ORIGINAL ARTICLE 764

Increased Smad1 Expression andTranscriptional Activity EnhancesTrans-Differentiation ofHepatic Stellate Cells

HONG SHEN,1 JIANGHONG FAN,2 FRANK BURCZYNSKI,2 GERALD Y. MINUK,3

PETER CATTINI,4 AND YUEWEN GONG2,3*1Medical Research Center, Xiangya Hospital, Central South University, Changsha, Hunan, China2Faculty of Pharmacy, University of Manitoba, Winnipeg, Canada3Department of Internal Medicine, University of Manitoba, Winnipeg, Canada4Department of Physiology, Faculty of Medicine, University of Manitoba, Winnipeg, Canada

Smad1 is a receptor-activated intracellular signaling protein, whichmediates signal transduction of bonemorphogenetic proteins. Currentstudy investigated the expression and transcriptional activity of Smad1 during hepatic stellate cell (HSC) activation. Rat HSCs wereisolated from rats at 1, 2, 3 and 4 days after gavagedwith carbon tetrachloride (CCl4) or corn oil. RT-PCR,Western blot, gel-shift assay andluciferase assay were employed to examine Smad1 expression and transcriptional activity, respectively. CCl4-cirrhotic liver fat-storingcells-8B (CFSC-8B) cells were infected with recombinant adenoviruses of Smad1 and/or Smad1 shRNA. BothmRNA and protein levels ofSmad1were significantly increased at 48 h after gavage of CCl4. Gel shift assays demonstrated a significant increase in nuclear Smad1 in day9 HSCs. Transfection of HSCs with Smad1 responsible luciferase indicated an increase in Smad1 transcriptional activity in day 6 HSCs(1.563� 0.229 in day 6 versus 0.785� 0.192 in day 3).When CFSC-8B cells were infected with adenoviruses with Smad1 or Smad1 shorthairpin RNA (shRNA), there was an increase or decrease in Smad1 mRNA and protein, respectively. Smooth muscle a-actin expressionwas increased or decreased according to induction or reduction of Smad1. In conclusion, there were significantly increases in Smad1expression and transcriptional activity during in vivo activation of hepatic stellate cells.J. Cell. Physiol. 212: 764–770, 2007. � 2007 Wiley-Liss, Inc.

*Correspondence to: Yuewen Gong, Faculty of Pharmacy,University of Manitoba, 50 Sifton Road, Winnipeg, Manitoba,Canada, R3T 2N2. E-mail: ygong@cc.umanitoba.ca

Received 18 October 2006; Accepted 6 February 2007

DOI: 10.1002/jcp.21074

Hepatic stellate cells (HSCs) are nonparenchymal cells of theliver constituting about 1.4% of total liver volume and 5% oftotal liver cells (Blouin et al., 1977; Klinger et al., 1988). It iscommonly believed that HSCs are the key fibrogenicprogenitor cells in the liver (Davis and Kresina, 1996; Pinzaniand Gentilini, 1999; Brenner et al., 2000). In normal liver, HSCscontain lipid droplets of retinoid and express glial fibrillaryacidic protein (GFAP) (Hendriks et al., 1985; Buniatian et al.,1996; Neubauer et al., 1996). One of the important features ofHSCs is the activation of HSCs from quiescent phenotype toproliferative, fibrogenic, contractile and myofibroblast-likephenotype (Bachem et al., 1993; Pinzani, 1995). The activationof HSCs is a complex process and involves a variety ofinteractions among hepatocytes, endothelial cells, Kupffer cellsand inflammatory cells (Kmiec, 2001). Moreover, the activationof HSCs can be initiated by cytokines (Marra, 2002; Yang et al.,2003;Marra et al., 2004), reactive oxygen species (Bataller et al.,2003) and products of ECM (Imai and Senoo, 1998).Furthermore, the activation process could be reproduced byeither in vitro culture of HSCs on uncoated plastic dishes orsingle gavage of CCl4 in the rats (Burt et al., 1986; Friedmanet al., 1992). ActivatedHSCs express smoothmuscle alpha actin(a-SMA) and different extracellular matrix (ECM) proteins ascompared to quiescent HSCs. These different ECM proteinsare the major cause of liver fibrosis (Friedman, 1993).

Smads are intracellular proteins in mammalian cells. Theywere first identified in Drosophila and C. elegans as MAD andSma, respectively (Derynck, 1998). At present, at least eightSmads have been identified and they are divided into threedifferent groups: (1) receptor-activated Smad (R-Smad) whichinclude Smads 1, 2, 3, 5 and 8; (2) a common-mediated SmadSmad4; (3) inhibitory Smad including Smads 6 and 7 (Attisano

� 2 0 0 7 W I L E Y - L I S S , I N C .

and Wrana, 2000; Massague and Wotton, 2000). It is generallyknown that Smads 1, 5 and 8 mediate bone morphogeneticprotein (BMP) signaling and Smads 2 and 3 involve transforminggrowth factor beta (TGF-b) signal transduction (Miyazonoet al., 2000). However, TGF-b1 may also signal through Smad1pathway in some cancer cells (Liu et al., 1998) and other factorsmay also activate Smad2 and Smad3 (Feng et al., 1998; Hymanet al., 2003).

Smad1 is an R-Smadwhichmediates BMP signal transduction,and contains two highly conserved domains; the N terminus ofSmad1 contains the MH1 domain and is responsible for DNAbinding while the C terminus contains the MH2 domain and isresponsible for transcriptional activation. These domains arejoined in the middle by a more variable proline-rich linker.Smad1 mediates its effect through protein–protein andprotein–DNA interactions. Several DNA-binding motifs forSmad1 have been identified. It was reported that a GCATmotifwas responsible for Smad1 binding in the Xvent-2B promoter(Henningfeld et al., 2000). Smad1 binding sequence—GCCGnCGC (GCCG motif) was considered as the consensusbinding motif in mammalian BMP-regulated genes (Langlands

H E P A T I C S T E L L A T E C E L L A C T I V A T I O N 765

et al., 1997; Park and Morasso, 2002). Smad1 directly bindsmulti-copies of the consensus GCCG motif and its bindingaffinity is highly dependent on the copy number of this motif(Kusanagi et al., 2000). Previously, we have reported that Smad1expression was increased during in vitro activation of HSCs(Shen et al., 2003a). In current study, we investigate the role ofSmad1 in HSC activation.

