Loss-of-function genomic variants highlight potential therapeutic targets for cardiovascular disease

Jonas B. Nielsen, Oren Rom, Ida Surakka, Sarah E. Graham, Wei Zhou, Tanmoy Roychowdhury, Lars G. Fritsche, Sarah A. Gagliano Taliun, Carlo Sidore, Yuhao Liu, Maiken E. Gabrielsen, Anne Heidi Skogholt, Brooke N. Wolford, William Overton, Ying Zhao, Jin Chen, He Zhang, Whitney Hornsby, Akua Acheampong, Austen Grooms, Amanda M. Schaefer, Gregory J. M. Zajac, Luis Villacorta, Jifeng Zhang, Ben Brumpton, Mari Løset, Vivek Rai, Pia R. Lundegaard, Morten S. Olesen, Kent D. Taylor, Nicholette D. Palmer, Yii-Der Chen, Seung Hoan Choi, Steven A. Lubitz, Patrick T. Ellinor, Kathleen C. Barnes, Michelle Daya, Nicholas Rafaels, Scott T. Weiss, Jessica Lasky‐Su, Russell P. Tracy, Ramachandran S. Vasan, L. Adrienne Cupples, Rasika A. Mathias, Lisa R. Yanek, Lewis C. Becker, Patricia A. Peyser, Lawrence F. Bielak, Jennifer A. Smith, Stella Aslibekyan, Bertha Hidalgo, Donna K. Arnett, Marguerite R. Irvin, James G. Wilson, Solomon K. Musani, Adolfo Correa, Stephen S. Rich, Xiuqing Guo, Jerome I. Rotter, Barbara A. Konkle, Jill M. Johnsen, Allison E. Ashley‐Koch, Marilyn J. Telen, Vivien Sheehan, John Blangero, Joanne E. Curran, Juan M. Peralta, Courtney G. Montgomery, Wayne H-h Sheu, Ren-Hua Chung, Karen Schwander, Mehdi Nouraie, Victor R. Gordeuk, Yingze Zhang, Charles Kooperberg, Alexander P. Reiner, Rebecca D. Jackson, Eugene R. Bleecker, Deborah A. Meyers, Xingnan Li, Sayantan Das, Ketian Yu, Jonathon LeFaive, Albert V. Smith, Tom Blackwell, Daniel Taliun, Sebastian Zöllner, Lukas Forer, Sebastian Schoenherr, Christian Fuchsberger, Anita Pandit, Matthew Zawistowski, Sachin Kheterpal, Chad M. Brummett, Pradeep Natarajan, David Schlessinger, Seunggeun Lee, Hyun Min Kang, Francesco Cucca, Oddgeir L. HolmenNature Communications 2020 · DOI

TL;DR

This study identifies genetic variants that protect against cardiovascular disease (CVD) without increasing the risk of liver disease. By analyzing a large Norwegian cohort, researchers found 76 variants, including 9 that cause a loss-of-function in their respective proteins, that are linked to improved blood lipid profiles. One promising target, ZNF529, when inhibited, led to increased LDL receptor expression and uptake in liver cells, suggesting its potential for treating dyslipidemia and CVD.

What Problem Does This Paper Address?

Cardiovascular disease (CVD) remains a leading cause of death globally, and lowering blood lipids is a key strategy for risk reduction. However, many existing lipid-lowering drugs can increase the risk of adverse effects like fatty liver disease. Furthermore, a significant number of novel drug candidates fail in clinical trials due to a lack of predictive preclinical models and insufficient understanding of long-term consequences. This paper aims to address this by identifying genetic variants that confer a beneficial lipid profile and reduce CVD risk, while crucially avoiding detrimental impacts on liver function, thereby highlighting safer and more effective therapeutic targets.

