Progress in Neuropsychopharmacology & Biological Psychiatry 88 (2019) 375–387

Contents lists available at ScienceDirect

Progress in Neuropsychopharmacology
& Biological Psychiatry
journal homepage: www.elsevier.com/locate/pnp

The influence of the serotonin transporter gene 5-HTTLPR polymorphism on T

suicidal behaviors: a meta-analysis
⁎
Giuseppe Fanelli, Alessandro Serretti
Department of Biomedical and NeuroMotor Sciences, University of Bologna, Bologna, Italy

A R T I C LE I N FO A B S T R A C T

Keywords: Suicidal Behavior (SB) is the second leading cause of death among youths worldwide and the tenth among all age
Suicide groups. Inherited genetic differences have a role in suicidality with heritability ranging from 30 to 55%. The
5-HTTLPR SLC6A4 5-HTTLPR gene variant has been largely investigated for association with SB, with controversial results.
Meta-analysis In this work, we sought to determine whether the results of previous meta-analyses were confirmed or
Serotonin transporter gene
modified subsequent to the inclusion of more recent literature data.
Depression
Personality traits
An electronic literature search was performed to identify relevant studies published until July 2018. Data
were analysed through RevMan v5.3. Subgroup and sensitivity meta-analyses were performed considering dif-
ferent SB sub-phenotypes, ethnicity, gender and psychiatric diagnostic categories.
Our literature search yielded 1186 articles; among these, we identified 45 pertinent case-control studies
(15,341 subjects). No association was found between low-expressing alleles or genotypes (S + LG alleles or S′
carrier genotypes) and SB in the primary analyses. However, low-expressing alleles (S + LG) were associated
with an increased risk of Violent Suicide Attempt (OR = 1.44, C.I. 1.17–1.78, p = .0007). An effect of the same
alleles on SB was found in a subpopulation of substance abusers, but this result was not confirmed after the
exclusion of healthy subjects from the control group. The other sensitivity meta-analyses did not show any
significant effect.
Our findings contribute to clarify the conflicting previous evidence by suggesting an association between the
5-HTTLPR and Violent SB. Nonetheless, many other modulators, including environmental factors and epigenetic
mechanisms may act to further increase the level of complexity.

1. Introduction Tidemalm et al., 2011), with heritability ranging from 30 to 55%

According to World Health Organization (WHO, 2014) Suicidal
Behavior (SB) may be defined as “a range of behaviors that include
thinking about suicide (or ideation), planning for suicide, attempting
suicide and suicide itself”. Every year, about 800,000 people die by
suicide worldwide (WHO, 2018) with an estimated age-adjusted rate of
13.4 per 100,000 population (21.3 for males and 6 for females) (CDC,
2016) and many more people attempt suicide. Therefore, suicide may
be considered a serious public health issue, as it is the second leading
cause of death among people aged 10 to 34 and the tenth among all age
groups (CDC, 2016; WHO, 2018). SB does not exclusively concern the
most developed nations, as more than 78% of suicides occur in low and
middle-income countries (WHO, 2018) with a substantial variation
across ethnicities.
Inherited genetic differences have a relevant role in suicidality, as
shown by family, adoption and twin studies (Brent and Melhem, 2008;
(Voracek and Loibl, 2007). In fact, in a large case-control study,
Mittendorfer-Rutz et al. (2008) pointed out an increased risk of suicide
attempts (in a range from 2.7 to 9.8 times) in the presence of family
members with a previous history of SB. Furthermore, a meta-analysis
showed that monozygotic twins' concordance for the completed suicide
is notably higher than in dizygotic twin pairs (respectively, 24.1% and
2.8%) (Voracek and Loibl, 2007). These data suggest that the likelihood
of occurrence of SB is significatively greater in those who share a si-
milar genetic background than in the general population. It should be
noted that, although a genetic effect may explain the higher con-
cordance among monozygotic twin pairs, a concordance approaching
100% would be expected if suicide was only due to a genetic aetiology.
It means that the environmental component also plays at least an equal
role in the onset of SB. Indeed, distal (predisposing) factors seem to
interact with proximal (precipitating) factors in determining the sui-
cidal event; in this way, genetic predisposition as well as early-life

⁎
Corresponding author at: Department of Biomedical and Neuromotor Sciences, University of Bologna, Viale Carlo Pepoli 5, 40123 Bologna, Italy.
E-mail address: alessandro.serretti@unibo.it (A. Serretti).

