Hepatocyte collagen production in vivo in normal rats

Histone deacetylase inhibition suppresses the transforming growth factor beta1-induced epithelial-to-mesenchymal transition in hepatocytes

Transforming growth factor beta1 (TGFbeta1) plays a crucial role in the induction of the epithelial-to-mesenchymal transition (EMT) in hepatocytes, which contributes to the pathogenesis of liver fibrosis. The inhibition of the TGFbeta1 cascade suppresses EMT and the resultant fibrosis. In this study, we focus on EMT-induced fibrosis in hepatocytes and the epigenetic regulation of the type I collagen gene. Histone acetylation is an important, major epigenetic mechanism that modulates gene transcription. We evaluated the epigenetic regulation of type I collagen in alpha mouse liver 12 hepatocytes (an untransformed mouse cell line) that had undergone EMT after treatment with TGFbeta1. The histone deacetylase inhibitor trichostatin A (TSA) inhibited EMT; this was reflected by the preservation of epithelial markers and function (E-cadherin and albumin). Fibrosis, the ultimate outcome of EMT, was abolished by TSA; this was indicated by the inhibition of type I collagen deposition. TSA exerted its anti-EMT effects by deactivating the mothers against decapentaplegic homolog 3 (Smad3)/Smad4 transcription complex and by interfering with p300, a coactivator of the type I collagen promoter, and preventing its binding to Smad3. TSA also restored Friend leukemia virus integration 1, an inhibitor of the type I collagen gene. TGFbeta1-induced EMT and its inhibition by TSA were replicated in human primary hepatocytes.

CONCLUSION

Histone deacetylase inhibition abrogates TGFbeta1-induced EMT in hepatocytes and reverses EMT-induced fibrosis by epigenetic modulation of type I collagen.

Murine cirrhosis induces hepatocyte epithelial mesenchymal transition and alterations in survival signaling pathways

Hepatocytes that reside in a chronically-injured liver have altered growth responses compared to hepatocytes in normal liver. Transforming growth factor beta (TGFbeta) is upregulated in the cirrhotic liver, and cirrhotic hepatocytes, unlike normal hepatocytes exposed to this cytokine, exhibit decreased apoptosis. In fetal hepatocytes, TGFbeta also induces epithelial-mesenchymal transition (EMT) and signaling changes in cell survival pathways. Here, chronic murine liver injury was induced by twice-weekly carbon tetrachloride administration for 8 weeks. Normal liver-derived hepatocytes (NLDH) and cirrhotic liver-derived hepatocytes (CLDH) were examined for EMT and the small mothers against decapentaplegic homolog (Smad), phosphatidylinositol-3-kinase (PI3K/Akt), and mitogen activated protein kinase (MAPK) pathways were investigated. Immunofluorescence imaging of cirrhotic livers demonstrated increased vimentin expression, which was confirmed by immunoblot analysis. In vitro, CLDH exhibited increased vimentin and type 1 collagen expression within cellular extensions consistent with EMT. Treatment with TGFbeta augmented the EMT response in CLDH. In contrast, untreated NLDH did not display features of EMT but responded to TGFbeta with increased vimentin expression and EMT characteristics. In response to PI3K/Akt inhibition, CLDH had decreased basal and insulin-stimulated p-Akt expression and decreased apoptosis compared to NLDH. In both NLDH and CLDH, vimentin expression was dependent on PI3K/Akt activity. CLDH demonstrated increased basal p-extracellular signal-regulated kinase expression that was independent of Smad and PI3K/Akt signaling. Inhibition of the MAPK pathway produced a marked increase in CLDH apoptosis.

CONCLUSION

CLDH have increased vimentin and type 1 collagen expression and morphologic features consistent with EMT. In addition, compared to NLDH, the cellular signaling phenotype of CLDH changes from a MAPK-independent pathway to a MAPK-dependent cell survival pathway. These findings may have clinical implications for chemoprevention of hepatocellular carcinoma in the cirrhotic liver.

Transforming growth factor-beta1 induces an epithelial-to-mesenchymal transition state in mouse hepatocytes in vitro

Liver fibrosis is a progressive pathologic process that involves deposition of excess extracellular matrix leading to distorted architecture and culminating in cirrhosis. The role of transforming growth factor-beta (TGF-beta) as a key molecule in the development and progression of hepatic fibrosis via the activation of hepatic stellate cells, among other fibroblast populations, is without controversy. We hereby show that TGF-beta1 induces an epithelial-to-mesenchymal transition (EMT) state in mature hepatocytes in vitro. EMT state was marked by significant upregulation of alpha(1)(I) collagen mRNA expression and type I collagen deposition. Similar changes were found in a "normal" mouse hepatocyte cell line (AML12), thus confirming that hepatocytes are capable of EMT changes and type I collagen synthesis. We also show that in hepatocytes in the EMT state, TGF-beta1 induces the snail-1 transcription factor and activates the Smad2/3 pathway. Evidence for a central role of the TGF-beta1/Smad pathway is further supported by the inhibition of EMT by Smad4 silencing using small interference RNA technology. In conclusion, TGF-beta1, a known pro-apoptotic cytokine in mature hepatocytes, is capable of mediating phenotypic changes and plasticity in the form of EMT, resulting in collagen deposition. Our findings support a potentially crucial role for EMT in the development and progression of hepatic fibrogenesis.

Dynamic changes of type I, III and IV collagen synthesis and distribution of collagen-producing cells in carbon tetrachloride-induced rat liver fibrosis

AIM: To find out the relationship between the gene transcription of different types of procollagen and the deposition of the relevant collagens in the liver tissue and to confirm the types of collagen producing cells in liver fibrogenesis.

METHODS: Dynamic changes of the expression of α1(I), α1(III) and α1(IV) procollagen mRNA and relevant collagens and the distribution of collagen producing cells during liver fibrogenesis of rat induced by CCl₄ (20 weeks) were investigated with Northern blot analysis, in situ hybridization and immunohistochemical techniques.

