Social neuroscience and mechanisms of risk for mental disorders.

Abstract

In their thoughtful paper, Cacioppo et al 1 emphasize the role of social neuroscience as an integrative meeting ground of a variety of methodological approaches, “neural, hormonal, cellular, molecular and genetic”, and call for a truly “interdisciplinary” approach. We fully endorse this viewpoint and would like to emphasize the importance to apply this strategy at another important, but underexamined juncture between disciplines: the connection between psychiatric epidemiology and psychiatric neuroscience. Specifically, we propose that there is growing evidence that an important facet of risk for severe mental illness can be understood as altered neural processing of specifically social stimuli, and that this may have therapeutic and preventive applications in the future. It is well known that impoverished or abusive social contexts contribute to the risk for mental illness 2. Recent work also supports the view that epidemiologically validated risk factors such as urban living and upbringing and migration have a social component, as proposed by the social defeat hypothesis 3. Specifically, work from our laboratory has suggested a specific impact of social stress on activation in a perigenual cingulate-amygdalar circuit in healthy populations exposed to urban living and up-bringing 4 and migration. Since previous work, as highlighted by Cacioppo et al, has shown that the same circuit is impacted by (serotonergic) candidate genes that show gene-environment interactions 5,6, we have recently proposed that this circuit, the limbic regions (such as ventral striatum and hippocampus) linked to it, and regions of evolutionarily more recent medial and lateral prefrontal cortex that regulate it, may be a core convergence pathway for risk for mental disorders arising through social stressors 7. In fact, given preclinical evidence, it also appears possible that socially adverse environments early in brain development are causal for the dysfunction specific to social stress observed in adults exposed to risk. Since this circuit and those linked to it are found abnormal and related to symptomatology in patients (as demonstrated by the three clinical conditions discussed by Cacioppo et al), they may also be important for treatment. Seen this way, the social environment remains relevant throughout the trajectory of mental illness, as it is configured, unfolds and gets addressed. Clarifying these interacting genetic, neural and environmental mechanisms further is thus important for therapy, and possibly even prevention. This suggests an ambitious, but actionable work program spanning several layers of description. Evidence for mechanisms supporting gene-environment interaction on the systems level should be supplemented by research to clarify its cellular basis in humans. This is challenging as the primary tissue of interest is not available in man, but methods such as genome wide methylation analysis in peripheral tissues may help to identify some of the genes and pathways modified by social environmental stress 8. A further level of description may be added through the study of human neurons derived through induced pluripotent stem cell technology 9, although most epigenetic programming is removed in the process of generating these cells. Gene-environment interaction effects in brain should be confirmed directly by combining social exposures with imaging in genotyped subjects. Based on these data, the neural underpinnings of subcomponents of the social environment that could contribute to the epidemiological evidence we have highlighted need to be analyzed further (examining effects such as social defeat, discrimination, and loneliness – as emphasized in Cacioppo et al's paper). This will require close interactions with the social sciences in providing scales that measure these facets of the social landscape with precision. In addition to laboratory paradigms, this will also require field studies that combine neuroimaging and biomarker ascertainment with experience-based assessment, mobile neuropsychological testing and tracking of subjects in spatially and socially well defined real-life contexts 10. Based on epidemiological evidence that time periods such as the perinatal period, early childhood and adolescence are especially vulnerable for these environmental exposures, an account must be developed on which neurobiological changes underlie vulnerability across the lifespan. This requires longitudinal and ecological human studies that combine neuroscience approaches with the power of epidemiological methods, such as the European IMAGEN study 11. We have recently begun a study that uses these methods combined with imaging and urban geography to highlight aspects of the urban social environment. A better understanding of these neural circuits will also enable back-translation into a generation of more specific animal models that can be used in early drug discovery 12, and could enable the identification of molecular targets for social dysfunction through cellular and (epi-)genetic approaches. Further into therapy development, this neuroscience information can also constrain pharmacological and psychotherapeutic strategies targeting the identified circuits. For example, the prosocial neuropeptides oxytocin and vasopressin can be shown to act on precisely the core perigenual cingulate-amygdalar regulatory circuit for gene-environment interaction, and common genetic variants in the brain receptors for these prosocial neuropeptides modulate the activity and even structure of these regions in humans 13. As we have recently discussed for the specific case of oxytocin in combination with behavior therapy 13, this opens up a mechanism-guided approach for interfacing the usually separate domains of biological and psychological therapies, as predicted at the conclusion of Cacioppo et al's paper. As emphasized in that paper, a truly interdisciplinary approach to social neuroscience in psychiatry has therefore much to offer for people suffering from mental illness, their clinicians, and science.