Studies of cell-specific gene expression in central nervous system neurons have been limited by the paucity of homologous neuronal cell lines that adequately represent the phenotypes of specific central neurons. Hence, for most of these neuronal phenotypes relevant information has been mainly obtained by studying the specific neurons in transgenic mice. Transgenic mouse studies on oxytocin (OT) and vasopressin (VP) gene expression in magnocellular neurons (MNCs) of the hypothalamo-neurohypophysial system (HNS) in our own and other laboratories have led us to propose the intergenic region (IGR) hypothesis, which states that the 3?flanking region of the VP gene (the IGR) separating the OT- and VP-genes, contains critical enhancer sites for cell-specific expression of these neuropeptide genes. We tested this hypothesis using transgenic mice containing mouse genomic DNA-derived constructs linked to chloramphenicol acetyltransferase (CAT) reporters. VP gene' expression was studied using constructs containing 3.8 kbp of the 5' flanking region and all the exons and introns in the mouse VP gene which was fused at the end of exon 3 to a CAT reporter followed by a 2.1kb IGR fragment. A similar construct for the oxytocin CAT transgene was used which contained the full-length mouse IGR(3.6 kbp). Both transgenic constructs produced cell-specific expression of the CAT-reporter in the appropriate OT or VP magnocellular neurons. Electron microscopic immunocytochemistry showed that the CAT fusion proteins produced from the OT- and VP-transgenes were efficiently trafficked through the regulated secretory pathways in their respective magnocellular neurons, packaged into large dense core vesicles (LDCVs) and transported to nerve terminals in the posterior pituitary. Further tests of the IGR hypothesis are now being done using a novel high throughput method recently developed in our laboratory which uses biolistic transfections of identified neurons in organotypic hypothalamic cultures. We recently demonstrated that the extensive apoptosis of magnocellular neurons (MCNs) in the hypothalamus that occurs in organotypic cultures due to axotomy could be rescued by CNTF. More recently, we demonstrated that the anti-apoptotic agents, Bcl-XL and z-VAD, could also rescue VP and OT MCNs in vitro, consistent with the view that the VP and OT MCN cell death is occurring by apoptotic mechanisms. The use of CNTF to rescue the MCNs has for the first time allowed us to perform long-term molecular and physiological studies of both MCN phenotypes in vitro. Biolistic transfections of hypothalamic organotypic slice explants were done using plasmid constructs containing an EGFP reporter to study VP gene expression in the HNS. Using this high throughput strategy we found that reducing the 5?UTR from 3.5kbp to 288bp and the IGR from 3.6 to 2.1 kbp in the mouse VP gene construct did not alter the efficacy of its expression in the hypothalamic slices. All subsequent constructs were based on this 288 bp 5? UTR VP gene construct with changes made only to the IGR region. Our first construct used to test the IGR hypothesis eliminated the IGR completely, and gave negative results in transfection assays, consistent with the hypothesis. The 2.1 kb IGR was then divided into 5 segments of 176, 374, 298, 446, and 834 bp based on our bioinformatic anayses of conserved sequences found in the human, rat and mouse IGRs. Experiments using the 5 constructs containing the above IGR segments are now in progress. In addition to the above studies on the MCNs, we have also studied the expression of the endogenous VP gene in parvocellular neurons in the hypothalamus. VP is synthesized in and secreted by the suprachiasmatic nucleus (SCN) in a circadian pattern. We studied transcription of the AVP gene in organotypic cultures of rat SCN by using intronic in situ hybridization. AVP gene transcription in the cultured SCN maintained a daily rhythm with a peak in the daytime. Inhibition of spontaneous activity by the sodium channel blocker, tetrodotoxin (TTX), dramatically decreased AVP heteronuclear RNA levels and suppressed rhythmicity. In addition, the MAP kinase pathway inhibitor, PD98059, profoundly decreased VP transcription and abolished its daily rhythm. Hence, neural activity and a functional MAP kinase signaling pathway appears to be critical for AVP gene expression in the SCN. In another study on VP gene transcription in the rat paraventricular nucleus (PVN) in organotypic cultures using intronic in situ hybridization we could demonstrated that cAMP directly activates AVP transcription in parvocellular neurons of the PVN. Finally, the cell-specific gene expression data has allowed us to target the gene expression of specific molecules to LDCVs in the MCNs in the HNS in vivo and to directly visualize neurosecretory processes in the MCN nerve endings. We showed that the AI-03 transgene which contains enhanced green fluorescent protein (EGFP) fused to the end of the neurophysin at the C-terminus of the OT pre-prohormone, is expressed selectively in OT-magnocellular neurons,and is trafficked to secretory granules in transgenic mice. The EGFP-containing secretory granules are then transported to OT-neurosecretory terminals in the neurohypophysis, where the EGFP fluorescence could be seen to undergo depolarization-induced calcium-dependent secretion. We also studied the cell-specific expression of tyrosine hydroxylase in the CNS using a 9kbp upstream region of the gene, coupled to an EGFP reporter, both in vitro and in vivo. Three transgenic mouse lines which express EGFP in a cell-specific manner in catecholaminergic neurons in the CNS have been produced and are being backcrossed. Analysis of EGFP expression in these mice shows that dopamine neurons in the arcuate nucleus and norepinephrine neurons in the locus coeruleus show very robust expression of the GFP.
Showing the most recent 10 out of 33 publications
