Bioinformatic analysis of microRNA and mRNA Regulation in peripheral blood mononuclear cells of patients with chronic obstructive pulmonary disease

Background Chronic obstructive pulmonary disease (COPD) is a progressive, irreversible chronic inflammatory disorder typified by increased recruitment of monocytes, lymphocytes and neutrophils. Because of this, as well as the convenience of peripheral blood nuclear cells (PBMCs) assessments, miRNA profiling of PBMCs has drawn increasing attention in recent years for various disease. Therefore, we analyzed miRNA and mRNA profiles to understand their regulatory network between COPD subjects versus smokers without airflow limitation. Methods miRNA and mRNA profiling of PBMCs from pooled 17 smokers and 14 COPD subjects was detected by high-throughput microarray. The expression of dysregulated miRNAs were validated by q-PCR. The miRNA targets in dysregulated mRNAs were predicted and the pathway enrichment was analyzed. Results miRNA microarray showed that 8 miRNAs were up-regulated and 3 miRNAs were down-regulated in COPD subjects compared with smokers; the upregulation of miR-24-3p, miR-93-5p, miR-320a and miR-320b and the downregulation of miR-1273 g-3p were then validated. Bioinformatic analysis of regulatory network between miRNA and mRNA showed that NOD and TLR were the most enriched pathways. miR-24-3p was predicted to regulate IL-18, IL-1β, TNF, CCL3 and CCL4 and miR-93-5p to regulate IκBα. Conclusions The expression of miRNA and mRNA were dysregulated in PBMCs of COPD patients compared with smokers without airflow limitation. The regulation network between the dysregulated miRNA and mRNA may provide potential therapeutic targets for COPD.


Background
Chronic obstructive pulmonary disease (COPD) is a progressive, irreversible chronic inflammatory disorder. Caused predominantly by cigarette smoking, COPD is one of the leading causes of mortality globally [1].
Although cigarette smoking is the major cause of COPD, there is currently no satisfactory therapy to treat individuals once the disease is established. Inflammation plays a pivotal role in the disease process, with CD8 + T lymphocytes, neutrophils and macrophages being the main type of immune cells of local inflammatory milieu of COPD [2]. Different immunoregulatory properties of T cells and monocytes have been demonstrated in COPD patients [3], and the inflammatory response in the lungs of COPD patients is strongly linked to tissue destruction and alveolar airspace enlargement [4], in part due to the loss of lung structural cells due to heightened apoptosis. microRNAs (miRNAs) are a growing family of small non-coding RNAs (approximately 19 to 25 nucleotides long) with a regulatory function on gene expression [5]. Through binding to the 3′ untranslated region (3′ UTR) of target messenger RNAs, miRNAs can lead to direct inhibition of protein translation or degradation of messenger RNA [5]. In addition, miRNAs can alter gene expression by targeting transcription factors and DNA methyltransferases. In this way, miRNAs work as posttranscriptional regulators of gene expression and control various cellular processes such as differentiation, proliferation, and cell-cell interaction. The dysregulation of miRNAs is linked to a wide spectrum of diseases, including proliferative vascular disease, cardiac disorders, lung diseases, kidney diseases, diabetes mellitus, fibrosis and cancer [6][7][8].
A few studies have been conducted to reveal the dysregulation and role of miRNAs in COPD. Ezzie et al. [9] compared the miRNA expression in lung tissue derived from smokers with and without COPD and identified 70 differentially expressed miRNAs. A report by Pottelberge et al. [10] demonstrated that 34 miRNAs were differentially expressed between never-smokers and current smokers without airflow limitation, and 8 miRNAs were expressed at a significantly lower level in current-smoking patients with COPD compared with never-smokers without airflow limitation. Another study showed that microRNA-34c is associated with emphysema severity in COPD [11]. Sato et al. [12] observed reduced miR-146a expression in lung fibroblasts of patients with COPD and showed that miR-146a deficiency increased the expression of PGE 2 through depression of the miR-146a target COX-2. Finally, Lewis et al. [13] found downregulated miR-1 expression in quadriceps muscles and speculated that this is linked to the muscle weakness observed in COPD.
On account of the convenience of peripheral blood nuclear cells (PBMC) assessments, the miRNA profiling of PBMC has drawn increasing attention in recent years in various diseases such as cancer [14,15], Alzheimer's Disease [16], diabetes [17] and autoimmune disease [18]. However, there is no report regarding miRNA expression profiles of PBMC in COPD patients. Therefore, in this study, we sought to determine if miRNAs were differentially expressed in PBMCs of COPD patients and if miRNA expression may be linked to dysregulated mRNA expression relevant to the pathogenesis of COPD. We analyzed miRNA and mRNA expression profiles in PBMCs from COPD patients versus smokers without airflow limitation. The mRNA targets of the dysregulated miRNA were predicted and pathway enrichment was analyzed. We identified a signature of COPD-associated miRNA, such that miR-24-3p, miR-93-5p, miR-320a, miR-320b and miR-1273g-3p were significantly dysregulated in COPD patients. Bioinformatic analysis between miRNA and predicted mRNA showed that NOD and TLR were the most enriched pathways. In NOD pathway, miR-24-3p was predicted to regulate IL18/IL1B/TNF, and miR-93-5p to regulate NLRP3/IL6/NFKBIA. In TLR pathway, miR-24-3p was predicted to regulate CCL3/CCL4/IL1B/TNF, and miR-93-5p to regulate IL6/CXCL10/NFKBIA.

