Introduction

The neurotransmitter serotonin (5-hydroxytryptamine, 5-HT) plays an important role in neurodevelopment, functioning and plasticity of the brain. It is also involved in sleep, mood, aggression, affective disorder, obsessive-compulsive disorder (OCD), anxiety (Klauck et al. 1997; Guhathakurta et al. 2006) and altered neuroendocrine function (Williams et al. 2003). Serotonin is produced from its precursor 5-hydroxytryptophan (5-HTP). Once secreted into the synaptic spaces between axons and dendrites of neurons it binds to various pre- and post-synaptic receptors and transporters that allow activation or repression of serotonin signaling to other neurons as well as changes in gene expression. The serotonin transporter (SERT) strongly controls the duration and magnitude of serotonin in the synapse. Furthermore, it plays a critical role in regulating the homeostasis of spatial distribution and intensity of 5-HT signals with 5-HT receptors (Murphy et al. 2004; Hariri and Holmes 2006).

The serotonin transporter gene localized to chromosome 17q11 encodes the serotonin transporter protein. This gene spans a 31-kb region, consisting of 14 exons and codes for the 630 amino acid transporter protein (Bartlett et al. 2005). Polymorphisms in this gene that have been reported include: a 44-bp insertion/deletion polymorphism in promoter region and is known as 5-HTTLPR; a ∼380-bp deletion just before the start codon; a 17-bp variable number of tandem repeats (VNTR) in intron 2 and a single nucleotide polymorphism (SNP) in the 3′ untranslated region (UTR) (Murphy et al. 2004; Bartlett et al. 2005; Guhathakurta et al. 2006). The first two exons, exon 1a and 1b, are also alternatively transcribed (Ramoz et al. 2006). Both the promoter polymorphism and the 17 bp VNTR have been intensively studied and were found to have an effect on the expression levels of the serotonin transporter gene and ultimately serotonin signaling. The short and long alleles of the 5-HTTLPR have different transcription rates, which leads to different SERT mRNA, protein levels and 5-HT uptake activity (Murphy et al. 2004). The 5-HTTLPR promoter polymorphism contains different numbers of a 20–23 bp irregular repeat element (Hariri and Holmes 2006; Hu et al. 2006). In the case of long (L) allele 16 repeat elements are found while in the short (S) allele only 14 repeat elements are present (Hu et al. 2006; Murphy et al. 2004). Alleles with extra repeat elements, containing up to 20 copies, have also been found in some individuals (Hu et al. 2006; Murphy et al. 2004). In studies of individuals with psychiatric disorders like depression or autism, an improvement of symptoms was noted in individuals homozygous for L allele after treatment with selective serotonin reuptake inhibitors (SSRIs) (Pollock et al. 2000; White et al. 2005). SSRIs drugs are structurally similar to serotonin and serve to increase serotonin in the synapse by competing for binding to the SERT protein. This results in an increase in the duration and amount of serotonin in the synapse resulting in better 5-HT signaling between neurons. The disruption of the serotonergic system often leads to a range of psychiatric and medical disorders. Lower serotonin levels have been detected in children with autism in comparison to the normal children. The SSRI’s have been found to alleviate some autism-related symptoms like aggression and sleep disorders (Hollister 2000). Recently, another group of heterogenous disorders, Temporomandibular disorder (TMD), which affect temporomandibular joints and or masticatory muscles has been linked with L alleles (L and XL) of 5-HTTLPR (Ojima et al. 2007).

The S allele is associated with increased response to emotional stimuli and behavior (shyness, aggression, anxiety, etc.), depression and suicide. The S allele is also associated with depression in patients suffering from coronary heart disease, Parkinson’s disease, stroke and hip fracture (Otte et al. 2007). When 121 Chinese patients suffering from depression were treated with SSRIs, a significantly better response occurred in patients homozygous for the L allele compared to those who carried at least one S allele (Yu et al. 2002). Individuals with S allele showed either no symptom improvements or a worsening of symptoms when treated with these drugs (White et al. 2005). It is therefore important to determine an individual’s genotype before SSRIs are prescribed to patients.

Studies performed on various populations like Japanese, Americans and Europeans found different allele frequencies of L and S alleles in different populations (Guhathakurta et al. 2006; Hariri and Holmes 2006; Hu et al. 2006). Williams et al. (2003) reported an L-allele frequency of more than 70% in African and African–Americans, a frequency of roughly 50% in Europeans and a 30% or lower L-allele frequency in the Japanese population.

The allelic frequency in SA population for 5-HTTLPR polymorphism has not been determined to date. Furthermore, the SA population is very diverse due to its various religions and cultures. For this reason the frequencies of L and S alleles have to be determined for different ethnic groups in the healthy SA population before association studies regarding the 5-HTTLPR can be performed on disorders in which it is implicated. In the present study, the genotype and allelic frequency of the promoter region of the 5-HTTLPR were determined in the SA population and compared to population groups from several other countries.

