Lung Tissue Multi-Layer Network Analysis Uncovers the Molecular Heterogeneity of COPD

BACKGROUND

Chronic Obstructive Pulmonary Disease (COPD) is a heterogeneous condition. We hypothesized that the unbiased integration of different COPD lung omics using a novel multi-layer approach may unravel mechanisms associated with clinical characteristics.

METHODS

We profiled mRNA, miRNA and methylome in lung tissue samples from 135 former smokers with COPD. For each omic (layer) we built a patient network based on molecular similarity. The three networks were used to build a multi-layer network, and optimization of multiplex-modularity was employed to identify patient communities across the three distinct layers. Uncovered communities were related to clinical features.

RESULTS

We identified five patient communities in the multi-layer network which were molecularly distinct and related to clinical characteristics, such as FEV1 and blood eosinophils. Two communities (C#3 and C#4) had both similarly low FEV1 values and emphysema, but were molecularly different: C#3, but not C#4, presented B and T cell signatures and a downregulation of secretory (SCGB1A1/SCGB3A1) and ciliated cells. A machine learning model was set up to discriminate C#3 and C#4 in our cohort, and to validate them in an independent cohort. Finally, using spatial transcriptomics we characterized the small airway differences between C#3 and C#4, identifying an upregulation of T/B cell homing chemokines, and bacterial response genes in C#3.

CONCLUSIONS

A novel multi-layer network analysis is able to identify clinically relevant COPD patient communities. Patients with similarly low FEV1 and emphysema can have molecularly distinct small airways and immune response patterns, indicating that different endotypes can lead to similar clinical presentation.

Epigenome-Wide Association Studies of COPD and Lung Function: A Systematic Review

BACKGROUND

Chronic Obstructive Pulmonary Disease (COPD) results from gene-environment interactions over the lifetime. These interactions are captured by epigenetic changes, such as DNA methylation. This systematic review synthesizes evidence from epigenome-wide association studies (EWAS) related to COPD and lung function.

METHODS

Systematic literature search on PubMed, Embase and CINAHL databases, identified 1947 articles that investigated epigenetic changes associated with COPD/lung function; 17 of them met our eligibility criteria from which data was manually extracted. Differentially methylated positions (DMPs) and/or annotated genes, were considered replicated if identified by ≥2 studies with a p<1 x 10-4.

RESULTS

Ten studies profiled DNA methylation changes in blood and 7 in respiratory samples, including surgically resected lung tissue (n=3), small airways epithelial brushings (n=2), bronchoalveolar lavage (n=1) and sputum (n=1). Main results showed: (1) high variability in study design, covariates and effect sizes, which prevented a formal meta-analysis; (2) in blood samples, 51 DMPs were replicated in relation to lung function and 12 related to COPD; (3) in respiratory samples, 42 DMPs were replicated in relation to COPD but none in relation to lung function; and, (4) in COPD vs. control studies, 123 genes (2.6% of total) were shared between ≥1 blood and ≥1 respiratory sample and associated with chronic inflammation, ion transport and coagulation.

CONCLUSIONS

There is high heterogeneity across published COPD/lung function EWAS studies. A few genes (n=123; 2.6%) were replicated in blood and respiratory samples, suggesting that blood can recapitulate some changes in respiratory tissues. These findings have implications for future research.

Lung immune signatures define two groups of end-stage IPF patients

BACKGROUND

The role of the immune system in the pathobiology of Idiopathic Pulmonary Fibrosis (IPF) is controversial.

METHODS

To investigate it, we calculated immune signatures with Gene Set Variation Analysis (GSVA) and applied them to the lung transcriptome followed by unbiased cluster analysis of GSVA immune-enrichment scores, in 109 IPF patients from the Lung Tissue Research Consortium (LTRC). Results were validated experimentally using cell-based methods (flow cytometry) in lung tissue of IPF patients from the University of Pittsburgh (n = 26). Finally, differential gene expression and hypergeometric test were used to explore non-immune differences between clusters.

RESULTS

We identified two clusters (C#1 and C#2) of IPF patients of similar size in the LTRC dataset. C#1 included 58 patients (53%) with enrichment in GSVA immune signatures, particularly cytotoxic and memory T cells signatures, whereas C#2 included 51 patients (47%) with an overall lower expression of GSVA immune signatures (results were validated by flow cytometry with similar unbiased clustering generation). Differential gene expression between clusters identified differences in cilium, epithelial and secretory cell genes, all of them showing an inverse correlation with the immune response signatures. Notably, both clusters showed distinct features despite clinical similarities.

CONCLUSIONS

In end-stage IPF lung tissue, we identified two clusters of patients with very different levels of immune signatures and gene expression but with similar clinical characteristics. Weather these immune clusters differentiate diverse disease trajectories remains unexplored.

