Advertisement
Research Article
Transparent Process
Open Access

De novo discovery of traits co-occurring with chronic obstructive pulmonary disease

Evgeniia Golovina, Tayaza Fadason, Rachel K Jaros, Haribalan Kumar, Joyce John, Kelly Burrowes, Merryn Tawhai, Justin M O’Sullivan  Correspondence email
Published 27 December 2022. DOI: 10.26508/lsa.202201609

Abstract

Chronic obstructive pulmonary disease (COPD) is a heterogeneous group of chronic lung conditions. Genome-wide association studies have identified single-nucleotide polymorphisms (SNPs) associated with COPD and the co-occurring conditions, suggesting common biological mechanisms underlying COPD and these co-occurring conditions. To identify them, we have integrated information across different biological levels (i.e., genetic variants, lung-specific 3D genome structure, gene expression and protein–protein interactions) to build lung-specific gene regulatory and protein–protein interaction networks. We have queried these networks using disease-associated SNPs for COPD, unipolar depression and coronary artery disease. COPD-associated SNPs can control genes involved in the regulation of lung or pulmonary function, asthma, brain region volumes, cortical surface area, depressed affect, neuroticism, Parkinson’s disease, white matter microstructure and smoking behaviour. We describe the regulatory connections, genes and biochemical pathways that underlay these co-occurring trait-SNP-gene associations. Collectively, our findings provide new avenues for the investigation of the underlying biology and diverse clinical presentations of COPD. In so doing, we identify a collection of genetic variants and genes that may aid COPD patient stratification and treatment.

Introduction

Chronic obstructive pulmonary disease (COPD) is a heterogeneous group of chronic lung conditions that are characterized by persistent respiratory symptoms and airflow limitation due to airway and/or alveolar abnormalities (2020 Global Initiative for Chronic Obstructive Obstructive Lung Disease, 2021). These abnormalities are caused by a combination of distinct pathophysiological processes that result in diverse clinical presentations, responses to treatment, and patterns of progression. According to the World Health Organization, COPD accounted for more than 3.23 million deaths in 2019 and remains the third leading cause of death worldwide (2020 Global Initiative for Chronic Obstructive Obstructive Lung Disease, 2021).

Given the widespread exposure to the environmental factors (e.g., smoking, indoor and outdoor air pollution, childhood respiratory infections) that contribute to the development of COPD, it is striking that most individuals will never develop COPD. The variance in individual susceptibility to COPD can be partly explained by genetic factors. The estimated genetic heritability of COPD ranges from 20% to 40% (Zhou et al, 2013; Gim et al, 2020; Stolz, 2020).

Co-occurring conditions are widely recognised as impacting on COPD patient outcomes (Cavaillès et al, 2013; Burke & Wilkinson, 2021). As such, a better understanding of COPD co-occurring conditions is essential to enable effective management, therapeutic optimization and reduce the costs of managing COPD patients (Mannino et al, 2015). Epidemiological and genetic studies have reported that beyond respiratory impairment COPD-associated co-occurring conditions include coronary artery disease (CAD), lung cancer, osteoporosis, mental health problems such as anxiety, unipolar depression (UD), Alzheimer’s disease (AD), and Parkinson’s disease (PD) (Cavaillès et al, 2013; Li et al, 2015; Ställberg et al, 2018; Xia et al, 2020; Burke & Wilkinson, 2021; Carmona-Pírez et al, 2021; Martucci et al, 2021). The presence of these conditions in COPD patients indicates that common or interacting biological mechanisms underlie these conditions.

To date, genome-wide association studies (GWASs) have identified common single nucleotide polymorphisms (SNPs) that are associated with COPD, or its individual co-occurring conditions (Sakornsakolpat et al, 2019; Shrine et al, 2019; Zhu et al, 2019; Kim et al, 2021b). Most of the COPD-associated SNPs are located within the non-coding genome. Therefore, the impacts that these SNPs have on the biological pathways and processes underlying the development of COPD remain unclear. It is possible that the COPD-associated SNPs mark regulatory regions (i.e., expression quantitative trait loci [eQTLs]) that are associated with tissue-specific gene expression. eQTLs can interact with their target genes in three dimensions, forming spatial eQTL–gene regulatory connections that span the genome (e.g., cis, ≤1 Mb on the same chromosome; trans-intrachromosomal, >1 Mb on the same chromosome; or trans-interchromosomal, between different chromosomes). These spatial interactions are cell and tissue type–specific (GTEx Consortium, 2020; Halow et al, 2021). As lung is the primary affected tissue in COPD, integrating lung-specific spatial chromatin interactions and eQTL information may help us understand how SNPs impact biological pathways that increase an individual’s risk of developing COPD.

Little is known about functional relationships between genes and phenotypes in the lung. However, gene regulation is widely understood to occur through the combinatorial action of regulatory elements, transcription factors and genes within complex networks (i.e., gene regulatory network [GRN]) (Buenrostro et al, 2018; Chen et al, 2021a, 2021b; Zaied et al, 2022 Preprint). Moreover, genes encode proteins that physically interact with each other to form a complex protein–protein interaction network (PPIN) that responds to biological and environmental signals. Here, we integrated COPD-associated SNPs with: (1) information on the genome organization within the lung; and (2) lung-specific eQTL information to identify genes that are spatially regulated within the lung tissue. We integrated information across a lung-specific GRN and PPIN to identify conditions that were co-occurring with COPD. Collectively, our results highlight potential regulatory mechanisms and pathways important for COPD etiology. These results open a new avenue towards understanding the diverse clinical presentations of COPD and patient stratification.

Results

COPD-associated SNPs mark putative regulatory regions in the lung

COPD-associated SNPs (P < 5 × 10−8, n = 263) were downloaded from the GWAS Catalog (Tables S1 and S2) and run through the CoDeS3D pipeline (Fig 1A). ∼96% of the identified eQTLs were located within non-coding genomic regions, with 66.02% and 18.45% of them being intronic and intergenic, respectively (Fig S1A and B [wANNOVAR annotation], Table S2 [original GWAS Catalog annotation]). Analysis of these SNPs using the CoDeS3D (Fadason et al, 2018) pipeline (Fig 1A) identified 103 eQTLs and 107 genes that are involved in 151 significant (FDR < 0.01) eQTL–gene interactions within the lung (Fig S1 and Table S3). Most COPD-associated eQTLs (n = 67) are involved in one-to-one, 26 eQTLs—in one-to-two and 8 eQTLs—in one-to-three eQTL–gene regulatory interactions (Fig S1C and Table S3). Only two eQTLs (i.e., rs2277027 and rs9435731) were associated with the regulation of ≥4 genes (i.e., ADAM19, CTB-109A12.1, CTB-47B11.3, CYFIP2 and ATP13A2, CROCC, MFAP2, RP1-37C10.3, respectively; Fig S1C and Table S3). Most of the identified eQTL–gene regulatory interactions (n = 148) were cis-acting (Fig S1D and Table S3). One trans-intrachromosomal (i.e., rs2077224-NAV2) and two trans-interchromosomal (i.e., rs12894780-LIPC and rs2128739-KALRN) eQTL–gene interactions were identified within the lung (Fig S1D and Table S3). Collectively, COPD-associated eQTLs are associated with changes in transcription levels of 84 protein-coding genes, 22 non-coding RNA genes and one pseudogene (Fig S1E).

Figure 1.
Figure 1. Overview of the analytical pipelines used in this study.

(A) The CoDeS3D pipeline was used to identify lung-specific gene regulatory networks (GRNs). First, 263 GWAS SNPs associated (P < 5 × 10−8) with COPD were run through the CoDeS3D pipeline to identify 151 spatial eQTL–gene regulatory interactions in the lung (COPD-associated lung-specific GRN). Next, all GTEx SNPs (MAF ≥ 0.05, n = 43,174,097) were downloaded from dbGaP (Table S1) and analysed using CoDeS3D to identify “all” significant lung-specific spatial eQTL–gene regulatory interactions (whole lung-specific GRN). The resultant whole lung-specific GRN is comprised of 873,133 spatially constrained regulatory interactions involving 740,028 eQTLs and 15,855 genes (Figs S2 and S3). (B) The Multimorbid3D pipeline was used to identify potential co-occurring conditions associated with COPD. * Hi-C datasets for primary lung cells were obtained from Schmitt et al (2016) (GEO accessions: GSM2322544 and GSM2322545). ǂ eQTL datasets for lung was obtained from GTEx v8 (dbGaP accession: phs000424.v8.p2).

