mTOR Overactivation in Mesenchymal cells Aggravates CCl4− Induced liver Fibrosis

Hepatic stellate cells are of mesenchymal cell type located in the space of Disse. Upon liver injury, HSCs transactivate into myofibroblasts with increase in expression of fibrillar collagen, especially collagen I and III, leading to liver fibrosis. Previous studies have shown mTOR signaling is activated during liver fibrosis. However, there is no direct evidence in vivo. The aim of this study is to examine the effects of conditional deletion of TSC1 in mesenchymal on pathogenesis of liver fibrosis. Crossing mice bearing the floxed TSC1 gene with mice harboring Col1α2-Cre-ER(T) successfully generated progeny with a conditional knockout of TSC1 (TSC1 CKO) in collagen I expressing mesenchymal cells. TSC1 CKO and WT mice were subjected to CCl4, oil or CCl4+ rapamycin treatment for 8 weeks. TSC1 CKO mice developed pronounced liver fibrosis relative to WT mice, as examined by ALT, hydroxyproline, histopathology, and profibrogenic gene. Absence of TSC1 in mesenchymal cells induced proliferation and prevented apoptosis in activated HSCs. However, there were no significant differences in oil-treated TSC1 CKO and WT mice. Rapamycin, restored these phenotypic changes by preventing myofibroblasts proliferation and enhancing their apoptosis. These findings revealed mTOR overactivation in mesenchymal cells aggravates CCl4− induced liver fibrosis and the rapamycin prevent its occurance.

activation. In response to numerous intracellular and extracellular stimuli, mTORC1 phosphorylates eukaryotic initiation factor 4E-binding protein-1 (4E-BP1) and S6 kinase 1 (S6K1), which exerts an essential role in increasing translation of a subset of mRNAs and accelerating growth and proliferation 6,8 . Under certain circumstances, inhibition of mTOR activity by rapamycin accelerates apoptosis 9,10 . Induction of mutations of mTOR can contribute to apoptotic resistance and might contribute to cellular transformation 11 .
Activation of the mTOR pathway has been reported in several fibrotic diseases. The mTOR pathway plays an important role in cardiac fibrosis and rapamycin is a potential therapeutic treatment that can be used to attenuate cardiac fibrosis 12 . The rapamycin analogue SDZ RAD was shown to promote dramatic inhibitory effects on collagen accumulation in the lungs in a bleomycin model of pulmonary fibrosis 13 , thus suppression of mTOR may be a viable treatment for pulmonary fibrosis 14 . mTORC1 signaling promotes the activation of kidney fibroblasts and contributes to the development of interstitial fibrosis 15 . There is growing evidence to support the idea that the mTOR signaling pathway plays a key role in liver fibrosis 16 . Activation of HSCs is regarded as a critical step in the pathogenesis of liver fibrosis 3 . HSCs are a resident mesenchymal cell type located in the subendothelial space of Disse that are interposed between the sinusoidal endothelium and hepatocytes 4 . Based on the findings of these studies, we speculate that activation of mTOR signaling in mesenchymal cells might also play a role in liver fibrosis.
To date, the specific effect of mTOR on liver fibrosis has not yet been reported. Therefore, the aim of the present study was to investigate the function of mTOR overactivation in mesenchymal cells in CCl 4 − induced liver fibrosis and to reveal possible mechanisms underlying its participation in fibrotic diseases.

Results
Generation of TSC1-conditional knockout mice in the mesenchymal compartment. Homozygous floxed TSC1 mice (TSC1 fl/fl ) were cross-bred with mice harboring a Cre-ER(T) recombinase gene to generate Cre/TSC1 heterozygous mice. The second cross generated [TSC1 fl/fl , Col1α2-CreER(T)+ /0] mice. Genomic DNA Generation of TSC1 conditional knockout (CKO) mice in the mesenchymal compartment A: Mouse genotype was detected by PCR. Genomic DNA samples from the indicated strains of mice were analyzed using primers specific for wild-type (WT) and floxed (fl) TSC1, or Cre-ER (T) mice. The PCR products and 100-bp DNA ladder were separated by agarose gel electrophoresis. The bands corresponding to the WT (193 bp) and fl (230 bp) TSC1 alleles are indicated, and confirm the correct TSC1 genotype for each strain. The correct genotype with respect to Cre-ER(T) (500 bp) gene is confirmed in the respective strain. B: TSC1 knockout efficiency was detected by western blotting after daily injection of tamoxifen for 8 days. Liver tissue samples from WT and TSC1 CKO mice were analyzed for p-S6 (s235/236). The phosphorylation status of S6 was elevated, indicating TSC1 knock-down in the livers of TSC1 CKO mice. S6 and GAPDH were included as loading controls.
