AJP - GI Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Gastrointest Liver Physiol 293: G154-G164, 2007. First published May 17, 2007; doi:10.1152/ajpgi.00432.2006
0193-1857/07 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
293/1/G154    most recent
00432.2006v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dodig, M.
Right arrow Articles by McCullough, A. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dodig, M.
Right arrow Articles by McCullough, A. J.

LIVER AND BILIARY TRACT

Differences in regulation of type I collagen synthesis in primary and passaged hepatic stellate cell cultures: the role of {alpha}5beta1-integrin

Milan Dodig,1 Ben Ogunwale,2 Srinivasan Dasarathy,1 Min Li,1 Bingcheng Wang,2,3 and Arthur J. McCullough1

1Department of Gastroenterology and Hepatology, Cleveland Clinic and Cleveland Clinic Lerner College of Medicine; 2Division of Gastroenterology, MetroHealth Medical Center, School of Medicine at Case Western Reserve University; and 3Rammelkamp Research Center, MetroHealth Medical Center, Cleveland, Ohio

Submitted 20 September 2006 ; accepted in final form 10 May 2007


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Hepatic stellate cells (HSC) differ in their phenotype depending on the initiation and progression of their activation. Our hypothesis was that different mechanisms govern type I collagen synthesis depending on stage of HSC activation. We investigated the role of {alpha}5beta1-integrin as a regulator of type I collagen gene COL1A1 expression in primary and passaged HSC cultures using transgenic mouse containing type I collagen gene COL1A1 promoter linked to the chloramphenicol acetyltransferase (CAT) reporter gene. The {alpha}5beta1 protein levels increased during the activation and were highest in day 6 primary cultures but decreased in passaged HSC. CAT activity, reflecting COL1A1 expression, was upregulated by {alpha}5beta1-integrin. Inhibition of {alpha}5beta1-integrin by echistatin and blocking antibody resulted in reduced transgene activity only in early primary cultures (compared with the control, 53.3 ± 12% echistatin and 58.8 ± 7% blocking antibody, respectively, P < 0.05). Treatment of passaged HSC with either echistatin or blocking antibody had no effect. Fibronectin, an {alpha}5beta1-integrin ligand, increased transgene activity in primary (210 ± 33%, P < 0.05) but not in passaged HSC cultures (119 ± 8%). This {alpha}5beta1-integrin effect appears to be at least in part mediated by CCAAT enhancer binding protein-beta (C/EBPbeta), because fibronectin increased and {alpha}5-gene silencing by small interfering RNA decreased C/EBPbeta levels. In addition, C/EBPbeta knockout mice showed reduced type I collagen synthesis compared with wild-type littermates. Therefore {alpha}5beta1-integrin is an important regulator of type I collagen production in early primary HSC cultures but appears to have no direct role once the HSC are fully activated.

integrins; C/EBPbeta; {alpha}5-integrin siRNA

ACTIVATION OF HEPATIC STELLATE CELLS (HSC) is a dynamic process, commonly divided into an early initiation phase and a late perpetuation phase (33). Molecular mechanisms governing HSC proliferation and extracellular matrix (ECM) synthesis differ in activated HSC depending on how far they are in the progression of the activated phenotype. Older, nonreplicating cells have increased expression of inflammatory genes and decreased expression of ECM genes compared with younger, replicating HSC (35).

We hypothesized that activated HSC employ different mechanisms to regulate type I collagen synthesis depending on the stage of activation. Animal experiments studying spontaneous regression of early and advanced liver fibrosis showed less reversal of fibrosis with longer duration of injury (34). Clinical data also suggest that early stages of fibrosis may be reversible whereas late, more advanced cirrhosis demonstrates little or no reversibility (28). Although this phenomenon has been postulated to result from lower collagenase activity in advanced cirrhosis (24), differences in the regulation of type I collagen transcription could also play a role in reversibility of hepatic fibrosis. This may be achieved by two possible mechanisms. Cellular dedifferentiation in other cell types has been linked to activation or silencing of various nuclear transcription factors (41). During dedifferentiation, cis-elements in the gene promoters become accessible to previously inactive transcription factors by chromatin reorganization and modifications of the histone proteins (46). Alternatively, nuclear transcription factors, not present before, may become induced by changes in cellular phenotype. These changes can affect transcription of genes encoding ECM proteins (26).

The main difference between early and late liver fibrosis is the composition of the hepatic ECM. ECM serves as a reservoir of biologically active soluble molecules, including cytokines that regulate HSC activation. Different structural matrix components also have the potential to stimulate or curtail HSC activation (38, 43), and interactions between ECM composition and cell phenotype have been identified in a variety of tissues (22, 39, 42).

In an effort to identify the mechanism responsible for the differences in behavior of activated HSC, we elected to examine fibronectin, a component of the ECM, and its corresponding membrane receptor. Fibronectin is an ECM molecule that stimulates HSC activation. At the time of liver injury, fibronectin is released into the circulation as well as into the local extracellular milieu by injured hepatocytes, endothelial cells, and activating HSC (31). Acetaldehyde and TGF-beta, two important stimulants of HSC activation, have been found to stimulate fibronectin expression by HSC (10). In an animal model of chronic alcohol-induced liver injury, fibronectin is deposited in the liver in increasing quantities prior to HSC activation (40). Therefore we hypothesized that fibronectin-mediated signaling contributes to HSC activation and type I collagen synthesis during the early phase of the fibrotic process, but not in the later stages when HSC activation has been established and liver ECM has been rearranged. Changes in fibronectin content of the ECM are signaled to the HSC through the fibronectin receptor, {alpha}5beta1-integrin, which plays a role in both HSC attachment to fibronectin and in the activation process (16, 30). Additional evidence for this role of {alpha}5beta1-integrin in HSC activation is derived from studies on blockade of {alpha}5beta1-integrin by Arg-Gly-Asp peptides that ameliorates CCl4-induced liver fibrosis (21).

Integrins are adhesion receptors located on the cell surface and the major membrane proteins involved in cell-ECM interactions (8). Each integrin is a heterodimer consisting of noncovalently bound {alpha}- and beta-subunits. The ligand specificity depends primarily on the extracellular component of the {alpha}-subunit. However, both subunits are required for an integrin to function. Multiple integrins have been reported to exist on the HSC membrane, including the fibronectin receptor {alpha}5beta1-integrin (4). Integrins have also been implicated as regulators of type I collagen synthesis in other tissues and organs (1, 9).

The mechanism of increased expression of type I collagen COL1A1 gene by {alpha}5beta1-integrin is unclear. We hypothesized that {alpha}5beta1-mediated signaling may upregulate expression of a transcriptional enhancer in activating HSC. One such transcription factor previously described to play a role in a type I collagen expression is CCAAT enhancer binding protein-beta (C/EBPbeta). In HSC, C/EBPbeta has been reported to mediate acetaldehyde and hydrogen peroxide-induced upregulation of type I collagen transcription (2, 11).

