The immune landscape in SSc-ILD and tow genes are potential risk factors for pulmonary fibrosis

doi:10.21203/rs.3.rs-2251740/v1

Download PDF

Article

The immune landscape in SSc-ILD and tow genes are potential risk factors for pulmonary fibrosis

https://doi.org/10.21203/rs.3.rs-2251740/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Interstitial lung disease (ILD) is the most common cause of death in patients with systemic sclerosis (SSc-ILD) and immune cells are crucial in the onset and development of ILD. The aim of this study was to compare the molecular fingerprint of lung tissue from patients with SSc-ILD with that of lung tissue from normal donors, and to determine the immune landscape according to their gene expression profiles.

Methods

Two gene expression omnibus (GEO) datasets were merged as a test set, and another dataset was selected as the validation set. Lung biopsies and alveolar macrophages from 2 SSc-ILD patients and 2 healthy controls were obtained for further validation. Machine-learning algorithms were used to filter and identify potential diagnostic biomarkers of SSc-ILD in the test set. These biomarkers were examined in a validation dataset and further validated by quantitative real-time PCR and western blotting. CIBERSORT was used to quantify the proportions of immune cells in lung samples from SSc-ILD patients and healthy controls. The link between potential biomarkers and immune infiltration cells was established using a logistic regression approach.

Results

CDH3 upregulated and TNFAIP3 downregulated in SSc-ILD, and their encoded proteins (Cadherin 3 and TNFAIP3, respectively) also showed the same trend of changes. TNFAIP3 protein in alveolar macrophages derived from the alveolar lavage fluid of patients with SSc-ILD was decreased too. The proportion of M2 macrophages in SSc-ILD was significantly higher. TNFAIP3 was negatively correlated with M2 macrophages. CDH3 was positively correlated with plasma cells, M0 macrophages, and resting mast cells, and negatively correlated with M1 macrophages, resting NK cells, activated mast cells, eosinophils, and monocytes.

Conclusions

TNFAIP3 and CDH3 are two potential factors for pulmonary fibrosis. In particular, the lower expression of TNFAIP3 in alveolar macrophages of SSc-ILD patients may be linked to the maintenance of the profibrotic phenotype of macrophages. This research offers a fresh viewpoint on how SSc-ILD manifests itself at the transcriptomic and immune cell level, and may be useful in future therapeutic strategies.

Biological sciences/Immunology

Health sciences/Rheumatology

systemic sclerosis

pulmonary fibrosis

macrophage

TNFAIP3

CDH3

gene microarray

machine learning

Interstitial lung disease is the most common cause of death in patients with systemic sclerosis, with a prevalence of 30% and a 10-year mortality rate of up to 40% [1]. The precise molecular and cellular mechanisms remain unclear despite increased research efforts, and the conventional combination of immunosuppressive drugs and anti-fibrotic therapy, even in hematopoietic stem cells, have not provided the anticipated results in clinical practice.

Gene chip microarray or next-generation sequencing technology to analyze the gene expression profile of lung biopsy tissue has facilitated the elucidation of new biomarkers and new signaling pathways. Here, two gene expression omnibus (GEO) datasets were merged as a test set and another dataset was selected as the validation set. Machine-learning algorithms were used to filter and identify candidate diagnostic biomarkers of interstitial lung disease associated with systemic sclerosis (SSc-ILD) in the test set, and these biomarkers were then examined in a validation dataset and further examined in four individuals (2 SSc-ILD patients and 2 healthy controls from our medical center) by quantitative real-time PCR and western blotting.

As is the case with other autoimmune diseases, immune cells are crucial to the emergence and progression of SSc-ILD. It is unfortunate that established techniques such as immunohistochemistry and flow cytometry are unable to precisely determine the immunological makeup of SSc-ILD. According to a new study, CIBERSORT, an in-silico method for identifying cell subsets of interest, can reliably predict the immunological composition of tumor biopsies [2]. It partially compensates for the drawbacks of conventional methods. For the first time, gene expression profiling of SSc-ILDs and healthy controls (HC) was employed in conjunction with CIBERSORT to assess the proportions of immune cells in lung samples.

