Sustained fibrosis with deposition of excessive extracellular matrix proteins leads to cirrhosis. Alcohol abuse is one of the main causes of severe liver disease. We established an ethanol-induced zebrafish fibrotic liver model to study the mechanisms and strategies of promoting hepatocyte regeneration upon alcohol-induced injury.
Sustained liver fibrosis with continuation of extracellular matrix (ECM) protein build-up results in the loss of cellular competency of the liver, leading to cirrhosis with hepatocellular dysfunction. Among multiple hepatic insults, alcohol abuse can lead to significant health problems including liver failure and hepatocellular carcinoma. Nonetheless, the identity of endogenous cellular sources that regenerate hepatocytes in response to alcohol has not been properly investigated. Moreover, few studies have effectively modeled hepatocyte regeneration upon alcohol-induced injury. We recently reported on establishing an ethanol (EtOH)-induced fibrotic liver model in zebrafish in which hepatic progenitor cells (HPCs) gave rise to hepatocytes upon near-complete hepatocyte loss in the presence of fibrogenic stimulus. Furthermore, through chemical screens using this model, we identified multiple small molecules that enhance hepatocyte regeneration. Here we describe in detail the procedures to develop an EtOH-induced fibrotic liver model and to perform chemical screens using this model in zebrafish. This protocol will be a critical tool to delineate the molecular and cellular mechanisms of how hepatocyte regenerates in the fibrotic liver. Furthermore, these methods will facilitate potential discovery of novel therapeutic strategies for chronic liver disease in vivo.
Despite the remarkable regeneration capacity of hepatocytes1, which are the major parenchymal cell type of the liver, chronic liver failure impairs this ability, leading to hepatic progenitor cell (HPC)-dependent regeneration2.
Chronic liver damage is mainly derived from alcohol abuse, chronic hepatitis C virus (HCV) infection3 and non-alcoholic fatty liver disease (NAFLD)4. It leads to sustained liver fibrosis, which is associated with the accumulation of extracellular matrix (ECM) proteins. Persisting ECM accumulation distorts intact hepatic architecture by forming a fibrous scar tissue5, subsequently resulting in cirrhosis with high morbidity and mortality. Many attempts have been made to mitigate the fibrotic response mainly by focusing on inhibiting profibrogenic cytokines and activated myofibroblasts6. The latter is primarily derived from hepatic stellate cells (HSCs), the principle hepatic non-parenchymal cells responsible for liver scar formation4. Nevertheless, regenerative therapies that stimulate endogenous cellular sources including HPCs to regenerate hepatocytes in the presence of sustained fibrogenic insults await further investigation.
Many experimental models of hepatic fibrosis have been described in mammals. Repetitive injection of carbon tetrachloride (CCl4) has been widely used to induce liver fibrosis in murine and rat models7. When combined with a high-fat (HF) diet, alcohol led to a substantial upregulation of profibrogenic gene expression and hepatic fibrosis8. While steatosis (lipid accumulation) results from acute alcohol exposure, it makes the liver susceptible to more severe hepatic injury9.
The zebrafish, Danio rerio, has emerged as an invaluable vertebrate model system for studying regeneration. Though other lower vertebrates such as newts and axolotls have a remarkable capacity for regeneration, the zebrafish has advantages over other model systems in regards to the gene manipulation and visualization strategies needed to manipulate potential regenerative factors10. The zebrafish also represents an attractive vertebrate model for studying alcoholic liver disease (ALD) by simply adding ethanol (EtOH) to their water. Acute EtOH exposure to larval and adult zebrafish caused hepatic steatosis11-13. When adult zebrafish received extended EtOH exposure, collagen deposition was observed with upregulation of fibrosis-related genes14. However, a need exists for developing models to study liver regeneration in response to EtOH as a fibrogenic stimulus.
