Bioactive compound reveals a novel function for ribosomal protein S5 in hepatic stellate cell activation and hepatic fibrosis


  • Supported by the National Natural Science Foundation of China (No. 91013014, 81170404, 81270508), National Major Special Science and Technology Project (No. 2012ZX09103101-043) and Shanghai Municipal Science and Technology Commission Innovation Action Plan of Biomedical Science and Technology Project (No. 10431900200).

  • Potential conflict of interest: Nothing to report.


Liver fibrosis and its endstage, cirrhosis, represent a major public health problem worldwide. Activation of hepatic stellate cells (HSCs) is a central event in hepatic fibrosis. However, the proteins that control HSC activation are incompletely understood. Here we show that (6aS, 10S, 11aR, 11bR, 11cS)-10-methylamino-dodecahydro-3a, 7a-diaza-benzo [de]anthracene-8-thione (MASM) exhibits potent inhibitory activity against liver fibrosis in vitro and in vivo associated with the reduction of Akt phosphorylation. Furthermore, ribosomal protein S5 (RPS5) was identified as a direct target of MASM, which stabilized RPS5 in cultured HSCs and in the liver of experimental animals after dimethylnitrosamine (DMN) or bile duct ligation (BDL). Functional studies revealed that RPS5 could prevent HSC activation. RPS5 overexpression in HSCs resulted in Akt dephosphorylation at both Ser473 and Thr308, and led to subsequent dephosphorylation of GSK3β or P70S6K. Progression of DMN- and BDL-induced hepatic fibrosis was aggravated by Rps5 knockdown and alleviated by RPS5 overexpression, which correlated with the modulation of Akt phosphorylation and HSC number in the fibrotic livers. Moreover, RPS5 was substantially reduced in the transdifferentiated HSCs, experimental fibrotic livers, and human cirrhosis samples. Conclusion: These results demonstrate that RPS5 is implicated in hepatic fibrogenesis and may represent a promising target for potential therapeutic intervention in liver fibrotic diseases. (Hepatology 2014;60:648–660)


α-smooth muscle actin


the adenovirus expressing RPS5 shRNA


bile duct ligation




extracellular matrix


glial fibrillary acidic protein


green fluorescent protein


glycogen synthase kinase 3β


hepatic stellate cells


mitogen-activated protein kinase


(6aS, 10S, 11aR, 11bR, 11cS)−10-methylamino-dodecahydro-3a, 7a-diaza-benzo [de]anthracene-8-thione


negative control


Nuclear factor kappa B


platelet-derived growth factor


phosphoinositide-dependent kinase 1


phosphoinositide 3-kinase


p70S6 kinase


small interfering RNA


small mothers against decapentaplegic homolog


ribosomal protein S5


transforming growth factor-β1

Liver fibrosis, a common consequence of different chronic liver injuries, is characterized by excess deposition of extracellular matrix (ECM).[1] Activation of hepatic stellate cell (HSC) is generally accepted as a critical event in hepatic fibrosis and considered an attractive target for treatment.[2, 3] HSC is activated upon liver injury, with changes in gene expression and major phenotypical transformation to α-smooth muscle actin (α-SMA)-positive myofibroblast that increases cell proliferation and produces large amounts of ECM including collagen I.[4, 5] HSC activation is associated with stimulation of several intracellular signaling cascades. For example, platelet-derived growth factor (PDGF), a potent proliferative cytokine for HSCs, stimulates mitogen-activated protein kinase / extracellular signal-regulated kinase (MAPK/ERK) and PI3K/Akt/P70S6K signaling in the HSCs. Transforming growth factor beta (TGFβ), a highly potent fibrogenic cytokine, simulates both small mothers against decapentaplegic homolog (Smad) and MAPK signaling in HSCs, stimulating collagen I gene expression. Other signaling pathways, including that which leads to nuclear factor kappa B (NFκB) activation, are induced during HSC activation.[3, 5] However, these events are probably not the primary events that trigger the activation of HSC.

Chemical genetics approaches are useful in identifying specific target proteins of bioactive chemicals. A wide range of different technologies including affinity chromatography has been successfully employed to identify the specific targets of many bioactive chemicals.[6] The identification of the molecular targets is important for elucidating the biological mechanisms of diseases and will greatly aid disease-related new drug development.

