Antitumor effects of hyaluronic acid inhibitor 4-methylumbelliferone in an orthotopic hepatocellular carcinoma model in mice

Glycobiology, Mar 2012

Liver cirrhosis is characterized by an excessive accumulation of extracellular matrix components, including hyaluronan (HA). In addition, cirrhosis is considered a pre-neoplastic disease for hepatocellular carcinoma (HCC). Altered HA biosynthesis is associated with cancer progression but its role in HCC is unknown. 4-Methylumbelliferone (4-MU), an orally available agent, is an HA synthesis inhibitor with anticancer properties. In this work, we used an orthotopic Hepa129 HCC model established in fibrotic livers induced by thioacetamide. We evaluated 4-MU effects on HCC cells and hepatic stellate cells (HSCs) in vitro by proliferation, apoptosis and cytotoxicity assays; tumor growth and fibrogenesis were also analyzed in vivo. Our results showed that treatment of HCC cells with 4-MU significantly reduced tumor cell proliferation and induced apoptosis, while primary cultured hepatocytes remained unaffected. 4-MU therapy reduced hepatic and systemic levels of HA. Tumors systemically treated with 4-MU showed the extensive areas of necrosis, inflammatory infiltrate and 2–3-fold reduced number of tumor satellites. No signs of toxicity were observed after 4-MU therapy. Animals treated with 4-MU developed a reduced fibrosis degree compared with controls (F1-2 vs F2-3, respectively). Importantly, 4-MU induced the apoptosis of HSCs in vitro and decreased the amount of activated HSCs in vivo. In conclusion, our results suggest a role for 4-MU as an anticancer agent for HCC associated with advanced fibrosis.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://glycob.oxfordjournals.org/content/22/3/400.full.pdf

Antitumor effects of hyaluronic acid inhibitor 4-methylumbelliferone in an orthotopic hepatocellular carcinoma model in mice

Flavia Piccioni 0 1 Mariana Malvicini 0 1 Mariana G Garcia 0 1 Andrs Rodriguez 1 Catalina Atorrasagasti 0 1 Nestor Kippes 1 Ignacio T Piedra Buena 1 Manglio M Rizzo 1 Juan Bayo 1 3 Jorge Aquino 0 1 Manuela Viola 2 Alberto Passi 2 Laura Alaniz 0 1 Guillermo Mazzolini 0 1 0 CONICET (Consejo Nacional de Investigaciones Cientificas y Tecnicas) , Buenos Aires , Argentina 1 Gene Therapy Laboratory, School of Medicine, Austral University , Avenida Presidente Peron 1500 (B1629ODT) Derqui-Pilar, Buenos Aires , Argentina 2 Dipartimento di Scienze Biomediche Sperimentali e Cliniche, Universita degli Studi dell'Insubria , via J.H. Dunant 5, 21100 Varese , Italy 3 Agencia Nacional de Promocion Cientifica y Tecnologica (ANPCyT) , Buenos Aires , Argentina - Liver cirrhosis is characterized by an excessive accumulation of extracellular matrix components, including hyaluronan (HA). In addition, cirrhosis is considered a pre-neoplastic disease for hepatocellular carcinoma (HCC). Altered HA biosynthesis is associated with cancer progression but its role in HCC is unknown. 4-Methylumbelliferone (4-MU), an orally available agent, is an HA synthesis inhibitor with anticancer properties. In this work, we used an orthotopic Hepa129 HCC model established in fibrotic livers induced by thioacetamide. We evaluated 4-MU effects on HCC cells and hepatic stellate cells (HSCs) in vitro by proliferation, apoptosis and cytotoxicity assays; tumor growth and fibrogenesis were also analyzed in vivo. Our results showed that treatment of HCC cells with 4-MU significantly reduced tumor cell proliferation and induced apoptosis, while primary cultured hepatocytes remained unaffected. 4-MU therapy reduced hepatic and systemic levels of HA. Tumors systemically treated with 4-MU showed the extensive areas of necrosis, inflammatory infiltrate and 23-fold reduced number of tumor satellites. No signs of toxicity were observed after 4-MU therapy. Animals treated with 4-MU developed a reduced fibrosis degree compared with controls (F1-2 vs F2-3, respectively). Importantly, 4-MU induced the apoptosis of HSCs in vitro 1To whom correspondence should be addressed: Tel: +54-2322-482618; Fax: +54-2322-482204; e-mail: (G.M.)