key: cord-0005372-potmzdfs
authors: Sun, Dong; Wan, Xin; Pan, Bin-bin; Sun, Qing; Ji, Xiao-bing; Zhang, Feng; Zhang, Hao; Cao, Chang-chun
title: Bioinformatics Analysis of Genes and Pathways of CD11b(+)/Ly6C(intermediate) Macrophages after Renal Ischemia-Reperfusion Injury
date: 2018-03-15
journal: Curr Med Sci
DOI: 10.1007/s11596-018-1848-7
sha: e8c6b43bd13717e442f126f271f19c9a887c0578
doc_id: 5372
cord_uid: potmzdfs

Renal ischemia-reperfusion injury (IRI) is a major cause of acute kidney injury (AKI), which could induce the poor prognosis. The purpose of this study was to characterize the molecular mechanism of the functional changes of CDllb+/Ly6C(intermediate) macrophages after renal IRI. The gene expression profiles of CDllb+/Ly6Cintermcdiate macrophages of the sham surgery mice, and the mice 4 h, 24 h and 9 days after renal IRI were downloaded from the Gene Expression Omnibus database. Analysis of mRNA expression profiles was conducted to identify differentially expressed genes (DEGs), biological processes and pathways by the series test of cluster. Protein-protein interaction network was constructed and analysed to discover the key genes. A total of 6738 DEGs were identified and assigned to 20 model profiles. DEGs in profile 13 were one of the predominant expression profiles, which are involved in immune cell chemotaxis and proliferation. Signet analysis showed that Atp5al, Atp5o, Cox4i, Cdc42, Rac2 and Nhp2 were the key genes involved in oxidation-reduction, apoptosis, migration, M1-M2 differentiation, and proliferation of macrophages. RPS18 may be an appreciate reference gene as it was stable in macrophages. The identified DEGs and their enriched pathways investigate factors that may participate in the functional changes of CD 1lb(+)Ly6C(intermediate) macrophages after renal IRI. Moreover, the vital gene Nhp2 may involve the polarization of macrophages, which may be a new target to affect the process of AKI

bule injury. A subsequent switch to M2 macrophages can suppress the inflammatory response and induce a proliferative repair phase [11] . M2 macrophages are involved in producing extracellular matrix components, but may also contribute to the tissue fibrosis, should this process become dysregulated [12] .

The marker of macrophages may be diverse in different organs, therefore defining macrophages according to their function has become challenging [13] . According to the whole genome microarray analysis data, the CD11b + /Ly6C high population was associated with the onset of renal injury and produced proinflammatory cytokines. In addition, the CD11b + / Ly6C int population demonstrated a wound healing phenotype [14] . Dragomir et al reported that CD11b + / Ly6C lo cells were larger than CD11b + /Ly6C hi cells, and more irregularly shaped. Moreover, CD11b + /Ly6C lo cells contained a highly vacuolated cytoplasm and an increased cytoplasm:nuclear ratio. RT-PCR analysis revealed that mRNA expression levels related to proinflammatory proteins (for example, TNF-α, iNOS, and the chemokine receptor CCR2) were significantly higher in CD11b + /Ly6C hi cells than in CD11b + / Ly6C lo cells. In contrast, the mRNA expression level of anti-inflammatory cytokine IL-10 was reduced in CD11b + /Ly6C hi cells when compared to CD11b + / Ly6C lo cells [15] . Clements et al investigated the genes that were uniquely expressed in each population. It is necessary to consider of genes that were regulated in each phenotype over time [14] .

The purpose of our study was to evaluate the proportion of macrophages that changed at different time points after IRI. We used microarray analysis to identify the differentially expressed genes (DEGs) in CD11b + /Ly6C int macrophages of C57BL/6 mice and mice undergoing sham surgery or IRI for 4 h, 24 h or 9 days. We used The Series Test of Cluster (STC) analysis, STC-Gene Ontology analysis and pathway analysis to identify changes in function and pathways in macrophages in the different groups. The protein-to-protein interaction (PPI) network was applied to select key genes by degree, which may help explain how macrophages were influenced at different time points. Our study may provide further insight into a new target that affects the process of AKI by changing the macrophage function.

