Genetic Factors and the Directionality of Comorbid Disorders University of Pittsburgh Michael Vanyukov Supported by NIDA grants R01DA011922, P50DA005605, R01DA019157, K02DA

Getting to the root(s) “The active ingredient in Roundup moves through the weed to kill the root.”

Phenotyping SUD Dx of substance dependence:  3 out of nine symptoms  466 combinations Arbitrary phenotypes—ICD-m, DSM-n Staging—opportunistic Mixed effects: stimulants & depressants And a little bit of this'd get you up And a little bit of that'd get you down Mark Knopfler “Junkie Doll”

Liability, the concept “...to express not only the individual innate tendency to develop or contract the disease, i.e., his susceptibility in the usual sense, but also the whole combination of external circumstances that make him more or less likely to develop the disease...” Falconer (1965) Falconer (1965)

Liability distribution Liability Affected threshold phenotype

Shared liability variance Tsuang et al., Harvard Rev Psychiatry, 9: , 2001

More sharing Kendler et al., Am J Psychiatry 160: 687–695, Cannabis Cocaine Halluc Sedat Stimul Opiate AcAcAcAc S c /U c 0.01/ / / / / /0.64

Copyright restrictions may apply. Kendler, K. S. et al. Arch Gen Psychiatry 2003;60: Psychiatric and drug abuse disorders

Common sources of variation Pre-use Temperament Temperament Personality Personality Cognition Cognition Behavior Behavior Self-medication Self-medication Post-use Positive reinforcement Positive reinforcement Negative reinforcement Negative reinforcement Biobehavioral self-regulation

Prefrontalcortex Nucleusaccumbens Substantia nigra Hippocampus Locus coeruleus Cerebellum Ventral tegmental area Colliculus Periaqueductal gray gray area area ALCOHOL OPIATES COCAINE & AMPHETAMINES Striatum Dopamine pathway Amygdala The drug-activated mesocorticolimbic dopamine pathway

Sources of SUD risk variation

Comorbidity models Chance, bias, stratification Chance, bias, stratification Alternate forms (single liability) Alternate forms (single liability) Multiformity (having one increases probability of another, threshold-dependent) Multiformity (having one increases probability of another, threshold-dependent) Three independent Three independent Correlated liabilities (threshold-independent) Correlated liabilities (threshold-independent) –correlated –one causes another –reciprocal causation

Subsample of models Rhee et al. J Abnorm Psychol, 114: , 2005

Genetics in SUD risk

It’s a long way …

Dynamic liability: Tracking etiology and comorbidity of SUD Risk Drug useSUD Time 0 Initial liability SUDthreshold Liability Age R Time 2 Time 1 R R Epigenetic trajectory

Genetics & comorbidity Behavior genetics Behavior genetics –genetic and environmental correlations –developmental tracking Molecular genetics Molecular genetics –association/linkage –mediation by intermediate traits “Hybrids” “Hybrids”

Family history Faster physiological maturation Faster physiological maturation Detachment from parents Detachment from parents Homophilic peer selection/contagion Homophilic peer selection/contagion Dysregulation/disinhibition Dysregulation/disinhibition Maladjustment Maladjustment SU/SUD SU/SUD

LFLF 11 22 33 Transmissible causal variables (e.g., genetic polymorphisms, neurochemical processes, transmissible environment) 44 LPLP nontransmissible environment Psychological variables L P – parental liability L F – son’s liability Determination of SUD liability

DiagnosisHR95% CIP ADHD CD ODD DBD and the rate of SUD dysregulation

transmissible causal variables (eg, genetic polymorphisms, neurochemical processes, transmissible environment) L P – parental liability L F – son’s liability Measurement of SUD liability LFLF LPLP nontransmissible environment SUDLI indicator variables (items) y4y4 y3y3 y1y1 y2y2 y5y5 y6y6 y7y7 y8y8 y9y9 y 10 y 11 y 12 Psychological constructs 11 22 33 44

DRD4-ERP-Disinhibition Family-based analysis (FBAT) Family-based analysis (FBAT) Promoter region (-521) SNP Promoter region (-521) SNP ERP—ND: p=.02 ERP—ND: p=.02 SNP—>P300: p=.004 SNP—>P300: p=.004 SNP—>ND: p=.003 SNP—>ND: p=.003 SNP—>ND|P300: p=.85 SNP—>ND|P300: p=.85

Parenting & MAOA Vanyukov et al. (in press) Psychiatric Genetics

MAOA & SUD Low activity increase in the risk increase in the risk limbic volume reductions, hyperresponsive amygdala during emotional arousal, diminished reactivity of regulatory prefrontal regions (Meyer-Lindenberg et al., 2006) limbic volume reductions, hyperresponsive amygdala during emotional arousal, diminished reactivity of regulatory prefrontal regions (Meyer-Lindenberg et al., 2006)

Drug use severity Drug Use Screening Inventory substance use problems scale 94 male Caucasian adolescents (12-18 years of age) with a DSM-IV diagnosis of Substance Dependence related to illicit drugs Polymorphisms in DRD1-DRD5 genes Two-way ANOVA, testing main effects and interactions between D1- and D2-family genes Trait Sample Genes Analysis

Interactions as expected

Candidate system genes Comorbidity helps: multiple hits cross-verify identify pathways support the gene’s role

c3 GcGcGcGc h3 e3 E c1 E c1 GpGpGpGp h e EpEpEpEp p E c3 E c3 E c2 E c2 c2 h2 e2 c1 h1 e1 T1T2T3 DynamicDynamic frozen birth time Note: G×E omitted

Genetic roots … environmental “Roundup”?