Materials and MethodsMaterials

Collagenase D, pronase, DNase, transfection reagent Fugene-6 andLightCycler-DNA Master SYBR Green I kit were purchased fromRoche Diagnostics (Laval, QC, Canada). Dual-Luciferase ReporterSystem and all restriction enzymes were purchased from Promega(Madison,WI). All reagents for cell culture (Dulbecco’s modified Eaglemedium (DMEM), fetal bovine serum (FBS), trypsin–EDTA andphosphate-buffered saline) were purchased from Invitrogen. Thenuclear extract isolation kit of NuCLEARTM Extraction and all otherchemicals were obtained from Sigma–Aldrich Canada Ltd. (Oakville,ON) unless otherwise indicated. Rabbit polyclonal antibodies againstSamd1 and phospho-Smad1 were purchased from UpstateBiotechnology (Lake Placid, NY). Mouse monoclonal antibody againstBMP4 and antibody against b-actin were purchased from Santa CruzBiotechnology (San Francisco, CA). Trizol Reagent and PCR primerswere purchased from GIBCO/BRL (Burlington, ON). [a-32P]-dCTP,donkey anti-rabbit IgG, sheep anti-mouse IgG and the enhancedchemiluminescence (ECL) Western blotting kit were purchased fromAmersham-Pharmacia Biotech Inc. (Baie d’Urfe, Quebec). TheAdvantage RT-for-PCR Kit, Advantage-HF PCR Kit, Advantage cDNAPCR Kit, BD Adeno-X Expression System 2 and Adeno-X Rapid TiterKit were purchased from BD Biosciences (Palo Alto, CA).

Animal and isolation of HSCs

Male Sprague–Dawley rats (200–250 g body weight) were purchasedfrom the Central Animal Breeding facility of the University of Manitobaand maintained under temperature controlled conditions (22� 28C)of an artificial 12-h light/dark cycles with food and water ad libitum. Inconducting the research described in this report, all animals receivedhumane care in compliance with the Institution’s guidelines (AnimalProtocol No. F2003-011), which is in accordance with the CanadianCouncil on Animal Care’s criteria. Rats were treated with a single doseof carbon tetrachloride (CCl4) (0.4ml/kg bodyweight, 1:1 dilutionwithcorn oil) by oral gavage. Control rats were treated with an equalvolumeof corn oil only. At 1, 2, 3, and 4 days after gavagingCCl4 or cornoil, rats were sacrificed and HSCs were isolated from these rats asdescribed previously (Shen et al., 2002). All isolated HSCs werecultured for only overnight and thenHSCswere collected for RNA andprotein extraction.

Cell culture

HSCs were cultured in DMEM with 10% FBS. HSCs were eitheremployed for the isolation of nuclear and cytosol proteins immediatelyor cultured for 3, 6, and 9 days. For cultured cells, they were preparedfor the extraction of nuclear proteins or transfected with a luciferasereporter vector. All culture media were supplemented with penicillin(100 IU/ml)/streptomycin (100 mg/ml) and cells were maintained at

TABLE 1. Polymerase chain reaction primers and oligonucleotides

No. Sequence

1 UP 5(-CCGCCTGCTTACCTGCCTCCTGAA-3( (685–DP 5(-GAACGCTTCGCCCACACGGTTGT-3( (863–8

2 UP 5(-TTCTTGTGCAGTGCCAGCCTCGTC-3( (815–DP 5(-GCCGTTGAACTTGCCGTGGGTAGA-3( (994

3 UP 5(-GTATTTCCTACCTTTCCGAACC-3(DP 5(-GGCCATCTCTTTTCTAACTATTCA-3(

4 UP 5(-CAGCCGCTATGAATGTGACCAG-3(DP 5(-AGCTCCCCATCCTGTCTGACTTCT-3(

5 US 5(CACCGCAACTACCACCATGGCTTTCCGAAGDS 5(AAAAGCAACTACCACCATGGCTTTCTTCGG

UP indicates upstream primer; DP means downstream primer. US indicates up-strand oligonucl

JOURNAL OF CELLULAR PHYSIOLOGY DOI 10.1002/JCP

378C and 5% CO2 in a humidified atmosphere. The transformed ratHSC cell line [CCl4-cirrhotic liver fat-storing cells-8B (CFSC-8B)] waskindly provided by Dr. Marcos Rojkind in the George WashingtonUniversity Medical Center. CFSC-8B cells were incubated in DMEMwith 10% FBS (Inagaki et al., 1995).

Protein isolation and western blot analyses

HSCs protein was extracted from freshly isolated HSCs at 12, 24, 48,72 and 96 h after CCl4 gavage by cellular protein extraction solution(1�¼ 10mmol/L Tris—HCl pH7.5, 1mmol/L EDTA pH8.0, 10mmol/L NaCl, 1% SDS, 1 mmol/L PMSF, 0.25 mol/L sucrose). Proteinconcentrations were measured by the Lowry method (Fryer et al.,1986). One hundred micrograms of protein was then mixedwith 4� gel loading buffer, separated on 12% sodium dodecylsulfate-polyacrylamide (SDS-polyacrylamide) gel under reducingconditions, and transferred onto Nitroplus-2000 membrane (MicronSeparations Inc., Westborough, MA). Nonspecific antibody bindingwas blocked by pre-incubation of the membranes in 1� Tris-buffered-saline (TBS) containing 5% skim milk for 1 h at room temperature.Membranes were incubated overnight at 48C with antibodies againstSmad1, phospho-Smad1 or a-SMA at 1:500, 1:500 or 1: 2000 dilutions,respectively, in 1� TBS containing 0.1% Tween-20 and 2% skim milk.After washing, they were incubated with donkey anti-rabbit IgG orsheep anti-mouse at 1:1000 dilutions for 1 h at room temperature.Bands were visualized by employing the enhanced chemiluminescencekit according to the manufacturer’s instructions.

