Abstract
Background Idiopathic pulmonary fibrosis (IPF) is a devastating disease characterised by myofibroblast proliferation and abnormal extracellular matrix accumulation in the lungs. Transforming growth factor (TGF)-β1 initiates key profibrotic signalling involving the SMAD pathway and the small heat shock protein B5 (HSPB5). Tripartite motif-containing 33 (TRIM33) has been reported to negatively regulate TGF-β/SMAD signalling, but its role in fibrogenesis remains unknown. The objective of this study was to elucidate the role of TRIM33 in IPF.
Methods TRIM33 expression was assessed in the lungs of IPF patients and rodent fibrosis models. Bone marrow-derived macrophages (BMDM), primary lung fibroblasts and 3D lung tissue slices were isolated from Trim33-floxed mice and cultured with TGF-β1 or bleomycin (BLM). Trim33 expression was then suppressed by adenovirus Cre recombinase (AdCre). Pulmonary fibrosis was evaluated in haematopoietic-specific Trim33 knockout mice and in Trim33-floxed mice that received AdCre and BLM intratracheally.
Results TRIM33 was overexpressed in alveolar macrophages and fibroblasts in IPF patients and rodent fibrotic lungs. Trim33 inhibition in BMDM increased TGF-β1 secretion upon BLM treatment. Haematopoietic-specific Trim33 knockout sensitised mice to BLM-induced fibrosis. In primary lung fibroblasts and 3D lung tissue slices, Trim33 deficiency increased expression of genes downstream of TGF-β1. In mice, AdCre-Trim33 inhibition worsened BLM-induced fibrosis. In vitro, HSPB5 was able to bind directly to TRIM33, thereby diminishing its protein level and TRIM33/SMAD4 interaction.
Conclusion Our results demonstrate a key role of TRIM33 as a negative regulator of lung fibrosis. Since TRIM33 directly associates with HSPB5, which impairs its activity, inhibitors of TRIM33/HSPB5 interaction may be of interest in the treatment of IPF.
Abstract
TRIM33 has a protective role against fibrogenesis, inhibiting the TGF-β1 pathway through a direct association with HSPB5. Interactions between TRIM33, SMAD4 and HSPB5 may represent key targets to prevent the progression of pulmonary fibrosis. http://bit.ly/3aVCuxc
Introduction
Idiopathic pulmonary fibrosis (IPF) is a rare, chronic and progressive disease of the lung parenchyma of unknown origin, with a median survival of 3–5 years [1]. With the exception of pirfenidone and nintedanib, which exhibit moderate efficacy on disease progression, no pharmacological treatment is currently available [2]. IPF development associates abnormal alveolar repair with proliferation of activated fibroblasts named myofibroblasts, which leads to abnormal extracellular matrix (ECM) accumulation and tissue remodelling. Transforming growth factor (TGF)-β1 is a key profibrotic growth factor, and the adenoviral vector-mediated gene transfer of active TGF-β1 (by instilling/injecting the recombinant adenovirus, AdTGF-β1) in rodent lungs leads to progressive and severe fibrosis [3]. TGF-β1 is responsible for fibroblast activation into myofibroblasts. Mechanistically, TGF-β1 classically signals via the SMAD proteins, a crucial pathway in fibrogenesis [4].
We previously demonstrated that the small heat shock protein B5 (HSPB5; also termed heat shock protein αB-crystallin) is overexpressed in human IPF lungs. HSPB5 interferes with SMAD4 localisation by inhibiting its export from the nucleus. At the molecular level, HSPB5 hampers SMAD4 ubiquitination and, consequently, SMAD4 accumulates in the nucleus and promotes TGF-β1 signalling and fibrogenesis [5]. Tripartite motif-containing 33 (TRIM33) is an E3 ubiquitin ligase responsible for SMAD4 ubiquitination, inducing its export from the nucleus and therefore inhibiting SMAD transcriptional activity [6]. TRIM33 is involved in several processes, such as embryogenesis, haematopoiesis and tumorigenesis [6–8]. Specifically, TRIM33 is involved in the turnover of TGF-β receptor I [8] and in the innate immune response by regulating a subset of genes, including TGF-β1 or interferon-β in macrophages [9, 10]. Anti-TRIM33 antibodies are elevated in humans with myositis, a set of diseases with increased prevalence of cancer and interstitial lung disease (ILD) [11]. However, the impact of TRIM33 in lung fibrosis and IPF remains unstudied.
