Here, we present a protocol for the isolation of healthy and functional primary mouse hepatocytes. Instructions for detecting hepatic nascent protein synthesis by non-radioactive labeling substrate were provided to help understand the mechanisms underlying protein synthesis in the context of energy-metabolism homeostasis in the liver.
Hepatocytes are parenchymal cells of the liver and engage multiple metabolic functions, including synthesis and secretion of proteins essential for systemic energy homeostasis. Primary hepatocytes isolated from the murine liver constitute a valuable biological tool to understand the functional properties or alterations occurring in the liver. Herein we describe a method for the isolation and culture of primary mouse hepatocytes by performing a two-step collagenase perfusion technique and discuss their utilization for investigating protein metabolism. The liver of an adult mouse is sequentially perfused with ethylene glycol-bis tetraacetic acid (EGTA) and collagenase, followed by the isolation of hepatocytes with the density gradient buffer. These isolated hepatocytes are viable on culture plates and maintain the majority of endowed characteristics of hepatocytes. These hepatocytes can be used for assessments of protein metabolism including nascent protein synthesis with non-radioactive reagents. We show that the isolated hepatocytes are readily controlled and comprise a higher quality and volume stability of protein synthesis linked to energy metabolism by utilizing the chemo-selective ligation reaction with a Tetramethylrhodamine (TAMRA) protein detection method and western blotting analyses. Therefore, this method is valuable for investigating hepatic nascent protein synthesis linked to energy homeostasis. The following protocol outlines the materials and methods for the isolation of high-quality primary mouse hepatocytes and detection of nascent protein synthesis.
Protein is an important nutritional element and approximately 50% of the dry weight of a human body is composed of proteins which have several biological traits and functions1. Consequently, protein synthesis is one of the most energy consuming events and an alteration in protein metabolism is highly associated with the development of diseases, including metabolic diseases2,3,4. In the liver, protein biosynthesis accounts for approximately 20–30% of total energy consumption5,6. In addition, proteins function as not only passive or building blocks of the liver, but also active signal-mediating factors intracellularly or extracellularly to regulate the systemic metabolism7. For instance, reduced levels of serum albumin, which is synthesized and secreted by the liver and the most abundant protein in the plasma8, increases the risk of type 2 diabetes development9,10,11, whereas a higher concentration of serum albumin is protective against developing metabolic syndrome12. Furthermore, disturbed or disruption of secretory or membrane-bound hepatic proteins, which modulate cholesterol homeostasis, including lipoproteins, LDLR, and LRP1, can lead to the development of insulin resistance, hyperlipidemia, or atherosclerosis13. Therefore, the identification of molecular pathophysiological mechanisms that are involved in protein metabolism disturbance in the liver and its associated metabolic complications might be useful for discovering novel pharmacological approaches to retard onset or treat metabolic diseases such as insulin resistance, diabetes, and non-alcoholic fatty liver.
Protein synthesis is tightly linked to the cellular energy status (e.g., the formation of one peptide bond during the elongation step of protein synthesis requires 4 phosphodiester bonds14) and is regulated by molecular pathways that sense inter- and intra-cellular nutrient availability15,16. AMP-activated protein kinase (AMPK) is one of the intracellular energy sensors that maintain energy homeostasis17. Once AMPK is activated when cellular energy levels become lower, AMPK and its targeted substrates function to stimulate catabolic pathways and inhibit anabolic processes including protein synthesis18,19. The regulation of protein synthesis is mediated by phosphorylation of multiple translation factors and ribosomal proteins20. Of note, the mammalian target of rapamycin complex 1 (mTORC1), a major driver of protein synthesis, is one of the main targets of AMPK21. Activation of the mTORC1 pathway enhances the cell growth and proliferation by stimulating protein translation and autophagy20,21. Therefore, it is logical that activation of AMPK can inhibit mTORC1-mediated protein synthesis22. Indeed, activation of AMPK counteracts and directly phosphorylates mTORC1 on threonine residue 2446 (Thr2446) leading to its inactivation23 and suppression of protein biosynthesis24. Moreover, AMPK can indirectly inhibit mTORC1 function by phosphorylation and activation of tuberous sclerosis complex 2 (TSC2)25 which is the upstream regulator of mTORC1 signaling cascade. In short, dysregulation of these pathways in the liver is often linked to the development of metabolic diseases and therefore there is a critical need to establish effective experimental tools to investigate the role of these pathways in the regulation of energy and protein metabolism in hepatocytes.
