Journal of Molecular Neuroscience

, Volume 44, Issue 1, pp 59–67

Candidate-Gene Approach in Posttraumatic Stress Disorder After Urban Violence: Association Analysis of the Genes Encoding Serotonin Transporter, Dopamine Transporter, and BDNF

Authors

    • Departamento de PsiquiatriaUniversidade Federal de São Paulo
  • Homero Vallada
    • Departamento de Psiquiatria e LIM-23da Faculdade de Medicina da Universidade de São Paulo
  • Quirino Cordeiro
    • Departamento de Psiquiatria e LIM-23da Faculdade de Medicina da Universidade de São Paulo
  • Karen Miguita
    • Departamento de Psiquiatria e LIM-23da Faculdade de Medicina da Universidade de São Paulo
    • Instituto Adolfo Lutz
  • Rodrigo Affonseca Bressan
    • Departamento de PsiquiatriaUniversidade Federal de São Paulo
  • Sergio Baxter Andreoli
    • Departamento de PsiquiatriaUniversidade Federal de São Paulo
  • Jair Jesus Mari
    • Departamento de PsiquiatriaUniversidade Federal de São Paulo
    • Health Service and Population Research Department, Institute of PsychiatryKing’s College London
  • Marcelo Feijó Mello
    • Departamento de PsiquiatriaUniversidade Federal de São Paulo
Article

DOI: 10.1007/s12031-011-9513-7

Cite this article as:
Valente, N.L.M., Vallada, H., Cordeiro, Q. et al. J Mol Neurosci (2011) 44: 59. doi:10.1007/s12031-011-9513-7

Abstract

Posttraumatic stress disorder (PTSD) is a prevalent, disabling anxiety disorder marked by behavioral and physiologic alterations which commonly follows a chronic course. Exposure to a traumatic event constitutes a necessary, but not sufficient, factor. There is evidence from twin studies supporting a significant genetic predisposition to PTSD. However, the precise genetic loci still remain unclear. The objective of the present study was to identify, in a case–control study, whether the brain-derived neurotrophic factor (BDNF) val66met polymorphism (rs6265), the dopamine transporter (DAT1) three prime untranslated region (3′UTR) variable number of tandem repeats (VNTR), and the serotonin transporter (5-HTTPRL) short/long variants are associated with the development of PTSD in a group of victims of urban violence. All polymorphisms were genotyped in 65 PTSD patients as well as in 34 victims of violence without PTSD and in a community control group (n = 335). We did not find a statistical significant difference between the BDNF val66met and 5-HTTPRL polymorphism and the traumatic phenotype. However, a statistical association was found between DAT1 3′UTR VNTR nine repeats and PTSD (OR = 1.82; 95% CI, 1.20–2.76). This preliminary result confirms previous reports supporting a susceptibility role for allele 9 and PTSD.

Keywords

PolymorphismDAT1BDNF5-HTTLPRPosttraumatic stress disorderPTSDViolence

Introduction

Brazil has one of the highest rates of urban violence. Data from Health Ministry National Database of Mortality Information System showed that in 2007, there were 25 homicides per 100,000 inhabitants, which mean more than 47,000 homicides each year, or about 131 Brazilian citizens were killed every day (Waiselfisz 2010). Data from Brazilian Institute of Statistics and Geography show that 47.2% of the population report a feeling of continuous fear and insecurity (IBGE 2010).

Posttraumatic stress disorder (PTSD) is an anxiety disorder marked by behavioral, physiological, and hormonal alterations that occurs after experiencing or witnessing a traumatic event, which is necessary but not sufficient to determine the disorder development. The PTSD lifetime prevalence in the USA is around 6% of the general population. A chronic course commonly follows PTSD, leading to occupational difficulties, troubled social relationships, marital instability, and substance abuse (Ballenger et al. 2000).

Traumas related to combat exposure, intense interpersonal violence (such as sexual abuse and kidnapping), and atrocities as those experienced from those living in conflict zones may trigger the development of the PTSD in a higher level than those traumatic events related to natural disasters (Breslau et al. 1991, 1998; Breslau 2001; de Jong et al. 2001; Kessler et al. 2005). The risk of PTSD increases with the severity of trauma, so the environment plays an important role among persons who have experienced the most severe traumas.

