- Email: [email protected]
Epistatic effects between variants of kappa-opioid receptor gene and A118G of mu-opioid receptor gene increase susceptibility to addiction in Indian population
Progress in Neuro-Psychopharmacology & Biological Psychiatry 36 (2012) 225–230
Contents lists available at SciVerse ScienceDirect
Progress in Neuro-Psychopharmacology & Biological Psychiatry journal homepage: www.elsevier.com/locate/pnp
Epistatic effects between variants of kappa-opioid receptor gene and A118G of mu-opioid receptor gene increase susceptibility to addiction in Indian population Deepak Kumar a, Japashish Chakraborty b, Sumantra Das a,⁎ a b
Neurobiology Division, Indian Institute of Chemical Biology, 4 Raja S. C. Mullick Road, Jadavpur, Kolkata 700032, India Baulmon, 34 Jadavpur Central Road, Kolkata 700032, India
a r t i c l e
i n f o
Article history: Received 5 July 2011 Received in revised form 1 October 2011 Accepted 31 October 2011 Available online 28 November 2011 Keywords: Addiction Alcohol Heroin Kappa opioid receptor Single nucleotide polymorphism
a b s t r a c t Objective: Unequivocal evidence suggests contribution of κ-opioid receptor (KOR) in addiction to drugs of abuse. A study was undertaken to identify the single nucleotide polymorphisms (SNP) at selective areas of kappa opioid receptor 1 (OPRK1) gene in heroin as well as in alcohol addicts and to compare them with that in control population. The potential interaction of the identified KOR SNPs with A118G of μ opioid receptor was also investigated. Methods: Two hundred control subjects, one hundred thirty heroin and one hundred ten alcohol addicts, all male and residing in Kolkata, a city in eastern India, volunteered for the study. Exons 3 and 4 of OPRK1 and the SNP, A118G of mu opioid receptor 1 (OPRM1) in the DNA samples were genotyped by sequencing and restriction fragment length polymorphism respectively. The SNPs identified in the population were analyzed by odds ratio and its corresponding 95% confidence interval was estimated using logistic regression models. SNP–SNP interactions were also investigated. Results: Three SNPs of OPRK1, rs16918875, rs702764 and rs963549, were identified in the population, none of which showed significant association with addiction. On the other hand, significant association was observed for A118G with heroin addiction (χ2 = 7.268, P = 0.0264) as well as with alcoholic addition (χ2 = 6.626, P = 0.0364). A potential SNP–SNP interaction showed that the odds of being addicted was 2.51 fold in heroin subjects [CI (95%) = 1.1524 to 5.4947, P = 0.0206] and 2.31 fold in alcoholics [CI (95%) = 1.025 to 5.24, P = 0.0433] with the OPRK1 (rs16918875) and A118G risk alleles than without either. A significant interaction was also identified between GG/AG of A118G and GG of rs702764 [O.R (95%) = 2.04 (1.279 to 3.287), P = 0.0029] in case of opioid population. Conclusion: Our study suggests that set associations of polymorphisms may be important in determining the risk profile for complex diseases such as addiction. © 2011 Elsevier Inc. All rights reserved.
1. Introduction There has been an increasing interest in the understanding of the role of κ-opioid receptors (KOR) in addiction to drugs of abuse. Although the μ-opioid receptor (MOR) mediates most of the opioid actions, including analgesia, tolerance and reward (Matthes et al., 1996; Reisine and Pasternak, 1996; Zadina et al., 1997), KOR has been receiving much attention because of its functional interaction with the MOR. In animal models, an extensive body of pharmacological data demonstrates that the KOR opposes a variety of MOR-mediated actions like analgesia (Friedman et al., 1981; Ramarao et al., 1988;
Abbreviations: OPRK1, kappa opioid receptor; OPRM1, mu opioid receptor; CI, confidence interval; LD, Linkage disequilibrium; OR, Odds ratios; SNP, single nucleotide polymorphism; KOR, κ-opioid receptors; MOR, μ-opioid receptor. ⁎ Corresponding author at: Neurobiology Division, Indian Institute of Chemical Biology, 4, Raja S.C. Mullick Road, Jadavpur, Kolkata 700032, India. Tel.: + 91 33247 35197/ 24736793/24724049; fax: + 91 33247 35197/24730284. E-mail addresses: [email protected], [email protected] (S. Das). 0278-5846/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.pnpbp.2011.10.018
Sora et al., 1997; Tao et al., 1994; Tulunay et al., 1981), tolerance (Bhargava, 1994; Rothman, 1992), dependence (Suzuki et al., 1992), withdrawal (Cowan et al., 1998) as well as in the reinforcing effects of morphine in both self-administration (Glick et al., 1995; Kuzmin et al., 1997) and conditioned place-preference paradigms (Bolanos et al., 1996; Funada et al., 1993). Like MOR, the KOR is localized in several areas of the dopaminergic nigrostriatal and mesolimbic– mesocortical systems which are the sites of known actions of drugs of abuse (Zhang et al., 2004). However, unlike MOR ligands, the endogenous KOR agonists, dynorphins, decrease basal and drug induced dopamine levels in several areas of dopaminergic nigrostriatal and mesolimbic–mesocortical system and inhibit morphine-withdrawal symptoms induced by naloxone precipitation or morphine discontinuation in morphine dependent animals (Suzuki et al., 1992). The KOR– dynorphin system may therefore be considered to be a part of the counter-modulatory mechanisms of the brain and studies suggest that KOR agonists may be potential therapeutic agents for treatment of drug addiction and disorders that result from alterations in mesolimbic dopamine transmission (Shippenberg and Rea, 1997).
226
D. Kumar et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 36 (2012) 225–230
Considering that several genetic studies using family, twin and linkage and association designs have consistently demonstrated genetics factors as contributing to vulnerability to drug addiction (Bierut et al., 1988; Xuei et al., 2007), there have been attempts to identify mutations in the KOR gene as possible risk factors for the development of addiction. KOR gene polymorphisms have been reported to contribute to predisposition to voluntary alcohol-drinking behavior in experimental animals (Saito et al., 2003). In humans, the 36G>T SNP at exon 1 of the KOR gene (OPRK1), which results in a silent mutation, has been shown to exhibit a significant association with a population of heroin addicts of West European, Caucasian origin (Gerra et al., 2007) as well of African-American, Caucasian and Hispanic origin (Yuferov et al., 2004). However, contrasting reports excluding a possible role of the opioid receptor genes for the proclivity to dependence (Loh el et al., 2004) in Taiwanese population, suggest influence of the different racial or ethnic stratification of the samples. The human OPRK1 gene is located on chromosome 8q11.2. The human OPRK1 gene has four exons and three introns and untranslated (UTR) regions of 660 and 3576 nucleotides respectively at the 5′ and 3′ ends. Altogether about 24 number of SNPs of KOR gene has so far been investigated (Fig. 1). Although some studies have explored association of some of these SNPs with drug dependence, no definite identification of risk allele has been detected (Yuferov et al., 2004; Edenberg et al., 2008). Our main objective was to investigate the contribution of OPRK1 gene polymorphisms to addiction in exonic regions 3 and 4 of both alcohol as well heroin addicts and compared with control subjects. Three known SNPs in exonic region 4 were identified in the population. Since in many digenic disorders, gene–gene polymorphic interactions can influence the risk of any complex disease, we studied locus–locus interaction between OPRK1 and OPRM1. Considering the complex nature of addiction, SNP–SNP interaction studies are expected to provide useful information on the inter-dependence of specific variants of genes in contributing towards the development of diseases. For our studies, we investigated the interaction of the SNP, A118G of OPRM1 with the SNPs of OPRK1 identified in our study. A118G was considered because it has been previously shown to be highly associated with addiction in the population studied (Deb et al., 2010).
2. Materials and methods 2.1. Subjects One hundred thirty heroin addicts, one hundred ten alcoholic and two hundred non-addict or control individuals participated in this
study and provided written informed consent. Volunteers were unrelated male of mean age of 39.2 years. All subjects, both cases and controls, were ethnically Indians belonging to the Kayastha and Brahmin castes of the Bengali-Hindu ethnic background and were recruited from the city of Kolkata. Ethnicities of subjects were determined with the help of physicians' admission notes and the hospital data, which was based on information regarding family origin, ethnic background including geographical and place of birth. They were all unrelated male from various backgrounds like students, workers and general population of different income groups. Addicted patients were under treatment at Baulmon, a psychiatric institute situated in Kolkata. The institute was selected to carry out this study because of their capacity to administer structured DSM interview, to produce a diagnostic evaluation and to use routinely the psychometric instruments. Demographic description of the cases is provided in Table 1. Subjects were examined by a psychiatrist specializing in substance abuse disorders and interviewed using the Addiction Severity Index (ASI) questionnaire that covers medical, employment, drug, legal, family history, family and social relationships and psychiatric status. All heroin as well as alcohol-dependent subjects conformed to the criteria for dependency to drugs of abuse as defined by DSM IV (Diagnostic and Statistical Manual of Mental Disorders, 4th edition). Individuals with multidrug dependency or comorbidity to heroin and alcohol were excluded from this study. All the patients, meeting the criteria for DSM IV opioid dependence, submitted positive urine for morphine metabolites, before the beginning of the treatment. Only cases where the patients were dependent upon drugs for 5 years or more with history of at least one time relapse were selected for the study. Control subjects were healthy volunteers, all males having similar caste and ethnic background and matched for age. They were also interviewed and checked for history and family history for addiction. All the subjects included in this study were recruited at the same time period. The study had the approval of the institutional ethical committee for protection of research risks to humans.
2.2. Selection of SNPs While the 36G>T SNP at exon 1 of OPRK1 has been linked with addiction (Gerra et al., 2007; Loh el et al., 2004; Xuei et al., 2006; Yuferov et al., 2004), initial studies with a small population of subjects from both control and cases could not detect the presence of 36G>T SNP in the Indian population. In present study we selected exonic regions 3 and 4 of OPRK1 which contain functional receptor domains which are important for pharmacological treatment. To investigate the effect of gene–gene
Fig. 1. Locations of SNPs in OPRK1 gene drawn on the basis of published studies (Edenberg et al., 2008; Yuferov et al., 2004) as well as from descriptions in Vega Genome Browser (http://vega.sanger.ac.uk/Homo_sapiens/Transcript/Summary?db=core;g=OTTHUMG00000164276;r=8:54138284–54164257;t=OTTHUMT00000378048). Exons, introns and UTRs are not drawn to scale. *Indicates SNPs found to be present in the population studied. Numbers within parenthesis indicate frequency of occurrence as reported in dbSNP.