We then used a logistic regression approach to develop a diagnostic prediction model based on the connection between possible biomarkers and immune infiltration cells. This study offers a new viewpoint on the mechanism of SSc-ILD at the transcriptomic and immune cell level, with the hope that it will aid in its treatment.

Identification of DEGs and diagnostic biomarkers in lung biopsy

The gene expression data of a total of 29 SSc-ILD and 9 HC samples from the test set (GSE81292 and GSE76808) were retrospectively analyzed in this study.

First, the DEGs of the metadata were analyzed using the “limma” package after removing the batch effects. A total of 46 DEGs were obtained: 10 genes were significantly upregulated and 36 genes were significantly downregulated (Fig. 1A). Next, two different algorithms were used to screen potential biomarkers. The DEGs were narrowed down using the LASSO regression algorithm, resulting in the identification of 6 variables as diagnostic biomarkers for SSc-ILD (Fig. 1B), and a subset of 10 features among the DEGs were determined using the SVM-RFE algorithm (Fig. 1B). Then, four overlapping genes (CDH3, TNFAIP3, FOSB, and C10orf10) between these two algorithms were ultimately selected (Fig. 1B). Furthermore, to obtain more accurate and reliable results, the validation dataset (GS48149) was used to verify the expression levels of these four genes in SSc-ILD lung biopsies. Consistent with the test set, CDH3 was upregulated in SSc-ILD (p < 0.001; Fig. 1C), while TNFAIP3 was notably deregulated compared with those in the HC group (p = 0.0006; Fig. 1D). However, there was no significant difference between the two groups of FOSB and C10orf10 expression. Validation of SSc-ILD lung tissues by qPCR and western blotting showed that, along with the changes in these two genes (CDH3 upregulation and TNFAIP3 downregulation), their encoded proteins (Cadherin 3 and TNFAIP3, respectively) also showed the same trend of changes in SSc-ILD (Fig. 1E–H). Furthermore, the expression of TNFAIP3 in alveolar macrophages derived from the alveolar lavage fluid of patients with SSC-ILD was decreased too (Fig. 1I).

On the basis of the above findings, we chose to conduct more detailed analyses of CDH3 and TNFAIP3. These two genes were applied to establish a diagnostic model using logistic regression. In the test set, the diagnostic ability of TNFAIP3 and CDH3 in discriminating SSc-ILD from the control samples displayed a favorable diagnostic value, with an AUC of 0.98 (95% CI 0.931–0.972) for TNFAIP3 and an AUC of 0.98 (95% CI 0.904–0.968) for CDH3 (Fig. 1J). A powerful discriminative ability was confirmed in the validation set, with an AUC of 0.90 (95% CI 0.766–0.972) for TNFAIP3 and an AUC of 0.91 (95% CI 0.745–0.995) for CDH3 (Fig. 1K), which indicated that the featured biomarkers had potential diagnostic ability.

Enrichment Analysis

GO and KEGG enrichment analyses of the DEGs were performed to obtain more information about TNFAIP3 and CDH3. The GO-enriched sets associated with these two genes were drawn in a chord plot (Fig. 2A). These GO sets are mainly involved in the following bioprocesses: response to lipopolysaccharide, response to molecules of bacterial origin, negative regulation of cytokine production, regulation of transforming growth factor beta 1 (TGF-β1) production, regulation of endothelial cell apoptotic processes, tissue homeostasis, negative regulation of immune response, regulation of inflammatory response, regulation of B cell activation, and regulation of epithelial cell proliferation, all of them were associated with pulmonary fibrosis [3–5], in particular, TGF-β1 is the master regulator of fibrosis. Furthermore, GO sets such as regulation of cytokine production, immune response, inflammatory response, and B cell activation indicated humoral immunity was dominant in the occurrence and development of pulmonary fibrosis in SSc patients. In addition, the enriched KEGG pathways that included CDH3 and/or TNFAIP3 were as follows: IL-17 signaling, TNF signaling, cytokine–cytokine receptor interaction, rheumatoid arthritis, viral protein interaction with cytokine and cytokine receptor, NF-κB signaling pathway, chemokine signaling pathway, AGE-RAGE signaling in diabetic complications, and NOD-like receptor signaling (Fig. 2B).

