Arsenic-based life finding fails follow-up

Tests see no evidence that microbe uses element in cellular machinery

The controversial claim that one microbe can use arsenic in its cellular machinery is mired in scientific quicksand after scientists attempting to duplicate the finding have come up empty-handed. Though the microbe in question clearly thrives in the presence of the usually toxic substance, there is no evidence that the bacterium requires arsenic to live or incorporates the element in its DNA, researchers report in a paper posted February 1 at

The original study began when researchers led by Felisa Wolfe-Simon, a NASA astrobiology research fellow, cultured a microbe now known as GFAJ-1 from eastern California’s Mono Lake. The lake has an unusual aquatic chemistry rich in carbonates, phosphorus, arsenic and sulfur. Wolfe-Simon and her colleagues starved GFAJ-1 of phosphate, a combination of phosphorus and oxygen that’s an essential building block of life, and force-fed the critter arsenate, an almost identical arrangement of arsenic and oxygen.

It’s the chemical similarity of the two elements that makes arsenic such an effective poison. Cells can be tricked into absorbing arsenate, but the stuff doesn’t quite fit into the molecular machinery. That makes it unsuitable as a building block for DNA or cell membranes, jobs readily taken on by phosphate.

Using arsenate in the place of phosphate is “like putting a round peg in a square hole,” says microbiologist Jim Cotner of the University of Minnesota in St. Paul.

Yet Wolfe-Simon and her colleagues reported that GFAJ-1 thrived when deprived of phosphate and fed only arsenate. “We report the discovery of an unusual microbe, strain GFAJ-1, that exceptionally can vary the elemental composition of its basic biomolecules by substituting [arsenic] for [phosphorus],” the team wrote in their December  2010 online report in Science.

The scientific community was astounded and perplexed. The notion that arsenate could do the work of phosphate was astonishing, says chemist Steven  Benner of the Foundation For Applied Molecular Evolution in Gainesville, Fla. Chemically speaking, “so much of what we think is true would have to be false,” he says. And Wolfe-Simon and her colleagues didn’t do some experiments deemed obvious by other researchers, such as attaching a radioactive tag to the arsenate and locating exactly where it turned up in GFAJ-1’s DNA.

Now researchers led by microbiologist Rosemary Redfield of the University of British Columbia in Vancouver have tried to replicate the original growth experiments. Redfield, her colleague Marshall Reaves of Princeton University, and others grew GFAJ-1 in a test tube. After confirming with a genetic test that they had the right microbe, the researchers couldn’t get GFAJ-1 to be fruitful and multiply until they added a dash of glutamate, an amino acid that Wolfe-Simon did not use in her own experiments.

Then the team tried growing GFAJ-1 with no additional additives. The microbe didn’t grow nearly as much as Wolfe-Simon and her colleagues had reported. But she and her colleagues did note that the culture might have been contaminated by a little phosphate. So Redfield and her team added a sprinkling of phosphate, a comparable amount to what Wolfe-Simon’s team thought might have been in their culture anyway.

GFAJ-1 then grew much better. In fact, it grew in densities similar to those Wolfe-Simon and her colleagues had reported for GFAJ-1 when arsenic was added. And adding or removing arsenic from the cultures made no difference in growth, Redfield and her team report.

“At this point the discussion is essentially over,” says Benner, who was not involved with the work.

Nonetheless, Redfield, who reported much of her work on her research blog as it was being conducted, then extracted and purified GFAJ-1’s DNA. The samples did contain trace amounts of arsenate, but not in the ratios one would expect if the microbe had incorporated arsenate into its cellular machinery. More likely, says Redfield, is that GFAJ-1 can tolerate a bit of arsenic here or there without any serious effects.

“You can grow a bug in arsenic-rich media and you will see some arsenate,” says geobiologist Tanja Bosak of MIT. But that does not mean the arsenate is in the DNA, she says.

Wolfe-Simon, who says she can’t comment in detail until Redfield’s results appear in a peer-reviewed journal, wrote in an email that her original paper never actually claimed that arsenate was being incorporated in GFAJ-1’s DNA, but that others had jumped to that conclusion. “As far as we know, all the data in our paper still stand,” she wrote. “Yet, it may take some time to accurately establish where the [arsenic] ends up.”

More Stories from Science News on Humans