Figure S1.
Figure S1. Characteristics of COPD-associated lung-specific GRN.

(A) Of 263 COPD-associated GWAS SNPs, 103 SNPs are involved in spatially constrained eQTL–gene interactions in the lung. (B) wANNOVAR annotations identified ∼96% of the COPD-associated eQTLs are located within non-coding genomic regions (Table S3). (C) Most COPD-eQTLs (n = 67) associate with the transcript levels of one gene. Two eQTLs (i.e., rs2277027 and rs9435731) control four genes (i.e., ADAM19, CTB-109A12.1, CTB-47B11.3, CYFIP2 and ATP13A2, CROCC, MFAP2, RP1-37C10.3, correspondingly). (D) Most of the eQTL–gene interactions (n = 148) are cis-acting. (E) Most COPD-eQTLs are associated with changes in transcript levels of protein-coding genes (n = 84) within the lung.

Figure S2.
Figure S2. Characteristics of the lung-specific gene regulatory network.

(A) Of 43,174,097 genetic variants present within the GTEx database, only 1.71% (n = 740,028) are involved in 873,133 spatially constrained eQTL–gene regulatory interactions in the lung tissue. (B) Of the 56,200 genes present in the GTEx database, 28.21% (n = 15,855) are involved in spatially constrained eQTL–gene regulatory interactions within the lung tissue. (C) Of the spatially constrained eQTL–gene regulatory interactions, 850,093 are cis (between 720,083 eQTLs and 15,044 genes), 16,414 are trans-intrachromosomal (between 16,376 eQTLs and 2,743 genes), and 6,626 are trans-interchromosomal (between 6,469 eQTLs and 2,024 genes). There are significantly more cis interactions than trans-interactions (t test, P ≤ 0.0001). Mean (red line), median (white line), and SD of the number of spatially constrained eQTLs per gene grouped by eQTL interaction type (i.e., cis, trans-intra, and trans-inter). (D) Genes associated with cis-acting eQTLs have higher expression levels than those associated with trans-interactions (t test, P ≤ 0.01 and P ≤ 0.001, respectively). Mean (red line), median (white line), and SD of gene transcript levels grouped by eQTL interaction type. (E) Correlation analysis between the number of spatially constrained eQTL–gene interactions from the lung-specific gene regulatory network and the number of all common (MAF ≥ 0.05) GTEx genetic variants identifies chromosomes X, 18, 13, 8, and 4 as having fewer eQTLs than expected. By contrast, chromosomes 19, 17, 6, and 1 tend to have more eQTLs than expected in the lung tissue, with chromosomes 6 and 17 exhibiting the greatest deviation from the prediction curve. t test: ns, not significant (P > 0.05), ** significant (P ≤ 0.01), *** significant (P ≤ 0.001), **** significant (P ≤ 0.0001).

Figure S3.
Figure S3. Characteristics of lung-specific protein–protein interaction network.

The proteins encoded by all genes (n = 15,855) within the whole lung-specific GRN were mapped to the curated STRING human protein–protein interaction database (combined score ≥ 0.7). (A) Of 505,968 STRING human PPIs, 41.54% (n = 210,192) are present within the lung-specific GRN. (B) Of 16,814 human unique proteins in the STRING database, 60.59% (n = 10,188) are present in the lung tissue.

COPD-associated genes are enriched for diverse biological processes in the lung

Functional gene ontology (GO) enrichment analysis identified metabolic, behavioural, regulatory and protein modification processes (e.g., “phosphorus metabolic process,” “behavioral response to nicotine,” “regulation of postsynaptic membrane potential,” “protein acetylation,” and “protein acylation”) as being significantly enriched (FDR < 0.05) enriched within the 107 COPD-associated genes (Table S4 and Fig S4). These 107 genes encoded proteins that formed nine COPD-associated lung-specific protein–protein interaction subnetworks (Fig 2). Pathway analysis of these 107 COPD-associated genes identified biological pathways that were enriched (FDR < 0.05) for regulation of actin cytoskeleton, insulin signaling and resistance, focal adhesion, phagosome, immune processes, infections and diseases, alcoholism, long-term depression (Table S5).

Figure S4.
Figure S4. Top 10 biological processes, molecular functions, and cellular components for genes that are targeted by spatially constrained COPD-associated eQTLs within the lung.

Gene ontology enrichment analysis was performed using g:Profiler. The threshold for significance (red line) is P < 0.05.

Figure 2.
Figure 2. COPD-associated protein–protein interactions.

Green colour, the protein encoding gene is associated with COPD (DOID:3083); blue colour, the gene is associated with lung disease (DOID:850); red colour, the gene is associated with nicotine dependence (DOID:0050742) according to the DISEASES database (https://diseases.jensenlab.org/, 06/10/2022). Red text, the eQTLs is associated with up-regulation of the gene; blue text, the eQTL is associated with down-regulation of the gene transcript; black text, eQTLs are associated with up- and down-regulation of the gene transcript levels.

COPD has associations with co-occuring traits in the lung, brain, and blood

Patients with COPD often also suffer from cardiovascular disease, osteoporosis, lung cancer, sleep disorders and mental health problems (Burke & Wilkinson, 2021; Carmona-Pírez et al, 2021). Yet the biology of these interactions is rarely known. The multimorbid3D algorithm was used to integrate COPD-associated genes, the lung-specific PPIN, the lung-specific GRN and all catalogued GWAS SNP-trait associations (30/03/2022) to identify co-occurring conditions and potential regulatory mechanisms underlying these associations with COPD (Fig 1B). We identified 39 GWAS traits that are significantly (FDR ≤ 0.05) enriched for eQTLs that target the COPD-eQTL associated genes (“level 0”; Fig 3 and Tables S6 and S7). Most of the level 0 co-occurring traits were “lung-” (i.e., COPD, lung function, pulmonary function, post bronchodilator FEV1, asthma), or “mood/brain-related” (i.e., brain region volumes, cortical surface area, depressed affect, neuroticism, PD, white matter microstructure, smoking behaviour). eQTLs that regulate genes encoding proteins within levels 1–4 of the expanded COPD lung protein interaction network were enriched within traits that have and have not been previously recognized as being co-occurring with COPD (Fig 3).

Figure 3.
Figure 3. Network analysis identified co-occurring conditions that are associated with COPD.

We identified 39 GWAS traits that are enriched (FDR ≤ 0.05) for eQTLs associated with COPD-eQTL target genes (level 0). Most of these co-occurring traits are “lung-related” (i.e., COPD, lung function, pulmonary function, post bronchodilator FEV1, and asthma) and “mood/brain-related” (i.e., brain region volumes, cortical surface area, depressed affect, neuroticism, Parkinson’s disease, white matter microstructure, and smoking behaviour). Genes interacting with COPD-associated genes (level 1) within LSPPIN are regulated by eQTLs that have previously been associated with Alzheimer’s disease, atopic asthma, and post bronchodilator FEV1 or FEV1/FVC ratio. Level 2-genes are only regulated by eQTLs previously associated with “mood/brain-related” (e.g., depressed affect, neuroticism, and Parkinson’s disease) or “blood-related” (e.g., blood protein levels) traits. Genes within levels 3 and 4 are mostly associated with eQTLs enriched within “blood-related” traits.

COPD shows associations with lung function, CAD, AD, and brain region volumes

Proteins encoded by four COPD-associated genes (i.e., MSL1, MOCS2, NUPR1, and SGF29) form a PPI subnetwork (level 0). These genes are associated with eQTLs linked to COPD and lung function (Fig 4A). Notably, MSL1, MOCS2, NUPR1, and SGF29 interact with proteins encoded by three genes that are associated with post bronchodilator FEV1, atopic asthma and AD (Figs 4A and S5A).

