This is part 8 of my review of
The Myth of Junk DNA. For a list of other postings on this topic see the link to
Genomes & Junk DNA in the "theme box" below or in the sidebar under "Themes."
Pseudogenes are the classic example of junk DNA and, as pointed out by many evolutionary biologists, they represent a difficult challenge for Intelligent Design Creationists. It's especially difficult to explain pseudogenes that are located in the same place in different species.
Chapter 5: Pseudogenes—Not so Pseudo After AllChapter 5 is
Pseudogenes—Not so Pseudo After All. This is the chapter where Jonathan Wells takes the standard creationist approach to the problem of pseudogenes—he denies that they exist!
Wells begins the chapter by reminding us that several evolutionary biologists have challenged the IDiots to come up with an explanation for pseudogenes, especially those that are found in closely related species. The usual suspects are quoted: Ken Miller, Douglas Futuyma, Jerry Coyne, Richard Dawkins, and John Avise. All of these challenges are based on solid evidence that most pseudogenes are actually pseudogenes (non-functional, degenerate copies of functional genes). But Wells says, "Yet there is growing evidence that many pseudogenes are not functionless, after all."
Types of PseudogenesThere are three kinds of pseudogenes [
Pseudogenes]. The first category contains genes that used to be functional in our ancestors but currently are non-functional. The best example is the human
GULOP pseudogene that used to encode a key enzyme in the pathway for synthesis of vitamin C [
Human GULOP Pseudogene]. This gene is active in most animals but it has become a pseudogene in primates and, independently, in a few other animals.
The second category includes genes that arise from a gene duplication event followed by inactivation of one of the copies. These pseudogenes tend to be located near their active siblings and they retain most of the features of the original gene except they can't produce an active product. Many of them are transcribed, especially if they have only recently become pseudogenes.
The third category is called "processed" pseudogenes. They arise when a mature mRNA molecule is copied into DNA by reverse transcriptase and the resulting DNA is integrated into the genome. Processed pseudogenes will usually not have any introns and when they integrate they will not be near a promoter. Many of them are truncated because the copying process was not complete. Processed pseudogenes were never able to produce their original functional product (protein or RNA) and they accumulate mutations at the rate expected from fixation of neutral alleles by random genetic drift.
The first two categories of pseudogene will also accumulate mutations once they become inactive. It's a characteristic of pseudogenes that the older they are the more mutations they have. Thus a pseudogene that is only found in chimpanzees and humans will have fewer mutations than one found in monkeys and apes and even fewer that one found in both rodents and primates.
The human genome contains about 20,000 pseudogenes, about the same as the number of genes. Many of these pseudogenes belong to the same family so not every gene has a corresponding pseudogene. About 6,000 of these pseudogenes arose from duplication events and 14,000 are processed. (The first category is rare.)
The pseudogenes in the "duplicated" category tend to be associated with large families of related genes. For example, in the human genome there are 414 pseudogenes in the olfactory receptor gene family [
Olfactory Receptor Genes,
The Evolution of Gene Families]. It's difficult to imagine how any substantial number of these genes could have a function.
As expected, the processed pseudogenes are scattered thoughout the genome because they insert at random. Roughly 2,000 of them are found in introns and this is further evidence that introns are mostly junk. Of course, if a processed pseudogene plunks down in an intron sequence it will be transcribed. Processed pseudogenes tend to come from functional genes that are abundantly expressed in the germ line. That's because the processed pseudogene has to be integrated into germ line DNA in order to be passed on to the progeny. Most of these genes are standard housekeeping genes. For example, there are 1300 pseudogenes derived from ribosomal protein genes.
Jonathan Wells describes the three categories but doesn't explain any of the other things I just mentioned.
Transcribed PseudogenesThe first important section of his chapter is "Transcribed Pseudogenes." He quotes a number of papers showing that some pseudogenes are transcribed in humans, cows, and plants. In humans he claims that about one-fifth of pseudogenes are transcirbed at some time or another. This is probably an upper limit since most workers suggest that only 10% are transcribed.
I'm sure that some pseudogenes are transcibed. Many of the pseudogenes derived from gene duplication events will still be transcribed from active promoters even though they may not produce a functional product. Many of the processed pseudogenes will be transcribed because they are found within a gene (introns) or have fortuitously integrated near a promoter.
'Pseudogenes' That Encode ProteinsThe key question is whether any pseudogenes produce a functional product and that's addressed in the next section: "'Pseudogenes' That Encode Proteins." Wells describes five studies where presumed pseudogenes were found to be genes after all (three in humans and two in fruit flies). Interesting but irrelevant. These genes are not junk. What about the other 20,000 real pseudogenes? Here's what Wells says at this point in the chapter ...
To be sure, only a relatively small proportion of known pseudogenes have been shown to encode proteins. But there is growing evidence that RNAs transcribed from pseudogenes perform essential functions in the cell
Here's where we get to the most important part of the chapter. It's in a section called "RNA Interference."
