Study 03 – Molecular Genetics of Heroin Dependence in China

Investigators:  Ming Tsuang, Steve Faraone

Release Date:  Available at NIDA and dbGaP


Although the available data strongly suggest that a substantial component of the etiology of drug dependence is mediated by gene expression in the central nervous system (Tsuang et al., 1996; Tsuang et al., 1998), a detailed understanding of the mechanisms involved has remained elusive.  Genetic studies have not been able to determine the mode of inheritance.  The main goal of the proposed program of research is to detect one or more genes responsible for the genetic transmission of heroin dependence.  We have selected heroin dependence because our previous research (Tsuang et al.,1998) has demonstrated that heroin is the most “heritable” form of illicit drug dependence and that heroin has the greatest degree of genetic variance not shared with other drugs (see Preliminary Studies).  Merikangas et. al. (1998) have also adduced evidence for familial specificity of heroin abuse.  Our principal goal is to determine the chromosomal locations of genetic variations associated with increased vulnerability to heroin dependence.  This goal will be attained by identifying specific DNA markers that are genetically linked to heroin dependence.

Specific Aims: 1) To collect and clinically characterize a large sib-pair sample with adequate statistical power for identifying genomic regions that may harbor loci conferring susceptibility to heroin dependence; 2)  To conduct a whole-genome scan to establish the chromosomal localization of such loci;  3) To follow-up regions of interest from the whole-genome scan and evaluate candidate genes; and 4.)  To make the biological, clinical and genotypic data quickly available to other investigators in the scientific community.

We will attain these aims by achieving the following objectives: 1)  From ascertainment sites in Yunnan Province, China we will collect blood and diagnostic information from 1000 sib-pairs having DSM-IV defined heroin dependence as well as from their parents and other affected and unaffected siblings; 2) We will complete a genome scan using 350 markers spaced at an average of 10 cM intervals; 3)  We will follow-up regions of interest with a denser set of markers and evaluate candidate genes; and 4.)  All clinical data will be made available to the scientific community by the end of the funding period.  All genotypes will be available one year after they are generated, but no later than a year after the end of the funding period.


The minimum pedigree ascertained will consist of an affected sibling pairs (ASP)  and two parents who will provide a blood sample.  We will exclude ASPs if both parents have heroin dependence.  We expect this to be rare, but we exclude such families because the ascertainment of ASPs increases the chance of parental homozygosity and, consequently, of uninformative families.  Excluding bilineal matings protects against that situation.  When two parents are available, the unaffected siblings are not needed for the linkage analyses of DSM-IV defined heroin dependence.  We will, however, collect unaffected siblings for several reasons.  First, many of these siblings may have heroin dependence-related phenotypes that will be useful for exploratory linkage analyses as described in our data analysis section.  Second, the unaffected siblings will be useful for analyses that we or other users of our data may find useful.  Examples include the linkage analysis of discordant sib-pairs (Risch et al., 1995) and the use of unaffected siblings for studying quantitative traits.  In cases in which it is impossible to obtain genotypic data on both parents, unaffected sibs may be useful for establishing identity by state.  Our preference will be for families in which both parents can provide DNA.  However, families with one available parent and at least two unaffected siblings, in addition to the affected sib-pair, will be included.  We estimate that there will be an average of one sibling per family in addition to the affected sib-pair.  Larger pedigrees may be collected if available.


Following the requirements of NIDA’s RFA, we will use DSM-IV diagnostic criteria to ascertain pedigrees.  Thus, each family must contain at least two siblings meeting DSM-IV criteria for heroin dependence.  Subjects will be eligible based on either a current or lifetime diagnosis of heroin dependence.  Siblings, other than those comprising the ascertained affected sib-pair, will be classified as: 1) “affected” if they meet DSM-IV diagnostic criteria for heroin dependence; 2) “unaffected” if they have never been dependent on heroin and are past the age of 24 {the majority of subjects of subjects with heroin dependence developed dependence between the ages of 16 and 24 (Wu et al., 1996)}; or 3) “phenotype unknown” if they have never met criteria for heroin dependence and are under the age of 24.