In the present study we examined the relationship between {alpha}5beta1-integrin and type I collagen expression in primary (day 6) and passaged HSC cultures. Our data suggest that type I collagen expression is regulated by different mechanisms, depending on the phase of HSC activation. {alpha}5beta1-Integrin is important for high levels of type I collagen expression in primary cultures. However, passaged HSC do not require either fibronectin or {alpha}5beta1-integrin signaling to maintain high levels of type I collagen expression. Our data also suggest that the {alpha}5beta1-integrin effect on type I collagen synthesis in primary HSC cultures is in part mediated by the transcription factor C/EBPbeta.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Animals. Transgenic mice harboring construct pOBColCAT3.6 were used to evaluate type I collagen expression in HSC. pOBColCAT3.6 contains 3.6 kb of rat type I collagen gene COL1A1 promoter with parts of the first intron, driving expression of the reporter gene chloramphenicol acetyltransferase (CAT). Another transgenic line harboring a similar construct containing the same upstream regulatory COL1A1 promoter sequence but with green fluorescence protein as a reporter gene (pOBColGFP3.6) was used to evaluate number of type I collagen-expressing cells in the culture upon activation. Transgenic mice and measurements of reporter gene were used to focus on changes in COL1A1 gene expression independent of any posttranscriptional and posttranslational processing involved in deposition of type I collagen by HSC. The use of transgenic animals simplified type I collagen expression measurement and permits quantification of changes following experimental manipulations. Animals were kindly provided by Dr. David Rowe and Dr. Alexander Lichtler from University of Connecticut Health Center, Farmington, CT. These transgenic lines have been characterized previously, and promoter constructs have been shown to support endogenous type I collagen like tissue expression pattern (3, 17). In addition, our data indicate that changes in the reporter gene closely resemble changes in mature collagen synthesis estimated by Western blot analysis of the growth media.

C/EBPbeta –/– mice were used to investigate the role of C/EBPbeta in HSC activation and resultant type I collagen transcription. C/EBPbeta –/– mice and heterozygous (wild-type phenotype) littermates were generated by Dr. Valeria Poli (6) and provided by Dr. Colleen M. Croniger and Dr. Richard W. Hanson from Case Western Reserve University, Cleveland, OH. The generation of C/EBPbeta –/– mice and their genetic background has been described previously (6).

All animal experiments were reviewed and approved by Institutional Animal Care and Use Committee at Case Western Reserve University, Cleveland, OH.

HSC isolation and culture. Primary HSC were isolated from the liver as previously described (13). Briefly, after animals were killed, livers were removed, washed in ice-cold HBSS (HyClone, Logan, UT), minced, and incubated in HBSS without calcium and magnesium with 0.2% Pronase (Sigma, St. Louis, MO) and 0.02% Collagenase H (Sigma) for 30 min at 37°C. The digest was passed through a fine wire mesh and the resultant cell suspension was washed three times with HBSS. HSC were further purified by density gradient centrifugation. Cells were resuspended in HBSS with 3% BSA (Sigma) and mixed with Histodenz (Sigma) in HBSS without NaCl. This mixture was then layered under HBSS with 3% BSA. After centrifugation, the white interphase layer containing HSC was removed, washed, and plated at 250,000 cells/cm2 in DMEM with 10% fetal calf serum and 10% horse serum. All media and sera for cell culture were purchased from Hyclone (Logan, UT). The medium was changed daily during the experiment. Cells were passaged on days 7-8 of primary cultures and plated at 25,000/cm2 density. These cells were actively replicating without signs of replicative senescence, secreted large amount of type I collagen, and exhibited all other features of the fully activated phenotype.

The purity of the cultures was evaluated by morphology, vitamin A autofluorescence at 328 nm on day 1 of culture, {alpha}-smooth muscle actin ({alpha}SMA), desmin, and glial fibrillary acidic protein (GFAP) immunofluorescence. In addition, activated cells isolated from 3.6ColGFP transgenic mice were examined for fluorescence daily during the culture period.

Fibronectin (Sigma) was used to coat culture dishes in a concentration of 10 µg/ml for 24 h before subgroups of either primary or secondary HSC cultures were plated. Echistatin (Sigma) in doses of 100, 10, and 1 nM, as well as {alpha}5beta1 blocking antibody (Chemicon, Temecula, CA) (10 µl/ml), was added to primary cultures on day 4, and cells were harvested after 24 h. For passaged HSC, echistatin and {alpha}5beta1 blocking antibody were added to secondary cultures 3 days after passage.

{alpha}5-integrin gene silencing in HSC culture. HSC were isolated as described above. At 24 h after plating, HSC were transfected with 10 nM of {alpha}5-integrin small interfering RNA (siRNA) in HiPerFect transfection reagent. Cells were harvested and analyzed at day 5 of culture. The {alpha}5-integrin, type I collagen, and C/EBPbeta protein levels were determined in the cell lysate. Both {alpha}5-integrin siRNA and transfection reagent were purchased from Qiagen (Valencia, CA) and used according to the manufacturer's recommendations.

RNA isolation and real-time RT-PCR. Total RNA was isolated using TRI reagent (Sigma). Reverse transcriptase kit was purchased from Invitrogen (Invitrogen, Carlsbad, CA) and used to generate cDNA per manufacturer recommendations. Specific primers were designed and analyzed by use of commercial software (Light Cycler Probe Design, Roche Diagnostics). The primer sequences for {alpha}5-integrin are as follows: {alpha}5-forward primer 5'-ACCAAGACGGCTACAATGATG-3' and {alpha}5-reverse primer 5'-CTGCTTGGAAGTCAGGAACAG-3'. The GAPDH primers are as follows: GAPDH forward 5'-GATGACATCAAGAAGGTGGTGA-3'and GAPDH reverse 5'-GGTCCAGGGTTTCTTACTCCTT-3'. Primers for COL1A1 gene were COL1A1 forward 5'-ATGTTCAGCTTTGTGGACCTC-3' and COL1A1 reverse 5'-AGTTTGAAGCACAGCACTCG-3'.

Real-time PCR for quantification of RNA was carried out by using SYBR protocol on the fluorescence temperature cycler (Light Cycler; Roche Molecular Diagnostics, Indianapolis, IN). The reaction conditions were optimized at different temperature ranges and magnesium concentrations. Real-time reactions were carried out in duplicate, and amplicons were analyzed by generating melting curves with continuous measurement of fluorescence. Results were calculated as relative differences in target threshold cycle values normalized to GAPDH. All real-time PCR products were separated on a 1.5% Tris-acetic acid agarose gel to confirm product presence and size. Values were expressed as fold increase in the target mRNA at days 4, 6, and 8 of primary culture compared with day 2.

CAT assay. CAT assay was performed by using a standard fluor diffusion protocol. Cells were scraped from the culture in 0.3 ml of extraction buffer (0.25 M Tris·HCl, pH 7.8, containing 0.5% Triton X-100), subjected to three cycles of freezing and thawing, and then heated at 65°C for 15 min to inactivate endogenous deacetylases. After centrifugation the supernatants were used to determine CAT activity and protein concentration. CAT activity was measured by a modified fluor diffusion assay (23) and normalized to the protein content of the extract as measured by the BCA assay (Pierce, Rockford, IL).

Proliferation assay. Proliferation of the HSC was measured with Alamar blue (Serotec, Westbury, NY) according to manufacturer recommendations. We added 900 µl of fresh medium and 100 µl of Alamar blue to 12-well cell culture plates, mixed gently, and incubated for 4 h. Cell proliferation was determined by a colorimetric change in the medium measured as a difference in culture media absorbance at 570 and 600 nm. Using primary and passaged HSC cultures, we demonstrated that this method showed a high correlation with the total cell number in culture (r2 = 0.992, P < 0.01) and bromodeoxyuridine incorporation (r2 = 0.998, P < 0.001) (data not shown). This corresponded to the manufacturer's data supplied in the Alamar Blue Datasheet and previously published studies (7).

