仁木 利郎

Laminin-5 is an extracellular matrix protein that plays a key role in cell migration and tumor invasion. Cox-2 is an induced isoform of cyclooxygenases that plays an important role in carcinogenesis, suppression of apoptosis, angiogenesis, and metastasis of colon cancer. We report frequent co-expression of cox-2 and laminin-5 at the invasive front of earlystage lung adenocarcinomas. We investigated the expression of cox-2 and laminin-5 immunohistochemically in 102 cases of small-sized lung adenocarcinoma (maximum dimension, 2 cm or less). Cox-2 and laminin-5 were expressed in 97 (95.1%) and 82 (80.4%) cases, respectively. Both were preferentially localized in cancer cells at the cancer-stroma interface, although cox-2 tended to show a diffuse staining pattern in some cases. A comparison of their staining patterns revealed a striking similarity in their distribution in 24 cases, and a partial overlap between their localization in another 20 cases. Moreover, an overall correlation was found between the expression levels of cox-2 and laminin-5 (P = 0.018). To gain insight into the mechanisms that regulate the expression of these proteins, we additionally studied their expression in 58 cases of stage I lung adenocarcinoma, in which P53 status was determined by immunohistochemistry, polymerase chain reaction-single strand conformation polymorphism analysis, and direct sequencing. The results showed that tumors with mutant p53 tended to express more cox-2 than those with wild-type p53 (P = 0.080). Also, tumors that overexpressed P53 had higher levels of cox-2 and laminin-5 than those without P53 overexpression (P = 0.032 and 0.047, respectively). Further immunohistochemical analysis showed that tumors that over-expressed both epidermal growth factor receptor (EGFR) and erbB-2 had higher levels of cox-2 and laminin-5 than those without concomitant overexpression of these proteins (P = 0.014 and P = 0.018, respectively). To see whether EGFR signaling is involved in cox-2 and laminin-5 expression, we further conducted in vitro analyses using six lung adenocarcinoma cell lines (A549, HLC-1, ABC-1, LC-2/ad, VMRC-LCD, and L27). Western blot analyses showed that cox-2 mRNA levels, and to a lesser extent laminin-5 gamma(2) mRNA levels, correlated with the expression levels of erbB-2 and the phosphorylated form of MAPK/ERK-1/2 protein. The addition of transforming growth factor-alpha increased both cox-2 and laminin-5 gamma2 n3RNA levels in A549, ABC-1, and L27 with different kinetics; the induction of cox-2 occurred earlier th an that of laminin-5 gamma2. Finally, the migration of ABC-1 cells was inhibited by MAP kinase kinase inhibitor PD98059 and a selective cox-2 inhibitor NS-398. In contrast, the migration of A549 cells was inhibited by PD98059, but much less effectively by NS-398. These results suggest that co-stimulatory mechanisms may exist that increase the expression of cox-2 and laminin-5 at the invasive front of lung adenocarcinomas and that EGFR signaling could be one of the mechanisms. Further investigations are warranted concerning the role of cox-2 and laminin-5 in cancer cell invasion and the significance of p53 and EGFR signaling in the regulation of cox-2 and laminin-5 expression.

Hepatocyte growth factor (HGF) plays important rots in tumor development and progression. It is currently thought that the main action of HGF is of a paracrine nature: HGF produced by mesenchymal cells acts on epithelial cells that express its receptor c-MET. In this investigation, we explored the significance of c-MET expression in myofibroblasts, both in culture and in patients with lung adenocarcinoma. We first showed that human myofibroblasts derived from primary lung cancer expressed c-MET mRNA and protein by reverse transcription-polymerase chain reaction and Western blot analysis. Proliferation of myofibroblasts was stimulated in a dose-dependent manner by exogenously added recombinant human HGF whereas it was inhibited in a dose-dependent manner by neutralizing antibody to HGF. The addition of HGF in the culture medium stimulated tyrosine phosphorylation of c-MET. The c-MET protein was immunohistochemically detected in myofibroblasts in the invasive area of lung adenocarcinoma. Finally, the prognostic significance of c-MET expression in stromal myofibroblasts was explored in patients with small-sized lung adenocarcinomas. c-MET-positive myofibroblasts were observed in 69 of 131 cases (53%). A significant relationship between myofibroblast c-MET expression and shortened patient survival was observed in a whole cohort of patients including ah pathological stages (two-sided P = 0.0089 by log-rank test) and in patients with stage IA disease (two-sided P = 0.0019 by log-rank test). These data suggest that the HGF/c-MET system constitutes an autocrine activation loop in cancer-stromal myofibroblasts. This autocrine system may play a role in invasion and metastasis of lung adenocarcinoma.