Figure 4.
Figure 4. Identified trait-eQTL–gene associations for two COPD-associated PPI subnetworks (NUPR1-MSL1-SGF29-MOCS2 and TESK2-SSH2).

(A) Within the “NUPR1-MSL1-SGF29-MOCS2” PPI subnetwork four genes (i.e., MSL1, MOCS2, NUPR1, and SGF29) are associated with eQTLs linked to COPD and lung functioning. Proteins encoded by these four genes interact with the products of three genes (“level 1”) that are associated with post bronchodilator FEV1, atopic asthma, and Alzheimer’s disease. (B) SSH2 and TESK2 within the “TESK2-SSH2” PPI subnetwork are associated with eQTLs linked to COPD, lung functioning, brain region volumes, and CAD. There were no trait-eQTL–gene associations identified within level 1.

Figure S5.
Figure S5. Protein-protein interaction networks for NUPR1 and TESK2.

Protein-protein interaction networks for NUPR1 and TESK2. “Level 0” and “level 1” genes within (A) “NUPR1-MSL1-SGF29-MOCS2” and (B) “TESK2-SSH2” PPI subnetworks.

COPD-associated SSH2 and TESK2 genes form another PPI subnetwork (level 0). Within this subnetwork they are associated with eQTLs linked to COPD, lung functioning, brain region volumes and CAD (Fig 4B). SSH2 interacts with SSH1, SSH3, LIMK1, CFL1, CFL2, PPP3CA, and PPP3CC proteins at the “level 1” (Fig S5B), but there were no GWAS traits identified for the eQTLs that are associated with expression changes of the genes encoding these proteins.

Reversing the analysis, using CAD-associated (n = 804) and UD-associated (n = 932) SNPs (P < 5 × 10−8) confirmed COPD was co-occurring with CAD (“level 0,” Fig S6 and Table S8). By contrast, analysis of UD identified general lung function (FEV1/FVC), asthma and “mood/brain-related” (i.e., bipolar disorder, depression, autism spectrum disorder, and schizophrenia) as being linked to UD-associated genes (Fig S7 and Table S8).

Figure S6.
Figure S6. Co-occurring conditions that are putatively associated with CAD.

We identified 58 GWAS traits that are enriched (FDR ≤ 0.05) for eQTLs associated with CAD-eQTL targeted genes (level 0) (Table S8). Among these co-occurring traits are “blood- and heart-related” (i.e., CAD, coronary heart, and cardiovascular diseases). COPD is also found to be associated with CAD-eQTL targeted genes (level 0, “chronic obstructive pulmonary disease or coronary disease [pleiotropy]”).

Figure S7.
Figure S7. Co-occurring conditions that are putatively associated with UD.

We identified 58 GWAS traits that are enriched (FDR ≤ 0.05) for eQTLs associated with UD-eQTL targeted genes (level 0) (Table S8). As expected, most of the identified co-occurring traits on level 0 are “mood/brain-related” (i.e., bipolar disorder, depression, autism spectrum disorder, or schizophrenia). However, a few “lung-related” are also found to be associated with UD-eQTL targeted genes (level 0, Lung function [FEV1/FVC], Asthma).

Discussion

We integrated genes that are targeted by spatially constrained COPD-associated eQTLs with a lung-specific GRN, lung-specific PPIN and all GWAS SNP-trait associations to identify traits that are co-occurring with COPD. The results of this integration provide insights into the regulatory mechanisms underlying these associations. We identified co-occurring traits that have been previously linked to COPD (e.g., lung function, asthma, depressed affect, CAD, AD, smoking behaviour, PD [Cavaillès et al, 2013; Li et al, 2015; Ställberg et al, 2018; Xia et al, 2020; Burke & Wilkinson, 2021; Carmona-Pírez et al, 2021; Martucci et al, 2021]) and those that have not (brain region volumes and white matter microstructure). We contend that the eQTLs we identified, as impacting on COPD and its co-occurring traits, represent the population-based genetic burden that may contribute to an individual’s risk of developing COPD and its diverse clinical presentations.

The results of this study should be interpreted in view of its strengths and limitations. The main strength of this study is the integration of independent datasets: lung-specific 3D genome structure (Schmitt et al, 2016), common GWAS SNPs, genotypes and gene expression data (Aguet et al, 2019 Preprint), and protein-protein interactions (PPIs) (Szklarczyk et al, 2019). Integrating these datasets enabled the identification of the impact of spatially constrained COPD-associated eQTLs on genes and biological pathways that link to co-occurring conditions. Indeed it is possible that the “total” genetic burden we identified can be used to stratify COPD and its multiple clinical presentations, which have previously led to questions about the validity of classifying it as a single diagnostic category (Sakornsakolpat et al, 2019; Corlateanu et al, 2020; Alabi et al, 2021). However, this study also has several limitations. Firstly, this study was focused on the regulatory roles of common genetic variants (MAF ≥ 0.05) ignoring the impacts of other genetic (e.g., rare SNPs) and environmental factors, which will undoubtedly contribute to the risk of COPD and its co-occurring conditions. Secondly, we focused on the extended PPIN within the lung GRN, as the lung represents the primary affected tissue in COPD. However, it is possible that genetic variation will impact on COPD risk through other tissues (e.g., blood) (Burke & Wilkinson, 2021). Thirdly, the tools and datasets used in this study are potentially biased (e.g., lung-specific eQTL and Hi-C data were not obtained from identical samples). Furthermore, the identification of the co-occurring conditions was limited to the traits that have been studied and were present in the GWAS Catalog. Finally, mapping to Ensembl gene identifiers potentially causes a loss of data specificity, since alternative splicing typically produces multiple transcripts and protein variants.

Co-occurring conditions are commonly associated with COPD and increase the risk of hospitalisation (Schnell et al, 2012). Our analysis of co-occurring conditions has identified risk variants and protein interactions that connect COPD with smoking (Hopkinson, 2022), asthma (Maselli & Hanania, 2018), CAD (Xia et al, 2020), lung cancer (Parris et al, 2019), multiple sclerosis (Egesten et al, 2008), kidney failure (Trudzinski et al, 2019), UD (Li et al, 2019a), AD (Wang et al, 2019), PD (Li et al, 2015), and personality traits (e.g., neuroticism) (Chetty et al, 2017; Terracciano et al, 2017; Caille et al, 2021). These conditions have been previously reported as being co-occurring conditions in COPD patients but the biological basis of the connections was unknown.

Brain region volumes, white matter microstructure, left-right brain asymmetry and cortical surface area (global PC1) were additional traits that we have identified as being co-occurring with COPD. Notably, patterns of brain structural alteration have been reported in COPD with different levels of pulmonary function impairment and cognitive deficits (Yin et al, 2019; Wang et al, 2020). This suggests that COPD patients may exhibit progressive structural impairments in both the grey and white matter, along with impaired levels of lung function (Yin et al, 2019; Wang et al, 2020). In addition to the trait associations, the results of our study provide putative evidence for the existence of genetic and biological connections between these traits (Higbee et al, 2021). In so doing our results stop short of providing causal evidence for these putative connections and two sample or multivariable Mendelian randomization (e.g., Transcriptome-Wide Mendelian Randomization) or mediation analyses are required to clarify which if any of the interactions are causal (Russ et al, 2021; Porcu et al, 2019).

Impaired lung function as measured by forced vital capacity or forced expiratory volume in the first second (FEV1) has been previously associated with insulin resistance (Sagun et al, 2015; Piazzolla et al, 2017; Machado et al, 2018; Kim et al, 2021a). The evidence that supports the existence of a relationship between COPD and insulin resistance is complex. Our independent identification that PPP1CB and PPP1R3B are COPD-associated genes is notable because these genes are directly involved in controlling glycogen synthesis and glucose homeostasis in insulin signaling (Li et al, 2019b) and have been linked to glycaemic traits (Niazi et al, 2019).