/laualaniz@ yahoo.com.ar (L.A.) and decreased the amount of activated HSCs in vivo. In conclusion, our results suggest a role for 4-MU as an anticancer agent for HCC associated with advanced fibrosis. Introduction Hyaluronan (HA) is a linear, large and ubiquitous nonsulfated glycosaminoglycan (GAG) of extracellular matrix (ECM) that can also be associated with cell surface (Toole 2004). The cellular effects of HA are mediated by different receptors, such as CD44 and RHAMM (Turley et al. 2002). HA is found in almost all tissues but especially increased in those undergoing cell proliferation, regeneration and repair, such as embryonic, inflamed and tumor stroma tissues (Toole 2002; Alaniz et al. 2009). HA can be present in different molecular weight (MW) species and associated with different biological functions. High MW (HMW) HA (106107 Da) can play a physiological role participating in tissue homeostasis and in protein distribution in cell membrane. However, in certain pathological conditions such as injured tissues, low MW products (0.11 106 Da) of HA are generated by differential expression of hyaluronidases (Stern 2005) as well as by oxidation process (Noble 2002; Tammi et al. 2002). Moreover, these products are supposed to contribute to scar formation and fibrogenesis (McKee et al. 1996; West et al. 1997; Guo et al. 2010). The liver is responsible for the uptake and degradation of HA (Fraser et al. 1981). Under physiological conditions, systemic HA levels are low or undetectable; however, serum HA levels are increased in patients with liver fibrosis or cirrhosis (Vrochides et al. 1996). Liver fibrosis is a wound healing process that occurs in response to a number of hepatic diseases such as severe alcohol intake and chronic hepatitis (Bataller and Brenner 2005; Friedman 2008). As a result, there are changes in ECM composition and organization where activated hepatic stellate cells (HSCs) play a key role as the major collagen-ECM-producing cells in the damaged livers (Safadi and Friedman 2002; Friedman 2008). The accumulation of ECM components and advanced liver fibrosis may result in cirrhosis, hepatic failure and increased portal pressure and often require liver transplantation (Bataller and Brenner 2005). In addition, cirrhosis is a major risk factor for the development of hepatocellular carcinoma (HCC; El-Serag and Rudolph 2007). Curative treatments can only be applied to a minority of patients at the time of diagnosis (Rampone et al. 2010). Therefore, new therapeutic options are urgently needed for advanced disease. 4-Methylumbelliferone (4-MU) is a 7-hydroxy-4-methyl coumarin, a molecule of vegetal origin that has been used safely in humans as a cholagogue orally as well as in clinical trials in patients with chronic hepatitis B and C (ClinicalTrials.gov identifier: NCT00225537; Takeda 1981). 4-MU has been used as a specific HA synthesis inhibitor in many in vitro studies (Yoshihara et al. 2005; Morohashi et al. 2006; Kultti et al. 2009). Its effect was first demonstrated in vitro in human fibroblasts, where the inhibition of HA synthesis but not other GAGs was observed (Kakizaki et al. 2004). Similar results were observed in B16F10 melanoma and KP1-NL human pancreatic cancer cells (Morohashi et al. 2006). Even though 4-MU action mechanism is not welldefined, this molecule can act through the reduction in HA synthase mRNA levels and by the depletion of HA synthesis precursors UDP-glucuronic acid as it has been observed in melanoma, breast and ovarian cancer cells (Kultti et al. 2009). It has been demonstrated in mice with melanoma that 4-MU selectively inhibits hepatic HA synthesis, resulting in the inhibition of liver metastases. The authors argued that HA formation on the cell surface is critical for tumor dissemination (Yoshihara et al. 2005). In addition, antitumoral effects of 4-MU have been reported for breast, esophageal and prostate carcinoma (Lokeshwar et al. 2010; Twarock et al. 2011; Urakawa et al. 2011). Therefore, we thought that the inhibition of HA synthesis by 4-MU in the injured liver and in HCC cells should help to control tumor growth and fibrosis progression. Results HA production and binding activity of HCC cells We first analyzed HA-binding capacity, CD44 expression and HA production on HCC cells including the Hepa129 cell line. Hepa129 cells expressed CD44 (60.3%) and bound to HA (59.2%; Figure 1A). Binding is partially bloquead to CD44, because pretreatment with a monoclonal antibody (mAb) anti-CD44 (KM81, HA-blocking antibody) affected HA binding (from 59.2% to 45.4%; Figure 1A). In addition, we observed that Hepa129 cells highly secreted HA as detected by an enzyme-linked immunosorbent assay (ELISA)-like assay in culture supernatants (1940 90 ng/mL/24 h; Figure 1B). We also analyzed these parameters in Hep3B and Huh7 