There is a stronger similarity between the functional properties of isolated primary hepatocytes and in vivo hepatocytes than with in vitro liver-derived cell lines26,27,28. It has been shown that primary human hepatocytes share 77% similarity with that of liver biopsies, whereas HepG2 cells, which are well-differentiated hepatic cancerous cells and widely used to investigate hepatic functions, display less than 48% in the context of gene expression profiles29. Therefore, utilization of primary hepatocytes, rather than immortalized culture cells, is of vital importance in investigating hepatic function and physiology, and several protocols are available for the isolation and culture of primary hepatocytes especially from rats30,31. While the rat hepatocytes are useful with a relatively higher yield of cells, mouse hepatocytes have a greater potential in many scientific aspects because of the wide availability of genetically modulated mice. However, there are several technical challenges in isolating healthy and abundant primary hepatocytes from mice for cellular- and molecular-based assessments: first, cannula insertion to perfuse the liver with buffer reagents is very difficult to handle because of the small and thin mouse portal vein or inferior vena cava; second, a longer manipulation time of cells during the isolation can cause reduction in cell quantity and quality; third, non-enzymatic mechanical separation methods can result in severe damage and produce a low yield of viable isolated primary hepatocytes32,33. In the 1980s, collagenase perfusion technique was introduced for isolating hepatocytes from the livers of animals34. This method is based on collagenase perfusion of the liver35,36,37, infusion of the liver with calcium chelator solution38,39, enzymatic digestion and mechanical dissociation of the hepatic parenchyma35. In the first step, a mouse liver is perfused with a calcium [Ca2+] free buffer containing a [Ca2+] chelator (ethylenediamine tetraacetic acid, EDTA). In the second step, the mouse liver is perfused with a collagenase-containing buffer to hydrolyze the cellular-extracellular matrix interactions. Unlike the buffer used in step one, the presence of [Ca2+] ions in the buffer of the second step is required for effective collagenase activity, after which the digested liver has to be further gently and mechanically separated using forceps between the hepatic capsule and parenchymatous tissue. Finally, connective tissue is removed by filtering, and subsequent centrifugation separates viable hepatocytes from both non-parenchymal cells and non-living hepatocyte with the use of density gradient buffer40,41,42. In the present study, we show a modified two-step collagenase perfusion technique to isolate primary hepatocytes from a mouse liver for the analysis of protein synthesis.
The radiolabeling of proteins is widely used to quantify the expression levels, turnover rates, and determine the biological distribution of proteins43 due to the high sensitivity of detection of radioactivity44. However, the use of radioactive isotopes requires highly controlled research circumstances and procedures45. Alternative non-radioactive methods have been developed and have increasingly gained in popularity. The chemo-selective ligation reaction with a Tetramethylrhodamine (TAMRA) protein detection method is one of them and is based on the chemo-selective reaction between an azide and alkyne groups46, which can be utilized to analyze cellular events such as detecting nascent protein synthesis and subclasses of glycoproteins modified with an azide group. For nascent protein synthesis, L-Azidohomoalanine (L-AHA, an azide-modified amino acid) can be metabolically incorporated into proteins and detected by using the TAMRA protein detection method47. By using this assay in primary mouse hepatocytes, we show that the nascent protein synthesis rate is tightly linked to the availability of ATP from mitochondrial and AMPK activation (Figure 1).
In summary, utilization of primary mouse hepatocytes is crucial to investigating the protein and energy metabolism and quantifying nascent protein synthesis is valuable for gaining insights into the physiological role of pathways relevant to the development and cure of hepatocyte-related diseases.
This protocol contains the use of laboratory mice. Animal care and experimental procedures were performed according to procedures approved by the animal care committees of Cincinnati Children Hospital Medical Center.
1. Isolation of Primary Mouse Hepatocytes
2. Chemo-selective Ligation Reaction Assay for the Detection of Nascent Protein Synthesis
Primary mouse hepatocytes isolation results in a yield of approximately 20 x 106 total cells/mouse. Histologically, live and attached primary hepatocytes appear polygonal or typical hexagonal in shape with clearly outlined membranous boundary after 24 h incubation (Figure 2).
