Signature in the Pseudogenes, Part 2
As we indicated in Part 1 of this series, pseudogenes are remnants of once-functional genes. Since they are segments of a DNA molecule, they are faithfully copied and passed along generation after generation through the millennia of time. For this reason, they serve as excellent markers. They allow us to trace ancestry.
Consider, for example, the lineage shown in the diagram at right. In this example, Species A and B diverged from a single ancestor (red) fairly recently. Since there has been little time for them to evolve, the species are similar to each other.
Species A, B, C are also derived from a common ancestor (blue). That ancestor lived much longer ago, so the three species have had more time to diverge. Thus they may look quite different from each other. Finally, a very long time ago, there lived an even more ancient ancestor (yellow). This one gave rise to all four of these species. Having had lots of time to evolve, this ancestor’s descendents have become increasingly different from one another.
With regard to pseudogenes, the theory of common descent makes a prediction. Let’s say that in sequencing a genome, one finds a specific pseudogene (we’ll call it ‘y’) in Species A, but it is not found in Species B (see below).
If the theory of common descent is true:
- The event which gave rise to pseudogene ‘y’ occurred recently. It could not have been present in the common ancestor highlighted in red in the diagram to the left; otherwise both Species A and B would have had it.
- Since the Species A/Species B common ancestor didn’t have the pseudogene, earlier ancestors could not have had it either.
- Species C and D, because they are derived from those earlier ancestors, would not have pseudogene ‘y’ either.
With the sequencing of many genomes, it is now a straightforward matter to test this hypothesis. This can be done not by examining one or two genes, but by examining hundreds, even thousands. Do all pseudogenes fit into this sort of pattern? We will examine this question by considering one particular subset.
Our sense of smell is made possible through a set of proteins, the olfactory receptors, found on the surface of cells lining the nasal cavity. Airborne compounds bind to these receptors, thereby sending signals to the brain, which then interprets the pattern of binding as a particular odor.
Recently it has become apparent that many mammals have lost some of their olfactory receptor proteins through mutation of the genes which produce them: the mutated genes have been converted into non-functional pseudogenes. It is possible to compare the distribution of numerous olfactory receptor pseudogenes in several primate species.
Let’s first consider 15 pseudogenes1 present in humans but not in chimpanzees. According to the theory of common descent, these 15 pseudogenes have arisen since humans and chimps last had a common ancestor about six million years ago. If this is so, we would predict that none of the 15 pseudogenes will be present in primates believed to have diverged even earlier. As illustrated at right, this is exactly what we find when examining gorilla and orangutan genomes.
What about other olfactory pseudogenes? Do they follow the same sort of pattern? Are they in the “correct” places? Indeed they are – every one:
Six pseudogenes with identical inactivating mutations are found in all four species. Humans and chimpanzees share twelve identical pseudogenes (6 plus 3 plus 3) in common, but humans and gorillas share only nine (6 plus 3). These nine, as predicted, are a subset of the twelve shared by humans and chimpanzees.
Using pseudogene evidence alone, in the absence of any other line of evidence (gene homology, shared synteny, anatomy, etc), it would assemble these species into the same pattern of relatedness as any of the others. Indeed for the 47 pseudogenes studied, not one is out of place. We can tell when in the ancestry each arose relative to the others, and no cases exist where the same pseudogene appears in a manner inconsistent with the proposed lineage. Also recall that this is only 47 pseudogenes within a single family of genes: many, many more have been analyzed and they give parallel results.
As compelling as this pattern is, pseudogene data can also be extended to much more distantly-related species. All mammals, for instance, are predicted to be the evolutionary descendents of egg-laying ancestors. Indeed, the fossil record contains species classified as “mammal-like reptiles” as well as “reptile-like mammals” that blur the distinction between these groups. The evolutionary prediction that mammals are descended from egg-laying ancestors was tested recently using the hypothesis of shared synteny to look for the inactivated remains of a gene devoted to egg-yolk production in the human genome. This gene, called the vitellogenin gene, is used as a component of egg yolk in a wide array of egg-laying species. This research group wondered if it would be possible to find the remains of the vitellogenin gene in the human genome. To help in their search, they employed the prediction of shared synteny.
You may recall from our previous post on synteny that over time, blocks of genes in diverging species are increasingly “broken up” into smaller and smaller blocks. Very closely linked genes, however, can stay together for a very, very long time. Using this knowledge, the researchers:
- Located the (functional) vitellogenin gene in the chicken genome,
- Took note of the gene “next door” to the vitellogenin gene in the chicken,
- Looked to see if this neighboring gene was present in the human genome (it was – a functional version of this gene is found in humans),
- Looked in the same relative spot next door to this gene in the human genome, and
- Discovered the mutated remains of the vitellogenin gene in the human genome in precisely this location.
While it might be possible to present a (strained) argument for the presence of olfactory receptor pseudogenes in humans, the mere presence of the mutated remains of a gene required for making egg yolk in the human genome should give even the most ardent anti-evolutionist pause. That this gene was found using the prediction of shared synteny between humans and chicken only adds to the impact.
Common ancestry is an elegant, parsimonious explanation for the pattern of pseudogenes that we observe, yet many Christians reject common ancestry for theological reasons. The challenges for a non-evolutionary explanation of this data, however, are many:
- Why do humans (or any species for that matter) have so many inactivated genes?
- Why does the distribution of these inactivated genes match precisely the pattern of relatedness (phylogenies) predicted by other, independent criteria? Why are there no “out-of-place” pseudogenes?
- Why are pseudogenes found in the precise locations predicted by shared synteny?
- Why are some of these inactivated genes dedicated to functions that make no sense for the species that harbors them (e.g. defective genes for egg-yolk production in placental mammals like humans)?
To be blunt, if this pattern is not to be accepted, why did God put it in place for us to discover? And if this pattern is not to be trusted, how can anything in genetics be certain? As a colleague once commented, this pattern “would deceive all honest investigators” if in fact it is not accurate.
Many believers are troubled by the idea of humans sharing ancestry with other forms of life (and its attendant theological issues). Consider the opposite, however: suppose that common ancestry is in fact incorrect. The trouble is this: the data doesn’t go away. In this case, one still has to explain why the data looks the way it does: why did God choose to create independent species with this pattern? Even among anti-evolutionists there is no satisfying answer to this question. Time and again, what we see from Christian anti-evolutionary organizations is not an attempt to wrestle with the data, but rather to obfuscate it.
These lines of evidence are becoming more and more widely known among believers and non-believers. If Christians continue to insist on denying the implications of this (very solid) science, we greatly risk setting a stumbling block before our brothers and sisters in Christ, or bringing the faith into disrepute with those whom we seek to reach with the Good News.
1. Of course, there is always a possibility that the same gene may acquire a mutation and become a pseudogene independently in other species. In this case, the inactivating mutations in the two species will be different, and they will be classified as “species-specific” events. It is also possible that species-specific mutations can obscure previously shared mutations (for example, the complete deletion of a previously-mutated pseudogene).
Brawand, D., Wali, W. and Kaessmann H. 2006. Loss of Egg Yolk Genes in Mammals and the Origin of Lactation and Placentation. PLoS Biology 6: 0507-0517. Available free here
Gilad, YG., Man, O., Paabo, S., and Lancet, D. 2003. Human specific loss of olfactory receptor genes. Proc. Natl. Acad. Sci. 100: 3324-3327. Available free here
Zhang, ZD, Cayting, P., Weinstock, G. and Gerstein, M. 2008. Analysis of of nuclear receptor pseudogenes in vertebrates: how the silent tell their stories. Mol. Biol. Evol. 25: 131-143. Available free here