Materials and methods

Ethical clearance

Ethical clearance was obtained from the UWC ethics committee and permission to work with healthy individuals was obtained from the Western Cape Education department. Since the results of this study will be used to compare the allelic distribution of 5-HTTLPR in children with autism, it was appropriate to obtain the samples from healthy children of approximately the same age. Approval was obtained from principals of various schools and consent forms were issued to the parents of learners. Only those children whose parents gave consent were used in our study group. Consent was given for 347 healthy SA individuals that consisted of 96 Caucasians, 70 Africans and 181 colored/Mixed groups.

Sample collection

Samples were collected from healthy individuals from primary and high schools all around Cape Town, Western Cape. Individuals participating in this study were swabbed by taking a sterile swab, rubbing it against the inside of their cheek for 1 min and then placing it into a sterile 15-ml grinder tube. Swabs were taken from both cheeks.

DNA extraction

The applicator (swab) tip was placed into an eppendorf tube to which 300 μl of 1× Phosphate buffer saline pH 7.4 (1.4 M NaCl; 0.027 M KCl; 0.101 M Na2HPO4; 0.018 M KH2PO4) was added. DNA was extracted from the swabs using either the cells to DNA (CTD) kit according to the manufacturers’ specifications (Genecare Molecular Genetics (Pty) Ltd.) or a modified protocol using the FlexiGene Qiagen kit. Only the first four steps of the FlexiGene protocol were changed, as follows:

  1. 1.

    The eppendorf tubes containing the swab tip and 1× PBS were vortexed until the solution turned milky

  2. 2.

    A measure of 300 μl was transferred to a clean 1.5 ml eppendorf tube

  3. 3.

    The tubes were then centrifuged at 12,000 rpm for 5 min and the supernatant was removed

  4. 4.

    A total of 300 μl of recommended buffer was added to the cell pellet and was mixed by pippeting.

The remaining steps were carried out according to the manufacturers’ specifications (Southern Cross Biotechnology (Pty) Ltd).

Sample analysis

The concentration of each DNA sample was determined using the NanoDrop® ND-1000 spectrophotometer (NanoDrop Technologies). Following this the samples were used in Polymerase Chain Reaction (PCR).

Genotyping

Primers used by Klauck et al. (1997) (forward: 5′-GGCGTTGCCGCTCTGAATGC-3′ and reverse: 5′-GAGGGACTGAGCTGGACAACCAC-3′) were used to determine the genotype of individuals. Primers were purchased from Inqaba Biotech (Pretoria, South Africa). The PCR reaction contained the following reagents in a final volume of 25 μl: 0.4 ng/μl BSA; 3 picomolar forward primer; 3 picomolar reverse primer; 1× PCR Buffer; 100 μM dNTPs; 1.5 mM MgCl2; 1 U BioTaq; 20–150 ng DNA and de-ionized autoclaved water. PCR conditions used were as follows: initial denaturation for 5 min at 95°C; 35 cycles of 95°C for 30 s; 63°C for 30 s; 72°C for 1 min and a final extension at 72°C for 10 min.

Gel electrophoresis

A total of 10 μl PCR product was mixed with loading dye (Fermentas O’ GeneRuler™ 6x Orange Loading Dye) and loaded next to a 100-bp DNA ladder (Fermentas O’ GeneRuler™ DNA Ladder Mix) in the gel. PCR products of samples were electrophoresed on a 1.2% agarose gel at a voltage of 70 V for one and a half hours.

Statistical analysis

Allele frequencies for the L and S alleles were determined using the Hardy–Weinberg equation. Chi-square testing was also performed at a confidence interval of 95%, to compare and determine the significance between different populations.

Results

The SA control population was subdivided into three major ethnic groups: Mixed, African and Caucasian population. After genotyping the 5-HTTLPR, five different genotypes were observed namely, L/L, L/S, S/S, L/XL and S/XL (Fig. 1). The extra-long (XL) allele has an additional 44-bp insertion. In total 3 L/XL and 2 S/XL variants out of the 347 control samples were identified. Variants viz., L/XL and S/XL genotypes were not included in the genotype and allele frequency calculations because of their low observed frequencies and since they are not present in all populations, and the L/L, L/S and S/S genotypes are the most commonly occurring genotypes in any given population. Thus, the SA control group (342 individuals) consisted of 178 Mixed, 68 African and 96 Caucasian individuals.

Fig. 1
figure 1

Gel photo of different genotypes observed in the populations. The sizes of the PCR products of the alleles were as follows: long allele 528 bp; short allele 484 bp and the extra-long allele 572 bp. Lane 1 100 bp MW DNA ladder, Lane 2 L/L, Lane 3 L/S, Lane 4 S/S, Lane 5 S/XL, Lane 6 L/XL

Full size image

Genotype distribution using the Hardy–Weinberg equation was determined for all the three SA ethnic groups. The genotype distribution observed in SA population along with other populations is presented in Table 1. Chi-square testing was performed to compare the number of alleles between different ethnic groups and also to other populations studied in different regions of the world at a confidence interval of 95% and df = 1 (Table 2).

