At the British Neuroscience Association (BNA)’s Festival of Neuroscience in April 2019, we were lucky enough to sit down with some influential neuroscientists to discuss their work. We’ve assembled these transcripts into our BNA Interview Series. Here we interview Professor Michael Meaney, head of the Translational Neurosciences Program at the Singapore Institute of Clinical Sciences. Professor Meaney’s work explores the intersection between stress, maternal care and gene expression.
Ruairi Mackenzie (RM): What do we know about the genetic basis of stress?
Michael Meaney (MM): Actually, not so much and I’ll tell you one of the reasons why. We have studies for the most part about environmental conditions that influence stress reactivity, the way people respond to stress, rendering people more reactive to stressors. Then we have genetic studies that are, for the most part, linked to pathological conditions but there are a few out there that are starting to creep beyond pathology, because one of the big trends in genetics these days is moving research a little way from disorders and starting to look more at traits that associate with disorders. For the sake of your question, one of the ones that’s really interesting is a trait called neuroticism. So, that’s a term we all use right?
RM: I have been called it before.
MM: Neuroticism is pretty much what people understand it to be. It refers to individuals who are much more reactive in general, but clearly much more reactive to stressors in their lives. So we are starting to move into the zone where we can study populations rather than individuals who we would call cases, people will specific disorders. That’s nice because we know that there are certain traits that render people more likely to develop diseases under conditions of adversity and neuroticism is a perfect example of that. Of course what you can do is you can take the genetics of neuroticism and say, well okay what are the individual genes here? But neuroticism is a complex trait, it’s not going to be associated with any single variant etc. For the most part it’s going to be defined by a whole series of genes that come together and form an operative network in certain areas of the brain.
So we can take that information and say, what networks are formed by those genes? Are those networks, for example, that influence certain neurotransmitters, are they networks that influence the growth or connectivity between certain structures in the brain. We can actually use the genetic information to inform us a little bit about the biology. Now, what we get from that are ideas, thoughts, hypotheses. So they need to be explored in a more detailed way, but I think more and more people are starting to say we can use genetic data to try to understand the biology and there are some key questions there.
One of the things about neuroticism that makes it so interesting to think about is that we know that exposure to an endangering environmental event, a serious chronic stressor in adulthood or in early life, that information alone is not enough to predict a particular health outcome. So individuals exposed even to severe forms of childhood maltreatment are at greater risk, but many of those individuals do well. They’re healthy and productive over the course of their lives. So there is this massive difference between resilient and susceptible individuals. Neuroticism is one of the avenues that we can use to say, well how does it develop and how do some people become more neurotic, more reactive to stress and others less so given a common exposure to early life adversity?
Additionally, we have got a wealth of studies done over 70 or 80 years, largely with rodent models, understanding their neurobiology with stress and so now we can start to put those together. Then you get neuroimaging data, mostly from humans as well and they start to converge on particular networks, so you start to find patterns of where connectivity in the brain can influence the way people respond to stress. So I think in terms of the biology of stress, it’s not going to be any single source of data that’s going to illuminate completely our ability to understand these individual differences, it’s going to be a collection of framing all of those particular types of data, but genetics is suddenly becoming part of that equation.
RM: Which factors most significantly influence cognitive outcomes in childhood development?
MM: Well there are two in my opinion. I’m heavily influenced by the idea that the primary influence in a developing organism is always going to be the maternal organism. There is a simple evolutionary reason for that. The mother is heavily invested in the success of her offspring. She is, by design, the most prominent source of influence, she is the largest source of influence over that period of time when the brain is maturing, in utero and in early postnatal care in humans. She is, I wouldn’t say the exclusive influence, but she is certainly the most proximal influence in development. So that’s one of the reasons that we focused on maternal care.
The idea was really very simple, it was that there is individual variation in the way genes are expressed and particularly in brain regions that regulate our responses to stress. Could it be that one of the features that is influencing that is the maternal organism? One of the ways to understand this is not to think of the maternal effect as one that necessarily is good or bad. Parents should be invested in producing kids that are best adapted to the circumstances in which they’re going to live and, in some cases, that might mean producing a child that is somewhat more fearful and anxious. A child growing up in a very severely challenging environment might be better adapted if they express a little more fearful and wariness, as it were. We never went in with the idea that these effects were necessarily good or bad, but rather that they were producing variation. I think that helped us to understand across the continuum how it was that parental influence could in fact exert long-term changes in gene expression and how those changes are necessarily producing stable individual differences that are probably designed to best fit the challenges that the offspring is going to face in adulthood.