In the current study, we demonstrate for the first time that TRIM33 is overexpressed in the lungs of IPF patients and has a protective effect against pulmonary fibrosis in vivo. In addition, we highlight a mechanism by which HSPB5 interferes with the antifibrotic properties of TRIM33, paving the way for the development of novel therapeutic strategies targeting the interaction between HSPB5 and TRIM33.
Materials and methods
Human tissue samples
Lung tissue samples (n=5) were obtained by open lung biopsy (INSERM U700, Paris, France). IPF was diagnosed according to the American Thoracic Society/European Respiratory Society consensus criteria [12], including clinical, radiographic and characteristic histopathological features. Control non-IPF lung tissue samples were obtained from smokers who underwent thoracic surgery for localised primary lung carcinoma (n=5). The local ethics committee (Comité de Protection des Personnes, Ile de France 1) approved the study, and patients provided informed consent before lung surgery.
Human microarray data
GSE110147 was performed and analysed by Cecchini et al. [13] and GSE49072 was performed and analysed by Shi et al. [14].
Animal procedures
SV129 wildtype mice (Charles River, Saint Germain-sur-l'Arbresle, France) and SV129 knockout mice for the HSPB5 gene were housed in pathogen-free conditions. We generated mice selectively deficient for TRIM33 by breeding floxed Trim33 mice with cFes-Cre transgenic mice. Trim33f/f-CrecFes (hsTrim33−/−) mice thus corresponded to haematopoietic tissue-restricted knockout mice, as previously described [15]. Mouse food and water were provided ad libitum. The animals were treated according to the guidelines of the Ministère de l'Enseignement Supérieur et de la Recherche (Ministry of Higher Education and Research, Paris, France). All experiments were approved by the Comité d'Ethique de l'Expérimentation Animale du grand campus Dijon (Animal Experimentation Ethics Committee of the large Dijon campus, Bourgogne, France). Intratracheal instillation of bleomycin (BLM) at 1.5 mg·kg−1 (Santa Cruz Biotechnology, Dallas, TX, USA) was performed as previously described [16]. AdTGF-β1 (5×108 plaque-forming units (PFU) per mouse) and adenovirus Cre recombinase (AdCre; 5×108 PFU per mouse) were instilled following the same protocol. BLM and adenovirus were diluted in 0.9% NaCl. Mice were euthanised by abdominal aortic bleeding at day 21 after administration. Bronchoalveolar lavage fluid (BALF) was gathered as previously described [4].
Collagen quantification
For histomorphometric assay, the amount of collagen in paraffin-embedded tissue sections was quantified by staining with Picro Sirius Red as previously described [17, 18].
For colorimetric assay, the Sircol assay was performed on lung left lobe extracts using the Sircol kit (Biocolor Ltd, Carrickfergus, UK) and following the manufacturer's recommendations.
Precision-cut lung slice preparation
3D lung tissue slices were generated as previously described [19] and cultured for 72 h in DMEM +10% serum, 1% l-glutamine and 1% penicillin/streptomycin. Three slices were pooled for each mRNA extraction.
Statistical analysis
Comparisons between different groups were performed using the nonparametric Mann–Whitney test and Prism 6 software (Graphpad, San Diego, CA, USA). p-values below 0.05 were considered statistically significant. All results are representative of at least three different experiments.
Results
TRIM33 is upregulated during pulmonary fibrosis
Histological examination of IPF lungs showed overexpression of TRIM33 when compared to non-IPF control lungs, in which TRIM33 was barely detectable (figure 1a–c). TRIM33 was overexpressed preferentially in myofibroblasts expressing smooth muscle actin (α-SMA; figure 1d–f) and in macrophages expressing macrosialin (CD68; figure 1g–i). TRIM33 overexpression could also be observed in ATII cells expressing pulmonary surfactant-associated protein C (SP-C; figure 1j–l). Analysis of publicly available microarray datasets showed that the TRIM33 mRNA level was upregulated in total lung tissue from IPF patients compared to control lung tissue (dataset GSE110147; figure 1m) but also specifically in alveolar macrophages from IPF patients (dataset GSE49072; figure 1n). The analysis of our patient cohort confirmed the increase of TRIM33 expression in the whole lung tissue of IPF patients compared to control tissue (supplementary figure S1).
