DRD4 Interacts with Adverse Life Events in Predicting Maternal Sensitivity via Emotion Regulation

Abstract

We examined whether and how the dopamine receptor D4 (DRD4) interacts with adverse life events to predict maternal sensitivity directly and indirectly via emotion regulation. The sample included 207 (105 European American, 102 African American) mothers and their children (52% female). Sensitive maternal behavior was rated and aggregated across five stress-free and stress-inducing tasks when children were about 2 years old, when mothers also retrospectively reported on their adverse life experiences and transitions throughout childhood from birth to age 20. When children were about 1year old, mothers reported on their difficulties with emotion regulation. Results from path analysis indicated that mothers who carried the long allele of DRD4 and experienced more adverse life events were less sensitive in interactions with their children. These mothers were also more likely to have difficulties with emotion regulation, which in turn predicted lower maternal sensitivity. These effects were significant above and beyond the effects of maternal education, coherence of mind, race, and infants’ DRD4 genotype, and did not vary for African American and European American mothers. Results suggest that genetic predispositions modify the effects of maternal experience of adverse life events on maternal sensitivity and that emotion regulation serves as one mechanism by which genetic factors and gene-environment interactions affect maternal behavior.

Behavioral genetic research demonstrates that genes play an important role in individual differences in parenting behavior (Klahr & Burt, 2014). A number of studies have examined associations between specific genotypes and parenting outcomes, including genes involved in the dopamine, serotonin, oxytocin, and vasopressin neurobiological systems (Mileva-Seitz Bakermans-Kranenburg, & van IJzendoorn, 2016). While some studies found significant associations (e.g., Cent et al., 2014), many others failed to find simple direct effects of genotypes on parenting (e.g., Mileva-Seitz et al., 2013; Van IJzendoorn, Bakermans-Kranenburg, & Mesman, 2008), pointing to the importance of examining mechanisms by which genes are associated with parenting and environmental factors that may interact with genetic factors to influence sensitive parenting behavior. In this study, we examine the association between dopamine receptor D4 (DRD4) and maternal sensitivity, test the mediating role of maternal emotion regulation on this association, and examine interaction effects between DRD4 and adverse life events in relation to emotion regulation and maternal sensitivity.

Maternal Sensitivity and the Role of DRD4

Maternal sensitivity refers to a mother’s ability to perceive and interpret her infant’s signals accurately and respond promptly and appropriately (Ainsworth, Blehar, Waters, & Wall, 1978). Maternal sensitivity predicts infants’ subsequent social-emotional development, such as attachment security (Verhage et al., 2016), adaptive emotion regulation (Perry, Calkins, & Bell, 2016), and fewer behavior problems (Leerkes, Blankson, & O’Brien, 2009). Thus, identifying the factors that promote or undermine sensitive maternal behaviors has important implications for prevention and intervention efforts aimed at promoting healthy child development.

The dopamine system is involved in behavioral activation, motivation, and reward processing (Ikemoto & Panksepp, 1999) and proposed to be important for caregiving behaviors as it activates a motivational system controlled by the nucleus accumbens that promotes appropriate responses to significant stimuli (Numan, 2010). DRD4 is one of several dopamine-related genes that have been studied in relation to human behavior. DRD4 contains a 48-bp Variable Number Tandem Repeat polymorphism in the third exon (van Tol et al., 1992), which results in ten allelic products comprised of 2–11 repeat units. The longer variants (7 repeats or more) blunt the intracellular response to dopamine in vitro as compared with other variants (Asghari et al., 1995). Individuals with one or two long alleles show lower gene expression (Schoots & Van Tol, 2003), which may undermine reward processing (Blum, Cull, Braverman, Chen, & Comings, 1997). The DRD4 long allele has also been shown to undermine social cognition, such as theory of mind (Lackner, Sabbagh, Hallinan, Liu, & Holden, 2012) and maternal cognitions about infant crying (Leerkes, Su, Calkins, Henrich, & Smolen, 2017).

The proposed important role of dopamine in caregiving behavior, as well the demonstrated role of DRD4 in reward processing and social cognition, which are important underlying skills for sensitive parenting (Barrett & Fleming, 2011), suggest DRD4 as a prime candidate gene to study for maternal sensitivity. However, to our knowledge, only four published studies have examined the associations between DRD4 and maternal sensitivity among mothers of infants or young children and none found main effects of DRD4 on sensitive parenting behaviors (Fortuna et al., 2011; Kaitz et al., 2010; Leerkes et al., 2017; van IJzendoorn et al. 2008). However, three of these studies showed that DRD4 moderated the extent to which other risk factors influenced maternal sensitivity. Specifically, daily hassles were associated with lower maternal sensitivity only among mothers with DRD4 long alleles coupled with another dopamine risk polymorphism (COMTval158met) (van IJzendoorn et al. 2008). Likewise, higher levels of child risk at birth (Fortuna et al., 2011) and infant fussiness (Kaitz et al., 2010) were associated with less sensitive parenting only among mothers carrying the DRD4 long allele. Thus, carrying the long DRD4 alleles appears to place mothers at elevated risk for compromised parenting in the face of other risk factors, suggesting the importance of DRD4 as a moderator.