Vascular endothelial growth factor receptor 3 (VEGFR-3) has been proposed as a marker for lymphatic endothelial cells. This study investigated the expression of VEGFR-3 in the tumour vessels of lung adenocarcinoma and evaluated whether VEGFR-3 staining was useful for identifying lymphatic vessels within the tumo. It also explored whether active growth of lymphatic vessels occurred in lung adenocarcinoma. Formalin-fixed, paraffin-embedded specimens obtained from 60 cases of lung adenocarcinoma, including five cases of pure bronchiolo-alveolar carcinoma (BAC) without stromal, vascular, and pleural invasion, were examined. No VEGFR-3-positive vessels were observed in pure BAC, but varying numbers of VEGFR-3-positive vessels were found in 39 of 55 (70.9%) invasive adenocarcinomas. A comparison of serial sections stained for VEGFR-3, CD31, and laminin-1 showed that most of the VEGFR-3-positive vessels appeared to be blood vessels (CD31-positive, laminin-1-positive), but some had the characteristics of lymphatic vessels (variable staining for CD31, little or no staining for laminin-1). VEGFR-3 staining highlighted lymphatic invasion by cancer tells; this invasion could not be detected by CD31 or haematoxylin and eosin (H&E) staining. Active growth of lymphatic vessels (as indicated by nuclear Ki-67 labelling of the endothelium) was observed in five tumours, four of which showed a high level of lymphatic invasion by cancer cells. It was concluded that VEGFR-3 immunostaining did not discriminate clearly between vascular and lymphatic endothelial cells, since expression of VEGFR-3 can be up-regulated in tumour blood vessels. However, VEGFR-3 staining combined with laminin-1 and CD31 staining would be useful for identifying lymphatic vessels and their invasion by tumour cells in a more objective way. Finally, proliferation of lymphatic endothelial cells may occur in association with lymphatic invasion by cancer cells. Copyright (C) 2001 John Wiley & Sons, Ltd.

Increased desmin synthesis and formation of desmin-containing intermediate filaments (IFs) is one of the hallmarks of transdifferentiation of hepatic stellate cells into myofibroblast-like cells. These desmin-enriched myofibroblast-like cells are the major sources of fibrotic extracellular matrix in chronically diseased liver. Myofibroblast-like cells are also involved in the contraction of sinusoids, which leads to increased intrahepatic pressure and portal hypertension. To address the requirements for the formation of desmin-containing IFs both in quiescent and in transdifferentiated stellate cells, we used mice deficient for glial fibrillary acidic protein (GFAP) and/or vimentin, which are additional IF proteins present in stellate cells. In this study, we show that desmin cannot form full-length bundles of IFs in the absence of both GFAP and vimentin. Quiescent and transdifferentiated GFAP(-/-)vim(-/-) stellate cells are devoid of normal bundles of IFs. Instead, they exhibit only residual IF bundles restricted to subcortical cytoplasm, although these cells contain equal desmin mRNA and protein levels as wild-type cells. The absence of vimentin alone restricts formation of desmin-containing IF bundles to the perinuclear region, while both the distal processes in quiescent stellate cells and the subcortical zone in myofibroblast-like cells remain free of desmin-containing IF bundles. The absence of GFAP alone does not interfere with the formation of desmin-containing IFs. Thus, to form normal IFs in stellate cells, desmin is required to partnerize with vimentin. In addition, these mouse models will prove to be instrumental in addressing the role of IFs in the process of stellate cell transdifferentiation.

