Maternal Prenatal Social Experiences & Offspring Epigenetic Aging

Drs. Zachary Laubach and Wei Perng investigate associations between adverse maternal social experiences and offspring epigenetic aging among 205 mother-offspring pairs of minoritized racial and ethnic groups. "Maternal prenatal social experiences and offspring epigenetic age acceleration from birth to mid-childhood” is published in the February 2024 issue (Vol. 90) of Annals of Epidemiology.

Read the full article here:

https://www.sciencedirect.com/science/article/pii/S1047279723001928

Episode Credits:

• Executive Producer: Sabrina Debas
• Technical Producer: Paula Burrows
• Annals of Epidemiology is published by Elsevier.
  
  
  

Transcript

Patrick Sullivan: 0:12

Hello, you're listening to EPITalk: Behind the Paper, a monthly podcast from the Annals of Epidemiology. I'm Patrick Sullivan, Editor-in-Chief of the journal, and in this series we take you behind the scenes of some of the latest epidemiologic research featured in our journal. Today we're talking with Drs. Zach Laubach and Wei Perng about their article "Maternal, prenatal Social Experiences and Offspring Epigenetic Age Acceleration from birth to mid-childhood. You can find the full article online in the February 2024 issue of the journal at www. annalsofepidemiology. org. So I'll briefly introduce our guests.

Patrick Sullivan: 0:57

Dr. Zach Laubach is a research associate in the Department of Ecology and Evolutionary Biology at the University of Colorado, Boulder. He's a behavioral ecologist and evolutionary biologist interested in how and why (proximate and ultimate) early life environments influence the phenotypes upon which selection acts. And Dr. Wei Perng is an Associate Professor of Epidemiology at the Colorado School of Public Health and the Life Courseourse Epidemiology of Adiposity and Diabetes, or LEAD Center. She is a nutritional and life course epidemiologist whose research leverages omics science as a tool to study early origins of excess adiposity and metabolic risk in youth. Doctors, fascinating work that you do individually, and so I can't wait to talk about what you've done together and thank you so much for joining us today.

Wei Perng: 1:48

Yeah, we're happy and excited to be here.

Patrick Sullivan: 1:51

Great. So,Dr. Perng, can you start out just by giving us some background on the problem that's described in your paper? What is the aging epidemic and what is the EAA, and why is this an important issue to study?

Wei Perng: 2:04

Yes, thank you for this question and I'll just start by saying that I don't know if I'm a big fan of the term aging epidemic, acknowledging that we use it in the paper. But you know, this term reminds me of other terms like geriatric pregnancy. That makes me cringe to describe people who are older than 35 when they get pregnant. I just think this term is not ideal because it really assigns the notion of a disease state to what is actually a normal, ongoing process of senescence, whether that's biological, functional and or reproductive. But regardless of that, the term is referring to this recent increase in older populations, and here we're typically thinking of folks who are 65 years or older. I believe we've observed an increase in this population over the last century, which is a good thing, because it's really resulted from a general increase in overall population size, but also the improvements in health of older persons worldwide due to better sanitation, in health of older persons worldwide due to better sanitation, healthcare, reduced infectious disease burden and better access to nutritious foods, for example.

Wei Perng: 3:11

And the growth of older populations worldwide, I think is really relevant to the field of public health because we seek not just to prevent disease but also to optimize health across the life course.

Wei Perng: 3:23

So primordial prevention and, as my co-author, who is also on this call, Dr. Zach Laubach, will tell you, a key reason why aging coincides with higher incidence of chronic conditions like cardiovascular disease, cancer and diabetes is because natural selection will act on reproductive fitness, so optimizing health through the reproductive years, and that leaves very little selective pressure after this life stage. So I would say the latter one third of our lives do tend to be more fraught with poor or declining health and with the growing number of aging persons worldwide, a better understanding of the aging process, both in terms of molecular markers and mechanisms, of which epigenetic age acceleration is one of them, and just a quick recap, epigenetic age acceleration is this difference between your biological age, based on DNA methylation markers, and your actual calendar or chronological age. So a better understanding of these types of markers or mechanisms and also the upstream determinants of aging, can help to inform public health strategies to promote healthy aging, so that we can compress morbidity at the latter stage of the lifespan.

