Last year: Arsenic life. This year: Chlorine life?

[Note: Some folks don’t like the phrase “chlorine-based life.” I welcome suggestions in the comments for a better shorthand descriptor]

Last year, a team of NASA-funded scientists claimed to have found bacteria that could use arsenic to build their DNA, making them unlike any form of life known on Earth. That didn’t go over so well. (See my two pieces for Slate for a quick recap: #1, #2.)  One unfortunate side-effect of the hullabaloo over arsenic life was that people were distracted from all the other research that’s going on these days into weird biochemistry. Derek Lowe, a pharmaceutical chemist who writes the excellent blog In the Pipeline, draws our attention today to one such experiment, in which E. coli is evolving into a chlorine-based form of life.

As I wrote in Microcosm, scientists have been contorting E. coli in all sorts of ways for years now to figure out what the limits of life are. Some researchers have rewritten its genetic code, for example, so that its DNA can encode proteins that include amino acids that are not used by any known organism.

Others have been tinkering with the DNA itself. In all living things, DNA is naturally composed of four compounds, adenine, cytosine, guanine, and thymine. Thymine, shown in the upper left panel here, is a ring of carbon and nitrogen atoms, with oxygen and hydrogens atoms dangling off the sides. Since the 1970s, some scientists have tried to swap thymine for other molecules, such as the one show to the lower left here. This compound is called 5-chlorouracil, the “chloro-” referring to the chlorine marked here in red. No natural DNA contains chlorine.

A team of German scientists recently published the details of an experiment that has taken them a long way towards E. coli that live only on chlorouracil. They didn’t simply sit down and type out a new sequence for E. coli’s genome–we just aren’t smart enough to make such predictions. Instead, they harnessed evolution. The scientists fed a population of E. coli chlorouracil, supplementing their dieet with a little thymine to keep them from starving to death. Bacteria that picked up mutations that allowed them to use some chlorouracil instead of thymine were favored by natural selection. They lowered the thymine levels to increase the pressure on the bacteria to evolve more.

After five months of this evolution, the E. coli underwent a noticeable change. For one thing, they changed shape. They started out in the normal capsule shapes in the top photo, turned into the worm-like shapes below, and then later became capsule-shaped again. The scientists also found that the bacteria were slurping up chlorouracil. Ninety percent of the thymine in their DNA was replaced by chlorouracil. As Lowe writes, the scientists found that the bacteria were still incorporating some thymine into their DNA through a pathway no one had discovered before (which just shows how marvelously mysterious E. coli remains after a century of intensive study). The scientists shut down this new pathway, and found that the bacteria still grew happily. But now they only had 1.5% thymine in their DNA.

This E. coli is different from ordinary E. coli, with over a thousand mutations that enable it to take up chlorouracil. But it is not quite yet off the grid. If it is fed thymine and no chlorouracil, it can switch back to a conventional way of life. So now we’ll see if scientists can further evolve the bacteria so that they can only live on chlorouracil, and starve on thymine.

I checked in with Steven Benner, a chemist who has raised a lot of concerns about the arsenic-life research last year. What did he think of the new research?

“It looks true,” he said.

Benner also pointed out that using chlorine in DNA is a pretty modest changed compared to what would have been required to substitute arsenic for phosphorus in the backbone of DNA. Swapping in chlorine doesn’t change how the compound reacts with other compounds, doesn’t change its size much, doesn’t change its stability much, and so on.

On the other hand, it appears to have the advantage of being real.

[Update: Rosie Redfield, an outspoken critic of the arsenic-life research, just got back to me with nice things to say as well:

“This looks quite solid to me (I read it quickly). The technology is nice, very appropriate for the project. No red flags at all.”]

21 thoughts on “Last year: Arsenic life. This year: Chlorine life?