Frequent allelic losses on chromosome 22q in small cell lung carcinomas (SCLCs) and advanced non-small cell lung carcinomas indicate the presence of tumor suppressor gene(s) on this chromosome arm. We detected a homozygous deletion at 22q12.1 in a SCLC cell line, Lu24. Cloning of the breakpoints of the Lu24 deletion revealed that the deletion was interstitial and 428, 131 bp in size. The deleted region contained the SEZ6L (Seizure 6-like) gene, whose structure had been partially determined by the chromosome 22 sequencing project, We determined the full length cDNA sequence for the SEZ6L gene based on the genomic sequence for the SEZ6L locus using the GENSCAN program and the RT-PCR method. The deduced SEZ6L protein was a transmembrane protein of 1024 amino acids with multiple domains involved in protein-protein interaction and signal transduction, SEZ6L expression was detected in a variety of human tissues, including lung, while its expression was detected in 14 (30%) of 46 lung cancer cell lines examined, Missense mutations were detected in three (7%) of the 46 cell lines, and a 1 bp deletion in the polypyrimidine tract preceding exon 4 was detected in one (2%) of 46 primary lung cancers. Therefore, it is possible that genetic and/or epigenetic SEZ6L alterations are involved in the development and/or progression in a subset of lung cancer, although functional analysis of the SEZ6L gene as well as molecular analysis of other genes in the homozygously deleted region is necessary to understand the pathogenetic significance of 220 deletions in human lung carcinogenesis.

Radixin is a member of the ERM (ezrin/radixin/moesin) protein family that is proposed to function as a membrane-cytoskeletal linker. Using differential display analysis, we have identified radixin as a gene down-regulated in primary lung adenocarcinoma. Real-time quantitative reverse transcription polymerase chain reaction confirmed that radixin mRNA was decreased, both in 10 early-stage bronchioloalveolar carcinomas and in 16 invasive lung adenocarcinomas, by 69% (p = 0.0002) and 82% (p < 0.0001), respectively, compared with 9 nontumor lung tissues. Similarly, moesin and ezrin mRNA levels were reduced in lung adenocarcinoma. Immunohistochemistry confirmed that cancer cells expressed very little radixin and moesin, whereas non-neoplastic alveolar and bronchiolar epithelial cells, and endothelial cells, including those within the tumor stroma, were consistently positive for these two proteins. Ezrin was localized in the apical surface of non-neoplastic bronchiolar and alveolar epithelial cells and, in contrast to radixin and moesin, the majority of tumor cells retained expression of ezrin. Localization of ezrin was altered in a significant proportion of tumor cells: whereas tumor cells forming lumina displayed membranous staining on the apical side, tumor cells with disorganized structures were either negative or diffusely positive for ezrin in the cytoplasm. Furthermore, a fraction of tumor cells invading the stroma in a scattered manner were strongly positive for ezrin. In conclusion, expression of radixin and moesin is down-regulated in lung adenocarcinoma, including early-stage bronchioloalveolar carcinoma. An intriguing implication of this finding is that these two genes may function as tumor suppressors in lung adenocarcinoma oncogenesis. Although structurally related to radixin and moesin, ezrin may have a distinct function in tumor-cell invasion.

To identify genes differentially expressed between small cell lung carcinoma (SCLC) cells and non-SCLC cells, mRNA differential display was applied to 3 SCLC cell lines and 6 non-SCLC cell lines. The LAMBS gene was identified as being expressed only in non-SCLC cells and not in SCLC cells. The LAMB3 gene encodes the laminin beta 3 chain, which is a unique component of laminin-5. Laminin-5 is a heterotrimer protein consisting of the alpha 3, beta 3, and gamma 2 chains, and another unique component of laminin-5 is the gamma 2 chain encoded by the LAMC2 gene. RT-PCR analysis of the LAMBS and LAMC2 genes in 45 lung cancer cell lines revealed that both the LAMB3 and LAMC2 genes were co-expressed in 21 of 32 non-SCLC cell lines (66%) but only in one of 13 SCLC cell lines (8%). Coexpression of the LAMB3 and LAMC2 genes was also observed in all 4 cases of primary non-SCLC cells examined but not in the corresponding non-cancerous lung cells. Since alpha 6 beta 4 integrin, the specific laminin-5 binding receptor, is known to be expressed only in non-SCLC cells and not in SCLC cells, it was indicated that laminin-5 is a critical microenvironmental factor for the growth of non-SCLC cells but not of SCLC cells, The differences in the expression of integrins and laminins would be critical factors to distinguish SCLC and non-SCLC cells, and such differences might be associated with the unique biological properties of SCLC cells, including metastatic potential and drug sensitivity. (C) 2000 Academic Press.

