Nature has published a correction to a 2002 article identifying cells from bone marrow that were surprisingly multipotent, or able to form a variety of other cell types1,2. The main findings of the paper are unaffected by the correction, but descriptions of proteins on the cells' surface were flawed, according to the paper's authors and independent experts who reviewed data at Nature's request.

Led by Catherine Verfaillie, who is now at the Catholic University of Leuven (KUL) in Belgium, the researchers say measurements of a control protein on cells' surface varied so much that the measurements for other surface markers cannot be trusted. Additionally, they wrongly identified measurements for one antigen (MHC1) as those for another antigen (Mac-1). The researchers subsequently assessed these markers on another line of the so-called multipotent adult progenitor cells (MAPCs) and obtained more reliable data, which are presented in the Nature correction.

But Irving Weissman, director of the Stanford Institute of Stem Cell Biology, says mistakes in the markers are not the main issue. He worked with Verfaillie on a different but similarly characterized set of cells than those isolated in the Nature paper. These experiments, he says, “failed to confirm pluripotency of the cells, even though it was in a situation where pluripotency should have been revealed.”

Weissman did find that the isolate of mesenchymal cells that his laboratory tested could form blood, an unexpected ability. However the results, published this year in the Journal of Experimental Medicine3, showed “nothing like the robust contribution to tissues seen in the 2002 paper,” he said.

It's possible that MAPCs are a heterogeneous cell population that exhibits different properties depending on culture conditions and accumulated mutations, says Natalie DeWitt, Senior Editor at Nature. “The difficulty of replicating the 2002 Nature paper is probably that it was not yet possible to define the changes that occurred to the cells as they were cultured for six months.” She cites recent articles by Yamanaka and others as evidence that reprogramming differentiated cells to an embryonic-like state is possible. (Disclosure: Natalie DeWitt is editor at large of Nature Reports Stem Cells.)

On top of the technical difficulties, other problems face the team that pioneered the MAPC studies. Data from Verfaillie and colleagues were called into question after New Scientist reported that the researchers had presented the same data as coming from different experiments in the scientific journal Blood and in a patent application4,5. Paul De Boeck, vice-president for research at KUL, said that an inquiry was being conducted into the matter both at his institution and at the University of Minnesota, where Verfaillie previously worked.

Care to comment? Blog here on The Niche