10 AHA Sessions Recap and FGTB YIA

Getting Personal: Omics of the Heart

Jane Ferguson: Hello, I'm Jane Ferguson and you are listening to Getting Personal: Omics of the Heart, the podcast from Circulation: Cardiovascular genetics, and the functional genomics and translational biology council of the AHA. This is episode ten, from November 2017.

November is always a big month for AHA and the annual Scientific Sessions were held in Anaheim, California, November 11th through 15th. For those of you who were able to attend, hopefully you came away feeling refreshed and invigorated and with your desired level of Disney merchandise. For those of you who could not attend, or who didn't make it to all of the genomic sessions, this month's episode should catch you up.

For the past several years, the FGTB Council has been organizing boot camps at AHA sessions to give people a chance for hands on learning in a flipped classroom model. This year was no exception and in addition to a clinical genomics boot camp focused on patient centric genomics including single gene testing, whole genome sequencing and pharmacogenomics there was also a new boot camp focused on tackling big data network systems analysis for high input data interpretation.

These boot camps are always very well attended and popular, so if you're interested in attending one next year, make sure to get in early and sign up during registration. There was also a hands on session in collaboration with the AHA's Precision Medicine Institute to teach people how to use the precision medicine platform to further their research.

In addition to this, there was a full day of programming related to precision medicine in the precision medicine summit, which is held on the Tuesday of Sessions. That covered topics ranging from big data, electronic health records, collaborations and the All of Us initiative to rapid fire reports from ongoing consortium, large scale analysis to disease specific approaches in cardiomyopathy.

We were planning to have an in depth focus on the Institute for Precision Cardiovascular Medicine in a future podcast episode, so stay tuned for more on that coming soon. There were a number of individuals who were recognized for their contributions to science and we would like to congratulate all of these outstanding individuals.

The FGTB medal of honor was awarded to Stuart Cook from the Duke National University of Singapore. The FGTB mentoring award was awarded to Robert Gerszten from Beth Israel Deaconess Medical Center. The FGTB distinguished achievement award went to Sekar Kathiresan from the Broad Institute. And the functional genomics and epidemiology mid-career research award went to Kiran Musunuru from the University of Pennsylvania. Congratulations to all of these.

One of the highlights for the FGTB council at sessions is the FGTB young investigator award. This award celebrates early career investigators and recognizes outstanding research in basic science, populations science, genetic epidemiology, clinical genetics and translational biology. Four finalists presented their research on the Sunday afternoon sessions and I had the chance to chat with all four of them before and after their presentations. So listen on for a behind the scenes over view of the finalists research and the announcement of the winner.

Mark Benson is a cardiology fellow at Brigham and Women's Hospital and is working on post-doctoral research at the Beth Israel Deaconess Medical Center in Boston with Dr. Robert Gerszten. His talk was entitled "The Genetic Architecture of the Cardiovascular Risk Proteum."

Mark Benson: My name's Mark Benson. I'm just finishing up a cardiology fellowship at Brigham and Women's Hospital and am in the middle of post doc in Robert Gerszten’s lab at Beth Israel.

Jane Ferguson: Great, and congratulations on being chosen as a finalist for the FGTB Young Investigator Award. We would love to hear a little bit more about what you’re working on and what you're gonna be telling us.

Mark Benson: Yeah, absolutely. So the goal of the project was really to integrate proteomic data with genomic data, with the idea that we may be able to use the overlap between those data sets to identify potentially novel biological pathways that underlie very early cardiovascular disease risk.

And the thinking behind that was that the lab had just finished up applying DNA-aptamer-based proteomic platform to profile over 110 proteins and the Framingham-Offspring Cohort and from that work, we had identified a very specific signature of 156 proteins in plasma that were each very strongly associated with cardio-metabolic risk.

The idea was while those associations were very strong, it was unclear if we were capturing cart or horse or how these associations were fitting together. We wanted to incorporate the genomic data to try to get a better handle on that, to try to connect those pathways to see how these proteins might actually associate with the end phenotype of risk.

Jane Ferguson: It's a sort of Mendelian randomization-esque.

Mark Benson: Exactly, yeah. So what we were able to find in doing this, we were able to use peripheral blood samples from participants at the Framingham-Offspring study. With a validation in participants of the Swedish Malmo Cancer and Diet Study. Then we did protein profiling using commercial DNA aptamer platform, soma scan. What we were able to find is we were able to detect very strong associations between these circulating cardio metabolic risk-proteins and genetic variance.

