In plain English
A cell culture model is a set of living cells kept in a dish or flask, fed with nutrient medium and held at a steady temperature. Researchers use these models to study how cells behave and how they respond to a test condition, such as exposure to a peptide at a set concentration [10, 12].
There are three broad kinds:
- Cell lines keep dividing almost indefinitely. They are cheap, uniform and easy to share, but they have changed a great deal from the tissue they came from [14, 15].
- Primary cells are taken directly from a tissue sample. They are closer to the original tissue, but they stop dividing after a limited number of divisions [11].
- Organoids are small three-dimensional structures grown from stem cells that copy some of the architecture of an organ [13].
The biggest practical risk is simple: the cells may not be what the label says. Tens of thousands of published papers have reported work on misidentified cells [4]. So identity testing, contamination checks and passage records are part of the result, not paperwork on the side.
The short version
- Cell lines are convenient but genetically altered from their source [14, 15].
- Primary cells are closer to tissue but divide only a limited number of times in culture [11].
- Organoids add three-dimensional structure but are harder to standardise [12, 13].
- STR profiling confirms the identity of human cell lines [6, 7].
- Mycoplasma contamination is common and needs dedicated tests to detect [8, 9].
- An in vitro result describes cells in a dish, nothing more.
What is a cell culture model?
Cell culture is the growth of cells outside the organism they came from, in a controlled environment. It has become a standard tool for studying how cells assemble into tissues and how those tissues function [12]. Most work still uses flat, two-dimensional (2D) culture on plastic, although research has shifted toward three-dimensional (3D) systems with more realistic chemical and mechanical surroundings [12].
The Good Cell and Tissue Culture Practice guidance (GCCP 2.0, 2022) sets out six principles for reproducible cell-based work: characterising cells and maintaining their essential characteristics, quality management, documentation and reporting, safety, education and training, and ethics [10]. It describes potential quality problems rather than prescribing fixed procedures [10].
The main model types compared
| Model | Where it comes from | Strengths | Main limits |
|---|---|---|---|
| Immortalised cell line (for example HeLa, HEK293) | A single original culture, adapted to divide indefinitely | Unlimited supply, uniform, widely shared, large published literature | Abnormal chromosome numbers and gene expression; drift over time; misidentification risk [4, 14, 15] |
| Primary cells | Freshly isolated from a donor tissue sample | Closer to the source tissue | Limited number of divisions; donor-to-donor variation; supply depends on tissue access [11] |
| Stem cell-derived cells (including induced pluripotent stem cells) | Reprogrammed adult cells or stem cells, then differentiated | Can produce many cell types from one source | Differentiation varies between batches; characterisation is demanding [3, 10] |
| Organoids (3D) | Stem cells grown in a 3D matrix | Recreate some organ architecture and cell-to-cell contact | Variable size and composition; nutrient and oxygen gradients [12, 13]; harder to examine and measure |
Immortalised cell lines
An immortalised (or transformed) cell line is a population that has escaped the normal limit on cell division. Two lines show both the value and the problems.
HeLa, a cervical epithelial line, is the most widely used human cell line in cell and molecular biology [14]. When its genome was sequenced in 2013, the HeLa Kyoto line showed a remarkably high level of aneuploidy (abnormal chromosome numbers), many large structural rearrangements, and gene expression patterns that differ from normal human tissue. The authors stressed that these aberrant features must be taken into account when designing and interpreting experiments [14].
HEK293, a line derived from human embryonic kidney cells, is widely used in cell biology and biotechnology [15]. Whole-genome sequencing of six 293 lines showed an aneuploid genome and measurable genetic differences between common derivatives such as 293T, 293S and 293SG. Under standard culture conditions the genome structure of most of these lines was stable [15].
The lesson is that a cell line is a model of itself. It can answer well-posed mechanistic questions, but it is not a stand-in for a tissue.
Primary cells
Primary cells are isolated directly from tissue, so they have not been adapted to grow forever. In 1965 Leonard Hayflick reported that normal human diploid cell strains can be kept in culture only for a limited time [11]. That limit is why primary cells must be sourced repeatedly, and why results can vary from one donor to the next.