Gene expression analysis was performed on an Affymetrix Mouse Genome 430 2.0 Array platform (Affymetrix, USA) for which the transcription profile GSE75808 from the Gene Expression Omnibus (GEO) database, an open-access functional genomics data repository was downloaded. Twenty-five male C57BL/6 mice (8 to 10 week-old) were divided into several groups and underwent either bilateral renal IRI for 28 min or sham surgery followed by reperfusion. For RNA isolation and amplification for cDNA production, macrophage populations were sorted on a BD FACSARIA II cell sorter (BD Bioscience,usA) based on CD11b + /Ly6C high , CD11b + /Ly6C int or CD11b + / Ly6C low as previously described [14] . We analysed the expression of CD11b + /Ly6C int macrophages of the sham (GSM1968232 and GSM1968235), 4-h IRI (GSM1968226, GSM1968228 and GSM1968230), 24 h IRI (GSM1968216, GSM1968217 and GSM1968218) and 9 day IRI (GSM1968219, GSM1968221 and GSM1968223) groups to identify genes that are associated with this subset of macrophages.

The probe-level data were converted into expression measures so that by taking the average expression value, the expression values of all probes for a given gene in each sample were reduced to a single value [16] . Next, the principal component analysis (PCA) was performed and DEGs were identified as previously described [17] . In our study, we used the effective statistical method for small samples to identify differentially expressed mRNAs among the four groups of macrophages by GCBI (https://www.gcbi.com.cn/gclib/html/ index). These values include false discovery rate-adjusted P values and were considered significant when P<0.05.

To validate the most probable set of clusters of four-time series, we used the STC algorithm of gene expression dynamics to profile the gene expression time series as previously described [18] . The STC algorithm identified which profiles have a significant number of genes assigned and the result may indicate the change rule of samples at different time points.

To identify functional changes in macrophages, we analysed the role of DEGs in each significant expression profile. Analyses were based on the Gene Ontology (GO) database in Funrich 3.0 software as previously described [19] . The DEGs of each significant expression profile was classified into a group of biological process categories from GO annotation. GO terms were considered significant at P<0.05.

We identified the significant pathways that were changed in CD11b + /Ly6C int macrophages. To identify the main biological function of CD11b + /Ly6C int macrophages, pathways of genes with similar expression trend were analysed. Analyses were based on the Reactome database in Funrich. The threshold of significance was considered as P<0.05.

We created a PPI network to analyse the key genes which regulate other genes. Based on the 927 DEGs in profile 13, PPI was analysed by STRING 10.5 and the cytoHubba app of Cytoscape software (version 3.5.1) as previously described [20] . The combined score of the PPI value was>0.4. Genes that showed a high degree were identified as key genes as previously described [21] . CytoHubba is an approach that has frequently been used to select hub genes, and provides 11 topological analysis methods, including Degree, Maximum Neighborhood Component, Edge Percolated Component, Maximal Clique Centrality, Density of Maximum Neighborhood Component, and six centralities (Bottleneck, EcCentricity, Closeness, Betweenness, Radiality, and Stress) based on the shortest paths [22] . Genes that appeared in the top 50 genes by more than 6 ways in CytoHubba were identified. The top 50 hub forming genes were output by each of the ranking methods as a measure of significance. Genes were considered significant when they were identified by both methods. We especially focused on the genes of the top 20 degrees in the PPI network as previously described [23] . 

In this study, we analysed DEGs from CD11b + / Ly6C int macrophages, which were isolated from kidneys of mice undergoing sham surgery (n=2), and IRI at 4 h, 24 h, and 9 days (n=3 per group). Figure 1A and 1B present the data before and after normalization. In fig. 1C , the PCA score plots of the four groups are shown. A total of 6738 normalized DEGs were identified (P<0.05 and q<0.05) at different time points.

To identify target genes among the 6738 genes, twenty expression profiles were evaluated by cluster analysis. Each profile contained genes with a similar expression pattern after IRI. Among the 20 profiles of genes, nine profiles (profile 13, 5, 10, 2, 4, 7, 11, 1 and 3) were significantly different (P<0.05) ( fig. 2 and table 1 ). 

