The DOPA decarboxylase ( DDC ) gene is associated with alerting attention Progress in Neuro-Psychopharmacology & Biological Psychiatry

The DOPA decarboxylase (DDC) gene is associated with alerting attention. DOPA decarboxylase ( DDC ) is involved in the synthesis of dopamine, norepinephrine and serotonin. It has been suggested that genes involved in the dopamine, norepinephrine, and cholinergic systems play an essential role in the ef ﬁ ciency of human attention networks. Attention refers to the cognitive process of obtaining and maintaining the alert state, orienting to sensory events, and regulating the con ﬂ icts of thoughts and behavior. The present study tested seven single nucleotide polymorphisms (SNPs) within the DDC gene for association withattention,which wasassessedbytheAttention NetworkTesttodetect three networksofatten-tion, including alerting, orienting, and executive attention, in a healthy Han Chinese sample ( N =451). Associa- tion analysis for individual SNPs indicated that four of the seven SNPs ( rs3887825, rs7786398, rs10499695 , and rs6969081) were signi ﬁ cantly associated with alerting attention. Haplotype-based association analysis revealed that alerting was associated with the haplotype G – A – T for SNPs rs7786398 – rs10499695 – rs6969081 . These associations remained signi ﬁ cant after correcting for multiple testing by max(T) permutation. No association was found for orienting and executive attention. This study provides the ﬁ rst evidence for the involvement of the DDC gene in alerting attention. A better understanding of the genetic basis of distinct attention networks would allow us to develop more effective diagnosis, treatment, and prevention of de ﬁ cient or underdeveloped alerting attention as well as its related prevalent neuropsychiatric disorders. © 2012 Elsevier Inc. All rights reserved.


Introduction
Attention is a central cognitive process involved in selectively concentrating on a certain aspect of the environment. Research on attention has potential applications in many areas, including education, human factors engineering, and treatment and rehabilitation of pathological conditions (Raz and Buhle, 2006). According to Posner and colleagues, there are three attention networks including alerting, orienting, and executive attention that can be measured by the attention network test (ANT) Posner and Petersen, 1990). Alerting refers to the ability to achieve and maintain response readiness in preparation for incoming stimuli. Orienting is the ability to select specific information from sensory stimuli. Executive attention involves the mechanism for monitoring and resolving conflicts among thoughts, feelings and responses.
The ANT is a suitable task for detecting the neural and genetic correlates of three attention networks (Fossella et al., 2002). Previous twin studies suggested that the heritability of the three attention networks ranged from 0 to .72 (i.e., low for orienting, moderate for alerting, and high for executive function) (Fan et al., 2001;Wang et al., 2012). The three attention networks showed distinct neural networks, and are differentially related to specific neurotransmitters. Alerting is associated with activation of brain areas in the locus coeruleus, right frontal lobe and right parietal lobe; and it is influenced by the level of norepinephrine, also known as noradrenaline (Oberlin et al., 2005;Witte and Marrocco, 1997). Orienting is associated with activation of brain areas in the superior colliculus, superior parietal lobe, temporal parietal junction, and frontal eye fields; and it is influenced by acetylcholine (Thiel et al., 2005). Finally, executive attention is related to activation of brain areas in the prefrontal cortex, anterior cingulate, and basal ganglia; and it is modulated by dopamine (Posner, 2008;Raz and Buhle, 2006). Several previous studies detected associations between various genes with relevance for neurotransmission and specific attention networks using the ANT. For example, in a study of 200 normal adults, Fossella et al. (2002) found that polymorphisms in the dopamine D4 receptor (DRD4) gene and the monoamine oxidase a (MAOA) gene were related to efficiency of executive attention. Other researchers found associations between the executive attention network and the tryptophan hydroxylase 2 (TPH 2) gene in normal subjects (Reuter et al., 2007), and between executive attention and the catechol-o-methyltransferase (COMT) Val108/158Met polymorphism in schizophrenic males (Opgen-Rhein et al., 2008). In addition, the dopamine transporter (DAT 1) gene showed a tendency for association with both alerting and executive attention in children with attention-deficit/hyperactivity disorder (ADHD), but these results did not survive the multiple testing corrections (Konrad et al., 2010).
The DOPA decarboxylase gene (DDC), which encodes the enzyme aromatic L-amino acid decarboxylase (AADC), is a very interesting candidate gene for mental activities and mental disorders, since it is involved in several neurotransmitter pathways (Borglum et al., 1999). The encoded protein catalyzes the conversion of Ldihydroxyphenylalanine (L-DOPA) to dopamine and of L-5 hydroxytryptophan to serotonin and L-tryptophan to tryptamine. Moreover, further downstream in the catecholamine pathway, dopamine transported into vesicles then converted to norepinephrine by dopamine β-hydroxylase. Although DDC is not a rate-limiting enzyme in the synthesis of dopamine, norepinephrine and serotonin, it is regulated at both pre-and post-translational levels (Ma et al., 2005). The DDC gene is located on the short arm of chromosome 7 and contains 15 exons and 14 introns.
Many previous studies showed that genetic variations in the DDC gene were significantly associated with ADHD, especially its inattentive symptoms. ADHD is a highly heritable and prevalent psychiatric disorder characterized by severe impairment in attention, hyperactivity and impulsivity (Banaschewski et al., 2010). Researchers reported that the DDC gene was strongly associated with ADHD in both childhood and adulthood (Ribases et al., 2009). A genome-wide association study also supported the association between the DDC gene and inattentive ADHD using over 900 ADHD Caucasian probandparent trios (i.e., International Multicenter ADHD Genetics [IMAGE] project dataset) (Lasky-Su et al., 2008). Moreover, this association was also confirmed in a Chinese Han sample (Guan et al., 2009). There were also a few non-significant or only trend results reported such as an earlier analysis using a subsample of patients from the IMAGE sample (Brookes et al., 2006) or a study with a relatively small sample size (Hawi et al., 2001).
Thus, the DDC gene may act as a susceptibility gene for attention inefficiency. As addressed earlier, the DDC gene is involved with three important neurotransmitter systems (i.e., dopamine, norepinephrine, and serotonin), and these neurotransmitters are related to the attention networks. However, to date no study has examined possible associations between DDC variants and the attention networks. In this study, we analyzed seven SNPs selected to cover the whole DDC gene to explore whether there exists an association of the DDC gene with specific attention network inefficiency.