Patrick Sullivan: 4:44

Thanks for that explanation, and so here you're trying to understand the association of maternal prenatal social experiences and this sort of better marker or a very indicative marker of aging? Just for context, are there some other kinds of exposures that have already been proven to be associated to sort of accelerate aging in that way? That would just help us have a frame of reference for like what kinds of things speed up this particular clock.

Wei Perng: 5:12

Right-

Wei Perng: 5:13

That is a bit of a challenging question to answer because whether or not epigenetic age acceleration is happening or not, so if it's higher or lower, we don't always know what that means depending on the age of the population.

Wei Perng: 5:27

So most of this evidence has been generated in adults and we know that your typical unhealthy lifestyle like smoking and nicotine use, unhealthy diet, lack of physical activity, et cetera, that those are associated with accelerated aging and, in turn, higher morbidity and mortality as you age.

Wei Perng: 5:47

And we also know that the structural determinants of health or social determinants of health, including experiences of racial bias and discrimination, can get under the skin and in adults is associated with accelerated aging. In children, less is known because it is harder to draw blood from little people. And on top of that, we don't quite know yet what epigenetic age acceleration means before puberty and this is something that I think Zach will talk about later in one of his responses because from an evolutionary standpoint, there are different selective pressures that act right up until you hit puberty. So we as a group of co-authors, when we wrote this paper, tried hard not to assign a valence, if you will, to the direction of association that we observed, and it's really something that we can't do until we can link epigenetic age acceleration in young children to a clinically relevant health outcome that can provide us a better grasp on what our findings might mean.

Patrick Sullivan: 6:49

Yeah, it's so interesting because the method you know probably is relatively recent, just given the acceleration of understanding. So really that correlation with the long-term outcomes, there's just not sort of a history since this metric has been measured. So thanks for that answer. Dr. Laubach, I'm going to move to you now and just see if you could tell us a little bit more about the purpose of this particular study and the research question that you were seeking to answer.

Zach Laubach: 7:15

Yeah, sure. So this was an exploratory study involving data from about 200 mother-offspring pairs, where we asked a relatively simple question, which was do social experiences of the mom affect the development of her kids? And, more specifically, we looked at whether or not maternal experiences of racial bias or discrimination, social support or these indicators of socioeconomic disadvantage during these prenatal periods, whether they correspond with epigenetic age acceleration or this EAA marker, which is, as Dr. Perng mentioned, a biomarker of aging, and then we were asking if these differences in EAA were evident in the mom's kids across early and mid-childhood.

Patrick Sullivan: 7:56

So thanks for describing the question Dr. Perng. I wonder if you could walk us through what specific methods you used to answer this question and why that methodology this question and why that methodology?

Wei Perng: 8:07

Sure. So we didn't choose all of the methods per se, because we were capitalizing on extant data. This was secondary data analysis of the Project Viva cohort, which is an ongoing pre-birth cohort of what was originally over 2,000 mother-offspring pairs who were recruited during early pregnancy, at a median of nine gestational weeks, I believe, and then followed up across 18 years. Now, granted, this study didn't take advantage of all 18 years, but I think a major strength of this study is having this prospectively collected cohort data to work with, because it enhances our ability to make temporal and therefore causal inference although we know that is a tall ask right using observational data. In terms of the actual analytic design, which is where we did have more of a choice, we used information collected from the women at the early pregnancy visit at around nine gestational weeks, asking about their sources of social support, their experiences of racial bias and discrimination and various indicators of both structural as well as individual level socioeconomic status. So these exposures are happening during pregnancy or across the woman's life course prior to getting pregnant, and then what we have is repeated biosamples collected from the offspring at birth using cord blood, and then in early childhood as well as mid-childhood, so those represent ages three to six years and then ages six to 10 years.

