Knowledge-Based Discovery of Anti-Fibrotic and Pro-Fibrotic Activities from Chinese Materia Medica

Fibrosis, also known as scarring, sclerosis or cirrhosis, is characterised by excessive accumulation of extracellular matrix (ECM) proteins leading to tissue contraction, disruption of tissue architecture and eventually chronic organ failure (Wynn, 2007; Xu et al., 2007). Research and development of anti-fibrotic drugs are generally based on two distinct but interactive strategies, with one based on mechanism studies and another based on exploring efficacy. In principle, the mechanism-based strategy begins with identification of molecular targets through mechanistic studies, and then development of inhibitors or enhancers targeting the molecules. On the other hand, efficacy-based strategy starts with screening drug candidates in disease models to identify activities and efficacy, with less reliance on analysis of mechanisms of action. There are certain limitations in both the mechanism-based strategy and the efficacy-based strategy, which largely account for the lack of success in development of anti-fibrotic drugs. The former is often associated with identification of multiple molecular targets impeding development of a single drug that tackles multiple targets, while the latter is often hampered by establishment of apt models ideal for efficacy-driven drug screens. Efficacy-based strategy has been employed in development of both traditional and modern medicines. In the context of traditional medicine, the knowledge about efficacy of a given drug is largely derived from a trial-and-error process, namely by assessing patients’ response upon treatment with natural drug candidates. However, in modern medicine, it is impossible to directly test any new drugs in patients. Solid scientific evidence on efficacy and safety of a given drug in experimental models is required prior to clinical trials. Understandably, quality of these models would determine the specificity and efficiency of the tested drug.