In Australia, the ethical conduct of research involving human participants is guided by the NHMRC National Statement on Ethical Conduct in Human Research 2025, in effect from 23 June 2026 [17]. A laboratory planning to work with donor tissue checks with its human research ethics committee before the tissue is collected.
Stem cell-derived cells and organoids
In 2007, researchers showed that adult human fibroblasts could be reprogrammed into induced pluripotent stem (iPS) cells using four transcription factors. These cells resembled embryonic stem cells and could form cell types of all three germ layers in vitro [18].
Organoids are stem cell-derived 3D culture systems that can recreate aspects of the architecture and physiology of human organs in considerable detail [13]. They complement animal models and other cell systems, and a 2020 review set out both their advantages and the challenges that remain before they can substantially reduce animal experiments [13].
2D versus 3D culture
Flat 2D culture dominates because it is simple, cheap and easy to examine under a microscope. But cells on rigid plastic experience a very different environment from cells in tissue. A 2017 review comparing the two noted that 3D culture offers more realistic biochemical and biomechanical surroundings, while still facing open problems with tissue-to-tissue interfaces, the mechanical environment, and the uneven distribution of oxygen, nutrients and waste [12].
| Feature | 2D monolayer | 3D (spheroid, scaffold or organoid) |
|---|---|---|
| Cell contact | Mostly with plastic and neighbours on one plane | With neighbours in all directions and with a matrix |
| Access to medium | Every cell is near the medium | Gradients of oxygen and nutrients from surface to core [12] |
| Reproducibility | Generally high | Lower; size and composition vary [12, 13] |
| Microscopy and measurement | Straightforward | Harder; thick samples need specialised microscopy |
| Cost and throughput | Low cost, high throughput | Higher cost, lower throughput |
One practical consequence: a test compound applied at the same nominal concentration may not reach the inner cells of a 3D structure at that concentration. Concentration–response results from 2D and 3D formats are not directly interchangeable. See EC50, IC50, Kd and Ki.
Cell line misidentification and STR profiling
How big is the problem?
Cell lines can be swapped, mislabelled, or overgrown by a faster-dividing line that gets into the culture. A 2007 review estimated, from submissions to major cell repositories, that 18% to 36% of cell lines may be contaminated or misidentified [5]. A 2017 analysis found 32,755 published articles reporting research on misidentified cells, cited in turn by an estimated half a million other papers. The authors found no sign that this “contaminated” literature was shrinking [4].
The International Cell Line Authentication Committee (ICLAC) curates a Register of Misidentified Cell Lines: lines known to be misidentified through cross-contamination or mislabelling. Version 14 was released on 15 February 2026 [3]. Checking a line against this register before starting work is a basic step.
What is STR profiling?
Short tandem repeat (STR) profiling is a DNA genotyping method. It reads the number of repeats at a set of highly variable positions in the genome, giving each human individual, and each cell line derived from them, a characteristic profile. If applied routinely, STR genotyping, in combination with other methods, can greatly improve detection of cross-contamination [6].
STR profiling is the only cell line authentication method covered by a comprehensive consensus standard, ANSI/ATCC ASN-0002 [7]. That standard was revised in 2021 and gives the detailed criteria testing laboratories use to evaluate and interpret STR data [6]. Other genomic methods can add information about cell line quality, and a combination of methods gives the fuller picture [7].
Checking a profile
The Cellosaurus is a knowledge resource that aims to describe all cell lines used in biomedical research [1]. Release 57 (October 2026) lists 169,026 cell lines, including 124,299 human lines [2]. It flags problematic, contaminated and misidentified lines, and its CLASTR tool searches a laboratory’s STR profile against reference profiles [2]. Each entry carries an accession that doubles as a Research Resource Identifier: HeLa is RRID:CVCL_0030 and HEK293 is RRID:CVCL_0045 [2].
Worked example. A laboratory receives a vial labelled as a kidney epithelial line from a colleague.
- It checks the line’s name against the ICLAC Register [3] and the Cellosaurus entry [2].
- It sends a sample for STR profiling before any experiments [6].
- It compares the result with the reference profile using CLASTR [2].
- The profile matches HeLa instead. The vial is discarded, and the finding is recorded so earlier data from that stock can be reviewed.
Mycoplasma contamination
Mycoplasmas are small bacteria and notorious contaminants of cell culture. They can alter host cell biology by depriving cells of nutrients and causing broad changes in gene expression [8]. Routine sentinel testing over two decades has found widely ranging contamination rates in mammalian cultures [8].
