An integrated approach to identify environmental modulators of genetic risk factors for complex traits

Tradeoffs in Modeling Context Dependency in Complex Trait Genetics

Genetic effects on complex traits may depend on context, such as age, sex, environmental exposures or social settings. However, it is often unclear if the extent of context dependency, or Gene-by-Environment interaction (GxE), merits more involved models than the additive model typically used to analyze data from genome-wide association studies (GWAS). Here, we suggest considering the utility of GxE models in GWAS as a tradeoff between bias and variance parameters. In particular, We derive a decision rule for choosing between competing models for the estimation of allelic effects. The rule weighs the increased estimation noise when context is considered against the potential bias when context dependency is ignored. In the empirical example of GxSex in human physiology, the increased noise of context-specific estimation often outweighs the bias reduction, rendering GxE models less useful when variants are considered independently. However, we argue that for complex traits, the joint consideration of context dependency across many variants mitigates both noise and bias. As a result, polygenic GxE models can improve both estimation and trait prediction. Finally, we exemplify (using GxDiet effects on longevity in fruit flies) how analyses based on independently ascertained "top hits" alone can be misleading, and that considering polygenic patterns of GxE can improve interpretation.

The impact of exercise on gene regulation in association with complex trait genetics

Endurance exercise training is known to reduce risk for a range of complex diseases. However, the molecular basis of this effect has been challenging to study and largely restricted to analyses of either few or easily biopsied tissues. Extensive transcriptome data collected across 15 tissues during exercise training in rats as part of the Molecular Transducers of Physical Activity Consortium has provided a unique opportunity to clarify how exercise can affect tissue-specific gene expression and further suggest how exercise adaptation may impact complex disease-associated genes. To build this map, we integrate this multi-tissue atlas of gene expression changes with gene-disease targets, genetic regulation of expression, and trait relationship data in humans. Consensus from multiple approaches prioritizes specific tissues and genes where endurance exercise impacts disease-relevant gene expression. Specifically, we identify a total of 5523 trait-tissue-gene triplets to serve as a valuable starting point for future investigations [Exercise; Transcription; Human Phenotypic Variation].

Colocalization of corneal resistance factor GWAS loci with GTEx e/sQTLs highlights plausible candidate causal genes for keratoconus postnatal corneal stroma weakening

Background: Genome-wide association studies (GWAS) for corneal resistance factor (CRF) have identified 100s of loci and proved useful to uncover genetic determinants for keratoconus, a corneal ectasia of early-adulthood onset and common indication of corneal transplantation. In the current absence of studies to probe the impact of candidate causal variants in the cornea, we aimed to fill some of this knowledge gap by leveraging tissue-shared genetic effects. Methods: 181 CRF signals were examined for evidence of colocalization with genetic signals affecting steady-state gene transcription and splicing in adult, non-eye, tissues of the Genotype-Tissue Expression (GTEx) project. Expression of candidate causal genes thus nominated was evaluated in single cell transcriptomes from adult cornea, limbus and conjunctiva. Fine-mapping and colocalization of CRF and keratoconus GWAS signals was also deployed to support their sharing causal variants. Results and discussion: 26.5% of CRF causal signals colocalized with GTEx v8 signals and nominated genes enriched in genes with high and specific expression in corneal stromal cells amongst tissues examined. Enrichment analyses carried out with nearest genes to all 181 CRF GWAS signals indicated that stromal cells of the limbus could be susceptible to signals that did not colocalize with GTEx's. These cells might not be well represented in GTEx and/or the genetic associations might have context specific effects. The causal signals shared with GTEx provide new insights into mediation of CRF genetic effects, including modulation of splicing events. Functionally relevant roles for several implicated genes' products in providing tensile strength, mechano-sensing and signaling make the corresponding genes and regulatory variants prime candidates to be validated and their roles and effects across tissues elucidated. Colocalization of CRF and keratoconus GWAS signals strengthened support for shared causal variants but also highlighted many ways into which likely true shared signals could be missed when using readily available GWAS summary statistics.

Negative correlation between acetyl-CoA acyltransferase 2 and cetuximab resistance in colorectal cancer

The emergence of anti-EGFR therapy has revolutionized the treatment of colorectal cancer (CRC). However, not all patients respond consistently well. Therefore, it is imperative to conduct further research to identify the molecular mechanisms underlying the development of cetuximab resistance in CRC. In this study, we find that the expressions of many metabolism-related genes are downregulated in cetuximab-resistant CRC cells compared to their sensitive counterparts. Specifically, acetyl-CoA acyltransferase 2 (ACAA2), a key enzyme in fatty acid metabolism, is downregulated during the development of cetuximab resistance. Silencing of ACAA2 promotes proliferation and increases cetuximab tolerance in CRC cells, while overexpression of ACAA2 exerts the opposite effect. RTK-Kras signaling might contribute to the downregulation of ACAA2 expression in CRC, and ACAA2 predicts CRC prognosis in patients with Kras mutations. Collectively, our data suggest that modulating ACAA2 expression contributes to secondary cetuximab resistance in Kras wild-type CRC patients. ACAA2 expression is related to Kras mutation and demonstrates a prognostic role in CRC patients with Kras mutation. Thus, ACAA2 is a potential target in CRC with Kras mutation.