GSEA of all the normalized genes was performed for functional and pathway enrichment analyses in a broader and deeper perspective. All signaling pathways associated with collagen synthesis were activated (Fig. 2D). Puzzlingly, the top five enriched sets, which are known to promote fibrosis, were inhibited in SSc-ILD (Fig. 2C): tyrosine kinases receptor signaling [6], MAPK signaling [7], EGF–EGFR signaling [8], secreted factors signaling, and nuclear receptor meta signaling pathway [9]. This could be the result of negative feedback regulation at the end-stage of pulmonary fibrosis in an attempt to prevent further fibrosis.

Immune Cell Infiltration

We next explored the composition of immune cells in SSc-ILD and HC samples (Fig. 3A). The proportions of resting NK cells (p < 0.001), activated dendritic cells (p = 0.012), and eosinophils (p = 0.001) in SSc-ILD were significantly lower than normal, while the proportions of plasma cells (p < 0.001), M2 macrophages (p < 0.001), and resting mast cells (p < 0.001) in SSc-ILD were significantly higher (Fig. 3A). Subsequently, the correlation of 22 types of immune cells was calculated (Fig. 3B). M2 macrophages were significantly negatively correlated with activated mast cells (r = − 0.48), CD4 naïve T cells (r = − 0.37), activated dendritic cells (r = − 0.46), resting NK cells (r = − 0.37), eosinophils (r = − 0.22), memory B cells (r = − 0.32), gamma delta T cells (r = 0.22), and M1 macrophages (r = 0.17), and the resting mast cells were negatively correlated with activated mast cells (r = − 0.69), eosinophils (r = − 0.47), monocytes (r = − 0.63), resting NK cells (r = − 0.49), activated dendritic cells (r = − 0.43), neutrophils (r = − 0.38), and plasma cells (r = 0.49). Plasma cells were negatively correlated with activated mast cells (r = − 0.53), eosinophils (r = − 0.57), resting NK cells (r = − 0.43), activated dendritic cells (r = − 0.31), activated CD4 memory T cells (r = − 0.31), neutrophils (r = − 0.44), M0 macrophages (r = − 0.28), and M1 macrophages (r = − 0.4), while they were positively correlated with regulatory T cells (r = 0.4). Activated dendritic cells were significantly positively correlated with activated mast cells (r = 0.56), eosinophils (r = 0.56), resting NK cells (r = 0.78), and neutrophils (r = 0.42). Eosinophils were significantly positively correlated with activated mast cells (r = 0.75), activated dendritic cells (r = 0.56), gamma delta T cells (r = − 0.34), and regulatory T cells (r = 0.4).

Correlation Analyses Between The Two Candidate Genes And Infiltrating Immune Cells

As shown in Fig. 4A, CDH3 was positively correlated with plasma cells (r = 0.43, p = 0.007), M0 macrophages (r = 0.40, p = 0.01), and resting mast cells (r = 0.32, p = 0.049), and negatively correlated with M1 macrophages (r = − 0.35, p = 0.03), resting NK cells (r = − 0.40, p = 0.01), activated mast cells (r = − 0.60, p < 0.001), eosinophils (r = − 0.60, p < 0.001), and monocytes (r = − 0.60, p < 0.001).

As shown in Fig. 4B, TNFAIP3 was positively correlated with activated mast cells (r = 0.56, p < 0.001), resting NK cells (r = 0.36, p = 0.03), eosinophils (r = 0.34, p < 0.032), monocytes (r = 0.34, p = 0.34), and activated dendritic cells (r = 0.34, p = 0.039), and negatively correlated with macrophages 2 (r = 0.58, p < 0.001) and resting mast cells(r = − 0.72, p < 0.001).

Systemic sclerosis primarily affects young and middle-aged women and is characterized by immune dysregulation and progressive fibrosis. Pulmonary fibrosis is the most common cause of death in patients with systemic sclerosis, with an incidence of up to 30% and a 10-year mortality rate of up to 40% [10]. SSc-ILD is thought to be the result of interactions between inflammation, vascular injury, autoimmunity, and fibrosis [7]. With the development of novel therapeutic agents designed to target these immune cells, accurate assessment of immune cells appears to be increasingly important [11]. Recently, cancer studies found that T lymphocyte subsets such as CD8 + may predict the effects of existing and emerging immunotherapies, highlighting the importance of investigating immune cell infiltration in SSc-ILD and the relationship between immune cells and suspect biomarkers [12].

