Hepatocyte Cell Lines: Types, Applications, & Advancements

Hepatocyte cell lines are lab-grown liver cells used to study drug metabolism, liver disease, and toxicity without relying on fresh human tissue for every experiment. The field spans a range of models, from tumor-derived lines like HepG2 and Huh-7 that grow easily but lack full liver function, to newer stem cell-derived cells and sophisticated three-dimensional culture systems that better mimic what happens in a living liver. Each model comes with trade-offs between convenience and biological realism, and choosing the right one depends on the question being asked.

Primary Hepatocytes and Why Alternatives Are Needed

Freshly isolated primary human hepatocytes remain the benchmark for liver research. They express the full range of drug-metabolizing enzymes and transporters found in a real liver, making them the most accurate tool for studying how the body processes medications.¹ The catch is that once you plate these cells in a flat dish, they start losing their liver-like properties within about 72 hours. They flatten out, stop producing key enzymes at normal levels, and essentially forget how to be hepatocytes. This rapid decline, called dedifferentiation, means experiments with primary cells have a narrow window, and the cells themselves are expensive and variable from donor to donor.

That combination of limited lifespan, high cost, and inconsistent supply is exactly why researchers have spent decades developing cell lines that can fill the gap. The ideal replacement would grow indefinitely, behave like a mature liver cell, and respond to drugs the same way a real hepatocyte does. No single line has hit all three marks, but several come close enough to be genuinely useful.

HepG2 Cells

HepG2 is the workhorse of liver cell research and one of the most widely used cell lines in biology. Originally derived from a liver tumor in a teenage boy in the 1970s, these cells are easy to maintain and grow reliably. Their biggest limitation is that they do a poor job of metabolizing drugs the way a healthy liver would. The enzymes responsible for breaking down medications, particularly the cytochrome P450 family, are present at dramatically lower levels than in primary hepatocytes.² Across the board, messenger RNA levels for most of these enzymes were far below what you would find in a real liver cell.³

The picture is a bit more nuanced for the second wave of drug processing, known as phase II metabolism. Some of these enzymes, including certain sulfotransferases and glutathione transferases, are present in HepG2 cells at levels roughly comparable to primary hepatocytes. Others, particularly the glucuronosyl transferases, can be hundreds or even a thousand-fold lower in HepG2 than in primary cells.⁴ So HepG2 cells can tell you something about certain metabolic pathways, but they are unreliable as a general stand-in for how a human liver handles a new compound.

An underappreciated wrinkle is that not all HepG2 cells are the same. When researchers compared HepG2 lines obtained from different laboratories, enzyme activities varied wildly. Some enzyme activities that were present in one source were completely absent in another, and the way these cells conjugated metabolites differed between sources.⁵ This matters because a finding from one lab’s HepG2 stock may not replicate in another lab that nominally has the same cell line.

Cell Line Authentication and Contamination

The variation between HepG2 stocks is not just a quirk of biology. Cross-contamination during cell culture is a recognized problem across biomedical research, and hepatocyte lines are no exception. When investigators used genetic fingerprinting to compare HepG2 isolates from six different laboratories, two of the six turned out to be derived from a different individual entirely, meaning they were not actually HepG2 at all.⁶ This kind of mix-up can silently undermine years of published work. Most journals and funding bodies now require cell line authentication by STR profiling before results are accepted, but older studies often predate these standards.

Huh-7 and Its Sublines

Huh-7, another hepatoma-derived line, carved out its own niche largely because of viral research. In 2005, researchers discovered that Huh-7 cells could support the full life cycle of hepatitis C virus, something no other reliable lab system could do at the time.⁷ That breakthrough made Huh-7 the standard platform for studying HCV infection, replication, and drug candidates aimed at blocking the virus.