Subjects
Peripheral venous blood was taken in heparin-coated tubes from 17 smokers without airflow limitation and 14 COPD subjects at Department of Respiration, The First Affiliated Hospital, Xi'an Jiaotong University, Xi'an, China. COPD subjects were eligible for this study if they met the following criteria: age ≥ 50 and ≤76 years; smoking history (≥20 pack-years); post-bronchodilator FEV 1 ≥ 25% of predicted value and post-bronchodilator FEV 1 /forced vital capacity (FVC) ≤ 0.70; no history of asthma, atopy (as assessed by an allergy skin prick test during screening) or any other active lung disease. Patients on home oxygen or with raised carbon dioxide tension (>44 mmHg), α 1 -antitrypsin deficiency, recent exacerbation (in the last 4 weeks), an uncontrolled medical condition or hypersensitivity to inhaled corticosteroids and bronchodilators were not eligible for the study. All smokers without airflow limitation met the following criteria: age ≥ 42 and ≤75 years, Post-BD FEV1% predicted >80, no diagnostic cancer, diabetes, cardiovascular disease and hypertension, no use of inhaled or oral corticosteroids in the previous 6 months, no atopy, and no respiratory tract infection 1 month prior to the study. Patient characteristics are in Table 1. The experimental procedures were performed with ethical approval from the Research Ethics Boards of The First Affiliated Hospital, Xi'an Jiaotong University (2015-015).

Messenger RNA microarray
Samples were prepared for mRNA microarray analysis using Agilent Human Gene Expression Microarray V4.0 (Santa Clara, California, USA). Hybridized slides were then washed and scanned with Agilent Microarray Scanner System (G2565CA).

Data analysis
The miRNA and mRNA array data were analyzed for data summarization, normalization and quality control using GeneSpring V11.5 software (Agilent). To select differentially expressed genes, we used threshold values of >2 fold change. The data were Log2 transformed and median centred by genes using the Adjust Data function of CLUSTER 3.0 software. Further analysis was performed by hierarchical clustering with average linkages. Finally, we performed tree visualization using Java Treeview (Stanford University School of Medicine, Stanford, CA, USA).

Target prediction and network analysis
The target genes for miRNAs were predicted using mi-Randa, MirTarget2, PicTar, PITA and TargetScan. The regulation network diagram between miRNAs and mRNAs was generated using Cytoscape. Based on the data of mRNA array, the predicted target genes that negatively regulated by the validated dysregulated miR-NAs were selected for the further pathway enrichment analysis. The DAVID [19] online analysis tool was used and the significant enrichment threshold was P value of Modified Fisher exact less than 0.05 and enriched gene count more than 2.