human HCC cell lines and observed that they did not express CD44 and produced low amounts of HA in comparison with Hepa129 cells (1450 60 and 500 200 ng/mL/24 h, respectively) as assessed by an HA ELISA-like assay (Figure 1B). In vitro 4-MU treatment inhibits cell proliferation and induces apoptosis in HCC cell lines and inhibits HA production Effect of 4-MU on HCC growth is shown in Figure 2A (left panel). Hepa129 and Hep3B cells showed the inhibition of the proliferation in a dose-dependent manner. HCC cell proliferation was significantly reduced at a dose of 0.5 mM 4-MU for Hepa129 and Hep3B. In contrast, only higher doses of 4-MU (5 mM) were able to inhibit HuH7 cell proliferation (90% vs control). We next examined whether the growth inhibition by 4-MU was due to the induction of apoptosis (Figure 2A, right panel). 4-MU strongly induces apoptosis in Hepa129 and Hep3B cells at a dose of 5 mM (52 and 49%, respectively), whereas Huh7 were resistant to apoptosis in all tested doses. Apoptosis results were confirmed by an Annexin V assay (data not shown). Hepa129 cells secrete higher HA levels when compared with Hep3B and Huh7 cells (Figure 1B). We therefore measured the effect of 4-MU on HA synthesis in Hepa129, Hep3B and Huh7 cells. As shown in Figure 2, 4-MU inhibited HA synthesis in both Hepa129 and Hep3B cell lines. For murine Hepa129 cells, inhibition of HA synthesis was significant at 12 h in comparison with control (Figure 2B), but not significant differences were observed at 24 h (data not shown). Human Hep3B cells shown a significant reduction in the HA level at 24 h in comparison with control (Figure 2C), whereas no reduction in the HA level was observed in Huh7 cells for the analyzed times (Figure 2D). These results show that there is a correlation between HA synthesis and induction of apoptosis after treatment with 4-MU for Hepa129 and Hep3B cells, whereas the lowproducer Huh7 cells were resistant to 4-MU-induced apoptosis. Antitumoral effects of 4-MU in an orthotopic HCC model associated with advanced liver fibrosis HCC tumors are frequently associated with advanced fibrosis or cirrhosis. To assess the effects of HA synthesis inhibition using 4-MU on HCC tumor development, we established an orthotopic tumor model (as described in Materials and methods, Figure 3A) associated with advanced fibrosis [induced by thioacetamide (TAA)] that is characterized by the presence of a main tumor mass and macroscopic tumor satellites. 4-MU therapy inhibited HCC tumor mass growth in mice with fibrosis in comparison with untreated animals, although these results were not statistically significant (350 vs 250 mm3, respectively; Figure 3B). However, the number of satellite tumor nodules, as an indicator of tumor aggressiveness, was found 23-fold reduced in 4-MU-treated mice (HCC + TAA + 4-MU group; Figure 3B). Interestingly, the majority of untreated mice (HCC + TAA group) developed severe ascites (including hemorrhagic ascites), whereas in the HCC + TAA + 4-MU group, ascites was almost absent or mild, as assessed by laparotomy. Importantly, microscopic examination of hepatic tumors revealed that mice treated with 4-MU showed increased mononuclear inflammatory cell infiltrate, which was more abundant in tumor edges and extensive areas of necrosis (Figure 3C) when compared with non-treated mice. Toxicity studies We then evaluated 4-MU toxicity in vitro and in vivo. Healthy mice treated with 4-MU showed no changes in body weight during the treatment course in comparison with healthy untreated mice (not shown). Serum chemistry analysis showed no significant differences in alanine transaminase (ALT) levels after treatment with 4-MU, indicating the absence of significant necrosis of hepatocytes after 4-MU administration (Figure 3D). 4-MU treatment did not cause any sign of hepatocellular damage and liver histology was normal (Figure 3E). To confirm these results, primary cultured hepatocytes treated with 4-MU were assessed for apoptosis and cytotoxicity. We observed that 4-MU did not affect hepatocytes in culture, since there were no significant differences in apoptosis between 4-MU doses (not shown) and cytotoxicity was lower than 5% (Figure 3F). In addition, 4-MU also showed no cytotoxic effect on Hepa129 cells (Figure 3G). 