The association between childhood maltreatment and empathic perspective taking is moderated by the 5-HTT linked polymorphic region: Another example of "differential susceptibility".

Previous research has suggested that the short (S)-allele of the 5-HT transporter gene-linked polymorphic region (5-HTTLPR) may confer "differential susceptibility" to environmental impact with regard to the expression of personality traits, depressivity and impulsivity. However, little is known about the role of 5-HTTLPR concerning the association between childhood adversity and empathy. Here, we analyzed samples of 137 healthy participants and 142 individuals diagnosed with borderline personality disorder (BPD) focusing on the 5-HTTLPR genotype (S/L-carrier) and A/G SNP (rs25531), in relation to childhood maltreatment and empathy traits. Whereas no between-group difference in 5-HTTLPR genotype distribution emerged, the S-allele selectively moderated the impact of childhood maltreatment on empathic perspective taking, whereby low scores in childhood trauma were associated with superior perspective taking. In contrast, L-homozygotes seemed to be largely unresponsive to variation in environmental conditions in relation to empathy, suggesting that the S-allele confers "differential susceptibility". Moreover, a moderation analysis and tests for differential susceptibility yielded similar results when transcriptional activity of the serotonin transporter gene was taken into account. In conclusion, our findings suggest that the S-allele of the 5-HTTLPR is responsive to early developmental contingencies for "better and worse", i.e. conferring genetic plasticity, especially with regard to processes involving emotional resonance.

Introduction

Social interaction requires reciprocal understanding of verbal and nonverbal signals, which entails the ability to understand others’ emotional states. The mechanism involved in this process is commonly referred to as “empathy”, which can be conceptualized as a semi-automatic sharing of another’s feelings, combined with the ability to differentiate between own and others’ affect [1-3]. Empathic deficits have been described in several neuropsychiatric conditions including autism [4, 5], schizophrenia [6, 7], psychopathy [8] and personality disorders, with mixed results for borderline personality disorder [9-13]. Our own research revealed that patients with borderline personality disorder (BPD) showed selectively increased empathy for psychological pain compared to somatic pain [14], which was associated with childhood trauma, alexithymia and emotional empathy.

A plethora of studies suggests that the activity of the serotonergic system is critically involved in social behavior [15, 16]. With regard to gene-environment interaction, for instance, research in nonhuman primates demonstrated that length variations located in the promoter of the serotonin transporter gene (5-HTTLPR) affects, together with rearing experiences, the level of serotonin metabolites in the cerebrospinal fluid and primate social behavior [17, 18]. In humans, the short (S) allele of the 5-HTTLPR has been associated with reduced serotonin transporter expression and function; it has also been found to be related to trait-anxiety, depression and impulsivity [19-23]. With regard to the processing of social cues, previous studies reported increased emotional reactivity, especially towards negatively biased stimuli [24-28] as well as heightened physiological stress responses in carriers of the S-allele [27, 29]. However, this polymorphic variation does not generally occur more frequently in clinical samples compared to the general population [30-32]. The general population is heterozygous, whereas the LL-genotype is less common, and the SS-variant relatively rare, in part depending on ethnicity [33, 34]. In addition, controversy exists about the effect of rs25531, a SNP within the 5-HTTLPR repetitive element, with the A-variant of the L-allele being associated with greater transcriptional activity and thus more efficient serotonin turnover [35-37], whereby a linkage disequilibrium between 5-HTTLPR and rs25531 has been described, with the rarer G-variant of rs25531 occurring more frequently together with the L-allele than with the S-allele [36, 38]. In recent years, researchers have become interested in the question, raised from an evolutionary point of view, why genes conferring increased risk to psychological dysfunction may be conserved in the genepool of human populations or have even been positively selected in recent millennia (see, for instance, [39]). As an alternative account to the widely-known “diathesis-stress-model” [40], the groups of Ellis [41] as well as Belsky and colleagues have suggested that genetic variants may not one-sidedly convey risk to the development of psychological dysfunction if associated with adverse life events, but that the very same polymorphic variation may confer lower than average risks if met with superior environmental conditions, foremost empathic parental care and emotional availability of care-givers [42]. Therefore, the “differential susceptibility” or “genetic plasticity” model emphasizes the difference between plasticity and resilience (i.e. unresponsiveness to environmental conditions; [43]).

With regard to the 5-HTTLPR, several studies reported an association of stressful live events and depression in SS-homozygotes or S-carrying heterozygotes ([44]; for meta-analyses, see [45, 46]), while others did not confirm these findings (for meta-analyses, see [31, 47]). Conversely, and in line with the “differential susceptibility” model, Pluess and colleagues reported that SS-carriers had higher scores in neuroticism when exposed to negative life events (within the last six months), whereas more positive life events were related to less than average neuroticism. This association was absent in L-carriers [48]. Similar findings were reported by Kuepper et al. [49] who also found an association of negative life events (over the life span) and neuroticism in S-allele carriers.

However, whether or not traumatic events during childhood specifically influence the development of empathic abilities, moderated by the 5-HTTLPR, is still unclear. Accordingly, the present study aimed to investigate the impact of the 5-HTTLPR on trait empathy in a sample of healthy participants and patients with BPD. We deliberately chose the two samples, because one was characterized by relatively few adverse childhood experiences, while the clinical group was coined by relatively high degrees of early maltreatment. We specifically hypothesized that the S-allele of the 5-HTTLPR would differentially impact on the association of childhood trauma with empathic perspective taking, whereas the L-allele would be unresponsive to environmental variation, with some potential modification according to the transcriptional activity of the serotonin transporter gene conveyed by the rs25531 polymorphism.

Material and methods

Participants

For the current study we recruited female in-patients with BPD, diagnosed according to DSM-5 criteria [50] from the LWL-University Hospital Bochum and female healthy control participants via advertisement. In total, 142 patients with BPD and 137 control participants were included. The age of participants was between 18 and 50 years. All participants were fluent in German, free of somatic illnesses and not pregnant (see Table 1 for comorbid disorders and medication of BPD patients). The control participants were free of medication and psychiatric disorders. Regarding the ethnical background, 95.2% were Caucasians, 4.4% originated from the Middle East (mainly of Turkish origin) and 0.4% from the Far East (Vietnamese). The study was approved by the Ethics Committee of the Medical Faculty of the Ruhr-University Bochum (project number 4639–13). The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. All participants gave their full informed consent in writing.