https://doi.org/10.1016/j.pnpbp.2018.08.007
Received 17 June 2018; Received in revised form 31 July 2018; Accepted 13 August 2018
Available online 17 August 2018
0278-5846/ © 2018 Elsevier Inc. All rights reserved.
G. Fanelli, A. Serretti
adversities and associated epigenetic modifications, which together
modulate behavior and personality traits, provide the basis for the in-
tervention of proximal triggers, such as recent life events and acute
major psychiatric disorders episodes (Mann, 2003; Turecki, 2014).
Noteworthy, roughly 90% of suicide completers may be diagnosed with
a mental disorder prior to their death (Arsenault-Lapierre et al., 2004);
however, it is worth considering that the family clustering of suicide,
although partially coincident, is independent of that of psycho-
pathology (Brent et al., 1996; Kim et al., 2005; McGirr et al., 2009).
In the last decades, many efforts have been made in order to dis-
entangle the genetics and the neurobiological basis underpinning SB.
Particular attention has been paid to the serotonergic system (Antypa
et al., 2013). Earlier studies demonstrated a lower concentration of the
serotonin metabolite 5-hydrxyindolacetic acid (5-HIAA) in the cere-
brospinal fluid of depressed patients prone to develop SB (Asberg et al.,
1976) as well as lower levels of serotonin (5-HT) and 5-HIAA in the
brainstem of suicide victims (Mann et al., 1989). Moreover, a phar-
macologic challenge with Fenfluramine, a serotonin agonist which
provokes an increase of the serotonin in brain synapses (DrugBank,
http://www.drugbank.ca/drugs/DB00574), has shown blunted pat-
terns of prefrontal cortex (PFC) activation in SB (Desmyter et al., 2011),
suggesting an impaired function of the serotonin neurotransmission.
Among the serotonin system candidate genes for SB, many genetic
association studies have focused on the SLC6A4 (Solute Carrier Family
6, Member 4) gene (Antypa et al., 2013; Gonda et al., 2011), located on
chromosome 17 (17q11.2). This gene encodes for the serotonin trans-
porter, a transmembrane presynaptic protein involved in the reuptake
of the released serotonin from the synaptic cleft (Daws and Toney,
2007; Rudnick and Clark, 1993). In humans, the transcriptional activity
of SLC6A4 gene is modulated by a 44 base-pair insertion/deletion
polymorphism, commonly known as 5-HTTLPR (Serotonin Transporter
Linked Polymorphic Region polymorphism - rs4795541), located up-
stream of the transcription start site. Most frequently, it is composed by
the repetition of 14 (for the S - short - allele) or 16 (for the L - long -
allele) repeated elements (Murphy et al., 2004). The S allele causes
reduced expression of the SLC6A4, in a dominant way compared to the
L allele, which is characterised, conversely, by higher levels of ex-
pression (Heils et al., 1996; Lesch et al., 1996). Nevertheless, the effect
of rs25531 SNP (A- > G substitution), located within the L allele of the
5-HTTLPR (Nakamura et al., 2000), may result in a similar transcription
activity of the LG and S allele (Hu et al., 2006; Wendland et al., 2006).
Thus, the LA and LG allele, together with the S allele, may be considered
as the components of a triallelic SLC6A4 locus. Post-mortem and in-vivo
studies have also investigated the role of the 5-HTTLPR on SB. Of note,
it has been shown that depressed suicide victims had a smaller amount
of serotonin transporters in the PFC, hypothalamus and brainstem
(Purselle and Nemeroff, 2003). Moreover, a SPECT study pointed out a
lower serotonin transporter availability in the frontal cortex of male
suicide attempters carrying the S allele when compared to healthy
controls (Bah et al., 2008). Most importantly, a number of studies have
indicated an increased risk of SB among people carrying the S allele and
reporting a previous history of stressful life events or early life adver-
sities (Mandelli and Serretti, 2013); this is consistent to the existence of
a reasonable gene-environment interaction effect.
Controversial results have been obtained from numerous case-con-
trol association studies, which have so far examined the influence of the
5-HTTLPR on SB. As a consequence, a number of meta-analytic studies
have attempted to summarize previous single study data in order to
elucidate the role of the 5-HTTLPR polymorphism on SB. At first
(Anguelova et al., 2003), a significant association for the S allele with
SB was reported (p = .009); the association was particularly valid
among attempters (SAs), though only three studies for suicide com-
pleters (SCs) were available at that time. A subsequent meta-analysis
(Lin and Tsai, 2004), confuted the prior results and pointed out the lack
of association between the 5-HTTLPR polymorphism and SB (p = .379);
nevertheless, they found a significant association between the
376
Progress in Neuropsychopharmacology & Biological Psychiatry 88 (2019) 375–387
genotypes carrying the S allele and violent suicide, when cases were
compared to healthy controls. A few years later, with an increased
sample size, Li and He (2007) described a robust positive association
(p = .007) of the 5-HTTLPR with SB, both for pooled and sub-grouped
samples; the association was independent of the psychiatric diagnoses,
as suggested by the stronger association obtained comparing SAs and
non-attempter patients rather than with healthy controls. Noteworthy,
Clayden and colleagues' meta-analysis (Clayden et al., 2012) showed a
significant association between attempted suicide and the S allele, but
not with the overall SB, suggesting that it would be better to consider
SB as distinct phenotypes. A very small statistically significant effect of
the 5-HTTLPR L/S on SB was achieved in Schild and coworkers' meta-
analysis (Schild et al., 2013) only when cases were compared to healthy
controls. Lastly, de Medeiros Alves et al. (2015), in another meta-ana-
lysis, confirmed an association between the 5-HTTLPR and SB, showing
an increased risk of SB for the L allele among patients affected by
psychiatric disorders.
Given the abovementioned discrepancies, we re-evaluated the pos-
sible modulatory effect of the 5-HTTLPR on SB with the inclusion of
more recent literature data. Furthermore, we tested whether the effect
of the 5-HTTLPR polymorphism may vary depending on gender, eth-
nicity, a shared psychiatric diagnosis or by using more specific suicidal
phenotypes as an outcome. The clarification of the role of genetic
variability in SB may be of particular importance in order to help
clinicians to better stratify subjects based on their suicidal genetic risk
as well as to develop better prevention and therapeutic strategies.
2. Methods
2.1. Search strategy
An electronic search of the literature was performed to identify
association studies investigating the link between the 5-HTTLPR and
SB. Medline/PubMed, Web of Science, Scopus and PsycINFO databases
were searched for papers published until July 2018. We used search
terms related to the serotonin transporter gene and its polymorphism,
such as “5-HTTLPR” or “HTTLPR” or “SLC6A4” or “SERT gene” or “5-
HTT gene” or “5HTT gene”, and terms referring to the outcomes, such
as “suicide” or “suicidal” or “suicidal behavior” or “suicide attempt” or
“suicide attempters” or “suicide completers” or “suicide completer” or
“completed suicide” or “attempted suicide”, excluding reviews as
publication type and including only papers written in English language.
The reference lists of the selected articles were inspected to retrieve
additional papers not indexed by the scientific literature databases.
2.2. Study selection and data extraction
The two reviewers independently screened searches to identify po-
tentially relevant studies. The full text of the selected studies was ob-
tained and evaluated to detect pertinent studies. Studies were included
if: 1) they were genetic association studies evaluating the relationship
between the 5-HTTLPR polymorphism and SB; 2) the study did not
overlap with others taking into account a larger sample; 3) they were
written in English. For each selected study, the following information
was extracted: first author, publication year, location/ethnicity, sample
size, characteristics of the control sample (presence of healthy or non-
healthy subjects), 5-HTTLPR genotypic or allelic frequencies, triallelic
or biallelic genotyping approach, gender ratio, psychiatric diagnosis
and comorbidities, suicide phenotypes tested. Nevertheless, data were
not available for some studies. When genotypic or allelic frequencies
were partial or entirely not available in the research paper or the
supplementary information, we extracted data from the presented
graphs by g3data software (https://github.com/pn2200/g3data) or we
requested data from the corresponding author.
G. Fanelli, A. Serretti
2.3. Outcomes and comparisons
The primary outcome was defined as Suicidal Behavior (SB), ac-
cording to the definition of Mann (2002), which includes only Com-
pleted Suicide (SC) and Attempted Suicide (SA); these two sub-pheno-
types were used for subgroup analyses and to compute the pooled OR.
Suicidal ideation and planning were excluded from the meta-analytic
process because they may act as confounding factors, since their higher
lifetime prevalence among the general population (respectively, 9.2%
and 3.1%) (Nock et al., 2008) and among patients with depression
(respectively, 47.7% and 22.6% in the Canadian population) (Patten
et al., 2015). Furthermore, there is evidence that suicidal ideation
should be considered distinct from the phenotypic spectrum of SA and
that familial liability to SB includes SA and SC and does not include
suicidal ideation (Brent et al., 1996; Mullins et al., 2014). The sec-
ondary outcome considered was Violent Suicide (VS), usually defined
according to the criteria proposed by Asberg (ÅSberg et al., 1986;
Asberg et al., 1976) who suggested that all suicides, other than drug
overdoses, should be considered as violent (e.g. hanging, drowning,
poisoning, the use of firearms, jumping from heights, run over by any
vehicle, cutting, car crashing, burning, etc.).
We tested the comparison of the S′ carrier genotypes (SS, SL) (low
expression) versus LL genotype (high expression), assuming that the S
allele had a dominant effect on the L allele, as previously reported
(Heils et al., 1996; Lesch et al., 1996). For studies examining the 5-
HTTLPR as a triallelic polymorphism, we included the SS, SL, SLG, SLA,
LGLA, LGLG in the S′ carrier genotypes group and the LALA was combined
with the LL genotype (Hu et al., 2006; Wendland et al., 2006). Because
LG drives expression nearly equivalently as S, studies that include many
LG alleles within the SL and LL genotypes may underestimate the effect
of the 5-HTTLPR (Hu et al., 2006). For the allele-wise analyses, the LG
allele was matched with the S allele and the LA allele was matched with
the L allele.
2.4. Statistical analysis
Data were entered and analysed with the Cochrane Collaboration
Review Manager Software, RevMan version 5.3 (RevMan v5.3, 2014).
Individual and pooled 95% confidence intervals (C.I.) were calculated.
Heterogeneity between studies was assessed with Tau-squared for
random effects model meta-analysis (τ2, or Tau2), Chi2-test of fit (Co-
chran Q test) and I2 measure. I2 values of 25%, 50% and 75% are
generally considered low, medium and high heterogeneity, respectively
(Higgins et al., 2003). The significance of the pooled effect size was
determined using a Z test. Data were analysed using a Mantel–Haenszel
random effects framework because of the assumption of the presence of
a significant between-study heterogeneity, in particular due to the
ethnic heterogeneity of the samples. A random effects framework as-
sumes that between-study variation is due to both chance or random
variation and an individual study effect. Furthermore, a funnel plot was
created to reveal selective reporting (publication bias) by plotting the
natural logarithm of individual study effect size.
Taking into account the heterogeneity between the studies included
in the primary meta-analyses, we conducted separate meta-analyses
considering case-control subjects homogeneous for the ethnicity,
gender or a triallelic genotyping approach. We also performed separate
sensitivity meta-analyses for three psychiatric diagnostic categories
(mood disorders, schizophrenia spectrum disorders and substance
abuse disorders) comparing suicidal patients versus non-suicidal pa-
tients affected by the same psychiatric disorder, firstly by considering
healthy subjects among the control group and then excluding them
from the analyses. Besides, using Violent Suicide as the secondary
outcome, we tested Violent Suicide versus non-Violent Suicide subjects
and Violent Suicide versus non-healthy control subjects + healthy
control subjects. We could not exclude possible other sources of het-
erogeneity among the included studies. Indeed, stratification factors
377
Progress in Neuropsychopharmacology & Biological Psychiatry 88 (2019) 375–387
can derive from the inclusion of different subethnic groups among the
principal ones, different patients' psychiatric comorbidities and dif-
ferent clinical-demographic features (e.g. age of onset of the illness,
number of hospitalisations, number of previous attempts, years of
treatment, personality traits and different degree of impulsivity and
aggression). Given the prior evidence and the confirmatory nature of
the present study, we did not apply the correction for multiple testing.
However, analyses were divided into primary and secondary ones.
3. Results
The initial literature search yielded 1186 articles; these articles have
been revised in order to investigate if they fulfilled the inclusion cri-
teria. Initially, we selected 102 studies through the title and abstract
inspection. In the following step, after a careful examination of the full
text, 57 articles were excluded because they investigated other SLC6A4
