Harold Gainer - US grants

An important issue that we continue to address in the laboratory is how to quantitatively measure gene expression in the CNS. Previous studies of gene expression in the HNS have been performed by in situ hybridization histochemistry (ISHH) using exon-specific probes, and measured the steady-state levels of mRNA, which is determined by both gene transcription and mRNA degradation processes. In contrast, measurements using intron-specific probes measure pre-mRNA or heteronuclear RNA (hnRNA) levels in the neuron which, because of the rapid turnover of the primary transcript and intermediate forms of RNA in the cell nucleus, is believed to primarily reflect the transcription rate of the gene. Since an effective intronic probe for VP hnRNA had previously been developed, we developed an effective intron-specific OT probe de novo, and then used both these intronic probes, together with other well established exonic OT and VP probes to reevaluate OT and VP gene expression in the hypothalamus under three classical physiological conditions, acute and chronic osmotic stimulation and lactation. We found that while there was the expected large increase in VP hnRNa after acute salt loading, there was no change in OT hnRNA, indicating that acute hyperosmotic stimuli produce increased VP but not OT gene transcription which was surprising. Since both neuropeptides are robustly and equivalently secreted from the neurohypophysis following acute salt-loading it had always been assumed that the gene expression responses of the OT and VP MCN phenotypes would be equivalent. It has always been believed that OT and VP gene expression and secretion are closely coupled. We then extended these observations over a wide range of times and found that VP hnRNA levels in the SON increased to near maximal levels after 2 hours following the NaCl injection (p<0.05), and reached maximal levels by 24 hours (p<0.001), which was sustained thereafter. The VP gene transcriptional activity in the SON is as rapid and sensitive to increases of plasma osmolality as is VP secretion. In contrast, the salt-loading stimulus did not produce statistically significant changes in OT hnRNA levels after 2, 24 or 48 hours but did increase the OT hnRNA to a maximal level by 72 hours (p<0.05), which was maintained for 120 and 168 hours (p<0.001). Since both OT and VP are secreted equivalently under these conditions, one possible interpretation is that OT gene expression is not closely correlated with secretion as is VP gene expression. If this is the case, this would further suggest that there are significant differences in excitation-transcription coupling mechanisms that regulate the OT and VP genes in the rat magnocellular neurons in the SON. [unreadable] [unreadable] Given the above findings that the OT- and VP MCNs gene expression responses to acute osmotic stimuli appear to be dramatically different, we sought to determine whether direct application of the presumed neurotransmitter signals for secretion on to the SONs would also differentiate between the OT and VP MCNs. For this purpose, we used a novel experimental paradigm in which ALZET mini osmotic pumps attached to pre-positioned cannulae position over each SON in male rats were used to infuse control (PBS only) solutions over the left SON and an excitatory cocktail consisting of of a mixture of NMDA, AMPA and bicuculline (NAB) were infused over the right SON. In this way, each animals left SON serves as a control for the experimental right SON, and the hnRNA measurements on the experimental side are expressed as a percentage of the values measured on the control side. We found that there was an increase of VP hnRNA in the NAB-treated SON (right SON) as compared to the PBS-control infused SON, indicating that the NAB was a very effective stimulus to increase VP hnRNA expression in the stimulated SON, but there was no change in OT hnRNA in the NAB-treated SONs as compared to the PBS-control SON. Thus, these data are consistent with the view that the VP hnRNA is regulated by excitatory amino acid input, but that under the same conditions the OT hnRNA is not. In summary, we find that direct excitatory neurotransmitter and acute systemic osmotic stimuli both activate VP gene transcription in the SON, whereas neither stimulus effects OT gene transcription. While NAB activates both MCN phenotypes as measured by an increase in c-fos expression and increased VP gene expression as well, it clearly is an inadequate stimulus for increasing OT gene expression. In addition, the ALZET mini osmotic pumps experimental paradigm used as described above, especially when combined with Laser Capture Microdissection and Microarrays, is a novel approach we will continue to use in our future studies of the regulation of gene expression in the CNS in vivo.[unreadable] [unreadable] With regard to signal-transduction issues we previously reported that the SCN in organotypic culture exhibits a robust circadian rhythm in VP gene transcription and that the daytime peak of VP transcription is completely inhibited to reduced night-time levels by a 2 hour exposure to tetrodotoxin (TTX) in the culture medium. We found that VIP activating a VPAC2-receptor maintains VP gene transcription at peak levels, and in addition, that potassium depolarization was as effective as a stimulus of VP gene transcription in the SCN in TTX as was Forskolin. This suggested that there might also be a depolarization-evoked, presumably calcium ion-dependent pathway that could increase VP gene transcription in the SCN), i.e., a VIP-cAMP-MAPkinase pathway and a depolarization-calcium ion influx (L-channel)-CaCAM kinase pathway, possibly acting co-operatively to phosphorylate CREB, and possible activate other transcription factors. We are currently applying a similar strategy to study the signal transduction of OT and VP in the SON but in vivo, by using stereotaxic injections into and AZET miniosmotic pump infusions into the SON to affect the MNCs as described above.