Table 1

Comorbid disorders and medication of patients with BPD in absolute (n) and relative (in %) amounts.

	n	%
Comorbid disorders of patients with BPD
Depressive episode	74	52.1
Posttraumatic Stress Disorder	21	14.8
Phobic/ anxiety Disorder	7	4.9
Substance misuse	42	29.8
Medication
without regular medication	59	41.5
antidepressant	51	35.9
antipsychotic	22	15.5
antidepressant and antipsychotic drugs	22	15.5
antiepileptic	8	5.6
Other psychoactive drugs	6	4.2

Questionnaires

Premorbid or general intelligence was estimated using the Mehrfachwahl-Wortschatz-Intelligenz-Test (MWT-A; [51]). Empathic abilities were measures using the German version of the Interpersonal Reactivity Index [52], called “Saarbrücker Persönlichkeits-Fragebogen” [53]. This questionnaire comprises four subscales, namely “perspective taking” (PT), “fantasy” (FS), “‘empathic concern” (EC) and “personal distress” (PD), and has proven reliable with a Cronbach’s alpha of 0.78. The present analysis focused on the “perspective-taking” score of the Interpersonal Reactivity Index (IRI), because this score is suggested to reflect cognitive empathy traits. In contrast, emotional empathy is more context dependent [54] and therefore emotional empathy scores are not appropriate for trait analyses (the validity of the other cognitive empathy score of the IRI, “fantasy”, is debated and therefore not included into the present analyses; [55]).

The short German version of the Childhood Trauma Questionnaire (CTQ) was used to assess the experience of maltreatment during childhood. The CTQ contains 28 questions tapping into the history of emotional abuse, physical abuse, sexual abuse, emotional neglect and physical neglect. Participants were asked to rate the occurrence of maltreatment on a 5-point Likert scale (1 = never, 5 = very often) pertaining to their childhood and youth. Cronbach’s alpha values for the German version were high for all subscales (0.80), except for physical neglect [56]. In addition, we used the Beck’s Depression Inventory to assess the self-rated level of depressivity [57].

Genotyping

The DNA samples of participants were collected using Oragene OG-500 collection kits (DNA Genotek, Inc., Ottawa, ON, Canada) and by mouthwash with a commercially available mouthwash solution (Listerine). The DNA extraction was conducted according to the manufacturer's instructions of the Oragene Kit and by an adapted version for the mouthwash samples, using a standard salting-out procedure proposed by Miller et al. [58]. The DNA samples were diluted to a concentration of (20 ng/μL). The 5-HTTLPR and the rs25531 genotypes were determined as described by Wendland et al. [36].

Statistical analyses

We conducted a power analysis for interaction effects, i.e. the differences between slopes for the moderation model with the genotypes SS+SL and LL, using G*Power, Version 3.1.9.2. [59]. Power calculation for the current sample of 205 participants was determined by the following model: t-test-linear bivariate regression, two groups, difference between slopes, with α set at 0.05. Standard deviation of the residuals, the sample size and the difference between the slopes were also considered. Accordingly, the statistical power coefficient was 0.75.

In accordance with previous studies, we divided the sample into S-carriers (SS+SL pooled) and LL-carriers (e.g. [60-62]). This approach was justified, because previous studies reported no differences between SS and SL-carriers with regard to personality traits, suggesting a dominant-recessive type of association of the S-allele with personality (21).

Similarly, following previous research (e.g., [37]), subjects were further divided into groups according to the”transcriptional activity”(TA) of the rs25531.

Independent two-sample t-tests were used for comparisons of questionnaire data between groups. The distributions of genotypes were assessed by chi-square tests, whereas calculations were performed for SS, SL and LL genotypes, and for both groups, i.e. patients and controls. Since we did not find any effect of group for the genotype distribution, further analyses were performed without the factor group. Because of the difference between groups regarding age and IQ, we included these variables as covariates into further analyses. In order to investigate the effect of the genotype, we calculated a multivariate analysis of covariance (MANCOVA) with the covariates IQ and age and the between-subject factor genotype (SS+SL vs. LL) and the independent variables were the IRI scores. The moderation analysis was conducted by means of the SPSS macro tool PROCESS developed by Hayes [63]. The moderation was calculated for the dependent variable “perspective taking” (Y; outcome) and the independent variable CTQ total score (X; predictor) and the moderator (M; susceptibility factor), i.e. the 5-HTTLPR genotype (SS+SL vs. LL), as well as controlled for IQ and age (Table 2).

Table 2

Summary of moderation analyses conducted.

The predictor and outcome variables remained constant across calculations whereas the moderator and control variables were exchanged.

Predictor	Outcome	Moderator	Covariates
CTQtotal score	Perspectivetaking	SS+SL vs. LL	age, IQ
			age, IQ, BDI
		SS vs. SL vs. LL	age, IQ
			age, IQ, BDI
		TA groups: low/low, high/low, high/high	age, IQ
			age, IQ, BDI
		BPD vs. HC	age, IQ

In order to explore whether the data were in accordance with the differential susceptibility model, we further investigated the association of the predictor (CTQ) and the susceptibility factor, the moderator, by correlation analyses (partial correlations corrected for age and IQ between CTQ total score and the genotype). In addition, we tested for an association of the susceptibility factor with the outcome variable by calculating the correlation of genotype with perspective taking (partial correlation). Finally, we calculated the difference between the slopes of the associations of the moderation analysis between the two genotype groups.

The same analyses for moderating effects and differential susceptibility were also carried out according to differences in transcriptional activity. Additional moderation analyses were performed for the SS, SL and LL Genotypes and for group (BPD vs. HC) as the moderator instead of Genotype. In order to examine the impact of depressivity, we calculated additional moderation analyses with the additional covariate “BDI score” for the moderators “SS+SL vs. LL”, “SS, SL and LL” and transcriptional activity (Table 2). The statistical analyses were performed using the software SPSS version 25 (IBM Corp. Released 2017. IBM SPSS Statistics for Windows, Version 25.0. Armonk, NY: IBM Corp.).

Results

Patients differed in age and IQ from control participants and reported more severe experiences of childhood maltreatment as well as more depressive symptoms. Patients reached higher scores in the personal distress score of the IRI, whereas healthy controls scored higher in perspective taking and fantasy scores (Table 3).

Table 3

Psychometric properties of patients with BPD and healthy participants.

Results are reported as mean (M) values and standard deviations (SD). t, p and df of T tests between groups are shown.