Matrine (Fig. 1A) is one of the major alkaloids found in traditional Chinese medicine Sophra leguminosae and possesses significant antiinflammatory and antifibrotic properties.[7-11] MASM [(6aS, 10S, 11aR, 11bR, 11cS)−10-Methylamino-dodecahydro-3a, 7a-diaza-benzo (de)anthracene-8-thione] (Fig. 1A) is one of the recently synthesized derivatives of matrine and has potent antiinflammatory activity.[12] In this study we found that MASM exhibited a powerful inhibitory effect on HSC activation and selectively targeted the Akt signaling pathway. Moreover, MASM bound specifically to ribosomal protein S5 (RPS5) and stabilized RPS5 in HSCs and in the fibrotic livers. An investigation of RPS5 revealed an unexpected function in the inhibition of HSC activation and hepatic fibrosis associated with the reduction of Akt phosphorylation. Interestingly, RPS5 was substantially reduced in the transdifferentiated HSCs, experimental fibrotic livers, and human cirrhosis samples.

Figure 1.

MASM inhibits HSC activation. (A) Chemical structures of matrine and MASM. (B,C) Expression of acta2 and col1a1 mRNA in HSCs (B) and HSC-T6 cells (C) treated with MASM in the presence or absence of TGFβ1 (2 ng/mL) for 24 hours. mRNA was normalized against β-actin and corrected to control. N = 3. *P < 0.05 versus the control. (D) The morphological change and α-SMA expression in HSCs treated with MASM (5 μM) and immunostained with α-SMA antibody at days 5, 8, and 12 after isolation. Scale bars = 25 μm. (E) Effect of MASM on hepatic stellate cell and hepatocyte proliferation. HSCs, activated HSCs (day 9), HSC-T6 cells, and rat primary hepatocytes were cultured for 72 hours in the presence of MASM. (F) Apoptosis of HSCs treated with MASM for 24 hours.

Materials and Methods

See the detailed Supporting Materials and Methods.

Recombinant Adenoviruses

The recombinant adenoviruses of AdRPS5 and AdGFP (control adenovirus) and of AdshRPS5 and AdshNC (control adenovirus) were obtained from Cyagen Biosciences (Guangzhou, China). Rps5 small interfering RNA (siRNA) sequences are in Supporting Table S1.

Cell Isolation and Treatment

Primary HSCs were prepared from male Sprague-Dawley rats and immortalized rat HSC-T6 lines were used as previously described.[13] Purity of HSCs was 90-95% as assessed by glial fibrillary acidic protein (GFAP) immunostaining (Fig. S1). Unless otherwise indicated, experiments described in this study were performed on HSCs undergoing transdifferentiation at day 3 after isolation. HSCs were infected by the adenoviral vectors at a multiplicity of infection (MOI) of 50 for 48 hours.

Cell Proliferation Assay, Apoptosis, Quantitative Real-time polymerase chain reaction (PCR) Analysis, Western Blot, and Immunofluorescence

Cell proliferation assay, quantitative real-time PCR analysis, western blot, and immunofluorescence were performed as described previously.[13-15] Antibodies used are in supporting materials. Primer sequences of the detected genes are in Supporting Table S2.

Affinity Chromatography

Serial affinity chromatography was performed as described.[16] For competition experiments, HSC-T6 cells were treated with MASM-biotin with or without MASM for 2 hours. Streptavidin agarose affinity assay was then performed.

Experimental Animals and Treatment

Male Sprague-Dawley rats, weighing about 200 g each, were purchased from Shanghai SLAC Laboratory Animal (Shanghai, China). Hepatic fibrosis was induced by dimethylnitrosamine (DMN) injection or bile duct ligation (BDL) as described previously.[13, 14] Adenoviruses (4 × 109 pfu) were injected by way of the tail vein. MASM was given orally by gavage. All procedures involving animals and their care in this study were approved by the Animal Care Committee of our institution in accordance with institutional guidelines for animal experiments.


Liver tissues were obtained from the liver tissue bank of the Eastern Hepatobiliary Surgery Hospital of our university. The normal liver samples (n = 7) were from individuals undergoing hepatic resection for angeioma or liparomphalus, and fibrotic samples (n = 12) were from patients undergoing liver transplantation and hepatic resection for hepatocellular carcinoma with clearly established diagnosis of hepatic cirrhosis. Informed consent was obtained from all subjects. The study protocol was approved by the Scientific Investigation Board of Second Military Medical University.