What was fascinating was we were able to see many things. We were able to start mapping where are these associations, where are these genetic variance in relation to, for example, the gene that's coding the protein that we're measuring. That had some interesting implications because for about half of the protein that had significant associations, we could track those genetic variance back to the gene. It was coding the protein that we were measuring, which was interesting because it's validating the specificity of the proteomic platform that we're using.

Jane Ferguson: Right that's nice, because so often you found a gene that's nothing related to what you think it's going to be so it's nice actually the gene you expect.

Mark Benson: Yeah, it's very reassuring too when you're looking at rows and rows and rows of data. When the top association of the p value of 10 in the minus 300 is the actual gene you thought would be coding the protein that you're measuring. So that was very reassuring, but we also found dozens and dozens and dozens of associations that were totally unexpected and that may point to completely unexplored biological pathways in cardiovascular disease. So that was obviously very exciting.

That actually led us to do two things. One was to make all these data available publicly on dbGaP because as a resource for cardiovascular research there is just way too much data for one group or a handful of groups to digest. The other thing that was fun about the project, is we were able to take one association that was particularly interesting for a number of reasons and experimentally validate it in a tissue-culture model.

Jane Ferguson: So how did that work?

Mark Benson: So this was an interesting challenge where we all of a sudden got all of these hits back, which was probably to be expected, but to try to figure out which of these dozens and dozens and dozens of new, unexpected hits, what do you do? There was one hit, one association, that was particularly strong and it was between several variance around this gene. That's a phosphatase called PPM1G. It's a transcription factor.

These variants, which was interesting, were associated with several different circulating cardio metabolic risk proteins. So our idea was, isn't that interesting? Is it possible that this is mapping to some central regulator? And so it fit that that would be ... that the nearest gene to these variants was a transcription factor and could be a central regulator.

What made it more interesting to us was that several variants in the GLGC had recently been described that were highly associated with circulating levels of total cholesterol and triglycerides and they were located around this PPM1G locus as well. The association between those variants and circulating cholesterol didn't have a clear biological connection.

So what our work had shown is that those same variants were associated with circulating levels of apolipoprotein E. So wouldn't that be interesting if these variants mapped to PPM1G, the transcription factor, this PBM1G in turn regulated circulating apolipoprotein E and that would provide some insight into the biology behind the GLGC findings.

So sure enough we were able to knock down PPM1G using SRNA and hepatocytes and then see that that led to a significant down regulation of the transcription of Apo-B and extra-cellular presumably secreted Apo-B in this model, which is kind of a nice proof of principal that this idea of integrating proteomics and genomics may lead to some novel biological pathways.

Jane Ferguson: Yeah, it's really interesting. So what's next. There are probably a lot more associations that you're going to have to go after?

Mark Benson: Yeah, I think that what this showed us is that this seems like a powerful tool. Joining these orthogonal data sets to find new pathways and so we're continuing to pursue that with an increasing number of proteins for example, so we're doing genome-wide association studies and x-gamma rays. We've gone from 156 to 1100 to 1300 and are now going beyond that and so as those numbers get higher, you start to see these central nodes come together and more interesting targets and potential pathways. It's also interesting to use these data to find new associations or new tools that you would never think to look for as ways to modulate protein levels.

So you can imagine, for example, one thing that we've been exploring for the last few months is can we identify, for example, SNP associated with an interesting circulating protein. That SNP maps to an enzyme or some other druggable mechanism and very preliminary studies, it seems like the answer is probably yes, but there is still a lot of work to be done.

Jane Ferguson: Well that's cool. That sounds really interesting.

Mark Benson: Yeah, I think the key thing is that all these data will soon be out there and so it's a very rich data set and I think there are many ways that we could use the data.

Jane Ferguson: So is that the genomic data and all the proteomic data or it's the summary of the those associations?

Mark Benson: All the genomic data, all the proteomic data and the associations as well. You can do the associations yourself if you'd like to.

Jane Ferguson: We can find that dbGaP. Awesome, well thank you for talking to us.

Mark Benson: Thank you. It's been fantastic.

Jane Ferguson: Congratulations again.

Mark Benson: Thanks so much. ...