Recently, we developed an EtOH-induced fibrotic liver model in zebrafish15. We combined a hepatocyte-specific genetic ablation system with EtOH treatment in larval and adult zebrafish. We generated two transgenic lines, Tg(fabp10a:CFP-NTR)gt1 and Tg(fabp10a:mCherry-NTR)gt2, in which E.coli nitroreductase (NTR) are fused to the cyan and mCherry fluorescent protein, respectively, under the control of the hepatocyte-specific fatty acid binding protein 10a, liver basic (fabp10a) promoter. In this system, NTR converts a nontoxic prodrug metronidazole (MTZ) into a DNA inter-strand cross-linking agent16, inducing explicit death of hepatocytes. Using this model, we demonstrated that a population of hepatic cells, which are responsive to Notch signaling, converted into hepatocytes in the near absence of hepatocytes and in the excess of ECM. We designated these cells as HPCs. Furthermore, through chemical screens, we identified small molecule activators of Wnt signaling and inhibitors of Notch signaling that augment hepatocyte regeneration in the fibrotic liver. Therefore, our fibrotic liver model in zebrafish represents a superb chemical screening system compared to cell culture- or mammalian-based screening system. It is an in vivo system with significant cost- and time-saving benefits. Here we describe the detailed procedures for establishing an EtOH-induced fibrotic liver model and for performing chemical screens using this model in zebrafish. Furthermore, time-course analyses were performed to investigate how hepatocyte regeneration occurs in the fibrotic liver. This protocol will provide an invaluable tool to study the mechanisms and strategies of enhancing hepatocyte regeneration in the fibrotic liver.
Zebrafish were raised and bred using a standard protocol that meets the criteria of the National Institutes of Health and approved by the Georgia Institute of Technology Institutional Animal Care and Use Committee.
1. Preparation of Solutions
2. Preparation of Larval Zebrafish
3. Ethanol, Metronidazole Treatment and Liver Regeneration in Larval Zebrafish
4. Chemical Screens in EtOH/MTZ-treated Larval Zebrafish
5. Larval Zebrafish Fixation, Immunostaining, and Confocal Imaging
6. Preparation of Adult Zebrafish
7. Ethanol, Metronidazole Treatment, and Liver Regeneration in Adult Zebrafish
8. Adult Zebrafish Liver Fixation, Immunostaining, and Confocal Imaging
Figure 1 shows the development of an EtOH-induced fibrotic liver model in larval zebrafish. To optimize a protocol for exposing zebrafish larvae to EtOH, we first assessed EtOH toxicity. 2.5 days-post-fertilization (dpf) larvae were exposed to EtOH concentration 1%, 1.5%, or 2% for 24 hr followed by a concurrent 24 hr EtOH/MTZ treatment. Exposure to 2% EtOH caused high mortality, while nearly all larvae exposed to 1% EtOH or less showed minimal fibrogenic changes with rare deposition of extracellular matrix proteins like collagen. Based on these results, the larvae were pretreated with 1.5% EtOH for 24 hr (2.5-3.5 dpf) and then simultaneously treated with MTZ for 24 hours (3.5-4.5 dpf) (Figure 1A). The EtOH/MTZ-treated larvae showed morphological abnormalities including upward curvature of the trunk and tail, pericardial edema, and failure to inflate swim bladder. (Figure 1B). We used three transgenic lines to analyze fibrogenic changes in the EtOH/MTZ-treated larval livers: 1) Tg(fabp10a:CFP-NTR)gt1 line expresses the NTR gene fused with cyan fluorescence protein under the hepatocyte-specific fabp10a promoter15; 2) Tg(Tp1:mCherry)jh11 line marks cells with active Notch signaling (Notch-responsive cells, NRCs) or biliary epithelial cells (BECs) with nuclear mCherry17, which expression is under the control of the TP1 module containing multiple RBP-Jk-binding sites; and 3) Tg(hand2:EGFP)pd24 line, which labels hepatic stellate cells (HSCs)13. DMSO-treated control [Tg(fabp10a:CFP-NTR)gt1; Tg(Tp1:mCherry)jh11; Tg(hand2:EGFP)pd24 ] livers