RNA InterferenceRNA interference arises when one RNA molecule interferes with the expression of another. The easiest example to understand is when part of a gene is transcribed in the opposite direction producing what's called "antisense" RNA. This antisense RNA will hybridize to the functional mRNA and either block translation or induce degradation. In either case, less protein is made.
If a pseudogene is transcribed in the opposite direction then its antisense RNA could interfere with the expression of the normal gene. Wells gives us three examples of this phenomenon: one famous one from snails (1999), one from mouse oocytes (2008), and one from rice (2009). I don't know whether all these results have been confirmed but even if they have it doesn't amount to much. All kinds of strange things happen in biology and the fact that a few pseudogenes might have acquired a regulatory function isn't shocking. What about the other 20,000 pseudogenes?
Pseudogene Enhancement of Gene ExpressionTheme
Genomes
& Junk DNAWell, there's always the possibility that pseudogenes could
enhance gene expression. The next section is "Pseudogene Enhancement of Gene Expression." Jonathan Wells begins this section with a description of the results on the mouse
Mkrn-1-p1 pseudogene. The authors of this famous 2003
Nature paper claim that transcripts from the pseudogene protect the functional mRNA by shielding it from degradation. The title of their paper is "An expressed pseudogene regulates the messenger-RNA stability of its homologous coding gene" (Hirotsune et al., 2003).
Wells devotes three paragraphs to this important paper. There's only one slight problem. This work has been discredited in a 2006 PNAS paper with the title "The putatively functional
Mkrn-1-p1 pseudogene is neither expressed nor imprinted nor does it regulate its source in trans." (Gray et., 2006) Oops!
Wells knows about this 2006 paper because he discusses it in Chapter 8 when he attacks the defenders of junk DNA. In chapter 5 he adds the following remark in parentheses, "Other biologists later challenged the
Makorin-1 pseudogene results, which remain controversial." I think this is the only time when he mentions any legitimate scientific controversy.
There are three other examples of pseudogenes that might have an enhancement function: one from plants and two from humans. I don't know if these studies have been confirmed but even if they have they don't have much of an impact on the possible functions of the remaining 20,000 pseudogenes.
Sequence ConservationThe only significant evidence of widespread functionality of pseudogenes comes from two studies on sequence conservation. Wells covers this in a one-page section on "Sequence Conservation." The studies purport to show that pseudogene sequences are much more conserved than expected if they are really junk DNA. These studies have not been reproduced, as far as I know, and they fly in the face of much evidence to the contrary—evidence that Wells forgets to mention.
The first paper is a review by Balakirev and Ayala (2003). They review the possible functions of some pseudogenes in
Drosophila and mammals. Some of them show evidence of sequence conservation. They conclude that most pseudogenes are probably functional. A study published that same year looked at all the known pseudogenes in the human genome and concluded that 95% of them evolved as though they had no function (i.e. they were not conserved) (Torrents et al., 2003)
The second paper that Wells mentions is Khachane and Harrison (2009). They identify 68 human pseudogenes whose sequences appear to be conserved in at least two other mammals. These are good candidates for functional genes.
The strange thing about this argument is that Wells doesn't believe in common descent so the evidence of sequence conservation really shouldn't have any meaning for him. Nevertheless, he says ...
How odd! As we saw in Chapter 2, Kenneth Miller, Richard Dawkins, Douglas Futuyma, Michael Shermer, Jerry Coyne and John Avise argue that pseudogenes confirm Darwinism because they are non-functional. But if we assume that Darwinism is true and then compare the DNA of unrelated organisms, sequence similarities imply that many of their pseudogenes are functional. So nonfunction supposedly implies Darwinism, but Darwinism plus sequence conservation implies function. When it comes to conserved pseudogenes, it seems, Darwinism saws off the very branch on which it sits.
For the record, if the majority of pseudogenes really are more conserved than expected from random mutations and fixation by random genetic drift then this would, indeed, be evidence that something is going on. Maybe they do have a function we don't know about. I don't think the evidence points in this direction at all—in fact much of the evidence contradicts it. Balakirov and Ayala (2003) are just speculating. I prefer the evidence of Torrents et al. (2003) suggesting that only a small number of potential pseudogenes have a function. This is consistent with the results of Khachane and Harrison (2009).
[If all 1,000 presumed pseudogenes turned out to be real genes then this moves about 0.06% of the genome from the junk category to the functional category. This isn't enough to save the IDiots.]
The Vitamin C PseudogeneFinally, there's a section I haven't covered. It's titled "The Vitamin C Pseudogene." Wells has to address this particular pseudogene because it's the one that comes up most often when evolutionary biologists (e.g. Ken Miller and Jerry Coyne) criticize Intelligent Design Creationism. Here's what Wells says,
The evidence is not as straightforward as Miller and Coyne make it out to be, however, and their argument is ultimately circular. In any case, common ancestry and intelligent design are two different issues, and the vitamin C story would take us on a detour from the issue of junk DNA that's the focus of this book, so the details are omitted here and included in an appendix.
