Introduction

Lung transplantation (LTx) represents a therapeutic modality for patients afflicted with end-stage lung disease. The present median post-transplant survival stands at 6 years, with 5-year graft and patient survival rates approximating 65% [1]. Long-term survival remains constrained by both nonimmunologic and immunologic occurrences, including primary graft dysfunction (PGD), anastomotic complications, infections, acute rejection (AR) mediated by cellular (ACR) or antibody mechanisms (AMR), and chronic lung allograft dysfunction (CLAD). Despite advancements in immunosuppressive strategies, lung recipients exhibit elevated incidences of acute and chronic allograft rejection in contrast to recipients of other frequently transplanted solid organs.

The airway epithelium serves as the primary interface with the external environment and is perpetually exposed to the exposome. Traditionally viewed as providing a barrier function that regulates airway surface liquid and mucus secretion through ion transport, the respiratory epithelium is now recognised for its pivotal role in immunity [2]. Apart from physical protection, it offers chemical defence against harmful environmental agents, in collaboration with adjacent immune cell networks [3]. Therefore, the airway epithelium is involved in all potential complications after LTx. It both protects the lung and renders it vulnerable. Airway epithelial cells can engage in innate immunity via pattern-recognition receptors [4, 5], can act as antigen-presenting cells in certain contexts [6, 7], demonstrate direct microbial activity through secretion of antimicrobial products [8–11] or may mitigate inflammation by expressing immunotolerant molecules, as human leukocyte antigen (HLA)-Ib molecules (HLA-G, HLA-E, HLA-F) [12, 13].

However, despite growing recognition of the airway epithelium's role after LTx, there remains a scarcity of specific literature on the topic. Given the prognostic implications of immunologic and infectious events following LTx impacting the airway epithelium, this review aims to thoroughly investigate the pivotal role of these structural cells in the genesis of post-transplant complications. In the first part, we will describe major events critical to patient survival after LTx and delineate the involvement of the airway epithelium in each. Subsequently, in the second and third parts, we will elucidate the potential of the airway epithelium as a research model, encompassing both clinician and scientist perspectives, in shaping the future landscape of LTx.

Search strategy

Data for this narrative review were identified by searches of PubMed (https://pubmed-ncbi-nlm-nih-gov.proxy.insermbiblio.inist.fr), as well as references from relevant articles and reviews, using the following search terms: “lung transplant AND infection/anastomotic complications/primary graft dysfunction/CLAD/acute rejection/immunity”; “bronchial epithelium/ airway epithelium AND infection/anastomotic complications/primary graft dysfunction/CLAD/acute rejection/immunity”; and “lung transplant AND senescence/epithelial-to-mesenchymal transition”. Only articles published in English from inception to March 2024 were included.

Involvement of the airway epithelium in major post-transplant lung complicationsPrimary lung graft dysfunction and anastomotic complications

PGD represents a severe form of lung injury occurring within the initial 72 h following LTx. Clinical diagnosis relies on the presence of multilobar infiltrates on chest X-ray and a reduced ratio of arterial oxygen to inspired oxygen below 300, with obligatory exclusion of alternative causes such as left ventricular failure. The incidence of PGD ranges from 10 to 25%, constituting the primary cause of early mortality after LTx, contributing to nearly half of the short-term deaths [14–16]. Survivors of PGD exhibit decreased long-term mortality, particularly among those experiencing severe grade 3 PGD. This reduced long-term mortality is associated with an increased risk of developing bronchiolitis obliterans syndrome (BOS) [17, 18]. The mechanisms underpinning PGD primarily involve ischaemia–reperfusion injury (IRI), which is contingent upon the severity of ischaemic insult, encompassing warm and cold ischaemia during the surgical procedure. The absence, in most LTx, of re-anastomosis of bronchial arteries during transplantation results in bronchial perfusion via a retrograde flow from the pulmonary circulation, exacerbating ischaemic injury [19] and potentially leading to anastomotic complications [20, 21].

