In this Papers Podcast, Dr. Charlotte Cecil and Dr. Mannan Luo discuss their co-authored JCPP paper ‘DNA methylation at birth and lateral ventricular volume in childhood: a neuroimaging epigenetics study’ (https://doi.org/10.1111/jcpp.13866).
There is an overview of the paper, methodology, key findings, and implications for practice.
Discussion points include:
- Definitions of ‘lateral ventricular volume enlargement’ and ‘epigenetics process of DNA methylation’.
- The advantage of capturing DNA methylation at different life stages.
- How this approach will further knowledge in this field.
- Implications of the findings for researchers, and clinicians.
- Potential follow-up studies.
In this series, we speak to authors of papers published in one of ACAMH’s three journals. These are The Journal of Child Psychology and Psychiatry (JCPP); The Child and Adolescent Mental Health (CAMH) journal; and JCPP Advances.
Subscribe to ACAMH mental health podcasts on your preferred streaming platform. Just search for ACAMH on; SoundCloud, Spotify, CastBox, Deezer, Google Podcasts, Podcastaddict, JioSaavn, Listen notes, Radio Public, and Radio.com (not available in the EU). Plus we are on Apple Podcasts visit the link or click on the icon, or scan the QR code.
Dr. Mannan Luo is a postdoctoral researcher at the Department of Psychiatry, Interdisciplinary Center Psychopathology and Emotion Regulation (ICPE), University Medical Center Groningen, the Netherlands. Her current research explores the intergenerational transmission of substance use and substance use disorders, by integrating family and genomic data from both parents and offspring. Dr. Luo received a PhD in Developmental Psychopathology from Erasmus University Rotterdam. The PhD thesis focuses on understanding the complexity of genes-environment (G-E) interplay, including G-E interaction/correlation and epigenetics, in children’s neurobehavioral development and mental health.
Dr. Charlotte Cecil is an Associate Professor in Biological Psychopathology at the Department of Child and Adolescent Psychiatry and Head of the inDEPTH Lab based at the Erasmus Medical Centre, the Netherlands. Her work aims to better understand factors influencing brain development and mental health from prenatal life to adolescence, in order to improve strategies for early risk detection and the prevention of chronic mental illnesses later in life. A key interest in her group is to identify how, at a biological level, genetic and environmental factors beginning in pregnancy come together to shape a child’s development, behavior, and health, with a particular focus on epigenetic mechanisms. She has authored more than 90 scientific articles and expert reviews on this topic, and (co)leads multiple international efforts dedicated to better understanding and enhancing mental health during development, including the European collaborative projects TEMPO, MIND, FAMILY, EarlyCause, and HappyMums.
Transcript
[00:00:01.339] Jo Carlowe: Hello, welcome to the Papers Podcast series for the Association for Child and Adolescent Mental Health, or ACAMH for short. I’m Jo Carlowe, a Freelance Journalist, with a specialism in psychology. In this series, we speak to authors of papers published in one of ACAMH’s three journals. These are the Journal of Child Psychology and Psychiatry, commonly known as JCPP, the Child and Adolescent Mental Health, known as CAMH, and JCPP Advances.
Today, I’m interviewing Dr. Mannan Lou, Postdoctoral Research Fellow at the Department of Psychiatry, University Medical Center Groningen, the Netherlands, and Dr. Charlotte Cecil, an Associate Professor within the Department of Child and Adolescent Psychiatry and the Department of Epidemiology, at Erasmus Medical Center, Rotterdam, the Netherlands. Mannan and Charlotte are co-authors of the paper, “DNA Methylation at Birth and Lateral Ventricular Volume in Childhood: a Neuroimaging Epigenetics Study,” recently published in the Journal of Child Psychology and Psychiatry. This paper will be the focus of today’s podcast.
If you’re a fan of our Papers Podcast series, please subscribe on your preferred streaming platform, let us know how we did, with a rating or review, and do share with friends and colleagues.
Mannan and Charlotte, welcome, thank you for joining me. Can you each start with an introduction about who you are and what you do?
[00:01:29.909] Dr. Mannan Luo: Yes, sure. Hi, Jo, and many thanks for having us. I’m Mannan, a Postdoc at the University Medical Center Groningen, in Netherlands. I currently focus on substance use disorders, by using genomic data of the parents and the offspring, and looking at how substance use behaviours are passed from parents to their children. And I did my PhD in the Child Psychology and Psychiatry, mainly on genes and the environment interplay on child neurobehaviour development, together with Charlotte, as my PhD Supervisor.