4-MU treatment reduced the amount of hepatic HA HA expression was evaluated in liver tissue sections of mice from different groups by staining with an HA-binding protein (HA-BP). In healthy mice, the HA label was almost absent (Figure 4A). Contrarily, high HA expression was observed in liver sections from mice with established fibrosis or with both HCC and fibrosis (Figure 4A). As a result of 4-MU therapy, mice showed a remarkable reduction in HA expression, mainly inside HCC tumor nodules (Figure 4A). We then quantified HA by an ELISA-like assay in the liver tissue and, as expected, we observed a very significant reduction in the hepatic HA content after treatment with 4-MU in comparison with the livers from non-treated mice. This reduction was also seen when compared with mice with fibrosis alone (TAA group; Figure 4B). Accordingly, serum HA levels were also reduced after 4-MU therapy (Figure 4C). 4-MU treatment reduces liver fibrosis induced by TAA We next decided to investigate whether 4-MU treatment may influence liver fibrogenesis and other associated processes in our animal model. After 6 weeks of TAA administration, TAA-treated livers from tumor-bearing mice showed the extensive appearance of portalportal and centralportal fibrous septae, distortion of liver architecture and regenerative nodules (Figure 5A). However, a marked reduction in the amount of fibrous septae and of regenerative nodules was found in 4-MU-treated animals. In order to assess the extent of liver fibrosis, the Metavir score and Sirius Red staining were performed. We observed that untreated animals showed distortion of liver architecture, portalportal and centralportal fibrous linkages, correlated with an F2-3 fibrosis degree, whereas 4-MU-treated mice showed a reduced amount of collagen with occasional portalportal bridges (F1-2; Figure 5A). In order to quantify the hepatic content of collagen, the ECM protein most abundantly accumulated in the fibrotic livers, tissue sections from TAA-treated animals were stained with Sirius Red and morphometric analysis was performed. A significant reduction in the Sirius Red+ area was found in 4-MU-treated mice when compared with non-treated mice (0.65 0.08 vs 1.48 0.11%; Figure 5B). Liver fibrogenesis is characterized by trans-differentiation of HSCs. In addition, HSC-derived myofibroblasts, regarded as the main sources of ECM proteins as well as of pro-inflammatory cytokines in a fibrotic liver, are known to increase -smooth muscle actin (-SMA) expression levels according to the degree of their activation state. In order to address whether either the number of myofibroblasts in fibrous septae or/and the activation state of myofibroblasts might be affected by 4-MU treatment, the liver tissue obtained from 6-week TAA-treated mice was immunostained for -SMA detection. A significant reduction in the -SMA+ immunostained area was found in 4-MU when compared with non-treated mice (0.17 0.02 vs 0.40 0.04%; Figure 5C). These results suggest that the 4-MU therapy reduces the number and/or the activated state of myofibroblasts, cells playing a key role in liver fibrogenesis. In vitro 4-MU treatment inhibits cell proliferation and induces apoptosis in fibroblast cell lines In order to study the effect of 4-MU treatment in fibrosis development, we analyzed proliferation and apoptosis of hepatic fibroblasts and HSC as described in Materials and Methods section (Figure 6A). Proliferation was inhibited in all the studied doses of 4-MU in murine hepatic fibroblasts IZA2.1, rat CFSC-2G and human LX2 hepatic stellate cell lines. However, the apoptosis index was increased only with higher doses of 4-MU (5 mM) in the three cell lines. Discussion In this work, we show that 4-MU, a well-known HA synthesis inhibitor and a dietary supplement, is an effective antiproliferative and pro-apoptotic agent with antitumoral and antifibrotic effects both in vitro and in vivo. It is well known that endogenous HA synthesis is up-regulated during liver fibrosis, cirrhosis (Engstrom-Laurent et al. 1985; Frebourg et al. 1986) and cancer progression (Boregowda et al. 2006). Almost 90% of circulating HA is metabolized in the liver by HSC and sinusoidal cells; therefore, hepatic failure deteriorates HA clearance (Vrochides et al. 1996; Fraser et al. 1997). We hypothesize that hepatic HA accumulation plays a role in liver fibrogenesis and, consequently, in HCC development. The aim of our study was to determine whether 4-MU could be able to ameliorate liver fibrosis and HCC growth in an orthotopic mouse model. First, we analyzed the expression of CD44, HA synthesis and binding capacity of Hepa129 