We conducted an integrated analysis of the test set. A total of 46 DEGs were identified: 10 upregulated genes and 36 downregulated genes. Six genes were screened out by LASSO regression algorithm, and five genes were determined using the SVM-RFE algorithm. All overlapping genes (CDH3, TNFAIP3, FOSB, and C10orf10) between these two algorithms were obtained, and their expression levels were confirmed in the validation dataset GSE41849; hereafter, we found that only CDH3 and TNFAIP3 showed consistent expression trends in the test and validation sets. Also encouraging was that along with the changes in these two genes (CDH3 upregulation and TFAIP3 downregulation), their encoded proteins (Cadherin3 and TNFAIP3, respectively) also showed the same changes in SSc-ILD lung tissues, as validated by qPCR and western blotting.

CDH3 encodes a classical cadherin of the cadherin superfamily, Cadherin-3. Cadherin-3 is a calcium-dependent, cell–cell adhesion glycoprotein that plays a vital role in protecting the structural integrity of epithelial tissues. Cadherin-3 regulates a variety of cellular homeostatic processes involved in cell differentiation, shape, polarity, proliferation, and migration [13]. Overexpression of Cadherin-3 is associated with the aggressive behavior of tumors such as breast [14], ovarian [15], and gastric cancer [16]. GO enrichment also demonstrated that CDH3 is involved in the regulation of endothelial cell apoptotic processes, tissue homeostasis, epithelial cell proliferation, and epithelial cell proliferation. We tested patients at our medical center and confirmed that along with the upregulation of CDH3, Cadherin-3 was also significantly increased in SSc-ILD. On this basis of this evidence, we speculated that CDH3 and its encoded protein Cadherin-3 play an important role in the pathogenesis of SSc-ILD.

TNFAIP3, whose expression is rapidly induced by tumor necrosis factor (TNF), has been characterized as a ubiquitin-editing enzyme and has been shown to inhibit NF-κB activation as well as TNF-mediated apoptosis in endothelial cells [17] and fibroblasts [18]. Several lung histological changes are obvious in patients with SSc-ILD, including gaps, vacuolization, apoptosis of endothelial cells, perivascular inflammation, and fibrosis, which suggests that the clinical manifestations of SSc-ILD are the result of vascular lesions developing in lung [19]. Following injury to endothelial cells and alveolar epithelial cells, fibroblasts play a major role in the pathogenesis of SSc-ILD [10]. GO and KEGG enrichment also proved that TNFAIP3 is involved in the regulation of endothelial cell apoptotic processes, epithelial cell proliferation, tissue homeostasis, and NF-κB signaling. It is encouraging that the decline of TNFAIP3 and its encoded protein were further confirmed by qPCR and western blotting in SSc-ILD patients. In view of these findings, it was reasonable to infer that the inadequate expression of TNFAIP3 is associated with persistent NF-κB signaling activation, which led to uncontrolled apoptosis of endothelial and alveolar epithelial cells and excessive proliferation of fibroblasts in SSc-ILD.

We then investigated immune cell infiltration and the correlation between immune cells with CDH3 or TNFAIP3 expression. As a result of CIBERSOTR analysis, six immune cell subtypes were found to be closely involved in important biological processes of SSc-ILD. An increased infiltration of plasma cells, M2 cells, and resting mast cells, and a decreased infiltration of resting NK cells, activated dendritic cells, and eosinophils were found to be potentially related to the occurrence and development of pulmonary fibrosis. Plasma cell gene signatures were elevated in SSc patients’ blood and lung tissue, and highly correlated with collagen gene expression [20]. Macrophages are at the crossroads of inflammation and fibrosis and play a crucial role in SSc. M1 macrophages promote inflammation and M2 macrophages participate in fibrosis through the secretion of profibrotic mediators, such as transforming growth TGF-β and platelet-derived growth factor [21]. Here, immune cell infiltration analysis revealed that the increase in M2 macrophages was accompanied by decreased TNFAIP3. Furthermore, we further confirmed TNFAIP3 was decreased in alveolar macrophages isolated from alveolar lavage fluid of SSc-ILDs. Coincidentally, only 2 years prior to this study, Hu and colleagues proved in detail that changes in TNFAIP3 in alveolar macrophages led to continuous accumulation of C/EBPβ transcription factor in alveolar macrophages of pulmonary fibrosis mice and idiopathic pulmonary fibrosis patients, which promoted fibrosis development [22].

