RESEARCH ARTICLE Neuropsychiatric Genetics Association Between CYP2C19 Polymorphism and Depressive Symptoms Sarah C. Sim,1* Linn Nordin,1 Therese M.-L. Andersson,2 Susanne Virding,1 Marita Olsson,2,3 and Nancy L. Pedersen2 1 Section of Pharmacogenetics, Department of Physiology and Pharmacology, Karolinska Institutet, Stockholm, Sweden Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden 2 3 Department of Mathematical Sciences, Chalmers University of Technology, Gothenburg, Sweden Received 20 May 2009; Accepted 10 February 2010 Cytochrome P450 2C19 (CYP2C19) is a polymorphic enzyme active in the metabolism of for example diazepam and the antidepressants sertraline, citalopram, and escitalopram, whereby allelic variants cause increased (CYP2C19*17) or abolished (mainly CYP2C19*2) enzymatic activity in drug metabolism. In light of the importance of CYP2C19 in the metabolism of psychoactive substances we considered it of interest to investigate the relationship between CYP2C19 polymorphisms and depressive symptoms in 1,472 subjects of European ancestry (45–98 years old) from the Swedish Twin Registry. Depressive symptoms were assessed using the Center of Epidemiologic Studies Depression (CES-D) scale. We found that poor metabolizers lacking CYP2C19 activity (PMs, CYP2C19*2/*2) had signiﬁcantly lower levels of depressive symptoms than extensive metabolizers (EMs, CYP2C19*1/*1) (P ¼ 0.0018). The size of this difference was in the same range as that between subjects reported taking antidepressants (n ¼ 104) and those without antidepressant treatment (P < 0.0001). Our results suggest for the ﬁrst time that the CYP2C19 polymorphism might be of importance for depressive symptoms, as here shown for older European adults. 2010 Wiley-Liss, Inc. Key words: neurosteroids; serotonin; depressed mood INTRODUCTION Cytochrome P450s (CYPs) compose a large family of phase I metabolizing enzymes active in the metabolism of clinically used drugs, other xenobiotics and endogenous substances. Most of these P450s are polymorphic, giving rise to a wide spectrum of metabolizing phenotypes ranging from ultra-rapid metabolizers (UMs), through extensive metabolizers (EMs) to poor metabolizers (PMs), who lack enzymatic activity. Due to the polymorphisms, there are large interindividual as well as interethnic differences in the metabolism of different cytochrome P450 substrates [Ingelman-Sundberg et al., 2007]. CYP2C19 is active in the metabolism of clinically used drugs, several of which are psychoactive such as the antidepressants sertraline [Rudberg et al., 2008a], escitalopram [Rudberg et al., 2008b] and citalopram [Yin et al., 2006], as well as diazepam [Fukasawa et al., 2010 Wiley-Liss, Inc. How to Cite this Article: Sim SC, Nordin L, Andersson TM-L, Virding S, Olsson M, Pedersen NL, IngelmanSundberg M. 2010. Association Between CYP2C19 Polymorphism and Depressive Symptoms. Am J Med Genet Part B 153B:1160–1166. 2007]. There is a large interethnic variation in the frequency of the different CYP2C19 phenotypes: the PM phenotype encompasses 3–5% of Swedes, 20% in Japanese and is as high as 79% on the island of Vanuatu in the Paciﬁc Ocean [Desta et al., 2002]. In Europeans and Asians, the defective CYP2C19*2 allele is responsible for the majority of the PM phenotypes, whereas in Asia, the defective CYP2C19*3 allele also contributes [Desta et al., 2002]. The CYP2C19*17 allele is associated with a rapid metabolizer (RM) phenotype in vivo for drugs such as omeprazole and escitalopram and has an allele frequency of 18–27% in different European populations [Kurzawski et al., 2006; Sim et al., 2006; Schroth et al., 2007; Baldwin et al., 2008; Ohlsson Rosenborg et al., 2008; Rudberg et al., 2008b], thus giving rise to a homozygous RM genotype frequency of 3–7%. Recently, the relationship between CYP2C19 polymorphism and personality traits, as measured by the Temperament and Character Inventory (TCI), has been investigated in Japanese healthy volunteers [Ishii et al., 2007; Yasui-Furukori et al., 2007]. Ishii et al.  