Summary
Primary hepatocytes are a valuable tool to study liver response and metabolism in vitro. Utilizing commercially available reagents, an improved time- and labor-efficient protocol for mouse primary hepatocyte isolation was developed.
Abstract
Primary hepatocytes are used extensively in liver in vitro research, especially in glucose metabolism studies. A base technique has been adapted based on different needs, like time, labor, cost, and primary hepatocyte usage, resulting in various primary hepatocyte isolation protocols. However, the numerous steps and time-consuming reagent preparations in primary hepatocyte isolation are major drawbacks for efficiency. After comparing different protocols for their pros and cons, the advantages of each were combined, and a rapid and efficient primary hepatocyte isolation protocol was formulated. Within only ~35 min, this protocol could yield as much, if not more, healthy primary hepatocytes as other protocols. Further, glucose metabolism experiments performed using the isolated primary hepatocytes validated the usefulness of this protocol in in vitro liver metabolism studies. We also extensively reviewed and analyzed the significance and purpose of each step in this study so that future researchers can further optimize this protocol based on needs.
Introduction
The liver serves as one of the most important organs in the vertebrate body due to the vital role it plays in numerous life-supporting functions like food digestion, blood circulation, and detoxification. Usage of mouse primary hepatocyte in vitro culture is increasingly popular in studies of carbohydrate metabolism and hepatic carcinoma. Therefore, it is important to develop a convenient method for mouse primary hepatocyte isolation while maintaining its innate physiological function. Due to its function as a hub of glucose metabolism, the liver is also central to glucose production and storage1. Experiments with primary hepatocytes in vitro are a must to most glucose metabolism studies. Therefore, for years, various research groups have developed protocols for mouse primary hepatocyte isolation.
The general procedure of mouse hepatocyte isolation is to first flush out blood in the liver with an isosmotic liquid such as phosphate-buffered saline (PBS) or Hanks' Balanced Salt Solution (HBSS) and then use collagenase-containing solution to dissociate hepatocytes. These protocols share a general procedure but differ in reagents and steps based on different needs. However, preparing required reagents and performing isolation steps take time. In developing the present protocol, efficiency was set as a priority, with all reagents ready-to-use and available from the market, and as few steps as possible. The overall goal of this protocol is to provide a fast and labor-efficient method to isolate primary hepatocytes from mouse, without jeopardizing the isolated primary hepatocyte purity and viability.
Subscription Required. Please recommend JoVE to your librarian.
Protocol
All procedures were approved by the Johns Hopkins Animal Care and Use Committee. C57BL/6 female mice (8-week old) were used in this study.
1. Preparation:
- Mix William's E Medium (GlutaMAX Supplement) with 10% FBS and 1% antibiotic-antimycotic solution to make the culture medium.
- Filter 25 mL of collagenase-dispase medium (e.g., Liver Digest Medium) through a 0.45 µm syringe filter to remove particle debris.
- Warm 50 mL of double-distilled H2O (ddH2O), 35 mL of perfusion Medium (e.g., Liver Perfusion Medium) (or 50 mL at the first time using this protocol), and 25 mL of filtered collagenase-dispase medium in a 45 °C water bath for 30 min.
- Within a sterile tissue culture hood, mix 2 mL of 10x HBSS and 18 mL of Percoll in a 50 mL tube to make 20 mL 1x Percoll-HBSS and keep on ice or 4 °C.
NOTE: 1x Percoll-HBSS can be kept at 4 °C for at least 6 months. - Within a sterile tissue culture hood, pour 30 mL of wash medium (e.g., Hepatocyte Wash Medium) into a clean Petri dish, and keep on ice.
- Submerge the pumping tube in the water of a 45 °C water bath. Results are most reliable if the room temperature is at 25 °C.
- Prepare 2 mL of 1x anesthetics by mixing 225 µL of Ketamine HCL, 93.75 µL of Xylazine, and 1681 µL of 1x PBS.
- Anesthetize one mouse using an approved method. Here the mouse was intraperitoneally injected with 150 µL of 1x anesthetics. Perform the tests for loss of reflexes such as reaction to toe pinching to ensure full anesthesia.
- Secure the mouse on its back onto the dissection pad by four limbs, by either pining or using water-proof tape or other methods approved by the institution's Animal Care and Use Committee (or equivalents).
- Prepare sterilized forceps and scissors for dissection. To avoid possible contamination, conduct all steps within a sterile hood.
2. Procedure:
- Using a peristaltic pump, start pumping warmed-up ddH2O at a speed of 4 mL/min for 5 min. Change the pumping tube from water to a warmed-up perfusion medium.
- Disinfect the anesthetized mouse's abdomen with 70% EtOH and cut open with a scissor to expose the liver, portal vein, and inferior vena cava (IVC).
- Stop the peristaltic pump. Insert a 24 G catheter (e.g., Closed IV Catheter, 24 G, 0.75 IN) into IVC. Start pumping and cut the portal vein open.
- Continue pumping until the flushed-out liquid is clear (around 3-5 min). Press the portal vein every minute to let liquid reach every corner of the liver. Be careful not to let air bubbles enter the IVC.