showed no fibrillar type I collagen deposition4,6,9 (Figure 1C, C'), whereas EtOH/MTZ-treated livers displayed elevated type I collagen accumulation at 25 and 50 hr-post-ablation (hpa) (Figure 1D, D', E, E'). Additionally, compared to DMSO-treated control livers (Figure 1C''), HSCs increased in number with the altered morphology from a star-like configuration to a myofibroblast-like shape with lost cytoplasmic processes in the EtOH/MTZ-treated regenerating livers (Figure 1D'', E''). We observed two discrete populations of mCherry-expressing NRCs at 25 hpa in the EtOH/MTZ-treated regenerating livers: dim red NRCs (Figure 1D''', inset, white arrows) and bright red NRCs (Figure 1D''', inset, yellow arrowheads). At 50 hpa, hepatocyte-specific CFP began to co-express in NRCs with dim mCherry expression throughout the regenerating livers (Figure 1E''', inset, white arrows). These CFP and dim mCherry-coexpressing NRCs are evidently distinguishable from bright red NRCs that are negative for CFP (Figure 1E''', inset, yellow arrowheads). Previous studies showed that when hepatocyte proliferation was suppressed after partial hepatectomy (PHx), HPCs expanded with concomitant proliferation of HSCs21. Additionally, the HPCs and BECs shared histochemical markers22. Hence, our results indicate that the CFP and dim mCherry co-expressing NRCs portray a zebrafish-counterpart of HPCs that differentiate into hepatocytes by encountering Notch downregulation. The NRCs maintaining higher levels of Notch signaling may remain as cholangiocytes, which are differentiated BECs23.
Figure 2 shows the development of an EtOH-induced fibrotic liver model in adult zebrafish. First, we exposed adult zebrafish to 1%, 1.5%, or 2% EtOH to assess viability. Most adult zebrafish did not survive more than 72 hr of exposure to EtOH concentrations greater than 1% (unpublished data). Therefore, for adult zebrafish, we pretreated the fish with 1% EtOH for 72 hr and followed with MTZ treatment (Figure 2A). By performing time-course analyses, we showed significantly elevated type I collagen deposition in the EtOH/MTZ-treated regenerating livers at 2, 3, and 4 days-post-ablation (dpa) (Figure 2C', D', E') compared to DMSO-treated livers (Figure 2B'). Similar to that observed in the larvae, the CFP and dim mCherry co-expressing cells were detected in the EtOH/MTZ-treated regenerating livers of 12-month-old [Tg(fabp10a:CFP-NTR)gt1; Tg(Tp1:mCherry)jh11] adult fish at 3 and 4 dpa (Figure 2D, E, insets, white arrows). These data indicate that the HPCs, responsive to Notch signaling, maintain their capacity to regenerate as hepatocytes in the presence of fibrogenic insult in adult zebrafish.
Figure 3 shows the representative chemical screening results using our ethanol-induced fibrotic liver model in larval zebrafish. To pinpoint bioactive compounds that accelerate hepatocyte regeneration in the fibrotic liver, we performed chemical genetic screens. We screened a library of 75 small molecules with well-characterized biological and pharmaceutical activities (Stem Cell Signaling Compound Library, Selleckchem) and a library of 1,000 less-characterized small molecules (ActiProbe-1K Library, TimTec) using [Tg(fabp10a:CFP-NTR)gt1; Tg(Tp1:mCherry)jh11] larvae. We ablated the hepatocytes in the presence of 1.5% EtOH/15 mM MTZ and then exposed the larvae to 50 µM of the compounds for 50 hr (Figure 3A). A number of Wnt agonists such as SB 415286 and CHIR-99021, inhibitors of glycogen synthase kinase-3, promoted hepatocyte regeneration (Figure 3C, D) compared to DMSO (Figure 3B). Furthermore, [4-(1H-1,2,3,4-tetraazol-5-yl)-1,2,5-oxadiazole-3-ylamine], abbreviated as HTOA, a novel Wnt pathway activator, enhanced hepatocyte regeneration in the fibrotic liver as indicated by bigger regenerated liver (Figure 3E). These data suggest that our sustained fibrotic model provides an invaluable tool to discover small molecules that enhance hepatocyte regeneration.
