5.1. Presynaptic Localization of SNAP-25
SNAP-25 is one of the three core presynaptic proteins that combine to form the SNARE complex. Structural members of the SNAP-25 protein family contribute two of the four alpha helices that compose the SNARE complex, which is necessary for Ca2+-triggered exocytic transmitter release. The first characterized SNAP-25 proteins were the two isoforms SNAP-25A and SNAP-25B, which represent highly homologous proteins that differ by only nine amino acids44. SNAP-23 represents another member of this protein family, which shares an approximately 60% identity with SNAP-25, although the cellular distribution of the former protein is substantially more ubiquitous than the latter45. There is evidence that SNAP-23 may also be involved in the positive regulation of neurotransmitter release46, unlike SNAP-29 - another member of the protein family - which appears to act as a negative modulator of neurotransmitter release, possibly by slowing recycling of the SNARE-based fusion machinery and synaptic vesicle turnover47.
Current theories of vesicular membrane fusion emphasize the role of SNARE proteins. According to these models, syntaxin 1, synaptobrevin 2, and SNAP-25 regulate neuroexocytosis by forming a complex that forces vesicle and plasma membranes together, allowing the transmitter contained in the synaptic vesicle to enter the synaptic cleft (for review, see ref. 48). The SNARE proteins form a highly stable complex in vitro from helices in their cytoplasmic domains, with the formation of a 4-helix bundle - with 1 helix donated each by syntaxin 1 and synaptobrevin 2, and 2 helices donated by SNAP-25. The SNARE proteins are also specific substrates for the proteolytic action of the Clostridial neurotoxins, which inhibit neurotransmitter release.
Data from in situ and immunohistochemical studies indicate that SNAP-25 displays a regional pattern of expression in the brain that changes developmentally. For example, some regions, such as the olfactory bulbs, exhibit high levels of SNAP-25 mRNA expression early during development, which decline to negligible levels in adulthood, whereas other brain regions, such as the neocortex, display developmental increases and persistent adult neuronal immunoreactivity49. The cellular distribution of SNAP-25 may also change during development, as in the striatum, where SNAP-25 in the caudate nucleus is initially concentrated in axons, but subsequently localized in presynaptic regions of these axons49. During development, SNAP-25 plays a key role in neurite elongation and axonal growth. Highest levels of SNAP-25 expression in adulthood are observed in neurons of the neocortex, hippocampus, piriform cortex, anterior thalamic nuclei, pontine nuclei, and granule cells of the cerebellum50.
The primary function of SNAP-25 in the brain, in adulthood, is the regulation of calcium-dependent exocytosis of neurotransmitter, although it should be noted that SNAP-25 also regulates exocytosis in non-neuronal systems, such as insulin secretion by pancreatic beta cells51, and histamine release by gastric enterochromaffin-like cells52. The exocytic activity of SNAP-25 is regulated by protein kinase-dependent phosphorylation at a number of different sites53. In addition to its role as part of the SNARE complex, SNAP-25 also displays a number of direct protein-protein interactions, including binding to voltage-gated K+ and Ca2+ channels54,55, whose functional implications are continuing to be determined. While SNAP-25 serves as one of the three core proteins of the SNARE complex, details regarding its physiological activity during exocytosis indicate a more complex role than may be initially predicted. Recent evidence demonstrates that in the large, glutamatergic calyx of Held presynaptic terminal from rats, loss of SNAP-25 produces a graded loss of calcium sensitive transmitter release, unlike loss of syntaxin or synaptobrevin, which generates an "all-or-nothing" block of release56. Thus, while SNAP-25 remains essential for most calcium-dependent neurotransmitter release, it is likely, based on the findings described above, that our understanding of this molecule will continue to evolve.
Given the ubiquitous nature of SNAP-25, the protein is presumably of functional importance for numerous brain processes. Loss of SNAP-25, through genetic engineering of SNAP-25 null mutant mice, results in embryonic lethality57. However, heterozygous SNAP-25 mutant mice are robust and fertile. Insights into the role of SNAP-25 in behavioral processes have been provided by the heterozygous Coloboma mutant mouse (for review, see ref. 58). This mutant, which exhibits a contiguous chromosomal deletion encompassing 1.1 to 2.2 cM that includes the SNAP-25 gene, displays discrete physiological and behavioral deficits. The more obvious of these include spontaneous locomotor hyperactivity and head bobbing, and heterozygous Coloboma mice also exhibit delays in some tests of complex motor skills, such as righting reflex and bar holding. Locomotor hyperactivity in these mice has been hypothesized to result from regional alterations in monoamine neurotransmission, including decreased depolarization-evoked release of dopamine and serotonin in the dorsal striatum, and increased norepinephrine release in the striatum and nucleus accumbens59. Hippocampal-related deficits in heterozygous Coloboma mice include decreased induction of theta rhythmic activity after tail pinch, and reduced LTP in the dentate gyrus following high-frequency stimulation of the perforant path.
Evidence of a role for SNAP-25 in neural plasticity has been obtained from additional studies of LTP in nonmutant rats. Levels of both isoforms of SNAP-25 mRNA were increased 2 h after the induction of LTP in granule cells of the dentate gyrus following high frequency stimulation of the perforant path in vivo60. Hippocampal LTP was also associated with a significant increase in levels of SNAP-25 phosphorylation at multiple sites on the protein61. More recently, SNAP-25 was detected as one gene differentially expressed in the hippocampus in a learning and memory study in rats62. This same research group demonstrated that antisense oligonucleotides for SNAP-25 that were infused into the CA1 region of the hippocampus impaired long-term contextual fear memory and spatial memory, as well as interfered with the LTP of synaptic transmission in the CA1 region.
