellinor P. T., Inman I., Zhou M., Tsien R. W. and Schwarz T. L. (1993) Proc. Natn. Acad. Sci. U.S.A. 90, 3787-3791]. When expressed in Xenopus oocytes, doe-1 forms a high voltage-activated (HVA) Ca2+ channel [Ellinor P. T., Zhang J.-F., Randall A. D., Zhou M., Schwarz T. L., Tsien R. W. and Horne W. (1993) Nature 363, 455-458]. It inactivates more rapidly than any previously expressed calcium channel and is not blocked by dihydropyridine antagonists or omega-Aga-IVA. Doe-1 current is reduced by omega-CTx-GVIA, but the inhibition is readily reversible and requires micromolar toxin, in contrast to this toxin's potent and irreversible block of N-type channels. Doe-1 shows considerable sensitivity to block by Ni2+ or Cd2+. We have identified components of Ca2+ channel current in rat cerebellar granule neurons with kinetic and pharmacological features similar to alpha 1A and doe-1 in oocytes [Randall A. D., Wendland B., Schweizer F., Miljanich G., Adams M. E. and Tsien R. W. (1993) Soc. Neurosci. Abstr. 19, 1478]. The doe-1-like component (R-type current) inactivates much more quickly than L, N or P-type channels, and also differs significantly in its pharmacology.(ABSTRACT TRUNCATED AT 400 WORDS)

We used the fluorescent membrane probe FM 1-43 to label recycling synaptic vesicles within the presynaptic boutons of dissociated hippocampal neurons in culture. Quantitative time-lapse fluorescence imaging was employed in combination with rapid superfusion techniques to study the dynamics of synaptic vesicles within single boutons. This approach enabled us to measure exocytosis and to analyze the kinetics of endocytosis and the preparation of endocytosed vesicles for re-release (repriming). Our measurements indicate that under sustained membrane depolarization, endocytosis persists much longer than exocytosis, with a t1/2 approximately 60 s (approximately 24 degrees C); once internalized, vesicles become reavailable for exocytosis in approximately 30 s. Furthermore, we have shown that endocytosis is not dependent on membrane potential and, unlike exocytosis, that it is independent of extracellular Ca2+

Transcripts for the class A Ca2+ channel alpha 1 subunit (also known as BI) are present at high levels in many parts of the mammalian CNS and are widely assumed to encode the P-type Ca2+ channel. To characterize the biophysical and pharmacological properties of alpha 1A channels, macroscopic and single-channel recordings were made in Xenopus oocytes injected with alpha 1A cRNA. alpha 1-specific properties were identified by making systematic comparisons with the more familiar class C alpha 1 subunit under the condition of a standard ancillary subunit (alpha 2/delta + beta) makeup. alpha 1A currents activate and inactivate more rapidly and display steeper voltage dependence of gating than alpha 1C currents. Unlike alpha 1C, alpha 1A channels are largely insensitive to dihydropyridines and FPL 64176, but respond to the cone snail peptide omega-CTx-MVIIC(SNX-230), a potent and fairly selective inhibitor. In comparison with P-type Ca2+ channels in rat cerebellar Purkinje cells, alpha 1A channels in oocytes are approximately 10(2)-fold less sensitive to omega-Aga-IVA and approximately 10-fold more sensitive to omega-CTx-MVIIC. alpha 1A channels are not inhibited by Bay K 8644 and inactivate much more rapidly than P-type Ca2+ channels. Thus, alpha 1A is capable of generating a Ca2+ channel phenotype quite different from P-type current

Diverse types of calcium channels in vertebrate neurons are important in linking electrical activity to transmitter release, gene expression and modulation of membrane excitability. Four classes of Ca2+ channels (T, N, L and P-type) have been distinguished on the basis of their electrophysiological and pharmacological properties. Most of the recently cloned Ca2+ channels fit within this functional classification. But one major branch of the Ca2+ channel gene family, including BII (ref. 15) and doe-1 (ref. 16), has not been functionally characterized. We report here the expression of doe-1 and show that it is a high-voltage-activated (HVA) Ca2+ channel that inactivates more rapidly than previously expressed calcium channels. Unlike L-type or P-type channels, doe-1 is not blocked by dihydropyridine antagonists or the peptide toxin omega-Aga-IVA, respectively. In contrast to a previously cloned N-type channel, doe-1 block by omega-CTx-GVIA requires micromolar toxin and is readily reversible. Unlike most HVA channels, doe-1 also shows unusual sensitivity to block by Ni2+. Thus, doe-1 is an HVA Ca2+ channel with novel functional properties. We have identified a Ca2+ channel current in rat cerebellar granule neurons that resembles doe-1 in many kinetic and pharmacological features

In many neurons, transmitter release from presynaptic terminals is triggered by Ca2+ entry via dihydropyridine-insensitive Ca2+ channels. We have looked for cDNAs for such channels in the nervous system of the marine ray Discopyge ommata. One cDNA (doe-2) is similar to dihydropyridine-sensitive L-type channels, and two cDNAs (doe-1 and doe-4) are similar to the subfamily of dihydropyridine-insensitive non-L-type channels. doe-4, which encodes a protein of 2326 aa, most closely resembles a previously cloned N-type channel. doe-1, which encodes a protein of 2223 aa, is a member of a separate branch of the non-L-type channels. Northern blot analysis reveals that doe-1 is abundant in the forebrain. doe-4 is more plentiful in the electric lobe and, therefore, may control neurotransmitter release in motor nerve terminals. These results show that the familial pattern of Ca(2+)-channel genes has been preserved from a stage in evolution before the divergence of higher and lower vertebrates > 400 million years ago. The cloning of these channels may be a useful starting point for elucidating the role of the Ca2+ channels in excitation-secretion coupling in nerve terminals

