Recombination: Retrovirus Sex

There was a lot wrong with Micheal 'Lilo' Behes description of HIV-1 in last years book, 'Edge of Evolution'. There is a lot of discussion in the Evilutionist community as to what constitutes a Creationist, but there are a few mistakes Creationists continually make when trying to argue against evilution, especially regarding 'mutations'.

Despite Lilos acceptance of common descent and educational training as a biochemist, he made the same childish mistakes Creationists make when they are wrestling with 'mutation':

Because of the difference in mutation rates HIV has actually experienced about ten thousand times as many mutations as would a comparable number of malarial cells. The very many copies of HIV in the world would be expected to contain almost every imaginable kind of mutation. As one study put it, "Each and every possible single-point mutation occurs between 104 and 105 times per day in an HIV-infected individual."

Behes focus in 'Edge' with HIV-1 was just that-- a generic Creationist obsession with point mutations, and how they are all bad or unremarkable (except the ones that are good-- those were inserted by Jesus, personally).

Yes, a great source of diversity in HIV-1 is caused by mistakes in reverse transcription.

Another source of HIV-1 mutation is APOBEC3G-- a part of your innate immune system that tries to screw up retroviruses by plopping in uracil where there is supposed to be a cytidine. More point mutations.

But those arent the only sources of genetic variation in HIV-1. Retroviruses have sex. Kinda.

If Lilo had spent 30 seconds researching HIV-1 before he wrote 'Edge', he might have decided to put less of an emphasis on point mutations, and more than one sentence on that 'other sorts of mutation' (or maybe he would have realized how stupid it was to put HIV-1 in 'Edge' at all).:

In addition to all those point mutations, enormous numbers of insertions, deletions, duplications, and other sorts of mutations would occur as well.

That one sentence downplays how vital recombination is for HIV-1s genetic diversity. HIV-1 could actually be described as a diploid organism. Every virus contains two copies of the HIV-1 genome, even though it technically only needs one. Reverse Transcriptase can hop back and forth between the two genomes, creating chimeric viruses-- just like you are a chimera of your mom and your dad.

Sometimes these hopping events are very subtle:

If both of the genomes packaged within a virus are the same, hopping back and forth wont make 'drastic' changes. A lot of hopping occurs within my gene of interest, env, and we see lots of inserted/deleted chunks of genome, which changes the size and structure of the envelope gene-- the target for neutralizing antibodies. Not only do these 'tiny' changes add to the diversity of an already moving target for your immune system, some labs have found that the shorter envelope genes are actually the ones transmitted better (we didnt find it with our patients, but that doesnt mean its not so).

So these insertions/deletions might be subtle, but theyre still a big deal.

But just because the two packaged genomes are the same, doesnt mean there cant be HUGE genomic changes:

Unequal crossovers can lead to gene duplications (or deletions!). That is exactly what happened with Vpr and Vpx in HIV-2-- Vpx is just a copy of Vpr. Gene duplication and divergence is a classic route of evolution (look at the different kinds of antibodies we can make because of gene duplication and divergence!). Its also possible that Vpu in HIV-1 is the result of a same kind of event, but it has diverged so far from its parent sequence, we have no idea where it came from :(

Here is where retroviral sex becomes a huge problem-- When HIV-1 infects a cell, it normally down-regulates the receptor HIV-1 needs to infect cells. That means its basically peeing on the fire hydrant. THIS CELL IS MINE!!! It doesnt want to share with another virus, but that doesnt always work out. If an individual is infected with two kinds of HIV-1-- say, Subtype B and Subtype C-- its possible that both a B and a C virus infect the same cell. A B and C viral genome can be packaged into a progeny virus:

What if Virus 1 has already developed resistance to multiple reverse transcriptase inhibitors, and Virus 2 has already developed resistance to multiple entry inhibitors? Virus 3 is now resistant to both kinds of drugs. Statistically-impossible-totally-not-gonna-happen-through-point-mutation events are just a hop/skip/jump away. Literally.

Once again-- be very, very glad Mr. Lilo is not an HIV-1 researcher.

Tags

More like this

Very cool, I had no idea HIV did stuff like this. Is this kind of recombination common in other viruses, or is it something HIV is unique in coming up with?

Any retrovirus can do it!