Measurement of Hydroxyproline Content, Histological Examination, and Immunohistological Staining

Measurement of hepatic hydroxyproline content, hematoxylin-eosin (H&E) staining for histopathological examination, Sirius red, or Masson's trichrome staining for collagen determination and immunohistochemical staining was performed as previously described.[13, 14] The area of collagen and immunohistochemical positive and negative cells were determined using a computer-assisted automated image analyzer (Qwin Leica) as described.[17]

Statistical Analysis

The results are expressed as the mean with the SD and were analyzed by analysis of variance (ANOVA) followed by paired comparison, as appropriate. P < 0.05 was taken as the minimum level of significance.


MASM Inhibits HSC Activation

MASM substantially attenuated the expression of acta2 (encoding α-SMA) and col1a1 (encoding α1 (I) collagen) in cultured and TGFβ1-activated HSCs as well as TGFβ1-stimulated HSC-T6 myofibroblasts (Fig. 1B,C). The striking morphological alteration and α-SMA expression were inhibited by MASM as HSCs underwent in vitro activation (Fig. 1D). MASM treatment dose-dependently inhibited the proliferation of HSCs, activated HSCs, and HSC-T6 cells but not primary rat hepatocytes (Fig. 1E). The retardation in proliferation was not attributable to an increase in cell apoptosis (Fig. 1F).

MASM Reduces Akt/mTOR/p70S6K Phosphorylation in HSCs

MASM suppressed the phosphorylation of Akt (Thr308), glycogen synthase kinase 3β (GSK3β), and P70S6K in cultured activated HSCs and TGFβ1-activated HSCs (Fig. 2A,B). However, MASM rendered no evident influence on TGFβ1-induced phosphorylation of ERK and Smad2 (Fig. 2B). The inhibition of the Akt signal pathway by MASM was also observed in TGFβ1-stimulated HSC-T6 cells (Fig. 2C). Immunofluorescence staining revealed that MASM reduced the levels of Akt phosphorylation in nucleus and cytoplasm (Fig. 2D).

Figure 2.

MASM inhibits the Akt pathway. (A) Western blot analysis of HSCs treated with MASM for 24 hours with the indicated antibodies. (B) HSCs were pretreated with MASM for 5 hours, followed by exposing to 2 ng/mL of TGFβ1 for 0.5, 1, and 4 hours to detect Smad2, Erk, and Akt phosphorylation, respectively. (C,D) Western blot analysis and immunostaining of HSC-T6 cells treated with MASM and TGFβ1. Cells were fixed and stained with phospho-Akt T308 (red) and the nuclear dye DAPI (blue) (D). Scale bar = 25 μm. The band intensities were quantified. GAPDH or β-actin was used as control and results were normalized to untreated control. N = 3. *P < 0.05 versus untreated control or TGFβ only, #P < 0.05 versus untreated control.

MASM Ameliorates Liver Fibrosis Induced by DMN or BDL in Rats

MASM treatment dramatically ameliorated DMN or BDL-induced hepatic fibrogenesis as shown by H&E, Sirius red, or Masson's trichrome staining (Fig. 3A,B). Immunohistochemical staining demonstrated that MASM reduced α-SMA protein expression and Akt phosphorylation (Fig. 3A). Computer-assisted morphometric analysis of Sirius red staining revealed that MASM markedly reduced the ECM area (Fig. 3C). Biochemical determinations of hydroxyproline demonstrated to be significantly lower in MASM-treated rat livers (Fig. 3D). In addition, MASM treatment markedly decreased expression of acta2 mRNA in the fibrotic livers (Fig. 3E).

Figure 3.

MASM ameliorates experimental hepatic fibrosis. (A) Representative photographs of 10 animals are shown where extracellular matrix was confirmed by H&E, Sirius red staining, and Masson's trichrome staining (Masson) in liver sections. Immunohistochemical staining was carried out to detect the expression of α-SMA, pAkt S473 and Rps5. Scale bars = 25 μm. (B) Schematic representation of induction of hepatic fibrosis by DMN or BDL in rats. Rats were given for the last 21 days 25 mg/kg MASM orally by gavage. Control, normal, and sham control animals received only saline. (C) Quantitative data of Sirius red area was determined by computer-assisted image analyses. (D) Collagen content was measured by hydroxyproline (HYP) biochemical determinations. (E) Hepatic acta2 mRNA was determined by real-time PCR and normalized to β-actin. N = 10 rats per group. *P < 0.05 versus the control (Con).