There are increasing evidences of the role of the environment in the etiology of neuropsychiatric disorders; in most instances, such disorders arise through the interactions of genetic and environmental factors. Some types of trauma as natural disasters do not depend on individual characteristics but others do: the risk of trauma exposure is higher among persons who have personal or family histories of certain psychiatric conditions. Personal history of substance abuse disorder increases the risk of exposure to trauma by approximately 50% (Radant et al. 2001).

Although the etiology of PTSD is still unknown, there is evidence of a significant genetic component from family and twin studies (True et al. 1993; Sack et al. 1995; Yehuda et al. 1998, 2001, 2002). Therefore, a number of genetic investigations on PTSD have been published, particularly in the last 5 years. Candidate genes, particularly the ones based either on medication response or stress-related hormones, have been selected for molecular genetic studies on PTSD. Among these studies, the predominantly investigated loci for PTSD have been the ones related to the dopamine and serotonin systems, or the ones coding for glucocorticoid receptors and the brain-derived neurotrophic factor (BDNF) (Kolassa et al. 2010; Broekman et al. 2007; Kilpatrick et al. 2007; Binder et al. 2008; Koenen et al. 2008; Kolassa et al. 2010, 2009).

PTSD involves a dysfunction in acquisition, consolidation, evocation, and extinction of traumatic memories, with an excessive consolidation of traumatic memories (Lagarde et al. 2010), and the BDNF protein plays an important role on synaptic plasticity, interfering with the growth and survival of neurons (Martinowich et al. 2007). The BDNF gene presents an important polymorphism at the codon 66 (val66met, rs6265), a single nucleotide polymorphism (SNP) on human chromosome 11p14.1 leading to an exchange of amino acids from valine (val) to methionine (Met). BDNF not only exhibits neurotrophic actions but is also highly stress sensitive and involved in the regulation of hypothalamic neuropeptide (Tapia-Arancibia et al. 2004). One effect observed in carriers of the methionine allele is a difference in hippocampal morphology. In studies using structural magnetic resonance imaging scans, val/met individuals have repeatedly been shown to have a smaller hippocampal volume relative to controls who are homozygous for val allele (val/val) (Pezawas et al. 2004; Bueller et al. 2006). Since PTSD subjects show reduced hippocampus volume, BDNF val66met variant may be presumed to play a role in PTSD physiopathology as it is believed that changes in neuronal plasticity could compromise the ability to respond to stress (Kozlovsky et al. 2007) and also be associated to alterations in learning and memory observed in PTSD subjects. (Bremner et al. 2003). Little information exists for individuals who are homozygous for the Met allele (Met/Met) as this genotype is rare in the general population, comprising less than 4% of people in Caucasian populations (Gratacos et al. 2007). Despite these evidences, other studies have not found any association between polymorphisms in BDNF and PTSD (Olff et al. 2005; Lee et al. 2006; Zhang et al. 2006).

The dopamine transporter gene (DAT1), also known as SLC6A3, is located on chromosome 5p15, and its protein functions as the main regulator of dopamine levels in the synaptic cleft via reuptake of dopamine from extracellular to intracellular compartments, participating in the adjustment of dopamine availability. The level of dopamine varies in normal subjects and are reduced in Parkinson disease and elevated in attention deficit disorders and depression disorders. Miller and Madras (2002) identified a variable number tandem repeat (VNTR) element of 40 base pairs in the 3′UTR noncoding region of the DAT1 gene. Individuals can have alleles of this VNTR ranging from three to 12 copies, but the majority of individuals have alleles with either nine or ten copies. Despite being located in the untranslated region of the gene, this polymorphism has been associated with several clinical phenotypes associated with dysregulation of dopamine transmission. (Martinez et al. 2001). DAT is expressed in the brain in a region-specific manner, most abundantly in the striatum (the mesostriatal dopamine (DA) pathway), where its concentration is more than tenfold higher than in the frontal regions (Ciliax et al. 1999). In 2002, Segman et al. (2002) found an association between the presence of nine repeats in VNTR and chronic PTSD. More recently, a study showed that the presence of nine repeats is related to a risk six times higher to develop PTSD among pre-school children than ten-repeat carriers, and also that hypervigilance symptoms were significantly higher among the children who carried nine-repeat genotype (Drury et al. 2009). However, Bailey et al. (2010) found no association between this DAT1 polymorphism and neither to PTSD diagnosis nor to the total PTSD symptom severity in a multigenerational study of families exposed to Spitak earthquake (Armenia).