We also identified several COPD-associated genes enriched for immune processes and immune-related diseases (e.g., HLA-C and HLA-DQB1—antigen processing and presentation, HLA-DQB1—type I diabetes mellitus, influenza A, autoimmune thyroid disease) (Table S5). This finding is consistent with observations that proteins of the major histocompatibility complex of classes I and II (HLA-I and HLA-II) have been identified as potential markers of progression of systemic and local inflammation in patients with COPD (Kubysheva et al, 2018). Specifically, an increase in the level of HLA-I and HLA-II molecules in the exhaled breath condensate as well as an elevated serum level of HLA-II is observed in COPD patients when compared with healthy volunteers (Kubysheva et al, 2018). Previous studies have also identified elevated levels of plasma TGF-β, an important regulator of lung and immune system development, in COPD patients compared with healthy controls (Mak et al, 2009; Verhamme et al, 2015). Despite the apparent correlative support between our results and previously published fundings, the evidence for a role for the insulin resistance pathway and immune system in COPD remains putative until proven empirically.

COPD-associated eQTLs target cholinergic pathway genes (e.g., CHRM1 and CHRM3) that have previously been implicated as important susceptibility loci for lung diseases (e.g., asthma and COPD) (Palmberg et al, 2018; Rajasekaran et al, 2019). The cholinergic pathway mediated by the parasympathetic neurotransmitter, acetylcholine, is a predominant neurogenic mechanism contributing to bronchoconstriction (Ward, 2022). Notably, changes in the parasympathetic neuronal control of airway smooth muscle have been shown to increase bronchoconstriction in response to vagal stimulation, leading to airway hyperresponsiveness (Ward, 2022). At the biological level, these findings emphasize the effects that COPD-associated SNPs may have on the regulation of genes within specific biological pathways (e.g., through creating an imbalance in the concentration of specific proteins), which, in turn, can be associated with an increased risk of disease.

Integrative network-based methods have been used to explore complex SNP-gene interactions that can impact functional biological pathways and lead to complex phenotypes (Platig et al, 2016; Chen et al, 2021b; Zhu et al, 2021). The CONDOR algorithm was applied to study eQTLs in COPD (Platig et al, 2016). In brief, CONDOR built a bipartite network linking eQTLs and genes in 52 communities. 30 SNPs associated with COPD (Cho et al, 2014) mapped to three of these communities. Notably, despite only one gene (KANSL1) being identified in Platig et al (2016) and our current study, the loci involved were enriched in conserved biological processes including Alzheimer’s, Parkinson’s, and Asthma and immune responses. This is despite fundamental difference in the methods (e.g., eQTLs vs spatially constrained eQTLs, and the inclusion of the expanded protein network as additional information in multimorbid3D) and updates in the GWAS catalog that occurred between 2016 and 2022. As such, the apparent convergence of the results is consistent with the utility of these integrative approaches in the identification of the shared genetic and biological pathway information associated with COPD and its comorbid conditions.

In conclusion, we have integrated different levels of biological information (i.e., genes that are targeted by spatially constrained COPD-associated eQTLs, lung-specific GRN, LSPPIN, and all GWAS SNP-trait associations) to identify target genes, associated with COPD-associated eQTLs, that may interact to connect COPD to co-occurring conditions. Collectively, these results provide multiple new avenues for future investigation of the underlying biology and diverse clinical presentations of COPD. Empirical confirmation of the connections will suggest potential therapeutic COPD markers for follow-up patient stratification.

Materials and Methods

Hi-C data processing

Hi-C chromatin interaction libraries specific to primary lung cells (Table S1) were downloaded from GEO database (https://www.ncbi.nlm.nih.gov/geo/, accessions: GSM2322544 and GSM2322545) and analysed using the Juicer pipeline (v1.5) (Durand et al, 2016) to generate Hi-C libraries. The pipeline included BWA (v0.7.15) alignment of paired-end reads onto the hg38 (GRCh38; release 75) reference genome, merging paired-end read alignments and removing chimeric, unmapped and duplicated reads. We refer to the remaining read pairs as “contacts.” Only Hi-C libraries that contain >90% alignable unique read pairs, and >50% unique contacts (<40% duplication rate) within the total sequenced read pairs were included in the analysis. Files containing cleaned Hi-C contacts (i.e. *_merged_nodups.txt files) were processed to obtain Hi-C chromatin interaction libraries in the following format: read name, str1, chr1, pos1, frag1 mapq1, str2, chr2, pos2, frag2, mapq2 (str, strand; chr, chromosome; pos, position; frag, restriction site fragment; mapq, mapping quality score; 1 and 2 correspond to read ends in a pair). Reads where both ends had a mapq ≥30 were included in the final library. Hi-C chromatin interactions represent all captured pairs of interacting restriction fragments in the genome and were used by CoDeS3D to identify putative regulatory interactions between SNPs and genes.

Identification of SNPs associated with COPD, CAD, and UD

SNPs associated (P < 5 × 10−8) with COPD (n = 263), CAD (n = 804) and UD (n = 932) were downloaded from the GWAS Catalog (www.ebi.ac.uk/gwas/; 09/06/2021 and 11/04/2022; Table S2).

Identification of spatial regulatory interactions

The CoDeS3D (Fadason et al, 2018) pipeline was used to identify genes that spatially interact with putative regulatory regions tagged by SNPs (Fig 1A). Briefly, the human genome build hg38 (GRCh38; release 75) was fragmented in silico at HindIII sites (A^AGCTT), the restriction enzyme that was used in the preparation of the lung-specific Hi-C libraries (Schmitt et al, 2016). Disease associated SNP rsID numbers were cross-checked with the GTEx v8 lung eQTL database (GTEx Consortium, 2020) and restriction fragments that were tagged by the COPD associated SNPs were identified. Using lung-specific Hi-C libraries (Table S1), CoDeS3D identified the restriction fragments that were captured interacting with the SNP-tagged restriction HindIII fragments. Interacting fragments that overlapped annotated genes (GENCODE transcript model version 26) were identified. The resulting SNP-gene pairs were used to query the GTEx v8 lung eQTL database (GTEx Consortium, 2020) to identify cis- and trans-acting eQTLs (i.e. genes, whose expression levels are associated with the SNP identity). Finally, significant lung-specific eQTL–gene interactions were identified using the Benjamini–Hochberg (BH) FDR correction to adjust the eQTL P-values (FDR < 0.05).

CoDeS3D was used to build a lung-specific gene regulatory network (Fig S2). All SNPs (MAF ≥ 0.05, n = 43,174,097) present within GTEx lung-specific eQTL database (GTEx Consortium, 2020) were used to identify all significant (BH, FDR < 0.05) spatially constrained lung-specific eQTL–gene interactions (the lung-specific GRN). Multiple correction testing was performed across all interactions within individual chromosomes.

The lung-specific GRN was mined using the COPD-associated SNPs (n = 263; P < 5 × 10−8; GWAS Catalog; 09/06/2021) to identify all COPD-associated significant (BH, FDR < 0.05) lung-specific eQTL–gene interactions (COPD-associated GRN). This was repeated for the CAD-associated (n = 651; P < 5 × 10−8; GWAS Catalog; 11/05/2022) and UD-associated (n = 152; P < 5 × 10−8; GWAS Catalog; 11/05/2022) SNPs (CAD- and UD-associated GRNs, respectively).

Functional annotation of eQTL SNPs associated with COPD

The COPD-associated eQTLs were annotated using the wANNOVAR tool (Chang & Wang, 2012) (http://wannovar.wglab.org/, 09/06/2021) to obtain information about the locus they tagged. All genomic positions and SNP annotations were obtained for human genome reference build hg38 (GRCh38) release 75.

