One of the most serious issues facing the biomedicalresearch community today is the authentication of human cell lines used inresearch and drug development as models of normal and cancer tissue. Cell linesconstitute an important scientific resource, enabling investigators to unravelcancer mechanisms that were once intractable to our methods, and to screen morepotential drug candidates than ever before. Progress is being undermined,however, by cell line misidentification and cross-contamination. In thisarticle, we will address the history of the problem, its implications to bothcancer research and drug discovery and the options available to investigatorsconcerned with validating their materials. Lastly, we will highlight how a newANSI consensus standard and efforts by granting agencies and journals torequire cell line authentication as a condition for funding and publication,and offer hope for a future free from the plagues of misidentified andcontaminated cells.
Cell line misidentification and contamination is not a newproblem. In the 1970s, Walter Nelson-Rees, who then ran the Berkeley cell bankfor the National Cancer Institute, used karyotyping to determine that celllines submitted to the bank were cross-contaminated by HeLa cells. Nelson-Reesembarked on a crusade against contaminated cell lines, which culminated in a1981 Science paper listingpublications that had used the flawed material. His gambit was met with outragefrom authors of the implicated work, and his crusade faltered. In 2004, Roland Nardone,then a professor at Catholic University, picked up Nelson-Rees' mantle andchampioned efforts to raise awareness of the scope of the cell linemisidentification problem. Nardone has emphasized the need for training in cellauthentication to be added to conference agendas and is still garnering thesupport of professional societies and funding agencies to require cell lineauthentication.
Despite these renewed efforts, the consequences of usingunauthenticated cell lines are already being brought to bear on researchers.For example, in 2010, a 2005 CancerResearch article that suggested human adult stem cells were prone tospontaneous transformation in vitro1was retracted because the phenomenon was actually due to contaminatingimmortalized cells2. A year later, researchers hoping to developtherapeutic tools for adenoid-cystic carcinoma had their paper retracted whenit was found that the cell line used to perform the experiments was actuallyderived from cervical cancer3. Importantly, these papers werepublished in well-respected journals, which increase the likelihood that otherinvestigators based their research on the flawed data and compounded theoriginal error.
A handful of cases have been uncovered, but it is clear thatcountless more exist. A recent review of cell lines used to study esophagealadenocarcinoma found that many were actually derived from lung or gastriccancers. Data obtained through the use of these cell lines have been used tosupport clinical trials, grant applications, U.S. patents and publications4.Similarly, genomic profiling of 51 supposedly independent ovarian cancer celllines revealed considerable redundancy and that several were actually derivedfrom cervical cancer5. These studies refer to clinically relevantcell lines that are being used to design and test new drugs for cancer andother diseases, and suggest that patients are potentially being recruited toflawed drug trials. If we assume that the findings from these two studies areapplicable across all clinical fields, then the number of advances that standto be reversed is staggering.
An important first step towards remediation is to addressthe root cause and develop a plan for going forward. In 2004, a survey of 485investigators indicated that only 15 percent of investigators relied on DNA-basedassays to authenticate their cell lines6. Then, and most certainlytoday, many investigators rely on phenotypic traits to confirm cell identity.This is an unreliable method for confirming cellular identity, since phenotypictraits may change unpredictably as passage number increases7.Additionally, many laboratories make use of multiple cell lines or feeder cellco-cultures, which boost the probability that a cross-contaminating event willoccur, and make reliable methods of cell authentication, such as thosediscussed below, even more important to rule out both intra- and interspeciescontamination.
The gold standard in genomic-based methods of cellauthentication is short tandem repeat (STR) profiling. STR profiling is aPCR-based approach that can discriminate the origin of the cell line down tothe original donor. This technique takes advantage of variable microsatelliteregions within the genomic DNA. Primers are designed to amplify these shortrepetitive sequences and the resulting amplicons are used to establish aprofile. The profile acts as a genetic fingerprint that can be compared to areference database to confirm the identity of the cell line7. STRprofiling can be performed rapidly using commercial kits and services, whichmakes it an approachable option for all investigators. Additionally, becausethe microsatellite regions are defined, cell line profiles will be comparablebetween labs and easily collected into a centralized public database.
