Pogge,
Thanks for doing such a great job organising and moderating this difficult topic.
Monotreme at 23:44
On so-called Natural Selection . . .
Racter says it well . . .
Perhaps we should call it natural consequence or failure or just death.
Otherwise, we are invoking the chief metaphor of the religion of strict Darwinism.
We capitalise it as Natural Selection as if it is a mechanism that takes action, when, in fact, the metaphor describes/observes a post-event limitation, an intransitive that should be defined as such for the unknowing eyes of many of the readers here.
Yes, I’ve always been troubled by the propaganda and connotations behind the terms that we use so flippantly.
The logic is circular.
NS1:
“Natural Selection” simply refers to differing rates of reproductive success as the result of different physical or behavioral traits. Nothing circular about it. Clinging to the long-abandoned phrase: “survival of the fittest” is what produces the tautology: “survival of those most likely to survive”. Equating “failure” with “death” is at the root of this error. In a population of organisms which typically produce a thousand offspring during a lifetime, the one which produces only nine hundred is a failure (though the number of grandchildren is the real test).
I think it is time to move to real data. Monotreme failed to address the conservation of sequences or the recombination in Canadian swine. Sequences from H9N2 NS genes in Israel were just released. One isoalte is an exact match of an H5N1 Qingahi isoalte from Israel. However, the other 22 H9N2 sequences define some intersting polymorphism, which can then be used to create “treval-logs”. These travel log indicate the acquisition of polymorphisms is far from random, and instead link best buddies. In the best there has been a relationship between H9N2 (commonly found in wild birds) and mammalian sequences (commonly found in swine).
Here is one such polymorpshism found in the H9N2 in Israel, and two of teh Canadian swine that Montreme refuses to discuss:
DQ683029 A/chicken/Israel/416/2006 2006 H9N2 DQ683028 A/turkey/Israel/369/2006 2006 H9N2 DQ683026 A/turkey/Israel/89/2005 2005 H9N2 DQ683044 A/chicken/Israel/1808/2004 2004 H9N2 DQ683045 A/chicken/Israel/1953/2004 2004 H9N2 DQ683046 A/chicken/Israel/1966/2004 2004 H9N2 AY651562 A/Chicken/Viet Nam/C57/2004 2004 H5N1 AY818148 A/chicken/Vietnam/C58/04 2004 H5N1 DQ683043 A/turkey/Israel/1567/2004 2004 H9N2 DQ683038 A/chicken/Israel/1304/2003 2003 H9N2 DQ683039 A/chicken/Israel/1376/2003 2003 H9N2 DQ683041 A/chicken/Israel/1475/2003 2003 H9N2 DQ683040 A/ostrich/Israel/1436/2003 2003 H9N2 DQ280200 A/swine/Alberta/56626/03 2003 H1N1 DQ280216 A/swine/Ontario/53518/03 2003 H1N1 DQ683037 A/turkey/Israel/1209/2003 2003 H9N2 DQ683042 A/turkey/Israel/1562/2003 2003 H9N2 CY009384 A/swine/Spain/39139/2002 2002 H3N2 DQ683036 A/turkey/Israel/1013/2002 2002 H9N2 DQ683031 A/turkey/Israel/619/2002 2002 H9N2 DQ683035 A/turkey/Israel/965/2002 2002 H9N2 DQ683032 A/chicken/Israel/786/2001 2001 H9N2 DQ683034 A/chicken/Israel/811/2001 2001 H9N2 DQ683033 A/turkey/Israel/810/2001 2001 H9N2 DQ683047 A/chicken/Israel/90658/2000 2000 H9N2 DQ683025 A/turkey/Israel/90710/2000 2000 H9N2
The nails are quickly being pounded into the “random mutation” coffin (and no, pigs don’t fly).
It shoudl be noted taht the lab in Israel has no link to the Olsen lab in Wisconsin which generated the two Canadian swine sequences above.
The nails are also going into the “lab error” coffin.
The story is in the sequence, and the sequence says RECOMBINATION, loud and clear (for BOTH drifts and shifts).
Here is teh travel-log of “older” Israeli H9N2, which hops from human (going back to 1957 pandemic H2N2)
DQ683037 A/turkey/Israel/1209/2003 2003 H9N2 DQ683036 A/turkey/Israel/1013/2002 2002 H9N2 DQ683031 A/turkey/Israel/619/2002 2002 H9N2 DQ683035 A/turkey/Israel/965/2002 2002 H9N2 DQ683032 A/chicken/Israel/786/2001 2001 H9N2 DQ683033 A/turkey/Israel/810/2001 2001 H9N2 DQ683047 A/chicken/Israel/90658/2000 2000 H9N2 CY006019 A/Quail/Nanchang/2–0460/2000 2000 H9N2 DQ683025 A/turkey/Israel/90710/2000 2000 H9N2 AJ344029 A/swine/Italy/1513–1/98 1998 H1N1 AJ344026 A/swine/Cotes d’Armor/790/97 1997 H1N2 CY011148 A/Memphis/2/1986 1986 H3N2 CY004638 A/blue-winged teal/Alberta/452/1983 1983 H3N1 AY210164 A/Japan/170/62 1962 H2N2 AY210165 A/Netherlands/60/62 1962 H2N2 AY210166 A/Taiwan/1/62 1962 H2N2 AY210167 A/Yokosuka/3/62 1962 H2N2 AY210163 A/England/1/61 1961 H2N2 AY210162 A/Panama/1/61 1961 H2N2 M23968 A/Ann Arbor/6/60 1960 H2N2 AY210161 A/Ann Arbor/6/60 1960 H2N2 AY210159 A/SaoPaolo/3/59 1959 H2N2 AY210158 A/Victoria/15681/59 1959 H2N2 AY210157 A/Albany/6/58 1958 H2N2 AY210156 A/Malaya/16/58 1958 H2N2 AY210152 A/Albany/7/57 1957 H2N2 AY210153 A/Davis/1/57 1957 H2N2 CY008992 A/Denver/57 1957 H1N1 AY210155 A/ElSalvador/2/57 1957 H2N2 DQ508845 A/Japan/305/57 1957 H2N2 M81578 A/Leningrad/134/17/57 1957 H2N2
Niman,
Would you mind sharing the few steps required for recombination in step-by-step order in a short paragraph discussing your landing zones and template jumping for the readers. I think it would be very informative.
