polar bear
© Sean Kilpatrick/The Canadian Press
En isbjørn æder et stykke hvalkød, mens det går langs bredden af Hudson Bay i nærheden af Churchill Man.
Dette er den første af en serie af artikler, der svarer på den omfattende kritik af Darwin Devolves publiceret af Richard Lenski på hans blog Telliamed Revisited. Professor Lenski er måske den mest kvalificerede videnskabsmand i verden til at analysere argumenterne i bogen. Han er Hannah Distinguished Professor for mikrobebiologi [Eng: Microbial Ecology] ved Michigan State University, han er en MacArthur ("Genius Award") Fellow, og medlem af National Academy of Sciences. Han har hundredevis af publikationer bag sig og har også en stærk intereresse for videnskabens historie og filosofi. Hans eget laboratoriums arbejde med udviklingsteori er et centralt fokus i bogen. Jeg er taknemmelig over for Professor Lenski for at tage tid til at bedømme Darwin Devolves. Hans kommentarer vil tillade interesserede læsere hurtigt at vurdere den relative styrke af argumenterne vendt mod bogens tese.

Selv om det ikke var emnet for hans første post, vil jeg begynde med Lenskis diskussion af det eksempel hvormed jeg indleder bogen - isbjørnens genmateriale - fordi det illustrerer nogle princippper, som vil være nyttige, som vi skrider frem. For læsere, som ikke har tid til at læse indtil slutningen, er her et par konklusioner, de kan tage med hjem:
  • Eksperimentel evidens taler stærkt for min konklusion (der bestrides uden god grund af Lenski og andre) at strækt selekterede mutationer i isbjørnegenerne fungerer ved at bryde eller svække ftidligere eksisterende funktioner.
  • En "funktion" ved et protein er et molekylært træk eller en aktivitet på lavere niveau, så som at være et gear eller en kæde. Dette bør ikke forveksles med komplekse fænotypiske træk som "lavere cholesterol" eller "gør organismen lykkelig." At ignorere denne skelnen leder til megen forvirring.

Where We Agree

At the beginning of Darwin Devolves I discuss work by researchers who compared the genome of the brown bear (Ursus arctos) with that of the polar bear (Ursus maritimus). Those species separated from a common ancestor hundreds of thousands of years ago. By analyzing the DNA sequence data, the researchers were able to determine the genes whose selection most strongly adjusted the polar bear lineage to a frigid environment. One of those genes, called APOB, is involved in fat metabolism. As I wrote:
The scientists who studied the polar bear's genome detected multiple mutations in APOB. Since few experiments can be done with grumpy polar bears, they analyzed the changes by computer. They determined that the mutations were very likely to be damaging - that is, likely to degrade or destroy the function of the protein that the gene codes for.
In fact, about half of the mutations in the 17 most highly selected polar bear genes were predicted to be damaging. What's more, since many genes had multiple mutations, I noted that about two-thirds to four-fifths of selected genes had suffered at least one damaging mutation. I used this example to set the stage for the main theme of the book, that Darwin's mechanism works chiefly by degrading pre-existing genetic information, which sometimes helps a species survive.

Echoing blogged arguments by his lesser-known co-authors of the appalling review of my book in Science, Professor Lenski points out (as I repeatedly do in the book) that the computer analysis is a prediction that a particular mutation will or won't be damaging; it is not an experimental demonstration. In other words, the prediction could be wrong. Further, the program categorizes mutations into just three categories: probably damaging, possibly damaging, and benign. (Benign means simply that, as far as the program can tell, no damage has been done to the protein by that change; it does not mean the change is constructive.) Thus, as he stresses, the program is not set up to detect if in fact some new function had been gained by the protein. He goes on to emphasize that the polar bear is superbly adapted to its high-fat diet - much better in that regard than the brown bear. All of which, I happily agree, is true.

Where We Disagree

But then, without benefit of supporting data, Lenski waxes strongly optimistic. He quotes an author of the study and then stresses his own view in bold face:
In a news piece about this research, one of the paper's authors, Rasmus Nielsen, said: "The APOB variant in polar bears must be to do with the transport and storage of cholesterol ... Perhaps it makes the process more efficient." In other words, these mutations may not have damaged the protein at all, but quite possibly improved one of its activities, namely the clearance of cholesterol from the blood of a species that subsists on an extremely high-fat diet.
Lenski is almost certainly wrong about the bolded text. Here's why. In 1995 researchers knocked out (destroyed) one of the two copies of the APOB gene in a mouse model - the same gene as has been selected in polar bears. Although APOB is itself involved in the larger process of the transport of cholesterol, mice missing one copy of the APOB gene actually had lower plasma cholesterol levels than mice with two copies. (Mice missing both copies died before birth.) What's more, the researchers noted that "When fed a diet rich in fat and cholesterol, heterozygous mice were protected from diet-induced hypercholesterolemia."