Additional Supporting Information may be found in the online version of this article. Sarah C. Sim and Linn Nordin contributed equally to this study. *Correspondence to: Dr. Sarah C. Sim, Ph.D., Section of Pharmacogenetics, Department of Physiology and Pharmacology, Karolinska Institutet, SE-171 77 Stockholm, Sweden. E-mail: firstname.lastname@example.org Published online 7 April 2010 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/ajmg.b.31081 1160 SIM ET AL. found a signiﬁcant difference in three of the seven dimensions of the TCI between female genotypic CYP2C19 EMs and PMs, but no signiﬁcant difference when comparing males. Speciﬁcally, the female PMs had signiﬁcantly lower scores on reward dependence, cooperativeness and self-transcendence as compared to the EMs. Yasui-Furukori et al.  found a statistically signiﬁcant higher score in harm avoidance in genotypic PMs as compared to EMs without stratifying for sex. Thus, these data might possibly indicate an endogenous psychoactive function for CYP2C19, although these studies need to be reproduced in other Asian populations and investigated in Europeans. Because of the involvement of CYP2C19 in the metabolism of psychoactive drugs and in personality phenotypes, we investigated the relationship between CYP2C19 genotypes resulting in CYP2C19 RMs, EMs, and PMs and depressive symptoms using a large subset of elderly Swedish subjects from the Swedish Twin Registry (STR). As measured by CES-D, we found a signiﬁcantly lower level of depressive symptoms in PMs (CYP2C19*2/*2) as compared to EMs (CYP2C19*1/*1), with a stronger relationship in men and in the younger age group. Thus, our results might suggest that CYP2C19 polymorphism inﬂuences the level of depressive symptoms in European adults. MATERIALS AND METHODS Subjects In total, 1,555 subjects (nmen ¼ 653, nwomen ¼ 902) from the STR were included in the study. The participants originated from three different longitudinal studies of aging, which have all been described elsewhere: Origins of Variance in the Oldest Old/Octogenarian Twins (OCTO-Twin) [McClearn et al., 1997], Swedish Adoption/Twin Study of Aging (SATSA) [Pedersen et al., 1991; Finkel and Pedersen, 2004], and Gender [Gold et al., 2002]. At the time of blood sampling, the age distribution of our study population was 45–98 years with a mean age of 73 years (71 for men and 74 for women). Subjects taking antidepressant drugs were identiﬁed at the time of ﬁlling in questionnaires (SATSA, 1/3 of the material) or through review of medical records covering the last approximately 15 years (OCTOTwin and Gender, 2/3) [Jansson et al., 2003]. All drugs with the Anatomical Therapeutic Chemical (ATC) code N06A were classiﬁed as antidepressants. All participants were of European ancestry and born in Sweden. The study was approved by the Ethics Committee at Karolinska Institutet and the Swedish Data Inspection Board. All participants provided informed consent. Genotyping DNA samples were genotyped using the TaqMan Drug Metabolism Genotyping Assays for the CYP2C19*2 (C__25986767_70) and the CYP2C19*17 allele (C__469857_10) (Applied Biosystems, CA, Foster City). One microliter of genomic DNA (undetermined concentration) was included in a 5 ml reaction volume with PCR conditions as recommended by the manufacturer. The samples were run on a 7500 Real Time PCR System (Applied Biosystems) and the data were analyzed using the Sequence Detection Software 1161 version 1.3.1 (Applied Biosystems). Subjects with undetermined CYP2C19 alleles either from the CYP2C19*2 or CYP2C19*17 genotyping assays were excluded. Subjects were divided into the following genotype categories: PMs, CYP2C19*2/*2 (n ¼ 35); heterozygous EMs, CYP2C19*1/*2 (n ¼ 290); homozygous EMs, CYP2C19*1/*1 (n ¼ 637); heterozygous RMs, CYP2C19*1/*17 (n ¼ 380); homozygous RMs, CYP2C19*17/*17 (n ¼ 45); and CYP2C19*2/*17 (n ¼ 85). Assessment of Depressive Symptoms Depressive symptoms were self-reported using the Swedish translation of the Center of Epidemiologic Studies Depression Scale (CES-D) [Gatz et al., 1993]. This 20-item scale assesses the level of depressive symptom during the past week, but the CES-D scores are relatively stable over time [Fiske et al., 2003]. The CESD scale consists of four sub-scales that together sum to a total score (T1): depressed mood, psychomotor retardation and somatic complaints, well-being, and interpersonal difﬁculties. Each sub-scale is assessed by a number of questions with scores ranging from 0 to 3 with a maximal total score of 60 where higher total scores