Have you identified a short list (10–20) of the most likely landing zones on the Qinghai strains for H9N2 or H3N2?
Can you share these potential landing zones and your ideas on other potential SNP or string acquisitions?
Do you believe that a glutamic acid acquisition by H9N2 at NS1 position 92 will increase virulence as it does in H5N1?
Does the reassortment of the H5N1 with the H9N2 in Israel create a higher affinity for your G228S polymorphism or are your landing zones gene segment-dependent?
Reposted from yesterday in hopes of a deeper understanding being presented:
Niman and Monotreme,
Please take the list of eight words/topics, choose 6–8 as you are comfortable and make a paragraph that communicates to the average reader.
You may work together or separately. There is no time limit and you may use two number two pencils and one whiteboard no larger than 4 feet by 8 feet.
Feel free to submit drafts and a final version if you have several ideas. I’d love to see how your ideas develop(ed) on this one.
OK, NS1, here’s my attempt.
The RNA of influenza doesn’t look like a spring, it is stuck on the inside of a ribonucleoprotein complex. Transcription is not a simple complete-the-mirror-half-of-the-ladder trick, but involves using the whole RNP-RNA complex as the template. New RNA comes off the replication machinery first as a ribbon, then gets twisted gently, then gets wound tightly and is popped into the baby virion. The RNP helps twist the new strand together for packaging into the baby virion, as well as protect it from attack by the cell. In order to do all these things, the RNP has to change shape several times.
The polymerase complex reads the the parent sequence and recruits molecules to assemble the different amino acids for the copy. Chaperone molecules twist the incoming amino acids so that they line up right.
Now there are several ways to do this synthesis. One is to just lay the RNA out flat but stuck to the inside the RNP, and copy off it. For this to work, the inside of the RNP has to be pretty smooth, plus, you need a way to get nucleic acids in, so there has to be flow through. Another way is to float the whole strand free of the RNP, but this defeats the purpose of using RNP as part of the template. A third way is to have the whole RNA-RNP thing wadded up together for example,as one website says, fold it in half and twist it around itself. During replication, the whole thing locally untwists to serially expose parts of the RNA strand as it is being transcribed by the polymerase.
Now the first and second methods don’t lend themselves to error checking, but the third method does, because the 3-d structure of the complex could use lined-up parts of the twisted strands for locating landmarks and identifying errors. This would be consistent with ‘landing pads’ and ‘chunks’ of code that tend to stay constant. There is a marker laid at the beginning and end of the strand so they can be assembled in proper order.
Segmented viruses like flu replicate segment by segment, then assemble the segments into virions in the nucleus. That’s why reassortment usually involves whole segments getting swapped during coinfections. One of the things that triggers the shift between replication and assembly seems to be the concentration of parts lying around—when they get knee-deep, it’s time to assemble them. This is where segments can get swapped around by mistake—if an end code is a little off, mistakes can happen.
Recombination seems to work a a chunk level. I can imagine two different ways for template-hopping to occur during coinfection. One is to have two RNPs, one from each parent, side by side, and working in phase, and to have the chaperone molecules in between them. The chaperone accidently herds the wrong nucleotide to one of the two RNPs. Now Niman says that there is generally a symmetrical flip between the 2 parent strands, that is, “flipping T to C and A to G”, so #1, which should have recieved a T, instead got a C, and the extra T went to #2. Or something like that.
The other way is to mix up the polymerases from the two strands, so that #1 winds up in #2′s polymerase, and #2 gets #1′s polymerase, and tiny discrepancies between the two drive some of the mistakes.
In order to figure out whether there is a pattern encoded in the third position bits of the influenza genome, I need to understand a bit about the three-dimensional structure involved in RNP-RNA replication. I suspect that one error-control system would involve reading _across_ the lined-up coils. There might be some other way. I’m still thinking about it.
Well, my neck is stretched out far enough.
wetDirt,
Do you surmise that attractant genetic acquisition, in some way, may occur across only the exposed parts of the aligned coils? Does the chaperone molecule morphology or property drive the attraction? Can some chaperones interact with multiple parent / donor segments and the daughter segment?
Have you considered that viral strain 1 (vs1) may act as an influencer according to some ruleset that convinces viral strain 2 (vs2) to match a vs1.nucleotide or vs1.sting?