Identification of Rps5 as a Cellular Target for MASM

To identify the cellular target of MASM, MASM-Resin (MR), control resin (CR), and MASM-biotin (MB) were synthesized (Fig. 4A; Supporting Methods). In serial affinity chromatography, a protein with an apparent molecular mass of 23 kDa was found to bind to first MR (MR1) and almost none to either second MR (MR2) or CR (Fig. 4B). Mass spectrometry analysis identified this protein as a ribosomal protein S5 (Rps5) (Fig. 4C). This result was independently confirmed by western blotting with anti-RPS5 antibody (Fig. 4B). MB, a tagged MASM derivative, still retained the biological activity of the parent (Fig. 4D) and was colocalized to Rps5 (Fig. 4E). Moreover, streptavidin agarose affinity assay revealed that MB bound to RPS5 in HSC-T6 cells, and this binding could be eliminated by addition of MASM (Fig. 4F).

Figure 4.

Identification of RPS5 as a cellular target for MASM. (A) Chemical structures of MASM-Resin (MR), control resin (CR), and MASM-Biotin (MB). (B) Serial affinity chromatography for identifying MASM specific binding proteins. Arrow indicates the band containing RPS5 (upper panel). Western blot analysis of RPS5 on the MR-bound proteins (bottom panel). (C) The amino acid sequence of RPS5. The residues detected by mass spectrometry are shown in red. (D) Effects of MASM and MB on the proliferation and α-SMA expression of HSC-T6 cells. The levels of α-SMA were quantified by densitometry. GAPDH was used as control and the results were normalized to untreated control. N = 3. *P < 0.05 versus TGFβ only. (E) Colocalization of RPS5 to MB in HSC-T6 cells stained with RPS5 (green), biotin (red), and DAPI (blue). (F) Binding of MB to RPS5 could be eliminated by MASM in a streptavidin agarose affinity assay. The band intensities of three experiments were quantified using the input as control and normalized to untreated control.

RPS5 Suppresses HSC Activation and Is Required for the Inhibitory Effect of MASM on HSC Activation

Rps5 knockdown markedly increased α-SMA and α1 (I) collagen expression at the levels of both mRNA and protein (Fig. 5A,B; Fig. S2A). In contrast, overexpressing RPS5 in cultured HSCs decreased the mRNA and protein levels of α-SMA and α1 (I) collagen (Fig. S2B,C). High levels of RPS5 expression in rat HSC-T6 myofibroblasts also suppressed the cell proliferation (Fig. S2D). In addition, the increased acta2 and col1a1 mRNA in rps5-silenced HSC-T6 cells was almost abolished by overexpression of RPS5 using an adenoviruses carrying cDNA for human RPS5 (Fig. 5C), suggesting a cause-effect relationship between rps5 knockdown and HSC activation. As expected as a component of the ribosomal 40S subunit, mature 28S rRNA and the total protein concentration were reduced in rps5-silenced HSC-T6 cells (Fig. S3).

Figure 5.

RPS5 regulates HSC activation. (A,B) RPS5 knockdown increased expression of α-SMA and collagen I in HSCs. HSCs were infected with AdshRPS5 and AdshNC. mRNA was normalized against β-actin and corrected to control. The band intensities were quantified. N = 3. *P < 0.05 versus the AdshNC control. (C) A cause-effect relationship between rps5 knockdown and HSC activation. HSC-T6 cells were infected with AdshRPS5 for 48 hours and then with either AdRPS5 or AdGFP for another 48 hours. mRNA was corrected to AdshNC. (D) RPS5 was required for the inhibitory effect of MASM on HSC activation. MASM (10 μM) inhibited α-SMA expression in HSCs transfected with AdshNC, but not HSCs transfected with AdshRPS5. (E) Knockdown of rps5 abolished the inhibitory effect of MASM on TGFβ1-driven acta2 and col1a1 mRNA expression. HSC-T6 cells were infected with AdshNC or AdshRPS5 for 24 hours and then treated with MASM (10 μM) and TGFβ1 (2 ng/mL) for another 24 hours. mRNA was normalized against β-actin and corrected to AdshNC control. N = 3. *P < 0.05 versus the AdshNC control. (F) The levels of RPS5 in HSCs treated with MASM for 24 hours. Cells were visualized by immunofluorescence microscopy. Scale bars = 25 μm.