Similar to the results in humans, TRIM33 was upregulated in fibrotic areas of mice receiving repeated intravenous injections of BLM (figure 2a–c). This increase was confirmed by immunoblotting and reverse transcription quantitative PCR (figure 2d and e). TRIM33 upregulation was also observed in fibrotic areas in other models of lung and pleural fibrosis, such as in mice receiving a single intratracheal instillation of BLM, or in rats and mice receiving an intrapleural injection of AdTGF-β1 compared to control rodents (supplementary figure S2a). In the BLM model, TRIM33 was expressed in fibroblasts (vimentin) and macrophages (CD68) (supplementary figure S2b).
Trim33 inhibition impairs TGF-β1 secretion in macrophages
As TRIM33 is overexpressed in alveolar macrophages of IPF patients (figure 1), we investigated the effects of TRIM33 on bone marrow-derived macrophages (BMDM). BMDM isolated from Trim33-floxed (Trim33f/f) mice were depleted for Trim33 by ex vivo exposure to an adenovirus encoding the Cre recombinase (AdCre; figure 3a). Trim33 depletion was confirmed using quantitative PCR (supplementary figure S3a). BLM stimulation of BMDM expressing TRIM33 led to an increase in TGF-β1, tumour necrosis factor (TNF)-α and interleukin (IL)-6 production (figure 3b–d), with no impact on cellular viability (supplementary figure S3c). The depletion of Trim33 did not affect the level of TGF-β1 without stimulation, while it resulted in a three-fold increase in TGF-β1 secretion upon BLM stimulation (figure 3b). In parallel, we did not observe any significant difference in Tgfb1 gene expression (supplementary figure S3b), or in secretion of pro-inflammatory cytokines (TNF-α and IL-6; figure 3c and d) or other cytokines (supplementary figure S3d) between BMDM expressing TRIM33 (exposed to the control adenovirus vector, AdDL) and BMDM depleted for TRIM33 (AdCre), both at basal level or upon BLM exposure.
Next, we investigated the functional effect of Trim33 deficiency in the crosstalk between macrophages and lung fibroblasts (figure 3e–h). Conditioned media were collected from BMDM cultured in four distinct conditions: 1) unstimulated BMDM expressing Trim33 (AdDL NaCl); 2) BMDM expressing Trim33 stimulated with BLM (AdDL BLM); 3) unstimulated BMDM depleted for Trim33 (AdCre NaCl); and 4) BMDM depleted for Trim33 stimulated with BLM (AdCre BLM). The expression of mesenchymal markers was then investigated in mouse primary pulmonary fibroblasts incubated with these conditioned media. Conditioned media from unstimulated BMDM, expressing Trim33 or not (AdDL NaCl and AdCre NaCl), had no impact on the expression of mesenchymal markers in fibroblasts (figure 3e–h). However, conditioned media from BLM-stimulated BMDM expressing Trim33 (AdDL BLM) induced an increase in mesenchymal markers, which was further enhanced when fibroblasts were stimulated by conditioned media from BLM-stimulated BMDM depleted for Trim33 (comparison of AdDL BLM versus AdCre BLM: actin alpha 2, smooth muscle (Acta2) 1.87-fold increase; serpin family E member 1 (Serpine1) 1.46-fold increase; lymphoid enhancer-binding factor 1 (Lef1) 1.66-fold increase; vimentin (Vim) 1.2-fold increase; figure 3e–h).
Altogether, these results suggest that, although Trim33 deficiency has no impact on TGF-β1 secretion at the basal level, yet it can promote the production of active profibrotic factors upon BLM stimulation, which in turn activates lung fibroblasts.
Haematopoietic-specific Trim33−/− mice are more sensitive to BLM-induced pulmonary fibrosis
To confirm the in vivo significance of TRIM33 in immune cells during pulmonary fibrosis development, we used mice with cFes-driven Cre to generate conditional Trim33 knockout mice resulting in Trim33 depletion specifically in the haematopoietic system: Trim33f/f-CrecFes mice (hsTrim33−/− mice; figure 4a) [15]. Following BLM challenge, hsTrim33−/− mice developed more severe fibrosis compared to control mice (Trim33f/f mice), characterised by worsened pulmonary alveolar structure remodelling and increased collagen accumulation throughout the lungs (figure 4b). Picro Sirius Red staining of lung sections showed a 1.7-fold increase in lung collagen content in hsTrim33−/− mice compared to a 1.4-fold increase in control mice (supplementary figure S4a). This result was confirmed by a Sircol assay, in which we found that BLM induced a significant increase in collagen content in hsTrim33−/− mouse lungs compared to control lungs (figure 4c). In lung tissues, the increase in Acta2 mRNA level induced by BLM was significantly higher in hsTrim33−/− mice compared to control mice (supplementary figure S4b). Similar findings were observed after repetitive systemic intravenous BLM exposure (supplementary figure S5a–c).