What are the attractant chemical(s), mechanism or process that would be employed to drive vs2 to match the vs1 pattern?
Conservation, viability, so-called selection . . . all these observations should drive us to see that some rule is at play and some attractant influence may be defined, a true, pre-emergent mechanism at a detail level that we are not currently investigating.
Thought for Food.
Now I eat.
Chewing on that food lends another question . . .
What is the cause of a recombination?
Go ahead, anyone, please have a go at these very straightforward questions . . .
Homologous recombination requires homology and uses the Watson/Crick base pairing rules. Dual infections are required and control the opportunities (some strains frequently infect the same host and the prevelance of strains drives the frequency and combination of dual infections). Selection pressure drives the emergence of the sequences. Most acquisitions are third base transitions (which are silent).
(please go on all of you, and thanks)
NS1 – at 20:46
What is the cause of a recombination?
Is it agreed that before you get into your list of possible drivers of recombination that you must have:
Dual infection of the host; dual infection of a single cell within that host; replication of both viruses within the same capsid located in that single cell?
Dr. Niman, I have a very demanding, full-time job and other responsibilities. I’m not accepting homework assignments. As I’ve said before recombination does occur. I have acknowledged this about 10 times. That does not prove anything. There are anomalies in the sequences. I have acknowledged that as well. The people who generated the sequences have not explained them, period. Until they provide further information, we can’t know for sure what happened. Until then, we are all speculating.
NS1, just to be sure, you do believe darwinian evolution occurs, right? I have been assuming this, but I just want to make sure.
just another idea: assume, that sometimes some regions don’t mutate for quite some time. Then, how do we prove recombination ? We would need both parents and the child to demonstrate it, else we could always argue, that the apparant
recombination area just didn’t mutate while the other area did mutate,
thus mimicing recombination as in the Canadian swine.
Is that the reason, why Olson,Webster didn’t mention recombination although
the did examine the swine-sequences ? Well, they also fail to mention
the remarcable preservation of areas.
Recombination can’t well explain the preservation of areas either.
So we need another reason anyway.
PA-gene in : Tennessee26,Ontario48235,55383,11112,53518,57561
we could try to explain this by multiple recombinations plus low mutation rates
in some areas or by low mutation rates in some areas alone.
Both models have some weaknesses.
monotreme,ns1,niman,…, do you believe selection prefers some synonymous equivalents, sequences with identical amini-acids but different nucleotides , within one virus-generation ? I mean, can it happen that we take 2 viruses with identical amino acids, one is able to reproduce while the other isn’t ? Statistically one variant might do better in the long run, but this is not so significant in the next few generations (see the codon-bias … paper).
here are the pictures of the differences in those 6 sequences:
http://www.setbb.com/fluwiki2/viewtopic.php?p=269&mforum=fluwiki2#269
o:difference at that location
.:match at that location
-:either of the two sequences is not known at that position
Note, that the middle region remains conserved in all pictures
Monotreme, NS1, gs Don’t you stop your quest until the questions are answered appropriately and no matter what don’t be put put off.
pandemicflu – at 09:52 Very nice pictures. This is consistent with what I have been mulling on the mechanics of recombination: At least some authors suggest the RNP-RNA complex is not linear during replication, but is folded and twisted like a tie-dye t-shirt. Now a tie-dye t-shirt has regions with color, and regions without color, because the uncolored regions were inside the twist. Going back to the RNP-RNA complex, some portions of the coding regions are inside the twist, and some are outside the twist. It would sure be easier for some mechanisms such as homologous recombination to occur on the places outside the twist as compared with inside the twist, because other complexes can bump into the outside but not the inside. Do the constant regions in the code correspond with areas inside the twist?
there are too few examples of recombination to examine this statistically.
And then, we don’t know exactly at which position the recombination
occurs. That’s for influenza, but maybe it can be examined for other viruses first.
here is (again?)a link to the Olsen paper, describing how these viruses were isolated and about the other genes:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1393092
Monotreme, we’re all in the same place with you. : ) At least, I am.
Speculation is its own learning experience and has value. I am well-schooled and firmly centered in separating speculation and theoretical exercise from repeatable outcome.
The formal process of validating scientific information shouldn’t necessarily require inquiry, tutorial and didactic instruction. Presentation, publication and peer review processes to the scientific community require a cookbook and recipe, and demonstratable results. The clearer, more concise the information, the better the reception. If it can’t be repeated by an independent investigator, then it’s just theory.
On the other thread, I note that Niman says that the polymorphisms he is chasing consist of chunks of about 12–20 nucleotides, with just one carrying the mutation, often a silent third base type. So it LOOKS like a ‘point’ mutation, but it is actually a chunk of invariant code. What I need is a map of where the chunks carrying the polymorphism are, and whether there is a repeating pattern along the strand of code for the segment. Then I get a shoelace and a sharpie, mark the spots, and then start twisting up the shoelace.