from tail tissue samples of these mice was genotyped by PCR analysis to confirm the presence of the TSC1 and Cre genes (Fig. 1A). To delete the TSC1 gene in the mesenchymal compartments, mice with genotype [TSC1 fll/fl , Col1α2-CreER(T)+ /0] were treated with tamoxifen as described in the Materials and Methods section and were referred to as TSC1 CKO mice. Mice with genotype [TSC1 fl/fl ,Col1α2-CreER(T)0/0] under the same treatment were referred to as WT mice. Deficiency of TSC1 expression in the mesenchymal compartment was verified by western blot and immunofluorescence analyses. As shown in Figs 1B and 5B, expression levels of p-S6 (s235/236), mTOR, and the S6K1 downstream effector protein were strongly increased in TSC1 CKO mice.

TSC1 deletion augments CCl 4 − induced liver fibrosis in mice.
To evaluate the effect of TSC1 deficiency in mesenchymal cells on liver fibrosis in vivo, TSC1 CKO and WT mice were treated with CCl 4 , and the fibrotic response was evaluated. A histologic examination of liver sections using H&E and Sirius Red staining was performed, which showed that TSC1 CKO mice exhibited greater degrees of lobular architecture damage and severe bridging necrosis than WT mice ( Fig. 2A). TSC1 CKO mice presented exacerbated fibrosis by the morphometric assessment of the total area positively stained with Sirius Red (Fig. 2B,C) and by higher liver hydroxyproline(HYP) content (174.09 ± 4.12 μ g/mg vs. 199.78 ± 8.41 μ g/mg, P < 0.05; Fig. 2D). Meanwhile, TSC1 CKO Tamoxifen-treated control [wildtype (WT)] and TSC1 conditional knockout (CKO) mice received intraperitoneal injections of either CCl 4 or oil, as indicated. H&E staining (A) and Sirius Red staining (B) were performed 8 weeks later. TSC1 CKO mice had more severe liver fibrosis than WT mice. All were imaged at 100x or 400x magnification. (C) The Sirius Red staining area in the liver was calculated by Image-Pro Plus 6.0. in six different images taken at 100x magnification on each slide. (D) The liver samples were analyzed for hydroxyproline content. (E) Liver injury was determined by serum ALT in WT and TSC1 CKO mice. The results are presented as the mean ± SEM derived from three independent experiments. NS, p > 0.05; *p < 0.05; # p < 0.01 for TSC1 CKO vs. WT mice. mice exhibited higher serum ALT concentrations relative to WT mice (350.00 ± 5.77 U/L vs. 496.67 ± 8.82 U/L, P < 0.01; Fig. 2E). In comparison, there were no apparent differences in HYP (112.94 ± 9.74 μ g/mg vs. 118.92 ± 5.07 μ g/mg, P = 0.62) and ALT (30.00 ± 1.00 U/L vs. 31.67 ± 1.45 U/L, P = 0.40) detected after oil treatment between WT and TSC1 CKO mice. Collectively, these results demonstrated that TSC1 deletion in the mesenchymal compartment augments CCl 4 − induced liver fibrosis in mice.