[unreadable] [unreadable] Another major effort in our laboratory has been to establish a viral vector gene transfer methodology in our laboratory based on adeno-associated viral & lentiviral vectors that is being used to transduce MCNs in vivo and in vitro with various wild-type and mutant vectors driven by cell-specific promoters. In general we favor using the lentiviral vector initially because of its larger insert size capacity (8kb), versus the AAV which is significantly smaller (<4.3kb). Both viral vectors have been successfully used as vehicles for gene transfer in the CNS in vivo, and we have found that AAV is more efficient in organotypic culture. We are using the Lentilox 3.7 (pLL3.7) self-inactivating (third generation) lentivirus vector with a highly modified, partial HIV type 1 genome containing a CMV promoter driving an EGFP reporter, and a U6 promoter upstream of cloning sites for shRNA expression. We also have replaced the U6 promoter in this plasmid by a CMV promoter in order to express proteins from this site. In addition, we are presently modifying the LV-cre vector for this purpose. Thus far, we have succeeded in using standard protocols for the production, packaging, purification, and titering of both lentiviral and AAV-2 vectors. In the past year, we have been successful in doing stereotaxic injections of various viral vectors into the SON in vivo. Experiments currently in progress are to test our positive control constructs for OT and VP cell specific gene expression, i.e., OT III.EGFP.IGR 3.6 (insert length 5.9kb) and VPIII.EGFP.IGR 2.1 (insert length 8.3kb), in the lentivirus vector, and to do stereotaxic injections over the SON and PVN in vivo.

In this project we focus on epigenetic and signal-transduction mechanisms that regulate cell-specific OT and VP gene expression in the hypothalamus. In addition we do experiments in this project which continue to refine the organotypic culture of hypothalamus model that we previously developed in our laboratory, and which has proven to be so valuable for our signal-transduction mechanisms studies in the SCN (Rusnak et. al., 2007). A key issue in this project is to evaluate the chromatin status of the OT and VP genes in the SON. Our hypothesis is that certain chromatin modifications in the two genes are specifically expressed in the OT and VP neuronal phenotypes. One specific epigenetic mechanism that we attempted to study is the acetylation and methylation of histones on the OT and VP genes in the OT- and VP-MCNs. This type of chromatin modification is very dynamic &therefore reversible, so the conditional co-expression of OT &VP observed in some MCNs in the SON could occur under some circumstances (see Glasgow et al, 1999;Xi et al, 1999). To address this hypothesis, we did various epigenetic experiments focused on analyzing specific histone acetylation &methylation patterns in OT &VP chromatin in SON and in other specific genes, under various physiological conditions in vivo, and tested effects of acetylation/deacetylation inhibitors on OT &VP chromatin and hnRNA transcription in vivo. Unfortunately, these experiments were not successful, in large part, because the harvesting of SON tissues for the isolation of chromatin by conventional tissue punch techniques produced too much contaminating chromatin from non-MCN cells thereby creating an unfavorable signal to noise situation for the subsequent ChIP methodology. Similar experiments will be attempted again in the coming year, but will use laser microdissected (LCM) MCNs from the SON in an effort to improve the signal to noise situation for the ChIP protocol. The second specific epigenetic mechanism that we are studying is DNA methylation. We hypothesize that the silencing of these genes in most brain areas such as in cortex, striatum, cerebellum, etc, and likely also OT in the SCN, is due to the DNA methylation of these genes in the non-expressing cells. The stable silencing of the OT &VP genes in areas such as cerebral cortex, etc, and possibly of OT in the SCN could be due to DNA methylation, which is much less dynamic and in many cases irreversible. Experiments under way to test this hypothesis use the stereotaxic injection of AAV-LCM strategy developed and described in the summary of project No. 1 Z01 NS002723-24 LNC in order to identify and isolate OT- and VP-MCNs for RNA analysis. However, in this case we isolate DNA from the pools of the individual identified OT- and VP-MCNs for analysis of their methylation patterns by the bisulfate conversion procedure. These experiments require much larger numbers of LCM-isolated MCNs than those directed at RNA analysis by qPCR, and we are presently in the process of collecting the required number of neurons needed for this purpose. Organotypic cultures of mouse and rat magnocellular neurons (MCNs) in the hypothalamo-neurohypophysial system (HNS) and SCN neurons have served as important experimental models for the molecular and physiological study of these neuronal phenotypes. However, it has been difficult to maintain in vivo level numbers of the MCNs, particularly vasopressin MCNs, in these cultures for long periods of time in culture. The vulnerability of the MCNs to axotomy-induced programmed cell death that occurs in organotypic cultures is analogous to the extensive retrograde degeneration of these neurons that occurs in vivo after axonal damage (Shahar et al, 2004). This has led us to the hypothesis that this vulnerability of the MCNs is related to the loss of retrograde trophic factors from the posterior pituitary after axonal injury. Recently, we obtained some support for this hypothesis by employing two LIF-family neurotrophic factors found in the pituitary, leukemia inhibiting factor (LIF) and ciliary neurotrophic factor (CNTF), to rescue rat OT- and VPMCNs from axotomy-induced, programmed cell death in vitro. Quantitative data were obtained that showed the efficacy of the LIF family of neurotrophic factors to enhance the survival of MCNs in three nuclei, the paraventricular (PVN), supraoptic (SON), and accessory (ACC) nuclei in the mouse and rat hypothalamus in vitro (House et al, 2009). This optimization of the survival of the OT- and VP-MCNs in these cultures should greatly facilitate future cell-biological and physiologic studies of these important neuronal systems.