The lack of prior findings regarding a main effect of DRD4 on sensitive parenting may also suggest that DRD4 genotype influences parenting indirectly through complex mediating pathways. It has been theorized that social cognition and affective processes are mechanisms linking genes to parenting behavior (Leerkes et al., 2017). In this study, we examine emotion regulation as a mechanism linking DRD4 to maternal sensitivity, as discussed below.

Emotion Regulation as a Mediating Pathway Linking DRD4 to Maternal Sensitivity

Emotion regulation is the automatic or effortful behaviors, strategies, and skills that modulate the experience and expression of emotions, both positive and negative (Calkins & Hill, 2007). The regulation of negative emotions often focuses on down-regulation and can include behavioral strategies such as taking respite from a challenging situation or seeking emotional support from others; cognitive strategies such as reframing a problem or reflecting on the cause and consequences of emotions; and physiological processes that are automatic but also modulated by behavioral and cognitive strategies (e.g., deep breathing; meditation). Parents’ ability to maintain a regulated emotional state and “be patient” was proposed to underlie effective parenting decades ago (Belsky, 1984). Parental emotion regulation is likely of critical importance during early years in parenthood, a period of life that involves tremendous neurobiological, hormonal, and behavioral changes as parents navigate the unique demands of caring for a young child (Rutherford, Wallace, Laurent, & Mayes, 2015). Given that parenting is demanding and occurs in the context of various other challenges (e.g., work stress, marital discord), it has been argued that parents who are generally able to regulate their emotions are likely to parent more effectively (Crandall, Deater-Deckard, & Riley, 2015; Rutherford et al., 2015). When parental emotions are dysregulated, it may be difficult to focus on child cues, to interpret them accurately, and to modulate behavioral responses so they are temperate and appropriate (Leerkes & Augustine, in press). Consistent with this view, parents who have difficulties regulating their emotions engage in less supportive responses when their children and adolescents are distressed (Morelen, Shaffer, & Suveg, 2016), are more rejecting and hostile and less warm when interacting with their adolescents (Saritas et al., 2013), and are observed to be less sensitive, responsive, and positive when interacting with their young children (Shaffer & Obradovic, 2017) compared to parents who are more skilled at regulating their emotions. Thus, mothers’ emotion regulation difficulties are likely to undermine sensitive maternal behavior.

Emotion regulation may also serve as a mediating pathway through which DRD4 affects maternal sensitivity. Prior research demonstrates that dopamine and dopaminergic-related genes play important roles in self-regulation (Cómbita, Voelker, Abundis-Gutierrez, Pozuelos, & Rueda, 2017) and emotional processes (Salgado-Pineda, Delaveau, Blin, & Nieoullon, 2005). Numerous studies have examined the role of DRD4 in self-regulation, largely focusing on cognitive and attentional self-regulation in clinical samples of children with attention deficit hyperactivity disorder and yielding inconsistent findings regarding main effects of DRD4 on self-regulation (e.g., Johnson et al., 2008; Langley et al., 2004). Beyond main effects, there is evidence of moderating effects of DRD4 on environmental influences on regulation such that carriers of the DRD4 long allele are at elevated risk for difficulties with self-regulation in the presence of other risk factors. For example, hours of non-maternal care were associated with lower inhibitory control and more inattention and impulsive behaviors only among preschoolers carrying the DRD4 long allele (Berry, McCartney, Petrill, Deater-Deckard, & Blair, 2014). Likewise, low responsive and supportive parenting was associated with less self-regulation only among African American adolescents carrying the DRD4 long allele (Cho, Kogan, & Brody, 2016). Finally, low birthweight was associated with poorer emotion regulation, indexed by lower resting respiratory sinus arrhythmia, among adult carriers of the DRD4 long allele (Savoy et al., 2017). Furthermore, genetic factors related to self-regulation (e.g., DRD4) also play an important role in caregiving behaviors (Bridgett, Burt, Edwards, & Deater-Deckard, 2015). Taken together with prior evidence of the effect of emotion regulation on maternal behavior, these findings suggest that DRD4 may influence maternal sensitivity indirectly via emotion regulation.

Adverse Life Events

Adverse life experiences in childhood and adolescence (e.g., childhood maltreatment, parental death, parental divorce) have been associated with a host of negative psychosocial outcomes (Low et al., 2012), which can persist into adulthood (Chapman et al., 2004) and may undermine parenting. Of particular relevance to the current study, adverse life events are associated with emotion dysregulation (Poole, Dobson, & Pusch, 2017) and lower maternal sensitivity (Gonzalez, Jenkins, Steiner, & Fleming, 2012). Research also suggests that emotion dysregulation serves as a mediating mechanism linking early adverse life experiences to negative psychosocial outcomes (Boyes, Hasking, & Martin, 2015).