Increased α-SMA expression due to loss of TSC1. The results of the above analyses showed that loss of TSC1 resulted in exacerbation in CCl 4 − induced hepatic fibrosis. Consistent with enhanced liver fibrosis and collagen production, α -SMA positive staining was increased in CCl 4 − treated TSC1 CKO mice (Fig. 3B,C). Western blot analysis confirmed significant CCl 4 − induced increases in α -SMA protein expression in TSC1 CKO mice (Fig. 3A). However, there were no apparent differences in α -SMA expression detected after oil treatment between WT and TSC1 CKO mice. These results suggest that TSC1 deletion from mesenchymal cells promotes the development of liver fibrosis owing to the increasing number of α -SMA positive cells. Increased expression of profibrogenic markers in TSC1 CKO mice. RT-PCR was performed to quantify the expression of profibrogenic markers. As shown in Fig. 4A, levels of the profibrogenic markers α-SMA, Col1α1, TGF-β1, and TIMP2 were unchanged between the two genotypes following treatment with oil. In CCl 4 − induced animals, there was a trend towards increased levels of α-SMA(11.2-fold; P < 0.01), Col1α1 (8.3-fold; P < 0.01), TGF-β1(2.8-fold; P = 0.03), and TIMP2(4.2-fold; P < 0.01) in TSC1 CKO mice, as compared with WT mice (Fig. 4B). These results strengthened the evidence for the vulnerability of TSC1 CKO mice to CCl 4 − induced fibrosis.

Increased myofibroblasts in TSC1 CKO mice.
To further confirm the importance of mTOR activity in myofibroblast proliferation in CCl 4 − induced liver fibrosis, protein expression was monitored by western blot and immunofluorescence analyses. As expected, p-S6 (s235/236) expression was upregulated in myofibroblasts from CCl 4 − treated mice, but was higher in the TSC1 CKO mice than WT mice (Fig. 5A). Further detection of apoptotic proteins showed that expression levels of the cleaved form of caspase-3 and PARP were reduced in CCl 4 − treated mice, but was lower in the TSC1 CKO mice than the WT mice (Fig. 5A). Myofibroblast formation was measured by immunofluorescence analysis with α -SMA antibody. Consistent with the western blotting results, increased phosphorylation of S6 with CCl 4 treatment, at least partially, caused the number of myofibroblasts to increase in TSC1 CKO mice (Fig. 5B). Following CCl 4 treatment, ki67 and α -SMA double-positive cells, which are recognized as proliferative myofibroblasts, were significantly increased to a greater extent in TSC1 CKO mice than in WT mice (Fig. 5C). Meanwhile, the abundance of cleaved caspase-3 and α -SMA double-positive cells, which are recognized as apoptotic myofibroblasts, was decreased in TSC1 CKO mice, as compared with WT mice (Fig. 5D). Intriguingly, expression of α -SMA, ki67, and cleaved caspase-3 was not appreciably altered in oil group mice, indicating that under quiescent conditions, loss of TSC1 in cells expressing collagen type I was not accompanied by an increase in the abundance of myofibroblasts in the liver.

Rapamycin attenuates CCl 4 − Induced Liver Fibrosis and reverses phenotypes in TSC1 CKO mice.
It is known from the literature that the rapamycin has inhibitory effects on liver fibrosis 17 , and this is consistent with our results. In CCl 4 + rapamycin-treated CKO or WT mice, there was less severe liver histology injury and lower serum ALT concentrations compared with the mice merely treated with CCl 4 (Fig. 6A,B). CCl 4 + rapamycin-treated mice presented attenuated fibrosis by the morphometric assessment of the total area positively stained with Sirius Red (Fig. 6A,C). Immunohistochemical staining for α -SMA showed that α -SMA-positive staining was decreased in CCl 4 + rapamycin-treated mice (Fig. 6A,D). As shown in Fig. 6E, western blot analyses indicated that α -SMA and p-S6 protein expression were lower in the CCl 4 + rapamycin-treated mice in comparison with CCl 4 − treated mice, but the expression levels of the cleaved form of caspase-3 and PARP were   upregulated in CCl 4 + rapamycin-treated mice. Consistent with the western blotting results, decreased phosphorylation of S6 with rapamycin treatment, at least partially, caused the number of myofibroblasts to decrease (Fig. 6F). Ki67 and α -SMA double-positive cells were also significantly decreased to a greater extent in CCl 4 + rapamycin-treated mice (Fig. 6G). Meanwhile, the number of cleaved caspase-3 and α -SMA double-positive cells were increased in CCl 4 + rapamycin-treated mice, as compared with CCl 4 − treated mice (Fig. 6H). Intriguingly, there were no significant differences between the WT and the TSC1 CKO in CCl 4 + rapamycin-treated mice.