Introduction
Fibrosis, also known as scarring, sclerosis or cirrhosis, is characterised by excessive accumulation of extracellular matrix (ECM) proteins leading to tissue contraction, disruption of tissue architecture and eventually chronic organ failure (Wynn, 2007;Xu et al., 2007).Research and development of anti-fibrotic drugs are generally based on two distinct but interactive strategies, with one based on mechanism studies and another based on exploring efficacy.In principle, the mechanism-based strategy begins with identification of molecular targets through mechanistic studies, and then development of inhibitors or enhancers targeting the molecules.On the other hand, efficacy-based strategy starts with screening drug candidates in disease models to identify activities and efficacy, with less reliance on analysis of mechanisms of action.There are certain limitations in both the mechanism-based strategy and the efficacy-based strategy, which largely account for the lack of success in development of anti-fibrotic drugs.The former is often associated with identification of multiple molecular targets impeding development of a single drug that tackles multiple targets, while the latter is often hampered by establishment of apt models ideal for efficacy-driven drug screens.Efficacy-based strategy has been employed in development of both traditional and modern medicines.In the context of traditional medicine, the knowledge about efficacy of a given drug is largely derived from a trial-and-error process, namely by assessing patients' response upon treatment with natural drug candidates.However, in modern medicine, it is impossible to directly test any new drugs in patients.Solid scientific evidence on efficacy and safety of a given drug in experimental models is required prior to clinical trials.Understandably, quality of these models would determine the specificity and efficiency of the tested drug.
In vivo models are invaluable research tools but not without its limitation, especially in the development of drugs targeting fibrosis given the complex aetiologies and inflammatory processes involved in this pathological condition.In particular, drug leads displaying effectiveness in in vivo models of fibrosis may stem from (i) inhibition of the primary aetiological factors, (ii) inhibition of inflammation, a common inciting factor of fibrosis, (iii) inhibition of fibrosis per se, or (iv) a net effect of any combinations of the aforementioned.For the initial development of drugs with specific anti-fibrotic effects independent of antiinflammatory actions and any specific aetiological factors, in vitro models of fibrosis appear to be more appropriate.Our laboratory recently reported transforming growth factor-1 (TGF-1)-induced inflammation-free in vitro models of fibrosis in mesenchymal cells, featuring 2-dimentional (2D) ECM protein accumulation reflecting net collagen accumulation, and 3-dimentional (3D) nodular formation following ECM protein-induced disruption of cell monolayer, which can both be objectively quantified.By coating specific matrix on 96-well plates, which determines the 2D or 3D nature of the models, we are able to quantify readouts more reliably, permitting high-throughput screening of compounds with anti-fibrotic activities (Xu et al., 2007).In contrast to the lack of "anti-fibrotics" in Western medicine clinics, there is accumulating evidence suggesting anti-fibrotic effects of Chinese materia medicas (CMMs), i.e., medicinal materials used in traditional Chinese medicine (TCM) (Hu et al., 2007).However, most of these conclusions were drawn from animal models or patients and hence it was not clear whether the reported efficacy of those CMMs was secondary to inhibition of aetiological factors or inflammation, or whether they exerted genuine anti-fibrotic activities (Hu et al., 2009).We hypothesised that at least some of the "anti-fibrotic" herbal derivatives reported in the literature are indeed anti-fibrotic by antagonising TGF-1-specific pro-fibrotic pathways or common pathways of fibrosis.By employing the 2D in vitro model (Xu et al., 2007), we tested the anti-fibrotic activity of 21 CMM-derived compounds, 11 methanolic extracts of single CMMs and 27 formulae that contained two or more CMMs as mixtures, and found that five compounds, three single CMM extracts and 16 formulae had in vitro anti-fibrotic activities (Hu et al., 2009).Among the five CMM-derived compounds, three flavonoids (quercetin, baicalin and baicalein) showed similar dose-dependent in vitro antifibrotic activities while two non-flavonoids (salvianolic acid B and emodin) showed varied in vitro anti-fibrotic activities with poor dose dependency.Among the three CMM extracts showing significant in vitro anti-fibrotic activities, Huangqin (root of Scutellaria baicalensis Georgi) is rich in baicalin and baicalein, Danshen (root of Salvia miltiorrhiza Bunge) is rich in salvianolic acid B and Dahuang (root of Rheum palmatum L.) is rich in emodin.Among the 16 herbal formulae with in vitro anti-fibrotic effects, eight contained neither Huangqin, Danshen nor Dahuang, while the remaining eight contained at least one of the three CMMs (Hu et al., 2009).Following successful identification of in vitro anti-fibrotic activities in herbal entities, we have extended our work to focus on TCM knowledge-based discovery of novel anti-fibrotic drug leads from natural sources, especially CMMs.In TCM, fibrosis is diagnosed as a kind of "Jie Zheng" which means "lump or clot".Based on this concept, two senior TCM practitioners were invited to choose a collection of 27 CMMs, including 26 medicinal plant parts and one medicinal fungus, which they would consider using in patients with fibrotic diseases.Based on the traditional categories of CMMs, the 27 CMMs fall into three functional subgroups, namely "Huo Xue Hua Yu" ("promoting the circulation and resolving NRK-49F cells were sub-cultured before they became confluent to prevent transformation.The 2D model was employed for experiments in this project.Cells were seeded in collagen type I-coated 96-well plates (BD Biosciences, Oxford, UK) at a density of 110 4 cells per well in 200 µl DMEM supplemented with 2.5% FCS and 2.5% Nu-Serum TM V serum replacements (NU, BD Biosciences).After three days, the medium was changed to serum-free DMEM supplemented with 1% insulin-transferrin-selenium liquid media supplement (ITS, Sigma-Aldrich) for four days, and then changed to fresh ITS-supplemented serum-free medium containing 5 ng/ml TGF-1 in the presence of different concentrations of herbal extracts or a vehicle control (an equal volume of DMSO) for 48 h.In all the experiments, 1 µM IN-1130 was used as a positive control for anti-fibrotic activities.For initial screening, the concentrations of CMM extracts tested were 10, 20, 40, 80, 160 and 200 µg/ml, 3-6 wells per group, and the screening was repeated at least twice.For follow-up confirmation studies, four herbal extracts were selected based on the results from initial screening studies, and their effects were further confirmed using three different concentrations, 3-6 wells per group.Each follow-up experiment was performed four times.

Cell detachment index (CDI) and lactate dehydrogenase (LDH) release assay
For the screening studies, in vitro cytotoxicity at the end of 48 h treatment was assessed by phase-contrast microscopy using the same CDI criteria that we reported before (Hu et al., 2009).In brief, CDI reflects cell monolayer disruption, including cell detachment from the adherent surface and disorganisation of cell monolayer.Scores of 0, 0.5, 1, 1.5, 2, 2.5, 3, 3.5 and 4 represented an area of 0, 5%, 10%, 20%, 30%, 40%, 60%, 80% and 100%, respectively, of the total adherent surface not covered by cells.This method was used to approximate cytotoxicity, as well as to ensure the reliability of subsequent picro-Sirius red (PSR) staining results, which required minimum cell (and matrix) detachment and disruption of cell monolayer.For followup studies, LDH release assay was used to assess in vitro cytotoxicity at the end of 48 h treatment.Fifty microlitre of supernatant from each well was collected and tested for LDH release according to the manufacturer's instructions (Promega, Southampton, UK).LDH release was measured using a Dynex Technologies MRX spectrophotometer (Prior Laboratory Supplies Ltd., East Sussex, UK), at optical density (OD) value of 490 nm.

