A cell line early in the cross-contamination process. (Click to enlarge)

Genetically speaking, cancer cells are highly stable, but STR profiles can “drift” over time due to both loss of heterozygosity and deficiencies in DNA repair mechanisms, explains Wilhelm Dirks, who leads the molecular biology group at DSMZ. Primary tumors, in contrast, are relatively heterogeneous and subject to “bottlenecking selection,” which can lead to genetically distinct subclones, says Roderick MacLeod, DSMZ's head of cytogenetics. As a result, an 80% match is generally considered sufficient to prove identity.

The European Collection of Cell Cultures (ECACC) in the UK, with some 1500 mycoplasma-screened and authenticated cell lines, has been employing STR analysis of human cell lines for a decade now, says Jim Cooper, General Collection Project Manager. While some banks, including DSMZ and ATCC use 8-locus panels, ECACC currently uses Applied Biosystems' 16-locus Identifiler primer set, a service it also offers to customers, as well as a cytochrome oxidase I assay for cell line speciation.

Previously, ECACC authenticated lines with a combination of classical DNA finger printing (a restriction enzyme/Southern blotting–based test) and the assay Gartler used in 1966: isoenzyme analysis (a colori metric assay of protein mobility). The change to the new methods, says Cooper, reflects not only cost, but quality. “The cost of the fingerprinting and isoenzyme analysis services seemed too much to bear for our external customers,” says Cooper. “Because of concerns over the sourcing of reagents and our drive to consolidate STR profiling of all cell lines, whether human or non-human, we replaced both techniques.”

Perhaps no cell bank goes to greater authentication lengths than the DSMZ, which involves six different working groups in the process. Cells submitted to that bank are tested for contamination by a collection of viruses and mycoplasma. They then are tested for species origin, karyotyped, and immunoprofiled for surface and/or internal markers. Human lines are DNA fingerprinted. According to Drexler, this process takes up to two weeks to complete, and costs €500–5000; in one exceptional case, says Dirks, his team needed six weeks to confirm the authenticity of a particular line with karyotyping and cell surface marker analysis. Best Practices

As Gartler remembers it, the reaction to his 1966 announcement was essentially denial. “People ascribed all sorts of unusual properties to the HeLa cell,” he recalls: for instance, that it could survive in the open without culture media, or that it could fly through the air. “That's almost nonsense,” he says.

In reality, though, cross-contamination can be just as beguiling. Mistakes can happen: double-dipping with a pipet, working with multiple lines at once, pouring liquids (which creates droplets), and so on. Fortunately, researchers can minimize the damage using some common-sense procedures. Check your lines against the Database of Cross-Contaminated or Misidentified Cell Lines. Don't passage cells too long (Masters advises 3 months or 10 passages, whichever is shorter). And of course, test the cells.

Douglass advises testing new lines when they enter the lab (if obtained from somewhere other than a cell bank), after freezing down a batch, and before publication. These days, validation is relatively inexpensive. ATCC charges $195 for its Basic STR Profiling Service and $295 for its Verified Profiling Service, while ECACC charges £180 for species identification and £190.55 for STR-based genotyping.

For the moment, STR testing is available only for human lines. Because mouse and rat lines are so inbred, “variation is almost undetectable,” says MacLeod. That means researchers can determine whether their animal cells represent the correct species, but not if they are the correct cells. But ECACC plans to launch a murine cell line STR profiling service in July, following later with rat and primate panels as well, says Cooper. Already, the bank's murine database houses 42 unique profiles, which they hope to extend to 70 in the near future.

Perhaps most importantly, says MacLeod, get your cells from a trusted source: A cell bank. “If you get cells from colleagues and collaborators, then there's always a question mark over the identity of the cells.”

2.) Gartler, S.M. 1968. Apparent HeLa cell contamination of human heteroploid cell lines. Nature 217:750-751.[CrossRef] [PubMed]

3.) Cancer studies “wasted millions” 2007. BBC. .[CrossRef] [PubMed]

4.) Nardone, R.M. 2007. “An Open Letter Regarding the Misidentification and Cross-Contamination of Cell Lines: Significance and Recommendations for Correction”. .[CrossRef] [PubMed]

5.) Capes-Davis, A.. 2010. Check your cultures! A list of cross-contaminated or misidentified cell lines. Int. J. Cancer 127:1-8.[CrossRef] [PubMed]

6.) Rae, J.M.. 2007. MDA-MB-435 cells are derived from M14 melanoma cells––a loss for breast cancer, but a boon for melanoma research. Breast Cancer Res Treat 104:13-19.[CrossRef] [PubMed]

7.) Dirks, W.G., R.A. MacLeod, and H.G. Drexler. 1999. ECV304 (endothelial) is really T24 (bladder carcinoma): Cell line cross-contamination at source. In Vitro Cell Dev Biol Anim 35:558-589.[CrossRef] [PubMed]

8.) Patel, V.A.. 2009. Retraction. Am. J. Physiol. Endocrinol. Metab. 298:E562.[CrossRef]

9.) Notice Regarding Authentication of Cultured Cell Lines 2007..[CrossRef] [PubMed]

10.) Editorial 2009. Identity crisis. Nature 457:935-936.[CrossRef] [PubMed]