D. Kumar et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 36 (2012) 225–230 Table 1 Demographic and clinical characteristics of heroin and alcohol dependent patients. Characteristic
Heroin addict
Alcohol addict
Age Number of individuals Occupation (%) Employed Unemployed Marital status (%) Married Unmarried History of addiction (mean years) Average times of relapse Average daily intake of Heroin (g) Average daily ethanol intake (g/week) Route of heroin administration Smoked Intravenous Personal psychiatric disorders (%) a) Borderline personality traits b) Passive aggressive traits c) Antisocial personality traits d) Obsessive compulsive traits e) Hostile and aggressivea Personal non-psychiatric disorders (%) e.g. diabetes, gastric ulcer Family history—psychiatric (%) Family history—substance (%)
37.4 ± 0.82 130
38.9 ± 1.0 110
34 66
42 58
39 61 8.6 3.2 1.52
54 46 14.0 4.6 1105
96 4 43.1 24.6 19.2 7.7 5.4 5.4
40.9 21.8 23.6 6.4 7.3 10.9
11.5 27.7
18.2 41.8
a Did not fulfill the criteria for comorbid personality disorders. Mean age of control population was 40.4 ± 1.5 years.
combination we selected A118G SNP which is associated with addiction in our population (Deb et al., 2010). 2.3. Genotyping of SNPs Five milliliters blood was drawn from each subject. Isolation of DNA was carried out as performed earlier (Deb et al., 2010). Briefly, PCR amplification of exons 3 and 4 of OPRK1 gene was performed in 20 μl PCR reaction system consisting of 20 mM Tris–HCl pH 8.0, 50 mM EDTA, 0.2 mM dNTPs, 1.5 mM MgCl2, 0.5 μmol each primer (forward and reverse) and 2.5 units of Taq polymerase. The primers used are provided in Table 2. Following standardization of the PCR conditions, sequencing was carried out using an automated DNA sequencer ABI Prism 3130XL (Applied Biosystems, California, USA). To detect the polymorphism, A118G of OPRM1, restriction fragment length polymorphism (RFLP) was applied by analyzing the digestion profile of the PCR product in presence of Drd 1, as described earlier (Deb et al., 2010).
was performed with the help of PS v2.1.31 software (Dupont and Plummer, 1998). Additionally, genotype distribution was also analyzed by Hardy– Weinberg equilibrium by χ 2 test for studies for A118G as performed earlier (Deb et al., 2010). Odds ratios (OR) with Cornfield 95% confidence intervals (CIs) were computed by logistical regression using the SPSS statistical package (version 7.5; SPSS, Chicago, IL, USA) both for single SNP as well as for two SNP interactions. In SNP–SNP interaction studies, homozygous wild type genotypes (A) in the both the genes were taken as reference group while the AB and BB genotypes are pooled. A multiplicative interaction was performed and all the possible interactions were used to find out the risk of addiction employing the SPSS statistical package. Significance level was set at P b 0.05. For Linkage disequilibrium (LD) analysis, Haploview v4.1 was used for the analysis of LD blocks and haplotype. The numbers in each box correspond to r 2 values (multiplied by 100) between SNPs. 3. Results In the present study, we scanned exonic regions 3 and 4 of OPRK1. There are three SNPs in exon 3 and fourteen SNPs in exonic region 4 as reported in NCBI dbSNP (http://www.ncbi.nlm.nih.gov/snp). We observed the presence of three SNPs in exonic region 4, namely rs16918875, rs702764 present in exon 4 and rs963549 present in the 3′UTR region of exon 4 in the population studied. However, we did not find any of the reported SNP in the exonic region 3 in either cases or controls. The genotype distribution of rs16918875, rs702764 and rs963549 has been represented in Table 3. Application of Fisher's exact test showed that the distribution of rs16918875 (opioid population, P=0.3104; alcoholic population, P =0.3705), rs702764 (opioid population, P= 0.2982; alcoholic population, P=0.638) as well as rs963549 (opioid population, P=0.234; alcoholic population, P= 0.8809) in cases and controls was not significantly different suggesting that the SNPs were not associated with addiction. Pairwise linkage disequilibrium between the three SNPs of OPRK1 was measured by the standardized disequilibrium value (D′). Plot shows the confidence bound color scheme (Fig. 2), in which dark gray indicated a strong evidence of LD while light gray suggests uninformative status of LD. The observed value of D, a representation of allelic frequencies of SNPs, was not equal to 1, suggesting that the SNPs were not in perfect LD. The genotype distribution of A118G (rs1799971) of OPRM1 gene has been represented in Table 4. The genotypic frequency for homozygous Table 3 Association analysis of OPRK1 SNPs in addicted and control populations.
2.4. Statistical analysis The Fisher's exact test was generally used for comparing the frequencies of variables when the expected count was less than 5. Since the number of the counts for risk genotypes was small (less than 5) in OPRK1 polymorphism, we used the Fisher's exact test followed by odds ratio (OR) and 95% confidence intervals (CI) computation for evaluating the allelic and genotypic associations of the polymorphisms. P valuesb 0.05 were considered significant. The power study analysis
Table 2 Primer sequences. Gene
Region
Forward primer
Reverse primer
OPRK1
Exon 3 Exon 4 Exon 1
AGGCACTACATTTCCGTCGTC ACTGATCACGAACGTGGTCTG TGGCAGCGGCGAAAGGAAG
GCTTAAAGCCAAGCGACTCTG GTCGATGTCATTGAGTGCTCC TTCGGACCGCATGGGACGGAC
OPRM1a
227
a PCR product of OPRM1 was subjected to RFLP analysis for detecting A118G polymorphism.
Genotype rs16918875 TT TC + CC Fisher's exact test Odd ratio Power rs963549 CC CT + TT Fisher's exact test Odd ratio Power rs702764 GG GA Fisher's exact test Odd ratio Power
Control
Heroin
Addiction alcohol
178 22
110 20 P = 0.3104 1.471 (0.767 to 2.820) 0.221
94 16 P = 0.3705 1.377 (0.690 to 2.748) 0.157
162 38
112 18 P = 0.234 0.685 (0.372 to 1.262) 0.223
88 22 P = 0.8809 1.066 (0.593 to 1.915) 0.057
188 12
126 4 P = 0.2982 0.497 (0.157 to 1.577) 0.221
102 8 P = 0.638 1.229 (0.486 to 3.104) 0.078
Fisher's exact test (2 × 2 table, two tailed), P b 0.05.
228
D. Kumar et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 36 (2012) 225–230
TT allele of rs16918875 [OR = 1.660 (1.006 to 2.745), P = 0.0474]. For rs963549, there is a substantial risk of addiction where its risk alleles CT/TT interacted with the risk allele of A118G in the alcoholics [OR = 2.307 (1.074 to 4.955), P = 0.0321] and not in the heroin users. Surprisingly for rs963549, interactions of its risk alleles, CT/TT with the wild type allele, AA of A118G [OR = 0.305 (0.101 to 0.921), P = 0.0351] as well as interactions of its wild type allele, CC with the risk allele, GG/AG of A118G [OR 1.717 (1.037 to 2.844), P = 0.0357] were found to be significantly associated in heroin addicts. Similarly, studies of the interaction of the SNP, rs702764 with A118G also showed a significant association of its wild type allele, GG with the risk allele GG/AG in both the heroin [OR = 2.05 (1.279 to 3.287), P = 0.0029] as well as in alcohol [OR = 1.737 (1.048 to 2.879), P = 0.0322] users.
4. Discussion
Fig. 2. Pairwise linkage disequilibrium between three SNPs of OPRK1 in the studied population. The pattern of pair-wise LD between SNPs was measured by the standardized disequilibrium value (D′) as implemented in Haploview v4.1 (Barrett et al., 2005). Plot showing the confidence bound color scheme where dark gray indicates strong evidence of LD and light gray suggests uninformative status of LD.
wild AA, heterozygous mutant AG and homozygous mutant GG in heroin addicts was 54%, 41% and 5% respectively; in alcoholic subjects were 54.5%, 37% and 8.5% respectively and in the control population were 68.5%, 27.5% and 4% respectively. Genotype frequencies were significantly different between heroin and control populations (χ2 = 7.268, P = 0.0264), as well as between alcoholic and control populations (χ2 = 6.626, P= 0.0364). It was observed that the heterozygous genotype frequency was highest in heroin addicts and lowest in the controls. The allelic frequencies of A118G for A and G in the different groups were 74% and 26% respectively in heroin population, 73% and 27% respectively in alcoholic population and 82% and 18% respectively in control population (Table 4). The risk allele G of this SNP was more frequent in heroin dependent group [OR = 1.609 (1.102–2.348)] and as well as in alcoholic group [OR = −1.698 (1.146 to 2.517)] as compared to the control group. The heterozygous and homozygous carriers of the risk allele had higher O.R. in heroin 1.886 (1.173–3.032) and 1.713 (0.597 to 4.916) as well as in alcoholic 1.702 (1.027 to 2.822) and 2.569 (0.945 to 6.98) respectively as compared to the wild type genotype. The locus–locus interaction between OPRM1 and OPRK1 is shown in Table 5. There is a marked risk of addiction in subjects where both the risk alleles TC/CC of rs16918875 of OPRK1 and GG/AG of A118G of OPRM1 were present in both heroin [OR = 2.516 (1.152 to 5.495), P = 0.0206] and alcoholic [OR = 2.318 (1.025 to 5.24), P = 0.0433] cases. On the contrary, there is an increase of risk in addiction to only heroin addicts during interaction of GG/AG of A118G and the
Allelic variations in certain critical genes are likely to influence drug dependence in an individual since there is accumulating evidence suggesting that heritable factors play an important role in addictive behavior (Bierut et al., 1988). The present study investigated KOR as a candidate gene for genetic vulnerability to substance abuse. Three SNPs in OPRK1 were identified of which rs963549 is reported to have high FST, a measure of genetic differentiation, suggesting that it had undergone selection, and therefore might be of functional importance (Akey et al., 2002). However, rs963549 did not appear to be associated with drug dependence in the population, which is in agreement with a previous report of a similar lack of association with alcohol dependence in a population of European Americans (Xuei et al., 2006). It has been demonstrated that allelic variation at rs963549 did not moderate the response to the treatment for alcohol dependence in subjects by the opioid antagonist, naltrexone (Gelernter et al., 2007). Moreover, unlike association between alcohol dependence and the multiple SNPs across the large region located in introns of OPRK1, all the SNPs studied in the coding region including rs702764 did not show any risk for alcohol addiction (Xuei et al., 2006). Another SNP, rs702764, detected in the population, also did not show any moderating effect on the reduction of drinking by the opioid antagonist, nalmefene in a population of Caucasians of Finnish origin (Arias et al., 2008). In our study of OPRK1 polymorphism in rs16918875, we did not find any homozygous mutant as observed in the European and Asian population HAPMAP data (NCBI dbSNP). While we observed the presence of heterozygous mutants of rs16918875 in both cases and controls, studies in the Asian population did not report presence of either the heterozygous or the homozygous mutant. HAPMAP data also showed that in case of rs963549, the C is a major allele in the case of Asian, European and in the present studied population and it is the risk minor allele reported in sub-Sahara population (NCBI dbSNP). It appears, therefore, that there are considerable differences in the distribution of risk allele frequencies in the different ethnic groups.