Wei Perng: 9:46

And I think that having these repeated longitudinal bio samples for the epigenetic age acceleration assays is quite unique. You know I mentioned earlier how challenging it is to obtain blood from children, but to have these repeated measurements allowed us to look at patterns of change over time, look at stability in the associations over time, of change over time, look at stability in the associations over time. And you know, we know, that having more measurements provides more information about an outcome. And you know, in terms of other aspects of the analysis, we started off drawing directed acyclic graphs and we use prior knowledge as well as the principle of parsimony to determine which set of variables to account for in our multivariable models, with the primary goal of really accounting for confounding and some precision covariance. So for this study it was maternal prenatal smoking, which is known to affect epigenetic markers in offspring, the offspring's biological sex and then, as a precision covariate, cell type composition, because that can affect epigenetic age measures.

Patrick Sullivan: 10:55

Great, and so, when you take all those factors into account, what were some of the main takeaways from your analysis? How did you answer the question that you asked at the outset?

Zach Laubach: 11:06

So our main finding was that maternal experiences of racial bias and discrimination were associated with slower epigenetic aging in these children prior to puberty, and these findings suggest that social stresses contribute to this intergenerational embedding, if you will, of health disparities, possibly by slowing this tempo of development early in life. And the idea here is that changing this tempo of development may be a precursor to future pathologies that may develop in these children.

Patrick Sullivan: 11:35

Interesting. So is the thought that these exposures that may occur during gestation right, because you're measuring the exposures during the time of gestation but the effects of them play out then over the time period that you're actually doing the analysis as researchers who are collecting these specimens collected them over time. That's a pretty profound mechanism. I would just say, is it? Obviously you hypothesized it, but was it surprising based on how you went into this?

Patrick Sullivan: 12:05

that these biological markers would show these associations from exposures that happened, you know, so many years before.

Wei Perng: 12:14

I mean what I'll quickly say and put in a plug for that is that Zach and I published a prior paper using the same cohort and it was a different type of analysis. It was an epigenome-wide association analysis and with these types of studies what we know is that sometimes effects are more detectable as children get older. But it can also be the other way around, in that the epigenome becomes noisier when they get older, so it gets cluttered and it's really hard to tell the toss up. But if you're asking, we were surprised by the findings. I was, but Zach wasn't, and he can probably talk a little bit about why he wasn't.

Patrick Sullivan: 12:50

That's awesome. So, Zach, I have another comment about that. But, Zach, so you weren't surprised. Why were you less surprised, would you say?

Zach Laubach: 12:57

Well, so yeah, this idea that I guess I was, let me think about this.

Zach Laubach: 13:04

So the idea that these adversities were associated with slower epigenetic aging was somewhat surprising, I guess because of the fact that most of the literature is looking at these adversities and then these various environmental exposures that are associated with faster biological aging.

Zach Laubach: 13:22

And I think in an adult population that makes pretty good sense. We expect that if you're aging faster at the latter part of life, then you're likely going to not live as long, and so that makes sense there. The part that I guess that at least it can be that we think might be happening or where you know you can make some justification for the results that we found is that these deviations from an average may in and of themselves be indicative of something that is dysregulated or where the system is not functioning properly. For these children, if they are not aging as fast as maybe they should be, that could be indicative of underlying conditions, because at these early points in development there's coordination and rapid change of many biological systems. So this may, in and of itself, this lower epigenetic aging may be a marker of some underlying condition that is preventing the natural developmental pace.

Patrick Sullivan: 14:19

Yeah, and I think so often we have our preconceptions about what the results are. We can just call it an alternate and null hypothesis. But the real progress of science here is that when our questions sort of raise more hypotheses and lead to further exploration. Why do you think there is so much conflicting findings regarding EAA in the literature.

