A cis-Phase Interaction Study of Genetic Variants Within the MAOA Gene in Major Depressive Disorder

ARCHIVAL REPORTS

A cis-Phase Interaction Study of Genetic Variants Within the MAOA Gene in Major Depressive Disorder JieXu Zhang, YanBo Chen, KeRang Zhang, Hong Yang, Yan Sun, Yue Fang, Yan Shen, and Qi Xu Background: The genetic basis of major depressive disorder (MDD) has been explored extensively, but the mode of transmission of the disease has yet to be established. To better understand the mechanism by which the monoamine oxidase A (MAOA) gene may play a role in developing MDD, the present work examined the cis-phase interaction between genetic variants within the MAOA gene for the pathogenesis of MDD. Methods: A variable number tandem repeat (VNTR) and 19 single nucleotide polymorphisms (SNPs) within the gene were genotyped in 512 unrelated patients with MDD and 567 unrelated control subjects among a Chinese population. Quantitative real-time polymerase chain reaction analysis was applied to test the effect of genetic variants on expression of the MAOA gene in MDD. Results: Neither the VNTR polymorphism nor seven informative SNPs showed allelic association with MDD, but the cis-acting interactions between the VNTR polymorphism and four individual SNPs were strongly associated with MDD risk, of which the VNTR-rs1465107 combination showed the strongest association (p ⫽ .000011). Quantitative real-time polymerase chain reaction analysis showed that overall relative quantity of MAOA messenger RNA was significantly higher in patients with MDD than in control subjects (fold change ⫽ 5.28, p ⫽ 1.7 ⫻ 10⫺7) and that in the male subjects carrying the VNTR-L, rs1465107-A, rs6323-G, rs2072743-A, or rs1137070-T alleles, expression of MAOA messenger RNA was significantly higher in the patient group than in the control group. Conclusions: The cis-phase interaction between the VNTR polymorphism and functional SNPs may contribute to the etiology of MDD. Key Words: cis-regulation, expression, gene internal interaction, major depressive disorder, MAOA gene ajor depressive disorder (MDD) is a common psychiatric disorder with complex genetic architecture. The genegene/gene-environment interactions play a crucial role in developing the disease (1). The prevalence of MDD is approximately 17% in the general population (2) and its heritability reported to date is up to 42% (3). Genetic predisposition to complex disorders is thought to involve multiple genes of small effect, with an odds ratio of less than 1.5, and function of the disease-risk genes can be modified by environmental factors. Most severe psychiatric disorders are multifactorial complex diseases and their genetic mechanisms can be addressed by the common disease/common variant hypothesis. The onset of a psychiatric disease will be triggered when the additive effects of genetic and environmental factors exceed a particular threshold. In addition, gene-gene interactions (including epistatic effects) (4), gene internal interactions/ gene cluster additive effects (5), or gene-environment interactions (6) are likely to contribute to the pathogenesis of a complex mental illness. It is important to clarify how the development of these diseases involves different gene-gene or gene-environment interactions (7). The genes involved in neurotransmitter metabolism pathways have been a primary focus in previous studies of MDD. Monoamine oxidase A (MAOA) is a key enzyme that degrades many monoaminer-

M

gic neurotransmitters in the brain and has drawn much attention in investigation of severe psychiatric disorders like MDD (8,9). In addition, a number of studies have demonstrated that the MAOA gene may confer susceptibility to schizophrenia (10), depression (11), attentiondeficit/hyperactivity disorder (12), panic disorder (13), or some other conditions (14). It has been reported that function of the MAOA gene can be modulated by some environmental factors such as abuse, early traumatic life events, and familial adversity, resulting in violent, aggressive, and antisocial behaviors (15–17). The variable number tandem repeat (VNTR) polymorphism present in the 5=-flanking region of the MAOA locus has been a focus for the genetic analysis of mental disorders. The VNTR alleles comprise 2 repeats, 3 repeats, 3.5 repeats, 4 repeats, and 5 repeats. Because the VNTR polymorphism is related to MAOA activity, Sabol et al. (18) have defined the 2-, 3-, and 5-repeat alleles as the low-activity group and the 3.5- and 4-repeat alleles as the high-activity group, although controversial classifications have been proposed (13,18). Moreover, the VNTR polymorphism is also considered as a cis-regulatory element affecting phenotypic expression in humans (19), but little is known about why it could be involved in the development of different psychiatric conditions. The present study was then designed to test the following hypotheses: 1) whether the MAOA gene polymorphism could affect susceptibility to MDD in a Chinese population; 2) whether the cisacting regulation of multiple variants within the MAOA gene could play a role in predisposition to MDD; and 3) whether the cis-acting regulation could influence expression of the MAOA gene.

