SEARCH

SEARCH BY CITATION

Keywords:

  • bioinformatics;
  • hepatic stellate cells;
  • liver fibrosis;
  • microarray;
  • pathway

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Activation of hepatic stellate cells (HSCs), which is regulated by multiple signal transduction pathways, is the key event in liver fibrosis. Moreover, members of these pathways are important targets for microRNAs (miRNAs). To better understand the critical pathways of HSC activation, we performed comprehensive comparative bioinformatics analysis of microarrays of quiescent and activated HSCs. Changes in miRNAs associated with HSC activation status revealed that 13 pathways were upregulated and 22 pathways were downregulated by miRNA. Furthermore, mitochondrial integrity, based on highly upregulated Bcl-2 and downregulated caspase-9, was confirmed in HSCs and fibrotic livers by immnofluorescence assay, quantitative RT-PCR, and western blot analysis. These findings provide in vitro and in vivo evidence that the mitochondrial pathway of apoptosis plays a significant role in the progression of liver fibrogenesis via HSC activation.


Abbreviations
FDR

false discovery rate

GO

gene ontology

HE

hematoxylin/eosin

HSC

hepatic stellate cell

KEGG

Kyoto Encyclopedia of Genes and Genomes

KO

Kyoto Encyclopedia of Genes and Genomes orthology

MAPK

mitogen-activated protein kinase

miRNA

microRNA

SMA

smooth muscle actin

TGF-β

transforming growth factor-β

VEGF

vascular endothelial growth factor

VG

Van Gieson

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Liver fibrosis is the excessive accumulation of extracellular matrix that occurs in most types of chronic liver diseases. Hepatic stellate cells (HSCs), the major mesenchymal cells in liver, are widely accepted as playing a critically important role in liver fibrosis [1]. In the quiescent state, HSCs are lipid-storing cells located in the perisinusoidal endothelium. In contrast, they undergo myofibroblastic transdifferentiation, also known as activation, when stimulated by fibrogenic stimuli, which reflects the critical step of liver fibrogenesis [2].

Previous studies on HSCs have attributed their activation to the regulation of many signal transduction pathways, including transforming growth factor-β (TGF-β)/Smad, platelet-derived growth factor, mitochondrial pathway of apoptosis, mitogen-activated protein kinase (MAPK), Wnt, and vascular endothelial growth factor (VEGF) [3–8]. Furthermore, microRNA (miRNA)-mediated RNA interference has been identified as a novel mechanism that regulates protein expression at the translational level [9]. The differentially expressed miRNAs and their inhibitory effect on gene expression, especially those relevant to signal transduction, add a new level to our knowledge about the regulatory mechanisms of HSC activation [10]. However, the miRNAs often negatively modulate gene expression at the post-transcriptional level by incomplete binding to target sequences within the 3′-UTR, and generally do not affect mRNA levels [9]. Thus, high-throughput gene expression analysis by microarray is not suitable for exploring the target signaling pathways of miRNA during activation.

Fortunately, significant progress in data mining has provided a wide range of bioinformatics analysis options, such as gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) orthology (KO), to aid researchers in the interpretation of their data [11]. According to these techniques, KEGG acts as a bioinformatics resource for understanding higher-order functional meanings and utilities of the cell or the organism from its genome information. Integration of current knowledge on molecular interaction networks, such as signaling pathways and complexes (PATHWAY database), features the reference knowledge base (http://www.genome.ad.jp/kegg/) [12]. We therefore performed a global analysis of miRNA-regulated signaling pathways and related genes on the basis of miRNA expression profile and bioinformatic interpretation. The predicted signaling pathways, some of which had not been previously described in the activation of HSCs, were further selected and validated within both activated HSCs and fibrotic liver of rats.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Bioinformatics interpretation revealed the GOs and signaling pathways regulated by miRNAs

The purity of quiescent (> 95%) and activated (> 95%) HSCs was confirmed immunofluorescently, using desmin (Fig. 1). Microarray hybridization preliminarily identified 21 miRNAs as being differentially expressed during HSC activation. A volcano plot provided further information about the significance and magnitude of expressive alteration of selected miRNAs (Fig. 2), which was helpful in judging the most significant candidates for follow-up studies.

image

Figure 1.  Characterization of HSCs isolated from rat liver (× 400). After isolation, the cells were cultured for 2 days (quiescent HSCs) or for 14 days (activated HSCs). (A) Immunofluorescence analysis of desmin expression in quiescent HSCs. (C) Desmin expression in activated HSCs. (E, K) Negative controls without primary antibody were performed in activated HSCs. (G, I) Negative controls with antibody against α-SMA (1 : 100; Santa Cruz, USA) were performed in activated HSCs. Hoechst 33258 nuclear staining for all conditions is shown in (B), (D), (F), (H), (J), and (L).

Download figure to PowerPoint

image

Figure 2.  The volcano plot shows the upregulated and downregulated miRNAs in activated HSCs. The horizontal axis represents the fold change between quiescent and activated HSCs. The vertical axis represents the P-value of the t-test for the differences between samples.

Download figure to PowerPoint

david gene annotation was used to interpret the biological effect of miRNAs filtered by volcano plot. According to the results of data mining, 19 upregulated GOs and 24 downregulated GOs were classified on the basis of the top 25% miRNA targets (Table 1). Additionally, miRNA–mRNA network analysis integrated these miRNAs and GOs by outlining the interactions of miRNA and GO-related genes (Fig. 3).

Table 1.   MicroRNA targets significant GO in HSCs.
Category IDCategory nameEnrichmentP-valueFDR
  1. a GOs targeted by upregulated miRNA. b GOs targeted by downregulated miRNA. All of these GOs show increased enrichment, P-values and FDRs.