	HC		BPD		T Test
	M	SD	M	SD	t	p	df
Age	24.8	5.6	27.6	7.9	3.41	0.001	253.6
IQ	108.8	17.1	101.4	16.9	-3.48	0.001	254
BDI	5.9	5.9	35.3	10.3	28.33	<0.001	207.4
Childhood Trauma Questionnaire
Total score	33.6	10.7	63.7	19.5	14.11	<0.001	166.3
Emotional abuse	7.5	3.7	17.1	5.8	14.50	<0.001	182.9
Physical abuse	5.6	2.1	9.6	5.2	7.41	<0.001	139.4
Sexual abuse	5.3	1.1	9.5	5.9	7.40	<0.001	115.1
Emotional neglect	8.5	3.9	17.2	5.7	13.21	<0.001	190.7
Physical neglect	6.2	2.1	10.3	4.3	8.85	<0.001	158.2
Interpersonal Reactivity Index
Perspective taking	19.3	4.3	13.8	5.97	-8.75	<0.001	221.7
Fantasy	19.1	5.5	16.4	7.0	-3.44	0.001	224.7
Empathic concern	20.4	4.1	19.8	5.4	-0.90	0.375	221.1
Personal distress	12.8	5.1	21.6	4.2	15.03	<0.001	249.4

Genotypes

The distributions of genotypes were in Hardy-Weinberg equilibrium in the whole sample (X2 = 0.096; p = 0.757) and were distributed as follows: LL genotype n = 103 (34.2%), SL genotype n = 131 (43.5%); SS genotype n = 45 (15%). When groups were analyzed separately, the distributions were also in Hardy-Weinberg equilibrium (BPD: X2 = 0.23; p = 0.631; HC: X2 = 0.02; p = 0.887) and there was no difference between allelic frequencies between groups (X2 = 2.84; p = 0.241; df = 2; Table 4). The division into S or LL-carriers resulted in n = 103 LL-carriers and n = 176 S-carriers. Another group formation was based on the rs25531, which offered the opportunity to form groups based on “transcriptional activity”(TA) (Table 4).

Table 4

Overview of 5-HTTLPR genotype distributions in the whole sample, and for patients with BPD and healthy controls (HC) separately.

The first column shows the distribution of the 5-HTTLPR genotypes regarding SS and LL homozygotes and SL heterozygotes. The second column shows the group formation according to the 5-HTTLPR and rs25531 genotypes, which results in high/high, high/low and low/low transcriptional activity groups.

5-HTTLPR genotypes				Transcriptional activity groups
	n total	BPD	HC			n total	BPD	HC
LL	103	59	44	high/ high		75	44	31
					LALA	75	44	31
SL	131	63	68
				high/low		140	68	72
SS	45	20	25		LALG	27	15	12
					LASA	112	52	60
					LASG	1	1	0
				low/ low		63	28	33
					LGLG	1	0	1
					LGSA	17	10	7
					SASA	45	20	25

Questionnaires and genotypes

The MANCOVA for the empathy scores revealed main effects for age and IQ (age F(4, 246) = 3.48, p = 0.009; IQ F(4, 246) = 6.69, p < 0.001), but no main effect of genotype or interaction with genotype.

Moderation analyses

In order to examine the nature of formation of specific social behavior associated with the 5-HTTLPR-genotype, we performed moderation analyses: We calculated the moderation for the independent variable CTQ total score (X; predictor), the dependent variable perspective taking (Y; outcome) and the moderator (M; susceptibility factor) the 5-HTTLPR genotype (SS+SL and LL). The overall model was significant (F(5, 199) = 6.48, p < 0.001, R2 = 0.1401), as was the interaction of CTQ by genotype (Interaction b = -0.0810, t(199) = -2.16, p = 0.032). According to Belsky et al. [64] and our own previous work [65], we investigated whether the data were compatible with the idea of differential susceptibility. Accordingly, we first tested if the predictor was associated with the moderator. Here, no correlation emerged between CTQ score and genotype (r = -0.130, p = 0.062). Second, the susceptibility factor (genotype) did not correlate with the outcome parameter (perspective taking; r = 0.084, p = 0.183). Third, we checked whether the simple slopes of the associations of CTQ with perspective taking differed significantly from zero. The significant difference to zero was only present for the slope of the regression in S-carriers (SS+SL b = -0.109, SE = 0.023, p < 0.001, LL: b = -0.028, SE = 0.031, p = 0.378; Fig 1; Table 5). Forth, we compared the simple slopes and found a significant difference between the groups with t = 3.08, p = 0.002 (SS+SL vs LL). In sum, these analyses are compatible with the “differential susceptibly” model, suggesting that the S-genotype may confer genetic plasticity to environmental variation.

Fig 1

Comparison of the regression lines of S-carriers (solid line) and LL-carriers (dashed line). The diagram supports the notion of differential susceptibly showing the crossing of the lines with the simple slopes differing between genotypes.

Table 5

Summary of moderation analyses performed for the predictor variable “CTQ” (total score) and the outcome variable “perspective taking”.

The table shows the moderator variables and covariates used and the respective model statistics and interactions between the predictor and moderator variables. The last two columns show tests for differential susceptibility, i.e. differences between slopes and differences from zero.

Moderator	Covariates	Model	Interaction moderator*CTQ	Differences of slopes from zero	Differences between slopes
SS+SL vs. LL	age, IQ	F(5, 199) = 6.48, p < 0.001, R2 = 0.1401	b = -0.0810, t(199) = -2.16, p = 0.032	SS+SL b = -0.109, SE = 0.023, p < 0.001	t = 3.08, p = 0.002
				LL: b = -0.028,SE = 0.031, p = 0.378
	age, IQ, BDI	(F(6, 196) = 9.20, p < 0.001, R2 = 0.2198	b = -0.0857, t(196) = -2.38, p = 0.018
SS vs. SL vs. LL	age, IQ	F(5, 199) = 6.77, p < 0.001, R2 = 0.1374	b = -0.0497, t(199) = -1.84, p = 0.067
	age, IQ, BDI		b = -0.0479, t(196) = -1.84, p = 0.067
TA groups: low/low, high/low, high/high	age, IQ	F(5, 199) = 6.77, p < 0.001, R2 = 0.1453	b = -0.0624, t(199) = -2.44, p = 0.015	high/high: b = -0.0378, SE = 0.0254, p = 0.1389	high/high vs. low/low: t = 3.30, p = 0.001
				high/low: b = -0.0808, SE = 0.0189, p < 0.001;	high/low vs. low/low: t = 2.00, p = 0.047
				low/low: b = -0.1237, SE = 0.0262, p < 0.001	high/high vs. high/low group t = 2.01, p = 0.046
	age, IQ, BDI	F(6, 196) = 9.43, p < 0.001, R2 = 0.2239	b = -0.0649, t(196) = -2.63, p = 0.009
BPD vs. HC	age, IQ	F(5, 199) = 11.71, p < 0.001, R2 = 0.2272	b = 0.0437, t(199) = 0.81, p = 0.419

We also calculated a moderation analysis for the three genotypes, SS, SL and LL. Here, the model was highly significant (F(5, 199) = 6.77, p < 0.001, R2 = 0.1374). However, the interaction of CTQ by genotype showed only a tendency toward statistical significance (Interaction b = -0.0497, t(199) = -1.84, p = 0.067), which could be related to the relatively small sample in the SS group (for comparison see Table 4).