Construction of lung-specific PPIN

The STRING (Szklarczyk et al, 2019) PPI database (version 11.5, https://string-db.org/, 15/03/2022) was downloaded and queried (STRING API) to identify potential PPIs (combined score ≥0.7). A lung-specific PPI network (LSPPIN) was constructed by filtering the STRING PPI network for the proteins encoded by the genes that were affected by eQTLs within the lung-specific GRN (Fig S3). Ensembl protein identifiers were mapped to Ensembl gene identifiers using EnsDb.Hsapiens.v86 R package. The LSPPIN represents a subnetwork of the entire STRING PPI network, in which a protein/node is only present if the encoding gene is associated with a spatially constrained eQTL within lung tissue. The size of each node depends on the protein expression levels (no missing values, TPM > 0.1 and ≥ 6 reads in a minimum of 20% of tested samples) within the GTEx v8 lung database (GTEx Consortium, 2020). The resulting LSPPIN contained 210,192 PPIs between 10,188 unique proteins. To build the COPD-specific LSPPIN, only interactions between genes targeted by COPD-associated eQTLs were extracted from the LSPPIN (Table S5).

Identification of potential co-occurring conditions

The multimorbid3D pipeline was used to identify traits that were co-occurring with COPD (Fig 1B) (Zaied et al, 2022 Preprint). At “level 0” within the LSPPIN, we first identify the proteins encoded by the genes targeted by the COPD-associated eQTLs. At “level 1,” the proteins interacting with the “level 0” proteins were identified. This process is iterated so that at each level, the proteins interacting with the proteins at the level minus 1 were identified. Next, we identify all eQTLs associated with the genes encoding the proteins at each level. The GWAS Catalog (www.ebi.ac.uk/gwas/, v1.0.2, 30/03/2022) was queried to identify traits that were enriched for the eQTLs from each level. The hypergeometric distribution test was used to identify significant enrichment of traits at each level within the GWAS Catalog traits (n = 17,841). Finally, significantly enriched traits were identified using the BH FDR correction to adjust the P-values (FDR ≤ 0.05; Tables S6 and S7). The same pipeline was also used to identify the co-occurring conditions for CAD and UD (Table S8).

GO enrichment and pathway analyses

GO enrichment and pathway analyses were performed using the g:GOSt module of the g:Profiler tool (Raudvere et al, 2019) (Tables S4 and S5). Enrichment was tested for within the biological process, molecular function and cellular component GO terms. All known human genes were chosen as the statistical domain scope. The significance level was determined using the BH algorithm (FDR < 0.05). The Kyoto Encyclopedia of Genes and Genomes database (Kanehisa & Goto, 2000) was used to identify impacted biological pathways.

Bootstrapping analysis

Bootstrapping analysis (n = 1,000 iterations) was performed to test the specificity of the identified genes to COPD. At each bootstrap iteration, the same number of SNPs as were present in the tested condition were randomly selected from the GWAS Catalog (www.ebi.ac.uk/gwas/; 20/09/2022) and run through the CoDeS3D pipeline. The P-value was calculated as the number of the occurrences of the identified COPD-associated genes in the iterations divided by 1,000.

Data Availability

Data access was approved by the dbGaP (https://www.ncbi.nlm.nih.gov/gap/) Data Access Committee(s) for total RNA-seq and WGS datasets across GTEx v8 tissues (project #22937: “Untangling the genetics of disease multimorbidity,” accession: phs000424.v8.p2) (Aguet et al, 2019 Preprint) (Table S1). Data analysis and visualisation were performed in Python (version 3.6.9) using miniconda (version 4.8.4), and R (version 4.0.2) through RStudio (version 1.2.5033). All datasets and software used in the analysis are listed in Table S1.

All findings, scripts and a reproducibility report are available https://github.com/Genome3d/genetic_regulation_in_COPD.

Juicer is available at https://github.com/aidenlab/juicer.

CoDeS3D is available at https://github.com/Genome3d/codes3d-v2.

Multimorbid3D is available at https://github.com/Genome3d/multimorbid3D.

Acknowledgements

The authors would like to thank the Genomics and Systems Biology Group (Liggins Institute, University of Auckland) for useful discussions. The authors wish to acknowledge the use of New Zealand eScience Infrastructure (NeSI, https://www.nesi.org.nz) high-performance computing facilities, consulting support and/or training services as part of this research. New Zealand’s national facilities are provided by NeSI and funded jointly by NeSI’s collaborator institutions and through the Ministry of Business, Innovation & Employment’s Research Infrastructure programme. This work was funded by the Dines Family Charitable Trust. The Genotype-Tissue Expression (GTEx) Project was supported by the Common Fund of the Office of the Director of the National Institutes of Health, and by NCI, NHGRI, NHLBI, NIDA, NIMH, and NINDS.

Author Contributions

  • E Golovina: data curation, formal analysis, investigation, methodology, and writing—original draft.

  • T Fadason: resources, software, investigation, and writing—review and editing.

  • RK Jaros: resources, data curation, formal analysis, and writing—review and editing.

  • H Kumar: conceptualization and writing—review and editing.

  • J John: conceptualization and writing—review and editing.

  • K Burrowes: conceptualization and writing—review and editing.

  • M Tawhai: conceptualization, funding acquisition, and writing—review and editing.

  • JM O’Sullivan: conceptualization, resources, formal analysis, supervision, funding acquisition, project administration, and writing—original draft, review, and editing.

Conflict of Interest Statement

The authors declare that they have no conflict of interest.

  • Received July 15, 2022.
  • Revision received December 6, 2022.
  • Accepted December 8, 2022.
Creative Commons logoCreative Commons logohttps://creativecommons.org/licenses/by/4.0/

This article is available under a Creative Commons License (Attribution 4.0 International, as described at https://creativecommons.org/licenses/by/4.0/).