MicroRNAs are dysregulated in PBMCs of COPD patients
We investigated the miRNA profiling in pooled PBMCs from 17 smokers without airflow limitation and 14 COPD patients (Fig. 1). Compared with smokers, there were 103 up-regulated and 34 down-regulated miRNAs in COPD patients. We selected the dysregulated miR-NAs with differences in the fluorescence intensity higher than 1000 between the smokers and COPD patients. As shown in Table 3, there were a total of 8 up-regulated and 3 down-regulated miRNA; these were selected for the further validation and analysis.

Validation of dysregulated miRNAs in PBMCs of COPD patients
Among the selected 11 dysregulated miRNAs in COPD patients versus smokers, we validated the expression of 5 miRNAs (miR-24-3p, miR-93-5p, miR-320a, miR-320b and miR-1273 g-3p) in smokers and COPD patients by qRT-PCR, as shown in Fig. 2a. MiR-1273 g-3p was the most highly expressed miRNA in PBMCs of smokers, while miR-320a and miR-320b had the higher relative abundance in PBMCs of COPD patients (Fig. 2b). The correlation analysis between miRNA expression and FEV1% predicted showed that significant negative relevance appeared in the expression of miR-24-3p, miR-320a and miR-320b and positive relevance appeared in miR-1273 g-3p (Fig. 2c). However, there is no difference in miRNA sets between ex-and current smokers in the smokers and COPD groups.

Verification of expression of dysregulated miRNAs in different cell types of PBMCs
PBMCs consist mainly of T lymphocytes, B lymphocytes and monocytes. We therefore analyzed the expression of dysregulated miRNAs in the isolated different cell types including CD4 + T cells, CD8 + T cells, CD20 + B cells and CD14 + monocytes from PBMCs of smokers and COPD patients. MiR-24-3p was consistently expressed higher in monocytes. In smokers, miR-93-5p was mainly expressed  The miRNAs with difference of fluorescence intensity higher than 1000 was selected in monocytes, while it was equally expressed in CD4 + T cells, CD8 + T cells and monocytes in COPD patients. CD8 + T cells mainly contribute to the increased expression of miR-320a and miR-320b in COPD patients. In COPD patients, more miR-1273 g-3p was expressed in monocytes, suggesting that the expression was decreased in CD4 + T cells, CD8 + T cells and CD20 + B cells in COPD (Fig. 3).

mRNAs are dysregulated in PBMCs of COPD patients
We performed a parallel mRNA microarray study to compare the mRNA expression between the smokers and COPD patients. The results revealed a total of 1508 genes that differed in expression between the two groups (fold change >2; Fig. 4), where 164 mRNAs were upregulated and 137 were down-regulated when filtered by (See figure on previous page.) Fig. 2 Validation of differentially expressed miRNAs. a Expression of selected miRNAs in PBMCs of smokers and COPD patients. qRT-PCR was performed on the same RNA samples (17smokers and 14 COPD patients) as microarray analysis. Data are presented as 2 (−σσCt) relative to U6. * P < 0.05, ** P < 0.01 compared with smokers by Mann Whitney U test. b Relative abundance of differentially expressed miRNAs in PBMCs of smokers and COPD patients. * P < 0.05, ** P < 0.01 compared with smokers. c Correlation analysis between miRNA expression and FEV1% predicted Fig. 3 Expression of miRNAs in the isolated different cell types of PBMCs from smokers (a) and COPD patients (b). The expression of miR-24-3p, miR-93-5p, miR-320a, miR-320b and miR-1273g-3p was examined by qRT-PCR on CD4 + T lymphocytes, CD8 + T lymphocytes, CD20 + B lymphocytes and CD14 + monocytes from smokers and COPD patients. Data are presented as 2 (−σσCt) relative to β-actin fluorescence intensity difference higher than 1000. Table 4 shows the 20 genes with the largest fold changes and their known biological functions.