PGD can impact both short- and long-term outcomes after LTx by targeting bronchial epithelium. Indeed, early IRI stages entail significant oxidative stress, precipitating airways epithelial and endothelial dysfunction. Airway epithelial cells release chemokines and damage-associated molecular patterns (DAMPs) [22–24], predictive factors of PDG such as club cell secretory component [25], fostering immune cell recruitment, notably neutrophils and macrophages, and facilitating antigen presentation [26, 27]. Transcriptomic alterations and gene expression changes in endothelial and airway epithelial cells induced by IRI promote endothelial lesions and immune cell recruitment. This highlights that the epithelium is not only a target of IRI but also plays an amplifying role in PGD-associated allograft injury [28]. Immune cells and especially natural killer (NK) cells seem to be critical in translating epithelial cell stress during IRI to allograft damage because, in LTx recipients following PGD, NK cells are observed in and around airways [29]. Furthermore, NK cell-mediated host-versus-graft activity may confer benefits following LTx, particularly in cases of HLA mismatch, potentiating HLA or neo-antigen presentation by bronchial epithelium and ensuing immune responses [22, 30–32]. Regarding long-term outcomes, PGD may accelerate cellular senescence by decreasing DNA methylation in airway epithelial cells, isolated and cultured by transbronchial brush [33]. Biologic or epigenetic age reflecting DNA methylation patterns may not always align with chronological age. Notably, epigenetic age was found to be higher in airway epithelial cells of LTx patients who had experienced PGD, independent of donor and recipient characteristics, engendering enduring long-term repercussions for lung transplant recipients [33]. Additionally, IRI may upregulate anti-donor anti-major histocompatibility complex expression and autoantigen antibody production, constituting risk factors for BOS [34].

Anastomotic complications are defined as any injuries involving bronchial anastomoses. These complications are associated with significant morbidity and mortality, with risks such as mediastinitis or acute respiratory failure. Incidence varies widely, ranging from 1.4 to 33%, reflecting inconsistent definitions [20, 35]. The pathophysiology of these complications is primarily attributed to donor bronchial ischaemia, as allograft airway perfusion relies heavily on retrograde flow from the low-pressure, poorly oxygenated pulmonary arterial system [20, 21]. Various risks factors have been identified, with compromised blood perfusion of the bronchial epithelium and mucosa emerging as the principal suspected common pathway [36, 37]. These factors include hypoperfusion due to surgery or intensive care unit management [38], as well as anatomical considerations such as right-sided anastomosis, where reduced blood perfusion may occur due to the singular bronchial artery supply [39]. Organ preservation techniques [37], mechanical ventilation because of high positive end-expiratory pressure level, conditions such as PGD and IRI leading to interstitial oedema [40–42], as well as microbiologic contamination, especially with fungi such as Aspergillus, Candida, Rhizopus or Mucor [43–45], can also negatively affect bronchial mucosa blood flow [46, 47] and then amplify airway epithelium ischaemia and necrosis. ACR is a controversial risk factor, but reduction in graft perfusion during ACR has been documented [48]. Ischaemic time and surgical techniques play roles in airway ischaemia, with alternative approaches, such as bronchial arterial revascularisation, showing potential in reducing anastomotic complications [49, 50]. However, evidence drawing a direct link between anastomotic complications and airway epithelium, without implication of blood flow reduction, is scarce.

Infection

Solid organ transplant (SOT) recipients, particularly LTx recipients, face major susceptibility to infections. The transplanted airway epithelium, serving as the primary interface between host and environment, exhibits early compromised mucociliary function regarding beating frequency and amplitude [51], lymphatic drainage and cough reflex, compounded by immunosuppression therapy, fostering interactions between multiple pathogens and airway epithelium [52]. Moreover, infections of the small airways typically manifest as peribronchiolar inflammation that can trigger or overlap AR [7, 53]. Consequently, patients encounter opportunistic pathogens, such as bacteria, viruses, fungi or mycobacteria, precipitating significant morbidity and mortality after LTx [54, 55]. Infections may originate from the donor or manifest as reactivated latent infections in the recipient, particularly prominent in the initial months after LTx. Although the literature is still scarce, airway epithelial cells, being the first cells in contact with pathogens and capable of enhancing or repressing immunity [2, 3], are rationally posited as potential risk or protective factors in the interaction with pathogens and the development of complications.

