The persistent inflammation in COPD: is autoimmunity the core mechanism?

Abstract

COPD poses a significant global public health challenge, primarily characterised by irreversible airflow restriction and persistent respiratory symptoms. The hallmark pathology of COPD includes sustained airway inflammation and the eventual destruction of lung tissue structure. While multiple risk factors are implicated in the disease's progression, the underlying mechanisms remain largely elusive. The perpetuation of inflammation is pivotal to the advancement of COPD, emphasising the importance of investigating these self-sustaining mechanisms for a deeper understanding of the pathogenesis. Autoimmune responses constitute a critical mechanism in maintaining inflammation, with burgeoning evidence pointing to their central role in COPD progression; yet, the intricacies of these mechanisms remain inadequately defined. This review elaborates on the evidence supporting the presence of autoimmune processes in COPD and examines the potential mechanisms through which autoimmune responses may drive the chronic inflammation characteristic of the disease. Moreover, we attempt to interpret the clinical manifestations of COPD through autoimmunity.

Shareable abstract

Understanding the pivotal role of autoimmune responses in sustaining inflammation in COPD may guide targeted interventions and improve outcomes, but further research is crucial to unravel this complex interplay. https://bit.ly/3Hv42f2

Introduction

COPD, characterised by persistent respiratory symptoms and airflow limitation, is a worldwide public health challenge because of its high prevalence and related disability and mortality [1]. It has become the third leading cause of death [2, 3]. Persistent airway inflammation and ultimate destruction of lung tissue are the main pathological characteristics [1–3]. Multiple risk factors, including cigarette smoking, air pollution, occupational dust or chemicals, indoor products of biomass fuels, infections, early life events, airway remodelling, deficient lung repair and genetic susceptibilities are involved in the occurrence and progression of the disease [4]. Despite advancements in the understanding of the pathogenesis of COPD, major issues remain unresolved: 1) why do only some people develop COPD despite exposure to similar risk factors? 2) why does inflammation persist even after risk factor removal? 3) what causes disease exacerbations? and 4) why are conventional anti-inflammatory drugs like glucocorticoids ineffective for inflammation control?

The perpetuation of inflammation is key to the progression of COPD, making the exploration of these mechanisms essential for understanding the pathogenesis of the disease. Autoimmune responses serve as a significant mechanism for sustaining inflammation, and recent evidence increasingly supports their crucial role in the progression of COPD [5–7]. However, the specific mechanisms remain inadequately defined, and it is unclear whether they can offer insights into the unresolved complexities of the disease. In this comprehensive review, our primary focus is on elucidating the evidence and mechanisms of autoimmunity in COPD. More importantly, we attempt to use autoimmunity to explain the clinical phenomena associated with COPD, in hopes of providing fresh insights into the research of its pathogenic mechanisms.

The evidence of autoimmunity in COPD

Over the past two decades, accumulating evidence strongly supports the role of autoimmunity in COPD development. A pioneer survey revealed that 32% of nonsmoker COPD patients (22 in total) exhibited organ-specific autoimmune diseases and autoantibodies, particularly thyroid diseases [8]. The authors proposed that the link between organ-specific autoimmune diseases and COPD stems from the shared embryonic origin of the lungs and other organs [8]. Moreover, the majority of nonsmoker COPD patients are women, who often have higher B-cell counts and exhibit stronger antibody responses compared to men [8, 9].

Smoking, the most significant risk factor for COPD, is also strongly associated with the occurrence of various autoimmune diseases in humans, such as rheumatoid arthritis (RA), inflammatory bowel disease and systemic lupus erythematosus (SLE) [10–14]. A recent study involving 2396 subjects from the Genetic Epidemiology of COPD (COPDGene) revealed that the interferon (IFN)-stimulated gene signatures in bronchial brushings and peripheral blood of COPD patients correlated with lung function, exacerbations and airway wall thickness on chest computed tomography scans, and exhibited a similar blood transcriptomic signature to patients with SLE and respiratory viral infection [15]. The blood samples collected in the COPDGene study are obtained during periods of stable disease, at least 30 days after the last exacerbation, indicating that the identified gene signature is likely a result of sustained autoimmune responses rather than acute viral infection [15]. Besides, B-cell activating factor of tumour necrosis factor family (BAFF), which is an important autoimmune risk, is upregulated in COPD [16–18]. Targeting BAFF has demonstrated beneficial effects in treating both autoimmune diseases and COPD [19]. The autocrine or paracrine action of BAFF by B cells maintains their self-development and proliferation, and is potentially a key factor in the development of autoimmunity in COPD [19].

The generation of self-antigens

In COPD, adaptive immune responses can be activated by two types of antigens: those from the external environment and those from autologous tissue or cells. Foreign antigens primarily come from airway-colonising microorganisms, particulate matter or cigarette smoke. The activation of adaptive immune responses by foreign antigens is responsible for clearing invaders and is typically considered protective for the body. As a result, local inflammation should gradually resolve after eliminating foreign irritants, rather than progressing. Consistent with this, oral administration of bacterial lysates or vaccination, which mainly consist of bacterial antigens, is beneficial in preventing the progression of COPD, especially acute exacerbations [20–22]. However, the history of pathogen-induced acute exacerbations of COPD is a risk factor rather than a protective factor, indicating that foreign antigens are unlikely to be the primary inducers of pathogenic adaptive immune responses in COPD [23, 24]. In fact, self-antigens are likely to be the pathogenic antigens in COPD.

Self-antigens can emerge through various mechanisms, encompassing post-translational modifications of proteins, protein degradation and the release of cellular contents. These aforementioned mechanisms have been substantiated to occur in the context of COPD.