Methods

The researchers employed a multi-pronged genomic approach to identify potential drug targets. This included:

  • Low-coverage whole-genome sequencing (WGS): Used on a subset of participants to identify rare, region-specific variants.
  • Targeted genotyping: Expanded on WGS findings, including rare variants.
  • Deep genotype imputation: Leveraged a large multi-ethnic reference panel (TOPMed) to impute variants across the genome.
  • Genome-wide association studies (GWAS): Conducted on a large cohort to identify associations between variants and blood traits.
  • Stepwise conditional analyses: Refined associations to identify independent signals.
  • Trans-ancestry meta-analyses: Combined data from Norwegian, Japanese, and Sardinian populations to increase power and fine-map loci.
  • Custom array genotyping: Focused on specific variants of interest identified through WGS or clinical relevance.
  • In vitro functional characterization: Used human hepatoma cells (HepG2) to investigate the biological mechanisms of promising candidate genes, including siRNA-mediated gene silencing, RNA sequencing, qPCR, and western blotting.

Datasets

The primary dataset used was the HUNT Study from Norway, comprising approximately 69,479 participants for GWAS and up to 57,060 for custom array genotyping. Additional datasets used for trans-ancestry meta-analyses included:

  • SardiNIA cohort
  • Biobank Japan

For phenome-wide association studies (PheWAS), the UK Biobank (N = 408,961 participants) was utilized.

Key Insights and Ideas

  • Genetic support for drug efficacy: The study reinforces the idea that drugs with strong genetic support are more likely to succeed in clinical trials. By identifying genetic variants that naturally confer protection, they are pinpointing genes that are likely to be safe and effective targets.
  • Prioritizing safety alongside efficacy: A core innovation is the explicit focus on identifying variants that improve lipid profiles without negatively impacting liver function. This dual criterion is crucial for developing safer CVD therapies.
  • Loss-of-function (LoF) variants as key indicators: The identification of LoF variants is particularly valuable. These variants often represent a complete absence of protein function, making their associated phenotypes more definitive and their corresponding genes more attractive drug targets for inhibition.
  • ZNF529 as a novel therapeutic target: The detailed functional characterization of ZNF529:p.K405X highlights its role in regulating LDL cholesterol. Its inhibition leading to increased LDL receptor expression and uptake in liver cells is a significant finding, suggesting a novel pathway for lowering LDL-C.
  • Cross-trait analysis for comprehensive understanding: Integrating associations with blood traits, disease endpoints (via PheWAS), and in vitro experiments provides a robust framework for prioritizing targets and understanding potential side effects.

Key Results

  • 76 presumed causal protein-altering variants identified: These variants are associated with one or more CVD- or liver-related blood traits.
  • 11 previously unreported variants: This includes 9 variants predicted to cause loss-of-function (LoF).
  • ZNF529:p.K405X variant: Associated with significantly decreased LDL cholesterol (P = 1.3 × 10⁻⁸) without affecting liver enzymes or blood glucose.
  • ZNF529 silencing in HepG2 cells: Led to a significant upregulation of the LDL receptor (LDLR) mRNA and protein, and increased DiI-LDL uptake.
  • 24 variants associated with disease phenotypes: Out of 64 imputed variants, 24 showed statistically significant associations with one or more disease groups in the UK Biobank PheWAS.
  • Two variants with favorable lipid and disease profiles: COBLL1:p.N497D (decreased TG, reduced T2D risk) and LPL:p.S474X (decreased CAD, T2D, hypertension risk) were identified as promising targets.

Limitations and Future Work

  • Rodent homolog limitation for ZNF529: The absence of a ZNF529 homolog in rodents limits the use of traditional animal models for further mechanistic studies.
  • Low frequency of some variants: The low allele frequency of some promising variants (e.g., ZNF529:p.K405X) made replication in other large cohorts challenging, necessitating the use of specialized genotyping arrays.
  • Further mechanistic studies: While ZNF529's role in LDLR regulation is shown, further studies are warranted to fully elucidate the precise mechanisms by which ZNF529 regulates LDLR and LDL uptake in the liver.
  • Exploring other identified targets: The study identified numerous other promising variants and genes that warrant further investigation for their therapeutic potential in CVD and metabolic diseases.
  • Clinical validation: The ultimate validation of these targets will require preclinical and clinical studies to confirm their safety and efficacy in humans.