indicate higher levels of depressive symptoms [Radloff, 1977]. For the present analyses, only the total scale and the subscales ‘‘psychomotor retardation and somatic complaints,’’ and ‘‘depressed mood’’ were evaluated. Average CES-D scores were used for the 84% of the participants who had answered the test more than once at 2- to 4-year intervals [Jansson et al., 2003]. Information on antidepressant use came from two sources: medical records and self-report of the trade name of all medications taken during the last month which were then coded into the ATC system. All drugs with the ATC code N06A were classiﬁed as antidepressive (antidep). Use of antidepressants was included as a covariate, in order to be able to evaluate the effect of genotype on depressive symptoms independent of any confounding effect of the antidep variable. Statistics The association between the CES-D scores and the genotypic information was analyzed using linear regression. To account for dependence within the data due to including related individuals (twin pairs), we used a Generalized Estimating Equation (GEE) approach with an exchangeable correlation structure. Estimated effects of covariates are expressed as change in mean score. The main models, including males and females of all ages, were adjusted for sex and antidepressant drug intake as binary variables and age as a linear variable. The age and sex stratiﬁed models were adjusted for the same variables as the main models, except for age and sex, respectively. In the statistical analyses, everyone with missing genotype information or the outcome variable for that speciﬁc model was excluded. The study population used for statistical analyses included both members of a twin pair (monozygotic (MZ) females n ¼ 158, MZ males n ¼ 104, dizygotic (DZ) females n ¼ 362, DZ males n ¼ 702) as well as singletons (MZ females n ¼ 46, MZ males n ¼ 33, DZ females n ¼ 102, and DZ males n ¼ 48). All statistical analyses were performed in 1162 AMERICAN JOURNAL OF MEDICAL GENETICS PART B SAS 9.1 using the GENMOD procedure (SAS Institute, Cary, NC). RESULTS Out of 1,555 DNA samples, 1,472 (nmen ¼ 622, nwomen ¼ 850) were successfully genotyped for both CYP2C19*2 and CYP2C19*17. The genotype frequencies were in Hardy–Weinberg equilibrium. Due to missing information on some items, the ﬁnal numbers of participants were 1,416 for the total CES-D score (T1), 1,426 for psychomotor retardation and somatic complaints (SA), and 1,425 for depressed mood (DA). Individuals on antidepressants (antidep, n ¼ 104) had signiﬁcantly higher T1, SA and DA scores compared to subjects that had no antidepressant record (P < 0.0001, Table I, Fig. 1). Increasing age was associated with decreased T1 scores (P < 0.0001, Table I), and males had signiﬁcantly lower levels of depressive symptoms than females (P ¼ 0.0009, Table I). The same results were found in the SA and DA sub-scales, but the magnitude of the differences were smaller (Table I). In the entire study population, the CYP2C19*2/*2 genotype, which confers a PM phenotype, had signiﬁcantly lower T1 scores, and thus a lower level of depressive symptoms, as compared to the reference genotype CYP2C19*1/*1 with an EM phenotype (P ¼ 0.0018, Table I, Fig. 1). The SA and DA sub-scales displayed similar patterns of differences between genotypes as the T1 scale (Table I). The study population used for the analyses in Table I (as well as Tables II and III) included both members of a twin pair (monozygotic (MZ) n ¼ 262, dizygotic (DZ) n ¼ 1,064) as well as singletons, only one twin in a pair (MZ ¼ 79, DZ ¼ 150, see the Materials and Methods Section for further information on sex distribution). We performed the same analyses with only one member of a pair selected at random (i.e., n ¼ 815–820 depending on the scale tested) and the results were analogous: the magnitudes of the parameters were very similar, and because of the smaller sample, the P-values were somewhat increased but remained signiﬁcant (the only exceptions being lack of signiﬁcance for the DA sub-scale with CYP2C19*2/*2 and age with P-values of 0.07 and 0.19, respectively). After stratifying for sex, the same pattern of difference in the T1 scores between CYP2C19 genotype groups was found in the males (Table II), but not in the females (data not shown). Thus, in