Viral infections after transplantation can be divided into pathogens affecting extrapulmonary tissues, such as herpes viruses or community-acquired respiratory viruses (CARVs). Herpes viruses, such as cytomegalovirus (CMV), may arise from latent infections in donors or recipients, necessitating stringent donor–recipient matching pre-LTx. CARVs, including respiratory syncytial virus, parainfluenza virus, influenza virus, rhinovirus, adenovirus and human metapneumovirus, are frequent culprits in immunocompromised hosts, including LTx recipients. However, in the case of multiple clinical events affecting the airways, triggered or exacerbated by viruses, this direct effect is plausible and requires further investigation. CARVs occur in LTx patients at reported incidences ranging from 2 to 21% [56, 57]. Thanks to Toll-like receptors [5], bronchial epithelial cells can detect these pathogens, eliciting lung graft damage through direct cytopathic effects, causing bronchiolitis, bronchitis or pneumonia [58, 59], or indirectly through inflammatory cytokine stimulation and T-cell activation, culminating in AR or CLAD. Several studies have established associations between CARVs and AR [52, 57, 60] as well as CLAD and BOS. BOS can be favoured by both CARVs (since AR is a risk factor for BOS) and viruses such as CMV. The rate of BOS after CARVs ranges from 32 to 50% depending on the type of virus [61, 62]. CMV infection confers a significant BOS risk, with an odds ratio exceeding three in some studies, engendering poorer transplant recipient outcomes [63].

Invasive fungal infections exhibit an 8–10% yearly cumulative incidence. As mentioned above for viruses, despite a low evidence level for the tracheobronchial tropism of these pathogens, the direct cytopathogenic effect of fungi on airway epithelium needs to be explored. Predominant pathogens are Aspergillus and Candida, although Pneumocystis, cryptococcus, Mucormycosis or non-Aspergillus mould (Scedosporium and Fusarium) and endemic fungi (Histoplasma capsulatum and Blastomyces dermatitis) may also be involved [54, 64, 65]. Airway epithelial cells can sense these fungi thanks to the C-type lectin receptor family, including dectin-1, dectin-2 and mincle [4, 66]. Candida colonisation frequently occurs, but fewer than 10% of patients develop invasive disease [67]. Conversely, Aspergillus infections are associated with high mortality rates [68]. These fungi, originating from the donor's lung or inhaled from the environment, exhibit major tropism for tracheobronchial infections, with anastomotic infections leading to anastomotic failure, parenchymal lung infections and mediastinitis. Moreover, artificial bronchial stenting favours Aspergillus infection. Aspergillus colonisation post-LTx, affecting 20–40% of patients, may precipitate BOS, influencing the long-term prognosis [69, 70].

Bacterial pneumonia predominates as the leading cause of respiratory tract infections, accounting for 32–63% of all infections, with peak incidences in the initial 4–8 weeks post-surgery due to hospitalisation and mechanical ventilation [64]. Pathogens encompass both community-acquired and nosocomial strains, with nosocomial infection associated with heightened mortality risk, particularly from more resistant pathogens, including Pseudomonas aeruginosa, Escherichia coli, Klebsiella pneumoniae, Acinetobacter baumannii, Staphylococcus aureus, especially methicillin-resistant S. aureus, and Stenotrophomonas maltophiliae. TLRs and NOD-like receptors of the airway epithelium recognise and interact with bacteria to activate innate and adaptative immunity [71]. Numerous studies have shown that bacteria, particularly the Gram-negative pathogen P. aeruginosa, significantly contribute to the development of BOS [72–75]. The mechanism probably occurs through damage to the bronchial epithelium, leading to the production of DAMPs and interleukin (IL)-1α, driving fibroblast activation [76–78] and promoting epithelial-to-mesenchymal transition (EMT) in the airway [76]

Mycobacterium tuberculosis, nontuberculosis mycobacteria and parasites such as Toxoplasma gondii are other causes of respiratory tract infections. Like other pathogens, they can affect the long-term outcome of LTx [64, 79], albeit without clear evidence of the direct association between airway epithelium, AR or CLAD and these infections.

AR

ARs represent primary immunologic complications and stand as the early principal cause of morbidity and mortality post-LTx. According to an International Society for Heart and Lung Transplantation report, 30% of adult patients experience at least one treated episode of AR within the year following LTx. Furthermore, 3.6% and 1.8% of deaths, respectively within the initial 30 days or between 30 days and 1 year, are attributable to AR [53, 80]. The frequency and severity of these immunologic events constitute the major risk factor for BOS development, culminating in chronic graft failure [53, 81, 82]. AR can manifest as either cell- or antibody-mediated and may occur within days but even years after LTx. Both ACR and AMR serve as risks factors for CLAD [83]. ACR, primarily mediated by T-lymphocytes recognising foreign HLAs or other antigens, is more prevalent than AMR [84, 85].