Upregulated GOs by GO analysisa
 GO:0006813Potassium ion transport0.5614510662.26084E-050.002214733
 GO:0045817Positive regulation of transcription from RNA polymerase II promoter0.5329297132.15258E-100.000938004
 GO:0008283Cell proliferation0.5222098637.55331E-060.001902064
 GO:0001666Response to hypoxia0.5222098635.13583E-060.00181087
 GO:0045449Regulation of transcription0.5056317721.207E-120.000690475
 GO:0008284Positive regulation of cell proliferation0.4612411992.13311E-080.001107366
 GO:0042981Regulation of apoptosis0.3879273272.85018E-060.001719675
 GO:0007242Intracellular signaling cascade0.38254908600.000117251
 GO:0006350Transcription0.37941810300.000260557
 GO:0007049Cell cycle0.3698986531E-150.000508086
 GO:0007264Small GTPase-mediated signal transduction0.3686187273.2E-140.000586253
 GO:0007165Signal transduction0.36680692800.000247529
 GO:0008152Metabolic process0.35277615400.000143306
 GO:0006916Antiapoptosis0.3400436321.65703E-080.001081311
 GO:0045941Positive regulation of transcription0.3213599162.97119E-080.001159478
 GO:0007169Transmembrane receptor protein tyrosine kinase signaling pathway0.2534253752.3708E-110.00078167
 GO:0051056Regulation of small GTPase-mediated signal transduction0.1660004861E-150.000521114
 GO:0007186G-protein coupled receptor protein signaling pathway0.12551183009.11949E-05
 GO:0045840Positive regulation of mitosis0.1119021131.57057E-070.001315812
Downregulated GOs by GO analysisb
 GO:0006915Apoptosis−0.5088366566.68083E-100.000951032
 GO:0006629Lipid metabolic process−0.4888822771.19397E-060.001537285
 GO:0006281DNA repair−0.4747765964.28892E-060.001784814
 GO:0006886Intracellular protein transport−0.4705818718.61363E-060.001954176
 GO:0005975Carbohydrate metabolic process−0.4656021696.25054E-060.001849953
 GO:0006457Protein folding−0.4656021693.93087E-080.001172505
 GO:0006813Potassium ion transport−0.4602194851.25103E-050.002058399
 GO:0008285Negative regulation of cell proliferation−0.4589507091.39364E-060.001563341
 GO:0006468Protein amino acid phosphorylation−0.45723447900.000403863
 GO:0007275Multicellular organism development−0.4466532436E-150.000547169
 GO:0015031Protein transport−0.4462524681.7828E-110.000742587
 GO:0006816Calcium ion transport−0.4247046811.52674E-050.002097482
 GO:0030154Cell differentiation−0.4216774364E-150.000534141
 GO:0006512Ubiquitin cycle−0.4207952652.9346E-080.001133422
 GO:0006118Electron transport−0.4202780643.4327E-110.000807726
 GO:0006810Transport?−0.41788117100.000495058
 GO:0006470Protein amino acid dephosphorylation−0.4160700231.79328E-050.002136566
 GO:0007156Homophilic cell adhesion−0.4038958571.91987E-060.001602424
 GO:0006350Transcription−0.40376438100.000469002
 GO:0007165Signal transduction−0.40069248300.000455974
 GO:0007283Spermatogenesis−0.4000903061.48658E-070.001302784
 GO:0006917Induction of apoptosis−0.3990875736.34031E-080.001211589
 GO:0007155Cell adhesion−0.38094722900.000325696
 GO:0006955Immune response−0.30729743100.000364779
image

Figure 3.  GO network. Blue box nodes represent miRNA, and red cycle nodes represent mRNA. Edges show the inhibitory effect of microRNA on mRNA. Upregulated and downregulated microRNA have separate, specific targets. The upper subgraph shows under-expression microRNA–mRNA network and the lower subgraph is the overexpression microRNA–mRNA network. Four overexpressed miRNAs (miR-140, miR-207, miR-325-5p and miR-874) showed the most target mRNAs of 7 (degree 7). In contrast, rno-miR-16 are the highest degree in under-expression miRNAs.

Download figure to PowerPoint

Another functional analysis of miRNAs by KEGG revealed that 13 signal transduction pathways were upregulated and 22 were downregulated (Fig. 4). Many of these signaling pathways, such as VEGF, MAPK, and biosynthesis of steroids, have been shown to participate in the activation of HSCs (Table 2). A wide variety of cellular processes, including cell proliferation, differentiation, and stress responses, also featured the functions of significant signaling pathways (Table 2). However, some other signaling pathways have never been reported to play a role in resting or activated HSCs, e.g. folate-dependent one-carbon pool and carbon fixation. Among all these differentially regulated signaling pathways, apoptosis appeared to be the most enriched one. A similar phenomenon was observed in GO analysis. In detail, Bcl-2 and caspase-9, the critical members of the mitochondrial apoptosis pathway (http://www.genome.ad.jp/kegg/pathway.html), served as the significant targets of miR15b/16, and miR-138, respectively. This represents novel evidence for the modulatory effect of miRNAs on HSC function via signaling pathway.

image

Figure 4.  Pathway analysis based on miRNA-targeted genes. (A) and (B) show significant pathways targeted by upregulated and downregulated miRNA, respectively. The vertical axis is the pathway category, and the horizontal axis is the enrichment of pathways.

Download figure to PowerPoint

Table 2.   Regulation of target gene significant pathways by miRNA. GnRH, gonadotropin-releasing hormone; PPAR, peroxisome proliferator-activated receptor.
PathwayP-valueFDREnrichmentTarget genes
  1. a Pathways targeted by upregulated miRNA. b Pathways targeted by downregulated miRNA. All of these pathways show increased enrichment, P-values, FDRs, and predicted targeted genes.