Figure 1. Development of an ethanol-induced fibrotic liver model in larval zebrafish. (A) Scheme illustrating the periods of EtOH and EtOH/MTZ treatment for establishing a fibrotic liver model in larvae. (B) At 50 hours-post-ablation (hpa), EtOH/MTZ-treated larval zebrafish developed morphological abnormalities including upward curvature of the trunk and tail, pericardial edema, and failure to inflate the swim bladder. (C-C''') In the livers of [Tg(fabp10a:CFP-NTR)gt1; Tg(Tp1:mCherry)jh11; Tg(hand2:EGFP)pd24 ] larvae treated with DMSO, ECM protein collagen 1a was almost undetectable (C, C') and quiescent HSCs showed a star-like configuration (C''), whereas NRCs distributed among hepatocytes (C, C''', inset). (D-E''') In the livers of EtOH/MTZ-treated larvae, elevated collagen 1a deposition was observed (D, D', E, E'). HSCs increased in number and lost complex cytoplasmic processes (D'', E''). At 25 hpa, mCherry-expressing NRCs are composed of two distinct populations. Yellow arrowheads indicate bright red NRCs (D''', inset), whereas white arrows indicate dim red NRCs (D''', inset). At 50 hpa, Tg(fabp10a:CFP-NTR) and Tg(Tp1:mCherry) co-expressing cells started emerging throughout the regenerating livers (E''', inset, white arrows), while the bright red NRCs had no CFP expression (E''', inset, yellow arrowheads). All confocal images are single plane images except C'', D'', E'', which are projection images. Scale bars: B, 100 µm; C-E''', 20 µm. EtOH, ethanol; MTZ, metronidazole; hpa, hours-post-ablation; dpf, days-post-fertilization. Please click here to view a larger version of this figure.
Figure 2. Development of an ethanol-induced fibrotic liver model in adult zebrafish. (A) Scheme illustrating the periods of EtOH and MTZ treatment for establishing a fibrotic liver model in adult. (B-E') Tg(fabp10a:CFP-NTR), Tg(Tp1:mCherry), and collagen 1a expression in vibratome sections of DMSO- (B and B') and EtOH/MTZ-treated (C-E') adult zebrafish livers.(B-B') In the DMSO-treated controls, collagen 1a was almost undetectable (B') with no Tg(fabp10a:CFP-NTR) and Tg(Tp1:mCherry) co-expressing cells (B). (C-E') In the EtOH/MTZ-treated regenerating livers, collagen 1a deposition was markedly elevated at 2, 3, and 4 dpa (C', D', E') with a population of Tg(fabp10a:CFP-NTR) and Tg(Tp1:mCherry) co-expressing cells at 3 and 4 dpa (D, E, insets, white arrows). The bright red NRCs had no CFP expression (D, E, insets, yellow arrowheads). B-E, confocal single plane images. B'-E', confocal projection images. Scale bar, 20 µm. EtOH, ethanol; MTZ, metronidazole; dpa, days-post-ablation. Please click here to view a larger version of this figure.
