RSA 2007 Colorado

RSA 2007 Abstracts

University of Colorado Health Sciences Center

ANALYSIS OF “CANDIDATE GENES” FOR ALCOHOL CONSUMPTION IN HXB/BXH RATS: S. V. Bhave; L. Saba; M. Printz; P. Flodman; J. Gaydos; J. Mangion; N. Hubner; P. L Hoffman; B. Tabakoff. University of Colorado at Denver & Health Sciences Center, Aurora, CO 80045.

Alcoholism is a complex phenotype, but by breaking it into endophenotypes, we are able to focus on the genetics of different aspects of the disease. In this study, the endophenotype of interest was alcohol consumption in rats when given a choice between alcohol and water. We used a two step approach to identify “candidate genes” for alcohol consumption. First, gene expression was correlated with alcohol consumption. In the second step, we used behavioral QTLs to filter “candidate genes” by focusing only on those correlated genes whose expression level is controlled from within a behavioral QTL (bQTL). A panel of HXB/BXH recombinant inbred rats was used to measure alcohol consumption in the two-bottle choice paradigm and to measure alcohol naïve whole brain gene expression profiles using microarray technology. Transcript expression values were correlated with the quantitative phenotype of alcohol consumption using a bivariate random effects model. bQTL were determined for two different measurements of alcohol consumption that were also used in the correlation analysis and were compared to previously published bQTL for alcohol preference. Expression QTLs (eQTLs) were determined for individual transcripts indicating whether they were cis- or trans-regulated. Of the over 19,000 transcripts tested for correlation, 416 transcripts were correlated with alcohol consumption under both conditions used in our study. Out of 416 correlated transcripts, 16 transcripts had their expression regulated (significant eQTLs) from within a bQTL. Two transcripts, out of the 16, were cis-regulated and 14 showed trans-regulation. To understand the interactions between these 16 transcripts, we used the program PathwayAssist to cluster them in a pathway. Ten transcripts had functional annotation and remaining 6 transcripts were either ESTs or without any annotation. Eight of the annotated “candidate” transcripts were part of a pathway related to G protein signaling and protein phosphorylation. When considering all of the proteins in the identified pathway, the GO categories of signal transduction and G-protein coupled receptor protein signaling cascades were over-represented. Supported by the Banbury Foundation and NIAAA (INIA).

THE COLORADO INIA INFORMATICS WEBSITE – ANALYSIS OF “CANDIDATE GENES” FOR COMPLEX TRAITS: S. V. Bhave; C. Hornbaker; T. Phang; L. Saba; R. Lapadat; K. Kechris; J. Gaydos; D. McGoldrick; A. Dolbey; B. Soriano; A. Ellington; E. Ellington; K. Jones; J. Mangion; J. Belknap; R. W. Williams; L. E. Hunter; P. L. Hoffman; B. Tabakoff. University of Colorado at Denver & Health Sciences Center, Aurora, CO 80045.

            The informatics component of the Integrated Neuroscience Initiative on Alcoholism (INIA)-West has created a website (Colorado INIA Informatics, C-INIAi, http://inia.uchsc.edu) that provides users with a wide variety of tools particularly suited for combining QTL and micoarray data to search for “candidate” genes contributing to complex traits. Users can create new “in silico” experiments using their own data or data shared by other investigators. The website currently contains data from almost 1000 arrays, encompassing Drosophila, human, mouse and rat brain tissue. The website provides a wide variety of tools for quality control, normalization and filtering of the microarray data. Parametric and non-parametric statistical tests, including one-way and two-way ANOVA, and correlation analysis with a given phenotype, are available to analyze differential expression, with options for multiple comparison p-value adjustment. The user has access to tools for analysis and interpretation of differentially expressed gene lists: gene annotation is available from several sources, and the website provides the capacity to obtain information about the chromosomal location of each known gene; and about quantitative trait loci (QTLs) for various phenotypic traits; and localize genes of interest to these QTLs. Expression QTLs determined from the whole brain gene expression profiles in BxD recombinant inbred (RI) mouse strains or BXH/HXB RI rat strains are also available. The ability to identify differentially expressed transcripts with expression QTLs that overlap phenotypic QTLs represents a powerful method for identifying candidate genes that may contribute to the phenotype through their expression levels. Other tools available include “promoter analysis” and “literature search”. Supported by the Banbury Foundation and NIAAA (INIA).

PROTEOMIC ANALYSIS OF BRAINS OF MICE DISPLAYING ETHANOL WITHDRAWAL-INDUCED DRINKING (WID): P. L. Hoffman; C. C. Wu; B. J. Schiemann; A. Roberts; B. Tabakoff. University of CO at Denver and HSC, Aurora, CO 80045.