Protein isolation and Western blot analysis. Primary HSC cultures on days 2, 3, 4, 5, 6, 7, and 9, as well as passaged HSC cultures, were lysed with ice cold lysis buffer containing 25 mM Tris pH 7.4, 50 mM NaCl, 25 mM NaF, 10% glycerol, 1% Triton X-100, PMSF (1 mM), protease inhibitors cocktails 10 µl/ml (Sigma no. P8340, Sigma); scraped; and left on ice for 30 min. Protein concentration was determined by DC protein assay (Bio-Rad, Hercules, CA). Equal amounts of protein samples were resolved on a 7.5% SDS-PAGE under reducing conditions and then electrotransferred to polyvinylidene difluoride membrane. For {alpha}5beta1-integrin analysis, the membrane was blocked with 5% nonfat milk and 2% BSA in TBS with 0.1% Tween, incubated with goat polyclonal antibody to {alpha}5beta1 (Chemicon, Temecula, CA), washed, and then incubated with horseradish peroxidase-conjugated bovine anti-goat IgG antibody (Santa Cruz Biotechnology, Santa Cruz, CA). For {alpha}5-integrin assay, the membrane was blocked with 5% nonfat milk, in TBS with 0.1% Tween. The membrane was subsequently incubated with rabbit polyclonal antibody to {alpha}5-integrin, washed, and then incubated with horseradish peroxidase-conjugated bovine anti-rabbit IgG (Santa Cruz Biotechnology). The blots were visualized with the ECL detection (Amersham, Piscataway, NJ). A similar procedure was followed for Western blot analysis for C/EBPbeta protein (antibody SC-150X, Santa Cruz Biotechnology); the only difference was that samples were loaded on 12% SDS polyacrylamide gel.

Western blot for type I collagen was performed on cellular extracts as well as on harvested growth media. Cellular extracts from HSC were prepared by lysing HSC at designated time points in 2x sample buffer [0.125 M Tris pH 7.0, 4% SDS, 20% glycerol, 1 mM PMSF, 10 µl/ml protease inhibitor cocktail (Sigma), 2 mM activated Naorthovanadate]. For collection of growth medium, cells were incubated in serum-free DMEM enriched with 50 µg/ml ascorbic acid (Sigma) for 24 h. The medium was collected and spun through Ultracel YM-10 centrifugal filter unit (Millipore, Billerica, MA). After measurement of protein concentration, 1 mg of total protein was precipitated and loaded on 7.5% SDS polyacrylamide gel. Further Western blot analysis was performed as outlined above. Antibody against mature type I collagen (SC-8788) was used for analysis of growth media, and antibody against type I procollagen (SC-8787) was used for analysis of cell extracts.

Immunostaining. Desmin, {alpha}SMA, and GFAP antibodies for immunofluorescence were purchased from Sigma. HSC were grown on polylysine-treated slides and fixed in 3.7% formaldehyde in PBS. Slides were incubated in primary antibody overnight at 4°C. After incubation with secondary antibody, labeled with fluorophore, immunofluorescence was visualized by fluorescent microscope. For negative control we used serum corresponding to the primary antibody source.

Data analysis. Qualitative variables were compared by the {chi}2 test. Quantitative variables were compared by the nonparametric Mann-Whitney test to study the effect of different manipulations of HSC cultures. P value <0.05 was considered significant. Real-time PCR data were used for relative quantification of mRNA using GAPDH as the housekeeping gene. All values were expressed as arbitrary units which were derived from the ratios of cycle threshold values of the gene of interest to the house keeping gene.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Characterization of HSC cultures. HSC isolated from transgenic mice were identified according to their morphology and ability to synthetize {alpha}SMA. The purity of primary cultures, assessed by morphology and vitamin A autofluorescence measured in six independent cultures, was 85.2 ± 1.2% of the total cells in culture. Vitamin A autofluorescence and at phase photomicrographs are shown in Fig. 1, A and B. As seen in Fig. 1C, primary HSC on day 6 of culture were positive for GFAP, in accordance with their stellate cell phenotype. Day 8 primary cultures were strongly positive for {alpha}SMA, a marker of activated stellate cell phenotype (Fig. 1D). {alpha}SMA immunofluorescence also suggests that by day 8 of culture HSC represent more than 90% of cells in culture. In addition to these markers, we also stained passaged cells for desmin, another marker of HSC, reported to be positive in rodent HSC regardless of activation status. As seen in Fig. 1, Eand F, most of the cells in passaged cultures were positive for desmin. GFAP (for early stages of activation) and desmin (independent of HSC activation status) were previously described as markers of HSC lineage, almost completely absent from liver fibroblasts (20).


Figure 1
View larger version (71K):
[in this window]
[in a new window]

  		Fig. 1. Hepatic stellate cells (HSC) markers suggesting stellate cell phenotype in primary and passaged cultures. A and B: vitamin A autofluorescence at day 1 after HSC isolation. Phase photomicrograph (A) and autofluorescence at 328 nm (B) under low magnification (x100) showed that 85.2 ± 1.2% of cells in culture were positive for vitamin A content. Endothelial cell number, seen on photomicrograph, decreased with duration of culture owing to the lack of adequate matrix, growth factors, and nonoptimal growth media. This, together with rapid growth of stellate cells in the culture, ensures almost pure stellate cell population in later cultures, as evidenced by later panels in this figure. C: immunofluorescence staining for glial fibrillary acidic protein (GFAP), an early HSC marker, along with nuclear staining, at day 6 of primary culture (magnification x100). D: immunofluorescence for {alpha}-smooth muscle actin ({alpha}SMA; red), marker of activated HSC phenotype, at day 10 of culture (magnification x100). Majority of cells showed presence of {alpha}SMA filaments in cytoplasm, suggestive of activated HSC phenotype. E: same slide as seen in D, stained for GFAP (green), along with nuclear staining, shows majority of {alpha}SMA-positive cells also stain positive for GFAP, implicating stellate cell origin of myofibroblasts. F: composite double image of D and E, showing composite color (yellow-orange) in majority of visualized cells, implicating that visualized cells are positive for both {alpha}SMA and GFAP. G and H: immunofluorescence for desmin, HSC marker independent of activation status, in passaged cultures of HSC. G: low-power (x100) photomicrograph showing majority of cells in passaged cultures are positive for desmin. H: higher power (x200) photomicrograph showing typical filamentous cytoskeletal features of desmin.

 
HSC isolated from transgenic line harboring 3.6ColGFP demonstrated fluorescence once they became activated and synthesis of type I collagen was initiated. There was a gradual development of the fluorescence in cells, as well as a dramatic increase in number of fluorescent cells after 6–7 days in the culture, indicative of progressive activation of HSC and consequent type I collagen expression by cultured cells (Fig. 2).


