Volume 123, January 2021, 105026
aResearch Center for Child Mental Development, University of Fukui, Fukui, Japan
bJapan Society for the Promotion of Science, Tokyo, Japan
cDepartment of Child and Adolescent Psychological Medicine, University of Fukui Hospital, Fukui, Japan
dDivision of Developmental Higher Brain Functions, United Graduate School of Child Development, University of Fukui, Fukui, Japan
eBiomedical Imaging Research Center, University of Fukui, Fukui, Japan
Received 28 July 2020, Revised 16 September 2020, Accepted 16 October 2020, Available online 21 October 2020.
•Oxytocin gene (OXT) DNA methylation predicts maternal empathy (Personal Distress).
•OXT methylation is inversely correlated with the GMV of the right ITG.
•The indirect effect of ITG gray matter volumes was not significant.
Mother’s empathy is an important ability for parenting behavior. Many studies have confirmed that oxytocin affects empathy, but the epigenetic background of oxytocin in maternal empathy has not yet been examined. This study examined the relationship between the oxytocin gene methylation and empathy in mothers of children in early childhood. Additionally, in order to understand a comprehensive mechanism, we also investigated changes in gray matter volume as a function of oxytocin gene methylation and empathy. The Interpersonal Reactivity Index was used to assess cognitive and affective dimensions of empathy of the 57 mothers who participated in this study. Genetic data were collected via saliva samples and analyzed to quantify DNA methylation of oxytocin gene. Gray matter volumes were investigated by means of voxel-based morphometry across the whole brain. A positive correlation was found between oxytocin gene methylation and Personal Distress, an aspect of affective empathy. Moreover, we found an inverse correlation between oxytocin gene methylation and the volume of the right inferior temporal gyrus. In a relationship with oxytocin gene methylation and empathy, the indirect effect of the inferior temporal gyrus gray matter volumes was not significant. Our findings provide empirical evidence for an epigenetic mechanism linking the oxytocin gene, structural variation of brain, and empathy in mothers. Taken together, the current imaging epigenetic findings shed new light on the understanding of the epigenetic basis of oxytocin and parental empathy.
Many psychophysiological processes and abilities are involved in the formation of parental behaviors (e.g. Feldman, 2016). Among them, the capacity for empathy is a core component of parenting, especially during early childhood (Decety and Svetlova, 2012). Empathy is the ability to perceive, and understand others’ affective states, and motivates helping behaviors (Batson et al., 1987; Decety and Svetlova, 2012), and enables caregivers to be flexible in response to the needs of children (Rilling, 2013).
Empathy is a complex and multidimensional ability composed of affective and cognitive aspects (Shamay-Tsoory et al., 2009). Affective empathy implicates vicarious affect processes, in which the empathizer internally experiences feelings on behalf of the observed person. Affective empathy includes unconscious and quick reactions such as sharing of emotions and emotional contagion by observing the arousal, mood, and facial expressions of the target. The ability of affective empathy begins to emerge during infancy (Yu and Chou, 2018). On the other hand, cognitive empathy entails the ability to intellectually understand emotional states of others without undergoing emotional contagion. Cognitive empathy requires higher cognitive functions than affective empathy and continues to develop in adolescence. Inferior frontal gyrus, inferior parietal lobe, anterior cingulate cortex, and anterior insula are the key regions of affective empathy, whereas ventromedial prefrontal cortex, dorsomedial prefrontal cortex, temporoparietal junction, and medial temporal lobe are the key regions of cognitive empathy (Shamay-Tsoory, 2011). Affective empathy and cognitive empathy are considered as two-pathway models, both of which are bi-directionally related and ultimately motivate prosocial behavior, such as helping others in difficult situations (Shamay-Tsoory, 2011; Yu and Chou, 2018). Both, affective and cognitive empathy have a link to psychopathology and personality disorders. For example, narcissism and machiavellianism are associated with both affective and cognitive empathy, while psychopathy has been shown to be particularly associated with lack of affective empathy (Pajevic et al., 2018; Pfabigan et al., 2015). Both aspects of empathy are strongly related to parental behavior. Parental affective empathy is positively correlated with the degree of supportive parenting as rated by parents and children, respectively (Soenens et al., 2007) and parents with higher affective empathy are highly motivated to provide care in response to infant cries (Lin and McFatter, 2012). Similarly, parents with higher cognitive empathy are less likely to feel frustration to infant cries (Barr et al., 2014). It has also been shown that the brain networks involved in empathy are related to parenting behavior. While affective empathy-related regions such as anterior cingulate cortex, anterior insula, and inferior frontal gyrus are activated by facial and vocal cues of infant, cognitive empathy-related brain regions such as dorsomedial prefrontal cortex and dorsolateral prefrontal cortex suppress and regulate the bottom-up affective empathetic response (Kim et al., 2016; Rilling, 2013). It is also considered that these empathy networks may be inadequate in depressed parents. Parents with high depressive symptoms reduce subjective empathic response to their infants (Esposito et al., 2017; Salo et al., 2020), and depressive mothers show less brain activation in anterior cingulate cortex, which is part of the affective empathy, during listening to the infant cry (Laurent and Ablow, 2012). Additionally, both affective and cognitive empathy predict the risk of child maltreatment (De Paúl et al., 2008). Furthermore, parental empathy influences children’s psychological development and psychopathological symptoms (Psychogiou et al., 2008). Parental empathy determines the quality of parenting behavior and has a robust effect on the child’s psychological development. It is important to clarify the factors that influence parental empathy in order to understand the mechanisms of parental behavior.