tumor cells. We observed that Hepa129 cells bind to HA, in part through CD44, because HA binding was not completely abolished by anti-CD44 treatment, an HA-blocking antibody (KM81); therefore, it seems that other receptors may also be implicated, such as LYVE-1 and RHAMM (Day and Prestwich 2002). In addition, Hepa129 tumor cells produced higher amounts of HA as determined by the HA-BP and ELISA-like assays. These results are in agreement with other studies, where malignant tumors have high levels of HA (Toole 2002). Higher production of HA was associated with increased migration, metastases development and tumor survival (Toole 2002). HCC generally arises in a cirrhotic liver (Pons-Renedo and Llovet 2003); it is, therefore, of great importance to investigate HCC development in association with a fibrotic microenvironment. For that reason, we established an orthotopic HCC animal model in mice with advanced fibrosis induced by TAA chronic intoxication. This model allows us to investigate the effect of inhibition of HA synthesis by 4-MU in a scenario that more closely mimic clinical liver cancer (Pons-Renedo and Llovet 2003). 4-MU therapy inhibits the growth of the main tumor mass by 35%, although statistically no significant differences were achieved between groups. However, 4-MU treatment significantly reduced the number of satellites nodules when compared with untreated tumorbearing mice. In addition, histological analysis revealed that tumor tissue in animals treated with 4-MU presented extensive areas of necrosis, indicating tumor cell death, and tumor inflammatory infiltrate. Besides, we observed that 4-MU treatment was associated with a reduced development of severe ascites in mice with HCC and fibrosis. In order to support these results, we carried out in vitro proliferation and apoptosis assays to evaluate the effect of 4-MU on murine (Hepa129 cells) and human (Hep3B and Huh7) HCC cell lines. As expected, and most likely because of the great amount of HA produced by Hepa129, 4-MU inhibited proliferation and induce the apoptosis of these cancer cells. The same tendency regarding the inhibition of cell proliferation and the induction of tumor cell apoptosis was observed in Hep3B cells, which also produced HA although in low amount. Contrarily, Huh7 cells that synthesize low amounts of HA presented a partial resistance to 4-MU treatment, confirming that 4-MU activity depends on the inhibition of HA synthesis (Kultti et al. 2009; Lokeshwar et al. 2010). The toxicity of 4-MU was determined in vivo in healthy mice as well as in vitro in primary hepatocytes culture. The toxicity profile of 4-MU was investigated in mice for their general behavior and body weight with no signs of evident toxicity. No significant changes were observed in plasma ALT levels in all studied groups. In addition, hepatic architecture in healthy animals treated with 4-MU was well preserved and hepatocytes isolated from healthy mice were resistant to 4-MU toxicity in all doses analyzed. Therefore, we assume that 4-MU is well tolerated and has a very low profile of hepatic toxicity. In agreement with others, we observed that in fibrotic livers HA is scattered and presented within fibrotic bridges (Urashima et al. 1999; George et al. 2004). Both fibrosis and HCC increased the amount of HA present in the liver tissue and serum, which could be a result of increased synthesis and/or deficient clearance (Fraser et al. 1997). As expected, 4-MU treatment reduced the hepatic HA content and serum levels in mice with advanced fibrosis with HCC. These results might indicate that the effects induced by 4-MU treatment are closely related to its capability to inhibit HA synthesis. Hepa129 cells produced larger amounts of HA and 4-MU reduced not only intratumoral but also non-tumoral HA deposition. It has been observed that HA MW may determine its biological function (Toole 2004). We analyzed the MW of hepatic HA by fast protein liquid chromatography (FPLC) and observed that 4-MU seems not to affect HA function by altering its weight, but only its amount (not shown). ECM synthesis and composition, in particular collagen fibrillar formation, are critical steps during hepatic fibrogenesis. In this sense, we showed in this work that 4-MU therapy ameliorates the degree of fibrosis in mice chronically intoxicated with TAA. Accordingly, tumor-bearing animals that received TAA showed the F2-3 fibrosis stage and those animals treated with 4-MU showed