polymorphisms for the association with SB (n = 8), they used a sample
overlapping with previous studies (n = 13) or they lacked a control
group (n = 8), data were not available or partly reported (n = 19), they
used continuous measures or scales of suicidality (n = 3) or they mainly
focused on a different outcome such as suicidal ideation (n = 2).
Moreover, we excluded a number of studies which had Deliberate Self-
Harming (DSH) or self-harming or self-mutilation as primary outcomes
(n = 4). Noteworthy, the term DSH is used to describe self-harming acts
regardless of a real suicidal intent, that often follow irresistible impulses
of harming oneself with a subsequent psychic relief from internal ten-
sion and anger (Kreitman et al., 1969; Lauw et al., 2015; Pattison and
Kahan, 1983; Skegg, 2005). For this reason, we have excluded Pooley
et al., 2003 (Pooley et al., 2003), whose study was included in the
previous meta-analyses. When samples were overlapping, only the
study with the larger sample size was included. Conversely, we re-
trieved some GxE studies (n = 5) (see Fig. 1 - Study Flow). Vincze et al.,
2008 (Vincze et al., 2008) was included in previous meta-analyses,
although we decided to exclude it to avoid overlaps with Bellivier et al.,
2000, Courtet et al., 2001, Courtet et al., 2003 (Bellivier et al., 2000;
Courtet et al., 2001; Courtet et al., 2003); overall, the sample of Vincze
et al., 2008 was smaller than the sum of the other three samples. After
the inclusion and the exclusion criteria were applied, 45 studies were
included in the current meta-analyses (overall for the genotype-wise
and the allele-wise analyses) (see Table 1).
3.1. Primary analyses: allele-wise and genotype-wise meta-analyses of the
5-HTTLPR polymorphism on Attempted Suicide and Completed Suicide
For the allele-wise analysis, a total of 41 studies were included
(14,842 subjects in all, 12 studies and 4252 subjects for the SC sub-
group analysis; 30 studies and 10,771 subjects for the SA subgroup
analysis; Akar et al. (2010) sample was split into two subgroups, re-
spectively for SC and SA). For the genotype-wise analysis, a total of 41
studies were included (14,610 subjects in all, 9 studies and 3707 sub-
jects for the SC subgroup analysis; 32 studies and 10,903 subjects for
the SA subgroup analysis).
No significant association was found neither in the allele-wise
analysis between S + LG alleles (low-expressing alleles) and SB
(OR = 0.97, C.I. 0.88–1.06, p = .45) (Fig. 2), nor in the genotype-wise
analysis between S′ carrier genotypes and SB (OR = 1, C.I. 0.87–1.14,
p = .99), pooling S′ carrier genotypes versus LL + LALA genotypes (Fig.
S1). Heterogeneity (between-studies variability) was moderate for both
the allele-wise and genotype-wise analyses. Moreover, no publication
bias was detected through the funnel plots inspection for both the
primary meta-analyses (Figs. S2 and S3).
Campi-Azevedo et al., 2003 (Campi-Azevedo et al., 2003), Gibb
et al., 2006 (Gibb et al., 2006), Neves et al., 2010 (Neves et al., 2010)
were considered only in the genotype-wise analysis because they re-
ported only genotype frequencies grouped for S carrier. Akar et al.,
2010 (Akar et al., 2010), Mann et al., 2000 (Mann et al., 2000) and
G. Fanelli, A. Serretti
Fig. 1. Study flow.
Linkowska et al., 2010 (Linkowska et al., 2010) have been taken taken
into account only in the allele-wise analysis because of missing geno-
type frequencies. For Akar et al., 2010, the same control group was
included two times, respectively once in the SC and once in the SA
subgroup analysis in order to compute a separate OR for its completed
and attempted suiciders subsamples. Data from Courtet et al. 2001 and
2003 (Courtet et al. 2001, 2003) were merged to avoid duplication of
control groups.
3.2. Sensitivity analyses (allele-wise method)
3.2.1. Studies with the triallelic 5-HTTLPR polymorphism
In this analysis, we included only those studies in which the rs25531
polymorphism was used to distinguish the three alleles (S, LA and LG) at
the 5-HTTLPR locus. Data derived from a total of 9 studies (4744 sub-
jects in all, 2 studies and 1851 subjects for the SC subgroup analysis; 7
studies and 2893 subjects for the SA subgroup analysis). No significant
effect was observed when we pooled S + LG alleles versus LA allele for
SA and SC subgroups (OR = 0.86, C.I. 0.73–1.01, p = .06).
Heterogeneity tests revealed a moderate between-studies variability
(Fig. S4).
3.2.2. Asian population
A total of 7 studies, which only encompass subjects who attempted
suicide, were included (3329 subjects in all); no studies considering SC
have been found. No significant association between S + LG alleles and
SB was obtained (OR = 0.97, C.I. 0.84–1.12, p = .66). No heterogeneity
was observed, indicating that all the variability in effect size estimates
was due to sampling error within studies (Fig. S5).
3.2.3. Caucasian population
A total of 27 studies were included (10,127 subjects in all, 10 stu-
dies and 3898 subjects for the SC subgroup analysis; 19 studies and
6410 subjects for the SA subgroup analysis). No association was found
between S + LG alleles and SB (OR = 0.98, C.I. 0.86–1.1, p = .7). The
specific statistic tests assessed a moderate to high heterogeneity. For
Akar et al., 2010 (Akar et al., 2010) the same control group was in-
cluded two times, respectively for the Completed and the Attempted
subgroup analysis (Fig. S6).
3.2.4. Violent suicide phenotype
In order to test the association between the S + LG alleles and the
Violent Suicide phenotype, two strategies were used. Firstly, we com-
pared people with Violent Suicide phenotype versus non-suicidal pa-
tients and healthy controls; then, we compared Violent Suicide subjects
with non-violent suicide subjects. A total of 11 studies were included in
378
Progress in Neuropsychopharmacology & Biological Psychiatry 88 (2019) 375–387
the first analysis (2739 subjects in all, 5 studies and 1259 subjects for
the SC subgroup analysis; 6 studies and 1480 subjects for the SA sub-
group analysis). Conversely, 10 studies were included in the second
analysis (1208 subjects in all, 3 studies and 467 subjects for the SC
subgroup analysis; 7 studies and 741 subjects for the SA subgroup
analysis). A significant effect of the S + LG alleles was observed on
Violent Suicide phenotype both when subjects with a history of Violent
Suicide were compared with non-suicidal patients and healthy controls
(OR = 1.27, C.I. 1.01–1.58, p = .04 - Fig. 3) and when the same sub-
jects were compared with subjects who had a non-violent suicide be-
havior (OR = 1.35, C.I. 1.12–1.62, p = .001 - Fig. S7). Measures of
heterogeneity revealed a moderate heterogeneity for the first compar-
ison (I2 = 61%) and no between-studies variability for the second one
(I2 = 0%). In particular, for the first comparison, considering only the
SA subgroup, we obtained an OR = 1.44 (C.I. 1.17–1.78) at p = .0007
with no observed heterogeneity (Fig. 3). With regard to the most sig-
nificant results we obtained, we further investigated the association by
analysing data from a genotype-wise perspective. Thus, adding two
studies that could not be included in the allele-wise analysis (Campi-
Azevedo et al., 2003; Segal et al., 2006) and without Linkowska et al.
(2010), which lacks genotypic frequencies, we have seen that the re-
sults were also significant (for the first comparison: OR = 1.99, C.I.
1.23–3.20, p = .005, I2 = 77% - Fig. S8; for the second one: OR = 2.46,
C.I. 1.46–4.15, p = .0007, I2 = 66% - Fig. S9).
3.2.5. Male and female gender
When male subjects were considered alone, 11 studies were col-
lected and included (2499 subjects in all, 2 studies and 458 subjects for
the SC subgroup analysis; 9 studies and 2041 subjects for the SA sub-
group analysis). By contrast, a total of 6 studies were included when
only female sex was considered (959 subjects in all, 1 study and 288
subjects for SC subgroup analysis; 5 studies and 671 subjects for the SA
subgroup analysis). No significant associations were found between the
5-HTTLPR alleles and SB neither for the male sex (OR = 0.99, C.I.
0.87–1.13, p = .88), nor for the female sex (OR = 0.94, C.I. 0.66–1.35,
p = .74). Heterogeneity was moderate for both the analyses (Fig. S10
and S11). Of note, although Roy et al., 2007 (Roy et al., 2007) was
partially overlapped with the sample of Enoch et al., 2013 (Enoch et al.,
2013), it was included in the sensitivity analysis for the male gender
because the latter lacked gender-specific genetic data.
3.2.6. Different diagnostic groups
3.2.6.1. Schizophrenia spectrum. A total of 5 studies were included
(1947 subjects in all, with previous Suicide Attempts, including
healthy controls; 1383 subjects in all, after the exclusion of healthy
controls). No significant effect of the S + LG alleles was found on SB,
 Table 1
Characteristics of the included studies.
Reference
(Akar et al., 2010)
(Baca-Garcia et al., 2004)
(Bah et al., 2008)
(Bayle et al., 2003)
(Bellivier et al., 2000)
(Benedetti et al., 2014)
(Bondy et al., 2000)
(Bozina et al., 2012)
(Buttenschon et al., 2013)
(Campi-Azevedo et al.,
2003)
(Chen et al., 2007)
(Chong et al., 2000)
(Courtet et al., 2001)
(Courtet et al., 2003)
(Dell'osso et al., 2013)
379
(Du et al., 2000)
(Enoch et al., 2013)
(Fitch et al., 2001)
(Gaysina et al., 2006)
(Geijer et al., 2000)
(Gibb et al., 2006)
(Gorodetsky et al., 2016)
(Gorwood et al., 2000)
(Helbecque et al., 2006)
(Hranilovic et al., 2003)
(Hung et al., 2011)
(Lee et al., 2015)
(Limosin et al., 2005)
(Linkowska et al., 2010)
(Lopez De Lara et al., 2006)
(Mann et al., 2000)
(Neves et al., 2010)
(Ohara et al., 1998)
(Pompili et al., 2017)
(Preuss et al., 2001)
(Pungercic et al., 2006)
(Roy et al., 2007)
Main Outcomes (diagnosis) Sample investigated Populations Findings
SA, SC 86 SAs + 96 SCs + 181 HC Caucasian (Turkish) No association
G. Fanelli, A. Serretti
SA/impulsivity-aggressive behavior 216 SAs + 223 HC Caucasian (Spanish) No association with impulsivity-aggressive behavior
(SUD, SCZ spectrum, MDD, PD)
SA/5-HTT availability (PD, MDD, AD) 9 SAs + 9 HC – S allele associated with lower 5-HTT availability in SA
SA, VSA (SCZ spectrum) 70 SAs (18 VSAs) + 112 NHC + 159 HC Caucasian (French); 2 patients with one S carrier genotypes associated with VSA
Caribbean grandparent
SA, VSA (MDD, BD) 99 SAs (26 VSAs) + 138 NHC + 187 HC (233 Caucasian (French) S allele associated with VSA
patients +184 HC genotyped)
SA (BD) 32 SAs + 104 NHC Caucasian (Italian) LL genotypes associated with SA. Early stress predicts the risk of SA in
S carrier genotypes.
SC, VSC 58 SCs (53 VSCs) + 110 HC Caucasian (Southern German) S carrier genotypes associated with SC
SA/SI (SCZ spectrum); SI excluded 161 SAs, 174 SIs, 184 NHC Caucasian (Croatian) LA allele associated with SA
SC (SCZ spectrum, Mood, SUD, AD, 572 SCs + 1049 controls Caucasian (Danish) Medium-expressing genotypes associated with SC in individuals aged
others) 35–49 yo
SA, VSA (SCZ, MDD) 93 SAs + 144 NHC + 75 HC Brazilian (African-Brazilian and Caucasian- S carrier genotypes associated with SA
Brazilian)
SA (SUD, substance induced psychosis or 94 SAs + 294 NHC Chinese No association
depression)
SA (SCZ) 76 SAs + 262 NHC Chinese No association
VSA (Mood, SCZ, SUD) 51 VSAs +139 NHC (including 44 with a Caucasian (West European) S allele and SS genotype associated with VSA
Mood disorder)
NVSA (Mood, SCZ, SUD) 166 NVSA +139 NHC (including 44 with a Caucasian (West European) No association
Mood disorder)
SA, SMB 86 SAs + 22 SM + 39 SAs/SM – SS genotype associated only with a temperamental factor (which
includes impulsive-aggressive and anxiety-related traits)
SC (Mood) 35 SCs + 84 controls Caucasian (Hungarian) LL genotype associated with SC
SA (SUD) 141 SAs + 689 HC African-Americans Higher risk of SA in subjects with high Childhood Trauma and Low-
expressing genotypes
SC (misc) 95 SCs + 120 HC Caucasian (French Canadian) No association
SA (PD, MDD, BD, SUD, SCZ spectrum, 144 SAs + 224 HC Caucasian (Russian/Slavic) LL genotype associated with SA in females, but not in males
PTSD)
SA (MDD, PD, SUD, AD) 165 SAs + 99 HC Caucasian (European) No association
SA (Mood, SCZ spectrum, AD, SUD) 15 SAs + 15 NHC misc S carrier genotypes moderate the link between childhood physical
and sexual abuse and SA
SA, SMB (MDD, AnxD, BD, SUD) 57 SM, 41 SAs, 47 SAs with SM, 557 controls Caucasian High- and Intermediate-expressing genotypes associated with SA and/
(non self-harming) or SMB
SA (SUD with comorbid MDD, GD,PD, 55 SAs + 55 NHC + 61 HC Caucasian (French) S carrier genotypes associated with SA
AnxD)
SC, VSC 62 SCs + 72 controls Caucasian-(92.5%)/African-American No association
SC, VSC 135 SCs + 299 HC Caucasian (Croatian/Southern Slavic) No association
SA (SCZ) 60 SAs + 108 NHC + 302 HC Chinese LA carrier genotypes associated with SA in males
SA (MDD) 132 SAs + 122 HC Korean No association
SA (SUD) 45 SAs + 55 NHC Caucasian (French) S allele associated with SA in males
VSC 66 VSCs +51 controls Caucasian (Polish) No association
SC (Mood) 106 SCs + 152 NHC 98–99% Caucasian (French Canadian) No association
SC/5-HTT binding (Mood, SCZ spectrum, 82 SCs + 138 NHC misc No association
SUD)
SA, VSA (BD) 86 SAs (16.67% VSAs) + 112 NHC Caucasian-Brazilian S carrier genotypes associated with VSA
SA (Mood) 13 SAs + 67 NHC + 92 HC Japanese No association
SA (PD, MDD, BD, SUD, SCZ spectrum) 54 SAs + 57 NHC Caucasian (Italian) No association after Bonferroni correction
SA (SUD, MDD) 52 SAs + 111 NHC + 117 HC Caucasian (Southern German) S carrier genotypes associated with SA
SC, VSC 235 SCs + 233 HC Caucasian (Slovenian) No association
SA (SUD) 81 SAs + 176 NHC + 132 controls African-American Low-expressing genotypes associated with SA in the presence of
higher Childhood Trauma scores
(continued on next page)
Progress in Neuropsychopharmacology & Biological Psychiatry 88 (2019) 375–387
G. Fanelli, A. Serretti Progress in Neuropsychopharmacology & Biological Psychiatry 88 (2019) 375–387