Participants
451 healthy Han Chinese undergraduates were recruited (mean age = 19.95 years, SD=.89, range 18-22 years old; 57% female). The present sample included all subjects who completed the ANT from the Gene-Brain-Behavior Project at Beijing Normal University in China. All subjects were Han Chinese with normal or corrected-to-normal vision and had no neurological or psychiatric history based on their self-report 1 . They all signed written informed consent. This study was approved by the Institutional Review Board (IRB) of Beijing Normal University, China.

Genotyping
A 4 ml venous blood sample was collected from each subject. Genomic DNA was extracted according to standard methods within two weeks after the blood sample was collected. All samples were genotyped using the standard Illumina GoldenGate Genotyping protocol (see www.southgene.com.cn for details).
As depicted in Fig. 1 and Table 1, seven SNPs in the DDC gene on chromosome 7 were selected based on the HapMap data (www.hapmap.org), including rs11238131, rs11238133, rs3887825, rs3807566, rs7786398, rs10499695, and rs6969081. All seven SNPs met the criteria of a call rate of> 95%, Minor Allele Frequency (MAF) of >0.05, and Hardy-Weinberg equilibrium (HWE) of p> 0.05. The allele frequencies in our sample were very similar to those of the Chinese in the HapMap dataset (see Table 1 and supplemental materials Figs. S2-S6). In order to sample the genetic diversity of the DDC gene, we selected the tag SNPs (tSNPs) defined by the HapMap project (www.hapmap.org [phase 3]), which are the minimum set of SNPs needed to sample most genetic diversity through linkage disequilibrium (LD). The tSNPs were defined by HapMap in 2007 using the four populations investigated at that time (European, African-Yoruban, Chinese, and Japanese ancestry), and a general r 2 value of 0.8 was used for identification. These seven SNPs were chosen to cover 84% of LD blocks in the DDC gene, as defined for the Chinese sample included in the HapMap Project (http://www.hapmap.org) and in the 1000 Genomes Project (http://browser.1000genomes.org). For SNPs rs11238131 and rs11238133, there was one subject with missing genotype data in the current study.