Zach Laubach: 14:43

Yeah, I'll just follow up. It's sort of related to what I just said. I think a lot of the work that has been done has focused and for good reason on these epigenetic marks in adults, and so there's now more work coming out looking at EAA and determinants of this in younger. Just the timing of when these epigenetic markers are measured could be some reasons why there are some conflicting results. Secondarily, these markers are developed in different populations and because of that these algorithms that use these DNA methylation data they vary, so there are many different kinds of epigenetic locks and they are developed in different populations or validated in different populations, and so these two factors also could contribute to a lack of generalizability, just because of how the algorithms are developed or the validation population in which the algorithm is developed.

Patrick Sullivan: 15:43

Yeah, thank you, and I think it really just reminds us that, even in a really specialized field like this, that the questions about you know how we measure and our confidence in the measurement is one of the issues we have to consider, and another one you know that I always sort of lurking for me is bias.

Patrick Sullivan: 15:59

And so for either one of you here you have, the biological outcomes are, you know, are very empiric, but the data on the maternal social experiences are self-reported. So how do you sort of weigh the potential biases, I guess, in the self-report aspect of the exposures?

Wei Perng: 16:18

That is a great question and it is one I think about all the time in the context of nutritional epidemiology, where we know that there's differential recall bias, often with respect to the health outcome of interest, when you are assessing someone's dietary intake using a food frequency questionnaire or a recall right, and then we have the gold standard of having people weigh their foods and keep a diet record and et cetera. In this case, what we're interested in is the perceived experience of bias and discrimination, and there is no gold standard. If there is, that is the person's self-report, and so I'm not too worried about bias in this particular setting, especially because it's an observational cohort study right now. It's not as though, you know, we found this and then went back and asked the women. You know what did you perceive? So I don't think it's an issue and I think when we think about these types of structural social experiences, it is the lived experience that really is the gold standard.

Patrick Sullivan: 17:20

Great, thank you. So I'm going to pivot a little bit. And the sort of other thing we always try and get at is understanding how scientists work, work together, how you come to ask the questions that you did, and here the two of you worked on this together and I'm so grateful to have both of you here to talk little bit about this. So I wonder if you can just talk about- in the analysis that was done, what roles you played.

Patrick Sullivan: 17:35

I mean, you have really interesting and complementary, I think, kinds of training. So what roles did you play in the research process and how do you think those different perspectives let you ask this particular question? What it would have been like missing one of those expertise?

Wei Perng: 18:08

Well, I'm a co-investigator of the Project Viva cohort and so Zach is often interested in asking questions that parallel his research in spotted hyenas in Kenya. He has a cohort of spotted hyenas there. So when I see opportunities and ones that I'm also interested in, I'll often relay it to him. And I would say our roles have shifted over time, because initially I was primarily the methodologist and he was more the question asker, and we try and figure out how we can operationalize the questions that he was asking in a quantitative fashion. Now, you know, over the years Zach has taken epidemiology courses.

Wei Perng: 18:48

At one point in time wanted to get an EPI certificate but decided it was more important to just finish his dissertation, and so he's actually mastered a lot of the EPI concepts.

Wei Perng: 18:58

Like he lives and breathes DAGs. People from his lab go to him to ask about, you know, what's the best analytic approach for this, and so for this project I would say it was really split 50-50. You know, aging is something that we both become interested in. It's kind of at the other end of the spectrum of what I study, which is early origins and developmental origins of health and disease, but it's very much a part of life course research and we both believe that this social aspect of you know our experiences is increasingly recognized as one of the strongest drivers of health and disease. So we were both interested in the topic. We came up with the analytic approach together and then Zach implemented the analysis this time so, and we we split the paper writing. So you know, the methods came easily to me because I know the cohort so well, but he really drafted the intro and the discussion and we worked together on it and sent it to co-authors.

Patrick Sullivan: 19:57

Thanks. Dr. Laubach, any other thoughts about, like how you work together on this?

Zach Laubach: 20:01

Yeah, so, as Dr. Perng mentioned, we've been working on science related questions for a while.

