Endogenous retroviruses

[Jet Black]

No I still disagree. The initial infection could happily take place in the cells from which the sperm and the oocyte are derived. If this were the case then my scenario stands and it still describes initial infection leading to huge infection in sperm or oocytes.

h2

you're talking about infection of the germ line cells early on in the organism's development?

 

[h2whoa]

you're talking about infection of the germ line cells early on in the organism's development?

Well, in the male, anytime before fertilising a female, yeah. In the female it would have to be very early on (i.e. embryonic) to affect all gametes.

h2

 

[Jet Black]

it's not like males have only got one germline cell which spawns all the sperm.

 

[h2whoa]

No, that's true. But if you've got a viral infection are you really suggesting that only one cell gets infected? Multiple cells get infected and then spread to others, even in locally disease.


h2

 

[Jet Black]

you are looking for a germ line cell that gets infected and then the infection breaks leaving you with an ERV, a much rarer occurance than a normal functioning infection which kills the cell - I haven't heard of any viruses which ultimately don't kill the cell, though granted many retroviruses do sit in the cell for a while. while I accept your point that a number of germline cells could be infected, 50% being left with an ERV seems rather too high. It would indicate either an extremely high failure rate of viral insertion, at the very least 50% if my thinking is correct and every single germline cell gets infected.

 

[Ondoher]

you are looking for a germ line cell that gets infected and then the infection breaks leaving you with an ERV, a much rarer occurance than a normal functioning infection which kills the cell - I haven't heard of any viruses which ultimately don't kill the cell, though granted many retroviruses do sit in the cell for a while. while I accept your point that a number of germline cells could be infected, 50% being left with an ERV seems rather too high. It would indicate either an extremely high failure rate of viral insertion, at the very least 50% if my thinking is correct and every single germline cell gets infected.

All this is still missing the point:

• First we need an infection of a germ cell
• That has to go on to produce offspring
• That offspring has to ultimately become an ancestor of all members of that species.

That's all well and good, there are lots of retroviruses in the world, and there will always be somebody to whom all individuals can trace a lineage. Genetic bottle necks increase this tendency. It gets tricky when you start to explain homology of ERV's between species as due to anything other than common ancestry. Here are the factors:

• Retroviral integration is an essentially random process. With 3 billion basepairs in the human genome, that's lots of places to insert
• The chance of any specific germ cell leading to an offspring is also pretty rare.
• The odds of any specific individual becoming a common ancestor to all members of his species is pretty low.
• That trees drawn from ERV presence in different species would happen to match trees drawn from anatomical data, and pseudogene data, and protein data, and gene expression data, etc. just seems to make it too astounding to consider.

Compare that to common ancestry, where this result would be the natural, expected outcome, and see which explanation is more parsimoneous.

 

[Jet Black]

All this is still missing the point:

• First we need an infection of a germ cell
• That has to go on to produce offspring
• That offspring has to ultimately become an ancestor of all members of that species.

That's all well and good, there are lots of retroviruses in the world, and there will always be somebody to whom all individuals can trace a lineage. Genetic bottle necks increase this tendency. It gets tricky when you start to explain homology of ERV's between species as due to anything other than common ancestry. Here are the factors:

• Retroviral integration is an essentially random process. With 3 billion basepairs in the human genome, that's lots of places to insert
• The chance of any specific germ cell leading to an offspring is also pretty rare.
• The odds of any specific individual becoming a common ancestor to all members of his species is pretty low.
• That trees drawn from ERV presence in different species would happen to match trees drawn from anatomical data, and pseudogene data, and protein data, and gene expression data, etc. just seems to make it too astounding to consider.

Compare that to common ancestry, where this result would be the natural, expected outcome, and see which explanation is more parsimoneous.

oh I know the point, and of course you are correct, I am just enjoying picking biological nits with a proper biologist.... these creationists get so dull after a while.

 

[ed_m]

thanks guys, i've heard enough.

