George R. Uhl - US grants

Individual differences in drug abuse vulnerabilities among humans are likely to display genetic as well as environmental components. During this year, these investigators continued to explore possible roles of allelic variants at candidate gene loci in possibly contributing to human individual differences in drug abuse vulnerability, made substantial progress with genome scanning approaches to identifying previously-unanticipated gene loci conferring drug abuse vulnerabilities, developed and validated microarray-based strategies for assessing single nucleotide polymorphisms in pooled samples, worked to overcome the limitations of the sibship collections possible in the intramural setting, and continued to made major advances in providing simulations and modeling for the power of genomme-wide and focused association/linkage-disequilibrium based genome scanning. Work during this year identified candidate regions on several human chromosomes using high-density SNP and SSLP-based linkage-disequilibrium based genome scanning. Data is consistent with the emerging models that genetic influences are polygenic with few major gene influences. Chromosome 1 and 11 regions previously nomninated in alcoholism studies may also play roles in polysubstance abuse vulnerability. This work represents one of the first high-density genome scans for genomic regions containing polysubstance abuse vulnerability alleles, and a novel approach to genome scanning for complex disorcer genetics.

There are large individual differences among humans and animals in behavioral, physiological and toxicological responses to drugs of abuse. Many of these individual differences in behavioral responses to drugs display substantial genetic components. Transgenic animals provide means for approaching several interrelated goals: 1)Ascertainment of biochemical and behavioral consequences of the introduction of or disruption of specific genes; 2)Ascertainment of the consequences of over- or under-expressing candidate genes identified in human studies; 3) Elucidation of gene elements yielding cell-type specific expression and trans-synaptic gene regulation; 4) Studying influences of interactions between variants at different genomic loci; 5)Elucidating haplotype-specific levels of expression differences in vivo, and 6) allowing us to monitor synaptic connectivities and their modulation by drug administration and alterations in haplotypes in cell adhesion and other impoprtant brain molecules. Interest mechanisms of reward, reinforcement and learning have led to continuing focus on these systems during this year, as we have established much of the groundwork for identification of wiring differences in brain slices. In continuing studies of knockouts of candidate genes that are interesting because they encode drug targets and/or are expressed in circuits of interest, we have reported novel data concerning the differential influences of these knockouts on a) adaptive responses to cocaine (neurotensin KOs) b) stress engendered by social defect (mu/OPRM1 knockouts)c) antidepressants and analgesics (OPRM1 knockouots) d) substance-induced dopamine release (mu/OPRM1 knockouts) e) brain circuitry (DAT/SLC6A3 knockouts) and f) pain responses (DAT/SLC6A4, SERT/SLC6A4 and mu/OPRM1 knockouts). We have moved forward with studies of the influences of cell adhesion gene alterations by studies of mice with altered expression of the four genes for which our human genome wide association datasests provide the msot consistent replication. We have submitted data that mice with homozygous or heterozygous deletion of the gene for which we have the strongest evidence in humans, CDH13, dramatically reduce preference for cocaine paired environments. These results display specificity, since cocaine retains locomotor stimulation and since these knockout mice can perform at normal (or supranormal) levels in Morris water maze and other comparison behaviors. Initial data supports the possibility that the magnitude of extinction and reinstatement are also reduced in CDH13 knockouts. These animals thus provide one of the best mouse confirmations of a genome wide association result from a complex, polygenic human locus.

The mu opiate receptor (OR) has been identified as the principal brain receptor site best correlated with the rewarding and euphoric properties of opiate drugs. This year investigators in this Branch have continued work with mu receptor knockout mice, reducing work in direct phosphorylation of the muOR due to personnel attenuation. Work on mu knockout mice during this year documented the lack of influence of genetic background on almost all of the reward and pain asessments made on the mixed C57/129 mice. It documented striking effects of mu knockout on ethanol consumption. The gene expression changes in these animals were compared to those produced by morphine administration, with alterations in expression of genes such as tyrosine hydroxylase found in both settings.

Activities of the principal brain vesicular monamine transporter, VMAT2, are key to understanding the cellular compartmentalization of monoamines. They may play key roles in modulating the actions and neurotoxicities induced by amphetamine and by each of the toxins that selectively kills dopaminergic neurons to provide the best current models of Parkinson's disease. They could even play roles in normal age-related alterations in these systems. In this year, workers in this Branch continued to describe the poperties of knockout mice with deletions of the VMAT2 gene and other plasma membrane monaamine transporters. Aging studies documented clear reductions in locomotion, in amphetamine responsiveness and exaggerated age-related losses of dopaminergic markers in the heterozygous VMAT2 knockouts. Mice with deletions of VMAT2 and the plasma membrane transporters for DAT and SERT are viable and fertile and may display altered amphetamine responses. VMAT2 knockout mice continue to substantially enhance our understanding of mechanisms of age- related alterations in dopaminergic systems and actions of psychostimulants and locomotor systems. Studies completed suring this year also defined the two common haplotypes in the human VMAT2 gene, the patterns of imprinting through which they are inherited in humans, and in vitro expression patterns that provided more than 20-25% differences in cell lines expressing luciferase fusion constructions made with the two alternative human VMAT2 promoter regions.