In addition to main effects, adverse life events may interact with DRD4 in relation to emotion regulation and maternal sensitivity. Different conceptual perspectives regarding the nature of gene by environment (GXE) effects have been proposed in the literature. The diathesis-stress model posits that individuals carrying genetic “risk” predispositions are at increased vulnerability to environmental adversity, whereas the differential susceptibility perspective proposes that individuals with genetic “susceptibility/plasticity” predispositions are more susceptible to negative environments but are also more responsive to positive environments (Ellis, Boyce, Belsky, Bakermans-Kranenburg, & van Ijzendoorn, 2011). Our measure of adverse life events represented adverse but not positive environments, so we did not anticipate a pattern of differential susceptibility based on this variable. In addition, prior evidence of GXE effects involving adverse life events generally supported the diathesis stress model. For example, adverse life events were associated with negative outcomes such as higher risk for smoking (Pampel et al., 2015), attention deficit hyperactivity disorder (van der Meer et al., 2014), and greater risk for mood disorder (Mandelli et al., 2007) only among individuals carrying purported “risk” alleles (i.e., serotonin transporter linked polymorphic region (5-HTTLPR) short allele, val allele of the catechol-O-methyltransferase (COMT)). Notably, adult male carriers of the DRD4 long allele reported higher levels of emotional instability than others, but only when they experienced several adverse life events (Reiner & Spangler, 2011). Thus, individuals who have experienced adverse life events are at elevated risk for adverse psychosocial outcomes if they carry the DRD4 long allele. To our knowledge, no prior research has examined the interaction between DRD4 and adverse life events in relation to emotion regulation and parenting behaviors.

The Current Study

The goal of this study is to test the processes through which DRD4 may be linked with maternal sensitivity. We test a mediated moderation model where DRD4 interacts with adverse life events to predict maternal sensitivity directly and indirectly via emotion regulation. Based on prior research and the diathesis-stress model, we hypothesize that carriers of the DRD4 long allele who also experienced more adverse life events will have lower maternal sensitivity. These individuals will also have higher difficulties with emotion regulation, which in turn predict lower maternal sensitivity. We control for race to adjust for potential population stratification (Barnholtz-Sloan, McEvoy, Shriver, & Rebbeck, 2008), and adult attachment coherence and maternal education as prior research has shown that each is related to emotion regulation and/or maternal sensitivity (Leerkes et al., 2017). In addition, we control for infant’s DRD4 genotype to take into account potential infant evocative effects. Lastly, we test race as a moderator of proposed pathways given half of our participants are African-American and half are European-American and differences in genetic allele frequencies across groups can have implications for associations between genotypes and phenotypes (Haberstick et al., 2015).

Materials and Methods

Participants

Participants in the current study were 209 primiparous mothers and their infants (106 European American, 103 African American) from the southeastern United States drawn from a larger sample of 259 mothers initially recruited during the prenatal period. Mothers in the analytic sample ranged in age from 18 to 44 years (M = 25.5) at recruitment. Twenty-three percent had a high school diploma or less, 31% had attended but not completed college, and 46% had a 4-year college degree. The majority (59%) of mothers were married or living with their child’s father, 23% were in a relationship but not living with their child’s father, and 16% were single. Annual family income ranged from less than $2,000 to over $100,000 (median =$35,000). All participating infants were healthy; 52% were female. Initial participants who did not provide DNA (due to attrition, not refusal) were younger and less educated than mothers who provided DNA, but they did not differ on race, income, or adult attachment ratings.

Procedures

Participants were recruited through childbirth classes, breastfeeding classes, obstetric practices, and word-of-mouth and provided written consent. Relevant to the current study, mothers completed the Adult Attachment Interview and reported demographic information during a prenatal interview 6–8 weeks prior to their due date, and responded to the Difficulties in Emotion Regulation Scale when their infant was about 1 year old (M = 13.90 months). When the infant was about 2 years old (M = 27.32 months) mothers and infants participated in a videotaped laboratory observation used to assess maternal sensitivity, both provided DNA samples, and mothers responded to the Adverse Life Events Questionnaire. The research protocol was approved by the Institutional Review Board at the University of North Carolina at Greensboro (IRB No. 09–0035, Triad Child Study).

Maternal sensitivity tasks.