Behavioral assessment
The attention network test (ANT) assesses the abilities of alerting, orienting, and executive function (see Fan et al., 2002, and Figure S1 for details). Subjects were seated 65 cm in front of the 17-in. computer screen, and responses were collected via two response buttons on the keyboard. They saw several arrows on the computer screen and had to decide the direction of the arrow in the middle. RT (ms) and performance accuracy (almost 100% accuracy in our study [mean = .98, SD = .02]) were recorded. It was a 4 × 2 × 3 × 2 design. The four factors were cue (no-cue, center-cue [i.e., a cue appeared in the middle], double-cue [i.e., two cues appeared up and down], spatial-cue [i.e., a cue appeared up or down]); the position of the target arrow (up or down); the direction of the arrows around the target arrow (congruent, incongruent, neutral); and the direction of the target arrow (right or left). The ANT consisted of a 24-trial practice block and two 96-trial experimental blocks. The alerting effect was calculated by subtracting the mean RT of the double-cue conditions from that of the no-cue conditions. The orienting effect was calculated by subtracting the mean RT of the spatial cue conditions (i.e., cues appeared up or down) from that of the center 1 Although we did not administer a systematic battery of objective assessments for neurological diseases and mental disorders, the larger project included measures of two main mental health issues relevant to college students: depression and anxiety. They were measured with Beck's depression and anxiety inventories: BDI-II and BAI (Beck & Steer, 1990;Beck et al., 1996). According to the cutoff scores and interpretive labels provide by Beck and Steer (1990) and Beck et al. (1996), three subjects scored in the "severe" range (>29) of BDI-II and 24 subjects scored in the "severe" range (>26) of BAI. Because depression and anxiety were not significantly correlated with any indices of the attention network test in the current data, we did not exclude these subjects from our analyses. In addition, we administered Wechsler Adult Intelligence Scale-Revised Chinese version (WAIS-RC) (Gong, 1992). Mean IQ was 126 (S.D. = 8), with a range of 101 to 147. IQ was significantly associated with orienting attention (r = .15, p = .001), but not with either alerting (r = .04, p = .41) or executive attention (r = .05, p = .28). Even after controlling for IQ, however, the same four SNPs in the DDC gene (see Results) remained to be significant correlates of alerting attention. cue (i.e., the cue appeared in the center). The executive effect was calculated by subtracting the mean RT of all congruent flanking conditions, summed across cue types, from the mean RT of the incongruent flanking conditions. It should be noted that alerting and orienting efficiency both represent benefits (i.e., decreases in RT), but executive represents the cost that is incurred (increased RT). Therefore, higher scores for indexes of alerting and orienting indicated higher abilities of alerting and orienting, but higher scores for the index of executive attention indicated poorer executive attention.

Data analyses
Quantitative trait genetic association analysis was carried out by using Plink v1.07 (Purcell et al., 2007), including allelic associations tests between individual SNPs and attention measures, and associations between haplotypes and attention measures. In order to test the group differences between different genotypes, ANOVA and the Fisher's least significant difference (LSD) post hoc tests (t-tests) were performed in SPSS 17.0. Pair-wise LD between all SNP markers was assessed using the Haploview 4.2 program (Barrett et al., 2005).
In a preliminary analysis, several separate two-way ANOVAs were conducted to test the effects of gender, each SNP, and their interaction on each attention measure. Results showed that the main effects of gender were non-significant: the p values were .33 for alerting, .50 for orienting, and .37 for executive scores. Perhaps due to the restricted age range, age was also not correlated with attention. Therefore, we did not include gender and age in subsequent analysis. The three attention measures (alerting, orienting, and executive scores) were analyzed individually. All significant associations were corrected for multiple testing by the max(T) permutations permutation approach in Plink (10000 permutation) for individual SNP analysis and haplotype-based association analysis considering all tests done for all three traits.

Results
In the current study, the mean scores for alerting, orienting, and executive attention were 6.67 ± 25.61, 33.52 ± 29.49, and 81.09 ± 36.02, respectively. Pair-wise correlations were not significant (i.e., p > .05), suggesting three independent attention networks as found in a previous study (Fossella et al., 2002). The mean reaction times for each condition in the ANT are shown in the supplementary materials Table S1.
Individual SNP analysis using Plink revealed significant associations for rs3887825, rs3807566, rs7786398, rs10499695, and rs6969081 with alerting (see Table 2 for details). Except for rs3807566, the other four associations remained significant after correcting for multiple testing by max(T) permutation (corrected empirical p-value [max(T)/familywise]).
As shown in Fig. 1, the mean pair-wise r 2 values of seven SNPs within DDC were 0.57. According to the criteria of confidence intervals (Gabriel et al., 2002), two haplotype blocks across the DDC gene were revealed from the linkage disequilibrium (LD) data for these seven SNPs. The first block contained three SNPs rs11238133, rs3887825, and rs3807566, which covered 21 kb. The second block contained three other SNPs rs7786398, rs10499695, and rs6969081, which covered 11 kb.
Haplotype-based association analysis was performed for different combinations of SNPs within DDC in the current sample. As shown in Table 3, we found a major haplotype G-A-T (with a frequency of 52%) for rs7786398-rs10499695-rs6969081 that showed a significant   inverse association with alerting (t = −7.39, df= 450, p = 0.0068). This association remained significant after correcting for multiple testing by max(T) permutation (corrected empirical p-value [max(T)/ familywise] = 0.044). The haplotype A-G-A (with a frequency of 42%) for rs7786398-rs10499695-rs6969081 showed a significant positive association with alerting (t =6.06, df=450, p = 0.0142), but this association was no longer significant after multiple comparison correction. Additional haplotype analysis based on a different SNP combination (rs11238133-rs3887825-rs3807566) revealed a major C-G-A haplotype (with a frequency of 33%) in the current sample, which was positively associated with alerting (t =5.08, df= 450, p =0.0248), but this association was no longer significant after correction. No significant associations for seven SNPs in the DDC gene with orienting or executive attention measures were found (p > .05). In addition, the overall mean reaction time in ANT was 541 ± 67 ms. There were no significant differences among genotypes in the overall mean reaction time. The overall mean reaction time was significantly associated with orienting (r = −0.11, df=449, p b 0.05), and executive attention (r =0.27, df =449, p b 0.05), but not with alerting (r =0.03, df=449, p = 0.05). Therefore, we also conducted individual SNPs analysis association with three attention networks, using the overall mean reaction time as a covariate. The same four SNPs in the DDC gene were still significantly associated with alerting attention after controlling for the covariate. There were no significant associations between the DDC gene and the orienting and executive attention measures with or without the covariate.