Abbreviations: SA, Attempted Suicide; SAs, Suicide Attempters; SC, Completed Suicide; SCs, Suicide Completers; SI, Suicidal Ideation; SIs, subjects with SI; SMB, Self-Mutilation Behavior; SM, subjects with SMB; VSA/

Disorders; AnxD, Anxiety Disorders; ED, Eating disorders; GD, Gambling Disorders; GMC, mental disorders due to a General Medical Condition; Mood, Mood disorders; OCD, Obsessive Compulsive Disorders; PD,
VSAs/VSC/VSCs, − with Violent methods; 5-HTT, Serotonin Transporter; BD, Bipolar Disorder; MDD, Major Depressive Disorder; SCZ, Schizophrenia; HC, Healthy Controls; NHC, Non-Healthy Controls; AD, Adjustment
both by including healthy subjects in the control groups and after their

No association (a significant difference in 5-HTTLPR found between

among older SA, the S allele was associated with high lethality in a
exclusion (respectively, OR = 0.92, C.I. 0.78–1.07, p = .27 and
OR = 0.91, C.I. 0.77–1.08, p = .27). The heterogeneity was null or
approaching to zero for the first and the second comparison,
respectively (Figs. S12 and Fig. S13).

3.2.6.2. Mood disorders. At first, a total of 15 studies were included

(3340 subjects in all, 2 studies and 370 subjects for SC subgroup
analysis; 13 studies and 2970 subjects for the SA subgroup analysis).
After the exclusion of healthy subjects from the control group of each
SS genotype associated with VSC

study, it was possible to combine data derived from 10 of the initial 15

LL genotype associated with SA

studies (1615 subjects in all), because 5 studies lacked non-healthy

single SA and recurrent SA)

control subjects. No association was found between the 5-HTTLPR and

SB in both the analyses (respectively, OR = 0.99, C.I. 0.81–1.22,
p = .93 and OR = 0.95, C.I. 0.70–1.30, p = .75). The heterogeneity
dose–effect way
No association

No association

No association
No association

tests showed a moderate between-studies variability (Figs. S14 and Fig.

S15).
Findings

3.2.6.3. Substance abuse disorders. At first, a total of 6 studies were

included (1465 subjects in all, with previous Suicide Attempts). After
the exclusion of healthy subjects from the control groups, it was
possible to combine data derived from 5 of the initial 6 studies (1004
subjects in all). An association was found between the low-expressing
alleles (S + LG) and SB only when suicidal patients were compared
versus non-suicidal patients and healthy subjects (OR = 1.24, C.I.
Brazilian (heterogeneous)

Misc (94.6% Caucasian)

1.02–1.52, p = .03). Conversely, no association was found when

Caucasian (German)

Caucasian (Spanish)

healthy subjects were excluded from the control groups (OR = 1.24,
Caucasian (Finnish)

Caucasian (Danish)

C.I. 0.96–1.60, p = .10). A low heterogeneity between studies was

observed (Fig. S16 and Fig. S17). Of note, although Roy et al., 2007
Populations

(Roy et al., 2007) was partially overlapped with the sample of Enoch
Japanese
Chinese

et al., 2013 (Enoch et al., 2013), it was included in these sensitivity

analyses because the latter study lacked specific genetic data for
substance abusers.

4. Discussion
32 VSAs +118 NVSAs +200 NHC
349 SCs (137 VSCs) + 284 NHC

The present study aimed to clarify the previous discrepant results

227 SAs + 686 NHC + 420 HC

272 SAs + 275 NHC + 628 HC

28 SAs + 188 NHC + 213 HC

concerning the influence of the 5-HTTLPR polymorphism (SLC6A4

gene) on SB, by updating the previous meta-analytic studies with the
latest published data. In our study, we have included 45 studies re-
158 SAs + 264 NHC
Sample investigated

124 SAs + 185 HC

84 SAs + 152 HC

levant to our research topic. No association was found between the low-
expressing alleles and genotypes (S + LG alleles and S′ carrier geno-
Personality Disorders; SUD, Substance Abuse Disorders; misc., miscellaneous.

types) and SB in the general primary analyses, which include all the
studies we previously selected, without taking into account any source
of heterogeneity. The sensitivity analyses did not show any significant
effect of the 5-HTTLPR polymorphism either when we considered only
SA, VSA (MDD, BD, PD, AD, AnxD, OCD,

SA (Mood with comorbid SUD 30.2% and

SA (Mood, SCZ spectrum, AnxD, AD, PD,

studies which used a triallelic genotyping approach or homogeneous

SA (SCZ spectrum, Mood, GMC, SUD,
SA, VSA (Mood, with comorbid SUD

data for ethnicity, gender, schizophrenia spectrum and mood disorder

diagnoses. An effect of the low-expressing alleles (S + LG) on SB was
found in a subpopulation of substance abusers, although this result was
Main Outcomes (diagnosis)

not confirmed after the exclusion of healthy subjects from the control
SA, VSA (SCZ spectrum)

group. However, when we considered the Violent Suicide phenotype,

the low-expressing alleles (S + LG) were associated with an increase in
the risk of SB, mainly in the SA subgroup.
SUD, others)

AnxD 50%)

SA (MDD)

Overall, we may hypothesise that the possible effect of the 5-

ED, SUD)
SC, VSC

others)

HTTLPR polymorphism on SB may be not detected after the inclusion of

31%)

an increasing number of subjects and studies in both the allele-wise and

the genotype-wise primary analyses. It might be due to the presence of
a strong variation in the effect sizes and a substantial between-studies
(Rahikainen et al., 2017)

(Watanabe et al., 2015)

(Shinozaki et al., 2013)

variability, which is probably derived from the occurrence of con-

(Rujescu et al., 2001)
Table 1 (continued)

(Wang et al., 2009)

(Segal et al., 2006)

(Shen et al., 2004)

siderable sociodemographic and clinical differences (such as the pre-

(Saiz et al., 2011)

sence of different psychiatric diseases and comorbidities, the severity of

symptoms, the ongoing treatments) along with the presence of different
Reference

allelic frequencies between studies. A similar phenomenon was recently

reported in Culverhouse et al., 2018 (Culverhouse et al., 2018), who
applied de novo statistical analyses to 31 datasets containing more than

380
G. Fanelli, A. Serretti Progress in Neuropsychopharmacology & Biological Psychiatry 88 (2019) 375–387

Fig. 2. Forest plot (allele-wise analysis) for the effect of the 5-HTTLPR polymorphism on Suicidal Behavior, pooling the S + LG alleles versus the L + LA alleles.

