I wrote the following in an online, one on one, debate at Skeptic Times. I am posting this here for an open debate.
---------------------------- HERVís: Evidence of Common Ancestory
Certain types of viruses are able to insert their DNA or RNA into the host DNA. These types of viruses are called retroviruses. By doing so, they are able to take over the machinery of the host cell to make more virus particles. Retrovirus DNA is made up of various genes including an integrase which inserts the viral DNA into the host genome, an envelope gene which makes the outer protein coat of the virus, and various other genes necessary for reproducing viral particles. Also, on each end of the virus are sections called LTRs, or long tandem repeats. These are long stretches of a repeated sequence of DNA. All retroviruses have these characteristics.
A retrovirus has a particular mode of insertion. Different viruses tend to insert into different parts of the host genome. HIV, for instance, prefers to insert into areas with active host genes. These sites, called integration sites, are abundant within the host genome for each virus, numbering anywhere from 500 to 2,000 integration sites per host genome. Also, each integration site is usually between 100,000 to 250,000 DNA letters long (letters being the A, G, T, and C of the genetic code). Taking the low number of 500 and the lower size of 100,000, this limits retrovirus insertion to about 50 million bases in the genome. This would mean that we could expect that one letter of DNA would be used twice every 50 million insertions. This figure shows 3,000 insertions by three different retroviruses. As you can see, the retroviral insertions are spread over all of the chromosomes in a random pattern. Not one nucleotide is used twice. There has never been a retrovirus that has been shown to insert at the same letter in the host genome even 0.001% of the time.
An analogy that helps me to visualize this process is an unabridged dictionary. Letís pretend that each letter of the dictionary (the Aís, Bís, etc.) represent a different chromosome. Letís also pretend that each letter in the dictionary represents a letter in the DNA sequence of the host genome. Now, close your eyes and flip through the pages. Randomly stop on a page, keeping your eyes shut, put your finger down on the page. Open your eyes and record the page number and the letter that you put your finger on. Next, pass the dictionary to someone else and have them repeat the procedure. What are the chances that you would both put your finger on the same letter on the same page? I would think this event is highly unlikely, almost like winning the lottery.
However, not all insertions go well for retroviruses. During rare events only part of the viral genome is inserted or a mutation occurs in one of the viral genes. This prevents an infection of the cell. What results is a cell that is not killed by the virus and a partial viral sequence, or a whole viral genome including a mutation, permanently implanted into the host genome of that cell. If this rare event happens in an egg or sperm, it is possible that that egg or sperm could be involved in the production of an offspring. As you can imagine, the odds of sperm or egg used in reproduction with a partial viral genome are quite low, but given enough time it does happen. What results is an organism with a partial viral sequence, called an endogenous retrovirus (ERV), present in every cell of their body including half of their eggs or sperm. Think of it as a scar left in the genome that is passed on to the next generation when it occurs in an egg or sperm.
Now we have a sequence that is part of the genome and is passed on from generation to generation. After numerous generations we look for ERVís in a personís genome. What we find is that certain people share the same insertion at the same spot in the DNA sequence, that is at the same letter of DNA. How do we explain this? The most obvious reason is that they share a common ancestor. Why donít we ascribe this to two different viral insertion events? Remember that 1 in 50 million odds of two viral insertion happening at the same letter of DNA from above, and the dictionary example? This is where these come in. The odds of an effective viral insertion occurring at the same letter of DNA in two different infections is 1 in 50 million, and this is for your run of the mill, full blown, cell killing infection. What we are talking about now is a rare event of a viral misfire. Not only that, but a misfire that happens in an egg or sperm, and even more improbable a misfire in an egg or sperm that leads to living offspring who themselves reproduce at a later time. This multiplies the chances of two people having the same ERV at the same letter of DNA due to separate viral insertions as being highly, and I mean highly, unlikely. Therefore, we can conclude that they share something like a great, great, great grandparent.
