This protocol provides a simple and reliable method for the production of viable precision-cut liver slices from mice. The ex vivo tissue samples can be maintained under laboratory tissue culture conditions for multiple days, providing a flexible model to examine liver pathobiology.
Understanding the mechanisms of liver injury, hepatic fibrosis, and cirrhosis that underlie chronic liver diseases (i.e., viral hepatitis, non-alcoholic fatty liver disease, metabolic liver disease, and liver cancer) requires experimental manipulation of animal models and in vitro cell cultures. Both techniques have limitations, such as the requirement of large numbers of animals for in vivo manipulation. However, in vitro cell cultures do not reproduce the structure and function of the multicellular hepatic environment. The use of precision-cut liver slices is a technique in which uniform slices of viable mouse liver are maintained in laboratory tissue culture for experimental manipulation. This technique occupies an experimental niche that exists between animal studies and in vitro cell culture methods. The presented protocol describes a straightforward and reliable method to isolate and culture precision-cut liver slices from mice. As an application of this technique, ex vivo liver slices are treated with bile acids to simulate cholestatic liver injury and ultimately assess the mechanisms of hepatic fibrogenesis.
The pathogenesis of most chronic liver diseases (i.e., viral hepatitis, nonalcoholic steatohepatitis, cholestatic liver injury and liver cancer) involves complex interactions between multiple different liver cell types that drive inflammation, fibrogenesis, and cancer development1,2. To understand the molecular mechanisms underlying these chronic liver-based diseases, the interactions between multiple liver cell types must be investigated. While multiple hepatic cell lines (and more recently, organoids) can be cultured in vitro, these models do not accurately emulate the complex structure, function, and cellular diversity of the hepatic microenvironment3. Furthermore, cultured liver cells (in particular, transformed cell lines) may deviate from their original source biology. Animal models are used experimentally to investigate the interactions between multiple liver cell types. However, they may become significantly reduced in scope for experimental manipulation, due to significant off-target effects in extrahepatic organs (e.g., when testing potential therapeutics).
The use of precision-cut liver slices (PCLS) in tissue culture is an experimental technique first used in drug metabolism and toxicity studies, and it involves the cutting of viable, ultrathin (around 100−250 µm thick) liver slices. This permits the direct experimental manipulation of liver tissue ex vivo4. The technique bridges an experimental gap between in vivo animal studies and in vitro cell culture methods, overcoming many drawbacks of both methods (i.e., practical limits on the range of experiments that can be performed in whole animals as well as loss of structure/function and cellular diversity with in vitro cell culture methods).
Furthermore, PCLS vastly increases experimental capacity compared to whole animal studies. As one mouse can produce more than 48 liver slices, this also facilitates the use of both control and treatment groups from the same liver. In addition, the technique physically separates the liver tissue from other organ systems; therefore, it removes potential off-target effects that can occur in whole animals when testing the effects of exogenous stimuli.
In this protocol, PCLS are generated using a vibratome with a laterally vibrating blade. Other studies have successfully used a Krumdieck tissue slicer, as described in Olinga and Schuppan5. In the vibratome, lateral vibration of the blade prevents tearing of the ultrathin tissue caused by shear stress, as the blade is pushed into the tissue. Both the vibratome and Krumdieck tissue slicer work effectively without structural embedding of liver tissue, which streamlines the slicing procedure. This technique can also be used to create PCLS from diseased livers, including those from mouse models of fibrosis/cirrhosis6 and hepatic steatosis7.
In addition to demonstrating the techniques required for preparation and tissue culture of PCLS, this report also examines the viability of these ex vivo tissues by measuring adenosine triphosphate (ATP) levels and examining tissue histology to assess necrosis and fibrosis. As a representative experimental procedure, PCLS are treated with pathophysiological concentrations of three different bile acids (glycocholic, taurocholic, and cholic acids) to simulate cholestatic liver injury. In the context of cholestatic liver injury, taurocholic acid in particular has been shown to be significantly increased in both the serum and bile of children with cystic fibrosis-associated liver disease8.
Liver progenitor cells have also been treated in vitro with taurocholic acid to simulate the elevated taurocholic acid levels observed in patients, and this treatment caused increased proliferation and differentiation of liver progenitor cells towards a biliary (cholangiocyte) phenotype9. Subsequently, PCLS were treated ex vivo with elevated levels of taurocholic acid, and increased cholangiocyte markers were observed. This supports the in vitro observation that taurocholic acid drives biliary proliferation and/or differentiation in the context of pediatric cystic fibrosis-associated liver disease9.
