ASPP2 Inhibits the Profibrotic Effects of Transforming Growth Factor-β1 in Hepatic Stellate Cells by Reducing Autophagy

Human HSCs (LX-2 cells) were grown in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal bovine serum. After pre-incubation with GFP-adenovirus (Ad) or ASPP2 adenovirus (AdASPP2) at a concentration of 1 × 10⁶ PFU/ml for 24 h, the LX-2 cells were treated with or without TGF-β1 (10 ng/ml, PeproTech).

Total RNA was extracted from cultured LX-2 cells and liver tissue using an RNeasy kit (Qiagen) and reverse transcribed using the SuperScript^® III First-Strand Synthesis System (Invitrogen). Real-time RT-PCR was performed in the ViiA 7DX sequence detecting system (Applied Biosystems). Relative transcript levels of target genes were normalized with 18S rRNA levels. Primers were as follows: 18S rRNA: sense, 5′-GTA ACC CGT TGA ACC CCA TT-3′; antisense, 5′-CCA TCC AAT CGG TAG TAG CG-3′. Human Col a1(I): sense, 5′-CCC CAC TCA GCC CAG TGT-3′; antisense, 5′-ACC AGA CAT GCC TCT TGT CCT T-3′. Human Col a1(III): sense, 5′-GGT TTT GCC CCG TAT TAT GGA-3′; antisense, 5′-GAA GTC ATA ATC TCA TCG GTG TTG A-3′. Human a-SMA: sense, 5′-TGC CTG ATG GGC AAG TGA T-3′; antisense, 5′-GTC TCT GGG CAG CGG AAA-3′. Human Bax: sense, 5′-AGT GGC AGC TGA CAT GT-3′; antisense, 5′-AGG GCC TTG AGC ACC AGT-3′. Human Bcl-2: sense, 5′-ATG GCG CAC GCT GGG AGA A-3′; antisense, 5′-ATG GAT GTA CTT CAG CAC TAT-3′. Mouse Col a1(I): sense, 5′-AGG GCG AGT GCT GTG CTT T-3′; antisense, 5′-CCC TCG ACT CCT ACA TCT TCT GA-3′. Mouse Col a1(III): sense, 5′-TGA AAC CCC AGC AAA ACA AAA-3′; antisense, 5′-TCA CTT GCA CTG GTT GAT AAG ATT AA-3′. Mouse α-SMA: sense, 5′-ATG CTC CCA GGG CTG TTT T-3′; antisense, 5′-TTC CAA CCA TTA CTC CCT GAT GT-3′. Mouse ASPP2: sense, 5′-AGC TGC CAT GGA GAC CAT CT-3′; antisense, 5′-ACT GTT CTC CGT ACT GGC AC-3′.

Western blot analysis was performed using 50 μg of protein extract obtained as described previously [15]. Antibodies against a-SMA, ASPP2, and LC3B were purchased from Sigma Aldrich; polyclonal antibodies against caspase3 were purchased from Abcam. Peroxidase-conjugated goat anti-mouse or goat anti-rabbit IgG antibodies were used as secondary antibodies. Protein expression was visualized using an enhanced chemiluminescence (ECL Plus) assay kit according to the manufacturer’s instructions. The results were normalized relative to the expression of glyceraldehyde-3-phosphate dehydrogenase (GAPDH).

Liver samples were fixed in 4% paraformaldehyde and embedded in Tissue Tek OCT compound (Sakura Finetek USA, Inc.); 7-μm frozen sections were used for immunofluorescence. They were blocked with 2% bovine serum albumin for 1 h and then incubated with anti-α-SMA monoclonal antibody (1:500 sigma) and anti-ASPP2 polyclonal antibody(1:500 invitrogen) followed by Alexa Fluor 488- or 546-conjugated secondary antibody (Sigma-Aldrich). The sections were covered with Vectashield mounting medium containing 4,6-diamidino-2-phenylindole and observed under a confocal microscope (Leica TCS SP8, Germany). For negative controls, sections were processed as described earlier, except that incubation with the primary antibody was omitted.

LX-2 cells were fixed with 4% paraformaldehyde and permeabilized with 0.5% Triton X-100 for 15 min. Then, cells were blocked with 2% BSA for 1 h and incubated with anti-M30 monoclonal antibody (produced by our laboratory) and Alexa Fluor 546-conjugated secondary antibody. DAPI was used to stain the nuclei.

Immunohistochemical analysis with Masson stain was performed to detect hepatic collagen deposition.

Small interfering RNA (siRNA) specific to human ASPP2 (target sequence: 5-AAGTTGCTGAGCAGGAGAAAC-3) and a nonspecific control were synthesized by Genepharmer. LX-2 cells of 40–50% confluency were prepared for siRNA transfection. Transient transfection of siRNA (35 nM) was performed using FuGENE HD transfection reagent (Promega, Madison, WI, USA).