NOTE: This step is to flush out as much blood as possible from the liver. - Change the pumping tube from the perfusion medium to the collagenase-dispase medium. Continue pumping until all 25 mL of the collagenase-dispase medium is depleted while doing step 2.6.
- Press the portal vein every minute to let liquid reach every corner of the liver.
NOTE: At this stage, the complete loss of blood is fatal to the mouse. The death of the mouse can be confirmed by a lack of heartbeat after the experiment. Dispose of the carcass as per facility policies. - Isolate the whole liver out without gallbladder to the 30 mL wash medium in the Petri dish on ice.
- Tear it up into pieces with forceps to release primary hepatocytes into solution. This step would turn the wash medium into a cloudy solution full of released primary hepatocytes and small liver pieces.
- Filter the cloudy solution in step 2.8 through a 70 µm cell strainer into the 20 mL 1x Percoll-HBSS in a 50 mL tube on ice. Mix by inverting the tube 20 times.
- Centrifuge at 300 x g for 10 min at 4 °C.
- Within the tissue culture hood, aspirate the supernatant. Wash the pellet with cold 30 mL of wash medium.
- Centrifuge at 50 x g for 5 min at 4 °C.
- Remove the supernatant and resuspend the pellet in 25 mL of the culture medium (or appropriate other volumes) within the tissue culture hood.
- Count the cell number and plate the cells on desired culture plates according to the experimental design.
NOTE: Primary hepatocytes properly isolated from one 8-week-old mouse are usually sufficient to be plated on four 6-well plates or four 12-well plates.
Subscription Required. Please recommend JoVE to your librarian.
Representative Results
In order to test the efficiency, the present primary hepatocyte isolation protocol was performed on 8-week old female C57BL/6 mice. The attachment and purity of isolated primary hepatocytes were tested. Primary hepatocyte isolation is used for a wide variety of experiments on liver physiology, such as hepatic drug effects and glucose metabolism, pharmaceutical biomarker activity2, insulin sensitivity, and glucose production. Therefore, the activities of the primary hepatocytes isolated with this protocol were tested in the following experiments.
Coating was not required for primary hepatocyte attachment in the present protocol
Primary hepatocytes isolated were plated at 5 x 105 cells/well on a 6-well plate. After 1 h, a firm attachment was observed, and primary hepatocytes fully expanded 12 h after plating (Figure 3). This indicated that 12 h after plating, cells were ready to be utilized for experiments. Based on the nucleus morphology of mouse hepatocytes within the liver, mononuclear hepatocytes were enriched at borders, while binuclear/polynuclear hepatocytes, a signature of terminal differentiation, were in the middle3,4. A significant amount of cells imaged displayed typical dual-nucleus (diploid) morphology, indicating the success of isolating and purifying live primary hepatocytes. This also indicates that collagen coating is not a requirement for primary hepatocyte attachment with this protocol.
Purity was enriched in the isolated primary hepatocyte population
Various cell type-specific gene markers have been used in previous protocols to check isolated primary hepatocyte purity (Table 1). TTR (Transthyretin), CD95 (Cluster of differentiation 95, also known as Fas), ASGR1 (Asialoglycoprotein receptor 1), and ASGR2 are markers for hepatocytes. After isolation, the mRNA levels of these hepatocyte markers were significantly increased, compared to the whole liver (Figure 4A-D,H).
This protocol also greatly reduced the interference from other hepatic cells. The presence of immune cells, stellate cells, and endothelial cells were lower, shown by the sharp decrease of mRNA levels of CD45 (immune cell marker), COL1A1 (Collagen, type I, alpha 1, stellate cell marker), and TIE2 (Tunica interna endothelial cell kinase, endothelial cell marker), compared to the whole liver (Figure 4E-H). These suggest that this protocol could purify the primary hepatocyte population from hepatic cells and thus reduce the possible interference from other cell types in experiments.
Activity of pharmaceutical biomarkers was preserved
Biomarkers on hepatocytes have been extensively used for drug targeting and delivery. The activity preservation of pharmaceutical biomarkers thus is a key point in primary hepatocyte isolation and is a standard to test the usefulness of primary hepatocyte isolation2. Hepatocyte markers ASGR1 and ASGR2 are used in this manner2. We first tested the time-course expression level of these two markers before and after primary hepatocyte plating. After plating, their mRNA level decreased considerably with time, but levels remained considerable compared to the whole liver until the 12 h time point after plating, especially for ASGR1 (Figure 5A,B). The expression trend was consistent with a previous report2 and indicated comparable primary hepatocyte healthiness. Various pathogens target CD81, a hepatocyte membrane-bound protein, to facilitate their entrance into cells and infection, like hepatitis virus5, Plasmodium falciparum, and Plasmodium yoelii6. Other hepatocyte membrane-located proteins, like TLR4 (Toll-like receptor 4), are also targeted by pathogens and important for hepatocyte immune response7. After plating, the expression level of CD81 was consistent until 48 h (Figure 5C). TLR4 expression level generally increased, but not until after 48 h, when it reached a level higher than in vivo (Figure 5D). These suggest that primary hepatocytes isolated by this protocol can also be used for CD81 and TLR4 studies within at least 48 h after plating. Together, these results indicate that primary hepatocytes isolated are valid for use in studies related to pharmaceutical biomarkers. It is worth noting that RNA and protein levels may be inconsistent because of influences from post-transcriptional activities like signal peptide-induced RNA migration, posttranslational modification and/or protein degradation. Therefore, protein level and bioactivity verification of pharmaceutical biomarkers identified by mRNA may be necessary if required by the experimental paradigm.