MASM inhibited α-SMA expression in HSCs transfected with AdshNC, but not HSCs transfected with AdshRPS5 (Fig. 5D). Similarly, an inhibitory effect of MASM on TGFβ1-driven acta2 and col1a1 mRNA expression was lost in HSC-T6 cells transfected with AdshRPS5 (Fig. 5E). These data support RPS5 as a cellular target of MASM.

Effect of MASM on RPS5 Levels in HSCs and Fibrotic Livers

To further support RPS5 as a target of MASM, we investigated the effect of MASM on the amount of Rps5 in vitro and in vivo. Immunofluorescence study revealed higher fluorescent Rps5 signals in cultured HSCs treated for 24 hours with varying concentrations of MASM (Fig. 5F). However, treatment with MASM did not affect rps5 mRNA levels (data not shown). Accordingly, MASM administration caused an increase in the amount of Rps5 protein in the livers of experimental animals after DMN or BDL (Fig. 3A).

RPS5 Reduces Akt Phosphorylation

Activation of the Akt signaling pathway promotes the proliferation and collagen expression of HSCs.[18-20] In our experiments, overexpression of RPS5 in HSCs resulted in a dramatic reduction of phosphorylation of Akt at Thr308 and Ser473 (Fig. 6A), GSK3β, and P70S6K (Fig. 6B). This effect was also observed in RPS5 overexpressing HSC-T6 cells upon TGFβ1 stimulation (Fig. 6C). Rps5 knockdown, in contrast, increased Akt phosphorylation at Thr308 and Ser473 under both basal condition (10% fetal bovine serum [FBS] Dulbecco's modified Eagle's medium [DMEM]) and serum-free conditions, and increased a subset of potential Akt substrates (Fig. 6D,E).

Figure 6.

RPS5 regulates Akt phosphorylation. (A,B) RPS5 overexpression reduces Akt, GSK3β, and P70S6K phosphorylation in HSCs. The band intensities were quantified. N = 3. *P < 0.05 versus the AdGFP control. (C) HSCs were infected with AdGFP or AdRPS5 for 36 hours. Cells were then serum-starved for 16 hours and stimulated with TGF β1 (2 ng/mL) for 4 hours. N = 3. #P < 0.05 versus untreated control, *P < 0.05 versus AdGFP + TGFβ1. (D) HSC-T6 cells were infected with AdshNC or AdshRPS5 for 24 hours and then cultured in the medium with or without 10% FBS for another 24 hours. N = 3. *P < 0.05 versus the AdshNC control, #P < 0.05 versus AdshNC + serum. (E) Increased Akt phospho-substrates by rps5 knockdown in HSC-T6 cells by immunoblotting with specific antibodies against pan-Akt phospho-substrate. (F) HSC-T6 cells were infected with AdshNC or AdshRPS5 in the presence or absence of MK-2206 (0.5 μM) for 48 hours. N = 3. *P < 0.05.

Akt phosphorylation is the result of an equilibrium between PI3K/PDK1-mediated phosphorylation and serine/threonine protein phosphatases-mediated dephosphorylation.[21] The results showed that Akt phosphorylation at Thr308 and Ser473 was slightly reduced by the PI3K inhibitor LY294002, but not by the PDK1 inhibitor BX-912 in rps5-silenced cells (Fig. S4A). Rps5 knockdown also led to a remarkable increase in the PDGF-stimulated phosphorylation of Akt on Thr308 and Ser473. In the presence of LY294002 that blocks rephosphorylation of Akt by inhibiting PI3K, this increase remained higher and lasted longer relative to that of control (Fig. S4B), suggesting a loss of dephosphorylation in rps5-silenced cells.

MK-2206, an allosteric Akt inhibitor,[22] diminished the elevated levels of Akt, GSK3β, and P70S6K phosphorylation as well as α-SMA expression in rps5-silenced cells (Fig. 6F; Fig. S4C), suggesting that inhibition of HSC is a result of Rps5 to suppress Akt activity.

Experimental Hepatic Fibrosis Is Aggravated by rps5 Knockdown, and Alleviated by RPS5 Overexpression

Progression of hepatic fibrosis induced by DMN or BDL in rats was evaluated by adenovirus-mediated knockdown of rps5 (Fig. 7A,B), as reflected as inhibited expression of Rps5 in the livers by immunohistochemistry (Fig. 7B) and by RT-PCR analysis (Fig. 7C). AdshRPS5 injection greatly aggravated experimental hepatic fibrosis as shown by Sirius red staining and Masson's trichrome staining (Fig. 7B). The area of fibrosis was increased by 170% and 157% in DMN- and BDL-treated rats, respectively (Fig. 7C). AdshRPS5 also led to an increase in hepatic hydroxyproline content, α-SMA protein, and the phosphorylated Akt level (Fig. 7B,C).