The Diagnostic Interview for Genetic Studies (DIGS): All subjects will be personally interviewed with the DIGS, which was created by two of the investigators (Tsuang and Faraone) and their colleagues from the NIMH Human Genetics Initiative. This interview generates diagnoses according to several diagnostic systems: the Research Diagnostic Criteria (RDC), Modified RDC, DSM-III, DSM-III-R, DSM-IV, Washington University Criteria and the International Classification of Diseases, 10th Revision (ICD-10).  Although we will be using DSM-IV diagnoses for our analyses, the availability of a polydiagnostic assessment will be very useful when other researchers utilize the data.  The DIGS makes a detailed assessment of the course of illness, and makes a careful assessment of substance dependence.  This detailed assessment helps protect against false positive diagnoses.  We view the use of the DIGS as ideal for two reasons: 1)The PI and a co-investigator (Faraone) have extensive experience using the DIGS in genetic linkage studies and in training personnel to use these instruments and 2) the interview was translated into Mandarin and has been used in genetic linkage research.


Clinical experience demonstrates that individuals are differentially vulnerable to the initiation, persistence, severity, and cessation of substance abuse.  Not everyone has the opportunity to use an addictive substance, not everyone who has the opportunity uses it, not everyone who uses an addictive substance becomes addicted, and not everyone who becomes addicted remains addicted for life.  Variability is also seen in reports of symptoms of drug dependence, even among regular users. For example, forty-two percent of regular cocaine users do not report symptoms of cocaine dependence. Observations such as these suggest that both genetic and environmental influences, operating via behaviorally complex pathways, differentially predispose individuals to drug-taking behavior, drug abuse, and dependence.  Genetic effects on drug use and dependence might operate at a variety of levels.  Genetic influences that contribute to the initiation of drug use may differ from those which contribute to heavier drug use or dependence.  Drug use vulnerability might also be modified by protective factors that contribute to drug abstinence or protect from development of regular use patterns or drug dependence.  Allelic variants of specific genes might mediate differential drug reinforcing properties, alter drug pharmacodynamics or pharmacokinetics, influence personality traits (e.g., “sensation seeking”) which may facilitate exposure to drugs, exacerbate drug toxicities, or minimize “protective” factors such as hangovers. 


Heroin is perhaps the most difficult substance to study among human populations because of the low prevalence of heroin abuse/dependence in the community and the frequent polysubstance abuse observed in heroin abusers (Schuckit, 1994).  There are some indications that genetic determinants influence twins’ reaction to pain and its perceived alleviation by morphine (Liston, 1981). Outside of this report, one by Merikangas et al. (1998) described below, and our data indicating significant genetic influence on heroin dependence (Tsuang et al., 1996; 1998), the literature in this area is sparse.  A recent report by Kotler et al. (1997) was the first to find an association between a specific genetic polymorphism and opioid addiction. Kotler et al. found an association between the personality trait of novelty seeking and the long alleles of the D4 dopamine receptor (D4DR) exon III polymorphism. With that finding in hand, they investigated the prevalence of this D4DR polymorphism among a group of heroin addicts and found the polymorphism to be significantly over-represented in the addict sample compared with a control sample (Kotler, 1997).

Merikangas and her colleagues (1998) published results from a study of familial transmission of substance use disorders.  Their results demonstrated a significantly elevated risk of drug dependence among the relatives of probands with dependence on a range of illicit drugs.  Moreover, their data also provide support for a meaningful degree of specificity of the type of drug abused by the proband and the type of drug abused by the relative.  Merikangas et. al’s study indicated that the relatives of opiate abusers have a risk of opiate abuse that is 25.5 times higher than the risk of opiate abuse among the relatives of controls, yielding a lambda statistic of 25.5 (10.2/0.4).  The adjusted odds ratios indicated that opiates exhibit the highest degree of specificity of any drug of abuse examined.  Recent data from our Harvard Twin Study of Drug Abuse and Dependence indicated that heroin dependence is the most heritable type of drug dependence and that there is greater genetic specificity for heroin than for any other drug.

We found evidence for a shared or common vulnerability factor that underlies the dependence of marijuana, sedatives, stimulants, heroin/opiates, and psychedelics.  This shared vulnerability is influenced by genetic, family environmental, and non-family environmental factors, but not every drug is influenced to the same extent by the shared vulnerability factor.  Marijuana, more than other drugs, was influenced by family environmental factors.  Each drug, except psychedelics, had genetic influences unique to itself (i.e., not shared with other drugs).  Heroin abuse had the largest amount of unique genetic variance (70%) and the least amount of shared genetic variance (30%) of any of the drugs.  In contrast, abuse of psychedelic drugs shared 100% of its genetic variance with the common vulnerability.  For four of the five categories of drug abuse, genetic influences are responsible for between 26% and 33% of the variance.  Heroin abuse is the one exception to this pattern with  54% of the variance being due to genetic factors.  Heroin abuse had the largest amount of unique genetic variance (38%) and the least amount of shared genetic variance (16%) of any of the drugs.   Therefore, it is for heroin abuse that it is most likely that a genetically influenced characteristic of the nervous system unique to the mechanism of action of a single drug is important.  However, different drugs may have specific receptors in the brain associated with their mechanism of action, but further “downstream” in the brain, their effects might converge.

