J Physiol Volume 539, Number 2, 332-, March 1, 2002 DOI: 10.1113/jphysiol.2002.017160
Journal of Physiology (2002), 539.2, p. 332
© Copyright 2002 The Physiological Society
DOI: 10.1113/jphysiol.2002.017160
Synaptophysins: vesicular cation channels?
Oussama El Far and Heinrich Betz
Department of Neurochemistry, Max-Planck-Institute for Brain Research, Deutschordenstrasse 46, 60528 Frankfurt, Germany
During the past 15 years, the presence of ion channels in synaptic vesicle and neurosecretory granule membrane preparations has been demonstrated repeatedly by different investigators. The physiological roles and molecular identity of these ion channels remained, however, enigmatic. The report of Yin et al. in this issue of The Journal of Physiology strengthens previous suggestions that at least one of these channels could be associated with, related to, or identical to, a major membrane protein of synaptic vesicles, synaptophysin.
Synaptophysin was originally identified as an abundant and highly immunogenic membrane protein of small synaptic vesicles that is also found in dense-core chromaffin and neurosecretory granules. Synaptophysin and its homologues synaptoporin (or synaptophysin II) and pantophysin share a common transmembrane organization, with four membrane-spanning regions and cytoplasmic amino and carboxy termini. Upon incorporation into lipid bilayers, purified synaptophysin has been found to form an ion channel. This together with its oligomeric nature and structural similarities to gap junctions proteins led Thomas et al. (1988) to suggest that synaptophysin may be a component of the fusion pore that forms during neurotransmitter release. Moreover, Alder et al. (1992) have shown that antisense oligonucleotides complementary to the synaptophysin mRNA reduce Ca2+-dependent glutamate secretion from Xenopus oocytes induced by injection of total brain mRNA. Similarly, microinjection of synaptophysin antibody into motor neurons blocked neuromuscular transmission. These data are consistent with synaptophysin being essential for neurotransmitter secretion. However, genetic approaches to identify the function of synaptophysin have not been successful; mutant mice lacking synaptophysin show a normal phenotype (Eshkind & Leube, 1995; McMahon et al. 1996). This may reflect compensation by synaptoporin or other synaptophysin family members. Indeed, mice doubly deficient in synaptophysin and the structurally related vesicle protein synaptogyrin display defects in synaptic plasticity (Janz et al. 1999).
In addition to its putative function as an ion channel, synaptophysin has been proposed to play a structural role in vesicle formation. Based on its high capacity to bind cholesterol, synaptophysin has recently been implicated in the generation of membrane curvature during synaptic vesicle biogenesis (Thiele et al. 2000). Also, synaptophysin is known to interact tightly with other proteins of the synaptic vesicle membrane, i.e. synaptobrevin (Calakos & Scheller, 1994) and the 38 kDa subunit of the vacuolar H+-ATPase (Siebert et al. 1994). These interactions are thought to regulate exocytotic membrane fusion at the level of the SNARE complex or fusion pore formation. The latter idea is supported by recent studies on yeast vacuole fusion which implicate the vacuolar ATPase directly in membrane fusion (Peters et al. 2001). Finally, Daly et al. (2000) have shown that microinjection of the unique cytoplasmic C-terminal tail region of synaptophysin into the squid giant terminal produces an activity-dependent reduction in transmitter release, which is accompanied by the accumulation of clathrin-coated vesicles. Synaptophysin therefore has been proposed to play a role in synaptic vesicle recycling. This is consistent with an as yet uncharacterized Ca2+-dependent interaction of synaptophysin with dynamin I (Daly et al. 2000), a GTPase known to be essential for the membrane fission step of clathrin-mediated retrieval of synaptic vesicles from the presynaptic plasma membrane.
In their study, Yin et al. present immunological evidence that one of the ion channels detected in bilayers fused with purified neurosecretory granule membranes from rat neurohypophysis may be synaptophysin. Previous work by the same authors had shown that such membrane preparations contain both a small and a large cation channel. The open probability of the large but not the small channel is now found to be strongly reduced in the presence of the monoclonal antibody SY-38 that recognizes the most degenerate out of ten pentapeptide repeats within synaptophysin's C-terminal tail region (Knaus & Betz, 1990). Notably, the same antibody has been previously demonstrated to block channels formed from reconstituted synaptophysin preparations (Thomas et al. 1988). Inhibition is specific for this particular synaptopyhsin antibody, since another antibody that also recognizes synaptophysin, p38, had no effect. Importantly, the same antibody specificity was recovered in arginine vasopressin release assays with permeabilized nerve terminals purified from the rat neurohypophysis. Again, only SY-38, but not the p38, antibody potently inhibited Ca2+-triggered peptide secretion. These data indicate that synaptophysin or a closely related molecule participates in the formation of a cation channel that is essential for neurosecretion. Whether this channel is formed by synaptophysin alone or involves associated proteins like subunits of the H+-ATPase, which bind synaptophysin and have been implicated in membrane fusion, remains to be established.
A particularly intriguing feature of the large neurosecretory granule channel is its regulation by micromolar concentrations of Ca2+. In their report, Yin et al. report that elevating the Ca2+ concentration on the cis, i.e. putative cytoplasmic side, of the bilayer increases both the open probability and conductance of the channel. Notably, higher concentrations of Ca2+ (10 µM) reverse the inhibitory effects of the SY-38 antibody on both reconstituted channel activity and arginine vasopressin secretion from permeabilized neurohypophysial terminals. This may indicate that Ca2+ can directly or indirectly change the conformation of synaptophysin's tail region such that the antigenic epitope is no longer accessible. However, since synaptophysin binds Ca2+ only with rather low affinity, the Ca2+ regulation seen is likely to be provided by an associated Ca2+ binding protein.
In conclusion, the study of Yin et al. revitalizes old observations that synaptophysin could form an ion channel, and provides indirect evidence for the involvement of such channels in arginine vasopressin secretion from neurohypophysial terminals.
 |
REFERENCES |
| ALDER, J. et al. (1992). Science 257, 657-661 |
|
| CALAKOS, N. & SCHELLER, R. H. (1994). Journal of Biological Chemistry 269, 24534-24537 |
[Abstract] |
| DALY, C. et al. (2000). Proceedings of the National Academy of Sciences of the USA 97, 6120-6125 |
[Abstract/Full Text] |
| ESHKIND, L. G. & LEUBE, R. E. (1995). Cell and Tissue Research 282, 423-433 |
[Medline] |
| JANZ, R. et al. (1999). Neuron 24, 687-700 |
[Medline] |
| KNAUS, P. & BETZ, H. (1990). FEBS Letters 261, 358-360 |
[Medline] |
| MCMAHON, H. T. et al. (1996). Proceedings of the National Academy of Sciences of the USA 93, 4760-4764 |
[Abstract] |
| PETERS, C. et al. (2001). Nature 409, 581-588 |
[Medline] |
| SIEBERT, A. et al. (1994). Journal of Biological Chemistry 269, 28329-28334 |
[Abstract] |
| THIELE, C. et al. (2000). Nature Cell Biology 2, 42-49 |
[Medline] |
| THOMAS, L. et al. (1988). Science 242, 1050-1053 |
|
| YIN, Y. et al. (2002). Journal of Physiology 539, 409-418 |
|