Methods and Materials From the National Laboratory of Medical Molecular Biology (JZ, YC, YF, YSh, QX), Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences, and Peking Union Medical College, Tsinghua University, Beijing; and Department of Psychiatry (KZ, HY, YSu), The First Hospital Shanxi Medical University, Taiyuan, China. Address correspondence to Qi Xu, Ph.D., National Laboratory of Medical Molecular Biology, Institute of Basic Medical Sciences, Chinese Academy of Medical Science and Peking Union Medical College, Tsinghua University, Dong Dan San Tiao No. 5, Beijing 100005, China; E-mail: [email protected]. Received Apr 13, 2010; revised Jun 7, 2010; accepted Jun 8, 2010.

0006-3223/$36.00 doi:10.1016/j.biopsych.2010.06.004

Subjects This study recruited 512 unrelated patients with MDD, of whom 251 were male and 261 were female, aged 28.87 ⫾ 8.32 (SD) years, through the Department of Psychiatry, First Hospital of Shanxi Medical University, Taiyuan, China, in the period between January 2006 and August 2008. These patients were diagnosed as having MDD by at least two consultant psychiatrists according to the DSM-IV criteria for MDD (20). Potential participants who were pregnant or had severe medical conditions, abnormal laboratory baseline values, BIOL PSYCHIATRY 2010;68:795– 800 © 2010 Society of Biological Psychiatry

796 BIOL PSYCHIATRY 2010;68:795– 800 unstable psychiatric features (e.g., suicidal), history of alcoholism or drug abuse, epilepsy, brain trauma with loss of consciousness, neurological illness, or a concomitant Axis I psychiatric disorder were excluded from this study. These patients were also assessed by the same psychiatrists with the 17-item Hamilton Rating Scale for Depression (HAMD-17) and Hamilton Anxiety Scale (HAMA) during the day of sampling. The HAMD-17 rating scale was divided into four subscales, including the anxiety/somatization subscale composed of items 10, 11, 12, 13, 15, and 17; the cognitive disturbance subscale composed of items 2, 3, and 9; the retardation subscale composed of items 1, 7, 8, and 14; and the sleep disturbance subscale composed of items 4, 5, and 6. Each item in both the HAMD-17 and the HAMA was scored in five grades, i.e., 0 for absent, 1 for doubtful to mild, 2 for mild to moderate, 3 for moderate to severe, and 4 for maximum severity. To increase sample power for statistical tests, the scored symptoms were converted into a binary dataset, in which 0 was defined as the absence of symptoms and 1 as the presence of symptoms. Meanwhile, 567 unrelated healthy subjects, of whom 308 were male and 259 were female, aged 28.44 ⫾ 9.96 (SD) years, were also recruited as control subjects from local communities. These control subjects were asked to give detailed information about medical and family histories. Those who had history of major psychiatric or neurological disorders, psychiatric treatment or drug abuse, or family history of severe forms of psychiatric disorders were excluded. All patients and control subjects were of Chinese Han origin from the northern area of China. They all gave written informed consent to attending this study as approved by the Ethics Committee of Chinese Academy of Medical Sciences and Peking Union Medical College. These samples have also been used in our previous association studies (4,21).

J. Zhang et al. quenced and the DNA sequencing data were aligned using the SeqMan II software (DNASTAR, Inc., Madison, Wisconsin), showing that all resulting genotypes were concordant.

Selection of DNA Markers The VNTR polymorphism and 19 single nucleotide polymorphisms (SNPs) present in the MAOA locus were genotyped, including 11 coding SNPs and 8 intronic SNPs, of which 2 tag SNPs were selected by analyzing the HapMap database of the Han Chinese in Beijing population (http://snp.cshl.org/) (Figure 1). Six of the 11 coding SNPs were synonymous and 5 were nonsynonymous. All the SNPs selected for further analysis had a minor allele frequency of ⬎5%.

Quantitative Analysis of Gene Expression Ninety control subjects (63 male subjects and 27 female subjects) and 35 patients with MDD (22 male patients and 13 female patients) were selected to analyze expression of the MAOA gene in the peripheral WBCs. All the patients recruited for analysis of gene expression were drug-naive. Total RNA was extracted from the peripheral WBCs. The purity and integrity of total RNA were evaluated by measuring optical density and by electrophoresis on a denaturing agarose gel, respectively. An aliquot of 1 ␮g total RNA was used to synthesize complementary DNA (cDNA) in a 50 ␮L reaction volume containing Oligo (dT)18 primers (Invitrogen, Beijing, China), M-MLV reverse transcriptase (Promega, Madison, Wisconsin), and RNasin ribonuclease inhibitor (Promega). Expression of MAOA messenger RNA (mRNA) was measured by relative quantitative analysis using the ABI PRISM 7500 real-time PCR (RT-PCR) system (Applied Biosystems). The primers used for RT-PCR amplification were designed by Primer Express version 2.0 (Applied Biosystems). A 20 ␮L reaction volume containing 10 ␮L of 2 ⫻ SYBR Green master mix (Applied Biosystems), 1 ␮L of mixed primers (5 ␮mol/L), and 2 ␮L of cDNA was used for RT-PCR amplification. All RT-PCR reactions were performed in triplicate. The primers used for RT-PCR amplification of MAOA cDNA samples are 5=-GAGCGGCTACATGGAAGGG-3= (forward) and 5=-TCACCTTCCCGAGACCATTTA-3= (reverse). Glyceraldehyde-3-phosphate dehydrogenase was used as an internal control gene, and the primers used for amplification of glyceraldehyde-3-phosphate dehydrogenase cDNA samples are 5=-ACTTCAACAGCGACACCCACT-3= (forward) and 5=-GCCAAATTCGTTGTCATACCAG-3= (reverse). The detailed RT-PCR conditions comprised 95°C for 10 minutes, followed by 40 cycles of 95°C for 15 seconds and 60°C for 1 minute. The comparative Ct (2⫺⌬⌬Ct) method was employed to quantify MAOA transcripts (22). The data of gene expression were presented as the means of ⌬Ct for a statistical test and the fold change (FC) was calculated using the mean ⌬Ct values of the two groups tested to represent alteration of gene expression.