All animal experiments were performed in accordance with the Australian code for the care and use of animals for scientific purposes at QIMR Berghofer Medical Research Institute with approval from the institute animal ethics committee. Male C57BL/6 mice (15−20 weeks old) were obtained from the Animal Resources Centre, WA, Australia.
NOTE: All solutions, media, instruments, hardware, and tubes that contact the samples must be sterilized or thoroughly disinfected with a 70% ethanol solution and handled using sterile techniques to minimize the risk of culture contamination.
1. Setup of the vibratome
2. Liver removal and preparation
3. Production of liver slices
4. Tissue culture
NOTE: All tissue culture work must be performed in a sterile laminar flow hood.
5. Example application of PCLS
To determine the cell viability of PCLS over time, tissue ATP levels were measured. ATP levels are typically proportional to viability. PCLS (around 15 mm2 in area) were cultured in normal William's E medium with 10% FBS, then at specific timepoints, liver slices were removed from tissue culture and homogenized with both ATP and protein (for normalization) concentrations (Table of Materials) being measured (Figure 1A). For biochemical assays like this, normalization is important, as the cut liver slices are not necessarily identical in dimensions. ATP levels (relative to protein) were suppressed immediately post-isolation and after 1 h (Figure 1A), suggesting short-term metabolic stress from the cooling and cutting procedures. However, ATP levels recovered by 3 h. ATP levels remained elevated at up to 5 days of tissue culture, indicating no significant decrease in viability. Hematoxylin and eosin (H&E) staining of liver slices suggested that limited tissue necrosis (characterized by nuclear pleomorphism) occurred in culture from around days 2 and 3. Tissue necrosis levels progressed to severe by day 5 (Figure 1B). Considering these morphological data, taken together with the ATP results, it is recommended to use this experimental tissue model for up to 3 days.
PCLS also displayed increasing collagen accumulation at later culture timepoints, as shown by thickening of Picro-sirius red-stained collagen fibers at day 5 (Figure 2). This thickening of collagen fibers suggests that spontaneous fibrogenic processes are active in PCLS obtained using this method. This process appears independent of PCLS isolation methodology with the thickening of collagen fibers6 and profibrogenic gene expression14 occurring from both vibratome and Krumdieck tissue slicers, respectively, over time. The development of these spontaneous fibrogenic processes needs be taken into account when interpreting PCLS biology, particularly with experiments associated with fibrotic processes.
We have previously shown the significant induction of cholangiocyte-specific gene connexin 43 (Cx43) and secretion of glutamyl transpeptidase (Ggt1) protein by TCA in PCLS9. The expression of cholangiocyte-specific genes relative to housekeeping controls (glyceraldehyde 3-phosphate dehydrogenase [Gapdh] and hypoxanthine phosphoribosyltransferase 1 [Hprt1]) in PCLS treated with TCA, GCA, or CA were examined by qPCR using specific primers. Consistent with a previous report, a significant induction in the expression of cholangiocyte-specific genes cytokeratin 19 (CK19; Figure 3A) and connexin 43 (Cx43; Figure 3B) were observed by both GCA and TCA. Ggt1 expression was increased by both bile acids; although, this did not reach statistical significance, possibly due to experimental variation (Figure 3C). The expression of CA was unaffected by any bile acid. The induction of cholangiocyte-specific genes suggests that GCA may also be involved in cholestatic liver injury, as previously reported for TCA8,9.
Figure 1: Tissue viability of precision-cut liver slices (PCLS). (A) ATP and protein levels in PCLS were measured immediately (T + 0) and at 1 h, 3 h, 6 h, and 1−5 days post-isolation. (B) To examine cell morphology, PCLS were fixed, paraffin-embedded, sectioned, and stained with H&E immediately (day 0) and at 1, 2, and 5 days post-isolation. A Kruskal-Wallis test with Dunn's multiple comparisons test was performed relative to T + 0. All data are represented as mean ± SEM (n = 2−3 mice) with *p < 0.05. The scale bars represent 100 µm. Please click here to view a larger version of this figure.
Figure 2: Assessment of collagen deposition in PCLS. PCLS were fixed, paraffin-embedded, sectioned, and stained with Picro-Sirius Red to visualize collagen fibers at days 0, 1, 2, and 5 post-isolation. The scale bars represent 200 µm. Please click here to view a larger version of this figure.