Table 4 Distribution of genotype and allele frequencies of A118G (rs1799971) in addicted and control populations. Control (n%)
Heroin (n%)
OR (95%CI)
Alcoholic (n%)
OR (95%CI)
AA AG GG
137(63.5) 55(27.5) 8 (4) χ2HW = O.67 P = 0.410
70(54) 53(41) 7(5) χ2HW = 0.560 P = 0.454
1 (referent) 1.886 (1.173 to 3.032) 1.713 (0.597 to 4.916) Ptrend = 0.0138⁎ χ2 = 7.268 P = 0.0264⁎
60 (54.5) 41 (37) 9 (8.5) χ2HW = 0.279 P = 0.596
1 (referent) 1.702 (1.027 to 2.822) 2.569 (0.945 to 6.98) Ptrend = 0.0101⁎ χ2 = 6.626 P = 0.0364⁎
A G
329(82) 71(18)
193(74) 67(26)
Power = 0.766 0.622 (0.426 to 0.907) 1.609 (1.102 to 2.348)
161(73) 59(27)
χ2(1), Chi-square analysis of genotype frequency. OR, odds ratio; CI, 95% confidence interval; n, number of individuals/n, number of alleles; χ2HW , Hardy–Weinberg-Equilibrium calculated by χ2test. ⁎ Indicates significance.
Power = 0.683 0.589 (0.397–0.873) 1.698 (1.146–2.517)
D. Kumar et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 36 (2012) 225–230
229
Table 5 Locus–locus interactions between OPRK1 and OPRM11 on addiction risk. Genotype
Genotype
OPRM1 AA AA GG/AG GG/AG OPRM1 AA AA GG/AG GG/AG OPRM1 AA AA GG/AG GG/AG
rs16918875 TT TC/CC TT TC/CC rs963549 CC CT/TT CC CT/TT rs702764 GG GA/AA GG GA/AA
Control
Cases (heroin)
OR
Cases (alcoholic)
OR
127 9 51 13
66 3 44 17
1 (referent) 0.641 (0.168 to 2.45) 1.660 (1.006 to 2.745) 2.516 (1.152 to 5.495)
P = 0.5160 P = 0.0474 P = 0.0206
59 2 35 14
1 (referent) 0.478 (0.1 to 2.283) 1.477 (0.867 to 2.507) 2.318 (1.025 to 5.24)
P = 0.3551 P = 0.1487 P = 0.0433
114 23 48 15
65 4 47 14
1 (referent) 0.305 (0.101 to 0.921) 1.717 (1.037 to 2.844) 1.637 (0.743 to 3.605)
P = 0.0351 P = 0.0357 P = 0.2212
56 5 32 17
1 (referent) 0.443 (0.156 to 1.226) 1.357 (0.783 to 2.352) 2.307 (1.074 to 4.955)
P = 0.1168 P = 0.2764 P = 0.0321
134 5 54 7
69 0 57 4
1 (referent) 0.176 (0.01 to 3.228) 2.05 (1.279 to 3.287) 1.11 (0.314 to 3.922)
P = 0.2418 P = 0.0029 P = 0.8716
60 1 42 7
1 (referent) 0.447 (0.051 to 3.906) 1.737 (1.048 to 2.879) 2.233 (0.750 to 6.649)
P = 0.4663 P = 0.0322 P = 0.1489
Odds ratios (ORs) and 95% CIs for the different genotype combinations are shown. P in bold demonstrates significantly different (b 0.05) from controls.
Significant advances have been made during the past few decades in determining the molecular genetic basis for human inherited diseases. A plethora of genes have been identified, mutations of which have been linked with specific clinical conditions. Such mutations are generally associated with disorders that demonstrate Mendelian inheritance characterized by “single-gene” disorders although in many cases, the genotype–phenotype correlations due to the mutation have proved elusive (Beaudet et al., 2001; Scriver, 2002). Various reasons like reduced penetrance of the mutant gene, contributions of important genetic modifiers for the expression of the affected genes, have been attributed to this lack of correlation (Dipple and McCabe, 2000; Haldane, 1949; Scriver and Waters, 1999). This is more glaring for complex diseases where the genetic etiology is indicative of multifactorial syndrome. Contributions of gene–gene and gene–environment interactions for the variable phenotypic outcome are indicative from various studies. Recent reports indicate a digenic inheritance due the association of pten-induced kinase 1 (PINK1) and DJ-1 for early-onset Parkinson's disease (Tang et al., 2006). Cell transfection studies demonstrate that DJ-1 and PINK1 physically associate and collaborate to protect cells against stress via complex formation suggesting that DJ-1 and PINK1 are genetically linked as well as physically associated. In patients suffering from retinitis pigmentosa (RP), an association of the Asn306Ser variant of the SP4 transcription factor and an intronic variant in the beta-subunit of transducin has been reported (Gao et al., 2007). Families have been identified with mutations in the unlinked photoreceptor-specific genes ROM1 and peripherin/RDS, in which only double heterozygotes develop RP (Kajiwara et al., 1994). In RP it is apparent that a mutation in a second gene is necessary for the disease phenotype whereby the two mutants combine to prevent the formation of complexes which affect the integrity of photoreceptors. Some of the other digenic inheritance reported are for diseases like deafness (Adato et al., 1999) as well as SNP–SNP interaction in breast cancer (Onay et al., 2006), prostate cancer (Lin et al., 2008; Beuten et al., 2009.), type 2 diabetes (Keshavarz et al., 2008; Qu et al., 2011) etc. However, there are also contrary reports showing lack of clinical significance and digenic inheritance due to mutations of more than one gene (Azmanov et al., 2011; Vilella et al., 2008). Nevertheless, all these studies demonstrate that the presence of mutations in one familial gene should not serve as exclusion criteria for the screening for further genetic variation. Likewise, addiction is also a complex, multifactorial disease and the hypothesis that multiple genes are involved in the predisposition of addiction is well supported by understanding the biology of addiction. OPRK1 and OPRM1 are both involved in addiction pathway but their functions are opposite in nature. In gene–gene co-association study, LD based statistic has suggested that SNP–SNP interactions in OPRK1 and OPRM1 were significant in heroin dependent subjects (Peng et al., 2010). The present study of OPRK1 polymorphisms and their interaction
with A118G of the OPRM1 gene is perhaps the first case–control study in an Indian population. Interestingly, increased risk of addiction was generally associated with those genotype combinations where one or both the risk alleles were present. In almost all the risk combinations, the OPRM1 risk allele GA/GG was invariably present suggesting an important contribution of OPRM1 in the digenic inheritance for addiction. Opioids regulate a variety of brain functions related to addictions like reward, reinforcement, mood and psychomotor stimulation. While the biological basis of mood is not well understood, KOR systems tend to compensate for the continued exposure to drugs acting at MOR and have been hypothesized to underlie the protracted abstinence syndrome, with dysphoric mood in opioid-dependent individuals (Rothman, 1992). Treatment with selective KOR agonists is reported to cause a psychotomimetic effect and dysphoria, both in clinical studies and experimental animal models (Hasebe et al., 2004) while KOR agonists may have a potential role in the treatment of mood disorders (Carlezon et al., 2009). The lack of depressed patients, in the present study, may have been influenced by the high rate of KOR polymorphisms in the samples. Since A118G polymorphism is expected to undermine reward experience in the cases, possibilities that bad mood, induced by KOR variants, becoming problematic only when A118G mutant is present, cannot be ruled out. There are some limitations in this study. Firstly, the sample size is moderate and there is a need for confirmation of these data in larger populations. Secondly, we investigated only interactions of two genes, but a large number of candidate genes have been implicated in the addiction pathway. Thirdly, the participants in the study were males only, the underlying reason may be the unavailability of female volunteers in such conservative society. This may result in gender bias. Lastly, after multiple testing the interaction between OPRM1 (GG/AG) and rs702764 remained significant while all other results lost their significance (See Supplementary Table for details). In summary, this study provides evidence that gene–gene interaction between KOR and OPRM1 can influence the risk of addiction to narcotics and alcohol. Results also suggest a disease association of rare genotypes which may otherwise be missed in the absence of interactions with known risk alleles. Our approach of evaluating moderate sample size is one of the first efforts in explaining the effect of gene– gene interaction in Indian population and might be helpful in improving the ability and accuracy in assessing individual risk towards addiction. Supplementary materials related to this article can be found online at doi:10.1016/j.pnpbp.2011.10.018. Acknowledgments The authors thank Mr. Pinaki Mondal, Senior Research Fellow, Cancer and Cell Biology Division, Indian Institute of Chemical Biology, Kolkata for his help in the statistical analysis of the data.