38 thousand depressed subjects; in that meta-analysis we did not con-
firm the previously reported interaction between stress and the 5-
HTTLPR genotype in increasing the risk of major depression. Thus, in-
dependent from the outcome considered, the effect of 5-HTTLPR poly-
morphism may be more difficult to be observed as the number of the
included subjects and independent studies increases; this phenomenon
was observed in the results of previous meta-analyses, both on SB and
MDD in a gene-environment interaction model.
We also attempted to evaluate the effect of the 5-HTTLPR/rs25531
polymorphism on SB by considering only studies which used a 5-
HTTLPR triallelic genotyping approach. To our knowledge, our work
was the first one to compute a similar sensitivity analysis. The specific

381
G. Fanelli, A. Serretti
Fig. 3. Forest plot (allele-wise analysis) for the effect of the 5-HTTLPR polymorphism on Violent Suicidal Behavior, pooling the S + LG alleles versus the L + LA
alleles. Subjects with a history of Violent Suicide are compared with non-suicidal patients and healthy controls.
purpose was to reduce the bias introduced by studies which use a
biallelic genotyping approach and do not distinguish the LA from the LG
allele, known for having opposite effects on gene expression.
Nevertheless, this sensitivity analysis did not reach the statistical sig-
nificance. Of note, it included just 9/45 studies and it did not contribute
to reducing the heterogeneity between studies compared to our primary
analyses.
In relation to gender, we know that SCs are more frequently males
while SAs are predominantly females (Fushimi et al., 2006; Liu et al.,
2016; Parra Uribe et al., 2013). Besides that, there are other substantial
gender differences in SB and different risk factors between the two
gender groups (Hawton, 2000; Qin et al., 2000). In addition, we pre-
viously (Gressier et al., 2016) observed that the 5-HTTLPR poly-
morphism could modulate different gender-specific features. However,
in our gender-related sensitivity analyses, we did not find any 5-
HTTLPR effect on SB that was able to explain the aforementioned
gender differences. Conversely, using data from 6 datasets, Li and He,
2007 (Li and He, 2007) found a weak association of the L allele with an
increased risk of SB in males (OR = 1.35, C.I. 1–1.82, p = .49) and a
more consistent association in the female sex (OR = 0.68, C.I. 0.51–0.9,
p = .007) but in the risk direction of the S allele. It should be noted that,
in our sensitivity analyses, we use 11 datasets for the male sex and 6 for
the female sex.
Furthermore, we sought to address the issue of the heterogeneity in
the primary analyses assuming the coexistence of different ethnic
groups among the included subjects as one of the causes. Indeed, it is
well known that the allele frequencies between Caucasians and Asians
differ since the S allele is present in 42% and 79% of the subjects, re-
spectively (in our sample, 46% and 75%, respectively). Nevertheless,
we did not find any significant effect of the 5-HTTLPR polymorphism on
SB in both the Caucasian and Asian ethnic groups. This result is con-
sistent with Lin and Tsai, 2004 (Lin and Tsai, 2004), who did not find
any effect of the 5-HTTLPR when they considered only subjects of
382
Progress in Neuropsychopharmacology & Biological Psychiatry 88 (2019) 375–387
Caucasian ethnicity, although the authors did not perform any analysis
for the Asiatic ancestry. Conversely, Li and He, 2007 (Li and He, 2007)
obtained a substantial effect of the S allele on SB by considering a
European population, but no effect was pointed out when subjects of
Asiatic ethnicity were analysed; noteworthy, they did not take into
account potential overlapping samples between individual studies. The
substantial heterogeneity in our sensitivity analysis for Caucasians
could be explained considering that differences in the 5-HTTLPR allele
frequencies may also exist among Caucasian healthy subjects from
different European populations, perhaps due to a moderate Asian in-
fluence existing in some regions (Noskova et al., 2008).
Moreover, other sensitivity analyses which included subjects
homogeneous for three different diagnostic groups, such as mood dis-
orders, schizophrenia spectrum disorders and substance abuse dis-
orders, have not helped us to unveil the effect of the 5-HTTLPR on SB. In
particular, we obtained only a significant effect of the low-expressing
alleles (S + LG) on SB when we compared subjects with SB and a co-
existent substance abuse disorder versus healthy controls combined
with non-suicidal patients belonging to the same diagnostic category.
This result may be due to an interference of the underlying psychiatric
morbidity on SB, which may become more apparent when the com-
parison group also contains healthy controls. Indeed, the effect dis-
appeared when we excluded healthy subjects from the same compar-
ison. Wu et al. (2004) found that SAs were strongly associated with
alcohol abuse and addiction, even after controlling for depression. The
acute effects of intoxication may also represent a proximal risk factor
for SB (Hufford, 2001). Furthermore, the presence of a substance abuse
disorder was the only psychiatric diagnostic predictor of a greater
lethality of SAs in adolescents (McManama O'Brien and Berzin, 2012).
Additionally, we know that alterations in the serotonergic system play
an essential role in the development and maintenance of addiction
(Kirby et al., 2011), similar to what happens for SB. Of note, there is
evidence that the 5-HTTLPR low-expressing variants may be associated
G. Fanelli, A. Serretti
with an increased risk for substance abuse disorders (Serretti et al.,
2006), especially in those with higher impulsivity, aggression and
violent behaviors (Gerra et al., 2004). Consistent to our findings, by
comparing only patients belonging to the same diagnostic category with
the exclusion of healthy subjects, Lin and Tsai, 2004 (Lin and Tsai,
2004) found that the association of the S allele with SB was not sig-
nificant in patients with schizophrenia spectrum or mood disorders,
but, in contrast to our result, the association was significant in patients
with alcoholism. Worthy of note, in our sensitivity analysis, which in-
cludes patients affected by substance abuse disorders and excludes
healthy controls, we pooled a total of five studies encompassing not
only alcohol-dependent subjects; in fact, compared to Lin and Tsai's
sensitivity meta-analysis (Lin and Tsai, 2004), we added the studies of
Chen et al., 2007, Limosin et al., 2005 and Roy et al., 2007 (Chen et al.,
2007; Limosin et al., 2005; Roy et al., 2007). Furthermore, Lin and Tsai
(2004) combined all the eight studies previously used in the distinct
analyses for the separate diagnostic categories; as a result, they pointed
out a trend of significance for the association of the 5-HTTLPR and SB.
Using the same method, Li and He (2007) achieved a more significant
effect when SAs were compared with non-suicidal patients than with
healthy controls, supporting the hypothesis that genetic liability for SB
was independent of the underlying psychiatric morbidity.
The most interesting finding of our present study is related to
Violent Suicide (VS) behavior. We have updated the meta-analyses of
Lin and Tsai, 2004 (Lin and Tsai, 2004) and Li and He, 2007 (Li and He,
2007) concerning the phenotype of VS, adding studies both among the
SA and the SC subgroups. Lin and Tsai, 2004 (Lin and Tsai, 2004) found
a significant association both by comparing Violent Suicide subjects
(VSs) with healthy controls and with non-VSs. Li and He, 2007 (Li and
He, 2007) did not observe an effect when they compared VSs against
non-VSs, but they obtained a weaker significance for the effect of the S
allele when they compared VSs against healthy controls and a stronger
one comparing the same VSs against non-suicidal patients. At first, in
our sensitivity analyses for VS, we compared VSs against merged non-
suicidal patients and healthy controls. Noteworthy, we obtain a more
evident significance for the effect of the S allele in the subgroup of the
SAs compared to that of the SCs. Overall, the effect for both pooled
groups barely reaches the significance level, due to the detrimental
contribution of the SC subgroup. Notable is the fact that we have in-
cluded, comparable with previous meta-analyses, the study of
Hranilovic et al. (2003) in which 98% of the cases are referable to VSs;
the inclusion of this study might cause a negligible approximation bias
on the allele frequencies of VSs for the unification of a slight share of
suiciders with non-violent or unknown characteristics. However, our
results were strengthened by those obtained through the allele-wise
comparison between VSs versus non-VSs as well as the genotype-wise
sensitivity analyses for the VS phenotype. We should note that some
clinical, personality and sociodemographic features may allow distin-
guishing SAs from completers. Indeed, SCs seem less mentally disturbed
than attempters, as they show lower mean scores at MMPI (Minnesota
Multiphasic Personality Inventory) and their profile correlates more
with the profile of non-suicidal controls than with the SAs' one (Daigle,
2004). Furthermore, compared to SAs, suicide completers are less likely
to have been treated by Mental Health Services, although they are more
prone to alcohol consumption and narcissistic personality disorders,
and have more often access to Primary Care, as they more frequently
suffer from physical diseases. From a sociodemographic point of view,
SCs tend to be older, retirees or people who live alone as well as having
occupational, financial or conjugal problems (DeJong et al., 2010;
Giner et al., 2013; Parra Uribe et al., 2013). These clinical, personality
and peculiar sociodemographic aspects suggest that the effect of the
low-expressing alleles (S + LG) is more evident where the psycho-
pathological dimension, underlying the SB, is more substantial and
robust, as it happens for SAs. On the other hand, the effect of the low-
expressing alleles (S + LG) seems to be irrelevant in SCs, who suffer
most from negative socio-economic aspects or physical health
383
Progress in Neuropsychopharmacology & Biological Psychiatry 88 (2019) 375–387
problems, which might play a predominant causal role in the SC sub-
group.
Personality traits can also be useful in differentiating subgroups of
suicidal subjects. In fact, SAs are characterised by higher scores in
neuroticism and lower in conscientiousness than SCs, which show a
healthier personality profile and they are therefore more deliberate and
decisive in their actions (Tsoh et al., 2005; Useda et al., 2007). Over the
years, there have been inconsistent results about the association of the S
allele with neuroticism (Munafo et al., 2006; Munafo et al., 2009;
Schinka et al., 2004; Sen et al., 2004; Willis-Owen et al., 2005), which
was predominant in SA group. An intervention of the environment was
shown as a possible mediator of the effect between the 5-HTTLPR
genotype and neuroticism (Pluess et al., 2010), although meta-analy-
tical evidence on depression, in which similarly neuroticism was im-
puted as an endophenotype (Goldstein and Klein, 2014), does not
support a 5-HTTLPR interaction with the environment in explaining an
increased risk of disease (Culverhouse et al., 2018). However, the effect
of 5-HTTLPR on neuroticism has been shown to be no longer evident
after adolescence (Harro et al., 2009), probably due to the subsequent
intervention of coping strategies on a maladaptive temperament, stra-
tegies which might fail in the proximity of the suicidal act. Further-
more, VS methods have been consistently associated with higher levels
of lifetime aggression and higher levels of impulsivity (Dumais et al.,
2005; Mann et al., 1999), although our group has partly limited the
impact of impulsive and aggressive personality traits on the choice of
the suicide method. In particular, we previously found that lower levels
of state anger and higher trait anger (angry temperament) were able to
better predict the use of violent means in SA (Giegling et al., 2009).
Besides, we pointed out that higher levels of anger expression were
associated with a more impulsive SA (Giegling et al., 2009).
Noteworthy, 5-hydroxyindolacetic acid (5-HIAA) levels in cere-
brospinal fluid (CSF) have been traditionally interpreted as markers of
the serotoninergic activity and serotonin turnover in the brain. Low 5-
HIAA levels in CSF have been related to more lethal and violent suicide
attempts in humans (Agren, 1980; Asberg et al., 1976; Banki and Arato,
1983), higher lifetime aggression (Brown et al., 1982; Brown et al.,
1979; Placidi et al., 2001) as well as a greater propensity towards im-
pulsivity (Linnoila et al., 1983). Furthermore, studying a population of
pyromaniacs and violent offenders, Virkkunen and colleagues suggested
that low concentrations of 5-HIAA may be associated with violent be-
havior through an altered control of impulsivity rather than with ag-
gression and violence as such (Virkkunen et al., 1987). Therefore, any
self-directed or hetero-direct violent behavior might be considered as a
resultant of a more complex interaction between different personality
traits, high impulsivity and a high propensity for aggressive behavior.
Also, at the peripheral level, in adolescents, serotonin plasma levels
were inversely correlated with the severity of suicidality and violence
as measured at the SPI (Suicidal Potential Inventory) scale (Tyano et al.,
2006). Besides that, experimental manipulation of peripheral serotonin
precursors and central serotoninergic function, through acute trypto-
phan depletion (ATD), resulted in higher levels of aggression in hu-
mans, while an opposite supplementation of plasmatic tryptophan re-
sulted in blunted aggressive responses (Bjork et al., 1999; Marsh et al.,
2002).
Unlike neuroticism, the onset of impulsive-aggressive traits seems to
be mediated by an environmental intervention. In fact, it is known that
Rhesus Monkeys moved away at birth from their mothers and grown up
into a peer-group have developed higher rates of impulsive-aggressive
traits as a function of 5-HTTLPR status. Therefore, Rhesus Monkeys LS
genotype carriers, compared to those with LL genotype, showed a sig-
nificant impairment of the metabolism of the serotonin and an extreme
aggressiveness, but only if peer-reared (Barr et al., 2003; Higley et al.,
1996). Epigenetic modifications, such as DNA methylation, histone