Microscopic examination
Microscopic examination was performed on a Nikon Eclipse TE2000-S microscope (Nikon Instruments Europe B.V., The Netherlands).Bright-field and phase-contrast images were captured with a DXM1200F Nikon digital camera (Nikon UK Limited, Surrey, UK) and processed with Adobe Photoshop (Adobe System Europe Ltd., London, UK).

PSR staining and spectrophotometric analysis
Total collagen accumulation was assessed qualitatively by microscopic examination of PSRstained cells and quantitatively by spectrophotometric analysis of PSR staining.After CDI determination and conditioned medium collected for LDH release assay, cell monolayer in 96well plates were fixed in ice-cold methanol (200 µl per well) overnight at -20 C.Cells were carefully washed twice (5 min each) with 1x phosphate buffered saline (PBS, 200 µl per well) and then stained with 0.1% w/v PSR solution (200 µl per well, Sigma-Aldrich) at room temperature for 46 h.The staining solution was then removed and excessive PSR stain was carefully washed off with 0.1% v/v acetic acid (200 µl per well, VWR International Ltd, Lutterworth, UK) three times (5 min each).The stained wells were left to air dried for 2448 h.PSR stain was then observed under bright-field microscopy and microscopic pictures were taken.Finally, PSR stain was eluted in 0.1 N NaOH (200 µl per well) on a rocking platform at room temperature for 1 h.The plate was then subjected to spectrophotometric analysis of OD at 540 nm on a Dynex Technologies MRX spectrophotometer (Hu et al., 2009).

PPAR, PPAR/ and PPAR agonistic activities of Shiliuhua extract and effect of Shiliuhua extract in the presence of PPAR antagonists
We commissioned Tebu-bio Laboratories, Le Perray en Yvelines, France, to perform a pilot test on Shiliuhua extract for its isotype-specific PPAR activation.In brief, the assay was carried out in vitro in three HeLa cell lines stably expressing a chimeric protein containing the yeast transactivator GAL4 DNA binding domain fused to ligand binding domain regions of human PPAR, PPAR/ or PPAR, and a luciferase reporter gene driven by a pentamer of the GAL4 recognition sequence in front of the β-globin promoter (Seimandi et al., 2005).The PPAR reporter cell lines were seeded in a 96-well plate in triplicates, and treated with 40 g/ml Shiliuhua extract, or an equal volume of DMSO as negative control, for 24 h.Luciferase activity was determined with a luminometer and relative light units (RLU) were recorded.Three independent experiments were performed and the fold changes of RLU were normalised to the mean of negative control.

Statistical analysis
Results of PSR and LDH OD values were expressed as mean  SEM unless stated otherwise.Statistical differences were computed with Prism 4.0 (GraphPad Software, San Diego, CA, USA), by one-way analysis of variance and Dunnett post test for comparison between a control group and all other groups; for PPAR reporter activity, one-tail paired t test was performed on log-transformed fold changes.p<0.05 was regarded as statistically significant.

Initial screening
The screening results of CMM extracts were summarised in Table 2. Anti-fibrotic and profibrotic effects were defined as PSR OD values significantly lower and higher than that of TGF-1-treated group, respectively; cytotoxicity was defined as CDI significantly higher than that of TGF-1-treated group.Doses at which extracts exhibited reproducible effects were indicated in the table.The CMMs were categorised into four groups, i.e., eight in Group A showed in vitro anti-fibrotic activities; three in Group B showed pro-fibrotic activities; six in Group C showed prominent cytotoxicity; 11 in Group D did not have profound effects on total collagen accumulation nor integrity of cell monolayer.In Group A, Liedang, Meiguiqie and Gusuibu extracts were well-tolerated agents with anti-fibrotic activities noted within the range of 80-200 µg/ml; extracts of Shiliuhua, Baibeiyegen, Jixueteng, Moyao and Lingzhi showed anti-fibrotic effects at concentrations ranging from 1080 µg/ml, beyond which cytotoxicity was noted.In Group B, Chuanwutou showed a pro-fibrotic effect at concentrations as low as 20 µg/ml whereas pro-fibrotic effect of Dangshen and Yimucao were observed only at higher concentrations (160200 µg/ml).