Self-antigens produced by post-translational modification of proteins

Oxidative stress has been conclusively established as a critical factor in the initiation and progression of COPD. One of its notable effects is the induction of oxidative modifications in proteins, with one of the most significant and irreversible modifications being protein carbonylation (figure 1) [25, 26]. Carbonylation of proteins can result from the oxidation of various amino acid residues, including lysine, arginine, proline and serine, showcasing a notable degree of diversity [26]. The serum antibody titre against carbonyl-modified proteins is significantly elevated in COPD patients (Global Initiative for Chronic Obstructive Lung Disease (GOLD) 3) compared with healthy controls [25]. These antibodies, predominantly IgG1 isotype, possess high specificity and affinity and have the potential to trigger autoimmune diseases [25]. An ozone-induced COPD mouse model also exhibits increased antibody levels against carbonyl-modified proteins [25]. Although oxidative stress has been well documented as a pivotal factor in disease progression, the potential impact of these self-reactive carbonylated proteins may have been considerably underestimated.

FIGURE 1.

The production of self-antigens leads to the occurrence of autoimmune responses in COPD. Risk factors trigger the production of reactive oxygen species (ROS), epithelial cell injury and the activation of inflammatory cells, including macrophages and neutrophils, resulting in the release of inflammatory factors and proteases. ROS are able to mediate protein carbonyl modifications. Proteases, such as macrophage-derived matrix metalloproteinases (MMPs) and neutrophil elastases, play a role in breaking down elastic protein fibres and promoting the generation of elastin fragments. Carbonylated proteins, elastin fragments and extracellular traps produced by activated neutrophils can act as self-antigens, leading to the occurrence of autoimmune responses. This process is facilitated by the presentation of self-antigens by conventional dendritic cells type 2 (cDC2s), leading to the activation of T-helper 1 (T_H1), T-helper 17 (T_H17) and plasma cells. Ultimately, this cascade leads to the production of autoantibodies and inflammatory factors, including interferon-γ (IFN-γ), interleukin (IL)-17A and IL-21. The activation of autoimmunity perpetuates airway inflammation independently of further cigarette smoking because it causes additional epithelial damage, recruitment of inflammatory cells and production of self-antigens, forming a positive feedback loop. iBALT: inducible bronchus-associated lymphoid tissue; NETs: neutrophil extracellular traps; T_FH: T follicular helper cells.

Citrullinated proteins, a characteristic feature of RA, are slightly elevated in the lung tissue of COPD patients, irrespective of smoking [27]. Consistently, protein-arginine deiminase type-2 (PAD2) and PAD4, enzymes responsible for citrullination and typically upregulated in conditions of hypoxia or low pH, are also upregulated in lung samples from COPD patients [27]. The presence of citrullinated proteins in COPD lungs is strongly associated with the inflammatory state of the disease [27]. Intriguingly, the level of citrullination in COPD lung tissue surpasses that of other organs, including lymph nodes, liver, kidneys, spleen, ovaries, heart and skeletal muscle, indicating a heightened susceptibility of lung tissue to protein citrullination and potential autoimmune-related disorders [27].

Self-antigens derived from degradation of the extracellular matrix

Destruction of lung extracellular matrix, especially elastin fibres, is a key characteristic of COPD. Neutrophils and macrophages can release proteolytic enzymes like elastase and matrix metalloproteinases (MMPs), leading to lung tissue damage and the development of emphysema (figure 1) [28, 29]. When researchers coculture blood CD4⁺ T cells from COPD patients or healthy controls with components of the lung extracellular matrix (albumin, collagen I and elastin peptides), they observe that only T cells from COPD patients are specifically activated by elastin peptides but not by albumin or collagen I [30]. This activation results in the secretion of IFN-γ, indicating a predominant T-helper 1 (T_H1) immune response triggered by elastin peptides [30]. Additionally, COPD patients display increased levels of elastin antibodies in their plasma, while elastin-responsive B cells are also elevated in lung tissue [30].

Elastin fragmentation is involved in COPD development, leading researchers to propose that elastin fragments may act as neo-self-antigens to initiate autoimmune responses in the disease. A study has confirmed in mice that just 2 weeks of cigarette smoke exposure increases MMP-12 secretion by alveolar macrophages, promoting elastin degradation and fragment generation [31]. After the cessation of exposure, these elastin fragments function as self-antigens, triggering the generation of immune memory specifically directed towards self-proteins [31]. Consequently, re-exposure of elastin or elastin peptides leads to stronger COPD-like phenotypes in mice, including short-term exposure-induced T_H17 immune response, neutrophilic inflammation and increased mucus secretion, as well as long-term exposure-induced epithelial cell death, airway remodelling, lung function decline and emphysema (figure 2) [31]. This explains the continued progression of COPD even after smoking cessation. The proportion of T_H1 and T_H17 cells, which are responsive to elastin peptides, has also been shown to increase in the peripheral blood of COPD patients [31].

FIGURE 2.

Matrix metalloproteinase-12 (MMP-12)-cleaved elastin fragments drive smoke-induced autoimmunity in a mouse model of COPD. Exposure to cigarette smoke in mice for 2 weeks triggers the activation of alveolar macrophages and the secretion of MMP-12, resulting in the cleavage of elastin into active fragments. The primary impact of these elastin fragments is to stimulate lymphocytes, leading to the production of memory T cells and B cells within the 2-week period following cigarette smoke exposure. The presence of memory T cells and B cells enhances the immune response upon re-introduction of elastin, resulting in more robust secondary immune responses in mice. Acute challenge with elastin induces human bronchitis-like features, including T-helper 17 (T_H17) immune responses, neutrophilic inflammation and excessive mucus production. Sub-chronic challenge with elastin produces emphysema-like characteristics, such as epithelial cell death, airway remodelling, decline in lung function and enlargement of airspaces.