Confirming our in vitro results, TGF-β1 levels were significantly increased in the BALF from hsTrim33−/− mice compared to control wildtype mice exposed to BLM (two-fold increase) (figure 4d). In parallel, there was no difference in TNF-α or IL-6 protein level in BALF from hsTrim33−/− versus control mice after BLM exposure (figure 4e and f).
Altogether, these results suggest that TRIM33 significantly represses the establishment of a profibrotic microenvironment, specifically by inhibiting the production of active TGF-β1 by immune cells, including macrophages.
Trim33 depletion ex vivo promotes signalling pathways downstream of TGF-β1
As TRIM33 was also overexpressed in fibroblasts from IPF patients (figure 1), we investigated the effects of TRIM33 expression in primary pulmonary fibroblasts. We isolated primary pulmonary fibroblasts from Trim33f/f mice and depleted the Trim33 gene by infection with AdCre (or AdDL as control; figure 5a). The loss of TRIM33 expression was confirmed by Western blot (figure 5b). Treatment of control (AdDL) fibroblasts with recombinant TGF-β1 induced an upregulation of the SMAD pathway downstream gene mRNAs (Acta2, Serpine1, Lef1 and twist family BHLH transcription factor 1 (Twist1); figure 5c–f). Interestingly, the upregulation of those genes was further increased in Trim33-depleted fibroblasts (AdCre) compared to AdDL fibroblasts (2.98-fold increase for Acta2, 2.05 for Serpine1, 3.2 for Lef1 and 1.8 for Twist1; figure 5c–f). These results were confirmed using 3D lung tissue slices generated from Trim33f/f mice and treated with AdCre (figure 5g–j). Trim33 depletion was confirmed by quantitative PCR (figure 5g) and we showed that this depletion induced, upon TGF-β1 treatment, a significant increase in Acta2, Serpine1 and collagen type I alpha 1 chain (Col1a1) mRNA (1.9-fold increase expression for Acta2, 1.3 for Serpine1 and 1.54 for Col1a1; figure 5h–j).
Therefore, TRIM33 appears to negatively control downstream signalling of TGF-β1 by blocking overactivation of the TGF-β1 pathway in fibrotic conditions.
Trim33 inhibition in the lungs worsens BLM-induced fibrosis in mice
To confirm the in vivo significance of Trim33 expression in the lungs, we depleted Trim33 in Trim33f/f mouse lungs by intratracheal administration of AdCre (figure 6a). We observed in our model that 29.7% of the lung cells underwent recombination 48 h after AdCre intratracheal instillation, as measured using flow cytometry (supplementary figure S6a). AdCre instillation was able to induce recombination mostly in CD326 (epithelial cell adhesion molecule)-positive cells (epithelial cells) but also in ER-TR7-positive cells (fibroblasts) and in some CD68-positive cells (macrophages) (supplementary figure S6b and c). 3 days after AdCre exposure, mice were challenged with BLM and fibrosis was assessed at 21 days post BLM. AdCre-mediated Trim33 depletion in mice resulted in more severe fibrosis after BLM intratracheal challenge, compared to the AdDL mice expressing normal levels of TRIM33 (figure 6b and c; supplementary figure S7a). The upregulation of fibrosis-related genes induced by BLM at the mRNA level was significantly higher in AdCre mice compared to AdDL mice (1.64-fold increase for Col1a1, 1.41 for Serpine1 and 1.3-fold for Twist1; supplementary figure S7a). Upregulation of fibrosis-related markers was also found at the protein level for plasminogen activator inhibitor 1 (PAI-1), type I collagen and HSPB5 (figure 6d and f–h). As expected, mice treated with AdDL showed an increase in TRIM33 protein level upon BLM challenge, whereas mice treated with AdCre did not (figure 6d and e). Interestingly, whereas depletion of Trim33 potentiated fibrogenesis marker induction, in nonfibrotic conditions it did not appear to have any effect on collagen accumulation, mesenchymal differentiation or on the immune cell profile of the BALF (supplementary figure S7a and b).