Five mother-child interaction tasks were observed during the 2-year lab visit. First was a transition period after arrival to the laboratory, during which mothers had electrodes placed on their chest to measure their heart rate and velcro strips placed on two adjacent fingers to measure skin conductance; children had electrodes place on their chest to measure their heart rate. Next was a free play task in which mothers were instructed to play with their child as they normally would for 7 minutes using toys that were already set up in the laboratory. This was followed by a clean-up task; the experimenter brought in two containers and instructed mothers to get their child to clean-up all the toys. Mothers were told they could handle the task however they wanted but they must involve their child. The task ended after 5 minutes or when all the toys were in the containers. Next was the locked box task. The experimenter allowed the child to choose between two attractive toys and play with that toy momentarily. Once children seemed interested, the experimenter placed the toy in a clear plastic box and locked it. Children were then given a set of keys that did not open the lock, and were prompted by the experimenter to open the lock and get their chosen toy for 4 minutes. The final task was the spider task, during which a stuffed spider attached to a remote control car moved around the room for 4 minutes. During the last 30 seconds the experimenter came back into the room and asked children to touch the spider 3 times. For the first minute of the locked box and spider tasks, mothers were instructed to sit on a couch in the room and remain neutral. For the remaining 3 minutes mothers could behave however they wanted except for opening the locked box or picking up or touching the spider.

DNA collection.

At the end of the 2-year visit, DNA was collected from mothers and children via buccal samples using Oragene™ (DNAgenotek, Ottawa, Ontario, Canada) materials. Mothers deposited 2ml of saliva into a vial (#OG-500). A swab format (#OG-575) was used with children. An experimenter used a small sponge-tipped swab to soak up 2ml saliva from the child’s mouth and deposited it into a vial using an attached funnel. A stabilizing lysis buffer was released when the vials were capped; vials were stored at 15–30°C prior to DNA extraction.

Measures

Emotion regulation.

The Difficulties in Emotion Regulation Scale (DERS; Gratz & Roemer, 2004) was utilized as a measure of emotion dysregulation in mothers. The DERS includes 36 items rated on a 5-point scale, 1 = “almost never” to 5 = “almost always.” Items assess: nonacceptance of reactions to distress (6 items); difficulty enacting goals when upset (5 items); impulse control difficulties when upset (6 items); limited awareness of emotions (6 items); limited regulation strategies (8 items); and poor clarity about emotions (5 items). A total DERS score was calculated as the average all items (Cronbach’s α = .82).

Adverse life events.

The Adverse Life Events Questionnaire (ALE) was adapted from the Life History Calendar (Caspi, Moffitt, Thornton, & Freedman, 1996). Mothers indicated whether or not they experienced 9 adverse events from birth to age 20: the death of an important caregiver (one item each for mother, father, and up to 2 additional caregivers), parental divorce, parent/caregiver had a serious psychiatric disorder (e.g., schizophrenia, bipolar disorder, post-traumatic stress disorder), parent/caregiver abused alcohol or drugs, physical abuse, and/or sexual abuse. A total ALE score was calculated by summing the 9 dichotomized items.

Maternal sensitivity.

Each 2-year interactive task was rated using Ainsworth’s 9-point sensitivity scale (1 = “highly insensitive” to 9 = “highly sensitive”) (Ainsworth et al., 1978). Fifteen percent of videos were double-coded to assess inter-rater reliability; intraclass correlations ranged from .74 to .93 across tasks with a mean of .87. An overall sensitivity composite measure was computed as the average of sensitivity ratings for all tasks (α = .89).

Genotyping.

As detailed in Leerkes et al. (2017), DNA was prepared at the Molecular/Cellular Biology Core Laboratory at the University of North Carolina at Greensboro using methods described by Oragene. Then, DNA was quantified by spectrophotometry (Nanodrop Spectrophotometer) and standardized to a working concentration of 20 ng/μl. Genotyping was performed at the Institute for Behavioral Genetics at the University of Colorado under the supervision of Dr. Andrew Smolen. Two individuals scored genotypes independently, and inconsistencies were reviewed and rerun when necessary.

Genotyping was performed for the 48 bp Variable Number Tandem Repeat (VNTR) polymorphism in the third exon of the D4 dopamine receptor gene (van Tol et al., 1992). The assay followed the method of Anchordoquy, McGeary, Liu, Krauter, and Smolen (2003). The primer sequences were forward: 5’-VIC-GCT CAT GCT GCT GCT CTA CTG GGC-3’; and reverse: 5’-CTG CGG GTC TGC GGT GGA GTC TGG-3’, which yield PCR products from 279 (2 repeat) to 663 (10 repeat) bp. Genotype groups were formed based on previous studies that grouped participants on the basis of carrying a long allele or a short allele, and molecular genetic work suggesting that the 7-repeat allele confers a functional difference in D4 receptors (e.g., Asghari et al., 1995; Kaitz et al., 2010). Accordingly, participants were classified to a DRD4 L (i.e., long; homozygous or heterozygous for at least one allele of ≥ 7 repeats; S/L or L/L) or as DRD4 S (i.e., short; both alleles < 7 repeats; S/S). Frequencies for DRD4 genotype groups in the whole sample and by participant race are presented in Table 1.