TNFAIP3 and CDH3 are two potential factors for pulmonary fibrosis. In particular, the lower expression of TNFAIP3 in alveolar macrophages of SSc-ILD patients may be linked to the maintenance of the profibrotic phenotype of macrophages. To achieve safe and effective treatment, precise control over various types of immune cells is needed. Therefore, bioinformatic analyses to identify novel biomarkers of SSc-ILD associated with the magnitude of immune cell infiltration will be helpful in its treatment.

Data processing and screening of differentially expressed genes

In this study, we downloaded three microarray datasets (GSE81292, GSE76808, and GSE48149) of SSc-ILD from the GEO database (https://www.ncbi.nlm.nih.gov/geo/). All lung samples from the three datasets were integrated by batch normalization using the “sva” package in R software (version 3.6.3) to reduce batch effects and heterogeneity. The first two datasets GSE81292 and GSE76808 were merged (29 SSc-ILDs vs 9 HCs) and selected as the test set. Analysis of differentially expressed genes (DEGs) (log₂FC >|2|, p < 0.05) was performed between the SSc-ILD and HC samples. GSE48149 (13 SSc-ILD vs 9 HC: 7 SSc-ILD samples were eliminated due to the lack of ILD evidence) was chosen as the validation set (Table 1).

Table 1

Microarray Data
GEO assession	Platform	Health control	SSc-ILD	Source tissue	Attribute
GSE81292	GPL18991	GSM2149865, GSM2149866 GSM2149867, GSM2149868 GSM2149869	GSM2149850, GSM2149851, GSM2149852, GSM2149853, GSM2149854, GSM2149855, GSM2149856, GSM2149857, GSM2149858, GSM2149859, GSM2149860, GSM2149861, GSM2149862, GSM2149863, GSM2149864	lung	Test set
GSE76808	GPL571	GSM2038281, GSM2038282 GSM2038283, GSM2038284	GSM2038267, GSM2038268, GSM2038269, GSM2038270, GSM2038271, GSM2038272, GSM2038273, GSM2038274, GSM2038275, GSM2038276, GSM2038277, GSM2038278, GSM2038279, GSM2038280,	lung	Test set
GSE48149	GPL16221	GSM1169973, GSM1169974 GSM1169975, GSM1169976 GSM1169977, GSM1169978 GSM1169979, GSM1169980 GSM1169981	GSM1169990, GSM1169991, GSM1169992, GSM1169993, GSM1169994, ,GSM1169995, GSM1169996, GSM1169997, GSM1169998, GSM1169999, GSM1170000, GSM1170001, GSM1170002	lung	Validation set

Candidate Biomarker Screening

Two machine-learning algorithms were applied to identify prognostic genes (biomarkers). The least absolute shrinkage and selection operator (LASSO), a regression analysis algorithm that uses regularization to improve the prediction accuracy, was used to identify genes that were considerably related to the discrimination of SSc-ILDs and HCs. Support vector machine (SVM), a supervised machine-learning technique widely utilized for both classification and regression, was employed, and to avoid overfitting, an RFE algorithm was utilized to identify the optimal genes [23]. Thus, SVM recursive feature elimination (SVM-RFE) was applied to select the appropriate features.

Human Subjects

We obtained lung biopsies from 2 SSc-ILD patients and 2 healthy controls (tissues excised away from the cancer lesion after lung cancer surgery) (Table 2) from the first affiliated Hospital of FuJian Medical University for total RNA and protein extraction. Primary alveolar macrophages from these four individuals were obtained by alveolar lavage. All protocols using human specimens were approved by the Institutional Review Board of the Fujian medical university. informed consent was obtained from all subjects. The study conformed to the principles outlined in the Declaration of helsinki.