The human serotonin transporter gene (5-HTT) located on chromosome 17q11.1–q12 presents a 44-bp insertion/deletion polymorphism within the promoter region (5-HTTLPR), with two classical allelic forms, the long (l) variant and the short (s) variant, which has lower transcriptional activity than the l allele (Lesch et al. 1996).

Caspi et al. have shown that this functional polymorphism at the promoter region of the serotonin transporter gene has some effect on the risk for PTSD and depression: homozygote subjects with the s variant of this gene developed depression in a higher level after exposure to traumatic events. PTSD was also more frequent in the homozygote group than in the healthy controls (Caspi et al. 2003; Lee et al. 2005; Caspi and Moffitt 2006). In 2007, Kilpatrick et al. (2007)replicated the results of Caspi group, in which the “s” variant of the serotonin transporter gene increases the risk for depression and PTSD after traumatic environmental exposure (hurricane) in a condition of low social support (odds ratio = 4.5; 95% confidence interval (CI), 1.2–17.4). The same authors more recently evaluated how the influence of environmental factors such as unemployment and living in a place with high levels of criminality could interact with the genotype producing a higher risk to PTSD and found that both conditions also had a significant influence in the PTSD risk when associated to 5-HTTLPR genotype : crime rate showed an odds ratio = 2.68, (95% confidence interval: 1.09, 6.57) and unemployment rate showed an odds ratio = 3.67 (95% confidence interval: 1.42, 9.50). Interestingly, the s genotype was a risk factor for PTSD in high-risk environments and a protective factor on low-risk (less criminality and unemployment) environments (Koenen et al. 2009), suggesting that the social environment could modify the influence of the genotype in PTSD development.

Our present report aimed to further investigate the influence of the DAT1 3′UTR VNTR, 5-HTTPRL short/long variants and BDNF val66met polymorphism on the development of PTSD in an urban violence victim population.

Material and Methods

Subjects

There were two sources of subjects for the study: The first one was subjects (case-PTSD+ and controls-PTSD−) identified and selected through an epidemiological survey conducted in São Paulo, Brazil (Andreoli et al. 2009).

The inclusion criteria for this epidemiological survey were: being a victim of an urban violence that could be characterized as criterion A of Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria for a diagnosis of PTSD and being of both genders and aged between 18 to 60 years. After fulfilling these criteria, the World Mental Health WHO Composite International Diagnostic Interview was conducted. The presence of a lifetime history of bipolar disorder and psychotic disorders and the presence of substance dependence or abuse disorders (excluding nicotine and caffeine) in the previous 6 months were criteria for exclusion.

In this first source of subjects, two groups were created: individuals exposed to traumatic life experiences resulting in PTSD diagnosis (cases-PTSD+) and resilient subjects who are victims of traumatic life experiences but without PTSD diagnosis (controls-PTSD−). The two groups were compared in order to identify genetic variables that might protect or predispose to PTSD.

The second source of subjects was eligible patients admitted at the outpatient clinic from the violence program of the Federal University of São Paulo Hospital. A pre-request to be admitted to the violence program is having experienced a traumatic event as described in criterion A for PTSD diagnosis on DSM-IV.

Subjects from both sources were informed about the procedures of the study and were asked to formally consent to participate. The Structured Clinical Interview for DSM-IV Axis I and Axis II (SCID I and II) (Spitzer et al. 1992) was conducted on all patients by a trained psychiatrist. All instruments were translated to Portuguese and validated and adapted to the local cultural context.

Apart from the statistical analysis of PTSD+ and PTSD− groups, a third control group called “the community control group” also participated in the analysis. The community control group consisted of 335 subjects from the Genetic and Pharmacogenetics Program at the Institute of Psychiatry, University of São Paulo data bank, which were already genotyped controls for the three polymorphisms. This group was recruited from the Blood Transfusion Unit of the Hospital das Clínicas, University of São Paulo Medical School. A short interview was conducted, and subjects with a past history of drug abuse, use of an illegal drug, lifetime history of a psychiatric disorder, or suffering from a psychiatric condition at time of the evaluation were excluded. The presence of previous traumatic experiences was not evaluated in this group. The details of subject recruitment have been previously described (Valente et al. 2011).