Table 1.

Genotype Frequencies and Hardy-Weinberg Equilibrium Tests

	Mothers			Infants
DRD4 genotypes	Whole sample	European American	African American	Whole sample	European American	African American
L/L	10 (4.8%)	5 (4.7%)	5 (4.9%)	14 (6.8%)	8 (7.6%)	6 (5.9%)
L/S	72 (34.4%)	32 (30.2%)	40 (38.8%)	61 (29.6%)	27 (25.7%)	34 (33.7%)
S/S	127 (60.8%)	69 (65.1%)	58 (56.3%)	131 (63.6%)	70 (66.7%)	61 (60.4%)
HWE test	.96	.61	.56	.07	.09	.75

Note. n = 209 for mothers DRD4, and n = 206 for infants’ DRD4. Genotype frequencies are provided by maternal race. p values from Hardy-Weinberg Equilibrium tests are presented.

Covariates.

Based on previous research (e.g., Leerkes et al., 2017), the analysis controlled for mother race, infant DRD4 genotype group, maternal education (1 = some high school to 7 = graduate degree) and the coherence of mind rating from in the Adult Attachment Interview (AAI; George, Kaplan, & Main, 1984–1996). The AAI is a semi-structured interview in which participants describe their early childhood relationships with their primary caregivers and their perception of the influences those experiences have had on them. AAI transcripts were coded using the AAI Scoring and Classification System (Main & Goldwyn, 1998). Coherence of mind is a summary rating of participants’ ability to describe early attachment experiences and their influence on current functioning, made on a scale from 1 = not at all coherent to 9 = very coherent). Interrater reliability for this scale, based on 50 cases, was intraclass correlation = .75.

Analysis

Preliminary analyses were performed to examine the frequencies of DRD4 genotype for mothers and infants. We conducted chi-square tests using SPSS version 24 to examine whether genotype frequencies varied across racial groups. Chi-square tests were also conducted to examine deviations from Hardy-Weinberg equilibrium (HWE). Descriptive statistics and correlations between study variables were also examined. Path analysis was conducted using MPlus version 7 (Muthén & Muthén 2012) to evaluate the effect of DRD4 on maternal sensitivity, the mediating role of emotion regulation, and the interaction effects between DRD4 and adverse life events. In the path model, DRD4 was specified as an exogenous variable that predicted emotion regulation and maternal sensitivity. Emotion regulation was specified as predicting maternal sensitivity. To examine the interaction effects between DRD4 and adverse life events, an interaction term was created by multiplying DRD4 and mean-centered adverse life events; the interaction term, along with adverse life events were specified as additional exogenous variables that predicted emotion regulation and maternal sensitivity. Maternal education and coherence of mind were specified as exogenous control variables linked to maternal sensitivity. Race was specified as a covariate associated with emotion regulation and maternal sensitivity to account for potential population stratification effects. Maternal education, coherence of mind, and race were allowed to be correlated with each other in the path model, because preliminary analysis indicated that they were significantly correlated. Infant’s DRD4 genotype was also included as a covariate associated with emotion regulation and maternal sensitivity to take into account potential child evocative effects. Infants’ DRD4 genotype was specified to be correlated with mothers’ DRD4 genotype.

Hypotheses related to indirect associations were evaluated using bias-corrected bootstrapped 95% confidence intervals (CI) (MacKinnon, Lockwood, & Williams 2004). CI not spanning zero indicate statistical significance of the indirect associations. Significant G×E interaction effects were probed by calculating simple slopes of adverse life events for each DRD4 genotype group. Additional analyses were conducted to evaluate whether mothers with the DRD4 long allele and those without the DRD4 long allele significantly differed in their emotion regulation difficulties and maternal sensitivity under conditions of low and high adverse life events. Regions of significance for DRD4 with respect to adverse life events were also calculated based on Roisman et al.’s (2012) recommendations using a supplemental online utility (available at http://www.yourpersonality.net/interaction), in order to better understand the nature of GXE effects. To examine possible differences between European-American and African-American mothers, multigroup analysis was conducted by removing race from the path model and then (a) comparing a model with all remaining paths constrained to equality with one that had all paths freely estimated across African-American and European-American women and (b) comparing each path coefficient across groups one by one, using chi-square difference tests. Missing data were accounted for using full information maximum likelihood estimation method.

Results

Preliminary Analysis

Genotype frequencies for mothers and infants for the whole sample and by maternal racial groups are presented in Table 1. Chi-square tests indicated that DRD4 genotype frequencies did not vary across racial groups for mothers (χ2 =1.44, df =2, p =0.49) or infants (χ2 =1.63, df=2, p =0.44). DRD4 genotype frequencies were in HWE for the whole sample and for each racial group for mothers and infants. Descriptive statistics and intercorrelations are presented in Table 2. Mothers who were European-American, more highly educated, and had higher coherence of mind were rated higher on maternal sensitivity. Mothers’ and infants’ DRD4 long allele were associated with lower maternal sensitivity. Thus including mothers’ race, education, and coherence of mind, and infants’ DRD4 genotype as covariates was justified.

