If you work with cell lines you may have paid attention to the dramatic headline published last month in the online journal STAT, Thousands of studies used the wrong cells, and journals are doing nothing.” In their column TheWatchdogs (“Keeping an eye on misconduct, fraud, and scientific integrity”), Ivan Oransky and Adam Marcus call out the fact that scientists continue to publish research using cell lines that are contaminated or misidentified. Recent estimates have found that the percentage of misidentified cell lines used by scientists is as high as 20 to 36. The blame here is being placed on the peer reviewed journals for not blowing the whistle. The authors call for journals to put some “kind of disclaimer on the thousands of studies affected.”
This is not a new claim. The continuing problem of cell line misidentification, of lack of authentication, has been covered before in various channels. It’s easy to find news publicizing yet another retracted publication. Promega Connections has published a number of blog posts addressing this, one as recently as last year: Do You #Authenticate? This post describes the bold move by the journal Nature to adopt a new policy around cell line authentication. Beginning in May 2015 the journal required authors of all submitted manuscripts to confirm the identity of cell lines used in their studies and provide details about the source and testing of their cell lines. Continue reading “The Cell Line Identity Crisis: Old problems, new concerns”
On April 15, 2015 Nature announced a new policy around authentication of cell lines used in research studies that are published in its journals (1). Beginning in May 2015 they are asking all authors to confirm that they are not working on cells known to have been misidentified or cross-contaminated and to provide details about the source and testing of their cell lines.
The problem of cell line misidentification has been well documented in the literature with issues being reported with hematopoietic cell lines in 1999 (2) and a lymphoma cell line in 2001 (3). In 2006, one study suggested that 15–20% cells used in experiments have been misidentified or cross contaminated (4). And, in her book The Immortal Life of Henrietta Lacks, published in 2010, science writer Rebecca Skloot, notes that concern about cell line contamination dates back to 1958 (5). Promega has written about this problem and the power of STR analysis to assist you in assuring that your cell lines are what they should be (6–9). In fact, as John Masters, Professor of experimental pathology at University College, London points, out, there really is no excuse for the continuing problem of cell line contamination:
“For nearly 50 years, people have been using falsely identified cells totally unnecessarily because they haven’t checked.” (10)
The problem of cell line misidentification and contamination is not a new problem, and the calls for the scientific community to take extra care in understanding the identity of the cells that they are working with are not new either. Nature journals are not the first journals to take a stand to require authors to authenticate their cell lines. Journals including International Journal of Cancer , In Vitro Cellular and Developmental Biology and Cell Biochemistry and Biophysics previously put policies in place around this issue (11,12), and in 2012 a new standard (ASN-0002) was officially published by the American National Standards Institute regarding human cell line authentication using profiles generated from STRs (11) . Based on the work of the ASN-0002 work group, the International Cell Line Authentication Committee was formed to promote awareness and authentication testing worldwide (13), including creating a publicly available database of misidentified cell lines. However, as more and more high-profile cancer studies are retracted because of cell line issues (14,15), it has become apparent standards for cell line culture and authentication will need to become common place in life science research. Continue reading “Do You #Authenticate?”
Back in 2009, we reported on the problem of cell line contamination (1). In that article we reported the statistics that an estimated 15–20% of the time, the cell lines used by researchers are misidentified or cross-contaminated with another cell line (1). This presents a huge problem for the interpretation of data and the reproducibility of experiments, a key pillar in the process of science. We have revisited this topic several times, highlighting the issues cell and tissue repositories have discovered with cell lines submitted to them (2) and discussing the new guidelines issued by ANSI (3,4) for researchers regarding when during experimental processes cell lines should be authenticated and what methods are acceptable for identifying cell lines.
Just recently two papers were voluntarily retracted by their authors because of cross contamination among cell lines used in the laboratories. The first that came to my attention represented the first retraction from Nature Methods in its nine years of publication. In this paper, cross contamination of a primary gliomasphere cell lines with HEK cells expressing GFP resulted in “unexplained autofluorescence” associated with tumorigenicity (5). The second paper, retracted from Cancer Research by the original authors, was also another cross contamination story involving HEK cells (6). In this story a gene was incorrectly described as a tumor suppressor, that when silenced led to the formation of tumors in nude mice. It turns out that the contaminating HEK cells also failed to express this same gene.
So because of cross contamination of cell lines, two groups have voluntarily retracted papers. Being open and honest about what had happened with the cell lines and reaching the decision to retract the papers could not have been an easy thing, but these decisions benefit the scientific community in many ways. Obviously they benefit the researchers doing work on the specific research questions addressed by the papers by preventing researchers from pursuing paths that lead to dead ends. But in the bigger picture these retractions reinforce the argument that cell line authentication needs to become a routine and accepted part of any experimental process that depends on cell culture if we are to have confidence in the experimental results.
Researchers working with immortalized cell lines would readily agree when I state that it is almost impossible to look at cells under the microscope and identify them by name. There are phenotypic traits, however they do change with change in media composition, passage number and in response to growth factors. I remember the pretty arborizations my neuroblastoma cell line SH-SY5Y exhibited in response to nerve growth factor treatment. Thus physical appearance is not a distinguishing feature. Currently, in many labs, researchers typically use more than one cell line, and more than likely, share the same lab space to passage cells and the same incubator to grow the cells. In such scenarios, it is not difficult to imagine that cell lines might get mislabeled or cross-contaminated. For example HeLa cells, one of the fastest growing cell lines have been shown to invade and overtake other cell lines.
Misidentification of cell lines has deep and severe implications. A review of cell lines used to study esophageal adenocarcinoma found that a large number of the cell lines were actually derived from lung or gastric cancers. Unfortunately, by the time this error was discovered, data from these cell line studies were already being used for clinical trials and other advanced studies and publications. Moreover, the cell lines were being to screen and design and test specific cancer drugs which ended up in flawed clinical trials. Continue reading ““Fingerprinting” Your Cell Lines”