A total of 99 subjects exposed to traumatic life experiences were selected for the study. Sixty-eight of them came from the epidemiological study (40 PTSD+ and 28 PTSD−), and 31 came from the outpatient clinic (25 PTSD+ and 6 PTSD−).

Ethical Approval

All of the subjects included in this study gave written informed consent, and this project was approved by the Ethical Committee of the Federal University of São Paulo.

Clinical Data

Each selected subject was evaluated using the following instruments:
  • The Structured Clinical Interview for DSM-IV-Axis I and II (SCID I and II) (Spitzer et al. 1992; First et al. 1995) is a semi-structured interview that allows for the diagnosis of Axis I and II disorders, respectively, according to DSM-IV criteria (American Psychiatric Association 1994);

  • Clinician-Administered PTSD Scale (Blake et al. 1995) A clinician rating scale for assessing current and lifetime PTSD: the CAPS-1. The Clinician-Administered PTSD Scale (CAPS) is a structured clinical interview administered by trained clinicians. It is a structured interview developed to diagnose PTSD and rate its severity. It is composed of 30 items to assess PTSD-related symptom frequency and severity. Scores range from 0 to 136, with scores classified as follows: subclinical, from 0 to 19; mild, from 20 to 39; moderate, from 40 to 59; severe, from 60 to 79; and extreme, 80 and above. Recently, it was validated and culturally adapted to Portuguese. (Pupo et al. 2011);

  • Beck Anxiety Inventory: The Beck Anxiety Inventory (BAI) is a self-administered 21-item questionnaire assessing intensity of anxiety symptoms (Beck et al. 1988);

  • Beck Depression Inventory (Beck et al. 1961): The Beck Depression Inventory (BDI) is also a 21-item self-report questionnaire used to assess depressive symptom severity in clinical settings. Scores range from 0 to 63, with depression classified as minimal when scores range from 0 to 11, mild from 12 to 19, moderate from 20 to 35, and severe from 36 to 63. It has also been validated for the Brazilian population (Gorenstein et al. 1999);

  • Early Trauma Inventory (Bremner et al. 2000): The Early Trauma Inventory (ETI) is a 56-item semi-structured interview that measures adults’ past traumatic experiences occurring during childhood and teenage years. The experiences are divided into four clusters: physical (9 items), sexual (11 items), psychological/emotional abuse (8 items), and general trauma (24 items). All ETI items are evaluated according to the frequency, stage of development, duration, and impact on the subject. The instrument allows us to calculate a general score, multiplying the frequency of each present item by the years of duration of them. The scores from each cluster are summed to obtain a total severity of early trauma. As there are differences on the number of items on each cluster, those with more items had higher weight on total score; for this reason, we also introduce weighted scores, dividing the sum of each positive answer by the number of items of each cluster. The total score was calculated by the sum of all positive items by the number of total number of items (56). In this way, the scores, either cluster or total, vary from 0 to 1, being 0 no early trauma, and 1 extreme early trauma. The ETI was also recently validated and culturally adapted to Portuguese (Mello et al. 2010);

DNA Extraction

Five milliliters of whole venous blood from each participant was collected and treated with sodium citrate anticoagulant. Genomic DNA was extracted by “salting out” method described by Miller et al. (1988).

Genotyping

Real-time PCR was performed in a volume of 1.0 ml at 50 ng/ul genomic DNA concentration. Tampon composition was ×10 PCR buffer : 500 mM KCl, 15 mM MgCl2, and 100 mM Tris–HCl, with a final volume of 25 μL. Primer sequences were: Primer F: 5′-TGT GGT GTA GGG AAC GGC CTG AG-3′ and Primer R: 5′-CTT CCT GGA GGT CAC GGC TCA AGG-3′. Amplification consisted in initial denaturation: at 94°C for 5 min followed by denaturation at 94°C for 30 s, hybridization at 62°C for 30 s, and extension at 72°C for 30 s for 35 times followed by a final extension at 72°C for 10 min. Fragment lengths were: 280 bp (5 repeats), 320 bp (6 repeats), 360 bp (7 repeats), 400 bp (8 repeats), 440 bp (9 repeats), 480 bp (10 repeats), 520 bp (11 repeats), and 600 bp (13 repeats). Amplification results were analyzed through electrophoresis and visualized by transillumination UV in 2% agarose gel embedded in etídeo bromide.