Diagnosis of ACR is confirmed through transbronchial biopsies, characterised by lymphocytic infiltrates in a perivascular (A grade) or peribronchiolar distribution (B grade). More recently, E grade, denoting lymphocytic inflammation in large airway (lymphocytic bronchitis (LB)) observed on bronchial biopsies has been identified as a significant CLAD risk factor [86, 87] with a twofold increased risk. Transcriptomic analysis of airway brushing samples from LB has unveiled upregulated genes common to nonlung allograft rejection [88]. Identified upregulated genes include indoleamine 2,3-dioxygenase 1 (IDO1) [89], HLA class I [90], Janus kinase 3 [91] and the C-X-C motif chemokine ligand (CXCL) 9–11 family, also known as monokines, induced by interferon (IFN)-γ [92]. Still using airway brushing, a pilot study identified 117 microRNAs that robustly segregated patients with ACR from those without. These results suggest that ACR is associated with a distinct epithelial microRNA signature that can provide insight into the pathogenesis of AR [93]. A proof-of-concept study has identified additional genes in bronchoalveolar lavage fluid (BALF) from transplant patients during ACR, such as thymocyte selection-associated high mobility group box protein, sterile alpha motif domain containing 3, IL-32 and killer cell lectin-like receptor K1, linked to T-cell activation [94].

AMR occurs due to the binding of pre-formed or de novo recipient antibodies targeting antigens expressed on the donor organ cells. Injury of the lung in AMR can occur by an antibody-dependent cell-mediated cytotoxicity mechanism, involving neutrophils, macrophages and NK cells. However, the precise role of airway inflammation and epithelium in AMR needs to be deciphered [56].

CLAD

CLAD encompasses a persistent decline in pulmonary function after LTx, comprising two recognised phenotypes, namely the predominant form, BOS, and the second phenotype, restrictive allograft syndrome (RAS).

BOS directly affects the epithelium of small airways. The underlying pathology of BOS is obliterative bronchiolitis, characterised by an accumulation of sub-epithelial fibrous tissue, extracellular matrix (ECM) deposition as a result of excessive fibroblastic response and luminal fibrosis resulting in smooth muscle remnants within obliterated bronchioles. CLAD BOS represent the principal cause of graft failure and mortality, with a 41% incidence at 5 years post-LTx and a current median survival term of 6 years after LTx [1, 95, 96]. The second phenotype, RAS, involves more lung parenchyma and progresses to end-stage pulmonary fibrosis with or without pleural engagement. Different histologic patterns can be found for RAS, such as pleuroparenchymal fibroelastosis, scattered in the lung parenchyma or even obliterative bronchiolitis [1, 94, 97, 98].

The understanding of CLAD's pathophysiology has evolved, but many specific mechanisms remain unclear. It is hypothesised that recurrent airway epithelium injuries resulting from both alloimmune and nonalloimmune triggers contribute to an abnormal healing of the epithelium, leading to the development of airway fibrosis and luminal obstruction. Nonalloimmune mechanisms targeting the airway epithelium may include pollution, gastro-oesophageal reflux disease or infections [99, 100]. Similar to AR, mounting evidence suggests that an airway epithelium gene expression profile associated with CLAD is involved. Genes upregulated during LB are further elevated during CLAD compared to controls, facilitating identification of patients at risk of graft failure [101]. Additionally, type-1 immune activation and IFN-dependent gene upregulation constitute part of CLAD's genetic signature, with the involvement of endogenous immune regulators such as IDO1 and tumour necrosis factor (TNF) receptor superfamily 6B [102]. Furthermore, other upregulated genes in BOS include IL-1α, IFN-γ, TNF-α, RANTES (regulated on activation, normal T-cell expressed and secreted) (C-C motif chemokine ligand 5), IL-8, B7-1 (CD80), B7-2 (CD86), IFN alpha and beta receptor subunit 2, CXCL9 (CXC chemokine receptor 3 family), HLA-B and beta-2 microglobulin [101, 103–106]. It is noteworthy that both type-2 regulatory cytokines (IL-5, IL-4 and IL-13) and IL-17 have been implicated in BOS and ACR, although a type-1 immune response appears to be predominant [102, 107, 108]. A dynamic interplay between airway epithelium and immunity is obvious, as airway epithelial cells under specific conditions can drive the differentiation of monocytes into macrophage-like cells rather than dendritic cells in bronchial epithelial cells cultured from stable lung allograft patients [109]. T-cells additionally promote the production of matrix metalloproteinases-9, a predictive biomarker for CLAD, by bronchial epithelial cells [110], illustrating this interaction.