Blood Immunophenotypes of Idiopathic Pulmonary Fibrosis: Relationship with Disease Severity and Progression

(1) The role of the immune response in the pathogenesis of idiopathic pulmonary fibrosis (IPF) remains controversial. We hypothesized that peripheral blood immune phenotypes will be different in IPF patients and may relate to the disease severity and progression. (2) Whole blood flow cytometry staining was performed at diagnosis in 32 IPF patients, and in 32 age- and smoking-matched healthy controls. Thirty-one IPF patients were followed up for one year and categorized as stable or progressors based on lung function, deterioration and/or death. At 18-60 months, immunophenotypes were characterized again. (3) The main results showed that: (1) compared to matched controls, at diagnosis, patients with IPF showed more neutrophils, CD8+HLA-DR+ and CD8+CD28- T cells, and fewer B lymphocytes and naïve T cells; (2) in IPF, circulating neutrophils, eosinophils and naïve T cells were associated with lung function abnormalities; (3) patients whose disease progressed during the 12 months of follow-up showed evidence of cytotoxic dysregulation, with increased CD8+CD28- T cells, decreased naïve T cells and an inverted CD4/CD8 ratio at baseline; and (4) blood cell alterations were stable over time in survivors. (4) IPF is associated with abnormalities in circulating immune cells, particularly in the cytotoxic cell domain. Patients with progressive IPF, despite antifibrotic therapy, present an over-activated and exhausted immunophenotype at diagnosis, which is maintained over time.

When GETomics meets aging and exercise in COPD

The term GETomics has been recently proposed to illustrate that human health and disease are actually the final outcome of many dynamic, interacting and cumulative gene (G) - environment (E) interactions that occur through the lifetime (T) of the individual. According to this new paradigm, the final outcome of any GxE interactions depends on both the age of the individual at which such GxE interaction occurs as well as on the previous, cumulative history of previous GxE interactions through the induction of epigenetic changes and immune memory (both lasting overtime). Following this conceptual approach, our understanding of the pathogenesis of chronic obstructive pulmonary disease (COPD) has changed dramatically. Traditionally believed to be a self-inflicted disease induced by tobacco smoking occurring in older men and characterized by an accelerated decline of lung function with age, now we understand that there are many other risk factors associated with COPD, that it occurs also in females and young individuals, that there are different lung function trajectories through life, and that COPD is not always characterized by accelerated lung function decline. In this paper we discuss how a GETomics approach to COPD may open new perspectives to better understand its relationship with exercise limitation and the ageing process.

Telomere Length but Not Mitochondrial DNA Copy Number Is Altered in Both Young and Old COPD

Accelerated ageing is implicated in the pathogenesis of respiratory diseases as chronic obstructive pulmonary disease (COPD), but recent evidence indicates that the COPD can have roots early in life. Here we hypothesise that the accelerated ageing markers might have a role in the pathobiology of young COPD. The objective of this study was to compare two hallmarks of ageing, telomere length (TL), and mitochondrial DNA copy number (mtDNA-CN, as a surrogate marker of mitochondrial dysfunction) in young (≤ 50 years) and old (>50 years) smokers, with and without COPD. Both, TL and mtDNA-CN were measured in whole blood DNA by quantitative PCR [qPCR] in: (1) young ever smokers with (n = 81) or without (n = 166) COPD; and (2) old ever smokers with (n = 159) or without (n = 29) COPD. A multivariable linear regression was used to assess the association of TL and mtDNA-CN with lung function. We observed that in the entire study population, TL and mtDNA-CN decreased with age, and the former but not the latter related to FEV₁/FVC (%), FEV₁ (% ref.), and DLCO (% ref.). The short telomeres were found both in the young and old patients with severe COPD (FEV₁ <50% ref.). In addition, we found that TL and mtDNA-CN were significantly correlated, but their relationship was positive in younger while negative in the older patients with COPD, suggesting a mitochondrial dysfunction. We conclude that TL, but not mtDNA-CN, is associated with the lung function impairment. Both young and old patients with severe COPD have evidence of accelerated ageing (shorter TL) but differ in the direction of the correlation between TL and mtDNA-CN in relation to age.

Gut T cell-independent IgA responses to commensal bacteria require engagement of the TACI receptor on B cells

The gut mounts secretory immunoglobulin A (SIgA) responses to commensal bacteria through nonredundant T cell-dependent (TD) and T cell-independent (TI) pathways that promote the establishment of mutualistic host-microbiota interactions. SIgAs from the TD pathway target penetrant bacteria, and their induction requires engagement of CD40 on B cells by CD40 ligand on T follicular helper cells. In contrast, SIgAs from the TI pathway bind a larger spectrum of bacteria, but the mechanism underpinning their production remains elusive. Here, we show that the intestinal TI pathway required CD40-independent B cell-activating signals from TACI, a receptor for the innate CD40 ligand-like factors BAFF and APRIL. TACI-induced SIgA responses targeted a fraction of the gut microbiota without shaping its overall composition. Of note, TACI was dispensable for TD induction of IgA in gut-associated lymphoid organs. Thus, BAFF/APRIL signals acting on TACI orchestrate commensal bacteria-specific SIgA responses through an intestinal TI program.

