Potential conflict of interest: Nothing to report.
Senescence marker protein 30 (SMP30), an important aging marker molecule that is highly expressed in the liver, has been known to protect hepatocytes from apoptosis by the synthesis of vitamin C. To explore the function of SMP30 in liver fibrosis, the effect of SMP30 deficiency on liver fibrosis was investigated in SMP30 knockout (KO) mice. Moreover, the in vivo results were further confirmed by way of hepatic stellate cell (HSC) isolation. We demonstrated that carbon tetrachloride (CCl4)-induced liver fibrosis and the nuclear translocation of p-Smad2/3, the immediate downstream of transforming growth factor beta (TGF-β), were significantly inhibited in the liver of SMP30 KO mice compared with wildtype (WT) mice. We also confirmed that both WT and SMP30 KO HSCs did not express SMP30. Finally, we further confirmed that up-regulation of peroxisome proliferator-activated receptor-gamma (PPAR-γ) caused by a lack of vitamin C was the pivotal factor in the mechanisms for attenuated liver fibrosis of SMP30 KO mice, and feeding with vitamin C restored CCl4-induced liver fibrosis in SMP30 KO mice. Conclusion: Vitamin C deficiency by SMP30 depletion attenuated liver fibrosis by way of up-regulated PPAR-γ expression in SMP30 KO mice. Our results provide, for the first time, the possible mechanisms underlying inhibition of HSC activation associated with vitamin C and PPAR-γ up-regulation in liver fibrosis of SMP30 KO mice. (HEPATOLOGY 2010.)
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
To date, it has been believed that the activation of hepatic stellate cells (HSCs) plays a pivotal role in the development of liver fibrosis.1–5 In reaction to liver injury by virus, chemicals, drugs, ischemia, or metabolic disorder, HSCs undergo phenotypic changes from a quiescent stage to an activated stage.
SMP30 is a 34-kDa aging marker protein that has high expression levels in the liver, kidney, and lung and decreases with the aging process.6–8 SMP30 contains gluconolactonase activity, which is involved in L-ascorbic acid (vitamin C) biosynthesis.8 Moreover, previous studies have shown that SMP30 prevents the apoptosis and necrosis of hepatocytes.9–11
According to our previous data,12 Smad3 knockout (KO) mice showed significantly increased levels of SMP30 and attenuated liver fibrosis as compared with WT mice. These data suggest the possibility that Smad3 expression might be related to SMP30 expression levels. The transforming growth factor beta (TGF-β)/Smad3 pathway is believed to be the most important pathway in the activation of quiescent HSCs to myofibroblasts.12, 13 The induction of collagen expression is mediated by the nuclear translocation of these phosphorylated Smads complexes composed of phosphorylated Smad2 (p-Smad2), phosphorylated Smad3 (p-Smad3), and Smad4.14–16
In contrast to the TGF-β/Smads signaling pathway activating HSCs, peroxisome proliferators-activated receptor-γ (PPAR-γ) has recently been identified as an important negative regulator in HSCs activation.17–20 PPAR-γ expression levels and activity are markedly down-regulated during the HSC activation process.17–21 Furthermore, stimulation of PPAR-γ not only inhibits HSC activation but also induces a phenotypic switch from activated HSCs to quiescent HSCs.19–22 Our previous unpublished data revealed significantly increased PPAR-γ levels and an elevated number of hypertrophic HSC in the liver of aged SMP30 KO mice compared with that of same-aged WT mice.
Taken together, these results suggest the possibility that SMP30 may act on TGF-β/Smad3 signaling and PPAR-γ expression. In order to ascertain the role of SMP30 in liver fibrosis, the present study was performed using SMP30 KO mice.
SMP30 KO mice were created as described.10 The WT C57BL/6 mice and SMP30 KO mice were housed and bred in a room at 22 ± 3°C, relative humidity 50 ± 10%, a 12-hour light-dark cycle, and were given food and water ad libitum. The genomic DNA was purified from mouse tail tissue using a combination of several procedures found in the literature.23 SMP30 KO mice genotyping was performed by polymerase chain reaction (PCR) as described.9 Animal procedures were performed in accordance with the National Institutes of Health (NIH) guidelines for the care and use of laboratory animals.
Experiment Design 1.