Isolated primary hepatocytes were insulin-sensitive
Primary hepatocyte performance in experiments of glucose metabolism was also analyzed. Insulin, a hormone playing a central role in glucose metabolism, decreases glucose level, promoting hepatic glucose uptake and storage through phosphorylating AKT and FOXO1 (Forkhead box O1). Therefore, an insulin sensitivity assay was carried out with isolated primary hepatocytes. After 16 h, cells were starved for 3 h, with a serum-free medium. At the last 0.5 h of starvation, 100 nM insulin was administrated to the culture medium. As shown in Figure 6A-C, insulin significantly promoted the phosphorylation of both AKT at Ser473 and FOXO1 at Ser256, indicating the sensitivity of primary hepatocytes to insulin. This suggests that the isolated primary hepatocytes from the present protocol are useful in insulin/glucose metabolism studies.
Isolated primary hepatocytes were capable of glucose production
Not only are they a center for glucose storage, but hepatocytes are also responsible for glucose production. To test whether the primary hepatocytes we isolated are useful in studies of glucose production, the cells were starved for 10 h in the presence of glucagon to stimulate glucose production. The starvation medium was then collected for glucose assay, while cells were harvested for western blot. Phosphoenolpyruvate carboxykinase (PEPCK) is an essential component in liver glucose production, controlling its rate8. The protein level of PEPCK was significantly increased after glucagon treatment, suggesting that the glucose production pathway was activated (Figure 7A,B). This activation was further confirmed by an increased level of glucose production (Figure 7C). This phenomenon was also confirmed with other glucose production stimulators like forskolin plus IBMX (Figure 7A-C). However, due to the limitation of this experiment, we could not verify whether glucose production was exclusive via gluconeogenesis or whether there is a component of glycogenolysis as well.
Figure 1: Bench setup. (A) The bench setup for primary hepatocyte isolation. (B) The cartooned bench setup for primary hepatocyte isolation. Please click here to view a larger version of this figure.
Figure 2: Mouse dissection, perfusion, and primary hepatocyte purification. (A) The dissected mouse, with arrows pointing to the liver, IVC, and portal vein. (B) Catheter insertion into IVC. (C) Pressuring portal vein causing enlargement and stiffness of liver lopes, indicating successful perfusion. (D) Softened liver lopes after perfusion, indicating the success of collagenase digestion. (E) The position of the gallbladder in the isolated liver (arrow pointing to the gallbladder). (F) Gallbladder removal. (G) Teared-up liver in hepatocyte wash medium. (H) Mixing 1x Percoll-HBSS with filtered primary hepatocyte before centrifuge. (I, J) primary hepatocyte pellet after centrifuge. (K) Primary hepatocyte resuspension within hepatocyte wash medium. (L, M) Primary hepatocyte pellet formed within hepatocyte wash medium after centrifuge. Please click here to view a larger version of this figure.
Figure 3: Primary hepatocytes after plating. Images were taken after (A)1 h, (B, C) 12 h, (D) 24 h, (E) 36 h, (F) 48 h, (G) 72 h, and (H) 96 h primary hepatocyte plating. Scale bar = 100 µm. Please click here to view a larger version of this figure.
Figure 4: Enhanced primary hepatocyte purity after isolation. RNA was isolated before (from the whole liver) and after (from isolated primary hepatocytes) primary hepatocyte isolation, followed by reverse transcription PCR, according to a previous protocol9. The primary hepatocyte purity was assessed by real-time PCR with primers for hepatocyte markers (A) TTR , (B) CD95, (C) ASGR1 and (D) ASGR2, while also with immune cell marker (E) CD45, (F) Stellate cell marker COL1A1, and (G) endothelial cell marker TIE2. (H) Heatmap was generated for the expression changes of cell type markers before and after primary hepatocyte isolation. GAPDH (Glyceraldehyde 3-phosphate dehydrogenase) was used as an internal control. Primers used here were for genes (Primer sequences are in Table 2. Sequence references are cited here): TTR10; CD9511; ASGR112; ASGR213; CD4514; COL1A115; TIE216; GAPDH17. Graphs and heatmap were generated with GraphPad Prism 8. Error bar indicates Standard Deviation, and two-tailed unpaired (since primary hepatocyte and whole liver samples were unpaired because this protocol requires intact liver to begin with) t-test significance is indicated by asterisks (* p < 0.05; ** p < 0.01; *** p < 0.001). N=7. Please click here to view a larger version of this figure.
Figure 5: Expression of pharmaceutical biomarkers. RNA was isolated from the whole liver and primary hepatocyte after plating, followed by reverse transcription PCR. The expression of pharmaceutical biomarkers was assessed by real-time PCR with primers for (A) ASGR1, (B) ASGR2, (C) CD81, and (D) TLR4. GAPDH was used as an internal control. New primers used here were for genes (Primer sequences are in Table 2. Sequence references are cited here): CD8118; TLR419. Graphs were generated with GraphPad Prism 8. N = 5. Please click here to view a larger version of this figure.