TRIM33 activity and protein level are impaired by HSPB5 overexpression
TRIM33 and HSPB5 are described as key regulators of the TGF-β1/SMAD pathway. HSPB5 was upregulated following AdCre-mediated Trim33 depletion, suggesting a mechanistic regulation between both proteins. In the pulmonary tissue of Hspb5-deficient mice (Hspb5 knockout; supplementary figure S8a), TRIM33 protein level was increased compared with wildtype mice (figure 7a). This difference was further amplified following BLM challenge. At the same time, the depletion of Hspb5 induced no change in Trim33 mRNA level in total lung tissue (figure 7b). Similar results were found using primary pulmonary fibroblasts from Hspb5 knockout mice treated ex vivo with TGF-β1 (supplementary figure S8b). However, HSPB5 overexpression in human alveolar epithelial A549 cells induced a decrease in TRIM33 protein level (figure 7c), again without interfering with its mRNA level (figure 7d). Our data suggest that, in parallel to the induction of its expression after Trim33 depletion, HSPB5 negatively regulates TRIM33 protein level in a post-translational manner.
Next, we further investigated the protein–protein interaction between HSPB5 and TRIM33. Biolayer interferometry experiments using recombinant purified proteins showed that TRIM33 was able to bind directly to both HSPB5 (with a dissociation constant (Kd) of 2.91×10−9 M) and SMAD4 (with a Kd of 1.28×10−9 M) (supplementary figure S8c and d). Furthermore, by fluorescence lifetime imaging microscopy - Förster resonance energy transfer (FLIM-FRET) analysis, we showed that HSPB5 overexpression decreased SMAD4–TRIM33 interaction (supplementary figure S8e). In line with these results and the previously described inhibitory role of HSPB5 on TRIM33 activity, we demonstrated that HSPB5 was able to reduce the sumoylation of TRIM33, which is essential for its E3 ubiquitin ligase activity on SMAD4 (figure 7e).
Altogether, these results demonstrate a feedback loop between HSPB5 and TRIM33, respectively positive and negative regulators of the SMAD pathway. We have shown that the TRIM33 level is impaired upon HSPB5 overexpression. Furthermore, we have uncovered that HSPB5 promotes TRIM33 sumoylation and thus inhibits its binding to SMAD4, leading to enhanced TGF-β1 signalling.
Discussion
TRIM33 is an E3 ubiquitin ligase known as a negative regulator of TGF-β1 signalling, one of the most potent profibrotic signals, through SMAD4 ubiquitination and turnover of TGF-β receptors [6, 8]. We have shown that TRIM33 is upregulated in both IPF patients and rodent models of lung fibrosis. Trim33 depletion in vitro and in vivo caused an exacerbation of fibroblast activation, collagen deposition and pro-fibrogenic factor secretion under profibrotic conditions (i.e. on BLM or TGF-β1 exposure), suggesting a protective role of TRIM33 in lung fibrosis. Therefore, the upregulation of TRIM33 may be viewed as a failing attempt to prevent fibrosis progression in IPF and models of lung fibrosis. TRIM33 inhibitory properties on TGF-β1 signalling might be due to an insufficient level of TRIM33 expression or an intrinsic inhibition of its activity. The present works sheds light on this regulation.
Analysis of publicly available datasets demonstrated an upregulation of TRIM33 in IPF lungs. However, we hypothesise that the level of upregulation compared to control lungs remains moderate and insufficient to exert its antifibrotic action. Nevertheless, datasets only recapitulate mRNA expression, which may not be representative of the real level of TRIM33, either at the protein level or in terms of activity in the IPF lungs. Our data showed that TRIM33 protein is markedly upregulated, mainly in myofibroblasts and macrophages and also in alveolar epithelial cells, suggesting that TRIM33 may encounter post-translational regulation that prevents its antifibrotic functions. Our team previously demonstrated that HSPB5 upregulation in IPF induced a decrease in the interaction between TRIM33 and SMAD4, hampering the ubiquitination and the nuclear export of the latter [5]. In this context, HSPB5 may serve as a physiological inhibitor of the TRIM33 negative control on TGF-β1 signalling. Furthermore, we showed that TRIM33 is able to bind to both HSPB5 and SMAD4 directly. Within this complex, HSPB5 appears to reduce the level of sumoylation of TRIM33, therefore reducing its binding and ability to ubiquitinate SMAD4 and resulting in an exacerbation of the profibrotic TGF-β1 signalling. We believe that the inhibition of the interaction between TRIM33 and HSPB5 may be able to restore TRIM33 repressor effects on the TGF-β1 pathway and may be of interest for the treatment of IPF.