Regulated by upregulated miRNAa
 Folate-dependent one-carbon pool0.0006460.0054.47172298Atic, Ftcd, Gart, Mtfmt, Mthfd1, Shmt1, Tyms
 Carbon fixation0.0005080.0053.73708278Aldob, Aldoc, Fbp2, Got1, Gpt1, Mdh2, Pgk1, Pklr, Pkm2
 Biosynthesis of  steroids0.0041780.0053.16366796Ebp, Idi1, Lss, Mvd, Nqo1, Nsdhl, Tm7sf2, Vkorc1
 Glycolysis/ gluconeogenesis0.001150.0052.58366217Adh4, Adh7, Aldh1a1, Aldh1a3, Aldob, Aldoc, Eno1, Eno2, Fbp2, G6pc, Pfkm, Pgk1, Pklr, Pkm2
 VEGF signaling pathway0.0005040.0052.35504387Casp9, Hspb1, Kras, Map2k2, Mapk12, Mapkapk2, Mapkapk3, Pik3cb, Pla2g6, Plcg2, Ppp3cb, Ppp3r1, Ppp3r2, Prkcb1, Ptk2, Pxn, Rac1, Sphk1, Src
 Gap junction0.0002410.0052.21456757Adrb1, Csnk1d, Drd2, Egfr, Gja9, Gna11, Gnai2, Gucy1a2, Gucy2c, Htr2c, Kras, LOC500319, Map2k2, Map2k5, Npr1, Pdgfd, Pdgfra, Prkcb1, Prkg2, Src, Tuba6, Tubb2c, Tubb5, Tubb6
 Pyrimidine metabolism0.001790.0052.19828399Dck, Dhodh, Ecgf1, Entpd1, Entpd6, LOC682744, Nme3, Nme7, Pnpt1, Pola1, Pold1, Pold2, Polr3d, Prim2, Rpa1, Rrm2, Tyms, Umps
 Purine metabolism0.0001040.0052.0761571Allc, Atic, Dck, Ecgf1, Entpd1, Entpd2, Entpd6, Fnta, Gart, Gucy1a2, Gucy2c, LOC682744, Nme3, Nme7, Npr1, Pde10a, Pde1c, Pde2a, Pde3a, Pde3b, Pde4a, Pklr, Pkm2, Pnpt1, Pola1, Pold2, Polr3d, Prim2, Rpa1, Rrm2
 Apoptosis0.0080880.00751.90105951Aim1, Birc3, Cad, Casp6, Casp8, Casp9, Dffb, Fadd, Il1rap, Ntrk1, Pik3cb, Ppp3cb, Ppp3r1, Ppp3r2, Tnfrsf10b, Tnfrsf1a, Tnfsf10, Tp53, Tradd
 Neuroactive ligand–receptor interaction7.54E-070.0051.85470034Adcyap1r1, Adra1a, Adra1b, Adra2c, Adrb1, Agtrl1, Avpr2, Chrm3, Crhr1, Drd2, Edn2, Ednra, F2rl1, Fshb, Gabbr1, Gabrb1, Gal, Galr3, Gcg, Ghsr, Gip, Gnrhr, Gpr35, Gria2, Gria4, Grid1, Grik1, Grik3, Grin2a, Grin2d, Grm7, Hcrtr2, Htr1b, Htr2c, Kiss1r, Lep, Lgr8, Lhcgr, Ltb4r, Ltb4r2, Mtnr1a, Nbpwr1, Nmur1, Npffr1, Npffr2, Oprd1, Oxt, P2rx1, P2rx5, P2rx7, P2ry13, P2ry2, Ppyr1, Prl, Prlhr, Ptger2, Pthr1, Sct, Sstr2, Sstr3, Taar5, Tac2, Tac4, Trhr2, Trpv1, Tshr
 Glycan structures – biosynthesis 10.0046780.0051.82854203Alg6, Alg8, B4galt3, Chst1, Chst3, D1bwg1363e, Extl3, Galnt10, Galnt11, Galnt13, Gcnt3, Gcs1, H2afx, Hs3st1, Hs3st2, LOC683264, LOC687718, Mgat5, Ndst1, Pomt1, Rpn2, St3gal2, St3gal3, Xylt1
 MAPK signaling pathway8.86E-050.0051.74129305Aim1, Arrb2, Cacna1b, Cacna1e, Cacna1h, Cacna2d1, Cacna2d2, Cacng8, Ddit3, Dusp7, Egfr, Fgf1, Fgf11, Fgf17, Fgf22, Fgf5, Fgfr2, Hspb1, Jund, Kras, Map2k2, Map2k5, Map2k7, Map3k1, Map3k10, Map3k12, Mapk12, Mapk4, Mapk8ip, Mapk8ip3, Mapkapk2, Mapkapk3, Mapt, MGC116327, Mras, Myc, Ntrk1, Pdgfra, Pla2g6, Ppm1b, Ppp3cb, Ppp3r1, Ppp3r2, Ppp5c, Prkcb1, Ptk7, Rac1, Rap1b, Stmn1, Tgfbr1, Tnfrsf1a, Tp53
 Regulation of actin cytoskeleton0.0020640.0051.64448087Arhgef1, Arhgef6, Arhgef7, Arpc1b, Arpc5, Bcar1, Cfl1, Chrm3, Egfr, Fgf1, Fgf11, Fgf17, Fgf22, Fgf5, Fgfr2, Gsn, Itgad, Itgam, Itgb1, Itgb6, Itgb7, Kras, Map2k2, Mras, Myh10, Mylk2, Pak4, Pdgfd, Pdgfra, Pik3cb, Pip5k1a, Ppp1ca, Ptk2, Pxn, Rac1, Scin, Ssh3, Tiam1, Was
 Cytokine–cytokine receptor interaction0.0022620.0051.57856573Blr1, Bmpr2, Ccl17, Ccl21b, Ccl24, Ccl4, Ccl6, Ccr1, Ccr3, Ccr5, Cd40lg, Cxcl1, Cxcl14, Cxcl5, Cxcl7, Cxcl9, Cxcr6, Egfr, Flt1, Flt3, Flt4, Gnrhr, Hgf, Il10, Il11ra1, Il13ra1, Il17b, Il1rap, Il2, Il23a, Il2rg, Il5ra, Il7, Lep, LOC679119, LOC688065, Osm, Pdgfd, Pdgfra, Prl, Tgfbr1, Tnfrsf10b, Tnfrsf1a, Tnfrsf8, Tnfsf10, Tpte2
Regulated by downregulated miRNAb
 Apoptosis0.00812310.0054.279734875Akt2, Akt3, Apaf1, Atm, Bcl-2, Bid, Birc2, Capn2, Csf2rb1, Faslg, Ikbkb, Il1a, Il1b, Nfkbia, Pdcd8, Pik3r2, Pik3r3, Prkar2a
 Cell cycle0.00254570.0053.894558736Anapc7, Atm, Ccnd1, Ccne2, Ccnh, Cdc25a, Cdc2a, Cdk7, Plk1, Skp1a, Smad2, Tgfb1, Tgfb2, Ywhag, Ywhah, Ywhaq
 Adipocytokine signaling pathway0.00016210.0052.781827668Acsl5, Acsl6, Adipor2, Akt2, Akt3, Cpt2, Ikbkb, Jak2, Mapk8, Mapk9, Nfkbia, Pck1, Ppargc1a, Prkaa1, Prkaa2, Ptpn11, Tnfrsf1b