BUT! Some retroviruses pair up their genomes as they are leaving the host cell nucleus (like MLV). They are more likely to pair two 'like' genomes together than a retrovirus that pairs genomes up at the plasma membrane (HIV-1).

The last bit of your post might give the erroneous assumption that the recombination of the viruses automatically leads to progeny with the best of both parents. Clearly it's more subtle than that.

From my much more rudimentary understanding of what goes on here, sometimes they'll recombine and make for a stronger virus. Sometimes they'll recombine to make a weaker one. The important thing is that those relatively uncommon stronger recombinations are the ones that end up reproducing more.

By Jason Dick (not verified) on 13 Oct 2008 #permalink

Damn. Diploid viruses? Viral pr0n?

I didn't know any of this (not that that's saying much). You're a great communicator. Thank you.

I vaguely recall there being something about certain types of hepatitis only infecting cells that are already infected with one of the more common type. Would that have anything to do with this?

Yeah, Jason, the likelihood is that any particular recombination event is not likely to be good for the virus. But lots of viruses are being made, so...

Actually, retroviruses are really good for gene therapy - they stably put genes in cells, and the cells don't need to be actively dividing for them to do this (most of the cells in your body are not actively dividing). But you don't want the virus to insert its full genome into you - then you're infected. So the first gene therapy systems used mutated virus - not good enough - the virus could simply revert to its normal form. Second generation systems actually separated some of the viral genes so they weren't on the same piece of RNA. Not good enough - recombination came in and made full length virus. Now we're into fourth generation systems that separate the virus into three or four different pieces.

I guess the point is - if the selective pressure is strong enough, life can do almost anything. Viruses (and especially retroviruses) just do it real quick.

You are so ditzy and clueless about HIV, it is pathetic to watch.

First, HIV has 9,000 base pairs, which makes it, genetically, one of the dumbest and smallest "live" entities on planet earth. A grain of wheat is millions of time more complex.

Second, when you prattle on about mutation rates of this virus, you further embarrass yourself. Here's a clue by 4: mutations that don't alter phenotype, nor alter the function (please study this word), don't freakin' matter. Do you not get this, you clueless, ditzy, Sarah Palin-esque, grad school geek?

Endogenous retrovirus HAVE NO FUNCTION. They merely waste space in the cellular genome. HIV is a stupid, trivial, unworthy retrovirus (arguably exogenous), that cannot do the things you claim it does, because it has a whopping 8 or 9 genes. It replicates and nothing more. It has no mode of locomotion, it doesn't spray toxins, it can barely exist, what?, 48 or 72 hours outside of a cell.

Please quit the lab and get a sales job at Bloomingdales or something.

Ky Sanderson

By Ky Sanderson (not verified) on 13 Oct 2008 #permalink

Ky Sanderson,

It is actions like this that utterly ruin any chance you have of getting a date. We can only hope your species of derangement dies with you.

Having gotten that (#8) out of my system, I have a question for our host.

ERV, to what extent does the exchange of DNA between virus and cell---whenever and however it occurs---affect the virus itself?

I always thought most viruses that integrate into the host DNA take bits and pieces from other genes when they depart...

Aside from your characteristic disdain and arrogance, I don't see how you extend the edge of evolution as outlined by Behe. Admittedly, I'm a stupid creationist by your standards. But folks don't become smarter when you call them stupid, nor do they magically disappear (Praise Science).

So, my stupid question is this:

How does what you talk about here differ significantly from what can be accomplished by Mendelian genetics in terms of the scope of what can be accomplished by evolution?

To #11:

Did you just read the same post I did?

If you weren't paying attention, it just gave you a beautiful, succinct, and highly interesting answer to the 'stupid question' you just asked.

Since you were apparently asleep for that part, it not only replicates mendelian genetics, it extends it substantially as well, by providing a mechanism for systematically generating new genes (by duplication, as often seems to have happened in higher organisms at much, much lower rates). I.e., the virus has a mechanism with which to readily build additional complexity and functionality, as well as providing for gene mixing.

It's missing the concepts of dominance and recession (so far as this article is concerned), but that's mostly a detail. Genetic diversity in its offspring, biased towards functional offspring, is driving HIV's "success".