Among lung cells, we demonstrated that alveolar macrophages are one of the main cell types expressing TRIM33 in human and rodent lung fibrosis. While the role of inflammation in pulmonary fibrosis remains controversial [20] and despite the lack of efficacy of anti-inflammatory therapies on IPF progression, unbiased approaches have unravelled a significant immune component in IPF and alveolar macrophages appear to be critical for the development of lung fibrosis [21, 22]. Macrophage activation is heavily involved in fibrogenesis regulation in various tissues, including the liver, kidney and lungs [23], mainly through the secretion of TNF-α, IL-6 and, most importantly, TGF-β1 [24], the latter being able to modulate fibroblast proliferation and ECM production [25]. As macrophages are able to produce a variety of profibrotic signals, we hypothesised that TRIM33 may have a key role in controlling the macrophage secretory profile in a fibrotic context. While TRIM33 has been demonstrated to play a role in the inflammatory response in BMDM [26], we have shown here, in fibrotic conditions, that Trim33 deficiency promotes active TGF-β1 secretion by BMDM. This depletion does not modify the secretion of pro-inflammatory cytokines TNF-α and IL-6, demonstrating a rather specific mechanism towards the TGF-β1 pathway. Remarkably, we demonstrated that Trim33 deficiency does not interfere with Tgfb1 gene expression. TGF-β1 is primarily produced in an inactive form that requires processing inside and outside the cell in order to be secreted and activated. As a consequence, TGF-β1 activity is mainly regulated by post-translational modifications. Therefore, our data suggest that Trim33 depletion may interfere indirectly with the mechanisms involved in TGF-β1 maturation rather than with Tgfb1 gene expression.
Myofibroblasts are of major importance in pulmonary fibrosis, as they are the main secretor of ECM. They can originate from various cell types, including resident fibroblasts, fibrocytes and differentiated fibroblasts via cellular reprogramming of epithelial cells [27]. TRIM33 has been demonstrated to be involved in mesenchymal differentiation in the oncology field. For instance, TRIM33 depletion in epithelial cells has been linked to enhanced mesenchymal differentiation in several cancer models including nonsmall cell lung cancer [28], NMuMG mammary epithelial cells [29] and renal cell carcinoma [30]. In lung fibrosis conditions, we have demonstrated that Trim33 depletion, in primary pulmonary fibroblasts as well as in ex vivo cultured lung tissue, promotes mesenchymal differentiation. In addition, depletion of Trim33 in the lungs by intratracheal administration of AdCre increased sensitivity to BLM-induced fibrosis with an increase in ECM deposition. These results may suggest a crucial role of TRIM33 in the activation of mesenchymal differentiation, leading to an increase in the amount of myofibroblasts and subsequent exaggerated ECM deposition.
It is noteworthy that TRIM33 antibodies are found in the serum of patients in pathological conditions such as polymyositis and dermatopolymyositis. These diseases are associated with cancer but also with ILDs [11]. These ILDs present as nonspecific interstitial pneumonia or with a pattern of usual interstitial pneumonia, as observed in IPF [31]. Although the pathogenic role of TRIM33 antibodies is beyond the scope of the current work, we believe that they should be further explored to better understand their protective or aggravating role in IPF.
Thus, we hypothesise that TRIM33 is induced in profibrotic conditions as part of a negative feedback loop to slow down an exacerbated activation of TGF-β1 signalling. However, in several pathological conditions, such as IPF, the negative control exerted by TRIM33 is swallowed, due either to an inefficient upregulation or to an inhibition by other upregulated proteins in fibrotic conditions, such as HSPB5.
Conclusion
Our work demonstrates that TRIM33 is overexpressed in the lung during fibrotic conditions and we have shown that TRIM33 has a protective role against fibrogenesis by inhibiting the TGF-β1 pathway independently of inflammation. We believe that TRIM33 expression may represent an attempt to resolve the fibrotic process, which is overwhelmed during IPF. We believe that the complex interactions between TRIM33, SMAD4 and HSPB5 may represent key targets in the prevention of progression of fibrosis in cases of induced lung fibrosis, as in iatrogenic diseases or in IPF.