 PPAR signaling pathway0.00016210.0052.781827668Acaa1, Acadm, Acsl5, Acsl6, Angptl4, Apoa5, Cpt2, Cyp4a14, Cyp4a22, Ehhadh, Fabp7, Me1, Pck1, Plin, Ppard, Pparg, Ubc
 SNARE interactions in vesicular transport0.00757770.0052.781827668Bet1, Bet1l, Epim, Gosr2, Stx17, Stx3, Sybl1, Vamp1, Vamp8
 Pancreatic cancer0.00012310.0052.74372044Akt2, Akt3, Ccnd1, Erbb2, Figf, Ikbkb, Jak2, Mapk8, Mapk9, Pgf, Pik3r2, Pik3r3, Smad2, Tgfa, Tgfb1, Tgfb2, Tgfbr2, Vegfa
 mTOR signaling pathway0.00132210.0052.729340354Akt2, Akt3, Eif4b, Figf, Ins1, LOC684368, Pgf, Pik3r2, Pik3r3, Prkaa1, Prkaa2, Rps6kb1, Vegfa
 GnRH signaling pathway0.00017070.0052.528934244Adcy3, Adcy4, Atf4, Calm1, Calm3, Cga, Gnaq, Itpr1, Itpr2, Jun, Map2k4, Map2k6, Mapk14, Mapk8, Mapk9, Pla2g2a, Pla2g5, Plcb1, Prkca, Prkcd
 Wnt signaling pathway0.00070180.0052.413151712Ccnd1, Fzd6, Jun, Mapk8, Mapk9, MGC112790, Nfatc4, Plcb1, Ppard, Ppp2r2a, Ppp2r2d, Prkca, Rock2, Skp1a, Smad2, Wif1, Wnt2b
 Prostate cancer0.00051750.0052.402487532Akt2, Akt3, Atf4, Bcl-2, Ccnd1, Ccne2, Creb1, Creb3l2, Creb3l3, Erbb2, Ikbkb, Ins1, Nfkbia, Pdgfc, Pik3r2, Pik3r3, Srd5a1, Srd5a2, Tgfa
 Fc epsilon RI signaling pathway0.00240770.0052.384423716Akt2, Akt3, Gab2, Inpp5d, Map2k4, Map2k6, Mapk14, Mapk8, Mapk9, Pik3r2, Pik3r3, Pla2g2a, Pla2g5, Prkca, Prkcd
 Insulin signaling pathway4.593E-050.0052.347167095Akt2, Akt3, Calm1, Calm3, Exoc7, Fbp1, Ikbkb, Inpp5d, Ins1, Insr, LOC361377, LOC684368, Mapk8, Mapk9, MGC112775, Pck1, Phka1, Pik3r2, Pik3r3, Ppargc1a, Ppp1cc, Ppp1r3b, Prkaa1, Prkaa2, Prkar2a, Pygm, Rps6kb1
 Type I diabetes mellitus0.00514750.0052.290916903Faslg, Gad1, Gad2, Hspd1, Ifng, Il1a, Il1b, Ins1, RT1-A2, RT1-CE10, RT1-CE2, RT1-CE4, RT1-Dob, RT1-Ha
 Long-term depression0.00375730.0052.2864337Crh, Gnai3, Gnaq, Gucy2e, Itpr1, Itpr2, Nos3, Npr2, Pla2g2a, Pla2g5, Plcb1, Ppp2r2a, Ppp2r2d, Prkca, Prkg1
 TGF-β signaling pathway0.00201090.0052.279087728Acvr2a, Acvr2b, Amh, Fst, Gdf7, Ifng, Inhbc, Ppp2r2a, Ppp2r2d, Rock2, Rps6kb1, Skp1a, Smad2, Smad9, Tgfb1, Tgfb2, Tgfbr2
 Colorectal cancer0.00362790.00752.225462135Akt2, Akt3, Bcl-2, Ccnd1, Fzd6, Jun, Mapk8, Mapk9, Met, MGC112790, Pik3r2, Pik3r3, Smad2, Tgfb1, Tgfb2, Tgfbr2
 Small cell lung cancer0.0047340.0052.171182571Akt2, Akt3, Apaf1, Bcl-2, Birc2, Ccnd1, Ccne2, Col4a1, Fn1, Ikbkb, Itga6, Nfkbia, Nos3, Pias4, Pik3r2, Pik3r3
 Focal adhesion0.0002930.00751.978188564Akt2, Akt3, Arhgap5, Bcl-2, Birc2, Capn2, Ccnd1, Col4a1, Col5a2, Erbb2, Figf, Fn1, Ibsp, Itga6, Itgb4, Jun, Mapk8, Mapk9, Met, Myl2, Pak1, Pdgfc, Pgf, Pik3r2, Pik3r3, Ppp1cc, Ppp1cc, Ppp1r12a, Prkca, Rock2, Sgpp1, Vegfa, Vwf
 Calcium signaling pathway0.00043750.0051.959924039Adcy3, Adcy4, Adra1d, Adrb2, Atp2a2, Atp2b4, Bdkrb1, Calm1, Calm3, Chrm2, Chrm5, Chrna7, Erbb2, F2r, Gna14, Gnaq, Grin1, Grpr, Hrh1, Htr2b, Htr5a, Itpr1, Itpr2, LOC361377, Nos3, Oxtr, Phka1, Plcb1, Pln, Prkca, Slc25a5
 MAPK signaling  pathway0.00085280.0051.74985934Akt2, Akt3, Atf4, Cacnb1, Cacnb2, Cacnb4, Cacng5, Daxx, Dusp5, Faslg, Fgf2, Fgf6, Fgfr3, Hspa2, Ikbkb, Il1a, Il1b, JIK, Jun, Map2k1ip1, Map2k4, Map2k6, Map4k3, Mapk14, Mapk6, Mapk8, Mapk9, MGC112775, Nfatc4, Ntf5, Pak1, Pla2g2a, Pla2g5, Prkca, Ptpn5, Rasa1, Tgfb1, Tgfb2, Tgfbr2
 Regulation of actin cytoskeleton0.00878020.0051.652570892Abi2, Bdkrb1, Chrm2, Chrm4, Chrm5, Enah, F2r, Fgf2, Fgf6, Fgfr3, Fn1, Ins1, Itga6, Itgb4, Limk2, LOC683685, LOC684227, Myh9, Myl2, Nckap1, Pak1, Pdgfc, Pik3r2, Pik3r3, Pip5k1c, Pip5k2a, Ppp1cc, Ppp1r12a, Rock2, Slc9a1
 Cytokine–cytokine receptor  interaction0.00693670.0051.609321792Acvr2a, Acvr2b, Amh, Ccl2, Ccl7, Clcf1, Csf1r, Csf2rb1, Csf3, Cxcl11, Cxcr3, Faslg, Figf, Gdf7, Ifng, Il1a, Il1b, Il24, Il2ra, Il4ra, Inhbc, Kit, Lif, LOC681692, Ltb, Met, Pdgfc, Pgf, Tgfb1, Tgfb2, Tgfbr2, Tnfrsf1b, Tnfrsf5, Tpo, Vegfa