Thus, TSC1 deficiency in the mesenchymal compartment resulted in a significant increase in the abundance of myofibroblasts, thereby contributing to the noted exacerbation of liver fibrosis. Conversely, rapamycin can alleviated CCl 4 − induced liver fibrosis by inhibiting myofibroblast proliferation and inducing apoptosis, and rescued the specific phenomenon by TSC1 deficiency in the mesenchymal compartment.

Discussion
Fibrosis is a pathological process characterized by excessive accumulation of connective tissue components in organs and tissues. Liver fibrosis develops as a consequence of liver injury and activation of HSCs, which represents the imbalance between ECM production and degradation. Previous studies have reported that mTOR was involved in the process of liver fibrosis in human tissues and animal models of CCl 4 − mediated fibrosis 18,19 . Understanding the importance of mTOR signaling in different cell compartments is critical for future studies to develop effective drugs targeting this signaling pathway with minimal off-target or negative adverse effects. In this study, the role of aberrant mTOR activity in the mesenchymal compartment in the pathogenesis of liver fibrosis in a conditional TSC1 knockout mouse model was investigated, which showed that TSC1 deletion in mesenchymal cells aggravated liver fibrosis and was associated with myofibroblast proliferation and apoptosis.
Growing evidence supports the hypothesis that mTOR overactivation is involved in the pathogenesis of fibrotic diseases, including liver fibrosis 16,[20][21][22] . We specifically focused on the role of mesenchymal mTOR overactivation in liver fibrosis. Our data further indicated that the mTOR signaling pathway was activated in CCl 4 − mediated liver fibrosis (Fig. 5A,B), which is consistent with a previous study that reported mTOR activation in liver fibrosis in an animal model 19 . After obtaining evidence supporting our hypothesis that mTOR overactivation was involved in the process of liver fibrosis, we further demonstrated that mTOR activation resulting in increased myofibroblast accumulation in the liver could be a major mechanism underlying fibrosis formation.
When the Cre-ER(T) recombinase was activated by tamoxifen, TSC1 expression was selectively absent only in mesenchymal cells. HSCs, first described by Kupffer in the 19th century, have emerged in the past 25 years as remarkably versatile mesenchymal cells 4 . The mTOR signaling pathway may be activated in HSCs in TSC1 CKO mice (Fig. 1C). As previously reported using other methodologies, HSCs are the major source of myofibroblasts (> 87%) in CCl 4 − induced liver injury 23 .
Myofibroblasts are the major source of fibrogenic cytokines and ECM 24 . Increasing myofibroblast accumulation in TSC1 CKO mice in response to CCl 4 treatment is thought to contribute to the noted significant increase in liver fibrosis. This result is also consistent with a previous study in which rapamycin appeared to inhibit proliferation and differentiation of corneal myofibroblasts in vitro 25 . mTOR regulates renal myofibroblast activity and proliferation, and also plays an important role in preventing apoptosis of activated HSCs and promoting progression of fibrosis 26 . Apoptosis of HSCs may be involved in the termination of this response 27 . In a microarray study of human HSCs, cells became senescent and switched from a primarily fibrogenic to an inflammatory phenotype, with decreased proliferation and gene expression, and increased apoptosis 28 . PI3K/AKT as the upstream of the mTOR signaling pathway, inhibition of PI3K/Akt signaling pathway has been shown to induce HSCs apoptosis and attenuate liver fibrosis in vitro 29 . These findings suggest that reduced myofibroblast proliferation and increased myofibroblast apoptosis could effectively reduce fibrosis.
Interestingly, the CCl 4 − treated TSC1 CKO mice exhibited obvious differences in liver fibrogenesis relative to control mice. However, there were no significant differences in the oil group between TSC1 CKO and WT mice. These results imply that the effect of mTOR pathway activation in mesenchymal cell in normal mice is balanced by tissue homeostasis. In the experimental hepatic fibrosis model, after liver cell injury, mTOR pathway activation in mesenchymal cell enhanced the wound healing response.
