Recall a few weeks ago, when I was discussing how a change in regulation in a single gene in Streptococcus pyogenes led to a huge change in phenotype? (I know you do--you probably have it memorized.) A new study shows a similar phenomenon's occurred during human evolution, and probably is the cause of much of the differences between humans and apes:
Scientists have known for three decades that humans and chimpanzees share 99 percent of the same genes, but they have been at a loss to explain what causes the two to be so obviously different physically, behaviorally and mentally.Now a team of geneticists from Yale University and the University of Chicago have evidence indicating that humans experienced a high rate of mutations in key master genes that turn arrays of other genes on and off.
In many cases the changes in these master or regulatory genes wound up endowing the other genes with new properties--something like a conductor making a four-piece band play like a symphony orchestra. This, they say, may be the process that permits rapid evolutionary changes in body structure and function.
"It involves very little effort to create a huge effect," said the U. of C.'s Yoav Gilad.
Francis Collins, director of the National Human Genome Research Institute, said the study helps explain why humans and chimps can have genes that are nearly identical but perform different operations. It is part of the unfolding story of human evolution, he said.
- Log in to post comments
From the story quoted:"humans and chimpanzees share 99 percent of the same genes"
I know that you don't specialize in the hairier end of the biosphere, but I wonder if you might know off the top of your head...
While I know that the 99% number is a rough figure, I was wondering if it includes only genes that are either identical or code for identical proteins. Specifically, if a gene codes for the same protein in humans and chimps but has different introns does it get counted as one of the shared genes?
I seem to remember an Scientific American story about how some introns might serve as regulatory elements through RNA interference to shut off other genes.
Either way it is fascinating how, as you said, little changes yield big differences, even at such distant ends of the evolutionary shrub of life as S. pyogenes and great apes.
Sorry to bring in math, but my physics-teacher mind wants to verify a hunch. Is it valid to argue that evolution is not a linear process? For example, in x years we should expect y genetic changes to happen? I read this article to mean that at times evolution is more like a step function, with sudden widespread genetic changes as x years pass by. So humans ended up several steps above chimpanzees because of these genetic arrays being modified? (Sounds like computer coder talk ...)
I ask because the anti-evolution windbags carp about how gradual genetic change could not have resulted in prokaryotes evolving into humans. The chimp-human dichotomy would then be a great example of evolution actually not being gradual at times.
I suspect (and stand to be corrected as I only have one undergrad genetics course) that while certain forms of mutations are regular, whether or not the changes are conserved in the population would depend on whether the species was moving into an new niche or staying in a stable one. Much of human evolution coexists with periods of climate change and geological changes (the Great African Rift formation). The theory of 'punctuated equilibrium' (that most evolutionary change happens during certain periods after ecological changes or mass extinctions and the rest of evolution being more of a fine tuning to the environment) has a lot of support if I am not mistaken.
Other forms of mutations like neutral mutations are more regular.
In short, evolution can be somewhat non linear but I would suspect that the best answer against gradualism not being able to explain evolution would be that so much time has been involved. Multi-celled organisms are so late on the scene that just about every cellular mechanism had literally billions of years to develop everything else is fairly simple after the basic cell functions of multi-celled eukaryotes develop. One great difficulty of people with anti-evolution beliefs (and all humans in general) is thinking about the vast numbers involved in single celled organisms reproducing (sometimes exponentially when conditions permit) over vast stretches of time or in thinking about how much complexity bacteria and then eukaryotes had ready to build on by the time they became easily visible in the fossil record.
Hi Apesnake,
I think the 99% figure is based on DNA-DNA hybridization experiments. The better the match between two strands of DNA, the more tightly they cling together; the more tightly they cling together, the hotter the temperature has to be before they'll dissociate. So experimenters threw a bunch of human and chimp DNA in a tube and watched to see what temperature the double-stranded hybrid DNA "melted" at. So this figure (I've seen 98.4% to 99%) is based on DNA differences rather than protein differences, but it doesn't tell us anything about which sites are different, exons or introns.
There might be a more recent study that has looked at exactly which sites in the two genomes differ. I know of a 2002 paper that suggests that there's actually about another 3% difference between chimp and human due to deletions and insertions, which don't show up in a DNA-DNA hybridization study. A lot of these insertions and deletions are in noncoding DNA, but that's all I know. Maybe somebody out there can help me out with more information on this?
You and wheatdogg are both right about gradualism (I think). Mutation rate does seem to stay (roughly) constant for a given gene over time, though it does appear to change from one group of organisms to the next, for reasons that, AFAIK, are not well-understood. And, yes, whether or not those changes get *kept* depends critically on what sort of change they produce and what selection is doing at the moment. Using statistics to pick out genes that are under strong stabilizing selection or selection for change is a hot topic right now, as the paper Tara cites makes clear!
Neutral changes accumulate in a relatively "clock-like" fashion in many cases. My sense-- though this isn't my area of expertise-- is that there's no reason to expect changes at the protein level to accumulate linearly. In fact, it seems to me that they're much more likely to occur in clumps or bursts. Punctuated evolution at the morphological level is definitely visible in certain fossil lineages.
The most recent resource that I know of here is the Sept 1, 2005 issue of Nature, devoted entirely to the chimp (including the first chimp fossils). Our genes are 96-99% identical, a small difference, but as the French say, vive le difference! What number you get depends on the method you use for comparison, but every method rates the chimp-human similarity as far closer than any other animal.
It would be interesting to see if anyone has done a comparison between common chimps and bonobo chimps. Speciation seems to be possible with only a separation of pre existing alleles, before any new mutations (I don't have a reference for that, it's just what I have gleaned from magazines like American Scientist and Scientific American, or is that the other way round). If that is the case animals like lions and tigers (which produce fertile offspring but are considered different species because their offspring suck at being ether lions or tigers) might be extremely similar. It would be interesting to compare rate at which species diverged once they stop interbreeding.
Just to emphasize for those who might be wondering, the difference in the numbers, 96-99% (you might find 95 in some sources) is not a measuring error. These people are measuring different things. For instance, you get the lower numbers if you include indels.
Don't know about bonobos specifically, although I would suppose that since they and common chimps are part of the same branch, that both would be equidistant from humans, barring some slight statistical differences. I do know that in the triad chimp-human-gorilla that chimps are more closely related to humans than chimps are to gorillas. Indeed, gorillas are also about equidistantly related from chimps as they are from humans. Not identically of course, since we don't expect constant divergence rates or identical selective forces between lineages.
I'd also be very interested in seeing a bonobo-chimp comparison. Or a comparison between the bonobo and humans. Does anyone know if those comparisons have been made?
B. Spitzer says:
Surfing around on the Web (not for porn this time) I found THIS. The bonobo-common chimp geneome difference is about 0.7%, which seems to be large considering the chimp-human distance is only about twice that, but there it is.
The fact that chimps are closer (in amount of shared genes) to humans than to gorillas and the fact that this is surprising to most people is a good example of how human judgement needs to be checked against the evidence. While it makes sense that gorillas (vegetarians living on the ground) and chimps, (omnivores spending some of the time in trees) would have had to vary significantly since their last shared ancestor, all we see is two types of hairy primates walking on four limbs and make the assumption that they must be closer than chimps and humans.
I think that even people who understand evolution feel that the buffer of time and cognitive development isolate us from the reality that humans are not just descended from apes but actually are bipedal apes. (Cool and sometimes clever apes though, plus we are better looking than other great apes IMO. Most of us anyway - and with the possible exception of the orangutan.)