24 January 2019

Getting Personal: Omics of the Heart

Jane Ferguson: Hello everyone, and happy new year. Welcome to episode 24 of Getting Personal: Omics of the Heart. It's January 2019, I am Jane Ferguson, an assistant professor at Vanderbilt University Medical Center and an associate editor at Circulation Genomic and Precision Medicine. We have a great line-up of papers this month in the journal, so let's jump right into the articles.

First up, a paper from Christopher Nelson, Nilesh Samani, and colleagues from the University of Lester entitled, "Genetic Assessment of Potential Long-Term On-Target Side Effects of PCSK9 Inhibitors."

I think most listeners are well aware of the efficacy of PCSK9 inhibition in reducing cardiovascular risk. However, as a relatively new treatment option, we do not yet have data on potential long-term side effects of PCSK9 inhibition. In this study, they utilized genetics as a proxy to understand potential long-term consequences of lower PCSK9 activity. They examined a PCSK9 variant that associates with lower LDL, as well as examining two LDL-lowering variants in HMGCR, the target of statins, which served as a positive control of sorts. They used data from over 479,000 individuals in the UK Biobank and looked for associations between the three LDL-lowering variants and 80 different phenotypes.

For the PCSK9 variant, the allele which is associated with lower LDL was significantly associated with the higher risk of type 2 diabetes, higher BMI, higher waist circumference, higher waist-hip ratio, higher diastolic blood pressure, as well as increased risk of type 2 diabetes and insulin use. The HMGCR variants were similarly associated with type 2 diabetes as expected.

Mediation analysis suggested that the effect of the PCSK9 variant on type 2 diabetes is independent of its effect on obesity. There were nominal associations between the PCSK9 variant and other diseases, including depression, asthma, chronic kidney disease, venous thromboembolism, and peptic ulcer.

While genetics cannot fully recapitulate the information that would be gained from long-term clinical follow up, these data suggest that like statins, PCSK9 inhibition may increase the risk of diabetes and potentially other disease. Overall, the cardiovascular efficacy of PCSK9 inhibition may outweigh these other risks, however, future studies should carefully examine these potential side effects.

Next up, we have a paper from Xiao Cui, Fang Qin, Xinping Tian, Jun Cai, and colleagues from Peking Uni and Medical College, on "Novel Biomarkers for the Precise Diagnoses and Activity Classification of Takayasu's Arteritis."

They were interested in identifying protein biomarkers of Takayasu arteritis, to improve diagnosis and understanding of disease activity in this chronic vascular disease. They ran a proteomic panel including 440 cytokines on 90 individuals, including individuals with active disease, inactive disease, and healthy controls. They found a number of candidates and validated one protein, TIMP-1, as a specific diagnostic biomarker for Takayasu arteritis. For assessing disease activity, there was no single biomarker that could be used for classification, however, the combination of eight different cytokines identified through random forest-based recursive feature elimination and [inaudible] regression, including CA 125, FLRG, IGFBP-2, CA15-3, GROa, LYVE-1, ULBP-2, and CD 99, were able to accurately discriminate disease activity versus inactivity.

Overall, this study was able to identify novel biomarkers that could be used for improved diagnosis and assessment of Takayasu arteritis, and may give some clues as to the mechanisms of pathogenesis.

Our next paper is entitled, "Familial Sinus Node Disease Caused By Gain of GIRK Channel Function," and comes from Johanna Kuß, Birgit Stallmeyer, Marie-Cécile Kienitz, and Eric Schulze-Bahr, from University Hospital Münster. They were interested in understanding novel genetic underpinnings of inherited sinus node dysfunction.

A recent study identified a gain of function mutation in GNB2 associated with sinus node disease. This mutation led to enhanced activation of the G-protein activated inwardly rectifying potassium channel, or GIRK, prompting the researchers to focus their interest on the genes encoding the GIRK subunits, KCNJ3 and KCNJ5. They sequenced both genes in 52 patients with idiopathic sinus node disease, and then carried out whole exome sequencing in family members of patients with potential disease variants in either gene. They identified a non-synonymous variant in KCNJ5, which was not present in the EVS or ExAC databases, and which segregated with disease in the affected family. This variant was associated with increased GIRK currents in a cell system, and in silico models, predicted the variant altered or spermine binding site within the GIRK channel. Thus, this study demonstrated that a gain of function mutation in a GIRK channel subunit associates with sinus node disease, and suggests that modulation of GIRK channels may be a viable therapeutic target for cardiac pacemaking.

Our next paper, "Key Value of RNA Analysis of MYBPC3 Splice-Site Variants in Hypertrophic Cardiomyopathy," comes from Emma Singer, Richard Bagnall, and colleagues from the Centenary Institute and the University of Sydney.

They wanted to understand the impact of variants in MYBCP3, a known hypertrophic cardiomyopathy gene, on splicing. They recruited individuals with a clinical diagnosis of hypertrophic cardiomyopathy and genetic testing of cardiomyopathy-related genes. They further examined individuals with a variant in MYBCP3 which had an in silico prediction to affect splicing. They sequenced RNA from blood or from fixed myocardial tissue and assessed the relationship between each DNA variant and gene splicing variation.