References

  1. ↵
    1. Aguet F,
    2. Barbeira AN,
    3. Bonazzola R,
    4. Brown A,
    5. Castel SE,
    6. Jo B,
    7. Kasela S,
    8. Kim-Hellmuth S,
    9. Liang Y,
    10. Oliva M, et al.
    (2019) The GTEx Consortium atlas of genetic regulatory effects across human tissues. BioRxiv. doi:10.1101/787903 (Preprint posted October 3, 2019).
    OpenUrlAbstract/FREE Full Text
  2. ↵
    1. Alabi FO,
    2. Alkhateeb HA,
    3. DeBarros KM,
    4. Barletti Benel PS,
    5. Sanchez-Martez RL,
    6. Zeper ML,
    7. Ismail RA,
    8. Umeh F,
    9. Medina-Villanueva N
    (2021) The heterogeneity of COPD patients in a community-based practice and the inadequacy of the global initiative for chronic obstructive lung disease criteria: A real-world experience. Chronic Obstr Pulm Dis 8: 396–407. doi:10.15326/jcopdf.2021.0229
    OpenUrlCrossRef
  3. ↵
    1. Buenrostro JD,
    2. Corces MR,
    3. Lareau CA,
    4. Wu B,
    5. Schep AN,
    6. Aryee MJ,
    7. Majeti R,
    8. Chang HY,
    9. Greenleaf WJ
    (2018) Integrated single-cell analysis maps the continuous regulatory landscape of human hematopoietic differentiation. Cell 173: 1535–1548.e16. doi:10.1016/j.cell.2018.03.074
    OpenUrlCrossRefPubMed
  4. ↵
    1. Burke H,
    2. Wilkinson TMA
    (2021) Unravelling the mechanisms driving multimorbidity in COPD to develop holistic approaches to patient-centred care. Eur Respir Rev 30: 210041. doi:10.1183/16000617.0041-2021
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Caille P,
    2. Alexandre F,
    3. Molinier V,
    4. Heraud N
    (2021) The role of personality traits in inpatient pulmonary rehabilitation response in patients with chronic obstructive pulmonary disease. Respir Med 190: 106680. doi:10.1016/j.rmed.2021.106680
    OpenUrlCrossRef
  6. ↵
    1. Carmona-Pírez J,
    2. Poblador-Plou B,
    3. Ioakeim-Skoufa I,
    4. González-Rubio F,
    5. Gimeno-Feliú LA,
    6. Díez-Manglano J,
    7. Laguna-Berna C,
    8. Marin JM,
    9. Gimeno-Miguel A,
    10. Prados-Torres A
    (2021) Multimorbidity clusters in patients with chronic obstructive airway diseases in the EpiChron Cohort. Sci Rep 11: 4784. doi:10.1038/s41598-021-83964-w
    OpenUrlCrossRef
  7. ↵
    1. Cavaillès A,
    2. Brinchault-Rabin G,
    3. Dixmier A,
    4. Goupil F,
    5. Gut-Gobert C,
    6. Marchand-Adam S,
    7. Meurice J-C,
    8. Morel H,
    9. Person-Tacnet C,
    10. Leroyer C, et al.
    (2013) Comorbidities of COPD. Eur Respir Rev 22: 454–475. doi:10.1183/09059180.00008612
    OpenUrlAbstract/FREE Full Text
  8. ↵
    1. Chang X,
    2. Wang K
    (2012) wANNOVAR: Annotating genetic variants for personal genomes via the web. J Med Genet 49: 433–436. doi:10.1136/jmedgenet-2012-100918
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Chen C,
    2. Jiang L,
    3. Shen B,
    4. Wang M,
    5. Griffin CH,
    6. Chinchilli VM,
    7. Wu R
    (2021a) A computational atlas of tissue-specific regulatory networks. Front Syst Biol 1: 764161. doi:10.3389/fsysb.2021.764161
    OpenUrlCrossRef
  10. ↵
    1. Chen X,
    2. Zhou J,
    3. Zhang R,
    4. Wong AK,
    5. Park CY,
    6. Theesfeld CL,
    7. Troyanskaya OG
    (2021b) Tissue-specific enhancer functional networks for associating distal regulatory regions to disease. Cell Syst 12: 353–362.e6. doi:10.1016/j.cels.2021.02.002
    OpenUrlCrossRef
  11. ↵
    1. Chetty U,
    2. McLean G,
    3. Morrison D,
    4. Agur K,
    5. Guthrie B,
    6. Mercer SW
    (2017) Chronic obstructive pulmonary disease and comorbidities: A large cross-sectional study in primary care. Br J Gen Pract 67: e321–e328. doi:10.3399/bjgp17x690605
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Cho MH,
    2. McDonald M-LN,
    3. Zhou X,
    4. Mattheisen M,
    5. Castaldi PJ,
    6. Hersh CP,
    7. Demeo DL,
    8. Sylvia JS,
    9. Ziniti J,
    10. Laird NM, et al.
    (2014) Risk loci for chronic obstructive pulmonary disease: A genome-wide association study and meta-analysis. Lancet Respir Med 2: 214–225. doi:10.1016/s2213-2600(14)70002-5
    OpenUrlCrossRef
  13. ↵
    1. Corlateanu A,
    2. Mendez Y,
    3. Wang Y,
    4. Garnica RdJA,
    5. Botnaru V,
    6. Siafakas N
    (2020) Chronic obstructive pulmonary disease and phenotypes: A state-of-the-art. Pulmonology 26: 95–100. doi:10.1016/j.pulmoe.2019.10.006
    OpenUrlCrossRef
  14. ↵
    1. Durand NC,
    2. Shamim MS,
    3. Machol I,
    4. Rao SSP,
    5. Huntley MH,
    6. Lander ES,
    7. Aiden EL
    (2016) Juicer provides a one-click system for analyzing loop-resolution Hi-C experiments. Cell Syst 3: 95–98. doi:10.1016/j.cels.2016.07.002
    OpenUrlCrossRef
  15. ↵
    1. Egesten A,
    2. Brandt L,
    3. Olsson T,
    4. Granath F,
    5. Inghammar M,
    6. Löfdahl C-G,
    7. Ekbom A
    (2008) Increased prevalence of multiple sclerosis among COPD patients and their first-degree relatives: A population-based study. Lung 186: 173–178. doi:10.1007/s00408-008-9081-y
    OpenUrlCrossRefPubMed
  16. ↵
    1. Fadason T,
    2. Schierding W,
    3. Lumley T,
    4. O’Sullivan JM
    (2018) Chromatin interactions and expression quantitative trait loci reveal genetic drivers of multimorbidities. Nat Commun 9: 5198. doi:10.1038/s41467-018-07692-y
    OpenUrlCrossRefPubMed
  17. ↵
    1. Gim J,
    2. An J,
    3. Sung J,
    4. Silverman EK,
    5. Cho MH,
    6. Won S
    (2020) A between ethnicities comparison of chronic obstructive pulmonary disease genetic risk. Front Genet 11: 329. doi:10.3389/fgene.2020.00329
    OpenUrlCrossRef
  18. ↵
    1. 2020 Global Initiative for Chronic Obstructive Obstructive Lung Disease
    (2021) Global strategy for the diagnosis, management, and prevention of chronic obstructive obstructive lung disease. Available at: https://goldcopd.org/wp-content/uploads/2020/11/GOLD-REPORT-2021-v1.1-25Nov20_WMV.pdf.
  19. ↵
    1. GTEx Consortium
    (2020) The GTEx Consortium atlas of genetic regulatory effects across human tissues. Science 369: 1318–1330. doi:10.1126/science.aaz1776
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Halow JM,
    2. Byron R,
    3. Hogan MS,
    4. Ordoñez R,
    5. Groudine M,
    6. Bender MA,
    7. Stamatoyannopoulos JA,
    8. Maurano MT
    (2021) Tissue context determines the penetrance of regulatory DNA variation. Nat Commun 12: 2850. doi:10.1038/s41467-021-23139-3
    OpenUrlCrossRefPubMed
  21. ↵
    1. Higbee D,
    2. Granell R,
    3. Walton E,
    4. Korologou-Linden R,
    5. Davey Smith G,
    6. Dodd J
    (2021) Examining the possible causal relationship between lung function, COPD and Alzheimer’s disease: A mendelian randomisation study. BMJ Open Respir Res 8: e000759. doi:10.1136/bmjresp-2020-000759
    OpenUrlAbstract/FREE Full Text
  22. ↵
    1. Hopkinson NS
    (2022) COPD, smoking, and social justice. Lancet Respir Med 10: 428–430. doi:10.1016/s2213-2600(22)00130-8
    OpenUrlCrossRef
  23. ↵
    1. Kanehisa M,
