More Recent Comments

Friday, November 20, 2015

Different kinds of pseudogenes: Unitary pseudogenes

The most common types of pseudogenes are processed pseudogenes and those derived from gene duplication events [duplicated pseudogenes].

The third type of pseudogene is the "unitary" pseudogene. Unitary pseudogenes are genes that have no parent gene. There is no functional gene in the genome that's related to the pseudogene.

Unitary psedogenes arise when a normally functional gene becomes inactivated by mutation and the loss of function is not detrimental to the organism. Thus, the mutated, inactive, gene can become fixed in the population by random genetic drift.

The classic example is the gene for L-glucono-γ-lactone oxidase (GULO), a key enzyme in the synthesis of vitamin C (L-ascorbate, ascorbic acid). This gene is functional in most vertebrate species because vitamin C is required as a cofactor in several metabolic reactions; notably, the processing of collagen [Vitamin C]. This gene has become inactive in primates so primates cannot synthesize Vitamin C and must obtain it from the food they eat.

A pseudogene can be found at the locus for the L-glucono-γ-lactone oxidase gene[GULOP = GULO Pseudogene]. It is a highly degenerative pseudogene with multiple mutations and deletions [Human GULOP Pseudogene]


This is a unitary pseudogene. Unitary pseudogenes are rare compared to processed pseudogenes and duplicated pseudogenes but they are distinct because they are not derived from an existing, functional, parent gene.

Note: Intelligent design creationists will go to great lengths to discredit junk DNA. They will even attempt to prove that the GULO pseudogene is actually functional. Jonathan Wells devoted an entire chapter in The Myth of Junk DNA to challenging the idea that the GULO pseudogene is actually a pseudogene. A few years ago, Jonathan McLatchie proposed a mechanism for creating a functional enzyme from the bits and pieces of the human GULOP pseudogene but that proved embarrasing and he retracted [How IDiots Would Activate the GULO Pseudogene] Although some scientists are skeptical about the functionality of some pseudogenes, they all accept the evidence showing that most psuedogenes are nonfunctional.


19 comments :

Unknown said...

I guess I have a problem with the whole "Unitary pseudogenes are genes that have no parent gene" because probably 99.9999% of genes in existence have a parent gene: they arose from the duplication of an ancestral gene that might or might not still be present in the genome, as opposed to arising de-novo from previously non-coding DNA. Furthermore, there are many examples of processed genes which have existed as functional genes for millions, if not billions of years, and have in recent history become deactivated. Would you consider an olfactory receptor pseudogene a processed or a unitary pseudogene? Presumably processed pseudogenes are DOA.

I find it much simpler to define a unitary pseudogene as a pseudogene that was once a functioning gene that has become deactivated.

Larry Moran said...

I don't know if you are just quibbling or if you really didn't understand the context of my statement. The other types of pseudogenes are formed by copying a functional gene that still exists in the genome. Those genes are called the "parent genes" to distinguish them from the derived pseudogene.

I find it much simpler to define a unitary pseudogene as a pseudogene that was once a functioning gene that has become deactivated.

That would apply to one of the daughter genes of a gene duplication event but we don't call them "unitary" pseudogenes because they are not unitary.

Furthermore, there are many examples of processed genes which have existed as functional genes for millions, if not billions of years, and have in recent history become deactivated.

What do you mean by "processed gene"? Can you give an example of one that has existed for billions of years?

Would you consider an olfactory receptor pseudogene a processed or a unitary pseudogene?

Neither. Most of them are derived from gene duplication events.

Unknown said...

Sorry, my mistake, olfactory receptors are generated through duplication and are not processed genes.

That slightly rephrases my question: Do you consider olfactory receptor pseudogenes to be duplicate pseudogenes or unitary pseudogenes? I'm assuming from your comment you consider them to be duplicate pseudogenes.

Unknown said...

"Neither. Most of them are derived from gene duplication events."

I'm relatively confident that all genes in the human genome where generated due to duplication events. Even GULOP at one time had a paralog.

Larry Moran said...

Do you consider olfactory receptor pseudogenes to be duplicate pseudogenes or unitary pseudogenes? I'm assuming from your comment you consider them to be duplicate pseudogenes.

Of course I do. There's no possible meaning of "unitary" that could apply to the dozens of olfactory receptor pseudogenes in the human genome.

Larry Moran said...

I'm relatively confident that all genes in the human genome where generated due to duplication events.

Now you're just quibbling. What point are you trying to make?

Even GULOP at one time had a paralog.

Really? Where is it? I'm certain that there are ORTHOLOGOUS genes in other species that happen to be functional but I'm not aware of any PARALOGOUS genes in the human genome or any other genome that incudes a GULO pseudogene..

Unknown said...

Well, that makes the distinction between unitary and duplicate pseudogenes rather trivial doesn't it? In the former, the paralog is simply no longer present in the genome, in the latter it is still identifiable.

Larry Moran said...

I notice that you were a post-doc in Mark Gerstein's lab and you are one of the authors on the Sisu et al. (2014) paper. In the last paragraph of that paper the authors (i.e. you) say ...

... we suggest that some pseudogenes may play active roles in genome biology and warrant further investigation. We realize the notion of functional pseudogene is, in a sense, an oxymoron. However, here we focus only on tabulating and enumerating these potential functional candidates.

You looked at 12,358 pseudogenes and discovered that only 12% were "potentially transcribed" (p. 13363). Of those 1,441 (12%) pseudogenes, how many do you think have a biologically significant function? You don't have to give an exact number, just a ball-park estimate. Is it 10, 100, or 1000?

Larry Moran said...

The Susi et al. paper that includes your name in the list of authors describes three categories of pseudogene including ...

.. unitary (unprocessed pseudogenes with an active ortholog in another species).