Discussion
In the present study, we chose seven SNPs in the DDC gene to investigate their genetic effects on three components of the attention network in normal Chinese adults. Results showed that four SNPs (rs3887825, rs7786398, rs10499695 and rs6969081) were significantly associated with individual differences in the alerting efficiency of attention after max (T) permutation for multiple testing correction. Further, we identified a major haplotype G-A-T (with a frequency of 52%) for rs7786398-rs10499695-rs6969081, which showed a significant negative association with alerting, even after max (T) permutation for multiple testing correction. Moreover, two additional haplotypes were found to be positively related to alerting (i.e., A-G-A [with a frequency of 42%] for rs7786398-rs10499695-rs6969081 and C-G-A [with a frequency of 33%] for rs11238133-rs3887825-rs3807566), which did not survive the correction. There was no association for the seven SNPs in the DDC gene with orienting and executive attention. These effects were independent of subjects' overall mean reaction time, age, and gender. These results further confirmed the critical role of the DDC gene in the alerting attention network.
These findings can be integrated with multiple lines of human and animal research (including biochemical, pharmacological, genetic, and brain imaging studies) on the DDC gene, the related neurotransmitters, and specific attention networks. The enzyme of the DDC gene is required in the synthesis of dopamine, norepinephrine, and serotonin (Christenson et al., 1972). It is regulated at both pre-and posttranslational levels (see Berry et al., 1996 for details). Three attention networks were related to three neurotransmitters respectively, alerting with norepinephrine, orienting with acetylcholine, and executive attention with dopamine. Pharmacological studies with human and monkeys showed that norepinephrine influences alerting but not orienting, while acetylcholine-related drugs influence orienting but not alerting (Posner, 2008). These findings may explain the association between the DDC gene and alerting but not orienting found in the current study.
Moreover, the DDC gene was found to be related to ADHD inattention symptoms (Guan et al., 2009;Lasky-Su et al., 2008;Ribases et al., 2009). There was direct evidence of altered DOPA decarboxylase activity in children and adults with ADHD using the positron emission tomography (Ernst et al., 1998(Ernst et al., , 1999. Using the ANT, several researchers reported alerting and executive attention deficits in children with ADHD (Johnson et al., 2008;Mullane et al., 2011). However, some other researchers suggested that the alerting network can best detect inattentive ADHD (Booth et al., 2007). For example, Booth et al. (2007) found that children with inattentive type ADHD showed a stronger alerting effect than those with the combined type ADHD, but no group differences were found in orienting and executive attention. A neuroimaging study also suggested that children with ADHD had a deficit in alerting, and this deficit was related to abnormal activities in the frontal and parietal brain regions (Cao et al., 2008). In line with the evidence presented above, the DDC gene was associated with alerting but not with orienting and executive attention in the current sample.
In the current study, we identified four SNPs (located in introns 1 and 10) and a three-marker haplotype (located in intron 1) in the DDC gene that were significantly related to alerting. These significant markers have not been used in previous studies. There were studies, however, linking markers in the same intron or adjacent location  with certain LD to ADHD and other disorders. For example, researchers reported that ADHD was associated with several genetic variations in the DDC gene such as polymorphisms on exon 1 (4 bp insertion), intron 3 (rs1466163), intron 8 (rs11575454) and intron 9 (rs6592952), and a haplotype block (rs11238131-rs6592961-rs1982406-rs2044859) on introns 1, 4, 6, 11 (Brookes et al., 2006;Guan et al., 2009;Hawi et al., 2001;Kirley et al., 2002;Ribases et al., 2009). Based on the LD data from the Asian samples in the HapMap Project and the 1000 Genomes Project, the four significant SNPs and the haplotype on introns 1 and 10 found in the current study are located in the LD blocks, which encompassed the associated SNPS reported in the previous studies about the DDC gene and attention-related phenotypes. Genetic variations located in intron 1 of the DDC gene were also found to be related to nicotine dependence (Ma et al., 2005;Yu et al., 2006), and interestingly nicotine could potentiate alerting attention in humans (Hahn et al., 2007;Thienel et al., 2009). In addition, the DDC gene has a high level of expression in the basal ganglia, which have been shown