What happens when two different SPECIES share the same ERV at the same letter of DNA? The very same logic applies. Given the improbable event of two separate infections leading to the same ERV the most likely scenario is that the two species share a common ancestor. Taxonomy, through the study of fossils, has come to the conclusion that apes and humans share a common ancestor. Therefore, knowing the implications of ERV production, we should find ERVís at the same letter of DNA in each of these species. This is a prediction made by the theory of evolution. Not only that, but the patterns of similarities should also match cladistics. Cladistics is what many call ďthe tree of lifeĒ which show species branching off from one another. One such clade, constructed through the study of fossils, proposes that humans, chimps, gorillas, and orangutans all share a common ancestor. The first species to branch off were orangutans, the second were gorillas, the third were chimps, and the final branch resulted in humans. This allows us to make very precise predictions. If humans and orangutans share a common ERV at the same letter of DNA, then chimps and gorillas should also have that same ERV at the same letter of DNA because all of these species share one common ancestor. Since orangutans branched off before the other three, we should see ERVís occuring after this branching. That is, there should be ERVís common between gorillas, chimps, and humans that orangutans do not have. Since gorillas split off next, we should see ERVís shared between chimps and humans that are not seen in gorillas or orangutans. In fact, there are seven ERVís between humans and chimps that can only be explained by common ancestory, as well as the other ERVís shared by humans and other apes. Here are a couple maps showing where ERVís occured in the evolution of apes and humans:
Above from: Lebedev, Y. B., Belonovitch, O. S., Zybrova, N. V, Khil, P. P., Kurdyukov, S. G., Vinogradova, T. V., Hunsmann, G., and Sverdlov, E. D. (2000) "Differences in HERV-K LTR insertions in orthologous loci of humans and great apes." Gene 247: 265-277.
Above from: Isolation and phylogeny of endogenous retrovirus sequences belonging to the HERV-W family in primates. Kim HS, Takenaka O, Crow TJ. J Gen Virol. 1999 Oct;80 ( Pt 10):2613-9.
As to the falsification of evolution, if you were able to find a sequence shared by gorillas and humans that was not found in chimps then the theory of evolution would be in serious doubt. Additionally, find an ERV only shared by orangutans and humans and not chimps or gorillas, you would again cast serious doubt on the theory of evolution. However, these potential falsifications have never been observed. Only recently has the human genome been decoded, and even more recently the chimp genome. Soon, the gorilla genome will be complete, so even more ERVís may show up. As more genomes are completed this test can be continually applied as new ERVís are discovered in other primate and ape species, not to mention other non-primate species. Therefore, ERVís are a fine example of a repeatable and falsifiable data set that can be used to test the theory of evolution.
Some have argued that ERVís are not the result of viral infections but instead are the result of a supernatural design process. Firstly, the viral genes found in mammals today are very similar (99% to 75% homology) to viral genes that we see today. This evidence supports the origins of ERVís as viral. No other mechanism, besides viral infection, has ever been observed that results in an ERV. Secondly, the argument could be made that if ERVís are all focused on the design of the organism that a difference in ERVís would make a noticeable difference in design. As it turns out, humans differ in genomic ERV content. That is, some people have ERVís that other people donít.  Since we donít see any difference in design between these groups of people it would seem that ERVís are not necessarily there for design purposes only. However, ERVís have been suspected as playing a role in many functions, such as placental development in mammals. On the other hand, ERVís have also been suspected of causing multiple sclerosis  and susceptibility to cancer. Most scientists have come to the conclusion that ERVís act as random mutation events in that most ERVís are neutral and some have beneficial or detrimental effects on the individual. If they are part of some design process then they are ineffective in causing design differences in most cases. Also, a designer would not be forced to make the pattern of similar ERVís between species follow what is found in the fossil record. That is, a designer would not be forced to follow the rules set forth by common ancestory and the theory of evolution.
In summary, ERVís are a potent test of common ancestory between species. Commonalities between ERVís in separate species is easily explained by simple heredity of a mutation. The chances of an ERV occurring at the same letter of DNA in separate genomes is extremely improbable due to the random nature of retrovirus insertion and the rarity of ERV production. Given that we share seven such ERVís with chimps rules out one single improbable event. This would be similar to winning the lottery seven times in only a few hundred tries. The insinuation by some in the creationist movement that ERVís are the fingerprints of design is not supported, nor is it substantiated by any data. ERVís are random mutations and viral in origin.