the males, the CYP2C19*2/*2 genotype group showed signiﬁcantly lower T1 values compared to the CYP2C19*1/*1 genotype group (Table II, Fig. 1). Noteworthy, the size of the difference in T1 scores between the antidep group and subjects not treated with antidepressant drugs was higher in the male study population (Table II) as compared to the mixed study population (Table I). The results from analyses of the sub-scales SA and DA were similar to that of the T1 total score but with smaller differences (Table II). Furthermore, we stratiﬁed the study population for age, dividing the data set into two age groups, a younger (mean age (range) ¼ 65 (45–72) years, nmen ¼ 344, nwomen ¼ 348) and an older group (mean age (range) ¼ 80 years (73–98), nmen ¼ 264, nwomen ¼ 460). The CES-D scores differed by genotype in the younger subgroup (Table III, Fig. 1) but not the older subgroup (data not shown). The magnitude of difference in T1 scores between CYP2C19*2/*2 and CYP2C19*1/*1 was greater in the younger subgroup as compared to the entire study population. As was also observed in the entire study population, males in the younger age group displayed a lower level of depressive symptoms than females, and subjects on antidepressant medication scored higher than those without (Table III). Again, the difference in T1 scores between genotype groups in the younger age group was also observed for the sub-scales SA and DA (Table III). DISCUSSION The present study investigated the relationship of CYP2C19 polymorphism with depressive symptoms in 1,472 subjects with Euro- TABLE I. CES-D Scores for the Entire Study Population CES-D total score: T1 Category CYP2C19*2/*2 CYP2C19*1/*2 CYP2C19*2/*17 CYP2C19*1/*1 CYP2C19*1/*17 CYP2C19*17/*17 Antidep Age Male Female n 33 277 84 613 365 44 91 — 608 808 Estimate 2.9743 0.2155 0.0639 (Ref) 0.1550 1.5395 4.2803 0.1071 1.2497 (Ref) Standard error P-value n 0.9529 0.0018 33 0.4904 0.6603 279 0.8520 0.9402 84 (Ref) (Ref) 617 0.4733 0.7433 369 1.0720 0.1510 44 1.0964 <0.0001 95 0.0219 <0.0001 — 0.3754 0.0009 611 (Ref) (Ref) 815 CES-D sub-scale: SA (somatic complaints) Estimate 1.0069 0.1129 0.1416 (Ref) 0.2697 0.4742 1.8159 0.0203 0.5496 (Ref) Standard error P-value n 0.3372 0.0028 33 0.1759 0.5208 279 0.2924 0.6281 84 (Ref) (Ref) 617 0.1638 0.0996 368 0.2988 0.1126 44 0.3764 <0.0001 95 0.0071 0.0043 — 0.1312 <0.0001 611 (Ref) (Ref) 814 CES-D sub-scale: DA (depressed mood) Estimate 0.7057 0.0561 0.2146 (Ref) 0.0491 0.3908 1.3534 0.0193 0.7551 (Ref) Standard error P-value 0.2858 0.0135 0.1610 0.7275 0.2795 0.4426 (Ref) (Ref) 0.1584 0.7564 0.3335 0.2412 0.3260 <0.0001 0.0073 0.0084 0.1241 <0.0001 (Ref) (Ref) Estimates are deviations from the intercept (reference group). The CYP2C19 genotype categories were analyzed with CYP2C19*1/*1 as the reference group. The antidep line compares subjects reported to take antidepressant medication with subjects not taking antidepressant drugs (the reference group). Age is presented as unit difference per extra year of age, and males were compared with females. Since the statistical models contained several variables (CYP2C19 genotype, antidep, age, and sex), each variable analyzed was adjusted for the other variables. P-values in bold refer to values below 0.05 and are considered statistically signiﬁcant. SIM ET AL. 1163 TABLE II. CES-D Scores for Males, All Ages CES-D sub-scale: DA (somatic complaints) CES-D total score: T1 Category CYP2C19*2/*2 CYP2C19*1/*2 CYP2C19*2/*17 CYP2C19*1/*1 CYP2C19*1/*17 CYP2C19*17/*17 Antidep Age n 17 123 31 260 160 17 30 — Estimate Standard error 4.6120 1.0378 0.3330 0.6810 0.8579 1.2719 (Ref) (Ref) 0.1994 0.6259 2.1727 1.2971 6.0744 1.9360 0.0553 0.0298 P-value <0.0001 0.6249 0.5000 (Ref) 0.7500 0.0939 0.0017 0.0636 n 17 124 31 261 161 17 31 — Estimate Standard error 1.3509 0.4047 0.0563 0.2518 0.2328 0.3974 (Ref) (Ref) 0.0961 0.2326 0.3379 0.4000 2.3364 0.7024 0.0044 0.0102 CES-D sub-scale: SA (depressed mood) P-value 0.0008 0.8231 0.5580 (Ref) 0.6795 0.3983 0.0009 0.6659 n 17 124 31 261 161 17 31 — Estimate Standard error 0.9433 0.3138 0.1506 0.1940 0.4362 0.4299 (Ref) (Ref) 0.1564 0.1959 0.5931 0.3268 1.5282 0.5746 0.0037 0.009 P-value 0.0026 0.4377 0.3103 (Ref) 0.4246 0.0696 0.0078 0.6784 Estimates are deviations from the intercept (reference group). The