to modulate frontal-posterior connectivity involved in attention (van Schouwenburg et al., 2010). Our analysis suggested that the particular haplotype (G-A-T) of rs7786398-rs10499695-rs6969081 located on intron 1 decreased alerting in the attention network test. Although these three SNPs are intronic and hence their biochemical functions are not straightforward, several previous studies have shed light on this genetic variant. One previous study found that a DNA sequence upstream of DDC possesses enhancer-link properties and is essential for normal neuron-specific expression in Drosophila (Johnson et al., 1989). Mutations that abolish the activity of the DDC gene have been described in a recessively inherited disease, called aromatic L-amino acid decarboxylase deficiency (Lee et al. 2009). As the second enzyme in the catecholamine biosynthetic pathway, the deficiency of the enzyme DOPA decarboxylase leads to a reduced level of downstream biogenic amines, including dopamine, norepinephrine, epinephrine and serotonin. Previous studies also described chromosomal aberrations in the region encompassing the DDC gene, e.g., a duplication (dup(7)p11.2-p12) associated with mild cognitive deficit (Leach et al. 2007). Taken together, these studies provided evidence that common variants in the DDC gene may be involved in the genetics of alerting.
Finally, given that DDC is involved in the synthesis of dopamine and subsequently norepinephrine and that dopamine is believed to modulate executive function while norepinephrine is believed to modulate alerting, it is somewhat counterintuitive that the DDC gene is linked only to alerting, not to executive function. There may be several explanations of this apparent paradox. First, the dopamine system is a complex system involving several subsystems and many genes. It is possible that different genes play different roles in modulating dopamine levels in different brain regions, which in turn affect different components of the attention network. For example, previous studies showed that DRD4 VNTR was related to executive attention (e.g., Fossella et al., 2002). Second, we studied healthy young college students, who were likely to have normal executive function. Indeed, as Fig S7 shows, we had a restricted distribution of scores on executive function (but not for alerting and orienting), which might have contributed to our not finding an association between the DDC gene and executive function. It is plausible that when clinical samples such as children with ADHD who have a special deficit in executive attention are included (e.g., Johnson et al., 2008;Mullane et al., 2011), one may find an association between the DDC gene and executive function. Finally, because the relationship between the dopamine level and cognitive performance follows an "inverted-U-shaped" function (Cools and D'Esposito, 2011), complex patterns of associations between dopamine-related genes and cognitive performance occur depending on environmental factors as well as gene-gene and gene-environment interactions. Future research is needed to untangle these complex relationships.
Although the current findings add to the growing body of evidence for genetic bases of individual differences in attention (Fossella et al., 2002;Greenwood et al., 2005;Parasuraman et al., 2005), several limitations of this study need to be noted. First, the biochemical and physiological functions of the haplotype polymorphisms were not directly explored in the present study. Additional molecular functional studies are needed to investigate the details of mechanism of this association. Second, it should be noted that the current study is based on a healthy Han Chinese sample. Previous studies have inconsistent results about the racial specificity (e.g., African Americans versus European Americans) of the association between the DDC gene and certain behavioral phenotypes (Ma et al., 2005;Yu et al., 2006). The SNPs in the DDC gene have different minor allele frequency (MAF) and constructed different LD blocks in different ethnic populations based on the HapMap Data (www. hapmap.org, see Table 1 and Figs S2 to S6). In order to obtain confirmatory evidence, this association should be replicated in future studies using different ethnic samples. Third, although alerting is a particularly important component of attention, it is not well studied as compared to other components of attention, such as executive attention (Raz and Buhle, 2006). Finally, future studies should also map the DDC genetic variation of alerting attention onto brain activity.

Conclusions
In conclusion, this study provided the first evidence of the association between the DDC gene and alerting attention in healthy Chinese individuals and excluded several confounding factors including overall mean reaction time, age, and gender.