230
D. Kumar et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 36 (2012) 225–230
References Adato A, Kalinski H, Weil D, Chaib H, Korostishevsky M, Bonne-Tamir B. Possible interaction between USH1B and USH3 gene products as implied by apparent digenic deafness inheritance. Am J Hum Genet 1999;65:261–5. Akey JM, Zhang G, Zhang K, Jin L, Shriver MD. Interrogating a high-density SNP map for signatures of natural selection. Genome Res 2002;12:1805–14. Arias AJ, Armeli S, Gelernter J, Covault J, Kallio A, Karhuvaara S, et al. Effects of opioid receptor gene variation on targeted nalmefene treatment in heavy drinkers. Alcohol Clin Exp Res 2008;32:1159–66. Azmanov DN, Dimitrova S, Florez L, Cherninkova S, Draganov D, Morar B, et al. LTBP2 and CYP1B1 mutations and associated ocular phenotypes in the Roma/Gypsy founder population. Eur J Hum Genet 2011;19:326–33. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005;21:263–5. Beaudet A, Scriver C, Sly W, Valle D. Genetics, biochemistry, and molecular bases of variant human phenotypes. In: Scriver C, Beaudet A, Sly W, Valle D, editors. The metabolic and molecular bases of inherited disease. 8th ed. New York: McGraw-Hill; 2001. p. 3-45. Beuten J, Garcia D, Brand TC, He X, Balic I, Canby-Hagino E, et al. Semaphorin 3B and 3F single nucleotide polymorphisms are associated with prostate cancer risk and poor prognosis. J Urol 2009;182(4):1614–20. Bhargava HN. Diversity of agents that modify opioid tolerance, physical dependence, abstinence syndrome, and self-administrative behaviour. Pharmacol Rev 1994;46:293–324. Bierut LJ, Dinwiddie SH, Begleiter H, Crowe RR, Hesselbrock V, Nurnberger Jr JI, et al. Familial transmission of substance dependence: alcohol, marijuana, cocaine, and habitual smoking: a report from the Collaborative Study on the Genetics of Alcoholism. Arch Gen Psychiatry 1988;55:982–8. Bolanos CA, Garmsen GM, Clair MA, McDougall SA. Effects of the kappa-opioid receptor agonist U-50,488 on morphine-induced place preference conditioning in the developing rat. Eur J Pharmacol 1996;317:1–8. Carlezon Jr WA, Béguin C, Knoll AT, Cohen BM. Kappa-opioid ligands in the study and treatment of mood disorders. Pharmacol Ther 2009;123:334–43. Cowan A, Zhu XZ, Mosberg HI, Omnaas JR, Porreca F. Direct dependence studies in rats with agents selective for different types of opioid receptor. J Pharmacol Exp Ther 1998;246:950–5. Deb I, Chakraborty J, Gangopadhyay PK, Choudhury SR, Das S. Single-nucleotide polymorphism (A118G) in exon 1 of OPRM1 gene causes alteration in downstream signaling by mu-opioid receptor and may contribute to the genetic risk for addiction. Neurochem 2010;112:486–96. Dipple KM, McCabe ER. Phenotypes of patients with “simple” Mendelian disorders are complex traits: thresholds, modifiers, and systems dynamics. Am J Hum Genet 2000;266:1729–35. Dupont WD, Plummer Jr WD. Power and sample size calculations for studies involving linear regression. Control Clin Trials 1998;19:589–601. Edenberg HJ, Wang J, Tian H, Pochareddy S, Xuei X, Wetherill L, et al. A regulatory variation in OPRK1, the gene encoding the κ-opioid receptor, is associated with alcohol dependence. Hum Mole Genet 2008;17:1783–9. Friedman HJ, Jen MF, Chang JK, Lee NM, Loh HH. Dynorphin: a possible modulatory peptide on morphine or beta-endorphin analgesia in mouse. Eur J Pharmacol 1981;69:357–60. Funada M, Suzuki T, Narita M, Misawa M, Nagase H. Blockade of morphine reward through the activation of kappa-opioid receptors in mice. Neuropharmacology 1993:1315–23. Gao YQ, Danciger M, Ozgul RK, Gribanova Y, Jacobson S, Farber DB. Association of the Asn306Ser variant of the SP4 transcription factor and an intronic variant in the beta-subunit of transducin with digenic disease. Mol Vis 2007;13:287–92. Gelernter J, Gueorguieva R, Kranzler HR, Zhang H, Cramer J, Rosenheck R, et al. VA Cooperative Study #425 Study Group. Opioid receptor gene (OPRM1, OPRK1, and OPRD1) variants and response to naltrexone treatment for alcohol dependence: results from the VA Cooperative Study. Alcohol Clin Exp Res 2007;31:555–63. Gerra G, Leonardi C, Cortese E, D'Amore A, Lucchini A, Strepparola G, et al. Human kappa opioid receptor gene (OPRK1) polymorphism is associated with opiate addiction. Am J Med Genet B Neuropsychiatr Genet 2007;144:771–5. Glick SD, Maisonneuve IM, Raucci J, Archer S. Kappa opioid inhibition of morphine and cocaine self-administration in rats. Brain Res 1995;681:147–52. Haldane JBS. Disease and evolution. Ric Sci Suppl 1949;19:3-10. Hasebe K, Kawai K, Suzuki T, Kawamura K, Tanaka T, Narita M, et al. Possible pharmacotherapy of the opioid kappa receptor agonist for drug dependence. Ann N Y Acad Sci 2004;1025:404–13. Kajiwara K, Berson EL, Dryja TP. Digenic retinitis pigmentosa due to mutations at the unlinked peripherin/RDS and ROM1 loci. Science 1994;264:1604–8.
Keshavarz P, Inoue H, Nakamura N, Yoshikawa T, Tanahashi T, Itakura M. Single nucleotide polymorphisms in genes encoding LKB1 (STK11), TORC2 (CRTC2) and AMPK alpha2-subunit (PRKAA2) and risk of type 2 diabetes. Mol Genet Metab 2008;93: 200–9. Kuzmin AV, Semenova S, Gerrits MA, Zvartau EE, Van Ree JM. Kappa-opioid receptor agonist U50,488H modulates cocaine and morphine self-administration in drugnaive rats and mice. Eur J Pharmacol 1997;321:265–71. Lin HY, Wang W, Liu YH, Soong SJ, York TP, Myers L, et al. Comparison of multivariate adaptive regression splines and logistic regression in detecting SNP–SNP interactions and their application in prostate cancer. J Hum Genet 2008;53:802–11. Loh el W, Fann CS, Chang YT, Chang CJ, Cheng AT. Endogenous opioid receptor genes and alcohol dependence among Taiwanese Han. Alcohol Clin Exp Res 2004;28: 15–9. Matthes HW, Maldonado R, Simonin F, Valverde O, Slowe S, Kitchen I, et al. Loss of morphine-induced analgesia, reward effect and withdrawal symptoms in mice lacking the mu-opioid-receptor gene. Nature 1996;383:819–23. Onay VU, Briollais L, Knight JA, Shi E, Wang Y, Wells S, et al. SNP–SNP interactions in breast cancer susceptibility. BMC Cancer 2006;6:114. Peng Q, Zhao J, Xue F. A gene-based method for detecting gene–gene co-association in a case–control association study. Eur J Hum Genet 2010;18:582–7. Qu L, He B, Pan Y, Xu Y, Zhu C, Tang Z, et al. Association between polymorphisms in RAPGEF1, TP53, NRF1 and type 2 diabetes in Chinese Han population. Diabetes Res Clin Pract 2011;91:171–6. Ramarao P, Jablonski Jr HI, Rehder KR, Bhargava HN. Effect of kappa-opioid receptor agonists on morphine analgesia in morphine-naive and morphine-tolerant rats. Eur J Pharmacol 1988;156:239–46. Reisine T, Pasternak G. Opioid analgesics and antagonists. In: Hardman JG, Limbird LE, editors. Goodman & Gilman's the pharmacological basis of therapeutics. 9 ed. McGraw-Hill; 1996. Rothman RB. A review of the role of anti-opioid peptides in morphine tolerance and dependence. Synapse 1992;12:129–38. Saito M, Ehringer MA, Toth R, Oros M, Szakall I, Sikela JM, et al. Variants of kappaopioid receptor gene and mRNA in alcohol-preferring and alcohol-avoiding mice. Alcohol 2003;29:39–49. Scriver CR. Why mutation analysis does not always predict clinical consequences: explanations in the era of genomics. J Pediatr 2002;140:502–6. Scriver CR, Waters PJ. Monogenic traits are not simple: lessons from phenylketonuria. Trends Genet 1999;15:267–72. Shippenberg TS, Rea W. Sensitization to the behavioral effects of cocaine: modulation by dynorphin and kappa-opioid receptor agonists. Pharmacol Biochem Behav 1997;57:449–55. Sora I, Takahashi N, Funada M, Ujike H, Revay RS, Donovan DM, et al. Opiate receptor knockout mice define mu receptor roles in endogenous nociceptive responses and morphine-induced analgesia. Proc Natl Acad Sci 1997;94:1544–9. Suzuki T, Narita M, Takahashi Y, Misawa M, Nagase H. Effects of nor-binaltorphimine on the development of analgesic tolerance to and physical dependence on morphine. Eur J Pharmacol 1992;213:91–7. Tang B, Xiong H, Sun P, Zhang Y, Wang D, Hu Z, et al. Association of PINK1 and DJ-1 confers digenic inheritance of early-onset Parkinson's disease. Hum Mol Genet 2006;15: 1816–25. Tao PL, Hwang CL, Chen CY. U-50,488 blocks the development of morphine tolerance and dependence at a very low dose in guinea pigs. Eur J Pharmacol 1994;256: 281–6. Tulunay FC, Jen MF, Chang JK, Loh HH, Lee NM. Possible regulatory role of dynorphin on morphine- and beta-endorphin-induced analgesia. J Pharmacol Exp Ther 1981;219: 296–8. Vilella E, Costas J, Sanjuan J, Guitart M, Diego YD, Carracedo A, et al. Association of schizophrenia with DTNBP1 but not with DAO, DAOA, NRG1 and RGS4 nor their genetic interaction. J Psychiatr Res 2008;42:278–88. Xuei X, Flury-Wetherill L, Bierut L, Dick D, Nurnberger Jr J, Foroud T, et al. The opioid system in alcohol and drug dependence: family-based association study. Am J Med Genet B Neuropsychiatr Genet 2007;144:877–84. Xuei X, Dick D, Flury-Wetherill L, Tian HJ, Agrawal A, Bierut L, et al. Association of the kappa-opioid system with alcohol dependence. Mol Psychiatry 2006;11:1016–24. Yuferov V, Fussell D, LaForge KS, Nielsen DA, Gordon D, Ho A, et al. Redefinition of the human kappa opioid receptor gene (OPRK1) structure and association of haplotypes with opiate addiction. Pharmacogenet 2004;14:793–804. Zadina JE, Hackler L, Ge LJ, Kastin AJ. A potent and selective endogenous agonist for the mu-opiate receptor. Nature 1997;386:499–502. Zhang Y, Butelman ER, Schlussman SD, Ho A, Kreek MJ. Effects of the kappa opioid agonist R-84760 on cocaine-induced increases in striatal dopamine levels and cocaineinduced place preference in C57BL/6j mice. Psychopharmacol 2004;173:146–52.
