|Home | About | Journals | Submit | Contact Us | Français|
Tryptophan hydroxylase (TPH), the rate limiting enzyme in serotonin biosynthesis, is one of the most important regulating factors in the serotonergic system. Recently, polymorphisms of the TPH gene have been identified as being associated with suicide, but the evidence is inconsistent. To investigate the role in suicide of one of the isoforms, TPH1, we examined the association of five single nucleotide polymorphisms (SNPs) in the promoter region and in intron 7 of the TPH1 gene based on a sample from the Chinese population of 810 subjects, of whom 329 had made no suicide attempts (NSA), 297 had made suicide attempts (SA), and 184 were healthy subjects (HS). In this study, we observed statistically significant differences between NSA and HS subjects in allele distributions on one marker, −6526A (p=0.0329; odds ratio (OR) 1.36; 95% confidence interval (CI) 1.01 to 1.81). No significant difference in genotype distribution or allele frequencies of other polymorphisms was found between the suicide victims and the controls. The overall haplotype frequency was significantly different between cases and healthy controls (p=0.000024 NSA v HS; p<0.000001, SA v HS; p<0.000001, cases v HS). We found the haplotype TCAAA of −7180/−7065/−6526/218/779 to be strongly associated with suicidal behaviour and psychiatric disorders (p=0.00243; OR=1.62; 95% CI 1.17 to 2.24 and p=0.018; OR=1.41; 95% CI 1.05 to 1.91), which suggests an association of TPH1 with suicidal behaviour and indicates that TPH1 may play a significant role in the aetiology of psychiatric disorders in the Han Chinese population.
The human gene for tryptophan hydroxylase (TPH; MIM 191060) is situated on chromosome 11p15.3–14, and expressed in the gut, pineal gland, spleen, and thymus.1 It has two isoforms, TPH1 and TPH2. TPH has been the candidate gene for suicidal behaviour in recent years and although there have been several positive studies associating TPH and suicidal behaviour the evidence has not been entirely consistent. The TPH allele A218C was found to be associated with suicidal behaviour in Koreans and white populations,6,7 while Pooley found that the TPH A779 allele is a risk factor for deliberate self harm.9 Rujescu has suggested that both A218C and A779C are associated with anger related traits.10 The TPH promoter loci −7180T/G, −7065C/T, −6526A/G, and −5806G/T have been detected in the Finnish population, and −6526A/G has also been observed to be significantly associated with suicidal behaviour.11 Turecki found haplotype −6526G/−5806T/218C of TPH1 to be significantly more common among suicide cases than in normal controls .12 However, in other investigations, no association with 218A/C was found in a European sample13,14 nor was there any association between 218A/C, A779C and suicidal tendencies in white,15,16 Japanese,17 or Chinese populations.18 Rotondo et al found no association between A218C and the personality traits of attention deficit hyperactivity disorder, or those elicited by the Tridimensional Personality Questionnaire in the Chinese population,19,20 and A218C variants have been found not to be associated with major psychoses.21
TPH catalyses the 5‐hydroxylation of tryptophan, which is the first step in the biosynthesis of indoleamines (serotonin and melatonin).4 We selected five loci (three polymorphisms in the promoter region (−7180G/T, −7065C/T, and −6526A/G) and two in intron 7 (218A/C, 779A/C)) explored in earlier studies to investigate whether these variants were associated with suicidal behaviour and psychiatric disorders in the Chinese population. In this study of a cohort of case control Chinese samples, we identified an association between suicidal behaviour and haplotypes in the region of TPH1, which strongly supports previous findings.
Written informed consent, reviewed and approved by the Shanghai ethics committee of human genetic resources, was obtained from all participating subjects.
In total, 297 patients who had made suicide attempts (SA; 171 men and 126 women; mean (SD) age 46.10 (13.14) years), 329 patients who had not made suicide attemts (nonsuicide attempts; NSA; 217 men and 112 women, age 43.22 (12.27) years), and 184 healthy subjects (HS; 98 men and 86 women, age 42.85 (14.46) years) were recruited for this study. The combined SA and NSA subjects constituted the case group (SA+NSA). The demographic and clinical characteristics of these subjects are listed in table 11.. Of the 297 SA patients in the case group, 196 (66%) had schizophrenia, as had 303 (92.1%) of the 329 NSA patients. The other patients had one or more of anxiety disorders, organic mental disorders, personality disorders, or alcohol and drug misuse disorders. Of the SA patients, 67% had tried to end their lives by violent means. About 26% of SA patients and 30% of NSA patients had a family history of psychiatric disorders. All the samples were Han Chinese in origin, from the Shanghai area. The diagnoses conformed with the criteria of the Diagnostic and statistical manual of mental disorders, third revised edition.22 All the patients were individually interviewed by two independent senior psychiatrists.