Table 2

Patients enrolled for validation
	Diagnosis	Age	Male/Female	Immunosuppressive agents	Source tissue
Patient 1	SSc-ILD	36	Female	Initial diagnosis, no treatment	Lung, bronchoalveolar lavage fluid	Validation
Patient 2	SSc-ILD	51	Male	Initial diagnosis, no treatment		Validation
Patient 3	Lung cancer	61	Female	none		Health control
Patient 4	Lung cancer	46	Male	none		Health control

Diagnostic Value Of Feature Biomarkers In Ssc-ild

To calculate the predicted value of overlapping biomarkers obtained by performing LASSO and SVM-RFE, a receiver operating characteristic (ROC) curve was generated. The area under the ROC curve (AUC) value was utilized to determine the diagnostic effectiveness in discriminating SSc-ILDs from control samples. Furthermore, the expression and ROC curves of candidate genes were further validated in the validation dataset, GSE48149.

Functional Enrichment Analysis

Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis of the DEGs, and Gene Set Enrichment Analysis (GSEA) of all the normalized genes in the test dataset of SSC-ILD and HC samples were performed by R software (version 3.6.3). The analysis was conducted using the Molecular Signatures Database (MSigDB) of KEGG gene sets (c2.cp.kegg.v7.2.symbols). A gene set was regarded as significantly enriched if (p-adj < 0.05 and q-value < 0.25) and the false discovery rate was < 0.025.

Immune Landscape In Ssc-ild Lung Biopsies And Correlation With Identified Genes

To quantify the relative proportions of infiltrating immune cells in lung biopsies according to gene expression, CIBERSORT (https://cibersortx.stanford.edu/) was employed to estimate immune cell infiltration in SSc-ILD lung biopsies using a reference set with 22 immune cell subtypes (LM22) with 1,000 permutations [24]. Correlation analysis and visualization of 22 types of infiltrating immune cells were performed using the R package “corrplot.” Violin plots were drawn using the “vioplot” package in R to visualize the differences in immune cell infiltration between SSc-ILD and HC samples .The association of the identified gene biomarkers with the levels of infiltrating immune cells in SSc-ILD lung biopsies was explored using Spearman's rank correlation analysis in R.

Rna Extraction And Quantitative Real-time Pcr

Total RNA from SSc-ILD lung tissues was extracted and cDNA was prepared by reverse transcription (Thermo Fisher Scientific, former Savant, MA, USA). TNFAIP3 gene expression was determined by quantitative reverse-transcription polymerase chain reaction (qRT-PCR) assay according to the manufacturer's instructions (Thermo Fisher ABI QuantStudio® 5). TNFAIP3-specific primers used for qRT-PCR were as follows: TCCTCAGGCTTTGTATTTGAGC (forward), TGTGTATCGGTGCATGGTTTTA (reverse). CDH3-specific primers: ATCATCGTGACCGACCAGAAT (forward), GACTCCCTCTAAGACACTCCC (reverse). All samples were measured in triplicate in independent experiments.

Alveolar Macrophages Were Extracted And Cultured From Alveolar Lavage Fluid

A total of 40ml alveolar lavage fluid per people (two SSc-ILD and two HCs) was collected in sterile 50ml centrifuge tubes and centrifuged at 400 ×g for 10 min at 4°C. The supernatant was discarded, leaving a thick layer of phlegm and cell residue, and three times the volume of 0.1% DTT was added. The samples were then incubated in a 37°C water bath to split sputum for 15 minutes, which was diluted with equal volume of PBS, strained through a 100-mesh strainer, and centrifuged at 400 ×g for 5 minutes, before resuspension in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS). The cells were then seeded into dishes and non-adherent cells were washed with PBS 2 hours after adherence. Pure alveolar macrophages were obtained and cultured in RPMI1640 medium supplemented with 10% FBS, 100000 U/ml penicillin, and 1000 mg/ml streptomycin at 37°C under 5% CO₂ in a high-humidity atmosphere.