Figure 6: Insulin sensitivity preserved after primary hepatocyte isolation. (A) Western blot of p-AKT (S473), AKT, p-FOXO1 (S256), FOXO1, and GAPDH after insulin treatment. (B) p-AKT (S473)/AKT in densitometry of western blot. (C) p-FOXO1 (S256)/FOXO1 in densitometry of western blot. Insulin sensitivity assay was carried out 16 h after primary hepatocyte plating. Primary hepatocytes were starved for 3 h in William's E Medium (GlutaMAX Supplement) without FBS but with 1% antibiotic-antimycotic solution and with 100 nM insulin treatment at the last 30 min, before being harvested for protein lysis with 1x RIPA buffer. Graphs were generated with GraphPad Prism 8. Western blot imaging was carried out with LI-COR Odyssey CLx. Error bar indicates Standard Deviation, and two-tailed paired t-test significance is indicated by asterisks (* p < 0.05; ** p < 0.01). N = 4. Please click here to view a larger version of this figure.
Figure 7: Glucose production assay. (A) Western blot of PEPCK and GAPDH after glucagon or forskolin+IBMX treatment for 10 h. (B) Densitometry of PEPCK protein level comparing to GAPDH after either glucagon or forskolin+IBMX treatment for 10 h. (C) Glucose level comparing to protein level after either glucagon or forskolin+IBMX treatment for 10 h. Glucose production assay was carried out 16 h after primary hepatocyte plating. Primary hepatocytes were starved for 10 h in glucose- and phenol red-free DMEM (added with 2 mM L-Glutamine, 2 mM Sodium Pyruvate, 20 mM Sodium L-Lactate, 1% Pen Strep), with either 50nM glucagon or 20 µM forskolin and 200 µM IBMX. The medium was harvested for glucose concentration measurement with Glucose Assay Kit, while protein was harvested for western blot. Western blot imaging was carried out with LI-COR Odyssey CLx. Error bar indicates Standard Deviation, and two-tailed paired t-test significance is indicated by asterisks (* p < 0.05; ** p < 0.01; *** p < 0.001). N = 4. Please click here to view a larger version of this figure.
| Centrifugation Times | Rib Cage Removal | Perfusion Buffers self-making | Surgical Knot | Lamp Heating | Plate Coating | Isolated cell quantity/mouse | Gradient Centrifugation | Purity Marker Genes | |
| Present protocol | 2 | No | No | No | No | No | 2.5–4 x 107 | Yes | TTR, CD95, ASGR1, ASGR2, CD45, COL1A1, TIE2 |
| Severgnini et al.2. | 4 | Yes | Yes | Yes | Yes | Yes | 1.8–2 x 107 | No | TTR, CD45, COL1A1, TIE2 |
| Gonçalves et al.24. | 5 or 6 | No | No | No | No | Yes | 1–3 x 106 | Yes | CD45, CD95 |
| Li et al.20. | 4 or 5 | No | Yes | Yes | No | Not Mentioned | 1–4 x 107 | No | N/A |
| Salem et al.21. | 3 | No | Yes | No | No | Not Mentioned | 2 x 107 | Yes | N/A |
| Cabral et al.22. | 3 or 6 (with gradient Centrifugation) | No | Yes | Yes | No | Yes/Recommended | N/A | Optional | N/A (Purity assessed by light microscopy) |
| Korelova et al.23. | 3 | No | Yes | Yes | No | Yes | N/A | Yes | N/A |
Table 1: Comparison of primary hepatocyte protocols.
| Primer (Forward) | Primer (Reverse) | Reference | ||
| TTR Forward 5’-3’: AGCCCTTTGCCTCTGGGAAGAC | TTR Reverse 5’-3’: TGCGATGGTGTAGTGGCGATGG | 10 | ||
| CD95 Forward 5’-3’: ATGCACACTCTGCGATGAAG | CD95 Reverse 5’-3’: CAGTGTTCACAGCCAGGAGA | 11 | ||
| ASGR1 Forward 5’-3’: GAGTCGAAGCTGGAAAAACAG | ASGR1 Reverse 5’-3’: CCTTCATACTCCACCCAGTTG | 12 | ||
| ASGR2 Forward 5’-3’: CTACTGGTTTTCTCGGGATGG | ASGR2 Reverse 5’-3’: CAAATATGAAACTGGCTCCTGTG | 13 | ||
| CD45 Forward 5’-3’: GAACATGCTGCCAATGGTTCT | CD45 Reverse 5’-3’: TGTCCCACATGACTCCTTTCC | 14 | ||
| COL1A1 Forward 5’-3’: GAAGCACGTCTGGTTTGGA | COL1A1 Reverse 5’-3’: ACTCGAACGGGAATCCATC | 15 | ||
| TIE2 Forward 5’-3’: ATGTGGAAGTCGAGAGGCGAT | TIE2 Reverse 5’-3’: CGAATAGCCATCCACTATTGTCC | 16 | ||
| GAPDH Forward 5’-3’: CGACTTCAACAGCAACTCCCACTCTTCC | GAPDH Reverse 5’-3’: TGGGTGGTCCAGGGTTTCTTACTCCTT | 17 | ||
| CD81 Forward 5’-3’: CCAAGGCTGTGGTGAAGACTTTC | CD81 Reverse 5’-3’: GGCTGTTCCTCAGTATGGTGGTAG | 18 | ||
| TLR4 Forward 5’-3’: ACCTGGCTGGTTTACACGTC | TLR4 Reverse 5’-3’: CTGCCAGAGACATTGCAGAA | 19 | ||
Table 2: List of primers: Primers used for genes TTR, CD95, ASGR1, ASGR2, CD45, COL1A1, TIE2, GAPDH, CD81, and TLR4
Subscription Required. Please recommend JoVE to your librarian.