Real-time PCR for BDNF (rs6265) was performed for a final volume of 7 μL using 100 ng/μl of DNA. The equipment used was the 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA). Each reaction compromised 0.35 μL of Taq Man SNP Genotyping Assay (Applied Biosystems, Foster City, CA) and 3.5 μL de TaqMan Universal PCR Master Mix (Applied Biosystems, Foster City, CA). Cycling conditions were 50°C for 2 min, 95°C for 10 min, 45 cycles of 95°C for 15 s, and 60°C for 1 min.

For the serotonin transporter (5-HTTLPR), polymorphic region was amplified by polymerase chain reaction (PCR) with oligonucleotide primers: forward 50-GAG GGA CTG AGC TGG ACA AC-30, reverse-50-GCA GCA GAC AAC TGT GTT CAT C-30. PCR started with an initial denaturation at 95°C for 3 min, followed by 45 s at 95°C, 45 s at 61.2°C, 45 s at 72°C for 30 cycles, and a final extension at 72°C for 3 min. PCR products were separated on a 3% agarose gel containing ethidium bromide, and bands were visualized under UV light. For the 5-HTTLPR genotype, the resulting product with 585 bp carries the short allele (s), whereas the resulting product with the 629 bp carries the long allele (l).

Statistical Analysis

Chi-square tests were performed to verify Hardy–Weinberg equilibrium (HWE) on groups and also to compare allelic and genotypic frequencies. For continuous variable comparison, the Kruskal–Wallis test was performed. Software R (Team RDC 2008) was used for the statistical analysis, and significance level used for all tests was 5%.

Results

Demographic Data

The socio-demographic characteristics of the sample are summarized in Table 1. Both groups (cases: PTSD+ and controls: PTSD−) did not differ statistically concerning marital status, gender, religion, and years of education. However, the case group (patients or PTSD+) was younger than the control (PTSD−) group (p < 0.02; Table 1).
Table 1

Summarized demographic data from our 99 subjects

 

PTSD+, n = 65 (%)

PTSD−, n = 34 (%)

p

Marital status

 Single

34.8%

23.5%

0.59

 Married

54%

64.7%

 

 Divorced

10.6%

8.9%

 

 Widows

1.5%

3.1%

 

Gender

 Men

33.3%

17.6%

0.11

 Women

66.7%

82.4%

 

Religion

 Catholic

44.6%

50%

0.48

 Evangelical

30.3%

14.7%

 

 Spiritist

10.7%

17.6%

 

 Other

8.9%

11.8%

 

 No religion

5.3%

2.9%

 

Average age

 Years (±SD)

37.9(±8.7)

44(±13.8)

0.02 (T)

Years of education

 <8 years

8 (12.2)

4 (12.1)

0.24

 8< × <12 years

15 (22.8)

3 (9.1)

 

 >12 years

43 (65.1)

26 (78.8)

 
The distribution of different types of traumatic event differs between groups, for “kidnapping” (p < 0.006) and “being witness of violence” (p < 0.05), which were more prevalent in the PTSD+ group (Table 2).
Table 2

Distribution of the different types of traumatic events in case and control groups

Traumatic event

PTSD+

%

PTSD−

%

p

Sexual abuse

8

13.8

3

9.8

 

Robbery

13

20

8

23.5

 

Kidnapping*

10

15.4

1

2.9

*0.006

Physical aggression

5

7.7

4

11.8

 

Violent death in family

11

17

9

26.5

 

Domestic violence

1

1.5

4

11.8

 

Witness of violence*

6

9.2

1

2.9

*0.05

Threatening of physical integrity

4

4.6

2

5.9

 

Severe accident

7

10.7

2

5.8

 

Total

65

100

34

100

 

*p values

Clinical Data

The frequencies of other psychiatric disorders in both groups were higher than expected in normal population. The presence of a current depressive episode (major depressive disorder (MDD)) was the most common comorbidity, as shown on Table 3.
Table 3