Figure 2
View larger version (53K):
[in this window]
[in a new window]

  		Fig. 2. Low-power (x100) photomicrographs represent green fluorescent protein (GFP) fluorescence of HSC on different days of primary culture. GFP is fused to type I collagen COL1A1 gene promoter (3.6 pOBColGFP) as described previously. Cells were isolated and cultured as described in MATERIALS AND METHODS. A: HSC in primary culture containing 3.6 pOBColGFP are visualized at day 2 (D2) for GFP expression. Phase micrograph (top) was shown to document presence of the cells in culture; fluorescent micrograph (bottom) showed almost completely absent GFP fluorescence at day 2. B: HSC in primary culture containing 3.6 pOBColGFP are visualized at day 4 (D4) for GFP expression. Phase micrograph (top) was shown to document presence of the cells in culture; fluorescent micrograph (bottom) showed minimal GFP fluorescence at day 4. C: HSC in primary culture containing 3.6 pOBColGFP are visualized at day 6 (D6) for GFP expression. Phase micrograph (top) was shown to document presence of the cells in culture; fluorescent micrograph (bottom) showed increasing number of cells expressing GFP fluorescence at day 6. D: HSC in primary culture containing 3.6 pOBColGFP are visualized at day 9 for GFP expression. Phase micrograph (top) was shown to document presence of the cells in culture; at this time in culture cells formed dense monolayer. Fluorescent micrograph (bottom) showed that most of the cells exhibit GFP fluorescence at day 9 (D9) of the primary culture, therefore identifying them as type I collagen-producing cells.

 
The time course of CAT transgene, representing COL1A1 promoter activity (HSC isolated from transgenic line harboring pOB3.6COLCAT construct where CAT reporter gene is driven by the same COL1A1 promoter as in pOB3.6COLGFP transgenic mice), showed minimal activity in the first 3 days of culture. CAT levels in cultures harvested on day 3 were not different from the background. COL1A1 promoter activity was consistently higher on day 5 of culture and increased further as activation progressed. Maximal activity was observed at day 7 and plateaued thereafter (Fig. 3A). This increase of COL1A1 promoter activity during activation mimics the endogenous COL1A1 gene expression that we observed in the primary cultures using real time quantitative RT-PCR (Fig. 3B). It also parallels levels of endogenous type I collagen protein, as seen on the Western blot of the growth media from quiescent (day 3) and activated (day 7) HSC (Fig. 3C).


Figure 3
View larger version (12K):
[in this window]
[in a new window]

  		Fig. 3. Increase in chloramphenicol acetyltransferase (CAT) activity (representing COL1A1 expression), COL1A1 mRNA, and type I collagen protein in the growth media during primary culture of HSC. HSC were isolated from transgenic mice livers and plated as described in MATERIALS AND METHODS. Cultures were fed daily for the duration of experiment and harvested at designated time points. A: CAT activity increases as HSC activation progresses. CAT was extracted at days 3, 5, 7, 8, 16, 23, and 28 of the primary culture by harvesting cells followed by 3 cycles of freezing and thawing. CAT activity was determined by incorporation of [3H]acetate from acetyl CoA into chloramphenicol as described in MATERIALS AND METHODS. The figure is a result of 3 independent experiments with samples from each of independent cultures done in duplicates. There is a significant increase in CAT activity at day 5 and day 7 compared with day 3 (P < 0.05). Day 7 CAT activity was also higher than that of day 5 (P < 0.05). After day 7 there is a small decrease in CAT activity; however, CAT activity essentially plateaued after day 7. The small decrease in CAT activity after day 7 may be explained by the fact that cells at day 7 are confluent and following time points contain significant amount of cellular debris, suggesting some cell death in very confluent cultures. B: real-time PCR analysis of type I collagen COL1A1 mRNA in the primary cultures of HSC at days 4, 6, and 8 compared with the expression at day 2 (quiescent HSC). Values are expressed as a fold change compared with levels measured in quiescent HSC. COL1A1 mRNA increases as the activation progresses, with the highest levels observed in late primary cultures. C: Western blot analysis of the growth media harvested from primary HSC cultures 3 and 7 days in culture. Media was harvested and concentrated through Ultracell YM-10 centrifugal filter unit and TCA precipitated as detailed in MATERIALS AND METHODS. An equal amount of protein was loaded on 7.5% SDS polyacrylamide gel, transferred to polyvinylidene difluoride (PVDF) membrane, and probed with antibody against type I collagen (SC-8788). There is no significant type I collagen signal in day 3 cultures representing mostly quiescent HSC. In contrast, type I collagen is easily detected in media harvested from day 7 cultures.

 
HSC activation and {alpha}5beta1-integrin. Levels of {alpha}5beta1-integrin have been reported to increase during the HSC activation, but its kinetics in the later stages of activation is largely unknown. As shown in Fig. 4A, Western blot analysis revealed that {alpha}5beta1-integrin levels increase as HSC activate; however, once activated (days 6 and 7), the amount of {alpha}5beta1-integrin decreases to the levels found in quiescent HSC. Low levels are observed in passaged cultures of HSC, as well as in "older" primary cultures (day 9), indicating that the decrease in the {alpha}5beta1-integrin is not an artifact related to passage. There was an increase in the expression of {alpha}5-integrin mRNA in primary HSC cultures from day 2 to day 6, with decrease on day 8 (Fig. 4B), that paralleled the changes in {alpha}5-integrin protein levels. We chose to investigate {alpha}5-mRNA rather than beta1-integrin because {alpha}5-dimerizes only with beta1-integrin subunit. beta1-integrin, on the other hand, dimerizes with multiple other integrin subunits and changes in its expression could reflect events different than those mediating {alpha}5beta1-integrin changes. Furthermore, beta1-integrin protein levels remained constant through the culture period, showing no difference between quiescent HSC, early primary cultures, and passaged HSC (data not shown). Changes in the levels of {alpha}5-integrin protein, however, closely paralleled the pattern observed with {alpha}5beta1-integrin (Fig. 4C).


Figure 4
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 4. Dynamic changes of the levels of {alpha}5beta1-integrin protein, {alpha}5-integrin mRNA, and {alpha}5-integrin protein in activating stellate cells. This figure shows a representative result of all experiments that were done in triplicates. The common pattern found in all experiments is increasing levels of {alpha}5beta1-integrin as well as {alpha}5-integrin in early primary HSC cultures and decreased levels in passaged HSC and late (past day 7) primary cultures. A: Western blot analysis of HSC extracted at days 2, 3, 4, 5, 6, 7, and 9 of primary cultures, as well as tertiary (Tert) HSC cultures. Total proteins were extracted as detailed in MATERIALS AND METHODS. Total protein (40 µg) was loaded per lane on 7.5% SDS polyacrylamide gel, transferred to PVDF membrane, and probed with antibody against {alpha}5beta1-integrin (Chemicon, Temecula CA). There is an increase in {alpha}5beta1-integrin signal as cells activate in primary culture (gradual increase in signal intensity from day 2 to day 7). In contrast, {alpha}5beta1-integrin protein levels were reduced at day 9 of primary culture and in passaged HSC cultures. B: real-time PCR analysis of type {alpha}5-integrin mRNA in the primary cultures of HSC at days 4, 6, and 8 compared with the expression at day 2 (quiescent HSC). Values are expressed as a fold change compared with levels measured in quiescent HSC. {alpha}5-Integrin mRNA increases as the activation progresses in early primary cultures of HSC. The highest levels were observed at day 6. By day 8 expression of {alpha}5-integrin returns to the levels similar to one observed in quiescent HSC (day 2). C: Western blot analysis of HSC extracted at days 3, 5, 7 and 9 of primary cultures, as well as tertiary HSC cultures. Total proteins were extracted as detailed in MATERIALS AND METHODS. Total protein (40 µg) was loaded per lane on 7.5% SDS polyacrylamide gel, transferred to PVDF membrane, and probed with antibody against {alpha}5-integrin kindly provided by Dr. Bingcheng Wang, MetroHealth Medical Center, Cleveland, OH. As seen with {alpha}5beta1-integrin, there is an increase in {alpha}5-integrin signal as cells activate in primary culture (gradual increase in signal intensity from day 3 to day 7). {alpha}5-Integrin levels were reduced in day 9 primary cultures and passaged HSC cultures.