Genotyping of DNA Markers Genomic DNA was extracted from peripheral white blood cells (WBCs) using the phenol-chloroform method. The primer sequences and polymerase chain reaction (PCR) conditions are given in Table S1 in Supplement 1. The PCR products were sequenced bi-directionally on an ABI 3700 DNA sequencer (PerkinElmer, Applied Biosystems, Foster City, California) and SNP genotype calling was performed using the Chromas program (Technelysium Pty Ltd., version 2.22, Tewantin, Australia). The VNTR polymorphism was detected using the PCR-based genotyping method as described by Sabol et al. (18). To validate the accuracy of genotyping work, all homozygous samples were se-

Data Analysis The Haploview program (version 4.1, Broad Institute of MIT and Harvard, Cambridge, Massachusetts) was applied to test the genotypic distributions of SNPs for Hardy-Weinberg equilibrium, to estimate linkage disequilibrium (LD) between paired SNPs, and to define the haplotype blocks (23). Because the MAOA gene is located on the X chromosome, only female subjects were used to perform the Hardy-Weinberg equilibrium test. To ensure full coverage of the tested gene, LD between paired SNPs was estimated based on the four-gamete rule. Primary analyses were performed for allelic and haplotypic association under the chromosome X model with the UNPHASED

Figure 1. Schematic representation of the MAOA gene and the location of tested single nucleotide polymorphisms. Exon is represented by black vertical lines and UTR is indicated by brown bar. VNTR, variable number tandem repeat.

www.sobp.org/journal

BIOL PSYCHIATRY 2010;68:795– 800 797

J. Zhang et al. Table 1. The cis-Phase Interactions Between the VNTR and SNPs in the MAOA Gene MDD Combination

a

VNTR-rs1465107c

VNTR-rs6323d

VNTR-rs202743e

VNTR-rs1137070f

CTR

Haplotype

n

%

n

%

␹2

p Valueb

OR

H-A H-G L-A L-G H-G H-T L-G L-T H-A H-G L-A L-G H-C H-T L-C L-T

15 288 450 16 27 274 445 28 26 274 441 27 276 26 30 436

1.98 37.42 58.49 2.11 3.45 35.44 57.54 3.57 3.33 35.73 57.47 3.47 35.98 3.34 3.87 56.81

40 296 402 37 54 290 389 55 49 295 389 54 291 51 59 384

5.17 38.19 51.86 4.78 6.86 36.79 49.36 6.99 6.2 37.51 49.45 6.84 37.06 6.51 7.53 48.9

10.7 .13 6.6 7.66 8.35 .43 9.97 8.14 6.28 .66 9.61 8.23 .29 7.34 8.72 9.22

.001 .72 .01 .006 .004 .51 .002 .004 .012 .42 .002 .004 .59 .007 .003 .002

1 2.56 2.94 1.15 1 1.92 2.32 1.02 1 1.77 2.16 .94 1 .53 .53 1.20

95% CI

1.38–4.75 1.60–5.42 .50–2.64 1.16–3.17 1.42–3.79 .53–1.95 1.06–2.96 1.30–3.58 .48–1.84 .32–.88 .33–.85 .97–1.48

CI, confidence interval; CTR, control group; MAOA, monoamine oxidase A; MDD, major depressive disorder; OR, odds ratio; SNP, single nucleotide polymorphism; VNTR, variable number tandem repeat. a The global p value was .015 after 10,000 permutations. b The 1 ⫺ df ␹2 test. c 2 ␹ ⫽ 19.31, df ⫽ 1, p ⫽ .000011. d 2 ␹ ⫽ 17.44, df ⫽ 1, p ⫽ .00003. e 2 ␹ ⫽ 15.14, df ⫽ 1, p ⫽ .0001. f 2 ␹ ⫽ 17.01, df ⫽ 1, p ⫽ .000037.

program (version 3.1.3, Dudbridge F., MRC Biostatistics Unit, Cambridge, United Kingdom) (24). Only those haplotypes with a frequency of ⬎ .01 in either control subjects or cases were considered. To test an underlying interaction between functional DNA markers within the MAOA gene, cis-phase interaction analysis was performed under the X-linked gene-gene interaction model of UNPHASED, using the VNTR polymorphism as a conditional marker against functional SNPs within the gene (25). To circumvent the problem of multiple testing for disease association, a permutation test was performed for a global p value corrected for all the DNA markers tested. The permutation test is a built-in program of UNPHASED and 10,000 permutations were employed to empirically estimate a statistical significance. Comparison of a difference in MAOA mRNA expression between two groups was analyzed by the Student t test with SPSS for windows 13.0 (SPSS Inc., Chicago, Illinois). The Bonferroni correction was applied to correct a p value from individual haplotype association (df ⫽ 1) and symptoms analysis. The significance level was set at a corrected p value of .05 (two-tailed).