To confirm whether isolated cells are primary hepatocytes, we compared expression levels of albumin protein in isolated primary mouse hepatocytes (PMHs), mouse embryonic fibroblasts (MEFs), a mouse hepatoma cell line (Hepa 1-6), and mouse livers. The albumin protein expression was detected in primary mouse hepatocytes and mouse livers but was not detectable in either MEFs or Hepa 1-6 cells (Figure 3).
To determine whether the effects of rotenone on hepatic AMP-activated protein kinase (AMPK) activation is cell-autonomous, we compared the time-dependent response of primary hepatocytes isolated from a liver of wild-type mouse. Primary hepatocytes were treated without or with 10 μM rotenone for 0, 4 or 6 h. In these cells, rotenone treatment robustly induced AMPK activation, as detected by increased phosphorylation levels on AMPK-T172 similar to those seen in vivo48 (Figure 4).
To directly measure the effects of rotenone treatment on nascent protein synthesis in primary hepatocytes, the incorporation of non-radioactive AHA into protein over 5 h was assessed by performing the chemo-selective ligation reaction assay. Treatment with 10 μM rotenone for 5 h had profound inhibitory effects on nascent protein synthesis in treated primary hepatocytes when compared to untreated primary hepatocytes (Figure 5).
Figure 1: An illustration shows the steps of primary mouse hepatocytes isolation, western blot analysis, and chemo-selective ligation reaction procedures.
Figure 2: Phase contrast morphological images of primary mouse hepatocytes. Representative pictures of isolated primary hepatocytes were acquired at 2, 12, 24, and 72 h after plating. Scale bars = 100 μm.
Figure 3: Western blot analysis for albumin protein expression in primary mouse hepatocytes. Western blot analysis was conducted with the loading of 10 μg protein/lane from primary mouse hepatocytes (PMHs), mouse embryonic fibroblasts (MEFs), Hepa 1-6 cells, and mouse livers. One-way ANOVA showed a significant albumin protein expression in PMHs compared to MEF or Hepa 1-6 cells. Values are means ± SEM of group size (n = 3). *P, #P <0.001 vs. Hepa 1-6 or MEFs, respectively. $P <0.001 vs. liver lysate.
Figure 4: Induction of phosphorylation of AMPK in rotenone-treated primary mouse hepatocytes. Primary hepatocytes from a wild-type mouse liver were treated with 10 μM rotenone for 0, 4 or 6 h. Phosphorylation levels of AMPK were detected by western blotting and β-actin was used as a loading control (10 μg protein was loaded/lane). Repeated measurements 2-way ANOVA showed that treatment caused a significant increase in protein expression of phospho-AMPK in comparison to untreated primary hepatocytes. Values are means ± SEM of group size (n = 3). *P <0.001 vs. Untreated hepatocytes.
Figure 5: Reduction of hepatic nascent protein expression after rotenone treatment. Primary hepatocytes from a wild-type mouse liver were treated with 10 μM rotenone for 5 h, followed by a nascent protein synthesis assay with non-radioactive AHA method. Protein synthesis levels were assessed by a laser scanner detecting the TAMRA fluorescence. Total protein loading levels for this assay were detected by the coomassie-dye reagent.
Isolation of primary mouse hepatocytes | Chemoselective ligation reaction assay for detection of nascent protein synthesis |
1. HBSS (-) Buffer 500 ml: | 1. Protein analysis detection Kit, TAMRA alkyne |
1.1. HEPES: 1.1915 g | 1.1. Component A (TAMRA alkyne) |
1.2. D-glucose: 0.9 g | 1.2. Component B (reaction buffer) |
1.3. EGTA: 0.095 g | 1.3. Component C (CuSO4) |
1.4. Anti-Anti: 5 ml | 1.4. Component D (reaction buffer additive 1) |
1.5. HBSS (10 X): 50 ml | 1.5. Component E (reaction buffer additive 2) |
1.6. Add distilled deionized water (DDW) up to 500 ml | 2. AHA: L-azidohomoalanine |
(Note: Adjust pH to 7.4, Filter, and Store at 4 oC) | 3. Methionine-free DMEM medium |
2. HBSS (+) Buffer 500 ml: | 3.1. DMEM: 500 mL |
2.1. HEPES: 1.1915 g | 3.2. FBS: 56 mL |
2.2. CaCl2.2H2O: 0.3675 g | 3.3. Anti-Anti: 5.6 mL |
2.3. Anti-Anti: 5 ml | 3.4. L-Cystine Dihydrochloride |
2.4. HBSS (10 X): 50 ml | 3.5.L-alanyl-L-glutamine dipeptide: 5.6 mL |
2.5. Add DDW up to 500 ml | (Note: Store at 4 oC) |
(Note: Adjust pH to 7.4, Filter, and Store at 4 oC) | 4. Warm PBS |
3. 40% density gradient buffer 450 ml: | 5. Lysis buffer |
3.1. density gradient buffer: 180 ml | 5.1. 50 mM Tris-HCl |