 
To investigate the role of {alpha}5beta1-integrin in HSC activation and type I collagen expression, we treated primary HSC (days 4-6 of primary culture) with echistatin, an {alpha}5beta1-integrin blocker (45). Primary HSC cultures treated with echistatin showed a significant dose-dependent decrease in COL1A1 promoter activity (P < 0.05; Fig. 5A). A similar decrease was observed in the amount of secreted endogenous type I collagen protein (Fig. 5B). Decreased COL1A1 promoter activity was also observed in the primary HSC (day 6 culture) with the use of {alpha}5beta1-integrin blocking antibodies. Echistatin and {alpha}5beta1-integrin blocking antibody did not significantly affect COL1A1 promoter activity in secondary (passaged) HSC cultures (Fig. 5C). Secondary cultures were treated 3 days after passage to allow cells to recover most of their membrane components that may have been damaged during trypsin treatment of the primary cultures. Proliferation was not significantly affected by echistatin treatment in either primary or passaged HSC cultures (data not shown).


Figure 5
View larger version (13K):
[in this window]
[in a new window]

  		Fig. 5. Effect of integrin inhibitor, echistatin, on CAT activity and endogenous type I collagen synthesis in HSC cultures. A: inhibition of CAT activity by echistatin and blocking antibody against {alpha}5beta1-integrin in day 5 primary HSC cultures. HSC were incubated with different echistatin concentrations (100, 10, and 1 nM; Echi 1, Echi 10, and Echi 100, respectively) or with {alpha}5beta1-integrin blocking antibody [10 µl/ml; Ctrl(Ab)] in serum-free media for 48 h. CAT activity was measured as described above. Figure represents compilation of 3 different experiments with at least 3 independent samples per experiment. Changes in CAT activity are shown as percentage changes compared with the control values (CTRL). There is a dose-dependent inhibition of CAT activity with echistatin. Concentrations of 100 and 10 nM showed significant reduction compared with the control values (a, P < 0.05). There was no difference between control values and HSC treated with 1 nM of echistatin. Incubation with {alpha}5beta1-integrin blocking antibody also significantly inhibited CAT activity compared with the control (b, P < 0.05). B: echistatin decreases synthesis of endogenous type I collagen in primary HSC cultures. Western blot analysis of the growth media harvested at day 6 of primary cultures incubated in presence and absence of the {alpha}5beta1-integrin blocking antibody (10 µl/ml) showed reduced amount of type I collagen secreted in antibody treated cultures. C: lack of inhibitory effect of echistatin and {alpha}5beta1-integrin blocking antibody in passaged HSC cultures. Primary HSC cultures were passaged at confluency (day 7 of primary culture). At 3 days after passage cells were incubated with different echistatin concentrations (100 nM, 10 nM and 1 nM) or with {alpha}5beta1-integrin blocking antibody (10 µl/ml) in serum free media for 48 h. No difference between control values and any of the treated groups was detected.

 
The next series of experiments was designed to investigate the effect of fibronectin, the natural {alpha}5beta1-integrin ligand, on proliferation and COL1A1 promoter activity of the primary (day 6) and passaged HSC. Both primary and passaged HSC were plated in wells previously coated with fibronectin. Fibronectin stimulated COL1A1 promoter activity in primary HSC cultures (day 6) (Fig. 6A). In contrast, COL1A1 promoter activity of the passaged HSC cultures was not affected by fibronectin (Fig. 6B).


Figure 6
View larger version (13K):
[in this window]
[in a new window]

  		Fig. 6. Effect of fibronectin on CAT activity in primary and passaged HSC cultures. HSC were isolated as described. For early activated HSC, cells were plated on plastic or fibronectin 10 µg/ml and harvested at day 6 of primary culture. For late activated HSC, cells were passaged after 7 days in primary culture and replated on plastic or fibronectin 10 µg/ml. Passaged HSC were harvested for CAT activity analysis after 6 days. A: primary (day 6) HSC cultures plated on fibronectin showed significant increase in CAT activity compared with control cells plated on plastic (*P < 0.05). Figure represents compilation of 3 independent experiments, each done on 3 samples. B: passaged HSC cultures plated on fibronectin did not show any significant difference in CAT activity compared with control HSC plated on plastic. Figure represents compilation of 3 independent experiments, each done on 3 samples.

 
These data suggest that type I collagen synthesis is regulated differently in early primary HSC cultures compared with the HSC in the fully established activated state represented by passaged cultures. HSC early in their activation utilize signaling through {alpha}5beta1-integrin to increase their type I collagen synthesis. That pathway is likely activated by increased fibronectin synthesis and deposition in the ECM surrounding HSC early in liver injury. Once HSC are fully activated, fibronectin and {alpha}5beta1-integrin do not appear to play a significant role in regulation of type I collagen synthesis.

C/EBPbeta and type I collagen synthesis. C/EBPbeta is a transcription factor reported to be involved in the control of type I collagen gene expression in HSC. In addition, C/EBPbeta has been implicated in regulation of {alpha}5-integrin gene expression. This regulation was cell type specific, as C/EBPbeta was repressor of {alpha}5-promoter activity in keratinocytes but stimulator in hepatoma HepG2 cells (5). C/EBPbeta mRNA is translated into two principal isoforms of C/EBPbeta: liver-enriched activating protein (LAP, ~35 kDa) and liver-enriched inhibitory protein (LIP, ~20 kDa). As shown in Fig. 7A, there was an increase in C/EBPbeta in primary HSC culture on days 5 and 7. C/EBPbeta paralleled changes in {alpha}5beta1-integrin described earlier. On day 2 of primary culture (mostly quiescent phenotype) as well as in passaged HSC cultures, C/EBPbeta levels were lower than in primary HSC cultures (days 5 and 7) (Fig. 7A). HSC plated on fibronectin showed increased levels of C/EBPbeta protein on day 6 of primary culture, further supporting a possible role of C/EBPbeta in fibronectin/{alpha}5beta1-integrin-mediated upregulation of type I collagen synthesis in primary HSC culture (Fig. 7B). We did not observe any difference in LIP-to-LAP ratio in fibronectin-stimulated cells (mean ± SD plastic 1.382 ± 0.348, fibronectin 1.543 ± 0.417; n = 11; P = 0.337).