Supplement 1), analysis of the cis-phase interaction demonstrated that of the seven SNPs tested, four showed strong association with MDD when conditioning on the VNTR polymorphism (Table 1), in which the VNTR-rs1465107 combination had the strongest association (␹2 ⫽ 19.31, df ⫽ 1, p ⫽ .000011). Interestingly, the combinations of VNTR-L alleles with the major alleles of these four individual SNPs showed increased risk of MDD (Table 1). The conditional test did not show combined effects of the SNPs paired between rs1465107, rs6323, rs2072743, and rs1137070 on MDD (Table S5 in Supplement 1), suggesting that these four SNPs may share an LD signal from a disease-causing variant (25). Analysis of the VNTR-four SNP haplotypes confirmed that the L-AG-A-T haplotype, which harbors the major alleles of four disease-associated SNPs (Table 2), was associated with increased risk of MDD (corrected p ⫽ .0054); analysis of the haplotypes composed of all eight

Results

Haplotypea

n

Frequency

n

Frequency

␹2

p Valueb

Of 19 SNPs tested, 12 had a minor allele frequency of ⬍5% and were removed from subsequent analysis. Analysis with the Haploview program showed that the genotypic distributions of all seven highly informative SNPs did not deviate from Hardy-Weinberg equilibrium in female subjects (Table S2 in Supplement 1) and that of these seven SNPs, rs2235186, rs2235185, rs2072744, and rs2072743 were present in the same LD block (D= ⬎ .9 and r2 ⬎ .8), while rs1465107, rs6323, and rs1137070 were present outside the four-SNP LD block (Figure S1 in Supplement 1). The frequencies of VNTR alleles and genotypes in both control subjects and patients with MDD are listed in Table S3 in Supplement 1. While the VNTR polymorphism and seven highly informative SNPs did not show allelic association with MDD (Table S4 in

H-A-G-A-T H-G-G-A-T H-G-T-G-C L-A-G-A-T L-A-T-G-C L-G-T-G-C

13 7 252 415 6 15

.02 .01 .35 .59 .01 .02

34 8 269 360 18 35

.05 .01 .36 .5 .03 .05

8.37 .04 .47 11 5.8 7.05

.004c .84 .5 .0009d .016 .008e

Table 2. Analysis of Five DNA Markers for Haplotypic Association With MDD MDD

CTR

CTR, control group; MDD, major depressive disorder; VNTR, variable number tandem repeat. a The order of DNA markers tested is as follows: 5=-VNTR-rs1465107rs6323-rs2072743-rs1137070-3=. b 2 ␹ ⫽ 27.19, df ⫽ 5, p ⫽ .00005. c Corrected p ⫽ .024. d Corrected p ⫽ .0054. e Corrected p ⫽ .048.

www.sobp.org/journal

798 BIOL PSYCHIATRY 2010;68:795– 800

J. Zhang et al.

Table 3. Analysis of Eight DNA Markers for Haplotypic Association With MDD MDD Haplotypea H-A-G-T-T-A-A-T H-G-G-T-T-A-A-T H-G-T-C-C-G-G-C L-A-G-T-T-A-A-T L-A-T-C-C-G-G-C L-G-T-C-C-G-G-C

CTR

n

Frequency

n

Frequency

␹2

p Valueb

13 6 244 411 6 15

.02 .01 .35 .59 .01 .02

32 8 258 350 13 34

.05 .01 .37 .5 .02 .05

7.45 .28 .73 10.41 2.6 6.85

.006c .6 .39 .001d .11 .009

CTR, control group; MDD, major depressive disorder; VNTR, variable number tandem repeat. a The order of DNA markers tested is as follows: 5=-VNTR-rs1465107rs6323-rs2235186-rs2235185-rs2072744-rs2072743-rs1137070-3=. b 2 ␹ ⫽ 22.66, df ⫽ 5, p ⫽ .0004. c Corrected p ⫽ .036. d Corrected p ⫽ .006.