As shown in table 2, GO terms that were downregulated after renal IRI included cell cycle, inflammatory response, apoptosis, oxidation-reduction process, autophagy, and cell proliferation. Profile 13 consisted of genes that were stable at early time points and then rapidly decreased at later time points. To get insights into the biological effects of the genes in profile 13, we analysed the involved GO terms using the GO annotation in Funrich 3.0 (the threshold of GO terms was P<0.05). Biological processes such as transport and apoptosis showed the most notable enrichment of the target genes. Moreover, chemotaxis of several immune cells was also decreased at day 9 after IRI. Among the cellular component, extracellular vesicular exosome and mitochondrion showed a maximum enrichment, whereas protein binding and RNA binding showed the highest enrichment of molecular function ( fig. 3 ). 

We analysed the pathway enrichment using Reactome database for DEGs of profile 13. P<0.05 was considered for pathway analysis. The highest enrichment of pathway was demonstrated for neutrophil degranulation (P=1.01e-41) and respiratory electron transport (P=1.47e-17) ( fig. 4 ). 

STRING analysis was used to obtain the PPI of the 927 DEGs in profile 13. The minimum required interaction score was 0.4. We gained a total of 6903 edges and 902 nodes, accounting for 97.30% of all DEGs. Results demonstrated that Gapdh (degree: 138), Hsp90aa1 (degree: 110), Actb (degree: 90), Actg1 (degree: 81), Atp5a1 (degree: 74), Atp5o (degree: 69), Cdc42 (degree: 68), Pcna (degree: 66), Uqcr11 (degree: 65) and Cox4i1 (degree: 60), which all have a high degree, were identified as hub genes in the PPI network ( fig. 5) . Moreover, Rhoa, Rac2, Nhp2, Rplp0, Rpl5, Rpl7, Ppp2ca, Mrpl2, and Rps6 were among the top twenty degrees in the PPI network. In addition to STRING analysis, key genes were also identified by CytoHubba. Table 3 presents genes that were considered to be the top 50 key genes in more than 6 ways of the 11 topological analysis methods described above. Because the changes of the genes with either a high degree or selected by Cytohubba could regulate the expression of multiple genes, they were identified as key genes for further study.

As AKI could result in high mortality, it is important to take back to homeostasis by a suitable way [4] . The CD11b + /Ly6C int population is identified as wound healing population, which carries membrane receptors and chemokines associated with inflammation at the same time [14] . Understanding how the population changes may provide a new way to treat AKI by modulating the immune system.

In our study, we identified nine gene profiles, most of which were downregulated after IRI. The proportion of CD11b + /Ly6C int macrophages was at a maximum one day after IRI and was markedly reduced at day 9 after IRI. Taking this into consideration, the genes in the profile 13 were involved in infiltration and proliferation. GO analysis of profile 13 showed that the genes related to the extracellular vesicular exosome were significantly changed, which was consistent with the finding that CD11b + /Ly6C int macrophage could secrete anti-inflammation cytokines [15] . GO and pathway analysis indicated that immune cell chemotaxis (for example, neutrophil, leukocyte, lymphocyte, monocyte, and macrophage) and G1/S progression were downregulated at later time points. In mitochondria, the oxidation-reduction process and respiratory electron transport were significantly involved after IRI as At-p5a1, Atp5o, and Cox4i1 are also key genes in the PPI network. It would be required to further assess the vitality, switch and infiltration ability of CD11b + /Ly6C int macrophages, given that Cdc42 and Rac2 were found to have an effect on apoptosis, migration, and M1-M2 differentiation [24, 25] . Protein phosphatase 2A (Pp2a) is a bona fide tumor suppressor gene that is involved in mitosis and apoptosis [26] . Homologues Ppp2ca (Pp2ac) and Ppp2cb are the catalytic subunits of Pp2a [27] . Previous studies have shown that upregulation of Ppp2ca in systemic lupus erythematosus decreased IL-2 [28] , resulting in the generation of effector and memory T cells and the maintenance of regulatory T cells [29] . Moreover, it was suggested that reduction of Ppp2ca may play a role in anti-inflammatory progression by increasing osteoprotegerin (OPG) expression and decreasing receptor activator of nuclear factor κB ligand (RANKL) expression [30] . In prostate cancer cells, loss of Ppp2ca facilitated epithelial-to-mesenchymal transition [31] . Taken together, the reduction of Ppp2ca in profile 13 may be a reason why CD11b + /Ly6C int macrophages decreased inflammation.