/es

ES @ 12. So it doesn't require 2 parents to have the right combination of genes. Seriously, what is the advantage here? First you have to infected with 2 strains. With Mendellian genetics--two parents with the right combo have to breed. You're right, I continue to be stupid even after reading your 'intelligent' answer.

TCS: First you have to infected with 2 strains.

No.

Each virus contains two copies of its genome. If they're the same, cool, it works just like a regular virus. If they're not, then it gets to mix and match.

There are, implicitly then, two ways for a virus to get mismatched genome copies. First is for a cell to be re-infected with a different virus. The virus however, attempts to shut this down as part of infection, so it isn't as common as it might otherwise be. The second is for the transcription process to make errors (as described). Any erroneously transcribed genomes are likely to be paired up with correctly transcribed partners before getting their protein coats and being sent out into the world.

In (rare?) cases, two erroneously copied genomes may be packaged together to form a virus.

In the first two cases (one or two good genomes in the virus), the virus should stay infectious and produce viable "offspring". In the last case, it'll be more of a crapshoot and infectivity is likely to be poor.

In all infectious cases, however, the virus should reproduce itself well, and in addition, generate a significant amount of genetic variability in its offspring. This is key -- every organism "wants" (evolutionarily-speaking) to maximize the genetic variability of its offspring while maximizing the number of viable offspring. In small/parasitic organisms, this maximizes the probability of avoiding host defensive systems; in large organisms (e.g., mammals) this maximizes the probability of disease/parasite resistance.

/es

Ky Sanderson:

function (please study this word)
*snip*
HIV is a stupid, trivial, unworthy retrovirus (arguably exogenous)

Do you know what exogenous, or even endogenous, actually means? Maybe you should study those words.

By Thomas S. Howard (not verified) on 16 Oct 2008 #permalink

Thanks ES for explaining things a bit more clearly. However, my initial question has not been answered:

How does what you talk about here differ significantly from what can be accomplished by Mendelian genetics in terms of the scope of what can be accomplished by evolution?

Let me help you atheists (and rare theistic evolutionists), with how to construct a response.

It will start with something like this:

"What ERV talks about, accomplishes the following beyond Mendelian genetics, and extends what evolution can accomplish by: x" (And x is the variable. Your response. Please fill in the blank.) And, "This extends the edge of what evolution can accomplish beyond the assertions of Behe by: y." (I think you get the picture). ERV, calling something 'stupid' is not an intelligent argument (it is adolescent or pre-adolescent). You must actually extend the edge of evolution outlined by Behe in order to make a valid argument. You have not done so in this post.

TCS:

ERV, calling something 'stupid' is not an intelligent argument (it is adolescent or pre-adolescent).

Calling something stupid, when it is indeed stupid, isn't an argument of any sort at all. It's merely description. For example, it's stupid not to notice that ERV is describing how things like gene duplication can occur even in lowly little HIV, something Behe doesn't seem to think makes any difference:

So in just the past few decades HIV has actually undergone more of certain kinds of mutations than all cells have endured since the beginning of the world. Yet all those mutations, while medically important, have changed the functioning virus very little. It still has the same number of genes that work in the same way. There is no new molecular machinery. If we see that Darwin's mechanism can only do so little even when given its best opportunities, we can decisively conclude that random mutation did not build the machinery of life.

But it does. All that mutation, point or duplication or whatever, has yielded results Behe claims could not and did not happen:

Vpu: Very useful to HIV-1 (follow the link to learn how) and which may very well have originated by way of a duplication of the sort dealt with in this post.

Vpx: Important to HIV-2, which did.

By Thomas Howard (not verified) on 16 Oct 2008 #permalink

The way I'm reading this seems to be:

1) Behe made fundamental assumptions about the variability of HIV.
2) Behe's assumptions were wrong.
3) Being wrong pretty much invalidates the rest of his point.

I'm not seeing why there's confusion here.

Just found your blog via Reddit and I LOVE IT. I work on the social side of HIV prevention and it's just brilliant finding someone who can explain this stuff so clearly. Hope you'll ignore (or better, delete) morons like Ky Sanderson - their goal is to toss in enough plausible but wrong information that you waste your time responding to a comment that was only ever designed to waste your time.