Reverse transcription polymerase chain reaction (RT-PCR)

Total RNA was isolated by Trizol from fresh isolated HSCs at 12, 24,48, 72, and 96 h after CCl4 gavage. The first strand cDNA wassynthesized by the Advantage RT-for-PCR kit as described previously(Shen et al., 2003a). PCR was performed using the Advantage PCR kitand the oligonucleotides synthesized by GIBCO/BRL. The specificprimers (Table 1 primer pair no. 1) for rat Smad1 were designed fromthe rat Smad1 sequence (GenBankAF067727) byOligo 5.1 programona Macintosh computer. The specific primers for rat glyceraldehydes3-phosphate dehydrogenate (GAPDH) (Table 1 primer pair no. 2)were designed from the rat GAPDH sequence (GenBankNM_017008). PCR amplification was carried out by applying 28 cyclescomprising: denaturation at 948C for 1 min, 648C (Smad1) or 608C(GAPDH) for 30 sec, elongation at 728C for 2 min, followed by a finalelongation at 728C for 8 min using an Eppendoff MasterCycler(Eppendoff, Westbury, NY). PCR products were analyzed byelectrophoresis on a 1.5% agarose gel. Identity of PCR products wasconfirmed by DNA sequencing at the DNA sequencing facility ofManitoba Institute of Cell Biology.

Real-time RT-PCR quantitation of mRNA

Quantization of mRNA was performed by Light-Cycler following thereal-time RT-PCR protocol provided by Roche MolecularBiochemicals. Briefly, rat full length Smad1 cDNA was inserted intopCR2.1 vector (Invitrogen Canada Inc., Burlington, ON). The sensemRNA of rat Smad1 was synthesized by Riboprobe CombinationSystem-T3/T7 kit (Promega Corporation, Madison, WI) with T7polymerase. The in vitro transcribed sense mRNA of rat Smad1 wasemployed as template for standard curve.Onemicrogramof total RNAfrom each sample was employed for real-time RT-PCR quantization ofSmad1 with the primers indicated above. Real-time RT-PCR wasperformed with the LightCycler-DNA Master SYBR Green I kit in the

Length of product

709) 201 bp86)839) 203 bp–1018)

AAAGCCATGGTGGTAGTTGC3(AAAGCCATGGTGGTAGTTGC3(

eotide; DS means down-strand oligonucleotide.

Fig. 1. Expression of Smad1 and a-SMA during in vivo activation ofHSCs. Upper panel represents typical Western blot pictures ofa-SMA and Smad1 protein expression in primary rat HSCs at 1, 2, 3and 4 days after one dose administration of CCl4 or corn oil. Freshisolated HSCs were cultured for overnight only and protein wasextracted for Western blot analyses with antibodies againsta-Smooth muscle actin (a-SMA), Smad1 and b-actin, respectively.b-Actin was used as protein loading control. Lower panel displaysrelative expression of rat a-SMA and Smad1 protein level in HSCsafter CCl4 or corn oil administration. ‘‘R’’ represents CCl4 treatmentand ‘‘S’’ represents corn oil gavage as the sham control. Datarepresent meanWSEM from six rats. MP< 0.05. [Color figure can beviewed in the online issue, which is available at www.interscience.wiley.com.]

766 S H E N E T A L .

following protocol: denaturation at 958C for 2 min, amplificationat 40 cycles of 1 sec at 948C, 10 sec at 628C and 16 sec at 728C.Fluorescent intensity was determined at 588C for 10 sec at the end ofeach cycle. The amount of Smad1mRNAwas calculated and expressedas ng/mg of standard RNA.

Construction and expression of Adenoviruscontaining rat Smad1

The full-length rat Smad1 cDNA was cloned from the rat liverMarathon-Ready cDNA (BD Biosciences, Palo Alto, CA) by nestedhigh-fidelity polymerase chain reaction with respective primer pairs(Table 1 primer pairs nos. 3 and 4). The purified nested-PCR productwas cloned into pCR-Blunt II TOPO (Invitrogen) following by DNAsequencing. After sequencing confirmation of Smad1 cDNA, theSmad1 cDNA was cut and sub-cloned into the shuttle plasmid ofpDNR-CMV and then transferred to the acceptor vector of pLP-Adeno-X-CMV. The recombinant adenovirus with rat Smad1 wasproduced in HEK-293 cells by the transfection of pLP-Adeno-X-CMVwith FuGene-6. After quantification by a titration test, the recombinantSmad1 adenoviral stocks were employed to infect theHSCs at the titerof 20 pfu/cell. A recombinant adenovirus of the same construct withenhanced green fluorescence protein (EGFP) instead of rat Smad1(BD Biosciences) was used as the adenovirus infection control.

Smad1 short-hairpin RNA (shRNA) adenovirus construction

BLOCK-iT Adenoviral RNAi Expression System (Invitrogen) wasemployed to construct the recombinant adenoviruses for RNAi thattarget the rat Smad1mRNA sequence according to themanufacturer’sinstruction. Briefly, the double-stranded oligonucleotides (Table 1, no.5) were ligated with the linearized pENTR/U6 vector. An LRrecombination reaction was performed to transfer the U6 RNAicassette from the pENTR/U6 vector into a full-length pAd/Block-iT-DEST adenoviral vector. The PacI-digested plasmid was used totransfect HEK-293 cells with FuGene-6. The crude viral lysate wasamplified in HEK-293 cell and the recombinant adenovirus withshort-hairpin Smad1 RNA was titrated. Recombinant adenovirus withshort-hairpin Smad1 RNAwas employed to infect CFSC-8B cells at thetiter of 20 pfu/cell.