A 2015 survey of sequencing data from 9,395 rodent and primate samples, spread across 884 data series from hundreds of laboratories, found that 11% of the series were contaminated. The authors also found 61 host genes whose expression tracked the level of mycoplasma reads [8]. In other words, contamination does not just add noise; it can change the answer.
Polymerase chain reaction (PCR) and fluorescent in situ hybridisation (FISH) are fast and sensitive detection methods. Both need built-in control reactions to guard against false negatives from reaction inhibition and false positives from contamination or background staining [9].
Passage number and genetic drift
Each time cells are split into fresh vessels, the passage number goes up by one. Over many passages, selective pressure and genetic drift change the population. A 2007 review concluded that cell lines kept in culture too long show reduced or altered key functions and often no longer represent their source material; it recommended using verified lines at low or defined passage numbers [5].
Good practice is to:
- record the passage number on every vessel and in the laboratory record;
- work from a characterised master cell bank and a working cell bank, frozen at low passage;
- set an upper passage limit for each line and return to the bank when it is reached;
- note the passage number in methods sections, so others can compare like with like [5, 10].
What in vitro models cannot show
Cell culture is powerful for mechanism, but its limits are structural.
- No whole organism. A dish has no circulation, no immune system and no organ-to-organ signalling.
- Abnormal cells. Immortalised lines carry large genomic changes that set them apart from normal tissue [14, 15].
- Artificial environment. Rigid plastic, high oxygen and rich medium differ from tissue, and even 3D systems only partly close that gap [12].
- Concentration is not exposure. A nominal concentration in medium can differ from what cells actually see, because peptides can stick to plastic or break down. See Peptide Adsorption to Labware.
- Contaminants confound results. Mycoplasma contamination is linked to changes in host gene expression [8]. Bacterial endotoxin in a test material is a separate concern; see Endotoxin (LAL) Testing.
- Reproducibility. One economic analysis estimated that the cumulative rate of irreproducible preclinical research exceeds 50%, costing about US$28 billion a year in the United States alone [16]. Unverified cell lines are one of the avoidable contributors [5].
For these reasons, the evidence level label “In vitro” on this site means exactly what it says: an observation in cells or tissue culture, which cannot be assumed to apply to a living organism.
Reading a cell culture study: a checklist
When reading a paper that relies on cell culture, look for:
- The cell line’s name and RRID, so it can be checked in the Cellosaurus [2].
- A recent STR authentication result, with the date and the method [6, 7].
- A mycoplasma test result and the detection method used [9].
- The passage range used in the experiments [5].
- The culture format (2D, spheroid, organoid) and why it suits the question [12].
- Concentrations with units and exposure times, plus vehicle controls.
- Replicates, and whether they are independent cultures or wells from the same plate. See Statistics in Research Papers.
What this means in Australia
- Human research ethics. Research involving human participants is guided by the NHMRC National Statement, current version 2025 [17]. Work with donor tissue is a question for a human research ethics committee.
- Regulatory status. A cell culture result is laboratory evidence. It does not change the regulatory status of any substance. The TGA has stated that a “research use only” disclaimer does not make supply lawful or change a product’s regulatory status [19].
This is general information, not legal advice. Laws change; check current NHMRC and TGA guidance.
Frequently asked questions
What are cell culture models?
They are cells grown outside the organism under controlled conditions, used to study cell behaviour and responses to test conditions. The main types are cell lines, primary cells and 3D organoids [10, 12, 13].
What is the difference between a cell line and primary cells?
Cell lines divide almost indefinitely but are genetically altered from their source. Primary cells come directly from tissue samples and divide only a limited number of times in culture [11].
What is STR profiling?
STR profiling is a DNA genotyping method that gives a human cell line a characteristic profile, which can be compared with reference profiles to confirm its identity [6, 7].
How common is mycoplasma contamination?
In one survey of public sequencing data, 11% of 884 data series showed mycoplasma contamination [8].
Are organoids better than 2D culture?
Organoids recreate more of an organ’s structure, but they are harder to standardise and quantify. The better choice depends on the question [12, 13].