By Daniel Reeders (not verified) on 18 Oct 2008 #permalink

I don't think you have much to worry about, Daniel, because our host does a good job of both presenting the facts in a manageable way and ridiculing the willfully ignorant. Ridicule is the best weapon in those cases, because taking them seriously lends the credence -- far better to point and laugh.

Indeed Ransom and Daniel. If ERV's felt her previous arguments were that sound, why does she continue to address Dr. Behe? My question has not been answered, and at this point, I assume this means you cannot answer my question. It's been 6 days now (the time needed to create the Earth), and yet none of you atheists have been able to answer a simple question.

Country Shrink-- Honestly, I have no idea what you are talking about to answer you. I think you think HIV-1 should behave in a Mendelian fashion, thus this isnt surprising, but it shouldnt.

Lets step back and look at another virus-- HPV (warts). HPV has a double stranded DNA genome. It does not code for its own DNA Polymerase, it just uses the host cells (the same DNA polymerase your cells use to replicate our DNA). Thus the wart genome doesnt change all that much (it doesnt need to, it hides from our immune system a different way than HIV-1).

Now lets look at Hepatitis C. It is an RNA virus, like HIV-1. It codes for its own RNA-RNA polymerase (our cells have no use for RNA-RNA polymerases), and it messes up a lot. So RNA viruses operate as a 'quasispecies'-- there is a general 'consensus sequence', but there is a cloud of like-but-not-like HepC genomes around, because of all the mistakes the RNA-RNA-pol makes.

Then we have HIV-1. Its an RNA virus, like HepC. It operates as a quasispecies, like HepC. But unlike HepC, HIV-1 could be considered a diploid organism. Thus it doesnt just mutate a lot because of reverse transcription. It mutates a lot because the RT enzyme can hop back and forth between the two genome templates. That is *not* how humans, other mammals, other diploid organisms operate. It is not Mendelian :) We get one set of chromosomes from each of our parents. We dont have one set of chromosomes that is half mom, half dad.

TCS:
Let me help you atheists (and rare theistic evolutionists), with how to construct a response.

It will start with something like this:

"What ERV talks about, accomplishes the following beyond Mendelian genetics, and extends what evolution can accomplish by: x" (And x is the variable. Your response. Please fill in the blank.) And, "This extends the edge of what evolution can accomplish beyond the assertions of Behe by: y." (I think you get the picture). ERV, calling something 'stupid' is not an intelligent argument (it is adolescent or pre-adolescent). You must actually extend the edge of evolution outlined by Behe in order to make a valid argument. You have not done so in this post.

Um, no. Not at all. And talking down to us "atheists" doesn't help your point, since it's pretty clear that you're the one who's confused.

The problem is that "Mendelian Genetics" is entirely inadequate to describe the replication behavior of HIV. ERV's description, I thought, was on-point and quite illuminative. Given the very small number of genes that HIV has, Mendelian genetics aren't a very good model. There's also (so far as described), no concept of dominant/recessive traits, nor any reason to think they would be appropriate or adaptive. Nor is HIV's reproductive strategy, so far as described, biased towards the preservation of individual genes (coding or otherwise). However, HIV's reproductive strategy does mimic some of the characteristics of sexual reproduction, perhaps giving it an outwardly Mendelian aspect.

The advantages of HIV's reproductive scheme include: (1) greater genetic diversity in the virus' offspring; (2) greater viability/survivability in its offspring, and (3) variability directed towards potentially "useful" genetic changes.

If you continue to insist on forcing HIV's replication model in terms of the confining 'mendelian genetics' model, you will grossly misunderstand what was said here, and indeed fail to understand the value of the research that ERV described. That model is simply inadequate.

Nor is it a matter of 'extending evolution'. Evolution is what it is, and what we observe. What we discern in HIV's mechanisms for driving its own evolution, however, are the subject of a great deal of research (including ERV's). And what ERV tells us is that HIV has developed a simple and interesting scheme for driving 'useful variability' in its own offspring. This scheme gives it many of the advantages of sexual reproduction (variability and survivability in offspring) without many of the accompanying costs and difficulties.

Ultimately, this shouldn't be surprising: HIV's hosts' immune systems drive its survival strongly towards genetic variability. What's interesting is the simplicity and elegance of HIV's scheme.

The Country Shrink

Indeed Ransom and Daniel. If ERV's felt her previous arguments were that sound, why does she continue to address Dr. Behe?