the F1-2 fibrosis stage as assessed by the Metavir score. Densitometric analyses of collagen deposition from tissues stained with Sirius Red were significantly reduced in 4-MU-treated mice when compared with untreated mice. Myofibroblasts do normally accumulate in the parenchyma close to liver injury areas. Many of them most likely derive from resident HSCs, the main source of fibrillar collagen. By mean of immunohistochemical studies, a decrease in the number of -SMA+ myofibroblasts was shown in fibrotic livers of 4-MU treated mice. In addition, when the action of 4-MU on hepatic myofibroblast-like or fibroblast cells was analyzed, we observed that all doses of 4-MU induced significant apoptosis and inhibited cell proliferation, indicating a high dependence on the synthesis of HA for survival, supporting the notion that 4-MU has an antifibrogenic activity. Recent studies have shown that HA is essential in differentiation of myofibroblast cells by regulating transforming growth factor (TGF)- response, main mediator of fibrotic progression (Webber, Jenkins, et al. 2009; Webber, Meran, et al. 2009). Thus, further studies are needed to evaluate a possible mechanism involving 4-MU as a modulator of TGF- response in the context of liver fibrosis. In conclusion, our study shows that 4-MU, a well-known HA synthesis inhibitor, is a non-toxic compound that demonstrates strong anticancer effects and antifibrotic properties. Therefore, we suggest a role for 4-MU as an anticancer agent in HCC associated with advanced fibrosis. Materials and methods Animals, cells and reagents Six-to-eight-week-old male C3H/He mice were purchased from Centro Atmico Ezeiza, Bs. As., Argentina. Animals were maintained at our Animal Resources Facilities (School of Biomedical Sciences, Austral University) in accordance with the experimental ethical committee and the NIH guidelines on the ethical use of animals. Hepa129 cells (HCC syngeneic with C3H/He mice) were kindly provided by Dr. Volker Schmitz (Bonn University, Germany) and grown in RPMI 1640 (GIBCO, Invitrogen Argentina, Buenos Aires, Argentina) with 10% fetal bovine serum (FBS). Hep3B (ATCC, Manassas, VA 20108, USA), Huh7 and LX2 (human HSCs were kindly provided by Dr. Scott Friedman, Mount Sinai, School of Medicine, NY) were grown in Dulbeccos modified Eagles medium (DMEM; GIBCO) with 5% FBS. CFSC-2G cells (activated HSC lines) kindly provided by Dr. Marcos Rojkind (Albert Einstein College of Medicine) were grown in MEM (GIBCO) with 10% FBS. IZA2.1 (an immortalized hepatic fibroblast cell line) provided by Dr. Ignacio Melero (University of Navarra, Spain) was grown in DMEM (GIBCO) with 10% FBS. Primary culture of murine hepatocytes was obtained through liver perfusion with collagenase type V (Sigma-Aldrich, Buenos Aires, Argentina) and maintained in 70% DMEM30% F12 medium with 10% FBS, 8.5 ng/mL choleric toxin, 10 ng/mL epidermal growth factor and 0.8 g/mL hydrocortisone (De Smet et al. 1998). For in vitro toxicity study, these cells (5000/well) were incubated for 24 h and then treated with 4-MU (Sigma-Aldrich) at different concentrations (0.25, 0.5, 1 and 5 mM) or control vehicle solution [Hanks balanced salt solution (HBSS)] for additional 48 h. CD44 expression and HA binding Briefly, cells (5 105) were incubated with different mAbs: anti-mouse CD44 (an HA-binding blocker antibody, KM81, 1 g/100 L) and anti-mouse PE for Hepa129 cells and antihuman CD44-PE (dilution 1:11, DB105, Miltenyi-Lab Systems S. A. Buenos Aires, Argentina) for Hep3B and Huh7, respectively, or isotype control (BD Biosciences, Becton Dickinson Argentina, Buenos Aires, Argentina) on ice for 30 min. Cells were washed thoroughly with PBS1% bovine serum albumin (BSA), fixed with 1% paraformaldehyde and subjected to flow cytometry (FACSCalibur, Becton Dickinson Immunocytometry Systems, San Jose, CA). HA binding was detected by incubation with HA-fluorescein isothiocyanate (FITC) (20 g/100 L, Calbiochem SA, Bs. As., Argentina), and CD44 displacement was performed by blockade with 10 L/100 L anti-CD44 (KM81) for 30 min. Hep3B and Huh7 as growth as adherent monolayer for all assay cells were harvested using a cell scraper in order to avoid cell-surface receptor cleavage. Analysis by flow cytometry was performed with a FACSAria flow cytometer (Becton Dickinson Immunocytometry Systems), and data were analyzed by WinMDI software (The Scripps Research Institute, La Jolla, CA). Cell proliferation and apoptosis