RESULTS: The increased expression of α₁(III) procollagen mRNA by Northern blot analysis was the most predominant one among the three mRNAs during fibrogenesis. However, the enhanced expression of α₁(IV) procollagen mRNA occurred very early while the expression of α1(I) mRNA was not enhanced much until the middle stage of the experiment. Desmin (Dm) positive hepatic stellate cells (HSCs) and few myofibroblasts (MFs) in and around the necrotic areas expressed α1(I), α1(III) and α1(IV) procollagen mRNA signals detected by in situ hybridization at the early stage of the experiment. All the three procollagen mRNA signals thereafter mainly localized in fibroblasts (Fbs) and MFs in fibrotic septa during the middle and late stages of fibrosis, which distributed parallel to the corresponding collagens detected by immunohistochemical study. In addition, the endothelial cells of sinusoids and the small blood vessels within the septa also showed α1(IV) procollagen mRNA and type IV collagen expression

CONCLUSION: It is considered that “HSC-MF-Fb” effect cell system is the major cellular source of collagen production in liver fibrosis, in which HSCs are collagen producing precursor cells in the early liver fibrogenesis, thereafter the synthesis of type I, III and IV collagens (Col I, Col III and Col IV) mainly derives from MFs and Fbs, which play a very important role in the progress of liver fibrosis. The endothelial cells along sinusoids, as another source of Col IV production, might participate in the capillization of liver sinusoids.

Selective inhibition of hepatic collagen accumulation in experimental liver fibrosis in rats by a new prolyl 4-hydroxylase inhibitor

Fibrosis and cirrhosis of the liver are often the result of chronic liver damage by a variety of different agents. Pathological accumulation of collagen, disruption of the lobular structure, and impaired hepatocellular function frequently lead to systemic involvement and fatal complications. Drugs inhibiting collagen hydroxylation and accumulation are expected to improve this situation, making prolyl 4-hydroxylase (P4H), the key enzyme of intracellular collagen processing, a rational target for pharmacological intervention. S 4682, a novel inhibitor of purified P4H (Ki = 155 nmol/L), reduced hydroxyproline (Hyp) synthesis in chicken embryo calvaria (IC50 = 8.2 micromol/L) and in cultured hepatic stellate cells (HSC) (IC50 = 39 micromol/L). S 4682 inhibited hepatic collagen hydroxylation in vivo after metabolic labeling with [14C]proline. In the CCl4 model of chronic hepatic injury, characterized by histologically and biochemically evident fibrosis and highly elevated levels of serum procollagen type III N-peptide, S 4682 reduced hepatic collagen accumulation, decreased prevalence of ascites, and lowered serum procollagen type III N-peptide (PIIINP) levels. The hepatic Hyp content of drug-treated animals was closely correlated with serum levels of PIIINP S 4682 had no influence on Hyp content of heart, lung, and kidney.

TNF-alpha inhibits liver collagen-alpha 1(I) gene expression through a tissue-specific regulatory region

Although tumor necrosis factor-alpha (TNF-alpha) inhibits collagen-alpha 1(I) gene expression in cultured hepatic stellate cells, assessment of its effects on hepatic collagen expression is complicated by the confounding variables of tissue necrosis and inflammation. Therefore, we analyzed whether chronically elevated serum TNF-alpha affects constitutive hepatic collagen metabolism in vivo by inoculating nude mice with Chinese hamster ovary (CHO) cells secreting TNF-alpha (TNF-alpha mice) or with control CHO cells (control mice). Before the onset of weight loss, collagen synthesis and collagen gene expression were inhibited in the liver of TNF-alpha mice. In transgenic mice, after 8 h, TNF-alpha (500 ng at 0 and 5 h) inhibited the liver expression of the collagen-alpha 1(I)-human growth hormone (hGH) transgene containing the first intron and -440 bp of the 5' region. Similarly, in cultured hepatic stellate cells isolated from these transgenic animals, the -440 bp collagen-alpha 1(I)-hGH transgene was responsive to TNF-alpha treatment independent of the activation of these cells. Transfection studies in stellate cells allowed further characterization of this TNF-alpha-responsive segment to -220 bp of the 5' region. Because in the skin the inhibitory effect of TNF-alpha involves a regulatory region of the collagen-alpha 1(I) gene beyond -440 bp, we herein identify a novel tissue-specific regulation of collagen-alpha 1(I) gene by TNF-alpha.

Tumor necrosis factor-alpha inhibits collagen alpha1(I) gene expression and wound healing in a murine model of cachexia

The mechanisms responsible for impaired wound healing in patients with cachexia-associated infection, inflammation, and cancer are unknown. As tumor necrosis factor (TNF)-alpha is elevated in these diseases, and TNF-alpha inhibits collagen alpha1(I) gene expression in cultured fibroblasts, we analyzed whether chronically elevated serum TNF-alpha affects collagen metabolism in vivo by inoculating nude mice with Chinese hamster ovary cells secreting TNF-alpha (TNF-alpha mice) or control Chinese hamster ovary cells (control mice). Before the onset of weight loss, TNF-alpha mice had a selective decrease in collagen synthesis and collagen alpha1(I) mRNA in the skin. In addition, TNF-alpha mice displayed impaired healing of incisional and excisional skin wounds, compared with control animals, before the onset of cachexia. The expression of transforming growth factor-beta1, a potent fibrogenic factor, was inhibited by TNF-alpha in the skin. In studies with transgenic mice expressing the human growth hormone under the direction of 5' regulatory regions of the human collagen alpha1(I) gene, TNF-alpha treatment inhibited the expression of the collagen alpha1(I) human growth hormone transgene containing -2.3 kb of the 5' region, whereas transgene expression directed by -0.44 kb of the 5' region was not affected. These experiments suggest that TNF-alpha may play an important role in the impaired wound healing of chronic diseases that are characterized by a high production of this cytokine and provide insights for potential therapeutic approaches.

Liver stellate cells in chronic viral hepatitis: the effect of interferon therapy

BACKGROUND/AIMS

Liver stellate cell proliferation and differentiation into myofibroblast-like cells is related to the development of liver fibrosis. Several cytokines, including interferons, regulate liver stellate cell proliferation and phenotypic modulation. Recent studies indicate that human liver stellate cells express the alpha-isotype of actin, specific to smooth muscle cell differentiation. We aimed to evaluate the expression of alpha-smooth muscle actin-positive liver stellate cells in patients with chronic viral hepatitis and to evaluate whether and how such expression can be modified by alpha-interferon treatment.