Jane Ferguson: Jenny Lin is an instructor at the University of Pennsylvania, working with Dr. Kiran Musunuru. Her presentation was entitled, "RNA binding protein A1CF Modulates Plasma Triglyceride Levels through Transcriptomic Regulation of Stress-Induced BLDL Secretion".

Jenny, can you take a moment to introduce yourself?

Jenny Lin: Yes, hi. Thank you for this opportunity to participate. I'm Jenny Lin. I'm an instructor of medicine at the University of Pennsylvania, a nephrologist by clinical training, but training in cardiovascular research in Kiran Musunuru's lab.

Jane Ferguson: So congratulations for getting selected as a finalist for the Young Investigator Award. We'd love to hear a little bit more about what you've been presenting and what you've been working on.

Jenny Lin: Thank you. So basically, what I've been working on over the past year is functional follow-up of this A1CF locus, which is a novel locus for triglycerides. So say Sek Kathiresan's group recently published in Nature Genetics and x and y association study on plasma lipids involving more than 300,000 individuals.

One of the key findings from that study is this strong association between a lo-frequency coding variant and elevated plasma triglycerides. So we wanted to delve more deeply into the biology for why we have that genotype/phenotype connection. One of the key things that we wanted to do was ... A1CF is not a stranger to lipo-protein metabolism, but we wanted to see what else it may be doing outside of its canonical role of facilitating the editing of Apo-B messenger RNA.

It really took us on a little bit of a wild journey using different unbiased approaches to try to figure out some of the mechanisms that could be behind it.

Jane Ferguson: So you had to do a lot of different types of experiments to really get at this question.

Jenny Lin: Yeah. So again, one thing we wanted to see was: if you lose A1CF function, whether or not you would have differences in Apo-B 100-B48. We actually found that A1CF isn't even needed for that editing reaction and that our mice that we were able to create with crispr cas9 genome editing, so knocking in the mutation and knocking out the gene, actually have the phenotype even though they don't have changes in editing.

But what surprised us was that we know that A1CF as an RNA binding protein binds Apo-B transcript, yet it somehow does not alter transcriptional abundance of the Apo-B messenger RNA. And it has nothing to do with Apo-B synthesis so we basically had to think, what is A1CF doing outside of Apo-B biology?

We found that you have A1CF loss of function, you have increased triglycerides secretion. There is more Apo-B secretion, but that seems to be a downstream effect of other processes going on in the cell and to really try to figure out what those processes are, we had to take an unbiased approach using enhanced clipseek to figure out binding targets and also doing some transcriptional profiling with RNA sequencing and found that it's not necessarily regulating that transcriptum on a differential expression level, but there are some key alternative splicing events as well as messenger RNA binding to affect translational efficiency of some key targets that could be driving the biology.

Jane Ferguson: That's really interesting and you wouldn't have been able to find that by just looking at levels of protein or levels of mRNA, you really had to do these additional clipseek and some experiments to really get at this splicing.

Jenny Lin: Yeah, so it's been interesting. Clipseek is not as commonly performed method, so we had to collaborate with some brilliant people over at UCSD, to help us facilitate this. But again, finding that A1CF binds many more transcripts than Apo-B itself is a novel finding and the fact that it can regulate alternative splicing is also a very novel finding as well.

Jane Ferguson: So what was the most challenging part of this whole project?

Jenny Lin: I think the challenging part was that when we saw there wasn't necessarily a direct effect on Apo-B abundance and having to then cast this wide net and then figure out from all of the different unbiased data we have and integrating it find different pathways that may be relevant. In this case, it may all be relevant to ER stress, which is a field that is a little bit controversial in VLDL secretion in terms of directionality, but certainly is important in the biology.

Jane Ferguson: So is that something that you're going to have to start doing in the future? Are you going to start looking at ER stress or what kind of other experiments do you think you're going to keep doing to move this project forward?

Jenny Lin: Yeah, so actually, I think focusing in on A1CF as an RNA-binding protein and pursuing some of these additional targets will also be relevant, so I think in terms of ER stress, we could be looking at different targets, but there other processes going on in the cell that's mediated by A1CF, that could contribute maybe doing some isoform specific studies just to really prove that these alternative-splicing changes are driving some of the biology.

There's a lot of work to do as I would joke to anyone on study section listening to this, perhaps four to five years of work for an RO1.

Jane Ferguson: Sounds very appropriate.