We are addressing two key issues and associated hypotheses related to molecular mechanisms that regulate cell-specific OT and VP gene expression in the hypothalamus. The first issue is to evaluate which cis-elements in the OT &VP genes are responsible for the cell-specific regulation. Previous transgenic studies (Young et al,1990;Jeong et al, 2001;Davies et al, 2003) &in vitro analyses using biolistics and organotypic cultures (Fields et al, 2003) show that the expression of OT &VP is due to the coordinate action of specific cis-elements found 554 bp 5upstream of the transcription start site (TSS) in the OT gene, and <3.4 kbp 5 upstream of the TSS in the VP gene, and 178, and 430 bp 3downstream of exon 3 in the VP and OT genes, respectively. Recent in vitro studies showed that the 178 and 430 bp downstream regulatory elements (REs) are interchangeable with regard to expression of the the OT &VP genes (Fields, unpublished), indicating that these REs are not responsible for the cell-specific expression. More recent experiments using AAV vectors to transduce (transfect) neurons in the rat SON in vivo with the 563-OT-III-EGFP-520 sequence have confirmed that the DNA sequence 563bp upstream of the TSS in the OT gene is able to produce robust expression selectively in OT magnocellular neurons (MCNs) (Fields &Ponzio, unpublished). Additional experiments have indicated that AAV vectors containing 448 bp, 325bp and 216bp upstream sequences of the OT gene can support cell specific OT gene expression (vs VP cell expression). However, the 216 bp upstream region also leads to expression in a non-MCN population of neurons dorsal to the SON. In contrast, the 100bp upstream region produces clear expression but with no specificity at all, as might be expected of a core promoter region. Given these data, we hypothesize that specific cis regulatory elements (REs) located in the 325bp 5upstream region in the OT gene, are critical for its cell-specific and regulated expression in OT magnocellular neurons (MCNs) in the SON. More specifically, we hypothesize: 1) that there is a repressor RE in the -216 to -100 region of the OT gene preventing VP cell expression, 2) ) that there is an enhancer RE in the -216 to -100 region of the OT gene specific for OT cell expression, 3) that there is another repressor RE in the -325 to -216 region of the OT gene which prevents expression in the supra SON (non-MCN) population of neurons, and 4) there may be additional enhancer REs in the -440 to -216 region of the OT gene specific for the OT cell expression. Finally, the putative ROR alpha activation site at -180 to -160bp upstream previously described in the OT gene (Chu &Zingg, 1999) is a candidate for the putative enhancer RE in the -216 to -100 region of the OT gene specific for OT cell expression. Experiments in progress that are testing these assertions are: 1) Determine if the minimal REs (325 vs 216) essential for basal cell specific expression in the SON will also support regulated expression during acute and chronic osmotic stimulation. 2) Determine if the minimal REs (325 vs 216) essential for basal cell specific expression in the SON will also work in the PVN &SCN. 3) Do more deletion experiments actual &in silico to further dissect the putative REs in regions -325 to -216 and -216 to -100. (It might be possible to test whether there is an additional OT enhancer in -440 to -216 and only one VP repressor, i.e, in -216 to 100bp, by removing -216 to -100 from p440bp and seeing if the resulting sequence without the -216 to -100bp region is robust in both the OT &VP cells,etc.) 5) Begin deletion analyses of the 3.4kbp 5 regions in the VP gene to determine which REs are critical for cell-specific and regulated expression in the SON. The second issue is to determine which transcription factors and co-regulators are present &functioning in the OT &VP MCNs in the SON With respect to this issue, we previously used differential analyses of single cell OT &VP expression libraries (Yamashita et al, 2002) and laser microdissection of the SON and microarrays (Mutsuga et al, 2004;2005;Yue et al, 2006),and we identified a number of specific molecules, in addition to OT and VP, that are preferentially expressed in the MCNs in the SON, and which are substantially regulated in expression during osmotic adaptation. We hypothesize that cell specific expression of the OT &VP and other genes in the MCNs will depend on specific proteins in the transcriptosome (e.g, transcription factors, enhancers,suppressors, coactivators, cosuppressors , etc) in each cell type. Experiments in progress that address this hypothesis are: 1) To examine whether specific candidate transcription factors, enhancers, suppressors, coactivators, cosuppressors genes are in fact present in the SON, and if they are selectively expressed in OT vs VP MCNs. One approach will be to perform laser microdissection of the SON, isolation of its RNAs and to determine the presence of candidate transcription factor mRNAs in the SON by qRTPCR. Some of the candidate transcription factors we will test as regulators of OT &VP transcription in the SON are CREB, ER beta, GR, and RORalpha. The strategy will be to use lentiviral and/or AAV vector transduction to express the wildtype genes or to knockdown their expression (using shRNA, or dominant negative specific genes) in the SON in vivo, and to then measure changes in the OT &VP hnRNAs by qRTPCR.