modifications and the effect of non-coding RNA, may be the main de-
terminants of the long-term effects of early-life adversities on SB and
the endophenotypes associated with it (Labonte and Turecki, 2012;
G. Fanelli, A. Serretti
Lutz et al., 2017; Palma-Gudiel and Fananas, 2017). For this reason,
personality traits may be hypothesised as mediators in the diathesis
towards SB as they can mediate the adaptation to adverse life events
and the development of psychopathology (Dean et al., 1996). More-
over, they are partially under genetic control, heritable, state in-
dependent, and co-segregated with SB in families and therefore they
may be used to dissect SB in intermediate phenotypes more proximal to
the pathogenic bases of SB (Baud, 2005; Gottesman and Gould, 2003;
Turecki, 2005).
With the aim of reaching the explained variance suggested by
formal genetic studies, Genome-wide Association studies have also been
performed to identify more genetic determinants of SB (Bani-Fatemi
et al., 2016; Mirkovic et al., 2016; Stein et al., 2017). A pattern of in-
consistent results has derived, with the majority of studies that did not
find any association and a minority which have found few associated
loci which reached the genome-wide significance threshold, although
with an impossibility to reproduce the findings in a replication sample.
Some authors used a Polygenic Risk Score (PRS) approach to combine a
series of small effects of single common variants derived from GWAS
and in order to increase the power to detect the effect of multiple genes
in smaller samples (Laursen et al., 2017; Mullins et al., 2014;
Sokolowski et al., 2016; Stein et al., 2017). Such a cumulative score of
genetic risk is more accurately able to predict the risk of developing a
specific phenotype and to explain a greater component of heritability
(International Schizophrenia et al., 2009; Lee et al., 2012; Mullins et al.,
2014). Using this approach, Sokolowski et al. (2016) found that the
polygenic risk in SAs was mainly driven by SNPs of neurodevelop-
mental genes. Furthermore, they demonstrated an association of a PRS
for schizophrenia in SAs, although a partially discordant result was
obtained by Laursen et al. (2017). Similarly, Mullins et al. (2014) and
Stein et al. (2017) reported that a PRS for major depressive disorder and
bipolar disorder was capable of predicting the SA phenotype, sug-
gesting the presence of pleiotropy between several major psychiatric
disorders and SB. The omnigenic model, theorised by Boyle et al.
(2017), would be able to explain this pleiotropy by claiming that each
variant with a regulatory role in a given tissue would have a small effect
on all the disorders modulated through that tissue where its gene is
expressed. In this way, most of the heritability would be due to many
regulatory genes of a small effect, the so-called “peripheral” genes since
they are not directly related to the aetiology of the disease, which
would influence a few “core genes” of great effect with a direct influ-
ence on the aetiopathogenesis of the disease.
Some limitations may have affected the results of our study. In fact,
it should be noted that the definition of SB and its sub-phenotypes has
been extremely controversial and a source of discussion among scien-
tists. The same definition of VS may vary according to the authors,
including or not drowning, poisoning with corrosive agents, CO poi-
soning together with drug overdoses among non-violent suicides
(Asberg et al., 1976; Denning et al., 2000; Dumais et al., 2005;
Stenbacka and Jokinen, 2015). Furthermore, some studies that could
potentially be included in our meta-analyses did not show allele/gen-
otype frequencies or showed partially incomplete data; in some cases,
genetic data were combined without distinguishing between various
phenotypic subgroups. Last but not least, there is diversity in the gen-
otyping methods, with the majority of studies using a biallelic rather
than a triallelic approach, which includes the differentiation of the L
allele according to the status of the rs25531 polymorphism; this issue
could underestimate the effect of the low-expressing alleles for the
component of the L allele with a G in correspondence of the rs25531
(i.e. the LG allele), which is embedded in the L allele according to the
biallelic approach.
In conclusion, we did not observe any overall effect of the 5-HTTLPR
on SB in the primary analyses. Conversely, we observed an interesting
effect of the S allele of 5-HTTLPR on VS, both when we compared VSs to
the general population, including healthy subjects and non-suicidal
patients, and when we compared VSs to non-VSs. An effect of the same
384
Progress in Neuropsychopharmacology & Biological Psychiatry 88 (2019) 375–387
alleles on SB was found in a subpopulation of substance abusers, but
this result was not confirmed after the exclusion of healthy subjects
from the control group. The other sensitivity meta-analyses did not
show any significant effect.
Many levels of complexity must be therefore taken into account in
the research of the neurobiological bases of SB. Among the main ones,
we emphasize the modulatory effect of the environment in gene ex-
pression through tissue-specific epigenetic mechanisms, the effect of
copy number variations and rare variants of major effect alongside
common variants of small effect. The advancement of biomolecular and
statistical analysis technologies in genetics and a higher propensity to
data sharing among research groups as well as the use of animal models
and endophenotype techniques will help us to better elucidate the
biological bases of complex traits such as SB. Research advances might
allow better prevention of this serious mental health issue through
genetic stratification in addition to more effective and targeted ther-
apeutic interventions in populations at extreme risk.
Role of funding source
This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors.
Contributors
G. Fanelli performed the electronic search of the literature, the
analysis of data and wrote the first draft of the manuscript. A. Serretti
designed the study, performed the electronic search of the literature
and reviewed the first draft of the manuscript. All authors contributed
to the interpretation of the data and the improvement of the final
version of the paper. All authors approved the final manuscript.
Declaration of interest
A. Serretti is or has been consultant/speaker for: Abbott, Abbvie,
Angelini, Astra Zeneca, Clinical Data, Boheringer, Bristol Myers Squibb,
Eli Lilly, GlaxoSmithKline, Innovapharma, Italfarmaco, Janssen,
Lundbeck, Naurex, Pfizer, Polifarma, Sanofi, Servier. G. Fanelli has no
conflicts to declare.
Acknowledgement
The Network on Suicide Research and Prevention of the European
College of Neuropsychopharmacology (ECNP) commissioned this
manuscript and contributed by acting as Reference Group and by pro-
viding comments and critical review to the manuscript. The Chair of the
Network is Professor in psychiatry and suicidology Danuta Wasserman,
Head of the National Centre for Suicide Research and Prevention of
Mental Ill-Health (NASP), Karolinska Institutet, Stockholm, Sweden.
The co-Chair is Dr. Vladimir Carli, Senior Lecturer in Prevention of
Suicide and Mental-Ill Health. Secretary of the network is Professor
Pilar A. Saiz, Department of Psychiatry, School of Medicine, University
of Oviedo, CIBERSAM, Oviedo, Spain. Members of the Network are:
Professor Alan Apter, Feinberg Child Study Centre, Schneider Children's
Medical Centre, Tel Aviv University, Tel Aviv, Israel; Professor Enrique
Baca Garcia, Department of Psychiatry, Fundacion Jimenez Diaz
University Hospital, Autonomous University of Madrid, Madrid, Spain;
Professor Judit Balazs, Eotvos Lorand University, Budapest, Hungary;
Professor Julio Bobes, Department of Psychiatry, School of Medicine,
University of Oviedo, CIBERSAM, Oviedo, Spain; Professor Philippe
Courtet, Head of the Department of Emergency Psychiatry at the
Academic Hospital, Montpellier, France; Professor Lars Hakan Thorell,
University of Linkoping, Linkoping, Sweden; Professor Zoltan Rihmer,
Faculty of Medicine, Department of Clinical and Theoretical Mental
Health and Department of Psychiatry and Psychotherapy, Semmelweis
University, Budapest, Hungary; Professor Dan Rujescu, Molecular and
G. Fanelli, A. Serretti
Clinical Neurobiology, Department of Psychiatry, University of Munich,
Munich, Germany; Professor Marco Sarchiapone, Department of Health
Sciences, University of Molise, Campobasso, Italy; Alessandro Serretti,
Section of Psychiatry, University of Bologna, Bologna, Italy; Associate
Professor Marcus Sokolowski, National Centre for Suicide Research and
Prevention of Mental Ill-Health (NASP), Karolinska Institutet,
Stockholm, Sweden; Professor Gil Zalsman, Department of Child and
Adolescent Psychiatry at Geha Hospital, Tel Aviv University, Tel Aviv,
Israel; Professor Jerzy Wasserman, National Centre for Suicide Research
and Prevention of Mental Ill-Health (NASP), Karolinska Institutet,
Stockholm, Sweden; Professor Avi Weizman, Felsenstein Medical
Research Center (FMRC), Tel Aviv University, Petah Tikva, Israel.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.pnpbp.2018.08.007.
References
Agren, H., 1980. Symptom patterns in unipolar and bipolar depression correlating with
monoamine metabolites in the cerebrospinal fluid: II. Suicide. Psychiatry Res. 3 (2),
225–236.
Akar, T., Sayin, A., Bakkaloglu, Z., Cabuk, D.K., Kucukyildirim, S., Demirel, B.,
Candansayar, S., Ozsoy, E.D., Mergen, H., 2010. Investigation of serotonin trans-
porter gene promoter (5-HTTLPR) and intron 2 (variable number of tandem repeats)
polymorphisms with suicidal behavior in a Turkish population. DNA Cell Biol. 29 (8),
429–434.
Anguelova, M., Benkelfat, C., Turecki, G., 2003. A systematic review of association studies
investigating genes coding for serotonin receptors and the serotonin transporter: II.
Suicidal behavior. Mol. Psychiatry 8 (7), 646–653.
Antypa, N., Serretti, A., Rujescu, D., 2013. Serotonergic genes and suicide: a systematic
review. Eur. Neuropsychopharmacol. 23 (10), 1125–1142.
Arsenault-Lapierre, G., Kim, C., Turecki, G., 2004. Psychiatric diagnoses in 3275 suicides:
a meta-analysis. BMC Psychiatry 4, 37.
Asberg, M., Traskman, L., Thoren, P., 1976. 5-HIAA in the cerebrospinal fluid. A bio-
chemical suicide predictor? Arch. Gen. Psychiatry 33 (10), 1193–1197.
Åsberg, M., Nordström, P., Träskman-Bendz, L.I.L., 1986. An overview. Ann. N. Y. Acad.
Sci. 487 (1), 243–255.
Baca-Garcia, E., Vaquero, C., Diaz-Sastre, C., Garcia-Resa, E., Saiz-Ruiz, J., Fernandez-
Piqueras, J., de Leon, J., 2004. Lack of association between the serotonin transporter
promoter gene polymorphism and impulsivity or aggressive behavior among suicide
attempters and healthy volunteers. Psychiatry Res. 126 (2), 99–106.
Bah, J., Lindstrom, M., Westberg, L., Manneras, L., Ryding, E., Henningsson, S., Melke, J.,
Rosen, I., Traskman-Bendz, L., Eriksson, E., 2008. Serotonin transporter gene poly-
morphisms: effect on serotonin transporter availability in the brain of suicide at-
tempters. Psychiatry Res. 162 (3), 221–229.
Bani-Fatemi, A., Graff, A., Zai, C., Strauss, J., De Luca, V., 2016. GWAS analysis of suicide
attempt in schizophrenia: main genetic effect and interaction with early life trauma.
Neurosci. Lett. 622, 102–106.
Banki, C.M., Arato, M., 1983. Amine metabolites, neuroendocrine findings, and person-
ality dimensions as correlates of suicidal behavior. Psychiatry Res. 10 (4), 253–261.
Barr, C.S., Newman, T.K., Becker, M.L., Parker, C.C., Champoux, M., Lesch, K.P.,
Goldman, D., Suomi, S.J., Higley, J.D., 2003. The utility of the non-human primate;
model for studying gene by environment interactions in behavioral research. Genes
Brain Behav. 2 (6), 336–340.
Baud, P., 2005. Personality traits as intermediary phenotypes in suicidal behavior: genetic
issues. Am. J. Med. Genet. C: Semin. Med. Genet. 133c (1), 34–42.
Bayle, F.J., Leroy, S., Gourion, D., Millet, B., Olie, J.P., Poirier, M.F., Krebs, M.O., 2003.
5HTTLPR polymorphism in schizophrenic patients: further support for association
with violent suicide attempts. Am. J. Med. Genet. B Neuropsychiatr. Genet. 119B (1),
13–17.
Bellivier, F., Szoke, A., Henry, C., Lacoste, J., Bottos, C., Nosten-Bertrand, M., Hardy, P.,
Rouillon, F., Launay, J.M., Laplanche, J.L., Leboyer, M., 2000. Possible association
between serotonin transporter gene polymorphism and violent suicidal behavior in
mood disorders. Biol. Psychiatry 48 (4), 319–322.
Benedetti, F., Riccaboni, R., Poletti, S., Radaelli, D., Locatelli, C., Lorenzi, C., Pirovano, A.,
Smeraldi, E., Colombo, C., 2014. The serotonin transporter genotype modulates the
relationship between early stress and adult suicidality in bipolar disorder. Bipolar
Disord. 16 (8), 857–866.
Bjork, J.M., Dougherty, D.M., Moeller, F.G., Cherek, D.R., Swann, A.C., 1999. The effects
of tryptophan depletion and loading on laboratory aggression in men: time course
and a food-restricted control. Psychopharmacology 142 (1), 24–30.
Bondy, B., Erfurth, A., de Jonge, S., Kruger, M., Meyer, H., 2000. Possible association of
the short allele of the serotonin transporter promoter gene polymorphism (5-
HTTLPR) with violent suicide. Mol. Psychiatry 5 (2), 193–195.
Boyle, E.A., Li, Y.I., Pritchard, J.K., 2017. An expanded view of complex traits: from
polygenic to omnigenic. Cell 169 (7), 1177–1186.
Bozina, N., Jovanovic, N., Podlesek, A., Rojnic Kuzman, M., Kudumija Slijepcevic, M.,
Roguljic, A., Dimitrovic, A., Bozina, T., Lovric, J., Ljubic, H., Medved, V., 2012.
385
Progress in Neuropsychopharmacology & Biological Psychiatry 88 (2019) 375–387
Suicide ideators and attempters with schizophrenia–the role of 5-HTTLPR, rs25531,
and 5-HTT VNTR Intron 2 variants. J. Psychiatr. Res. 46 (6), 767–773.
Brent, D.A., Melhem, N., 2008. Familial transmission of suicidal behavior. Psychiatr. Clin.
North Am. 31 (2), 157–177.
Brent, D.A., Bridge, J., Johnson, B.A., Connolly, J., 1996. Suicidal behavior runs in fa-
milies: a controlled family study of adolescent suicide victims. Arch. Gen. Psychiatry
53 (12), 1145–1152.
Brown, G.L., Goodwin, F.K., Ballenger, J.C., Goyer, P.F., Major, L.F., 1979. Aggression in
humans correlates with cerebrospinal fluid amine metabolites. Psychiatry Res. 1 (2),
131–139.
Brown, G.L., Ebert, M.H., Goyer, P.F., Jimerson, D.C., Klein, W.J., Bunney, W.E.,
Goodwin, F.K., 1982. Aggression, suicide, and serotonin: relationships to CSF amine
metabolites. Am. J. Psychiatry 139 (6), 741–746.