Jenny Lin: Yeah, there's a lot of exciting work to do. A1CF is actually also a locus for other cardio-metabolic relevant traits such as uric acid, gout and kidney function so there could be something very interesting going on. There could be cross talk among cellular processes that could lead to these different phenotypes.

Jane Ferguson: Really interesting project and a lot of really great work. Congratulations again on being selected as finalist and on this really interesting paper.

Jenny Lin: Thank you.

Jane Ferguson: Thanks.

Sarah Parker is based in Cedar Sinai Medical Center in LA and her mentor is Dr. Jenny Van Eyk. The title of her presentation was "Identification of Putative Fibrous Plaque Marker Proteins by Unsupervised Deconvolution of Heterogeneous Vascular Proteomes ". And I apologize in advance for the quality of this recording. The background noise wasn't that noticeable at the time, but that recording really gives you that full immersive audio experience of a busy hotel lobby.

Hey Sarah. Thank you for joining us. Could you just take a few moments to introduce yourself to the audience?

Sarah Parker: So I'm Sarah Parker. I'm a project scientist at Cedar Sinai Medical Center where I'm doing work to study the basic mechanisms of vascular biology of various indolent conditions.

Jane Ferguson: So congratulations on being selected as a finalist for the Young Investigator Award. It's a great achievement. I'd love to hear a bit more about your project, how that started and what you found.

Sarah Parker: The work that I did was under the overarching umbrella of a project called the Genomic and Proteomic Architecture of Atherosclerosis. So with this project, we're using tissues that we're able to obtain from individuals who are young and have passed away from traumatic and violent and so non-cardiovascular causes of death. Because of the presence of atherosclerosis in the population, we get this range of lesion, both fatty streak and fibrous-plaque lesions in these asymptomatic or non-diseased individuals and this gives us this opportunity to do some molecular profiling to really try to find protein-signatures of early stage plaque formation, that could ultimately and hopefully be used for biomarker development.

Jane Ferguson: That's really cool and that's such a valuable sample resource.

Sarah Parker: Yeah so we've essentially, in this project I was able to set up a pipeline that enabled us to do these proteomics on such a large scale, because that's actually really difficult in label free quantitative proteomics and to use other forms becomes very expensive and cost-limiting.

So we were able to find a panel of proteins that we think are a putative early set of fibrous plaque markers and with this panel, we took them to see if any of these tissue derived markers would then be detectable and informative in plasma, because that's the next really big translational leap with these discovery-type data sets. Of our 58 initial candidates, we were able to detect 39 of them and about a handful 10-13 are showing informative behavior in the plasma of initial cohort of women with known coronary-artery disease.

Jane Ferguson: So out of the 58 that you first found, how many of them were potentially known to be involved in disease and how many were novel?

Sarah Parker: I would say, going through the list, it was probably about 50/50 in terms of background data that shows role as a biomarker, so there are a lot of apolipoproteins, which have all been characterized as potential biomarkers. There were a lot that could feasibly be linked through the literature to atherosclerosis. Most of them made a lot of sense, but having been proposed as potential biomarkers, some of them were more rare.

Jane Ferguson: Were there any of them that were sort of in different directions, let's say were elevated in tissue, but then were lower in plasma?

Sarah Parker: Funny you should ask. That actually has us scratching our heads a little bit right now. There were a couple of apolipoproteins that are more associated with HDL biology that we saw as being elevated in the tissue but then lower in the plasma [inaudible 00:23:34] so that's a really interesting observation so something about the role of these proteins to scavenge cholesterol and then once they're in the blood, they're cleared really quickly relative to normal, or something. So we're really trying to figure out what that biology means.

Jane Ferguson: Maybe if they're building up in the tissue, that's bad. But while in circulation, they're fine.

Sarah Parker: Yeah, maybe they're trapped in the circulation. We have a lot of exciting hypotheses to test along that front.

Jane Ferguson: So what's next? Are you following up some of these proteins?

Sarah Parker: Yep, so we have a huge discovery arm to the project where we're looking for more molecular mechanisms like why do we have these things in the tissue versus plasma and then we are working to really validate and optimize these multi-plexes in much more generalized large-scale populations to determine whether this strategy of instead of one or two biomarkers, more of a signature-style panel can be informative, especially as we try to press towards a precision medicine approach where different substratum might be informed by different protein signatures.

