To our knowledge, this is the first study on TOLLIP expression in lung samples from patients with CTD-ILD. We studied the association between TOLLIP and clinical characteristics and indexes, and found that the expression ratio of TOLLIP was negatively correlated with changes in G-score. We further analysed the rs3750920 genotype of TOLLIP, and found that the frequency of the minor allele T was lower than the previously reported frequency.
In the present study, the expression of TOLLIP significantly increased in fibrotic areas, in the cytoplasm of both alveolar epithelial cells and bronchiolar epithelial cells. Based on previous reports, TOLLIP widely expressed in the human respiratory tract, including macrophages, primary human nasal, bronchial, and type II alveolar epithelial cells, where diffuse punctate staining was observed in the cytoplasm and cell membrane .
Although TOLLIP expression and localization in the lung specimens of patients with CTD-ILD has not been previously reported, the location of TOLLIP protein in IPF has been described in a previous study . This study revealed that TOLLIP abundantly expressed in “aberrant basaloid cells” in the lungs of patients with IPF, whereas the total expression of TOLLIP was lower in patients with IPF than in healthy controls. Co-staining of TOLLIP and E-cadherin confirmed that the positive cells were epithelial cells.
TOLLIP expression in lung tissue of patients with CTD-ILD could be different from patients with IPF. Previous studies have demonstrated that there are some differences in the pathogenesis of pulmonary fibrosis between IPF and CTD-ILD [21–23]. For instance, endothelial cells are primarily involved in SSc-ILD, whereas the injury of alveolar epithelial cells is mainly observed in IPF . Since autoimmunity is an essential feature of CTD-ILD, immune-mediated inflammation of lung tissue is common, but the mechanisms by which the lung tissue becomes inflamed remain unclear for IPF. Therefore, the assessment of TOLLIP expression in lung tissue of patients with CTD-ILD is necessary.
We hypothesize that the increased expression of TOLLIP observed in fibrotic areas might be associated with its role in pulmonary fibrosis. Previous studies have indicated that TOLLIP participates in the inflammatory response, autophagy, and nuclear interactions [6–8, 24–25]. As an endogenous inhibitor of Toll-like receptor signalling, TOLLIP can prevent pro-inflammatory signalling by binding to IL-1 receptor-associated kinase (IRAK-1) . The negative role of TOLLIP in pathogen-associated inflammation is essential for immune tolerance. Dysfunction of TOLLIP leads to overactivation of the canonical MyD88-dependent TLR2 and TLR4 signalling pathway [26–27]. The pro-fibrotic effect of TLR2 and TLR4 activation is supported by several previous studies [28–29]. The activation of TLR4 enhances the process of fibrosis in the lung by transforming danger signals into myofibroblasts, and increasing the secretion of ECM molecules . Another crucial role of TOLLIP is the promotion of autophagy, which it facilitates by bridging ubiquitinated protein aggregates with LC3 and inhibiting the phosphorylation of vacuolar protein sorting 34 (VPS34), an important protein involved in autophagy . Impaired autophagy and excessive ROS generation contribute to IPF progression. Anti-fibrotic agents can attenuate fibrosis in bleomycin-exposed mice by improving autophagy . Although the exact molecular mechanisms remain unclear, mounting evidence suggests that TOLLIP is a protective regulatory protein in pulmonary fibrosis. Therefore, the observed TOLLIP overexpression in the lung specimens of the present cohort might be a compensatory response to pulmonary fibrosis associated with CTD.
We observed a significant correlation between the change in G-score and TOLLIP expression in both alveolar epithelial cells and bronchiolar epithelial cells, indicating that patients with higher expression of TOLLIP had better improvement of ground glass opacity signs during follow-up by CT. This correlation was not identified with the change in F-score which evaluates fibrosis and TOLLIP expression. It is possible that TOLLIP could protect lung tissue from fibrosis in the early stages of the disease. In the present study, there was no correlation between TOLLIP expression and histological patterns of ILD or CTD diagnosis, which implies that increased TOLLIP expression is closely related to pulmonary fibrosis, regardless of the cause.