Vascular endothelial growth factors (VEGFs) C and D are novel members of the VEGF family that show some selectivity toward lymphatic endothelial cells. Recent studies suggest that VEGF-C may be involved in lymphangiogenesis and spread of cancer cells via lymphatic vessels. However, whether other VEGF family members play a role in lymph node metastasis is largely unknown. The aim of the present study was to explore whether expressions of VEGF-A, VEGF-B, VF,GF-C, and VEGF-D are correlated with lymph node status in lung adenocarcinoma. Total RNA was isolated from 60 surgical specimens of lung adenocarcinoma with (n = 27) or without (n = 33) lymph node metastasis. The relative mRNA abundance of VEGF-A, VEGF-B, VEGF-C, and VEGF-D was measured by real-time reverse transcription-PCR analysis based on TaqMan fluorescence methodology. We found that, as single factors, expression of none of the four VEGF family members clearly correlated with the presence of lymph node metastasis. The only tendency noted was for higher VEGF-B and VEGF-C and lower VEGF-D levels in the node-positive group. However, two-way scatterplot analysis revealed that tumors with lymph node metastasis were associated with a pattern of low VEGF-D and high VEGF A, VEGF-B, or VEGF-C, such that the ratios of VEGF-D:VEGF-A, VEGP-D:VEGF-B, or VEGF-D:VEGF-C were significantly lower in the node-positive group. Strikingly, none of the 11 tumors with high VEGP-D levels metastasized to lymph nodes. Furthermore, a low VEGF-D:VEGF-C ratio correlated with the presence of lymphatic invasion, and six of seven tumors with a pattern of very high expression of VEGF-C and low expression of VEGF-D displayed lymph vessel invasion that extended along the bronchovascular tree beyond the main tumor. Finally, levels of VEGF-A, but not VEGF-B or VEGF-C, were higher in tumors with large nodal metastasis (greater than or equal to 1 cm) than in those with small (<1 cm) nodal metastasis. These results support the hypothesis that two VEGF family members are involved in lymph node metastasis at two distinct steps; VEGF-C facilitates entry of cancer cells into the lymph vasculature, whereas VEGF-A promotes the growth of metastatic tumor through angiogenesis. The results also suggest that the balance between VEGF-C and VEGF-D could be important rather than the level of VEGF-C alone. Whether a low VEGF-D level plays a causative role in lymph node metastasis requires further investigation.

BACKGROUND, The laminin-5 gamma 2 chain plays an important role in cell migration during tumor invasion and tissue remodeling. METHODS. Laminin-5 gamma 2 chain expression in squamous cell carcinomas of the tongue in 67 patients with Stage II, III, or IVA,B (excluding the cases with distant metastasis) was examined immunohistochemically to determine its associations with the clinicopathologic features of each tumor. The predominant staining patterns were categorized as follows: A, few or no tumor cells were positive; B, part of the tumor nest periphery was positive; C, the tumor nest periphery was circumferentially positive; or D, almost all the tumor cells were positive. RESULTS. Laminin-5 gamma 2 chain expression was observed clearly in tumor cell cytoplasm. Of the 67 tumors examined, 6 (9%), 31 (46%), 19 (28%), and 11 (17%) showed staining patterns A, B, C, and D, respectively. With progression from staining pattern A to D, the number of immunopositive tumor cells increased significantly (P < 0.0001), and the tumor histology showed significantly more infiltrative growth (P < 0.0001) and poorer differentiation (P = 0.0021). Furthermore, both univariate (P = 0.0019) and multivariate (P = 0.0003; hazard ratio = 3.132) analysis of the patients' survival revealed that the prognosis became significantly poorer with progression from staining pattern A to D. CONCLUSIONS. Increased laminin-5 gamma 2 chain immunoreactivity, which may reflect a high invasive potential of cancer cells, is a factor indicative of a poor prognosis for patients with squamous cell carcinoma of the tongue. Cancer 1999;85:2315-21. (C) 1999 American Cancer Society.