NS1 – at 20:42 “Do you surmise that attractant genetic acquisition, in some way, may occur across only the exposed parts of the aligned coils? Does the chaperone molecule morphology or property drive the attraction? Can some chaperones interact with multiple parent / donor segments and the daughter segment? “ Yes. I can’t see any other way to do it if the replicating strands are not straight, with all parts equally accessible. It’s like trying to zip up a jacket through a wrinkle. The zipper jams. (Of course, RNA viruses only get half the zipper.) I have no idea about the chaperones, my impression is just that they are preprocessors for aligning the next tooth of the zipper. They appear to have a definite polarity, and that makes me wonder whether they would be of much use here. If they only line up one way, it’s hard to imagine how they can point to two different RNP-RNA complexes at the same time. Naaah, I can’t see it. (feel free to set me straigt on this, anyone)
“Have you considered that viral strain 1 (vs1) may act as an influencer according to some ruleset that convinces viral strain 2 (vs2) to match a vs1.nucleotide or vs1.sting? “ I haven’t seen anything yet that makes me think so. What Niman says he sees is that it’s a popularity contest: the prevalence of polymorphisms about matches the popularity of the strains. Certain polymorphisms prefer ducks, others geese. Until I hear different, I will go with that.
glo – at 12:10 I disagree. Recipes are for labwork. What you are describing doesn’t work for observational sciences, and there are a few out there. Before you can even get to labwork, you need some observations that need explaining. What all your fancy words are saying is, us Bigheads see no need to explain our reasoning to you peons, we only need to talk Latin to each other. Well, us peons have noticed that the emperor’s clothes are a little thin in a few places. If you won’t explain it, we will dig it out ourselfs. Thanks for your time.
“Monotreme, NS1, gs Don’t you stop your quest until the questions are answered appropriately and no matter what don’t be put put off.”
In this post earlier, in my haste I left some of the members off our genetics team at flu wiki…
…including: wetDirt, birdman, glo, beehiver,LMWatbullrun and Racter…sorry for the inadvertent ommisions.
I don’t get this stuff at all but I hope you will all continue the quest and not be diverted by outside comments. Thanks.
wetdirt,I don’t understand some things with this niman-short recombinations. Why do only short sequences occur ? Why does it have to be homologous ? How do the genes align ? maybe same questions again and again… And I think, nobody except niman believes in it. (at least for influenza, it seems) You can find something with searching “copy choice recombination”. So, when we check all point mutations for matching 20-bp subsequences, wouldn’t that give statistical evidence ? Compare it with other viruses or bacteria etc. and see, how flu compares. niman should have already done it… that would have been one of the first things I’d done in his place.
pandemicflu – at 13:27 Do you have access to the 21 July 2006 issue of Science? Check out p 357 et seq. Look really hard at the ribbon diagrams. This is a cousin of the Influenza virus, and it works a little differently, but some things are the same.
My guess, homology is required because it’s easier to trick the system if the sections look, feel, and taste practically identical. I think that alignment is really critical: the two strips have to be lined up for the recombination to occur. The thing I’m trying to figure out is, at what point in the reproduction process does recombination occur? Is it when the template is made, or when the copy is made? If it is when the template is made, then the template cranks out a significant population of recombinat children. If it is at the copy stage, then it only happens once in awhile, and the population in the cell nucleus is mostly faithful copies and just a few recombinants. Now, I don’t see that anyone has taken a single coinfected cell and characterized a stastically significant number of the children to see what the population statistics are. Once the recombinant is generated, it is then subject to all the fates that flesh is heir to, and sinks or swims on its own merits.
As for your statistical evidence, I think probably it’s best to work from the big pieces down to the little pieces. My bet is that the segments where fidelity is critical to function (NP comes to mind) are twisted into the center of the RNP where fewer accidents can occur, and the segments where you want variability are more to the outside. So first I would look at whether there is a systematic bias in the size of ‘immovable’ objects depending on what segment it’s in, then look at the frequency of the smaller ones, and whether they are spread out over the segment, or cluster here or there. But the first thing, is just get a map showing where they are.
wetDirt,
Your last contribution is very important and a path where I’ve begun to dig as well. Keep the good ideas coming.
Your analysis of why and how recombination biochemically occurs is extremely beneficial to this discussion.
Hi everyone, I have also been searching over the last few weeks for additional information on the mechanisms of recombination in viruses, since there seems to be quite a lot of bandwidth on the subject.
The first article that was very helpful was [[http://www.pubmedcentral.gov/articlerender.fcgi?tool=pubmed&pubmedid=1579113|this 1992 item], titled “RNA recombination in animal and plant viruses”, by M.M. Lai. It discusses the “copy choice mechanism” mentioned above by pandemicflu.
But what got my attention was the fact that nearly every single recombination example was given on positive-sense single-stranded RNA viruses. As we might remember, infl A virus is a negative-sense virus. This means that the initial viral segments which are relased into the cell, are not capable of acting as messenger RNA and producing proteins. A positive-sense strand has to be produced before any process of infection and replication can proceed. Fields Virology (4th ed.) p. 1487 affirms that “The genomic RNA of negative-stranded RNA viruses has to serve two functions: first as a template for synthesis of mRNAs and second as a template for synthesis of the antigenome positive strand…mRNAs are only synthesized once the virus has been uncoated in the infected cell.” Other sources state that the purpose of the second function given above, is to allow replication of the full-length genomic segments, to be later packaged into new viral particles.