CYP2C19 genotype categories were analyzed with CYP2C19*1/*1 as the reference group. For other details, see Table I. P-values in bold refer to values below 0.05 and are considered statistically signiﬁcant. pean ancestry from the Swedish Twin Registry. CYP2C19 PMs had a signiﬁcantly lower level of depressive symptoms than homozygous EMs both when analyzing the total CES-D score (T1) and when analyzing the sub-scales of psychomotor retardation and somatic complaints (SA) and depressed mood (DA). This difference was even more prominent in men and in the younger age group (summarized in Fig. 1). The other genotype groups (including subjects homozygous or heterozygous for CYP2C19*17) did not differ from the reference group (EMs, CYP2C19*1/*1). Interestingly, the differences in depressive symptoms between genotypically classiﬁed PMs and EMs were accentuated in males and in the younger group of twins. This might be due to differences in the etiology of depressive symptom as indicated by previous studies demonstrating different etiologies for clinical depression in younger people as compared to older [Kendler et al., 2007], and in males as compared to females [Kendler et al., 2006]. It is thus possible that the consequences of poor metabolism of CYP2C19 activity (CYP2C19*2/*2) has differential effects in different depressive phenotypes. Our study suggests a link between depressive symptoms and CYP2C19 genetic polymorphism, yet the mechanism is not clear. Today, medical treatment for depression is directed towards manipulating the monoaminergic neurotransmission in the brain, such as that of serotonin. The commonly prescribed selective serotonin reuptake inhibitors (SSRIs) function by increasing the levels of serotonin (5-hydroxytryptamine, 5-HT) in the postsynaptic space. The increased level of 5-HT is associated with relief of the depressive symptoms, although the exact mechanisms are not clear. Thus far, a possible link between CYP2C19 and 5-HT has only been described in vitro, suggesting that CYP2C19 participates in the biotransformation of 5-HT [Fradette et al., 2004]. However, the afﬁnity of CYP2C19 for 5-HT was shown to be very low and the transformation by no means speciﬁc for CYP2C19 since both CYP2B6 and CYP2C9 had identical capacities in this reaction. TABLE III. CES-D Scores for Both Sexes, Younger Age Group CES-D sub-scale: DA (somatic complaints) CES-D total score: T1 Category *2/*2 *1/*2 *2/*17 *1/*1 *1/*17 *17/*17 Antidep Male Female n 15 132 36 317 166 26 21 344 348 Estimate 5.6424 0.3557 0.2359 (Ref) 0.4726 0.3660 7.6179 1.5526 (Ref) Standard error 1.2314 0.6622 1.1338 (Ref) 0.7235 1.0768 2.1366 0.5120 (Ref) P-value <0.0001 0.5912 0.8352 (Ref) 0.5136 0.7340 0.0004 0.0024 (Ref) n 15 133 36 318 167 26 21 346 349 Estimate 1.9557 0.2521 0.2712 (Ref) 0.0487 0.3687 2.7555 0.6281 (Ref) Standard error 0.3877 0.2290 0.3728 (Ref) 0.2468 0.3694 0.7242 0.1785 (Ref) P-value <0.0001 0.2708 0.4670 (Ref) 0.8434 0.3182 0.0001 0.0004 (Ref) CES-D sub-scale: SA (depressed mood) n 15 133 36 318 167 26 21 346 349 Estimate 1.3826 0.0495 0.0007 (Ref) 0.2856 0.4654 2.1718 0.8651 (Ref) Standard error 0.3795 0.2243 0.4244 (Ref) 0.2469 0.3140 0.6679 0.1777 (Ref) P-value 0.0003 0.8254 0.9986 (Ref) 0.2474 0.1382 0.0011 <0.0001 (Ref) Estimates are deviations from the intercept (reference group). The CYP2C19 genotype categories were analyzed with CYP2C19*1/*1 as the reference group. For other details, see Table I. P-values in bold refer to values below 0.05 and are considered statistically signiﬁcant. 1164 FIG. 1. The CES-D total scores (T1) represent deviations from the intercept (reference group). The value of the antidep group (subjects having taken antidepressant treatment) is shown as the deviation from subjects reported not to take antidepressant drugs as the reference group. The CYP2C19*2/*2 subjects are compared with the CYP2C19*1/*1 genotype as the reference group. The graph displays results from analyses using all subjects, the males, and the younger age group, respectively. All differences were statistically signiﬁcant with *P < 0.01, **P < 0.001, ***P < 0.0001. Furthermore, the proposed enzymatic conversion did not follow true Michaelis–Menten kinetics but most probably merely reﬂected oxidation as a result of auto-oxidation by cytochrome P450-generated