 "Retroviral DNA Integration: ASLV, HIV, and MLV Show Distinct Target Site Preferences," Mitchell RS et al., PLoS Biol. 2004 August; 2 (8): e234 (full text)
 "Insertional polymorphisms of full-length endogenous retroviruses in humans," Turner G et al, Curr Biol. 2001 Oct 2;11(19):1531-5 (abstract)
 "Endogenous retroviruses and MS: using ERVs as disease markers," Clausen J, Int MS J. 2003 Apr;10(1):22-8 (abstract)
First, I don't understand that this is not in the Contributor's forum. It certainly merits that status.
quote:As to the falsification of evolution, if you were able to find a sequence shared by gorillas and humans that was not found in chimps then the theory of evolution would be in serious doubt. Additionally, find an ERV only shared by orangutans and humans and not chimps or gorillas, you would again cast serious doubt on the theory of evolution. However, these potential falsifications have never been observed. Only recently has the human genome been decoded, and even more recently the chimp genome. Soon, the gorilla genome will be complete, so even more ERVís may show up. As more genomes are completed this test can be continually applied as new ERVís are discovered in other primate and ape species, not to mention other non-primate species. Therefore, ERVís are a fine example of a repeatable and falsifiable data set that can be used to test the theory of evolution.
There is a problem with this for example,
Curr Biol. 2001 May 15;11(10):779-83. Related Articles, Links
A HERV-K provirus in chimpanzees, bonobos and gorillas, but not humans.
Barbulescu M, Turner G, Su M, Kim R, Jensen-Seaman MI, Deinard AS, Kidd KK, Lenz J.
Department of Molecular Genetics, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461, USA.
Evidence from DNA sequencing studies strongly indicated that humans and chimpanzees are more closely related to each other than either is to gorillas [1-4]. However, precise details of the nature of the evolutionary separation of the lineage leading to humans from those leading to the African great apes have remained uncertain. The unique insertion sites of endogenous retroviruses, like those of other transposable genetic elements, should be useful for resolving phylogenetic relationships among closely related species. We identified a human endogenous retrovirus K (HERV-K) provirus that is present at the orthologous position in the gorilla and chimpanzee genomes, but not in the human genome. Humans contain an intact preintegration site at this locus. These observations provide very strong evidence that, for some fraction of the genome, chimpanzees, bonobos, and gorillas are more closely related to each other than they are to humans. They also show that HERV-K replicated as a virus and reinfected the germline of the common ancestor of the four modern species during the period of time when the lineages were separating and demonstrate the utility of using HERV-K to trace human evolution.
Note, some portion of HERV-K shows closer affinity of Pan and Gorilla as opposed to Pan/Homo. I don't think this falsifies Pan/Homo as a grouping because HERV-K is extremely active (there are novel integrations that are human specific..and presumably novel chimp, gorilla, etc. integrations), they tend to homogenize by gene conversion, and HERVs tend to excise themselves by recombination so that humans and gorillas may share a HERV whereas in the chimp lineage, it was deleted and the deletion fixed. So, I think you would have to look at the specific history of specific HERVs to be able to determine whether a specific association falsifies the current concepts of primate phylogeny or not (I have a paper in review on this which I will link to...if it gets accepted ).
quote:First, I don't understand that this is not in the Contributor's forum. It certainly merits that status.
I didn't approach Percy with the idea of putting this in the Columnist's Corner. It was my choice to have it in one of the other forums.
quote:Note, some portion of HERV-K shows closer affinity of Pan and Gorilla as opposed to Pan/Homo.
Are there other examples other than the one you listed?
As a counter-hypothesis, could the affinity of this HERV-K to the Pan/Gorilla be due to polymorphism? In my post I listed an HERV-K polymorphism in humans. If this same condition was present in the common ancestor of Pan/Gorilla/Human it may cloud phylogenetics. For the HERV-K provirus in the abstract you cited, if it was heterozygous in the common ancestor of gorilla and pan/human it could have resulted in heterozygous conditions in both branches. In pan, it only became homozygous after the human/pan split and disappeared or was non-existent in the human lineage. Of course, the occurence of this HERV-K provirus in each population would have to be rare before the human lineage diverged from the pan lineage. Am I barking up the wrong tree?
This message has been edited by Loudmouth, 10-20-2004 12:23 PM
Here is are reference to human specific HERVs (again supporting Pan/Gorilla)
Barbulescu M, Turner G, Seaman MI, Deinard AS, Kidd KK, Lenz J. Related Articles, Links Many human endogenous retrovirus K (HERV-K) proviruses are unique to humans. Curr Biol. 1999 Aug 26;9(16):861-8.
You are not barking up the wrong tree. The authors hypothesize that the proviral insertions were heterozygous and by chance disappeared in our lineage but became fixed in Pan and Gorilla. I guess the fact that such elements are rare (for inactive HERVs) supports this scenario i.e. as opposed to huge numbers of HERVs showing totally conflicting distribution patterns.