Psychiatric comorbidity

Psychiatric comorbidity

PTSD+

PTSD−

Current major depressive disorder

48.2%

39%

Past major depressive disorder

23.2%

24%

Panic disorder

8.9%

0

Generalized anxiety disorder

5.3%

12.1%

Psychotic symptoms

5.3%

0

Obsessive complusive disorder

1.8%

0

As expected, the CAPS [PTSD+ = 67.8(±25.3), PTSD− = 29.2(±21.2), T = 7.557, p < 0.001], BDI [PTSD+ = 24.63(±13.6), PTSD− = 14.7(±11.9), T = 3.4 p < 0.001], and BAI [PTSD+ = 30.2(±13.6), PTSD− = 17.8(±15.5), T = 3.63, p < 0.001] scores were significantly higher in PTSD group. Although early life stress has been related to risk factor for depression and PTSD, we did not find a difference regarding the presence of a previous traumatic event during childhood or adolescence, measured by early trauma inventory (ETI) administration on either group [PTSD+ = 161.7(±197), PTSD− = 160.11(±165), T = 0.46, p = 0.64).

The presence of current and lifetime major depressive disorder was correlated with significant differences on CAPS scores. Subjects with diagnosis of current depressive disorder scored higher on CAPS than those without depression [PTSD+/MDD+ = 61.6(±25), PTSD+/MDD− 47.2(±17), p = 0.02].

Genotyping Data

We did not find statistically significant differences on genotypic and allelic frequencies of the BDNF met/val polymorphism and short/long variants of serotonin transporter gene on PTSD+ and PTSD− groups and community control group. For the DAT1 3′UTR VNTR, a difference was found between PTSD+, PTSD−, and community control group (p = 0.02; Table 4).
Table 4

BDNF, DAT1, and 5-HTTLPR genotypes on PTSD+, PTSD−, and community control groups

BDNF

PTSD+

PTSD−

Controls

p value

H–W

Val/val

48

29

555

0.65

0.543a

Val/met

15

5

164

 

0.901b

Met/met

2

0

14

 

0.643c

DAT1

10|10

20

15

168

0.028

0.538a

9|10

30

16

101

 

0.661b

9|9

8

3

28

 

0.030c

5-HTTLPR

ll

21

5

100

0.27

0.610a

ls

29

20

134

 

0.261b

ss

13

9

72

 

0.041c

aPTSD+

bPTSD−

cControls

Between PTSD+ and community control group, a difference was found regarding allelic frequencies at the level of p = 0.008 and genetic frequencies at the level of p = 0.004 (Table 5).
Table 5

DAT1 genotype on PTSD+ and community control group

DAT1

PTSD+

Controls

p value

10|10

20

168

0.008a

9|10

30

101

0.004b

9|9

8

28

 

aGenotype

bAlelle

When comparing PTSD+ and PTSD− groups, no statistical difference was observed (p = 0.3).

Statistical analysis of the genotype and clinical measurement scores were also performed. No correlation could be made between genotype and severity of symptoms. However, subjects carrying the ten-repeat allele had higher scores of early trauma than homozygote for nine-repeat allele, even though not statistically significant (Table 6).
Table 6

Clinical measures and DAT1 3′UTR VNTR

DAT1

10|10

9|10

9|9

 

Average

SD

Average

SD

Average

SD

p value

CAPS

53.40

33.04

53.37

30.29

56.20

18.90

0.8970

CapsB

15.00

10.99

13.24

9.97

19.00

7.72

0.2340

CapsC

20.33

14.39

22.29

14.31

21.40

11.02

0.8440

CapsD

18.26

12.46

17.84

10.27

15.80

6.46

0.8270

BDI

19.78

14.87

21.49

14.30

25.11

9.48

0.4100

BAI

26.15

16.38

23.95

18.47

25.22

13.02

0.7020

ETI

152.05

178.83

182.61

188.01

64.96

60.06

0.0930

Discussion

The present study investigates the BDNF val66met (rs6265), the DAT1 3′UTR VNTR, and the 5-HTTPRL short/long variants and the development of PTSD on individuals exposed to urban violence. The demographic data from our sample are similar to previous epidemiological studies, which report that more women than men are affected by PTSD (Kessler et al. 1995; Breslau et al. 1997; Breslau 2001). Specific types of violence were more related to PTSD development, such as being kidnapped (p = 0.006) or being a witness of violence (p = 0.05).