Self-antigens resulting from the release of cellular contents

Intracellular antigens are normally sequestered within the cell membrane, preventing their recognition by the immune system. However, pathological conditions can lead to the extracellular exposure of intracellular antigens, triggering adaptive immune responses and the development of autoimmune diseases. Cell death plays a pivotal role in releasing intracellular components. During apoptotic cell death, intracellular components are redistributed onto the surface of dying cells, allowing them to be recognised as self-antigens by the immune system [32]. Injection of syngeneic apoptotic cells into normal mice induces the production of autoantibodies, including antinuclear autoantibodies, anticardiolipin and anti-single-stranded DNA antibodies [33]. Flow cytometry experiments using healthy or apoptotic cells cocultured with autoantibodies reveal that the autoantibodies selectively bind to apoptotic cells, predominantly relying on caspase activity [34]. It is hypothesised that the binding of autoantibodies to apoptotic cell surfaces also affects the recognition and clearance of apoptotic cells by the innate immune system, leading to impaired B-cell tolerance and the development of autoimmune diseases [34, 35]. Cell death has been implicated in the production of self-antigens and the progression of autoimmune diseases, such as SLE, and is also significantly increased in lung tissues of COPD patients [33–40]. Intraperitoneal injection of xenogeneic endothelial cells in rats elicits antibody production against endothelial cells, resulting in CD4⁺ T-cell accumulation in the lungs and the development of emphysema [41]. Considering the increased accumulation of dead cells in the lungs of COPD patients, it is reasonable to assume that self-antigens derived from dead cells may activate autoimmune responses and contribute to the progression of COPD. Indeed, anti-pulmonary epithelial cell antibodies have been detected in the lungs of COPD patients [42].

The formation of neutrophil extracellular trap (NETs), also called NETosis, is characterised by the release of intracellular DNA and proteins [43, 44]. It is speculated to induce the production of anti-ribonucleoprotein and anti-DNA antibodies in the serum of SLE patients, and is closely associated with increased autoimmune responses against insulin-producing cells in type 1 diabetes [45–47]. NETs are also significantly elevated in the induced sputum of COPD patients and correlate with the severity of airflow limitation [48–50]. Research has confirmed that cigarette smoke exposure-induced production of NETs can subsequently mediate the maturation and activation of dendritic cells (DCs), as well as T-cell immune responses and the occurrence of chronic inflammation in COPD (figure 1) [51]. Higher levels of extracellular DNA in NETs have been linked to the pathobiological characteristics of neutrophils in COPD-derived sputum [52].

The reduction in immune tolerance

The key to immune tolerance is to distinguish foreign from self and not to react to self. Immune tolerance is achieved through central and peripheral tolerance mechanisms. Central tolerance involves negative selection of T cells and B cells in specific lymphoid organs, namely the thymus for T cells and the bone marrow for B cells [53–55]. In the thymus, medullary thymic epithelial cells express autoimmune regulator, a transcriptional activator that allows the expression of numerous self-antigens specific to peripheral tissues. These self-antigens are presented to developing T cells, ensuring the elimination of self-reactive T cells [56, 57]. Autoimmune regulator (AIRE) also regulates the development of thymic DCs and regulatory T (Treg) cells, both of which contribute to immune tolerance [58]. Similarly, B cells undergo negative selection in the bone marrow, where immature B cells recognise self-antigens through the B-cell receptor, leading to the apoptosis of self-reactive B cells [59–63]. However, in healthy humans and mice, a significant number of autoreactive T cells and B cells can escape negative selection and enter the peripheral tissues [64]. To ensure self-tolerance in these leaked cells, peripheral tolerance mechanisms are necessary [53]. Peripheral tolerance involves specialised cell populations, such as Treg cells, T_H3 cells, type 1 Treg (Tr1) cells and regulatory B (Breg) cells [65]. Among these subsets, Treg cells have been extensively studied and consist of naturally producing Treg cells in the thymus (tTreg) and induced Treg cells in the periphery (pTreg) [66]. Thymic negative selection produces mature self-reactive tTreg cells, which play a crucial role in suppressing self-attack and facilitating tissue repair and regeneration [67–70]. pTreg cells differentiate from CD4⁺ T cells in the periphery under the influence of suppressive factors like transforming growth factor β (TGF-β), interleukin (IL)-10 and bacterial-derived metabolic products [71]. Breg cells contribute to immune tolerance by facilitating Treg cell differentiation and secreting suppressive cytokines like IL-10, TGF-β and IL-35 [72]. Breg cells have been proved to limit the inflammatory reactions in autoimmune diseases, including colitis, experimental autoimmune encephalomyelitis and arthritis [73–75]. Targeting costimulatory signalling to block T-cell activation has also been proven to induce tolerance [76–79]. Abnormal immune tolerance can lead to the development of autoimmune diseases.

The IL-7–IL-7 receptor and BAFF–BAFF receptor signalling pathways are pivotal in autoimmune disease development, supporting the survival of autoreactive T cells and B cells, respectively (figure 3) [80, 81]. Research has confirmed their notable activation in COPD T cells and B cells [16–18, 82]. Neo-self-antigens, which are able to restore autoreactive T-cell and B-cell pathogenicity, are also increased in COPD [83–85]. Pathogens can diminish B-cell immune tolerance by prompting B cells to erroneously ingest nearby cell membranes close to the pathogen [86]. Because infection is an important risk factor for acute exacerbation of COPD, it may also be important in the progression of the disease. Studies on mouse models have linked activation of autoreactive T cells and B cells with pattern recognition receptors, including toll-like receptors (TLRs), nucleotide oligomerisation domain-like receptor family pyrin domain containing 3 (NLRP3) inflammasome and cyclic GMP-AMP synthase, which are all prominently activated in COPD [87–95]. TLR7, located on the X chromosome, exhibits higher expression and stronger responses in women, potentially contributing to the female predisposition to autoimmune diseases [96–98].

FIGURE 3.

