The evolution of brand new genes: Gene birth.

[wattsr1]

Hi Folks,


I saw this one in a very recent Nature:-

Anne-Ruxandra Carvunis, Thomas Rolland, Ilan Wapinski, Michael A. Calderwood, Muhammed A. Yildirim, Nicolas Simonis, Benoit Charloteaux, César A. Hidalgo, Justin Barbette, Balaji Santhanam, Gloria A. Brar, Jonathan S. Weissman, Aviv Regev, Nicolas Thierry-Mieg, Michael E. Cusick & Marc Vidal, Proto-genes and de novo gene birth, Nature (2012), Received 02 November 2011 Accepted 08 May 2012 Published online 24 June 2012


Novel protein-coding genes can arise either through re-organization of pre-existing genes or de novo1, 2. Processes involving re-organization of pre-existing genes, notably after gene duplication, have been extensively described1, 2. In contrast, de novo gene birth remains poorly understood, mainly because translation of sequences devoid of genes, or ‘non-genic’ sequences, is expected to produce insignificant polypeptides rather than proteins with specific biological functions1, 3, 4, 5, 6. Here we formalize an evolutionary model according to which functional genes evolve de novo through transitory proto-genes4 generated by widespread translational activity in non-genic sequences. Testing this model at the genome scale in Saccharomyces cerevisiae, we detect translation of hundreds of short species-specific open reading frames (ORFs) located in non-genic sequences. These translation events seem to provide adaptive potential7, as suggested by their differential regulation upon stress and by signatures of retention by natural selection. In line with our model, we establish that S. cerevisiae ORFs can be placed within an evolutionary continuum ranging from non-genic sequences to genes. We identify ~1,900 candidate proto-genes among S. cerevisiae ORFs and find that de novo gene birth from such a reservoir may be more prevalent than sporadic gene duplication. Our work illustrates that evolution exploits seemingly dispensable sequences to generate adaptive functional innovation.

Figures at a glance

It’s probably worth writing about, although I’m unsure as to how far I can get with it. It looks a bit harder than normal. On the other hand, several things caught my eye while flipping through the article.

There is a bit of background to this. The organisms used in the study are yeast, in particular Saccharomyces cerevisiae. This yeast is a model organism in that it has several attributes that make it most attractive to scientists as a tool for researching cells, genes and evolution. Accordingly, scientists often use it in their studies. Roughly a decade ago, a team of scientists demonstrated that S. cerevisiae arose from a duplication of the whole genome of an ancestral yeast. You can see many of the figures here:-

Yeast genome duplication

- which were taken from this research article:-

Proof and evolutionary analysis of ancient genome duplication in the yeast Saccharomyces cerevisiae.

Essentially what they found was that following the whole genome duplication, there was a massive silencing or deletion of one of the pair of each duplicated gene in S. cerevisiae. However, of the pairs remaining, most underwent what is called “subfunctionalisation”. Genes code for proteins which have several different functions. In subfunctionalisation, each of the pair of genes codes for proteins which kind of split these functions between themselves.

Very few of the pairs of genes that remained underwent what is called “neofunctionalisation” - that is, forming genes which coded for functions not seen in the ancestral organism.

Gene and genome duplication appears to be very important in evolutionary terms and it had been thought that it is the main vehicle by which new genes evolve.

Maybe not so if this latest study is anything to go by.


So it’s worth a bit of a write up as I attempt to make sense of the paper.

I’ll be away for a few weeks at the end of the week. Hence there might be some delays.

Nevertheless, here goes as I plunge into yet another attempt to describe scientists doing science.


To be continued ....



Regards, Roland

[lucaspa]

Roland:

Very good. This describes another method of getting a brand new gene. It's been known for a long time that one of the errors of DNA copying is gene duplication. Where there was one copy of a gene, now there are 2. Since the first copy is still doing the function, the 2nd copy can now be modified to give a new gene with a different function. There are over 13 different types of the protein collagen. They arose by gene duplication over the history of life on earth.