Follow-up studies
Based on the initial screening results, Shiliuhua (SLH), Liedang (LD), Meiguiqie (MGQ) and Chuanwutou (CWT) extracts were selected for follow-up studies in view of their minimum CDI changes indicating low cytotoxicity, and the results are shown in Fig. 1.In contrast to SLH, LD and MGQ extracts, which showed varying degrees of in vitro anti-fibrotic effects, CWT extract exhibited a marked pro-fibrotic effect.Representative effects of SLH, LD, MGQ and CWT extracts on PSR staining, relative PSR OD values and LDH release are shown in Fig. 2 and Fig. 3.

Activation of PPARs by SLH extract
As SLH extract is one of the CMM extracts with the most potent in vitro anti-fibrotic activity and it was previously reported to interfere with the PPAR signalling pathway (Li et al., 2008), we hypothesised that the anti-fibrotic effect of SLH extract was at least in part mediated by activation of one or more PPAR receptors.Indeed, SLH extract induced PPARand PPAR-mediated reporter activity; it also marginally activates PPAR/ but the induction of reporter activity was just above the threshold of significance (p=0.066) (Fig. 4a).In order to determine if the agonistic effect of SLH extract has an impact on TGF-1-induced fibrogenesis, cells were treated with SLH extract and individual PPAR antagonists.PPAR/ antagonist, GSK0660 (IC 50 160 nM) (Shearer et al., 2008), did not affect the anti-fibrotic activity of SLH extract at all concentrations tested (up to 6.25-fold higher than its IC 50 ); PPAR antagonist, GW6471 (IC 50 240 nM) (Xu et al., 2002), also did not show any significant effect at concentrations up to 4.2-fold of its IC 50 , but at 10 M (42-fold higher than its IC 50 ), it did moderately suppress the anti-fibrotic activity of SLH extract; PPAR antagonist, T0070907 (IC 50 1 nM in inhibiting rosiglitazone binding to PPAR and 3.2-24.3μM in inhibiting proliferation of different cancer cell lines) (Lee et al., 2002;Burton et al., 2007), did not show any significant effect at concentrations of 0.1, 1 and 10 μM, but further increased the anti-fibrotic effect of SLH extract at 25 μM (Fig. 4b).
Table 2. Initial screening results of CMMs in TGF-1-induced in vitro fibrogenesis.Minus (-) and plus (+) percentages represent relative reduction and increase in TGF-1-induced PSR OD values, respectively."-": Cytotoxicity, assessed from CDI, was not significantly different from TGF-1-treated group at all concentrations tested (10-200 µg/ml).Further to the evidence-based approach that we used in our earlier screens (Hu et al., 2009), in which we selected CMMs that had been reported to reduce fibrotic lesions in vivo, we used a knowledge-based approach in this study, namely, to screen candidate CMMs that were believed to be beneficial in treating fibrotic diseases based on the theory and practice of TCM.Of the 28 CMM extracts examined, eight (28.5%)showed reproducible in vitro antifibrotic activities, i.e., Baibeiyegen, Liedang, Gusuibu, Jixueteng, Lingzhi, Meiguiqie, Moyao and Shiliuhua.Identification of these CMMs with anti-fibrotic activities in vitro underscores their inflammation-independent activities to inhibit TGF-1-induced total collagen accumulation, perhaps through inhibiting TGF-1-specific signalling or common pathways of fibrogenesis.Among the eight CMM extracts with anti-fibrotic activities, Baibeiyegen (Zhao et al., 2002), Lingzhi (Lin et al., 2006;Wang et al., 2009;Wu et al., 2010), Meiguiqie (Liu et al., 2006), Moyao (Massoud et al., 2004) and Shiliuhua (Huang et al., 2005) have been reported antifibrotic in animal models of liver fibrosis or in hepatic stellate cells; Gusuibu was one of the weakest and its effect on fibrosis has never been previously reported either in vivo or in vitro; Liedang has not been reported on its effect on fibrosis either, but the ethanolic extract of its closely related family member Boschniakia rossica (Cham.& Schltdl.)B. Fedtsch.has been reported to mitigate liver fibrosis in a rat model of dimethylnitrosamine-induced liver fibrosis (Piao et al., 2005); Jixueteng is used in some herbal formulae, such as Huangqijixuetengtang (Li et al., 2006;Li et al., 2007) and Herbal Compound 861 (also known as Fufangdanshenheji) (Wang 2000;Wang et al., 2008), which were reported to be antifibrotic in patients and in vitro, but has never been reported as an individual anti-fibrotic herb.
Although only a quarter of the selected CMMs showed anti-fibrotic activities in our model, we do not exclude the possibilities that some of these CMMs might have a favourable effect in treating fibrotic diseases by interfering with other factors e.g.inflammation, TGF-1 production and activation, molecules upstream of TGF-1 signalling and TGF-1independent pro-fibrotic signalling pathways.Of note, CMMs which were identified as cytotoxic in the NRK-49F renal fibroblast model might be of therapeutic value in fibrotic diseases since loss of fibroblasts, the main producers of pathological matrices, may result in reduced ECM accumulation hence promoting regression of fibrosis.It is also worth reiterating that the results presented here were derived from extracts of CMMs in boiling 80% ethanol.Different extraction methods, including extraction solutions and efficiency, might have different impact on in vitro activities of the materials.Nevertheless, it would be of particular interest to isolate compounds contained in the eight ethanolic CMM extracts with in vitro anti-fibrotic activities, and to further test individual compounds for their in vitro anti-fibrotic activities.