 

[Jet Black]

well when you want to know more, don't be afraid to ask.

 

[h2whoa]

you are looking for a germ line cell that gets infected and then the infection breaks leaving you with an ERV, a much rarer occurance than a normal functioning infection which kills the cell - I haven't heard of any viruses which ultimately don't kill the cell, though granted many retroviruses do sit in the cell for a while. while I accept your point that a number of germline cells could be infected, 50% being left with an ERV seems rather too high. It would indicate either an extremely high failure rate of viral insertion, at the very least 50% if my thinking is correct and every single germline cell gets infected.

Hi Jet, sorry for not getting back sooner but I was little caught up last night in another thread...

The reason I raised the 50% figure is this: each cell that the sperm originate from is diploid, however each sperm is only haploid.

So lets say the RV inserted into chromosome 14 (randomly chosen) of the original cell. Now the are two chromosome 14s in that cell, one from the organisms mother (call it Cr14i), one from the father (Cr14ii). Only one of those chromosomes will be present in each sperm, so statistically speaking that's 50% of sperm contain Cr14i and 50% contain Cr14ii . If the RV was inserted only into, lets say, Cr14i then only the sperm carrying this chromosome will carry the ERV.

However if the RV manages to insert itself into both Cr14i and Cr14ii then every sperm will carry the ERV regardless of which chromosome 14 copy it has.

h2

 

[Loudmouth]

Hi Jet, sorry for not getting back sooner but I was little caught up last night in another thread...

The reason I raised the 50% figure is this: each cell that the sperm originate from is diploid, however each sperm is only haploid.

The problem is that sperm arise from a population of progenitor cells. In order for 50% of the sperm to have the same ERV it would require the RV to insert into the same nucleotide of 50% of the progenitor cells, of which there are thousands. Also, the progentior cells themselves are replaced, so the 50% would only be a transient characteristic of the sperm cell population.

So lets say the RV inserted into chromosome 14 (randomly chosen) of the original cell. Now the are two chromosome 14s in that cell, one from the organisms mother (call it Cr14i), one from the father (Cr14ii). Only one of those chromosomes will be present in each sperm, so statistically speaking that's 50% of sperm contain Cr14i and 50% contain Cr14ii . If the RV was inserted only into, lets say, Cr14i then only the sperm carrying this chromosome will carry the ERV. However if the RV manages to insert itself into both Cr14i and Cr14ii then every sperm will carry the ERV regardless of which chromosome 14 copy it has.
h2

Firstly, Cr14i and Cr14ii, since they are different insertion events, would occur at separate nucleotide positions on chromosome 14. Therefore, they would not be detected as the same insertion event. Secondly, if there are thousands of progenitor cells then less than one percent of the sperm population would be carrying the specific ERV.

 

[Jet Black]

Hi Jet, sorry for not getting back sooner but I was little caught up last night in another thread...

The reason I raised the 50% figure is this: each cell that the sperm originate from is diploid, however each sperm is only haploid.

So lets say the RV inserted into chromosome 14 (randomly chosen) of the original cell. Now the are two chromosome 14s in that cell, one from the organisms mother (call it Cr14i), one from the father (Cr14ii). Only one of those chromosomes will be present in each sperm, so statistically speaking that's 50% of sperm contain Cr14i and 50% contain Cr14ii . If the RV was inserted only into, lets say, Cr14i then only the sperm carrying this chromosome will carry the ERV.

However if the RV manages to insert itself into both Cr14i and Cr14ii then every sperm will carry the ERV regardless of which chromosome 14 copy it has.

h2

indeed, but I was talking about the primary infection event, which isn't going to be that common. i.e. the infection of a germline cell which happens to be fertilised or fertilise, or hypothetically, the infection of the embryonic cell which goes on to make the gonads and gametes at a later stage. Once this happens and the offspring has an ERV copy in every cell of their body, then the odds of it being inherited are 50% as you describe, but then we are dealing with the next stage of the game, which is the establishment of the ERV in the population due to genetic drift.

