 |
|
 |
|
|
 |
 |
 |
 |
 |
 |
 |
 |
|
 |
 |
 |
|
 |
 |
|
 |
 |
 |
 |
 |
RSA 2006 Abstracts
University of Colorado Health Sciences Center
“MAGIC-B”: COMBINING GENETICS AND GENE EXPRESSION PROFILING TO IDENTIFY CANDIDATE GENES FOR ALCOHOL PREFERENCE BEHAVIOR:
P. L. Hoffman; S. V. Bhave; L. Saba; J. K. Belknap; P. Bice; N. Grahame; B. Tabakoff.
University of Colorado at Denver and Health Sciences Center, Aurora CO 80045Alcohol preference represents an endophenotype that may be related to susceptibility to alcohol dependence. We describe the application of the technique “MAGIC-B” (microarray ascertainment of genes that influence complex behaviors) to identify candidate genes associated with alcohol preference, as measured in a two-bottle choice paradigm. This technique makes use of selectively bred lines of mice that differ in voluntary alcohol consumption (high alcohol preferring [HAP] mice, and low alcohol preferring [LAP] mice). Whole brain gene expression profiles from two replicate lines of these mice (6 male mice from each line, 70-90 days of age, RNA from each mouse hybridized to a separate array) were determined using Affymetrix whole genome oligonucleotide arrays. Expression data were normalized by two different procedures (RMA and MAS 5), and differential expression was determined by two statistical procedures (permutation analysis and t-test noise distribution). Transcripts found to be differentially expressed in the same direction in both replicate lines, following both normalization procedures and both statistical analyses, were considered for further analysis. The expression QTLs for these transcripts were determined with R/qtl, using a database of gene expression profiles in whole brains from 25 BxD recombinant inbred strains (4-6 male mice, 70-90 days old, per strain, RNA from each mouse hybridized to a separate array). We then determined transcripts for which the eQTLs overlapped with behavioral QTLs determined for voluntary alcohol consumption in the two-bottle choice paradigm (including QTLs determined from C57BL/6 and DBA/2 mice, and from an F2 generation of HAP and LAP mice). These genes were considered as candidate genes for alcohol preference, and could be organized into signal transduction pathways that include intracellular signaling systems, as well as pathways involved in neuronal survival and differentiation. Supported by NIAAA/NIH, INIA project; VA; Banbury Fund
ADENYLYL CYCLASE TYPE VII AND ANXIETY: I. Goncharov; S. Pronko; P. L. Hoffman; N. Thomas; B. Tabakoff. University of Colorado at Denver and Health Sciences Center, Aurora, CO 80045.
Much evidence implicates the cyclic AMP signal transduction system (adenylyl cyclase, cyclic AMP, PKA, CREB) in acute and chronic actions of ethanol. Of the various isoforms of adenylyl cyclase, the Type VII isoform (AC7) is most sensitive to activation by ethanol. AC7 has also been implicated in the neurobiology of stress, i.e., in the pituitary corticotroph, the coupling of the CRF1 receptor to AC7 is increased under stressful conditions, leading to increased ACTH synthesis/release (Antoni et al., Mol End. 17:692). Ethanol’s known effect to increase plasma glucocorticoids may be due to ethanol-induced potentiation of CRF stimulation of AC7. It is also known that CRF in extra-hypothalamic regions produces an anxiogenic response (likely a postsynaptic effect). CRH can also release GABA in the amygdala (presynaptic effect). The behavioral implications of this event are not clear. Ethanol-induced increases in GABA transmission in the amygdala are mediated via CRF1 receptors (Nie et al., Science 303:1512). We postulate that the effects of ethanol are determined both by the presynaptic and postsynaptic localization of the CRF1 receptors, and by the coupling of these receptors to the various isoforms of adenylyl cyclase, i.e., the function of CRF1 receptors coupled to AC7 is most sensitive to ethanol. To test our hypotheses, we have created transgenic mice that overexpress AC7 in brain, and heterozygous AC7 knockout mice. These mice display gender-specific differences in anxiety, compared to wild-type controls. AC7 transgenic mice also display more severe ethanol withdrawal signs than wild-type mice following chronic ethanol exposure. Initial in situ hybridization analysis demonstrates increased AC7 expression in the paraventricular nucleus of the transgenic mice, suggesting the presence of this isoform in CRF-producing cells, and current work is demonstrating alterations in plasma glucocorticoid levels in the AC7 transgenic and heterozygous knockout mice following ethanol treatment. These studies begin to elucidate the role of AC isoforms in ethanol-induced anxiolytic effects and in “paradoxical” stress-like responses to ethanol. Supported by NIAAA (INIA) and Banbury Fund.
