Candidate allele & gene hypotheses developed from pharmacological & genetic findings are tested. We also follow up on whole genome linkage findings. Candidate gene linkage investigations included: DRD2 haplotype/alcoholism and opioid addiction, HTR1B/antisocial alcoholism, serotonin transporter(SLCA4)/anxiety, SLCA4/OCD, BDNF Val66Met/episodic memory, NPY/anxiety, COMT/opioid addiction/Schizophrenia/anxiety/dysphoria/Executive cognition, GTP cyclohydrolase/chronic pain, TPH2/anxiety-dysphoria, GABAA alpha 2 and alpha 6/alcoholism, and DISC1/schizophrenia. Phenotype, sampling framework, genetic structures [e.g. TDT, sibpairs], & power are evaluated. Assessment is generally by structured interview, & usually with intermediate phenotypes and/or additional endophenotypes. The Finnish dataset was ascertained from criminal alcoholic probands & is thus enriched for Type II [early onset] alcoholism. SW Indian, Plains Indian, Ten Tribes, & Finnish datasets are derived from isolates; psychiatric interviewed controls are available from source populations. Chinese and German case/control datasets are in use as is a predominantly African American cocaine dataset [N=1000]. Remaining datasets are almost entirely Caucasian. Some 980 psychiatrically interviewed Caucasian population controls are available. The SW Indian dataset includes 600 sibpairs, the Finnish dataset used in the SLCA4 study included 366 sibpairs, & the Plains Indian EEG dataset contains a large number of sib & relative pairs. Both schizophrenia datasets [Egan & Weinberger, Malhotra] enable TDT/sTDT analysis. A publicly available portion of the COGA datset (Collaboration on the Genetics of Alcoholism) is also used. Power for case/control association is dependent on allele frequency, association strength, & desired level of alpha. Functional alleles tested have moderate [HTR2C, HTR2A, mu opioid receptor: 0.1] or high [SLCA4 & COMT: 0.4] frequencies, & functional polymorphisms known to affect alcoholism ADH2 Arg47His & ALDH2 Glu487Lys have 5-10 fold effects. Therefore, our smallest datasets [n=50] have 58% power at p< 0.01; power is essentially 100% for larger datasets available for alcoholism [several datasets], schizophrenia, OCD, anorexia nervosa, SAD, & bipolar. For lower effect size, power at p<0.01 remains > 0.8 for the association strength of 0.1, in the larger datasets. Relationship of candidate genes and alleles to behavior: DRD2. We early tested the DRD2 dopamine receptor, """"""""Reward Deficiency Gene"""""""" hypothesis in ethnically matched, psychiatrically interviewed alcoholics & controls, in severe alcoholics, & in alcoholics with parental alcoholism finding no evidence for a substantial role for role of the well-known Taq1A locus nor even an effect of Ser311Cys, a functional polymorphism that severely impairs signal transduction & that is uniquely abundant in Indians, our dataset actually including 15 Cys311/Cys311 homozygotes. To monitor unknown functional variants at DRD2 we used a ten-locus haplotype, finding linkage to opioid dependence in two large datasets. A functional promoter polymorphism (HTTLPR) of the serotonin transporter, which inactivates serotonin and is a major antidepressant drug target, has been extensively tied to behavior in studies in which LNG led or was integral and at the levels of complex behavior (anxiety, OCD and depression clinical course) and brain intermediate phenotypes including serotonin transporter density and amygdala metabolic response to emotional stimuli. Our discovery of unrecognized functionality of a common HTTLPR allele (Hu et al, 2006) enabled linkage of gain-of-function genotypes to OCD in two populations. Earlier, we also linked a rare serotonin transporter missense gain-of-function variant to OCD and other severe psychopathology, in two families (Ozaki et al). HTR1B, the termninal autoreceptor modulating serotonin release, was implicated in mouse alcohol preference by a QTL & subsequent discovery that HTR1B knockout mice exhibit increased aggression & preference for alcohol. We built on these mouse model findings to identify a potential role for 5HT1B in a subtype of alcoholism. Evidence for i.b.d. linkage was found in families from two isolates, Finns and SW Indians [p=0.04 & p=0.01] & association to HTR1B was also detected in Finns. TPH2, rate limiting for serotonin synthesis in brain, was linked to anxiety/dysphoria in multiple populations and the linked haplotype was shown to be associated with low serotonin turnover as indexed by 5HIAA levels, except Indians where the protective haplotype is absent. Meanwhile, we led a consortium effort that showed that a TPH2 missense variant prominently reported to be linked to depression is apparently quite rare in frequency (Zhou et al, Neuron). A DISC1 (Disrupted in Schizophrenia) haplotype linkage to schizophrenia by Peltonen et al was replicated and we proceeded to explore the role of DISC1 binding partners such as FEZ and the identity of functional loci at DISC1. COMT Val158Met & Frontal lobe function: Executive cognitive function is impaired in several psychiatric diseases: alcoholism, ADHD, & schizophrenia (SZ). Patients with SZ, & to a lesser extent their healthy siblings, have deficits in working memory & show excess cortical activity [& are thus said to be inefficient] during these tasks. We found an allele-dosage relationship of a functional COMT variant to frontal cognitive function. The relationship is seen across several populations differing in baseline cognitive function: SZ, moderate-severe head injury (Lipsky et al), & controls. Dopamine enhances prefrontal cortical efficiency. Met158, a common COMT variant, leads to four-fold reduction in COMT activity & was thus a candidate allele for frontal lobe cognitive function. We discovered a remarkable allele-dosage relationship to to measures of frontal performance, and in different datsets, and in an imaging genetic study the Val158 allele was associated with diminished frontal cortical efficiency. LNG then developed several lines of evidence that the Val158 allele has a counter-advantage, namely stress resiliency. In two populations Met158 predicted anxiety in women and decreased frontal EEG coherence, and in two populations the Met allele was associated with lower resiliency to pain/stress, and these findings included a brain imaging study in which the Met158 allele predicted an inability to activate endomorhin release after a pain stress challenge. The GABAA alpha 2 (chr 4 cluster) and alpha 6 (chr 5 cluster) recptors were linked to alcoholism. These linkages followed up our earlier whole genome scan in SW Inddians. Our alpha 2 finding replicated the haplotype found by Edenberg and Kranzler and showed the linkage may be mediated through the anxiety dimension. Extending studies in the pain/stress domain and in collaboration with investigators at NIDCR and Harvard, a functional polymorphism of GTP cyclohydrolase (a gene implicated in pain by an array expression experiment) was linked to chronic back pain (Tegeder et al, Nature Medicine). In pharmacogenetic studies, we are directly investigating the predictive power of polymorphisms for treatment and drug response, and these studies include naltrexone, acamprosate and behavioral intervention in alcoholism with the COMBINE study group, naltrexone and sertraline in alcoholism in Alaskan Natives with S. O'Malley (Yale) with at least partial confirmation of the role of OPRM1 Asn40Asp in naltrexone response. In a multicenter study on citalopram response in depression, Robert Lipsky helped find a role for 5HT2A and replicated the effect of HTTLPR via the effect of this polymorphism on treatment tolerability. In a large study of first episode schizophrenics, we helped show that a functional DRD2 promoter polymorphism influenced antipsychotic treatment response (Lencz et al, 2006).
Showing the most recent 10 out of 105 publications