Rpl5 can induce the cell cycle, and the downregulation of Rpl5 at day 9 after IRI may provide an explanation of the reduction of CD11b + /Ly6C int macrophages. Moreover, the depletion of Rpl5 strongly suppresses cell cycle progression in primary human lung fibroblasts [32] . Rpl5 and Rpl11 promote apoptosis and reduce cellular proliferation in tumor in vitro [33] . However, it has also been reported that a heterozygous deletion or mutated Rpl5 occurred in 11% of glioblastoma, 28% of melanoma and 34% of breast cancer cases [34] .

Nhp2, one of the key nodes in PPI, was significantly decreased when 1-day and 9-day CD11b + /Ly-6C int macrophages were compared. Nhp2 is a part of H/ACA ribonucleoprotein particles (RNPs), containing four common proteins, including the pseudouridine synthase Cbf5, Nhp2, Nop10, Gar1a and a substrate-specific H/ACA RNA. H/ACA is considered to play a role in the biogenesis of spliceosomal small nuclear RNA (snRNA) and ribosomal RNA (rRNA) [35] . Depletion of Nhp2 reduced all H/ACA snoRNAs and impaired global rRNA pseudouridylation, which play a role in the uridine selection or isomerization processes as they are required for the synthesis and stability of particles [36, 37] .

Nhp2 is highly expressed in spleen, thymus, small intestine, testis, ovary, prostate, colon (mucosal lining), skeletal muscle, kidney, heart, pancreas, placenta, and brain, whereas the expression levels in the liver are low [38] . Nhp2 mRNA was barely detectable in peripheral blood leukocytes and lung. During the differentiation of U937 cells into monocytes and macrophages induced by 12-O-tetradecanoylphorbol-13-acetate (TPA), the expression of Nhp2 mRNA was markedly decreased, but was upregulated during the retro-differentiation [38] . Nhp2, hTR, hTERT and Nop10 could control telomere homeostasis, which is important for apoptosis and cell-cycle arrest [39, 40] . Thus, Nhp2 may affect the polarization of macrophage.

Finally, it is important to identify which genes could serve as reference genes. We analysed the expression levels of Gapdh, Actb, Β2m, Hmbs, Hprt, Rplp0, Tbp, Gusb, Ppia, Oaz1, Nono, Tfrc, Eef2, Hs-b90ab1, Rps18, Sdha, Ywhaz, Ubc, Rps17, Rplp0, Rp-l37a, Pum1, Psmc4, Pop4, Pgk1, Pes1, Mrpl9, Ipo80, Gadd45a, Elf1, Eif2b1, Cdkn1b, Cdkn1a, Casc3, Abl1, and Pol2a which are widely used [41] [42] [43] . The data showed that RPS18 may be an appropriate reference gene as its expression was stable in macrophages after renal IRI.

In our study, a total of 6738 DEGs were found. We analysed the genes that were stable at early time, but were abruptly downregulated at late time, investigated the function and molecular pathways that they were involved, and studied the critical genes involved. A possible explanation why CD11b + /Ly6C int macrophages were reduced on day 9 may be due to the fact that genes involved in the chemotaxis and proliferation were decreased. On the other hand, one of the key genes, Nhp2, may be involved in the polarization of macrophages. Taken together, our study will increase insight into the functional changes of macrophages; therefore, identifying the critical genes involved may provide novel targets for regulating the quantity and phenotype of macrophages.