In order to investigate the impact of the rs25531 within the 5-HTTLPR we built “transcriptional activity” (TA) groups (Table 4). We re-calculated the moderation analysis for the independent variable CTQ total score (X; predictor), the dependent variable perspective taking (Y; outcome) and the TA group as the moderator (M; low/low, high/low and high/high) The overall model was significant with (F(5, 199) = 6.77, p < 0.001, R2 = 0.1453), as was the interaction of CTQ by genotype (Interaction b = -0.0624, t(199) = -2.44, p = 0.015). Moreover, no correlations were found between CTQ score and TA group (r = -0.108, p = 0.123) or between perspective taking and TA group (r = 0.094, p = 0.138). Next, the analysis of slopes showed that only the slopes of TA groups high/low and low/low differed significantly from 0 (high/high: b = -0.0378, SE = 0.0254, p = 0.1389, high/low: b = -0.0808, SE = 0.0189, p < 0.001; low/low: b = -0.1237, SE = 0.0262, p < 0.001). The comparisons of slopes revealed that the slopes of the regression lines of high/high and high/low groups were significantly different from the slope of the low/low group (high/high vs. low/low: t = 3.30, p = 0.001; high/low vs. low/low: t = 2.00, p = 0.047). The slope of high/high also differed from the slope of high/low group (t = 2.01, p = 0.046; see Fig 2). These results confirm the differential susceptibility model and extend our results to the level of transcriptional activity.

Fig 2

Graphical representation of the association of childhood maltreatment (CTQ) and perspective taking ability in high/high (small dashed line), high/low (solid line) and low/low (longer dashed line) transcriptional activity groups based on 5-HTTLPR and rs25531.

We further aimed to examine whether the association of childhood trauma with perspective talking was related to diagnosis. Therefore, we performed the same moderation analysis, but tested the factor “group” (BPD vs. HC) as the moderator. The model was also significant (F(5, 199) = 11.71, p < 0.001, R2 = 0.2272), but the interaction of CTQ by group was not (Interaction b = 0.0437, t(199) = 0.81, p = 0.419), suggesting that a diagnosis of BPD was not the sole factor impacting on the association of childhood trauma with empathic perspective taking. Since depressivity is highly prevalent among individuals suffering from BPD, we repeated the moderation analyses and included also the BDI score as a covariate into the calculation. Here, the same results were obtained as in the previous analyses without the covariate BDI (see Table 5).

Discussion

In the present study we aimed to investigate the role of the 5-HTLLPR concerning the association between childhood trauma and trait empathy. The moderation analysis and tests for differential susceptibility showed that the influence of childhood maltreatment on empathic perspective-taking seemed to be specific for S-carriers, suggesting that this allele confers genetic plasticity to environmental variation. Put differently, in people with at least one S-allele childhood maltreatment seems to be related to reduced perspective taking, whereas the absence of childhood maltreatment is associated with well-preserved perspective taking capacities. As noted by Belsky et al. [66], the mere absence of maltreatment is not equivalent to high-quality parenting. When applying this idea to the present findings, it can tentatively be hypothesized that emotional warmth and availability during early development may lead to even better-than-average perspective-taking abilities (in this case, the slopes of the regression lines shown in Fig 1 may diverge to a greater extent if extrapolated to the left). In contrast, LL-carriers seemed to be unresponsive to childhood adversity in terms of consequences for trait empathy. The relation of trauma and reduced perspective-taking was already shown in previous studies and shown to be related to alexithymia and stress [14]. A possible explanation for reduced perspective taking in S-allele carriers exposed to childhood trauma may be the negative effect of stress, induced by the traumatic history and the following consequences (e.g. unsuccessful coping strategies), on the development of cognitive empathic perspective taking. Our findings are consistent with previous studies reporting that S-carriers showed increased attention towards emotional stimuli and especially negative stimuli when compared to L-carriers [24, 28, 67]. Owens and colleagues further reported that S-homozygotes (adolescents) were impaired in emotion recognition of negative and neutral stimuli and had more difficulties in responding to ambiguous negative feedback [67].

When looking at prosocial behavior, Stoltenberg and colleagues reported that S-allele carriers scored higher in social anxiety and lower in prosocial behavior [62]. Moreover, lower levels of sensitive responsiveness to their own toddlers were found in parents with the SS-allele [42]. Unfortunately, these studies did not include measures of the participants’ own experiences during childhood. Moreover, Gyurak et al. [27] found that SS-homozygotes showed greater levels of emotional reactivity accompanied by an increased psychosocial stress response. Similarly, another group reported that SS-carrier showed the greatest increase in cortisol levels following the Trier Social Stress Test. In addition, the association between genotype and cortisol reactivity was strongest when receiving negative feedback. The authors concluded that carrying the SS-allele may make the individuals more vulnerable to stressful life events, which leads to a greater risk for the severe psychological and physical health consequences associated with heightened cortisol exposure [29]. In support of this assumption Gotlib and colleagues examined the association between stress, 5-HTTLPR and depression in children. They demonstrated that girls, who were homozygous for the S-allele showed higher and prolonged cortisol levels in response to a stressor (mental arithmetic and Ewart Social Competence Interview) compared to L-allele carrying girls [68]. Additionally, it was shown that acute stress exposure led to a significant impairment in the inhibition of negative affective information only in SS-carriers. The authors concluded that a cognitive-attentional bias for negative emotional information may make an individual more vulnerable for stress-induced depressive symptoms [69]. With regard to our study, increased stress-reactivity and altered emotion processing in S-allele carriers may cause, together with the experiences of childhood adversity, impairment in taking the perspective of another individual. With regard to depression, the exact role of the 5-HTTLPR in the development of depression is unclear. This was shown by recent meta-analyses, which reported an association of stressful life events and depression in SS-homozygotes or S-carrying heterozygotes [45, 46], whereas other studies failed to determine such an association [31, 47]. For example, Culverhouse and colleagues found a significant main effect of sex and life stressor (high risk factor for depression), but they did not found an effect of genotype on the association of stress and depression, even if they included only studies with large sample sizes [47]. The authors concluded that, if any interaction would exist, it would not be a generalizable effect, only detectable in limited situations and of modest effect size. In our study, the inclusion of depressivity as a covariate did not affect the interaction, which suggests that the effect on perspective taking was not due to depressive symptoms. This further implies that our study does also not support the link between 5-HTTLPR, stress and depression. Eventually, the 5-HTTLPR, together with stress, may induce stress and emotion processing impairments (as described above), which lead only in a subset of individuals to the development of depression. This subset may bear additional risks, which are currently not in the focus of interaction studies. One potential factor could be the transcriptional activity of the serotonin transporter gene, which could be altered by epigenetic processes or other SNPs, as for example the rs25531 [35, 36, 70].

In our study, additional analyses according to the transcriptional activity of the serotonin transporter gene revealed similar results, i.e. differential susceptibility in the low/low and high/low groups akin to what emerged in S-allele carriers of the 5-HTTLPR. This finding is in accordance with previous studies, which also did not report modulation of the associations between 5-HTTLPR and phenotypes by the rs25531 [37, 71]. Interestingly, however, we found a graded effect of the transcriptional activity, indicating that the lower the activity of the serotonin transporter gene, the greater the genetic plasticity with regard to the effect of childhood adversity on empathic perspective-taking. When calculating the additional moderation analysis with the three 5-HTTLPR groups, SS, SL and LL, the interaction failed to reach significance. This could be due to a lack of statistical power due to the small sample size in the SS-group, or be related to the fact that the exact genotype may not be as relevant as the existence of at least one “risk allele”. In any event, even though the statistical power for detecting significant interaction effects was sufficiently large with α = 0.75, it is warranted to replicate the study in an independent, and preferably larger, sample.