assays For the proliferation assay, cells (2.5 105/well) were incubated in a 24-well plate with 4-MU at different concentrations (0.25, 0.5, 1 and 5 mM). After 48 h, cultures were pulsed with 5 Ci/mL [methyl-3H] thymidine for the last 6 h. Then, cells were lysed and the incorporation of radioactivity was measured in a liquid scintillation -counter (Beckman LS 6500). Morphological features associated with apoptosis were analyzed by acridine orange and ethidium bromide staining. Cells (Hepa129, Hep3B, Huh7, IZA2.1, CFSC-2G, LX2 and freshly isolated hepatocytes) were treated with 4-MU as in the proliferation assay. Forty-eight hours later, cells were resuspended in the dye mixture (100 g/mL acridine orange and 100 g/mL ethidium bromide in PBS) and visualized by fluorescence microscopy (Nikon Eclipse E800, Nikon, Buenos Aires, Argentina). The percentage of apoptotic cells or apoptotic index was calculated as: apoptotic cells (%) = (total number of cells with apoptotic nuclei/total number of cells counted) 100. The percentage of apoptosis for each treatment was calculated by subtraction of spontaneous apoptosis from induced apoptosis [treated apoptotic cells (%) untreated cells (%)] (Alaniz et al. 2006). In order to confirm apoptosis results by acridine orange and ethidium bromide staining, Annexin V kit detection (BD Pharmingen, Becton Dickinson, Argentina) was used (Alaniz et al. 2006). Analysis by flow cytometry was performed with a FACSAria flow cytometer (Becton Dickinson Immunocytometry Systems), and data were analyzed by WinMDI software (The Scripps Research Institute). HA quantification by an ELISA-like assay HA from cell-free supernatants, serum and liver tissues were measured using a competitive ELISA-like assay as described elsewhere (Cordo-Russo et al. 2009). Briefly, 96-well microplates (Nunc, Lobov y Cia S.A, Ciudad de Buenos Aires, Argentina) were coated with 100 g/mL of HA (CPN spol.s.r. o., Czech Republic). Then, wells were incubated with 25 L of sample (serum diluted 1/50, tissues diluted 1/800) or standard HA (01 g/mL), in the presence of 0.75 g/mL of biotinylated HA-BP (bHA-BP; # 385911, Calbiochem) at 37C for 4 h and then washed with PBS0.05% Tween-20. The bHA-BP bound to the wells was determined using an avidin biotin detection system (Sigma-Aldrich, Buenos Aires, Argentina). Sample concentrations were calculated from a standard curve generated by plotting the absorbance at 490 nm against the concentration of HA (Cordo-Russo et al. 2009). Cell-free culture supernatants were obtained from tumor cell treated or not treated with 4-MU at different concentrations (0.25, 0.5, 1 and 5 mM) at 12 h for Hepa129 (106 cells/mL) or 24 h for Hep3B and Huh7 (106 cells/mL); the HA level was normalized to the supernatant volume. For liver HA analysis (100 mg of tissue), the GAG extract was obtained as described (Cordo-Russo et al. 2009). Briefly, frozen samples of a liver (50 mg each) were suspended in 0.1 M ammonium acetate, pH 7.0, and digested with 20 U/mL of protease K (EC 3.4.21.64 Finnzymes Espoo, Finland), at 60 C for 2 h followed by enzyme inactivation by heating at 95C for 5 min. GAGs were precipitated with ethanol 96% at 20 C overnight. After centrifugation at 10,000 g at 4C for 45 min, pellets were washed with ethanol and resuspended in PBS. HA from serum was obtained from blood with drawn by retro-orbital bleeding, centrifuged at 0.8 g for 20 min at 4C, and the decanted serum was stored at 80C. Experimental model Fibrosis was induced by intraperitoneal (i.p.) injections of TAA (at a dose of 200 mg/kg; Sigma-Aldrich) thrice weekly for 42 days (Figure 3A; Salguero Palacios et al. 2008). On day 30, fibrotic mice were anesthetized and orthotopic tumors were established by subcapsular inoculation of 1.25 105 Hepa129 cells into the left liver lobe by laparotomy (Kornek et al. 2008). Two days before tumor implantation, a group of mice received 4-MU treatment by i.p. injection at a dose of 20 mg/kg per day, for 14 days (HCC + TAA + 4-MU group). A group of mice with fibrosis induced by TAA was also included as control (TAA group). Twelve days after tumor implantation, mice were sacrificed, and ascites grade (mild, moderated, severe and hemorrhagic) and tumor volume and numbers of HCC satellites were quantified. Pathology Formol-fixed liver samples were embedded in paraffin and 5 m sections were stained by hematoxylin/eosin (H&E), Massons trichrome and Sirius Red. Fibrosis was assessed