scigirl

Yesterday I attended a seminar given by Dr. Anthony Furano from the NIH entitled The evolution and impact of L1 retrotransposons in the human lineage.

A brief lesson on transposons here (see this link at <a href="http://www.talkorigins.org/faqs/comdesc/section4.html#transposons" target="_blank">talkorigins</a> for more information).

Transposons are mobile pieces of DNA, found in species from bacteria to humans, that copy themselves and insert their progeny into new places in the genome. There are two types of transposons: transposons that replicate through a DNA intermediate, and those that replicate through an RNA intermediate and thus require a special enzyme, since our cells are not used to making RNA turn into DNA. The latter are termed “retrotransposons.”

Retrotransposons come in two flavors: lines and sines. SINEs are Short INterspersed Elements, and are also called “alu” sequences. These have been useful as markers for the Human Genome Project. LINEs are Long INterspersed Elements, and are also called L1 elements. It is estimated that 30% of the human genome has been generated by L1 activity. They can also cause genomic rearrangements, and therefore these elements have probably played a significant role in evolution, although nobody really knows how yet.

Here’s the anatomy of an L1 element, usually about 6 to 7 thousand bases long:

5’ UTR------ORF1 ---------ORF2-------------3’UTR

UTR stands for “untranslated region,” and when this part of the DNA is made into RNA, only the ORFS, or “open reading frames” will actually contribute to protein sequence. ORF1 codes for an RNA binding protein, and ORF2 codes for the replicase. Both proteins are needed for further replication of this element. A full-length L1 element will contain all four of these features – but many L1 elements get messed up when the replicase is replicating the mRNA back into the DNA (a process that does not normally occur in our nucleus) so you will see fragments of L1s all over the place.

L1 elements are great for confirming evolutionary trees. Why? Well once an L1 element has inserted itself into the genome of a species, it is very difficult to get rid of (they can be ridded from the genome which I’ll talk about later, but it’s hard to do). So the evidence is still there to analyze. Also, the manner in which L1 elements replicate make it easy to determine their age. Young families have more polymorphic inserts than old families. They will often replicate incompletely, leaving partial sequences behind as a marker. Eventually they will no longer have any full-length progeny and thus go extinct.

These LINES can be sequenced and placed into various families. Dr. Furano’s lab did an interesting study. They did a phylogenetic analysis of the L1 families, and obtained a very good tree. By very good, I mean lots of 100% numbers at the branch points – in every run out of 1000, the computer gave that exact same branch point. Ideally we want 100% at all the branches, but for reasons beyond the scope of this thread, it’s hard to get 100% all the time. Anyway, then they estimated the time of each branch split, done by mutation rate analysis, and overlaid their tree on a time line.

So they got something that looked like this (I’m making up the numbers, but this is what it looked like so be aware!):

L1A------&gt;L2A----------&gt;L3, L4
L1A------&gt;L2B----------&gt;L5, L6

35 MYA 25 MYA 10 MYA

(Their tree had lots more branches and different numbers. MYA stands for Million Years Ago.)

(grr I can't get a tree to look good so imagine a split in the tree above, where you see arrows ok!!)

Then they took another evolutionary tree – the current phylogenetic model for primate evolution, derived NOT from LINES, but from fossil record and other things – and superimposed their LINE tree. Guess what – they correlated perfectly.

What do I mean by correlation? Well say the split from the blue monkey to the red and purple monkeys (again I’m making this up, but their data was real) occurred at the same time as the L1A split above (the place with the two stars). This means that all the descendents of one split should contain L2A but not L2B, and the descendents of the other one should contain L2B but not L2A. Guess what – they analyzed dozens of these LINES and they ALL correlated perfectly with our current evolutionary tree for primates.

Can we say independent verification of primate evolution? Slowly, for the creationists. . . “IN-DEE-PEN-DENT…..” Oh and one more point on this analysis – if scientists are so dogmatic about evolution, than why do they continue to test evolution trees with every new weird genetic phenomena they discover? Huh? Huh? (I guess the evilutionist conspiracy meetings aren’t very effective – I keep missing them, so I really wouldn’t know).