Table 2.

Descriptive Statistics and Intercorrelations

	1	2	3	4	5	6	7	8
1. Race (EA)	--
2. Maternal education	.37**	--
3. Coherence of mind	.30**	.37**	--
4. Mother DRD4 long allele	−.09	−.12	−.04	--
5. Infant DRD4 long allele	−.07	.02	−.01	.40**	--
6. Adverse life events	−.10	−.24**	−.22**	−.02	.00	--
7. DERS 1Y	−.08	−.17*	−.09	.15*	.07	.04	--
8. Maternal sensitivity 2Y	.46**	.55**	.36**	−.26**	−.16*	−.10	−.26**	--
M or %	51%	3.94	5.39	39%	36%	1.25	1.86	6.13
SD	--	1.78	1.43	--	--	1.25	.50	1.48

Note. n ranges from 196 to 209. EA = European-American; DERS = difficulties with emotion regulation; Y = years.

^*P < 0.05;

^**P < 0.01.

Predicting Maternal Sensitivity

The path model predicting maternal sensitivity demonstrated excellent fit (χ² = 5.13, df = 2, p = .08; CFI = .98; RMSEA = .09, 90% CI [.00, .18]; SRMR = .02). Path coefficients are presented in Figure 1 and supplementary Table 1. Consistent with preliminary analysis and prior research, higher coherence of mind and maternal education were associated with higher maternal sensitivity. Consistent with prediction, mothers with a long DRD4 allele had more emotion regulation difficulties and engaged in lower maternal sensitivity. Difficulties with emotion regulation were associated with lower maternal sensitivity above and beyond the effects of covariates, albeit marginally so (B = −.30, CI [−.598, −.010], β = −.10, p = .065). Furthermore, results indicated that the indirect effect of DRD4 on maternal sensitivity via emotion regulation was significant (95% CI [−.154, −.003], B = −.05, SE = .04, β = −.02).

Figure 1.

Effects of DRD4 and Adverse Life Events on Maternal Sensitivity via Emotion Regulation

Note. ALE = Adverse life events. Standardized coefficients are presented. Race was coded 1 = European American, 0 = African American. Statistically significant paths related to the effect of DRD4 are bolded. *p < .05, **p < .01, ***p < .001, ^ap = .066, 95% CI = [.02, .21], ^bp = .065, 95% CI = [−.60, −.01]. N = 209.

In addition to these main effects, mothers’ DRD4 genotype moderated the effects of adverse life events on emotion regulation (B = .11, CI [.019, .205], β = 0.15, p = .066, marginally significant) and maternal sensitivity (B = −.54, CI [−.775, −.321], β = 0.25, p < .001). The pattern of interaction effects between DRD4 and adverse life events for emotion regulation is illustrated in Figure 2 Panel A. Simple slope analysis indicated that experience of adverse life events was associated with more emotion regulation difficulties for mothers with the DRD4 long allele (B = .09, SE = .05, β = .23, p = .07, marginally significant) but not for mothers without the DRD4 long allele (B = −.02, SE = .03, β = −.05, p = .53). Among mothers who experienced low adverse life events (−1 SD), there was no significant difference in their emotion regulation difficulties as a function of DRD4 genotype (B = .02, SE = .11, β = .02, p = .88). However, among mothers who experienced high adverse life events (+1 SD), those carrying the DRD4 long allele had significantly more difficulties with emotion regulation (B = .30, SE = .12, β = .29, p < .01). The lines for each DRD4 genotype group crossed at the mean-centered adverse life events value of −1.39 (−1.11 SD), which corresponds to a nonexistent raw value below 0. The regions of significance indicated that mothers carrying the DRD4 long allele had more difficulties with emotion regulation when they were above a value of −.04 on the mean-centered adverse life events variable (a raw score of 1.21 events). However, given the location of the crossover point, the DRD4 genotype groups did not differ at any observable low level of adverse life events.

Figure 2.

Interaction between DRD4 and Adverse Life Events in Relation to Emotion Regulation and Maternal Sensitivity

Note. Scale for adverse life events is centered at the mean of 1.25.