3.2. HBSS (10 X): 45 ml | 5.2. 1% SDS |
3.3. DDW: 225 ml | 5.3. Protease inhibitor |
(Note: Filter and Store at 4 oC) | 5.4. Phosphatase inhibitor |
4. DMEM: | |
4.1. DMEM: 500 ml | |
4.2. FBS: 56 ml | |
4.3. Anti-Anti: 5.6 ml | |
(Note: Store at 4 oC) | |
5.Williams’ Medium E: | |
5.1. Williams’ Medium E: 500 ml | |
5.2. FBS: 30 ml | |
5.3. Anti-Anti: 5.4 ml | |
5.4. L-alanyl-L-glutamine dipeptide: 5.4 ml | |
(Note: Store at 4 oC) |
Table 1
Although several immortalized hepatic cell lines have been proposed and used to investigate liver functions49,50,51,52, these cells generally lack the important and fundamental functions of normal hepatocytes, such as the expression of albumin (Figure 3). It is widely recognized, therefore, that utilizing primary hepatocytes is a valuable option for examining liver physiology and metabolism in culture, despite the challenges of culturing and maintaining of these cells. Herein, our study details the successful isolation of primary mouse hepatocytes by conducting a two-step collagenase method in a relatively convenient and efficient manner. Through these procedures, the majority of isolated primary hepatocytes within the culture retain their metabolic functionality and morphological stability and can be used for investigating hepatic functions including protein metabolism, gluconeogenesis, and mitochondrial respiration.
Isolated primary hepatocytes are a powerful experimental tool to examine molecular mechanisms of hepatic energy metabolism. Rotenone interferes with the electron transport chain complex I activity in mitochondria and blocks oxidative phosphorylation with the limited synthesis of ATP53, as evidenced by the activation of AMPK (Figure 4). The AMPK activation triggers the cellular adaptive responses that suppress the anabolic mechanism to compensate the rotenone-induced energy-deficient stress. Under this condition, we successfully detected the drastic reduction of nascent protein synthesis levels in primary hepatocytes using a non-radioactive, chemo-selective ligation reaction with a TAMRA protein detection method (Figure 5). These results confirm the importance of energy supply from the mitochondria for appropriate protein synthesis in primary hepatocytes. These techniques are useful to provide the information regarding a drug action or toxicity within therapeutically relevant concentrations in the context of hepatic protein synthesis of healthy and diseased conditions. In addition, these techniques are applicable for examining the role of specific pathways modulating protein and energy metabolism by analyzing primary hepatocytes isolated from genetically manipulated mice.
In general, low yield and cell viability are the major drawbacks of the use of primary hepatocytes. These issues are due to several reasons such as initial cannulation, insufficient washing with HBSS, inappropriate concentration of collagenase, or long-perfusion period. Our current protocol optimizes these multiple steps and leads to gaining a higher yield and viable cells around 20 million/liver/adult mice. Especially, there are two critical steps: one is to adjust the temperature and the flow rate of digested buffers with proper collagenase concentration to digest the hepatic parenchyma efficiently in 15–20 min from cannula insertion until liver excision, and another is to process these collagenases treated samples quickly as described in order to keep cells in healthy condition. While isolated hepatocytes-based studies have been conducted within 48 h time frame in many published data, primary hepatocytes isolated with our procedures endure the longer-term culture (Figure 2). With these cells, we show the robust labeling of hepatic nascent protein synthesis through the chemo-selective ligation reaction assay, although the concentration and incubation time with AHA in the protocol may have to be re-optimized in different biological conditions of cells (e.g. physiological vs. pathological, wild-type vs. transgenic or knockout/down cells).
In summary, we have established procedures to analyze primary mouse hepatocytes for nascent protein synthesis in a non-radioactive manner. These cells respond well to mitochondrial stress and exhibit suppression of protein synthesis with activation of the AMPK pathway. Thus, these isolated hepatocytes are valuable models for investigating the pathophysiology of hepatic protein and energy metabolism ex vivo.