Hepatic stellate cells are the major cellular sources of extracellular matrix in chronic liver diseases leading to fibrosis, We explored the antifibrogenic effect of two histone deacetylase inhibitors, sodium butyrate and trichostatin A (TSA), on this cell type in vitro, Primary hepatic stellate cells as well as culture activated cells were exposed to butyrate (0.01-1 mmol/L) or TSA (1-100 nmol/L); their effect on collagen types I and III and smooth muscle ol-actin was examined by quantitative immunoprecipitation and by Northern analysis. Their antiproliferative effect was examined by H-3-thymidine incorporation and cell counting, Hyperacetylation of histones was demonstrated by acid urea/Triton-X-100 (AUT) polyacrylamide gel electrophoresis. Possible cytotoxic effects were judged on stellate cells by evaluating de novo total protein synthesis, and on hepatocytes by measuring lactate dehydrogenase (LDH) leakage, albumin secretion, and epoxide hydrolase and ethoxycoumarin O-deethylase activity, TSA at 100 nmol/L and butyrate at 1 mmol/L retarded the morphological changes characteristic for activation of primary stellate cells. TSA at 100 nmol/L inhibited synthesis of collagen types I and III and smooth muscle alpha-actin by 62%, 70%, and 88%, Butyrate at 1 mmol/L showed a modest inhibitory effect on collagen type III and smooth muscle alpha-actin, but had no effect on collagen type I. Northern analysis suggested that these inhibitory effects on collagen type III and smooth muscle alpha-actin were transcriptional, while the effect on collagen type I was largely posttranscriptional, At 100 nmol/L, TSA strongly suppressed proliferation of primary hepatic stellate cells. Inhibition of activation of stellate cells was preceded by hyperacetylation of histone H4, When tested on cells at day 14 in culture, butyrate had no inhibitory effects on the synthesis of collagens or smooth muscle alpha-actin. One hundred or 10 nmol/L TSA modestly inhibited the synthesis of collagens type I (-24%,-22%) and III (-34%,-22%), and smooth muscle alpha-actin (-27%,-12%). We conclude that TSA inhibits transdifferentiation of stellate cells into myofibroblasts by interfering with the level of acetylation of histone H4.

Hepatic stellate cells are considered to be liver-specific pericytes that play a key role in liver fibrosis. Because these cells express desmin and smooth muscle alpha-actin, they were assumed to be of myogenic origin. This hypothesis became doubtful when it was reported that stellate cells also express glial fibrillary acidic protein and neural cell adhesion molecule. In the present study, we show that activated stellate cells express nestin, a class VI intermediate filament protein originally identified as a marker for neural stem cells. Expression of nestin was first studied during spontaneous activation of stellate cells in culture. Immunohistochemistry showed that nestin-positive stellate cells already appeared at day 3, and nearly all the cells became positive for nestin at day 6 and 15. The immunoreaction was present in filaments except in dividing cells. The presence of messenger RNA transcript for nestin was shown by reverse transcription polymerase chain reaction and sequencing of amplified complementary DNA. We then compared the presence of nestin with that of other intermediate filament proteins and smooth muscle alpha-actin, Immunoblotting showed that the relative concentrations of nestin, desmin, and vimentin increased between day 2 and 6 in primary culture. After the initial increase vimentin leveled off, while nestin and desmin showed a tendency to decrease. This pattern was quite different from that of glial fibrillary acidic protein, which kept declining, and smooth muscle alpha-actin, which increased continuously up to day 13 in culture. We then studied the presence of nestin in normal and CCl4-injured rat liver. In normal liver, minimal immunoreaction for nestin was observed within the liver parenchyma. During induction of fibrosis by carbon tetrachloride, nestin-positive stellate cells appeared at 6 weeks, which was late in comparison with the induction of desmin and smooth muscle alpha-actin. We conclude that nestin is induced in stellate cells during transition from the quiescent to the activated phenotype; culture activation is a stronger stimulus than in vivo activation by injection of CCl4. Taken together with reports on expression of glial fibrillary acidic protein and neural cell adhesion molecule by stellate cells, new experimental studies on the embryonic origin of these cells are required.

