5 tips for choosing the right cell line for your experiment



Cell line selection

1. Is it authentic?

It is important when you begin with any new cell line to be certain that it is what you believe it to be, and exclude the possibility that it’s been contaminated by other lines cultured in proximity.

The risk is very real – there are now over 400 misidentified cell lines registered in the Database of cross-contaminated or misidentified cell lines.

Ensure you obtain your cell line from a trusted cell bank such as ATCC, DSMZ or ECACC. If this isn’t possible, the identity of cell lines can be initially confirmed (and then annually assessed) using STR profiling at these same banks.

2. Is it free of contamination?

A common problem in cell culture labs is contamination of cell cultures with mycloplasma. An infection can significantly alter the biology of a cell, and compromise experimental results. mycoplasma infections are often missed however, due to the minute size of the micro-organism as well as their structural differences from other bacteria (they lack a solid wall, instead possessing a plasma like form).

Trusted cell sources will supply cell stocks as verified mycoplasma free. If the contamination status of a cell stock is not known, it is good practice to culture in a quarantine environment, and establish “clean” status prior to moving into general circulation. In this way the risk that mycoplasma will infect a whole cell culture lab is kept as low as possible.

3. Is it low passage?

Due to their inherent genetic instability, cancer cell lines will over time drift genetically. A recent paper from Domcke et al found “pronounced differences in molecular profiles between commonly used ovarian cancer cell lines and high-grade serous ovarian cancer tumor samples”. This is almost certainly in part due to the age (and consequent extended culture) of some of these lines, leading to their divergence from their originating tumors.

To avoid this it is recommended to maintain stocks of cells at as low a passage as possible, and regularly (every 2-3 months) restart your cell culture with a fresh vial. Changes to cell lines over time can in some cases be visually apparent – extended culture of U87 cells for example can lead to less of a monolayer, and cells growing as spheroids from the flask surface.

4. Does it exhibit the right biology?

Your choice of cell line will almost certainly be most dependent on the question or problem you’re trying to solve. If you’re studying a particular disease state then the more closely the cell line exemplifies this disease the better.

And even within disease types, careful selection is key. For example, if you’re studying breast cancer you may be interested in just triple negative breast cancer, or a panel of different subtypes may be preferable. Your research goals will inform your choices.

If you are studying a disease, an excellent place to identify a cell line that expresses a specific biomarker, or contains a certain genetic aberration is the Cancer Cell Line Encyclopaedia (CCLE). The CCLE provides public access to genomic data, analysis and visualization for about 1000 cell lines, allowing you to not only select an appropriate cell line, but then frame any results you generate in the context of the CCLE’s extensive data set. 

If your focus is basic biology rather than disease biology, then quite often there will be a model line with which a body of work has been developed. By choosing this same line, results can again be set in the context of existing scientific understanding.

This approach to selection however creates a self-perpetuating bias towards published cell lines, which may not have been originally selected for being the most suitable. Unsurprisingly a cell line’s age can influence how frequently it is found in the literature - as with the first immortalised cell line, HeLa cells, which have been used extensively (with their ease of culture no doubt perpetuating their use).

Best practice therefore is to identify multiple different cell lines, and demonstrate results across a panel. This not only builds confidence in results, it excludes the possibility that they are simply a quirk of an individual line.

5. Is it suitable for your experiment?

Choice of cell line for any experiment requires striking a balance between choosing the most biologically appropriate model and selecting a cell line that can be worked with.

Factors such as culture conditions, growth rate, transfectability, morphology and subcellular structure (for microscopy) and indeed cost can have a big impact on the experiments that can be performed, and their likely success.

For example, while induced pluripotent stem cells (iPSCs) may represent a more “normal” model than a cancer cell line, the arduous cell culture regimen and associated costs may mean these lines are not suitable for some labs.

The trick therefore is to find the best compromise – a cell line the exhibits the same or close to the right biology, but that enables good science to be done robustly and rapidly, while standing up to verification in other labs.

Culture conditions and growth rates for many commonly used cell lines can be found on the ATCC website.

How did Horizon select its cell lines?

At Horizon we have adopted the Hap1 human cell line as our model of choice for high throughput gene editing, and a panel of established colorectal cancer cell lines (HCT116, RKO, DLD1, SW48) as workhorse lines for cost effective generation of isogenic cell pairs.

To understand the HAP1 line better, we’ve performed expression profiling using RNASeq to confirm what is and isn’t expressed. We have subsequently built an extensive catalog of valuable engineered cell lines. These easy access CRISPR tools allow rapid hypothesis testing, and validation of in-house knockout line data as a secondary independent model. 

And from our customers and their publications using both the Hap1 cell line, we can learn where this model can be appropriately applied, including the study of:

To learn whether Hap1 cells can be suitable for your experiments take a look at the growing number of peer-reviewed articles using the HAP1 model system.