The authors have nothing to disclose.
We thank Drs. Joonbae Seo and Vivian Hwa for their scientific input and discussion. This work was supported by National Institute of Health (NIH) (R01DK107530). T.N. was supported by the PRESTO from the Japan Science and Technology Agency. A part of this study was supported by a grant from NIH (P30DK078392) for the Digestive Disease Research Core Center in Cincinnati.
HEPES buffer | Fisher Scientific | BP310-500 | |
D-glucose | Fisher Scientific | D16-500 | |
Ethylene glycol-bis(β-aminoethyl ether)-tetraacetic acid | AmericanBio | AB00505-00025 | |
Antibiotic-Antimycotic (100X) | Gibco | 15240-062 | |
HBSS (10X) no calcium, magnesium, phenol red | Gibco | 14185-052 | |
Calcium Chloride Dihydrate (CaCl2.2H2O) | Fisher Scientific | C79-500 | |
Density gradient buffer | GE Healthcare | 17-0891-02 | |
DMEM (Dulbecco's Modified Eagle Medium) low glucose, pyruvate | Gibco | 11885-084 | |
Fetal Bovine Serum | Hyclone | SH30910.03 | |
Phosphate Buffered Saline (PBS) (1X) | Gibco | 1897141 | |
Williams medium E, no glutamine | Gibco | 12551-032 | |
L-alanyl-L-glutamine dipeptide supplement | Gibco | 35050-061 | |
Collagenase Type X | Wako Pure Chemical Industries | 039-17864 | |
Perfusion pump | Cole-Parmer | Masterflex L/S | Equipment |
IV administration set | EXELINT | 29081 | Equipment |
A water bath | REVSCI | RS-PB-200 | Equipment |
Tube heater | Fisher Scientific | Isotemp | Equipment |
Ethanol | Decon Lab, Inc | 0-39613 | |
Isoflurane | PHOENIX | 10250 | |
Autoclaved Cotton Tips | Fisherbrand | 23-400-124 | |
100 mm Petri Dish | TPP | 93100 | |
Connector (Male Luer Lock Ring) | Cole-Parmer instrument | EW-4551807 | |
24G catheters | TERUMO | Surflo 24Gx3/4' | |
100 μm Filter (CELL STRAINERS) | VWR | 10199-658 | |
15 ml conical-bottom centrifuge tubes | VWR | 89039-666 | |
50 ml conical-bottom centrifuge tubes | VWR | 89039-658 | |
Chemoselective ligation reaction PROTEIN ANALYSIS DETECTION KIT, TAMRA ALKYNE | Invitrogen | C33370 | |
AHA (L-azidohomoalanine) | Invitrogen | C10102 | |
DMEM (methionine free) | Gibco | 21013024 | |
L-Cystine Dihydrochloride | SIGMA | C2526 | |
Laemmli sample buffer | BioRad | 161-0737 | |
Protease Inhibitor Cocktail | SIGMA | P9599 | |
SDS solution (20%) | BioRad | 161-0418 | |
Tris-HCL (1M) | American Bioanalytical | AB14044-01000 | |
Phosphatase Inhibitor Cocktail | SIGMA | P5726 | |
Protein concentration measuring Kit (Bovin Serum Albumin-BSA) | BioRad | 500-0207 | |
6-well tissue culture plate | TPP | 92006 | |
Digital Heatblock | VWR | 12621-092 | Equipment |
Multi-Rotator | Grant-bio | PTR-60 | Equipment |
Ultrasonic Sonicator | Cole-Parmer | GE130PB | Equipment |
Standard Heavy-Duty Vortex Mixer | VWR | 97043-566 | Equipment |
A variable mode laser scanner | GE Healthcare Life Science | FLA 9500 | Equipment |
Coomassie-dye reagent | Thermo Scientific | 24594 | |
Inverted microscope | Olympus | CKX53 | Equipment |
Western Blotting apparatus | BioRad | 1658004 | Equipment |
Centrifuge | Eppendorf | 5424R | Equipment |
Automated cell counter | BioRad | TC20 | Equipment |
FluorChem R system | proteinsimple | – | Equipment |
p-Ampka (T172) antibody | Cell signaling | 2535 | |
Total-AMPK antibody | Cell signaling | 5832 | |
Albumin antibody | Cell signaling | 4929 | |
beta actin antibody | Santa Cruz | sc-130656 | |
Fine scissors and forceps |