For this paper, people might find the site on open reading frames useful: http://bioweb.uwlax.edu/genweb/molec...anslation.html
What this paper says is that a non-genic portion of the DNA can be transcribed and sent to the ribosome for translation, making a new protein. Notice that their new hypothesis -- transcription of non-genic open reading frames can lead to new genes -- has observational consequences. These observational consequences are also known as "predictions" -- predictions of new data that should be there if the hypothesis is true:

"This evolutionary model leads to the following predictions: (1) the structural and functional characteristics of S. cerevisiae ORFs (for example, length, expression level or sequence composition) should reflect an evolutionary continuum ranging from non-genic ORFs to genes; (2) many non-genic ORFs should be translated; and (3) ORFs that emerged recently should occasionally have adaptive functions retained by natural selection."

The rest of the paper is testing whether these predictions are, in fact, true or whether they are false:
"To examine these predictions, we estimated the order of emergence of S. cerevisiae ORFs "

Now comes the critical evaluation to see whether they did the tests appropriately. That is a big part of the peer-reviewer's job. Since the paper was accepted, the peer-reviewer thought that 1) the predictions were accurate and 2) the tests were done well and supported the predictions.

New" activities of old genes can arise by mutations within the gene. The gene is still there, but now it does something different than it used to.

One famous example of a new gene arose from an insertion mutation within an existing gene. This changes every amino acid downstream of the insertion and gives a brand new protien. In this case, the new protein had the enzymatic ability to degrade nylon. Since it happened in a bacterium living in a waste pool of a nylon factory, there was plenty of nylon around. Suddenly that lucky bacterium had a whole additional food source:
1.Birth of a unique enzyme from an alternative reading frame of the pre-existed, internally repetitious coding sequence", Ohno, S, Proc. Natl Acad. Sci. USA 81:2421-2425, 1984.  Frame shift mutation yielded random formation of new protein, was active enzyme nylon linear oligomer hydrolase (degrades nylon) http://www.nmsr.org/nylon.htm
2. http://home.earthlink.net/~misaak/guide/CB/CB904.html
3.http://home.earthlink.net/~misaak/guide/CB/CB101_2.html

[wattsr1]

Roland:

Very good. This describes another method of getting a brand new gene. It's been known for a long time that one of the errors of DNA copying is gene duplication. Where there was one copy of a gene, now there are 2. Since the first copy is still doing the function, the 2nd copy can now be modified to give a new gene with a different function. There are over 13 different types of the protein collagen. They arose by gene duplication over the history of life on earth.

For this paper, people might find the site on open reading frames useful: http://bioweb.uwlax.edu/genweb/molec...anslation.html
What this paper says is that a non-genic portion of the DNA can be transcribed and sent to the ribosome for translation, making a new protein. Notice that their new hypothesis -- transcription of non-genic open reading frames can lead to new genes -- has observational consequences. These observational consequences are also known as "predictions" -- predictions of new data that should be there if the hypothesis is true:

"This evolutionary model leads to the following predictions: (1) the structural and functional characteristics of S. cerevisiae ORFs (for example, length, expression level or sequence composition) should reflect an evolutionary continuum ranging from non-genic ORFs to genes; (2) many non-genic ORFs should be translated; and (3) ORFs that emerged recently should occasionally have adaptive functions retained by natural selection."

The rest of the paper is testing whether these predictions are, in fact, true or whether they are false:
"To examine these predictions, we estimated the order of emergence of S. cerevisiae ORFs "

Now comes the critical evaluation to see whether they did the tests appropriately. That is a big part of the peer-reviewer's job. Since the paper was accepted, the peer-reviewer thought that 1) the predictions were accurate and 2) the tests were done well and supported the predictions.

New" activities of old genes can arise by mutations within the gene. The gene is still there, but now it does something different than it used to.