Conversely, the airway epithelium may exert protective effects against CLAD onset. After LTx, heightened HLA-G expression in the bronchial epithelium correlates with stability in recipients compared to those with AR or BOS [111]. De novo donor-specific antibody in the recipient is inversely correlated with HLA-G positivity in lung grafts [49]. Furthermore, club cells may play a beneficial role after LTx. Club cells are secretory cells and can act as stem cells at a bronchiolar level, playing a central role in the small airway repair process and epithelium integrity [112, 113]. Club cell loss leads to the augmentation of an adaptative immune response coupled to BOS risk in a mouse model. In the same model, CD8+ T-cells inhibit club cell proliferation and their depletion restores the reparative function of club cells to prevent BOS [114]. This protective role against CLAD could be due to the production of club cell secretion protein (CCSP or CC16). A reduction of CCSP level correlates with an increased CLAD risk and reduced overall survival [115–117]. The potential involvement of airway epithelium in CLAD is summarised in figure 1.

FIGURE 1

Role of the airway epithelium in the development of chronic lung allograft dysfunction (CLAD). External triggers, such as pathogens, smoking, pollution or gastro-oesophageal reflux disease (GORD), can activate airway epithelial cells through pattern-recognition receptors that recognise pathogen-associated molecular patterns (PAMPs) or damage-associated molecular patterns (DAMPs). Depending on epigenetic factors, immunity may be enhanced through the upregulation of pro-inflammatory genes, particularly those involved in interferon (IFN) and T-helper cell (Th) 1 inflammation pathways. These immune responses can precipitate epithelial-to-mesenchymal transition (EMT), sub-epithelial fibrosis and the onset of CLAD, notably bronchiolitis obliterans syndrome (BOS). The pathogenesis may also involve Th2, Th17 or donor-specific antibody production, although to a lesser extent. Moreover, the airway epithelium may exert a protective function, dampening inflammation, as evidenced by the expression of molecules such as human leukocyte antigen (HLA)-G or club cell secretion protein (CCSP). AMR: antibody mechanism rejection; CARV: community-acquired respiratory virus; CCL5: C-C motif chemokine ligand 5; CLR: C-type lectin receptor; CMV: cytomegalovirus; CXCL9: C-X-C motif chemokine ligand 9; DSA: donor-specific antibody; IDO1: indoleamine 2,3-dioxygenase 1; IL: interleukin; P.: Pseudomonas; MMP-9: matrix metalloproteinase-9; NLR: NOD-like receptor; TLR: Toll-like receptor; TNF-α: tumour necrosis factor-α; TNFRS6B: tumour necrosis factor receptor superfamily 6B. Figure created with BioRender.com.

Several studies have indicated an association between EMT and BOS, demonstrated through various models, including bronchial epithelial cell culture, murine models and human post-transplant sampling [118–122]. EMT is a biological process wherein epithelial cells undergo phenotypic and functional changes, adopting characteristics typical of mesenchymal cells. This transformation involves the downregulation of epithelial markers, including cytokeratin and E-cadherin, and the upregulation of mesenchymal markers, such as vimentin and fibronectin. Moreover, cells gain migratory potential, secrete ECM proteins including collagen I and fibronectin, and produce matrix metalloproteinases [118, 119, 123]. EMT markers have been observed in patients without significant loss of function, yet they are also associated with the onset of BOS. Transforming growth factor-β1 (TGF-β1) plays a pivotal role in driving EMT and recent findings suggest that pro-inflammatory cytokines such as TNF-α and IL-1α can exacerbate TGF-β1-driven EMT. Lung macrophages may play a central role in this process by secreting these pro-inflammatory cytokines [119, 124]. Consequently, airway cell EMT could serve as the converging pathway linking alloimmune and nonalloimmune mechanisms in the development of BOS.