Several sublines have been derived from Huh-7 to boost its utility. Huh-7.5, for example, carries a defect in its innate antiviral signaling that lets HCV replicate especially efficiently.⁸ Testing across nine Huh-7 stocks from different labs found that eight of nine were highly permissive for HCV infection, though some produced more infectious virus particles than others, suggesting variation in how well these cells support the later stages of viral assembly.⁹

Beyond virology, Huh-7 has shown some promise for drug metabolism studies. A comparative analysis of five hepatic cell lines found that Huh-7 cells, particularly between early passages, had the drug-metabolizing profile closest to primary hepatocytes. Their enzyme activities were still only around 5 to 15 percent of what primary cells produce, but Huh-7 delivered the most consistent results over time compared to other hepatoma lines.¹⁰

HepaRG Cells

HepaRG cells occupy an unusual and valuable space among hepatocyte models. Derived from a liver tumor in a patient with hepatitis C, these cells are bipotent: under the right culture conditions, they can differentiate into either hepatocyte-like cells or bile-duct-like (cholangiocyte) cells. When pushed toward the hepatocyte fate, they express functional levels of most major drug-metabolizing enzymes and maintain those levels stably for weeks.¹¹ Among all the commonly available cell line models, HepaRG cells show the greatest similarity to actual liver tissue and primary hepatocytes when it comes to the expression of drug-metabolizing genes.¹²

Their enzyme activities respond to prototypical inducers in much the same way primary hepatocytes do. Rifampicin, phenobarbital, and omeprazole all trigger the expected increases in their target enzymes.¹³ HepaRG cells also express a functional complement of drug transporters on both the blood-facing and bile-facing sides of the cell, closely matching the pattern found in primary hepatocytes. By contrast, HepG2 cells only showed notable activity for a couple of transporter types.¹⁴

They are not perfect. Two enzymes, CYP2A6 and CYP2E1, are notably underexpressed in HepaRG cells. And because they come from a single donor, they reflect one person’s genetic background for drug metabolism, which limits the ability to study population-level variation. Still, their combination of realistic enzyme profiles and indefinite growth makes them a go-to choice for many drug screening and toxicity labs.

Stem Cell-Derived Hepatocyte-Like Cells

Induced pluripotent stem cells have opened a different route to hepatocyte models. By reprogramming a patient’s own skin or blood cells into stem cells and then coaxing those toward a liver cell fate, researchers can generate hepatocyte-like cells with the patient’s exact genetic makeup. This theoretically provides an unlimited supply of cells that can model patient-specific liver diseases.¹⁵ The approach has been used for disease modeling, toxicity testing, and drug discovery.¹⁶

The persistent challenge is maturity. When tested head to head against fresh liver tissue, these hepatocyte-like cells consistently look more like fetal hepatocytes than adult ones. Across multiple tests comparing gene expression, protein production, and functional activity, cells from two independent laboratories both demonstrated fetal characteristics and fell short of the adult liver phenotype.¹⁷ They produce albumin and metabolize some drugs, but at levels you would expect from a developing liver rather than a fully mature one. Pushing these cells past the fetal stage remains one of the central technical problems in the field.

Three-Dimensional Culture and Organoids

Growing hepatocytes in flat, two-dimensional dishes is convenient but artificial. Liver cells in the body exist in a complex three-dimensional architecture with specific neighbors, blood flow, and structural proteins surrounding them. Researchers have increasingly moved toward 3D formats, including spheroids, scaffolds, and organoids, that better preserve mature hepatocyte traits and extend how long cells survive in culture.¹⁸

A landmark advance came with the development of long-term 3D organoid cultures from adult primary hepatocytes. These organoids can be established from single cells and maintained for multiple months while retaining key structural features, gene expression, and function.¹⁹ That longevity directly addresses the dedifferentiation problem that plagues conventional primary hepatocyte cultures.

The surrounding structural proteins also matter. When Huh-7.5 cells were grown in microporous scaffolds coated with different extracellular matrix proteins, collagen and fibronectin each boosted albumin production and liver-specific gene expression, but they did so in distinct ways, producing different cell aggregation patterns and activating different adhesion-related genes.²⁰ Hepatoma cells grown on these scaffolds also showed enhanced expression of albumin and other mature hepatocyte markers, suggesting that the 3D environment actively pushes tumor-derived cells toward a more primary-hepatocyte-like state.²¹

Liver-on-a-Chip Systems

Even 3D cultures usually sit in still liquid, which is nothing like the constant blood flow a real liver experiences. Liver-on-a-chip devices use tiny channels to pump fluid past hepatocytes, recreating the mechanical forces and nutrient gradients that cells encounter inside the body. These microphysiological systems provide perfusion, shear stress, and 3D cell-cell communication that static cultures cannot offer.²² Maintaining continuous flow has been shown to better preserve hepatocyte function and the metabolic zonation that exists across a real liver lobule.²³