HERVs are difficult to study because they come in many different flavors. There are active HERVs like those of the HERV-K family which are still actively transposing around the genome. There are long dead ones even to the point of hardly being recognizable as HERVs anymore, there are HERVs that are serving critical function like syncytin so their evolution is much more highly conserved than one would expect, some have recombined to form novel elements, some have become homogenized by gene conversion, some are partially deleted because of recombination. The problem is HERVs are defined like cancer. One thinks of them as a single entity or locus as cancer is often thought of as a single disease. The fact is that HERVs make up about 10% of the genome which is far greater than the content of actual human genes. So it is millions of entities just as cancer referrs to a pantheon of diseases.
Then there is the issue of horizontal transfer. HERVs enter the genome by HGT. But then are transmitted vertically. Since different HERV groups entered the genome at different times and then transposed in (some cases) species specific ways, it makes studying their evolution challenging to say the least.
quote:You are not barking up the wrong tree. The authors hypothesize that the proviral insertions were heterozygous and by chance disappeared in our lineage but became fixed in Pan and Gorilla. I guess the fact that such elements are rare (for inactive HERVs) supports this scenario i.e. as opposed to huge numbers of HERVs showing totally conflicting distribution patterns.
From my brief research on HERV's, HERV-K insertions are the youngest insertions. You would then expect heterozygosity to correlate with HERV-K's and not with older HERV's or SINES found in other old world or new world apes, not to mention other primates. The fact that we find heterozygousity in HERV-K's even in humans seems to support this. And as you say, the rarity of HERV-K that support a Pan/Gorilla branch could be explained through fixation after a possible Pan/Human branching. Nice to know that my hypothesis is supported by others as well.
quote:Then there is the issue of horizontal transfer. HERVs enter the genome by HGT. But then are transmitted vertically. Since different HERV groups entered the genome at different times and then transposed in (some cases) species specific ways, it makes studying their evolution challenging to say the least.
Studying ERV's in a single species would be difficult. Wouldn't you say that comparing ERV's in related species helps to elucidate insertion events and transpositions?
But in any case, that is a part of what I am doing in my research and what a large portion of the HERV community does i.e. Jens Mayer, Eugene Sverdlov, et al. The distributions however, tend to focus on HERV-K (with a bit of work on HERV-L to) but most other HERV's and their distributions are not known for non-human primates (until my paper gets accepted )
This message has been edited by Adminnemooseus, 10-22-2004 04:07 PM
Hehe, I ran across that paper during recent research for the above post. Congrats.
From the abstract: "CONCLUSIONS: In general, the evidence supports the hypothesis that the complement had a single origin within basal Uranotheria."
Your work and others have illustrated the usefullness of ERV (and other genomic sequences) in constructing phylogenies and clades. Mayr in a recent paper spoke about the recent "DNA revolution" and how it has supported the Evolutionary Synthesis. Of course, as with anything in biology, there are exceptions to the rule such as the HERV-K insertion supporting a contrarian Pan/Gorilla clade which has to be dealt with and explained. Can't remember who said it, but biology doesn't make sense without evolution, and such is the case with ERV's.
Here is a brand spanking new example of a polymorphic HERV within a species, namely ours,
Genomics. 2004 Sep;84(3):596-599. Related Articles, Links
A rare event of insertion polymorphism of a HERV-K LTR in the human genome.
Mamedov I, Lebedev Y, Hunsmann G, Khusnutdinova E, Sverdlov E.
Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences, 16/10 Miklukho-Maklaya, 117997 Moscow, Russia.
Human endogenous retroviruses (HERVs), which constitute a significant part of the human genome, might have a serious impact on primate evolution. Over a hundred insertions of HERV-K(HML-2) family members distinguish the human genome from other primate genomes. However, only three cases of insertion polymorphisms have been reported so far, all for endogenous HERV-K proviruses. This suggests that some retroviral integrations occurred rather recently in human genome evolution. In this report, we describe a very rare case of true insertion polymorphism of a solitary HERV-K LTR in the human genome. Distribution of the LTR-containing allele was tested in 5 Africans and 83 individuals from three Russian populations. The allele frequency appeared to be relatively high in populations of both European and Asian origin. The detected polymorphic LTR could be a useful molecular genetic marker of the corresponding genomic region.
So it would appear it is a process that is still in action.
COMMON ANCESTRY does not evidence humans; uncommon ancestry has a better chance. Unless someone was expecting humans to be made of foreign material - like their skeletal and biological imprints should be from an ET?
Consider an exam question asking to list the UNCOMMON factors between humans and all other life forms? Variances are not differentials.