TOLLIP is associated with some clinical characteristics, the treatment response, and prognosis of ILD. A genome-wide association study (GWAS) correlated additional common variants with susceptibility and mortality of IPF in three stages. Three TOLLIP SNPs (rs111521887, rs5743894, and rs5743890) were identified in this GWAS, of which the minor alleles rs111521887_G and rs5743894_G presented a risk for IPF . The minor allele rs5743890_G was protective for susceptibility to IPF, but was also a risk factor for mortality, whereas the major allele rs5743890_A affected the progression of IPF. Interestingly, this study demonstrated an association between TOLLIP SNPs and TOLLIP expression. Compared with homozygous carriers of common alleles, TOLLIP expression decreased in the lung tissue of minor allele carriers, by 20% in patients with rs5743890_G, by 40% in those with rs111521887_G, and by 50% in those with rs5743894_G. The decreased TOLLIP expression observed in minor allele carriers indicates that TOLLIP deficiency might contribute to the pathogenesis of IPF.
In addition to GWAS, some clinical retrospective studies have also demonstrated the function of TOLLIP in IPF. A retrospective study demonstrated that the minor allele (C) in TOLLIP rs5743890 was associated with worse survival and disease progression in patients with IPF . In a post-hoc exploratory analysis of patients enrolled in the clinical trial “Evaluating the Effectiveness of Prednisone, Azathioprine, and N-Acetylcysteine in Patients with Idiopathic Pulmonary Fibrosis”, a significant correlation was observed between the response to N-acetylcysteine (NAC) therapy and rs3750920 of TOLLIP . These results were replicated in an independent cohort of patients drawn from the University of Chicago (UChicago) and “The INSPIRE Trial: A Study of Interferon Gamma-1b for Idiopathic Pulmonary Fibrosis”, which included 405 individuals from the two cohorts, and a significant interaction between NAC therapy and rs3750920 was confirmed. In another retrospective study of patients with IPF and IPAF, rs3750920 T/T showed an increased frequency in the positive ANA group, in which NAC exposure seemed to be beneficial for transplant-free survival . In a Japanese population study, the frequency of the minor T allele of rs3750920 was low in patients with fibrotic ILD, especially in non-IPF patients, which might be an indicator of better survival in these patients . In an inception cohort study on early rheumatoid arthritis (RA), TOLLIP rs111521887 was associated with pulmonary fibrosis in RA .
In the present study, we analysed the association between the rs3750920 genotype distribution and clinical characteristics. The frequencies of C/C and C/T were 43% and 57%, respectively, whereas T/T was not identified. MAF was 0.28, which was obviously lower than that from 1000 Genomes data. A Japanese study on rs3750920 genotype of patients with fibrotic ILD showed C/C, C/T, and T/T frequencies of rs3750920 were 63%, 30%, and 7%, respectively. We hypothesize that the minor T allele of rs3750920 may present a lower frequency in Asian patients with fibrotic ILD. No association between rs3750920 and CTD-ILD progression was observed in the present study.
The critical function of TOLLIP has been evaluated in several diseases, including neurodegenerative diseases, pulmonary diseases, cardiovascular diseases, gastrointestinal diseases, and the pathogen immune system response [9–13, 37–38]. The role of TOLLIP in CTD remains unclear. However, the genotypes of rs3750920 might be associated with the presence of ANA in patients with fibrotic ILD . In in vivo animal model research, SSc individuals presented high TLR4 expression in skin fibrosis, and might show an optimal therapeutic response to selective inhibitors of TLR4 . Based on previous studies and the functional importance of TOLLIP, we postulate that it may also participate in the pathogenesis of rheumatoid diseases, which warrants ongoing investigation.
The present study has several limitations. First, the small sample size limits the association analysis and the generalizability of the results. Second, the disease severity of the recruited participants was mild, so they were able to withstand TBLC. This could explain why all patients survived during the follow-up period. Finally, although we observed increased TOLLIP expression in patients with CTD-ILD and associated this with clinical characteristics, further in vivo and in vitro studies are required to explore the mechanisms by which TOLLIP participates in CTD-ILD.