METHODS

Using immunohistochemistry, and a semi-quantitative scoring method, we evaluated alpha-smooth muscle actin expression in liver stellate cells before and after alpha-interferon therapy in a series of liver biopsies from 44 patients with chronic viral hepatitis.

RESULTS

Before therapy, a large number of liver stellate cells expressing alpha-smooth muscle actin were present throughout all acinar zones. A significant reduction in alpha-smooth muscle actin expression by liver stellate cells was demonstrated in biopsies performed after suspending the interferon treatment. The drop in the number of alpha-smooth muscle actin-labelled cells after therapy correlated closely with the improvement in the histological index of activity.

CONCLUSIONS

The results suggest a specific effect of interferon on liver stellate cells, possibly related to its anti-inflammatory action.

Characterization and functional studies on rat liver fat-storing cell line and freshly isolated hepatocyte coculture system

We developed and characterized a coculture system composed of a fat-storing cell clone (CFSC-2G) and freshly isolated hepatocytes that can reproduce in vitro some of the physical and functional relationships observed in vivo. Hepatocytes in the coculture are polarized, are smaller in size than hepatocytes plated on plastic, maintain a cuboidal shape, and have a tendency to form cords. Fat-storing cells, which are initially extended, retract and leave spaces that resemble liver sinusoids. Both cell types in the coculture system are functional for at least two weeks as determined by the expression of high levels of liver-specific protein mRNAs as well as by the production and secretion of liver-specific proteins into the culture medium. The hepatocytes maintain relatively high levels of asialoglycoprotein receptor on their cell surface and form functional gap junctional complexes with fat-storing cells. Hence, this coculture system retains a number of differentiated functions of hepatocytes, making it a useful model to study cell-cell interactions in culture and to analyze regulation of hepatocyte functions.

Coexpression of the collagen-binding stress protein HSP47 gene and the alpha 1(I) and alpha 1(III) collagen genes in carbon tetrachloride-induced rat liver fibrosis

HSP47 is a collagen-binding stress protein and is assumed to act as a collagen-specific molecular chaperone during the biosynthesis and secretion of procollagen in the living cell. The synthesis of HSP47 has been reported to correlate with that of collagen in several cell lines. We examined the expression of HSP47 mRNA during the progression of carbon tetrachloride (CCl4)-induced liver fibrosis in rats. Northern blot analysis revealed that the expression of HSP47 mRNA was markedly induced during the progression of fibrosis in parallel with alpha 1(I) and alpha 1(III) collagen mRNAs. By in situ hybridization, the distribution of HSP47 transcripts was similar to that of alpha 1(I) collagen and was observed only in cells lining collagen fibrils. These collagen-positive cells were confirmed to be Ito cells by immunohistochemistry for desmin. The absence of high levels of HSP47 mRNA in the liver of rats treated with only a single administration of CCl4 indicated that the induction of HSP47 mRNA was not due to the direct effect of CCl4 as a stressor, but was due to the progression of liver fibrosis. The function of HSP47 in liver fibrosis will also be discussed.

LAP (NF-IL6) transactivates the collagen alpha 1(I) gene from a 5' regulatory region

Although collagen is known to enhance hepatocyte differentiation and hepatocytes produce collagen in vivo, the transcriptional factors responsible for collagen type I gene expression in hepatic cells are not known. LAP (Liver Activator Protein) is a member of the C/EBP family, which in differentiated hepatocytes contributes to the high levels of liver-specific gene expression. In this study we show that LAP binds to the collagen alpha 1(I) promoter at both reverse CCAAT motifs and activates transcription. Furthermore, an upstream element, collagen element I (-370/-344), which shares homology with the LAP binding cis-element of the albumin promoter (9 of 13 bp) is described. This collagen element I stimulates transcription in both orientations and when placed in front of either a homologous or a heterologous chimeric report construct. These experiments suggest that LAP may be important in the expression of collagen in differentiated hepatocytes through both the promoter and a newly described upstream element.

Identification of connective tissue gene transcripts in freshly isolated parenchymal, endothelial, Kupffer and fat-storing cells by northern hybridization analysis

The aim of the present study was to identify the cell types that express collagen alpha 1(I), alpha 1(III) and alpha 1(IV), fibronectin and laminin B1 genes in normal rat liver. Parenchymal, sinusoidal endothelial, Kupffer and fat-storing (Ito) cells were isolated and purified. Total RNA of the freshly isolated cells was subjected to Northern hybridization analysis. We also compared the steady state levels of specific mRNAs in freshly isolated fat-storing cells to the levels in myofibroblast-like cells obtained from purified fat-storing cells cultured for two passages. The average purity of each cell preparation, and the percentage of contaminating cells, were determined by transmission electron microscopy and by examining the presence of vitamin A-autofluorescent cells. Fibronectin and collagen alpha 1(III) mRNAs were detected in total RNA of purified parenchymal cells. In poly(A)+ enriched RNA, small amounts of collagen alpha 1(I) mRNA were also present. In total RNA of freshly isolated fat-storing cells, collagen alpha 1(III), alpha 1(IV), and laminin B1 transcripts were found, whereas collagen alpha 1(I) and fibronectin mRNAs were not detected. Cultured fat-storing cells, however, did contain high levels of collagen alpha 1(I) and fibronectin mRNAs. The molecular size of the latter transcript was larger than the fibronectin transcript found in parenchymal cells and the whole liver. Endothelial cells contained small amounts of alpha 1(IV) mRNA. Kupffer cells did not contain the investigated transcripts. We conclude that normal parenchymal, fat-storing and endothelial cells each express a typical pattern of connective tissue molecules. When fat-storing cells are allowed to differentiate into myofibroblast-like cells, they express high levels of collagen alpha 1(I) and fibronectin mRNAs, in addition to collagen alpha 1(III) and alpha 1(IV), and laminin B1 chain mRNAs.