Electrophoretic mobility shift assay (EMSA)

The nuclear extracts of HSCs were isolated by NuCLEARTM

Extraction. Protein concentrations were determined using theBradford Assay (Bio-Rad) and the nuclear extracts stored at �808Cuntil study. A double-stranded oligonucleotides representing theconsensus binding site of Smad1 (9xGCCG) (Kusanagi et al., 2000)were labeled with g-32P adenosine triphosphate using T4-polynucleotide kinase. Five micrograms of nuclear protein extractswere incubated on ice for 20 min with a mixture containing 0.1 ml of5 mg/ml poly(dI-dC), 1 ml of 5 mg/ml single-stranded oligonucleotide,1� binding buffer (10 mmol/L HEPES, pH 7.8, 50 mmol/L KCl,0.1 mmol/L EDTA, 2 mmol/L MgCl2, 1 mmol/L DTT, 10% glycerol)and 105 cpm 32P-labeled oligonucleotide in a final volume of 20 ml. Forspecific gel mobility analysis, the nuclear extracts were incubated with2 mg antibody against Smad1 on ice for an additional 30 min prior toaddition of 32P-labeled probe. For competition analysis, different molarconcentrations of unlabeled probes were added to the solution. Afterincubation, the reaction mixtures were loaded on 4% non-denaturingacrylamide gel containing 0.25� Tris-buffered EDTA (TBE) and 2.5%glycerol. The gels were run in 0.25� TBE buffer at room temperature.After drying, gels were exposed to X-ray film.

Transient transfection and luciferase assay

Primary HSCs at days 3 and 6 of culture were transfected with 2 mgplasmid—pGL2-12xGCCG-CollagenX (a gift from Dr. KoheiMiyazono, The Japanese Foundation for Cancer Research, Tokyo,Japan) and pGL2-Basic (Promega Corporation, Madison, WI) usingFuGene-6 reagent, respectively. In addition, all cells wereco-transfected with pRL-TK vector as transfection efficiency control.After 24 h, culture medium was changed to DMEM with 10% FBS andHSCs were further cultured for 24 h. The transfected HSCs were thenlyzed and cytosol was prepared for luciferase assay by employing theDual-Luciferase Reporter System kit. The luciferase activity wasmeasured by a Luminometer (TD20/20; Turner Designs, Sunnyvale,CA). The activity of Firefly luciferase was normalized to that of Renilla

JOURNAL OF CELLULAR PHYSIOLOGY DOI 10.1002/JCP

luciferase (pRL-TK vector encoding Renilla luciferase, PromegaCorporation, Madison, WI) and expressed as relative fold induction.

Microphotography

CFSC-8B cells were infected with different adenoviruses for 72 hrs.At the end of infection, cell morphology was photographed on anOlympus inverted-phase microscope (CK-40) using a mountedOlympus 35-mm camera (Carsen Group Inc., Markham, ON) andTMAX 400 Kodak black-and-white film (Eastern Kodak Co.,Rochester, New York, USA).

Statistical analyses

To analyze differences in the treatment groups, we performed theANOVA and Fisher’s PLSD test as post hoc test using StatView(version 5.0) software (SAS Institute Inc., Cary, NC). Differences withP values below 0.05 were considered significant.

ResultsExpression of Smad1 during in vivo activation of HSCs

In order to investigate whether Smad1 expression is elevatedduring in vivo activation of HSCs, we employed an animal modelof acute HSCs activation by gavaging rats with CCl4. After oneadministration ofCCl4 or cornoil, HSCswere isolated from theliver at days 1, 2, 3 and 4, respectively. As shown in Figure 1,there was significant increase in a-SMA protein level at day 3

Fig. 2. The abundance of rat Smad1 mRNA in HSCs after CCl4administration. Expression of Smad1 mRNA in primary rat HSCs at 1,2, 3 and 4 days after one dose administration of CCl4 or corn oil. Freshisolated HSCs were cultured for overnight only and total RNA wasextracted for real-time quantitative RT-PCR analysis. Onemicrogram of total RNA from HSCs at 1, 2, 3, and 4 days after CCl4 orcorn oil administration was employed for the assay. Data representmeanWSEM from six rats. MP< 0.05.

Fig. 3. Transient transfection and luciferase assay in 3 and 6 daysprimary cultured HSCs. Rat HSCs were isolated by pronase/collagenase method and cultured on plastic dishes for 3 or 6 days. At3 or 6 days, primary HSCs were transfected with pGL2-12xGCCG-CollagenX (pGL2-12xGCCG) or pGL2-Basic (pGL2) with Fugene6,respectively. The cells were also co-transfected with pRL-TK vectoras transfection efficiency control. The relative luciferase activity wasdefined as the ratio of Firefly luciferase activity over Renilla luciferaseactivity. Data represent meanWSEM from four independentexperiments. MP< 0.05

Fig. 4. Electrophoretic mobility shift assay (EMSA) of rat Smad1 inprimary cultured HSCs on days 3, 6 and 9. Rat HSCs were isolated bypronase/collagenase method and cultured on plastic dishes for 3, 6, or9 days. At 3, 6, or 9 days, rat primary HSCs were subjected to isolationof nuclear extract. The nuclear extracts (5 mg) from HSCs at 3, 6, or9 days were incubated with Smad1 consensus sequence 9xGCCG onice and then subjected to electrophoresis. Smad1 binding intensityincreased from days 3 to 9 after in vitro culture. The specificity ofSmad1 binding was documented by either incubation of Smad1antibody or unlabelled Smad1 consensus sequence 9xGCCG motif onice before incubation with labeled probe. NE represents nuclearextract.