Our study has several other limitations. For one, since we included only female participants, our conclusions cannot be generalized for males. Second, the clinical and the non-clinical group differed significantly with regard to the experience of childhood trauma. However, since our main interest pertained to the influence of childhood adversity on trait empathy and its moderation by genotype, we were much less concerned with the presence or absence of specific effects of a diagnosis of BPD. In support of this idea, the moderation analysis with the moderator group (BPD/HC) did not show a specific effect of BPD on the association of childhood trauma with perspective taking. Third, as already pointed out, for a more substantial corroboration of the differential susceptibility hypotheses, it would have been desirable to expand measures of adversity in the direction of parental warmth and caregiver availability to better reflect the whole spectrum ranging from superior to poor environmental conditions [66]. Forth, since we assessed only adversity during childhood, we were unable to exclude confounding effects of recent negative life events, which may also have impact on present social behavior. Thus, future studies may investigate childhood, as well as recent adversity, in order to define the concrete contribution of these factors on social behavior and the development of psychopathology.

Together, the present study is the first to show that the association of empathic perspective-taking with childhood adversity is moderated by the 5-HTTLPR, and the transcriptional activity of the serotonin transporter gene. It therefore corroborates previous findings suggesting that the S-allele of the 5-HTTLPR conveys differential susceptibility to environmental cues, as does low transcriptional activity.

Data from genotyping, CTQ and IRI questionnaires for each participant.

(PDF)

Click here for additional data file.

9 Aug 2019

PONE-D-19-14573

The association between childhood maltreatment and empathic perspective taking is moderated by the 5-HTT linked polymorphic region. Another example of “differential susceptibility”

PLOS ONE

Dear Prof. Brüne,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by Sep 23 2019 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Huiping Zhang

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

3. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This study focuses on the effect of 5-HTT linked polymorphic region in the association between childhood maltreatment and empathic perspective taking. In 279 female patients with BPD or healthy participants, childhood maltreatment and empathic abilities were measures using standardized questionnaires and genotypes in 5-HTT linked polymorphic region were determined and further categorized by different levles of transcriptional activities. Moderator analysis revealed that S-allele carriers and corresponding lower transcriptional activities showed significant response to childhood mental environmental impact in empathic perspective taking.

Previous studies mainly investigated the association negative environmental impact in empathic perspective taking while this study demonstrated the bidirectional empathic perspective taking outcome from childhood treatment.

Major:

1.How was the effect of recent major life events in adulthood evaluated for the 279 participants, which may serve as controls in this study.

2.Moderator analysis of CTQ total score, perspective taking and the BPD/ healthy participants.

Minor:

1.How does the CTQ questionnaire define age range of childhood?

2.Moderator analysis on three genotypes (Line 264-268) can be moved after moderator analysis on SS+SL vs LL (after line 244).

3.Typo: genotype should be genotypes in line 264.

4.In Table 3 and regarding description, the total n in the SS+SL group was stated as 176, however, only 175 samples can be identified.

5.In Table 3, LGSG group or SGSG group are also not presented in patients or healthy participants. Either remove the SASG group from the table or add these groups.

6.In the discussion (lines 287-291), the author’s statement regarding the benefit of emotional warmth and availability during childhood is confusing.

Reviewer #2: In this manuscript, authors investigate the role of 5-HTTLPR on the effects of child maltreatment on empathy. In a sample of 279 women (142 with BPD and 137 healthy controls), they studied the effects of 5-HTTLPR genotype on empathic perspective taking. The manuscript is well-written, but not rigorous in the study design and methods. Further, the conclusions made are premature, given the lack of replication and the numerous weaknesses identified.

1. There is a lack of power due to the small sample size of their cohort, particularly for the interaction analysis. Authors should include a power analysis to justify the sample size used in the study.

2. All subjects should have trauma information, including controls, since one of the main analysis performed is an interaction analysis.

3. There is no information about race. This should be also accounted for in the analysis.

4. Depression is highly prevalent among BPD. Analysis should account for depression as a covariate.

5. No information about healthy controls group is provided. What is the prevalence, if any, of disorders and medication listed Table 1 for BPD group.

6. No efforts for replication were made. Given the sample size, it is critical to conduct a replication analysis in an independent cohort.

7. In the largest meta-analysis studying the interaction between stress and 5-HTTLPR genotype in the context of depression, the Culverhouse et al found no association. Authors should discuss why studying 5-HTTLPR genotype is still relevant.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

6 Sep 2019

Revision 1

PONE-D-19-14573

The association between childhood maltreatment and empathic perspective taking is moderated by the 5-HTT linked polymorphic region. Another example of “differential susceptibility”

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

3. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

� Response: We revised the manuscript with regards to the journal requirements. We also add the study findings as supporting information.

________________________________________

Reviewer #1:

This study focuses on the effect of 5-HTT linked polymorphic region in the association between childhood maltreatment and empathic perspective taking. In 279 female patients with BPD or healthy participants, childhood maltreatment and empathic abilities were measures using standardized questionnaires and genotypes in 5-HTT linked polymorphic region were determined and further categorized by different levles of transcriptional activities. Moderator analysis revealed that S-allele carriers and corresponding lower transcriptional activities showed significant response to childhood mental environmental impact in empathic perspective taking.

Previous studies mainly investigated the association negative environmental impact in empathic perspective taking while this study demonstrated the bidirectional empathic perspective taking outcome from childhood treatment.

Major:

1. How was the effect of recent major life events in adulthood evaluated for the 279 participants, which may serve as controls in this study.

Response: Unfortunately, we cannot exclude or control for effects of recent/present life events, because we did not assess traumatic experiences during adulthood. Since the reviewer raised an important point with this concern, we add this point to the limitation section. This part reads as follows: “Forth, since we assessed only adversity during childhood we were unable to exclude confounding effects of recent negative life events, which may also have impact on present social behavior. Thus, future studies may investigate childhood, as well as recent adversity, in order to define the concrete contribution of these factors on social behavior and the development of psychopathology.”