CCl4 administration induced liver fibrosis in rats

Histopathological analysis revealed little fibrosis in the liver of normal rats. On the contrary, fatty degeneration, necrosis and infiltration of inflammatory cells were obvious in the fibrosis model group. Moreover, there was nodular fibrosis with extensive collagen deposition and well-delineated fibrosis septa, which were continuous and extended in each section, sometimes even bridging portal regions (Fig. 5). In contrast to the normal rats (stage 0), the Ishak staging for the fibrosis model group reached 5.2 ± 1.2.

image

Figure 5.  HE and VG staining of liver tissue (× 200). HE staining of normal and CCl4-treated liver tissue is shown in (A) and (B), respectively. VG staining of normal and CCl4-treated liver tissue is shown in (C) and (D), respectively.

Download figure to PowerPoint

Members of apoptosis pathways were differentially expressed during HSC activation and liver fibrosis

As evaluated by immnofluorescence assay, Bcl-2 expression was virtually undetectable in quiescent HSCs. However, dramatic increases in Bcl-2 level occurred in HSCs after their activation (P < 0.05). The opposite was seen for caspase-9, another member of the mitochondrial apoptosis pathway (Fig. 6). Its level was reduced by a statistically significant amount throughout HSC activation (P < 0.05).

image

Figure 6.  Immunofluorescence staining of quiescent and activated HSCs for Bcl-2 and caspase-9 (× 400). (A) and (B) show the expression of Bcl-2 on day 2 and day 14, respectively. (C) and (D) show the expression of caspase-9 on day 2 and day 14, respectively. Positive cells per microscopic field: *statistically significant differences. P < 0.05 versus control group.

Download figure to PowerPoint

These findings were confirmed in CCl4-induced experimental hepatic fibrosis. Bcl-2, which can rarely be detected in the normal liver, was expressed increasingly in intrahepatic HSCs after CCl4 injury (P < 0.05) (Figs 7 and 8). The expression of caspase-9, however, was decreased significantly in fibrotic liver when compared to that of normal controls (P < 0.05) (Figs 7 and 8).

image

Figure 7.  Double immunofluorescence staining of Bcl-2, caspase-9 and desmin in liver tissue (× 200). Double staining of desmin (green), Bcl-2, caspase-9 (red) and Hoechst in nuclei (blue) was in the liver tissue. The arrowheads indicate the expression of Bcl-2 and caspase-9 in HSCs showing red/green double-stained cytoplasm and blue-stained nuclei.

Download figure to PowerPoint

image

Figure 8.  Expression of Bcl-2 and caspase-9 in intrahepatic HSCs after CCl4 injury. (A) Double immunofluorescence staining assay; the positive rate is determined by comparing the number of red/green double-stained cells with the number of desmin-positive cells. (B) The mRNA levels of Bcl-2 and caspase-9 in intrahepatic HSCs after CCl4 injury, detected by quantitative real-time PCR. (C) The protein levels of Bcl-2 and caspase-9 in intrahepatic HSCs after CCl4 injury, analyzed by western blotting. A statistically significant difference between control and liver fibrosis is indicated by P < 0.01. *Statistically significant differences.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

MicroRNAs, a set of small, noncoding RNAs, 21–22 nucleotides in length, have recently been recognized to be deeply involved in various crucial cell processes, such as mitosis, differentiation, oncogenesis, and apoptosis, by regulating signal transduction pathways [13]. With the use of in vitro cell activation and miRNA microarray hybridization [10], many differentially expressed miRNAs, 12 upregulated ones (miR-874, miR-29C*, miR-501, miR-349, miR-325-5p, miR-328, miR-138, miR-143, miR-207, miR-872, miR-140, and miR-193) and nine downregulated ones (miR-341, miR-20b-3p, miR-15b, miR-16, miR-375, miR-122, miR-146a, miR-92b, and miR-126), were also identified in rat HSCs during activation. Taking into account the aberrant phenotypes that are closely related to activated HSCs, including myofibroblastic transdifferentiation, active proliferation, and apoptosis resistance [14], an indispensable role of miRNAs was hypothesized throughout their activation on the basis of signaling pathway alternation.