  1. I don’t understand why it is being called chlorine-based (nor why the previous story was called arsenic-based). They’re still quite clearly carbon-based lifeforms. Carbon is the core component, forming a backbone of structure and reactability- the base. These chlorouracil-uptake coli are neat, and I wish the researchers the best, but you can’t call it chlorine-based life. Chlorine-uptake life? chlorouracil-uptake? The other lot could be called arsenic-dependent, although the new ones discussed here are not -yet- dependent on chlorouracil

  2. Very interesting … One minor quibble, however; you write “This E. coli is different from ordinary E. coli, with over a thousand mutations that enable it to take up chlorouracil” when it is very unlikely that all of the noted mutations are actually involved in that change. I would guess that many are what might be termed “collateral damage,” and it will take further study to identify the key mutations. As for a better shorthand descriptor, perhaps some play on “When life gives you …”?

  3. I’m surprised you didn’t mention that 5-bromouracil uptake has been used for years as a way to label metabolically active cells (because it is incorporated into the DNA and can be detected using an antibody). This has been used by environmental microbiologists to identify the metabolically active group in a mixed sample for metagenomics.

    Immunochemical Detection and Isolation of DNA from Metabolically Active Bacteria
    Ena Urbach,* Kevin L. Vergin, and Stephen J. Giovannoni
    Applied and Environmental Microbiology, March 1999, p. 1207-1213, Vol. 65, No. 3

    The differences here are 1) a different halogen, obviously, and 2) much higher concentrations. So the biochem strategy is pretty clear cut (grow them on slowly increasing concentrations, select for fast growth with lots of chlorine). It will be interesting to see what mutations are shown to be important in this adaptation, for various reasons, but as your other microbiology inclined correspondents said, this doesn’t seem nearly as contentious of a report.

  4. That’s some of real good works!! Paul’s post enlightens its connection with base alterations. And it might give a new hope to know future predictions of evolution of living organism, and that too starting with the earth’s first creature ‘bacteria’. Its important to know here that due to this alterations the organic compound becomes more lipophilic and any changes brought about in their membrane properties.

    Better to call it ‘chlorine-life’ knowing its future prospects.

  5. Back in the 70’s and probably earlier Bromo deoxy uracil was used to mark newly synthesized DNA from bulk DNA by taking advantage of the fact that the BUdR was denser than deoxy thymidine. As a result, in an ultracentrifuge at equilibrium, the BUdR DNA banded at a lower portion of the tube than normal DNA.

    It was not possible to do this over the long term, but the bugs had the enzymes to incorporate the BromoT as well as normal.

  6. “One minor quibble, however; you write “This E. coli is different from ordinary E. coli, with over a thousand mutations that enable it to take up chlorouracil” when it is very unlikely that all of the noted mutations are actually involved in that change. I would guess that many are what might be termed “collateral damage,” and it will take further study to identify the key mutations.”

    Actually when I spoke to the researchers last week that’s exactly what they said. They’re now going back and trying to work out which mutations confer this ability and are necessary. They’re also trying to make the bug so it can’t survive on T, which at the moment it can if no chlorouracil available

  7. GCACl (pronounced gee-cackle).

    I know nor- is not correct but chlorinorinated sounds better than chlordesmethylated life.


    E. chlori.

    Betting they don’t use “nuchlor mutant”.

  8. As Mike McK & Paul Orwin point out 5-Br-uracil was known to be incorporated into DNA early on. In fact, the early study of the molecular basis of mutagenesis used it to generate mutations at a time before the genetic code was known, in the late 1950s. It was shown to be incorporated into bacteriophage DNA by Dunn & Smith in 1954 (Nature, 174:304). Another halogenated base, 5-Fl-uracil, was used by Benzer & Champe to demonstrate the existence of nonsense mutations in 1962 (PNAS, 48:1114).

  9. Chlorinated gene pool, cool! If they try hard enough they can turn the thymine into hexachlorobenzene. But we may already have that going on in old chemical dumps.

Leave a Reply

Your email address will not be published. Required fields are marked *