Ok enough creationist teasing (for this post anyway). Another interesting finding of this group was that in human evolution, only one L1 family can be replicatively competent at one time. This implies that there is some type of filter that is weeding out the other families. So not only do the LINES have to fight to survive in the host, they also have to fight against other LINES to survive. Sort of like two levels of selection are going on here.

It was discussed that L1 elements may be imposing a genetic load on its host. Why do we think this? Well, first of all, ancestral L1 familes have far fewer full-length elements. Second, full-length ancestral elements are missing from regions of our chromosomes that frequently recombine. So, our genome has tried to get rid of them (via recombination during meiosis). Why would this happen? Well they must have conferred a disadvantage to our genome. Another interesting point is that since the rate of L1 removal is less than the rate of L1 generation, our genomes are getting bigger.

He also pointed out that the only species that does not have L1 elements is a rare form of conifers – which it turns out – doesn’t participate in sexual selection. That’s right – they don’t get busy! Therefore – the only reason we can tolerate L1 elements at all is because we have sex, and eventually we can rid ourselves of the buggers when our sperm and egg cells are being produced.

A question you may have is--why do we even tolerate them at all? Well for one, they are tough to get rid of – probably they are remnants of viruses. And we all know how smart viruses are! But there’s another more interesting reason why we even have these things in our genomes. They may have conferred a survival advantage and played a role in evolution.

At one particular time in evolutionary history, a certain sequence in the L1 element underwent a strong positive selection (our genomes kept it around on purpose). This sequence makes up a “coiled coil” protein domain in ORF1. How do we know about positive selection? By analyzing the ratio of non-synonymous mutations (base pair changes which affect protein sequence) to synonymous mutations (base pair changes which do not affect protein sequence). This ratio is called w, or omega. If w = 1, then the genome is indifferent to the sequence. If w is less than one, negative selection is going on, and if w is greater than 1, you have positive selection. The number he gave for omega here was like 68 (a huge huge positive selection!) Interestingly, this was all occurring right before the split between Asian and African primates. Unfortunately, the speaker would not speculate as to whether there could be causation between the two events. However, his lab is currently doing a huge screening of the types of proteins or genes that this particular domain could have interacted with to affect the evolutionary history of primates. I think those results will be very fascinating.

I found some abstracts from this lab if anyone is interested in reading further.

scigirl

P.S. I have never seen any speaker get this excited about sequence data before – he was very cool!

<a href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=973672 8&dopt=Abstract" target="_blank">Determining and dating recent rodent speciation events by using L1 (LINE-1) retrotransposons.</a>


<a href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=117195 68&dopt=Abstract" target="_blank">Adaptive evolution in LINE-1 retrotransposons.</a>

[ October 04, 2002: Message edited by: scigirl ]

[ October 04, 2002: Message edited by: scigirl ]</p>

lpetrich

Nice article. Though a common creationist rebuttal is that all seemingly nonfunctional genetic material is somehow functional.

scigirl

It doesn't matter if it's functional, in my view, to be a support for the primate tree. Why are the sequences that diverged after time X only found in species that we think diverged after time X?

scigirl

lpetrich

I agree that that asks too much of coincidence, and one does have to ask why some creator does not simply give him/her/itself away by creating something that clearly does not look like evolution.

lpetrich

And the functional bit is about the creationist argument that human and chimpanzee proteins have to be made much alike because of the similar lifestyles of the two species, with similar arguments for the dissimilarity in more distantly related species.

That may be why creationists often argue that the concept of homology (structural resemblance greater than justified by functionality) is a tautology.

At least when they are not arguing that "the creator chose a common plan".

Bubba

Amazing how nicely evolution fits the evidence, isn't it??

Bubba

Thanks for the info.