Advanced donor age increases the risk of allograft failure in various types of SOT [125]. Specifically, in LTx, survival rates in the first 3 years decrease when donors are over 60 years old [126, 127]. While cellular senescence is a physiological part of aging, it can also occur prematurely in certain diseases [128–131]. Telomeres, which are nucleoproteins located at the ends of chromosomes, safeguard against chromosomal shortening during cell replication. However, telomere length diminishes with age, potentially triggering airway epithelial cell senescence [33]. In LTx, a shortened donor telomere length correlates with worse CLAD-free survival, independent of donor age, whereas the recipient's telomere length was associated with an increased incidence of post-transplant leukopenia [128]. Moreover, decreased telomere length within the lung, particularly in club cells, may directly contribute to CLAD pathology [132]. Telomere dysfunction in airway epithelial cells leads to increased expression of senescence markers (β-galactosidase, p16, p53 and p21) and histopathological features consistent with CLAD, suggesting a role for telomere dysfunction in airway tissue remodelling and the development of different CLAD phenotypes [132, 133].

Airway epithelial cell culture: a model in LTx research

The airway epithelium has now been placed in a central position, orchestrating airway remodelling, scarring and epithelial–mesenchymal interaction, and thus is involved in numerous pathologies, such as lung fibrosis, cystic fibrosis, asthma and COPD [22, 23]. Literature on the classical biology of the airway epithelium following LTx is sparse but raises many important questions. After injury, studies in animals and in vitro models have shown that the remodelled allograft airway epithelium contains a mixture of donor and recipient epithelial cells [134–137]. This chimerism develops within days of LTx and can persist for years [135]. More recently, both donor and recipient tissue stem cells have been identified in anastomotic sites and bronchial airways of cystic fibrosis lung transplant recipients using air–liquid interface (ALI) models of cultured cells [138]. This chimerism may be involved in the repair of alloimmune-mediated tissue injury after transplantation; however, existing data do not clearly distinguish between protective and reparative chimerism.

As previously mentioned, in LTx, the airway epithelium is either a target or central to alloimmune and nonalloimmune mechanisms leading to epithelial activation, EMT and subsequently to the development of BOS and graft failure [139]. While primary fibroblast cultures have been established from LTx recipients [140], the first description of human primary bronchial epithelial cells (hPBECs) derived from lung allografts with bronchial brushes was made by Forrest et al. [141] in 2005. hPBEC development from both stable lung allografts and patients with BOS has allowed a better understanding of its pathogenesis and the ability to assess the effects of therapeutics such as immunosuppressants [142]. Although hPBECs from LTx recipients provide a model that is closer to reality, hPBECs from controls cultured under ALI conditions allow the obtention of pseudostratified epithelium with differentiated cells with a mucociliary phenotype. This model provides a polarised epithelium with apical and basolateral sides that can be analysed under physiological conditions or under activation with poly I/C or lipopolysaccharide, for example [143, 144].

Different techniques can be used to produce hPBECs and certain considerations for their use are important. First, the source of the cells can vary from brushes of large or small airways to endobronchial biopsies through bronchoscopy of LTx recipients [145]. Controls can also be obtained with airway brushing, bronchial biopsies, donor lungs before transplant (bronchial rings) or cadavers [146, 147]. Conversely, using brushes from the lungs from living patients not only allows the collection of a much higher numbers of samples, but also facilitates longitudinal analyses and enables a more rapid translation of in vitro observations to the clinical outcomes of the recipients. It is important to note that hPBECs from airway brushes in LTx recipients can be compromised by contamination of pathogens, insufficient cell yield or BOS [145]. Second, culture methods with medium and growth factors can lead to differences on cell composition, gene expression profile, cell signalling and epithelial morphology [148]. Third, it is important to work on differentiated epithelium in ALI conditions to better reproduce the real physiology of the lung. Moreover, methods to extend the lifespan of hPBECs have been developed, such as treatment with the Rho-associated protein kinase inhibitor Y-27632 [149] or exogenous induction of human telomerase reverse transcriptase [150]. However, it is unknown if the characteristics of hPBECs remain the same after immortalisation [151].