Discussion
Various primary hepatocyte isolation protocols have been developed. They also have been kept optimized and adapted based on different needs (Table 1). Isolation protocols are generally composed of two parts: perfusion (including enzyme digestion) and purification.
The perfusion can be performed with the entire liver in vivo2,20,21,22,23 or with dissected liver lobes24. The perfusion of dissected lobes was generally restricted to the left and medium lobes for technical ease. Theoretically, in vivo perfusion should yield more primary hepatocytes due to the usage of the entire liver with all lobes. Keeping the liver in place during perfusion may also reduce cell death since hepatocytes in vivo can be bathed in fluid, either blood or perfusion buffers, at all times during this step.
The maintenance of sterile conditions also plays an important role in this step. If the condition permits, it is better to conduct perfusion in a clean hood. All the tools with direct touch to tissues should be sterile, which can be achieved by autoclave. In order to prevent any contamination from bacteria and fungi, an antibiotic-antimycotic solution was added into the culture medium.
The IVC and portal vein are two main vessels connecting the liver with other organs. Most protocols utilize either of these veins to insert needles for perfusion: IVC2,20,21,23, portal vein22. The position of the portal vein is at a skewed angle, which leads to difficulty in positioning and stabilizing the inserted needle. A suture is thus usually needed to tie the needle in place with a surgical knot, which must be done very carefully as the vein is prone to damage. Generally, the diameter of the portal vein is also smaller than IVC, complicating needle insertion. Considering this, the use of IVC can be more convenient, although in some protocols, the needle is also surgically knotted to the IVC20. It has been reported that perfusion with IVC insertion resulted in less viable primary hepatocytes than portal vein25. However, in the present protocol, we successfully used IVC insertion, and the isolated primary hepatocyte viability (>96% in optimized conditions, measured with Trypan Blue staining according to manufacturer's protocol) was as high, if not more, as other protocols (varies between 80% and 96% based on protocols and conditions2,22,23,24).
There are two aims for the perfusion step: to flush blood out of liver lobes and digest the liver to release primary hepatocytes. In order to achieve these objectives, at least two buffers should be used, one for blood flux (Flux Buffer), and one for digestion (Digestion Buffer). Most of the protocols prepare HBSS-based Flux Buffer and add collagenase within to make it capable for digestion, as Digestion Buffer. The preparation of these buffers is time-consuming, and the buffers could vary slightly in some delicate properties, like pH, from batch to batch, therefore introducing uninvited variables. Considering this, utilizing commercially available buffers saves valuable labor and time while minimizing these variables. Gibco developed a protocol for primary hepatocyte isolation from adult rats26, based on the use of commercially available Liver Perfusion Medium (as Flux Buffer), and Liver Digestion Medium (as Digestion Buffer). Mouse and rat, as rodents, share high similarities in body properties; one may integrate these two rat-optimized buffers into mouse primary hepatocyte isolation. Indeed, Gonçalves et al.24, successfully carried out the perfusion of dissected liver lobes with these buffers, raising the promise of their use in liver perfusion in vivo. Here we successfully tested Liver Perfusion Medium and Liver Digestion Medium in the present mouse primary hepatocyte protocol, which yielded high-quality viable primary hepatocytes (Figure 3 and Figure 4).
The temperature of the perfusion buffer determines how well primary hepatocytes survive the isolation. If the temperature is too high, the collagenase within Digestion Buffer may have reduced activity; if too low, the primary hepatocytes may suffer cold shock, causing possible compromised isolation yield. The time that buffer takes to flow through perfusion tubing also determines the temperature of buffer when it reaches the liver. In previous protocols, 40 °C2 or 42 °C21 water bath was used. This temperature may be subject to change based on lab conditions, like the environment temperature, length of the perfusion tube, and the type of Peristaltic Pump. In the present protocol, after multiple times of testing, we optimized the water bath temperature to be 45 °C to warm up the buffers.