In vitro anti-fibrotic activities and TCM categorisation of the CMMs
CMMs are traditionally characterised based on their function, nature, taste and channel tropism.By focusing on functional groups of the selected CMMs, we found that four out of 15 (27%) CMMs of "Huo Xue Hua Yu" group, two out of six (33%) CMMs of "Hua Tan" group and three out of seven (43%) CMMs of "Bu Xu" group, had in vitro anti-fibrotic activities.In view that "Hua Tan" and "Bu Xu"groups had rarely been the focus of previous evidence-based studies of anti-fibrotic CMMs, we consider our findings significant in guiding future studies of anti-fibrotic herbal medicines and in selecting therapeutic options in the clinic.However, due to the small sample numbers involved in this study, we do not intend to conclude that CMMs of certain functional groups have a higher chance of being anti-fibrotic than another.For example, results of Danshen, a drug in the "Huo Xue Hua Yu" group, was excluded from this project as it has been tested and reported elsewhere (Hu et al., 2009).In fact, Danshen is the most used herb in the 16 herbal formulae that showed in vitro anti-fibrotic activities (Hu et al., 2009).On the other hand, our results do not negate the fact that other functional groups of CMMs may possess anti-fibrotic activities.For example, Dahuang (root of Rheum palmatum L.) and Huangqin (root of Scutellaria baicalensis Georgi), both not in these three functional groups, had been previously reported to be anti-fibrotic and their activities had been confirmed in our earlier studies (Hu et al., 2009).

Shiliuhua and PPAR
In this study, Shiliuhua extract was one of the most potent CMMs in reducing TGF-1induced matrix accumulation.Extracts of Shiliuhua had been previously reported to activate PPAR and induce PPAR expression.We found that ethanolic extract of Shiliuhua significantly activated both PPAR and PPAR, which might account for its anti-fibrotic activity since agonists of PPAR (Toyama et al., 2004;Iglarz et al., 2003) and PPAR (Milam et al., 2008;Kawai et al., 2008;Iglarz et al., 2003) were previously reported to suppress fibrosis of liver, lung, heart and kidney.Contrary to our hypothesis, we found that the antifibrotic activities of Shiliuhua extract can only be partially blocked by GW6471, a PPAR antagonist, at a dose 42-fold of its reported IC 50 .Since the dose of PPAR antagonists were selected based on their IC 50 reported in other cell types, it is possible that the doses we employed were sub-optimal in blocking PPARs in NRK-49F renal fibroblast cells.More interestingly, T0070907, a PPAR antagonist, further increased the anti-fibrotic effect of shiliuhua extract when used at high dose (25 M).Thus, it appears that Shiliuhua extract can activate both PPAR and PPAR, but this property could not explain the anti-fibrotic activity of Shiliuhua in full.Further studies are required to elucidate the involvement of PPAR signalling pathway in anti-fibrotic activities of shiliuhua, for example to establish if these two PPAR isotypes have opposing functions in regulating fibrogenesis in renal f i b r o b l a s t s i n v i e w t h a t t h e y d i d h a v e o pposing effects on monocyte chemotaxis in endometriosis (Hornung et al., 2001).