One famous example of a new gene arose from an insertion mutation within an existing gene. This changes every amino acid downstream of the insertion and gives a brand new protien. In this case, the new protein had the enzymatic ability to degrade nylon. Since it happened in a bacterium living in a waste pool of a nylon factory, there was plenty of nylon around. Suddenly that lucky bacterium had a whole additional food source:
1.Birth of a unique enzyme from an alternative reading frame of the pre-existed, internally repetitious coding sequence", Ohno, S, Proc. Natl Acad. Sci. USA 81:2421-2425, 1984.  Frame shift mutation yielded random formation of new protein, was active enzyme nylon linear oligomer hydrolase (degrades nylon) http://www.nmsr.org/nylon.htm
2. http://home.earthlink.net/~misaak/guide/CB/CB904.html
3.http://home.earthlink.net/~misaak/guide/CB/CB101_2.html

Neat post lucuspa.

I find gene duplication a fascinating subject and there is a lot of research around on the subject. It appears that the process of gene birth I describe here is very much up and coming, thanks I guess, to advances in gene technology giving researchers the ability to peer even deeper into the genome, and to test ideas.

[wattsr1]

The abstract - 1




Let us begin by looking at the first few sentences of the abstract, posted at the OP:-

Novel protein-coding genes can arise either through re-organization of pre-existing genes or de novo1, 2. Processes involving re-organization of pre-existing genes, notably after gene duplication, have been extensively described1, 2. In contrast, de novo gene birth remains poorly understood, mainly because translation of sequences devoid of genes, or ‘non-genic’ sequences, is expected to produce insignificant polypeptides rather than proteins with specific biological functions1, 3, 4, 5, 6.

In the first sentence the authors, hereafter referred to as Carvunis et al. list the two processes by which it is understood that brand new genes arise. They can arise by rearranging pre-existing genes, or they can arise from areas of DNA where no genes existed previously. They refer to the fairly recent reports in my references area below. See 1) and 2).

They continue by noting that there is a lot of literature dealing with the first mechanism, particularly that dealing with gene duplication, whereby a gene duplicates, and following the duplication, either one of the pair mutates. If this mutation is not neutral, then it will lock the other gene in to maintaining the original function of the genes, while the mutated gene wanders off “exploring” the space of mutations against the constraints of the external environment, as it accumulates more and more mutations. Again the researchers defer to references 1) and 2) for descriptions of this.

However, the birth of genes from regions of DNA where no genes existed previously is poorly understood. There are reasons for this. Often (I think), cellular translating machinery will translate regions of DNA between actual genes producing strings of amino acids (polypeptides) that have no function. They are meaningless. This introduces two problems for their study. One is to identify them, given that genes are often determined by looking at some function they might have. That is, even with existing technology, identifying such meaningless polypeptides has been difficult. The second problem is that a cell has many proteins whose job it is to go around cleaning out rubbish. That is, such meaningless polypeptides don’t necessarily remain in existence for that long, thereby denying their identification and investigation. The authors cite references 1), 3), 4), 5) and 6)

And while these non coding areas might produce polypeptides of no significance for the cell’s current situation, in terms of evolution, they might be very relevant. In essence, it seems, not only do genes in a cell code for proteins, but bits and pieces of non gene regions also get coded, for a bit, and it’s in these regions that evidence for the emergence of new genes is being found.

To be continued ...




REFERENCES

1)

Diethard Tautz & Tomislav Domazet-Lošo, The evolutionary origin of orphan genes, Nature Reviews Genetics 12, 692-702 (October 2011)

Absract
Gene evolution has long been thought to be primarily driven by duplication and rearrangement mechanisms. However, every evolutionary lineage harbours orphan genes that lack homologues in other lineages and whose evolutionary origin is only poorly understood. Orphan genes might arise from duplication and rearrangement processes followed by fast divergence; however, de novo evolution out of non-coding genomic regions is emerging as an important additional mechanism. This process appears to provide raw material continuously for the evolution of new gene functions, which can become relevant for lineage-specific adaptations.

2)
Henrik Kaessmann, Origins, evolution, and phenotypic impact of new genes, Genome Res. 2010. 20: 1313-1326.’