Do you agree with what you said in 2014?

Unknown said...

You're saying that in the billions of years that life has existed on earth there were no duplication events in the tree leading to the present day GULOP psdueogene in Humans? If I were a betting man I'd wager against that.

I'm not quibbling, either the difference between unitary and duplicate pseudogenes is trivial: being able to identify a "parent" or "sister" gene; or there is something more important (and maybe implicit) in the distinction between the two classes of pseudogenes.

Unknown said...

"Do you agree with what you said in 2014? "

No, I don't. There were actually very lively discussion about this. Unfortunately a camel is a horse by committee.

Unknown said...

"how many do you think have a biologically significant function"

Transcription is not an indication of function. I don't have to tell you that.

Unknown said...

but I do agree with what was said about effective population size and recombination rates.

Larry Moran said...

Why did you find it so difficult to answer the question? Is it because you want to include some non-transcribed pseudogenes in the functional category?

Unknown said...

Not a difficult question to answer. Probably not many are functional. No ulterior motives here. Just quibbling about the definition of unitary pseudogenes.

Larry Moran said...

You're saying that in the billions of years that life has existed on earth there were no duplication events in the tree leading to the present day GULOP psdueogene in Humans? If I were a betting man I'd wager against that.

Zhang, Z. D., Frankish, A., Hunt, T., Harrow, J., and Gerstein, M. (2010) Identification and analysis of unitary pseudogenes: historic and contemporary gene losses in humans and other primates. Genome Biol, 11(3), R26. [doi:10.1186/gb-2010-11-3-r26]

These authors discuss the loss of gene function in members of gene families. These are genes with multiple functional paralogs in the mouse genome but the gene at the orthologous location in the human genome is a pseudogene. They found 47 examples. Then they say,

At one extreme, pseudogenes of GULO, major urinary protein (MUP), nephrocan (NEPN), neurotrophin receptor associated death domain (NRADD), threonine aldolase I (THA1), and [urate oxidase] UOX do not have any closely related paralogs. These genes are particularly intriguing as there are no alternatives with similar sequences and, as such, they represent unequivocal losses of biological functions.

Do you disagree with Zheng et al.? In addition to Mark Gerstein, there are two other authors on the Zheng et al. paper who were also authors on your paper (Adam Frankish and Jennifer Harrow of the Sanger Institute).

The GULO pseudogene in primates first appears about 60 My ago and as far as I know there are no examples of a GULO gene family in any of the primates including those that still have functional GULO genes. What evidence makes you so confident about your bet?

Larry Moran said...

Hmmmm ... I see one of your points. It's possible to envisage a situation where there are two member of a gene family in one species and both of them are selected for function.

If one of them is inactivated in a related species this should count as a unitary pseudogene. I think that's what you are referring to, no?

If so, that makes my statement about olfactory receptor pseudogenes incorrect. There could, indeed, be examples of unitary pseudogenes in that cluster.

However, your suggestion that all unitary pseudogenes could be derived from paralogs (ancient gene duplication events) doesn't make sense since it adds an extra layer of complexity with no evidence to support it and no theoretical basis to propose it.

Unknown said...

Well, I'm not sure what part you're asking me whether I disagree with. I agree these are unitary pseudogenes. They were functional genes that died. But I also believe many olfactory receptors should be classified as unitary pseudogenes. They were genes that were functional for a very long time and died. Being able to identify a paralog or an ortholog has nothing to do with the definition, but can be useful in clarifying the chain of events that led to the pseudogene. What matters is how long the gene was functional. As you said (although you said additional things I disagree with) a "Unitary psedogenes arise when a normally functional gene becomes inactivated by mutation." Simple as that. A duplicate pseudogene arises when a duplication occurs and one copy is DOA (dead on arrival) or shortly thereafter.

Furthermore, I am confident in me bet because I have billions of years on my side. Just because you can no longer identify a paralog in the genome does not mean one never existed. It may be gone, or so divergent that it is unidentifiable. It is highly likely there is a family of enzymes that act on a slightly different substrate as L-glucono-γ-lactone oxidase and share a common ancestor. There are probably other classical unitary pseudogenes which this is easier to do.

Unknown said...

I'm referring to genes that were functional genes for a long period of time and then became non-functional. If you look at the provenance of the term, this is generally what has been described when people give a definition of a "unitary" pseudogene.

Olfactory receptors are an example of what you are referring to. Although they arose from a large number of duplication events, they were functional for a very long time before being lost. Olfactory receptors are referred to as being unitary pseudogenes in many of the earlier uses of the term.

In my humble opinion, the "extra layer of complexity" comes from pedantic definitions and unfortunate nomenclature. Not finding a paralog is often used as criteria for excluding a gene from being a processed or duplicate pseudogene. Furthermore, finding a functional ortholog can also be used to identify unitary pseudogenes. These are important factors to consider when trying to sift though all of the junk in our genomes in order to identify categories of pseudogenes; but it doesn't take away from the fact that the primary purpose of the unitary pseudogene category is to facilitate the analysis of "vestigal" genes.

Regarding my assertions about duplications, the percentage of genes in the human genome with an identifiable paralog is quantifiable (would 15% be [very] conservative?), so there is evidence to support the assertion that many genes, as well as "unitary" pseudogenes, will have paralogs. Unless for some reason you wouldn't count it, if one is referring to the genome of any organism that has experienced a whole genome duplication along its lineage, then the argument could be made that all genes arose from a duplication.

I'm not trying to call anyone out, or unnecessarily quibble. I think unitary pseudogenes are interesting. Categorizing large numbers of one time functional genes that have "died" as either duplicate or processed pseudogenes (for whatever reason) takes away from the evolutionary story that unitary pseudogenes can tell about an organism (or group of organisms).