Searching through other literature on the subject primarily revealed recombination research on positive-sense RNA viruses, and yes there are numerous articles available. I won’t list them here in order to keep this message coherent, but if there is interest I can post links or accession numbers later. There was however one free-access article from 2002 describing research on homologous recombination in a negative-sense RNA virus, titled “Transfection-mediated generation of functionally competent Tula hantavirus with recombinant S RNA segment”, by A. Plyusnin et al, and available at this link. The abstract begins: “Since the discovery of RNA recombination in polioviruses, there has been a general belief that this mechanism operates only in positive-sense RNA viruses. Recently, studying wild-type Tula hantavirus, we observed a mosaic-like structure of the S RNA segment that was consistent with generation by recombination between viruses from two genetic lineages. Here we show transfection-mediated rescue of Tula virus carrying recombinant S RNA segment.”
The entire article is worth a good read, because the authors did locate recombination “hotspots”…which begins to address some of the questions that have arisen on this board regarding the viral RNA structure as it might relate to recombination. More on that below. But we might note from the discussion section the following thoughts: “How widely HRec [homologous recombination] occurs in negative-sense RNA viruses remains to be seen. One could speculate that viruses with a high level of interference, short reproduction time and/or strong cytopathic capacity (like vesicular stomatitis virus, a classical example of a negative-sense virus) would have relatively small chances to recombine, simply due to the extremely low frequency of double infection, which is an obligatory prerequisite for recombination. On the other hand, persistent, less cytopathic and interfering viruses (like bunyaviruses, including hantaviruses) might have reasonable chances to recombine. However, for viruses with a segmented genome, those chances are expected to be smaller than (or even uncomparable with) the chances for reassortment.” (bold is added)
There is another article abstract here, “Evidence for recombination in Crimean-Congo hemorrhagic fever virus”, and this virus is also negative-sense single-stranded RNA. This article however is not free access, although some of you may have access.
There is one additional (free access!) article that may give depth to the current discussion. Titled “A universal BMV-based RNA recombination system—how to search for general rules in RNA recombination” by M. Alejska et al, it can be found here. This article discusses template switching and copy-choice mechanism.
This article is a long and involved read. Here are a few sentences from the discussion at the end, to give you an idea of their thoughts, and to whet your appetite.
I will have to add that I cannot answer technical questions on the above material, as I am still studying it myself…but wanted to pass it along to this group. One question I have though, if someone would like to jump in. Does anyone know if the homologous recombination that might be occurring in negative-sense RNA viruses (such as infl A) take place between the two negative-sense strands; or later on once the complementary positive-sense strands have been created?
Also I noted that even though the virus descriptions at GenBank say “negative-sense ssRNA virus”, the actual sequence given is the positive sense version.
Finally, apologies for such a long post, but one thing led to another. :-)
One additional detail to the first bulleted item in my post above (from the Alejska article). The authors’ abbreviation “RAS” means “recombinationally active sequence”. Sorry for the omission!
beehiver,
thanks for that post. Yes I have found that there is very little material on recombination in negative sense RNA viruses, especially flu. So please let us know if you find any more.
Mods, could someone please check the Niman- Israel thread?
At least 3 posters have posted but their posts do not appear, nor is there a place to reply. Thanks.
NS1,
Are you a proponent of creationism/intelligent design, or do you believe darwinian evolution occurs? I am not a scientist, but I am following the thread. Your direct answer would help clear up some confusion on my part. Thank you in advance.
Beehiver. I honestly don’t know how you keep coming up with this stuff but thanks again…again!!
Hi Tom DVM, actually I don’t keep up with some things very well at all. I often just start reading, then let my mind & intuition ask the questions that are bothering me, then try to chase down some answers…at PubMed clicking on related links can sometimes turn up gems. It really does chew up a lot of time, to the neglect of other things that should be getting done!
May I add I have a very great appreciation for your contributions to the fluwiki, and I have learned a lot from you. As well as from many others, not to exclude anyone.
anon_22 at 21:25, yes there were a couple articles about recombination in influenza virus, but I couldn’t access part of them (Science journal). I will post about that in the morning as it’s getting a bit late. Maybe you or someone else has an online subscription to Science?
thanks for the search and the articles. Although… I don’t feel that I see clearer now.
can we coordinate efforts with the papers ? Several people read one paper each and then report a summary and whether it applies to flu. So we needn’t read all the long papers each.
beehiver,
“anon_22 at 21:25, yes there were a couple articles about recombination in influenza virus, but I couldn’t access part of them (Science journal). I will post about that in the morning as it’s getting a bit late. Maybe you or someone else has an online subscription to Science?2
I do. Just post the link and I’ll find it. Thanks.
anonymous – at 09:05 “can we coordinate efforts with the papers ? Several people read one paper each and then report a summary and whether it applies to flu. So we needn’t read all the long papers each.”
I’ve just added a Bibliography page to the wikie linked from here. You can either add directly to the list or email them to me and I will organize them.
Eventually. :-)
…hoping that beehiver tells us, which are most interesting. I downloaded #1 and #3 of his list, but haven’t yet printed+read them
Apologies to all for delay in posting! Saturday chores and lower temps outdoors (for once), have ruled the day thus far.
This is the 2001 article titled “Recombination in the hemagglutinin gene of the 1918 “Spanish flu”, by MJ Gibbs et al. (Science. 2001 Sep 7;293(5536):1842–5). Abstract is posted below, which indicates the authors apparently used a combination of sequence and phylogenetic analysis to arrive at their conclusions. The article is not free access. However, oddly enough one of the replies is free access, see below. This link also shows the abstract as well as references in the article, which may provide additional relevant reading material (I have not checked those yet).