H E P A T I C S T E L L A T E C E L L A C T I V A T I O N 767

afterCCl4 administration indicating a trans-differentiation of ratHSCs from quiescent phenotype to myofibroblast-likephenotype. Moreover, Smad1 protein expression wasincreased significantly. A significant elevation of Smad1 wasobserved at day 2 after CCl4 administration, which was earlierthan that of a-SMA. The increase in Smad1 was alsodemonstrated at mRNA level starting from day 2 and peakingon day 3 after CCl4 administration. The real-time quantitativeRT-PCR demonstrated 1.7-and 2.1-fold increases in Smad1mRNA abundance on days 2 and 3 in CCl4 treated rats ascompare to that of corn oil treated rats, respectively (Fig. 2). Inaddition, there were no changes of a-SMA and Smad1 proteinlevels in corn oil treated rats (Figs. 1 and 2).

Transcriptional activity of Smad1 during in vitroactivation of HSCs

To demonstrate whether there was an increase in Smad1transcriptional activity during HSC activation, we performedtransient transfection with luciferase assay and electrophoreticmobility shift assay (EMSA) with primary cultured HSCs. ThepGL2 luciferase plasmid including 12xGCCG motifs in front ofmouse type X collagen promoter (pGL2-12xGCCG) or pGL2promoterless plasmid (pGL2-Basic) was transiently transfectedinto rat primary isolated HSCs with FuGene-6 for 48 h, Smad1transcriptional activity was examined at days 3 and 6 primarycultured HSCs, respectively. Although transfection efficiencywas low in HSCs, detectable luciferase activities weredetermined in these cells. After being normalized with Renillaluciferase reading, the Firefly luciferase activity was about one-fold higher in day 6 HSCs than that in day 3 HSCs (Fig. 3). Tofurther demonstrate the transcriptional activity of Smad1, theg-32P-labeled Smad1DNA binding consensus motif (9xGCCG)was employed to incubatewith 5mg of nuclear extracts isolatedfrom primary cultured HSCs on days 3, 6 and 9. The bindingintensity of Smad1 DNA binding consensus motif (9xGCCG)and HSCs nuclear extracts gradually increased and peaked at

JOURNAL OF CELLULAR PHYSIOLOGY DOI 10.1002/JCP

day 9. This result indicated that the Smad1 displayed the highesttranscriptional activity in day 9 primary cultured HSCs.Moreover, the band intensity was decreased after an addition ofspecific antibody against Smad1, indicating that the 9xGCCGconsensus motif was bound with Smad1. Furthermore,competition with different concentrations of unlabeled9xGCCG motifs was observed in the gel-shift assay (Fig. 4)indicating the specificity of Smad1 binding.

Fig. 5. Smad1 expression in CFSC-8B after infection of recombinant adenoviruses. CFSC-8B cells were infected with different adenoviruses for72h.Attheendof infection,totalRNAandproteinwereextractedforRT-PCRandWesternblotanalyses,respectively.AdenoviruscontainingGFPwas used as control for Smad1 and non-specific short-hairpin RNA (shRNA) was employed as control for Smad1 shRNA. One microgram of totalRNAwasusedforRT-PCRassayasshowninpanelAwhile50mgoftotalcellularproteinwasusedforWesternblotanalysisasshowninpanelB(blackbar indicates the resultof Smad1 andwhite bar indicates the resultofa-SMA). Both GAPDH mRNAand protein were employedas loading controlfor RT-PCR and Western blot analyses, respectively. Data represent meanWSEM from four independent experiments. MP< 0.05.

768 S H E N E T A L .

Recombinant adenoviruses construction and transienttransfection Smad1 in HSC line (CFSC-8B)

To further demonstrate the biological activity of Smad1 inHSCs, we constructed the adenoviruses expressing Smad1(Adeno-Smad1) and short-hairpin RNA of Smad1(Adeno-Smad1-shRNA). After CFSC-8B cells were infectedwith Adeno-Smad1 and Adeno-Smad1-shRNA for 72 h in thepresence of 10% FBS, there was a significant increase in Smad1mRNA expression in the cells infected with Adeno-Smad1while a significant decrease (60%) in Smad1 mRNA expressionwas documented in the cells infected with Adeno-Smad1-shRNA (Fig. 5A and B). After isolation of cytoplasmic proteinfrom these cells, Western blot demonstrated that there wasan increase or decrease of Smad1 protein in the cells infectedwith either Adeno-Smad1 or Adeno-Smad1-shRNA,respectively.

Smad1 regulation of CFSC-8B cell trans-differentiation

To investigate whether Adeno-Smad1 induced HSCtrans-differentiation, we examined the abundance of a-SMAprotein in CFSC-8B cells, which were infected withAdeno-Smad1 or Adeno-Smad1-sh-RNA. As shown inFig. 5B, there was a significant increase in a-SMA in the cellsinfected with Adeno-Smad1 while a significant decrease ina-SMA was documented in the cells infected withAdeno-Smad1-shRNA. Moreover, morphology of CFSC-8Bcells was examined after infection of different adenoviruses. Asshown in Figure 6, infection of GFP did not changeCFSC-8B cellmorphology (Fig. 6B), while infectionAdeno-Smad1-shRNAdidcause CFSC-8B cells to cluster together with a big hole among

JOURNAL OF CELLULAR PHYSIOLOGY DOI 10.1002/JCP

cells (Fig. 6D). The change of CFSC-8B cells after infection ofAdeno-Smad1 was not significant (Fig. 6C).

Discussion

Hepatic stellate cells represent the major fibrogenic cell typein the liver. Activation of HSCs is a key event in hepaticfibrogenesis and includes two steps of initiation andperpetuation (Friedman, 1996). Several factors have beenidentified to play a critical role during activation ofHSCs, such asinflammatory cells and cytokines. In our previous study, we haveidentified that BMPs significantly increaseda-SMA abundance inHSCs, indicating a role of BMPs in HSC trans-differentiation(Shenet al., 2003b). BMPandTGF-b1belong to the sameTGF-bsuperfamily. It is generally considered that TGF-b1 is the majorinducer of the myofibroblastic-like phenotype and the principalprofibrogenic cytokine (Rosenbaum et al., 1995; Mezzano et al.,2000). Molecular signaling of BMP and TGF-b1 are mediated byintracellular proteins—Smads, therefore, the aim of currentstudy is to delineate the role of Smads especially Smad1 in theactivation of HSCs.

