Antibiotics, Creationism, and Evolution

I have to admit that creationists are a creative bunch, if not accurate. From the files of the Mad Biologist comes this post about a creationist explanation of antibiotic resistance. It's pretty remarkable. And I hope nobody tells the Coultergeist about this argument... (originally published April 18, 2005).

Google is an amazing thing (so is Gizoogle). I typed in "evangelical" and "antibiotic" and found a "creationist explanation" of the evolution of antibiotic resistance in bacteria. What strikes me is the selective use of facts to support a preconceived notion. While creationists do this all the time, they rarely delve into the scientific literature in any detail. That's what makes this article so insidious: if you don't know the literature in this area, the argument sounds reasonable. They actually do a very good job of describing the molecular mechanisms of antibiotic resistance. As you can imagine, it's when they hit the evolutionary parts, everything goes haywire (and I'll show how in a bit). Essentially, they argue that the spread of antibiotic genes through horizontal transfer means that the antibiotic resistance hasn't evolved, but instead results from a transfer of pre-existing genetic variation (horizontal transfer is just like it sounds--genes are transferred from bacterium to another without reproduction or cell division). The second argument they make is that because antibiotic resistance can be costly in environments without any antibiotics present, these genes will go extinct and hence there's no evolution. So here's the horizontal transfer argument they make:

Bacteria don't appear to be evolving new genes; they are acquiring previously existing antibiotic resistance genes through lateral gene transfer.

The problem is this isn't true. While transfer is a key component in the spread of antibiotic resistance, evolution "of new genes" is taking place. First of all, Staphylococcus aureus (the skin bacterium that can cause "staph" infections) has acquired methicillin resistance genes from a related Staphylococcus species, S. sciuri, which is found on wild animals. What's interesting is that the 'sciuri' form of these genes does not confer resistance to methicillin. The methicillin sensitive form had to be converted into a methicillin resistant form--in other words, evolution happened. Second, resistance to triclosan in S. aureus is due to mutational changes in the fabI locus as well as other genes. There is compelling evidence that triclosan resistance arose through mutation--evolution happened. Third, many of the genes that are transfer horizontally are called 'multidrug efflux pumps', which basically pump nasties out of the bacterial cell. However, these pumps have changed in both what compounds they pump and how well they pump them-again, evolution happened. The counter argument I'm making is that the transferred genes themselves arose through evolution. The supposition that an Intelligent Designer created allelic variants is just daffy, and is required to refute the evidence. Here's the nonsense about evolution and mutation:

Furthermore, a mutation that confers resistance to one antibiotic may make the bacterium more susceptible to other antibiotics.[15] These deleterious effects are what would be expected from a creationist model for mutations. The mutation may confer a benefit in a particular environment, but the overall fitness of the population of one kind of bacterium is decreased as a result of a reduced function of one of the components in its biological pathway. The accumulation of mutations doesn't lead to a new kind of bacterium--it leads to extinction.

There is so much wrong here, I'm going to have to list it:

1. Not all evolution involves speciation. In fact, much of evolution doesn't involve speciation. While the microevolution/macroevolution debate has always struck me as misguided, there is something important in that debate. Populations can evolve, or, in order words, experience changes in gene composition, without undergoing separation into distinct evolutionary entities.
2. The phenomenon of pleiotropy is hardly shocking. The pioneering population geneticist Sewall Wright wrote in Evolution and the Genetics of Populations, "Pleiotropy is ubiquitous." (Pleiotropy is where one genetic change results in multiple phenotypic changes; for example, a mutation that results in slower growth in the absence of antibiotics, but confers resistance to antibiotics is a pleiotropic mutation). These tradeoffs are what gives rise to the phenomenon known as stabilizing selection: selection favors intermediate phenotypes and winnows away the extremes of genetic and phenotypic variation. This is one reason why species and populations usually can be recognized.
3. The notion of global or universal fitness is silly. An antibiotic mutation is more or less fit than the wild-type (the original genotype) with respect to a particular environment. There is no reason to think that antibiotic resistance mutations are destined to become extinct. Furthermore, given the spread of evolved antibiotic resistance (see above), it's not clear to me that antibiotic resistance will become extinct. And a question for our creationist friends: if the rise of antibiotic resistance is due to the transfer of existing antibiotic genes, then why didn't these genes go extinct in the pre-antibiotic era? Maybe it's because these genes were useful in certain habitats (oh, that pesky natural selection).
4. Let's combine points 2 and 3. I did my thesis work on the evolution of resistance in E. coli to an interesting group of antibiotics known as colicins. Here are some take home points. First, exposure to one colicin always resulted in resistance to at least two colicins; pleiotropy is, as Wright said, ubiquitous. Second, the patterns of resistance found in nature are consistent with pleiotropy and inconsistent with a 'one mutation-one resistance' model. Third, different kinds of mutations have different costs and benefits. The least costly mutants (or, in fact, those with no cost in the conditions I measured) were those that conferred the resistance to the most types of colicins. The implication of all of this is that the origin of a colicin resistance mutation (i.e., previous exposure to colicin X) can be decoupled from the maintenance of that mutation (i.e., current exposure to colicin Y). Pleiotropy isn't always a bad thing. In fact, most strains of E. coli turn out to be resistant to most colicins.
5. There is good evidence of the evolution of compensatory mutations. These mutations ameliorate or eliminate the costs of resistance mutations. Once again, resistance mutations are not a dead end.

So I have no idea how to sum this up, except to say if you still think antibiotic resistance follows a creationist 'model', then you're an idiot.