Closing the Case on STAP?

Several reports offer an inside look into the stem-cell research controversy.

Sep 23, 2015
Karen Zusi

HARUKO OBOKATALast year, two Nature papers in which researchers reported having created pluripotent stem cells by stressing differentiated mouse cells were published, questioned, and subsequently retracted. The confusion led to multiple misconduct investigations and a media firestorm, and reportedly contributed to the death of one of the studies’ authors. A set of papers published today (September 23) in Nature brings to light additional details regarding the controversial stimulus-triggered acquisition of pluripotency—or STAP—research.

In their original articles (published in January 2014; retracted in July 2014), lead author Haruko Obokata of Japan’s RIKEN Center for Developmental Biology and her colleagues reported that exposure to an acid bath could reprogram mature cells into pluripotent STAP cells. Within weeks, other researchers reported difficulty reproducing the published protocols. (Obokata resigned from RIKEN in December 2014.)

As it turns out, contamination of the supposed STAP cells with embryonic stem cells may have been in part to blame. In one of the Nature commentaries published today, researchers from RIKEN and the Tokyo Institute of Technology examined this aspect of the STAP reproducibility problem.

“The publication and retraction of STAP papers strongly influenced the national scientific community, especially the stem-cell community. We felt that we were responsible to notice what was real,” study coauthor Fumio Matsuzaki of the RIKEN Center for Developmental Biology told The Scientist.

Using whole-genome sequencing genomic analysis to “fingerprint” the chromosomal details of STAP-like cells, Matsuzaki and his colleagues found that these pluripotent cells were derived from established embryonic stem cell lines. “This is probably the first probe to investigate [STAP cells] using whole-genome sequencing,” said Matsuzaki. “During the investigation, this approach was very powerful.”

These data were included in RIKEN’s own investigative committee reports released at the end of 2014, but this latest report marks the introduction of this information to the peer-reviewed literature. Matsuzaki told The Scientist that his team provided comment at 14 of RIKEN’s investigative committee meetings. “It was very hectic work.”

“It helps sort things through in a new way that’s much more thorough and points to specific things that may have gone wrong,” said Paul Knoepfler, a stem-cell biologist from the University of California, Davis, who was not involved with the work.

A team led by George Daley of Harvard Medical School also published a report analyzing the failure of seven independent stem-cell research labs to produce pluripotent stem cells via the STAP protocols. Daley had contacted one of the STAP authors, Charles Vacanti of Brigham and Women’s Hospital in Boston, shortly after the original papers were published. “I wrote Chuck Vacanti and said, ‘We’d love to collaborate with you on the STAP phenomenon and dissect it on the single-cell levels,’” recalled Daley. “He was receptive and accepted one of my trainees into his lab. And within a very short period of time, my trainee started to recognize artifacts in the process and was unable to replicate the process within the Vacanti lab.”

“It was significant that they were able to attempt the validation in Vacanti’s lab with one of the senior authors and it still didn’t work. That kind of resonates with Obokata herself trying to replicate STAP under supervision, and it didn’t work,” said Knoepfler. “The consensus is that STAP is not a real phenomenon.”

Daley cited the observation of autofluorescence, essentially a “false positive” for pluripotency, as an early indication that something might have been amiss with the results. “In a very short period of time there were a lot of us communicating and trying to determine whether anybody else had success,” he said.  The researchers included in Daley’s report, hailing from China and Israel as well as several US stem-cell labs, tested 133 experimental replicates in total using different STAP protocols. None succeeded.

“It’s always impossible to prove the negative, but you can at least say—under the conditions that were reported—that these were neither robust nor reproducible in the hands of labs that should’ve been expert enough to do it,” said Daley.

“I think these papers are a positive step in resolving what went on,” said Knoepfler.

A. De Los Angeles et al., “Failure to replicate the STAP cell phenomenon,” Nature, doi:10.1038/nature15513, 2015.

A. De Los Angeles et al., “Hallmarks of pluripotency,” Nature, doi:10.1038/nature15515, 2015.

D. Konno et al., “STAP cells are derived from ES cells,” Nature, doi:10.1038/nature15366, 2015.