Zach Laubach: 20:07

We met in grad school and, yeah, we met teaching a physiology lab as grad students and we're actually- I guess it's worth mentioning that we're partners as well and I think it's sort of interesting and not so uncommon. There's a number of scientists who, I know, at least, that have formed these collaborations. It lends to lots of science discussions over dinner and that may sound like a fairly dry dinner conversation, but I actually find it really really useful. As Wei mentioned, I'm just going to say Wei, the epi methods have always been very interesting to me because they apply ways or they're a method that can be used to test these causal hypotheses with observational data, which is the problem that many ecology and evolutionary biologists also face. So these dialogues that we've had have allowed us to think about ways in which there can be useful crosstalk between these disciplines that are otherwise, I think, fairly siloed or have been, that are otherwise, I think, fairly siloed or have been, but I think there's an increasing interest in combining the expertise and the theory across these two different fields.

Patrick Sullivan: 21:11

Yeah, the most interesting science happens at the margins, you know the little intersections of fields, I think. So these are really interesting ones and sometime although we don't have time on the podcast today, but sometime I want to hear about the hyenas. I mean, probably everybody wants to hear about the hyenas.

Wei Perng: 21:29

It's true, Zach gets to show pictures of the Serengeti and safaris on his job talks and I have nothing like that to show. I just have tables of numbers.

Patrick Sullivan: 21:39

Well, it takes all these pieces, and I do think that the combination of these methods and your mutual interest in epidemiology is a really interesting aspect of the paper that you brought forward. So I'll just ask either of you keeping, in mind that I think we're really interested in the science, but also in how science happens and how these fields come together, whether you have any last thoughts that you'd like to share with the listeners?

Patrick Sullivan: 22:05

If they read it before, this may give context. If they haven't read it, interesting to understand how it came about and how you worked on it. But any closing thoughts about the manuscript or the way that it came about that you want to share with our listeners?

Wei Perng: 22:18

I will just keep this brief. I think Zach has a much better response to this question, especially about what would you tell your younger self which I think is a great quI guess I would say and Zach reminds me of this a lot that the best science happens when it's interdisciplinary and there's challenges at the beginning, because it's as though you speak different languages, some terminology that epidemiologists are really specific about. It's like nails on the chalkboard when you hear someone misusing it, and similarly there's times when I've used the word evolve in completely the wrong way and Zach will remind me. So I think it's been a really nice synergistic collaboration and I've learned a lot.

Wei Perng: 22:59

I still struggle to think about how to place my research findings in the context of evolution, knowing that everything happens in the context of evolution. So I would like to be able to better use an evolutionary framework in the way that I think about my research, especially because some of my research is in precision medicine, which that really starts with the gene by environment interaction and our genetics is due to our ancestry, and so it is actually quite relevant to my work. So I do see future collaborations with Zach in that realm, and that's kind of what I have to say about that, but maybe I'll let Zach talk about. You know something that he'd like to share with the audience.

Zach Laubach: 23:38

I'll share two quick things if we have time for that. Just thinking a little bit about the question of, you know, epidemiology is this method-based approach, so it's very applicable and it's also very clear, and it extends naturally to other disciplines. But one piece of or sort of food for thought that I could give about how epidemiology or public health researchers can use evolution is more of a conceptual idea and it comes from this paper that was written by Randolph Nessie and Stephen Stearns, and they're arguing for the ways in which evolutionary biology is important for public health. And they're arguing for the ways in which evolutionary biology is important for public health, and I'll paraphrase them as I recall it. They say that we are humans. We are vulnerable to diseases because we're not these machines that are built from a plan, but rather we're just this bundle of compromises that have been shaped by natural selection that maximizes reproduction and not health, and that was something Wei mentioned earlier. So it can be the case that what is good for your fitness may not necessarily improve your health, and I think that is worth keeping in mind because while these mechanisms that may affect health are important, they're not acting in isolation from other important processes ontology and phylogeny and adaptive functions.