Regulation network of dysregulated miRNAs and mRNAs
Based on the genes negatively correlated with miRNAs from microarray, the predicted regulatory network between the dysregulated miRNAs (that were validated) and mRNAs in COPD patients was then analyzed (Fig. 5, Table 5). The KEGG pathway enrichment analysis indicated that the NOD − like receptor (NLR) and Toll − like receptor (TLR) signaling pathway relevant to the pathogenesis of COPD were significantly enriched (Fig. 6). The involved genes include IL18/IL1B/TNF predicted to be regulated by miR-24-3p, and NLRP3/IL6/NFKBIA by miR-93-5p for NLR pathway. The genes CCL3/CCL4/ IL1B/TNF were predicted to be regulated by miR-24-3p, and IL6/CXCL10/NFKBIA by miR-93-5p for TLR pathway. The expression level of relevant predicted target genes of individual subjects were further validated by qRT-PCR. As shown in Fig. 7, the expression of IL18, IL1B, TNF, NFKBIA, CCL3 and CCL4 was validated, but not for NLRP3, IL6 and CXLC10.

Discussion
MicroRNAs play important regulatory roles in cell differentiation, cell cycle and apoptosis. Due to the role of multiple gene regulation, miRNAs have received much attention as biomarkers and target for novel therapeutics. In COPD therefore, the role of miRNAs in disease pathogenesis is an attractive area of research. For the first time, we conducted a comprehensive analysis of both miRNA and mRNA expression in PBMCs from subjects with COPD and compared their expression profiles to smokers without airflow limitation. We identified 137 differentially-expressed miRNAs in PBMCs from COPD subjects compared with smokers without COPD. Among the selected 11 miRNAs, the dysregulated expression of 5 miRNAs including miR-24-3p, miR-93-5p, miR-320a, miR-320b and miR-1273g-3p were validated by qPCR.
Of the miRNA investigated in this study, miR-24-3p is of considerable interest. It has been reported that miR-24-3p was consistently upregulated during terminal differentiation of hematopoietic cells into a variety of lineages as well as during muscle and neuronal cell differentiation [20,21]. MiR-24-3p might also function in cell proliferation [22,23]. Upregulation of miR-24 is associated with a decreased DNA damage response upon etoposide treatment in highly differentiated CD8 + T cells [24], and miR-24 is a negative regulator of classical macrophage activation by LPS [25]. In this study, we found increased expression of miR-24-3p mainly in the T lymphocytes and monocytes, which might in part contribute to the increased number of CD8 + T cells in the lung [26] and to the impairment of host defenses in the lower respiratory tract due to the smoke related changes in the phenotype of alveolar macrophages of COPD patients [27].
As for the rest of dysregulated miRNAs reported in this study, we did not find the relevant evidences involving the pathogenesis of COPD, which implies the discrepancy of the miRNA profile between PBMCs and lung tissues. These miRNAs were mainly found dysregulated in cancers and other disorders like autoimmune diseases. For example, miR-93-5p was identified as a potential biomarker of various types of cancer such as acute myeloid leukemia [28] and laryngeal squamous cell carcinoma [29]. This could be due to the link between miR-93 and promotion of tumor growth, angiogenesis and metastasis [30,31]. miR-93 is up-regulated in PBMCs from adult T-cell leukemia patients, and suppresses the expression of a tumor suppressor protein, tumor protein 53-induced nuclear protein 1 (TP53INP1) [32]. Four different transcripts have been reported in the database for miR-320 (miR-320a, b, c, and d) [33]. The seed region considered crucial for target binding remains the same for miR-320a, b, c, and d. The plasma level of miR-320a was found increased in patients with systemic lupus erythematosus (SLE) [34], and miR-320-3p is increased in the plasma of non-small cell lung cancer (NSCLC) patients [35]. The expression of miR-1273g-3p was found remarkably changed in Human Umbilical Vein Endothelial Cells (HUVECs) under acute glucose fluctuations, which was demonstrated to contribute to endothelial dysfunction and autophagy [36]. MiR-1273 expression is also increased in the pancreas of mouse model of pancreatic cancer [37].