The pattern of interaction effects for maternal sensitivity is illustrated in Figure 2 Panel B. Simple slope analysis indicated that experience of adverse life events was associated with lower maternal sensitivity for mothers with the DRD4 long allele (B = −.31, SE = .12, β = −.27, p < .01) and higher maternal sensitivity for mothers without the DRD4 long allele (B = .23, SE = .08, β = .19, p < .01). Among mothers who experienced low adverse life events (−1 SD), maternal sensitivity did not differ as a function of DRD4 genotype (B = .28, SE = .23, β = .10, p = .21). However, among mothers who experienced high adverse life events (+1 SD), those with the DRD4 long allele engaged in lower maternal sensitivity than those without the DRD4 long allele (B = −1.06, SE = .26, β = −.37 p < .001). The lines for each DRD4 genotype group crossed at a value of −.72 (−.58 SD) on mean-centered adverse life events, suggesting that the left side of the crossover interaction affected a relatively small proportion (23%) of individuals in the sample. The regions of significance indicated that mothers carrying the DRD4 long allele engaged in lower maternal sensitivity when they were above a value of −.05 on the meancentered adverse life events variable (a raw score of 1.20 events). Despite the crossover pattern, the DRD4 long allele group did not engage in higher maternal sensitivity at any observable low level of adverse life events (this would occur below a nonexistent value of −1.87, or a raw score of −.62 adverse life events). Thus, both interactions fit the diathesis-stress pattern and not the differential susceptibility based on the procedures recommended by Roisman et al. (2012).

The indirect effect of DRD4 by adverse life events interaction on maternal sensitivity via emotion regulation was significant (95% CI [−.106, −.004], B = −.04, SE = .03, β = −.02). Thus, in addition to the direct association, the DRD4 long allele, when coupled with adverse life events, was linked with mothers’ greater difficulties in emotion regulation, which in turn predicted lower sensitivity. There were no statistically significant associations between infants’ DRD4 genotype and mothers’ emotion regulation and maternal sensitivity in the path model. Results of the multigroup analysis indicated that overall, the patterns of path coefficients did not differ significantly across racial groups ($\Delta \chi 2=12.37,\Delta df=11,p=0.34$). Follow-up analyses comparing path coefficients one by one across racial groups indicated only one significant difference involving a covariate. That is, coherence of mind was more strongly associated with maternal sensitivity for African American mothers than for European American mothers. Path coefficients for each racial group are presented in supplementary Table 2, along with the chi-square difference tests for each path coefficients across racial groups.

Discussion

The goal of this study was to examine the extent to which DRD4 predicts maternal sensitivity directly and indirectly via emotion regulation both as a main effect and in conjunction with adverse life events. Consistent with hypotheses, the results showed that mothers who carried the long allele of DRD4 and experienced more adverse life events were less sensitive in interactions with their infants. These mothers were also more likely to have difficulties with emotion regulation, which in turn predicted lower maternal sensitivity. These direct and indirect effects were significant above and beyond the effects of maternal education, coherence of mind, race, and infants’ DRD4 genotype.

Consistent with prediction and the view that dopamine related genes play an important role in parenting (Numan, 2010), the DRD4 long allele was associated with lower maternal sensitivity. In addition, DRD4 moderated the effects of adverse life events on maternal sensitivity, such that adverse life events were associated with lower maternal sensitivity among mothers carrying the DRD4 long allele. This finding is consistent with prior research suggesting that DRD4 long allele coupled with environmental risk factors are associated with poor psychosocial outcomes (Mandelli et al., 2007; Pampel et al., 2015) including compromised parenting (Fortuna et al., 2011; Kaitz et al., 2010; Mileva-Seitz et al., 2016; van IJzendoorn et al., 2008) and provides support for the diathesis-stress hypothesis (Ellis et al., 2011) for carriers of the DRD4 long allele. Interestingly, our results also indicated that adverse life events were associated with higher maternal sensitivity among mothers without the DRD4 long allele. Prior research suggests that mothers who experienced maladaptive parenting in childhood but went on to experience positive intervening experiences or relationships may become particularly attuned to their children’s emotional needs resulting in enhanced sensitivity, likely because they know firsthand how badly it feels to experience poor parenting or negative life events and have acquired skills in the mean time that help them respond sensitively to their children’s needs (Leerkes & Crockenberg, 2006). Perhaps mothers with the DRD4 S/S genotype are more likely to elicit support from others, which leads to high levels of sensitivity despite the experience of adverse life events. Taken together, these findings provide support for the importance of parenting intervention efforts targeting mothers with a history of early adverse life events; these intervention efforts may be most beneficial to individuals who carry the DRD4 long allele.