Figure 3: Bile acids induce cholangiocyte-specific gene expression. At 16 h post-isolation, medium on PCLS was changed, and new medium was added along with 150 μM glycocholic (GCA), taurocholic (TCA), or cholic (CA) acids (sodium salts). PCLS were harvested after 2 days, and the expression of cytokeratin 19 (CK19; A), connexin 43 (Cx43; B), and γ-glutamyl transpeptidase 1 (Ggt1; C) was examined relative to the geometric mean of Gapdh and Hprt1. Dunnett's multiple comparisons test was performed relative to untreated control tissue slices (n = 15). All data are represented as mean ± SEM (n = 4 mice, triplicate slices; *p < 0.05, **p < 0.01). Please click here to view a larger version of this figure.
Table 1: qPCR primers.
The protocol demonstrates the application of murine PCLS isolation and tissue culture, and the procedures are designed to assess both viability and utility as well as examine impacts of exogenous mediators of liver pathobiology using biochemical assays, histology, and qPCR. The experimental utility of PCLS tissue culture in rodents and humans has been demonstrated in a wide range of applications, including experimental investigations in microRNA15/RNA9/protein expression16, metabolism17, viral infection dynamics10,18, infection signaling19, tumor invasion12, toxicity studies4,13,15,20, DNA damage studies21, cell biology14, and secretion studies9.
While ATP levels indicate cellular viability in PCLS up to 5 days in culture, H&E staining suggest that severe necrosis was occurring by 5 days in culture. The main factor that appears to limit the viability of PCLS is oxygen availability6,11. Several studies have included enhanced oxygen availability and consequently increased the viability time of PCLS in tissue culture. Methods used to increase oxygen availability include the incubation of tissue in oxygen-rich atmospheres10,11, use of the oxygen carrier perfusion perfluorodecalin12, shaking/rolling the culture during incubation6,10,13, and tissue culture plates designed to maximize oxygenation6.
Recently, an innovative air-liquid interface tissue culture system has been described with functional utility stated at longer than 7 days in culture14. This protocol uses ambient atmospheric oxygen in a normal tissue culture incubator. The method allows PCLS to be accessible to laboratories that do not have highly specialized, custom equipment to safely enhance oxygen delivery. However, if such methods to enhance oxygen delivery are available, they may improve long-term PCLS viability in culture.
The accumulation of collagen in PCLS after day 3 in culture suggests that spontaneous fibrogenic processes are active within this model. Spontaneous fibrosis has also been observed previously in PCLS6,14,22,23,24 and is possibly mediated by damage-associated molecular patterns (DAMPs) released by the cutting process or tissue necrosis. DAMPs act as signaling molecules that subsequently activate pro-fibrotic signaling pathways25,26. Furthermore, spontaneous fibrosis in PCLS could be mediated by chemokines released from the activation of stellate cells and/or Kupffer cells (i.e., transforming growth factor beta [TGF-β]). In this context, a recent article by Bigaeva et al.24 suggests that spontaneous fibrosis in PCLS is in part mediated by TGF-β signaling, as incubation of PCLS with TGF-β inhibitor Galunisertib inhibited gene expression changes associated with spontaneous hepatic fibrosis. Another possible mechanism is that the slicing procedure initially induces cell proliferation signaling pathways and entry of hepatocytes into the cell cycle; however, this mechanism fails and results in cell cycle arrest in mid-G1 phase27. Cell cycle arrest is associated with hepatic fibrosis rather than hepatic regeneration28.
In the representative experimental procedure, PCLS are treated with three different bile acids (GCA, TCA, and CA) for 2 days to simulate cholestatic liver injury. Two of the bile acids (GCA and TCA) induce significant expression of CK19 and Cx43, which are genes associated with cholangiocyte function. This suggests the expansion or differentiation of cells towards a cholangiocyte lineage in PCLS treated with these bile acids. This is consistent with our previous work showing a similar effect using TCA9 on PCLS. Furthermore, it is also consistent with in vivo observations that show feeding taurocholate to rats increases cholangiocyte numbers29. The treatment of liver tissue slices with bile acids simulates hepatic cholestasis, and it is speculated that the increase in the expression of genes associated with cholangiocyte function is an attempt by the liver tissue to create additional bile ductules to increase bile secretion. Given the specific induction of these genes is produced by the conjugated bile acids (GCA and TCA) but not the unconjugated CA, it is suspected that these effects are mediated by the sphingosine-1-phosphate receptor 2, a cell surface receptor with preferential activation by conjugated bile acids30.