You really didn't pay much attention to the actual post did you? Since you didn't notice, ERV wasn't addressing Behe at all. Rather, she used him as an example of poor creationist thinking. All her references are pure third person, not second. IOW, she's addressing us, not Behe:

There was a lot wrong with Michael 'Lilo' Behes description of HIV-1 in last years book, 'Edge of Evolution'.

Despite Lilos acceptance of common descent and educational training as a biochemist,

Behes focus in 'Edge' with HIV-1 was just that

If Lilo had spent 30 seconds researching HIV-1 before he wrote 'Edge'

Once again-- be very, very glad Mr. Lilo is not an HIV-1 researcher

As to why: well, as she would know, if you're going to pick an example of creationists getting stuff wrong for use in the context of a discussion on HIV-1, Behe's a pretty good one to pick. Maybe even the best.

By Thomas S. Howard (not verified) on 21 Oct 2008 #permalink

Tom--is it appropriate to call you Tom? In your world, you may think she was not addressing Behe, but I think that's clear. There is such a thing as an indirect approach. You have all obviously misunderstood the question I am asking. ERV seems to be trying to point out a special case with viruses regarding genetics. I understood that what ERV was explaining differed from Mendelian genetics. What I was interested in, is how this makes "evolution" work beyond what is accomplished with Mendelian genetics (i.e., how is it superior?). How does it build new molecular machinery (apart from resembling hybridization)? And, the previous arguments between Behe and ERV are not relevant, because I'm not sure either side came out on top. Did I address ERV here? Only in the third person--so not at all by your reasoning.

TCS:

ERV seems to be trying to point out a special case with viruses regarding genetics. I understood that what ERV was explaining differed from Mendelian genetics. What I was interested in, is how this makes "evolution" work beyond what is accomplished with Mendelian genetics (i.e., how is it superior?). How does it build new molecular machinery (apart from resembling hybridization)? And, the previous arguments between Behe and ERV are not relevant, because I'm not sure either side came out on top.

Actually, Mendelian genetics are the special case, not viral replication. Huge amounts of cellular machinery in higher-level organisms is devoted to discovering and repairing transcription (reproduction) errors in DNA. Viruses, being that much simpler, necessarily use simpler mechanisms. E.g., their hijacking of cellular machinery for their own replication, their re-use of individual genes, &c.

Mendelian genetics is also a special case in that it cannot/does not explain genetic variation or mutation; it can only explain variations in the populations of extant genes. And it cannot explain or describe at all even a such a common genetic event as gene duplication -- even in higher animals, this is relatively (!) common.

Further, ERV points out that an understanding of evolution that limits itself to SNPs (a Behe favorite) is entirely inadequate to explain or describe HIV's evolutionary behavior, and such a model of HIV necessarily yields a very flawed understanding because it depends on an understanding of HIV that is not based in reality.

An understanding of the reality of HIV's reproduction strategy (as described by ERV) yields a much more reasonable understanding of how HIV's relatives -- which have duplicated and differentiated genes -- have come into being. Neither Mendelian genetics nor Behe's SNP model can plausibly explain these changes. ERV's description of HIV's replication mechanics makes it more than plausible.

As for whether the previous arguments between Behe and ERV had either coming out on top? Given that Behe's arguments depend critically on demonstrated falsehoods -- as ERV makes clear -- Behe clearly lost.

/es

The Country Shrink.

I haven't read anything by Behe since I had the misfortune of reading part 'Darwin's Black Box' a few years ago. But based on Thomas S. Howard's quote from Behe above (#17), it's clear that Behe still doesn't have a clue what he's talking about. His starting assumption that everything is designed is preventing him from seeing the elephant in the room.

Behe, like many creationists, seems to think that evolution is prescriptive; that it is (in the case of HIV) trying to design the perfect virus. But that is not the case. Organisms evolve to be 'good enough' - good enough to survive in their niche. In the case if HIV, this means that it has evolved to be good enough at evading the immune system of its host, and transmitting between individuals.

But Behe hasn't noticed - and this is the elephant in the room - that HIV is still a very successful virus. It currently infects ~30 million people worldwide. It is 'good enough' just the way it is.

By gregonomic (not verified) on 28 Oct 2008 #permalink