[00:02:06.070] Jo Carlowe: Great, thank you. And Charlotte?
[00:02:08.469] Dr. Charlotte Cecil: Yes, thank you very much, Jo, for having us here today. So, my name is Charlotte Cecil. I’m an Associate Professor in Biological Psychopathology at the Erasmus Medical Center in Rotterdam. And, as Mannan said, we’re really interested, and our group is to just better understand how children’s genes and their environmental exposures, beginning from pregnancy, actually come together at the biological level to shape their development, mental health, and also the risk of developing psychiatric disorders later in life. And one of our core focuses, as you will hear more about in this podcast, is epigenetics, and we look at this using, sort of, large-scale longitudinal datasets from the population.
[00:02:48.170] Jo Carlowe: Brilliant, thank you both. So, we’re going to look at your paper today, published in the JCPP, but before we go into the detail, I want to get to grips with some of the concepts and terminology used in the paper. So, this would be for listeners who may not have a background in genetics. Can you start with “lateral ventricular volume enlargement,” what is LVV enlargement, and why is it of interest to you as Researchers?
[00:03:15.470] Dr. Mannan Luo: The lateral ventricles are the two largest hollow structure in the brain, located under each side of your brain. And something is important to know is that the ventricles are the place to produce and secrete cerebrospinal fluid. This special fluid helps to keep the brain protected and healthy. For instance, it provides the fluid, the cushioning, for the brain. So, when you get a knock on the head, the brain doesn’t hit the inside of the skull.
And the reason why LVV is interesting for us is that we know, for many neuropsychiatric disorders, the volume of the lateral ventricles is enlarged. That’s a very strong finding, particularly in schizophrenia. Yet, we still don’t know much about what determines LVV size and whether we can identify some of the influence earlier in life, before the onset of the neuropsychiatric disorders.
[00:04:19.060] Jo Carlowe: Great, thank you, yeah, that’s really helpful. Sticking with the basics, your study refers to the “epigenetic process of DNA methylation,” can you explain what this means, in lay terms?
[00:04:31.150] Dr. Charlotte Cecil: So, epigenetics is quite a broad term, it’s an umbrella term, that literally means “above the genome.” So, it’s a layer of information that sits on top of your DNA. It’s actually – you can think of it as a collection of different, sort of, chemical tags, that regulate the expression of your genes, without actually changing the DNA sequence itself. So, you can think of the DNA sequence as something that’s really stable and, sort of, provides instructions for everything that your body needs to develop and survive and function. And, whereas this DNA sequence is very stable, the way that it’s actually used by your body, in different parts of your body, different cells, or at different times in your life, it’s very dynamic, and that’s thanks to epigenetic processes.
So, basically they are a set of processes that enable us to use our genetic code more flexibly, by regulating where in your body and when in time your genes are switched on or off. There’s different types of epigenetic processes, and in this paper, we focused on one of these processes called DNA methylation, that’s also because it’s just the type of epigenetic process that at the moment is by far better understood and most researched. But it’s important to know it’s not the only one, other processes exist, and they are likely very important as well, but we just don’t know so much about it.
So, briefly, DNA methylation, it involves these little chemical tags called methyl molecules, that basically stick to the DNA. And depending on the location and the amount of these molecules sticking to the DNA, that regulates how accessible your DNA is, and therefore how much it’s read and how much the genes are activated.
To conclude, why we are interested in it in this particular study, is because there’s more and more evidence suggesting that DNA methylation patterns are interesting, because a) they’re influenced by your genes, but they’re also influenced by your environment. So, environmental factors, again, beginning in early pregnancy can have an influence, can, sort of, get under your skin and change the way that these chemical tags are sticking and where they’re sticking. And that could explain how your environment can influence your long-term development, really at the level of changing the way your genes are expressed, so that’s really interesting to us. Secondly, it’s very essential for healthy development and brain development.
And, finally, alterations in these DNA methylation patterns have been linked to poor mental and physical health outcomes, including schizophrenia. So, to us, it’s a really nice biological system if we’re interested in better understanding how genes and environments come together to shape development and health.
[00:07:04.140] Jo Carlowe: That was excellent. That’s really, really clear. Let’s turn to the paper now. So, this is, “DNA Methylation at Birth and Lateral Ventricular Volume in Childhood: A Neuroimaging Epigenetics Study,” recently published in the JCPP. Can you give us an overview of the paper, to set the scene, what did you look at, and why?