The article for this can be found online here:-
Origins, evolution, and phenotypic impact of new genes

3)
François Jacob, Evolution and Tinkering, Science, New Series, Vol. 196, No. 4295. (Jun. 10, 1977), pp. 1161-1166.

This is something of a classic paper and it’s online at this link:-
http://www.gvsu.edu/cms3/assets/6D25...0tinkering.pdf

4)
Adam Siepel, Darwinian alchemy: Human genes from noncoding DNA , Genome Res. 2009 October; 19(10): 1693–1695.

The paper is online here:-
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2765273/

5)

Konstantin Khalturin,*Georg Hemmrich,*Sebastian Fraune,*René Augustin*and*Thomas C.G. Bosch, More than just orphans: are taxonomically-restricted genes important in evolution?, Trends in Genetics, Volume 25, Issue 9, 404-413, 28 August 2009

Abstract
Comparative genome analyses indicate that every taxonomic group so far studied contains 10–20% of genes that lack recognizable homologs in other species. Do such ‘orphan’ or ‘taxonomically-restricted’ genes comprise spurious, non-functional ORFs, or does their presence reflect important evolutionary processes? Recent studies in basal metazoans such as Nematostella, Acropora and Hydra have shed light on the function of these genes, and now indicate that they are involved in important species-specific adaptive processes. Here we focus on evidence from Hydra suggesting that taxonomically-restricted genes play a role in the creation of phylum-specific novelties such as cnidocytes, in the generation of morphological diversity, and in the innate defence system. We propose that taxon-specific genes drive morphological specification, enabling organisms to adapt to changing conditions.

6)
Benjamin A. Wilson and Joanna Masel, Putatively Noncoding Transcripts Show Extensive Association with Ribosomes, Genome Biology and Evolution, 2011.

And advance online copy can be found here:-

http://gbe.oxfordjournals.org/conten....full.pdf+html

[wattsr1]

Paper is online here


http://ccsb.dfci.harvard.edu/web/exp..._2012_ePub.pdf

Grab your copy quick, because I don't know how long these things remain up.

[wattsr1]

Continuing ....


Let’s look at the next few sentences of the abstract:-

Here we formalize an evolutionary model according to which functional genes evolve de novo through transitory proto-genes4 generated by widespread translational activity in non-genic sequences. Testing this model at the genome scale in Saccharomyces cerevisiae, we detect translation of hundreds of short species-specific open reading frames (ORFs) located in non-genic sequences. These translation events seem to provide adaptive potential7, as suggested by their differential regulation upon stress and by signatures of retention by natural selection.

In essence, this article is about the researchers testing a model for the evolution of new genes in the non coding regions of DNA. I will get to their model in an upcoming post, but basically it’s the opposite of what happens to genes, when genes “die”.

For the benefit of YECs, and putting it very simply, a chromosome is a humungously long stretch of DNA, along which genes are to be found. Genes are, in part, made up of chemicals called bases, and these define the way other chemicals called amino acids are to be strung together to make polypeptides. A polypeptide is just a string of amino acids. Different polypeptides chemically bind to each other to make functional protein which is used by the cell. There are whole set of proteins within a cell which, among other things, allow the cell to make other protein as well as clean up polypeptides that are made, but serve no use. Between genes on a chromosome are vast stretches of DNA that essentially do very little, or so it would seem. Even within DNA there are stretches of DNA that don’t define a protein and it’s a part of the job of a cell’s machinery to ensure that these regions don’t allow their corresponding amino acids to get incorporated into functional protein and that they too, end up getting cleaned up during the process of generating functional protein.

In this case, it’s not the unused DNA within a gene that is of importance here. Rather it’s the vast stretches of DNA between genes which might turn out to be highly significant when it comes to the evolution of new genes.

The researchers defined a model based on pre-existing research which provided evidence for recently evolved human genes from non coding regions in the common homonid ancestor of humans and chimps. This research is cited in my reference 4) in the previous post. Their model is diagramed and described in figure 1 here, and I will get to it later.