Immunoblotting Of Tnfaip3

Total protein was extracted from SSc-ILD lung tissues by CT-guided percutaneous lung puncture, and control lung tissues (tissues excised away from the cancer lesion after lung cancer surgery) and alveolar macrophages were treated with RIPA buffer (Cell Signaling Technology, Beverly, MA, USA). BCA protein assay kits (Beyotime Technology, Shanghai, China) were used to determine the protein concentrations. The protein lysates then underwent SDS-PAGE electrophoresis and nitrocellulose membrane transfer, followed by immunoblotting using anti-TNFAIP3 antibody (Abcam, Cambridge, MA, USA). The images were captured by a Bio-Rad Image Station (Bio-Rad Laboratories, Hercules, CA, USA)

Statistical analysis

All statistical analyses were conducted using R (version 3.6.3). For continuous variables data, Student's t-test was used to compare the normal distribution variables, and Mann–Whitney U test was used for abnormal distribution variables. “limma” package was used for DEG screening, LASSO regression analysis was performed using the “glmnet” package, and the SVM-RFE algorithm was performed using the “e1071,” “kernlab,” and “caret” packages. The “venn” package was used to identify overlapping genes, and the “pROC” package was used for ROC curve analysis. The “clusterProfiler” package was used for enrichment analysis, “Org.hs.eg. db package” for ID conversion, and “GOplot” package for calculating zscore. “corrplot,” “vioplot,” reshape2,”“ggpubr,” and “ggExtra” were used for immune cell infiltration and correlation analyses.“ggplot2” package was used for visualization. All statistical analyses were two-sided, with p < 0.05 regarded as statistically significant.

Acknowledgements

We thank all individuals participated in this study

Authors’ contributions

Conception and design of the work, Yang BeiBbei and Xiao Hua; acquisition and analysis of data, RuiHongbin and Xue Juan; interpretation of data, Yang BeiBei, Xiao Hua, and Liu Jizan; writing and preparing the original draft, Liu Jizan; writing, reviewing and editing the paper, YBB and XH; supervision, Rui Hongbin; project administration, Rui Hongbin and Xiao Hua. All authors have read and agreed to the published version of the manuscript and to have agreed to both be personally accountable for the author's contributions and ensure to answer any questions related to the accuracy or integrity of any part of the work. All authors read and approved the final manuscript.

Availability of data and materials

The [GSE datasets] data that support the findings of this study are available in the GEO database (https://www.ncbi.nlm.nih.gov/geo/) with the following data accession identifier(s): GSE81292, GSE76808, and GSE48149. The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests

The authors declare no conflicts of interest regarding this work. The corresponding authors have the right to and do speak on behalf of all authors.

Ethics approval and consent to participate

This study was reviewed and approved by the medical ethics committee of the First Affiliated Hospital of FuJian Medical University, China. Written informed consents were obtained from all patients before operation.

Consent for publication

Informed consent was obtained from all individual participants included in the study.

Funding

The National Natural Science Foundation of HuNan Province (No. 2020JJ8004).