Contents lists available at SciVerse ScienceDirect
Progress in Neuro-Psychopharmacology & Biological Psychiatry journal homepage: www.elsevier.com/locate/pnp
Epistatic effects between variants of kappa-opioid receptor gene and A118G of mu-opioid receptor gene increase susceptibility to addiction in Indian population Deepak Kumar a, Japashish Chakraborty b, Sumantra Das a,⁎ a b
Neurobiology Division, Indian Institute of Chemical Biology, 4 Raja S. C. Mullick Road, Jadavpur, Kolkata 700032, India Baulmon, 34 Jadavpur Central Road, Kolkata 700032, India
a r t i c l e
i n f o
Article history: Received 5 July 2011 Received in revised form 1 October 2011 Accepted 31 October 2011 Available online 28 November 2011 Keywords: Addiction Alcohol Heroin Kappa opioid receptor Single nucleotide polymorphism
a b s t r a c t Objective: Unequivocal evidence suggests contribution of κ-opioid receptor (KOR) in addiction to drugs of abuse. A study was undertaken to identify the single nucleotide polymorphisms (SNP) at selective areas of kappa opioid receptor 1 (OPRK1) gene in heroin as well as in alcohol addicts and to compare them with that in control population. The potential interaction of the identified KOR SNPs with A118G of μ opioid receptor was also investigated. Methods: Two hundred control subjects, one hundred thirty heroin and one hundred ten alcohol addicts, all male and residing in Kolkata, a city in eastern India, volunteered for the study. Exons 3 and 4 of OPRK1 and the SNP, A118G of mu opioid receptor 1 (OPRM1) in the DNA samples were genotyped by sequencing and restriction fragment length polymorphism respectively. The SNPs identified in the population were analyzed by odds ratio and its corresponding 95% confidence interval was estimated using logistic regression models. SNP–SNP interactions were also investigated. Results: Three SNPs of OPRK1, rs16918875, rs702764 and rs963549, were identified in the population, none of which showed significant association with addiction. On the other hand, significant association was observed for A118G with heroin addiction (χ2 = 7.268, P = 0.0264) as well as with alcoholic addition (χ2 = 6.626, P = 0.0364). A potential SNP–SNP interaction showed that the odds of being addicted was 2.51 fold in heroin subjects [CI (95%) = 1.1524 to 5.4947, P = 0.0206] and 2.31 fold in alcoholics [CI (95%) = 1.025 to 5.24, P = 0.0433] with the OPRK1 (rs16918875) and A118G risk alleles than without either. A significant interaction was also identified between GG/AG of A118G and GG of rs702764 [O.R (95%) = 2.04 (1.279 to 3.287), P = 0.0029] in case of opioid population. Conclusion: Our study suggests that set associations of polymorphisms may be important in determining the risk profile for complex diseases such as addiction. © 2011 Elsevier Inc. All rights reserved.
1. Introduction There has been an increasing interest in the understanding of the role of κ-opioid receptors (KOR) in addiction to drugs of abuse. Although the μ-opioid receptor (MOR) mediates most of the opioid actions, including analgesia, tolerance and reward (Matthes et al., 1996; Reisine and Pasternak, 1996; Zadina et al., 1997), KOR has been receiving much attention because of its functional interaction with the MOR. In animal models, an extensive body of pharmacological data demonstrates that the KOR opposes a variety of MOR-mediated actions like analgesia (Friedman et al., 1981; Ramarao et al., 1988;
Abbreviations: OPRK1, kappa opioid receptor; OPRM1, mu opioid receptor; CI, confidence interval; LD, Linkage disequilibrium; OR, Odds ratios; SNP, single nucleotide polymorphism; KOR, κ-opioid receptors; MOR, μ-opioid receptor. ⁎ Corresponding author at: Neurobiology Division, Indian Institute of Chemical Biology, 4, Raja S.C. Mullick Road, Jadavpur, Kolkata 700032, India. Tel.: + 91 33247 35197/ 24736793/24724049; fax: + 91 33247 35197/24730284. E-mail addresses: [email protected], [email protected] (S. Das). 0278-5846/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.pnpbp.2011.10.018
Sora et al., 1997; Tao et al., 1994; Tulunay et al., 1981), tolerance (Bhargava, 1994; Rothman, 1992), dependence (Suzuki et al., 1992), withdrawal (Cowan et al., 1998) as well as in the reinforcing effects of morphine in both self-administration (Glick et al., 1995; Kuzmin et al., 1997) and conditioned place-preference paradigms (Bolanos et al., 1996; Funada et al., 1993). Like MOR, the KOR is localized in several areas of the dopaminergic nigrostriatal and mesolimbic– mesocortical systems which are the sites of known actions of drugs of abuse (Zhang et al., 2004). However, unlike MOR ligands, the endogenous KOR agonists, dynorphins, decrease basal and drug induced dopamine levels in several areas of dopaminergic nigrostriatal and mesolimbic–mesocortical system and inhibit morphine-withdrawal symptoms induced by naloxone precipitation or morphine discontinuation in morphine dependent animals (Suzuki et al., 1992). The KOR– dynorphin system may therefore be considered to be a part of the counter-modulatory mechanisms of the brain and studies suggest that KOR agonists may be potential therapeutic agents for treatment of drug addiction and disorders that result from alterations in mesolimbic dopamine transmission (Shippenberg and Rea, 1997).
226
D. Kumar et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 36 (2012) 225–230
Considering that several genetic studies using family, twin and linkage and association designs have consistently demonstrated genetics factors as contributing to vulnerability to drug addiction (Bierut et al., 1988; Xuei et al., 2007), there have been attempts to identify mutations in the KOR gene as possible risk factors for the development of addiction. KOR gene polymorphisms have been reported to contribute to predisposition to voluntary alcohol-drinking behavior in experimental animals (Saito et al., 2003). In humans, the 36G>T SNP at exon 1 of the KOR gene (OPRK1), which results in a silent mutation, has been shown to exhibit a significant association with a population of heroin addicts of West European, Caucasian origin (Gerra et al., 2007) as well of African-American, Caucasian and Hispanic origin (Yuferov et al., 2004). However, contrasting reports excluding a possible role of the opioid receptor genes for the proclivity to dependence (Loh el et al., 2004) in Taiwanese population, suggest influence of the different racial or ethnic stratification of the samples. The human OPRK1 gene is located on chromosome 8q11.2. The human OPRK1 gene has four exons and three introns and untranslated (UTR) regions of 660 and 3576 nucleotides respectively at the 5′ and 3′ ends. Altogether about 24 number of SNPs of KOR gene has so far been investigated (Fig. 1). Although some studies have explored association of some of these SNPs with drug dependence, no definite identification of risk allele has been detected (Yuferov et al., 2004; Edenberg et al., 2008). Our main objective was to investigate the contribution of OPRK1 gene polymorphisms to addiction in exonic regions 3 and 4 of both alcohol as well heroin addicts and compared with control subjects. Three known SNPs in exonic region 4 were identified in the population. Since in many digenic disorders, gene–gene polymorphic interactions can influence the risk of any complex disease, we studied locus–locus interaction between OPRK1 and OPRM1. Considering the complex nature of addiction, SNP–SNP interaction studies are expected to provide useful information on the inter-dependence of specific variants of genes in contributing towards the development of diseases. For our studies, we investigated the interaction of the SNP, A118G of OPRM1 with the SNPs of OPRK1 identified in our study. A118G was considered because it has been previously shown to be highly associated with addiction in the population studied (Deb et al., 2010).
2. Materials and methods 2.1. Subjects One hundred thirty heroin addicts, one hundred ten alcoholic and two hundred non-addict or control individuals participated in this
study and provided written informed consent. Volunteers were unrelated male of mean age of 39.2 years. All subjects, both cases and controls, were ethnically Indians belonging to the Kayastha and Brahmin castes of the Bengali-Hindu ethnic background and were recruited from the city of Kolkata. Ethnicities of subjects were determined with the help of physicians' admission notes and the hospital data, which was based on information regarding family origin, ethnic background including geographical and place of birth. They were all unrelated male from various backgrounds like students, workers and general population of different income groups. Addicted patients were under treatment at Baulmon, a psychiatric institute situated in Kolkata. The institute was selected to carry out this study because of their capacity to administer structured DSM interview, to produce a diagnostic evaluation and to use routinely the psychometric instruments. Demographic description of the cases is provided in Table 1. Subjects were examined by a psychiatrist specializing in substance abuse disorders and interviewed using the Addiction Severity Index (ASI) questionnaire that covers medical, employment, drug, legal, family history, family and social relationships and psychiatric status. All heroin as well as alcohol-dependent subjects conformed to the criteria for dependency to drugs of abuse as defined by DSM IV (Diagnostic and Statistical Manual of Mental Disorders, 4th edition). Individuals with multidrug dependency or comorbidity to heroin and alcohol were excluded from this study. All the patients, meeting the criteria for DSM IV opioid dependence, submitted positive urine for morphine metabolites, before the beginning of the treatment. Only cases where the patients were dependent upon drugs for 5 years or more with history of at least one time relapse were selected for the study. Control subjects were healthy volunteers, all males having similar caste and ethnic background and matched for age. They were also interviewed and checked for history and family history for addiction. All the subjects included in this study were recruited at the same time period. The study had the approval of the institutional ethical committee for protection of research risks to humans.
2.2. Selection of SNPs While the 36G>T SNP at exon 1 of OPRK1 has been linked with addiction (Gerra et al., 2007; Loh el et al., 2004; Xuei et al., 2006; Yuferov et al., 2004), initial studies with a small population of subjects from both control and cases could not detect the presence of 36G>T SNP in the Indian population. In present study we selected exonic regions 3 and 4 of OPRK1 which contain functional receptor domains which are important for pharmacological treatment. To investigate the effect of gene–gene
Fig. 1. Locations of SNPs in OPRK1 gene drawn on the basis of published studies (Edenberg et al., 2008; Yuferov et al., 2004) as well as from descriptions in Vega Genome Browser (http://vega.sanger.ac.uk/Homo_sapiens/Transcript/Summary?db=core;g=OTTHUMG00000164276;r=8:54138284–54164257;t=OTTHUMT00000378048). Exons, introns and UTRs are not drawn to scale. *Indicates SNPs found to be present in the population studied. Numbers within parenthesis indicate frequency of occurrence as reported in dbSNP.