Role of mesenchymal cell populations in porcine serum-induced rat liver fibrosis

The role of liver mesenchymal cell populations in porcine serum-induced rat liver fibrosis were studied morphologically and immunohistochemically. Five-week-old rats were intraperitoneally injected with porcine serum twice a week and examined at various intervals between 3 and 24 wk after the initial injection. At an early phase, numbers of fibroblasts and extracellular matrix increased in the walls of central veins and in portal and capsular connective tissues. In the walls of central veins, the number of "second-layer cells" (i.e., the fibroblasts located at the second layer of the wall) increased. Connective tissue septa, accompanying some fibroblasts, extended from these interstitial tissues into the hepatic parenchyma, and their foremost edges came into direct contact with the perisinusoidal stellate cells. The sinusoids adjacent to the newly formed septa collapsed and later disappeared; this process resulted in the formation of hepatic limiting plates along the septa. At a more advanced stage, the interstitial fibroblasts and septal cells-which were derived from interstitial fibroblasts and the stellate cells-increased and became multilayered, constructing three-dimensional cell networks. These networks, together with increased collagen fibrils and elastic fibers, constitute the fibrotic dense connective tissue. In the control rat, smooth muscle cells were positive on vimentin, desmin and smooth muscle-alpha-actin staining. The stellate cells, second-layer cells, capsular and portal fibroblasts were shown to be vimentin and desmin positive and smooth muscle-alpha-actin negative. In the fibrotic liver, septal(fibroblastic) cells were vimentin and desmin positive and smooth muscle-alpha-actin negative. We conclude that not only the perisinusoidal stellate cells but also the interstitial fibroblasts, including the second-layer cells, play substantial role in the development of porcine serum-induced septal fibrosis in rat liver.

Intracellular cAMP determines the extent of degradation and not the synthesis of collagen by rat hepatocytes

Intracellular collagen degradation in normal rat hepatocytes was exponentially stimulated by db-cAMP (10-100 microM). The effect was manifested as a decrease (p less than 0.01) in net collagen production. The extent of degradation directly co-related with the intracellular cAMP levels, only up to a threshold concentration (16.2 +/- 1.3 p moles/10(6) cells) elicited by 100 microM of db-cAMP. Higher concentrations induced no further increment. Forskolin adenylate cyclase activator (10-50 microM), produced similar effects demonstrating cAMP dependence of the phenomenon. Both db-cAMP as well as Forskolin stimulated collagen degradation (p less than 0.05) in hepatocytes from rats administered CCL4. However, the extent of stimulation was significantly (p less than 0.01) less compared to that observed in normal hepatocytes. Our data demonstrates that elevated cAMP levels regulate net collagen content by signalling intracellular collagen degradation and not synthesis.

Cellular sources of matrix proteins in experimentally induced cholestatic rat liver

Collagens (I, III, and IV), fibronectin, and laminin were localized using the indirect immunoperoxidase technique 14 days after bile duct ligation, i.e., when extensive fibrosis and numerous neoformed bile ducts were observed. Extensive fibrous septa in enlarged portal spaces were stained for collagens I, III and IV, fibronectin, and laminin. Collagen IV and laminin were abundant around proliferative bile ducts. In addition, collagen IV was nearly continuous in the sinusoids. At the ultrastructural level, antigens were localized in the endoplasmic reticulum of several liver cell types. In portal spaces, bile duct cells and cells that form the transitional canal of Hering were strongly labelled for basement membrane components, particularly laminin, but not for collagens I and III and fibronectin, which were abundant in fibroblast-like cells. Inside the lobule, only Ito cells and, to a lesser extent, endothelial cells contained collagens, fibronectin, and laminin. Ito cells were found to be heavily stained for collagens III and IV, and laminin. Except for fibronectin, which was always abundant, precursors of extracellular matrix proteins were only slightly detectable in the endoplasmic reticulum of some hepatocytes, particularly those located close to altered areas. This study demonstrates that experimental extrahepatic cholestasis in the rat induces periportal fibrosis and continuous deposition of collagen IV in the sinusoids. Several cell types participate in the formation of extracellular matrix components, particularly bile duct cells and Ito cells, with a possible involvement of hepatocytes, thus suggesting that cholestasis provokes changes in the pattern of matrix protein production in liver cells.

Retinoic acid modulates rat Ito cell proliferation, collagen, and transforming growth factor beta production

Recent studies suggest that vitamin A plays an inhibitory role with respect to "activation" of the hepatic Ito cell, a likely effector of hepatic fibrogenesis. Ito cell "activation" during fibrogenesis is characterized by a decrease in intracellular vitamin A and an increase in cellular proliferation and collagen production. To explore the hypothesis that retinoids have the capacity to diminish Ito cell activation, cultured Ito cells were exposed to retinoic acid and its effects assessed on three key features: cell proliferation, collagen protein production and mRNA abundance, and transforming growth factor beta protein production. Retinoic acid was 100-1,000X more potent than retinol with respect to inhibition of Ito cell proliferation. Interstitial collagen and transforming growth factor beta production were also reduced by 10(-6) M retinoic acid. The relative abundance of type I collagen mRNA however, was not significantly altered. By contrast, retinoic acid administration to rats caused a marked reduction in the abundance of type I collagen mRNA in both total hepatic and purified Ito cell RNA. The relative abundance of rat hepatic fibronectin or apolipoprotein E mRNA was not significantly altered. These studies demonstrate that retinoic acid can differentially modulate several key features of hepatic fibrogenesis in vitro and in vivo.

