Misidentified cell line fells cancer paper

Screen Shot 2016-04-15 at 2.52.06 PMResearchers have retracted a paper about a new molecular target for cancer after realizing they had mistaken the identity of their cell line.

It’s all too easy to mix up cell lines, so we see plenty of retractions for that reason — and, according to an expert in the area, many more cases lurk uncorrected in the literature.

The retraction notice for “Knockdown of tumor protein D52-like 2 induces cell growth inhibition and apoptosis in oral squamous cell carcinoma” in Cell Biology International explains the authors’ perspective on this case:

The above article, published online on 13 October 2014 in Wiley Online Library (http://onlinelibrary.wiley.com/doi/10.1002/cbin.10388/abstract), has been retracted by agreement between the authors, the journal Editor, Sergio Schenkman, and John Wiley & Sons Ltd. The retraction has been agreed because the authors discovered after publication that one of the cell lines described in the article had been unintentionally misidentified. The experiments described in the article as being conducted on Human Oral Squamous Cell Carcinoma cell line KB were in fact conducted on a Human Oral Epidermal-like Cancer cell line.

The authors and publisher apologise for any inconvenience.

The paper has not been cited, according to Thomson Reuters Web of Science.

The retraction follows a PubMed commons comment from Amanda Capes-Davis, chair of the International Cell Line Authentication Committee, in which she posits that the cell line is HeLa:

KB is not an OSCC cell line – KB cells are actually HeLa, from cervical adenocarcinoma. Cross-contamination of KB was discovered by Stanley Gartler back in 1967. Unfortunately, KB is still widely used as a model for oral cancer. Cross-contaminated cell lines are extensively used in the scientific literature, with many scientists not aware of this important problem.

A list of cross-contaminated or otherwise misidentified cell lines can be found at http://iclac.org/databases/cross-contaminations/.

She also posted a comment on the retraction notice, noting why she believes the authors were using HeLa cells:

More than 600 articles have been published from 2000 to 2015 that refer to the KB cell line as “oral” or “epidermoid” (squamous cell carcinoma), when it is actually HeLa and thus derived from cervical adenocarcinoma. This is the first time I have seen a retraction or correction published in response. I would like to acknowledge the authors, editor Sergio Schenkman and publisher John Wiley & Sons Ltd for their integrity in correcting the scientific record.

In the comment, she explains why she believes some of the wording in the retraction notice is misleading:

The comment that KB is a “Human Oral Epidermal-like Cancer cell line” may cause confusion. More than 50 years of testing, starting with the work of Stanley Gartler in the 1960s, shows that KB is derived from HeLa. HeLa has been extensively described and we can be confident that it is cervical carcinoma. In the early stages of cross-contamination a mixed culture can occur, in which the original cells are present alongside the contaminating cells, resulting in a mixed phenotype. Typically this only lasts for a few passages; the faster growing culture will rapidly overgrow and replace the other population (Nims et al, 1998, PMID 9542633). Where HeLa is the contaminant it will typically outcompete other cell types, due to its higher rate of proliferation and resilience at low density. There is no evidence that KB is currently a mixed culture, or that it retains any characteristics from the original culture that were present before cross-contamination occurred.

The editor of the journal, Sergio Schenkman, told us he didn’t have more information on how the authors reached the conclusion that they were working with a Human Oral Epidermal-like Cancer (rather than HeLa):

As a Journal Editor, I’m not in a position to ask for a proof for a retraction.

Capes-Davis told us she,

came across this paper while doing a routine search for journal articles that use the KB cell line.  I track usage of KB to try and understand how misidentified cell lines continue to be used in the literature.

The authors could have prevented the mistake, Capes-Davis added:

To detect misidentified cell lines, researchers need to do authentication testing.  This involves comparing a DNA profile from their sample to others from the same person.  All samples from the same person should have corresponding DNA profiles.  If other samples from that donor are not available, researchers can compare to online databases that contain data from commonly used cell lines such as HeLa.

We have reached out to authors Fengshan Chen, Sihui Chen, and Ying Cai, affiliated with Tongji University in China. We could not find contact info for first author Yongchun He.

Hat tip: Rolf Degen 

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our new daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.

One thought on “Misidentified cell line fells cancer paper”

  1. Cell lines in general may reflect only part of the typical behaviour of similar tissue. In this case the authors showed good scientific behaviour; perhaps the data could be used in a different way and published with major corrections.

    I am generally wondering if one could publish fascinating data on cell adhesion with a somehow “wrong” cell line, but which was tested extensively to have the relevant cell adhesion receptor on its surface (and perhaps due to being the “wrong” cell line not to express the usually disturbing other receptors of the “right” cell type – in fact the clear experiments may have never been possible in our unpublished findings). I think one can easily mention this in the materials section and discuss this in the discussion section, as long as the results are new to the field and as long as one can see the wrong cell line as a model for the right ones – and with comparison of two other (in this case really right ones, but different species).

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.