oxyradicals [Ingelman-Sundberg and Johansson, 1984; Butura et al., 2009]. In addition, there was no selectivity in the action of different inhibitors like ketoconazole and omeprazole in inhibiting the different CYP-mediated 5-HT conversions, indicating that 5HT merely acted as a radical scavenger. However, it is possible that CYP2C19 metabolizes other endogenous compounds that could be of importance for depressive symptoms. In support of this ﬁnding is CYP2C19’s importance for clearance of escitalopram and other antidepressants, whereby a structure–function relationship might hold true for endogenous substrates with similar actions. In vitro studies suggest CYP2C19 is involved in the metabolism of steroid hormone structures such as progesterone, testosterone [Yamazaki and Shimada, 1997], estrone [Cribb et al., 2006], and estradiol [Cheng et al., 2001], thus raising the possibility of a mechanism involving neurosteroids [Pinna et al., 2009]. Today, the presence of CYP2C19 in the human brain remains controversial although presence in the brain may not be necessary for an impact on depressive symptoms. Hepatic CYP2C19 could potentially regulate levels of circulating neuroactive substances such as steroids, which in turn indirectly affect brain function. Furthermore, a genetic element that is associated with depressive symptoms may be in linkage disequilibrium with CYP2C19 (on chromosome 10q24), and thus CYP2C19 per se may not inﬂuence depressive symptoms. Apparently, the molecular link between the CYP2C19 genotype and depressive symptoms has to await further investigation. AMERICAN JOURNAL OF MEDICAL GENETICS PART B As could be expected, subjects on antidepressant treatment (the antidep group) identiﬁed either at the time of CES-D testing or through review of medical records had signiﬁcantly higher CES-D scores than subjects not on this type of medication, both when analyzing the total T1 score and the SA and DA sub-scales. The results are thus consistent with the assumption that subjects who receive antidepressant treatment would presumably show higher levels of depressive symptoms than healthy subjects despite their treatment. This expectation is supported by ﬁndings in elderly depressed in-patients receiving escitalopram for 4 weeks who improved from a Geriatric Depression Scale score of on average 9.4 to an average of 4.7, which however does not match the score of that of healthy controls (average 1.2) [Savaskan et al., 2008]. Further support comes from the observation that the hippocampal volume reduction that occurs in male in-patients with depression can be partially reversed with a ‘‘successful’’ antidepressant treatment, but will still not reach the volume of healthy controls [Kronmuller et al., 2008]. It is noteworthy that the magnitude of difference in CES-D scoring between subjects homozygous for the CYP2C19*2 (PMs) or the CYP2C19*1 allele (EMs) was in the same range as the difference between subjects who had been treated with antidepressants and those who had not. Recently, Rudberg et al. [2008b] found a strong gene dose effect on the steady state plasma level of escitalopram in relation to the CYP2C19 genotype in 166 psychiatric patients. The authors found a 15-fold difference in escitalopram concentration between subjects homozygous for CYP2C19*17 (RMs) and subjects homozygous for defect CYP2C19 alleles (PMs). When comparing the different CYP2C19 genotype groups, it is evident that the genotype with highest relative impact on the escitalopram level is that conferring a PM phenotype, whereas the RM phenotype (homozygous for CYP2C19*17) has a comparatively lower yet signiﬁcant effect. With EMs (CYP2C19*1/*1) as the reference, PMs had almost six times higher escitalopram levels (P < 0.001), whereas RM subjects only had 58% of the levels (P < 0.01) [Rudberg et al., 2008b]. In analogy, we observed an effect of the CYP2C19 PM genotype on the level of depressive symptoms, but no signiﬁcant effect of the RM genotype. Thus, the magnitude of difference in escitalopram plasma levels between genotypic PMs and RMs in relation to EMs as reported by Rudberg et al. [2008b] is compatible with the relative impact of the CYP2C19 genotype groups on the level of depressive symptoms shown here. Speciﬁcally, genotypic PMs were shown to have a lower