Which brings us to the Appendix: "The Vitamin C Pseudogene."
The main argument of scientists like Ken Miller and Jerry Coyne is not that the
GULOP pseudogene exists. It's that the
GULOP gene and its pseudogene are at the same location in the genomes of all mammals. In the primate lineage this gene is non-functional due to a number of mutations that make it impossible to produce a functional protein. Some of the same deactivating mutations are found in related species such as humans and chimpanzees. This suggests strongly that the non-functional pseudogene was inherited from a common ancestor. How do Intelligent Design Creationists deal with this evidence?
How does Wells respond?
... intelligent design and common ancestry are two different issues. Major ID proponents pointed this out before Miller wrote his book....Although some ID proponents (including me) question universal common ancestry on empirical grounds (as do some evolutionary biologists), intelligent design is not necessarily inconsistent with common ancestry.
I'm not sure what this means. Does it mean that people like Wells are completely bamboozled by this data since they can't refute either the evidence of common descent or the evidence of bad design? Other IDiots, like Michael Behe, only have to explain the bad design?
Jerry Coyne has published a similar argument but Wells attacks him on two fronts. First, he claims that human and chimpanzee Y chromosomes differ by 60 million nucleotide substitutions. If they really have a common ancestor then one would expect much greater sequence similarity. According to Wells, "If similarities in the vitamin C pseudogene are evidence for common ancestry, then differences in the Y chromosome are presumably evidence against it."
The Y chromosome paper is Hughes et al. (2010). Their results show that in orthologous regions of the Y chromosomes the human and chimp sequences are 98.3% identical. However, the chimp and human chromosomes differ in other regions because of large inserts and deletions. This is still evidence of common ancestry.
As usual, Wells is wanting to have his cake and eat it too.
The second attack is based on a number of quibbles. Coyne said that all primates need vitamin C in their diets but Wells points out that prosimians are primates and they can make vitamin C. Furthermore, according to Wells the requirement for vitamin C has only been established in nine species of monkeys. There are 251 other species and we don't know if they need vitamin C. Not only that, Coyne claimed that all primates have the same single nucleotide deletion in their
GULOP pseudogene but Wells is quick to point out that only five primate sequences have been published.
Put that in your pipe and smoke it, Jerry Coyne! I assume that Wells is completely incapable of answering the challenge that's been issued and that's why he resorts to red herrings.
Continuing with this shotgun approach we quickly encounter several other arguments that are designed to distract from the main topic.
- "Miller and Coyne rely on speculations about the motives of the designer or creator that have no legitimate place in natural science." (I hope you turned off your irony meter before reading that.)
- Miller and Coyne have not provided any evidence to justify their claim that the GLO pseudogene is completely nonfunctional.
- (Turn off your irony meter!) Their argument is circular. The similarities in sequence between chimp and human pseudogenes could be due to natural selection. "To break the circle, Miller and Coyne would either have to establish the recent ancestry of humans and chimps on other grounds (but why then bother invoking the vitamin C pseudogene at all?), or they would first have to establish that the vitamin C pseudogene has no function whatsoever (but this is impossible). So their argument not only fails to refute ID, but it also fails to establish that humans and chimps are descended from a common ancestor."
I feel a bit sorry for Ken Miller and Jerry Coyne. If this is the best the IDiots can do then why bother trying to argue with them in the first place?
Thus endeth Chapter 5.
Gray TA, Wilson A, Fortin PJ, Nicholls RD. (2006) The putatively functional Mkrn1-p1 pseudogene is neither expressed nor imprinted, nor does it regulate its source gene in trans. Proc. Natl. Acad. Sci. USA 103:12039-12044. [PDF]
Hirotsune, S., Yoshida, N., Chen, A., Garrett, L., Sugiyama, F., Takahashi, S., Yagami, K., Wynshaw-Boris, A., and Yoshiki, A. (2003) An expressed pseudogene regulates the messenger-RNA stability of its homologous coding gene. Nature 423:91-6. [PDF]
Hughes, J.F., Skaletsky, H., Pyntikova, T., Graves, T.A., van Daalen, S.K., Minx, P.J., Fulton, R.S., McGrath, S.D., Locke, D.P., Friedman, C., Trask, B.J., Mardis, E.R., Warren, W.C., Repping, S., Rozen, S., Wilson, R.K., and Page, D.C. (2010) Chimpanzee and human Y chromosomes are remarkably divergent in structure and gene content. Nature 463:536-539. [Nature]
Khachane, A.N., and Harrison, P.M. (2009) Assessing the genomic evidence for conserved transcribed pseudogenes under selection. BMC Genomics 15:435-449. [PDF]
Torrents, D., Suyama, M., Zdobnov, E., and Bork, P.. (2003) A genome-wide survey of human pseudogenes. Genome Res. 13:2559-2567. [PDF]
The Salem Conjecture is the work of Bruce Salem. He suggested in the mid 1990s on talk.origins that of those creationists who claim professional expertise in evolution, a substantial percentage are engineers.