The decline of immune tolerance leads to the occurrence of autoimmune responses in COPD. The inadequate negative selection of T cells and B cells within lymphoid tissues contributes to the emergence of self-reactive lymphocytes, closely linked to the pathogenesis of autoimmune disorders. Complement protein 3a (C3a) may potentiate the hyperactivation of antigen-presenting cells via the downregulation of peroxisome proliferator-activated receptor-γ (PPAR-γ) expression, leading to the amplified activity of self-reactive T cells and B cells. The signalling pathways for interleukin 7 (IL-7)–IL-7 receptor and B-cell activating factor of tumour necrosis factor family (BAFF)–BAFF receptor support the survival of autoreactive T cells and B cells, respectively. Regulatory T (Treg) cells and regulatory B (Breg) cells are mainly responsible for the periphery tolerance; they inhibit the self-reactive T cells and B cells directly or secrete suppressive factors, e.g. IL-10 and transforming growth factor-β (TGF-β), to inhibit the self-reactive cells indirectly. Dashed arrows in the diagram indicate potential signalling defects in COPD, while solid arrows suggest that the signalling may be improved or amplified. iBALT: inducible bronchus-associated lymphoid tissue; IFN: interferon; T_FH cell: T-follicular helper cell; T_H cell: T-helper cell.

The excessive activation of antigen-presenting cells (APCs) may also be related to the decrease in peripheral tolerance and the onset of autoimmune diseases. Peroxisome proliferator-activated receptor γ (PPAR-γ) deficiency in APCs or macrophages can result in spontaneous lung inflammation or emphysema (figure 3) [99–101]. This process is confirmed to be driven by the upregulation of the secreted phosphoprotein 1 (SPP1) gene, primarily encoding osteopontin, a versatile cytokine associated with autoimmune responses [100]. It is worth noting that cigarette smoke has been firmly established to diminish PPAR-γ expression [100]. Importantly, PPAR-γ agonists can effectively reverse emphysema in mice, even in the presence of ongoing cigarette smoke exposure [100]. Additionally, the deficiency of the bioactive molecule complement protein 3a (C3a) or its receptor, C3aR, impairs APC activation of autoreactive T cells (figure 3) [101, 102]. These findings highlight the pivotal role of the APC C3aR–PPAR-γ–OPN signalling pathway in contributing to the diminished immune tolerance during the progression of COPD.

CD4⁺CD25⁺ Treg cell frequency remains unchanged in peripheral blood but is sharply downregulated in COPD lung tissues [30]. Prostaglandin I₂ (PGI₂) signalling programmes Treg cells for suppressive capabilities [103]. PGI₂ analogues promote Treg cell differentiation in mice and humans [103]. Forkhead box p3 (Foxp3) mediates the regulatory or suppressive functions of Treg cells, and targeting Foxp3 is expected to treat autoimmune diseases or tumours [104]. Breg cells are responsible for the inhibition of T_H1/T_H17 responses [105, 106]. It is of great significance to investigate if there are dysfunctional Breg cell responses in COPD, because T_H1/T_H17 immune responses are characteristic immunological responses in this disease. All these findings support the likelihood of impaired immune tolerance in COPD, and restoring immune tolerance could potentially be beneficial for the disease.

The potential mechanisms of autoimmunity in COPD

Both innate and adaptive immune responses are pivotal in the development and exacerbation of COPD. However, the exact cellular contributors and their mechanisms in instigating COPD-associated autoimmunity are not clearly elucidated. We have conducted a detailed review of the mechanisms by which various immune cells mediate the development of COPD using existing literature, in hope of providing some clues to the mechanisms of COPD's autoimmune responses, as specified below.

The adaptive immune responses

Dendritic cells

DCs are professional APCs. There are three major subsets of DCs, including two conventional DC populations (cDC1s and cDC2s) and plasmacytoid dendritic cells (pDCs) [107–109]. cDC1s (class II major histocompatibility complex (MHC II)⁺CD11C⁺CD141⁺) are supposed to cross present intracellular antigens and prime CD8⁺ T-cell responses [110–114]. cDC2s (MHC II⁺CD11C⁺CD1c⁺) are highly heterogeneous and can be further categorised, and mediate multiple immune responses [113–115]. pDCs (Lin⁻MHC II⁺CD303⁺CD304⁺) are able to produce more type I IFNs than other subsets and have a pivotal role in the control of viral infections [114, 116, 117]. DCs originate from hematopoietic stem cells and initially enter tissues in an immature form [118, 119]. Immature DCs are relatively immobile but responsible for detecting invading pathogens [119, 120]. Upon antigen acquisition, resting DCs become activated, upregulating the expression of molecules such as C-C motif chemokine receptor 7, MHC class I and II, costimulatory molecules and cytokines, and migrate to the lymph node [113, 119].

Costimulatory molecules expressed on DCs are upregulated in COPD and strongly associated with the activation of CD4⁺ T cells in the disease [121, 122]. DCs contribute to the development of COPD by inducing the formation of inducible bronchus-associated lymphoid tissue (iBALT), with cDC2s being the main subset involved [123, 124]. Mechanistically, cDC2s facilitate the differentiation and IL-21 secretion of T follicular helper (T_FH) cells through the activation of the OX40 ligand (OX40L)–OX40 axis [124]. T_FH cells and IL-21 derived from them play a crucial role in B-cell isotype switching and iBALT formation in the lung tissue of COPD [124].

B cells and T cells

B cells and T cells are effector cells of the adaptive immune system. They originate from the bone marrow, with B cells maturing in the bone marrow while T cells mature in the thymus [19, 125, 126]. Once mature, they disperse throughout the body in their naive state, ready to respond to antigens.