Figure 7
View larger version (33K):
[in this window]
[in a new window]

  		Fig. 7. Western blot analysis of C/EBPbeta in HSC. A: changes in levels of C/EBPbeta in HSC at different time points during primary culture. Proteins were extracted at days 2, 5, 7, and 9 of primary HSC culture and a total of 40 µg of protein was loaded on 12% SDS-polyacrylamide gel. Equal loading was confirmed by Ponceau staining of the membrane after the transfer. The highest amount of C/EBPbeta is noted in primary HSC cultures at days 5 and 7, whereas in quiescent cells (day 2) and late primary cultures (day 9) HSC levels of C/EBPbeta were diminished. B: fibronectin induced C/EBPbeta in primary HSC cultures. HSC were plated on plastic or fibronectin (10 µg/ml) and cultured for 6 days. After 6 days, total proteins were extracted and 40 µg was loaded on 12% SDS-polyacrylamide gel. Top: Western blot showing increased levels of C/EBPbeta in HSC plated on fibronectin. Bottom: densitometry results (ratio of C/EBPbeta to actin). Figure is a representative blot of 3 independent experiments. NS, not significant; LAP, liver-enriched activating protein; LIP, liver-enriched inhibitory protein. C: effect of {alpha}5-integrin gene silencing with small interfering RNA (siRNA) on protein levels of type I collagen and c/EBPbeta in cell extract. HSC transfected with {alpha}5-integrin siRNA showed reduced expression of {alpha}5-integrin protein, type I collagen, and C/EBPbeta. Western blot is representative illustration of 3 independent transfection experiments.

 
{alpha}5-Integrin gene silencing with siRNA in primary cultures of HSC resulted in significant decrease in type I collagen as well as decrease in C/EBPbeta. Effectiveness of {alpha}5-integrin gene silencing was demonstrated by significant reduction of {alpha}5-integrin protein in HSC transfected with siRNA compared with control cultures (Fig. 7C).

To investigate the role of C/EBPbeta in type I collagen transcription, we extracted HSC from C/EBPbeta –/– mice and wild-type littermates. HSC were isolated and plated on plastic and fibronectin-coated wells as described above. When plated on plastic there was no detectable difference in morphology or proliferation of HSC isolated from C/EBPbeta –/– compared with wild-type littermates (data not shown). Immunofluorescence for {alpha}SMA was positive on day 6 HSC cultures isolated from C/EBPbeta –/– animals. The pattern of {alpha}SMA staining was similar to that seen in cells from wild-type littermates (Fig. 8A, 1 and 2). Type I collagen was reduced in the C/EBPbeta –/– derived HSC compared with wild-type; fibronectin matrix did not induce more type I collagen synthesis (Fig. 8B).


Figure 8
View larger version (49K):
[in this window]
[in a new window]

  		Fig. 8. Analysis of HSC isolated from C/EBPbeta –/– (KO) and wild-type (WT) littermates. HSC were isolated as described previously and cultured for 6 days in primary culture. A: immunostaining for {alpha}SMA, a marker of activated HSC, in cells isolated from C/EBPbeta null mice and wild-type littermates. After 6 days of primary culture HSC were replated on slides. Cells were incubated on slides for 3 more days to achieve full activation and than fixed with 4% formaldehyde. Top: immunostaining with {alpha}SMA antibodies in HSC of C/EBPbeta –/– genotype. {alpha}SMA fibers are present, suggesting that these cells activate. Staining pattern did not differ from HSC isolated from wild-type littermates (bottom). B: Western blot analysis of type I collagen from HSC isolated from C/EBPbeta-null (KO) mice and their wild-type littermates and plated on plastic (PL) and fibronectin (FBN). Total protein (40 µg) was separated on 7.5% SDS-polyacrylamide gel, transferred, and probed with antibodies against mature type I collagen (SC-8788, Santa Cruz Biotechnology). HSC isolated from C/EBPbeta –/– mice secreted reduced amount of type I collagen compared with HSC isolated from C/EBPbeta +/+ littermates; fibronectin did not have effect on reduced type I collagen synthesis by C/EBPbeta –/– HSC. C: Western blot analysis of {alpha}5-integrin in HSC isolated from C/EBPbeta null mice and their wild-type littermates and plated on plastic and fibronectin. Bottom: signal ratio ({alpha}5/beta actin) of different samples. Results shown are representative of 3 independent experiments. There is a blunted response of {alpha}5-integrin in HSC isolated from C/EBPbeta null mice plated on fibronectin compared with HSC isolated from wild-type littermates.

 
The levels of {alpha}5-integrin on day 6 did not differ in HSC isolated from C/EBPbeta –/– animals compared with the wild type. However, when cells were plated on fibronectin, levels of {alpha}5-integrin relative to beta-actin were higher in the wild-type HSC but failed to increase in HSC from C/EBPbeta –/– animals (Fig. 8C). Therefore C/EBPbeta plays a role not only in baseline expression of type I collagen genes but also in fibronectin induced changes in {alpha}5beta1-integrin synthesis. We described above that fibronectin induces type I collagen synthesis in early activated HSC through {alpha}5beta1-integrin. HSC of C/EBPbeta –/– genotype failed to increase {alpha}5-integrin levels when stimulated by fibronectin. Therefore C/EBPbeta is a likely regulator of both type I collagen and {alpha}5-integrin transcription and serves as part of complex transcriptional control machinery directing changes in ECM during the early activation of HSC.


   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Activation of HSC by liver injury of any type results in transition from a quiescent phenotype to a highly proliferative, matrix-synthesizing cell type. In the case of chronic injury, the result of this transformation is the development of liver fibrosis and ultimately cirrhosis. Not only does the amount of ECM increase severalfold in cirrhosis, the composition of the ECM matrix also changes. Low-density basement membrane-type matrix rich in type IV collagen is replaced by fibrillar type I collagen that is synthetized de novo by activated HSC. Activation of HSC is a dynamic process; however, sequential changes in transcriptional regulation leading to the fully activated phenotype are not fully understood. Because type I collagen is ultimately responsible for development of fibrosis in the liver, defining the differences in regulatory mechanisms controlling its synthesis during the different stages of HSC activation is a preface to any therapeutic intervention.

Our data indicate that different regulatory mechanisms are involved in the expression of type I collagen by HSC in a fashion that is dependent on the phase of HSC activation. Type I collagen was downregulated by {alpha}5beta1-integrin blocking antibodies and an {alpha}5beta1-integrin inhibitor in a dose-dependent manner only in primary HSC cultures. {alpha}5beta1-Integrin inhibitors were ineffective in decreasing type I collagen in passaged HSC, even though passaged HSC continue to synthesize large amount of type I collagen. In addition, fibronectin, the natural {alpha}5beta1-integrin ligand, increased the COL1A1 transgene activity only in primary HSC. CAT transgene activity was not affected by fibronectin in passaged HSC. These data support the observation that {alpha}5beta1-integrin-mediated regulation of type I collagen is operative only in early in activation process.

Fibronectin is released very early in liver injury by nonstellate cells, as well as synthesized by stellate cells early in their activation (28). Its deposition into ECM occurs early, even before the appearance of the {alpha}SMA (10). This information with the present data further supports our hypothesis that, as HSC activate, fibronectin binds to the increasing levels of the {alpha}5beta1-integrin and initiates signaling that results in upregulation of type I collagen gene transcription. In the later stages of activation, {alpha}5beta1-integrin presence on the membrane is downregulated and HSC utilize different mechanism(s) for continued expression of high levels of type I collagen.

At the present time it is unclear what mechanism(s) are responsible for this continued type I collagen synthesis. This switch in type I collagen regulation may be, at least in part, the result of changed composition of the ECM surrounding the HSC. Previous studies have shown that DDR2, which binds type I collagen, is upregulated in activated HSC and its stimulation contributes to proliferation and invasiveness of activated HSC (32). In addition, type I collagen- and laminin-binding integrins have been reported to exist on HSC membrane and to mediate HSC proliferation and apoptosis (8).

