In this section, we report our recent findings on the role of genetic variants that regulate the biological responses and synthesis of the neuronal survival and plasticity-associated factor, brain-derived neurotrophin factor (BDNF). Two studies were conducted. The first study describes the role of gene variation in response to the selective serotonin reuptake inhibitor citalopram in a large cohort of adults with major depression. In the second study, we determined an association of the human tyrosine kinase B receptor gene (NTRK2), the cognate receptor for BDNF, with sequence variants of NTRK2 that may have a role in alcohol response. ? ? Major depression. Strong evidence from family and twin studies demonstrates that major depressive disorder (MDD) is heritable, yet there has been limited progress in identifying the actual genes involved. A separate, perhaps overlapping set of genes is expected to play a role in individual variation in treatment response in MDD. In collaboration with the Sequenced Alternatives to Relieve Depression (STAR*D) project we performed genetic analyses using a large set of markers in many genes on unrelated patients who differ in their response to standard treatments as defined in the STAR*D protocol. The initial focus was on two sets of genes most likely to play a role in the etiology of major depression: 1) genes selected on neurobiological grounds because of their known involvement in pathways thought to be important in mood disorders; and 2) genes implicated primarily by their positions within genomic regions implicated by genetic linkage or association studies of major mood disorders and related conditions. These two approaches are complimentary since the neurobiological candidates arise directly from existing etiologic hypotheses while the positional candidates may lead to the discovery of unexpected pathways. We utilized state-of-the-art, high-throughput genotyping methods as well as sophisticated methods of genetic analysis that take into account haplotypes and multi-locus interactions in addition to standard, single-marker analyses. Primary comparisons were performed within the group of cases who respond to the selective serotonin reuptake inhibitor (SSRI),citalopram, and those who do not. In our first report, an association between the HTR2A gene, which encodes the serotonin 2A receptor and a known modulator of the neurotrophin BDNF, with patient response to citalopram was identified (McMahon et al., Am J Hum Genet 2006, 78: 804-814). ? In another study using the STAR*D cohort, we tested the hypothesis that the HTTLPR polymorphism, a frequently studied variation in the 5? flanking region of the serotonin transporter gene (HTT) was associated with treatment outcome. Conventionally, HTTLPR, which is composed of a variable number of tandem repeats, has two common variants: a long allele of 16 repeats and a short allele of 14 repeats. Recently, we determined that a single nucleotide polymorphism (SNP) within one of the repeats comprising the long allele causes the HTTLPR locus to be triallelic. The A allele of the long variant (LA)has increased HTT transcription, while the other allele, which contains a G (LG) binds the transcription factor AP2, that represses HTT transcription to levels that are similar to the short variant (Hu et al., Am J. Hum Genet 2006, 78: 815-826). A significant association with tolerability to citalopram was observed in a Caucasian sample based on LA allele frequency and genotype frequency. No significant association was observed between citalopram treatment response or symptom remission phenotypes and HTTLPR by genotype or allele frequencies. Because the LA allele predicts increased transcription of HTT, increased serotonin transporter levels in brain and other tissues may lead to increased tolerance to SSRI drugs targeting the transporter (Hu et al., submitted). Taken together, these results provide new insights into the genetic determinants of treatment response in MDD, and may also shed light on genes contributing to risk for MDD, with the promise of novel treatment approaches and improved targeting of existing therapies.? ? Alcoholism. Vulnerability to alcoholism is most likely due to multiple interacting genes, each have a modest contribution to the phenotype. Understanding the genetic basis of alcohol dependence and, in particular, understanding the contribution of allelic variation to alcohol dependence will aid in defining the how functional pathways are altered in the brain to produce addiction, which may be shared with other addictions in humans, such as cigarette smoking.? Linkage-based genome scans have identified multiple chromosomal regions as sources of potential susceptibility to alcohol dependence as well as other addictions, showing some convergent findings. Other convergent linkage findings in the addictions, particularly cigarette smoking with or without alcohol dependence have been reported for loci on chromosomes 9 and 11. Among other candidate addiction susceptibility genes defined by linkage studies are those that encode the neurotrophin, BDNF, (maps to chromosome 11), and its cognate receptor, neurotrophic tyrosine kinase receptor B (TrkB)(NTRK2, located on chromosome 9). ? Evidence supporting a role for BDNF signaling in mechanisms of alcohol and drug dependence has been well documented using different animal models. We thought that NTRK2 appeared to be a good candidate gene for alcohol vulnerability. We determined the association of the human tyrosine kinase B receptor gene (NTRK2) and identifed sequence variants of NTRK2 that may have roles in neuronal responses to alcohol. We resequenced the 5? flanking region, determined linkage disequilibrium (LD) values, haplotype structure, and performed association analyses using 43 SNPs in a Finnish Caucasian sample composed of 229 alcoholics and 287 healthy controls. Resequencing uncovered seven SNPs, two were novel. ? Association of 13 SNPs with alcohol dependence and alcohol abuse (AD) was observed (p-value from 0.002 to 0.047). Five regions of high linkage disequilbrium (LD blocks) were identified in the study sample, which correlated with HapMap Project results from a Utah CEPH panel. Case-control haplotype analyses supported similar regions of association as those found with individual markers. In particular, a common eighteen-locus haplotype within the largest LD block of NTRK2, embracing a region of at least 119 kb including the 5? flanking region and exons 1 through 15, was significantly overrepresented in control subjects compared to AD individuals (?2 = 5.898, p = 0.013). We also observed two common haplotypes within different LD blocks composed of five loci and seven loci and encompassing the middle portion of the gene were more abundant in the control group than in the AD group (?2 = 4.490, p = 0.034; ?2 = 5.858, p = 0.015 respectively). In contrast, the frequency of a haplotype from an LD block near the 3?end of NTRK2 that contained only two loci was significantly higher in the AD group compared to the controls. Based on allele, genotype, and haplotype-based associations with AD, our data provided the first evidence for a functional locus at NTRK2 (Xu et al., submitted).
Showing the most recent 10 out of 13 publications