Supplementary material
Supplementary Material
Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.
Supplementary material ERJ-01346-2019.SUPPLEMENT
Shareable PDF
Supplementary Material
This one-page PDF can be shared freely online.
Shareable PDF ERJ-01346-2019.Shareable
Acknowledgements
We thank V. Saint-Giorgio (Centre de Zootechnie, Université de Bourgogne, Dijon, France) and André Bouchot (Cellimap core facility, Université de Bourgogne, Dijon, France) for the microscopy and immunohistochemistry, and P. Winckler (Dimacell imaging facility, UMR PAM, Université de Bourgogne/Agrosup Dijon, France) for technical assistance in the FLIM-FRET experiments. We are grateful to L. Delva (INSERM U1231, Dijon, France) for providing the cFes-Cre and the floxed Trim33 mice. Illustrations were adapted from Smart Servier Medical Art.
Footnotes
This article has supplementary material available from erj.ersjournals.com
Author contributions: P-M. Boutanquoi, F. Goirand and P. Bonniaud designed the study. P-M. Boutanquoi, O. Burgy, G. Beltramo, P-S. Bellaye, L. Dondaine, G. Marcion, L. Pommerolle, A. Vadel, M. Spanjaard and A. Mailleux performed the experiments and data analyses. O. Demidov supervised the mouse breeding. M. Kolb provided adenoviruses. P-M. Boutanquoi prepared the manuscript. B. Crestani, M. Kolb and C. Garrido brought important intellectual content. O. Burgy, P-S. Bellaye, F. Goirand and P. Bonniaud provided critical revision of the manuscript. All authors have read, reviewed and approved the final submitted manuscript and agree to take public responsibility for it.
Conflict of interest: P-M. Boutanquoi has nothing to disclose.
Conflict of interest: O. Burgy has nothing to disclose.
Conflict of interest: G. Beltramo has nothing to disclose.
Conflict of interest: P-S. Bellaye has nothing to disclose.
Conflict of interest: L. Dondaine has nothing to disclose.
Conflict of interest: G. Marcion has nothing to disclose.
Conflict of interest: L. Pommerolle has nothing to disclose.
Conflict of interest: A. Vadel has nothing to disclose.
Conflict of interest: M. Spanjaard has nothing to disclose.
Conflict of interest: O. Demidov has nothing to disclose.
Conflict of interest: A. Mailleux has nothing to disclose.
Conflict of interest: B. Crestani reports personal fees for lectures and travel support for meeting attendance from AstraZeneca, grants, personal fees for lectures and travel support for meeting attendance from Boehringer Ingelheim and Roche, grants from MedImmune, personal fees for lectures and consultancy, and travel support for meeting attendance from Sanofi, personal fees for advisory board work from Genzyme, outside the submitted work.
Conflict of interest: M. Kolb reports grants and personal fees from Roche, Boehringer Ingelheim and Prometic, grants from GSK, Respivert, Alkermes, Pharmaxis and Canadian Institute for Health Research, personal fees from Genoa, Indalo, Third Pole and Pieris, outside the submitted work.
Conflict of interest: C. Garrido has nothing to disclose.
Conflict of interest: F. Goirand has nothing to disclose.
Conflict of interest: P. Bonniaud reports personal fees for advisory board work and travel support for meeting attendance from Roche and Novartis, personal fees for advisory board work and reimbursement of meeting registration from Boehringer, personal fees for advisory board work from TEVA and AstraZeneca, travel support for meeting attendance from Chiesi, reimbursement of meeting registration from Stallergene, outside the submitted work.
Support statement: This work received funding from the Agence Nationale de la Recherche (ANR-16-CE14-0004-01 SHOT-IPF). P-M. Boutanquoi is supported by the “Fonds de Recherche en Santé Respiratoire” Appel d'offres Formation par la Recherche 2018/Société de Pneumologie de Langue Française (SPLF). O. Burgy is supported by the French “Investissements d'Avenir” programme, project ISITE-BFC (contract ANR-15-IDEX-0003). The C. Garrido team has the “label d'excellence” from La Ligue National Contre le Cancer and L'Association pour la “Recherche sur le Cancer” and belongs to the LabEx LipSTIC and GR-Ex. Funding information for this article has been deposited with the Crossref Funder Registry.
- Received July 8, 2019.
- Accepted March 2, 2020.
- Copyright ©ERS 2020