The purity of isolated primary hepatocytes plays an important role in subsequent experiments, as the higher the purity ratio, the less possible interference. Although the liver is mainly composed of hepatocytes, accounting for 60%-80% by mass27, various types of other cells are also present, like immune cells, stellate cells, and endothelial cells, which are important for hepatic immunological activities. Each of these cells could post a potential interference in later-on experiments28. For instance, stellate cells in the liver also respond to pathogen invasions and liver injuries, involving in scar formation29. In numbers, 5%-8% of the total hepatic cell population is contributed by stellate cells30. Normally, stellate cells are in quiescence, but this status can be broken when stresses are present. Therefore, stellate cell activities may be triggered by stresses introduced during isolation or subsequent treatment if some of these cells remain in the final isolated primary hepatocyte pool. Other types of cells also contribute to a considerable portion of the hepatic cell population, like liver sinusoidal endothelial cells (LSECs), accounting for 20% of the total hepatic cell population31. How to eliminate the presence of these cells has been a puzzling part in primary hepatocyte isolation process. Therefore, purification is a key part in primary hepatocyte isolation, which usually takes advantages of the weight and size differences between different types of cells. By optimizing the centrifuge speed and/or purification buffer, primary hepatocytes can be pelleted down to the bottom of the tube.
Separating live cells from dead cells is also critical in obtaining healthy primary hepatocytes and accurate cell counting. Usually, counting cell number is a must after purification since the results of numerous experiments vary as cell confluence changes. Gradient centrifuge reagents can be used in this step to fulfill this mission, like Percoll. 36%-40% of Percoll in centrifuge pellets down live cells and keeps dead cells in supernatant. We successfully optimized centrifugation speed and time to reduce the non-primary hepatocyte cell population in the present protocol. Cell type-specific markers were used to test the purity of isolated primary hepatocytes here, like other protocols. The hepatocyte markers used here were TTR2, CD9524,32, ASGR133, and ASGR234. Markers for other types of cells include CD45 (immune cell marker), COL1A1 (Stellate cell marker), TIE2 (Endothelial cell marker)2,35. With these markers, the primary hepatocytes isolated with the present protocol showed a high level of purity (Figure 4), comparable to the previous protocols2,24.
During perfusion, collagenase in Digestion Buffer loosens attachment between cells by breaking down collagen within the extracellular matrix. This leads to difficulty in cell attainment during plating. Most of the previous protocols require/recommend pre-coating of plates with collagen2,22 or gelatine24 for better attachment of primary hepatocytes. To our knowledge, while Salem et al.21 and Li et al.20 did not discuss this step, other protocols assessed in this study clearly stated/recommended the usage of pre-coated plates2,22,23,24. In the present protocol, we found that the plate coating was not required for plates of certain types. While we are not sure whether this was because different types of plates, especially if they are from different manufacturers, have different surface smoothness, and whether this varied smoothness, if ever exists, plays an important role in primary hepatocyte attachment, it is beneficial to note so that another step (plate-coating) could be skipped for efficacy. It is also important to note that this protocol generates a significant proportion of terminally differentiated, e.g., diploid hepatocytes, along with mononuclear hepatocytes, similar to in vivo hepatic sinusoid.
In this study, the advantages of previous primary hepatocyte isolation protocols were combined, and the process was simplified as much as possible, using commercially available reagents and eliminating unnecessary steps. The centrifugation steps were successfully reduced to two, which is, to our knowledge, the fewest in published protocols. To confirm the purity and bioactivity of isolated primary hepatocytes, the level of various hepatic cell markers was assessed, confirming that the present protocol could greatly enhance primary hepatocyte purity and reduce other hepatic cell populations, such as immune cells, stellate cells, and endothelial cells. The activity of pharmaceutical biomarkers like ASGR1, ASGR2, CD81, and TLR4 was well-preserved in primary hepatocytes isolated with present protocol, which also had confirmed insulin sensitivity and glucose production activity. The main limitation of this protocol is the expense since all reagents were purchased commercially for efficiency. We did not specifically verify that glycogenosis was intact in primary hepatocytes isolated using the present protocol, and this may need further research for related studies. This protocol has similar perfusion steps to previous ones, like Salem et al.21, Severgnini et al.2 and Li et al.20, and Korelova et al.23, using IVC insertion. Their perfusion and digestion buffers, which may require extra labor to prepare, may also work with the present protocol with little modification of enzyme digestion time. Therefore, combining reagents of prior protocols with steps of the present protocol may also be beneficial, both time- and economically friendly.
In summary, an improved time- and labor-efficient protocol for primary hepatocyte isolation from mouse liver was developed. This protocol utilizes commercially available reagents entirely and can be completed in ~35 min, from dissecting mouse to plating primary hepatocytes, thus providing a useful technique to primary hepatocyte-related studies.
Subscription Required. Please recommend JoVE to your librarian.
Disclosures
The authors have nothing to disclose.
Acknowledgments
This work was supported by the National Institutes of Health (Grant 5R01HD095512-02 to S.W.).