What the researchers found were many Open Reading Frames (ORF) in these long, non coding regions. A gene must have a start point and an end point for the machinery which “reads” the gene in order to determine which amino acids to join to make a protein. The DNA from this start point, through to the termination point constitutes what is called the “reading frame”. An ORF then, is a stretch of DNA that has a start point but does not have a termination point.

Not only did they find hundreds of these ORFs in the DNA between genes, but they also found that they were species specific, and of varying lengths. There was already mounting evidence for them being translated sometimes, and conferring perhaps some kind of adaptive potential to the cells they belonged to and they cite reference 7 for this. I can only copy the abstract and so am unsure as to what the research paper actually describes. I suspect it shows that translation of non genetic regions can give rise to some kind of “quasi” proteins which allow an organism to survive in the event of the main gene undergoing a catastrophic mutation. In other words, mutation and selection are playing a survival role well outside of the conventional domain of the gene and so the suggestion is that new genes can indeed evolve from these regions.

It gets even more interesting. If we look at the next sentence from the abstract, we read:-

In line with our model, we establish that S. cerevisiae ORFs can be placed within an evolutionary continuum ranging from non-genic sequences to genes.

That is, these bits of DNA that get translated, range in length and complexity from mere ORFs, through to functioning genes. It’s as if, within S. cerevisiae they might be witnessing a range of ORFs that code for some kind of nonsense polypeptide which quickly gets removed once it is formed, to ORF’s that code for a polypeptide providing some kind of selective advantage, to ORF’s that cease to be such, but incorporate termination points, and hence are genes in their own right.

It’s the evidence that positive selection is acting on some of these ORFs that also piqued my interest. Furthermore, these are species specific, and the researchers have been able to place these into some kind of evolutionary continuum (a nested hierarchy, as well as the functional continuum just mentioned).

... that’s if I’m understanding it correctly. Let us see what happens as I continue ...


REFERENCES

7)

Daniel F. Jarosz,* Mikko Taipale,* and Susan Lindquist, Protein Homeostasis and the Phenotypic Manifestation of Genetic Diversity: Principles and Mechanisms , Annual Review of Genetics
Vol. 44: 189-216 (Volume publication date December 2010)

Abstract

Changing a single nucleotide in a genome can have profound consequences under some conditions, but the same change can have no consequences under others. Indeed, organisms can be surprisingly robust to environmental and genetic perturbations. Yet, the mechanisms underlying such robustness are controversial. Moreover, how they might affect evolutionary change remains enigmatic. Here, we review the recently appreciated central role of protein homeostasis in buffering and potentiating genetic variation and discuss how these processes mediate the critical influence of the environment on the relationship between genotype and phenotype. Deciphering how robustness emerges from biological organization and the mechanisms by which it is overcome in changing environments will lead to a more complete understanding of both fundamental evolutionary processes and diverse human diseases.

[lucaspa]

However, the birth of genes from regions of DNA where no genes existed previously is poorly understood. There are reasons for this. Often (I think), cellular translating machinery will translate regions of DNA between actual genes producing strings of amino acids (polypeptides) that have no function. They are meaningless.

To try to help explain this, let me say that genes have "start" and "stop" sequences. So there are regions that are not transcribed because they lie between the "stop" sequence of one gene and the "start" sequence of another. Also, these regions of DNA are often very important: they are the regulatory regions that help dictate when the gene is transcribed and when it isn't.

But apparently the idea that there are different reading frames -- giving different amino acid sequences -- is a fairly new idea. BTW, I got the paper at Nature itself:
http://www.nature.com/nature/journal...ture11184.html

So, if these different reading frames either start or end in the non-genic portions, that means a a different polypeptide would be translated by the ribosome. These polypeptides would be shorter than most proteins coded by genes, but activity is not dependent on length. For most enzymes, the active site is only 5-6 amino acids long.

And notice the importance of natural selection. Those polypeptides that have any type of selective advantage are going to benefit the individuals they are in. So individuals having that ORF and translating it are going to be selected. Over the course of many generation, you are going to get a population all of whom ttranscribe and translate that ORF. At this point, the ORF is becoming a new gene.