The expressions of Smad2, Smad3 and Smad4 during HSCactivation were examined. It was demonstrated that TGF-binduced phosphorylation and nuclear translocation of Smad2and Smad3 occurred in primary cultured HSCs but not in sub-cultured cells (Dooley et al., 2001). Moreover, by employing animmortalized cell line of activated rat HSCs, it wasdemonstrated that Smad2 was activated in response to TGF-bwhereas Smad3 was constitutively activated (Inagaki et al.,2001). Furthermore, in an in vitro model of HSC activation, theactivation ofHSCswas associatedwith a shift in TGF-b signalingpathway. It was documented that TGF-b activated Smad2 in

Fig. 6. Morphology of CFSC-8B cells after infection of recombinant adenoviruses. CFSC-8B cells were infected with different adenoviruses for72h. At the end of infection, cell morphology was photographed. Infection of shRNA of Smad1 changed CFSC-8B cells into a cluster of cells with bigholeinbetween,whichresemblesquiescentphenotypeofhepaticstellatecells.A: indicatesnon-infectedcells,B:representsCFSC-8Bcells infectedwith Adeno-GFP, C: indicates CFSC-8B cells infected with Adeno-Smad1, and D: represents CFSC-8B cells infected with Adeno-Smad1-shRNA.

H E P A T I C S T E L L A T E C E L L A C T I V A T I O N 769

quiescent cells and Smad3 in activated cells, respectively(Liu et al., 2003). By employing Smad3 null mice, it is clearlydemonstrated that Smad3 was not necessary for HSCactivation as assessed by a-SMA expression, but was necessaryfor inhibition of HSC proliferation by TGF-b (Schnabl et al.,2001).

Common-mediated Smad–Smad4 is required fortranslocation of heteromeric complex of R-Smads. Althoughthere was an increase in expression Smad1 during HSCactivation, no increased expression of Smad4 was observed inour experiment (data no shown). Similar observation wasnoticed by several investigators (Liu et al., 2003; Stopa et al.,2000). They observed transcription of Smad4 in quiescent andmyofibroblast-like HSCs and expression level of Smad4 wassimilar in both quiescent and activated HSCs. Moreover,TGF-b induced Smad2nuclear translocationwas observed in anin vitromodel of HSC activation and an immortalized cell line ofactivated HSCs (Inagaki et al., 2001). Although there was nonotable increase in expression of Smad4 during HSC activation,there was an significant increase in Smad4 in the nucleus ofactivated HSCs and immortalized cell line of activated HSCs(Inagaki et al., 2001; Liu et al., 2003). This suggests that Smad1might be able to associate with Smad4 in the nucleus as well asthe cytoplasm. Furthermore, our results indicated a significantelevation of Smad1 transcriptional activity during the in vitroactivation of HSCs.

After identification of Smad1 as an intracellular signalingprotein of decapentaplegic (a homologue of TGF-b inDrosophila), it was demonstrated that Smad1was involved in theintracellular BMP signals, which inhibited myogenicdifferentiation and induced osteoblast differentiation in C2C12cells (Yamamoto et al., 1997). In our previous study, increasedphosphorylation and nuclear translocation of Smad1 wereobserved during in vitro activation of HSCs (Shen et al., 2003a).In current study, we further demonstrate that there was notonly increased expression of Smad1 but also increasedtranscriptional activity of Smad1 during HSC activation.Regulation of Smad1 in tissue is not widely examined with few

JOURNAL OF CELLULAR PHYSIOLOGY DOI 10.1002/JCP

publications that demonstrated regulation of Smad1 by TGF-b1(Shen et al., 2003a). In current study, we observed that Smad1mRNA abundance was lower at day 4 after CCl4 administrationthan that at day 3 and mRNA level of collagen I (a1) was alsohigher at day 3 after CCl4 administration than that at day 4 (datanot shown). These observations are consistentwith the findingsof the other groups, which describe transient expression of a-SMA and collagen I (a1) after CCl4 administration(Kalinichenko et al., 2003; Friedman, 2000). Considering thatthere was an increased expression of TGF-b1 during HSCactivation (Hellerbrand et al., 1999; Yu et al., 2003), thedecreased mRNA level of Smad1 could be due to TGF-b1inhibition of Smad1 expression. Moreover, Smad1 wasconsidered as a convergent intracellular signaling mediator forSmad andMAPKpathway, whichmay indicate an important roleof Smad1 in cell differentiation and proliferation (Aubin et al.,2004). Targeted deletion of the Smad1 gene results in earlyembryonic lethality due to the failure of the allantois to fuse tothe chorion (Lechleider et al., 2001). Our adenovirus infectionstudy clearly indicated that Smad1 promoted HSCdifferentiation.

The enhancement of Smad1 transcriptional activity duringHSC activation indicates an increase or decrease in differentgene expression, which are regulated by Smad1. As shown inour result that abundance of a-SMAwas significantly increased.However, what is the mechanism involved in the Smad1increased a-SMA expression remained to be investigated. It isreported that regulation of a-SMA expression was dependenton CArG elements within the 50 and first intron promoterregion (Mack and Owens, 1999). Moreover, TGF-b1 was ableto induce a-SMA expression in aortic smooth muscle cellsthrough two CArG elements (Hautmann et al., 1997). Thisinformation indicates that Smad1 may regulate a-SMAexpression through TGF-b1 responsible CArG elements ona-SMA promoter region.

In conclusion, we have documented that there are increasedexpression of Smad1 during in vivo activation of hepatic stellatecells and increased transcriptional activity of Smad1 during in

770 S H E N E T A L .

vitro activation of hepatic stellate cells. Increased expression ofSmad1 enhanced the trans-differentiation and activation ofhepatic stellate cells.