Table 2: Proportions of variance in drug abuse variables from multivariate biometrical modeling under the latent phenotype model.1

 Total Genetic VarianceSpecific Genetic VarianceTotal Family Environ- mental VarianceSpecific Family Environ- mental VarianceTotal Non-family Environmental VarianceSpecific Non-family Environ- mental Variance

1 Parameter estimates derived from “Common Vulnerability” model. 


Cox NJ, Using interaction between genes in mapping genes for complex traits, Invited Speaker , International Genetic Epidemiology Society, 7th annual meeting 1998

Daly MJ, Kruglyak L, Pratt S, Houstis N, Reeve MP, Kirby A, Lander ES, GENEHUNTER 2.0 – a complete linkage analysis system . Am J Hum Genet  1998 63: A286

Goring HHH, Ott J, Relationship estimation in affected sib pair analyses of late-onset diseases. European Journal of Human Genetics, 1997;5:69-77.

Kong A, Cox NJ, Allele-sharing models: LOD scores and accurate linkage tests. Am J Hum Genet, 1997 61:1179-1188.

Kotler M, Cohen H, Segman R, Gritsenko I, Nemanov L, Lerer B, Kramer I, Zer-Zion M, Kletz I, Ebstein RP.  Excess dopamine D4 receptor (D4DR) exon III seven repeat allele in opioid-dependent subjects. Molecular Psychiatry, 1997; 2(3):251-254.

Kruglyak L, Lander ES, Complete multipoint sib-pair analysis of qualitative and quantitative traits. Am J Hum Genet 1995 57:439-454

Kruglyak L, Daly MJ, Reeve-Daly MP, Lander ES, Parametric and nonparametric linkage analysis: a unified multipoint approach. Am J Hum Genet, 1996; 58:1347-1363.

Lander E, Green P, Construction of multilocus genetic linkage maps in humans. PNAS, 1987; 84: 2363-2367.

Lander E, Kruglyak L.  Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results.  Nat Genet, 1995; 11:241-247.

Liston EH, Simpson JH, Jarvik LF, Guthrie D. Morphine and experimental pain in identical twins. Prog Clin Biol Res, 1981;69 Pt C:105-116.

Merikangas KR, Stolar M, Stevens DE, Goulet J, Preisig MA, Fenton B, Zhang H, O’Malley SS, Rounsaville BJ. Familial transmission of substance use disorders.  Arch Gen Psychiatry, 1998;55:973-979.

Tsuang, M.T., Lyons, M.J., Eisen, S.A., Goldberg, J., True, W., Meyer, J., & Eaves, L. Genetic Influences on Abuse of Illicit Drugs: A Study of 3,372 Twin Pairs American Journal of Medical Genetics (Neuropsychiatric Genetics), 1996;5, 473-477.

Tsuang MT, Lyons MJ, Meyer JM, Doyle TJ, Eisen SA, Goldberg J, True W, Lin N, Toomey R, Eaves L. Co-Occurrence of Abuse of Different Drugs in Men: The Role of Drug-Specific and Shared Vulnerabilities, Arch Gen Psychiatry, 1998; 55:967-972.

Whittemore AS, Halpern J, A class of tests for linkage using affected pedigree members. Biometrics,  1994, 50:118-127.

Wu, W., Holmans, P., Wavrant-DeVrieze, F., Shears, S., Kehoe, P., Crook, R., Booth, J., et al.  A. Failure to confirm a locus at chromosome 12p11-12 in late onset Alzheimer’s disease. JAMA, 1998; 280:619-622.


Diagnostic Interview for Genetic Studies (DIGS)


  • DIGS 2.0
  • DIGS Training Manual
  • Instrument Variables

Best Estimate Diagnoses

  • DSM-III-R Codes
  • DSM-IV Codes