This page was produced as an assignment for Genetics 677, an undergraduate course at UW-Madison.

Scientific Article Review

Open the file below to view the article under review:

File Size: 1054 kb
File Type: pdf
Download File


            In the paper “Glutamate receptor metabotropic 7 is cis-regulated in the mouse brain and modulates alcohol drinking," which was published in Genomics in August 2007, Vadasz et al. identify, for the first time, Grm7 as a gene for alcohol consumption. This paper is particularly significant in that, as pointed out in the introduction, even in animal studies for which the environment can be controlled, it is difficult to identify genes of small effect size that are involved in alcohol-related behaviors due to genetic background noise. Using a unique experimental design involving congenic mice that have drastically reduced levels of genetic background noise, as well as a number of genetic, transcriptome, and bioinformatic analyses, Vadasz et al., are able to first narrow down a previously identified QTL region for alcohol consumption to one candidate gene, Grm7, and then characterize the genetic variation in this gene that causes the different alcohol consumption phenotypes (see below for a detailed description of their results). While their analyses cannot completely rule out other genes in the QTL region as contributors to variation in alcohol consumption phenotypes, their results show, quite convincingly, that Grm7 plays a role in alcohol consumption. Vadasz et al. comment that “further rigorous work including fine mapping and mechanistic studies,” will confirm the role of Grm7 in modulating alcohol-related behaviors. But, while the authors of this paper provide very strong evidence for the role of the Grm7 gene in alcohol consumption in mice, thoroughly addressing any potential holes their results, they make a rather weak link to the role of this gene in alcoholism in humans. Presumably, the work done for this study was to meant to be transferable to alcoholism in humans; however, while pointing out in the discussion that human GRM7 has multiple mRNA isoforms and plays a similar role to that of mouse Grm7, Vadasz et al. do not mention the existence of multiple GRM7 alleles in humans, and therefore, do not provide particularly persuasive evidence that human GRM7 is likely to modulate alcohol consumption in the same way that the different alleles of mouse Grm7 do. Although it is unclear exactly how much this study will inform our understanding of alcoholism in humans, the wealth of evidence for the role of Grm7 in modulating alcohol consumption in mice presented in this paper lends support to the emerging idea that glutamate is an important player in alcoholism (Vadasz, et al., 2007).

Summary of Results and Figures

           In their paper entitled "Glutamate receptor metabotropic 7 is cis-regulated in the mouse brain and modulates alcohol drinking," Vadasz et al. describe the genetic, transciptome, and bioinformatic analyses that led them to uncover the role of Grm7 in modulating alcohol consumption in mice. In the first step of these analyses, they developed a congenic strain of mice using C57BL/6ByJ, a strain which has preference for alcohol (over water), as the background strain, and BALB/cJ, a strain which does not have a preference for alcohol, as the donor strain. This congenic strain carried a BALB/cJ donor segment for a particular region of Chr. 6 that was previously shown as a quantitative trait locus (QTL) for alcohol consumption. As shown in Figure 1A, the congenic mice (abbreviated B6By.C6), which carry a portion of the BALB/cJ (abbreviated C; its chromosomes are represented in white) Chr. 6, show significantly lower alcohol preference than the background C57BL/6ByJ (abbreviated B6By; its chromosomes are represented in black), suggesting that the donor region on Chr. 6 carries at least one genetic element that influences alcohol preference (Vadasz, et al., 2007).

   (Figure 1, Vadasz, et al.,2007)


          In the next step of the analyses, the set of candidate genes in the Chr. 6 segment of interest was narrowed down by identifying those regions of the segment that were identical by descent (IBD) in C57BL/6ByJ (the background strain) and BALB/cJ (the donor strain). Vadasz et al. reasoned that those regions of Chr. 6 for which the two strains IBD, and therefore, share a common ancestor, are unlikely to contain the genetic causes of the differences in alcohol preference between the congenic strain they created, and its background strain, C57BL/6ByJ. For the purpose of these analyses, IBD regions were defined as those regions having a low frequency of differing SNPs. As seen in Figure 2, which shows the number of SNP locations for which allele data was available for both C57BL/6ByJ and BALB/cJ as gray bars, and the number of SNPS differing between the two strains at each site as red bars, 11 IBD regions were contained in the Chr. 6 segment of interest. This reduced the number of candidate genes from 404 to 212, and “filtered out, for example, Slc6a1…, and otherwise likely candidate” (Vadasz, et al.,2007).