References
- Bairoch A. The Cellosaurus, a cell-line knowledge resource. J Biomol Tech. 2018;29(2):25–38. doi:10.7171/jbt.18-2902-002. PMID: 29805321
- SIB Swiss Institute of Bioinformatics. Cellosaurus: the reference resource on cell lines. Release 57. cellosaurus.org. Accessed 25 October 2026.
- International Cell Line Authentication Committee (ICLAC). Register of Misidentified Cell Lines, version 14. 15 February 2026. iclac.org. Accessed 25 October 2026.
- Horbach SPJM, Halffman W. The ghosts of HeLa: how cell line misidentification contaminates the scientific literature. PLoS One. 2017;12(10):e0186281. doi:10.1371/journal.pone.0186281. PMID: 29023500
- Hughes P, Marshall D, Reid Y, Parkes H, Gelber C. The costs of using unauthenticated, over-passaged cell lines: how much more data do we need? Biotechniques. 2007;43(5):575–586. doi:10.2144/000112598. PMID: 18072586
- Almeida JL, Korch CT. Authentication of human and mouse cell lines by short tandem repeat (STR) DNA genotype analysis. In: Markossian S, Grossman A, Baskir H, et al., eds. Assay Guidance Manual. Bethesda (MD): Eli Lilly & Company and the National Center for Advancing Translational Sciences; updated 17 January 2023. PMID: 23805434
- Almeida JL, Cole KD, Plant AL. Standards for cell line authentication and beyond. PLoS Biol. 2016;14(6):e1002476. doi:10.1371/journal.pbio.1002476. PMID: 27300367
- Olarerin-George AO, Hogenesch JB. Assessing the prevalence of mycoplasma contamination in cell culture via a survey of NCBI’s RNA-seq archive. Nucleic Acids Res. 2015;43(5):2535–2542. doi:10.1093/nar/gkv136. PMID: 25712092
- Uphoff CC, Drexler HG. Detection of mycoplasma contamination in cell cultures. Curr Protoc Mol Biol. 2014;106:28.4.1–28.4.14. doi:10.1002/0471142727.mb2804s106. PMID: 24733240
- Pamies D, Leist M, Coecke S, et al. Guidance document on Good Cell and Tissue Culture Practice 2.0 (GCCP 2.0). ALTEX. 2022;39(1):30–70. doi:10.14573/altex.2111011. PMID: 34882777
- Hayflick L. The limited in vitro lifetime of human diploid cell strains. Exp Cell Res. 1965;37:614–636. doi:10.1016/0014-4827(65)90211-9. PMID: 14315085
- Duval K, Grover H, Han LH, et al. Modeling physiological events in 2D vs. 3D cell culture. Physiology (Bethesda). 2017;32(4):266–277. doi:10.1152/physiol.00036.2016. PMID: 28615311
- Kim J, Koo BK, Knoblich JA. Human organoids: model systems for human biology and medicine. Nat Rev Mol Cell Biol. 2020;21(10):571–584. doi:10.1038/s41580-020-0259-3. PMID: 32636524
- Landry JJ, Pyl PT, Rausch T, et al. The genomic and transcriptomic landscape of a HeLa cell line. G3 (Bethesda). 2013;3(8):1213–1224. doi:10.1534/g3.113.005777. PMID: 23550136
- Lin YC, Boone M, Meuris L, et al. Genome dynamics of the human embryonic kidney 293 lineage in response to cell biology manipulations. Nat Commun. 2014;5:4767. doi:10.1038/ncomms5767. PMID: 25182477
- Freedman LP, Cockburn IM, Simcoe TS. The economics of reproducibility in preclinical research. PLoS Biol. 2015;13(6):e1002165. doi:10.1371/journal.pbio.1002165. PMID: 26057340
- National Health and Medical Research Council. National Statement on Ethical Conduct in Human Research 2025. Effective 23 June 2026. nhmrc.gov.au. Accessed 25 October 2026.
- Takahashi K, Tanabe K, Ohnuki M, et al. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell. 2007;131(5):861–872. doi:10.1016/j.cell.2007.11.019. PMID: 18035408
- Therapeutic Goods Administration. Understanding your responsibilities when importing, compounding and supplying unapproved peptide products. Safety advisory, 13 April 2026. tga.gov.au. Accessed 25 October 2026.