In order to gain insights into the function of miRNAs, GO term and KEGG pathway annotation were applied to their target gene pool. As a result, KEGG annotation showed that important proliferative (cell cycle, VEGF, MAPK, and Wnt), survival (TGF-β and mTOR), apoptotic (apoptosis), adhesive (gap junction and focal adhension,), oncogenic (pancreatic cancer, prostate cancer, colorectal cancer, and small cell lung cancer) and metabolic (biosynthesis of steroids, glycolysis/gluconeogenesis, pyrimidine metabolism, purine metabolism, glycan structure biosynthesis, adipocytokine signaling pathway, insulin signaling pathway) signaling pathways were abundant among the significantly enriched ones. Most of them have already been reported to take part in HSC activation and even hepatic fibrogenesis. For example, MAPK mediates mitosis and the synthesis of α1(I) collagen and matrix metalloproteinases in rat HSCs [15–17]. TGF-β signaling, the key pathway in fibrogenesis, has been found to be essential for myofibroblastic transdifferentiation of HSCs [18,19]. Signaling from the VEGF pathway also stimulates proliferation and type I collagen synthesis in activated HSCs subjected to hypoxia treatment [20–22]. The central role of the Wnt signaling pathway in HSC activation and survival has recently been discovered [5,23,24].

The GOs related to signal transduction (intracellular signaling cascade, small GTPase-mediated signal transduction, signal transduction, transmembrane receptor protein tyrosine kinase signaling pathway, regulation of small GTPase-mediated signal transduction, G-protein-coupled receptor protein signaling pathway, and signal transduction), cell growth (cell proliferation, positive regulation of cell proliferation, cell cycle, positive regulation of mitosis, negative regulation of cell proliferation, and multicellular organism development), apoptosis (antiapoptosis, apoptosis, and induction of apoptosis) and metabolism (lipid metabolic processes and carbohydrate metabolic process) represented up to 33% of the significantly enriched GO terms, which was in accordance with the KEGG analysis. This functional identity revealed by different bioinformatic interpretation confirmed that miRNAs have regulatory effects on HSC activation by affecting signaling pathways.

Enrichment ranking of both signaling pathways and GOs indicated apoptosis to be the most enriched. The miRNA–mRNA interaction network analysis further integrated the bioinformatic findings, and then outlined the major targets of miRNAs. Bcl-2 and caspase-9, both of which had the highest ratio and enrichment in the apoptosis-related pathway, were noted. In line with the in silico analysis, upregulated Bcl-2 and downregulated caspase-9 were identified in activated HSCs in vitro and fibrotic liver in vivo, using immunofluorescence assay, quantitative RT-PCR, and western blot analysis. Acting as an antiapoptotic member, Bcl-2 preserves mitochondrial integrity and potentially blocks the release of some soluble prodeath intermembrane proteins. Therefore, caspase-9-dependent apoptosis is inhibited. These results may provide more evidence for the reliability of bioinformatics analysis, and be helpful in shedding light on the mechanisms underlying HSC activation.

In conclusion, most of the signaling pathways involved in HSC activation may be regulated by miRNAs. Among these, the mitochondrial pathway of apoptosis is likely to take the critical place during activation by miRNA-targeted Bcl-2 and caspase-9.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Isolation and culture of rat HSCs

HSCs were isolated from Sprague–Dawley rats (350–400 g; Shanghai Laboratory Animal Center of the Chinese Academy of Sciences) by perfusion with collagenase and pronase, followed by centrifugation (1500 g, 17 min) over a Nycodenz gradient [25]. They were then cultured in DMEM supplemented with 10% fetal bovine serum. The quiescent and activated HSCs were then harvested on the second and the 14th day, respectively.

Immunofluorescence staining for desmin was performed using antibody against desmin. Cells were counterstained with fluorescein isothiocyanate-conjugated rabbit anti-(goat IgG) (1 : 100; Molecular Probes, Eugene, OR, USA). Nuclei were labeled with Hoechst 33258 (Roche, Germany). Negative controls were performed both with antibody against α-smooth muscle actin (SMA) (1 : 100; Santa Cruz, CA, USA) and without primary antibody.

GO terms and KEGG pathway annotation based on miRNA expression profile

Total RNA of HSCs was extracted and hybridized to the miRCURY LNA array, version 8.0 (Exiqon, Denmark). We selected the miRNAs measured as present in at least the smallest class in the dataset (25%) [26].

Thereafter, we pooled the reported and predicted targets of filtered miRNAs from the Sanger database (http://microrna.sanger.ac.uk/). The top 25% miRNA targets that had been assigned the highest numbers of miRNA interaction sites were collected, and subjected to GO term analysis. GO analysis was applied in order to organize genes into hierarchical categories and uncover the miR–gene regulatory network on the basis of biological process and molecular function; the network of miRNA–mRNA interaction, representing the critical miRNAs and their targets, was established according to the miRNA degree. Meanwhile, the top 25% miRNA targets were collected, and subjected to KEGG pathway annotation using the david gene annotation tool (http://david.abcc.ncifcrf.gov/) [27]. In detail, a two-sided Fisher’s exact test and chi-square test were used to classify the enrichment (Re) of pathway category, and the false discovery rate (FDR) was calculated to correct the P-value. Within a KO, the enrichment (Re) was given by

  • image

where nf and n represent the number of target genes and total genes, respectively, in the particular KO, and Nf and N represent the number of genes among the entire differential miRNA-corresponding target genes and the total number of genes on the pathway, respectively. We chose only pathways that had a P-value of < 0.01 and an FDR of < 0.01. The regulator pathway annotation was also performed on the basis of scoring and visualization of the pathways collected in the KEGG database (http://www.genome.jp/kegg/).

Animal model of liver fibrosis

Thirty Sprague–Dawley rats (250–400 g; Shanghai Laboratory Animal Center of Chinese Academy of Sciences) were divided into three groups (normal, control, and fibrosis model; n = 10 in each group). Fibrosis model rats were injected subcutaneously with 40% CCl4 (3 mL·kg−1; CCl4/olive oil ratio of 2 : 3) every 3 days for 8 weeks. Control rats received only olive oil in the same way. All rats received humane care according to the Guide for the Care and Use of Laboratory Animals of the Chinese Academy of Sciences.

Histological examination

Liver tissues were fixed in 40 g·L−1 solutions of formaldehyde in NaCl/Pi (pH 7.4) and embedded in paraffin. Five-micrometer thick section slides were prepared. All of the sections were stained with hematoxylin/eosin (HE) and standard Van Gieson (VG) stain, which was used to detect collagen fibers. Fibrosis was graded according to the Ishak modified staging system [28]. Histopathology was interpreted by two independent board-certified pathologists who were blind to the study.