Materials
| Name | Company | Catalog Number | Comments |
| 1x PBS | Gibco | 10010023 | |
| 10x HBSS | Gibco | 14065-056 | |
| 12-well Plate | FALCON | 353043 | Coating not required |
| 6-well Plate | FALCON | 353046 | Coating not required |
| anti-AKT | Cell Signaling | 2920S | |
| Antibiotic Antimycotic Solution (100x), Stabilized | Sigma-Aldrich | A5955 | |
| anti-FOXO1 | Cell Signaling | 97635S | |
| anti-GAPDH | Cell Signaling | 2118S | |
| anti-p-AKT (S473) | Cell Signaling | 9271L | |
| anti-PEPCK | Santa Cruz | SC-166778 | |
| anti-p-FOXO1 (S256) | Cell Signaling | 84192S | |
| Cell Strainer, 70 µm | CELLTREAT | 229483 | |
| Closed IV Catheter, 24 Gauge 0.75 IN | Becton Dickinson | 383511 | |
| DMEM, no glucose, no glutamine, no phenol red | ThermoFisher Scientific | A1443001 | |
| EnzyChrom Glucose Assay Kit | BioAssay Systems | EBGL-100 | |
| Fetal Bovine Serum (FBS) | Hyclone | SH30071.03 | |
| Forskolin | MilliporeSigma | F3917-10MG | |
| Glucagon | Sigma-Aldrich | G2044 | |
| Goat Anti-mouse IgG Secondary Antibody | LI-COR | 926-68070 | |
| Goat Anti-rabbit IgG Secondary Antibody | LI-COR | 926-32211 | |
| GraphPad Prism 8 | GraphPad Software | NA | |
| Hepatocyte Wash Medium | Gibco | 17704-024 | |
| IBMX | Cell Signaling | 13630S | |
| Insulin | Lilly | NDC 0002-8215-01 | |
| Ketamine HCL (100 mg/mL) | Hospira Inc | NDC 0409-2051-05 | |
| L-Glutamine | Gibco | 25030081 | |
| Liver Digest Medium | Gibco | 17703-034 | Aliquot within tissue culture hood to 25 mL each in 50 mL tube, and keep in -20 °C freezer |
| Liver Perfusion Medium | Gibco | 17701-038 | |
| Pen Strep | Gibco | 15140122 | |
| Percoll | GE Healthcare | 17-0891-01 | |
| Peristaltic Pump | Gilson | Minipuls 2 | Capable of pumping at 4 mL/min |
| Petri Dish | Fisherbrand | 08-757-12 | |
| Refrigerated Centrifuge | Sorvall | Legend RT | Capable to centrifuge 50 mL tube at 4 °C |
| Sodium L-Lactate | Sigma-Aldrich | L7022 | |
| Sodium Pyruvate | Gibco | 11360070 | |
| Syringe Filter, PVDF 0.45 µm 30mm diameter | CELLTREAT | 229745 | |
| Syringe, 0.5 mL | Becton Dickinson | 329461 | |
| Syringe, 60 mL | Becton Dickinson | 309653 | |
| Trypan Blue Solution, 0.4% | Gibco | 15250061 | |
| Tube, 15 mL | Corning | 430052 | |
| Tube, 50 mL | Corning | 430290 | |
| Water Bath Tank | Corning | CLS6783 | Or any water bath tank capable of heating up to 45 °C |
| William’s E Medium (GlutaMAX Supplement) | Gibco | 32551020 | |
| Xylozine (100 mg/mL) | Vetone Anased LA | NDC13985-704-10 |
References
- Han, H. S., Kang, G., Kim, J. S., Choi, B. H., Koo, S. H. Regulation of glucose metabolism from a liver-centric perspective. Experimental and Molecular Medicine. 48, 218 (2016).
- Severgnini, M., et al. A rapid two-step method for isolation of functional primary mouse hepatocytes: cell characterization and asialoglycoprotein receptor based assay development. Cytotechnology. 64 (2), 187-195 (2012).
- Guidotti, J. E., et al. Liver cell polyploidization: a pivotal role for binuclear hepatocytes. Journal of Biological Chemistry. 278 (21), 19095-19101 (2003).
- Morales-Navarrete, H., et al. A versatile pipeline for the multi-scale digital reconstruction and quantitative analysis of 3D tissue architecture. Elife. 4, 11214 (2015).
- Bruening, J., et al. Hepatitis C virus enters liver cells using the CD81 receptor complex proteins calpain-5 and CBLB. PLoS Pathogens. 14 (7), 1007111 (2018).
- Silvie, O., et al. Hepatocyte CD81 is required for Plasmodium falciparum and Plasmodium yoelii sporozoite infectivity. Nature Medicine. 9 (1), 93-96 (2003).
- Crispe, I. N. Hepatocytes as immunological agents. Journal of Immunology. 196 (1), 17-21 (2016).
- Liu, N. C., et al. Loss of TR4 orphan nuclear receptor reduces phosphoenolpyruvate carboxykinase-mediated gluconeogenesis. Diabetes. 56 (12), 2901-2909 (2007).
- Wang, Z., et al. Gonadotrope androgen receptor mediates pituitary responsiveness to hormones and androgen-induced subfertility. JCI Insight. 5 (17), 127817 (2019).
- Santos, S. D., et al. CSF transthyretin neuroprotection in a mouse model of brain ischemia. Journal of Neurochemistry. 115 (6), 1434-1444 (2010).
- Pinhu, L., et al. Overexpression of Fas and FasL is associated with infectious complications and severity of experimental severe acute pancreatitis by promoting apoptosis of lymphocytes. Inflammation. 37 (4), 1202-1212 (2014).
