What Doesn't Kill Them Makes Them Stronger: Bacteria That Live on Arsenic

Yesterday, NASA called a press conference and announced something extraordinary: bacteria that can substitute arsenic for phosphorus in their biochemistry — potentially including the backbone of their DNA.

The bacterium, dubbed GFAJ-1, was found in Mono Lake, California — a highly saline, arsenic-rich lake where unusual biochemistry had long been anticipated. The research team, led by astrobiologist Felisa Wolfe-Simon, reported in the journal Science that when starved of phosphorus and fed arsenic instead, the microbes not only survived but appeared to incorporate arsenic into their molecular structures in ways previously considered biochemically impossible.

If true, the claim would be genuinely revolutionary. All known life builds its biochemistry from the same six elements — carbon, hydrogen, nitrogen, oxygen, phosphorus, and sulfur. A microbe swapping in a seventh would expand the definition of life itself, with obvious implications for the search for life elsewhere in the universe. That is why NASA framed the announcement in astrobiological terms.

The skepticism was immediate. Within hours of the announcement, scientists were picking at the paper's methods online — questioning whether the bacteria were truly incorporating arsenic into DNA or merely surviving on trace phosphorus contamination while tolerating the arsenic around them. Extraordinary claims, the critics noted, require extraordinary evidence, and a press conference is not peer review.

Both reactions were appropriate. The history of science is full of revolutionary claims that collapsed under scrutiny — and a smaller number that survived it. GFAJ-1 deserved the excitement and the interrogation in equal measure.

Later context: follow-up studies published in 2012 found no evidence of arsenate in the bacterium's DNA — the microbes were extraordinarily good at scavenging trace phosphorus, not rewriting biochemistry. The episode became a landmark case study in science communication and the limits of press-release science.