Extracellular matrix gene expression increases preferentially in rat lipocytes and sinusoidal endothelial cells during hepatic fibrosis in vivo

Whether parenchymal or nonparenchymal liver cells play a predominant role in the pathophysiology of hepatic fibrosis has not been firmly established in vivo. We have addressed this question by quantitating the relative abundance of specific mRNAs for collagen types I, III, and IV, and laminin in purified populations of hepatocytes, sinusoidal endothelial cells, and lipocytes from normal and fibrotic rat liver. In normal liver, type I collagen gene expression was minimal in all cell types; mRNA for types III and IV collagen were apparent in endothelial cells and lipocytes, but not in hepatocytes. Laminin mRNA was present in all cell types. Induction of fibrogenesis by either bile duct ligation or carbon tetrachloride administration was associated with a substantial increase in mRNA for types I and III collagen in nonparenchymal cells. Lipocytes from fibrotic animals exhibited a greater than 30-fold increase in type I collagen mRNA relative to normal lipocytes, and greater than 40-fold relative to hepatocytes. Type III collagen mRNA reached 5 times that in normal lipocytes and greater than 120 times that in hepatocytes. Endothelial cells exhibited an isolated increase in type I collagen mRNA, reaching five times that in normal liver. Type IV collagen and laminin gene expression were not significantly increased in nonparenchymal cells during fibrogenesis; in fact, mRNA for type IV collagen and laminin decreased by up to 50% in endothelial cells. Despite the pronounced changes that occurred in matrix gene expression in nonparenchymal cells during fibrogenesis, no change was noted in hepatocytes. We conclude that nonparenchymal liver cells, particularly lipocytes, are important effectors of hepatic fibrosis in vivo.

Acetaldehyde selectively stimulates collagen production in cultured rat liver fat-storing cells but not in hepatocytes

Hepatocytes and fat-storing cells have been implicated in the production of collagen, under both normal and pathological conditions. In this study, short-term primary cultures of rat hepatocytes, maintained in a serum-free, hormonally defined medium without dexamethasone and cultured on a fibronectin-collagen type IV substratum, were used. Primary and passage 1 and 2 cultures of fat-storing cells maintained on tissue culture plastic were also studied. Hepatocytes produced significant amounts of collagen type III, but formation of collagen type I was not detectable. Laminin and collagen type IV production were very low. Hepatocytes maintained their ability to metabolize ethanol (at levels comparable to those observed at 2 hr) for at least 48 hr after plating and this metabolism was inhibited 86% to 95% by 4-methylpyrazole (1 mmol/L). Neither ethanol (50 mmol/L) nor acetaldehyde (175 mumol/L, initial concentration) had any effect on the production of collagen type III or laminin. Fat-storing cells (95% to 100% desmin-positive) produced significant amounts of both type I and type III collagen. Production of collagen type IV and laminin was very low. Metabolism of ethanol by these cultures was not detected. Addition of ethanol had no effect on collagen or laminin production in fat-storing cells. In contrast, acetaldehyde significantly increased the production of collagen type I, but did not alter the production of collagen type III, IV or laminin. Incorporation of 3H-proline into total protein was not affected by addition of ethanol or acetaldehyde to fat-storing cells or hepatocytes. Exposure of fat-storing cells to ethanol or acetaldehyde did not change 3H-collagen degrading activity in the media.(ABSTRACT TRUNCATED AT 250 WORDS)

Modulation of extracellular matrix proteins in rat liver during development

The expression and localization of extracellular matrix proteins in rat liver was investigated as a function of liver development. Levels of extracellular matrix proteins were measured by dot-blot or immunoblot protocols using monospecific antibodies prepared against collagen types I, III and IV; laminin; fibronectin; and fibronectin receptor. Proline and hydroxyproline levels from extracted liver collagen were quantitated by Pico Tag analysis. It was observed that the content of type IV collagen and fibronectin in the rat liver increased two to four times during the perinatal period. In contrast, levels of laminin and collagen types I and III decreased up to twofold in developing rat livers. The content of fibronectin receptor during ontogeny was decreased four times in an inverse relationship to fibronectin molecules. Fibronectin receptor and extracellular matrix proteins displayed no difference in apparent molecular weight as judged by sodium dodecyl sulfate-polyacrylamide gel electrophoresis immunoblots. Indirect immunofluorescence staining of frozen thin liver sections revealed that the pattern of localization of extracellular matrix proteins in the nonvascular regions of fetal liver was punctate rather than restricted to a specific region such as the perisinusoidal area of adult livers. Similarly, fibronectin receptor was also present, mainly in the sinusoidal area of adult livers, whereas fetal sections were diffusely stained. Our findings suggest that the differential modulation of extracellular matrix proteins and their localization in the developing rat livers undergo a dramatic alteration in the composition and structural organization of matrix material, which may act to modulate proliferation and to promote the differentiation of liver cells during development.

Cellular localization of type I III and IV procollagen gene transcripts in normal and fibrotic human liver

The authors have determined the cell types producing alpha 1 (I), alpha 2 (I), alpha 1 (III), and alpha 1 (IV) procollagen gene transcripts in adult human liver by in situ hybridization with [35S]-labeled RNA probes. The liver specimens comprised a total of 20 biopsies with normal histology and biopsies with fibrosis or cirrhosis at different clinical stages and of heterogeneous origins. In normal liver, procollagen type I, III, and IV transcripts were detected in stromal and vascular mesenchymal cells of portal tracts and central veins, as well as in some perisinusoidal cells of the lobule. In fibrotic liver, increased levels of these procollagen mRNAs were observed in the same locations, and particularly enhanced in stromal cells of fibrotic septa and portal tracts, as well as in perisinusoidal cells. Expression of alpha 1 (IV) procollagen RNA was additionally found in some vascular endothelial and bile duct epithelial cells. Although previously suggested as the major source of liver collagens, hepatocytes showed no significant procollagen transcript levels in any of our samples. Thus, procollagen synthesis does not appear to be a function of hepatocytes, but rather of mesenchymal, endothelial, and bile duct epithelial cells in adult human liver. These findings may have implications for the development of specifically targeted antifibrotic therapies.