Acknowledgments

The research was supported by a grant from the CanadianInstitute of Health Research to Dr. Yuewen Gong and aFellowship from the Canadian Association for the Study of theLiver to Dr. Hong Shen.

Literature Cited

Attisano L, Wrana JL. 2000. Smads as transcriptional co-modulators. Curr Opin Cell Biol12:235–243.

Aubin J, Davy A, Soriano P. 2004. In vivo convergence of BMP and MAPK signaling pathways:Impact of differential Smad1 phosphorylation on development and homeostasis. GenesDev 18:1482–1494.

Bachem MG, Meyer D, Schafer W, Riess U, Melchior R, Sell KM, Gressner AM. 1993. Theresponse of rat liver perisinusoidal lipocytes to polypeptide growth regulator changes withtheir transdifferentiation into myofibroblast-like cells in culture. J Hepatol 18:40–52.

Bataller R, Schwabe RF, Choi YH, Yang L, Paik YH, Lindquist J, Qian T, Schoonhoven R,Hagedorn CH, Lemasters JJ, Brenner DA. 2003. NADPH oxidase signal transducesangiotensin II in hepatic stellate cells and is critical in hepatic fibrosis. J Clin Invest 112:1383–1394.

Blouin A, Bolender RP,Weibel ER. 1977. Distribution of organelles andmembranes betweenhepatocytes and nonhepatocytes in the rat liver parenchyma. A stereological study. J CellBiol 72:441–455.

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 32:32–38.

Buniatian G, Gebhardt R, Schrenk D, Hamprecht B. 1996. Colocalization of three types ofintermediate filament proteins in perisinusoidal stellate cells: Glial fibrillary acidic proteinas a new cellular marker. Eur J Cell Biol 70:23–32.

Burt AD, Robertson JL, Heir J, MacSween RN. 1986. Desmin-containing stellate cells in ratliver; distribution in normal animals and response to experimental acute liver injury.J Pathol 150:29–35.

Davis BH, Kresina TF. 1996. Hepatic fibrogenesis. Clin Lab Med 16:361–375.Derynck R. 1998. SMAD proteins and mammalian anatomy. Nature 393:737–739.Dooley S, Streckert M, Delvoux B, Gressner AM. 2001. Expression of Smads during in vitrotransdifferentiation of hepatic stellate cells to myofibroblasts. Biochem Biophys ResCommun 283:554–562.

Feng XH, Zhang Y, Wu RY, Derynck R. 1998. The tumor suppressor Smad4/DPC4 andtranscriptional adaptor CBP/p300 are coactivators for smad3 in TGF-beta-inducedtranscriptional activation. Genes Dev 12:2153–2163.

Friedman SL. 1993. Seminars inmedicine of theBeth IsraelHospital, Boston. The cellular basisof hepatic fibrosis. Mechanisms and treatment strategies. N Engl J Med 328:1828–1835.

Friedman SL. 1996. Hepatic stellate cells. Prog Liver Dis 14:101–130.Friedman SL. 2000.Molecular regulation of hepatic fibrosis, an integrated cellular response totissue injury. J Biol Chem 275:2247–2250.

Friedman SL, Rockey DC, McGuire RF, Maher JJ, Boyles JK, Yamasaki G. 1992. Isolatedhepatic lipocytes and Kupffer cells from normal human liver: Morphological and functionalcharacteristics in primary culture. Hepatology 15:234–243.

Fryer HJ, Davis GE, Manthorpe M, Varon S. 1986. Lowry protein assay using an automaticmicrotiter plate spectrophotometer. Anal Biochem 153:262–266.

Hautmann MB, Madsen CS, Owens GK. 1997. A transforming growth factor beta (TGFbeta)control element drives TGFbeta-induced stimulation of smooth muscle alpha-actin geneexpression in concert with two CArG elements. J Biol Chem 272:10948–10956.

Hellerbrand C, Stefanovic B, Giordano F, Burchardt ER, Brenner DA. 1999. The role ofTGFbeta1 in initiating hepatic stellate cell activation in vivo. J Hepatol 30:77–87.

Hendriks HF, VerhoofstadWA, Brouwer A, de Leeuw AM, Knook DL. 1985. Perisinusoidalfat-storing cells are the main vitamin A storage sites in rat liver. Exp Cell Res 160:138–149.

Henningfeld KA, Rastegar S, Adler G, Knochel W. 2000. Smad1 and Smad4 are componentsof the bone morphogenetic protein-4 (BMP-4)-induced transcription complex of theXvent-2B promoter. J Biol Chem 275:21827–21835.

Hyman CA, Bartholin L, Newfeld SJ, Wotton D. 2003. Drosophila TGIF proteins aretranscriptional activators. Mol Cell Biol 23:9262–9274.

Imai K, SenooH. 1998.Morphologyof sites of adhesionbetween hepatic stellate cells (vitaminA-storing cells) and a three-dimensional extracellular matrix. Anat Rec 250:430–437.

Inagaki Y, Mamura M, Kanamaru Y, Greenwel P, Nemoto T, Takehara K, Ten Dijke P, NakaoA. 2001. Constitutive phosphorylation and nuclear localization of Smad3 are correlatedwith increased collagen gene transcription in activated hepatic stellate cells. J Cell Physiol187:117–123.

JOURNAL OF CELLULAR PHYSIOLOGY DOI 10.1002/JCP

Inagaki Y, Truter S, Greenwel P, Rojkind M, Unoura M, Kobayashi K, Ramirez F. 1995.Regulation of the alpha 2(I) collagen gene transcription in fat-storing cells derived from acirrhotic liver. Hepatology 22:573–579.

Kalinichenko VV, Bhattacharyya D, Zhou Y, Gusarova GA, KimW, Shin B, Costa RH. 2003.Foxf1 þ/� mice exhibit defective stellate cell activation and abnormal liver regenerationfollowing CCl4 injury. Hepatology 37:107–117.

Klinger W, Devereux T, Fouts JR. 1988. Functional and structural zonal hepatocyteheterogeneity—dynamics and ontogenic development. Exp Pathol 35:69–91.