[00:07:23.490] Dr. Mannan Luo: Previous studies have shown associations between DNA methylation measured in blood and other brain structure in patients with schizophrenia. However, they are mainly cross-sectional, in other words, epigenetics and the brain structure’s measured only once at the same timepoint. And, also, most of them use the clinical sample of adults, as such, that’s very unclear what the timing and the direction is of the DNA methylation during psychiatric symptoms associations. Then the question came up is that “Changes in DNA methylation lead to changes in brain structure, increasing risk for schizophrenia? Or is it that individual, already diagnosed with the schizophrenia, showed epigenetic and brain differences?” Example, because of a medication used, or other factors related to this condition.
So, I think it’s really crucial to address this direction issue, if we want to better understand the role of DNA methylation in psychiatric risk, and inform the potential translational applications. So, in this project, our goal was to use data from two prospective studies, to figure out three questions. First, “Would the DNA methylation early in life associate with developing brain, specifically LVV, in the general population, before psychotic symptoms manifest?” And, if so, we’re wondering, “Are the associations stable and dynamic over time?” As the DNA methylation can be very dynamic. Then we’re also wondering, “Can we trust, under our results, do these associations do exist in an independent sample?”
[00:09:24.480] Jo Carlowe: Thank you. Can you tell us a little about the methodology used for this study?
[00:09:29.279] Dr. Mannan Luo: In our main analysis, we used the data from over 800 children from Generation R study, which is a ongoing population-based study, that has been following the lives of the children in Rotterdam, from pregnancy to adolescence. And it’s currently one of the largest epigenetic and neurodevelopmental cohorts in the world. I should mention here, it’s quite rare to have this combination of epigenetics and the brain imaging data. Before we turned to epigenetics, which was our main focus, we first wanted to ensure the relevance of the LVV as a risk marker of the psychosis, in line with the previous clinical studies. As such, we test whether LVV in childhood prospectively associate with psychotic-like experience in adolescence.
Very important to emphasise here, because we are looking at the association in the general population, during development, we study psychotic-like experience, rather than clinically diagnosed psychotic symptoms. What we did is that we ask the adolescents to express, for example, “whether other people have ever read your thoughts?” And indeed, we found that association.
Then our analysis proceed in three steps. First, we performed a epigenome-wide association study, or EWAS for short, where we tried to test whether specific DNA methylation sites across the genome at birth associate with lateral – LVV sites during childhood. Second, we tested whether these epigenetic associations are stable or dynamic over time, using repeated measures of DNA methylation across development. Again, I want to point out that such repeat the epigenetic data during development is also very rare.
Last, we put our findings to the test, so there were, like, a polygenic score around the field of the genetics, which represent a total genetic signal for a particular phenotype. Here, in this case, we use our EWAS results, to construct a polyepigenetic score of LVV that would prevent a neonatal epigenetic markers of later LVV. And then we test whether this score actually associates with LVV in a different longitudinal cohort, ALSPAC in the UK.
And the other point I really should mention here, although Generation R and the ALSPAC are very similar cohorts, there is two key differences. ALSPAC measured LVV in adulthood, instead of the childhood, and the data was only available in males, not both sexes. As such, we cannot really call this a replication per se, but more test of how generalisable and reproducible findings are to a different cohort and age group.
[00:12:53.100] Jo Carlowe: Thank you. What key findings from the paper would you like to highlight?
[00:12:56.990] Dr. Mannan Luo: From the EWAS, using the DNA methylation at birth, we found the – that several epigenetic sites measured in cord blood at birth was prospectively associated with LVV at the age of ten years. And interestingly, these sites mapped to genes previously implicate in brain development, and also psychiatric disorders, including schizophrenia.
And we also find that, indeed, that epigenetic markers associated with LVV at birth still associate with LVV when DNA methylation measured later in life, support a degree of the stability. Last, we reviewed a polyepigenetic score of LVV at birth in ALSPAC, based on our discovery analysis in Generation R, we found that this epigenetic score did significantly predict LVV in young adulthood in an independent sample of older males. That’s the evidence of reproducibility, that was fantastic to see.
[00:14:07.040] Jo Carlowe: What is the advantage of capturing DNA methylation at different life stages? And how will this approach further knowledge in this field?
[00:14:15.730] Dr. Charlotte Cecil: So, as I mentioned earlier, the genetics processes, such as DNA methylation, are really important for allowing you to use your genetic information more flexibly across a life course. And that’s just really important for healthy development and healthy aging. So, these are really dynamic processes, and that’s a strength, but it’s also a challenge. Because it’s still very expensive to measure for genetic processes in large enough samples. So, the vast majority of epigenetic studies in the world just measure it at one timepoint, so you’re capturing a snapshot of something that’s very complicated and very time varying.