2. Moderator analysis of CTQ total score, perspective taking and the BPD/ healthy participants.

Response: We agree with the reviewer that is important to show that the associations found were not already coming up because of the BPD and HC groups. Therefore, we added another moderation analysis with Group (BPD/HC) as the moderator. We added the following part to the methods: “Another moderation analyses was performed for the SS, SL and LL Genotypes and for Group (BPD vs HC) as the moderator instead of Genotype. “

The following part was added to the results section: “We further aimed to examine whether the association of childhood trauma with perspective talking was related to diagnosis. Therefore, we performed the same moderation analysis but tested the factor group (BPD and HC) as the moderator. The model was also significant (F(5, 199) = 11.71, p < 0.001, R² = 0.2272), but the interaction of CTQ by group was not (Interaction b = 0.0437, t(199) = 0.81 , p = 0.419), suggesting that a diagnosis of BPD was not the sole factor impacting on the association of childhood trauma with empathic perspective taking “

We also extended the limitation section in the discussion: “However, since our main interest pertained to the influence of childhood adversity on trait empathy and its moderation by genotype, we were much less concerned with the presence or absence of specific effects of a diagnosis of BPD. In support of this idea, the moderation analysis with the moderator group (BPD/HC) did not show a specific effect of BPD on the association of childhood trauma with perspective taking.”

Minor:

1. How does the CTQ questionnaire define age range of childhood?

Response: The age range is not clearly defined by the questionnaire. The instruction on the questionnaire asks the participants to refer to their childhood and adolescence. This information was added to the methods: “Participants were asked to rate the occurrence of maltreatment on a 5-point Likert scale (1 = never, 5 = very often) pertaining to their childhood and youth.”

2. Moderator analysis on three genotypes (Line 264-268) can be moved after moderator analysis on SS+SL vs LL (after line 244).

Response: We agree with the reviewer and change the manuscript accordingly

3. Typo: genotype should be genotypes in line 264.

Response: Right, thanks for the hint!

4. In Table 3 and regarding description, the total n in the SS+SL group was stated as 176, however, only 175 samples can be identified.

Response: We verified the data shown in Table 3 several times, but could not find a mistake.

5. In Table 3, LGSG group or SGSG group are also not presented in patients or healthy participants. Either remove the SASG group from the table or add these groups.

Response: We agree with the reviewer and decided to remove the SASG group.

6. In the discussion (lines 287-291), the author’s statement regarding the benefit of emotional warmth and availability during childhood is confusing.

Response: we revised the part in order to make it more clearly: “As noted by Belsky et al. [63], the mere absence of maltreatment is not equivalent to high-quality parenting. When applying this idea to the present findings, it can tentatively be hypothesized that emotional warmth and availability during early development may lead to even better-than-average perspective-taking abilities (in this case, the slopes of the regression lines shown in Fig 1 may diverge to a greater extent if extrapolated to the left).”

Reviewer #2:

In this manuscript, authors investigate the role of 5-HTTLPR on the effects of child maltreatment on empathy. In a sample of 279 women (142 with BPD and 137 healthy controls), they studied the effects of 5-HTTLPR genotype on empathic perspective taking. The manuscript is well-written, but not rigorous in the study design and methods. Further, the conclusions made are premature, given the lack of replication and the numerous weaknesses identified.

1. There is a lack of power due to the small sample size of their cohort, particularly for the interaction analysis. Authors should include a power analysis to justify the sample size used in the study.

Response: Following up on the reviewer’s concern, we added a power analysis for the interaction effect: “We conducted a power analysis for the interaction, i.e. the differences between slopes for the moderation model with the genotypes SS+SL and LL. Therefore, we used G*Power Version 3.1.9.2. (56). For the calculation of the achieved power with the present sample for the interaction effects we applied the option “t-test-linear bivariate Regression, two groups, difference between slopes”, which showed, that we reached a statistical power of α = 0.749 with a sample size of 205 participants.”

We also extended the discussion accordingly: “In any event, even though the statistical power for detecting significant interaction effects was sufficiently large with α = 0.75, it, it is warranted to replicate the study in an independent, and preferably larger, sample.”

In addition, we would like to mention that other recent work utilized comparable sample sizes (for example Nardi et al., 2013; Owens et al., 2012; Wagner et al., 2009, for review on 5-HTTLPR and psychopathology see Kenna et al., 2012)

2. All subjects should have trauma information, including controls, since one of the main analysis performed is an interaction analysis.

Response: The moderation analyses were conducted for the whole sample, e.g. controls and BPD (except missing data). Participants of both groups received the CTQ (see also Table 2)

3. There is no information about race. This should be also accounted for in the analysis.

Response: The reviewer is absolutely right with expressing concern regarding the ethnical background in a genetic study. We added the information to the material and methods – participant section. “Regarding the ethnical background, 95.2 % were Caucasians, 4.4 % originated from the Middle East (mainly of Turkish origin) and 0.4 % from the Far East (Vietnamese).”

4. Depression is highly prevalent among BPD. Analysis should account for depression as a covariate.

Response: We agree with the reviewer and added moderation analyses, which include the covariate BDI score.

The methods were extended as following: “In addition, we used the Beck’s Depression Inventory to assess the self-rated level of depressivity [54 […] “In order to examine the impact of depressivity, we calculated additional moderation analyses with the additional covariate “BDI score” for the moderators “SS+SL vs. LL”, “SS, SL and LL” and transcriptional activity.”

We extended the results accordingly (see also Table 2): “Patients differed in age and IQ from control participants and reported more severe experiences of childhood maltreatment as well as more depressive symptoms.” […] “Since depressivity is highly prevalent among individuals suffering from BPD, we repeated the moderation analyses and included also the BDI score as a covariate into the calculation. Here, the same results were obtained as in the previous analyses without the covariate BDI (moderator = SS+SL vs. LL: the model (F(6, 196) = 9.20, p < 0.001, R² = 0.2198) and the interaction (b = -0.0857, t(196) = -2.38 , p = 0.018) were significant; moderator = SS vs.SL vs. LL: the model was significant (F(6, 196) = 8.81, p < 0.001, R² = 0.2124), the interaction showed only a tendency (b = -0.0479, t(196) = -1.84 , p = 0.067); moderator TA group: the model (F(6, 196) = 9.43, p < 0.001, R² = 0.2239) and interaction (b = -0.0649, t(196) = -2.63 , p = 0.009) were significant).

We also discussed the association with depressivity more extensively, as raised by Reviewer (Comment 7) and referred to our results concerning depression: “[…] In our study, the inclusion of depression as a covariate did not affect the interaction, which suggests that the effect on perspective taking was not due to symptoms of depression. This further implies that our study does also not support the link between 5-HTTLPR, stress and depression.[…]”

5. No information about healthy controls group is provided. What is the prevalence, if any, of disorders and medication listed Table 1 for BPD group.

Response: We agree with the reviewer that no information was given. We added the following sentence to the methods: “The control participants were free of medication and psychiatric disorders.”

6. No efforts for replication were made. Given the sample size, it is critical to conduct a replication analysis in an independent cohort.

Response: We absolutely agree with the reviewer that a replication in a larger sample is required and included this point in the discussion.

Discussion: “In any way, since the power for detecting significant interaction was 0.749, it is warranted to replicate the study in an independent, and preferably larger, sample.