level of depressive symptoms, whereas there was no signiﬁcant difference between EMs and RMs. Although the current study focuses on depressive symptoms rather than clinical depression, a cut-off value of 16 on the 20-item CES-D has a sensitivity of 100%, a speciﬁcity of 88%, and a positive predictive value of 13% for detecting major depression in the past year [Beekman et al., 1997]. Furthermore, Suthers et al.  found that the more stringent syndromal criteria of the CIDI-SF actually results in fewer detected cases of depression than a cut-off score on the short form of CES-D with 11 items. It should be emphasized that the CES-D scores from 84% of the subjects in the current study were based on an average of multiple tests with 2- to 4year intervals. Hence, the scores reﬂect more the ‘‘trait’’ of depressive symptoms rather than a transient state. Nevertheless, of all SIM ET AL. subjects in this study, 16% had a CES-D score of 16 or higher, indicative of a major depression. In conclusion, the present study based on 1,472 Swedes suggests that CYP2C19 polymorphism is associated with depressive symptoms as assessed by the CES-D scale. For the ﬁrst time, we have demonstrated that subjects of European ancestry with a CYP2C19 PM genotype have a signiﬁcantly lower level of depressive symptoms than EMs. The fact that CYP2C19 metabolizes psychoactive drugs implies a potential afﬁnity for substances active in the central nervous system. Further studies are needed however in order to investigate the functional link between CYP2C19 and depressive symptoms. ACKNOWLEDGMENTS This work was supported by grants from The Swedish Brain Foundation (Hj€arnfonden), Torsten och Ragnar S€ oderbergs Stiftelser and by the Swedish Research Council. The twin data were collected with support of NIH grants AG 04563, AG 10175, AG08861, the MacArthur Foundation Research Network on Successful Aging, and The Axel and Margaret Ax:son Johnson’s Foundation. REFERENCES Baldwin RM, Ohlsson S, Pedersen RS, Mwinyi J, Ingelman-Sundberg M, Eliasson E, Bertilsson L. 2008. Increased omeprazole metabolism in carriers of the CYP2C19*17 allele; a pharmacokinetic study in healthy volunteers. Br J Clin Pharmacol 65:767–774. Beekman AT, Deeg DJ, Van Limbeek J, Braam AW, De Vries MZ, Van Tilburg W. 1997. Criterion validity of the Center for Epidemiologic Studies Depression scale (CES-D): Results from a community-based sample of older subjects in The Netherlands. Psychol Med 27:231–235. Butura A, Nilsson K, Morgan K, Morgan TR, French SW, Johansson I, Schuppe-Koistinen I, Ingelman-Sundberg M. 2009. The impact of CYP2E1 on the development of alcoholic liver disease as studied in a transgenic mouse model. J Hepatol 50:572–583. Cheng ZN, Shu Y, Liu ZQ, Wang LS, Ou-Yang DS, Zhou HH. 2001. Role of cytochrome P450 in estradiol metabolism in vitro. Acta Pharmacol Sin 22:148–154. Cribb AE, Knight MJ, Dryer D, Guernsey J, Hender K, Tesch M, Saleh TM. 2006. Role of polymorphic human cytochrome P450 enzymes in estrone oxidation. Cancer Epidemiol Biomarkers Prev 15:551–558. Desta Z, Zhao X, Shin JG, Flockhart DA. 2002. Clinical signiﬁcance of the cytochrome P450 2C19 genetic polymorphism. Clin Pharmacokinet 41:913–958. 1165 Gatz M, Johansson B, Pedersen N, Berg S, Reynolds C. 1993. A cross-national self-report measure of depressive symptomatology. Int Psychogeriatr 5:147–156. Gold CH, Malmberg B, McClearn GE, Pedersen NL, Berg S. 2002. Gender and health: A study of older unlike-sex twins. J Gerontol B Psychol Sci Soc Sci 57:S168–S176. Ingelman-Sundberg M, Johansson I. 1984. Mechanisms of hydroxyl radical formation and ethanol oxidation by ethanol-inducible and other forms of rabbit liver microsomal cytochromes P-450. J Biol Chem 259:6447–6458. Ingelman-Sundberg M, Sim SC, Gomez A, Rodriguez-Antona C. 2007. Inﬂuence of cytochrome P450 polymorphisms on drug therapies: Pharmacogenetic, pharmacoepigenetic and clinical aspects. Pharmacol Ther 116:496–526. Ishii G, Suzuki A, Oshino S, Shiraishi H, Otani K. 2007. CYP2C19 polymorphism affects personality traits of Japanese females. Neurosci Lett 411:77–80. Jansson M, Gatz M, Berg S, Johansson B, Malmberg B, McClearn GE, Schalling M, Pedersen NL. 2003. Association between depressed mood in the elderly and a 5-HTR2A gene variant. Am J Med Genet Part B 120B:79–84. Kendler