A lot of studies of epigenetics and schizophrenia, for example, have been based on postmortem samples, it’s one timepoint and it’s postmortem. And while this can be particularly relevant mechanistically, because you are looking at the genetic properties in the brain, instead of as we do in living individuals in the blood, so it’s more relevant maybe mechanistically. But, at the same time, you’re really looking at this endpoint, so you cannot look at developmental processes in vivo, which is going to be much more relevant, especially in clinical uses.
And then, even in living individuals, a lot of studies, as Mannan mentioned before, really look at, say, the brain, or psychotic, or other psychiatric symptoms and epigenetics at the same timepoint, and then it’s really difficult to establish the direction of these associations. So, where Generation R and actually also ALSPAC are very unique I think in the world is that there are some of these few longitudinal studies with repeated measures of epigenetics, starting from birth, so before any onset of behavioural manifestations or psychiatric symptoms actually start to develop.
And because they are at repeated timepoints, you can ask specific questions that you can’t otherwise. For example, “Are your epigenetic patterns more sensitive to certain environmental insults at specific timepoints, maybe earlier in life but not so much later in life?” Or, also, “Are the genetic patterns at certain points in development more predictive of risk or future outcomes?” And that’s exactly what we’ve seen. So, in another related project, what we’re starting to see more and more, is that especially for certain neurodevelopmental conditions, epigenetic patterns at birth seem to be quite informative about your risk of developing neurodevelopmental conditions later on, and this signal can actually disappear over time.
So, there’s something about birth, maybe because it’s a closer proxy to foetal environmental factors, or foetal biology, that we know is very important for neurodevelopmental conditions, and actually schizophrenia is more and more recognised as a neurodevelopmental condition, and maybe it’s more reflective of genetic influences. And as time goes by, maybe there’s a lot of other things that are happening that wash out this early symptom, but I mean. So, having the genetics at repeated timepoints can give you that insight about when in time you should be measuring these patterns, and what’s irrelevant, maybe, for research and clinical intervention.
[00:17:10.770] Jo Carlowe: It’s really fascinating. And what are the implications of your findings for Researchers?
[00:17:15.500] Dr. Mannan Luo: For Researchers, I think there are two things to mention. One is, we really need to understand better DNA methylation, brain behaviour associations, and do more research from a developmental perspective. And that’s crucial, we want to dig more about the role of the DNA methylation and brain in the onset and the persistence of the psychiatric symptoms. I think our study provided a example of how to work with multiple time, varying types of data, epigenetic screen and behaviour, to map these developmental links.
And another thing I want to say is about the effort to improve the reproducibility. Although these differences between the two cohorts, as I said before, I feel confident that we really do our best, at least, to build this international collaboration and to test if the results still hold across time, across different cohorts and populations, across different measures of psychotic-like experience.
[00:18:27.860] Jo Carlowe: What about Clinicians? How might your findings be translated into practice for Clinicians? So, might your findings pave the way for more targeted treatments, or ways to mitigate risk for schizophrenia, or for a different approach to diagnosis?
[00:18:43.620] Dr. Charlotte Cecil: I think although I believe this is a really promising first step, our findings are still a bit too preliminary to be actually concretely translated into clinical recommendations. But I do believe that already what they show, what they can help us achieve, is a better understanding of the underlying biology and developmental pathways that might underly risk for – in developing psychotic-like experiences and psychotic symptoms. But I think what – the ambition, where this really leads the way to, is trying to identify better ways of detecting risk early on, before the development of symptoms. Because if we can detect things early on, we can put in place potential strategies to mitigate the risk, and try to shift the risk trajectory, and potential impairment that might result.
I do think, however, what we do see, from our results, is that even though we identify associations, they make a lot of sense biologically, we’re able to build the score, and it’s reproducible, and that’s fantastic, it’s still explaining very little variance in our ultimate [inaudible – 19:46]. I do not think that epigenetics will be the one answer. I think what is more likely is that epigenetics could be a very useful addition to a clinical toolkit or multimodal, sort of, algorithm, or a set of biomarkers that might include genetics, epigenetics, other clinical information, to help us really increase that predictive power.