Stem Cell Models for Context-Specific Modeling in Psychiatric Disorders

Genome-wide association studies reveal the complex polygenic architecture underlying psychiatric disorder risk, but there is an unmet need to validate causal variants, resolve their target genes(s), and explore their functional impacts on disorder-related mechanisms. Disorder-associated loci regulate transcription of target genes in a cell type- and context-specific manner, which can be measured through expression quantitative trait loci. In this review, we discuss methods and insights from context-specific modeling of genetically and environmentally regulated expression. Human induced pluripotent stem cell-derived cell type and organoid models have uncovered context-specific psychiatric disorder associations by investigating tissue-, cell type-, sex-, age-, and stressor-specific genetic regulation of expression. Techniques such as massively parallel reporter assays and pooled CRISPR (clustered regularly interspaced short palindromic repeats) screens make it possible to functionally fine-map genome-wide association study loci and validate their target genes at scale. Integration of disorder-associated contexts with these patient-specific human induced pluripotent stem cell models makes it possible to uncover gene by environment interactions that mediate disorder risk, which will ultimately improve our ability to diagnose and treat psychiatric disorders.

A variability in response of osteoclasts to zoledronic acid is mediated by smoking-associated modification in the DNA methylome

BACKGROUND

Clinical trials have shown zoledronic acid as a potent bisphosphonate in preventing bone loss, but with varying potency between patients. Human osteoclasts ex vivo reportedly displayed a variable sensitivity to zoledronic acid > 200-fold, determined by the half-maximal inhibitory concentration (IC50), with cigarette smoking as one of the reported contributors to this variation. To reveal the molecular basis of the smoking-mediated variation on treatment sensitivity, we performed a DNA methylome profiling on whole blood cells from 34 healthy female blood donors. Multiple regression models were fitted to associate DNA methylation with ex vivo determined IC50 values, smoking, and their interaction adjusting for age and cell compositions.

RESULTS

We identified 59 CpGs displaying genome-wide significance (p < 1e-08) with a false discovery rate (FDR) < 0.05 for the smoking-dependent association with IC50. Among them, 3 CpGs have p < 1e-08 and FDR < 2e-03. By comparing with genome-wide association studies, 15 significant CpGs were locally enriched (within < 50,000 bp) by SNPs associated with bone and body size measures. Furthermore, through a replication analysis using data from a published multi-omics association study on bone mineral density (BMD), we could validate that 29 out of the 59 CpGs were in close vicinity of genomic sites significantly associated with BMD. Gene Ontology (GO) analysis on genes linked to the 59 CpGs displaying smoking-dependent association with IC50, detected 18 significant GO terms including cation:cation antiporter activity, extracellular matrix conferring tensile strength, ligand-gated ion channel activity, etc. CONCLUSIONS: Our results suggest that smoking mediates individual sensitivity to zoledronic acid treatment through epigenetic regulation. Our novel findings could have important clinical implications since DNA methylation analysis may enable personalized zoledronic acid treatment.

The missing link between genetic association and regulatory function

The genetic basis of most traits is highly polygenic and dominated by non-coding alleles. It is widely assumed that such alleles exert small regulatory effects on the expression of cis-linked genes. However, despite the availability of gene expression and epigenomic datasets, few variant-to-gene links have emerged. It is unclear whether these sparse results are due to limitations in available data and methods, or to deficiencies in the underlying assumed model. To better distinguish between these possibilities, we identified 220 gene-trait pairs in which protein-coding variants influence a complex trait or its Mendelian cognate. Despite the presence of expression quantitative trait loci near most GWAS associations, by applying a gene-based approach we found limited evidence that the baseline expression of trait-related genes explains GWAS associations, whether using colocalization methods (8% of genes implicated), transcription-wide association (2% of genes implicated), or a combination of regulatory annotations and distance (4% of genes implicated). These results contradict the hypothesis that most complex trait-associated variants coincide with homeostatic expression QTLs, suggesting that better models are needed. The field must confront this deficit and pursue this 'missing regulation.'

SARS-CoV-2 infection impairs the insulin/IGF signaling pathway in the lung, liver, adipose tissue, and pancreatic cells via IRF1

BACKGROUND

COVID-19 can cause multiple organ damages as well as metabolic abnormalities such as hyperglycemia, insulin resistance, and new onset of diabetes. The insulin/IGF signaling pathway plays an important role in regulating energy metabolism and cell survival, but little is known about the impact of SARS-CoV-2 infection. The aim of this work was to investigate whether SARS-CoV-2 infection impairs the insulin/IGF signaling pathway in the host cell/tissue, and if so, the potential mechanism and association with COVID-19 pathology.