DNA markers also showed a strong association with MDD (␹2 ⫽ 22.66, df ⫽ 5, p ⫽ .0004), of which the L-A-G-T-T-A-A-T haplotype, containing the major alleles of the above four disease-associated SNPs, was associated with increased risk of MDD (Table 3). Quantitative RT-PCR analysis showed that overall relative quantity of MAOA mRNA was significantly higher in patients with MDD than in control subjects (FC ⫽ 5.28, t ⫽ 5.71, df ⫽ 83.76, p ⫽ 1.7 ⫻ 10⫺7); the significant change of MAOA mRNA expression was mainly shown in male subjects (FC ⫽ 4.69, t ⫽ 4.88, df ⫽ 83, p ⫽ 5.0 ⫻ 10⫺6) but not in the female subjects (FC ⫽ 5.5, t ⫽ 2.03, df ⫽ 37.41, p ⫽ .05). In the patient group, expression of MAOA mRNA was slightly higher in female patients than in male patients (FC ⫽ 2.71, t ⫽ 2.29, df ⫽ 33, p ⫽ .03), which did not survive the Bonferroni correction; in the control group, female subjects did not show significantly higher expression of MAOA mRNA than male subjects (FC ⫽ 2.28, t ⫽ 1.58, df ⫽ 31.39, p ⫽ .125). There was no significant difference in MAOA mRNA expression between patients with the HAMD-17 and HAMA scale symptoms and those without the symptoms (Table S6 in Supplement 1). Because male subjects only carry a single X chromosome, which makes it easy to analyze the effect of alleles or haplotypes on gene expression, we performed the test for association of these DNA markers with gene expression in male subjects. As shown in Table 4, expression of MAOA mRNA was significantly higher in the patient group than in the control group in male subjects carrying the VNTR-L, rs1465107-A, rs6323-G, rs2072743-A, or rs1137070-T alleles

Table 5. Effects of the VNTR-L Alleles on MAOA mRNA Expression by cis-interacting With rs1465107-A, rs6323-G, rs2072743-A, or rs1137070-T Alleles in Male Subjects Haplotype

MDD (⌬Ct)a

n

CTR (⌬Ct)a

n

FC

p Valueb

VNTR-L/rs1465107-A VNTR-L/rs6323-G VNTR-L/rs202743-A VNTR-L/rs1137070-T

5.85 ⫾ 1.60 5.75 ⫾ 1.58 5.75 ⫾ 1.58 5.95 ⫾ 1.46

13 14 14 13

8.48 ⫾ 2.12 8.68 ⫾ 2.09 8.69 ⫾ 2.05 8.68 ⫾ 2.09

22 25 26 25

6.19 7.60 7.68 6.62

.001 .00006 .00004 .00017

CTR, control group; FC, fold change; MAOA, monoamine oxidase A; MDD, major depressive disorder; mRNA, messenger RNA; VNTR, variable number tandem repeat. a Mean ⫾ SD. b Comparison of the ⌬Ct values between the patient group and the control group.

but not in those carrying the VNTR-H, rs1465107-G, rs6323-T, rs2072743-G, and rs1137070-C alleles. Two-marker haplotype analysis demonstrated that the individuals with haplotypes of the VNTR-L alleles coupled with rs1465107-A, rs6323-G, rs2072743-A, or rs1137070-T alleles had a significant increase in gene expression in the patient group compared with that in the control group (Table 5). Five-marker haplotype analysis showed that the L-A-G-A-T carriers had a higher level of MAOA mRNA expression in the patient group than in the control group (FC ⫽ 5.28, t ⫽ 3.50, df ⫽ 32, p ⫽ .001), whereas the H-G-T-G-C carriers did not show such a difference between the two groups (FC ⫽ 1.52, t ⫽ .85, df ⫽ 34, p ⫽ .402).

Discussion Genetic analysis frequently shows that a risk gene may be associated with several complex diseases. Thorgeirsson et al. (26) found that the CHRNA3 gene was associated not only with nicotine dependence but also with lung cancer. Coenen et al. (27) found that the TNFAIP3, IL-2/IL-21, SH2B3, LPP, MMEL1/TNFRSF14, and PFKFB3/PRKCQ genes could confer susceptibility to both rheumatoid arthritis and celiac diseases. Similar findings have also been reported in psychiatric disorders. For example, the NPAS3 gene was found to be associated with both schizophrenia and bipolar disorder (5). The role of the same gene in different diseases can be explained by several mechanisms, including the gene interacting with different environmental factors (28) or with distinct genes (29,30), as well as the cis-acting interaction between multiple variants within a gene (31). The present study showed that the MAOA gene could confer susceptibility to MDD in a Chinese population through a cis-acting

Table 4. Effect of the MAOA Gene Polymorphisms on mRNA Expression in Male Subjects Marker

Allele

MDD (⌬Ct)a

n

CTR (⌬Ct)a

n

FC

p Valueb

H L A G G T A G C T

8.01 ⫾ 1.77 5.91 ⫾ 1.70 5.85 ⫾ 1.60 7.75 ⫾ 1.90 5.75 ⫾ 1.58 7.75 ⫾ 1.90 5.75 ⫾ 1.58 7.75 ⫾ 1.90 7.25 ⫾ 2.34 5.95 ⫾ 1.46

6 16 13 8 14 8 14 8 9 13

8.71 ⫾ 1.63 8.71 ⫾ 2.03 8.47 ⫾ 2.03 8.65 ⫾ 1.58 8.66 ⫾ 2.05 8.75 ⫾ 1.66 8.67 ⫾ 2.01 8.74 ⫾ 1.68 8.75 ⫾ 1.66 8.66 ⫾ 2.05