Kmiec Z. 2001. Cooperation of liver cells in health and disease. Adv Anat Embryol Cell Biol161:1–151.

Kusanagi K, Inoue H, Ishidou Y, Mishima HK, Kawabata M, Miyazono K. 2000.Characterization of a bonemorphogenetic protein-responsive Smad-binding element. MolBiol Cell 11:555–565.

Langlands K, Yin X, Anand G, Prochownik EV. 1997. Differential interactions of Id proteinswith basic-helix–loop-helix transcription factors. J Biol Chem 272:19785–19793.

Lechleider RJ, Ryan JL, Garrett L, Eng C, Deng C, Wynshaw-Boris A, Roberts AB. 2001.Targetedmutagenesis of Smad1 reveals an essential role in chorioallantoic fusion. Dev Biol240:157–167.

Liu C, Gaca MD, Swenson ES, Vellucci VF, Reiss M, Wells RG. 2003. Smads 2 and 3 aredifferentially activated by transforming growth factor-beta (TGF-beta) in quiescent andactivated hepatic stellate cells. Constitutive nuclear localization of Smads in activated cellsis TGF-beta-independent. J Biol Chem 278:11721–11728.

Liu X, Yue J, Frey RS, Zhu Q, Mulder KM. 1998. Transforming growth factor beta signalingthrough Smad1 in human breast cancer cells. Cancer Res 58:4752–4757.

Mack CP, Owens GK. 1999. Regulation of smooth muscle alpha-actin expression in vivo isdependent on CArG elements within the 5’ and first intron promoter regions. Circ Res84:852–861.

Marra F. 2002. Chemokines in liver inflammation and fibrosis. Front Biosci 7:d1899–d1914.Marra F, DeloguW, Petrai I, Pastacaldi S, Bonacchi A, Efsen E, Aleffi S, Bertolani C, PinzaniM, Gentilini P. 2004. Differential requirement of members of the MAPK family for CCL2expression by hepatic stellate cells. Am J Physiol Gastrointest Liver Physiol 287:G18–G26.

Massague J,WottonD. 2000. Transcriptional control by theTGF-beta/Smad signaling system.EMBO J 19:1745–1754.

Mezzano SA, Droguett MA, Burgos ME, Ardiles LG, Aros CA, Caorsi I, Egido J. 2000.Overexpression of chemokines, fibrogenic cytokines, and myofibroblasts in humanmembranous nephropathy. Kidney Int 57:147–158.

Miyazono K, ten Dijke P, Heldin CH. 2000. TGF-beta signaling by Smad proteins. AdvImmunol 75:115–157.

Neubauer K, Knittel T, Aurisch S, Fellmer P, Ramadori G. 1996. Glial fibrillary acidicprotein—a cell type specific marker for Ito cells in vivo and in vitro. J Hepatol 24:719–730.

Park GT, Morasso MI. 2002. Bone morphogenetic protein-2 (BMP-2) transactivates Dlx3through Smad1 and Smad4: Alternative mode for Dlx3 induction in mouse keratinocytes.Nucleic Acids Res 30:515–522.

Pinzani M. 1995. Novel insights into the biology and physiology of the Ito cell. Pharmacol Ther66:387–412.

Pinzani M, Gentilini P. 1999. Biology of hepatic stellate cells and their possible relevance in thepathogenesis of portal hypertension in cirrhosis. Semin Liver Dis 19:397–410.

Rosenbaum J, Blazejewski S, Preaux AM, Mallat A, Dhumeaux D, Mavier P. 1995. Fibroblastgrowth factor 2 and transforming growth factor beta 1 interactions in human livermyofibroblasts. Gastroenterology 109:1986–1996.

Schnabl B, Kweon YO, Frederick JP, Wang XF, Rippe RA, Brenner DA. 2001. The role ofSmad3 in mediating mouse hepatic stellate cell activation. Hepatology 34:89–100.

ShenH,HuangG,HadiM,Choy P,ZhangM,MinukGY,ChenY,GongY. 2003a. Transforminggrowth factor-beta1 downregulation of Smad1 gene expression in rat hepatic stellate cells.Am J Physiol Gastrointest Liver Physiol 285:G539–G546.

Shen H, Huang GJ, Gong YW. 2003b. Effect of transforming growth factor beta and bonemorphogenetic proteins on rat hepatic stellate cell proliferation and trans-differentiation.World J Gastroenterol 9:784–787.

Shen H, Zhang M, Minuk GY, Gong Y. 2002. Different effects of rat interferon alpha, beta andgamma on rat hepatic stellate cell proliferation and activation. BMC Cell Biol 3:9.

Stopa M, Anhuf D, Terstegen L, Gatsios P, Gressner AM, Dooley S. 2000. Participation ofSmad2, Smad3, and Smad4 in transforming growth factor beta (TGF-beta)-inducedactivation of Smad7. THETGF-beta response element of the promoter requires functionalSmad binding element and E-box sequences for transcriptional regulation. J Biol Chem275:29308–29317.

YamamotoN, Akiyama S, Katagiri T, NamikiM, KurokawaT, Suda T. 1997. Smad1 and smad5act downstream of intracellular signalings of BMP-2 that inhibits myogenic differentiationand induces osteoblast differentiation in C2C12 myoblasts. Biochem Biophys ResCommun 238:574–580.

YangC, ZeisbergM,MostermanB, Sudhakar A, Yerramalla U, Holthaus K, Xu L, Eng F, AfdhalN, Kalluri R. 2003. Liver fibrosis: Insights intomigration of hepatic stellate cells in responseto extracellular matrix and growth factors. Gastroenterology 124:147–159.

Yu C,Wang F, Jin C, Huang X, Miller DL, Basilico C, McKeehanWL. 2003. Role of fibroblastgrowth factor type 1 and 2 in carbon tetrachloride-induced hepatic injury and fibrogenesis.Am J Pathol 163:1653–1662.