 

[h2whoa]

indeed, but I was talking about the primary infection event, which isn't going to be that common. i.e. the infection of a germline cell which happens to be fertilised or fertilise, or hypothetically, the infection of the embryonic cell which goes on to make the gonads and gametes at a later stage. Once this happens and the offspring has an ERV copy in every cell of their body, then the odds of it being inherited are 50% as you describe, but then we are dealing with the next stage of the game, which is the establishment of the ERV in the population due to genetic drift.

Hi Jet,

I still think that my argument stands. Once the original sperm/oocyte producing cell(s) become infected, the inheritance pattern would be the same as if the RV was in fact an ERV. Bearing (I never know if that should be baring) in mind that RVs can lie dormant for absolutely ages it could quite happily sit there getting replicated and passed on for years after original infection.

h2

P.S. Just wanted to say that regardless of whether I'm right or wrong, it's great to have a proper discussion!

 

[h2whoa]

The problem is that sperm arise from a population of progenitor cells. In order for 50% of the sperm to have the same ERV it would require the RV to insert into the same nucleotide of 50% of the progenitor cells, of which there are thousands. Also, the progentior cells themselves are replaced, so the 50% would only be a transient characteristic of the sperm cell population.


Firstly, Cr14i and Cr14ii, since they are different insertion events, would occur at separate nucleotide positions on chromosome 14. Therefore, they would not be detected as the same insertion event. Secondly, if there are thousands of progenitor cells then less than one percent of the sperm population would be carrying the specific ERV.

Well don't forget that the insertion events of the viral genome can be highly specific. So yeah, if the insertion events are random, then the chances of being inserted in exactly the same place time and time again would be nigh on incalculable. But if we remember the added factor of directed insertion it becomes much more apparent why it's not so impossible. For example, the RV is never going to insert into highly condensed, non-coding regions of DNA because it would never become activated.

However I do accept that my intial figure may have been a tad optimistic. I agree that probably not every cell would be infected and that there would be some chance insertion-variations.

So basically, I agree that you're right, probably not every progenitor cells will be infected in exactly the same way. But given the nature of viral infections I think that it's not unreasonable to allow for high levels of infection. And allowing for directed insertion the odds of the RV entering the genome in the same way multiple times seems less impossible.

h2

 

[Loudmouth]

Well don't forget that the insertion events of the viral genome can be highly specific. So yeah, if the insertion events are random, then the chances of being inserted in exactly the same place time and time again would be nigh on incalculable. But if we remember the added factor of directed insertion it becomes much more apparent why it's not so impossible.

What directed insertions? There is nothing at the nucleotide level that directs RV's to insert at the same nucleotide in separate infections. There is absolutely no evidence supporting this. At best, there is a preference for viruses to insert into general REGIONS of DNA, and these regions are several thousand base pairs in size. No one, and I mean no one, has observed a virus inserting at the same nucleotide even 0.01% of the time. This figure illustrates how 3,124 separate RV insertions by three different viruses inserted into a cell line. Not one nucleotide was used twice.

For example, the RV is never going to insert into highly condensed, non-coding regions of DNA because it would never become activated.

A minority of viruses do insert into these regions. In the case of HIV, about 25% of the time the virus inserts in areas that are distant from active genes. From this paper:

"For HIV the frequency of integration in transcription units ranged from 75% to 80%, while the frequency for MLV was 61% and for ASLV was 57%." Transcription units are areas with active genes.

However I do accept that my intial figure may have been a tad optimistic. I agree that probably not every cell would be infected and that there would be some chance insertion-variations.

I would call it unrealistic and not supported by any evidence.

So basically, I agree that you're right, probably not every progenitor cells will be infected in exactly the same way. But given the nature of viral infections I think that it's not unreasonable to allow for high levels of infection. And allowing for directed insertion the odds of the RV entering the genome in the same way multiple times seems less impossible.

h2

It is not the high level of infection. A high level infection would kill off vasts numbers of rapidly dividing progenitor cells. What we are talking about is a rare event, a failed insertion that leaves only a part of the viral genome in the host cell. I also talk about your "directed insertion" above.