The synthesis of glycosaminoglycans in isolated hepatocytes during experimental liver fibrogenesis

During human and experimental liver fibrogenesis, the pattern of glycosaminoglycans in fibrotic liver matrix is greatly changed by severalfold increases of hyaluronic acid, chondroitin sulfate, and dermatan sulfate, respectively. The present study aimed to determine whether hepatocytes take part during fibrogenesis in the alteration of the glycosaminoglycan profile in liver matrix. Rats received thioacetamide orally for 2 and 10 weeks, respectively. After 10 weeks a typical micronodular cirrhosis had developed. Hepatocytes isolated at these time points were characterized by light and electron microscopy and incubated for up to 4 h in suspension cultures in [35S]-sulfate and [3H]-glucosamine containing medium to study the synthesis and intra-/extracellular distribution of total and specific types of glycosaminoglycans. A biphasic change of glycosaminoglycan synthesis in hepatocytes was found. After 2 weeks of TAA-treatment parenchymal cells synthesized about 25% more labeled glycosaminoglycans than control liver cells, but at 10 weeks the synthesis was reduced by more than 40%. Thus, between 2 and 10 weeks of TAA-treatment hepatocellular glycosaminoglycan synthesis decreased by more than 50%. The major portion of newly synthesized glycosaminoglycans was nitrous acid labile and, hence, identified as heparan sulfate. Its fractional synthesis decreased from 0.90 in control cells to 0.84 (2 weeks TAA) and 0.76 (10 weeks TAA), respectively. Thus, the absolute synthesis of heparan sulfate was reduced by 50% in hepatocytes from cirrhosis liver. Eighty to 90% of labeled glycosaminoglycans remained cell-associated. Hyaluronic acid was detected neither in normal hepatocytes nor in hepatocytes from injured liver. We conclude from these data that parenchymal liver cells will not contribute actively to the accumulation of galactosaminoglycans (chondroitin sulfate, dermatan sulfate) and hyaluronic acid in the extracellular matrix during fibrogenesis. The diminished rate of synthesis of heparan sulfate in hepatocytes from cirrhotic liver might explain its fractional decrease in cirrhotic liver matrix.

Connective tissue biology and hepatic fibrosis: report of a conference

The past 15 years have seen major advances in the characterization of extracellular matrix proteins and structure of matrix. As a by-product of this work, investigators now have an array of molecular and immunological reagents for monitoring matrix metabolism. Progress in the isolation and culture of individual cell types from liver has made possible direct measurement of matrix protein production and also has opened the way to studies of matrix degradation. The expanding knowledge of soluble mediators of inflammation is being applied to the regulation of matrix protein synthesis and degradation. Finally, experimental models of fibrosis in vivo are available for defining the complexity of matrix metabolism in the intact tissue and for validating the findings from cell culture and in vitro systems.

Procollagen expression by nonparenchymal rat liver cells in experimental biliary fibrosis

To localize the cellular sources of the collagens excessively deposited in the liver in the course of secondary biliary fibrosis, we have analyzed by in situ hybridization the distribution of alpha 2(I), alpha 1(III), and alpha 1(IV) procollagen and albumin RNA transcripts in rat livers up to 6 wk following common bile duct ligation and scission. In normal liver, moderate amounts of alpha 2(I) and alpha 1(III) procollagen RNA were found in nonparenchymal cells, while alpha 1(IV) procollagen gene expression was at the threshold of detection. Following bile duct obstruction, increasing amounts of alpha 2(I), alpha 1(III), and alpha 1(IV) procollagen gene transcripts were observed in cells of the expanding portal tracts and in perisinusoidal cells in areas of excessive collagen deposition. Procollagen gene expressing perisinusoidal cells were colocalized with desmin-immunoreactive cells, suggesting that Ito cells and transitional cells were among the collagen-expressing cell types. Only alpha 1(IV) procollagen transcripts were found in epithelial cells of newly formed bile ducts. Neither normal nor fibrotic liver showed any hybridization signal above background over hepatocytes, indicating that hepatocytes are unlikely to be a major source of hepatic collagen.

Gamma-interferon treatment inhibits collagen deposition in murine schistosomiasis

Since interferons have been shown to affect the synthesis of matrix proteins such as collagen in several in vitro systems, the potential role of gamma-interferon in inhibiting hepatic fibrosis was investigated. Hepatic cells, consisting primarily of hepatocytes, were treated with recombinant gamma-interferon for 24 hr. Northern blot hybridization showed that gamma-interferon treatment caused a profound decrease in pro-alpha 2(I)collagen mRNA levels but an increase in beta-actin mRNA content. The effects of gamma-interferon were then studied in an in vivo model of hepatic fibrogenesis, murine schistosomiasis. Schistosoma-infected mice were treated with daily i.m. injections of gamma-interferon for a 4-week period starting 4 weeks after the initial infection. gamma-Interferon treatment decreased collagen deposition as determined by histologic evaluation and measurement of total liver collagen content. Northern blots showed Types I and III procollagen mRNA levels for treated, infected animals to be only 32 and 29% that of infected controls, but beta-actin mRNA levels were significantly elevated. These results indicate a potential role for gamma-interferon as an antifibrogenic agent in vivo.

In situ hybridization for procollagen types I, III and IV mRNA in normal and fibrotic rat liver: evidence for predominant expression in nonparenchymal liver cells

The expression of alpha 2(I), alpha 1(III) and alpha 1(IV) procollagen mRNA was analyzed in normal and CCl4-induced fibrotic rat liver by in situ hybridization using RNA probes. In normal liver, moderate amounts of alpha 2(I) and alpha 1(III) procollagen transcripts were found in sinusoidal cells, in stromal cells of the portal tracts and in the vicinity of central veins, whereas a1(IV) procollagen gene expression was below the threshold of detection. After 2 weeks of CCl4 treatment, increased transcription of alpha 2(I) and alpha 1(III) procollagen genes was observed in sinusoidal cells. At this stage, alpha 1(IV) procollagen mRNA was detectable in the same cell types and localization as alpha 2(I) and alpha 1(III) procollagen transcripts, although with a weaker signal. After 4 weeks, newly formed fibrous septa showed many cells intensely labeled by alpha 2(I), alpha 1(III) and alpha 1(IV) procollagen probes. Neither in normal liver nor at any stage of fibrosis was any hybridization signal above background observed in hepatocytes. These patterns suggest that in the liver Type I, Type III and Type IV procollagen expression takes place predominantly in nonparenchymal cells. Therefore, hepatocytes do not appear to be significantly involved in procollagen production in this experimental model of liver fibrosis.