In this study, we chose five single nucleotide polymorphisms (SNPs) (−7180G/T, −7065C/T, −6526A/G, 218A/C, and 779A/C) in and around the TPH1 gene locus, and genotyped them using allele specific PCR in separate PCR reactions (table 22).23 The PCR primers used in this study were designed using a tetra‐primer ARMS‐PCR primer design program.24 The assay combining kinetic (real time quantitative) PCR with allele specific amplification was performed as described by Germer et al.25 In the real time PCR, two PCR reactions for each sample were carried out in a total volume of 5 µl containing 10 ng genomic DNA, 2.5 µl Taqman Universal PCR Master Mix (Applied Biosystems), 0.2 µmol/l allele specific primer, 0.2 µmol/l common primer, and 0.2× SYBR Green I (Molecular Probe, Inc.) on a sequence detection system (ABI Prism 7900; Applied Biosystems). To reduce variability between wells in the PCR reactions, an automated dispenser (Hydra microdispenser; Robbins Scientific) and digital multichannel pipettes (Thermo Labsystems) were used. After an initial 2 minutes at 50°C to activate the AmpErase uracil‐N‐glycosylase (UNG) used and a step of 12 minutes at 95°C to deactivate the UNG and activate theAmpliTaq Gold enzyme, 50 cycles of 15 seconds at 95°C and 30 seconds at annealing temperature, followed by a final dissociation stage, were used.
In the haplotype association analysis, only haplotypes with an estimated frequency >1% were included for the χ2 test using CLUMP. The p values reported were two tailed, and significance was set at p<0.05. Tests for haplotypic association with schizophrenia were performed using the program FASTEHPLUS with a permutation test to obtain the empirical significance levels.28 The pairwise linkage disequilibrium (LD) values, as measured by D′, were estimated with the software 2LD.31,32 To test for departure from Hardy‐Weinberg equilibrium, an online calculator was used for both patient and control group data.
We observed statistically significant differences in allele distributions on one marker, −6526A, (p=0.0329; odds ratio (OR) 1.36; 95% confidence interval (CI) 1.01 to 1.81) between NSA and HS (table 33).
Linkage disequilibrium between each pair of SNPs is shown in table 44.. The five SNPs in the range between −7180 and 779 that revealed modest or strong LD were in an LD block (D′ 0.504 to 0.779).
Frequency analysis of the haplotypes with the five SNPs ((tablestables 5–8) in the 329 NSA, 297 SA and 184 control subjects showed that the overall frequency was significantly different (p=0.000024 NSA v HS; p<0.000001 SA/HS; p<0.000001 cases/HS; p=0.0198 SA/NSA). We analysed only haplotypes with frequencies >3% in patients or controls to avoid application of statistics to small numbers. Even taking into account a total of approximately 20 tests performed, the empirical p values from 300000 permutation tests were still significant after applying the Bonferroni correction.
The common haplotypes TCAAA and TTGCC of −7180/−7065/−6526/218/779 were found to be strongly associated with suicidal behaviour and psychiatric disorders (p=0.00243; OR=1.62; 95% CI 1.17 to 2.24 and p=0.0444; OR=0.33; 95% CI 0.91–12.41). In the case group, TCAAA was also associated with suicidal behaviour (p=0.0396; OR =1.28; 95% CI 0.99 to 1.64, SA/NSA). In the three marker haplotype analysis, the same significant association was found between HS subjects and suicidal behaviour or psychiatric disorders in the haplotypes: AAA of −6526/218/779 (p=0.01184), CAA of −7065/−6526/218 (p=0.0336), TCA of −7180/−7065/−6526 (p=0.0007, p=0.000984). In contrast, haplotype TCGAA of −7180/−7065/−6526/218/79 was strongly associated with HS rather than SA (p=0.0091; OR=0.26; 95% CI 0.08 to 0.84). Other haplotypes highly associated with HS subjects rather than SA subjects and the case group were GCA and GAC of −6526/218/779 (p=0.0004, 0.0004; p=0.0215, 0.0123), CGC of −7065/ −6526/218 (p=0.00002, 0.000001), and GCA and TCG of −7180/−7065/−6526 (p=0.000146, 0.0095; p=0.00002, 0.00009)
Each year, one million people die of suicide. Polymorphisms of genes can contribute to alterations in protein function in vivo, and twin and adoption family studies have provided evidence for familial transmission in suicide.33,34 The association of lower serotonergic function and suicidal behaviour indicate that genes related to the serotonergic system are candidates worthy of study as part of the genetic diathesis for suicidal behaviour. Tryptophan hydroxylase (TPH) catalyses the 5‐hydroxylation of tryptophan, which is the first step in the biosynthesis of indoleamines (serotonin and melatonin).4 Nielsen found the TPH genotype to be significantly associated with cerebrospinal fluid 5‐hydroxyindoleacetic acid (5‐HIAA) concentration in subjects suffering from impulsive behaviour. A genetic variant of the TPH gene has also been associated with violent suicide attempts. It may influence 5‐HIAA concentration in the cerebrospinal fluid and produce a predisposition to suicidal behaviour.3