Epigenetics is defined as the covalent modification of chromatin that influences activity-dependent changes in gene expression. The two main molecular epigenetic mechanisms are posttranslational histone modifications and DNA methylation. Our working hypothesis is that chromatin modification by DNA methylation of the OT and VP genes and their respective receptor genes is responsible for the selective expression of these genes in the brain. Methylation of DNA is a direct chemical modification of a cytosine catalyzed by a class of enzymes known as DNA methyltransferases (DNMTs). The DNMTs transfer methyl groups to cytosine residues, specifically at the 5th position of the pyrimidine ring. Cytosines must be immediately followed by a guanine to be methylated. These CpG dinucleotide sequences are highly underrepresented in the genome, and often occur in small clusters known as CpG islands. Of the three main DNMTs - 1, 3a and 3b, the latter two are thought to be responsible for de novo methylation on previously unmethylated CpG sites. Hypermethylation of CpG islands in the vicinity of genes is usually considered to be a transcription suppressing mechanism, although it was shown in some cases to be associated with transcription activation. The transcription regulating role of DNA methylation is mediated by methyl-DNA binding proteins (MBDs) such as MeCP2 whose loss of function is responsible for Rett syndrome. Of all the epigenetic mechanisms, DNA methylation is considered as the most stable, thus most suitable for long-term processes underlying maintenance and persistence of memory. Indeed, inhibiting brain DNMTs activity alters DNA methylation, blocks hippocampal LTP and impairs hippocampal-dependent memory formation. Recent studies have shown that DNA methylation is more dynamic than previously thought due to active demethylation by enzymes such as Gadd45b. DNA methylation was suggested to be involved in the regulation of the oxytocinergic system in the brain since the OT receptor gene was found to be hypermethylated on its promoter-located CpG island in the prefrontal cortex of autistic individuals. Since we proposed that there was a potential regulation of OT and VP expression by DNA methylation, we studied the differential methylation of specific CpG islands within the OT and VP genes in different brain areas to test this hypothesis. The results indicated that there were no differences in methylation patterns between the OT and VP genes in the CpG islands that we studied, when comparing DNA from brain regions that do not express these peptide genes (e.i., cortex) versus brain regions that do (i.e, in the SON). However, our studies on the oxytocin receptor (OXTR) gene did implicate methylation in the regulation of its gene expression. All central and peripheral actions of oxytocin are mediated through the oxytocin receptor (OXTR), which is the product of a single gene. Transcription of the OXTR is subject to regulation by gonadal steroid hormones, and is profoundly elevated in the uterus and mammary glands during parturition. Here, we hypothesized that methylation of the promoter of the OXTR gene (OXTRp) modulates its transcription, in a manner that is subject to physiological changes. Hypothalamus-derived GT1-7, and mammary-derived 4T1 murine cell lines displayed negative correlations between OXTR transcription and methylation of the gene promoter, and demethylation caused a significant enhancement of OXTR transcription in 4T1 cells. Using a reporter gene assay, we showed that methylation of specific sites in the gene promoter, including an estrogen response element, significantly inhibits transcription. Furthermore, methylation of the OXTRp was found to be differently correlated with OXTR expression in mammary glands and uterus of virgin and post-partum mice, suggesting that it plays a distinct role in OXTR transcription among tissues and under different physiological conditions. Together, these results support the hypothesis that epigenetic regulation by DNA methylation of the OXTR gene promoter does modifiy expression of the OXTR gene. This project will terminate in the next fiscal year.