7. In the largest meta-analysis studying the interaction between stress and 5-HTTLPR genotype in the context of depression, the Culverhouse et al found no association. Authors should discuss why studying 5-HTTLPR genotype is still relevant.

Response: We are grateful for the advice to extend the discussion to the depression debate and added the following section to the discussion: “With regard to depression, the exact role of the 5-HTTLPR in the development of depression is unclear. This was shown by recent meta-analyses, which reported on the one side an association of stressful live events and depression in SS-homozygotes or S-carrying heterozygotes [42, 43], and on the other side other studies found no significant association [31, 44]. Culverhouse and colleagues found a significant main effect of sex and life stressor (high risk factor for depression), but they did not found an effect of genotype on the association of stress and depression, even if they included only studies with large sample sizes [44]. The authors concluded that if any interaction would exist, it would not be a generalizable effect, only detectable in limited situations and of modest effect size. In our study, the inclusion of depression as a covariate did not affect the interaction, which suggests that the effect on perspective taking was not due to symptoms of depression. This further implies that our study does also not support the link between 5-HTTLPR, stress and depression. Eventually, the 5-HTTLPR, together with stress, may induce stress and emotion processing impairments (as described above), which lead only in a subset of individuals to the development of depression. This subset may bear additional risks, which are currently not in the focus of interaction studies. One potential factor could be the transcriptional activity of the serotonin transporter gene, which could be altered by epigenetic processes or other SNPs, as for example the rs25531 [33, 34, 67].

Submitted filename: Reponse to Reviewers.docx

Click here for additional data file.

19 Nov 2019

PONE-D-19-14573R1

The association between childhood maltreatment and empathic perspective taking is moderated by the 5-HTT linked polymorphic region. Another example of “differential susceptibility”

PLOS ONE

Dear Prof. Brüne,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

There are some concerns about the organization of the content in the manuscript. For example, some data analyses and the obtained results were placed in the Result section without describing the method in the Methods section.

We would appreciate receiving your revised manuscript by Jan 03 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Huiping Zhang

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #3: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #3: No

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Considering the study focuses on childhood adversity instead of lifetime adversity, better definition in age and evidence of adversity only in childhood are critical in the study design.

Reviewer #3: In this manuscript, Flasbeck et al. investigated whether, or not, the genetic variations in the 5HTTLP region (or SLC6A4) modify the effect of childhood trauma exposure on the empathy. To test this hypothesis, they collected the 205 sample of 142 BPD patients and 137 controls, and they measured PT and CTQ scores related with empathy and childhood trauma exposure, respectively. Also, they genotyped two genetic variations including 1) short/long alleles of 5HTTLPR and 2) rs25531 (A or G alleles). They conducted moderation analysis and they found that the effect of childhood trauma exposure on the empathy (PT score) among the carriers with specific alleles (short-5HTTLPR or A allele of rs25531) significantly differs from its effect among the non-carriers. Therefore, they concluded that the genetic variations of the 5HTTLP modifies the risk effect of childhood trauma exposure on the empathy.

Major comments:

• There were three moderation analyses with the same outcome and predictor variables by the different variables: 1) S allele carriers vs non-carriers; 2) transcriptional activity (L/A genotype vs others); and 3) the BPD status. Although all the summary statistics were written down in the result section, it would be much better for the audience if they summarize the statistics into a table.

• The author did power calculation using G*Power, but they did not explain about how they computed in detail at all. They should explicitly explain the parameter values (e.g. expected effect size and significance threshold) in the method section.

Minor comments:

• There are many different statistical tests in the manuscript. I think the main analysis is the moderation analysis. So, it would be nice if author better shows the three main moderation analysis models at once in the method section. It could be a formula format or a table. This will really improve understanding of this study.

• I think the full name of MANOCOVA should be defined in the page 8 (line number: 165).

• I think you meant the moderator (M: low/low, high/low and high/high), rather than (M: low/low, high/low and low/low), in the page 13 (line number: 253).

• I am not sure why the subtitle, “Genotyping” was written in Italic font in the page 7 (line number 145).

• Grammar error. “by calculation the correlation” in the page 8 (line number 176).

• What does “X2(2)” mean in the page 10 (line number: 204)?

• Author wrote why they divided the sample into S carriers vs. LL carriers in the result section in the page 10 (line number: 205-206). I think this should go into the method section. Also, it would be nice if author briefly explain about the rationale behind this stratification by S allele.

• I found that something should go into the discussion section other than in the result. The sentence in the page 12 (line number: 231-232) is the interpretation about the result from the moderation test.

• It was not described how the moderation analysis for the three genotypes was conducted. Did you use the additive model? It should be described in the method section.

• It would be nice if the authors show the LD information between the S/L variations and rs25531. Also, the allele frequencies of these variations in general population could be relevant information, which may exist in the gnomAD (https://gnomad.broadinstitute.org/) database.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

25 Nov 2019

Revision 2

PONE-D-19-14573

The association between childhood maltreatment and empathic perspective taking is moderated by the 5-HTT linked polymorphic region. Another example of “differential susceptibility”

Reviewer #1:

Considering the study focuses on childhood adversity instead of lifetime adversity, better definition in age and evidence of adversity only in childhood are critical in the study design

Response: We are grateful for the reviewer’s advice. Accordingly, we explicitly make the distinction between childhood trauma and other adverse life events in the discussion. In addition, we now point out, throughout the manuscript, that the present study deals with early adversity.

Reviewer #2:

Major comments:

* There were three moderation analyses with the same outcome and predictor variables by the different variables: 1) S allele carriers vs non-carriers; 2) transcriptional activity (L/A genotype vs others); and 3) the BPD status. Although all the summary statistics were written down in the result section, it would be much better for the audience if they summarize the statistics into a table.

Response: We are thankful for the comment and created table 5, which shows the moderator variables, covariates and statistics of the moderation analyses conducted (page 15). The table was a great advice, since results are more comprehensive and accessible for the reader. We therefore deleted the results of analyses with BDI as an additional covariate, since it can be found in the table.

Table 5. Summary of moderation analyses performed for the predictor variable “CTQ” (total score) and the outcome variable “perspective taking”. The table shows the moderator variables and covariates used and the respective model statistics and interactions between the predictor and moderator variables. The last two columns show tests for differential susceptibility, i.e. differences between slopes and differences from zero.