D. Kumar et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 36 (2012) 225–230 Table 1 Demographic and clinical characteristics of heroin and alcohol dependent patients. Characteristic
Heroin addict
Alcohol addict
Age Number of individuals Occupation (%) Employed Unemployed Marital status (%) Married Unmarried History of addiction (mean years) Average times of relapse Average daily intake of Heroin (g) Average daily ethanol intake (g/week) Route of heroin administration Smoked Intravenous Personal psychiatric disorders (%) a) Borderline personality traits b) Passive aggressive traits c) Antisocial personality traits d) Obsessive compulsive traits e) Hostile and aggressivea Personal non-psychiatric disorders (%) e.g. diabetes, gastric ulcer Family history—psychiatric (%) Family history—substance (%)
37.4 ± 0.82 130
38.9 ± 1.0 110
34 66
42 58
39 61 8.6 3.2 1.52
54 46 14.0 4.6 1105
96 4 43.1 24.6 19.2 7.7 5.4 5.4
40.9 21.8 23.6 6.4 7.3 10.9
11.5 27.7
18.2 41.8
a Did not fulfill the criteria for comorbid personality disorders. Mean age of control population was 40.4 ± 1.5 years.
combination we selected A118G SNP which is associated with addiction in our population (Deb et al., 2010). 2.3. Genotyping of SNPs Five milliliters blood was drawn from each subject. Isolation of DNA was carried out as performed earlier (Deb et al., 2010). Briefly, PCR amplification of exons 3 and 4 of OPRK1 gene was performed in 20 μl PCR reaction system consisting of 20 mM Tris–HCl pH 8.0, 50 mM EDTA, 0.2 mM dNTPs, 1.5 mM MgCl2, 0.5 μmol each primer (forward and reverse) and 2.5 units of Taq polymerase. The primers used are provided in Table 2. Following standardization of the PCR conditions, sequencing was carried out using an automated DNA sequencer ABI Prism 3130XL (Applied Biosystems, California, USA). To detect the polymorphism, A118G of OPRM1, restriction fragment length polymorphism (RFLP) was applied by analyzing the digestion profile of the PCR product in presence of Drd 1, as described earlier (Deb et al., 2010).
was performed with the help of PS v2.1.31 software (Dupont and Plummer, 1998). Additionally, genotype distribution was also analyzed by Hardy– Weinberg equilibrium by χ 2 test for studies for A118G as performed earlier (Deb et al., 2010). Odds ratios (OR) with Cornfield 95% confidence intervals (CIs) were computed by logistical regression using the SPSS statistical package (version 7.5; SPSS, Chicago, IL, USA) both for single SNP as well as for two SNP interactions. In SNP–SNP interaction studies, homozygous wild type genotypes (A) in the both the genes were taken as reference group while the AB and BB genotypes are pooled. A multiplicative interaction was performed and all the possible interactions were used to find out the risk of addiction employing the SPSS statistical package. Significance level was set at P b 0.05. For Linkage disequilibrium (LD) analysis, Haploview v4.1 was used for the analysis of LD blocks and haplotype. The numbers in each box correspond to r 2 values (multiplied by 100) between SNPs. 3. Results In the present study, we scanned exonic regions 3 and 4 of OPRK1. There are three SNPs in exon 3 and fourteen SNPs in exonic region 4 as reported in NCBI dbSNP (http://www.ncbi.nlm.nih.gov/snp). We observed the presence of three SNPs in exonic region 4, namely rs16918875, rs702764 present in exon 4 and rs963549 present in the 3′UTR region of exon 4 in the population studied. However, we did not find any of the reported SNP in the exonic region 3 in either cases or controls. The genotype distribution of rs16918875, rs702764 and rs963549 has been represented in Table 3. Application of Fisher's exact test showed that the distribution of rs16918875 (opioid population, P=0.3104; alcoholic population, P =0.3705), rs702764 (opioid population, P= 0.2982; alcoholic population, P=0.638) as well as rs963549 (opioid population, P=0.234; alcoholic population, P= 0.8809) in cases and controls was not significantly different suggesting that the SNPs were not associated with addiction. Pairwise linkage disequilibrium between the three SNPs of OPRK1 was measured by the standardized disequilibrium value (D′). Plot shows the confidence bound color scheme (Fig. 2), in which dark gray indicated a strong evidence of LD while light gray suggests uninformative status of LD. The observed value of D, a representation of allelic frequencies of SNPs, was not equal to 1, suggesting that the SNPs were not in perfect LD. The genotype distribution of A118G (rs1799971) of OPRM1 gene has been represented in Table 4. The genotypic frequency for homozygous Table 3 Association analysis of OPRK1 SNPs in addicted and control populations.
2.4. Statistical analysis The Fisher's exact test was generally used for comparing the frequencies of variables when the expected count was less than 5. Since the number of the counts for risk genotypes was small (less than 5) in OPRK1 polymorphism, we used the Fisher's exact test followed by odds ratio (OR) and 95% confidence intervals (CI) computation for evaluating the allelic and genotypic associations of the polymorphisms. P valuesb 0.05 were considered significant. The power study analysis
Table 2 Primer sequences. Gene
Region
Forward primer
Reverse primer
OPRK1
Exon 3 Exon 4 Exon 1
AGGCACTACATTTCCGTCGTC ACTGATCACGAACGTGGTCTG TGGCAGCGGCGAAAGGAAG
GCTTAAAGCCAAGCGACTCTG GTCGATGTCATTGAGTGCTCC TTCGGACCGCATGGGACGGAC
OPRM1a
227
a PCR product of OPRM1 was subjected to RFLP analysis for detecting A118G polymorphism.
Genotype rs16918875 TT TC + CC Fisher's exact test Odd ratio Power rs963549 CC CT + TT Fisher's exact test Odd ratio Power rs702764 GG GA Fisher's exact test Odd ratio Power
Control
Heroin
Addiction alcohol
178 22
110 20 P = 0.3104 1.471 (0.767 to 2.820) 0.221
94 16 P = 0.3705 1.377 (0.690 to 2.748) 0.157
162 38
112 18 P = 0.234 0.685 (0.372 to 1.262) 0.223
88 22 P = 0.8809 1.066 (0.593 to 1.915) 0.057
188 12
126 4 P = 0.2982 0.497 (0.157 to 1.577) 0.221
102 8 P = 0.638 1.229 (0.486 to 3.104) 0.078
Fisher's exact test (2 × 2 table, two tailed), P b 0.05.
228
D. Kumar et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 36 (2012) 225–230
TT allele of rs16918875 [OR = 1.660 (1.006 to 2.745), P = 0.0474]. For rs963549, there is a substantial risk of addiction where its risk alleles CT/TT interacted with the risk allele of A118G in the alcoholics [OR = 2.307 (1.074 to 4.955), P = 0.0321] and not in the heroin users. Surprisingly for rs963549, interactions of its risk alleles, CT/TT with the wild type allele, AA of A118G [OR = 0.305 (0.101 to 0.921), P = 0.0351] as well as interactions of its wild type allele, CC with the risk allele, GG/AG of A118G [OR 1.717 (1.037 to 2.844), P = 0.0357] were found to be significantly associated in heroin addicts. Similarly, studies of the interaction of the SNP, rs702764 with A118G also showed a significant association of its wild type allele, GG with the risk allele GG/AG in both the heroin [OR = 2.05 (1.279 to 3.287), P = 0.0029] as well as in alcohol [OR = 1.737 (1.048 to 2.879), P = 0.0322] users.
4. Discussion
Fig. 2. Pairwise linkage disequilibrium between three SNPs of OPRK1 in the studied population. The pattern of pair-wise LD between SNPs was measured by the standardized disequilibrium value (D′) as implemented in Haploview v4.1 (Barrett et al., 2005). Plot showing the confidence bound color scheme where dark gray indicates strong evidence of LD and light gray suggests uninformative status of LD.
wild AA, heterozygous mutant AG and homozygous mutant GG in heroin addicts was 54%, 41% and 5% respectively; in alcoholic subjects were 54.5%, 37% and 8.5% respectively and in the control population were 68.5%, 27.5% and 4% respectively. Genotype frequencies were significantly different between heroin and control populations (χ2 = 7.268, P = 0.0264), as well as between alcoholic and control populations (χ2 = 6.626, P= 0.0364). It was observed that the heterozygous genotype frequency was highest in heroin addicts and lowest in the controls. The allelic frequencies of A118G for A and G in the different groups were 74% and 26% respectively in heroin population, 73% and 27% respectively in alcoholic population and 82% and 18% respectively in control population (Table 4). The risk allele G of this SNP was more frequent in heroin dependent group [OR = 1.609 (1.102–2.348)] and as well as in alcoholic group [OR = −1.698 (1.146 to 2.517)] as compared to the control group. The heterozygous and homozygous carriers of the risk allele had higher O.R. in heroin 1.886 (1.173–3.032) and 1.713 (0.597 to 4.916) as well as in alcoholic 1.702 (1.027 to 2.822) and 2.569 (0.945 to 6.98) respectively as compared to the wild type genotype. The locus–locus interaction between OPRM1 and OPRK1 is shown in Table 5. There is a marked risk of addiction in subjects where both the risk alleles TC/CC of rs16918875 of OPRK1 and GG/AG of A118G of OPRM1 were present in both heroin [OR = 2.516 (1.152 to 5.495), P = 0.0206] and alcoholic [OR = 2.318 (1.025 to 5.24), P = 0.0433] cases. On the contrary, there is an increase of risk in addiction to only heroin addicts during interaction of GG/AG of A118G and the
Allelic variations in certain critical genes are likely to influence drug dependence in an individual since there is accumulating evidence suggesting that heritable factors play an important role in addictive behavior (Bierut et al., 1988). The present study investigated KOR as a candidate gene for genetic vulnerability to substance abuse. Three SNPs in OPRK1 were identified of which rs963549 is reported to have high FST, a measure of genetic differentiation, suggesting that it had undergone selection, and therefore might be of functional importance (Akey et al., 2002). However, rs963549 did not appear to be associated with drug dependence in the population, which is in agreement with a previous report of a similar lack of association with alcohol dependence in a population of European Americans (Xuei et al., 2006). It has been demonstrated that allelic variation at rs963549 did not moderate the response to the treatment for alcohol dependence in subjects by the opioid antagonist, naltrexone (Gelernter et al., 2007). Moreover, unlike association between alcohol dependence and the multiple SNPs across the large region located in introns of OPRK1, all the SNPs studied in the coding region including rs702764 did not show any risk for alcohol addiction (Xuei et al., 2006). Another SNP, rs702764, detected in the population, also did not show any moderating effect on the reduction of drinking by the opioid antagonist, nalmefene in a population of Caucasians of Finnish origin (Arias et al., 2008). In our study of OPRK1 polymorphism in rs16918875, we did not find any homozygous mutant as observed in the European and Asian population HAPMAP data (NCBI dbSNP). While we observed the presence of heterozygous mutants of rs16918875 in both cases and controls, studies in the Asian population did not report presence of either the heterozygous or the homozygous mutant. HAPMAP data also showed that in case of rs963549, the C is a major allele in the case of Asian, European and in the present studied population and it is the risk minor allele reported in sub-Sahara population (NCBI dbSNP). It appears, therefore, that there are considerable differences in the distribution of risk allele frequencies in the different ethnic groups.