The 12-week-old, male, specific pathogen-free SMP30 KO mice (n = 14) and WT mice (n = 14) weighing 23-25 g were used and both WT mice and SMP30 KO mice were divided into two groups. Liver fibrosis was induced by CCl4 (Sigma, St. Louis, MO) injections three times a week at a dose of 1 mL/kg body weight (10% CCl4) dissolved in olive oil (Sigma) for 16 weeks. The WT mice and SMP30 KO mice control groups received intraperitoneal olive oil injections (1 mL/kg body weight).
Experiment Design 2.
The 8-week-old, male, specific pathogen-free WT mice (n = 6) and SMP30 KO mice (n = 12) were divided into three groups: a WT group (n = 6), an SMP30 KO group without vitamin C (n = 6), and a vitamin C-treated SMP30 KO group (n = 6). All mice groups were given a vitamin C-free diet and vitamin C was provided in the drinking water (L-ascorbic acid, 1.5 g/L) during the experiment period, which lasted for 16 weeks.
Experiment Design 3.
The 8-week-old, female, WT mice (n = 21) and SMP30 KO mice (n = 15) were divided as follows: a WT group (n = 7), a CCl4-treated WT group (n = 7), a CCl4+vitamin C WT group (n = 7), an SMP30 KO group (n = 5), a CCl4-treated SMP30 KO group (n = 5), and a CCl4 + vitamin C SMP30 KO group (n = 5). Liver fibrosis was produced by intraperitoneal injection of 10% CCl4 (1 mL/kg) three times a week for 16 weeks. The WT mice and SMP30 KO mice control groups received the same volume of vehicle (olive oil, 1 mL/kg, intraperitoneal). Vitamin C was provided in drinking water (L-ascorbic acid, 1.5 g/L) during the experiment period of 16 weeks.
The other methods are described in the Supporting Materials as follows: histopathology and immunohistochemistry, immunoblotting, determination of hepatic hydroxyproline content, reverse transcription PCR (RT-PCR), measurement of reactive oxygen species (ROS), and lipid peroxidation, transferase-mediated dUTP nick-end labeling (TUNEL) assay, serum vitamin C measurement by high-performance liquid chromatography (HPLC) as well as isolation and culture of HSCs.
All results taken from each group are expressed as mean ± standard deviation (SD). The statistical significance between experimental groups was determined by Student's t test or one-way analysis of variance (ANOVA) using GraphPad InStat (v. 3.05, GraphPad Software). Statistical significance was set at P < 0.05 or P < 0.01.
Attenuated CCl4-Induced Liver Fibrosis in SMP30 KO Mice.
The CCl4-treated WT mice revealed significantly increased collagen accumulation, forming a bridging fibrosis between the central veins as compared with the CCl4-treated SMP30 KO mice (Fig. 1A,B). The WT mice also showed much greater hepatic micronodular changes, whereas SMP30 KO mice did not reveal significant changes (Fig. 1A). The results were confirmed by way of a hepatic hydroxyproline content analysis, which showed that the CCl4-treated SMP30 KO mice exhibited significantly lower levels in comparison to that of the CCl4-treated WT mice (Fig. 1C). Consequently, we investigated the expression of alpha smooth muscle actin (α-SMA), a marker of activated HSCs. As expected, the CCl4-treated SMP30 KO mice group exhibited much lower numbers of α-SMA immunopositive cells per field compared with that of the CCl4-treated WT mice (Fig. 1D,E). We confirmed identical immunoblot results (Fig. 1F,G). These data indicate that CCl4-induced liver fibrosis is inhibited in SMP30 KO mice and suggest that SMP30 might play an important role in the HSC activation.
Expression of SMP30 and Serum Vitamin C Levels in CCl4-Treated WT Mice and SMP30 KO Mice.
In the WT mice the CCl4 treatment induced a decreased level of SMP30 expression around the central vein characterized by necrotic hepatocytes and infiltration of inflammatory cells compared with that of the control group. However, in the SMP30 KO mice group we could not detect an SMP30 expression, confirmed with the SMP30 KO mice (Fig. 2A,B). The results, using immunoblotting and RT-PCR for SMP30 expression, were observed to be the same as the results obtained using immunohistochemistry (Fig. 2C-E). In serum vitamin C level measurements, the control group of the WT mice indicated normal serum vitamin C levels, whereas the serum vitamin C level of the CCl4-treated WT mice was significantly decreased. In SMP30 KO mice, the serum vitamin C was undetectable in both the control group and the CCl4-treated groups (Fig. 2F). These data reveal that the SMP30 expression significantly decreased due to CCl4-induced liver injury.