In vitro pro-fibrotic activities observed in this project
The in vitro pro-fibrotic activities of Chuanwutou, Dangshen and Yimucao deserve special attention.It rationalises pharmacovigilant studies to establish clinical relevance of these effects.Before clinical conclusions are drawn, it might be wise to avoid un-necessary, largedose and long-term use of these herbs, especially in patients prone to fibrotic diseases.This issue is important because Dangshen and Yimucao are commonly used in formulae for fibrotic diseases (Yao et al., 2003) and Chuanwutou is indicated for conditions such as osteoarthritis, muscular diseases and stroke that often are complicated by fibrosis as well.
As CMMs are rarely used individually, it is important to explore if their pro-fibrotic activities could be antagonised or eliminated when used in formulae.Of note, Yimucao had been reported in animal models to induce nephrotoxicity, including renal fibrosis, but its toxicity was reduced when used in formulae (Sun et al., 2005a(Sun et al., , 2005b)).

www.intechopen.com
Knowledge-Based Discovery of Anti-Fibrotic and Pro-Fibrotic Activities from Chinese Materia Medica 349 Among the three pro-fibrotic herbs, Chuanwutou appeared to be the most potent.Chuanwutou has well-known toxic effects that are believed to be reduced through a special processing procedure known as "Paozhi" (Chan et al., 1994;Singhuber et al., 2009), but as far as we know, this is the first report linking Chuanwutou to potent pro-fibrotic activities.
The Chuanwutou we examined in this project was a raw material that had not undergone any "Paozhi" before ethanolic extraction.It is important to establish if the potent pro-fibrotic effect of this toxic herb could be reduced or even eliminated through traditional "Paozhi" and if this newly identified adverse effect contributes to any known clinical toxicity and adverse effects, in view that only processed Chuanwutou is allowed to be used in TCM practice according to the Chinese Pharmacopeia.Interestingly, while extract of Dangshen, the root of Codonopsis pilosula (Franch.)Nannf.was found pro-fibrotic in this study, pollen of the same plant had been previously reported to be effective in preventing carbon tetrachloride-induced liver damage including fibrosis (Xiao et al., 1989).Furthermore, Fuzi, the daughter roots of the same plant as Chuanwutou, is a CMM categorised into a different functional group.We examined Fuzi in our in vitro model but it did not show any pro-fibrotic, anti-fibrotic or apparent cytotoxicity (data not shown).Thus, different parts of the same plant might have different and even opposite effects in regulating fibrogenesis.

Conclusion
Among the 28 herbal and fungal materials tested, eight showed in vitro anti-fibrotic activities while another three, especially Chuanwutou, showed pro-fibrotic activities.These results warrant further prudent investigations of their potential translation into clinical efficacy and their adverse effects.

Fig. 1 .
Fig. 1.In vitro anti-fibrotic and pro-fibrotic effects of four selected extracts, Shiliuhua (SLH) (a), Liedang (LD) (b), Meiguiqie (MGQ) (c) and Chuanwutou (CWT) (d) at three selected concentrations.Shown here are relative PSR OD changes of four independent experiments.The CDI changes were minimum and not shown.The average PSR OD values of negative control group and TGF-1-treated group were normalised to 0 and 100%, respectively, and changes in percentage of herbal extract-treated groups were relative to the TGF-β1 only group.*, **: p<0.05 and p<0.01 vs TGF-1 only group.

Fig. 2 .
Fig. 2. Effects of Shiliuhua (SLH) and Liedang (LD) extracts on PSR staining (a), relative PSR OD values (b) and LDH release (c).Results in (b) and (c) shown here are in Mean ± SEM from one representative experiment, ** p<0.01 vs TGF-β1 only group, n=6 wells.The average PSR OD values of negative control group and TGF-1 only group were normalised to 0 and 100%, respectively, and changes in percentage of herbal extract-treated groups were relative to the TGF-β1 only group.

Fig. 4 .
Fig. 4. Regulation of PPAR, PPAR/ and PPAR activation by Shiliuhua (SLH) extract and effects of PPAR, PPAR/ and PPAR antagonists on the anti-fibrotic effect of SLH extract.a. Effects of SLH extract on PPAR, PPAR/ and PPAR activation, n=3 independent biological experiments; b. effects of increasing doses of PPAR, PPAR/ and PPAR antagonists on the anti-fibrotic effect of SLH extract.** p<0.01 vs control group, n=3 independent biological experiments.Other statistical results are as indicated in the figures.