Updates on T helper type 17 immunity in respiratory disease

Summary

Interleukin‐17 (IL‐17)‐producing cells play a critical role in mucosal immunity including the respiratory tract. This review will highlight recent advances in our understanding of these cells in mucosal immunity in the lung as well as their potential pathogenic roles in respiratory diseases. The IL‐17‐producing cells include γδ T cells, natural killer cells, group 3 innate lymphoid cells, and T helper type 17 (Th17) cells. There have been recent advances in our understanding of these cell populations in the lung as well as emerging data on how these cells are regulated in the lung. Moreover, Th17 cells may be a key component of tissue‐resident memory cells that may be acquired over time or elicited by mucosal immunization that provides the host with enhanced immunity against certain pathogens.

Introduction

Interleukin‐17 (IL‐17) ‐producing cells play critical roles in mucosal immunity in the oropharynx, gut, skin and respiratory tract. The principal sources of IL‐17A in the lung include natural killer (NK) cells, γδ T cells, αβ T cells as well as group 3 innate lymphoid cells (ILCs). Many of these cells co‐express IL‐17F, which is in the same locus as IL‐17A in mice and humans.1 These cytokines signal to a receptor complex consisting of IL‐17RA, which is ubiquitously expressed, and IL‐17RC, which is expressed in lung epithelial cells2 and lung fibroblasts. In the primary immune response to pathogens such as Gram‐negative bacteria, IL‐17 is produced by NK cells and γδ T cells, as well as by a small population of group 3 ILCs.3, 4 However, after mucosal immunization, a pool of αβ ⁺ T helper type 17 (Th17) cells can be elicited.5 These latter cells may be important components of tissue‐resident memory cells, and several groups of investigators have shown that these cells can elicit serotype‐independent immunity to Gram‐positive bacteria (Streptococcus pneumoniae (S. pneumoniae)),6 Gram‐negative bacteria (Klebsiella pneumoniae (K. pneumoniae))5 and fungi.7, 8 Moreover, IL‐17 has been shown to be important in early host defense against intracellular pathogens such as Mycobacterium tuberculosis.9 Although IL‐17RA has been shown to be critical for opportunistic infections such as oropharyngeal candidiasis,10, 11 it is dispensable for the pulmonary opportunistic fungus, Pneumocystis pneumoniae.12 This review will highlight recent advances on how IL‐17⁺ cells signal in the lung as well as their role in host defense and lung diseases – highlighting recent advances in chronic obstructive pulmonary disease (COPD), asthma and cystic fibrosis (CF). Lastly, this review will highlight key unanswered questions in the field.

Immune cell sources of IL‐17 in lungs

Th17 cells have been widely recognized as a key source of IL‐17 in mucosal immunity and autoimmunity. Th17 cells develop under the control of transforming growth factor‐β and IL‐6 in mice through the transcription factor retinoic acid receptor‐related orphan nuclear receptor γt (RORγt),13 as well as signal transducer and activator of transcription 3 (STAT3). However, in addition to αβ TCR⁺ Th17 cells, additional cells can produce IL‐17. Recent studies have shown that γδ T cells, as well as group 3 ILCs (ILC3s), CD3⁺ invariant natural killer T cells and NK cells, can produce substantial IL‐17 in response to IL‐1β and IL‐23 without T‐cell receptor (TCR) stimulation.14 γδ T cells producing IL‐17 can mediate neutrophil recruitment into the lung at critical early stages of lung infection.15 Recently, γδ T‐cell‐derived IL‐17A has been shown to be required for host defense against neonatal influenza infection through IL‐33 up‐regulation,16 contrary to the previous reports suggesting the detrimental role of IL‐17A of leading to acute lung injury in adult mice.17 These characteristics of γδ T cells to trigger rapid responses to invading pathogens are intrinsic, namely determined during the thymic development. Specific subsets of γδ T cells acquired the potency of producing interferon‐γ (IFN‐γ) through suppressing transcription factors such as Sox13 and RORγt after activation of TCR signaling in the thymus.18 On the other hand, other subsets of γδ T cells are preprogrammed to secrete IL‐17 with no or weak TCR signaling.18 Additionally, IFN‐γ‐producing γδ T cells, as well as NKT cells, require CD27 co‐stimulation for cytokine production.19 These unique thymic processes appear to sub‐divide these cells into γδ T cells that are IL‐17‐ or IFN‐γ‐producing cells when entering the periphery. On the other hand, apart from the γδ T cells with an inborn ability to produce IL‐17, the γδ T cells resident in the secondary lymphoid organs need to develop after birth and require TCR activation to secrete IL‐17.20 In addition to γδ T cells, early inflammatory responses to pathogens can involve ILC3 cells, an identified subgroup of ILCs. Depletion of ILC3s rendered mice highly susceptible to oropharyngeal candidiasis, which is also dependent on IL‐17 and IL‐17RA.10, 21, 22 Invariant NKT cells, which are already committed to making IL‐17 in the thymus, have been shown to respond to glycolipids rapidly.23 In models of bacterial pneumonia, the respective contribution of these different IL‐17‐producing cell types remains to be determined.