Robert Webster makes a reply to the former article at this location. Title: “Virology. A molecular whodunit”, citation Science. 2001 Sep 7;293(5536):1773–5, not free access. Would love to know what thoughts Webster had on the subject.
Laver and Garman make a reply to the original article here. Title: “Virology. The origin and control of pandemic influenza”, citation Science. 2001 Sep 7;293(5536):1776–7.
Finally, M. Worobey et al make a reply to the original article here, and this article is free access. The authors of the originating article (Gibbs et al) make another counter-reply at the bottom of this page. Title: “Questioning the Evidence for Genetic Recombination in the 1918 “Spanish Flu” Virus”. Citation: Science. 2002 Apr 12;296(5566):211 discussion 211. It appears this particular discussion revolves around different ways of constructing phylogenetic trees, and the varying conclusions that might arise. Very interesting indeed.
Can someone recommend a webpage that is a primer on how to read and interpret phylogenetic trees? I will also look at google.
Here is the abstract of the original Gibbs et al article.
“When gene sequences from the influenza virus that caused the 1918 pandemic were first compared with those of related viruses, they yielded few clues about its origins and virulence. Our reanalysis indicates that the hemagglutinin gene, a key virulence determinant, originated by recombination. The “globular domain” of the 1918 hemagglutinin protein was encoded by a part of a gene derived from a swine-lineage influenza, whereas the “stalk” was encoded by parts derived from a human-lineage influenza. Phylogenetic analyses showed that this recombination, which probably changed the virulence of the virus, occurred at the start of, or immediately before, the pandemic and thus may have triggered it.”
PMID: 11546876
It occurred to me that in my post yesterday at 19:36, it was not immediately clear that the Tula hantavirus is indeed a negative-sense ssRNA virus, thus the interest in that particular study to see if the mechanism of recombination revealed there had applicability to infl A virus.
anon_22, happy to hear you have subscription to Science, and thank you very much.
in tat Worobey et.al letter, they write, that the evolution rate (=mutation rate?) for H1N1 is only half as high in human HA2 than in human HA1, but that in swine it’s the same. Remember the total conservation in most of PA for the swine/Tennessee/26/1977 virus ! What determines the evolution rate ? Is it host-specific or is it encoded in the virus genom ?
I am guessing it has to do with selection to overcome host immune responses.
beehiver,
I wrote briefly about the Gibbs and Worobey debate before, but I will review that again. But it might not happen quickly as I have at least 3 other big pieces that I want to post/write about:
1) updating the wiki on treatment/tamiflu
2) I have a whole big picture summary of current status of pandemic vaccine production/science that I have notes from several important sources but no time to write up
3) put up all of my bibliography with links and comments where possible
Plus jetting between Asia, Europe, and America in the next 2 weeks.
So watch this space…
:-)
anon_22 at 06:15
yes, antigenic sites are in HA1, but how does the polymerase know, that it should allow more mutations in HA1 ? I remember from the “codon bias…”
paper, that they assume a larger rate of significant mutations versus
synonymous ones (more mutations that change an amino acid).
But the mutation rate itself should be somehow encoded/triggered too. (IMO)
Hmm, when we understand this, we might combine it with antirals and thus
reduce the development of resistance.
anon_22 at 06:15 “yes, antigenic sites are in HA1, but how does the polymerase know, that it should allow more mutations in HA1 ?”
I don’t think it has anything to do with the polymerase ‘knowing’ anything. Rather, as people acquire immunity to a particular strain of the virus, those viruses with the same antigenic sites would no longer be able to circulate as effectively as new variants with new antigenic sites that people have no immunity to. Since the antigenic sites are in HA1 and not HA2, this process of selection causes a disproportionate favoring of strains with new HA1 but not HA2.
you could be right, that it’s just only selection. I’m not sure yet.
Not sure, how they define “evolution rate” in that W.-letter.
They compared human and swine HA1 and human HA1 showed more differences i.e.
in the amino acids at the antigenic sites.
(these sites were hard to find, so if someone else is interested, I list them below)
ratio of differences in HA
antigenic/all,HA1/HA2 protein - nucleotide
human:2.13,4.04 - 1.20,1.59
swine:1.59,2.39 - 1.14,1.25
avian:1.70,2.46 - 1.14,1.15
we also see a lower proportion of synonymous differences in human H1N1.
I assume, there are fewer immune attacks against swine and avian viruses,
just because birds,swine die younger ? I.e. swine in North America.
If the virus hasn’t learned in billions of years of evolution
how to mutate the mutation rate, there is hope that it will never learn it.
Except maybe one day in a military laboratory…
if monotreme,niman,beehiver,… are interested, we can maybe discuss this elsewhere
antigenic sites (=epitopes?)