Moderator Covariate Model Interaction moderator*CTQ Differences of slopes from 0 Differences between slopes

SS+SL vs. LL age, IQ F(5, 199) = 6.48, p < 0.001, R² = 0.1401 b = -0.0810, t(199) = -2.16 , p = 0.032 SS+SL b = -0.109, SE = 0.023, p < 0.001 t = 3.08, p = 0.002

LL: b = -0.028,

SE = 0.031, p = 0.378

age, IQ, BDI (F(6, 196) = 9.20, p < 0.001, R² = 0.2198 b = -0.0857, t(196) = -2.38 , p = 0.018

SS vs. SL vs. LL age, IQ F(5, 199) = 6.77, p < 0.001, R² = 0.1374 b = -0.0497, t(199) = -1.84 , p = 0.067

age, IQ, BDI b = -0.0479, t(196) = -1.84 , p = 0.067

TA groups: low/low, high/low, high/high age, IQ F(5, 199) = 6.77, p < 0.001, R² = 0.1453 b = -0.0624, t(199) = -2.44 , p = 0.015 high/high: b = -0.0378, SE = 0.0254, p = 0.1389 high/high vs. low/low: t = 3.30, p = 0.001

high/low: b = -0.0808, SE = 0.0189, p < 0.001; high/low vs. low/low: t = 2.00, p = 0.047

low/low: b = -0.1237, SE = 0.0262, p < 0.001 high/high vs. high/low group t = 2.01, p = 0.046

age, IQ, BDI F(6, 196) = 9.43, p < 0.001, R² = 0.2239 b = -0.0649, t(196) = -2.63 , p = 0.009

BPD vs. HC age, IQ F(5, 199) = 11.71, p < 0.001, R² = 0.2272 b = 0.0437, t(199) = 0.81 , p = 0.419

* The author did power calculation using G*Power, but they did not explain about how they computed in detail at all. They should explicitly explain the parameter values (e.g. expected effect size and significance threshold) in the method section.

Response: We extended the section on the power calculation as follows: Page 8 line 161-165: “We conducted a power analysis for interaction effects, i.e. the differences between slopes for the moderation model with the genotypes SS+SL and LL, using G*Power, Version 3.1.9.2. [59]. Power calculation for the current sample of 205 participants was determined by the following model: t-test-linear bivariate regression, two groups, difference between slopes, with α set at 0.05. Standard deviation of the residuals, the sample size and the difference between the slopes were also considered. Accordingly, the statistical power coefficient was 0.75. “

Minor comments:

* There are many different statistical tests in the manuscript. I think the main analysis is the moderation analysis. So, it would be nice if author better shows the three main moderation analysis models at once in the method section. It could be a formula format or a table. This will really improve understanding of this study.

Response: The authors completely agree with the reviewer that the statistical methods section was very long and difficult to follow. We therefore added another table showing the moderation analyses (page 10)

Table 2. Summary of moderation analyses conducted. The predictor and outcome variables remained constant across calculations whereas the moderator and control variables were exchanged.

Predictor Outcome Moderator Covariates

CTQ

total score

Perspective

taking

SS+SL vs. LL age, IQ

age, IQ, BDI

SS vs. SL vs. LL age, IQ

age, IQ, BDI

TA groups: low/low, high/low, high/high age, IQ

age, IQ, BDI

BPD vs. HC age, IQ

* I think the full name of MANOCOVA should be defined in the page 8 (line number: 165).

Response: We changed this sentence accordingly (now page 9 line 179): “In order to investigate the effect of the genotype, we calculated a multivariate analysis of covariance (MANCOVA) with the covariates IQ and age and the between-subject factor genotype (SS+SL vs. LL) and the independent variables were the IRI scores.”

* I think you meant the moderator (M: low/low, high/low and high/high), rather than (M: low/low, high/low and low/low), in the page 13 (line number: 253).

Response: Correct, this was an error, thank you for mentioning!

* I am not sure why the subtitle, "Genotyping" was written in Italic font in the page 7 (line number 145).

Response: Sorry, this shouldn’t be italic and was corrected.

* Grammar error. "by calculation the correlation" in the page 8 (line number 176).

Response: Thank you for the hint, we improved the sentence to “by calculating the correlation of…”

* What does "X2(2)" mean in the page 10 (line number: 204)?

Response: The (2) stands for the df of the Chi square test, we changed it to: “(X² = 2.84; p = 0.241; df = 2; Table 4)”

* Author wrote why they divided the sample into S carriers vs. LL carriers in the result section in the page 10 (line number: 205-206). I think this should go into the method section. Also, it would be nice if author briefly explain about the rationale behind this stratification by S allele.

Response: We transferred the section into the methods section and added the following part explaining the rationale (page 8 line 166-172): “In accordance with previous studies, we divided the sample into S-carriers (SS+SL pooled) and LL-carriers (e.g. [60-62]). This approach was justified, because previous studies reported no differences between SS and SL-carriers with regard to personality traits, suggesting a dominant-recessive type of association of the S-allele with personality (21).

Similarly, following previous research (e.g., [37]), subjects were further divided into groups according to the ”transcriptional activity“ (TA) of the rs25531, which is reported in detail in Table 4. “

The results were changed to (page 11 line 223-226): “The division into S or LL-carriers resulted in n = 103 LL-carriers and n = 176 S-carriers. Another group formation was based on the rs25531, which offers the opportunity to build “transcriptional activity“ (TA) groups, (see Table 4).”

* I found that something should go into the discussion section other than in the result. The sentence in the page 12 (line number: 231-232) is the interpretation about the result from the moderation test.

Response: The authors agree with the reviewer and deleted the interpretation in the results.

* It was not described how the moderation analysis for the three genotypes was conducted. Did you use the additive model? It should be described in the method section.

Response: The moderation analysis of the three genotypes can be found on page 9 line 194-196 and now in the table 2.

* It would be nice if the authors show the LD information between the S/L variations and rs25531. Also, the allele frequencies of these variations in general population could be relevant information, which may exist in the gnomAD (https://gnomad.broadinstitute.org/) database.

Response: We inserted the information into the introduction (page 3 line 69-71, page 4 line 73-76): “However, this polymorphic variation does not generally occur more frequently in clinical samples compared to the general population [30-32]. The general population is heterozygous, whereas the LL-genotype is less common, and the SS-variant relatively rare, in part depending on ethnicity [33, 34]. In addition, controversy exists about the effect of rs25531, a SNP within the 5-HTTLPR repetitive element, with the A-variant of the L-allele being associated with greater transcriptional activity and thus more efficient serotonin turnover [35-37], whereby a linkage disequilibrium between 5-HTTLPR and rs25531 has been described, with the rarer G-variant of rs25531 occurring more frequently together with the L-allele than with the S-allele [36, 38].

Submitted filename: Reponse to Reviewers.docx

Click here for additional data file.

6 Dec 2019

The association between childhood maltreatment and empathic perspective taking is moderated by the 5-HTT linked polymorphic region. Another example of “differential susceptibility”

PONE-D-19-14573R2

Dear Dr. Brüne,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Huiping Zhang

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

10 Dec 2019

PONE-D-19-14573R2

The association between childhood maltreatment and empathic perspective taking is moderated by the 5-HTT linked polymorphic region. Another example of “differential susceptibility”

Dear Dr. Brüne:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Huiping Zhang

Academic Editor

PLOS ONE