            Chronic intermittent exposure of C57BL/6 mice to ethanol vapor results in an increase in subsequent ethanol self-administration, called “withdrawal-induced drinking” (WID). This animal model can be used to investigate the neurobiological basis of increased alcohol intake in individuals with a history of alcohol dependence. We have used an unbiased method of “shotgun proteomics” to evaluate changes in brain proteins in animals that display WID. C57BL/6 mice drank alcohol in a limited access two-bottle choice paradigm for 7 days, then had 3 cycles of exposure to ethanol vapor (16 hr on/8hr off) or air (controls) in inhalation chambers. After the third cycle, mice were again given ethanol in the two-bottle choice paradigm for 7 days. This whole procedure was repeated, and resulted in WID. Mice were killed two weeks after the second vapor chamber exposure, when alcohol drinking was still increased over baseline. Brains were dissected into cortex, hippocampus, subcortex and cerebellum. For proteomic analysis, all regions except cerebellum, from eight animals per group, were pooled and subjected to subcellular fractionation. The synaptic membrane fraction was incubated at high pH and digested with trypsin. The resulting peptide mixture was analyzed using multidimensional protein identification technology (MudPIT). Peptides were separated by liquid chromatography and identified by mass spectrometry. Tandem mass spectra were analyzed using SEQUEST, and 1537 proteins could be identified (FDR=0). Proteins were quantitated based on relative spectral counts in the control and ethanol vapor-exposed mice (“WID mice”). Twelve hundred and forty two proteins were identified. Among the altered proteins were the glial high-affinity glutamate transporter and glutaminase (higher in WID mice) and glutamine synthetase and PSD-95 (lower in WID mice). These differences could result in more neurotransmitter glutamate in the WID mice, leading to a postsynaptic adaptation (decreased PSD-95). There were also changes in (several) proteins associated with neurotransmitter release, and structural proteins. The results (are consistent with) both pre- and postsynaptic changes in neurotransmitter function that may contribute to WID. Supported by NIAAA (INIA Project).

ROLE OF ADENYLYL CYCLASE TYPE VII IN HPA RESPONSE TO ALCOHOL:
S. P. Pronko; H. C. Pronko; L. Saba; P. L. Hoffman; B. Tabakoff. Department of Pharmacology, School of Medicine, UCHSC at Fitzsimons, Aurora, CO 80045.

            Though ethanol is considered an anxiolytic drug, acute ethanol administration increases glucocorticoid levels through increased synthesis of POMC and ACTH release. Studies using corticotrophs, indicated1a to adenylyl cyclase type 9 and type 7 (AC7), with AC7 becoming dominant under stress conditions. Our studies of ethanol’s action on AC activity have shown that ethanol enhances AC7 responsiveness to Gsa and AC7 is 2-3 times more responsive to ethanol than any other AC isoform. Therefore, we hypothesized that ethanol through its ability to increase the affinity of the AC7 for Gsa, can increase the responsiveness of the AC7 to CRF via CRF-R1a and Gsa. This increased responsiveness would lead to increased ACTH and corticosterone (CS) levels. To investigate this possibility we measured ACTH and CS levels after i/p injection of ethanol (1.75 and 2.25 g/kg) at different time points (0-120 min) in plasma of male and female AC7 transgenic (TG) mice in which AC7 is overexpressed in brain, heterozygous knockdown (HET) mice in which AC7 is underexpressed in brain and their respective littermate controls (WT). In female animals after ethanol was injected at a dose 2.25 g/kg ACTH and CS levels were higher in TG group compared to HETs, and the ACTH and especially CS levels were markedly higher compared to respective male groups. In male group after i/p injection of 2.25 g/kg TG male mice displayed significantly higher plasma ACTH levels compared to WT and HET groups while HETs displayed significantly lower ACTH levels compared to WT and TG animals. CS levels differed only between TG and HET mice. This phenomenon was however explained to be a result of a “ceiling” response to ACTH in its ability to stimulate CS secretion after 2.25 g/kg ethanol. When we utilized the lower (1.75 g/kg) dose of ethanol which produced lower ACTH levels, the TG animals had significantly higher CS concentration compared to WT animals. The data provide evidence for the integral role of AC7 in the increase of plasma ACTH and CS levels during alcohol intoxication. This work supported by NIAAA (INIA) and Banbury Foundation.

INIA HomeINIA StructureAdministrative Core
Genetic Animal Models CoreGene Expression CoreImaging CoreNeuroinformatics Core
SubgroupsCurrent GrantsConference AbstractsPublicationsLinks