The mechanism of increased expression of COL1A1 promoter transgene and the resultant increase in type I collagen synthesis by fibronectin and {alpha}5beta1-integrin is unclear. We hypothesized that {alpha}5beta1-integrin-mediated signaling may affect transcriptional regulation of type I collagen genes by upregulation of a transcriptional enhancer. {alpha}5beta1-Integrin as well as other integrins have been reported to modify activity of transcription factors in certain cell types (18, 19). One such transcription factor reported to play a role in regulation of type I collagen transcription in HSC is C/EBPbeta. C/EBPbeta is a member of the leucine zipper family of CCAAT enhancer-binding proteins. All members of this family are important regulators of gene expression, particularly involved in inflammatory response and energy metabolism (44, 36). C/EBPbeta is present in the majority of cells in two main isoforms. The isoform p35 (LAP) contains both DNA binding and activating domains, whereas the p20 isoform (LIP) contains only DNA binding domain. The p20 isoform does not contain an activating domain and therefore acts as a natural dominant negative isoform, binding to the cis-element on gene promoters but lacking transcription enhancement ability. Both isoforms are translated from a single mRNA molecule, using the same reading frame but different translation start sites (29). The ratio of p35 to p20 has been reported to be instrumental in the transcriptional regulation of {alpha}SMA, modulation of cell cycle during liver regeneration, as well as in PEPCK expression in hyperglycemic conditions in the liver (14, 25, 37). In HSC, C/EBPbeta has been reported to mediate acetaldehyde- and hydrogen peroxide-induced upregulation of type I collagen transcription (2). In addition, the p20 isoform has been shown to decrease type I collagen gene transcription in HSC in response to TNF-{alpha} treatment (15). Binding sites for C/EBP proteins exist throughout the type I collagen gene(s) promoter sequence. In particular, the sequence between –365 and –335 of murine {alpha}1(I) collagen promoter appears to play a role in acetaldehyde-induced and C/EBPbeta-mediated stimulation of type I collagen transcription (11). A previously reported analysis suggested that activated HSC nuclear extracts contain mainly C/EBPbeta isoform, with other C/EBP proteins being present in much smaller amounts (2).

Recent data using overexpression of C/EBPbeta dominant negative isoform in bone of the transgenic animals showed a decrease in bone formation and reduced COL1A1 expression, suggesting that the effect of C/EBPbeta on type I collagen transcription may be similar in all type I collagen-producing cells (12).

Our experiments suggest that fibronectin increases C/EBPbeta levels in primary HSC cultures, without consistent changes in the p35-to-p20 ratio. It appears that C/EBPbeta enhances both baseline and fibronectin-stimulated type I collagen transcription. HSC isolated from C/EBPbeta knockout mice showed lower baseline type I collagen synthesis as well as blunted response to fibronectin coating. C/EBPbeta knockout also abolished fibronectin-stimulated increase in {alpha}5-integrin, suggesting that ligand-induced upregulation of {alpha}5beta1-integrin may in part be mediated by C/EBPbeta. The CCAT/enhancer protein binding site has been previously identified in the {alpha}5-integrin promoter sequence and was reported to act as a transcriptional enhancer in keratinocytes during wound healing (5). Baseline expression of {alpha}5-integrin appears to be C/EBPbeta independent; its presence on membrane of HSC isolated from knockout animals did not differ from wild-type littermates. This dual action of C/EBPbeta on the transcription of both collagen and {alpha}5-integrin genes likely has a synergistic effect on the final amount of type I collagen synthesized and secreted by early activated HSC.

In conclusion, the results of this study suggest that type I collagen expression is regulated by activation phase-specific mechanisms. Differences in type I collagen control in the early and late phases of HSC activation might be particularly relevant since both animal and human data indicate that early fibrosis may be reversible. This study suggests that {alpha}5beta1-integrin upregulates type I collagen in primary HSC cultures. C/EBPbeta transcription factor may be instrumental in mediating this {alpha}5beta1-integrin effect, although details of this mechanism remain to be elucidated. Passaged HSC utilize a different mechanism to maintain high levels of type I collagen synthesis. Understanding this differences in regulation of type I collagen gene(s) transcription may be important in designing new therapeutic strategies for the management of different stages of liver fibrosis.