Comparison of sulphated glycosaminoglycan and hyaluronate synthesis and secretion in cultured hepatocytes, fat storing cells, and Kupffer cells

The extracellular matrix of normal liver contains several types of proteoglycans including heparan sulphate, chondroitin sulphate isomers, dermatan sulphate, and the glycosaminoglycan, hyaluronic acid. In the present study both the synthesis and secretion as well as the pattern of radioactively labeled proteoglycans and hyaluronic acid of hepatocytes, fat-storing cells (Ito cells), and Kupffer cells maintained in monolayer cultures under mostly identical conditions were compared to assess their relative contribution to hepatic proteoglycan synthesis. Fat-storing cells were identified as the main type of cell producing and secreting proteoglycan and hyaluronic acid. More than 70% of labeled proteoglycan and hyaluronic acid were secreted into the medium. Heparan sulphate is the main type of proteoglycan in hepatocytes, whereas in the medium of fat-storing cells, chondroitin sulphate and dermatan sulphate comprise the major fractions. Hyaluronic acid was not detectable in hepatocyte cultures and found only in low amounts in the medium of Kupffer cells. The results point to a stringent quantitative and qualitative cellular compartmentation of proteoglycan synthesis in liver with fat-storing cells as the most important cell type for matrix proteoglycan and hyaluronic acid production.

Hepatic structural correlates of liver fibrosis: a morphometric analysis

Liver fibrosis is an important feature of many liver diseases, and the assessment of fibrosis is essential for diagnosis and prognosis. Liver needle biopsy, however, tends to sample preferentially the soft parenchyma, which poorly reflects the true extent of fibrosis. Therefore, we have searched for capsule and parenchymal features that are correlated with the fibrosis. To test for these correlations, we used a morphometric analysis of a rat model of liver fibrosis. We demonstrated that, of the variables examined, capsule thickness was the best correlate of liver fibrosis. Our results also showed that there were parenchymal structures that underwent changes that were well-correlated with the development of fibrosis in the liver. These changes were identified as increases in hepatocyte nucleus, cytoplasm, and mitochondria. These facts suggest that, at least in a rat model of fibrosis, capsule thickness is a useful parameter by which to assess liver fibrosis and that specific and apparently adaptive parenchymal changes are well-correlated to fibrosis.

Massive pericellular collagen deposition in the liver of a young female with severe Crohn's disease

We report a patient with Crohn's disease in whom repeated liver biopsies taken at surgery for the bowel disease showed progressive hepatic fibrosis of an unusual type. Dense collagen was deposited in the perisinusoidal space often surrounding individual hepatocytes. The aetiology and pathogenesis of this abnormality remain speculative.

Hepatic ultrastructure after methotrexate therapy for rheumatoid arthritis

Twenty-six patients receiving long-term oral methotrexate (MTX) therapy for rheumatoid arthritis (24 patients) or psoriasis (2 patients) were prospectively evaluated for alterations in liver morphology by light microscopy, electron microscopy, and immunofluorescence microscopy. Although only 4 MTX-treated patients had light microscopic evidence of mild fibrosis, all had evidence of collagen deposition in the space of Disse near Ito cells and changes in hepatocyte lysosomes on electron microscopy. These findings were absent from control livers. Fibrinogen, fibronectin, and type IV collagen were identified by immunofluorescence in both MTX-treated patients and controls. We conclude that long-term MTX therapy for rheumatoid arthritis is associated with alterations in hepatic ultrastructure, including collagen deposition in the space of Disse and changes in hepatocyte lysosomes.

Different mechanisms decrease hepatic collagen and albumin production in fasted rats

Weight loss is correlated with a specific decrease in collagen synthesis in extrahepatic tissues, mainly through modulation of mRNA levels. Here, we investigated the response to weight loss in the rat liver. Male rats were either fed ad libitum or fasted for 92 hr; fasted animals lost approximately 20% of their initial body weight. Following i.p. injection of [5-3H]proline, hepatic collagen was extracted and de novo collagen production was measured. There was a decrease in the specific radioactivities of purified hepatic collagen (-75%) and albumin (-70%) relative to total hepatic protein, indicating that production of both of these proteins was specifically decreased. In fasted animals, the absolute hepatic collagen production was markedly decreased (-60%), while changes in absolute hepatic protein production were small (-15%). Using hybridization with specific DNA probes, we found that fasting causes about a 70% decrease in albumin mRNA, but the quantities of hepatic procollagen alpha 1(I) and alpha 2(I) mRNAs were unchanged. These results are consistent with regulation of albumin production during fasting by modulation of mRNA levels. The inhibition of hepatic collagen production in fasted animals, however, appears to be modulated at a posttranscriptional level or may result from increased degradation. This response differs from the pretranslational regulation of collagen synthesis in extrahepatic tissues during fasting. Furthermore, our results suggest that decreased body weight could be a potentially complicating variable in studies of collagen metabolism and fibrogenesis in the liver.

Increased production of collagen in vivo by hepatocytes and nonparenchymal cells in rats with carbon tetrachloride-induced hepatic fibrosis

We have shown, using the proline:ornithine dual label method, that in normal rats, hepatocytes contribute in vivo about 80 to 90% of the newly synthesized hepatic collagen. In order to quantify the contribution of hepatocytes and nonparenchymal cells to collagen synthesis in vivo in hepatic fibrogenesis, rats with CCl4-induced liver fibrosis were given [5(3H)]proline and [14C]ornithine intraperitoneally. About 80% of the 14C in albumin and transferrin was present as arginine, following conversion of [14C]ornithine via the urea cycle. In contrast to hepatocyte proteins, in nonparenchymal cells and serum a negligible percentage of the radioactivity was present as [14C]arginine. These combined findings indicate that, in spite of the hepatocellular damage, the labeling of hepatocyte proteins was efficient and specific, validating the use of the proline:ornithine method in this experimental model of hepatic fibrosis. We calculated the [3H]proline/[14C]arginine ratio in hepatic collagen (after correcting for the relative frequencies of amino acids) as a percentage of the same ratio in either albumin or transferrin, the index hepatocyte proteins. In this experimental model, during active fibrogenesis, both hepatocytes and nonparenchymal cells increase their production of collagen 2-fold when compared to normal animals, and hepatocytes produce the majority of the newly synthesized hepatic collagen.