    2. Goto S
    (2000) KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res 28: 27–30. doi:10.1093/nar/28.1.27
    OpenUrlCrossRefPubMed
  24. ↵
    1. Kim SH,
    2. Kim HS,
    3. Min HK,
    4. Lee SW
    (2021a) Association between insulin resistance and lung function trajectory over 4 years in South Korea: Community-based prospective cohort. BMC Pulm Med 21: 110. doi:10.1186/s12890-021-01478-7
    OpenUrlCrossRef
  25. ↵
    1. Kim W,
    2. Prokopenko D,
    3. Sakornsakolpat P,
    4. Hobbs BD,
    5. Lutz SM,
    6. Hokanson JE,
    7. Wain LV,
    8. Melbourne CA,
    9. Shrine N,
    10. Tobin MD, et al.
    (2021b) Genome-wide gene-by-smoking interaction study of chronic obstructive pulmonary disease. Am J Epidemiol 190: 875–885. doi:10.1093/aje/kwaa227
    OpenUrlCrossRef
  26. ↵
    1. Kubysheva N,
    2. Soodaeva S,
    3. Novikov V,
    4. Eliseeva T,
    5. Li T,
    6. Klimanov I,
    7. Kuzmina E,
    8. Baez-Medina H,
    9. Solovyev V,
    10. Ovsyannikov DY, et al.
    (2018) Soluble HLA-I and HLA-II molecules are potential prognostic markers of progression of systemic and local inflammation in patients with COPD. Dis Markers 2018: 3614341. doi:10.1155/2018/3614341
    OpenUrlCrossRef
  27. ↵
    1. Li C-H,
    2. Chen W-C,
    3. Liao W-C,
    4. Tu C-Y,
    5. Lin C-L,
    6. Sung F-C,
    7. Chen C-H,
    8. Hsu W-H
    (2015) The association between chronic obstructive pulmonary disease and Parkinson’s disease: A nationwide population-based retrospective cohort study. QJM 108: 39–45. doi:10.1093/qjmed/hcu136
    OpenUrlCrossRefPubMed
  28. ↵
    1. Li Q,
    2. Zhao Q,
    3. Zhang J,
    4. Zhou L,
    5. Zhang W,
    6. Chua B,
    7. Chen Y,
    8. Xu L,
    9. Li P
    (2019a) The protein phosphatase 1 complex is a direct target of AKT that links insulin signaling to hepatic glycogen deposition. Cell Rep 28: 3406–3422.e7. doi:10.1016/j.celrep.2019.08.066
    OpenUrlCrossRefPubMed
  29. ↵
    1. Li Z,
    2. Liu S,
    3. Wang L,
    4. Smith L
    (2019b) Mind-body exercise for anxiety and depression in COPD patients: A systematic review and meta-analysis. Int J Environ Res Public Health 17: 22. doi:10.3390/ijerph17010022
    OpenUrlCrossRef
  30. ↵
    1. Machado FVC,
    2. Pitta F,
    3. Hernandes NA,
    4. Bertolini GL
    (2018) Physiopathological relationship between chronic obstructive pulmonary disease and insulin resistance. Endocrine 61: 17–22. doi:10.1007/s12020-018-1554-z
    OpenUrlCrossRef
  31. ↵
    1. Mak JCW,
    2. Chan-Yeung MMW,
    3. Ho SP,
    4. Chan KS,
    5. Choo K,
    6. Yee KS,
    7. Chau CH,
    8. Cheung AHK,
    9. Ip MSM, Members of Hong Kong Thoracic Society COPD Study Group
    (2009) Elevated plasma TGF-beta1 levels in patients with chronic obstructive pulmonary disease. Respir Med 103: 1083–1089. doi:10.1016/j.rmed.2009.01.005
    OpenUrlCrossRefPubMed
  32. ↵
    1. Mannino DM,
    2. Higuchi K,
    3. Yu T-C,
    4. Zhou H,
    5. Li Y,
    6. Tian H,
    7. Suh K
    (2015) Economic burden of COPD in the presence of comorbidities. Chest 148: 138–150. doi:10.1378/chest.14-2434
    OpenUrlCrossRefPubMed
  33. ↵
    1. Martucci VL,
    2. Richmond B,
    3. Davis LK,
    4. Blackwell TS,
    5. Cox NJ,
    6. Samuels D,
    7. Velez Edwards D,
    8. Aldrich MC
    (2021) Fate or coincidence: Do COPD and major depression share genetic risk factors? Hum Mol Genet 30: 619–628. doi:10.1093/hmg/ddab068
    OpenUrlCrossRef
  34. ↵
    1. Maselli DJ,
    2. Hanania NA
    (2018) Asthma COPD overlap: Impact of associated comorbidities. Pulm Pharmacol Ther 52: 27–31. doi:10.1016/j.pupt.2018.08.006
    OpenUrlCrossRef
  35. ↵
    1. Niazi RK,
    2. Sun J,
    3. Have CT,
    4. Hollensted M,
    5. Linneberg A,
    6. Pedersen O,
    7. Nielsen JS,
    8. Rungby J,
    9. Grarup N,
    10. Hansen T, et al.
    (2019) Increased frequency of rare missense PPP1R3B variants among Danish patients with type 2 diabetes. PLoS One 14: e0210114. doi:10.1371/journal.pone.0210114
    OpenUrlCrossRef
  36. ↵
    1. Palmberg L,
    2. Sundblad B-M,
    3. Ji J,
    4. Karén J,
    5. Larsson K
    (2018) Cholinergic mechanisms in an organic dust model simulating an acute exacerbation in patients with COPD. Int J Chron Obstruct Pulmon Dis 13: 3611–3624. doi:10.2147/copd.s171495
    OpenUrlCrossRef
  37. ↵
    1. Parris BA,
    2. O’Farrell HE,
    3. Fong KM,
    4. Yang IA
    (2019) Chronic obstructive pulmonary disease (COPD) and lung cancer: Common pathways for pathogenesis. J Thorac Dis 11: S2155–S2172. doi:10.21037/jtd.2019.10.54
    OpenUrlCrossRef
  38. ↵
    1. Piazzolla G,
    2. Castrovilli A,
    3. Liotino V,
    4. Vulpi MR,
    5. Fanelli M,
    6. Mazzocca A,
    7. Candigliota M,
    8. Berardi E,
    9. Resta O,
    10. Sabbà C, et al.
    (2017) Metabolic syndrome and Chronic Obstructive Pulmonary Disease (COPD): The interplay among smoking, insulin resistance and vitamin D. PLoS One 12: e0186708. doi:10.1371/journal.pone.0186708
    OpenUrlCrossRef
  39. ↵
    1. Platig J,
    2. Castaldi PJ,
    3. DeMeo D,
    4. Quackenbush J
    (2016) Bipartite community structure of eQTLs. PLoS Comput Biol 12: e1005033. doi:10.1371/journal.pcbi.1005033
    OpenUrlCrossRef
  40. ↵
    1. Porcu E,
    2. Rüeger S,
    3. Lepik K, eQTLGen Consortium, BIOS Consortium,
    4. Santoni FA,
    5. Reymond A,
    6. Kutalik Z
    (2019) Mendelian randomization integrating GWAS and eQTL data reveals genetic determinants of complex and clinical traits. Nat Commun 10: 3300. doi:10.1038/s41467-019-10936-0
    OpenUrlCrossRefPubMed
  41. ↵
    1. Rajasekaran A,
    2. He D,
    3. Yue A,
    4. Singh A,
    5. Shannon CP,
    6. FitzGerald JM,
    7. Boulet L-P,
    8. O’Byrne PM,
    9. Gauvreau GM,
    10. Tebbutt SJ
    (2019) Cholinergic synapse pathway gene polymorphisms associated with allergen-induced late asthmatic responses. ERJ Open Res 5: 00107–02019. doi:10.1183/23120541.00107-2019
    OpenUrlCrossRef
  42. ↵
    1. Raudvere U,
    2. Kolberg L,
    3. Kuzmin I,
    4. Arak T,
    5. Adler P,
    6. Peterson H,
    7. Vilo J
    (2019) g:Profiler: a web server for functional enrichment analysis and conversions of gene lists (2019 update). Nucleic Acids Res 47: W191–W198. doi:10.1093/nar/gkz369
    OpenUrlCrossRefPubMed
  43. ↵
    1. Russ TC,
    2. Harris SE,
    3. David Batty G
    (2021) RESEARCH LETTER: Pulmonary function and risk of Alzheimer dementia: Two-sample Mendelian randomization study. Available at: https://discovery.ucl.ac.uk/id/eprint/10149131/2/Batty_FEV%20MR%20%28accepted%29.pdf, Accessed September 27, 2022.
  44. ↵
    1. Sagun G,
    2. Gedik C,
    3. Ekiz E,
    4. Karagoz E,
    5. Takir M,
    6. Oguz A
    (2015) The relation between insulin resistance and lung function: A cross sectional study. BMC Pulm Med 15: 139. doi:10.1186/s12890-015-0125-9
    OpenUrlCrossRef
  45. ↵
    1. Sakornsakolpat P,
    2. Prokopenko D,
