A Blog by Carl Zimmer

Mediocre Poison Eaters And The Imperfection of Evolution

It’s easy to forget sometimes that evolution is always a work in progress. We contemplate the eye or look upon an oak tree, and ask, how could they be any better? Somehow, in those moments of awe, we forget about detached retinas and sudden oak death. The evolutionary race is not in fact won by the perfect, but by the good-enough. And it just so happens that one of the best illustrations of evolution’s mediocrity is unfolding in front of us right now.

This episode of evolution is entirely of our own doing. In 1936, a chemical called pentachlorophenol went on the market. It was hugely popular as a way to preserve telephone poles and lumber against fungi and termites. Unfortunately, it also turned out to be toxic to humans, and once it got into the soil it could contaminate the ground for years. That’s because the molecule–five chlorine atoms decorating a ring of carbon atoms–had not previously existed in nature. Microbes had not evolved to feed on it before. It was as toxic to them as it was to us.

Starting in the 1970s,  however, scientists discovered some microbes that had begun to feed on pentachlorophenol. Pollution-eating bugs are popular in microbiology circles, because they can sometimes be deployed to clean up our messes. So a number of scientists have spent recent years dissecting the pentachlorophenol-eaters. Last year, for example, researchers published the genome of one such species, Sphingobium chlorophenolicum, which had been discovered in pentachlorophenol-laced soil in Minnesota in 1985.

When you first learn how Sphingobium eats pentachlorophenol, it inspires that same awe that eyes and oaks do. It uses a series of enzymes to pick off the chlorine atoms one at a time, like a gorilla removing spines from nettles. And yet, for all the complexity of Sphingobium‘s biochemistry, it does a pretty lousy job of feeding on pentachlorophenol.

Shelley Copley of the University of Colorado and her colleagues have tested out the individual enzymes that the bacteria use. They actually work far slower than typical enzymes involved in breaking down toxins. When they grab onto the molecule, they often lose their grip. Sometimes they grab onto an entirely different molecule instead. And while Sphingobium may be able to eat pentachlorophenol, they are not completely immune to its risks. Expose the bacteria to a high level of the pesticide, and they die.

A look at the genes that encode the enzymes reveals why they’re so mediocre: they’re new to the job. While they all act together like workers on an assembly line, they have different origins. Copley and her colleagues were able to gain some clues to those origins by comparing Sphingobium chlorophenolicum to closely related species that cannot break down pentachlorophenol. They have summed up their current understanding of the evolution of pentachlorophenol-feeding with a diagram, which I’ve reprinted below (click to enlarge). The molecules show pentachlorophenol being dismantled. The microbe’s enzymes are marked in red in each reaction arrow. (Spont. means that a reaction happens on its own–spontaneously.)

The oldest part of this pathway is marked in green. Related bacteria have PcpA and PcpE, and they use these enzymes to break down molecules that are similar to pentachlorophenol at this stage of the reactions. But the genes for the steps marked in blue and yellow were not present in that common ancestor. Instead, Sphingobium chlorophenolicum acquired them after it split off from its relatives.

Source: Genome Biol. Evol. 4(2):184–198. doi:10.1093/gbe/evr137
Click to enlarge. Source: Genome Biol. Evol. 4(2):184–198. doi:10.1093/gbe/evr137

Horizontal gene transfer, as this process is known, is common in the microbial world. Microbes slurp up DNA from dead neighbors, viruses shuttle genes to new hosts, and sometimes microbes even build tubes to inject their genes into other microbes. Scientists became aware of horizontal gene transfer when bacteria started trading genes for antibiotic resistance, rendering wonder drugs less than wonderful. But these cases were relatively simple: a single gene could, on its own, give bacteria better protection against antibiotics.

What’s been happening in Sphingobium is more complicated. Two sets of genes have moved into the bacteria, where they have linked together, as well as to a set of genes that was already there. Together, they took on an entirely new tasks that none of them could have handled before: breaking down pentachlorophenol.

Scientists don’t yet know where those pieces of the pentachlorophenol pathway came from, or what exactly they were doing in older microbes. PcpC, the enzyme in the yellow section, is closely related to enzymes that break down proteins. In fact, PcpC can still break down proteins, although not as well as more specialized enzymes. Breaking down proteins might have been its previous job, and only later did its ability to help break down chlorine-bearing molecules come to the fore.

The genes in this pathway have been continuing to evolve over the past few decades. Natural selection favors the microbes that can grow faster on pentachlorophenol than its competitors. But that competition has not produced any gold medalists just yet. The enzymes still aren’t very well adapted to breaking down this toxic molecule.