In COPD, B cells are primarily found in the third and fifth bronchioles and are associated with the reduction of alveolar attachments in the disease [127]. Mice with B-cell deficiency significantly resist smoke-induced emphysema [128]. The survival of mature B cells is mediated by BAFF signalling, which is upregulated in iBALT of COPD [127, 129]. Normally, BAFF is produced by stromal cells, airway epithelial cells, innate immune cells and T cells, but not B cells [16]. However, in COPD, B cells secrete significantly more BAFF, potentially establishing a self-sustaining loop that contributes to chronic injury in the disease [16]. Antagonising BAFF has been proved to attenuate inflammation and emphysema in cigarette smoke-exposed mice [17]. Additionally, the B-cell chemokine C-X-C motif chemokine ligand 13 (CXCL13) is found in COPD lungs and neutralising CXCL13 protects mice from cigarette smoke-induced lung inflammation and alveolar destruction [130, 131]. B cells carry out their functions through processes such as antibody production, immune memory establishment, class switching and antibody diversification. The secretion of antibodies by B cells is primarily achieved through their differentiation into plasma cells. Recent research has revealed that the gene expression characteristics of plasma cells are significantly upregulated in the lung tissues of COPD patients, displaying a clear correlation with the severity of emphysema in this patient population [132]. The levels of memory B cells in both peripheral blood and lung tissues of individuals with COPD show a notable increase compared to healthy controls [133]. Furthermore, even in the context of smoking, there is a preference for IgG in antibody class switching in COPD, whereas IgA is favoured in healthy controls [133]. This underscores the distinctions in adaptive immune responses between COPD patients and healthy individuals.

CD4⁺ T_H1, T_H17 and CD8⁺ T cells are all increased in the lung tissue of COPD patients and strongly correlated with disease severity [134–138]. Cigarette smoke exposure induces differentiation of naive CD4⁺ T cells into T_H17 cells and the secretion of IL-17A, leading to increased expression of macrophage chemokine monocyte chemoattractant protein-1 and MMP-12, neutrophil infiltration and emphysema development in mice [139, 140]. T_H1 and CD8⁺ T cells likely contribute to COPD progression by activating signal transducer and activator of transcription-4 and secreting IFN-γ [136]. Furthermore, a decrease in TLR4 expression on T_H1 cell surfaces, which is responsible for antibacterial responses, has been observed in COPD lung tissues compared to healthy controls [141]. Regarding Treg cells, a previous study identified three subsets: CD25⁺⁺CD45RA⁺ resting Treg (rTreg) cells, CD25⁺⁺⁺CD45RA⁻ activated anti-inflammatory Treg (aTregs) cells and CD25⁺⁺CD45RA⁻ cytokine-secreting (Fr III) cells [142]. In COPD, the anti-inflammatory subsets (aTreg, rTreg) are decreased, while the pro-inflammatory subset (Fr III Treg) is upregulated [142].

Inducible bronchus-associated lymphoid tissue

Adaptive immune responses primarily occur in secondary lymphoid tissues, such as lymph nodes and the spleen, to which pathogens have to travel [143, 144]. However, a rapid response is also necessary to address immediate threats. Tertiary lymphoid organs, which exist in nonlymphoid organs, fulfil this demand [145, 146]. In the lung tissue, these tertiary lymphoid organs, known as lymphoid follicles or iBALT, are typically absent in healthy individuals but form during infections or inflammation [123, 147]. They locate around distal airways and lung parenchyma [148].

The presence of iBALT is significantly increased in the lung tissue of COPD patients and is closely associated with emphysema in both human and mouse studies [16, 130]. Blocking the formation of iBALT in mice protects against smoke-induced emphysema [16, 17, 128, 130]. iBALT formation relies on the interaction between lymphotoxin β-receptor (LT-βR) on stromal organiser cells and lymphotoxin-α and lymphotoxin-β expressed on activated lymphocytes [149, 150]. Analysis of lung transcriptomic data from COPD patients has revealed the enrichment of the LT-βR signalling pathway [151]. Inhibiting LT-βR signalling not only disrupts cigarette smoke-induced iBALT formation but also prevents emphysema, promotes lung tissue regeneration and reverses airway fibrosis [151].

The innate immune responses

Airway epithelial cells

Once an intrusion is caught, the epithelial cells respond rapidly by secreting inflammatory mediators, leading to the infiltration of inflammatory cells into the lung and the activation of innate immune responses [134, 152]. The NF-κB signalling pathway is significantly activated in airway epithelial cells of COPD patients [153]. A recent study using single-cell transcriptome sequencing technology confirms that inflammatory gene expression in primary airway epithelial cells is significantly upregulated in COPD, including IL1A, IL1B, IL6, IL32, C-C motif chemokine ligand 20 (CCL20), colony-stimulating factor 1 (CSF1), NFKB inhibitor-ɑ (NFKBIA) and NFKB inhibitor-ζ (NFKBIZ), while genes with an anti-inflammatory function are significantly decreased, such as fatty acid binding protein 5 (FABP5) [154]. FABP5 also modulates corticosteroid responsiveness and the downregulation of this protein may, to some extent, explain glucocorticoid resistance in COPD [154]. The activation of innate immune responses in airway epithelial cells is considered an early indicator of inflammation.

Alveolar macrophages

Macrophages in the lung tissue can be classified into alveolar and interstitial macrophages. Among them, alveolar macrophages are considered more influential in the progression of COPD [28]. Alveolar macrophages are the most abundant cell in the alveolar lavage fluid of both healthy individuals and COPD patients [155, 156]. Unlike circulating monocytes, alveolar macrophages are long-lived cells that colonise the lungs shortly after birth and undergo self-renewal throughout life [157]. As tissue-specific macrophages, alveolar macrophages exhibit unique transcriptional and phenotypic characteristics that are largely thought to be determined by the microenvironment in which they are located [158–160]. Recent research has discovered that the microenvironment during the resolution phase of inflammation can induce epigenetic modifications and tolerogenic training in naive alveolar macrophages, leading to long-term immune paralysis of lung tissue [161]. Additionally, alveolar macrophages can generate immune memory in response to viral infections, facilitating the immune response against infections [162]. However, whether there is a trained immunity in alveolar macrophages in COPD is currently unclear.