The researchers admitted they did not know how it all came together - how that effect on the complex cholesterol-transport system resulted from breaking the gene. Nonetheless, there is no ambiguity about the mouse results. Simply by lowering the amount/activity of APOB, mice were protected from the effects of a high-fat diet. Deletion of one copy of the gene may have made the process of cholesterol removal more efficient, as Rasmus Nielsen speculated above about the polar bear, but it did so by decreasing the activity of mouse APOB.

Just to be extra clear about the relevance of the mouse results to the interpretation of the polar bear genome, let me state my reasoning explicitly. Given the experimental results with mice, it is most parsimonious to think APOB is broken or blunted in polar bears. For mice, having only half as much APOB activity protects them from a high fat diet. For polar bears, having mutated APOB genes protects them from a high fat diet. If those polar bear mutations decreased the activity of APOB by half or more, then we might expect a similar protective effect as was seen in the mouse. Given that computer analysis also estimated the APOB mutations in the polar bear as likely to be damaging, it is most reasonable to think the activity of the protein has been blunted by the mutations.

Thus there is no good reason to speculate about possible new activities of the coded protein in the polar bear. Rather, the simplest hypothesis is that the mutations in the polar bear lineage that were judged by computer analysis as likely to be damaging did indeed blunt the activity of the APOB protein in that species - that is, made it less effective. That molecular loss gave rise to a happy, higher-level phenotypic result - an increased tolerance of polar bears for their high fat diet.

The Way to Bet

The caveats mentioned above by Professor Lenski - about how computer-assignment of a mutation as "damaging" is not a guarantee, and that the protein may have secretly gained some positive new function - are correct. He is also quite right to say that without detailed biochemical and other experiments we cannot know for sure how the change affected the protein and the larger system at the molecular level. Nonetheless, computer methods of analyzing mutations are widely used because they are generally accurate. And they do not suddenly lose their accuracy when I cite their results. So, in the absence of specific information otherwise, that's the way for a disinterested scientist to bet. There is no positive reason - other than an attempt to fend off criticism of the Darwinian mechanism - to doubt the conclusion.

The APOB gene is exceptional in having such detailed research done on it. Most other genes haven't been so closely investigated. Nonetheless, in the absence of positive evidence to doubt a prediction for a specific case, the results of the computer analysis should be tentatively accepted for other genes to which it has been applied as well. Skepticism on the matter seems to stem less from the data than it does from reflexive defensiveness. (One of Lenski's co-reviewers actually talked himself into thinking that "it is entirely possible that none of the 17 most positively selected genes in polar bears are 'damaged.'" Now there's a great opportunity for someone to make a few dollars with a friendly wager.)

Lower-Level Functions Versus Higher-Level Purposes

I'd like to highlight one final critical point. Let me set it up with a homey analogy. When I was 14 I worked weekends at McDonald's, and sometimes I'd be assigned to operate the milkshake machine. The machine was broken down each night for cleaning. One of my tasks early in the morning before opening was to reassemble its parts. There were maybe a dozen parts to put together - sprockets, clamps, gaskets, and such. Shakes were very popular back then (mid 1960s) and made many customers happy for a while. Nonetheless, the function of the parts of a shake machine is not "to make people happy." The function of a sprocket or a clamp isn't even "to make a milkshake." Rather, they have lower-level mechanical duties that are subservient to the overarching higher purposes of the systems.

The same is true of APOB. Its function is not "to help polar bears survive," nor even "to clear cholesterol." Rather, it has one or more lower level functions that are subservient to those higher purposes. Thus the fact that cholesterol might be cleared more efficiently in polar bears does not at all mean that APOB hasn't been degraded, any more than breaking the off-switch of a shake machine so that it works continuously throughout lunch hour means some new improved function was added.

In both Darwin Devolves and my Quarterly Review of Biology paper on which it is based, I repeatedly stressed the need to look beneath higher-level, phenotypic changes to associated underlying molecular-level mutations. Did they help by constructing or by degrading what I termed Functional Coded elemenTs (FCTs)? Helpful higher level changes can often be misleading, because they might actually be based on degradative molecular changes. There is every reason to think that's what occurred in the evolution of the examples I cite in Darwin Devolves, definitely including the magnificent Ursus maritimus. The more effective clearance of its cholesterol allows the polar bear to thrive on a diet of seal blubber, but it is the result of a mutation that breaks or blunts APOB.