Federally funded US researchers were supplied with a prohibited stem-cell line, disguised as an approved line by researchers at Seoul's MizMedi hospital. Allegations that a cell line had been switched surfaced earlier this month (see Nature 441, 680; 2006). Sung-il Roh, chairman of the board of trustees at MizMedi Women's Hospital, has now admitted to Nature that this was indeed the case.

Research with the unapproved cell line has been halted. But the incident highlights how hard it is for researchers to know what they are working with. And in this case the problem was not only scientific but political: US president George W. Bush announced in 2001 that because of ethical concerns over human embryonic stem cells, only such cell lines established before 9 August 2001 would be eligible for federally funded research. One line established by researchers at MizMedi Hospital, Miz-hES1, was created before that date, and approved by the US National Institutes of Health (NIH). But the cell line sent out in its place, Miz-hES5, was created later.

Identity crisis: stem cells all look the same. Credit: D. SCHARF/SPL

The NIH has paid the hospital US$930,000 since 2002 to grow, characterize and distribute Miz-1. But Roh told Nature that in December 2003, researchers at the hospital identified a chromosomal abnormality in that line. In April 2004 they switched to Miz-5, but pretended they were shipping Miz-1.

The subterfuge came to light during the police investigation into the research fraud committed by cloning expert Woo Suk Hwang (see Nature 439, 122; 200610.1038/439122a). The human eggs used by Hwang were harvested at MizMedi. Roh admitted in November 2005 to paying for the eggs that he passed to Hwang, despite previously claiming they were from volunteers.

It's a healthy line. From a research perspective, it wouldn't cause trouble.

Roh says he knew nothing about the stem-cell switch, and only found out from the police report. But he says: “It's a healthy line. From a research perspective, it wouldn't cause any trouble.” According to Roh, around 80 Korean groups and more than 30 foreign groups were shipped the Miz-1 line. NIH spokesman John Burklow says three dozen US researchers received what they thought was Miz-1. But Roh says about 16 groups were shipped Miz-5 disguised as Miz-1.

The NIH has suspended research on MizMedi's cell line, and insists no one used it in federally funded work. But the episode raises the question of what can be done to ensure the integrity of stem-cell lines.

James Battey, head of the NIH Stem Cell Task Force, says that when the agency receives a stem-cell line, it checks that the cells have normal chromosomes, are free of contamination, and replicate in culture. These tests confirm a line's health but do not check its identity. As the two MizMedi lines were both male, ‘the NIH would have no easy way of knowing that Miz-5 was substituted for Miz-1’, he says.

Extensive DNA analysis would tell cell lines apart but is time-consuming, and reference information for the original line is often not available. Scientists at the Burnham Institute for Medical Research in La Jolla, California, working with a large group from the NIH and the DNA-analysis company Illumina of San Diego, have recently made a dent in this problem.

The work was spearheaded by Mahendra Rao, who left the NIH in October 2005 and is now at the company Invitrogen. While at the NIH, Rao offered to perform detailed genetic profiles of all president-approved stem-cell lines for free. He says suppliers weren't always eager to share their resources, but he has analysed the lines he could get agreements on, using bead-based microarrays developed by Illumina to compare single nucleotide polymorphisms (SNPs). Rao has published his analysis of 7 of 22 cell lines included in the NIH registry (http://www.biomedcentral.com/content/pdf/1471-213x-6-20.pdf), and says a paper in press covers ten more.

Rao says a quick, cheap profiling method is essential: “You need to know what you are working with, because mix-ups are possible.” The NIH says it knows of no other cases of deliberate fraud, but there have been accidental switches. Rao says he uncovered a case in which a US company was shipping out wrongly identified cells after a technician accidentally mislabelled vials: “We were able to inform the few people who got the mislabelled line.”

SNP analysis also reveals the genetic basis of different cell types. The work heralds a breakthrough in that respect, says Burnham's Evan Snyder: “Only now is the research evolving to the point where people are starting to profile these lines and compare them.”