Zach Laubach: 24:49

And then, moving to this question about what would you tell your younger self or what advice would you give for professionals that are interested in these interdisciplinary collaborations, and much like what you said about Dr. Sullivan, about the good signs happening at the margins, I would agree and say that people should embrace this uncertainty of what you don't know, because there's a lot to be learned outside of these silos of expertise.

Zach Laubach: 25:14

And I think here in these spaces we're sort of uninhibited by this established dogma or field or the expectations of what we maybe should know or how we think about a problem. And I think if we, yeah, embrace those margins, then we can think openly and inject creativity. And along those lines it's worthwhile to just take time and observe biology and nature and maybe quietly. So, some of the most notable, both public health achievements like John Snow's removal of the pump handle, or to stop the cholera outbreak, or Charles Darwin's description of evolution by natural selection. A common thread of all of these sort of monumental contributions to the field is that these came from scientists who were really astute observers of the natural world. So combining these two practices of embracing a bit of uncertainty and taking time to listen, I think it'd be pretty transformative and I guess I would agree with what you said. I think that's really where we find some innovative and often impactful science is in those margins.

Wei Perng: 26:17

I will just add one more thing to that, because I can't help it. But you know, in the context of thinking about experiences of racial bias and discrimination, I heard a speaker, Dr Sandro Galea, who you may know, Dr Sullivan. He came and talked at a steering committee meeting that I was at and he said the texture of the lived experience is much more informative than any fancy method, and I think that's a form of listening and observing and I think we would do ourselves a great service to, you know, do more of that. You know we're so boxed into quantitative analyses but maybe more qualitative work, and that's something that I'm learning as well.

Patrick Sullivan: 26:57

Yeah, thank you for sharing those words from Dr. Galea, and I'm actually just going to reflect back like what a rich conversation this has been, and the thing that I'm going to put on my coffee mug, with all credit given to Dr Lobach, is, if I got it right, a bundle of genetic mistakes shaped by natural selection/ Did.

Zach Laubach: 27:11

I get that righ hat's the paraphrase and that's from a physician. I think they're a physician, andolph Nessie and Steven Stern, who is a life history evolutionary biologist, and I do think, yeah, that life history biology is really integral into these questions of developmental plasticity and aging.

Patrick Sullivan: 27:30

So, thank you, got the attribution right, but I still want to put it on a coffee mug. If I make them, I'll make one for each of you as well. There you go.

Wei Perng: 27:38

Oh well, thank you for the coffee mug from this year. Oh right, those would be nice Y mugs, yes next year wit.

Patrick Sullivan: 27:49

OYes, ynext year with natural selection. So this has been a great conversation. We have done another podcast with like a pair of researchers, but I think when you comment things from different perspectives, that what really I've seen animated, heard in your voices for our listeners, but seen in our Zoom- because we also have a Zoom conversation is just how these ideas sort of ping off each other and scaffold to answer interesting questions. So I want to thank you for your work.

Patrick Sullivan: 28:14

I want to thank you for bringing it to Annals for publication and especially for the generosity of your time to answer these questions today, and although we sort of brainstorm a little bit beforehand about what we might ask. I'm also infamous for going off the script, so also thank you for rolling with the punches. But I think it's a great conversation and a pleasure to have you on the podcast, and we look forward to what your future research will be.

Wei Perng: 28:37

Likewise. Thank you for your time and I know it's close to dinner time for you if you're on the East Coast.

Patrick Sullivan: 28:43

Yeah, I am on the East Coast and dinner is looming. So, thank you, thank you both so much. Thank you very much. Thank you. Thank you both so much. Thank you very mu I'm your host, patrick Sullivan. Thanks for tuning in to this episode and see you next time on EpiTalk, brought to you by Annals of Epidemiology, the official journal of the American College of Epidemiology. For a transcript of this podcast or to read the article featured on this episode and more from the journal, you can visit us online at wwwannalsofepidemiologyorg.