Figure 3. Representative chemical screening results using the ethanol-induced fibrotic liver model in larval zebrafish. (A) Scheme for the hepatocyte regeneration screen in the fibrotic liver. (B-E) Known and novel Wnt activators augment hepatocyte regeneration in the fibrotic liver. Confocal projections of the EtOH/MTZ-treated livers in the [Tg(fabp10a:CFP-NTR)gt1; Tg(Tp1:mCherry)jh11] larvae that were administered with DMSO (B), SB415286 (C), CHIR-99021 (D), or a novel Wnt activator [4-(1H-1,2,3,4-tetraazol-5-yl)-1,2,5-oxadiazole-3-ylamine] (abbreviated as HTOA) (E). SB415286 and CHIR-99021 are glycogen synthase kinase-3 inhibitors. All images are confocal projection images. Scale bar, 20 µm. EtOH, ethanol; MTZ, metronidazole; hpa, hours-post-ablation; dpf, days-post-fertilization. Please click here to view a larger version of this figure.
We observed HPC-mediated hepatocyte regeneration in the EtOH/MTZ-treated recovering livers, suggesting that even in the presence of substantial amount of ECM proteins including fibrillar type I collagen, the HPCs retain their competency to regenerate as hepatocytes. The MTZ only-treatment did not increase deposition of ECM proteins significantly, whereas the EtOH only-treatment did not induce HPC activation15. By utilizing the combined EtOH/MTZ treatment, we were able to investigate HPC-driven regeneration in the fibrotic liver. As near-complete elimination of hepatocyte elicits HPC-driven hepatocyte regeneration, it is critical to sort out MTZ-treated larvae based on the criteria of absence of hepatocyte-specific fluorescence and significantly reduced liver volume by clustered NRCs. In fish, the blood vessels of the gill and skin absorb the alcohol24, so that there is no need to allow animals to drink ad libitum or require intragastric infusion as in mammalian models. We housed EtOH- and subsequent MTZ-treated adult fish in separate tanks without circulating the water. It is essential to transfer fish to fresh EtOH solution daily and keep the lid on to minimize the evaporation of EtOH. Although 48-72 hr EtOH/MTZ treatment efficiently induces ECM protein synthesis in our protocols, neither advanced fibrosis nor cirrhosis has been developed in these fish. Therefore, combination of MTZ with prolonged exposure of ethanol such as 12 weeks treatment14, specifically in adult fish, may induce more severe liver damage to properly simulate and interrogate HPC-driven regeneration in chronic liver failure.
Zebrafish serves as an outstanding animal model for in vivo compound testing due to its small size, large progeny, transparency, and permeability to small molecules25. Using the fibrotic liver model in zebrafish, we identified a number of known and novel Wnt activators and Notch inhibitors that accelerated hepatocyte regeneration15. Our current fluorescence-based chemical screening protocol consists of several steps all performed manually including preparation of chemicals, transferring EtOH-treated/hepatocyte-ablated larvae to 24-well plates, treating chemicals, and analysis of effects of chemicals on hepatocyte regeneration. It specifically employs epifluorescence and/or confocal microscope to capture images for individual animal, which is laborious and time consuming. High-throughput platforms that automatically handle and acquire cellular-resolution imaging of zebrafish in combination with quantitative data collection will facilitate large-scale screening26,27.These features will be strengthened if combined with automated drug dosing system that determines the range of chemical concentrations28, which can give minimum toxicity with maximum efficacy for enhancing hepatocyte regeneration. Despite these limitations, as many critical players and main cell types of the liver are conserved between zebrafish and mammals29, employing the in vivo-based small molecule screening using the zebrafish fibrotic liver model will catalyze valid discovery of novel therapeutic strategies for chronic liver disease.
