Slightly more than a year ago, the media was abuzz with news of a novel bacterium, culled from an arsenic-rich lake bottom in California. While all known life forms require phosphate to survive, scientists studying the microbe claimed that GFAJ-1 cells could use arsenate in place of phosphate as a nutrient and genetic building block.
“There were so many red flags in the original paper,” said Rosemary Redfield of the University of British Columbia, who spearheaded one of the efforts to repeat the experiments. “The two great big flaws both arose from lack of attention to issues of contamination.”
As described in the original June 2011 Science paper, NASA researcher Felisa Wolfe-Simon and colleagues grew GFAJ-1 cells in media, dubbed AML60, which mimicked the chemical composition of Mono Lake, the original source of the microbe. When they added arsenic, but not phosphate, to the mixture, they reported cell growth (1).
“The problem is that the medium had a background phosphate level of 3 µM,” said Redfield. “That is in fact very significant contamination and they should have known that those low amounts of phosphate would seriously compromise their results.”
The background phosphate levels could have been enough to sustain cellular growth whether or not the arsenic was present. When Redfield’s team repeated the experiment with a sixth less phosphate contamination, they found that while the microbe survived in the presence of arsenic—indicating a high tolerance for the often-toxic chemical—its growth was dependent on the presence of phosphate, not arsenate (2).
In the original research paper on GFAJ-1, the scientists also isolated DNA samples from cells grown in the presence of arsenate. They separated the DNA on an agarose gel and then used a NanoSIMS chemical imaging device to determine the ratio of arsenic to carbon in the gel band.
“Every molecular biologist knows to take that gel slice and purify it,” said Redfield. Otherwise, contaminants from the gel will be analyzed alongside the DNA. But Wolfe-Simon’s team analyzed the complete gel slice.
When Redfield’s team repeated the experiment, introducing further purification steps, the observed levels of arsenate never exceeded the detection limit—the numbers for arsenate in DNA samples were similar to the numbers seen in a blank water sample. Redfield’s new analysis, as well as a study by a group in Switzerland which also failed to find a dependence of GFAJ-1 on arsenic (3), were published online on July 8 in Science Express.
“There’s a long list of everybody who failed in this story,” said Redfield. “But I don’t think blame can really be attached at any level. Scientists are people, and we fail.” The researchers were likely too invested in their hypothesis to see potential weaknesses in their study, reviewers may have been poorly chosen or didn’t read the paper closely enough, and the news media got swept up in the story, said Redfield.
The authors of the original paper are still holding out that their results are valid, according to interviews with reporters over the past week. They are continuing work on how the microbe interacts with, and potentially relies upon, arsenate.
- F. Wolfe-Simon, J.S. Blum, T.R. Kulp, G.W. Gordon, S. E. Hoeft, J. Pett-Ridge, J. F. Stolz, S.M. Webb, P. K. Weber, et al. A bacterium that can grow by using arsenic instead of phosphorus. Published Online December 2, 2010. Science 332: 1163-1166
- M.L. Reaves, S. Sinha, J.D. Rabinowitz, L. Kruglyak, and R.J. Redfield. Absence of detectable arsenate in DNA from arsenate-grown GFAJ-1 cells. Online in Science Express July 8, 2012.
- T.J. Erb, P. Kiefer, B. Hattendorf, D. Günther, and J.A. Vorholt. GFAJ-1 is an arsenate-resistant, phosphate-dependent organism. Online in Science Express July 8, 2012.