LX-2 cells cultured on coverslips were infected with Ad-tf-LC3 at 20 MOI. Twenty-four hours after adenovirus infection, the cells were washed with PBS, fixed with 4% paraformaldehyde, and viewed with a fluorescence microscope (Nikon Ti-E, Japan). The number of GFP and mRFP dots was determined by manual counting of fluorescent puncta in five fields from 30 different cells using a 40× objective. Each experiment was repeated three times. The number of dots per cell was calculated by dividing the total number of dots by the number of cells in each microscopic field.

ASPP2^+/− Balb/c mice were kindly provided by Dr. Lopez et al. [16]. ASPP2 wild-type (WT) Balb/c mice (Academy of Military Medical Sciences, China), and ASPP2+/+ Balb/c mice (overexpressing ASPP2) were generated from ASPP2 WT Balb/c mice by injecting ASPP2-ad (60,000 viral particles/mice) into the tail vein. Control mice were generated from ASPP2 WT by injecting Control-ad (60,000 viral particles/mice) into the tail vein. All mice received intraperitoneal injections of a carbon tetrachloride (CCl₄)/olive oil (OO) mixture (1:9 v/v) twice a week (1 μl per gram body weight). Mice were killed after 6 weeks of CCl₄ treatment, and tissues were harvested the day after the final administration of CCl₄. All studies were approved by the Ethics Committee of the Youan Hospital Center and the hospital affiliated with Capital Medical University, Beijing. All animal care complied with applicable health guidelines.

The results are expressed as the mean ± SEM. Statistical significance was determined using the Student’s t test. A p value < 0.05 was considered significant.

The activation of HSCs plays a pivotal role in liver fibrogenesis [1]. TGF-β1 is the classic fibrogenic cytokine involved in accelerating the progression of liver fibrosis [6]. Therefore, it was of interest to investigate the effects of ASPP2 on the TGF-β1-induced fibrogenic activation of LX-2 cells. First, LX-2 cells were pre-treated with ASPP2-adenovirus (AdASPP2) or GFP-adenovirus (Ad) for 24 h and then treated with or without TGF-β1 (10 ng/ml). Next, to investigate the role of ASPP2 in the fibrogenic activation of LX-2 cells, we examined the expression of fibrotic markers [α-SMA, Colα1 (I), and Colα1 (III)]. Quantitative analysis showed that the mRNA expression of α-SMA, Colα1 (I), and Colα1(III) was significantly upregulated in TGF-β1 and Ad-treated cells compared to cells treated with Ad alone (Fig. 1A). However, the enhanced mRNA expression of α-SMA, Colα1(I), and Colα1(III) mediated by TGF-β1 was blunted by pre-incubation with AdASPP2. Likewise, TGF-β1-mediated increases in α-SMA protein levels were inhibited by AdASPP2 pre-incubation in LX-2 cells (Fig. 1B, C). We also observed that AdASPP2 treatment did not affect the viability of LX-2 cells (data not shown). In addition, we found that ASPP2 overexpression in LX-2 cells not only affects TGF-β1-induced fibrogenic activation, but also attenuates basal fibrogenic activation. To test this issue, siRNA technology was used to knock down ASPP2 expression, as shown in Fig. 1D. Infection of LX-2 cells with ASPP2 siRNA markedly increased the expression of α-SMA induced by TGF-β1 treatment. Taken together, these results suggest that ASPP2 may reduce the TGF-β1-induced fibrogenic activation of LX-2 cells.

To further confirm these results, we analyzed the changes in ASPP2 gene and protein expression during the activation of primary cultured mouse HSCs. Culture-activated HSCs expressed 2.3 times as much ASPP2 mRNA as quiescent HSCs, and ASPP2 protein levels were also clearly increased in culture-activated HSCs. (Supplementary Figure 1).