Two additional groups independently have demonstrated that after near total loss of hepatocytes without fibrogenic insults, BECs transdifferentiated into hepatocyte through a step of dedifferentiation30,31 with the assumption that cholangiocytes, fully differentiated BECs, primarily comprise NRCs/BECs in zebrafish. Although transdifferentiation can be one of the mechanisms of driving hepatocyte regeneration, our results indicate that the HPCs, which are known to constitute the majority of BECs in the zebrafish liver32, downregulate Notch activity to differentiate into hepatocytes. Meanwhile, it is plausible to speculate that the fully differentiated BECs, cholangiocytes, maintain higher levels of Notch activity without converting to hepatocytes during regeneration. Intriguingly, in the presence of EtOH, as previously reported13, we found that HSCs increased in number with the altered morphology from a star-like configuration to a myofibroblast-like shape with lost cytoplasmic processes. The activated HSCs characterized by synthesis of ECM proteins are the principle culprit responsible for abnormal wound healing during the integrated process of liver repair4. Although chronic injury often overrides the liver's remarkable capacity for regeneration, transplanted HPCs successfully repopulated the fibrotic/cirrhotic rat liver33. Hence, our zebrafish fibrotic liver model will be an essential tool for elucidating the molecular and cellular mechanisms that mediate the effects of sustained fibrosis on HPC-mediated hepatocyte regeneration. Furthermore, defining the multi-cellular crosstalk that balances regeneration and fibrosis by utilizing our zebrafish fibrotic liver model will further facilitate to design viable therapeutic strategies for chronic liver failure in vivo.
The authors have nothing to disclose.
This work was supported in part by grants from the GTEC (2731336 and 1411318), the NIH (K01DK081351), and the NSF (1354837) to C. H. S. We thank Alem Giorgis for critical reading of the manuscript.
Calcium sulfate hemihydrate (CaSO4) | Acros | AC385355000 | |
Magnesium sulfate (MgSO4) | EMD | MX0075 | |
1,4-Piperazinediethanesulfonic acid (PIPES) | Sigma-Aldrich | P6757 | |
Ethylene glycol-bis(2-aminoethylether)-N,N,N′,N′-tetraacetic acid (EGTA) | Sigma-Aldrich | E3889 | |
Ethanol | Sigma-Aldrich | E7023 | 200 proof |
Formaldehyde | Fisher Scientific | F79-500 | |
Metronidazole (MTZ) | Sigma-Aldrich | M3761 | |
1-phenyl-2-thiourea (PTU) | Sigma-Aldrich | P7629 | |
3-amino benzoic acid ethyl ester (Tricaine) | Sigma-Aldrich | A5040 | |
Phosphate-buffered saline (PBS) tablet | Amresco | E404 | Dissolve one tablet with 100 ml distilled water |
Dimethyl sulfoxide (DMSO) | Sigma-Aldrich | D2438 | |
Bovine serum albumin | Fisher Scientific | BP1600 | |
Triton X-100 | Fisher Scientific | BP151 | |
Low-melting agarose | Amresco | BP165 | |
Stem Cell Signaling Compound Library | Selleck Chemicals | L2100 | 10mM stock in DMSO |
ActiProbe-1K Library | Timtec | ActiProbe-1K | 10mM stock in DMSO |
SB 415286 | Selleck Chemicals | S2729 | Dissolve with DMSO to 10mM |
CHIR-99021 | Selleck Chemicals | S2924 | Dissolve with DMSO to 10mM |
Anti-Collagen I antibody | Abcam | ab23730 | Use at 1:100 for immunostaining, reacts with fish |
AlexaFluor 647 Donkey anti-rabbit IgG (H+L) | Molecular Probes | A31573 | Use at 1:200 for immunostaining |
Mounting media (Vectorshield) | Vector Laboratories | H-1400 | |
100 mm petri dish | VWR | 25384-088 | |
24-well plate | VWR | 10062-896 | |
Forceps | Fine Science Tools | 11255-20 | Dumont #55 |
Glass slide | VWR | 48312-003 | 75×25 mm |
Cover glass | VWR | 48366-045 | 18 mm |
Plastic wrap | Fisher Scientific | 22305654 | |
Aluminum foil | Fisher Scientific | 1213100 | |
Kimwipes | Kimberly-Clark | 34155 | |
Vibrotome | Leica | VT1000 S | |
Stereo microscope | Leica | M80 | |
Epifluoresent microscope | Leica | M205 FA | |
Confocol microscope | Zeiss | LSM700 |