33 30 24 35 26 37 27 36 37 26

1.63 6.96 6.15 1.87 7.52 2 7.57 1.99 2.83 6.54

.34 .00003 .0003 .17 .00005 .14 .00003 .15 .03 .0001

VNTR rs1465107 rs6323 rs2072743 rs1137070

⌬Ct, difference in threshold cycle; CTR, control group; FC, fold change; MAOA, monoamine oxidase A; MDD, major depressive disorder; mRNA messenger RNA; VNTR, variable number tandem repeat. a Mean ⫾ SD. b Comparison of the ⌬Ct values between the patient group and the control group.

www.sobp.org/journal

J. Zhang et al. effect of the VNTR polymorphism on a functional variant that is in strong LD with rs1465107, rs6323, rs2072743, and rs1137070. The VNTR polymorphism is located in the upstream region of the MAOA gene, approximately 1.2 kilobase of DNA away from the 5=-end. Deckert et al. (13) demonstrated that transcriptional activity was higher in the individuals with 3.5, 4, or 5 repeats than in those with 2 or 3 repeats, whereas Sabol et al. (18) found in an in vitro study that transcriptional efficiency was threefold to fivefold higher in the individuals carrying 3.5 or 4 repeats than in those carrying 2, 3, or 5 repeats. These findings raise the possibility that distinct VNTR alleles have different activities of binding to a transcriptional activator (18). The frequencies of VNTR alleles observed in our study are similar to those reported in a previous study with a Chinese population (10). We categorized the VNTR polymorphism based on the criteria suggested by Sabol et al. (18) and our work showed that the VNTR polymorphism was associated with MDD in the Chinese population, giving further supports of the initial finding reported by Sabol et al. (18). Our finding also suggests that the VNTR-L alleles may be involved in an increased risk of MDD through cis-acting regulation. Similarly, previous studies demonstrated that the individuals with VNTR-L alleles had an increased neural activity when a negative facial signal (angry or sad) was loaded as a stimulus, suggesting that the VNTR polymorphism could modulate brain limbic activation in response to negative emotional stimuli, possibly increasing the risk of depression (32). Other studies also showed that the VNTR-L carriers were more susceptible to emotional trauma, childhood abuse, and the adult forms of antisocial behaviors (15). The effect of an upstream VNTR polymorphism on transcriptional activities has also been observed in some other genes that are associated with susceptibility to paranoid schizophrenia and Parkinson’s disease, including those encoding the serotonin transporter, insulin, and dopamine D2 receptor (33–36). These findings suggest that the combined effect of the VNTR polymorphism and distinct disease-related variants within a gene could contribute to the pathogenesis of disease. The four SNPs that were associated with gene expression include two synonymous SNPs, rs6323 and rs1137070, and two intronic SNPs, rs1465107 and rs2072743. The combination of these four SNPs and the VNTR-L alleles may enhance MAOA activity by cis-acting upregulation of gene expression, although it is difficult to confirm which of these SNPs is functioning. The rs6323-G and rs1137070-T alleles have been found to be associated with increased MAOA activity (37). These two synonymous SNPs may therefore function in regulating gene expression or protein translation, especially by cis-phase interacting with the VNTR-L alleles in MDD. Although the two intronic SNPs, rs1465107 and rs2072743, do not influence protein function by changing amino acid sequence, it cannot be ruled out that they could alter primary transcriptional mRNA process by affecting the RNA stem-loop structure, resulting in alteration of gene expression as demonstrated in a previous study of the COMT gene (38). Nevertheless, our findings provide further evidence in support of the hypothesis that increased MAOA activity may be one of the major causes for the reduced serotonin activity observed in MDD (8). A question is whether gender and disease state affect MAOA mRNA expression in MDD. The levels of MAOA transcripts were significantly higher in male patients than male control subjects, but such a difference was not observed in female subjects. There is no significant difference in MAOA mRNA expression observed between female and male subjects in either the patient group or the control group. It is therefore difficult to draw a firm conclusion in a small female sample size that gender may significantly affect ex-