Nuclear Translocation of p-Smad2/3 Is Inhibited in the Liver of SMP30 KO Mice Compared With WT Mice.
It was noticed that the SMP30 KO mice exhibited higher TGF-β expression levels in comparison with those of the WT mice (Fig. 3A,B). To evaluate p-Smad3, downstream of TGF-β1, expression levels in CCl4-treated WT mice and SMP30 KO mice, immunoblotting and immunohistochemistry were performed. In immunoblot results, whole liver tissues of SMP30 KO mice exhibited an elevated total of p-Smad3 expression levels compared with that of WT mice (Fig. 3C,D). However, in immunohistochemistry, CCl4-treated WT mice, exhibited significantly higher numbers of nuclear p-Smad2/3-positive parenchymal cells and nonparenchymal cells, compared with CCl4-treated SMP30 KO mice (Fig. 3E-G). We observed more critical differences in nonparenchymal cells than in hepatocytes, which means the nuclear translocation of p-Smad2/3 was more severely inhibited in nonparenchymal cells, including HSCs and inflammatory cells. To confirm the immunohistochemistry results, we extracted nuclear proteins from the whole liver tissue for immunoblotting. The cytoplasmic p-Smad3 expression showed the same expression pattern as the total p-Smad3 expression pattern (Fig. 3C,H). Additionally, extracts of nuclear proteins also revealed well-matched results with the immunohistochemical nuclear p-Smad2/3 expression (Fig. 3C,I). Surprisingly, CCl4-treated SMP30 KO mice showed a significantly lower level of ROS generation and lipid peroxidation compared with CCl4-treated WT mice (Fig. 3J,K), suggesting that CCl4 metabolism is much lower in SMP30 KO mice than in WT mice. These data suggest that CCl4-induced liver fibrosis might be inhibited in SMP30 KO mice due to inhibition of the nuclear translocation of p-Smad2/3 and a lower level of ROS and lipid peroxidation as compared with WT mice.
HSCs Isolated From Both WT Mice and SMP30 KO Mice Do Not Express SMP30.
To determine if activated HSCs express SMP30 in the fibrotic liver, we performed immunohistochemistry using SMP30 antibody and α-SMA antibody on serial liver sections. As shown in Fig. 4A, nonparenchymal cells exhibited no expression of SMP30 (Fig. 4A, a, arrowheads and b, arrows), whereas hepatocytes revealed obvious nuclear and cytoplasmic expression of SMP30 (Fig. 4A, a and b, asterisk). In normal livers of WT mice, the quiescent HSCs containing lipid droplets in their cytoplasm also showed depletion of SMP30 (Fig. 4A, a, arrowhead). To confirm more clearly whether HSCs from WT mice express SMP30, we performed immunocytochemistry and RT-PCR analysis using isolated HSCs. The isolated HSCs were cultured for 6 days in serum-containing medium and the SMP30 messenger RNA (mRNA) expression was determined on day 0, day 3, and day 5. As expected, HSCs from the WT mice and SMP30 KO mice revealed obvious SMP30 deficiency (Fig. 4B,C). Immunocytochemistry also showed well-matched results with the RT-PCR analysis confirming HSCs from the WT mice and the SMP30 KO mice do not express SMP30 (Fig. 4B). These data demonstrated that SMP30 is not involved directly in the activation of HSCs, suggesting the possibility of the participation of other up-regulated or down-regulated factors affecting hepatocytes and HSCs in the liver of the SMP30 KO mice.
Enhanced Expression of PPAR-γ Is Associated With Inhibited HSC Activation in SMP30 KO Mice.