- Mi, Y., Lin, A., Fiete, D., Steirer, L., Baenziger, J. U. Modulation of mannose and asialoglycoprotein receptor expression determines glycoprotein hormone half-life at critical points in the reproductive cycle. Journal of Biological Chemistry. 289 (17), 12157-12167 (2014).
- Tanowitz, M., et al. Asialoglycoprotein receptor 1 mediates productive uptake of N-acetylgalactosamine-conjugated and unconjugated phosphorothioate antisense oligonucleotides into liver hepatocytes. Nucleic Acids Research. 45 (21), 12388-12400 (2017).
- Manczak, M., et al. Neutralization of granulocyte macrophage colony-stimulating factor decreases amyloid beta 1-42 and suppresses microglial activity in a transgenic mouse model of Alzheimer's disease. Human Molecular Genetics. 18 (20), 3876-3893 (2009).
- Zhang, Y., et al. JAK1-dependent transphosphorylation of JAK2 limits the antifibrotic effects of selective JAK2 inhibitors on long-term treatment. Annals of the Rheumatic Diseases. 76 (8), 1467-1475 (2017).
- Shih, S. C., et al. Molecular profiling of angiogenesis markers. American Journal of Pathology. 161 (1), 35-41 (2002).
- Liu, G., et al. miR-147, a microRNA that is induced upon Toll-like receptor stimulation, regulates murine macrophage inflammatory responses. Proceedings of the National Academy of Sciences of the United States of America. 106 (37), 15819-15824 (2009).
- Ding, Q., et al. Mice expressing minimally humanized CD81 and occludin genes support Hepatitis C virus uptake in vivo. Journal of Virology. 91 (4), 01799 (2017).
- Renshaw, M., et al. Cutting edge: impaired Toll-like receptor expression and function in aging. Journal of Immunology. 169 (9), 4697-4701 (2002).
- Li, W. C., Ralphs, K. L., Tosh, D. Isolation and culture of adult mouse hepatocytes. Methods in Molecular Biology. 633, 185-196 (2010).
- Salem, E. S. B., et al. Isolation of primary mouse hepatocytes for nascent protein synthesis analysis by non-radioactive l-azidohomoalanine labeling method. Journal of Visualized Experiments: JoVE. (140), e58323 (2018).
- Cabral, F., et al. Purification of Hepatocytes and Sinusoidal Endothelial Cells from Mouse Liver Perfusion. Journal of Visualized Experiments: JoVE. (132), e56993 (2018).
- Korelova, K., Jirouskova, M., Sarnova, L., Gregor, M. Isolation and 3D collagen sandwich culture of primary mouse hepatocytes to study the role of cytoskeleton in bile canalicular formation in vitro. Journal of Visualized Experiments: JoVE. (154), e60507 (2019).
- Goncalves, L. A., Vigario, A. M., Penha-Goncalves, C. Improved isolation of murine hepatocytes for in vitro malaria liver stage studies. Malar Journal. 6, 169 (2007).
- Yin, Z., Ellis, E. C., Nowak, G. Isolation of mouse hepatocytes for transplantation: a comparison between antegrade and retrograde liver perfusion. Cell Transplant. 16 (8), 859-865 (2007).
- Hepatocyte media, Thermofisher. , Available from: https://assets.thermofisher.com/TFS-Assets/LSG/manuals/Hepatocyte_Media_UG.pdf (2021).
- Sia, D., Villanueva, A., Friedman, S. L., Llovet, J. M. Liver cancer cell of origin, molecular class, and effects on patient prognosis. Gastroenterology. 152 (4), 745-761 (2017).
- Freitas-Lopes, M. A., Mafra, K., David, B. A., Carvalho-Gontijo, R., Menezes, G. B. Differential location and distribution of hepatic immune cells. Cells. 6 (4), 48 (2017).
- Schachtrup, C., Le Moan, N., Passino, M. A., Akassoglou, K. Hepatic stellate cells and astrocytes: Stars of scar formation and tissue repair. Cell Cycle. 10 (11), 1764-1771 (2011).
- Geerts, A. History, heterogeneity, developmental biology, and functions of quiescent hepatic stellate cells. Seminars in Liver Disease. 21 (3), 311-335 (2001).
- Poisson, J., et al. Liver sinusoidal endothelial cells: Physiology and role in liver diseases. Journal of Hepatology. 66 (1), 212-227 (2017).
- Peter, M. E., et al. The CD95 receptor: apoptosis revisited. Cell. 129 (3), 447-450 (2007).
- Peters, D. T., et al. Asialoglycoprotein receptor 1 is a specific cell-surface marker for isolating hepatocytes derived from human pluripotent stem cells. Development. 143 (9), 1475-1481 (2016).
- Willoughby, J. L. S., et al. Evaluation of GalNAc-siRNA conjugate activity in pre-clinical animal models with reduced asialoglycoprotein receptor expression. Molecular Therapy. 26 (1), 105-114 (2018).
- Hutchins, N. A., Chung, C. S., Borgerding, J. N., Ayala, C. A., Ayala, A. Kupffer cells protect liver sinusoidal endothelial cells from Fas-dependent apoptosis in sepsis by down-regulating gp130. American Journal of Pathology. 182 (3), 742-754 (2013).