Table 4 Distribution of genotype and allele frequencies of A118G (rs1799971) in addicted and control populations. Control (n%)
Heroin (n%)
OR (95%CI)
Alcoholic (n%)
OR (95%CI)
AA AG GG
137(63.5) 55(27.5) 8 (4) χ2HW = O.67 P = 0.410
70(54) 53(41) 7(5) χ2HW = 0.560 P = 0.454
1 (referent) 1.886 (1.173 to 3.032) 1.713 (0.597 to 4.916) Ptrend = 0.0138⁎ χ2 = 7.268 P = 0.0264⁎
60 (54.5) 41 (37) 9 (8.5) χ2HW = 0.279 P = 0.596
1 (referent) 1.702 (1.027 to 2.822) 2.569 (0.945 to 6.98) Ptrend = 0.0101⁎ χ2 = 6.626 P = 0.0364⁎
A G
329(82) 71(18)
193(74) 67(26)
Power = 0.766 0.622 (0.426 to 0.907) 1.609 (1.102 to 2.348)
161(73) 59(27)
χ2(1), Chi-square analysis of genotype frequency. OR, odds ratio; CI, 95% confidence interval; n, number of individuals/n, number of alleles; χ2HW , Hardy–Weinberg-Equilibrium calculated by χ2test. ⁎ Indicates significance.
Power = 0.683 0.589 (0.397–0.873) 1.698 (1.146–2.517)
D. Kumar et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 36 (2012) 225–230
229
Table 5 Locus–locus interactions between OPRK1 and OPRM11 on addiction risk. Genotype
Genotype
OPRM1 AA AA GG/AG GG/AG OPRM1 AA AA GG/AG GG/AG OPRM1 AA AA GG/AG GG/AG
rs16918875 TT TC/CC TT TC/CC rs963549 CC CT/TT CC CT/TT rs702764 GG GA/AA GG GA/AA
Control
Cases (heroin)
OR
Cases (alcoholic)
OR
127 9 51 13
66 3 44 17
1 (referent) 0.641 (0.168 to 2.45) 1.660 (1.006 to 2.745) 2.516 (1.152 to 5.495)
P = 0.5160 P = 0.0474 P = 0.0206
59 2 35 14
1 (referent) 0.478 (0.1 to 2.283) 1.477 (0.867 to 2.507) 2.318 (1.025 to 5.24)
P = 0.3551 P = 0.1487 P = 0.0433
114 23 48 15
65 4 47 14
1 (referent) 0.305 (0.101 to 0.921) 1.717 (1.037 to 2.844) 1.637 (0.743 to 3.605)
P = 0.0351 P = 0.0357 P = 0.2212
56 5 32 17
1 (referent) 0.443 (0.156 to 1.226) 1.357 (0.783 to 2.352) 2.307 (1.074 to 4.955)
P = 0.1168 P = 0.2764 P = 0.0321
134 5 54 7
69 0 57 4
1 (referent) 0.176 (0.01 to 3.228) 2.05 (1.279 to 3.287) 1.11 (0.314 to 3.922)
P = 0.2418 P = 0.0029 P = 0.8716
60 1 42 7
1 (referent) 0.447 (0.051 to 3.906) 1.737 (1.048 to 2.879) 2.233 (0.750 to 6.649)
P = 0.4663 P = 0.0322 P = 0.1489
Odds ratios (ORs) and 95% CIs for the different genotype combinations are shown. P in bold demonstrates significantly different (b 0.05) from controls.
Significant advances have been made during the past few decades in determining the molecular genetic basis for human inherited diseases. A plethora of genes have been identified, mutations of which have been linked with specific clinical conditions. Such mutations are generally associated with disorders that demonstrate Mendelian inheritance characterized by “single-gene” disorders although in many cases, the genotype–phenotype correlations due to the mutation have proved elusive (Beaudet et al., 2001; Scriver, 2002). Various reasons like reduced penetrance of the mutant gene, contributions of important genetic modifiers for the expression of the affected genes, have been attributed to this lack of correlation (Dipple and McCabe, 2000; Haldane, 1949; Scriver and Waters, 1999). This is more glaring for complex diseases where the genetic etiology is indicative of multifactorial syndrome. Contributions of gene–gene and gene–environment interactions for the variable phenotypic outcome are indicative from various studies. Recent reports indicate a digenic inheritance due the association of pten-induced kinase 1 (PINK1) and DJ-1 for early-onset Parkinson's disease (Tang et al., 2006). Cell transfection studies demonstrate that DJ-1 and PINK1 physically associate and collaborate to protect cells against stress via complex formation suggesting that DJ-1 and PINK1 are genetically linked as well as physically associated. In patients suffering from retinitis pigmentosa (RP), an association of the Asn306Ser variant of the SP4 transcription factor and an intronic variant in the beta-subunit of transducin has been reported (Gao et al., 2007). Families have been identified with mutations in the unlinked photoreceptor-specific genes ROM1 and peripherin/RDS, in which only double heterozygotes develop RP (Kajiwara et al., 1994). In RP it is apparent that a mutation in a second gene is necessary for the disease phenotype whereby the two mutants combine to prevent the formation of complexes which affect the integrity of photoreceptors. Some of the other digenic inheritance reported are for diseases like deafness (Adato et al., 1999) as well as SNP–SNP interaction in breast cancer (Onay et al., 2006), prostate cancer (Lin et al., 2008; Beuten et al., 2009.), type 2 diabetes (Keshavarz et al., 2008; Qu et al., 2011) etc. However, there are also contrary reports showing lack of clinical significance and digenic inheritance due to mutations of more than one gene (Azmanov et al., 2011; Vilella et al., 2008). Nevertheless, all these studies demonstrate that the presence of mutations in one familial gene should not serve as exclusion criteria for the screening for further genetic variation. Likewise, addiction is also a complex, multifactorial disease and the hypothesis that multiple genes are involved in the predisposition of addiction is well supported by understanding the biology of addiction. OPRK1 and OPRM1 are both involved in addiction pathway but their functions are opposite in nature. In gene–gene co-association study, LD based statistic has suggested that SNP–SNP interactions in OPRK1 and OPRM1 were significant in heroin dependent subjects (Peng et al., 2010). The present study of OPRK1 polymorphisms and their interaction
with A118G of the OPRM1 gene is perhaps the first case–control study in an Indian population. Interestingly, increased risk of addiction was generally associated with those genotype combinations where one or both the risk alleles were present. In almost all the risk combinations, the OPRM1 risk allele GA/GG was invariably present suggesting an important contribution of OPRM1 in the digenic inheritance for addiction. Opioids regulate a variety of brain functions related to addictions like reward, reinforcement, mood and psychomotor stimulation. While the biological basis of mood is not well understood, KOR systems tend to compensate for the continued exposure to drugs acting at MOR and have been hypothesized to underlie the protracted abstinence syndrome, with dysphoric mood in opioid-dependent individuals (Rothman, 1992). Treatment with selective KOR agonists is reported to cause a psychotomimetic effect and dysphoria, both in clinical studies and experimental animal models (Hasebe et al., 2004) while KOR agonists may have a potential role in the treatment of mood disorders (Carlezon et al., 2009). The lack of depressed patients, in the present study, may have been influenced by the high rate of KOR polymorphisms in the samples. Since A118G polymorphism is expected to undermine reward experience in the cases, possibilities that bad mood, induced by KOR variants, becoming problematic only when A118G mutant is present, cannot be ruled out. There are some limitations in this study. Firstly, the sample size is moderate and there is a need for confirmation of these data in larger populations. Secondly, we investigated only interactions of two genes, but a large number of candidate genes have been implicated in the addiction pathway. Thirdly, the participants in the study were males only, the underlying reason may be the unavailability of female volunteers in such conservative society. This may result in gender bias. Lastly, after multiple testing the interaction between OPRM1 (GG/AG) and rs702764 remained significant while all other results lost their significance (See Supplementary Table for details). In summary, this study provides evidence that gene–gene interaction between KOR and OPRM1 can influence the risk of addiction to narcotics and alcohol. Results also suggest a disease association of rare genotypes which may otherwise be missed in the absence of interactions with known risk alleles. Our approach of evaluating moderate sample size is one of the first efforts in explaining the effect of gene– gene interaction in Indian population and might be helpful in improving the ability and accuracy in assessing individual risk towards addiction. Supplementary materials related to this article can be found online at doi:10.1016/j.pnpbp.2011.10.018. Acknowledgments The authors thank Mr. Pinaki Mondal, Senior Research Fellow, Cancer and Cell Biology Division, Indian Institute of Chemical Biology, Kolkata for his help in the statistical analysis of the data.