Immunofluorescence staining of HSCs

The expression of Bcl-2 and caspase-9 in quiescent (2 days) and in culture-activated (14 days) HSCs was evaluated by immunocytochemistery. The adherent HSCs were fixed with 4% paraformaldehyde and permeabilized with 0.1% Triton X-100 (Sigma, St Louis, MO, USA). Following blocking in 10% preimmune goat serum for 2 h, cells were incubated with mouse monoclonal antibody against Bcl-2 (1 : 100; Santa Cruz, CA, USA) overnight at 4 °C. Cells were then incubated with tetramethylrhodamine isothiocyanate-conjugated donkey anti-(mouse IgG) (Sigma; 1 : 100) for 1 h. Tetramethylrhodamine isothiocyanate fluorescence were visualized using a fluorescence microscope. The positive cells of three randomly selected areas per slide from three slides was used to calculate the expression of Bcl-2 and caspase-9 in HSCs.

Double immunostaining on cryosections of rat liver

Double immunostaining on cryosections of rat liver were performed as previously described [29,30]. Liver tissue from five rats per group were blocked with 0.3% H2O2 in methanol for endogenous peroxidase activity. Double staining experiments on rat livers for desmin in combination with Bcl-2 or caspase-9 were performed. Immunohistochemical examination was carried out by a researcher blind to the experimental design. The percentage of cells coexpressing Bcl-2/desmin or caspase-9/desmin was determined by counting the number of Bcl-2-positive or caspase-9-positive cells in desmin-positive cells in three different fields per slide from three slides.

Quantitative RT-PCR analysis of apoptosis-related gene expression

HSCs from three rats per group were isolated from rat livers by perfusion of collagenase and pronase, followed by centrifugation (1500 g, 17 min) over a Nycodenz gradient, as described above. The extracted total RNA of HSCs from three rats per group was reverse transcribed into cDNA using an ExScript RT reagent Kit (Takara, Kusatsu, Japan). Real-time PCR was performed using the SYBR Premix ExTaq (Takara) on a LightCycler (Roche Diagnostics GmbH, Penzberg, Germany). The sense and antisense primers used in this study are as follows: Bcl-2 (NM-016993, 116 bp), 5′-TGAACCGGCATCTGCACAC-3′ and 5′-CGTCTTCAGAGACAGCCAGGAG-3′; caspase-9 (NM_031632, 206 bp), 5′-TGCACTTCCTCTCAAGGCAGGACC-3′ and 5′-TCCAAGGTCTCCATGTACCAGGAGC-3′; and glyceraldehyde-3-phosphate dehydrogenase (NM-002046, 450bp), 5′-ACCACAGTCCATGCCATCAC-3′ and 5′-TCCACCACCCTGTTGCTGTA-3′. PCR was performed as follows: 95 °C for 10 s, followed by 40 cycles of denaturation at 95 °C for 5 s and annealing/extension at 60 °C for 20 s. Each sample was run in triplicate. Independent experiments were repeated twice. Glyceraldehyde-3-phosphate dehydrogenase was used as endogenous control. Relative gene expression levels were calculated by the 2[-Delta Delta C(T)] method [19].

Western blot analysis of apoptosis-related gene expression

Total proteins were prepared by standard procedures and quantified by the bicinchoninic acid method (Pierce, Rockford, IL, USA). Thirty micrograms of protein per sample was loaded onto a 10% SDS/PAGE gel. After electrophoresis, the protein was transferred onto a poly(vinylidene difluoride) membrane (Millipore, Billerica, MA, USA) by electroelution. The membrane was incubated with antibody against Bcl-2 or caspase-9 (1 : 500; Santa Cruz, CA, USA) overnight at 4 °C, and with horseradish peroxidase-conjugated goat anti-(mouse IgG) (1 : 5000; Jackson ImmunoResearch) for 2 h at room temperature. After washing, the membrane was processed using SuperSignal West Pico chemiluminescent substrate (Pierce), and antibody against actin (Santa Cruz, CA, USA) (1 : 500) as an internal standard.

Statistical analysis

All of the results are expressed as mean ±standard deviation. Statistical analysis was performed with Student’s t-test for comparison of two groups, and with anova for multiple comparisons. In both cases, differences with P < 0.05 were considered to be statistically significant.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