Lessons from human IL‐17 deficiency syndromes

Several mutations in IL‐17, IL‐17RA or transcription factors that control Th17 development such as STAT3/STAT1 have been described that affect type 17 immunity in humans. A consistent phenotype of these deficiencies is chronic mucocutaneous candidiasis (CMC).11 This phenotype has been replicated in Il17ra ^−/− mice24 as well as conditional deletion of Il17ra ^−/− in oropharyngeal epithelium.10 In the murine studies, IL‐17RA signaling was required for the expression of mouse β‐defensin 3, an antimicrobial protein with fungicidal activity.10 Lévy et al.25 have recently reviewed 21 cases of human autosomal recessive IL‐17RA deficiency and a core phenotype is CMC and cutaneous Staphylococcus aureus (S. aureus) infection. Eight of the 21 cases also had sinusitis, bronchitis, or lobar pneumonia. Patients with mutations in ORC – a key transcription factor regulating IL‐17 production – also have increased susceptibility to M. tuberculosis infection.26 Another key transcription factor regulating Th17 development is STAT3.27, 28, 29 Patients with autosomal dominant STAT3 mutations or Hyper‐IgE syndrome share some clinical features with autosomal recessive IL‐17RA deficiency including CMC and cutaneous S. aureus infections. These patients can also develop pulmonary infections with S. aureus and S. pneumoniae, which can result in lung structural damage and pneumatocele formation. Interestingly, this is less common in patients with another form of Hyper‐IgE syndrome due to mutations in DOCK8,30 which also affects Th17 cell development. These data suggest that the pneumatocele phenotype may be due STAT3's role in mediating lung repair31 apart from its role in Th17 cell development. Pneumatoceles can be colonized and infected with fungi such as Aspergillus fumigatus (A. fumigatus), which can cause significant morbidity in these subjects.32 The cell‐intrinsic roles of these genes are important to understand as this may aid in determining the role of bone marrow transplantation for some of these disorders. Modeling S. aureus pneumonia in mice has shown that innate and adaptive lymphoid cells are dispensable to clear primary infection, but epithelial STAT3 is required.33

Interleukin‐17 and CF lung disease

Interleukin‐17 can be produced by a variety of cells in the lung including NK cells, ILC3s, γδ T cells and αβ T cells. Patients with CF undergoing lung transplantation who have chronic microbial infection with both bacteria and fungi, have robust Th17 responses to both microbial (Pseudomonas aeruginosa (P. aeruginosa)) and fungal (A. fumigatus) antigens in lung‐draining lymph nodes as well as in lung parenchymal cells.34 These data are from patients with end‐stage lung disease, but IL‐17⁺ cells have also been reported in younger subjects with CF, in addition to non‐CF bronchiectasis.24 In this immunohistochemistry study, the authors found Th17 cells in addition to IL‐17⁺ NK T cells, γδ T cells and neutrophils.35 Interleukin‐23 can be an important upstream regulator of these IL‐17 responses, and IL‐23‐deficient mice have reduced airway inflammation in response to acute or chronic P. aeruginosa infection.36, 37 Neutralization of IL‐17 in a murine model of CF lung inflammation, induced by the mucoid clinical P. aeruginosa isolate M5715, was shown to reduce lung inflammation and reduce the amount of recoverable bacteria in lung lavage fluid.38 However, using a similar agar‐based model of chronic P. aeruginosa infection with two clinical CF isolates (YH5 and NH57388A), Il17ra ^−/− mice (with a wild‐type Cftr allele) showed a higher rate of infection as well as higher mortality after infection.39 These data suggest that IL‐17RA is required for host defenses against these clinical isolates, but it remains unclear which IL‐17RA ligands are required for this effect. The murine studies are also difficult to interpret given that Cftr‐deficient mice do not develop spontaneous lung disease, in contrast to larger animal models such as the CF pig40, 41 or ferret42; hence, these larger animal models may be required to better understand the role of IL‐17 in CF lung inflammation. Consistent with the notion that Th17 cells contribute to disease, it has been reported that high circulating Th17 responses are strongly associated with poorer lung function.43 Conditional deletion of IL‐17RA or IL‐17RC in the conducting airway of mice attenuates IL‐17‐mediated neutrophil recruitment into the lung, demonstrating that airway epithelium is a key target of IL‐17RA/RC signaling.44 Additionally, in human bronchial epithelial cells, IL‐17 can augment HCO₃ ^– transport,45 which is defective in CF cells and the CF pig.46, 47 These data may explain why IL‐17 immunity is not sufficient to clear infection in the CF airway. Additionally, HLA class II alleles have been associated with the frequency48 and age49 of acquisition of P. aeruginosa as well as lung function in patients with CF due to homozygous F508del mutations.50 These data strongly suggest that class II‐restricted immunity plays a key role in these phenotypes, but the underlying immune basis of these observations remains to be determined.

