Clearly, only new data will resolve this controversy. Personally, what I would like to see is mass spectroscopic detection of arsenic-bearing nucleotides (both from the cellular pool and from hydrolyzing the DNA) and other arsenic-bearing metabolites that must be present. The nucleotides would be particularly key, as one of the critics (Steven Benner) points out that DNA containing arsenic implies the existence of nucleotide triphosphates with arsenate substituting in the alpha (closest to the sugar) position.
I'm also surprised nobody has jumped in to sequence the bug's genome. While it won't shed any immediate light on the controversy, it would be some easy publicity for somebody. Ion Torrent is already making hay of BGI using their instrument to knock off the German food poisoning bug in 3 days, though unfortunately their marketing department has already decided to emphasize the meaningless number of 2 hours for a run. Even if the bacterium turns out not to employ arsenic, it is quite tolerant of it and having the sequence around would be valuable.
On a lighter note, could we get a better name for the bug? GFAJ-1 just doesn't roll off the tongue. Would "Borgia monolakensis" be a possibility?
The coverage in the journals does have one really depressing angle. In a Nature news item, one researcher says
But some principal investigators are reluctant to spend their resources, and their students' time, replicating the work. "If you extended the results to show there is no detectable arsenic, where could you publish that?" asks Simon Silver of the University of Illinois at Chicago. "How could the young person who was asked to do that work ever get a job?"
This is a disappointing attitude; an issue has been raised in science and needs to be resolved by the scientific method. Disproving the paper, should that be the case, is an important publication. Clearly, Science has a moral obligation to publish quality attempts to rebut a paper in their pages (and indeed, just did this with the XMRV controversy). Furthermore, there is PLoS One, which would certainly publish a well-done paper in the field even if Science rejected it. In any case, as noted above, the bug has a high tolerance for arsenic and that could be a valuable line of inquiry. It would not be the first time the attempts to refute a controversial paper spawned a whole new field of work; note the John Cairns E.coli adaptive mutation affair: a claim was made that mutations favoring a bacterium were occurring at a higher than expected rate, and a lot of work illuminated new details of evolutionary selection on bacteria.