Buttenschon, H.N., Flint, T.J., Foldager, L., Qin, P., Christoffersen, S., Hansen, N.F.,
Kristensen, I.B., Mortensen, P.B., Borglum, A.D., Mors, O., 2013. An association study
of suicide and candidate genes in the serotonergic system. J. Affect. Disord. 148
(2–3), 291–298.
Campi-Azevedo, A.C., Boson, W., De Marco, L., Romano-Silva, M.A., Correa, H., 2003.
Association of the serotonin transporter promoter polymorphism with suicidal be-
havior. Mol. Psychiatry 8 (11), 899–900.
CDC, 2016. WISQARS Leading Causes of Death Reports. pp. 1981–2016.
Chen, C.K., Lin, S.K., Huang, M.C., Su, L.W., Hsiao, C.C., Chiang, Y.L., Ree, S.C., Chiang,
S.C., Liu, H.C., 2007. Analysis of association of clinical correlates and 5-HTTLPR
polymorphism with suicidal behavior among Chinese methamphetamine abusers.
Psychiatry Clin. Neurosci. 61 (5), 479–486.
Chong, S.A., Lee, W.L., Tan, C.H., Tay, A.H., Chan, A.O., Tan, E.C., 2000. Attempted
suicide and polymorphism of the serotonin transporter gene in Chinese patients with
schizophrenia. Psychiatry Res. 97 (2–3), 101–106.
Clayden, R.C., Zaruk, A., Meyre, D., Thabane, L., Samaan, Z., 2012. The association of
attempted suicide with genetic variants in the SLC6A4 and TPH genes depends on the
definition of suicidal behavior: a systematic review and meta-analysis. Transl.
Psychiatry 2, e166.
Courtet, P., Baud, P., Abbar, M., Boulenger, J.P., Castelnau, D., Mouthon, D., Malafosse,
A., Buresi, C., 2001. Association between violent suicidal behavior and the low ac-
tivity allele of the serotonin transporter gene. Mol. Psychiatry 6 (3), 338–341.
Courtet, P., Buresi, C., Abbar, M., Baud, P., Boulenger, J.P., Castelnau, D., Mouthon, D.,
Malafosse, A., 2003. No association between non-violent suicidal behavior and the
serotonin transporter promoter polymorphism. Am. J. Med. Genet. B Neuropsychiatr.
Genet. 116B (1), 72–76.
Culverhouse, R.C., Saccone, N.L., Horton, A.C., Ma, Y., Anstey, K.J., Banaschewski, T.,
Burmeister, M., Cohen-Woods, S., Etain, B., Fisher, H.L., Goldman, N., Guillaume, S.,
Horwood, J., Juhasz, G., Lester, K.J., Mandelli, L., Middeldorp, C.M., Olie, E.,
Villafuerte, S., Air, T.M., Araya, R., Bowes, L., Burns, R., Byrne, E.M., Coffey, C.,
Coventry, W.L., Gawronski, K.A.B., Glei, D., Hatzimanolis, A., Hottenga, J.J.,
Jaussent, I., Jawahar, C., Jennen-Steinmetz, C., Kramer, J.R., Lajnef, M., Little, K., Zu
Schwabedissen, H.M., Nauck, M., Nederhof, E., Petschner, P., Peyrot, W.J., Schwahn,
C., Sinnamon, G., Stacey, D., Tian, Y., Toben, C., Van der Auwera, S., Wainwright, N.,
Wang, J.C., Willemsen, G., Anderson, I.M., Arolt, V., Aslund, C., Bagdy, G., Baune, B.
T., Bellivier, F., Boomsma, D.I., Courtet, P., Dannlowski, U., de Geus, E.J.C., Deakin,
J.F.W., Easteal, S., Eley, T., Fergusson, D.M., Goate, A.M., Gonda, X., Grabe, H.J.,
Holzman, C., Johnson, E.O., Kennedy, M., Laucht, M., Martin, N.G., Munafo, M.R.,
Nilsson, K.W., Oldehinkel, A.J., Olsson, C.A., Ormel, J., Otte, C., Patton, G.C.,
Penninx, B., Ritchie, K., Sarchiapone, M., Scheid, J.M., Serretti, A., Smit, J.H.,
Stefanis, N.C., Surtees, P.G., Volzke, H., Weinstein, M., Whooley, M., Nurnberger, J.I.,
Jr., Breslau, N., Bierut, L.J., 2018. Collaborative meta-analysis finds no evidence of a
strong interaction between stress and 5-HTTLPR genotype contributing to the de-
velopment of depression. Mol. Psychiatry 23(1), 133–142.
Daigle, M., 2004. MMPI inmate profiles: suicide completers, suicide attempters, and non-
suicidal controls. Behav. Sci. Law 22 (6), 833–842.
Daws, L.C., Toney, G.M., 2007. High-speed chronoamperometry to study kinetics and
mechanisms for serotonin clearance In vivo. In: Michael, A.C., Borland, L.M. (Eds.),
Electrochemical Methods for Neuroscience. Boca Raton (FL).
Dean, P.J., Range, L.M., Goggin, W.C., 1996. The escape theory of suicide in college
students: testing a model that includes perfectionism. Suicide Life Threat. Behav. 26
(2), 181–186.
Dejong, T.M., Overholser, J.C., Stockmeier, C.A., 2010. Apples to oranges?: a direct
comparison between suicide attempters and suicide completers. J. Affect. Disord. 124
(1–2), 90–97.
Dell'osso, L., Mandelli, L., Carlini, M., Bouanani, S., Rotondo, A., Conversano, C., Serretti,
A., Marazziti, D., 2013. Temperamental and genetic predictors of suicide attempt and
self-mutilation. Neuropsychobiology 68 (4), 250–257.
Denning, D.G., Conwell, Y., King, D., Cox, C., 2000. Method choice, intent, and gender in
completed suicide. Suicide Life Threat. Behav. 30 (3), 282–288.
Desmyter, S., van Heeringen, C., Audenaert, K., 2011. Structural and functional neuroi-
maging studies of the suicidal brain. Prog. Neuro-Psychopharmacol. Biol. Psychiatry
35 (4), 796–808.
Du, L., Faludi, G., Palkovits, M., Bakish, D., Hrdina, P.D., 2000. Tryptophan hydroxylase
gene 218A/C polymorphism is not associated with depressed suicide. Int. J.
Neuropsychopharmacol. 3 (3), 215–220.
Dumais, A., Lesage, A.D., Lalovic, A., Seguin, M., Tousignant, M., Chawky, N., Turecki, G.,
2005. Is violent method of suicide a behavioral marker of lifetime aggression? Am. J.
Psychiatry 162 (7), 1375–1378.
Enoch, M.A., Hodgkinson, C.A., Gorodetsky, E., Goldman, D., Roy, A., 2013. Independent
effects of 5′ and 3′ functional variants in the serotonin transporter gene on suicidal
behavior in the context of childhood trauma. J. Psychiatr. Res. 47 (7), 900–907.
Fitch, D., Lesage, A., Seguin, M., Trousignant, M., Bankelfat, C., Rouleau, G.A., Turecki,
G. Fanelli, A. Serretti
G., 2001. Suicide and the serotonin transporter gene. Mol. Psychiatry 6 (2), 127–128.
Fushimi, M., Sugawara, J., Saito, S., 2006. Comparison of completed and attempted
suicide in Akita, Japan. Psychiatry Clin. Neurosci. 60 (3), 289–295.
Gaysina, D., Zainullina, A., Gabdulhakov, R., Khusnutdinova, E., 2006. The serotonin
transporter gene: polymorphism and haplotype analysis in Russian suicide attemp-
ters. Neuropsychobiology 54 (1), 70–74.
Geijer, T., Frisch, A., Persson, M.L., Wasserman, D., Rockah, R., Michaelovsky, E., Apter,
A., Jonsson, E.G., Nothen, M.M., Weizman, A., 2000. Search for association between
suicide attempt and serotonergic polymorphisms. Psychiatr. Genet. 10 (1), 19–26.
Gerra, G., Garofano, L., Santoro, G., Bosari, S., Pellegrini, C., Zaimovic, A., Moi, G.,
Bussandri, M., Moi, A., Brambilla, F., Donnini, C., 2004. Association between low-
activity serotonin transporter genotype and heroin dependence: behavioral and
personality correlates. Am. J. Med. Genet. B Neuropsychiatr. Genet. 126B (1), 37–42.
Gibb, B.E., McGeary, J.E., Beevers, C.G., Miller, I.W., 2006. Serotonin transporter (5-
HTTLPR) genotype, childhood abuse, and suicide attempts in adult psychiatric in-
patients. Suicide Life Threat. Behav. 36 (6), 687–693.
Giegling, I., Olgiati, P., Hartmann, A.M., Calati, R., Moller, H.J., Rujescu, D., Serretti, A.,
2009. Personality and attempted suicide. Analysis of anger, aggression and im-
pulsivity. J. Psychiatr. Res. 43 (16), 1262–1271.
Giner, L., Blasco-Fontecilla, H., Mercedes Perez-Rodriguez, M., Garcia-Nieto, R., Giner, J.,
Guija, J.A., Rico, A., Barrero, E., Luna, M.A., de Leon, J., Oquendo, M.A., Baca-Garcia,
E., 2013. Personality disorders and health problems distinguish suicide attempters
from completers in a direct comparison. J. Affect. Disord. 151 (2), 474–483.
Goldstein, B.L., Klein, D.N., 2014. A review of selected candidate endophenotypes for
depression. Clin. Psychol. Rev. 34 (5), 417–427.
Gonda, X., Fountoulakis, K.N., Harro, J., Pompili, M., Akiskal, H.S., Bagdy, G., Rihmer, Z.,
2011. The possible contributory role of the S allele of 5-HTTLPR in the emergence of
suicidality. J. Psychopharmacol. 25 (7), 857–866.
Gorodetsky, E., Carli, V., Sarchiapone, M., Roy, A., Goldman, D., Enoch, M.A., 2016.
Predictors for self-directed aggression in Italian prisoners include externalizing be-
haviors, childhood trauma and the serotonin transporter gene polymorphism 5-
HTTLPR. Genes Brain Behav. 15 (5), 465–473.
Gorwood, P., Batel, P., Ades, J., Hamon, M., Boni, C., 2000. Serotonin transporter gene
polymorphisms, alcoholism, and suicidal behavior. Biol. Psychiatry 48 (4), 259–264.
Gottesman, I.I., Gould, T.D., 2003. The endophenotype concept in psychiatry: etymology
and strategic intentions. Am. J. Psychiatry 160 (4), 636–645.
Gressier, F., Calati, R., Serretti, A., 2016. 5-HTTLPR and gender differences in affective
disorders: a systematic review. J. Affect. Disord. 190, 193–207.
Harro, J., Merenakk, L., Nordquist, N., Konstabel, K., Comasco, E., Oreland, L., 2009.
Personality and the serotonin transporter gene: associations in a longitudinal popu-
lation-based study. Biol. Psychol. 81 (1), 9–13.
Hawton, K., 2000. Sex and suicide. Gender differences in suicidal behaviour. Br. J.
Psychiatry 177, 484–485.
Heils, A., Teufel, A., Petri, S., Stober, G., Riederer, P., Bengel, D., Lesch, K.P., 1996. Allelic
variation of human serotonin transporter gene expression. J. Neurochem. 66 (6),
2621–2624.
Helbecque, N., Sparks, D.L., Hunsaker III, J.C., Amouyel, P., 2006. The serotonin trans-
porter promoter polymorphism and suicide. Neurosci. Lett. 400 (1–2), 13–15.
Higgins, J.P., Thompson, S.G., Deeks, J.J., Altman, D.G., 2003. Measuring inconsistency
in meta-analyses. BMJ 327 (7414), 557–560.
Higley, J.D., Mehlman, P.T., Poland, R.E., Taub, D.M., Vickers, J., Suomi, S.J., Linnoila,
M., 1996. CSF testosterone and 5-HIAA correlate with different types of aggressive
behaviors. Biol. Psychiatry 40 (11), 1067–1082.
Hranilovic, D., Stefulj, J., Furac, I., Kubat, M., Balija, M., Jernej, B., 2003. Serotonin
transporter gene promoter (5-HTTLPR) and intron 2 (VNTR) polymorphisms in
Croatian suicide victims. Biol. Psychiatry 54 (9), 884–889.
Hu, X.Z., Lipsky, R.H., Zhu, G., Akhtar, L.A., Taubman, J., Greenberg, B.D., Xu, K., Arnold,
P.D., Richter, M.A., Kennedy, J.L., Murphy, D.L., Goldman, D., 2006. Serotonin
transporter promoter gain-of-function genotypes are linked to obsessive-compulsive
disorder. Am. J. Hum. Genet. 78 (5), 815–826.
Hufford, M.R., 2001. Alcohol and suicidal behavior. Clin. Psychol. Rev. 21 (5), 797–811.
Hung, C.F., Lung, F.W., Chen, C.H., O'Nions, E., Hung, T.H., Chong, M.Y., Wu, C.K., Wen,
J.K., Lin, P.Y., 2011. Association between suicide attempt and a tri-allelic functional
polymorphism in serotonin transporter gene promoter in Chinese patients with
schizophrenia. Neurosci. Lett. 504 (3), 242–246.
International Schizophrenia, C, Purcell, S.M., Wray, N.R., Stone, J.L., Visscher, P.M.,
O'Donovan, M.C., Sullivan, P.F., Sklar, P., 2009. Common polygenic variation con-
tributes to risk of schizophrenia and bipolar disorder. Nature 460 (7256), 748–752.
Kim, C.D., Seguin, M., Therrien, N., Riopel, G., Chawky, N., Lesage, A.D., Turecki, G.,
2005. Familial aggregation of suicidal behavior: a family study of male suicide
completers from the general population. Am. J. Psychiatry 162 (5), 1017–1019.
Kirby, L.G., Zeeb, F.D., Winstanley, C.A., 2011. Contributions of serotonin in addiction
vulnerability. Neuropharmacology 61 (3), 421–432.
Kreitman, N., Philip, A.E., Greer, S., Bagley, C.R., 1969. Parasuicide. Br. J. Psychiatry 115
(523), 746–747.
Labonte, B., Turecki, G., 2012. Epigenetic effects of childhood adversity in the brain and
suicide risk. In: Dwivedi, Y. (Ed.), The Neurobiological Basis of Suicide. Boca
Raton (FL).
Laursen, T.M., Trabjerg, B.B., Mors, O., Borglum, A.D., Hougaard, D.M., Mattheisen, M.,
Meier, S.M., Byrne, E.M., Mortensen, P.B., Munk-Olsen, T., Agerbo, E., 2017.
Association of the polygenic risk score for schizophrenia with mortality and suicidal
behavior - a Danish population-based study. Schizophr. Res. 184, 122–127.
Lauw, M., How, C.H., Loh, C., 2015. PILL Series. Deliberate self-harm in adolescents.
Singap. Med. J. 56 (6), 306–308 (quiz 309).
Lee, S.H., DeCandia, T.R., Ripke, S., Yang, J., Schizophrenia Psychiatric Genome-Wide
Association Study, C, International Schizophrenia, C, Molecular Genetics of
386
Progress in Neuropsychopharmacology & Biological Psychiatry 88 (2019) 375–387
Schizophrenia, C, Sullivan, P.F., Goddard, M.E., Keller, M.C., Visscher, P.M., Wray,
N.R., 2012. Estimating the proportion of variation in susceptibility to schizophrenia
captured by common SNPs. Nat. Genet. 44 (3), 247–250.
Lee, H.Y., Hong, J.P., Hwang, J.A., Lee, H.J., Yoon, H.K., Lee, B.H., Kim, Y.K., 2015.
Possible Association between Serotonin Transporter Gene Polymorphism and Suicide
Behavior in Major Depressive Disorder. Psychiatry Investig. 12 (1), 136–141.
Lesch, K.P., Bengel, D., Heils, A., Sabol, S.Z., Greenberg, B.D., Petri, S., Benjamin, J.,
Muller, C.R., Hamer, D.H., Murphy, D.L., 1996. Association of anxiety-related traits
with a polymorphism in the serotonin transporter gene regulatory region. Science
274 (5292), 1527–1531.
Li, D., He, L., 2007. Meta-analysis supports association between serotonin transporter (5-
HTT) and suicidal behavior. Mol. Psychiatry 12 (1), 47.
Limosin, F., Loze, J.Y., Boni, C., Hamon, M., Ades, J., Rouillon, F., Gorwood, P., 2005.
Male-specific association between the 5-HTTLPR S allele and suicide attempts in
alcohol-dependent subjects. J. Psychiatr. Res. 39 (2), 179–182.
Lin, P.-Y., Tsai, G., 2004. Association between serotonin transporter gene promoter
polymorphism and suicide: results of a meta-analysis. Biol. Psychiatry 55 (10),
1023–1030.
Linkowska, K., Daca, P., Sykutera, M., Pufal, E., Bloch-Boguslawska, E., Grzybowski, T.,
2010. Search for association between suicide and 5-HTT, MAOA and DAT poly-
morphism in Polish males. Arch. Med. Sadowej Kryminol. 60 (2–3), 112–117.
Linnoila, M., Virkkunen, M., Scheinin, M., Nuutila, A., Rimon, R., Goodwin, F.K., 1983.
Low cerebrospinal fluid 5-hydroxyindoleacetic acid concentration differentiates im-
pulsive from nonimpulsive violent behavior. Life Sci. 33 (26), 2609–2614.
Liu, B.P., Liu, X., Jia, C.X., 2016. Characteristics of suicide completers and attempters in
rural Chinese population. Compr. Psychiatry 70, 134–140.
Lopez De Lara, C., Dumais, A., Rouleau, G., Lesage, A., Dumont, M., Chawky, N., Alda, M.,
Benkelfat, C., Turecki, G., 2006. STin2 variant and family history of suicide as sig-
nificant predictors of suicide completion in major depression. Biol. Psychiatry 59 (2),
114–120.
Lutz, P.E., Mechawar, N., Turecki, G., 2017. Neuropathology of suicide: recent findings
and future directions. Mol. Psychiatry 22 (10), 1395–1412.
Mandelli, L., Serretti, A., 2013. Gene environment interaction studies in depression and
suicidal behavior: An update. Neurosci. Biobehav. Rev. 37 (10), 2375–2397 Pt 1.
Mann, J.J., 2002. A current perspective of suicide and attempted suicide. Ann. Intern.
Med. 136 (4), 302–311.
Mann, J.J., 2003. Neurobiology of suicidal behaviour. Nat. Rev. Neurosci. 4 (10),