Interleukin‐17 and COPD

There is emerging data that IL‐17 may play a key role in the pathogenesis of COPD. Cigarette smoke contains ligands for the aryl hydrocarbon receptor (Ahr) that can skew T‐cell responses to Th17 responses.51, 52 Using gain‐of‐function and loss‐of‐function approaches has also implicated IL‐17 and IL‐17RA signaling in pathogenesis of emphysema. Mice overexpressing IL‐17 have more severe emphysema to cigarette smoke exposure53, and Il17ra ^−/− mice are protected against cigarette‐smoke‐induced emphysema.51 Both IL‐17A and the number of IL‐17⁺ CD4⁺ T cells have studies in COPD and this has been recently summarized in a review by Le Rouzic et al.54 The antigen specificity of these cells remains to be determined because some of these cells could be specific to the abnormal microbiome in the COPD lung55 or against host‐derived proteins such as elastin fragments,56 in which some investigators have considered COPD as an autoimmune disease. Further evidence of the involvement of IL‐17 is supported by data in this pathway regulating the formation of ectopic B‐cell follicles or inducible bronchial‐associated lymphoid tissue (iBALT),57, 58 which is a pathological feature in COPD.59 The amount of iBALT is associated with more severe forms of emphysema,59 and both lipopolysaccharide‐ and fungus‐induced iBALT formation requires IL‐17,57, 58 which can regulate the production of CXCL13 in lung fibroblasts, a chemokine necessary for follicle formation. However, COPD is heterogeneous disease with some patients having greater degrees of emphysema and others with a chronic bronchitis phenotype. How IL‐17 plays a role in these disease endotypes remains to be defined. Recently, lung epithelial transcriptomics has been used to phenotype patients with COPD60 and it appears that a subgroup of these patients have IL‐17‐regulated genes expressed in these samples, such as SAA4 and LCN2.60 Consistent with iBALT, there is also a recent report on the presence of B‐cell signature in COPD.61

Both CF and COPD can be characterized as a disease of exacerbations and remission, where exacerbations can be associated with further declines in lung function. Recent studies suggest that IL‐17 may play a role in viral and bacterial exacerbations.62 One agent implicated in COPD exacerbation is respiratory viruses, including respiratory syncytial virus. In testing of chemical‐induced emphysema with elastase, Mebratu et al.63 found that IL‐17 mediated virus‐induced neutrophilic inflammation as well as exacerbating airspace enlargement, suggesting that IL‐17 pathway may also be involved in viral‐mediated COPD exacerbations. Data supporting an interventional trial and the risk and benefits of targeting IL‐17 have been recently reviewed.54 A review of clinicaltrials.gov lists several observational trials to study T‐cell responses and IL‐17 and IL‐22 as biomarkers in COPD (NCT00281229 and NCT02655302). This type of phenotyping may be helpful in patient selection criteria for future interventional trials. There is some concern, given the role of IL‐17 in host defense, that targeting this pathway may exacerbate lung disease. A recent small interventional trial assessing the effects of Simvastatin in COPD was published. The investigators found an effect of Simvastatin on sputum IL‐17 levels (as well as IL‐6 and CXCL8), and there was also reduction in sputum macrophages, but not neutrophils. Whether this drug will continue to be studied in COPD is unclear. However, several anti‐IL‐17 drugs have been approved for psoriasis treatment, and infectious complications in this population were rarely severe and were manageable.64 Similar to CF, it is unclear why the IL‐17 pathway is not able to affect bacterial clearance in the lung. Part of this defect may be the effects of CS on airway ion transport,65 as well as effects that can result in metaplastic changes to the airway epithelium. Hence, if ion transport is defective, IL‐17 immunity may not be fully effective. Interestingly, conditional deletion of IL‐17R in the gut epithelium and lung epithelium results in reduced expression of the polymeric immunoglobulin receptor and sIgA is markedly reduced in the gut lumen.44, 66 However, IgA transport by polymeric immunoglobulin receptor requires adequate epithelial polarization, and so even regional areas of squamous metaplasia may set up an area of abnormal mucosal immunity and serve as the nidus for abnormal microbial communities in the lung. Understanding these processes at the epithelium may also be useful in identifying patient subgroups that may benefit from IL‐17 intervention.