Figure 7.

Experimentally induced hepatic fibrosis is aggravated by rps5 knockdown and attenuated by RPS5 overexpression. (A) Schematic representation of induction of experimental hepatic fibrosis by DMN or BDL in rats. AdshNC or AdshRPS5 were injected by way of the tail vein 2 days before DMN injection and 2 weeks after DMN injection or 2 days before BDL. AdGFP or AdRPS5 were injected 2 weeks after DMN injection. (B,C) RNA interference-mediated knockdown of rps5 expression aggravated hepatic fibrosis. (D,E) High levels of RPS5 expression attenuated DMN-induced hepatic fibrosis. Others were determined as in Fig. 3. Scale bars = 200 μm.

In contrast, increased RPS5 expression with a single injection of AdRPS5 by way of the tail vein after a 2-week treatment with DMN (Fig. 7A) increased RPS5 expression in the liver (Fig. 7D) and reduced ECM deposition, hydroxyproline content, and α-SMA protein as well as phosphorylated Akt level in the fibrotic livers (Fig. 7D,E).

Rps5 Expression Is Reduced in Transdifferentiated HSCs and Hepatic Myofibroblasts During Hepatic Fibrogenesis With Hyperphosphorylation of Akt

HSC culture on uncoated plastic was used as an experimental model for HSC activation. During the differentiation process of HSCs (from a quiescent state on day 1 to activated on day 5), the transcription of rps5 gene was gradually repressed (Fig. 8A). Rps5 protein also decreased as the cells underwent in vitro activation. Moreover, the increase of Akt phosphorylation at Ser473 was temporally consistent with the decrease of Rps5 protein (Fig. 8B).

Figure 8.

RPS5 was reduced in activated HSCs and hepatic myofibroblasts during hepatic fibrogenesis with hyperphosphorylation of Akt. (A) rps5 mRNA expression in cultured HSCs after isolation. mRNA was normalized against β-actin. N = 3. (B) Western blot analysis of Rps5, pAkt S473, and α-SMA expression at varying culture times. (C) Representative photographs of 10 animals are shown where RPS5 was decreased but pAkt S473 were increased in fibrotic livers. (D) Representative double immunostaining for Rps5 (green) and α-SMA (brown) in BDL-induced fibrotic liver (upper panel), and for pAkt S473 (brown) and α-SMA (green) in DMN-induced fibrotic liver (bottom panel). (E) Representative immunohistochemical staining for RPS5 in liver tissue sections obtained from control versus cirrhotic liver (upper panel), and representative double immunostaining for RPS5 (brown) and α-SMA (green) in cirrhotic liver (bottom panel). Scale bars = 200 μm.

Rps5 mRNA was significantly reduced in fibrotic livers, as assessed by H&E, Sirius red, or Masson's trichrome staining and α-SMA immunohistochemistry staining, of rats following DMN or BDL treatment (Fig. 8C; Fig. S5A,B). Immunohistochemistry showed faint Rps5 staining in the fibrotic livers in contrast to intense staining in normal and sham control. Along with decreased Rps5 expression, phospho-Akt Ser473 staining was increased (Fig. 8C). Double immunohistochemistry with Rps5 and α-SMA staining revealed substantially reduced Rps5 expression in myofibroblasts within fibrotic septa and moderate reduction in hepatocytes located at the scar-parenchyma interface (Fig. 8D, upper panels). Interestingly, increased phospho-Akt Ser473 staining was only observed in α-SMA-positive myofibroblasts but not in hepatocytes (Fig. 8D, bottom panels). In HSCs isolated from rats at 1 and 3 weeks after DMN treatment, rps5 mRNA expression was also dramatically decreased (Fig. S5C).

Fibrotic/cirrhotic human liver displayed faint RPS5 expression, in contrast to intense staining in control human liver (Fig. 8E, upper panel). RPS5 was colocalized to α-SMA (Fig. 8E, bottom panel).