METHODS

To determine the impact of SARS-CoV-2 on insulin/IGF signaling pathway, we utilized transcriptome datasets of SARS-CoV-2 infected cells and tissues from public repositories for a wide range of high-throughput gene expression data: autopsy lungs from COVID-19 patients compared to the control from non-COVID-19 patients; lungs from a human ACE2 transgenic mouse infected with SARS-CoV-2 compared to the control infected with mock; human pluripotent stem cell (hPSC)-derived liver organoids infected with SARS-CoV-2; adipose tissues from a mouse model of COVID-19 overexpressing human ACE2 via adeno-associated virus serotype 9 (AAV9) compared to the control GFP after SARS-CoV-2 infection; iPS-derived human pancreatic cells infected with SARS-CoV-2 compared to the mock control. Gain and loss of IRF1 function models were established in HEK293T and/or Calu3 cells to evaluate the impact on insulin signaling. To understand the mechanistic regulation and relevance with COVID-19 risk factors, such as older age, male sex, obesity, and diabetes, several transcriptomes of human respiratory, metabolic, and endocrine cells and tissue were analyzed. To estimate the association with COVID-19 severity, whole blood transcriptomes of critical patients with COVID-19 compared to those of hospitalized noncritical patients with COVID-19.

RESULTS

We found that SARS-CoV-2 infection impaired insulin/IGF signaling pathway genes, such as IRS, PI3K, AKT, mTOR, and MAPK, in the host lung, liver, adipose tissue, and pancreatic cells. The impairments were attributed to interferon regulatory factor 1 (IRF1), and its gene expression was highly relevant to risk factors for severe COVID-19; increased with aging in the lung, specifically in men; augmented by obese and diabetic conditions in liver, adipose tissue, and pancreatic islets. IRF1 activation was significantly associated with the impaired insulin signaling in human cells. IRF1 intron variant rs17622656-A, which was previously reported to be associated with COVID-19 prevalence, increased the IRF1 gene expression in human tissue and was frequently found in American and European population. Critical patients with COVID-19 exhibited higher IRF1 and lower insulin/IGF signaling pathway genes in the whole blood compared to hospitalized noncritical patients. Hormonal interventions, such as dihydrotestosterone and dexamethasone, ameliorated the pathological traits in SARS-CoV-2 infectable cells and tissues.

CONCLUSIONS

The present study provides the first scientific evidence that SARS-CoV-2 infection impairs the insulin/IGF signaling pathway in respiratory, metabolic, and endocrine cells and tissues. This feature likely contributes to COVID-19 severity with cell/tissue damage and metabolic abnormalities, which may be exacerbated in older, male, obese, or diabetic patients.

Integration of genetic colocalizations with physiological and pharmacological perturbations identifies cardiometabolic disease genes

BACKGROUND

Identification of causal genes for polygenic human diseases has been extremely challenging, and our understanding of how physiological and pharmacological stimuli modulate genetic risk at disease-associated loci is limited. Specifically, insulin resistance (IR), a common feature of cardiometabolic disease, including type 2 diabetes, obesity, and dyslipidemia, lacks well-powered genome-wide association studies (GWAS), and therefore, few associated loci and causal genes have been identified.

METHODS

Here, we perform and integrate linkage disequilibrium (LD)-adjusted colocalization analyses across nine cardiometabolic traits (fasting insulin, fasting glucose, insulin sensitivity, insulin sensitivity index, type 2 diabetes, triglycerides, high-density lipoprotein, body mass index, and waist-hip ratio) combined with expression and splicing quantitative trait loci (eQTLs and sQTLs) from five metabolically relevant human tissues (subcutaneous and visceral adipose, skeletal muscle, liver, and pancreas). To elucidate the upstream regulators and functional mechanisms for these genes, we integrate their transcriptional responses to 21 relevant physiological and pharmacological perturbations in human adipocytes, hepatocytes, and skeletal muscle cells and map their protein-protein interactions.

RESULTS

We identify 470 colocalized loci and prioritize 207 loci with a single colocalized gene. Patterns of shared colocalizations across traits and tissues highlight different potential roles for colocalized genes in cardiometabolic disease and distinguish several genes involved in pancreatic β-cell function from others with a more direct role in skeletal muscle, liver, and adipose tissues. At the loci with a single colocalized gene, 42 of these genes were regulated by insulin and 35 by glucose in perturbation experiments, including 17 regulated by both. Other metabolic perturbations regulated the expression of 30 more genes not regulated by glucose or insulin, pointing to other potential upstream regulators of candidate causal genes.

CONCLUSIONS

Our use of transcriptional responses under metabolic perturbations to contextualize genetic associations from our custom colocalization approach provides a list of likely causal genes and their upstream regulators in the context of IR-associated cardiometabolic risk.