In this study, we present a new method to obtain pure, viable, freshly isolated hepatic stellate cells. Stellate cells were purified by cell sorting using their high side scatter (SSC) of incident light. Purity of the cells was established by light and transmission electron microscopy (TEM). Starting from stellate cells that were 50% to 70% enriched by centrifugation in 11% Nycodenz, the cell purity after sorting was found to be 96.6% +/- 2.9%. Viability of the sorted cells was 90.8% +/- 2.2% as measured by the Trypan blue exclusion test and was confirmed by cell culturing. Per hour of sorting, 1.4 +/- 0.4 million stellate cells were obtained. Sorting runs of up to 4 hours were practically feasible, resulting in yields of 5 to 6 million cells per rat liver. Cells attached to plastic substratum within 24 hours. Subsequently, they spread and underwent spontaneous transition into myofibroblast-like cells. The purity of sorted cells was documented by reverse-transcriptase polymerase chain reaction (RT-PCR) experiments using specific primer pairs for messenger RNA (mRNA) species that were only present in parenchymal (preproalbumin), endothelial (endothelial cell nitric oxide synthase [eNOS]), stellate (desmin), or Kupffer cells (77- to 88-kd fucose receptor). Contaminating mRNA species were absent in sorted stellate cells. Next, we examined freshly sorted stellate cells by Western blotting to confirm the presence of relevant cytoskeletal proteins. Cells were positive for vimentin, desmin, and glial fibrillary acidic protein (GFAP), but negative for alpha-smooth muscle actin (alpha-SMA). Sorted and cultured cells were immunophenotyped for the presence of collagen types I, III, and IV, laminin, and the cytoskeletal proteins, alpha-SMA, desmin, vimentin, and GFAP. At 90 hours in culture, cells expressed all the investigated extracellular matrix proteins. Desmin was present in 82% +/- 1%, vimentin in 96% +/- 2.5%, and GFAP in 91% +/- 4.5% of cells. alpha-SMA was present in 91% +/- 2% of cultured cells. We conclude that cell sorting based on SSC of incident light is a convenient method to obtain virtually pure stellate cells that can be used for direct analysis or for culturing. Although the yields obtained with this method are lower than with standard methods, and additional equipment is required, SSC-activated sorting offers the possibility of very pure cells when essential for analyses based on sensitive detection methods such as RT-PCR.

Activins are dimeric proteins, members of the transforming growth factor beta (TGF-beta) gene superfamily, consisting of the beta-subunits of inhibin (beta(A) and beta(B)). Recently, it was shown that activin A (beta(A):beta(A)) inhibits DNA replication and induces apoptosis in rat parenchymal cells in vitro and in vivo. Cryostat sections of normal livers and livers of rats treated with intraperitoneal injections of CCl4 were stained for the different inhibin subunits and desmin, a marker for stellate cells (SC). Staining for inhibin-alpha mas invariably negative, both in normal and fibrotic rat liver. In normal liver, inhibin-beta(A) subunit immunoreactivity was localized in parenchymal cells (PC). Staining for inhibin-beta(B) was weaker but similarly distributed. In fibrotic livers, connective tissue septa were strongly immunoreactive for inhibin-beta A. Desmin-positive stellate cells (SC) accumulated in areas strongly immunoreactive for inhibin-beta A and several cells were positive for both desmin and inhibin-beta(A). staining for inhibin-beta was weaker but similarly distributed. As above data pointed to a possible role for PC and SC, we examined by Northern blot analysis, the expression of inhibin-alpha, -beta(A), and -beta(B) messenger RNA (mRNA) in total RNA extracted from freshly isolated SC and PC of normal and CCl4-treated liver and in cultured SC. Inhibin-beta(A) mRNA was predominantly expressed in PC of normal liver. Expression was lost in PC of CCL4-treated liver. Inhibin-beta(B) mRNA expression was induced in SC of CCl4-treated liver. Inhibin-beta(A) mRNA, and to a lesser extent, inhibin-beta(B) mRNA expression was rapidly induced in cultured SC. The presence of activin A in conditioned media of cultured SC was shown by Western blotting. Apoptotic cells, identified by terminal deoxy-transferase mediated X-dUTP nick end labeling (TUNEL)-staining, were found predominantly in and near the fibrotic septa. In conclusion: 1) while activin A was constitutively expressed in PC of normal liver, its expression was lost in PC of fibrotic liver; 2) expression of activins was induced in activated SC; and 3) apoptotic cells were found predominantly near the septa, in support of the hypothesis that activin A, derived from activated SC in the septa, contributes to the induction of cell death in neighboring PC.