Similar to findings regarding the Cxs, preclinical data linking SNAP-25 to psychiatric and neurological disorders are obtained from both pharmacological and in vivo studies. SNAP-25 is expressed at relatively high levels in the locus coeruleus, where it may play an important role in the regulation of norepinephrine release. Consistent with evidence that major depressive disorder may be due, in part, to overactivity of the locus coeruleus, we have shown previously that different modalities of antidepressant treatments in rats decrease SNAP-25 mRNA in this brain region63. More extensive investigation has measured the effects of antipsychotic drugs on levels of SNAP-25 in different brain regions in rodents. One study reported that there was no effect of chronic treatment with haloperidol on levels of SNAP-25 in multiple brain regions that are strongly innervated by dopamine neurons, including the prefrontal cortex, nucleus accumbens, striatum, substantia nigra, and ventral tegmental area64. Recently, we have assessed the effects of treatment for 21 days with the typical antipsychotic drugs haloperidol, chlorpromazine, and trifluoperazine on levels of SNAP-25 in the trisynaptic pathway of the hippocampus65. Levels of SNAP-25 were measured by quantitative immunohistochemistry. Our findings revealed that both haloperidol and chlorpromazine increased SNAP-25 throughout the hippocampus, with greatest effects for haloperidol-treated rats in the mossy fiber region, while chlorpromazine-treated rats exhibited largest increases of SNAP-25 in the mossy fiber and Schaffer collateral regions. However, these effects only remained significant for the haloperidol-treated rats following application of Bonferroni correction for multiple comparisons. These data, which are the first description of effects of treatment with antipsychotic drugs on levels of SNAP-25 within the hippocampus in rodents, suggest that SNAP-25 may be increased in the hippocampus following treatment with specific neuroleptics. This is particularly interesting, as schizophrenia has been associated with reduced levels of SNAP-25 in the hippocampus (see below).
Animal models have indicated a link between SNAP-25 and psychiatric disorders. We previously reported that levels of SNAP-25 were significantly greater in the amygdala of adult rats that were reared in isolation from weaning66. This experimental manipulation is known to induce a range of behavioral alterations, such as locomotor hyperactivity and deficits in PPI, that resemble homologous symptoms of psychotic disorders30. As isolation-reared rats also display increased anxiety, and the behavioral deficits in this paradigm have been associated with alterations in monoamine neurotransmission67, our findings may indicate a novel substrate for anxiety in this model. The behavioral deficits evident in the heterozygous Coloboma mutant mouse have been hypothesized to model various symptoms of psychiatric disorders. The hyperactivity associated with the mutant is decreased by treatment with low doses of J-amphetamine that are sufficient to induce increased activity in wild-type mice68. Thus, it has been hypothesized that the heterozygous Coloboma mutant mouse may represent a valid animal model of attention deficit hyperactivity disorder (ADHD). However, many of the behavioral sequelae in this SNAP-25 haploinsufficient mutant are also of relevance to schizophrenia, and it is of interest that certain symptoms of ADHD, including hyperactivity and aggressiveness, may be treated effectively by low doses of both haloperidol and chlorpromazine69 - the same drugs that increased levels of SNAP-25 in the hippocampus in our study.
Genetic studies have observed significant associations between SNAP-25 polymorphism haplotypes and ADHD70. These findings are consistent with the abnormal behavioral phenotype of the heterozygous Coloboma mutant mouse, and its differential response to low doses of j-amphetamine. Intriguingly, one group demonstrated that the association of SNAP-25 with ADHD is largely due to transmission of alleles from paternal chromosomes to affected probands, suggesting that this locus may be subject to genomic imprinting. By contrast, there has been no confirmed genetic link between SNAP-25 and schizophrenia, although it was recently noted that specific SNAP-25 polymorphisms were associated with a selective clinical response to antipsychotic drugs71.
However, a number of studies by independent research groups have reported decreased levels of SNAP-25 in schizophrenia. Reduced levels of SNAP-25 protein were noted in the cerebellum, anterior frontal cortex, inferior temporal cortex, and prefrontal association areas72-74, although CSF SNAP-25 is increased in schizophrenia75. We have previously used quantitative immunohistochemistry to measure SNAP-25 in the hippocampus. The multi-synaptic structure of the hippocampus, and its consequent cytoarchitectural complexity, allows for the measurement of subtle and regionalized changes. We have demonstrated that schizophrenia is associated with significant reductions in SNAP-25 immunoreactivity within the perforant path termination zones in the molecular layers of CA1 and CA2, as well as the outer molecular layer of the dentate gyrus, although no significant changes were observed in the presubiculum or CA3 region76. Similar, regionalized changes were observed by Fatemi and colleagues77 who noted general reductions in hippocampal SNAP-25 immunoreactivity and significant decreases within the stratum granulosum. Additionally, reduced levels of SNAP-25 have also been observed in schizophrenia when entire hippocampal homogenates were quantified using immunoblotting78.
5.5. Association Between Cognitive Function and Expression of SNAP-25 and Complexins in Schizophrenia
Previous work from our group examined presynaptic proteins in the hippocampus in schizophrenia, and we reported significantly lower levels of SNAP-25, Cx I, and Cx II in schizophrenia22,76. In the same series of cases compared with controls, we observed no significant differences in synaptophysin, a ubiquitously expressed presynaptic protein. Initial examination of the relationship between the Cxs and cognitive performance (in terms of memory, orientation, and judgment) revealed a number of significant associations. New analyses suggest there may be significant associations between SNAP-25 and cognitive function in these cases (Figure 26.2; Colorplate 12). Of note, there appeared to be some degree of specificity to the findings, for both individual proteins, and for individual hippocampal subfields.
Was this article helpful?