Addiction vulnerability genes include those that are likely to harbor allelic variants that contribute to human individual differences in vulnerability to addictions. During this and prior years, we identified dozens of candidates to play such roles based on the covergences between nominally-positive data derived from up to fourteen separate abuser/control or quit success comparisons. During the current year we have reported increasing support from outside samples/datasets for the genes that we have identified most consistently in our own data. We added to evidence supporting roles for many of these genes with novel data concerning rate of uptake of developmental use of addictive substances. Genes identified in this fashion include a disproportionate number of genes whose products are involved in cell adhesion molecule actions. During this year we identified additional functional variation at two loci that contain these genes, focusing on extended haplotypes that alters level of expression for CDH13 and PTPRD. These data provide the basis for animal models that replicate, findings in humans and predict stimulant dose-response relationship effects that were recently reported in humans at these two loci. During this year, we reported association between a specific addiction vulnerability phenotype, preference for mentholated cigarettes, and haplotypes in the TRPA1 channel that serves as a menthol receptor in the few African American subjects who prefere nonmentholated cigarettes. D

The dopamine transporter (DAT) has been a principal brain receptor site that has been correlated with the rewarding and euphoric properties of cocaine. MNB scientists cloned the DAT cDNA and gene, and found that deletion of both DAT and SERT are required to eliminate cocaine conditioned place preferences in mice. DAT is required for the actions of each of the current dopamine-selective toxins that produce the best models of Parkinsons disease. Analyses of the relationships between DAT expression and a number of pharmacological endpoints continued through this year, with emphasis on relationships with serotonin systems defined through the serotonin receptor subtypes. During this year, we reported detailed studies of effects of DAT deletion by itself or in combination with deletion of other sites on methamphetamine-induced locomotor activity and sensitization, methamphetamine-induced hyperthermia and lethal toxicity, Expression of pharmacologically-defined substates of DRD2 receptors. Effects of VMAT2 deletion on DAT expression were also reported. Work continued on double and ttriple knockout mice with alterations of DAT, 5HT1A and 5HT1B receptors. These insights should continue to help in identification of relationships between DAT gene expression and regulation, human individual differences in DAT levels of expression and a number of interesting features of stimulant pharmacology.

During this year, we reported (accepted for publication) genome wide association for smoking cessation success in data from an additional clincal trial. Smoking quit success genes identified in this fashion include genes that we have identified in studies of addiction vulnerability. A sizable number of these genes have products that are involved in cell adhesion molecule actions. Several are likely to be involved in learning and memory mechanisms. Several are implicated in indivdiual differences in regional brain volumes. During this year, we published a novel synthesis of animal models for smoking cessatin focused on mouse models. We have applied many of these tests to an animal model for one of the genes implicated from human studies, CDH13. Extinction/reinstatement of cocaine conditioned place preferences display trends toward difference in CDH13 KO mice; we are currently working with nicotine CPP in these animals. During this year, we had (accepted for publication) work that links, for the first time, individual differnces in abilities to quit smoking with individual differences in uptake of use of addictive substances, including cigaretes. During this year, we have also identified nominally-significant associations between TRPA1 haplotypes and success in smoking cessation. This data, motivated by our success in studying TRPA1 haplotype assocaitions with menthol cigarette preference, supports roles for noxious cigarette smoke influences in cessation success.