according to the Metavir scoring system (no fibrosis F0, cirrhosis F4; Beaussier et al. 2005). The analysis was performed in a blinded fashion by a single pathologist. HA staining HA staining was performed as described elsewhere (Jameson et al. 2005). Briefly, paraffin liver sections were incubated with 3% H2O2methanol for 30 min at room temperature (RT) to block endogenous peroxidase, followed by avidin, biotin and protein-blocking solution (Vector, BIOARS S.A. Ciudad de Buenos Aires, Argentina). Then, 5 g/mL of bHA-BP (# 385911, Calbiochem) diluted in 1% BSAPBS was applied for 1 h. Negative controls were stained with bHA-BP and pretreated with 100 U/mL of Streptomyces hyaluronidase (# 385931, Calbiochem) at 37C for 30 min. Peroxidase complex (Sigma) 1:10 in PBS was used as a revealing system. The signal was detected by 0.1% diaminobenzidine (Sigma-Aldrich, Buenos Aires, Argentina), 4% glucose, 0.08% ClNH4, 5% nickel ammonium sulfate in 0.2 M AcNa and 0.05% H2O2. Immunohistochemistry for -SMA Briefly, paraffin liver sections were incubated with 3% H2O2 methanol for 30 min at RT to block endogenous peroxidase, followed by avidin, biotin and protein-blocking solution (Vector). Sections were incubated with a rabbit polyclonal antibody to -SMA (Abcam, Tecnolab S.A. Buenos Aires, Argentina) 1/10 in 0.2% BSA0.1% Triton X-100 in PBS over night (O.N.) at 4C, and a biotinylated anti-rabbit secondary antibody (Vector) 1/100 in 0.2% BSA in PBS at RT for 1 h. The signal was detected as described for HA staining (Materials and Methods section). Collagen and -SMA quantification -SMA-positive fibroblasts and collagen fibrils from liver sections were detected by analyzing 50 images/section (n = 4) taken at 200 (Nikon) and percentage of the positive area was calculated with ImageJ software (National Institutes of Health, Bethesda, MD). Transaminases measurement Serum alanine transaminase (ALT) was measured using an ARCHITECT (Abbott, Buenos Aires, Argentina) autoanalyzer. Cytotoxicity assay Freshly isolated hepatocytes or Hepa129 cells were plated on 5000 hepatocytes/well, and Cytotoxicity Detection Kit PLUS (LDH, Roche, Buenos Aires, Argentina) was used to detect lactate dehydrogenase (LDH) activity. Absorbance was read at 490 nm. Toxicity (%) = (experimental value low control)/ (high control low control) 100, where low control means cells with assay medium and high control cells with assay medium and lysis buffer. Statistic analysis Values were expressed as the mean SEM. The Mann Whitney or KruskallWallis (ANOVA) tests were used to evaluate the statistical differences between two groups or more than two groups, respectively. A P-value of <0.05 was considered as significant. Prism software (Graph Pad, San Diego, CA) was employed for the statistical analysis. Funding This work was supported by grants from Mizutani Foundation for Glycoscience (80072), Austral University, Agencia Nacional de Promocin Cientfica y Tecnolgica (PICT-2006-1882; PICT-2005-34788, PICT 2007-00736, PICTO-CRUP 2005-31179 and PICT 2010-2818). Acknowledgements We would like to thank Soledad Arregui and Guillermo Gastn for their expert technical assistance. Conflict of interest None declared. Abbreviations ALT, alanine transaminase; bHA-BP, biotinylated HA-BP; BSA, bovine serum albumin; DMEM, Dulbeccos modified Eagles medium; ECM, extracellular matrix; ELISA, enzymelinked immunosorbent assay; FBS, fetal bovine serum; FITC, fluorescein isothiocyanate; FPLC, fast protein liquid chromatography; GAG, glycosaminoglycan; HA, hyaluronan; HA-BP, HA binding protein; HBSS, Hanks balanced salt solution; HCC, hepatocellular carcinoma; H&E, hematoxylin/ eosin; HMW, high MW; HSC, hepatic stellate cells; i.p., intraperitoneal; LDH, lactate dehydrogenase; mAb, monoclonal antibody; 4-MU, 4-methylumbelliferone; MW, molecular weight; O.N., over night; RT, room temperature; SMA, smooth muscle actin; TAA, thioacetamide; TGF, transforming growth factor.


This is a preview of a remote PDF: https://glycob.oxfordjournals.org/content/22/3/400.full.pdf

Flavia Piccioni, Mariana Malvicini, Mariana G Garcia, Andrés Rodriguez, Catalina Atorrasagasti, Nestor Kippes, Ignacio T Piedra Buena, Manglio M Rizzo, Juan Bayo, Jorge Aquino, Manuela Viola, Alberto Passi, Laura Alaniz, Guillermo Mazzolini. Antitumor effects of hyaluronic acid inhibitor 4-methylumbelliferone in an orthotopic hepatocellular carcinoma model in mice, Glycobiology, 2012, 400-410, DOI: 10.1093/glycob/cwr158