Fibronectins are multifunctional glycoproteins that are important components of the extracellular matrix in normal and fibrotic liver. Multiple fibronectin isoforms are generated from a single gene by alternative splicing of the primary transcript at the domains EIIIA, EIIIB, and V. The aim of this study was to investigate the fibronectin isoforms expressed by activated hepatic stellate cells, the most important connective tissue-producing cells in injured liver, Hepatocytes and skin fibroblasts were also studied for comparison, Activation of hepatic stellate cells in ville was induced by injecting CCl4 twice weekly for 3 weeks. Activation in vitro was achieved by culturing cells on plastic. The level of activation was evaluated by alpha-smooth muscle actin immunocytochemistry. Steady-state levels of fibronectin isoform messenger RNA were examined by Northern hybridization analysis using specific cDNA probes for the EIIIA, EIIIB, and V domains. The de novo synthesis of fibronectin isoforms was examined by metabolic labelling and immunoprecipitation using domain-specific monoclonal antibodies. Fibronectin transcripts were not detectable in freshly isolated hepatic stellate cells from normal liver. Cultured hepatic stellate cells, as web as skin fibroblasts, expressed EIIIA(+), EIIIB+, and V95(+) transcripts. They were detectable as early as day 3 and increased with time in culture, At 3 days in culture, more than 90 per cent of stellate cells were alpha-smooth muscle actin-positive, In vivo activated hepatic stellate cells expressed EIIIA(+) and V95(+) transcripts; EIIIB+ fibronectin mRNA was absent, Less than 20 per cent of in vice activated stellate cells expressed alpha-smooth muscle actin. Freshly isolated parenchymal cells from normal liver as web as from CCl4-treated liver expressed V95(+) transcripts, but were negative for EIIIA or EIIIB fibronectin mRNA, Immunoprecipitation results were in accordance with Northern hybridization analysis, Hepatic stellate cells in culture synthesized and secreted fibronectin molecules that contained EIIIA, EIIIB, and V fragments, Our results indicate that hepatic stellate cells synthesize and secrete fibronectin isoforms that are distinct from those of parenchymal cells. (C) 1997 by John Wiley & Sons, Ltd.

Glucocorticoids have been shown to suppress collagen synthesis and gene expression by fibroblasts. However, little is Known about their effects on fat-storing cells, the major matrix-producing cells in liver fibrosis. In this study we investigated the effect of dexamethasone on the extracellular matrix expression by cultured rat fat-storing cells. Fat-storing cells were isolated from male Wistar rats by collagenase/pronase digestion and purified by density gradient centrifugation. Pat-storing cells in early primary culture (3-day-old, representing a relatively quiescent phenotype) and in subculture (one passage, about a-week-old, representing an activated phenotype) were treated with 10(-6) mol/L dexamethasone for messenger RNA (mRNA) study or with 10(-8) to 10(-6) mol/L dexamethasone for protein study. Expression of collagen type I, III, IV, fibronectin, and laminin was analyzed at the mRNA level by Northern hybridization, and at the protein level by metabolic labeling and immunoprecipitation. Dexamethasone had a variable effect on the expression of collagen alpha 1(I) mRNA level. While a tendency for modest suppression was observed (5%-50%) in primary cells, the difference was not statistically significant, Variable response was observed in subcultured cells. Collagen alpha 1(III) mRNA level showed a tendency for stimulation, Dexamethasone stimulated the expression of collagen alpha 1(IV), fibronectin, and laminin B1 mRNA levels by 1.4-, 2.4-, and 1.6-fold respectively, in primary fat-storing cells. Subcultured cells showed a similar response, but the magnitude of stimulation was more variable than that of primary cells. Unexpectedly, at the protein level dexamethasone had no effect on the expression of these proteins. Our results indicate that glucocorticoids do not possess a net suppressive effect on extracellular matrix synthesis by fat-storing cells. Beneficial effects of glucocorticoids may be attributable to other mechanisms of action, such as their anti-inflammatory effect.