Summary/abstract: There are no FDA-approved medications for stimulant use disorders. Therapies for opiate use disorders remain suboptimal in ways that are now a focus of national attention. PTPRD (receptor type protein tyrosine phosphatase D) is now supported as a target for development of novel drugs to treat stimulant use disorders by genetic, molecular biologic and pharmacologic evidence in humans and mouse models. Data include: a) multiply-replicated human PTPRD genetic associations (clustered SNPs; 10-8 < p < 10-2) with addiction phenotypes including DSM dependence on opiates and stimulants; b) PTPRD SNP associations with levels of PTPRD expression in postmortem human brain; c) reduced cocaine conditioned place preference (CPP) and cocaine self-administration in mice with 50% reductions in PTPRD expression; d) little human or mouse model evidence for toxicity from reduced PTPRD expression; e) identification of 7-BIA (7-butoxyilludalic acid analog) as a PTPRD ligand that inhibits PTPRD's phosphatase; f) no identified 7-BIA toxicity; g) observations that 7-BIA attenuates both cocaine CPP and well-established cocaine self-administration. Human data supports PTPRD associations with vulnerability to opiate use disorders, though there is presently only preliminary mouse model/pharmacologic support for this association. These exciting results strongly support development of therapeutic PTPRD ligands for stimulant use disorders and studies to seek benefits for opiates. We will thus improve PTPRD ligands, identify effects on opiates and move the best novel, patentable PTPRD ligands toward human studies via i) synthesis of novel PTPRD ligand candidates; ii) testing in vitro activities at PTPRD, related phosphatases and off-target effects. We will seek a) improved potency, b) greater selectivity/few ?off target? effects, c) improved solubility, d) improved stability; e) predicted good half-life; f) predicted bioavailability after oral administration and g) predicted ability to cross the blood brain barrier. Selected compounds will be screened in vivo for toxicities from acute then chronic dosing in mice. Improved compounds will be tested in cocaine conditioned place preference and self-administration assays. In tests of opiate reward, we will examine effects of 7-BIA, improved compounds and heterozygous PTPRD knockout. The best compounds active in CPP/self-administration will be tested for biodistributions/ metabolism/stability, aversive or rewarding effects of their own, motor/memory/sensory effects and mouse/rat toxicity and preIND studies. We will thus generate novel, well tolerated and bioavailable PTPRD ligands that display in vitro potency, selectivity and stability, in vivo modulation of cocaine reward and, likely, modulation of opiate reward at doses that provide no significant toxicity. Grand Opportunity support will provide compounds and data to help underpin IND submission, intellectual property protection, pharmaceutical partnerships and progress from novel lead compound to exciting new anti-addiction drug candidates.

Project Summary/Abstract Alzheimer's disease (AD) receives pathogenic contributions from genetics [1, 2] and from environmental influences that include dietary intake of flavanols > flavones [4-8]. Neurofibrillary tangles (NFTs) rich in hyperphosphorylated tau protein [10] are prominent features of AD neuropathology. NFT densities correlate well with the degree of AD dementia [15] [18] and are influenced by variation in the ApoE and PTPRD genes [3]. One approach to altering tau/NFT pathophysiology is to reduce activities of the kinases that hyper- phosphorylate tau. The glycogen synthase kinases GSK3? and GSK3? are prominent tau phosphorylators [19]. GSK3? and GSK3? are activated by phosphorylation of their own tyrosines (pY279 and pY216) by known tyrosine kinases [20] [21, 22]. Increasing activity of tyrosine phosphatase(s) that dephosphorylate and reduce activities of brain GSK3? and GSK3? thus provides a novel approach to reducing tau pathology in AD. Evidence (much developed with support from our first NIA supplement) now supports roles for: a) the receptor type protein tyrosine phosphatase PTPRD as both a i) key physiological phosphatase for phospho (pY) GSK3? and GSK3? and ii) novel target for decreasing pathological AD tau hyper- phosphorylation and b) flavanols as lead compound PTPRD positive allosteric modulators (PAMs) that increase this desired PTPRD activity. We will enhance this evidence and move toward translation by testing hypotheses that a) increased PTPRD dephosphorylation of GSK3? and GSK3? reduces the activities of these tau-hyperphosphorylating kinases with specificity, underlying PTPRD's genetic associations with NFT densities in AD brains and b) flavanols whose intake reduces AD incidence in aging [4-8] serve as PAMs for PTPRD's phosphatase, increase GSK3 dephosphorylation with specificity and provide a pathway for development of improved, specific PTPRD PAMs that can reduce progression to AD deficits during aging. We will test these hypothesis and support development and translation of PTPRD PAMs in several ways: 1) we will characterize the specificity of PTPRD effects by comparing i: rates of PTPRD dephosphorylation of pYGSK3? and pYGSK3? and ii: quercetin effects on these rates vs those for each of > 80 candidate neuronal PTPRD substrates. 3) We will synthesize and test novel flavanol analogs as improved PTPRD PAMs, nominating novel structures by in silico docking to the PAM vs catalytic sites on PTPRD's phosphatase, testing these structures in vitro, refining our in silico models and nominating/synthesizing/testing new structures on the basis of these results. For the best candidate positive allosteric modulators, we will test specificities vs other PTPRD substrate phosphopeptides and off-target sites of action of currently marketed drugs. We will test the most promising PTPRD PAMs in vivo for gross, histological or behavioral toxicities, biodistribution (including brain) and target engagement. We will expand validation of quercetin effects in aging 3xTg- AD mice and initiate studies to test novel PTPRD PAM(s) in this model. This work will advance our understanding of AD pathophysiology, validate novel approaches to PTPRD positive allosteric modulation and provide a basis for development of interventions that can prevent and/or treat key aspects of AD pathophysiology.