BIOL PSYCHIATRY 2010;68:795– 800 799 pression of MAOA mRNA. In this study, we also detected association of MAOA mRNA expression with clinical phenotypes and the results suggest that MAOA mRNA expression is not state-dependent (Table S6 in Supplement 1). It is possible that a failure of MAOA homeostatic response to a primary disease trigger may act as a penetrance modulator in MDD. A cis-regulatory region/element means that a segment of DNA can regulate transcription on a neighboring DNA sequence. Normally, the cis-regulatory elements lie at the 5=-end of the transcription start site but some also reside in the 3=-untranslated region and introns, even up to several kilobases of DNA away from the gene (39). The cis-acting regulators can act on a functional allele at a regulatory DNA binding site or alter the binding affinity of regulatory proteins recruited to modulate expression of the gene (39). For example, a minisatellite in the human insulin gene plays a cis-regulatory role in the development of diabetes (34). Another example is the relationship between the DARC gene and malaria resistance (40). The VNTR polymorphism in the MAOA locus is considered as a cis-regulatory element affecting phenotypic expression in humans (19). The present study suggests that the cis-regulatory effect of the VNTR polymorphism on some functional SNPs within the MAOA gene may contribute to the development of MDD. In summary, the VNTR polymorphism and functional SNPs within the MAOA locus may confer susceptibility to MDD in the Chinese Han population. The VNTR-L alleles may act as a cis-acting regulator interacting with other variants in the MAOA gene to modify disease susceptibility. The present study provides useful information toward a better understanding of the genetic etiology of complex disorders, but further study is needed to explore the precise mechanisms involved. This work was supported by the research grants from the National Basic Research Program of China (2010CB529603), the National 863 program (2006AA02A407), the Fok Ying Tong Education Foundation, the National Natural Science Foundation of China (30721603, 30770770, and 30971054) and the Beijing Natural Science Foundation (Grant Number 7102109). We sincerely thank all the subjects for their support and participation and all the medical staff involved in collecting blood samples. All authors reported no biomedical financial interests or potential conflicts of interest. Supplementary material cited in this article is available online. 1. Lander ES, Schork NJ (1994): Genetic dissection of complex traits. Science 265:2037–2048. 2. Kessler RC, Berglund P, Demler O, Jin R, Koretz D, Merikangas KR, et al. (2003): The epidemiology of major depressive disorder: Results from the National Comorbidity Survey Replication (NCS-R). JAMA 289:3095–3105. 3. Kendler KS, Gatz M, Gardner CO, Pedersen NL (2006): A Swedish national twin study of lifetime major depression. Am J Psychiatry 163:109 –114. 4. Wang Y, Hu Y, Fang Y, Zhang K, Yang H, Ma J, et al. (2009): Evidence of epistasis between the catechol-O-methyltransferase and aldehyde dehydrogenase 3B1 genes in paranoid schizophrenia. Biol Psychiatry 65: 1048 –1054. 5. Pickard BS, Christoforou A, Thomson PA, Fawkes A, Evans KL, Morris SW, et al. (2009): Interacting haplotypes at the NPAS3 locus alter risk of schizophrenia and bipolar disorder. Mol Psychiatry 14:874 – 884. 6. Uher R (2008): The implications of gene-environment interactions in depression: Will cause inform cure? Mol Psychiatry 13:1070 –1078. 7. Neff CD, Abkevich V, Packer JC, Chen Y, Potter J, Riley R, et al. (2009): Evidence for HTR1A and LHPP as interacting genetic risk factors in major depression. Mol Psychiatry 14:621– 630. 8. Belmaker RH, Agam G (2008): Major depressive disorder. N Engl J Med 358:55– 68. 9. Meyer JH, Ginovart N, Boovariwala A, Sagrati S, Hussey D, Garcia A, et al. (2006): Elevated monoamine oxidase A levels in the brain: An explana-

www.sobp.org/journal

800 BIOL PSYCHIATRY 2010;68:795– 800

10. 11.

12. 13.

14.

15. 16. 17.

18. 19. 20. 21.

22. 23. 24.

tion for the monoamine imbalance of major depression. Arch Gen Psychiatry 63:1209 –1216. Qiu HT, Meng HQ, Song C, Xiu MH, Chen, da C, et al. (2009): Association between monoamine oxidase (MAO)-A gene variants and schizophrenia in a Chinese population. Brain Res 1287:67–73. Schulze TG, Muller DJ, Krauss H, Scherk H, Ohlraun S, Syagailo YV, et al. (2000): Association between a functional polymorphism in the monoamine oxidase A gene promoter and major depressive disorder. Am J Med Genet 96:801– 803. Jiang S, Xin R, Lin S, Qian Y, Tang G, Wang D, Wu X (2001): Linkage studies between attention-deficit hyperactivity disorder and the monoamine oxidase genes. Am J Med Genet 105:783–788. Deckert J, Catalano M, Syagailo YV, Bosi M, Okladnova O, Di Bella D, et al. (1999): Excess of high activity monoamine oxidase A gene promoter alleles in female patients with panic disorder. Hum Mol Genet 8:621– 624. Doornbos B, Dijck-Brouwer DA, Kema IP, Tanke MA, van Goor SA, Muskiet FA, Korf J (2009): The development of peripartum depressive symptoms is associated with gene polymorphisms of MAOA, 5-HTT and COMT. Prog Neuropsychopharmacol Biol Psychiatry 33:1250 –1254. Caspi A, McClay J, Moffitt TE, Mill J, Martin J, Craig IW, et al. (2002): Role of genotype in the cycle of violence in maltreated children. Science 297: 851– 854. Foley DL, Eaves LJ, Wormley B, Silberg JL, Maes HH, Kuhn J, Riley B (2004): Childhood adversity, monoamine oxidase A genotype, and risk for conduct disorder. Arch Gen Psychiatry 61:738 –744. Kim-Cohen J, Caspi A, Taylor A, Williams B, Newcombe R, Craig IW, Moffitt TE (2006): MAOA, maltreatment, and gene-environment interaction predicting children’s mental health: New evidence and a metaanalysis. Mol Psychiatry 11:903–913. Sabol SZ, Hu S, Hamer D (1998): A functional polymorphism in the monoamine oxidase A gene promoter. Hum Genet 103:273–279. Wray GA (2007): The evolutionary significance of cis-regulatory mutations. Nat Rev Genet 8:206 –216. American Psychiatric Association (2000): Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington, DC: American Psychiatric Press. Zhang K, Xu Q, Xu Y, Yang H, Luo J, Sun Y, et al. (2009): The combined effects of the 5-HTTLPR and 5-HTR1A genes modulate the relationship between negative life events and major depressive disorder in a Chinese population. J Affect Disord 114:224 –231. Livak KJ, Schmittgen TD (2001): Analysis of relative gene expression data using real-time quantitative PCR and the 2 (-Delta Delta C(T)) Method. Methods 25:402– 408. Barrett JC, Fry B, Maller J, Daly MJ (2005): Haploview: Analysis and visualization of LD and haplotype maps. Bioinformatics 21:263–265. Dudbridge F (2008): Likelihood-based association analysis for nuclear families and unrelated subjects with missing genotype data. Hum Hered 66:87–98.