Multi-level immune response network in mild-moderate Chronic Obstructive Pulmonary Disease (COPD)

Background

Chronic Obstructive Pulmonary Disease (COPD) is associated with an abnormal pulmonary and systemic immune response to tobacco smoking. Yet, how do immune cells relate within and between these two biological compartments, how the pulmonary infiltrate influences the lung transcriptome, and what is the role of active smoking vs. presence of disease is unclear.

Methods

To investigate these questions, we simultaneously collected lung tissue and blood from 65 individuals stratified by smoking habit and presence of the disease. The immune cell composition of both tissues was assessed by flow cytometry, whole lung transcriptome was determined with Affymetrix arrays, and we used Weighted Gene Co-expression Network Analysis (WGCNA) to integrate results.

Results

Main results showed that: (1) current smoking and the presence of COPD were both independently associated with a reduction in the proportion of lung T cells and an increase of macrophages, specifically those expressing CD80 + CD163+; (2) changes in the proportion of infiltrating macrophages, smoking status or the level of airflow limitation were associated to different WGCNA modules, which were enriched in iron ion transport, extracellular matrix and cilium organization gene ontologies; and, (3) circulating white blood cells counts were correlated with lung macrophages and T cells.

Conclusions

Mild-moderated COPD lung immune infiltrate is associated with the active smoking status and presence of disease; is associated with changes in whole lung tissue transcriptome and marginally reflected in blood.

Electronic supplementary material

The online version of this article (10.1186/s12931-019-1105-z) contains supplementary material, which is available to authorized users.

Do sputum or circulating blood samples reflect the pulmonary transcriptomic differences of COPD patients? A multi-tissue transcriptomic network META-analysis

BACKGROUND

Previous studies have identified lung, sputum or blood transcriptomic biomarkers associated with the severity of airflow limitation in COPD. Yet, it is not clear whether the lung pathobiology is mirrored by these surrogate tissues. The aim of this study was to explore this question.

METHODS

We used Weighted Gene Co-expression Network Analysis (WGCNA) to identify shared pathological mechanisms across four COPD gene-expression datasets: two sets of lung tissues (L1 n = 70; L2 n = 124), and one each of induced sputum (S; n = 121) and peripheral blood (B; n = 121).

RESULTS

WGCNA analysis identified twenty-one gene co-expression modules in L1. A robust module preservation between the two L datasets was observed (86%), with less preservation in S (33%) and even less in B (23%). Three modules preserved across lung tissues and sputum (not blood) were associated with the severity of airflow limitation. Ontology enrichment analysis showed that these modules included genes related to mitochondrial function, ion-homeostasis, T cells and RNA processing. These findings were largely reproduced using the consensus WGCNA network approach.

CONCLUSIONS

These observations indicate that major differences in lung tissue transcriptomics in patients with COPD are poorly mirrored in sputum and are unrelated to those determined in blood, suggesting that the systemic component in COPD is independently regulated. Finally, the fact that one of the preserved modules associated with FEV1 was enriched in mitochondria-related genes supports a role for mitochondrial dysfunction in the pathobiology of COPD.

The soluble pattern recognition receptor PTX3 links humoral innate and adaptive immune responses by helping marginal zone B cells

Cerutti and collaborators show that the humoral arms of the innate and adaptive immune systems are functionally interconnected by pentraxin 3, a soluble pattern recognition receptor that couples innate immune recognition with antibody-inducing function.

Pentraxin 3 (PTX3) is a fluid-phase pattern recognition receptor of the humoral innate immune system with ancestral antibody-like properties but unknown antibody-inducing function. In this study, we found binding of PTX3 to splenic marginal zone (MZ) B cells, an innate-like subset of antibody-producing lymphocytes strategically positioned at the interface between the circulation and the adaptive immune system. PTX3 was released by a subset of neutrophils that surrounded the splenic MZ and expressed an immune activation–related gene signature distinct from that of circulating neutrophils. Binding of PTX3 promoted homeostatic production of IgM and class-switched IgG antibodies to microbial capsular polysaccharides, which decreased in PTX3-deficient mice and humans. In addition, PTX3 increased IgM and IgG production after infection with blood-borne encapsulated bacteria or immunization with bacterial carbohydrates. This immunogenic effect stemmed from the activation of MZ B cells through a neutrophil-regulated pathway that elicited class switching and plasmablast expansion via a combination of T cell–independent and T cell–dependent signals. Thus, PTX3 may bridge the humoral arms of the innate and adaptive immune systems by serving as an endogenous adjuvant for MZ B cells. This property could be harnessed to develop more effective vaccines against encapsulated pathogens.