Alveolar macrophages in COPD produce an abundance of inflammatory mediators, including reactive oxygen species, cytokines, chemokines, growth factors and proteases; their own numbers are also increased [163]. Excess secretion of MMPs by alveolar macrophages, especially MMP-12, leads to pulmonary structural damage in COPD [164]. The loss of endolysosomal cation channel mucolipin 3 (TRPML3), which regulates MMP-12 re-uptake exclusively in alveolar macrophages, also contributes to lung injury and emphysema in COPD [165]. Inducing apoptosis in alveolar macrophages alleviates elastase-induced pulmonary emphysema in mice [28]. In addition to their enhanced capacity to promote lung injury, alveolar macrophages in COPD also exhibit dysfunctional responses to infection, showing lower phagocytic and antigen-presenting activity [166]. Oxidative stress and mitochondrial dysfunction of alveolar macrophages have been associated with defective phagocytosis in COPD [167, 168]. Single-cell sequencing analysis of alveolar macrophages in human and mouse lung tissue has revealed their significant heterogeneity, with clustering into eight sub-clusters [156]. Among them, cluster 5 is enriched among COPD patients while cluster 0 is enriched in healthy controls [156]. Regardless of the clustering, all macrophages in COPD exhibit increased chemotaxis and inflammatory features [156]. In summary, alveolar macrophages in COPD seem to have higher pro-inflammatory but lower antimicrobial capacity, and are pathogenic in the occurrence and progression of COPD.

Neutrophils

Under normal conditions, neutrophils migrate to the site of injury or infection using cell surface receptors like C-X-C motif chemokine receptor 2 (CXCR2) [169]. Their main functions include degranulation, engulfing damaged cells or pathogens, and releasing NETs [169]. In COPD patients, an increase in neutrophils is associated with disease severity instead of with lung protection, suggesting dysfunctional neutrophils in COPD [170–172]. Some researchers believe that it is mainly related to their impaired direction sensing and misplacement [169, 173]. Neutrophils lose their precise targeting ability with age, which may contribute to the higher prevalence of COPD in the older population [169]. Neutrophils can be categorised into pro-inflammatory and anti-inflammatory subsets [174]. However, it is still debated whether the diverse phenotypes observed in disease pathogenesis result from different neutrophil subtypes or the high plasticity of neutrophils.

Neutrophils secrete elastase, which, in the context of COPD, contributes to the degradation of the lung extracellular matrix and the development of emphysema [29]. Elastase attaches to the surface of neutrophil exosomes to avoid degradation by α₁-antitrypsin (AAT) and uses exosomes to accurately target the extracellular matrix [29, 175]. The degradation of the extracellular matrix generates N-acetyl pro-gly-pro (PGP), a neutrophil chemoattractant that attracts neutrophils to the lungs via CXCR2 [176]. PGP levels are significantly elevated in bronchoalveolar lavage fluid samples from COPD patients, and its administration to mice induces alveolar enlargement [176, 177]. Therefore, elastase alone is sufficient to cause the expansion and persistence of neutrophilic inflammation in COPD. Additionally, elastase can cleave C-X-C motif chemokine receptor 1 (CXCR1) and IL-22 receptor, impairing the precise targeting of neutrophils that is mediated by CXCR1 and the antimicrobial immune responses mediated by IL-22 [178, 179].

NETs are web-like structures composed of cytosolic and granule proteins, including elastase, bound to decondensed chromatin in neutrophils [180]. These structures, formed through NETosis regulated by CXCR2, are designed to ensnare and eliminate pathogens, playing a crucial role in host immune defence [181]. However, dysfunctional NETosis can also contribute to the occurrence of disease. Increased NETosis is observed in sputum from patients with severe COPD and correlates with disease severity [182]. Mechanistically, heightened NETosis in COPD reduces neutrophil phagocytosis and leads to decreased microbial diversity in the airway [182, 183]. It has been reported that different stimuli can influence the composition of NETs [184, 185]. However, whether the composition of NETs changes in COPD and whether such changes contribute to the development and progression of the disease remain unclear.

Eosinophils

A subset of COPD patients exhibits higher eosinophil counts and shows sensitivity to inhaled corticosteroid (ICS) treatment. Eosinophils are widely distributed in lung tissue, with a higher density observed in COPD patients, particularly in advanced stages (GOLD 4) associated with emphysema and lung remodelling [186–188]. Within the localised lung microenvironment of eosinophils in COPD patients, the presence of GATA3⁺ cells, including innate lymphoid cells type 2 (ILC2) cells and T_H2 lymphocytes, is significantly higher compared to non-adjacent regions [186]. Moreover, COPD patients with higher eosinophil counts exhibit increased levels of type 2 cytokines such as IL-5, IL-13 and IL-33 [187, 189]. Experimental studies in mouse models confirm that viral infections can enhance eosinophil activation in COPD, facilitated by the secretion of macrophage colony-stimulating factor and IL-33 by lung epithelial cells [142, 186, 188, 189]. Eosinophil-derived IL-13 can induce the production of macrophage MMP-12 and contribute to the development of emphysema [186, 188]. Consistent with this, MMP-12 levels are elevated in COPD patients with sputum eosinophilia and emphysema [188]. Additionally, eosinophil accumulation has been observed in emphysematous sites in a mouse model of pancreatic porcine elastase-induced emphysema [190]. Recent findings also suggest that eosinophils may promote pancreatic porcine elastase-induced emphysema through lysosomal protease cathepsin L [191]. Eosinophils do not appear to be involved in the pathogenesis of autoimmune responses in COPD, which will be clearly elaborated in the next section. Other innate immune cells are less investigated in COPD and are not introduced in this review.