Neurobiological studies also implicate serotonergic dysfunction in suicidal behaviour. Because genetic variants in the human TPH1 promoter and intron 7 can modulate gene transcription, these two loci have been widely studied in the TPH gene, but results have been mixed. Rujescu has provided strong evidence for an association of suicide related behaviour with A218 in TPH in a meta‐analysis.2 Shinkai also found that TPH may play a role in the negative symptoms in male patients with schizophrenia.5 In our study, −6526A/G showed significant differences between patients who had not attempted suicide (NSA) and HS (p=0.0329). The change of only one base pair (A→G) at −6526 strongly modifies the effect of this haplotype; TCAAA (p=0.00243) in −7180/−7065/−6526/218/779 seems to be a vulnerability haplotype , while TCGAA (p=0.0091) seems to be protective It suggests that the SNP −6526A/G is a potential suicide related site, with the A allele increasing the risk of suicide in patients, and it also explains the dose dependent effect of TPH1 A218 on the risk of suicidal behaviour shown in the meta‐analysis performed by Malafosse A.7 Patients with schizophrenia account for 92.1% of the NSA group, so we consider that the SNP −6526A/G may also be a locus for the aetiology of schizophrenia.
Our haplotype analysis suggests that the most common haplotype TCAAA, which was significantly more frequent in cases than in controls, might carry one or more predisposing variants. Interestingly, by comparison, the haplotype TCGAA, which was more frequent in controls than in cases, contained a variant that was likely to be protective against the disorder. Our case control study has independently replicated an association between TPH1 and suicidal behaviour. In our work, we observed the five marker haplotypes −7180G/T, −7065C/T, −6526A/G, 218A/C, and 779A/C to be significantly associated with psychiatric disorders in the Chinese population. The analysis of three marker haplotypes also show significance in every three adjacent marker that is in linkage disequilibrium ((tablestables 5–8), which suggests that there is complicated linkage and crossover between these five markers. It is difficult to interpret these conflicting results because of lack of power, ethnic heterogeneity, and variations in the sampling strategies (in particular for controls) and in polymorphisms of the TPH gene and others we studied. The relatively small sample size does not exclude the possibility of false positive results and the finding needs replication. Individual SNPs, which are not very informative, are more likely than haplotypes to give unstable frequencies among different populations in association studies of this complex disorder. The complexity of the phenotype of suicide may reflect multiple biological and social aetiological factors, and presents a worthwhile area for genetic studies. TPH or a gene in its vicinity may influence the psychiatry of suicidal behaviour. This region contains loci for several important genes, including those for Beckwith‐Wiedemann syndrome and tyrosine hydroxylase.37 The potentially confusing effect of population stratification is also a factor. Both environmental and genetic factors, as well as their interactions, are likely to contribute to the risk of suicide and psychiatric disorders.
In conclusion, our data were most consistent with a risk haplotype TCAAA in TPH1 for suicidal behaviour and psychiatric disorders. The lack of statistical significance with the individual markers probably implies that only one SNP genotyped in this work, −6526A/G, is directly involved in causing the disease. Replication of these findings in more independent samples is also essential to clarify the aetiology of suicide. Functional variations remain to be identified and subsequently tested for association with suicide related behaviour.
We are deeply grateful to all of the families participating in this study, as well as the psychiatrists and mental health workers who helped us with identification of the families. This work was supported by grants from the Ministry of Education, PRC, the national 973 and 863 Projects, the National Natural Science Foundation of China, and the Shanghai Municipal Commission for Science and Technology.
5‐HIAA - 5‐hydroxyindoleacetic acid
HS - healthy subjects
LD - linkage disequilibrium
NSA - no suicide attempts
SA - suicide attempts
SNP - single nucleotide polymorphism
TPH - tryptophan hydroxylase
UNG - uracil‐N‐glycosylase
Competing interests: there are no competing interests.