Types I and IV procollagen gene expression in cultured rat hepatocytes

The molecular mechanism involved in the expression of collagens by hepatocytes were investigated in both pure and co-culture with another rat liver epithelial cell type (RLEC). We measured the steady-state levels of mRNAs coding for pro alpha 1(I) and pro alpha 1(IV) chains by Northern analysis and by dot blotting, using specific recombinant cDNA probes. In freshly isolated hepatocytes, only small amounts of pro alpha 1(I) and pro alpha 1(IV) mRNAs were detected by dot-blot analysis. After 3 days in culture, the pro alpha 1(I) and pro alpha 1(IV) mRNA levels increased 2 to 5 times. The amount of pro alpha 1(IV) mRNAs was identical in hepatocyte cultured with RLECs while the pro alpha 1(I) mRNA level was 5 times that in pure hepatocyte culture. Hydrocortisone reduced pro alpha 1(I) mRNA in hepatocyte cultures, but had no effect on co-cultured cells. In both culture systems, this glucocorticoid did not act on the steady-state pro alpha 1(IV) mRNA level. Whatever the age and the type of culture (pure or mixed) RLECs exhibited the highest levels of pro alpha 1(I) and pro alpha 1(IV) mRNAs, which were reduced by hydrocortisone. These results show that procollagen gene expression by hepatocytes is not directly correlated with their functional state and that corticosteroids differently affect the expression of different collagen genes and collagen deposition.

The role of extracellular matrix in normal liver

Sustained expression of liver-specific function by hepatocytes in culture requires a basement membrane-like matrix as the culture substratum. Individual components of the matrix cannot substitute for the complex. The implications of these findings for the intact liver include the possibility that the extracellular matrix helps maintain normal parenchymal cell function and that alteration of the matrix underlies the loss of hepatocellular function that occurs in acute inflammation. Moreover, the low level of DNA synthesis by hepatocytes on the complex substratum, in contrast to that of cells cultured on collagen, suggests that extracellular matrix also modulates hepatocyte proliferation.

Estimation of the production rates of serum aminoterminal propeptide of type III procollagen and laminin in human fibrotic liver

The concentrations of laminin, a high molecular weight non-collagenous glycoprotein of basement membranes, and of the N-terminal propeptide of type III procollagen were determined in the serum of the liver outflow vascular region (hepatic vein) and of a peripheral vein (cubital vein) in patients with chronic liver diseases (fibrosis, cirrhosis, unspecified histology; n = 173), in order to determine their secretion rates from the injured livers. The mean levels of laminin (1.84 kU/l) and of procollagen peptide (28.0 micrograms/l) in hepatic vein were significantly higher (about 9.5% at p less than 0.02, and 37% at p less than 0.001, respectively) than those in the periphery (1.68 kU/l and 20.4 micrograms/l, respectively). In chronic liver diseases, however, laminin and procollagen peptide concentrations in the hepatic vein were lower than or equal to those in the cubital vein in 18% and 27% of patients, respectively. The highest regional differences of the concentrations were noted in cirrhotic subjects. The serum levels of laminin (rs 0.93) and of procollagen peptide (rs 0.73) in hepatic and in cubital vein are highly positively correlated (p less than 0.001), but the levels of procollagen peptide in hepatic vein are only weakly but still significantly statistically related with those of laminin (rs 0.446, p less than 0.001). Similarly, the hepatic-cubital venous concentration differences of both proteins are weakly (rs 0.312) but significantly (p less than 0.001) correlated. On the basis of several assumptions we estimated secretion rates from the livers of 120 U.min-1 for laminin, and 5.7 micrograms.min-1 for procollagen peptide.(ABSTRACT TRUNCATED AT 250 WORDS)

Measurement of connective tissue parameters in serum for diagnosis and follow-up of liver fibrosis

Hepatic fibrogenesis, i.e. activated synthesis and excessive intercellular deposition of connective tissue molecules (collagens, adhesive glycoproteins, proteoglycans) occurs in chronic alcoholic and viral liver injury and, less frequently, in some other conditions. The process may be monitored biochemically by the radioimmunoassay of some connective tissue molecules or their fragments and by the measurement of the activity of certain enzymes in serum. Currently, the radioimmunoassay of the aminoterminal propeptide of type III procollagen in serum reflects best the activity of liver fibrogenesis. The serum level of laminin, a high molecular weight basement membrane glycoprotein, was found to be correlated with an elevated portal venous pressure.

The concept of the "perisinusoidal functional unit" of the liver--importance to pathological processes

Non-hepatocytes account for 30 per cent of cells in the liver, with sinusoid-coating endothelial and Kupffer cells being the most common types. In addition, there are some scattered pit cells in the wall. Ito cells (perisinusoidal cells, lipocytes) are located in the Disse space. In this paper a review is given of structural and functional changes of cells in response to pathological conditions with reference to the importance of such changes due to diseases of the liver and the organism in large. Cooperation and interaction among non-hepatocytes, the vascular pole of hepatocytes, and Disse space as the site of reaction are summarised under the concept of the "Perisinusoidal Functional Unit", and the concept is explained in greater detail by examples (lipoprotein catabolism, processes of phagocytosis, vitamin A storage, virus infections, accumulation of pathological substances, fibrosis, and capillarization).

A [4,5-3H]lysine:[14C]lysine dual-label method to measure lysine hydroxylation in collagen

A new method has been developed to determine the extent of lysine hydroxylation in newly synthesized collagen. This method relies on the measurement of changes in the ratio of [3H]lysine:[14C]lysine in collagenase digests, resulting from loss of tritium from the C-5 position of lysine during hydroxylation. Lysine hydroxylation can be measured in the presence of large amounts of noncollagen proteins, and simultaneous quantitation of the relative rates of collagen and non-collagen protein production is obtained. The dual-label lysine method is simple, rapid, and accurate. There was a very good correlation between this method and column chromatography procedures currently used for the measurement of lysine hydroxylation.