These platforms are increasingly paired with hepatocyte cell lines and high-throughput screening. In one example, a microfluidic device containing HepG2 cells was used to test the toxicity of acetaminophen across a gradient of eight concentrations simultaneously, using only a few hundred microliters of drug solution.²⁴ Combining HepaRG cells with high-throughput imaging and machine learning has also been explored for predicting direct liver toxicity in humans, using phenotypic profiling across dozens of reference compounds.²⁵

Immune Co-Culture Models

Standard hepatocyte cultures miss a critical factor in many drug-induced liver injuries: the immune system. Some drugs cause liver damage not by directly poisoning hepatocytes but by triggering an immune attack against them. Recreating this in a dish requires growing liver cells alongside immune cells. Co-cultures of hepatocytes with Kupffer cells, the liver’s resident immune cells, have shown that certain toxicity signals only appear when inflammation is present. Trovafloxacin and acetaminophen, both linked to immune-mediated liver damage, produced toxicity markers in co-cultures activated by an inflammatory stimulus that were absent in hepatocyte-only dishes.²⁶

More recent work has gone further. A platform combining iPSC-derived liver organoids with the patient’s own immune cells has been used to model drug-induced liver injury that depends on specific immune genes. This system was able to recapitulate the injury caused by flucloxacillin in patients carrying a particular immune gene variant, including T-cell activation, hepatocyte death, and release of injury markers.²⁷ Immune-competent liver models are still in early development, but they address a real blind spot: roughly a third of all drug withdrawals from the market are due to liver toxicity, and many of those cases involve immune mechanisms that simple cell lines cannot detect.²⁸

CRISPR Gene Editing in Hepatocyte Models

Gene-editing tools, especially CRISPR, have turned hepatocyte cell lines into customizable disease models. Instead of waiting to find patient tissue with a particular genetic mutation, researchers can now engineer the mutation directly into an existing cell line and study its effects.

For Wilson’s disease, a genetic condition where copper accumulates in the liver, researchers corrected the responsible mutation in iPSC-derived hepatocytes. The gene-corrected cells restored normal copper-handling function, including the ability to shuttle the copper transporter protein to the right location in the cell and export excess copper.²⁹ In the opposite direction, knocking out a single gene in HepG2 cells created a model of primary hyperoxaluria type 1, a rare liver disease that causes kidney stones. The knockout cells lost the target enzyme activity and produced excess oxalate when challenged, validating them as a platform for studying the disease and screening drug candidates.³⁰

A newer application uses CRISPR prime editing, a more precise variant of the technology, to introduce a specific fat-metabolism gene variant into an immortalized hepatocyte line. The engineered cells accumulated lipid droplets and showed gene expression changes matching the fatty liver disease associated with that variant, providing a reproducible model that does not depend on variable patient-derived tissue.³¹

Bioartificial Liver Devices

Hepatocyte cell lines are also being used outside of basic research, in clinical devices designed to support patients with acute liver failure. A bioartificial liver is an external device that circulates a patient’s blood plasma over living hepatocytes packed in a bioreactor, aiming to keep the patient alive until their own liver recovers or a donor organ becomes available.³²

For these devices, cells need to perform a broad range of real liver tasks under demanding conditions. Two cell lines, HepaRG and HepG2/C3A (a derivative of HepG2 selected for higher albumin production), have reached advanced stages of development for this purpose. In a direct comparison, HepaRG cells outperformed C3A cells on drug metabolism, ammonia clearance, and lactate processing, while protein production was comparable between the two. Growing both cell types inside the three-dimensional bioreactor format improved their function compared to flat cultures.³³ Separate work confirmed that HepaRG cells achieved levels of ammonia elimination, protein production, and key enzyme expression on par with primary hepatocytes under optimized culture conditions.³⁴

Serum-Free Culture and Standardization

An underappreciated variable across all hepatocyte models is the culture medium itself. Traditional media contain animal serum, which introduces batch-to-batch variability and undefined growth factors that can influence cell behavior in unpredictable ways. For applications where consistency matters, including drug screening and clinical devices, chemically defined serum-free media offer a cleaner alternative. Optimized serum-free formulations have been shown to maintain key hepatocyte functions like glycogen storage, albumin secretion, and urea production while supporting cell growth and proliferation.³⁵ As regulatory expectations for reproducibility tighten, the shift toward defined media is likely to accelerate across hepatocyte research and its translational applications.