Fat-storing cells are the major producers of extracellular matrix in the liver. A good immunocytochemical marker is, however, still lacking for this cell type. Desmin, frequently used by most investigators, fails to stain many pericentral fat-storing cells in normal rat liver. The aim of the present study is to evaluate glial fibrillary acidic protein (GFAP) as an alternative marker of fat-storing cells. In normal rat liver, immunostaining of GFAP revealed numerous fat-storing cells with characteristic cytoplasmic extensions. Unlike desmin, which was preferentially expressed in periportal fat-storing cells, GFAP-positive fat-storing cells were distributed more evenly in the lobules, In a narrow periportal zone, however, GFAP-positive cells were occasionally absent. Dual GFAP/desmin staining revealed colocalization of these markers, but fat-storing cells positive only for GFAP or desmin were also present, Chronic carbon tetrachloride exposure induced a spatial change in the expression of GFAP and desmin. At 3 weeks, accumulation of GFAP/desmin double-positive cells was observed in developing fibrotic septa. At 8 weeks, the GFAP positivity in the septa persisted but became weak while desmin expression became stronger. In contrast, the expression of GFAP within the lobule was gradually decreased as fibrosis progressed. We conclude that GFAP is expressed by a subpopulation of fat-storing cells, which differs partially from the population that expresses desmin. Because in normal rat liver desmin-negative fat-storing cells can be identified by GFAP staining and vice versa, dual GFAP/desmin staining allows more complete identification of fat-storing cells. In chronically injured liver, GFAP may not be as useful as in normal rat liver. The coexpression of GFAP/desmin in developing septa and the subsequent downregulation of GFAP in an advanced stage of fibrosis may reflect different stages of fat-storing cell activation. Further investigation is required to determine the functional significance of alteration of GFAP expression in fat-storing cells.

Transforming growth factor beta (TGF-beta), a potent fibrogenic cytokine, is secreted in latent form. We examined which cell type in both normal and carbon tetrachloride (CCl4)-induced fibrotic rat liver bears surface type IIIGF/mannose 6-phosphate (IGF-II/M6P) receptor, known to facilitate activation of TGF-beta. In addition, the role of the IGF-II/M6P receptor in activation of latent TGF-beta was investigated in a coculture system with sinusoidal endothelial cells. Northern hybridization analysis for IGF-II/M6P receptor messenger RNA (mRNA) was performed on total RNA of different isolated and purified liver cell types. In normal liver, cells expressed little IGF-II/M6P receptor mRNA. In fibrotic liver, we found significant expression only in fat-storing cells. The presence of IGF-II/M6P receptors was established by [I-125]IGF-II binding assays on freshly isolated fat-storing cells from normal and CCl4-exposed rat livers. We found specific binding of [I-125]IGF-II only on CCl4 exposed fat-storing cells. As determined by polyacrylamide gel electrophoresis arter affinity labeling, the specific binding involved 220 kD type II IGF receptors. Scatchard analysis revealed the presence of 3,043 +/- 1,378 IGF-II/M6P high-affinity receptors/fat-storing cell, with a K-d of 387 +/- 165 pmol/L. With a mink lung epithelial cell (Mv(1)Lu) proliferation inhibition assay, inhibition of proliferation (a measure of active TGF-beta function) was determined using conditioned media of activated ht storing cells, cocultures of fat-storing cells, and endothelial cells and pure endothelial cell cultures. We found that conditioned media of cocultures of fat storing cells and endothelial cells inhibits the growth of Mv(1)Lu cells more strongly than conditioned media of homotypic cultures. Addition of neutralizing anti-TGF-beta antibodies neutralizes this effect. The contribution of IGF-II/M6P receptors on the cell membrane of activated fat-storing cells to the activation of latent TGF-beta was demonstrated by incubating the cells with M6P, or antibodies directed to the IGF-II/M6P receptor, both of which diminishes activation of TGF-beta. We conclude that IGF-II/M6P receptors are present on activated fat-storing cells and that the presence of this receptor at the cell surface is necessary, but not sufficient, for activation of latent TGF-beta. Other factors, derived from endothelial cells, are involved.