In summary, mTOR overactivation in the mesenchymal compartment augmented liver fibrosis induced by CCl 4 . The mechanism likely involved activation of the mTOR pathway, which induced myofibroblast proliferation and prevented apoptosis. Although these results confirmed the importance of mesenchymal mTOR activity on myofibroblast proliferation in vivo, additional and/or alternate mechanisms cannot be excluded, thus additional future studies are warranted. Inhibition of mesenchymal mTOR overactivation and the induction of myofibroblasts present potential treatments for liver fibrosis.
by Image-Pro Plus 6.0. in six different images taken at 100x magnification on each slide. (D) a-SMA staining area in the liver was calculated by Image-Pro Plus 6.0. in six different images taken at 400x magnification on each slide. (E) Liver tissue samples from CCl 4 − or CCl 4 + rapamycin-treated mice were analyzed for PARP, cleaved caspase-3, p-S6, and S6 protein by western blot analysis. Immunofluorescence(IF) staining were performed to evaluate the impact of rapamycin on myofibroblasts proliferation and apoptosis. (F) Sections were stained with α -SMA (green) and p-S6 (red) antibodies. There were decreased α -SMA and pS6 double positive cells in mice administered CCl 4 + rapamycin. (G) Sections were stained with α -SMA (red) and ki67 (green) antibodies.α -SMA-and ki67 double positive cells were decreased in mice following CCl 4 + rapamycin treatment. (H) Sections were stained with α -SMA (green) and cleaved caspase-3 (red) antibodies. α -SMA-and cleaved caspase-3 double positive cells increased in the CCl 4 + rapamycin-treated mice. Tissues were counterstained with DAPI (blue) to detect nuclei and imaged at 400x magnification. # p < 0.01; *p < 0.05 for CCl 4 − vs. CCl 4 + rapamycin-treated mice. To selectively delete TSC1 in Cre expressing cells, Col1α2CreERT /TSC1 fl/fl , mice (6-8 weeks old) received daily intraperitoneal injections of the tamoxifen suspension (0.1 ml of diluted stock) for 8 days (TSC1 CKO), while TSC1 wild-type (WT) mice, as controls, received the same treatment. After administration of the tamoxifen treatment regimen, 8-12-week-old male TSC1 CKO and WT mice were induced by intraperitoneal injection with 20%CCl 4 (5 μ l/g body weight; Sigma Chemicals, Heidelberg, Germany) dissolved in olive oil for 8 weeks. The control group received the same volume of olive oil only. In the CCl 4 + rapamycin group, after treatment with CCl 4 , the mice also received rapamycin (1 mg/kg body weight/day; Sigma-Aldrich) by intraperitoneal injection daily until sacrifice.
Hydroxyproline Assay. The hydroxyproline assay was used to estimate the percentage of degraded collagen 30 . To assess the extent of fibrosis, the collagen content of liver homogenates after alkaline hydrolysis was assayed using a colorimetric assay (Nanjing Jiancheng Bioengineering Institute, Nanjing, China).
Alanine transaminase (ALT). Serum was obtained by cardiac puncture from anesthetized mice following overnight fasting for detection of ALT levels using an IDEXX Catalyst Dx Biochemical Analyzer (IDEXX Laboratories, Westbrook, ME, USA).
Histological analysis. For histological analysis, liver samples were fixed for 24 h in 4% paraformaldehyde, embedded in paraffin, sliced into 5 μ m-thick sections, and stained with hematoxylin and eosin (H&E)or Sirius Red, as previously described 31 .
Real-time polymerase chain reaction (RT-PCR) analysis. Total RNA was isolated using TRI reagent (Sigma Chemicals) and cDNA was generated using Primescript RT master mix (Takara Bio, Inc., Ōtsu, Japan). Duplicate PCR amplifications were carried out using a Light Cycler LC480 Real-Time PCR System (Roche, Basel, Switzerland) with the SYBR Green PCR Kit (Takara Bio, Inc.). mRNA concentrations were calculated using the 2 −ΔΔCt method. The primers used in this study were produced by Sangon Biotech Co., Ltd. (Shangai, China) and the sequences are listed in Table 1.
Statistical analysis. All data are expressed as the mean ± standard error of the mean (SEM) from three individual experiments. The data for each group were analyzed using the t-test with SPSS ver. 20.0 software (IBM-SPSS, Inc., Chicago, IL, USA). A probability (p) value of < 0.05 was considered statistically significant.