CANDIDATE GENES AND SIGNAL TRANSDUCTION PATHWAYS FOR ALCOHOL TOLERANCE: S. V. Bhave; W. Hu; L. Saba; R. Lapadat; P. L. Hoffman; B. Tabakoff. University of Colorado at Denver & Health Sciences Center, Aurora, CO 80045.
We have developed a methodology, MAGIC-B (microarray ascertainment of genes that influence complex behaviors (Tabakoff et al., 2003, J Neurosci., 23:4491)) to elucidate “candidate genes” that may play a role in certain complex behaviors such as acute functional tolerance to the incoordinating effect of ethanol. This approach includes gene expression profiling with the addition of “filters” such as expression (e) QTL, pathway and promoter analysis. We have used mice bred selectively for high (HAFT) and low (LAFT) acute functional tolerance (AFT) to the incoordinating effect of ethanol for these studies. We used Affymetrix microarrays (MOE430 v2) to evaluate gene expression profiles in whole brain obtained from replicate lines of these selectively bred mice. Gene expression profiles in brains from 30 BXD recombinant inbred strains of mice were used to obtain expression (e) QTLs (using R/QTL). Differences in gene expression between the selected lines (HAFT and LAFT) of mice were analyzed statistically. A total of 275 genes were found to be differentially expressed between the HAFT and LAFT mice. This list of differentially expressed genes was significantly narrowed using eQTLs (chromosomal region regulating the expression of the given transcript) and behavioral QTLs. Behavioral QTLs were derived from our studies of BXD RI strains of mice (Kirsten et al., 2002, JPET, 302:1238). Our assumption is that genes that contribute to the phenotype through differential expression will have their expression regulated in regions of the genome (eQTL) that overlap with the behavioral QTLs. Out of 275 differentially expressed transcripts, 94 had “significant” or “suggestive” eQTLs. Out of these 94 transcripts, 22 had eQTLs located in behavioral QTLs for AFT. Four transcripts were apparently “cis-regulated” and 18 were “trans-regulated”. Pathway analysis, and clustering based gene ontology, indicated the importance of transcriptional regulation and protein binding in the AFT phenotype. Biological confirmation for the genes of interest was provided by analyzing correlations between brain gene expression profiles and AFT in 20 inbred strains of mice. Supported by the Banbury Foundation and NIAAA (INIA).MICROARRAY IDENTIFICATION OF CANDIDATE GENES FOR ALCOHOL PREFERENCE IN HAP AND LAP MICE: B. Tabakoff; S. V. Bhave; L. Saba; J. K. Belknap; P. Bice; N. Grahame; P. L. Hoffman. University of Colorado at Denver and Health Sciences Center, Aurora, CO 80045
We have previously described the “MAGIC-B” (microarray ascertainment of genes involved in complex behaviors) analysis that identifies candidate genes that contribute to complex traits. This procedure begins with an analysis of gene expression profiles in brains of animals that have been selectively bred for a particular trait. The differentially expressed genes are identified by a rigorous statistical analysis. Quantitative trait locus (QTL) analysis is used to determine regions of the genome that influence the trait under study (“behavioral QTLs”). If a gene affects the studied trait through a difference in expression, then one can postulate that the region of the genome that controls expression of the particular gene (“expression QTL”) should be localized within a behavioral QTL. We have used this procedure to identify candidate genes that influence alcohol preference as measured using the two-bottle choice paradigm. We determined gene expression differences (Affymetrix MOE 430 arrays) in brains of male mice selected for high vs low alcohol preference using this behavioral procedure (replicate selected lines of HAP and LAP mice; six mice per line on separate arrays). We determined expression QTLs by performing gene expression analysis (Affymetrix MOE430 arrays) on brains of 30 strains of BxD recombinant inbred mice (5-7 mice per strain). We then assessed the overlap of the expression QTLs of the differentially expressed genes with behavioral QTLs that had been determined for two-bottle choice alcohol drinking in C57 and DBA mice, as well as in an F2 population derived from the HAP and LAP selected lines. This procedure led to identification of 8 candidate genes for alcohol preference. Differential expression of several of these genes was confirmed by qRT-PCR. We also assessed correlation of gene expression with “alcohol preference” in BxD RI strains. This procedure provided a biological confirmation of the results. The genes identified define intracellular signaling pathways as key mediators of alcohol preference drinking. Supported by NIAAA (INIA) and Banbury Fund.
INIA Home INIA Structure Administrative Core
Genetic Animal Models Core Gene Expression Core Imaging Core Neuroinformatics Core
Subgroups Current Grants Conference Abstracts Publications Links