Oh, one thing I want to say is that even if we’re able to achieve that, so imagine that at birth, there is a way to meaningfully add some information about someone’s later risk, one thing is to establish whether this is useful. Another thing is ethics, like, this opens a lot of ethical potential considerations. Because we’re not talking about a diagnostic tool at birth, we’re talking about potentially a risk prediction, a way to increase your ability to predict this.
But like any type of risk prediction tool, it comes with potential error. It could be that biological risk never materialises into an actual impairment. What would that mean in terms of actionability? If you had a positive screen at birth, what could we do that really would help to shift that risk? And is that acceptable to stakeholders, to families, to individuals who experience these conditions, to Clinicians, to public health? These are all these kind of things that, further down the line as we’re more confident in the ability to have useful tools, the next question is, should we apply them, and how should we apply them?
[00:21:08.241] Jo Carlowe: Is there anything else in the paper that you would like to highlight?
[00:21:11.230] Dr. Mannan Luo: One thing which I personally find interesting is that we found that associations between DNA methylation at birth and LVV are stronger in males. The polyepigenetic score worked even better if it’s calibrated specifically for males, I mean, when only used the discovered – discovery findings from males, not the sex combined, actually these not only suggest the potential sex differences in the association between DNA methylation and the LVV, but it also shows that aligning discovery and the replication data as much as possible, can lead to better precision and improve the predictive performance. I think it’s – in future study it would be interesting to follow-up to see why that differences do exist.
[00:22:03.070] Jo Carlowe: Well, let’s talk about future studies, are you planning any follow-up research? Or is there anything else in the pipeline, for either of you, that you would like to share with us?
[00:22:13.940] Dr. Charlotte Cecil: Yes, I think this was one of our first, sort of, dipping our toes in the water, in the sense of, how can we actually integrate and examine, as Mannan said, what are essentially very complicated, high dimensional, time varying data types in multiple datasets? So, that was a bit of a – at the beginning, I think of a longer journey. I believe that in the field of psychiatric epigenetics, we’ve all been very interested in relating epigenetics directly to mental health outcomes, but we still know so little about how epigenetics relate to the brain itself in vivo.
And that’s why it’s so important, I think, to try to do these neuroimaging epigenetic efforts. But the reality is, as Mannan mentioned, that very few studies in the world actually have that combination of data, of having epigenetics and brain, across development, ideally at repeated timepoints, but that’s what we need.
So, in that line, what we’ve done recently, is actually we established a new consortium, a new international consortium called MIND, which stands for Methylation, Imaging and Neurodevelopment. And we’re actually bringing together ideally all Paediatric cohorts that have neuroimaging and epigenetic data at different timepoints, so that we can consolidate these international efforts, increase our statistical power to detect essentially what are small effects, and really map out, in the first 20 years of life, how your epigenetics and your brain and behaviour relate to one another. So, that’s our big ambition going forward.
[00:23:40.140] Jo Carlowe: Finally, Charlotte and Mannan, what is your take home message for our listeners?
[00:23:46.130] Dr. Mannan Luo: The study identifies, for the first time, DNA methylation markers at birth associated with LVV early in life, which can shed light on biology underlying this important brain feature. And, also, the polyepigenetic score might be more stable and reproducible than the single epigenetic sites.
[00:24:08.659] Jo Carlowe: Charlotte, do you have a – any comments?
[00:24:09.659] Dr. Charlotte Cecil: Yes, perhaps, just a more a general comment, I suppose, that I hope what emerges from this podcast is just how epigenetic processes, such as DNA methylation, can be such an interesting biological system to understand how our genes and our environment come together, really at a biological level, and how that can shape our development and later health. It also helps to explain potentially how environmental factors that occur very early in life can have such long-term effects, because it provides this, sort of, biological mechanism for embedding of experiences.
And that’s one thing that we didn’t look at in this paper, we – is why do we find this signal at birth? And I think that will be another important – next step, is, like, what is this signal, epigenetic signal, at birth actually reflecting? Is it due to genetics? Is it due to potentially certain environmental insults in pregnancy? Or other factors that could help point to potentially modifiable factors that are involved in the aetiology of schizophrenia risk that would also have important clinical implications.
[00:25:13.890] Jo Carlowe: Fantastic, thank you both ever so much. For more details on Dr. Mannan Luo, and Dr. Charlotte Cecil, please visit the ACAMH website, www.acamh.org, and Twitter @acamh. ACAMH is spelt A-C-A-M-H, and don’t forget to follow us on your preferred streaming platform, let us know if you enjoyed the podcast, with a rating or review, and do share with friends and colleagues.