Incidence, outcomes, and risk factors of community-acquired and hospital-acquired acute kidney injury: A Retrospective Cohort Study

Identifying acute kidney injury in the community--a novel informatics approach

Netrin-1 regulates the inflammatory response of neutrophils and macrophages, and suppresses ischemic acute kidney injury by inhibiting COX-2-mediated PGE2 production

Long-term outcome of severe acute kidney injury survivors followed by nephrologists in a developing country

Recognition and management of acute kidney injury in the International Society of Nephrology 0by25 Global Snapshot: a multinational cross-sectional study

Small interfering RNA targeting TNF-alpha gene significantly attenuates renal ischemia-reperfusion injury in mice

Double-negative alphabeta T cells are early responders to AKI and are found in human kidney

Vascular-resident CD169-positive monocytes and macrophages control neutrophil accumulation in the kidney with ischemia-reperfusion injury

MicroRNA 26a modulates regulatory T cells expansion and attenuates renal ischemia-reperfusion injury

GM-CSF promotes macrophage alternative activation after renal ischemia/reperfusion injury

Distinct macrophage phenotypes contribute to kidney injury and repair

Macrophage-mediated injury and repair after ischemic kidney injury

Renal F4/80+ CD11c+ mononuclear phagocytes display phenotypic and functional characteristics of macrophages in health and in adriamycin nephropathy

Differential Ly6C expression after renal ischemia-reperfusion identifies unique macrophage populations

Role of galectin-3 in classical and alternative macrophage activation in the liver following acetaminophen intoxication

Improved normalization of systematic biases affecting ion current measurements in label-free proteomics data

Time-series analysis in imatinib-resistant chronic myeloid leukemia K562-cells under different drug treatments

Differential gene expression profiling analysis in workers occupationally exposed to benzene

FunRich: An open access standalone functional enrichment and interaction network analysis tool

Proteomic Differences between Developmental Stages of Toxoplasma gondii Revealed by iTRAQ-Based Quantitative

Transcriptome analysis of chicken kidney tissues following coronavirus avian infectious bronchitis virus infection

cytoHubba: identifying hub objects and sub-networks from complex interactome

Effects of beta-catenin on differentially expressed genes in multiple myeloma

MIF inhibits monocytic movement through a non-canonical receptor and disruption of temporal Rho GTPase activities in U-937 cells

Rac2 controls tumor growth, metastasis and M1-M2 macrophage differentiation in vivo

SV40 small T antigen and PP2A phosphatase in cell transformation

Ppp2ca knockout in mice spermatogenesis

Protein phosphatase 2A is a negative regulator of IL-2 production in patients with systemic lupus erythematosus

Cutting edge: mechanisms of IL-2-dependent maintenance of functional regulatory T cells

Protein phosphatase 2A Calpha is involved in osteoclastogenesis by regulating RANKL and OPG expression in osteoblasts

Restoration of PPP2CA expression reverses epithelial-to-mesenchymal transition and suppresses prostate tumour growth and metastasis in an orthotopic mouse model

Loss of tumor suppressor RPL5/RPL11 does not induce cell cycle arrest but impedes proliferation due to reduced ribosome content and translation capacity

Ribosomal proteins L11 and L5 activate TAp73 by overcoming MDM2 inhibition

The ribosomal protein gene RPL5 is a haploinsufficient tumor suppressor in multiple cancer types

ACA small nucleolar RNA pseudouridylation pockets bind substrate RNA to form three-way junctions that position the target U for modification

Cb-f5p, a potential pseudouridine synthase, and Nhp2p, a putative RNA-binding protein, are present together with Gar1p in all H BOX/ACA-motif snoRNPs and constitute a common bipartite structure

Nhp2p and Nop10p are essential for the function of H/ACA snoRNPs

Expression of the human homologue of the small nucleolar RNA-binding protein NHP2 gene during monocytic differentiation of U937 cells

Structure of the Shq1-Cbf5-Nop10-Gar1 complex and implications for H/ACA RNP biogenesis and dyskeratosis congenita

Telomere maintenance mechanisms and cellular immortalization

Identification of optimal reference genes for RT-qPCR in the rat hypothalamus and intestine for the study of obesity

Appropriateness of reference genes for normalizing messenger RNA in mouse 2,4-dinitrobenzene sulfonic acid (DNBS)-induced colitis using quantitative real time PCR

Identification of optimal reference genes for normalization of RT-qPCR data in cancerous and non-cancerous tissues of human uterine cervix

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.