Even if hPBECs are an essential model to study LTx [119, 141, 152, 153], there are some limitations. The lungs fill up with air and deflate during breathing, creating stretching of the epithelium. To perform gas exchange and provide nutrients to the lung, endothelial blood vessels are connected to respiratory epithelial cells. Airways are normally colonised with different micro-organisms, referred to as the microbiome. Similarly, immune cells are present on the epithelium and can thus modify immunity triggered by the bronchial epithelium. Considering these limitations, other in vitro models are being developed. A lung-on-chip with hPBECs in contact with primary endothelial cells at the basolateral side has been used to create an alveolar–capillary system [154]. One method used to mimic air flow is to apply mechanical forces on hPBECs, recreating the movement of breathing [155]. Microbiome and immune cells can be mimicked with co-culture models with bacteria, viruses, dendritic cells, T-cells or macrophages [156–158]. However, these models are expensive and require specific expertise and equipment. Another way to perform bronchial epithelium culture is through stem cell based models, called organoid culture. Organoid culture is a novel three-dimensional culture first described in 2009 using organoids from the gut [159]. Airway organoids (AOs) were developed in 2012 [160, 161]. AOs can be generated from different stem cells and can overcome the problems associated with the limited lifespan of hPBECs. Additionally, AOs can be derived from nearly any part of the respiratory tract and genetic modifications can be performed in a similar way to those done with hPBECs [162]. In a differentiated spheroid AO, basal cells are on the outside of the sphere, whereas goblet cells excrete their mucus into the lumen. AOs represent a promising model with which to study the bronchial epithelium in LTx and other lung pathologies [151]. However, they are also influenced by the culture medium and do not interact with endothelial cells. In summary, airway epithelial cell culture using AOs, hPBECs or more complex models appear to complement each other, enhancing our understanding of LTx outcomes.

Targeting airway epithelial cells as a novel diagnostic and therapeutic approach

Airway epithelial cells play a role in the current diagnosis methods for complications and have the potential to serve as valuable diagnostic tools in the future. For example, AR including grade B and E diagnosis relies on transbronchial biopsies. Moreover, BALF can be a prognostic determinant, since neutrophils in BALF have been identified as a reversible risk factor for CLAD [163] and are included in the diagnosis criteria for azithromycin-responsive allograft dysfunction [164]. Lymphocytic airway inflammation and LB are now described as risk factors for CLAD. LB can be challenging to detect using standard-of-care histopathologic analysis on transbronchial biopsies, whereas airways brushes seem more relevant. Transcriptional analysis of airway brushes may allow more reliable detection of airway inflammation [22, 101, 102, 165]. Thus, airways brushes and transcriptional analyses show promise for the better understanding, earlier diagnosis and treatment of long-term complications such as CLAD.

In SOT and LTx, there is growing evidence concerning the genetic signatures of AR and CLAD, allowing a better understanding of immunologic complication mechanisms and making novel therapeutic approaches possible. JAK-1 inhibitor (itacitinib) has shown promise as an inhibitor of lymphocytic mucosal inflammation in the context of early CLAD [166]. Various techniques aim to reduce airway inflammation, such as the use of an adenosine A2A receptor antagonist, which has been studied in PGD to reduce NK cell-mediated inflammation (phase I clinical trial) [167] and even in a pre-clinical study before LTx with ex vivo lung perfusion [168]. Furthermore, novel therapeutic approaches directly targeting the bronchial epithelium are emerging. Host-directed therapy has the potential to enhance the innate immune response of epithelial cells and macrophages, improve epithelial barrier integrity, stimulate autophagy, and even restore mucociliary clearance and ion balance. This approach could be particularly beneficial for immunodeficient recipients with impaired mucosal immunity and offers a novel strategy to combat pathogens following LTx [3]. Host-directed therapies include vitamin D stimulated genes, aroylated phenylanediamines, bacterial lysates such as Broncho-Vaxom, host defence peptides and microbiota metabolites that can boost innate immune responses [169–172]. Another promising future therapeutic approach targeting airway epithelial cells could involve addressing club cell dysfunction, as it is a predictor of BOS onset and graft failure [116]. Understanding the crucial involvement of biological processes such as airway cell EMT and senescence in BOS pathogenesis also opens up avenues for novel therapeutic strategies. Firstly, targeting key molecules in the EMT pathway, such as TGF-β-activated kinase 1, which is activated by both TGF-β1 and TNF-α, presents a promising approach to mitigate inflammation-driven EMT and potentially halt the progression of BOS. This underscores the significance of EMT as a therapeutic target in combating BOS post LTx [68]. Secondly, the interplay between methylation aging and telomere shortening in the airways could serve as a tool for monitoring allograft health and identifying recipients at risk of adverse outcomes after LTx, whether early or late.