   (Figure 2, Vadasz, et al.,2007)

            The 212 candidate genes on the Chr. 6 segment of interest were filtered down to one, Grm7, in the next steps of the analyses. The first approach taken for this filtering was to search for coding-nonsynonymous SNPs (CNs) between the congenic strain, C57BL/6ByJ, and BALB/cJ. Although total of 11 potential CNs were identified, analysis of the Gene Ontology (GO) classifications of the genes containing these CNs did not reveal an obvious link between any of these genes and alcoholism, and these genes were not pursued further. Instead, Vadasz et al. decided to narrow down the 212 non-IBD genes by identifying those that were expressed in the hippocampus, an area of the brain involved in alcoholism, and also had genotype-dependent transcript abundance. Using hippocampal expression data for a cross between C57BL/6ByJ and BALB/cByJ (CXB), QTL mapping of mRNA abundance of the candidate 212 genes was conducted to identify those genes that showed genotype-dependant expression levels. This analysis narrowed the number of candidate genes to a mere seven, which are shown in Table 1. These seven genes were further narrowed to four through a microarray gene expression experiment that tested for the relative brain mRNA abundances of the seven candidate genes in the congenic strain as compared to the C57BL/6ByJ background strain. Finally, through estimation of the correlation between the SNP genotypes of these four remaining candidate genes and the alcohol phenotypes across commercially-available common inbred mouse strains, as well as analysis of the GO classifications of these four genes, Grm7 was identified as the primary candidate gene (Vadasz, et al., 2007).

   (Table 1, Vadasz, et al.,2007)

             Next, the specific genetic difference between C57BL/6ByJ and BALB/cJ present in Grm7, which is responsible the phenotypic differences in alcohol consumption in the mice under study, was identified. Based on their hypothesis that the alcohol-consumption phenotypes were based on differences in Grm7 expression levels, the promoter regions of the Grm7 gene were searched for SNPs differing between C57BL/6ByJ and BALB/cJ. Only one such SNP, whose genomic context is shown in Figure 3, was found. Interestingly, in silico analysis showed that when a T is present at this particular SNP, a binding site for the transcription factor Sox5 exists, while when a C at this SNP, the Sox5 binding site is abolished. As expected, commercially-available strains that carry a T at this site, and therefore, have the Sox5 binding site, have a higher relative Grm7 transcript abundance, while those carrying a C at this SNP have a lower relative Grm7 transcript abundance. Further experimental support for the involvement Sox5 in transcriptional regulation of Grm7 in the brain is needed, however (Vadasz, et al., 2007).

   (Figure 3, Vadasz, et al.,2007)

             In the final part of this study, the role of Grm7 in alcohol preference was confirmed through the use of two different analyses. First, to confirm that expression levels of Grm7 were genotype-dependant, and therefore, could be implicated in the phenotypic differences in alcohol consumption, the expression QTL mapping data obtained from their previous analysis of the CXB cross was validated by conducting expression QTL mapping on hippocampal dataset for a cross between C57BL/6J (B) and DBA/2J (D), as well as two striatal databases for BXD mice and B6D2F2 mice. The results of these analyses, shown in Figure 4, provide strong support that expression of Grm7 is genotype-dependent in both the hippocampal and striatal brain regions. Additionally, to confirm the association between alcohol consumption and Grm7 genotype, the F2 mice from a reciprocal cross between two quiasi-congenic strains, which both have greater than 97% C57BL/6J background: the alcohol preferring strain I5B25A, and the alcohol-nonpreferring strain C5B3. As shown in Figure 1B, F2 mice homozygous for the BALB/cJ genotype at a SNP lying in intron 7 of the Grm7 gene (these mice are T:T) show significantly lower alcohol consumption than those mice homozygous for the C57BL/6J genotype at that SNP (G:G). Interestingly the F2 mice heterozygous for the BALB/cJ and C57BL/6J genotypes (T:G) show intermediate alcohol consumption. This F2 reciprocal cross was also important in that it supported the exclusion of the other three genes on the Chr. 6 segment of interest that were identified earlier by the microarray as having differential expression, because the alcohol-nonpreferring C5B3 strain used for the cross carried the C57BL/6J alleles of these three genes (Vadasz, et al.,2007).

   (Figure 4, Vadasz, et al.,2007)



Vadasz, C., Saito, M., Gyetvai, B. M., Oros, M., Szakall, I., Kovacs, K. M., Prasad, V. V. T. S., Toth, R. (2007). Glutamate receptor metabotropic 7 is cis-regulated in the mouse brain and modulates alcohol drinking. Genomics, 90(6):690. doi:10.1016/j.ygeno.2007.08.006


Jennifer Wagner
Updated March 6, 2009