Consider the very first step in the pathway, where PcpB picks off the first chlorine atom. Usually, enzymes make molecules less toxic than before. But PcpB does the opposite. It turns pentachlorophenol into the truly nasty tetrachlorobenzoquinone, which you do NOT want to mess with.

There are other cases in which enzymes make molecules more toxic, rather than less. But in those cases–where evolution has had more time–the enzymes are adapted to protect the cell from their toxic creation. The molecule never gets a chance to float away, free to wreak havoc, because the enzyme binds to the next enzyme, carefully handing off the prisoner.

Sphingobium can’t do that handoff. The best it can manage is to have PcpB hold onto the molecule until the next enzyme, PcpD, happens to bump into it. That strategy keeps the nasty tetrachlorobenzoquinone from escaping and killing the microbe. But it slows down the whole process of breaking down the molecule enormously.

Will the mediocre Sphingobium evolve a hand-off? Stay tuned. If it only took a few decades for the microbe to get this far, maybe we’ll witness the next step in our lifetime.

10 thoughts on “Mediocre Poison Eaters And The Imperfection of Evolution

  1. I would vehemently disagree with your assertion that PCP is toxic because it isn’t natural. That’s one of the most absurd assertions in chemistry and fosters a belief in ‘natural = good’

    Every day new chemicals are produced by nature. Often just variations on theme (It is highly unlikely that two starch molecules will be exactly alike, each is its own distinct compound in most cases.) but also due tot he complex interactions of enzymes and environment.

    PCP is toxic for two reasons. Firstly it is not susceptible to the many breakdown pathways it is exposed to; pathways that usually deal with whole classes of compounds, not individual chemicals they were especially evolved to handle. The liver is a remarkable detoxifying organ for example.

    The second reason is it IS susceptible to either damaging cellular machinery or blocking metabolic pathways. The equally artificial hexchlorobenzene is much less toxic not by being more natural, but by lacking the reactive hydroxy group. Hexaflurobenzene is nearly harmless.

    [CZ: Either I didn’t express myself clearly enough, or you misunderstood my point. The fact that PCP is unnatural is not what makes it toxic. It’s part of the reason why PCP doesn’t get broken down in the environment. Let me quote from a 2009 review by Copley: “The persistence of anthropogenic compounds is due to their recent introduction into the environment; microbes in soil and water have had relatively little time to evolve efficient mechanisms for degradation of these novel compounds.” The whole review is free here.]

  2. I don’t believe the author stated that PCP was toxic because it wasn’t natural. He stated that because it wasn’t natural no microbes had evolved enzymes to break it down (because PCP wasn’t around until the 1930s).

  3. @Kudzu I think there is some confusion regarding the wording.
    You were likely referring to:
    “Unfortunately, it also turned out to be toxic to humans, and once it got into the soil it could contaminate the ground for years. That’s because the molecule–five chlorine atoms decorating a ring of carbon atoms–had not previously existed in nature.”
    Where you infer that the first half of the first sentence (toxic to humans) is justified by the second sentence (unnatural), rather than the second half of the first sentence (stays in the ground for years) being justified by the following sentence.
    Hence the point was not “toxic because unnatural” but “could contaminate the ground for years because it is unnatural”.

    Hope that clears up the confusion, I agree wholeheartedly that unnatural and toxic are not the same.

    1. @Andreas

      Your post helps greatly, yes. My point is that many, indeed most, unnatural chemicals are of no issue because living things have an immense capacity to deal with novel chemicals, natural or unnatural. Simply never having existed in nature before is absolutely no excuse.

  4. Are Copley et al planning to do any in-lab evolution to speed up the improvement of Sphingobium’s ability of to break down PCP? Shouldn’t be too hard to do.

  5. Um… well, one of the tenets of evolution is that evolution does not have a goal towards any form of perfection.

  6. Thank you, Carl, for writing about one of the many benefits of bacteria. When most people equate bacteria with disease, it is a huge obstacle for those of us who are trying to study microbes or communicate their overwhelming benevolence. Please write more bacterial posts in the future. Thanks!

  7. What you haven’t discussed is what is going to breakdown the contaminates that are created when the PCP is manufactured. specifically the chlorinated dioxins and dibenzofurans and HCB’s. The kidneys and liver do a great job of absorbing them and then the host prematurely dies. The million plus military personnel that have been exposed to Penta treated ammunition boxes for the last 40 years will live with the effects for years to come. The government won’t talk about this, but it would definitely like have someone create a bacterium that could make their disaster go away. Check out how many of these containers are stored at our Ammunition Depots. Maybe that is a story that you can write about and I can read before my kidneys and liver give up trying to eat one chlorine atom at a time? This is a story that needs to be told now!

Leave a Reply

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