   FOOTNOTES
 

Address for reprint requests and other correspondence: Address correspondence to: M. Dodig, Dept. of Gastroenterology and Hepatology, Cleveland Clinic and Cleveland Clinic Lerner College of Medicine, 9500 Euclid Ave., Cleveland, OH 44195 (e-mail: dodigm{at}ccf.org)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Asano Y, Ihn H, Yamane K, Kubo M, Tamaki K. Increased expression levels of integrin alphavbeta5 on scleroderma fibroblasts. Am J Pathol 164: 1275–1292, 2004.[Abstract/Free Full Text]
  2. Attard FA, Wang L, Potter JJ, Rennie-Tankersley L, Mezey E. CCAAT/enhancer binding protein beta mediates the activation of the murine alpha1(I) collagen promoter by acetaldehyde. Arch Biochem Biophys 378: 57–64, 2000.[CrossRef][ISI][Medline]
  3. Bogdanovic Z, Bedalov A, Krebsbach PH, Pavlin D, Woody CO, Clark SH, Thomas HF, Rowe DW, Kream BE, Lichtler AC. Upstream regulatory elements necessary for expression of the rat COL1A1 promoter in transgenic mice. J Bone Miner Res 9: 285–292, 1994.[ISI][Medline]
  4. Carloni V, Romanelli RG, Pinzani M, Laffi G, Gentilini P. Expression and function of integrin receptors for collagen and laminin in cultured human hepatic stellate cells. Gastroenterology 110: 1127–1136, 1996.[CrossRef][ISI][Medline]
  5. Corbi AL, Jensen UB, Watt FM. The alpha2 and alpha5 integrin genes: identification of transcription factors that regulate promoter activity in epidermal keratinocytes. FEBS Lett 474: 201–207, 2000.[CrossRef][ISI][Medline]
  6. Croniger CM, Millward C, Yang J, Kawai Y, Arinze IJ, Liu S, Harada-Shiba M, Chakravarty K, Friedman JE, Poli V, Hanson RW. Mice with a deletion in the gene for CCAAT/enhancer-binding protein beta have an attenuated response to cAMP and impaired carbohydrate metabolism. J Biol Chem 276: 629–638, 2001.[Abstract/Free Full Text]
  7. Desaulniers D, Leingartner K, Zacharevski T, Foster WG. Optimization of an MCF7-E3 cell proliferation assay and effects of environmental pollutants and industrial chemicals. Toxicology In Vitro 12: 409–422, 1998.[CrossRef][ISI]
  8. Eliceiri BP. Integrin and growth factor receptor crosstalk. Circ Res 89: 1104–1110, 2001.[Abstract/Free Full Text]
  9. Gardner H, Broberg A, Pozzi A, Laato M, Heino J. Absence of integrin {alpha}1beta1 in the mouse causes loss of feedback regulation of collagen synthesis in normal and wounded dermis. J Cell Sci 112: 263–272, 1999.[Abstract]
  10. Gillis SE, Nagy LE. Deposition of cellular fibronectin increases before stellate cell activation in rat liver during ethanol feeding. Alcohol Clin Exp Res 21: 857–861, 1997.[CrossRef][ISI][Medline]
  11. Greenwel P, Dominquez-Rosales JA, Mavi G, Rivas-Estilla AM, Rojkind M. Hydrogen peroxide: a link between acetaldehyde-elicited alpha1(I) collagen gene up-regulation and oxidative stress in mouse hepatic stellate cells. Hepatology 31: 109–116, 2000.[CrossRef][ISI][Medline]
  12. Harrison JR, Huang YF, Wilson KA, Kelly PL, Adams DJ, Gronowicz GA, Clark SH. Col1A1 promoter-targeted expression of p20 CCAAT/enhancer binding protein beta (C/EBPbeta), a truncated C/EBPbeta isoform, causes osteopenia in transgenic mice. J Biol Chem 280: 8117–8124, 2005.[Abstract/Free Full Text]
  13. Houglum K, Buck M, Alcorn J, Contreras S, Bornstein P, Chojkier M. Two different cis-acting regulatory regions direct cell specific transcription of the collagen {alpha}1(I) gene in hepatic stellate cells and in skin and tendon fibroblasts. J Clin Invest 96: 2269–2276, 1995.[ISI][Medline]
  14. Hu B, Wu Z, Jin H, Hashimoto N, Liu T, Phan SH. CCAAT/enhancer-binding protein beta isoforms and the regulation of alpha-smooth muscle actin gene expression by IL-1 beta. J Immunol 173: 4661–4668, 2004.[Abstract/Free Full Text]
  15. Iraburu MJ, Dominquez-Rosales JA, Fontana L, Auster A, Garcia-Trevijano ER, Covarrubias-Pinedo A, Rivas-Estilla AM, Greenwel P, Rojkind M. Tumor necrosis factor alpha down-regulates expression of the alpha1(I) collagen gene in rat hepatic stellate cells through a p20 c/EBP beta and C/EBP delta-dependent mechanism. Hepatology 31: 1086–1093, 2000.[CrossRef][ISI][Medline]
  16. Iwamoto H, Sakai H, Nawata H. Inhibition of integrin signaling with Arg-Gly-Asp motifs in rat hepatic stellate cells. J Hepatol 29: 752–759, 1998.[CrossRef][ISI][Medline]
  17. Kalajzic I, Kalajzic Z, Kaliterna M, Gronowitz G, Clark SH, Lichtler AC, Rowe D. Use of type I collagen green fluorescent protein transgene to identify subpopulations of cells at different stages of osteoblast lineage. J Bone Miner Res 17: 15–25, 2002.[CrossRef][ISI][Medline]
  18. Kettritz R, Choi M, Rolle S, Wellner M, Luft FC. Integrins and cytokines activate nuclear transcription factor-kappaB in human neutrophils. J Biol Chem 2279: 2657–2665, 2004.
  19. Klein S, de Fougerolles AR, Blaikie P, Khan L, Pepe A, Green CD, Koteliansky V, Giancotti FG. Alpha 5 beta 1 integrin activates an NF-kappa B-dependent program of gene expression important for angiogenesis and inflammation. Mol Cell Biol 22: 5912–5922, 2002.[Abstract/Free Full Text]
  20. Knittel T, Kobold D, Saile B, Grundmann A, Neubauer K, Piscaglia F, Ramodori G. Rat liver myofibroblasts and hepatic stellate cells: Different cell populations of the fibroblast lineage with fibrogenic potential. Gastroenterology 117: 1205–1221, 1999.[CrossRef][ISI][Medline]
  21. Kotoh K, Nakamuta M, Kohjima M, Fukushima M, Morizono S, Kobayashi N, Enjoji M, Nawata H. Arg-Gly-Asp (RGD) peptide ameliorates carbon tetrachloride-induced liver fibrosis via inhibition of collagen production and acceleration of collagenase activity. Int J Mol Med 14: 1049–1053, 2004.[ISI][Medline]
  22. Lelievre S, Weaver VM, Nickerson JA, Larabell CA, Bhaumik A, Petersen OW, Bissell MJ. Tissue phenotype depends on reciprocal interactions between the extracellular matrix and structural organization of the nucleus. Proc Natl Acad Sci USA 95: 14711–14716, 1998.[Abstract/Free Full Text]
  23. Lichtler A, Stover ML, Angilly J, Kream B, Rowe DW. Isolation and characterization of the rat alpha 1(I) collagen promoter. Regulation by 1,25-dihydroxyvitamin D. J Biol Chem 264: 3072–3077, 1989.[Abstract/Free Full Text]
  24. Lictinghagen R, Breitenstein K, Arndt B, Kuchbacher T, Boker KH. Comparison of matrix metalloproteinase expression in normal and cirrhotic human liver. Virchows Arch 432: 153–158, 1998.[CrossRef][ISI][Medline]
  25. Luedde T, Duderstadt M, Streetz KL, Tackle F, Kubicka S, Manns MP, Trautwein C. C/EBP beta isoforms LIP and LAP modulate progression of the cell cycle in the regenerating mouse liver. Hepatology 40: 356–365, 2004.[CrossRef][ISI][Medline]
  26. Maatta A, Glumoff V, Paakkonen P, Liska D, Pentinnen RP, Elima K. Nuclear factor binding to an AP-1 site is associated with the activation of pro-alpha 1(I)-collagen gene in dedifferentiating chondrocytes. Biochem J 294: 365–371, 1993.[ISI][Medline]
  27. Martinez-Hernandez A. The hepatic extracellular matrix. II. Electron immunohistochemical studies in rats with CCl4-induced cirrhosis. Lab Invest 53: 166–186, 1985.[ISI][Medline]
  28. Martinez-Hernandez A. The hepatic extracellular matrix. I. Electron immunohistochemical studies in normal rat liver. Lab Invest 51: 57–74, 1984.[ISI][Medline]
  29. McKnight SL. McBindall—a better name for CCAAT/enhancer binding proteins? Cell 107: 259–261, 2001.[CrossRef][ISI][Medline]
  30. Nejjari M, Couvelard A, Mosnier JF, Moreau A, Feldmann G, Degott C, Marcellin P, Scoazec JY. Integrin up-regulation in chronic liver disease: relationship with inflammation and fibrosis in chronic hepatitis C. J Pathol 195: 473–481, 2001.[CrossRef][ISI][Medline]
  31. Odenthal M, Neubauer K, Meyer zum Buschenfelde KH, Ramadori G. Localization and mRNA steady-state level of cellular fibronectin in rat liver undergoing a CCl4-induced acute damage or fibrosis. Biochim Biophys Acta 1181: 266–272, 1993.[Medline]
  32. Olaso E, Ikeda K, Eng FJ, Xu L, Wang LH, Lin HC, Friedman SL. DDR2 receptor promotes MMP-2-mediated proliferation and invasion by hepatic stellate cells. J Clin Invest 108: 1369–1378, 2001.[CrossRef][ISI][Medline]
  33. Reeves HL, Friedman SL. Activation of hepatic stellate cells—a key issue in liver fibrosis. Front Biosci 7: 808–826, 2002.[CrossRef]
  34. Rojkind M, Rojkind MH, Cordero-Hernandez J. In vivo collagen synthesis and deposition in fibrotic and regenerating rat livers. Coll Relat Res 3: 335–347, 1983.[ISI][Medline]
  35. Schnabl B, Purbeck CA, Choi YH, Hagedorn CH, Brenner D. Replicative senescence of activated human hepatic stellate cells is accompanied by a pronounced inflammatory but less fibrogenic phenotype. Hepatology 37: 653–664, 2003.[