(HA:
Epitope A
122,124,126,130,131,132,133,135,137,138,140,142,143,144,145,146,150,152,168
Epitope B
128,129,155,156,157,158,159,160,163,165,186,187,188,189,190,192,193,194,196,197,\\198
Epitope C
44,45,46,47,48,50,51,53,54,273,275,276,278,279,280,294,297,299,300,304,305,307,3\\08,309,310,311,312
Epitope D
96,102,103,117,121,167,170,171,172,173,174,175,176,177,179,182,201,203,207,
208,209,212,213,214,215,216,217,218,219,226,227,228,229,230,238,240,242,
244,246,247,248
Epitope E
57,59,62,63,67,75,78,80,81,82,83,86,87,88,91,92,94,109,260,261,262,265
NA:
Epitope A: 383–387,389–394,396,399,400,401,403
Epitope B: 197–200,221,222
Epitope C: 328–332,334,336,338,339,341–344,346,347,357–359,366–370
Here are some H1-sequences which are years apart but are similar.
This could be more significant than the preservation in other genes,
like Tennesse/1977 in PA , since we have the antigenic need for mutations
in HA:
21 A/MD/12/91 (H1N1) A/WI/4754/94(H1N1)
14 A/swine/St-Hyacinthe/106/91(H1) A/swine/Ontario/11112/04(H1N1)
21 A/swine/Iowa/15/30(H1N1) A/Alma Ata/1417/84(H1N1)
29 A/swine/St-Hyacinthe/148/90 (H1N1) A/swine/Iowa/15/30(H1N1)
21 A/Connecticut/9/56(H1N1) A/Fort Worth/50(H1N1)
20 A/New Caledonia/20/99) A/New York/497/2003(H1N1)
20 A/USSR/92/77(H1N1) A/swine/England/191973/92(H1N7)
11 A/USSR/92/77(H1N1) A/Mongolia/231/85 (H1N1)
19 A/Hong Kong/117/77(H1N1) A/swine/England/191973/92(H1N7)
27 A/Kiev/59/79 (H1N1) A/swine/England/191973/92(H1N7)
16 A/PR/8/34(HON1) A/Mongolia/111/91 (H1N1)
On the issue of whether mutations are random, here is an excerpt from the book ‘Influenza’ Potter (ed), chapter on antigenic drift by Alan Hampson (page 64), which IMO is probably easiest to understand:
“It has been argued that this high rate of mutation, which is maintained over time, is in itself sufficient to explain the antigenic changes observed in the viral surface proteins, without the need for participation of selective pressure from the host immune system, and that this is supported by the observation that synonymous nucleotide substitutions predominate over non-synonymous substitutions in the viral genome. However, for influenza A approximately 50% of nucleotide changes in the RNA coding for the HA and NA protein sequences result in amino acid changes, a considerably higher level than would be expected for random mutations. Additionally, analysis of large sequence data sets for the haemagglutinin gene of both H1 and H1 human influenza virusese has revealed that there are a number codons within the HA1 region where non-synonymous substitutions predominate, and these appear to be under positive Darwinian evolution.”
I wonder, how they measured the ratio of synonymous vs. nonsynonymous substitutions. When you just look at published sequences and compare these, then this data was already subject to selection. Non synonymous mutations were more likely to be included in the data.
average number of differences in my (n=136) H1-database:
HA1:
------human,swine,avian
human:095,225,276
swine:226,174,224
avian:276,224,097
SC/1918:149,179,215
HA2:
-----human,swine,avian
human:046,120,142
swine:120,101,134
avian:142,134,062
SC/1918:062,105,115
so, the 1918 virus was less human-like
in HA1 as well as in HA2 than average human H1-viruses.
The reason could be that the database is dominated by viruses
from 1998 and later, and these are all rather similar.
But the ratio 1918 vs. average is higher in HA1 than in HA2,
suggesting that HA1 in 1918 was indeed less human-like than HA2.
Supporting the findings of Gibbs et. al.
But whether we can conclude that there was recombination in 1918-HA,
I’m not sure.
So I couldn’t find evidence for recombination in 1918-HA either.
sorry, remove the last sentence. It was edited earlier and then I forgot to delete it.
best evidence for recombination in 1918 HA seems to be given by A/swine/Iowa/15/30, which is relatively more close to 1918HA in HA1 than in HA2. Even when you compare it with other human viruses to take into account the Worobey et.al argument.