Table 1 Genotype and allelic distribution of 5-HTTLPR in normal populations in different countries
Full size table
Table 2 Comparison of distribution of alleles in SA ethnic groups and other populations in different regions of world
Full size table

Allele frequency was calculated for the SA control group as a whole as well as the three major ethnic groups (Fig. 2). The comparison of Caucasian with African and Mixed population groups revealed significant difference in the frequency of S allele (P < 0.01). No significant difference was observed between allele (L and S) frequencies of African and Mixed groups.

Fig. 2
figure 2

Allele frequency of the 5-HTTLPR in the SA population

Full size image

Discussion

The present investigation aimed at identifying the allelic and genotype frequencies of 5-HTTLPR promoter polymorphism in healthy individuals of various ethnic groups residing in SA. The comparison of the genotype distribution of 5-HTTLPR in the SA population groups showed no significant differences between the African and Mixed population since both are of African descent/origin (Table 1). The L-allele frequencies for the Mixed and African groups were higher in comparison to Caucasians (Fig. 2). These results are in line with observations made in a study performed on 865 African–Americans and Afro-Caribbean’s by Lotrich et al. (2003), where they reported a high L-allele frequency of greater than 70%. The L and S allele frequencies for SA Caucasians (60.94 and 39.06%, respectively) is consistent with those reported for other Caucasian groups (Williams et al. 2003).

The genotypes of the three different ethnic groups in this study were also compared, using Chi-square testing, with other population studies reported from around the world (Table 2). No significant differences were noticed between the Caucasians from SA compared to Germans, Croatians, Europeans, British Caucasians, African–Americans and Israelis. Most of these populations are originally of European descent, so allele frequencies are not expected to differ significantly. A significant difference was expected between SA Caucasians and the African–Americans as studies have reported a higher L-allele frequency (>70%) in African–Americans; however, this was not the case. Possibly population interbreeding, classification or population size could be affecting this result. Significant differences were noticed between SA Caucasians and the Chinese, Indians and Japanese (Table 2). The large difference between SA Caucasians and the Japanese population indicates that these two populations are distantly related and diverse.

Both SA African and SA Mixed population showed significant differences in allele numbers to all the populations; namely Chinese, Germans, Croatians, Europeans, British Caucasians, Indians, Japanese and Israelis (Table 2). Once again the greatest difference occurred with the Japanese population, which suggests that the Japanese are the most distantly related to African populations. The theory of population evolution also accounts for this; namely that populations originated from Africa and the further they moved from Africa the further they diverged and evolved. Less difference was found between Africans from SA and African–Americans (Table 2). However, a greater difference was noted with the Mixed and African–American population in the S allele. A possible reason for this could be population interbreeding which caused a change in the number of S alleles in the populations.

Stoneking et al. (2007) performed an investigation on Alu insertion polymorphisms and human evolution. They found that in African populations Alu insert frequencies remained largely unchanged compared to the ancestral state. In non-African populations the frequencies for alleles differed from the ancestral state (Stoneking et al. 2007). Genetic drift of these alleles at a locus is influenced by population size and demographic history of a population. The possible explanation would be that non-African populations may have originated from African populations or that they diverged from the ancestral state by population expansion leading to allele frequencies drifting farther from the root (Stoneking et al. 2007). If this holds true for the 5-HTTLPR, then a possible explanation for the difference in allele frequencies is that the 5-HTTLPR polymorphism in the SA African and Mixed population is closer to the ancestral state. The heterogeneity amongst populations could be the cause of conflicting results reported by association studies like that of Klauck et al. (1997) and Cook et al. (1997). The higher frequency of L allele observed in African and Mixed group in the present study suggests that individuals of this ethnic background should respond to SSRIs better than Caucasian individuals in general. However, this aspect of the project needs expansion and more work is needed in support or negation of this observation.

An A to G SNP in the 5-HTTLPR was also reported to affect the expression of the SERT protein (Beitchman et al. 2006; Hu et al. 2006). The G nucleotide substitution is expressed at a lower level than in the case of an A nucleotide. This SNP can therefore make an L allele act like an S allele (Beitchman et al. 2006). This leads to the necessity for an in-depth investigation of the 5-HTTLPR as there are other regulatory elements that can affect its expression. Association studies therefore have to look at more than just the 5-HTTLPR.

The present study focused on identifying the frequencies of alleles in 5-HTTLPR polymorphism in the SA populations and comparison with other populations from around the world. Being the first study on 5-HTTLPR in the SA population it opens new doors for association studies. Future studies trying to link the 5-HTTLPR genotype to various disorders could benefit from this study and also the allele frequencies reported here for different ethnic groups could also act as a guideline for other studies aimed at determining the allele frequency of 5-HTTLPR in SA, or elsewhere. The shortcoming of this study is that the sample numbers used for the African and Caucasian population groups were small. Other polymorphisms of the SERT gene in larger population cohorts should also be investigated to discern the true effect of the polymorphism. This study will be expanded to estimate the allele frequencies of L and S alleles in individuals with autism in SA.