Family resemblance?

Why We Don't Look Like Chimps

Ann Gauger October 2, 2015 3:30 AM

Pictures showing the gradual transition of an ape-like creature to a human one are part of the cultural Gestalt. They are so ubiquitous that people use the idea for jokes. You've probably seen the picture describing the evolution of man from a bent-over ape to a bent-over office worker on the computer. But do such pictures represent the truth about our history?

Let's consider just one part of the transition that would be required -- to go from an ape-like face to a human one. The portraits of our supposed ancestors are in museums everywhere, and on the web pages ofNational Geographic -- the latest one, H. naledi, can be found there now. But whether or not that reconstructed face is a part of our history, the question remains: Is there enough time for us to have acquired by purely, Darwinian natural processes the differences that make us look human as opposed to ape-like?

A recent paper purports to tell us how it happened, at least from a genetic perspective. Prescott et al. report the regulatory changes that might have caused the change in morphology (the overall shape) of our faces. On the whole it's an interesting paper with a lot of hard work involved.

The title, "Enhancer Divergence and cis-Regulatory Evolution in the Human and Chimp Neural Crest," may sound a little cryptic to the uninitiated. Let me provide some background and define some terms. Enhancers are regulatory sequences found in the DNA near the gene(s) they regulate. They act by "enhancing" the expression of those nearby genes. They are also called cis-regulatory elements, hence the "Enhancer Divergence and cis-Regulatory Evolution" part of the title.

Neural crest cells come from a particular part of the early-developing brain, and they migrate out into the embryo to form the face and other structures in the body. Thus they have a major role to play in embryonic development of the face. Mutations that affect neural crest cells can cause major cranio-facial abnormalities.

The popular model is that changing enhancers is the way new morphological features evolved, changes such as from an ape-like face to a human one.

Now for the paper. Prescott et al. began by inducing stem cells of both chimp and human lineages to develop into neural crest cells. (They mostly used in vitro pluripotent stems cells (IPSCs), which are not of embryonic origin. Unfortunately, they did use some human embryo-derived cells to verify that their IPSCs were behaving normally.) From the IPSCs they identified enhancer-like sequences specific to neural crest cells. They then checked for patterns of gene expression driven by those enhancers that differed between chimp and human cells. Having identified those, they then used several methods to examine whether those enhancers did in fact act in a species-specific way in the neural crest. They also checked whether or not the genes they had identified caused human cranio-facial abnormalities when mutated. Some did.

Essentially, they looked for differences between humans and chimps in their enhancer sequences upstream of genes involved in neural crest development, and found them.

How convinced am I that their results are valid? They say they identified at least 208 enhancers that work together cooperatively to regulate neural crest gene expression. That I accept, though some of the details may be wrong. 208 is a reasonable number of enhancers to be involved in neural crest specification and behavior. They say that there are probably other enhancers they didn't identify, and the enhancers they did identify will no doubt have effects in other tissues. Fine. They also report having found multiple enhancer elements upstream of the same gene, each with a positive or negative effect on gene expression that work together. None of this is a surprise -- it confirms the idea that gene regulatory networks are complex and cooperative, and that changing nucleotides in an enhancer (a cis-regulatory element) can change its effect on nearby genes.

My difficulty is with their claim that anything like this evolved. There is plenty of use of the word evolution in the introduction and discussion sections of this paper. It's actually not surprising that they use evolution so often as an explanatory concept. Evolution is the dominant explanation right now for how things got to be the way they are, so of course these differences in enhancer sequences are explained by evolution...right?

Let's check the logic. Premise: evolution happens by changing enhancer sequences in a species-specific way. Observation: there are species-specific differences in enhancers. Conclusion: evolution is responsible for these species-specific changes.

See the problem? That's called circular reasoning. If you accept the premise, then the conclusion is automatic because the two statements are actually saying the same thing. If you don't accept the premise, of course, you will be unconvinced by this argument. Thus the paper assumes Darwinian evolution without demonstrating that it had anything to do with it.

So let's ask the question. Could naturalistic processes guided only by natural selection have accomplished any of this? Changes to the expression of one gene or another can affect morphology -- changing the expression of the gene ALX1 affects the size and shape of finch beaks, for example. Getting one regulatory change to an enhancer is notimpossible. But there are multiple enhancers controlling multiple genes involved in facial morphology (see the paper in Nature by Lamichhaneyet al.), and to get coordinated regulation of multiple genes, multipleenhancers must have changed in a coordinated fashion.

Is that possible? Durrett and Schmidt published a paper examining how long it would take to have two coordinated mutations (one inactivating and the other activating) take place in an evolving hominin population. They found it would require in excess of 100 million years. Obviously, the hominin population did not have that long to wait for regulatory change. We supposedly diverged from chimps six million years ago.

Nowhere in this paper is it demonstrated that anything like this regulatory network could have evolved step by step in the time available. All it establishes is that our faces are different from chimps because we have different regulatory sequences. In fact, I don't know of anywhere it has been demonstrated that regulatory changes on this scale could bethe product of evolution.

In the meantime, we do know something that can coordinate change on this scale. It's called intelligence.