As expected, the SMP30 KO mice liver tissue showed significantly enhanced PPAR-γ expression levels and mRNA levels compared with those of the WT mice (Fig. 5A,B). In order to compare the expression level of PPAR-γ, p-Smad2/3, α-SMA, and the activation degree of SMP30 KO HSC with WT HSC, HSCs were isolated and cultured in serum containing medium for 7 days. It was found that WT HSCs were activated faster compared with SMP30 KO HSCs until day 5 (Fig. 5C). Moreover, both the α-SMA expression and the p-Smad2/3 nuclear expression were much stronger in WT HSCs than in SMP30 KO HSCs (Fig. 5C). Additionally, it was observed that SMP30 KO HSCs contained a greater number of cytoplasmic lipid droplets compared with WT HSCs at the same time (Fig. 5D), which was well-matched with the HSC hypertrophy morphology in vivo in our previous unpublished data. For the sake of clarity, we used an RT-PCR analysis. On day 0, day 3, and day 5 the α-SMA mRNA expression levels of SMP30 KO HSCs were significantly inhibited compared with those of WT mice HSCs (Fig. 5E). The PPAR-γ expression levels showed time-dependent decreases in both WT mice HSCs and SMP30 KO HSCs. However, SMP30 KO HSCs revealed much greater PPAR-γ expression levels compared with WT HSCs at the same time (Fig. 5E). We observed that PPAR-γ negatively down-regulated α-SMA mRNA expression levels. These data demonstrate that the inhibition of SMP30 KO HSC activation was caused by an up-regulated PPAR-γ expression and this higher PPAR-γ level in freshly isolated SMP30 KO HSCs, compared with WT HSCs, might result in a significant reduction in both α-SMA mRNA expression and p-Smad2/3 nuclear expression.
Vitamin C Deficiency Induces Up-Regulated PPAR-γ Expression in the Liver Tissue of SMP30 KO mice.
To establish whether vitamin C deficiency induces up-regulation of PPAR-γ expression in the liver of SMP30 KO mice, we performed an additional animal experiment using 8-week-old WT mice and SMP30 KO mice as follows: a WT group (n = 6), a SMP30 KO group without vitamin C (n = 6), and a vitamin C-treated SMP30 KO group (n = 6) for a period of 16 weeks. Vitamin C was provided in the drinking water (L-ascorbic acid, 1.5 g/L) during the experimental period. Following immunoblot analysis, as expected, vitamin C-treated SMP30 KO mice revealed significantly decreased PPAR-γ expression levels in the liver tissue compared with nonvitamin C-treated SMP30 KO mice (Fig. 6A,B). These results indicate that vitamin C might be involved directly in the regulation of PPAR-γ expression in the liver. Therefore, it is believed that higher expression levels of PPAR-γ were caused by vitamin C deficiency in SMP30 KO mice.
Vitamin C Supplements Restored CCl4-Induced Liver Fibrosis in SMP30 KO Mice.
To assess reproducibility and whether vitamin C supplement restores CCl4-induced liver fibrosis in SMP30 KO mice, we performed another set of animal experiments using 8-week-old, WT mice, and SMP30 KO mice as follows; WT group (n = 7), CCl4-treated WT group (n = 7), CCl4+vitamin C WT group (n = 7), SMP30 KO group (n = 5), CCl4-treated SMP30 KO group (n = 5), and CCl4+vitamin C SMP30 KO group (n = 5), for an experiment period of 16 weeks. Interestingly, significantly increased liver fibrosis, measured by morphometry based on Masson's trichrome stain, was observed in the CCl4 + vitamin C SMP30 KO group compared with the CCl4-treated SMP30 KO group, whereas the WT mice showed no noticeable differences between the CCl4-treated WT group and the CCl4 + vitamin C WT group (Fig. 7A,B). These histological findings were further confirmed by measurement of the hydroxyproline content (Fig. 7C) and α-SMA expression level (Fig. 7D,E) in the liver, which demonstrated that vitamin C supplements restore CCl4-induced liver fibrosis in SMP30 KO mice. Taken together, these data suggest that vitamin C deficiency suppresses HSC activation following a CCl4-induced liver injury.
In this study we demonstrate for the first time that up-regulation of PPAR-γ expression by way of vitamin C deficiency inhibits HSC activation in SMP30 KO mice. We were led to accept that vitamin C deficiency caused by the absence of SMP30 can lead to: (1) ameliorated liver fibrosis; (2) inhibition of nuclear translocation of p-Smad2/3 in HSCs and hepatocytes; (3) higher PPAR-γ expression levels in SMP30 KO HSCs; (4) up-regulation of PPAR-γ, which is associated with vitamin C deficiency. Moreover, we confirmed that vitamin C supplement restores liver fibrosis in vitamin C-deficient SMP30 KO mice.
