The mechanism of volatile general anesthetics (VAs) is unclear. Biochemical and electrophysiological data suggest that VAs, such as isoflurane and halothane, may have both presynaptic and postsynaptic effects in vertebrate neurons. In C. elegans, we have shown that a specific partial reduction-of-function mutation in
unc-64, which encodes a homolog of syntaxin 1A, an essential component of the presynaptic transmitter release machinery, is completely resistant to the behavioral defects and to the reduction in neuromuscular synaptic transmission produced by VAs. This resistance is allele specific seen only in the splice donor mutant
unc-64(
md130), semidominant, rescuable with an
unc-64(+) array, and cannot be explained simply by a reduction in neurotransmitter release since other similarly severe mutations in
unc-64 and in other genes encoding release machinery do not produce VA resistance. Our working hypothesis to explain these results is that the
unc-64(
md130) mutant product, a truncated syntaxin without its normal transmembrane domain, interacts with another protein and thereby blocks VAs binding to or action against the other protein. A corollary of this hypothesis is that overexpression of this putative other protein would suppress the dominant negative effects
unc-64(
md130) has against VAs. Utilizing integrated arrays that result in over-expression of UNC-18 as assessed by Western blots (kindly provided by Mike Nonet and Erik Jorgensen), we tested whether UNC-18 could be the hypothesized syntaxin-interacting protein mediating VA action. Indeed, UNC-18 overexpression strongly suppresses the VA resistance produced by
unc-64(
md130). However, it has no effect on the Unc phenotype of
md130 and only weakly suppresses the aldicarb resistance of
md130. Thus, UNC-18 overexpression has little effect on improving the function of the small amount of wild type syntaxin produced in
md130, and this cannot explain the marked suppression of its VA resistance given that
md130/+ heterozygotes are VA resistant but non-Unc and normally sensitive to aldicarb. UNC-18 overexpression in the absence of the
md130 mutation does not alter VA sensitivity. We conclude that the antagonism of VA action produced by a truncated syntaxin is dependent on the levels of UNC-18. One attractive explanation for these results is that VAs target UNC-18, and VA affinity or efficacy is antagonized by interaction of the
md130 syntaxin product with UNC-18.