819–828.
Mann, J.J., Arango, V., Marzuk, P.M., Theccanat, S., Reis, D.J., 1989. Evidence for the 5-
HT hypothesis of suicide. A review of post-mortem studies. Br. J. Psychiatry Suppl.
(8), 7–14.
Mann, J.J., Waternaux, C., Haas, G.L., Malone, K.M., 1999. Toward a clinical model of
suicidal behavior in psychiatric patients. Am. J. Psychiatry 156 (2), 181–189.
Mann, J.J., Huang, Y.-y., Underwood, M.D., Kassir, S.A., Oppenheim, S., Kelly, T.M.,
Dwork, A.J., Arango, V., 2000. A serotonin transporter gene promoter polymorphism
(5-HTTLPR) and prefrontal cortical binding in major depression and suicide. Arch.
Gen. Psychiatry 57 (8) (729-738 %@ 0003-0990X).
Marsh, D.M., Dougherty, D.M., Moeller, F.G., Swann, A.C., Spiga, R., 2002. Laboratory-
measured aggressive behavior of women: acute tryptophan depletion and augmen-
tation. Neuropsychopharmacology 26 (5), 660–671.
McGirr, A., Alda, M., Séguin, M., Cabot, S., Lesage, A., Turecki, G., 2009. Familial ag-
gregation of suicide explained by cluster B traits: a three-group family study of sui-
cide controlling for major depressive disorder. Am. J. Psychiatr. 166 (10),
1124–1134.
McManama O'Brien, K.H., Berzin, S.C., 2012. Examining the impact of psychiatric diag-
nosis and comorbidity on the medical lethality of adolescent suicide attempts. Suicide
Life Threat. Behav. 42 (4), 437–444.
de Medeiros Alves, V., Bezerra, D.G., de Andrade, T.G., de Melo Neto, V.L., Nardi, A.E.,
2015. Genetic polymorphisms might predict suicide attempts in mental disorder
patients: a systematic review and meta-analysis. CNS Neurol. Disord. Drug Targets 14
(7), 820–827.
Mirkovic, B., Laurent, C., Podlipski, M.A., Frebourg, T., Cohen, D., Gerardin, P., 2016.
Genetic association studies of suicidal behavior: a review of the past 10 years, pro-
gress, limitations, and future directions. Front. Psychiatry 7, 158.
Mittendorfer-Rutz, E., Rasmussen, F., Wasserman, D., 2008. Familial clustering of suicidal
behaviour and psychopathology in young suicide attempters. A register-based nested
case control study. Soc. Psychiatry Psychiatr. Epidemiol. 43 (1), 28–36.
Mullins, N., Perroud, N., Uher, R., Butler, A.W., Cohen-Woods, S., Rivera, M., Malki, K.,
Euesden, J., Power, R.A., Tansey, K.E., Jones, L., Jones, I., Craddock, N., Owen, M.J.,
Korszun, A., Gill, M., Mors, O., Preisig, M., Maier, W., Rietschel, M., Rice, J.P.,
Muller-Myhsok, B., Binder, E.B., Lucae, S., Ising, M., Craig, I.W., Farmer, A.E.,
McGuffin, P., Breen, G., Lewis, C.M., 2014. Genetic relationships between suicide
attempts, suicidal ideation and major psychiatric disorders: a genome-wide associa-
tion and polygenic scoring study. Am. J. Med. Genet. B Neuropsychiatr. Genet. 165B
(5), 428–437.
Munafo, M.R., Clark, T.G., Roberts, K.H., Johnstone, E.C., 2006. Neuroticism mediates the
association of the serotonin transporter gene with lifetime major depression.
Neuropsychobiology 53 (1), 1–8.
Munafo, M.R., Freimer, N.B., Ng, W., Ophoff, R., Veijola, J., Miettunen, J., Jarvelin, M.R.,
Taanila, A., Flint, J., 2009. 5-HTTLPR genotype and anxiety-related personality traits:
a meta-analysis and new data. Am. J. Med. Genet. B Neuropsychiatr. Genet. 150B (2),
271–281.
Murphy, D.L., Lerner, A., Rudnick, G., Lesch, K.P., 2004. Serotonin transporter: gene,
genetic disorders, and pharmacogenetics. Mol. Interv. 4 (2), 109–123.
Nakamura, M., Ueno, S., Sano, A., Tanabe, H., 2000. The human serotonin transporter
gene linked polymorphism (5-HTTLPR) shows ten novel allelic variants. Mol.
G. Fanelli, A. Serretti
Psychiatry 5 (1), 32–38.
Neves, F.S., Malloy-Diniz, L.F., Romano-Silva, M.A., Aguiar, G.C., de Matos, L.O., Correa,
H., 2010. Is the serotonin transporter polymorphism (5-HTTLPR) a potential marker
for suicidal behavior in bipolar disorder patients? J. Affect. Disord. 125 (1–3),
98–102.
Nock, M.K., Borges, G., Bromet, E.J., Alonso, J., Angermeyer, M., Beautrais, A., Bruffaerts,
R., Chiu, W.T., de Girolamo, G., Gluzman, S., de Graaf, R., Gureje, O., Haro, J.M.,
Huang, Y., Karam, E., Kessler, R.C., Lepine, J.P., Levinson, D., Medina-Mora, M.E.,
Ono, Y., Posada-Villa, J., Williams, D., 2008. Cross-national prevalence and risk
factors for suicidal ideation, plans and attempts. Br. J. Psychiatry 192 (2), 98–105.
Noskova, T., Pivac, N., Nedic, G., Kazantseva, A., Gaysina, D., Faskhutdinova, G.,
Gareeva, A., Khalilova, Z., Khusnutdinova, E., Kovacic, D.K., Kovacic, Z., Jokic, M.,
Seler, D.M., 2008. Ethnic differences in the serotonin transporter polymorphism (5-
HTTLPR) in several European populations. Prog. Neuro-Psychopharmacol. Biol.
Psychiatry 32 (7), 1735–1739.
Ohara, K., Nagai, M., Tsukamoto, T., Tani, K., Suzuki, Y., Ohara, K., 1998. Functional
polymorphism in the serotonin transporter promoter at the SLC6A4 locus and mood
disorders. Biol. Psychiatry 44 (7), 550–554.
Palma-Gudiel, H., Fananas, L., 2017. An integrative review of methylation at the ser-
otonin transporter gene and its dialogue with environmental risk factors, psycho-
pathology and 5-HTTLPR. Neurosci. Biobehav. Rev. 72, 190–209.
Parra Uribe, I., Blasco-Fontecilla, H., Garcia-Pares, G., Giro Batalla, M., Llorens Capdevila,
M., Cebria Meca, A., de Leon-Martinez, V., Perez-Sola, V., Palao Vidal, D.J., 2013.
Attempted and completed suicide: not what we expected? J. Affect. Disord. 150 (3),
840–846.
Patten, S.B., Williams, J.V., Lavorato, D.H., Wang, J.L., McDonald, K., Bulloch, A.G.,
2015. Descriptive epidemiology of major depressive disorder in Canada in 2012. Can.
J. Psychiatr. 60 (1), 23–30.
Pattison, E.M., Kahan, J., 1983. The deliberate self-harm syndrome. Am. J. Psychiatry 140
(7), 867–872.
Placidi, G.P., Oquendo, M.A., Malone, K.M., Huang, Y.Y., Ellis, S.P., Mann, J.J., 2001.
Aggressivity, suicide attempts, and depression: relationship to cerebrospinal fluid
monoamine metabolite levels. Biol. Psychiatry 50 (10), 783–791.
Pluess, M., Belsky, J., Way, B.M., Taylor, S.E., 2010. 5-HTTLPR moderates effects of
current life events on neuroticism: differential susceptibility to environmental in-
fluences. Prog. Neuro-Psychopharmacol. Biol. Psychiatry 34 (6), 1070–1074.
Pompili, M., Gentile, G., Scassellati, C., Bonvicini, C., Innamorati, M., Erbuto, D.,
Montebovi, F., Ducci, G., Forte, A., De Pisa, E., Ferracuti, S., Serafini, G., De Luca, V.,
Amore, M., Simmaco, M., Girardi, P., 2017. Genetic association analysis of serotonin
and signal transduction pathways in suicide attempters from an Italian sample of
psychiatric patients. Neurosci. Lett. 656, 94–102.
Pooley, E.C., Houston, K., Hawton, K., Harrison, P.J., 2003. Deliberate self-harm is as-
sociated with allelic variation in the tryptophan hydroxylase gene (TPH A779C), but
not with polymorphisms in five other serotonergic genes. Psychol. Med. 33 (5),
775–783.
Preuss, U.W., Koller, G., Soyka, M., Bondy, B., 2001. Association between suicide at-
tempts and 5-HTTLPR-S-allele in alcohol-dependent and control subjects: further
evidence from a German alcohol-dependent inpatient sample. Biol. Psychiatry 50 (8),
636–639.
Pungercic, G., Videtic, A., Pestotnik, A., Pajnic, I.Z., Zupanc, T., Balazic, J., Tomori, M.,
Komel, R., 2006. Serotonin transporter gene promoter (5-HTTLPR) and intron 2
(VNTR) polymorphisms: a study on Slovenian population of suicide victims.
Psychiatr. Genet. 16 (5), 187–191.
Purselle, D.C., Nemeroff, C.B., 2003. Serotonin transporter: a potential substrate in the
biology of suicide. Neuropsychopharmacology 28 (4), 613–619.
Qin, P., Agerbo, E., Westergard-Nielsen, N., Eriksson, T., Mortensen, P.B., 2000. Gender
differences in risk factors for suicide in Denmark. Br. J. Psychiatry 177, 546–550.
Rahikainen, A.L., Majaharju, S., Haukka, J., Palo, J.U., Sajantila, A., 2017. Serotonergic
5HTTLPR/rs25531 s-allele homozygosity associates with violent suicides in male
citalopram users. Am. J. Med. Genet. B Neuropsychiatr. Genet. 174 (7), 691–700.
Review Manager (RevMan) [Computer program]. 5.3 ed, 2014. The Cochrane
Collaboration. The Nordic Cochrane Centre, Copenhagen.
Roy, A., Hu, X.Z., Janal, M.N., Goldman, D., 2007. Interaction between childhood trauma
and serotonin transporter gene variation in suicide. Neuropsychopharmacology 32
(9), 2046–2052.
Rudnick, G., Clark, J., 1993. From synapse to vesicle: the reuptake and storage of biogenic
amine neurotransmitters. Biochim. Biophys. Acta 1144 (3), 249–263.
Rujescu, D., Giegling, I., Sato, T., Moeller, H.J., 2001. A polymorphism in the promoter of
the serotonin transporter gene is not associated with suicidal behavior. Psychiatr.
Genet. 11 (3), 169–172.
Saiz, P.A., Garcia-Portilla, P., Paredes, B., Corcoran, P., Arango, C., Morales, B.,
Sotomayor, E., Alvarez, V., Coto, E., Florez, G., Bascaran, M.T., Bousono, M., Bobes,
J., 2011. Role of serotonergic-related systems in suicidal behavior: data from a case-
control association study. Prog. Neuro-Psychopharmacol. Biol. Psychiatry 35 (6),
1518–1524.
Schild, A.H., Pietschnig, J., Tran, U.S., Voracek, M., 2013. Genetic association studies
between SNPs and suicidal behavior: a meta-analytical field synopsis. Prog. Neuro-
387
Progress in Neuropsychopharmacology & Biological Psychiatry 88 (2019) 375–387
Psychopharmacol. Biol. Psychiatry 46, 36–42.
Schinka, J.A., Busch, R.M., Robichaux-Keene, N., 2004. A meta-analysis of the association
between the serotonin transporter gene polymorphism (5-HTTLPR) and trait anxiety.
Mol. Psychiatry 9 (2), 197–202.
Segal, J., Pujol, C., Birck, A., Gus Manfro, G., Leistner-Segal, S., 2006. Association be-
tween suicide attempts in south Brazilian depressed patients with the serotonin
transporter polymorphism. Psychiatry Res. 143 (2–3), 289–291.
Sen, S., Burmeister, M., Ghosh, D., 2004. Meta-analysis of the association between a
serotonin transporter promoter polymorphism (5-HTTLPR) and anxiety-related per-
sonality traits. Am. J. Med. Genet. B Neuropsychiatr. Genet. 127B (1), 85–89.
Serretti, A., Calati, R., Mandelli, L., De Ronchi, D., 2006. Serotonin transporter gene
variants and behavior: a comprehensive review. Curr. Drug Targets 7 (12),
1659–1669.
Shen, Y., Li, H., Gu, N., Tan, Z., Tang, J., Fan, J., Li, X., Sun, W., He, L., 2004. Relationship
between suicidal behavior of psychotic inpatients and serotonin transporter gene in
Han Chinese. Neurosci. Lett. 372 (1–2), 94–98.
Shinozaki, G., Romanowicz, M., Passov, V., Rundell, J., Mrazek, D., Kung, S., 2013. State
dependent gene-environment interaction: serotonin transporter gene-child abuse in-
teraction associated with suicide attempt history among depressed psychiatric in-
patients. J. Affect. Disord. 147 (1–3), 373–378.
Skegg, K., 2005. Self-harm. Lancet 366 (9495), 1471–1483.
Sokolowski, M., Wasserman, J., Wasserman, D., 2016. Polygenic associations of neuro-
developmental genes in suicide attempt. Mol. Psychiatry 21 (10), 1381–1390.
Stein, M.B., Ware, E.B., Mitchell, C., Chen, C.Y., Borja, S., Cai, T., Dempsey, C.L.,
Fullerton, C.S., Gelernter, J., Heeringa, S.G., Jain, S., Kessler, R.C., Naifeh, J.A., Nock,
M.K., Ripke, S., Sun, X., Beckham, J.C., Kimbrel, N.A., Va Mid-Atlantic Mental Illness
Research, E, Clinical Center, W, Ursano, R.J., Smoller, J.W., 2017. Genomewide as-
sociation studies of suicide attempts in US soldiers. Am. J. Med. Genet. B
Neuropsychiatr. Genet. 174 (8), 786–797.
Stenbacka, M., Jokinen, J., 2015. Violent and non-violent methods of attempted and
completed suicide in Swedish young men: the role of early risk factors. BMC
Psychiatry 15, 196.
Tidemalm, D., Runeson, B., Waern, M., Frisell, T., Carlström, E., Lichtenstein, P.,
Långström, N., 2011. Familial clustering of suicide risk: a total population study of
11.4 million individuals. Psychol. Med. 41 (12), 2527–2534.
Tsoh, J., Chiu, H.F., Duberstein, P.R., Chan, S.S., Chi, I., Yip, P.S., Conwell, Y., 2005.
Attempted suicide in elderly Chinese persons: a multi-group, controlled study. Am. J.
Geriatr. Psychiatry 13 (7), 562–571.
Turecki, G., 2005. Dissecting the suicide phenotype: the role of impulsive-aggressive
behaviours. J. Psychiatry Neurosci. 30 (6), 398–408.
Turecki, G., 2014. The molecular bases of the suicidal brain. Nat. Rev. Neurosci. 15 (12),
802–816.
Tyano, S., Zalsman, G., Ofek, H., Blum, I., Apter, A., Wolovik, L., Sher, L., Sommerfeld, E.,
Harell, D., Weizman, A., 2006. Plasma serotonin levels and suicidal behavior in
adolescents. Eur. Neuropsychopharmacol. 16 (1), 49–57.
Useda, J.D., Duberstein, P.R., Conner, K.R., Beckman, A., Franus, N., Tu, X., Conwell, Y.,
2007. Personality differences in attempted suicide versus suicide in adults 50 years of
age or older. J. Consult. Clin. Psychol. 75 (1), 126–133.
Vincze, I., Perroud, N., Buresi, C., Baud, P., Bellivier, F., Etain, B., Fournier, C., Karege, F.,
Matthey, M.L., Preisig, M., Leboyer, M., Malafosse, A., 2008. Association between
brain-derived neurotrophic factor gene and a severe form of bipolar disorder, but no
interaction with the serotonin transporter gene. Bipolar Disord. 10 (5), 580–587.
Virkkunen, M., Nuutila, A., Goodwin, F.K., Linnoila, M., 1987. Cerebrospinal fluid
monoamine metabolite levels in male arsonists. Arch. Gen. Psychiatry 44 (3),
241–247.
Voracek, M., Loibl, L.M., 2007. Genetics of suicide: a systematic review of twin studies.
Wien. Klin. Wochenschr. 119 (15–16), 463–475.
Wang, A.G., Rasmussenb, H.B., Sørensena, H.J., Hvida, M., Breddama, C., Hansena, B.,
Billea, V.H., Garsdala, O., Jacobya, A.S., Søebyb, K., 2009. Suicidal behavior and the
serotonin transporter gene polymorphism (5-HTTLPR) with novel subtypes, in Danish
schizophrenic patients. Open Psychiatry J. 3, 9–12.
Watanabe, S.Y., Iga, J., Numata, S., Umehara, H., Nishi, A., Kinoshita, M., Inoshita, M.,
Ohmori, T., 2015. Polymorphism in the promoter of the gene for the serotonin
transporter affects the age of onset of major depressive disorder in the Japanese
population. J. Affect. Disord. 183, 156–158.
Wendland, J.R., Martin, B.J., Kruse, M.R., Lesch, K.P., Murphy, D.L., 2006. Simultaneous
genotyping of four functional loci of human SLC6A4, with a reappraisal of 5-HTTLPR
and rs25531. Mol. Psychiatry 11 (3), 224–226.
WHO, 2014. Preventing Suicide: A Global Imperative.
WHO, 2018. Suicide Fact Sheet. Reviewed January 2018 ed.
Willis-Owen, S.A., Turri, M.G., Munafo, M.R., Surtees, P.G., Wainwright, N.W., Brixey,
R.D., Flint, J., 2005. The serotonin transporter length polymorphism, neuroticism,
and depression: a comprehensive assessment of association. Biol. Psychiatry 58 (6),
451–456.
Wu, P., Hoven, C.W., Liu, X., Cohen, P., Fuller, C.J., Shaffer, D., 2004. Substance use,
suicidal ideation and attempts in children and adolescents. Suicide Life Threat.
Behav. 34 (4), 408–420.