In addition, in our investigation, the PTSD group demonstrated a higher prevalence of other psychiatric disorders, depression (MDD) being the most common observed comorbidity.

Of those three investigated polymorphisms, only the allele 9 of the DAT1 was associated with an increased risk to develop PTSD after being exposed to urban violence. The DAT1 nine-repeat allele had been previously associated with higher risk for PTSD development (Segman et al. 2002). Our findings confirm that subjects carrying the nine-repeat allele present a higher risk to develop PTSD, with an OR of 1.82 (95% CI, 1.20–2.76). A possible explanation could be that nine-repeat allele could increase the risk to develop PTSD or be in strong linkage disequilibrium with a disorder allele. However, there were no statistical associations between the genotype and severity of symptoms in our sample, but it was observed that the subjects carrying nine-repeat allele showed higher scores in CAPS (specially CAPS B) scores and BDI scores. A previous study has found that class B symptoms in PTSD (intrusion) are the most heritable PTSD symptom category (0.75) (Bailey et al. 2010).

It is also interesting to observe that homozygous subjects for nine repeats had lower scores of early traumatic events than those heterozygote or homozygote for ten-repeat allele, which could interfere in the phenotype since early trauma had been previously associated with higher risk to develop PTSD (Yehuda et al. 2010).

Taken together, these findings and our previously reported association between COMT val158met polymorphism (Valente et al. 2011) in the same sample may suggest that dopamine may be involved in the response to stress and to the PTSD development in our sample. Other studies have already suggested that dopamine plays an important role in the pathogenesis of PTSD (Comings et al. 1996; Segman et al. 2002; Bressan et al. 2009).

It cannot be excluded that our observed positive association can be due to ethnic admixture (population stratification bias) between the tested groups. The Brazilian population has a high admixture, which could pose a problem for interpreting the results of this study since particular genetic admixture condition makes it more difficult to achieve ethnic matching in case–controls studies. In the Brazilian population, phenotypic information determined by physical evaluation is a poor predictor of genomic ancestry (Parra et al. 2003; Pimenta et al. 2006).

The fact that in the present study PTSD+ and PTSD− were in Hardy–Weinberg equilibrium may indicate that population stratification was not necessarily a problem. Nevertheless, we are aware that ethnic admixture may have been a limitation of the present work, and the results may not be generalized. It is also possible that the investigated BDNF and serotonin transporter polymorphisms have a smaller attributable effect in this population, or the size under study has not been large enough to detect an association. Another limitation is that the findings of this study may only be applicable to victims of urban violence, in an environment of higher risk. Another possible limitation of the present work is that the community control group is not in HWE for the DAT1 3′UTR VNTR and 5-HTTPRL polymorphisms. However, the HWE deviation of the DAT1 3′UTR VNTR polymorphism, which was associated to PTSD in the present work, is not very important (p = 0.04).

We should also mention that some of our subjects in the trauma control group can still develop PTSD since there are some evidence that previous assaultive violence could be a risk factor for subsequent PTSD after a new traumatic event (even if non-assaultive), specially in women (Breslau and Anthony 2007).

Other studies with larger samples and improved methodology will be necessary to confirm that allele 9 of the DAT1 polymorhism is a risk factor to develop PTSD on individuals exposed to urban violence.

Taking into account that PTSD is a complex disorder, other risk factors could be contributing for the development of the symptoms in our subjects. It is important to mention possible epigenetic influences since DAT expression is highly dynamic—the brain's DAT level adjusts to accommodate DA signaling (Zahniser and Sorkin 2004), changing markedly in response to environmental factors (Swanson et al. 2007).

Acknowledgments

This study was partly funded by Fundacão de Amparo a Pesquisa do Estado de São Paulo (grant: 2004/15039-0). NLMV received a scholarship from the Ministry of Education (CAPES). JJM is a CNPq level I researcher, and MFM, RAB, and SBA are CNPq level II researchers. We would like to thank Thais Chile and Katia Ichi for their technical contribution for the lab work.

Copyright information

© Springer Science+Business Media, LLC 2011