    3. Lamontagne M,
    4. Reeve NF,
    5. Guyatt AL,
    6. Jackson VE,
    7. Shrine N,
    8. Qiao D,
    9. Bartz TM,
    10. Kim DK, et al.
    (2019) Genetic landscape of chronic obstructive pulmonary disease identifies heterogeneous cell-type and phenotype associations. Nat Genet 51: 494–505. doi:10.1038/s41588-018-0342-2
    OpenUrlCrossRefPubMed
  46. ↵
    1. Schmitt AD,
    2. Hu M,
    3. Jung I,
    4. Xu Z,
    5. Qiu Y,
    6. Tan CL,
    7. Li Y,
    8. Lin S,
    9. Lin Y,
    10. Barr CL, et al.
    (2016) A compendium of chromatin contact maps reveals spatially active regions in the human genome. Cell Rep 17: 2042–2059. doi:10.1016/j.celrep.2016.10.061
    OpenUrlCrossRefPubMed
  47. ↵
    1. Schnell K,
    2. Weiss CO,
    3. Lee T,
    4. Krishnan JA,
    5. Leff B,
    6. Wolff JL,
    7. Boyd C
    (2012) The prevalence of clinically-relevant comorbid conditions in patients with physician-diagnosed COPD: A cross-sectional study using data from NHANES 1999-2008. BMC Pulm Med 12: 26. doi:10.1186/1471-2466-12-26
    OpenUrlCrossRefPubMed
  48. ↵
    1. Shrine N,
    2. Guyatt AL,
    3. Erzurumluoglu AM,
    4. Jackson VE,
    5. Hobbs BD,
    6. Melbourne CA,
    7. Batini C,
    8. Fawcett KA,
    9. Song K,
    10. Sakornsakolpat P, et al.
    (2019) New genetic signals for lung function highlight pathways and chronic obstructive pulmonary disease associations across multiple ancestries. Nat Genet 51: 481–493. doi:10.1038/s41588-018-0321-7
    OpenUrlCrossRefPubMed
  49. ↵
    1. Ställberg B,
    2. Janson C,
    3. Larsson K,
    4. Johansson G,
    5. Kostikas K,
    6. Gruenberger J-B,
    7. Gutzwiller FS,
    8. Jorgensen L,
    9. Uhde M,
    10. Lisspers K
    (2018) Real-world retrospective cohort study ARCTIC shows burden of comorbidities in Swedish COPD versus non-COPD patients. NPJ Prim Care Respir Med 28: 33. doi:10.1038/s41533-018-0101-y
    OpenUrlCrossRef
  50. ↵
    1. Stolz D
    (2020) Chronic obstructive pulmonary disease risk: Does genetics hold the answer? Lancet Respir Med 8: 653–654. doi:10.1016/s2213-2600(20)30191-0
    OpenUrlCrossRef
  51. ↵
    1. Szklarczyk D,
    2. Gable AL,
    3. Lyon D,
    4. Junge A,
    5. Wyder S,
    6. Huerta-Cepas J,
    7. Simonovic M,
    8. Doncheva NT,
    9. Morris JH,
    10. Bork P, et al.
    (2019) STRING v11: Protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res 47: D607–D613. doi:10.1093/nar/gky1131
    OpenUrlCrossRefPubMed
  52. ↵
    1. Terracciano A,
    2. Stephan Y,
    3. Luchetti M,
    4. Gonzalez-Rothi R,
    5. Sutin AR
    (2017) Personality and lung function in older adults. J Gerontol B Psychol Sci Soc Sci 72: 913–921. doi:10.1093/geronb/gbv161
    OpenUrlCrossRef
  53. ↵
    1. Trudzinski FC,
    2. Alqudrah M,
    3. Omlor A,
    4. Zewinger S,
    5. Fliser D,
    6. Speer T,
    7. Seiler F,
    8. Biertz F,
    9. Koch A,
    10. Vogelmeier C, et al.
    (2019) Consequences of chronic kidney disease in chronic obstructive pulmonary disease. Respir Res 20: 151. doi:10.1186/s12931-019-1107-x
    OpenUrlCrossRef
  54. ↵
    1. Verhamme FM,
    2. Bracke KR,
    3. Joos GF,
    4. Brusselle GG
    (2015) Transforming growth factor-β superfamily in obstructive lung diseases. more suspects than TGF-β alone. Am J Respir Cell Mol Biol 52: 653–662. doi:10.1165/rcmb.2014-0282rt
    OpenUrlCrossRefPubMed
  55. ↵
    1. Wang Y,
    2. Li X,
    3. Wei B,
    4. Tung T-H,
    5. Tao P,
    6. Chien C-W
    (2019) Association between chronic obstructive pulmonary disease and dementia: Systematic review and meta-analysis of cohort studies. Demen Geriatr Cogn Disord Extra 9: 250–259. doi:10.1159/000496475
    OpenUrlCrossRef
  56. ↵
    1. Wang W,
    2. Wang P,
    3. Li Q,
    4. Peng Z,
    5. Wang X,
    6. Wang G,
    7. Hou J,
    8. Fan L,
    9. Liu S
    (2020) Alterations of grey matter volumes and network-level functions in patients with stable chronic obstructive pulmonary disease. Neurosci Lett 720: 134748. doi:10.1016/j.neulet.2020.134748
    OpenUrlCrossRef
  57. ↵
    1. Ward SM
    (2022) Postjunctional M2 muscarinic receptors augment neurally mediated cholinergic contractions of Murine airway smooth muscle. Function (Oxf) 3: zqab067. doi:10.1093/function/zqab067
    OpenUrlCrossRef
  58. ↵
    1. Xia C,
    2. Rook M,
    3. Pelgrim GJ,
    4. Sidorenkov G,
    5. Wisselink HJ,
    6. van Bolhuis JN,
    7. van Ooijen PMA,
    8. Guo J,
    9. Oudkerk M,
    10. Groen H, et al.
    (2020) Early imaging biomarkers of lung cancer, COPD and coronary artery disease in the general population: Rationale and design of the ImaLife (imaging in lifelines) study. Eur J Epidemiol 35: 75–86. doi:10.1007/s10654-019-00519-0
    OpenUrlCrossRef
  59. ↵
    1. Yin M,
    2. Wang H,
    3. Hu X,
    4. Li X,
    5. Fei G,
    6. Yu Y
    (2019) Patterns of brain structural alteration in COPD with different levels of pulmonary function impairment and its association with cognitive deficits. BMC Pulm Med 19: 203. doi:10.1186/s12890-019-0955-y
    OpenUrlCrossRefPubMed
  60. ↵
    1. Zaied R,
    2. Fadason T,
    3. O’Sullivan J
    (2022) De novo identification of complex multimorbid conditions by integration of gene regulation and protein interaction networks with genome-wide association studies. Research Square. doi:10.21203/rs.3.rs-1313207/v1 (Preprint posted March 1, 2022).
    OpenUrlCrossRef
  61. ↵
    1. Zhou JJ,
    2. Cho MH,
    3. Castaldi PJ,
    4. Hersh CP,
    5. Silverman EK,
    6. Laird NM
    (2013) Heritability of chronic obstructive pulmonary disease and related phenotypes in smokers. Am J Respir Crit Care Med 188: 941–947. doi:10.1164/rccm.201302-0263oc
    OpenUrlCrossRefPubMed
  62. ↵
    1. Zhu Z,
    2. Wang X,
    3. Li X,
    4. Lin Y,
    5. Shen S,
    6. Liu CL,
    7. Hobbs BD,
    8. Hasegawa K,
    9. Liang L, International COPD Genetics Consortium, et al.
    (2019) Genetic overlap of chronic obstructive pulmonary disease and cardiovascular disease-related traits: A large-scale genome-wide cross-trait analysis. Respir Res 20: 64. doi:10.1186/s12931-019-1036-8
    OpenUrlCrossRefPubMed
  63. ↵
    1. Zhu X,
    2. Duren Z,
    3. Wong WH
    (2021) Modeling regulatory network topology improves genome-wide analyses of complex human traits. Nat Commun 12: 2851. doi:10.1038/s41467-021-22588-0
    OpenUrlCrossRef
Back to top
Download PDF
Email Article
Citation Tools
De novo discovery of COPD co-traits
Evgeniia Golovina, Tayaza Fadason, Rachel K Jaros, Haribalan Kumar, Joyce John, Kelly Burrowes, Merryn Tawhai, Justin M O’Sullivan
Life Science Alliance Dec 2022, 6 (3) e202201609; DOI: 10.26508/lsa.202201609
Twitter logo Facebook logo Mendeley logo
Issue Cover

In this Issue

Volume 6, No. 3
March 2023
Advertisement

Jump to section

Subjects

Related Articles

  • No related articles found.

Cited By...

  • No citing articles found.

More in this TOC Section

Similar Articles