Summary updated

Abstract

Pharmaceutical drugs targeting dyslipidemia and cardiovascular disease (CVD) may increase the risk of fatty liver disease and other metabolic disorders. To identify potential novel CVD drug targets without these adverse effects, we perform genome-wide analyses of participants in the HUNT Study in Norway (n = 69,479) to search for protein-altering variants with beneficial impact on quantitative blood traits related to cardiovascular disease, but without detrimental impact on liver function. We identify 76 (11 previously unreported) presumed causal protein-altering variants associated with one or more CVD- or liver-related blood traits. Nine of the variants are predicted to result in loss-of-function of the protein. This includes ZNF529:p.K405X, which is associated with decreased low-density-lipoprotein (LDL) cholesterol (P = 1.3 × 10<sup>-8</sup>) without being associated with liver enzymes or non-fasting blood glucose. Silencing of ZNF529 in human hepatoma cells results in upregulation of LDL receptor and increased LDL uptake in the cells. This suggests that inhibition of ZNF529 or its gene product should be prioritized as a novel candidate drug target for treating dyslipidemia and associated CVD.
BibTeX
@inproceedings{nielsen2020loss,
  author = {Jonas B. Nielsen and Oren Rom and Ida Surakka and Sarah E. Graham and Wei Zhou and Tanmoy Roychowdhury and Lars G. Fritsche and Sarah A. Gagliano Taliun and Carlo Sidore and Yuhao Liu and Maiken E. Gabrielsen and Anne Heidi Skogholt and Brooke N. Wolford and William Overton and Ying Zhao and Jin Chen and He Zhang and Whitney Hornsby and Akua Acheampong and Austen Grooms and Amanda M. Schaefer and Gregory J. M. Zajac and Luis Villacorta and Jifeng Zhang and Ben Brumpton and Mari Løset and Vivek Rai and Pia R. Lundegaard and Morten S. Olesen and Kent D. Taylor and Nicholette D. Palmer and Yii-Der Chen and Seung Hoan Choi and Steven A. Lubitz and Patrick T. Ellinor and Kathleen C. Barnes and Michelle Daya and Nicholas Rafaels and Scott T. Weiss and Jessica Lasky‐Su and Russell P. Tracy and Ramachandran S. Vasan and L. Adrienne Cupples and Rasika A. Mathias and Lisa R. Yanek and Lewis C. Becker and Patricia A. Peyser and Lawrence F. Bielak and Jennifer A. Smith and Stella Aslibekyan and Bertha Hidalgo and Donna K. Arnett and Marguerite R. Irvin and James G. Wilson and Solomon K. Musani and Adolfo Correa and Stephen S. Rich and Xiuqing Guo and Jerome I. Rotter and Barbara A. Konkle and Jill M. Johnsen and Allison E. Ashley‐Koch and Marilyn J. Telen and Vivien Sheehan and John Blangero and Joanne E. Curran and Juan M. Peralta and Courtney G. Montgomery and Wayne H-H Sheu and Ren-Hua Chung and Karen Schwander and Mehdi Nouraie and Victor R. Gordeuk and Yingze Zhang and Charles Kooperberg and Alexander P. Reiner and Rebecca D. Jackson and Eugene R. Bleecker and Deborah A. Meyers and Xingnan Li and Sayantan Das and Ketian Yu and Jonathon LeFaive and Albert V. Smith and Tom Blackwell and Daniel Taliun and Sebastian Zöllner and Lukas Forer and Sebastian Schoenherr and Christian Fuchsberger and Anita Pandit and Matthew Zawistowski and Sachin Kheterpal and Chad M. Brummett and Pradeep Natarajan and David Schlessinger and Seunggeun Lee and Hyun Min Kang and Francesco Cucca and Oddgeir L. Holmen},
  title = {{Loss-of-function genomic variants highlight potential therapeutic targets for cardiovascular disease}},
  booktitle = {Nature Communications},
  year = {2020},
  doi = {10.1038/s41467-020-20086-3},
}
Figure