KS, Gatz M, Gardner CO, Pedersen NL. 2006. A Swedish national twin study of lifetime major depression. Am J Psychiatry 163:109–114. Kendler KS, Gatz M, Gardner CO, Pedersen NL. 2007. Clinical indices of familial depression in the Swedish Twin Registry. Acta Psychiatr Scand 115:214–220. Kronmuller KT, Pantel J, Kohler S, Victor D, Giesel F, Magnotta VA, Mundt C, Essig M, Schroder J. 2008. Hippocampal volume and 2-year outcome in depression. Br J Psychiatry 192:472–473. Kurzawski M, Gawronska-Szklarz B, Wrzesniewska J, Siuda A, Starzynska T, Drozdzik M. 2006. Effect of CYP2C19*17 gene variant on Helicobacter pylori eradication in peptic ulcer patients. Eur J Clin Pharmacol 62: 877–880. McClearn GE, Johansson B, Berg S, Pedersen NL, Ahern F, Petrill SA, Plomin R. 1997. Substantial genetic inﬂuence on cognitive abilities in twins 80 or more years old. Science 276:1560–1563. Ohlsson Rosenborg S, Mwinyi J, Andersson M, Baldwin RM, Pedersen RS, Sim SC, Bertilsson L, Ingelman-Sundberg M, Eliasson E. 2008. Kinetics of omeprazole and escitalopram in relation to the CYP2C19*17 allele in healthy subjects. Eur J Clin Pharmacol 64:1175–1179. Pedersen NL, McClearn GE, Plomin R, Nesselroade JR, Berg S, DeFaire U. 1991. The Swedish Adoption Twin Study of Aging: An update. Acta Genet Med Gemellol (Roma) 40:7–20. Pinna G, Costa E, Guidotti A. 2009. SSRIs act as selective brain steroidogenic stimulants (SBSSs) at low doses that are inactive on 5-HT reuptake. Curr Opin Pharmacol 9:24–30. Radloff LS. 1977. The CES-D scale: A self-report depression scale for research in the general population. Appl Psychol Meas 1:385–401. Finkel D, Pedersen NL. 2004. Processing speed and longitudinal trajectories of change for cognitive abilities: The Swedish Adoption/Twin Study of Aging. Aging Neuropsychol Cogn 11:325–345. Rudberg I, Hermann M, Refsum H, Molden E. 2008a. Serum concentrations of sertraline and N-desmethyl sertraline in relation to CYP2C19 genotype in psychiatric patients. Eur J Clin Pharmacol 64:1181–1188. Fiske A, Gatz M, Pedersen NL. 2003. Depressive symptoms and aging: The effects of illness and non-health-related events. J Gerontol B Psychol Sci Soc Sci 58:P320–P328. Rudberg I, Mohebi B, Hermann M, Refsum H, Molden E. 2008b. Impact of the ultrarapid CYP2C19*17 allele on serum concentration of escitalopram in psychiatric patients. Clin Pharmacol Ther 83:322–327. Fradette C, Yamaguchi N, Du Souich P. 2004. 5-Hydroxytryptamine is biotransformed by CYP2C9, 2C19 and 2B6 to hydroxylamine, which is converted into nitric oxide. Br J Pharmacol 141:407–414. Savaskan E, Muller SE, Bohringer A, Schulz A, Schachinger H. 2008. Antidepressive therapy with escitalopram improves mood, cognitive symptoms, and identity memory for angry faces in elderly depressed patients. Int J Neuropsychopharmacol 11:381–388. Fukasawa T, Suzuki A, Otani K. 2007. Effects of genetic polymorphism of cytochrome P450 enzymes on the pharmacokinetics of benzodiazepines. J Clin Pharm Ther 32:333–341. Schroth W, Antoniadou L, Fritz P, Schwab M, Muerdter T, Zanger UM, Simon W, Eichelbaum M, Brauch H. 2007. Breast cancer treatment 1166 outcome with adjuvant tamoxifen relative to patient CYP2D6 and CYP2C19 genotypes. J Clin Oncol 25:5187–5193. Sim SC, Risinger C, Dahl ML, Aklillu E, Christensen M, Bertilsson L, Ingelman-Sundberg M. 2006. A common novel CYP2C19 gene variant causes ultrarapid drug metabolism relevant for the drug response to proton pump inhibitors and antidepressants. Clin Pharmacol Ther 79:103–113. Suthers KM, Gatz M, Fiske A. 2004. Screening for depression: A comparative analysis of the 11-item CES-D and the CIDI-SF. J Mental Health Aging 10:209–219. AMERICAN JOURNAL OF MEDICAL GENETICS PART B Yamazaki H, Shimada T. 1997. Progesterone and testosterone hydroxylation by cytochromes P450 2C19, 2C9, and 3A4 in human liver microsomes. Arch Biochem Biophys 346:161–169. Yasui-Furukori N, Kaneda A, Iwashima K, Saito M, Nakagami T, Tsuchimine S, Kaneko S. 2007. Association between cytochrome P450 (CYP) 2C19 polymorphisms and harm avoidance in Japanese. Am J Med Genet B Neuropsychiatr Genet 144B:724–727. Yin OQ, Wing YK, Cheung Y, Wang ZJ, Lam SL, Chiu HF, Chow MS. 2006. Phenotype-genotype relationship and clinical effects of citalopram in Chinese patients. J Clin Psychopharmacol 26:367–372.