The clinical implications of autoimmunity in COPD

Although substantial evidence has confirmed the role of autoimmune responses in the progression of COPD (table 1), whether they constitute the core mechanism behind the chronic inflammation of COPD requires further research and discussion. A crucial aspect of this investigation is to determine if they can account for the clinical phenomena of COPD, including its pathological phenotypes, inflammatory endotypes, population susceptibilities, acute exacerbations and the sensitivity to glucocorticoid treatment, all of which we will elaborate on in detail.

TABLE 1.

Evidence and mechanisms of autoimmune responses in COPD

Findings	Sources	Relevance with autoimmunity	Reference(s)
Antibodies of carbonyl-modified proteins	COPD patients Ozone-induced COPD mouse model	Autoantibodies	[22]
Citrullinated proteins	COPD patients	Self-antigens	[24]
Elastin antibodies	COPD patients Cigarette smoke-induced mouse model	Autoantibodies	[27, 28]
Neutrophil extracellular trap	COPD patients Cigarette smoke-induced mouse model	Self-antigens	[45–48]
Activation of IL-7–IL-7R on T cells	COPD patients	Decreased immune tolerance	[79]
Activation of BAFF–BAFFR on B cells	COPD patients	Decreased immune tolerance	[13–15]
Activation of C3a–C3aR on APCs	COPD patients Cigarette smoke-induced mouse model	Decreased immune tolerance	[98, 99]
Downregulation of PPAR-γ on APCs	COPD patients Cigarette smoke-induced mouse model	Decreased immune tolerance	[96–98]
Downregulation of CD4+CD25+ Treg cells	COPD patients	Decreased immune tolerance	[27]

APC: antigen-presenting cell; BAFF: B-cell activating factor of tumour necrosis factor family; BAFFR: B-cell activating factor of tumour necrosis factor family receptor; C3a: complement protein 3a; C3aR: complement protein 3a receptor; IL-7: interleukin-7; IL-7R: interleukin-7 receptor; PPAR-γ: peroxisome proliferator-activated receptor-γ; Treg cell: regulatory T cell.

Autoimmunity and pathological phenotypes

COPD is characterised by small airway remodelling, constriction and lung tissue damage, primarily driven by chronic lung inflammation. Current evidence strongly links autoimmunity to both the persistent inflammation and the various phenotypes in COPD. Notably, COPD patients endure ongoing lung inflammation even after smoking cessation, with autoimmune responses providing a compelling explanation for this phenomenon. Additionally, in clinical practice, some patients develop emphysema without airflow limitation, suggesting alternative mechanisms beyond gas retention. Studies have revealed a significant correlation between B-cell-related signals and the emphysema phenotype, while such signals are deficient in bronchitis [192]. Moreover, research shows a close link between the presence of lung tissue lymphoid follicles and the production of autoantibodies, particularly in relation to smoking status [5]. This phenomenon is observed not only in smokers with COPD but also in some individuals with normal lung function who smoke, whereas it is rare in nonsmokers [5]. When it comes to the chronic bronchitis phenotype, T-cell-related autoimmune responses appear to play a more prominent role. In COPD patients with disease primarily characterised by chronic bronchitis, bronchial biopsy samples demonstrate a significant increase in T-cell numbers, which is inversely correlated with the values of forced expiratory volume in 1 s, while there is no significant alteration in the number of B cells [136, 193, 194].

Autoimmunity and inflammatory endotypes

Not all COPD patients exhibit the same type of lung inflammation; some predominantly feature type 1 inflammation, characterised by neutrophil infiltration, while others primarily have type 2 inflammation, characterised by eosinophil infiltration. Based on our previous discussion, it becomes clear that autoimmune responses in COPD are primarily linked to type 1, rather than type 2, inflammation. Consistently, patients with two different inflammation endotypes also respond differently to treatment. Patients with a predominance of type 1 inflammation are resistant to glucocorticoid therapy, whereas those with type 2 inflammation are glucocorticoid sensitive. Hence, we posit that autoimmune-mediated COPD represents a subtype of the disease. As for the mechanism underlying type 2 inflammation in COPD, research has shown that virus-induced COPD-like phenotypes in mice are primarily mediated by macrophage IL-13–IL-13 receptor signalling, independent of adaptive immunity [195]. Multiple pieces of evidence also suggest that eosinophilic COPD, which is often IL-13 dependent, may be induced by viruses [142, 186, 188, 189, 196]. Therefore, it is highly likely that the self-perpetuation of chronic inflammation in eosinophilic COPD can be mediated by innate immune responses.

Autoimmunity and population susceptibility

Another notable characteristic of COPD is its varying susceptibility among populations. Despite exposure to identical risk factors, such as smoking, the onset of COPD does not occur uniformly among individuals. Autoimmune mechanisms seem to provide an explanation. Individuals who possess a more active immune system or lower immune tolerance, such as women, face an increased risk of developing autoimmune disorders. Research has confirmed that the susceptibility of women to autoimmune diseases arises from the dysregulation of X chromosome gene expression and the influence of oestrogen [197]. Similarly, within the context of COPD, women exhibit an amplified susceptibility and vulnerability when compared to their male counterparts. This encompasses earlier onset, a more rapid decline in lung function, greater severity of emphysema, a higher incidence of exacerbations and a less favourable prognosis [198, 199]. Research has uncovered numerous genetic loci associated with susceptibility in individuals with COPD [200]. Among these, AAT deficiency (AATD) caused by mutations in the Serpin family A member 1 (SERPINA1) gene is clearly associated with the development of emphysema [201]. Previous research primarily attributed this association to an imbalance between proteases and antiproteases. However, recent studies suggest that patients with AATD exhibit characteristics of adaptive immune responses in the lungs similar to those seen in patients with common COPD [202]. Furthermore, research has shown that patients with AATD exhibit elevated levels of neutrophil-related self-antigens and autoantibodies in their serum [203]. Treatment with AAT appears to correspondingly reduce the levels of these self-antigens and autoantibodies [203]. Consequently, the development of emphysema mediated by AATD may not solely be attributed to an imbalance between proteases and antiproteases, autoimmune factors may also play a crucial role in this process. However, the immunological changes driven by other genetic variations remain unclear. Exploring whether these genetic loci are located on the X chromosome and are associated with regulation of X chromosome gene expression may offer new insights.