Jane Ferguson: Right, so you might have to have a specific panel based on sex or age or race or some other demographic.

Sarah Parker: Yes and to find those signatures, it's going to be very big numbers, with very accurate, careful quantitation.

Jane Ferguson: So you have a lot of work to do.

Sarah Parker: Yes.

Jane Ferguson: Alright, well thank you for talking to us and congratulations again.

Louie Wang, a cardiologist and PhD student came all the way from the Victor Chang Cardiac Research Institute in Syndey, Australia. His mentor is Dr. Diane Fatkin. The title of his talk is "A novel zebrafish model of human A-band truncated titan exhibits alternated ventricular diastolic compliance in vivo and reveals enhanced susceptibility to the effects of volume overload in mutation carriers.

So thank you for joining me. Could you take a few minutes to introduce yourself?

Louie Wang: So I'm Louie Wang. I'm a cardiologist based in Australia. I work and live in Sydney. I'm a PhD student at the Victor Chang Cardiac Research Institute and I'm an NHMRC (National Health and Medical Research Council and National Heart Foundation of Australia post-graduate scholar). I have previously been based at St. Vincent's Hospital.

Jane Ferguson: Great. So we'd love to hear a little bit in advance of what you're working on and what you're planning to present.

Louie Wang: So basically what I'm presenting is what I think is a different form of functional of genomics. What we're actually looking at is the impact of genetic changes, specific genetic change on function of the heart at an organ level. So there is a problem out there that is very common in cardiology and it's a big problem in cardiology and that is there are mutations in the sarcomere protein titan, truncating variants which actually are associated with dilated cardiomyopathy.

Now they're pretty common in idiopathic dilated cardiomyopathy, present in about 15-20% of the cases depending on which cohort study you look at. But they're also widely prevalent in the general population. Somewhere between 0.3 to 1% of the general population carries this truncating variants or various forms of this truncating variant.

So it's not sure whether these are disease-causing in their own right or if it's just a genetic susceptibility factor for heart failure and so what our work involves is that we actually, by chance, at St. Vincent's Hospital and at Victor Chang Cardiac Institute, two families who had the identical genetic truncation in the A-band region of his human titan gene where the individuals in the family, typically who carried the gene, typically developed systolic heart failure, which is a mild phenotype and occurred at middle age, but in two individuals, they developed severe onset accelerated disease trajectory in a very severe phenotype when exposed to conditions associated with chronic volume overload.

We suspect and this was a hypothesis, not only was this genetic-truncation disease-causing, but at volume overload was disease-modifying and given that volume overload is a very common condition present in birth, a lot physiological processes like lung endurance, exercise, pregnancy as well as a lot of pathological disease states in cardiovascular disease, this was actually a very important modifiable factor.

So what we did, was we created a novel zebrafish model of this human A-band truncated variant. We then studied the animals when they became adults to look at their heart structure and function and we used zebrafish echocardiography. So reversed translated all the techniques you can do in human echocardiography so they can be used in the zebrafish.

What we found was, yes, this animal, or heterozygotes developed dilated cardiomyopathy but also the volume overload exacerbated this condition. So this is a phenomenon that has conserved this by four hundred million years of vertebrate evolution so this is a pretty important mechanism.

Jane Ferguson: So what kind of next steps do you see for this project?

Louie Wang: So one thing is that we obviously have shown that there is an association with volume overload in precipitous disease. The corollary of our work is that perhaps interventions that could reduce volume load in these genetic susceptible individuals or alternatively in people who can't avoid volume overload. Because a lot of volume overload conditions can be modifiable and perhaps this could be protective and that would have wide-ranging population benefits.

Jane Ferguson: Thank you for sharing that soundbite of your work and good luck. Congratulations again on becoming a finalist.

Louie Wang: Thank you. ...

Jane Ferguson: Each of these four finalists gave compelling presentations of their research and the judges were highly impressed of the quality of the research and level of accomplishments of these early career investigators.

Just getting selected as a finalist for this award is a huge accomplishment. But there did have to be one winner. I'm delighted to announce that Jenny Lin was selected as the 2017 FGTB Young investor award winner. Congratulations, Jenny, and thanks to all four finalists for agreeing to appear on this podcast.

And that's all for this month. We'll be back at the end of December with a new episode. Subscribe to the podcast through iTunes or your favorite podcast app. to get new episodes delivered automatically and thank you for listening.