Lee, D.M. & Weinblatt, M.E. Rheumatoid arthritis. Lancet 358, 903–911 (2001).
Chen, B., Khodadoust, M.S., Liu, C.L., Newman, A.M. & Alizadeh, A.A. Profiling Tumor Infiltrating Immune Cells with CIBERSORT. Methods Mol Biol 1711, 243–259 (2018).
Tyanova, S., et al. Proteomic maps of breast cancer subtypes. Nat Commun 7, 10259 (2016).
Newman, A.M., et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods 12, 453–457 (2015).
Hu, X., et al. PI3K-Akt-mTOR/PFKFB3 pathway mediated lung fibroblast aerobic glycolysis and collagen synthesis in lipopolysaccharide-induced pulmonary fibrosis. Lab Invest 100, 801–811 (2020).
Ramadhiani, R., Ikeda, K., Hirata, K.I. & Emoto, N. Endothelial Cell Senescence Exacerbates Pulmonary Fibrosis Potentially Through Accelerated Endothelial to Mesenchymal Transition. Kobe J Med Sci 67, E84-E91 (2021).
Meng, X.M., Nikolic-Paterson, D.J. & Lan, H.Y. TGF-beta: the master regulator of fibrosis. Nat Rev Nephrol 12, 325–338 (2016).
Grimminger, F., Gunther, A. & Vancheri, C. The role of tyrosine kinases in the pathogenesis of idiopathic pulmonary fibrosis. Eur Respir J 45, 1426–1433 (2015).
Zhang, Y., et al. Signaling Pathways Involved in Diabetic Renal Fibrosis. Front Cell Dev Biol 9, 696542 (2021).
Conese, M. & Di Gioia, S. Pathophysiology of Lung Disease and Wound Repair in Cystic Fibrosis. Pathophysiology 28, 155–188 (2021).
Trauner, M. & Fuchs, C.D. Novel therapeutic targets for cholestatic and fatty liver disease. Gut 71, 194–209 (2022).
Perelas, A., Silver, R.M., Arrossi, A.V. & Highland, K.B. Systemic sclerosis-associated interstitial lung disease. Lancet Respir Med 8, 304–320 (2020).
Huang, C., et al. Immune checkpoint molecules. Possible future therapeutic implications in autoimmune diseases. J Autoimmun 104, 102333 (2019).
Tumeh, P.C., et al. PD-1 blockade induces responses by inhibiting adaptive immune resistance. Nature 515, 568–571 (2014).
Cavallaro, U. & Dejana, E. Adhesion molecule signalling: not always a sticky business. Nat Rev Mol Cell Biol 12, 189–197 (2011).
Paredes, J., et al. P-cadherin overexpression is an indicator of clinical outcome in invasive breast carcinomas and is associated with CDH3 promoter hypomethylation. Clin Cancer Res 11, 5869–5877 (2005).
Patel, I.S., Madan, P., Getsios, S., Bertrand, M.A. & MacCalman, C.D. Cadherin switching in ovarian cancer progression. Int J Cancer 106, 172–177 (2003).
Akter, T., Silver, R.M. & Bogatkevich, G.S. Recent advances in understanding the pathogenesis of scleroderma-interstitial lung disease. Curr Rheumatol Rep 16, 411 (2014).
Wertz, I.E., et al. De-ubiquitination and ubiquitin ligase domains of A20 downregulate NF-kappaB signalling. Nature 430, 694–699 (2004).
Lee, E.G., et al. Failure to regulate TNF-induced NF-kappaB and cell death responses in A20-deficient mice. Science 289, 2350–2354 (2000).
Hamamdzic, D., Kasman, L.M. & LeRoy, E.C. The role of infectious agents in the pathogenesis of systemic sclerosis. Curr Opin Rheumatol 14, 694–698 (2002).
Lafyatis, R., O'Hara, C., Feghali-Bostwick, C.A. & Matteson, E. B cell infiltration in systemic sclerosis-associated interstitial lung disease. Arthritis Rheum 56, 3167–3168 (2007).
Morse, C., et al. Proliferating SPP1/MERTK-expressing macrophages in idiopathic pulmonary fibrosis. Eur Respir J 54(2019).
Liu, S.S., et al. Targeting Degradation of the Transcription Factor C/EBPbeta Reduces Lung Fibrosis by Restoring Activity of the Ubiquitin-Editing Enzyme A20 in Macrophages. Immunity 51, 522–534 e527 (2019).

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

The immune landscape in SSc-ILD and tow genes are potential risk factors for pulmonary fibrosis

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Introduction

Results

Identification of DEGs and diagnostic biomarkers in lung biopsy

Enrichment Analysis

Immune Cell Infiltration

Correlation Analyses Between The Two Candidate Genes And Infiltrating Immune Cells

Discussion

Conclusions

Materials And Methods

Data processing and screening of differentially expressed genes

Candidate Biomarker Screening

Human Subjects

Diagnostic Value Of Feature Biomarkers In Ssc-ild

Functional Enrichment Analysis

Immune Landscape In Ssc-ild Lung Biopsies And Correlation With Identified Genes

Rna Extraction And Quantitative Real-time Pcr

Alveolar Macrophages Were Extracted And Cultured From Alveolar Lavage Fluid

Immunoblotting Of Tnfaip3

Statistical analysis

Declarations

References

Additional Declarations

Status:

Version 1