Recent studies have shown that TGF-β1-induced autophagy is required for the fibrogenic response in human atrial myofibroblasts (hATMyofbs) [17] and that ASPP2 can inhibit autophagy through a p53-independent pathway [18]. Therefore, we speculated whether ASPP2 could decrease TGF-β1-induced profibrotic effects in LX-2 cells by inhibiting autophagy. Western blot analysis showed that α-SMA and LC3-II protein levels were markedly elevated in Ad and TGF-β1-treated cells, and this upregulation was abolished by AdASPP2 treatment (Fig. 2A). The graphical representation of this Western blotting data, including statistical analysis, is shown in Fig. 2B. To determine whether the ASPP2-mediated decrease of fibrogenic activation in LX-2 cells is associated with ASPP2-mediated inhibition of autophagy, we co-treated LX-2 cells with 3-methyladenine (3MA, an inhibitor of autophagy) and TGF-β1 for 24 h, and α-SMA and LC3-II expression was examined. The results were similar to those observed when cells were treated with both AdASPP2 and TGF-β1, as shown in Fig. 2A, B. These results support the hypothesis that ASPP2 decreases TGF-β1-induced profibrotic effects in LX-2 cells through inhibiting autophagy. To further confirm this hypothesis, we detected autophagic flux by using adenovirus harboring tf-LC3 (Ad-tf-LC3) [19, 24]. Next, the number of autophagosomes can be evaluated by determining the number of red dots that overlay green dots and appear yellow in merged images. The red dots that do not overlay green dots and appear red in merged images indicate autolysosome formation. As shown in Fig. 2C, D, the numbers of green and red dots per cell were both significantly decreased after treatment with AdASPP2 and TGF-β1 compared to Ad and TGF-β1 treatment. These results suggested that ASPP2 may reduce the fibrogenic activation of LX-2 cells by inhibiting TGF-β1-induced autophagy.

It has been shown that ASPP2 may induce tumor cell apoptosis [9]. Selective stimulation of apoptosis in stellate cells (rather than hepatocytes) may have anti-fibrotic effects [20‐23]. Thus, we questioned whether ASPP2 might induce apoptosis in HSCs and reduce liver fibrosis. First, we detected apoptosis by performing M30 staining, and we observed that the number of M30-positive cells was markedly increased in cells treated with AdASPP2 and TGF-β1 compared with Ad and TGF-β1-treated cells (Fig. 3A, B). Quantitative analysis of the expression of apoptosis-related genes Bax and Bcl-2 was performed by qRT-PCR. There was an approximately 2.5-fold increase in Bax mRNA (Fig. 3C) and a threefold decrease in Bcl-2 mRNA (Fig. 3D) in cells treated with AdASPP2 and TGF-β1 compared with Ad and TGF-β1-treated cells. To confirm this finding, we further analyzed apoptosis levels using immunoblotting. As shown in Fig. 3E, cleaved caspase3 fragment levels were markedly increased after ASPP2 overexpression. These data support the idea that ASPP2 may inhibit the anti-apoptotic effects of TGF-β1 in LX-2 cells.

Finally, we analyzed the effect of ASPP2 on mouse liver fibrosis induced by CCl₄. We used mice with low ASPP2 expression (ASPP2^+/−) [28] and mice overexpressing ASPP2 (ASPP2+/+), which were obtained by injecting AdASPP2 into the tail vein of ASPP2 WT mice. ASPP2 mRNA levels were significantly lower in ASPP2^+/− mice compared with ASPP2 WT mice after 6 weeks of CCl₄ treatment; meanwhile, ASPP2 mRNA levels were significantly higher in ASPP2+/+ mice compared with Ad-treated mice (Fig. 4A). Likewise, Western blot analysis revealed a similar change in ASPP2 protein levels (Fig. 4B). In addition, we found that ASPP2 is predominantly expressed in hepatocytes and HSCs during CCl₄-induced liver fibrosis (Fig. 4C). Next, we evaluated the effect of ASPP2 on hepatic collagen deposition with morphometric analysis of Masson staining and quantified any changes with digital image analysis. As shown in Fig. 4D, E, collagen deposition was significantly increased in ASPP2^+/− mice compared with ASPP2 WT mice after 6 weeks of CCl₄ administration (Fig. 4D-a, b, E). In contrast, collagen deposition was markedly decreased in ASPP2+/+ mouse liver tissues compared with control mouse liver tissues (Fig. 4D-c, d, E). We also examined the effect of ASPP2 on the expression of fibrosis markers [α-SMA, Colα1 (I) and Colα1 (III)] in mouse liver tissues. As shown in Fig. 4E, the levels of α-SMA, Colα1 (I), and Colα1 (III) mRNA were significantly increased in liver tissues of ASPP2^+/− mice compared with WT mice. However, compared with control mice, ASPP2 overexpression (ASPP2+/+) resulted in a marked reduction in α-SMA, Colα1 (I), and Colα1 (III) mRNA levels.

To identify the effects of ASPP2 on autophagy in liver tissue of CCL₄-treated mice, we performed an immunofluorescence assay to detect LC3 specks in fibrotic mouse liver tissue. LC3 specks were detected in most of the cells from ASPP2^+/− mice, but few were detected in ASPP2+/+ mice (Supplementary Figure 2). Finally, analysis of biochemical parameters (alanine aminotransferase, ALT) demonstrated that ASPP2 affected the extent of inflammation and necrosis in the liver. These results indicate that ASPP2 attenuates CCl₄-induced mouse liver fibrosis.