Although SMP30 is known as an antiapoptotic protein which rescues cells from necrosis and apoptosis, CCl4-treated SMP30 KO mice showed less progression of liver fibrosis (Fig. 1). Because SMP30 is expressed predominantly in the liver parenchymal cells, CCl4 administration causing liver injury induces a decreased SMP30 expression level and serum vitamin C level (Fig. 2). According to our previous studies,9 SMP30 KO mice exhibited more severe CCl4-induced acute centrilobular necrosis in comparison to WT mice. We may note that a substantial volume of previous research suggested that SMP30 plays a pivotal role as an anti-aging protein by way of the inhibition of oxidative stress, and also preventing cell apoptosis and necrosis.10, 24, 25 We pose the question: How do SMP30 KO mice exhibit more attenuated liver fibrosis in comparison to WT mice? In this study the CCl4-treated SMP30 KO mice showed significantly lower levels of ROS generation and lipid peroxidation compared with those of the CCl4-treated WT mice. In our previous acute single CCl4 administration study,9 the SMP30 KO mice showed a significantly lower CYP2E1 expression level compared with WT mice, which was induced by severe centrilobular necrosis of hepatocytes expressing CYP2E1 around the central vein. Because the hepatotoxicity of CCl4 depends on metabolism by CYP2E1, lowered CYP2E1 expression levels of CCl4-treated KO mice might be a pivotal factor in the lower ROS generation and lipid peroxidation compared with the CCl4-treated WT mice, as indicated in the present study. Moreover, stronger TGF-β, p-Samd3 expression levels (Fig. 3) and a greater number of apoptotic hepatocytes (Supporting Fig. 1) were observed in the livers of SMP30 KO mice in comparison to WT mice, although SMP30 KO mice showed a significantly lower ROS and lipid peroxidation level in this study. These data suggest that SMP30 KO mice have much more up-regulated p-Smad3 generation, although they revealed a significantly lower ROS generation compared with the WT mice after chronic CCl4 treatment. However, the present study showed that the nuclear translocation of p-Smad3 was inhibited in the SMP30 KO mice compared with the WT mice, which resulted in significantly attenuated liver fibrosis of the SMP30 KO mice in CCl4-induced liver fibrosis. We also confirmed that quiescent and activated HSCs do not express SMP30, which means that SMP30 is not involved directly in HSC activation. Therefore, we speculated that other up-regulated or down-regulated factors caused by vitamin C deficiency might affect the activation of HSCs in the liver of SMP30 KO mice.
PPAR-γ is one of the nuclear receptor superfamily of ligand-activated transcriptional factors, which is a critical factor in the development of adipose tissue in vivo and in vitro.26–28 Recently, PPAR-γ has been considered a potential molecular target for the inhibition of HSC activation. We observed significantly up-regulated PPAR-γ expression levels and increased numbers of hypertrophic HSCs containing excessive lipid droplets in the liver of the SMP30 KO mice compared with the WT mice (unpubl. data). Thus, we hypothesized that up-regulated PPAR-γ might inhibit liver fibrosis and HSC activation in the SMP30 KO mice.
In the present study the SMP30 KO mice revealed higher PPAR-γ expression levels and mRNA levels compared with the WT mice (Fig. 5A,B). In the culture of isolated HSCs, SMP30 KO HSCs showed delayed HSC activation, a higher PPAR-γ expression, a greater number of cytoplasmic lipid droplets, and inhibited α-SMA expression levels compared with WT HSCs. (Fig. 5C-E). Several previous studies have revealed that PPAR-γ ligands are associated with TGF-β/Smads signaling.29–31 In human HSCs, cotreatment with a synthetic PPAR-γ agonist revealed dose-dependent decreases of both Smad3 phosphorylation and collagen production.32 Moreover, a few previous studies have shown that treatment of a natural PPAR-γ agonist 15-PGJ2 or overexpression of PPAR-γ inhibited the nuclear translocation of p-Smad2/3 in rat kidney fibroblasts, mice ocular fibroblasts, and human fibrocytes.33–35 Consistent with previous studies, our study revealed decreased p-Smad2/3 nuclear translocation in the liver of SMP30 KO mice including parenchymal and nonparenchymal cells compared with those of WT mice (Fig. 3). These results can be explained by increased PPAR-γ expression in SMP30 KO mice livers (Fig. 5A,B). We also demonstrated inhibited p-Smad2/3 nuclear expression by way of immunocytochemistry in isolated SMP30 KO HSCs (Fig. 5C). Considered as a whole, our finding suggests that an up-regulated PPAR-γ level is the key negative regulator for a p-Smad2/3 nuclear translocation and an α-SMA expression in the SMP30 KO mice.