A growing number of empirical studies suggests that oxytocin has an important role as a neuroendocrine basis for empathy and parenting behavior (Feldman, 2016). Several studies that have quantified oxytocin levels from a peripheral sample have consistently found that parental oxytocin levels positively predict empathetic and sensitive parenting (Feldman et al., 2011; Gordon et al., 2010). In addition, intranasal oxytocin administration has been shown to increase emotional, but not cognitive, empathy in response to emotional stimulus (Hurlemann et al., 2010), and brain activation in the anterior insula and the inferior frontal cortex, which are parts of the empathy network, as adult female participants listen to infant cries (Riem et al., 2011). Moreover, some behavioral genetics studies have demonstrated that individuals with homozygous for the G allele of the oxytocin receptor gene (OXTR) polymorphism (rs53576) exhibited higher dispositional empathy (Rodrigues et al., 2009) and sensitive parenting (Bakermans-Kranenburg and Ijzendoorn, 2008; Klahr et al., 2015). Although these findings suggest that oxytocin and its genetic basis are closely related to empathy and parenting behavior, the mechanisms leading from oxytocin related genes to maternal phenotypes have not been fully investigated. In this study, we explored the process of the oxytocin gene (OXT) on the maternal phenotypes using an epigenetic approach.
DNA methylation, the addition of methyl residues (CH3) to the 5-carbon position of cytosine, typically occurs at the so-called CpG site. DNA methylation attenuates the binding of transcription elements, and thus, higher DNA methylation typically corresponds with silencing of gene activity. A growing body of evidence suggests that the epigenetic variability of OXTR is related to inter-individual differences in socio-behavioral phenotypes (Fujisawa et al., 2019; Kimmel et al., 2016). However, few studies have addressed OXT methylation. In a non-parental population, a previous study investigated the association between the OXT methylation and trait personality, behavior, and brain structure and function (Haas et al., 2016). They found that greater OXT methylation is associated with higher attachment anxiety and lower ability to recognize emotional facial expression. Additionally, they found that greater OXT methylation was associated with reduced gray matter volume (GMV) of right fusiform gyrus and reduced neural activity within brain regions including the superior temporal sulcus, fusiform gyrus, and inferior frontal gyrus during an empathizing task. A recent paper provides evidence that methylation at one CpG site (cg16887334) on promoter region of OXT is positively associated with intrusive parenting toward children (Toepfer et al., 2019). Intrusive parenting is observable in parents with high trait Personal Distress (Ewing et al., 2019), which is one component of affective empathy. To elucidate a more detailed mechanism by which OXT methylation influences trait empathy in mothers, the first aim of this study was to quantify the maternal OXT methylation and examine its relationship to maternal trait empathy.
The second aim of this study was to clarify the role of brain structure in the relationship between OXT methylation and empathy. Several studies have shown that DNA methylation is robustly associated with brain structure and/or function, and imaging epigenetics is a useful approach for a comprehensive understanding of the mechanisms between gene, brain, and phenotypes (Wheater et al., 2020). Some previous studies have considered that brain structure is an endophenotpye and that DNA methylation predicts the brain structure (e.g. Fujisawa et al., 2019). However, it has recently been pointed out that reverse causation from brain structure to DNA methylation can be assumed (Walton et al., 2020). For example, it has been shown that traumatic brain injury alters later DNA methylation of multiple genes in the hippocampus and leukocytes in a rodent study (Meng et al., 2017). In other words, a model in which brain structure is the independent variable and DNA methylation is a mediator should also be considered. In this study, we compared both models: 1) DNA methylation mediates the association between brain structure and phenotype, and 2) brain structure mediates the association between DNA methylation and phenotype.
The current study aimed to test whether OXT methylation is related to maternal brain structure and trait empathy. To this end, we quantified maternal OXT methylation from saliva. Structural neuroimaging data were collected using magnetic resonance imaging (MRI) and a voxel-based morphometry (VBM) approach, and trait empathy was measured using Interpersonal Reactivity Index (Davis, 1983). We analyzed the association between OXT methylation, empathy, and GMV. We hypothesized that OXT methylation would predict trait empathy. In particular, we hypothesized that OXT methylation would be associated with affective empathy, but not cognitive empathy (Shamay-Tsoory, 2011).