Autoimmunity and acute exacerbations

Acute exacerbations of COPD predict poor prognosis and are often triggered by infections [204]. The question arises as to why COPD patients fail to effectively eliminate certain pathogens upon secondary exposure and instead experience disease progression. It has been assumed that impaired lung clearance mechanisms contribute to the heightened vulnerability of COPD patients to pathogens, while the mechanisms underlying pathogen-induced disease progression remains unclear. The notion that the occurrence of acute exacerbations in COPD is mediated by autoimmune responses appears reasonable. As previously mentioned, infection can lead to a situation where B cells inadvertently engulf nearby cellular components of pathogens, resulting in a reduced level of self-immune tolerance [86]. Simultaneously, when pathogenic antigens exhibit a high degree of molecular structural similarity to self-antigens, the antibodies produced can react with self-antigens [205]. Furthermore, pathogens have the capacity to activate various immune cells through pathogen-associated molecular patterns. Given the presence of self-antigens in the lungs of COPD patients prior to infection, the pathogen-triggered production of additional self-antigens is expected to elicit a robust secondary immune response, potentially serving as the fundamental mechanism driving acute exacerbations of COPD. This process also helps explain why a history of previous acute exacerbations is a risk factor, but not a protective factor, in COPD. The specific mechanisms remain to be explored.

Autoimmunity and responsiveness to glucocorticoid therapy

The autoimmunity-associated lung inflammations in COPD are predominantly neutrophilic and often linked to glucocorticoid resistance [8, 206]. Glucocorticoids have been reported to promote the mobilisation and inflammatory responses of neutrophils, while inhibiting their apoptotic cell death [207–209]. Deficiency of glucocorticoid receptors in neutrophils from airway and lung tissue may contribute to glucocorticoid resistance in COPD [210]. The 2023 GOLD guidelines state that ICS treatment does not benefit COPD patients unless their blood eosinophil count exceeds 300 cells·μL⁻¹ [22]. Additionally, airway inflammation worsens after ICS withdrawal, despite long-term use attenuating inflammation and lung function decline [211]. In summary, autoimmunity-mediated lung inflammation in COPD is associated with insensitivity to ICS treatment.

Conclusions and perspectives

As stated above, autoimmune mechanisms offer explanations for unanswered questions in COPD research, including self-sustaining inflammation, population susceptibility, acute exacerbations and insensitivity to ICS treatment. Thus, we propose that autoimmunity is probably the core mechanism of the chronic inflammation in COPD, while innate immunity plays a crucial role in initiating autoimmune responses. Innate immune responses are the main source of newly generated self-antigens such as elastin fragments. Moreover, the resulting inflammatory microenvironment can restore the pathogenicity of autoreactive T cells and B cells, leading to a decline in immune tolerance. We hypothesise that targeting autoimmunity is likely to be effective in inhibiting the progression of COPD, while inhibiting the innate immune responses may alleviate the symptoms but exert little effect on disease progression, of which the exact mechanisms are still to be elucidated.

Diagnosis of COPD currently relies primarily on pulmonary function tests and imaging examinations. By the time a definitive diagnosis is established, patients often find themselves in the later stages of the disease, which poses challenges for intervention. Therefore, the search for a serum-based biomarker is of significant importance in enabling early detection of COPD. In the context of autoimmune diseases, the emergence of autoantibodies often precedes clinical symptoms or laboratory findings by several years [205]. Furthermore, these autoantibodies can serve as biomarkers for diagnosis, assessing disease activity and monitoring the efficacy of treatments in the autoimmune disease [205]. Therefore, autoantibodies associated with COPD may potentially serve as ideal biomarkers for the disease. A US patent application (20160097778) has introduced a method for diagnosing, assessing and selecting personalised treatment plans for COPD by detecting the reactivity of serum autoantibodies to carbonylated vimentin [212].

Therapeutic approaches targeting autoimmunity may decelerate or halt COPD progression. Monoclonal antibodies targeting CD20 to deplete B cells, e.g. rituximab, ocrelizumab and ofatumumab, are established in treating multiple sclerosis [213]. Belimumab, targeting BAFF, is employed in SLE management [214]. Antimalarial drugs with TLR-targeting activity, including chloroquine, hydroxychloroquine and quinacrine, have also been clinically successful in treating SLE [215]. Notably, both hydroxychloroquine and quinacrine also have the capacity to downregulate the expression of BAFF [216].

We propose that the production of self-antigens and the decline of immune tolerance could become two main promising directions in the mechanistic research of COPD. Future research in these two directions would enhance our comprehension of this intricate disease and, more importantly, guide the development of more effective treatment. Additionally, we can also identify potential autoimmunity-related biomarkers for early diagnosis and assessing prognosis, ultimately improving the management and outcomes of the disease.

Points for clinical practice and questions for future research

• COPD is a complex and heterogeneous chronic inflammatory disease, in which the self-perpetuating nature of inflammation plays a pivotal role in its progression.

• Elucidating the key factors maintaining chronic inflammation in COPD and developing targeted therapeutic approaches are the two most important aspects in the field of COPD research.

• The autoimmune response is likely to play a crucial role in maintaining chronic inflammation in COPD, and the development of targeted therapies in this regard holds promise for effectively inhibiting disease progression.

• The occurrence of the autoimmune response in COPD is primarily attributed to the generation of neo-self-antigens or the decline of immune tolerance. Future research should focus on these two aspects to uncover the correlation between autoimmune response and the progression of COPD.