Of 557 subjects, 10% carried rare splice site variants. Of 29 potential variants identified, they examined 9 which were predicted to affect splicing, and found that 7 of these were indeed associated with splicing errors.

Going back to the families, they were able to reclassify four variants in four families from uncertain clinical significance to likely pathogenic, demonstrating the utility of using RNA analysis to understand pathogenicity in genetic testing.

The next paper this issue comes from Catriona Syme, Jean Shin, Zdenka Pausova, and colleagues from the University of Toronto, and is entitled, "Epigenetic Loci of Blood Pressure: Underlying Hemodynamics in Adolescents and Adults."

A recent large meta epigenome-wide association study identified methylation loci that associate with blood pressure. In this study, they wanted to understand more about how these loci related to blood pressure and hemodynamics. They recruited adolescents and middle-aged adults and assessed 13 CPG loci for associations with hemodynamic markers, including systolic and diastolic blood pressure, heart rate, stroke volume, and total peripheral resistance, measured over almost an hour during normal activities. Several of the loci replicated associations with blood pressure, and two of these also showed age-specific associations with hemodynamic variables. One site in PHGDH was particularly associated with blood pressure and stroke volume in adolescents, as well as with body weight and BMI, where lower methylation resulting in higher gene expression associated with higher blood pressure. A second site in SLC7A11 associated with blood pressure in adults but not adolescents, with lower methylation and consequent higher gene expression associated with increased blood pressure.

Overall, this study indicates that methylation mediated changes in gene expression may modulate blood pressure and hemodynamic responses in an age-dependent manner.

Next up is a research letter from Ben Brumpton, Cristen Willer, George Davey Smith, Bjørn Olav Åsvold, and colleagues from the Norwegian University of Science and Technology, entitled, "Variation in Serum PCSK9, Cardiovascular Disease Risk, and an Investigation of Potential Unanticipated Effects of PCSK9 Inhibition: A GWAS and Mendelian Randomization Study in the Nord-Trøndelag Health Study, Norway."

As we heard about from the first study this issue, the long-term side effects of PCSK9 inhibition remain unknown. In this study, they also applied a genetic approach to understand potential unanticipated consequences of PCSK9 inhibition. They analyzed phenotypes from over 69,000 participants in the Nord-Trøndelag Health Study and measured serum PCSK9 in a subset.

In PCSK9 GWAS of over 3,600 people, with replication in over 5,000 individuals from the twin gene study. They defined a genetic risk score for serum PCSK9 and assessed the relationship between genetically predicted PCSK9 and outcomes. They saw the expected associations between lower PCSK9 and lower LDL and coronary heart disease risk. However, there was minimal evidence for associations with other outcomes.

While our first study in this issue, from Nelson, et al, found that lower PCSK9 from a single genetic variant was associated with higher diabetes risk, this risk was not found here using the genetic risk score. Differences in the genetic definitions and in the populations used can perhaps explain these differences between the two studies, but overall, the studies are consistent in suggesting that long-term PCSK9 inhibition is unlikely to be associated with major adverse outcomes.

Our second research letter comes from Young-Chang Kwon, Bo Kyung Sim, Jong-Keuk Lee, and colleagues from Asan Medical Center in Seoul, on behalf of the Korean Kawasaki Disease Genetics Consortium. The title is, "HLA-B54:01 is Associated with Susceptibility to Kawasaki Disease," and reports on novel Kawasaki disease variants.

HLA genes have been previously associated with disease, and in this report, the authors sequenced selected axons in HLA-DRB1, HLA-DQB1, HLA-A, HLA-B, HLA-C, and HLA-DBP1 in 160 Kawasaki disease patients and 278 controls. They find a significant association with HLA-B, and replicated this in a sample of 618 Kawasaki disease patients, compared with over 14,000 in-house controls. They identified specific amino acid residues conferring disease susceptibility, highlighting HLA-B as a potential modulator of Kawasaki disease.

Our third and final research letter concerns "Serum Magnesium and Calcium Levels and Risk of Atrial Fibrillation: a Mendelian Randomization Study," and comes to us from Susanna Larsson, Nikola Drca, and Karl Michaëlsson, from the Karolinska Institute.

Because magnesium and calcium are known to influence atrial fibrillation, this group was interested in whether genetic predictors of serum methyls associated with disease. They constructed genetic predictors from GWAS of calcium in over 61,000 individuals, and GWAS of magnesium in over 23,000 individuals. They applied these predictors to an AF GWAS including over 65,000 cases and over 522,000 controls. Genetically predicted magnesium was inversely associated with atrial fibrillation, while there was no association with genetically predicted calcium.

While this study does not definitively prove causality, future studies aimed at assessing whether dietary or other strategies to raise serum magnesium are protective against AF may yield novel strategies for disease prevention.

And that's it from us for this month. Thank you for listening, and come back next month for more from Circulation Genomic and Precision Medicine.

This podcast was brought to you by Circulation Genomic and Precision Medicine and the American Heart Association Council on Genomic and Precision Medicine. This program is copyright American Heart Association, 2019.