Our results revealed that the interaction effect between DRD4 and adverse life events on maternal sensitivity was in part indirect via emotion regulation. This finding is consistent with the wealth of literature that demonstrates the important role of emotion regulation in parenting behavior (Crandall et al., 2015; Leerkes & Augustine, in press; Rutherford et al., 2015). The interaction effect between DRD4 and adverse life events in relation to emotion regulation indicated that adverse life events was associated with greater emotion regulation difficulties only among mothers carrying the DRD4 long allele; mothers with the long allele of DRD4 who experienced more adverse life events had more difficulties with emotion regulation. These findings are consistent with the diathesis-stress hypothesis (Ellis et al., 2011), as well as prior findings that carriers of the DRD4 long alleles are at elevated risk for difficulties with self-regulation under adverse conditions (Berry et al., 2013; Savoy et al., 2017). We note, however, that prior research on the role of DRD4 in self-regulation has primarily focused on cognitive, attentional, and behavioral aspects of self-regulation, such as inhibitory control, executive functioning, and impulsivity (Berry et al., 2014; Cho et al., 2016). To our knowledge, this is among the first studies to show that DRD4 long allele is associated with poorer emotion regulation. These findings suggest that DRD4 long allele could be broadly associated with both top-down and bottom-up regulation-related processes (Bridgett et al., 2015; Nigg, 2017). It is also possible that other aspects of self-regulation, such as impulsivity and executive functioning may serve as intermediary processes linking DRD4 to emotion regulation difficulties and maternal behavior (Wante, Mezulis, Van Beveren, & Braet, 2017). Future research is warranted to examine this possibility. This indirect pathway is also consistent with the view that genes involved in the dopamine system are related to self-regulation and emotional processes (Cómbita et al., 2017; Salgado-Pineda et al., 2005) and play an important role in maternal behavior (Bridgett et al., 2015; Numan, 2010). It appears that mothers with the DRD4 long allele have greater difficulty regulating their emotions, which undermines their ability to respond sensitively to their infants’ needs. Our findings demonstrate that emotion regulation serves as an important mediating mechanism for genetic influences on maternal behavior, suggesting the promise of targeting emotion regulation deficits in prevention and intervention efforts aimed at promoting sensitive maternal behaviors, particularly among those genetically and/or environmentally at risk. For example, recent research has shown that mindfulness-based parenting intervention is effective in improving parenting quality among mothers, particularly among those who experienced higher childhood adversity (Gannon, Mackenzie, Kaltenbach, & Abatemarco, 2017).

Our findings need to be interpreted in light of several limitations. First, although our sample is relatively large for developmental studies with extensive observational measures, it is quite small for molecular genetic research that typically requires large sample sizes to detect small effects of specific genes on complex phenotypes such as maternal sensitivity, limiting our statistical power in analyses. Second, samples with homogeneous genetic ancestry are generally preferred in genetic research given concerns about potential bias due to population stratification. Our sample is not ideal in this regard as it included mothers who self-reported as European American or African American. However, that DRD4 allele frequencies in our sample did not significantly differ across race, and that we included race as a covariate, reduce concern about population stratification in this study. In addition, our sample allowed us to formally test race as a moderator of genetic effects. Our findings indicated that despite racial differences in mean levels of maternal sensitivity, pathways linking DRD4 to maternal sensitivity did not differ across racial groups. Third, mothers reported their experience of adverse life events retrospectively when their infants were about 2 years old. Thus, the assessment of adverse life events may be subjected to recall bias and potentially affect its associations with other constructs in this study. Prior evidence of consistency in mothers’ retrospective self-report of adverse childhood experiences (Cammack et al., 2016), however, reduces this concern.

Despite these limitations, this study has several notable strengths, including the longitudinal design, careful observation of maternal sensitivity, and inclusion of several important covariates to rule out potential bias and competing explanations. To our knowledge, we are the first to present evidence that emotion regulation serves as a mediating mechanism linking DRD4 genotype to maternal behaviors, although numerous studies have implicated genetic underpinnings for emotion regulation (Hawn, Overstreet, Stewart, & Amstadter, 2015) and the important role of emotion regulation in parenting (Shaffer & Obradovic, 2017). In the future, researchers should examine whether emotion regulation plays a role in linking other emotion-related genotypes to parenting behaviors. It is notable that DRD4, as well as its interaction with adverse life events, remained significantly associated with maternal sensitivity after emotion regulation and covariates were taken into account, suggesting the need for future studies to explore other mediating mechanisms for the effects of DRD4 and other genetic factors on parenting (e.g., executive function, mental health). This study is also the first to show interaction effects between DRD4 and adverse life events in predicting emotion regulation and maternal sensitivity, adding to the literature that suggests the important role of adverse life experiences in moderating genetic effects on psychosocial outcomes. The current findings are robust given that we controlled for maternal education and coherence of mind, both of which were highly correlated with maternal sensitivity, and infants’ DRD4 genotype to rule out the possibility of evocative gene-environment correlations.

In conclusion, our results show that the DRD4 long genotype, when coupled with the experience of adverse life events, is associated with less sensitive maternal behavior, both directly and indirectly through mothers’ compromised emotion regulation skills. These findings suggest that parenting is in part controlled by biological systems related to emotion processes, which is one way in which genes are linked with individual differences in maternal behavior. Our findings also further emphasize the importance of gene-environment interaction effects in relation to parenting. Future work is needed to replicate and extend the current findings.