A previous study indicated that vitamin C significantly down-regulates the expression of PPAR-α, γ genes within mononuclear cells.36 In the current research we demonstrated that the increased PPAR-γ expression was induced by vitamin C deficiency in the liver of SMP30 KO mice (Fig. 6). We observed significantly down-regulated serum vitamin C levels (Fig. 2E) and up-regulated PPAR-γ expression in SMP30 KO mice (Fig. 5A,B). As expected by us, with additional animal experiments we observed negative regulation between serum vitamin C levels and PPAR-γ expression levels (Fig. 6B,C). Finally, we proved that vitamin C treatment reinstated liver fibrosis levels in the vitamin C-deficient SMP30 KO mice (Fig. 7).
Recently, a decrease of PPAR-γ expression with age was demonstrated37, 38 and age-related chronic inflammation resulted in much greater decreases in PPAR-γ levels.39 Moreover, the growth hormone receptor/binding protein KO mice, showing significantly up-regulated PPAR-γ levels in the liver, were characterized by markedly extended life-spans compared with the WT mice.40 These results show that decreases of the PPAR-γ expression with age-related inflammation plays a pivotal role in the aging process and suggests the possibility of an anti-aging role for PPAR-γ. On the other hand, Guan et al.41 reported that PPAR-γ activation induces the expression of p16INK4α and G1 arrest in human bladder cancer cells. Moreover, Lu et al.42 revealed that a PPAR-γ agonist accelerates TRAIL-induced apoptosis and cell cycle arrest in cancer cells. Additionally, a recent study reported that PPAR-γ promotes cellular senescence by inducing p16INK4α expression in human diploid fibroblasts.43 These previous studies indicate that PPAR-γ might play a pivotal role in cellular senescence. Returning to our study, the SMP30 KO mice revealed elevated PPAR-γ levels relative to the WT mice, although the SMP30 KO mice are well known to have a shorter life span. Related to this aging factor, our findings show that the inhibition of p-Smad2/3 nuclear translocation by overexpressed PPAR-γ suggests the possibility that the chronic inflammatory cells were severely suppressed to induce age-related chronic inflammation. Recently, Krizhanovsky et al.44 reported that senescent activated HSCs down-regulate extracellular matrix production, which suggests that the senescence of activated HSCs limits liver fibrosis. Additionally, several studies demonstrated that PPAR-γ is a negative regulator of various inflammatory responses.45, 46 Therefore, based on the previous studies, it is believed that the decreases in cellular responses, including inflammatory reactions, might be caused by cellular senescence of the SMP30 KO mice.
To summarize, we can conclude that vitamin C deficiency ameliorated liver fibrosis by way of up-regulated PPAR-γ expression (Fig. 8). We confirmed that WT HSCs did not express SMP30, and vitamin C supplement reinstated CCl4-induced liver fibrosis in the SMP30 KO mice; therefore, we speculated that vitamin C deficiency caused by a lack of SMP30 might be more closely connected with the inhibited liver fibrosis than SMP30 itself. Finally, we demonstrated that up-regulated PPAR-γ expression induced by vitamin C deficiency is the pivotal factor in the mechanisms for attenuated liver fibrosis of the SMP30 KO mice. Although the present study describes that vitamin C deficiency ameliorates CCl4-induced liver fibrosis in the SMP30 KO mice, the WT mice did not show significant differences in the fibrosis grade and α-SMA expression level between the CCl4-treated WT group and the CCl4 + vitamin C supplement WT group. Thus, it appears that vitamin C does not promote liver fibrosis in the WT mice, although vitamin C treatment increased CCl4-induced liver fibrosis grade in the SMP30 KO mice. The above findings have not been published so far, and would be novel evidence that liver fibrosis is ameliorated in the vitamin C-deficient aging animal model.