Interleukin‐17 and asthma

Similar to COPD, there are many studies showing IL‐17 being up‐regulated in cohorts to asthma and asthma exacerbation.67, 68, 69 A recent study using airway transcriptomics similar to the approach of defining Th2 asthma70 showed that a subgroup of patients had a Th17 signature in the airway.69 This subgroup was characterized as having higher gene expression in the bronchial brushes of IL‐17‐regulated genes such as CSF3, CXCL1, ‐2, ‐3, and ‐8. This was in contrast to Th2 genes POSTN, CLCA1, and SERPINB2, which have been previously reported in patients with Th2‐high asthma.70 Both the Th1 and Th17 cohorts had eosinophilic inflammation, but the Th17 signature was associated with steroid‐dependent, moderate to severe asthma.69 Using a murine model, the authors saw some evidence for IL‐13 regulating the type 17 response in vivo and that the most efficacious intervention was blocking both IL‐13 and IL‐17.69 Due to the success of several drugs that target Th2 asthma, such as IL‐5, IL‐4Ra, and thymic stromal lymphopoietin, there may be an unmet group with non‐Th2 asthma. It remains unclear what proportion of these subjects would benefit from Th17 or dual Th2/Th17 intervention. This can only be addressed with clinical trials that define Th2 non‐responders and investigating new agents that may target Th1,71, 72 Th17 or other forms of non‐Th2 inflammation in the lung. A challenge of these trials will be phenotyping and endotyping of subjects. A current limitation of the gene signatures is that they are invasive and the field does not have information on the stability of these signatures. Hence, another need is to try to develop non‐invasive correlates that may aid in patient selection criteria.

Can we elicit Th17 cells to enhance mucosal immunity in the lung?

The genes IL17A and IL17F co‐evolved with genes that regulate TCR recombination.5 This strongly suggests evolutionary pressure of these genes to be expressed in memory cells. Although less is known about CD4⁺ T‐cell memory than CD8⁺ T‐cell memory, memory CD4⁺ IL‐17⁺ cells can be found in the lamina propria of the small intestine73 and human lung.74 Moreover, the cells in the lamina propria of the small intestine can be specific to the microbial flora – for example, segmented filamentous bacteria in mice.73 Elicitation of lung memory Th17 cells to (S. pneumoniae) 75 or K. pneumoniae 5 can provide T cells that can provide serotype‐independent immunity to a pulmonary challenge with those pathogens.5, 6, 75 This approach has also shown pre‐clinical success with Haemophilus influenzae, an important pathogen in COPD.76 Based on this work, there is the possibility to take advantage of this immunology to prevent invasive bacterial and fungal infection in the lung. The key questions are what adjuvants drive these responses. Can these responses be elicited by systemic immunization or will mucosal immunization be required? A recent study with lobar instillation of antigen suggests that these tissue‐resident memory responses are anatomically compartmentalized,75 and hence, generating these responses throughout the lung tissue may require systemic or inhaled approaches. If the latter, what would an inhaled vaccine look like and what regulatory issues would need to be addressed to assure that these approaches are safe. We have a breadth of human toxicology data with inhaled lipopolysaccharide that low doses are tolerated well,77 and hence the concept of inhaled vaccines may be achievable in a safe and effective manner. Another challenge to the field is actually measuring these types of immune responses. Most approved vaccines work by eliciting systemic antibody that is measurable in serum. Those assays will probably not reflect the successful elicitation of lung‐resident memory cell, so if we are to develop robust vaccines that work through the generation of lung TRM cells, we will need assays to measure their elicitation. Another point to consider is which individuals may be at risk by such an approach. Lessons from CF, asthma, and COPD suggest that individuals with epithelial defects due to genetic loss of ion transport or injury due to viral infection or smoking may have adverse effects from such an approach. Larger animal models that mimic human lungs such as the pig, ferret and primates may be extremely useful in trying to move this concept to the clinic.

Disclosures

The authors have no competing interests in relation to this work.