www.sobp.org/journal

J. Zhang et al. 25. Dudbridge F (2006): UNPHASED User Guide. Technical Report 2006/5. Cambridge, UK: MRC Biostatistics Unit. 26. Thorgeirsson TE, Geller F, Sulem P, Rafnar T, Wiste A, Magnusson KP, et al. (2008): A variant associated with nicotine dependence, lung cancer and peripheral arterial disease. Nature 452:638 – 642. 27. Coenen MJ, Trynka G, Heskamp S, Franke B, van Diemen CC, Smolonska J, et al. (2009): Common and different genetic background for rheumatoid arthritis and coeliac disease. Hum Mol Genet 18:4195– 4203. 28. Kilpatrick DG, Koenen KC, Ruggiero KJ, Acierno R, Galea S, Resnick HS, et al. (2007): The serotonin transporter genotype and social support and moderation of posttraumatic stress disorder and depression in hurricane-exposed adults. Am J Psychiatry 164:1693–1699. 29. Urwin RE, Nunn KP (2005): Epistatic interaction between the monoamine oxidase A and serotonin transporter genes in anorexia nervosa. Eur J Hum Genet 13:370 –375. 30. Urwin RE, Bennetts BH, Wilcken B, Lampropoulos B, Beumont PJ, Russell JD, et al. (2003): Gene-gene interaction between the monoamine oxidase A gene and solute carrier family 6 (neurotransmitter transporter, noradrenaline) member 2 gene in anorexia nervosa (restrictive subtype). Eur J Hum Genet 11:945–950. 31. Hennah W, Thomson P, McQuillin A, Bass N, Loukola A, Anjorin A, et al. (2009): DISC1 association, heterogeneity and interplay in schizophrenia and bipolar disorder. Mol Psychiatry 14:865– 873. 32. Lee BT, Ham BJ (2008): Monoamine oxidase A-uVNTR genotype affects limbic brain activity in response to affective facial stimuli. Neuroreport 19:515–519. 33. Heils A, Teufel A, Petri S, Stober G, Riederer P, Bengel D, Lesch KP (1996): Allelic variation of human serotonin transporter gene expression. J Neurochem 66:2621–2624. 34. Kennedy GC, German MS, Rutter WJ (1995): The minisatellite in the diabetes susceptibility locus IDDM2 regulates insulin transcription. Nat Genet 9:293–298. 35. Mossner R, Henneberg A, Schmitt A, Syagailo YV, Grassle M, Hennig T, et al. (2001): Allelic variation of serotonin transporter expression is associated with depression in Parkinson’s disease. Mol Psychiatry 6:350 –352. 36. Ohara K, Nagai M, Tani K, Nakamura Y, Ino A, Ohara K (1998): Functional polymorphism of -141C Ins/Del in the dopamine D2 receptor gene promoter and schizophrenia. Psychiatry Res 81:117–123. 37. Hotamisligil GS, Breakefield XO (1991): Human monoamine oxidase A gene determines levels of enzyme activity. Am J Hum Genet 49:383–392. 38. Nackley AG, Shabalina SA, Tchivileva IE, Satterfield K, Korchynskyi O, Makarov SS, et al. (2006): Human catechol-O-methyltransferase haplotypes modulate protein expression by altering mRNA secondary structure. Science 314:1930 –1933. 39. Knight JC (2005): Regulatory polymorphisms underlying complex disease traits. J Mol Med 83:97–109. 40. Tournamille C, Colin Y, Cartron JP, Le Van Kim C (1995): Disruption of a GATA motif in the Duffy gene promoter abolishes erythroid gene expression in Duffy-negative individuals. Nat Genet 10:224 –228.