Early H1N1:
001 002 003 004 005 006 007 008 009 010 011 012 013 014
1 A/Fort Monmouth/1/1947(H1N1) --- 044 056 082 089 123 130 116 128 126 145 136 236 175 1586 2 A/Cam/46(H1N1) 044 --- 075 087 093 130 135 123 137 132 154 142 255 189 1696 3 A/Fort Worth/50(H1N1) 056 075 --- 110 113 144 148 130 146 144 164 156 255 192 1833 4 A/Bel/42(H1N1) 082 087 110 --- 042 087 095 080 097 090 114 103 221 151 1359 5 A/Weiss/43(H1N1) 089 093 113 042 --- 097 103 090 106 099 126 113 227 159 1457 6 A/PR/8/34(HON1) 123 130 144 087 097 --- 014 059 082 075 101 086 215 138 1351 7 A/Puerto Rico/8/34(H1N1) 130 135 148 095 103 014 --- 066 086 079 103 088 220 143 1410 8 A/Melbourne/35(H1N1) 116 123 130 080 090 059 066 --- 065 060 085 073 206 129 1282 9 A/NWS-G70c(H1N9) 128 137 146 097 106 082 086 065 --- 013 042 044 214 134 1294 10 A/Wilson-Smith/33(H1N1) 126 132 144 090 099 075 079 060 013 --- 039 036 208 127 1228 11 A/WSN/1933 TS61(H1N1) 145 154 164 114 126 101 103 085 042 039 --- 059 226 147 1505 12 A/swine/Cambridge/39(H1N1) 136 142 156 103 113 086 088 073 044 036 059 --- 219 138 1393 13 A/swine/Iowa/15/30 (H1N1) 236 255 255 221 227 215 220 206 214 208 226 219 --- 114 2816 14 A/South Carolina/1/18 (H1N1) 175 189 192 151 159 138 143 129 134 127 147 138 114 --- 1936
HA1:
001 002 003 004 005 006 007 008 009 010 011 012 013 014
1 A/Fort Monmouth/1/1947(H1N1) --- 031 042 059 064 084 091 075 091 092 110 097 163 126 1125 2 A/Cam/46(H1N1) 031 --- 058 065 069 092 097 085 099 097 115 101 179 137 1225 3 A/Fort Worth/50(H1N1) 042 058 --- 083 084 105 109 089 107 108 124 114 180 141 1344 4 A/Bel/42(H1N1) 059 065 083 --- 034 061 069 053 071 067 087 074 153 111 987 5 A/Weiss/43(H1N1) 064 069 084 034 --- 069 075 061 078 074 097 083 161 119 1068 6 A/PR/8/34(HON1) 084 092 105 061 069 --- 012 040 064 060 082 068 151 104 992 7 A/Puerto Rico/8/34(H1N1) 091 097 109 069 075 012 --- 047 068 064 084 070 156 109 1051 8 A/Melbourne/35(H1N1) 075 085 089 053 061 040 047 --- 046 044 067 053 140 094 894 9 A/NWS-G70c(H1N9) 091 099 107 071 078 064 068 046 --- 010 035 033 150 100 952 10 A/Wilson-Smith/33(H1N1) 092 097 108 067 074 060 064 044 010 --- 035 028 147 096 922 11 A/WSN/1933 TS61(H1N1) 110 115 124 087 097 082 084 067 035 035 --- 049 161 114 1160 12 A/swine/Cambridge/39(H1N1) 097 101 114 074 083 068 070 053 033 028 049 --- 152 101 1023 13 A/swine/Iowa/15/30 (H1N1) 163 179 180 153 161 151 156 140 150 147 161 152 --- 075 1968 14 A/South Carolina/1/18 (H1N1) 126 137 141 111 119 104 109 094 100 096 114 101 075 --- 1427
HA2:
001 002 003 004 005 006 007 008 009 010 011 012 013 014
1 A/Fort Monmouth/1/1947(H1N1) --- 013 015 023 025 039 039 041 037 034 035 039 073 050 463 2 A/Cam/46(H1N1) 013 --- 018 022 024 038 038 038 038 035 039 041 076 053 473 3 A/Fort Worth/50(H1N1) 015 018 --- 028 030 040 040 042 040 037 041 043 076 051 501 4 A/Bel/42(H1N1) 023 022 028 --- 008 026 026 027 026 023 027 029 068 041 374 5 A/Weiss/43(H1N1) 025 024 030 008 --- 028 028 029 028 025 029 030 066 041 391 6 A/PR/8/34(HON1) 039 038 040 026 028 --- 002 019 018 015 019 018 064 035 361 7 A/Puerto Rico/8/34(H1N1) 039 038 040 026 028 002 --- 019 018 015 019 018 064 035 361 8 A/Melbourne/35(H1N1) 041 038 042 027 029 019 019 --- 019 016 018 020 066 036 390 9 A/NWS-G70c(H1N9) 037 038 040 026 028 018 018 019 --- 003 007 011 064 035 344 10 A/Wilson-Smith/33(H1N1) 034 035 037 023 025 015 015 016 003 --- 004 008 061 032 308 11 A/WSN/1933 TS61(H1N1) 035 039 041 027 029 019 019 018 007 004 --- 010 065 034 347 12 A/swine/Cambridge/39(H1N1) 039 041 043 029 030 018 018 020 011 008 010 --- 067 038 372 13 A/swine/Iowa/15/30 (H1N1) 073 076 076 068 066 064 064 066 064 061 065 067 --- 040 850 14 A/South Carolina/1/18 (H1N1) 050 053 051 041 041 035 035 036 035 032 034 038 040 --- 521
anonymous,
First of all, if you are going to put up more than occasional posts, especially posts that include technical stuff or data, please give yourself a handle.
Secondly, I looked over the Gibbs and Worobey papers. It seems to me that Gibbs built his whole case of recombination on matching sequences, and if there is an error in that methodology, (in rooting the trees), then it throws all of the rest of his arguments out.
Moreover, there is the paper by Chare which also analyzed Gibbs’ sequences and concluded there was no definite evidence of recombination based on their tests, not just on direct comparisons.
Personally, I am inclined to say at least that Gibbs’ thesis is in doubt, although I wouldn’t and can’t dismiss it either, not least because of my own lack of expertise.
Suffice it to say that Gibbs wrote the only paper on homologous recombination on influenza A virus in recent years, and his finding has been cast in doubt by 2 different researchers. No one else appears to have come up with any other data or analysis supporting Gibbs’ opinion.
So for now, the current weight of evidence still suggests that recombination if present does not play a large role in influenza virus evolution.
bump