By analyzing the regulation network between the dysregulated miRNA and mRNA, we predicted the negative regulatory role of miRNAs on a total of 36 over expressed and 61 under expressed mRNAs. The KEGG pathway enrichment analysis indicated that the NOD − like receptor (NLR) signaling pathway and Toll − like receptor (TLR) signaling pathway are the top 2 pathways likely involved in the pathogenesis of COPD, with mir-24-3p and miR-93-5p being predicted to regulate the relevant genes in both pathways. As down-stream factors of the NLR and TLR pathway, the mRNA levels of proinflammatory mediators including IL-18, IL-1β, CCL3, CCL4, and TNF were found down-regulated in PBMCs of COPD patients in this study. To the best of our knowledge, the down-regulation of these mRNAs in PBMCs of COPD patients has not been previously reported. In fact, a previous study performed on human peripheral lung tissue obtained from non-smokers, smokers and COPD patients revealed the similar trends of expression level of pro-inflammatory mediators, where the level of IL-8, IL-6, IL-1β and TNF-α showed the decreased trend in COPD patients compared with smokers [38]. Instead, most previous studies showed elevated expression levels of these mediators in serum or lung [39]. We suspect that certain changes in the cytokine expression profile may happen when the peripheral immune cells infiltrate the local inflammatory sites in the lung. The reduced expression of TLR2 has been found in the alveolar macrophages of smokers and COPD patients, which was associated with the impairment of host defenses in the lower respiratory tract [27]. Furthermore, decreased cytokine and chemokine mRNA expression in bronchoalveolar lavage cells from asymptomatic smokers has been reported [40].
In addition, NFKBIA predicted to be regulated by miR-93-5p-was the other down-regulated gene involved in enriched pathways. In unstimulated cells, NF-κB is found in the cytoplasm in an inactive non-DNA binding form, associated with its inhibitory protein κBα (IκBα, coded by NFKBIA gene), IκBα degradation unmasks the nuclear localization signal present in NF-κB, allowing it to enter the nucleus, bind DNA, and initiate gene transcription [41]. In the present study, the down-regulation of IκBα supposedly trigger the activation of NF-κB, whose expression was also higher in PBMCs of COPD patients. The coexistence of decreased level of IκBα and pro-inflammatory mediators in COPD patients was also reported on lung tissue [38]. Besides trigging the expression of pro-inflammatory mediators, NF-κB activation may also be related with the disordered apoptosis of Tcell hybridoma cell line [42]. Thus, the decreased expression of IκBα may contribute to the dysregulated apoptosis of T cells in COPD [43].
Overall, through miRNAs and mRNAs expression profiling in smokers and COPD patients, we identified the dysregulated miRNAs and mRNAs in PBMCs from COPD patients. We further analyzed the regulation network between miRNA and mRNA, where NLR and TLR was the most enriched pathways. Among them the regulation of IL-18, IL-1β, TNF, CCL3 and CCL4 by miR-24-3p, and IκBα by miR-93-5p may provide the clue for potential investigations.

Conclusions
The expression of miRNA and mRNA were dysregulated in PBMCs of COPD patients compared with smokers without airflow limitation. The regulation network between the dysregulated miRNA and mRNA may provide potential therapeutic targets for COPD. Fig. 5 Regulation network between miRNAs and mRNAs. The negatively regulation of miRNA on dysregulated mRNAs was predicted and the regulation network was drawn by using Cytoscape software. Red and green color represents up-and down-regulated genes, respectively  6 Pathway enrichment of dysregulated mRNAs regulated by miRNAs. The pathways in dysregulated mRNAs predicted by miRNAs were enriched by KEGG pathway enrichment analysis. "p adjust" represents the P value range. "Gene Ratio" represents the ratio of predicted target gene number in total gene number of each relevant pathway Fig. 7 Validation of predicted target genes of miRNAs. The expression of predicted target genes of miRNAs in PBMCs of smokers and COPD patients was examined by qRT-PCR. Data are presented as 2 (−σσCt) relative to β-actin. * P < 0.05, ** P < 0.01, *** P < 0.001 compared with smokers by Mann Whitney U test.