Although STR profiling is a high-resolution approach toconfirming the identity of a cell line, it does have some limitations. First,primers to detect STR from non-human species are limited, and currently thismethod is not capable of identifying interspecies cross-contamination. Second,donor information and tissue samples for older, commonly used cell lines maynot be available, therefore it can be difficult, if not impossible to developreference standards for older lines8. In rare situations, more thanone method may need to be used to fully validate the identity of a particularcell line. For example, karyotyping, isoenzyme analysis and human leukocyteantigen typing can be used to compliment STR profiling, and this may benecessary if the cell line is being grown on a mouse feeder cell layer or ifthe original cell stock has been lost.
The commercial availability of STR kits and authenticationservices make it easier for investigators to authenticate cell lines. However,the problem reaches so deep into the scientific community that institutionalsupport and the international collaboration of subject matter experts isnecessary to eradicate the use of unauthenticated cell lines. Responding to therecognized problem of cell line misidentification and the need for standardizedmethods that can be used to test cell lines early and often, a standardconsensus method for cell authentication was released earlier this year by theATCC Standards Development Organization (SDO) as ANSI Standard ASN-0002,"Authentication of Human Cell Lines: Standardization of STR Profiling."9The ANSI standard recommends that cell lines be tested when a cell line isfirst grown in a laboratory, when preparing the initial frozen cell stock,after two or three passages while cells are expanded during the course ofexperimentation and at the time the research using the cell samples ispublished. The most important aspects of the standard are the discussions onthe numbers and types of loci to be evaluated, quality control of the data,interpretation of the results (matching criteria, loss of alleles, etc.) andimplementation of an STR database.
An international workgroup of scientists representingacademia, regulatory agencies, major cell repositories, government agencies andindustry, chaired by John R. W. Masters of University College London and YvonneA. Reid of ATCC, worked together to develop the standard. The standardrepresents a collective experience and expertise that led to a refinement andconsolidation of methods that should be of critical value to investigators whoare working with human cell lines. Standardization fosters the reproducibilityand comparability of research and development employing human cells, leading toa marked decrease in the misidentification and contamination of human cellsused by the scientific community.
Scientific journals and funding agencies are called upon torequire proof that all human cell lines used in research studies have beenproperly authenticated. Professional scientific societies such as ASCB and AACRare encouraged to sponsor conferences, workshops, webinars and trainingactivities to facilitate the adoption of cell line authentication standards.The quality and validity of funded and published research will markedly improveas a result of the reduction in the use of misidentified and contaminated humancell lines.
Elizabeth Kerrigan is the director of standards, and Dr.Carolyn Peluso is a cell biology specialist, both at ATCC, a private, nonprofitbiological resource center and research organization whose mission focuses onthe acquisition, authentication, production, preservation, development anddistribution of standard reference microorganisms, cell lines and othermaterials for research in the life sciences.
References:
1. Rubin, L. and Haston, K.: "Stem cell biology and drugdiscovery," BMC Biology.
2. de la Fuente, R., etal.: "Retraction: Spontaneous human adult stem cell transformation," Cancer Research.
3. Retraction notice to "Nef from SIVmac239 decreasesproliferation and migration of adenoid-cystic carcinoma cells and inhibitsangiogenesis," Oral Oncology.
4. Boonstra, J., etal.: "Verification and unmasking of widely used human esophagealadenocarcinoma cell lines," Journal ofthe National Cancer Institute.
5. Korch, C., et al.:"DNA profiling analysis of endometrial and ovarian cell lines revealsmisidentification, redundancy and contamination," Gynecology Oncology.
6. Buehring, G., Eby, E. and Eby, M.: "Cell linecross-contamination: how aware are Mammalian cell culturists of the problem andhow to monitor it?" In Vitro CellDevelopment Biology Animal.
7. Kerrigan, L. and Nims, R.; "Authentication of humancell-based products: the role of a new consensus standard," Regenerative Medicine.
8. "Cell line misidentification: the beginning of the end," Nature Reviews Cancer.
9. ANSI/ATCC ASN-0002-2011: "Authentication of Human CellLines: Standardization of STR Profiling."