One key limitation of the application of PCLS is individual replicate variability. Since the liver is not homogenous and there is variability in production, liver slices tend to have larger biological variation than cell culture experimentation. In experimental studies with few variables, such as the representative procedure demonstrated above, this is overcome by using an increased number of replicates. However, this may become an issue for larger screening studies. In summary, the described protocol is a straightforward and reliable method to study aspects of liver pathobiology ex vivo, requiring little specialized equipment except for the vibratome.
The authors have nothing to disclose.
This work was supported by research grants from the National Health and Medical Research Council (NHMRC) of Australia (Grant No. APP1048740 and APP1142394 to G.A.R.; APP1160323 to J.E.E.T., J.K.O., G.A.R.). Grant A. Ramm is supported by a Senior Research Fellowship from the NHMRC of Australia (Grant No. APP1061332). Manuel Fernandez-Rojo was supported by the TALENTO program of Madrid, Spain (T1-BIO-1854).
10 cm Petri Dish | GREINER | 664160 | Sterile Dish |
12 Well Tissue Culture Plate Flat Bottom | Greiner Bio-one | 665180 | |
70% Ethanol Solution (made with AR Grade) | Chem-Supply Pty Ltd | EA043-20L-P | Disinfection solution |
Acetone | Chem-Supply Pty Ltd | AA008-2.5L | |
Cholic acid | Sigma-Aldrich | C1129-100G | |
Cyanoacrylate Super Glue | Parfix, DuluxGroup (Australia) | Other brands should work | |
Disposable Single Edge Safety Razor Blades | Mixed | ||
Dissection Board | Made in-house | Sterile material over polystyrene | |
Fetal Bovine Serum | GE Healthcare Australia Pty Ltd | SH30084.02 | |
Forceps sharp point 130 mm long | ThermoFisher Scientific | MET2115-130 | |
Forma Steri-Cycle CO2 Incubator | ThermoFisher Scientific | 371 | |
Glutamine | Life Technologies Australia Pty Ltd | 25030081 | |
Glycocholic acid hydrate | Sigma-Aldrich | G2878-100G | |
ISOLATE II RNA Mini Kit | Bioline (Aust) Pty Ltd | BIO-52073 | |
Ketamine 50 ml | Provet | KETAI1 | |
Krebs-Henseleit Buffer with Added Glucose 2000 mg/L | Sigma-Aldrich | K3753 | Can also be made in house |
Laminar Flow Hood | Hepa air filtration | ||
NanoDrop 2000/2000c Spectrophotometers | ThermoFisher Scientific | ||
Penicillin-Streptomycin, Liq 100 ml | Life Technologies Australia Pty Ltd | 15140-122 | |
Picro Sirius Red | ABCAM Australia Pty Ltd | ab246832 | |
Pipette Tips Abt 1000 µl Filter Interpath | Interpath | 24800 | |
Pipette Tips Abt 10 µl Filter Interpath | Interpath | 24300 | |
Pipette Tips Abt 200 µl Filter Interpath | Interpath | 24700 | |
Pipette Tips Abt 20 µl Filter Interpath | Interpath | 24500 | |
Precellys Homogeniser | Bertin Instruments | P000669-PR240-A | |
Protractor | Generic | To measure blade angle | |
Quantstudio 5 QPCR Fixed 384 Block | Applied Biosystems/ ThermoFisher Scientific | ||
Scalpel Blade | Mixed | ||
Scalpel Blade Holder | Mixed | ||
SensiFAST cDNA Synthesis Kit | Bioline (Aust) PTY LTD | ||
Small Paintbrush with Plastic Handle | Mixed | Plastic handle resists ethanol | |
Square-Head Foreceps | Mixed | ||
Sterile 50 ml Plastic Tubes | Corning Falcon | 352098 | |
Surgical Clamps | Mixed | ||
Surgical Forceps | Mixed | ||
Surgical Pins | Mixed | ||
Surgical Scissors | Mixed | ||
Taurochoic acid | Sigma-Aldrich | T-4009-5G | |
Vibratome SYS-NVSLM1 Motorized Vibroslice | World Precision Instruments | SYS-NVSLM1 | With thermoelectric cooling |
Williams Medium E | Life Technologies Australia Pty Ltd | 12551032 | 2.0 g/l glucose |
Xylazine 100 mg/mL 50 mL | Provet | XYLAZ4 |