This work was supported by the Foundation of Shanghai Commission of Science Technology of Research Program (O7JC14044).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • 1
    Brenner DA, Waterboer T, Choi SK, Lindquist JN, Stefanovic B, Burchardt E, Yamauchi M, Gillan A & Rippe RA (2000) New aspects of hepatic fibrosis. J Hepatol 32, 3238.
  • 2
    Friedman SL (2004) Mechanisms of disease: mechanisms of hepatic fibrosis and therapeutic implications. Nat Clin Pract Gastroenterol Hepatol 1, 98105.
  • 3
    Gressner AM & Weiskirchen R (2006) Modern pathogenetic concepts of liver fibrosis suggest stellate cells and TGF-beta as major players and therapeutic targets. J Cell Mol Med 10, 7699.
  • 4
    Godichaud S, Si-Tayeb K, Auge N, Desmouliere A, Balabaud C, Payrastre B, Negre-Salvayre A & Rosenbaum J (2006) The grape-derived polyphenol resveratrol differentially affects epidermal and platelet-derived growth factor signaling in human liver myofibroblasts. Int J Biochem Cell Biol 38, 629637.
  • 5
    Jiang F, Parsons CJ & Stefanovic B (2006) Gene expression profile of quiescent and activated rat hepatic stellate cells implicates Wnt signaling pathway in activation. J Hepatol 45, 401409.
  • 6
    Kanematsu M, Semelka RC, Osada S & Amaoka N (2005) Magnetic resonance imaging and expression of vascular endothelial growth factor in hepatocellular nodules in cirrhosis and hepatocellular carcinomas. Top Magn Reson Imaging 16, 6775.
  • 7
    Langer DA, Das A, Semela D, Kang-Decker N, Hendrickson H, Bronk SF, Katusic ZS, Gores GJ & Shah VH (2008) Nitric oxide promotes caspase-independent hepatic stellate cell apoptosis through the generation of reactive oxygen species. J Hepatol 47, 19831993.
  • 8
    Gäbele E, Brenner DA & Rippe RA (2003) Liver fibrosis: signals leading to the amplification of the fibrogenic hepatic stellate cell. Front Biosci 8, 6977.
  • 9
    Chitwood DH & Timmermans MC (2007) Target mimics modulate miRNAs. Nat Genet 39, 935936.
  • 10
    Guo CJ, Pan Q, Li DG, Sun H & Liu BW (2009) miR-15b and miR-16 are implicated in activation of the rat hepatic stellate cell: an essential role for apoptosis. J Hepatol 50, 766778.
  • 11
    Mao X, Cai T, Olyarchuk JG & Wei L (2005) Automated genome annotation and pathway identification using the KEGG Orthology (KO) as a controlled vocabulary. Bioinformatics 21, 37873793.
  • 12
    Kanehisa M, Goto S, Kawashima S, Okuno Y & Hattori M (2004) The KEGG resource for deciphering the genome. Nucleic Acids Res 11, 277280.
  • 13
    Ambros V (2004) The functions of animal microRNAs. Nature 431, 350355.
  • 14
    Iredale JP (2001) Hepatic stellate cell behavior during resolution of liver injury. Semin Liver Dis 21, 427436.
  • 15
    Popov Y, Patsenker E, Bauer M, Niedobitek E, Schulze-Krebs A & Schuppan D (2006) Halofuginone induces matrix metalloproteinases in rat hepatic stellate cells via activation of p38 and NFkappaB. J Biol Chem 281, 1509015098.
  • 16
    Tsukada S, Westwick JK, Ikejima K, Sato N & Rippe RA (2005) SMAD and p38 MAPK signaling pathways independently regulate alpha1(I) collagen gene expression in unstimulated and transforming growth factor-beta-stimulated hepatic stellate cells. J Biol Chem 280, 1005510064.
  • 17
    Saxena NK, Titus MA, Ding X, Floyd J, Srinivasan S, Sitaraman SV & Anania FA (2004) Leptin as a novel profibrogenic cytokine in hepatic stellate cells: mitogenesis and inhibition of apoptosis mediated by extracellular regulated kinase (Erk) and Akt phosphorylation. FASEB J 18, 16121614.
  • 18
    Dooley S, Hamzavi J, Ciuclan L, Godoy P, Ilkavets I, Ehnert S, Ueberham E, Gebhardt R, Kanzler S, Geier A et al. (2008) Hepatocyte-specific Smad7 expression attenuates TGF-beta-mediated fibrogenesis and protects against liver damage. Phosphorylation. Gastroenterology 135, 642659.
  • 19
    Marquez-Aguirre A, Sandoval-Rodriguez A, Gonzalez-Cuevas J, Bueno-Topete M, Navarro-Partida J, Arellano-Olivera I, Lucano-Landeros S & Armendariz-Borunda J (2009) Adenoviral delivery of dominant-negative transforming growth factor beta type II receptor up-regulates transcriptional repressor SKI-like oncogene, decreases matrix metalloproteinase 2 in hepatic stellate cell and prevents liver fibrosis in rats. J Gene Med 11, 207219.
  • 20
    Novo E, Cannito S, Zamara E, Valfrè di Bonzo L, Caligiuri A, Cravanzola C, Compagnone A, Colombatto S, Marra F, Pinzani M et al. (2007) Proangiogenic cytokines as hypoxia-dependent factors stimulating migration of human hepatic stellate cells. Am J Pathol 170, 19421953.
  • 21
    Liu Y, Lui EL, Friedman SL, Li L, Ye T, Chen Y, Poon RT, Wo J, Kok TW & Fan ST (2009) PTK787/ZK22258 attenuates stellate cell activation and hepatic fibrosis in vivo by inhibiting VEGF signaling. Lab Invest 89, 209221.
  • 22
    Ankoma-Sey V, Wang Y & Dai Z (2000) Hypoxic stimulation of vascular endothelial growth factor expression in activated rat hepatic stellate cells. Hepatology 31, 141148.
  • 23
    Cheng JH, She H, Han YP, Wang J, Xiong S, Asahina K & Tsukamoto H (2008) Wnt antagonism inhibits hepatic stellate cell activation and liver fibrosis. Am J Physiol Gastrointest Liver Physiol 294, G39G49.
  • 24
    Myung SJ, Yoon JH, Gwak GY, Kim W, Lee JH, Kim KM, Shin CS, Jang JJ, Lee SH, Lee SM et al. (2007) Wnt signaling enhances the activation and survival of human hepatic stellate cells. FEBS Lett 581, 29542958.
  • 25
    Friedman SL & Roll FJ (1987) Isolation and culture of hepatic lipocytes, Kupffer cells, and sinusoidal endothelial cells by density gradient centrifugation with Stractan. Anal Biochem 161, 207218.
  • 26
    Liu CG, Calin GA, Meloon B, Gamliel N, Sevignani C, Ferracin M, Dumitru CD, Shimizu M, Zupo S, Dono M et al. (2004) An oligonucleotide microchip for genome-wide microRNA profiling in human and mouse tissues. Proc Natl Acad Sci USA 101, 97409744.
  • 27
    Dennis G Jr, Sherman BT, Hosack DA, Yang J, Gao W, Clifford Lane H & Lempicki RA (2003) DAVID: database for annotation, visualization, and integrated discovery. Genome Biol 4, 3.
  • 28
    Ishak K, Baptista A, Bianchi L, Callea F, De Groote J, Gudat F, Denk H, Desmet V, Korb G, MacSween RNM et al. (1995) Histological grading and staging of chronic hepatitis. J Hepatol 22, 696699.
  • 29
    Cassiman D, van Pelt J, De Vos R, Van Lommel F, Desmet V, Yap SH & Roskams T (1999) Synaptophysin: a novel marker for human and rat hepatic stellate cells. Am J Pathol 155, 18311839.
  • 30
    Cassiman D, Denef C, Desmet V & Roskams T (2001) Human and rat hepatic stellate cells express neurotrophins and neurotrophin receptors. Hepatology 33, 148158.
  • 31
    Livak KJ & Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T))method. Methods 25, 402408.