G protein-mediated inhibition of voltage-activated calcium channels by neurotransmitters has important consequences for the control of synaptic strength. Single-channel recordings of N-type calcium channels in frog sympathetic neurons reveal at least three distinct patterns of gating, designated low-Po, medium-Po, and high-Po modes according to their probability of being open (Po) at -10 millivolts. The high-Po mode is responsible for the bulk of divalent cation entry in the absence of neurotransmitter. Norepinephrine greatly decreased the prevalence of high-Po gating and increased the proportion of time a channel exhibited low-Po behavior or no activity at all, which thereby reduced the overall current. Directly observed patterns of transition between the various modes suggest that activated G protein alters the balance between modal behaviors that freely interconvert even in the absence of modulatory signaling

The effect of protein kinase C (PKC) stimulation on Ca2+ channels was studied in frog sympathetic neurons. 12,13-Phorbol dibutyrate (PDBu) consistently augmented Ca2+ channel currents in whole-cell recordings. This enhancement was blocked by staurosporine and PKC(19-31), but not produced by 4 alpha-phorbol 12,13-didecanoate, indicating that PDBu acts via PKC. Both N- and L-type currents, as isolated pharmacologically, were increased. PKC enhancement was independent of the extent of G protein activation, indicating that it was not caused by removal of tonic G protein inhibition. In unitary recordings PDBu produced dramatic increases in single N- and L-type channel activity by sharply decreasing closed time intervals between adjacent openings, but did not alter the unitary current size or mean open time. This up-modulation by PKC may constitute a positive feedback mechanism in the regulation of neuronal Ca2+ channel activity

In many neurons, N-type Ca2+ channels are a major Ca2+ entry pathway and strongly influence neurotransmitter release. We carried out cell-attached patch recordings (110 mM Ba2+ as charge carrier) to characterize the rapid opening and closing kinetics of N-type Ca2+ channel gating in frog sympathetic neurons. Single channels display at least three distinct patterns of gating, characterized as low-, medium-, and high-rho o modes on the basis of channel open probability (rho o) during depolarizing pulses to -10 mV. Spontaneous transitions from one mode to another are infrequent, with an exponential distribution of dwell times and mean sojourns of approximately 10 sec in each mode. Thus, a channel typically undergoes hundreds or thousands of open-closed transitions in one mode before switching to a different mode. Transitions between modes during a depolarization were occasionally detected, but were rare, as expected for infrequent modal switching. Within each mode, the activation kinetics were well described by a simple scheme (C2-C1-O), as previously reported for other types of Ca2+ channels. Rate constants are strikingly different from one mode to another, giving each mode its own characteristic kinetic signature. The gating behavior at -10 mV ranges from brief openings (approximately 0.3 msec) and long closures (10-20 msec) for low-rho o gating to long openings (3 msec) and brief closures (approximately 1 msec) for high-rho o gating. Intermediate values for mean open durations (approximately 1.5 msec) and mean closed durations (approximately 3 msec) were found for medium-rho o gating. In addition to being kinetically distinct, channel openings in the low-rho o mode often exhibit a unitary current approximately 0.2 pA larger than in the medium- or high-rho o mode. Each mode is characterized by its own voltage dependence: activation occurs at relatively negative potentials and is most steeply voltage dependent in the high-rho o mode, while activation requires very strong depolarizations and is weakly voltage dependent in the low-rho o mode. The proportion of time spent in the individual modes varies greatly from one patch to another, suggesting that modal gating may be subject to cellular control

Sympathetic neurons display robust [Ca2+]i oscillations in response to caffeine and mild depolarization. Oscillations occur at constant membrane potential, ruling out voltage-dependent changes in plasma membrane conductance. They are terminated by ryanodine, implicating Ca(2+)-induced Ca2+ release. Ca2+ entry is necessary for sustained oscillatory activity, but its importance varies within the oscillatory cycle: the slow interspike rise in [Ca2+]i requires Ca2+ entry, but the rapid upstroke does not, indicating that it reflects internal Ca2+ release. Sudden alterations in [Ca2+]o, [K+]o, or [caffeine]o produce immediate changes in d[Ca2+]i/dt and provide information about the relative rates of surface membrane Ca2+ transport as well as uptake and release by internal stores. Based on our results, [Ca2+]i oscillations can be explained in terms of coordinated changes in Ca2+ fluxes across surface and store membranes

Glutamate application at synapses between hippocampal neurons in culture produces long-term potentiation of the frequency of spontaneous miniature synaptic currents, together with long-term potentiation of evoked synaptic currents. The mini frequency potentiation is initiated postsynaptically and requires activity of NMDA receptors. Although the frequency of unitary quantal responses increases strongly, their amplitude remains little changed with potentiation. Tests of postsynaptic responsiveness rule out recruitment of latent glutamate receptor clusters. Thus, postsynaptic induction can lead to enhancement of presynaptic transmitter release. The sustained potentiation of mini frequency is expressed even in the absence of Ca2+ entry into presynaptic terminals