We moved from that perspective and then asked, “Can we translate this into the woman’s situation?” So our studies have mostly looked either at specific outcomes, cognitive or emotional outcomes and of course brain imaging as well. We’ve seen, unsurprisingly in humans, evidence for these effects even at the time of birth. We can see changes, not simply in structure but in connectivity that are associated with maternal emotional wellbeing during pregnancy. So these effects, I think, are prominent. They are very early in life, they set the stage for potential trajectories and I suppose the last part to your question is if you are asking about cognitive and emotional outcomes, is that one of the really striking things, particularly when you start looking at the existing human data is the degree of gender differences.
There is a surprising level of support for the idea that early life adversity, either in the prenatal or early postnatal phase, and particularly in the prenatal phase perhaps, is associated more with cognitive outcomes in boys and more socioemotional outcomes in girls. Some of the strongest data comes from British cohort studies, such as the Avon Longitudinal Study of Parents and Children. There are reports from there and other large cohorts depicting these particular gender outcomes. One of the factors we need to look at with a far more serious lens when we examine outcomes across a range of children exposed to adversity is to factor gender into those outcomes. Why this would be the case I think is mystifying, but the evidence is actually surprisingly strong.
RM: Is this an effect we see inherited by the children of these boys and girls?
MM: In general yes. It’s not hard to see evidence of inter-generational effects, but they’re not strong. So statistically, for example, the children of parents who have known childhood maltreatment are more likely to experience adversity themselves, but the variation is enormous and there is no way that intergenerational transmission is even close to be inevitable. So the question is why not? What happens in between the cup and lip to produce that variability? If you look, for example, all the way from studies that look at if a child was seriously abused in early life, is that a risk factor for them emerging as abusing parents? Yes, it is, but the vast majority of parents who were sexually abused are not abusive parents. So somehow that chain is being broken at some particular point. I think it’s really critical to understand the processes by which that occurs and the stage of life it happens at. You could investigate this with regards to almost any trait; even obesity for our colleagues in metabolic disorders. These are not inevitable. There is a pattern of transmission but it’s not nearly as strong as we perhaps assume it to be.
RM: Have there been studies of the influence of care in societies where the burden of care might be moved between parents or is it reasonably constant across societies?
MM: No. We think it’s really interesting and it’s a dynamic. The psychologists speak, for example, of a child as being an organism that’s remarkably good at seeking out nurturance from whatever source may be there. We see in Asia that many children are raised by grandparents or in a more communal environment. In a cohort that we run in Singapore, I believe that in 50% of the cases the mother is not the primary care giver in terms of the person who spends the most amount of time with the child. That wouldn’t be true up to one year of life, but beyond one year it’s very often that children are raised by others, particularly grandparents. People are living in larger, extended families.
The irony there is that these care structures may come closer to the original organization of the family who lived in kin groups. In fact, it may be that we in the west are somehow the outliers having created these nuclear families in which parenting and child care is so closely linked to one or two individuals whereas in other societies it can be far more communal. That leads you to some interesting outcomes. For example, a colleague of ours in Singapore was looking at the association between a common feature of early maternal care, which is sensitivity – to what extent is the mother attuned to the cues of the infant? She had a predicted a relationship between development in the prefrontal cortex and maternal sensitivity. She’s seen a rather weak relationship, which surprised her, so she took this large sample she had and she split it and for those whom the mother was the primary care giver, she saw a very strong relationship between maternal sensitivity and prefrontal development, but for those children for whom the mother wasn’t the primary caregiver, it was absent. So I think in these particular instances, you’re seeing more diversity within the child care environment and I think it’s an excellent question to ask, does this in some ways buffer the child by having potentially more positive sources of resilience in their lives? It’s a question well worth noting.
Michael Meaney was speaking to Ruairi J Mackenzie, Science Writer for Technology Networks