 

[Jet Black]

Hi Jet,

I still think that my argument stands.

your argument is correct, but I think we are talking about different infection times.

Once the original sperm/oocyte producing cell(s) become infected, the inheritance pattern would be the same as if the RV was in fact an ERV.

yes, but only for those cells that have become infected. In the normal case of an adult being infected, there will be an awful lot of sperm/oocyte producing cells that are not infected. now of those cells that are infected, 50% of the sperm/oocytes they produce will carry the RV, but if only 10% of the cells were infected by an RV, that means 5% of the total number of sperm/oocytes will carry the RV - nowhere near the 50%. for 50% of all sperm/oocytes, every single sperm/oocyte producing cell would have to contain the RV, and the only way I can see that happening is either from parentel inheritance, or the stem cells that will later become the gonads being infected in a developing embryo.

Bearing (I never know if that should be baring)

Bearing is correct

in mind that RVs can lie dormant for absolutely ages it could quite happily sit there getting replicated and passed on for years after original infection.

I can't disagree with that

h2

P.S. Just wanted to say that regardless of whether I'm right or wrong, it's great to have a proper discussion!

me too I think we are both right, just talking about slightly different things.

 

[h2whoa]

What directed insertions? There is nothing at the nucleotide level that directs RV's to insert at the same nucleotide in separate infections. There is absolutely no evidence supporting this.

Ah. Yes. Well. Ahem.

It would be hypocritical of me if I didn't admit when I was wrong. You are quite right. I had got my wires crossed on this point. I thought I remembered something from my virology studies but when I came to look over the notes I couldn't find it. I emailed the Professor who had taught me virology and he confirmed that I must have been having a schizoid episode!

Game, set and match, Loudmouth!

h2

 

[gluadys]

Ah. Yes. Well. Ahem.

It would be hypocritical of me if I didn't admit when I was wrong. You are quite right. I had got my wires crossed on this point. I thought I remembered something from my virology studies but when I came to look over the notes I couldn't find it. I emailed the Professor who had taught me virology and he confirmed that I must have been having a schizoid episode!

Game, set and match, Loudmouth!

h2

You know, it would do us all good to sticky this post to our computers. It is a good model of how to reasonably and politely admit one has made an error. Hey, you don't even have to add the bit about a schizoid episode.

I would sure like to see more posts like this from people who have been shown their appraisal of the facts is wrong.

 

[Jet Black]

I was wrong.

haa haa, "your" stupid you got it wrong ....nyaa nyaaa....

just kidding of course

 

[Jet Black]

Ah. Yes. Well. Ahem.

It would be hypocritical of me if I didn't admit when I was wrong. You are quite right. I had got my wires crossed on this point. I thought I remembered something from my virology studies but when I came to look over the notes I couldn't find it. I emailed the Professor who had taught me virology and he confirmed that I must have been having a schizoid episode!

Game, set and match, Loudmouth!

h2

I tried to find that bit of the discussion, but I couldn't. you weren't thinking of this were you?

But although this concept of retrovirus selectivity is currently prevailing, practically all genomic regions were reported to be used as primary integration targets, however, with different preferences. There were identified ‘hot spots’ containing integration sites used up to 280 times more frequently than predicted mathematically. A recent study of the de novo retroviral integration demonstrated also preference for scaffold- or matrix-attachment regions (S/MARs) flanked by DNA with high bending potential. The S/MARs are thought to be important functional sequences of the genome that anchor chromatin loops to the nuclear matrix subdividing the genome into functional domains. They often neighbor regulatory elements involved in gene expression and DNA replication.

Sverdlov, Eugene D. 1998. Perpetually mobile footprints of ancient infections in human genome. FEBS Letters 428:1-6.