The present study demonstrates that MASM selectively inhibited HSC activation associated with the reduction of Akt/mTOR/p70S6K phosphorylation. Moreover, MASM administration markedly suppressed the fibrotic pathologic changes and the elevated hydroxyproline content in parallel with the reduction of HSC number and Akt phosphorylation in liver induced by DMN and BDL. The synthesis of tagged MASM derivative retaining the biological activity of the parent allowed us to identify RPS5 as a binding protein for MASM. Our finding that rps5 knockdown noticeably diminished the inhibitory effect of MASM on HSC activation indicates RPS5 is a molecular target of MASM. Furthermore, MASM treatment stabilized Rps5 levels in cultured HSCs and in the livers of experimental animals after DMN or BDL. The results suggest that MASM prevents hepatic fibrosis, at least in part, through binding to and stabilizing RPS5.

A growing body of evidence suggests that ribosomal proteins have extraribosomal functions.[23, 24] For example, RPS5 could interact with the hepatitis C virus[25] and cricket paralysis virus internal ribosome entry site.[26, 27] It is phosphorylated by casein kinase II when not assembled into ribosome and the phosphorylation seems to regulate protein trafficking from cytoplasm to nucleoli.[28] Interesting, RPS5 is tightly coupled with murine erythroleukemia cell differentiation.[29-31] Within the context of liver fibrosis, Rps5 has been reported to be differentially expressed between quiescent and activated HSCs.[32] The loss-of-function or gain-of-function experiments in the current study clearly demonstrated that RPS5 could suppress HSC activation. Our in vivo findings showed that liver fibrosis could be aggravated by rps5 knockdown, whereas high levels of RPS5 expression attenuated hepatic fibrosis with decreasing accumulation of HSC in the liver. In addition, Rps5 was highly reduced in the transdifferentiation of cultured HSCs and hepatic myofibroblasts in both toxic and biliary models of liver fibrosis as well as in hepatic myofibroblasts of patients with fibrosis or cirrhosis. We have to point out that we did not examine an HSC-specific promoter (such as GFAP or PDGFR) driving an RPS5 knockdown in vivo to show how specific RPS5 is for HSCs and whether MASM would show no further benefit in such a knockdown in vivo. Notwithstanding this limitation, this study does suggest that RPS5 is implicated in HSC activation and hepatic fibrogenesis and may represent a promising target for potential therapeutic intervention in liver fibrotic diseases. Interestingly, we failed to show a significant effect of RPS5 on hepatocellular injury (data not shown), suggesting that fibrogenesis inhibition is not secondary to reduced injury. Lastly, RPS5 expression in hepatocytes was also reduced in patients with fibrosis or cirrhosis. Whether changes of RPS5 in hepatocytes are also connected to liver fibrosis requires further study.

The Akt signaling pathway has been shown to regulate HSC activation and the progress of hepatic fibrosis.[18-20, 33] Our finding that the intense staining of phosphorylated Akt was paralleled with decreased Rps5 expression in myofibroblasts but not in hepatocytes within fibrotic septa is consistent with the notion that hepatic Akt hyperphosphorylation is associated with advanced liver fibrosis.[34] Functional experiments in the current study revealed that Akt phosphorylation was decreased by expression of exogenous RPS5 gene in HSCs, and vice versa. Moreover, rps5 knockdown enhanced phosphorylation levels of a subset of potential Akt substrates, whereas RPS5 overexpression reduced phosphorylation of GSK3β and P70S6K, two critical downstream signaling molecules in the Akt pathway. RPS5 was also involved in Akt dephosphorylation in HSCs stimulated with two important fibrogenic cytokines, TGFβ1 and PDGF. In addition, RPS5 suppressed progress of experimental hepatic fibrosis accompanied with a reduction of Akt phosphorylation. These results suggest that the effects of RPS5 on HSC activation and hepatic fibrosis is related to the ability of RPS5 to regulate Akt phosphorylation. Our findings could also well explain the inhibitory effects of MASM on the Akt pathway due to MASM stabilizing RPS5. Details of how this protein RPS5 regulates Akt phosphorylation remain to be elucidated, although we found that PDK1 inhibition failed to reduce Akt phosphorylation in rps5-silenced cells.

Taken together, the present study demonstrates that MASM targets and stabilizes RPS5, which leads to suppressed HSC activation and attenuated hepatic fibrosis associated with the reduction of Akt/mTOR/p70S6K phosphorylation. We have also revealed here for the first time that RPS5 plays a critical role in HSC activation and hepatic fibrosis. Our findings point to RPS5 as a previously uncharacterized druggable target for antihepatic fibrosis therapy.