230
D. Kumar et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 36 (2012) 225–230
References Adato A, Kalinski H, Weil D, Chaib H, Korostishevsky M, Bonne-Tamir B. Possible interaction between USH1B and USH3 gene products as implied by apparent digenic deafness inheritance. Am J Hum Genet 1999;65:261–5. Akey JM, Zhang G, Zhang K, Jin L, Shriver MD. Interrogating a high-density SNP map for signatures of natural selection. Genome Res 2002;12:1805–14. Arias AJ, Armeli S, Gelernter J, Covault J, Kallio A, Karhuvaara S, et al. Effects of opioid receptor gene variation on targeted nalmefene treatment in heavy drinkers. Alcohol Clin Exp Res 2008;32:1159–66. Azmanov DN, Dimitrova S, Florez L, Cherninkova S, Draganov D, Morar B, et al. LTBP2 and CYP1B1 mutations and associated ocular phenotypes in the Roma/Gypsy founder population. Eur J Hum Genet 2011;19:326–33. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005;21:263–5. Beaudet A, Scriver C, Sly W, Valle D. Genetics, biochemistry, and molecular bases of variant human phenotypes. In: Scriver C, Beaudet A, Sly W, Valle D, editors. The metabolic and molecular bases of inherited disease. 8th ed. New York: McGraw-Hill; 2001. p. 3-45. Beuten J, Garcia D, Brand TC, He X, Balic I, Canby-Hagino E, et al. Semaphorin 3B and 3F single nucleotide polymorphisms are associated with prostate cancer risk and poor prognosis. J Urol 2009;182(4):1614–20. Bhargava HN. Diversity of agents that modify opioid tolerance, physical dependence, abstinence syndrome, and self-administrative behaviour. Pharmacol Rev 1994;46:293–324. Bierut LJ, Dinwiddie SH, Begleiter H, Crowe RR, Hesselbrock V, Nurnberger Jr JI, et al. Familial transmission of substance dependence: alcohol, marijuana, cocaine, and habitual smoking: a report from the Collaborative Study on the Genetics of Alcoholism. Arch Gen Psychiatry 1988;55:982–8. Bolanos CA, Garmsen GM, Clair MA, McDougall SA. Effects of the kappa-opioid receptor agonist U-50,488 on morphine-induced place preference conditioning in the developing rat. Eur J Pharmacol 1996;317:1–8. Carlezon Jr WA, Béguin C, Knoll AT, Cohen BM. Kappa-opioid ligands in the study and treatment of mood disorders. Pharmacol Ther 2009;123:334–43. Cowan A, Zhu XZ, Mosberg HI, Omnaas JR, Porreca F. Direct dependence studies in rats with agents selective for different types of opioid receptor. J Pharmacol Exp Ther 1998;246:950–5. Deb I, Chakraborty J, Gangopadhyay PK, Choudhury SR, Das S. Single-nucleotide polymorphism (A118G) in exon 1 of OPRM1 gene causes alteration in downstream signaling by mu-opioid receptor and may contribute to the genetic risk for addiction. Neurochem 2010;112:486–96. Dipple KM, McCabe ER. Phenotypes of patients with “simple” Mendelian disorders are complex traits: thresholds, modifiers, and systems dynamics. Am J Hum Genet 2000;266:1729–35. Dupont WD, Plummer Jr WD. Power and sample size calculations for studies involving linear regression. Control Clin Trials 1998;19:589–601. Edenberg HJ, Wang J, Tian H, Pochareddy S, Xuei X, Wetherill L, et al. A regulatory variation in OPRK1, the gene encoding the κ-opioid receptor, is associated with alcohol dependence. Hum Mole Genet 2008;17:1783–9. Friedman HJ, Jen MF, Chang JK, Lee NM, Loh HH. Dynorphin: a possible modulatory peptide on morphine or beta-endorphin analgesia in mouse. Eur J Pharmacol 1981;69:357–60. Funada M, Suzuki T, Narita M, Misawa M, Nagase H. Blockade of morphine reward through the activation of kappa-opioid receptors in mice. Neuropharmacology 1993:1315–23. Gao YQ, Danciger M, Ozgul RK, Gribanova Y, Jacobson S, Farber DB. Association of the Asn306Ser variant of the SP4 transcription factor and an intronic variant in the beta-subunit of transducin with digenic disease. Mol Vis 2007;13:287–92. Gelernter J, Gueorguieva R, Kranzler HR, Zhang H, Cramer J, Rosenheck R, et al. VA Cooperative Study #425 Study Group. Opioid receptor gene (OPRM1, OPRK1, and OPRD1) variants and response to naltrexone treatment for alcohol dependence: results from the VA Cooperative Study. Alcohol Clin Exp Res 2007;31:555–63. Gerra G, Leonardi C, Cortese E, D'Amore A, Lucchini A, Strepparola G, et al. Human kappa opioid receptor gene (OPRK1) polymorphism is associated with opiate addiction. Am J Med Genet B Neuropsychiatr Genet 2007;144:771–5. Glick SD, Maisonneuve IM, Raucci J, Archer S. Kappa opioid inhibition of morphine and cocaine self-administration in rats. Brain Res 1995;681:147–52. Haldane JBS. Disease and evolution. Ric Sci Suppl 1949;19:3-10. Hasebe K, Kawai K, Suzuki T, Kawamura K, Tanaka T, Narita M, et al. Possible pharmacotherapy of the opioid kappa receptor agonist for drug dependence. Ann N Y Acad Sci 2004;1025:404–13. Kajiwara K, Berson EL, Dryja TP. Digenic retinitis pigmentosa due to mutations at the unlinked peripherin/RDS and ROM1 loci. Science 1994;264:1604–8.
Keshavarz P, Inoue H, Nakamura N, Yoshikawa T, Tanahashi T, Itakura M. Single nucleotide polymorphisms in genes encoding LKB1 (STK11), TORC2 (CRTC2) and AMPK alpha2-subunit (PRKAA2) and risk of type 2 diabetes. Mol Genet Metab 2008;93: 200–9. Kuzmin AV, Semenova S, Gerrits MA, Zvartau EE, Van Ree JM. Kappa-opioid receptor agonist U50,488H modulates cocaine and morphine self-administration in drugnaive rats and mice. Eur J Pharmacol 1997;321:265–71. Lin HY, Wang W, Liu YH, Soong SJ, York TP, Myers L, et al. Comparison of multivariate adaptive regression splines and logistic regression in detecting SNP–SNP interactions and their application in prostate cancer. J Hum Genet 2008;53:802–11. Loh el W, Fann CS, Chang YT, Chang CJ, Cheng AT. Endogenous opioid receptor genes and alcohol dependence among Taiwanese Han. Alcohol Clin Exp Res 2004;28: 15–9. Matthes HW, Maldonado R, Simonin F, Valverde O, Slowe S, Kitchen I, et al. Loss of morphine-induced analgesia, reward effect and withdrawal symptoms in mice lacking the mu-opioid-receptor gene. Nature 1996;383:819–23. Onay VU, Briollais L, Knight JA, Shi E, Wang Y, Wells S, et al. SNP–SNP interactions in breast cancer susceptibility. BMC Cancer 2006;6:114. Peng Q, Zhao J, Xue F. A gene-based method for detecting gene–gene co-association in a case–control association study. Eur J Hum Genet 2010;18:582–7. Qu L, He B, Pan Y, Xu Y, Zhu C, Tang Z, et al. Association between polymorphisms in RAPGEF1, TP53, NRF1 and type 2 diabetes in Chinese Han population. Diabetes Res Clin Pract 2011;91:171–6. Ramarao P, Jablonski Jr HI, Rehder KR, Bhargava HN. Effect of kappa-opioid receptor agonists on morphine analgesia in morphine-naive and morphine-tolerant rats. Eur J Pharmacol 1988;156:239–46. Reisine T, Pasternak G. Opioid analgesics and antagonists. In: Hardman JG, Limbird LE, editors. Goodman & Gilman's the pharmacological basis of therapeutics. 9 ed. McGraw-Hill; 1996. Rothman RB. A review of the role of anti-opioid peptides in morphine tolerance and dependence. Synapse 1992;12:129–38. Saito M, Ehringer MA, Toth R, Oros M, Szakall I, Sikela JM, et al. Variants of kappaopioid receptor gene and mRNA in alcohol-preferring and alcohol-avoiding mice. Alcohol 2003;29:39–49. Scriver CR. Why mutation analysis does not always predict clinical consequences: explanations in the era of genomics. J Pediatr 2002;140:502–6. Scriver CR, Waters PJ. Monogenic traits are not simple: lessons from phenylketonuria. Trends Genet 1999;15:267–72. Shippenberg TS, Rea W. Sensitization to the behavioral effects of cocaine: modulation by dynorphin and kappa-opioid receptor agonists. Pharmacol Biochem Behav 1997;57:449–55. Sora I, Takahashi N, Funada M, Ujike H, Revay RS, Donovan DM, et al. Opiate receptor knockout mice define mu receptor roles in endogenous nociceptive responses and morphine-induced analgesia. Proc Natl Acad Sci 1997;94:1544–9. Suzuki T, Narita M, Takahashi Y, Misawa M, Nagase H. Effects of nor-binaltorphimine on the development of analgesic tolerance to and physical dependence on morphine. Eur J Pharmacol 1992;213:91–7. Tang B, Xiong H, Sun P, Zhang Y, Wang D, Hu Z, et al. Association of PINK1 and DJ-1 confers digenic inheritance of early-onset Parkinson's disease. Hum Mol Genet 2006;15: 1816–25. Tao PL, Hwang CL, Chen CY. U-50,488 blocks the development of morphine tolerance and dependence at a very low dose in guinea pigs. Eur J Pharmacol 1994;256: 281–6. Tulunay FC, Jen MF, Chang JK, Loh HH, Lee NM. Possible regulatory role of dynorphin on morphine- and beta-endorphin-induced analgesia. J Pharmacol Exp Ther 1981;219: 296–8. Vilella E, Costas J, Sanjuan J, Guitart M, Diego YD, Carracedo A, et al. Association of schizophrenia with DTNBP1 but not with DAO, DAOA, NRG1 and RGS4 nor their genetic interaction. J Psychiatr Res 2008;42:278–88. Xuei X, Flury-Wetherill L, Bierut L, Dick D, Nurnberger Jr J, Foroud T, et al. The opioid system in alcohol and drug dependence: family-based association study. Am J Med Genet B Neuropsychiatr Genet 2007;144:877–84. Xuei X, Dick D, Flury-Wetherill L, Tian HJ, Agrawal A, Bierut L, et al. Association of the kappa-opioid system with alcohol dependence. Mol Psychiatry 2006;11:1016–24. Yuferov V, Fussell D, LaForge KS, Nielsen DA, Gordon D, Ho A, et al. Redefinition of the human kappa opioid receptor gene (OPRK1) structure and association of haplotypes with opiate addiction. Pharmacogenet 2004;14:793–804. Zadina JE, Hackler L, Ge LJ, Kastin AJ. A potent and selective endogenous agonist for the mu-opiate receptor. Nature 1997;386:499–502. Zhang Y, Butelman ER, Schlussman SD, Ho A, Kreek MJ. Effects of the kappa opioid agonist R-84760 on cocaine-induced increases in striatal dopamine levels and cocaineinduced place preference in C57BL/6j mice. Psychopharmacol 2004;173:146–52.