Don't Blame Birds for 1918 Flu

[super_flu]

Re: Don't Blame Birds for 1918 Flu

here is an older analysis from me for 1918 segments:
(always nucleotide-differences, since amino-acid differences can be
biased by functional constraints and are fewer)

http://www.setbb.com/fluwiki2/viewtopic.php?t=338&mforum=fluwiki2

assuming constant(time) nucleotide-mutation rates and
tracing the lines backward gives estimates for a common
ancestor of todays viruses and 1918 pandemic virus of:

segment H1N1 H3N2
-----------------------
 1      1880 1830
 2      1905 1700
 3      1905 1902
 4      1855
 5      1870 1893
 6      1900
 7      1900
 8

http://www.setbb.com/fluwiki2/viewtopic.php?t=424&mforum=fluwiki2

This linear regression approach to estimating ancestral divergence dates has pretty much been replaced with the method used in this paper. A constant mutational rate does not take into account substitution rate changes that may result from inter-species transmission events.

 

[AlaskaDenise]

Re: Don't Blame Birds for 1918 Flu

....changes that may result from inter-species transmission events.

What are those changes?

.

 

[super_flu]

Re: Don't Blame Birds for 1918 Flu

What are those changes?

.

Cut my quote from above a little short...

...substitution rate changes that may result from inter-species transmission events.

It is generally believed that after an interspecies transmission event, such as transmission of avian influenza to humans, that nucleotide substitutions will accumulate at a higher rate due to viral adaptation to a novel host.

This phenomenon is observed throughout the genome, even though selection only takes place on the antigenic sites. I believe that codon usage bias is the reason we see increased rates on the internal genes as well. Mutations accumulate to look more like the novel hosts own genome - viral host mimicry. mRNA's are often expressed on the surface of a cell. If a cell is infected with a virus it will express those mRNAs. If the mRNAs look similar to host sequences (ie. similar codon usage to that of the host) then the host will not recognize the infection.

So, there are two reasons we often see substitution rates increase after an inter-species transmission event.

1. Antigenic escape
2. Codon usage bias

I'm sure someone will correct me if I've forgotten something.

 

[AlaskaDenise]

Re: Don't Blame Birds for 1918 Flu

Is that higher mutation rate, in the process of host adaptation, one of the reasons it's been difficult to make enough novel H1N1 egg-based vaccine?, i.e., the swine virus is naturally mutating to become a chicken virus and it would slow down the mass-production of swine-like virus.

Can I assume these details also describe vaccination-driven mutation?

.

 

[AlaskaDenise]

Re: Don't Blame Birds for 1918 Flu

Mean TMRCA estimates of each gene segment of H1N1 viruses shows that the components of the 1918 pandemic strain were circulating in
mammalian hosts, i.e., swine and humans, at least 2 to 15 years
before pandemic occurrence.

Various sources say that severe swine flu wasn't known until 1918. I'm not sure about the correctness of that statement, but given the above excerpt from the new PNAS paper, would it make sense that a low path swine flu would circulate for 2 to 15 years and only become noticeably-severe at the same time as a severe human influenza?

.

 

[super_flu]

Re: Don't Blame Birds for 1918 Flu

Is that higher mutation rate, in the process of host adaptation, one of the reasons it's been difficult to make enough novel H1N1 egg-based vaccine?, i.e., the swine virus is naturally mutating to become a chicken virus and it would slow down the mass-production of swine-like virus.

Can I assume these details also describe vaccination-driven mutation?

.

Oooh, good question!

In egg we do not see immune driven selection. Rather, in this case everything has to do with efficient replication. More specifically, how well does the cellular machinery and the viral RNP complex work in the specific environment. Egg adapted mutations do exist. However, as far as I know, there has been no examination of viral substitution rates during egg adaptation or even quasi-species diversity within a single egg. Hopefully some enterprising young scientist will conduct some experiments to answer these questions. I think they're very interesting.

With vaccine development, the viruses used are adapted to grow well in mammalian cells. Often a virus needs to be engineered so that it's safe to work with and grows efficiently, but is still similar enough that it's still suitable for vaccine use. However, the evolutionary pressure of growing in egg is minimal - one reason to use eggs for vaccine development. Both mechanisms I described in the previous post are driven by the immune system.

 

[AlaskaDenise]

Re: Don't Blame Birds for 1918 Flu

Both mechanisms I described in the previous post are driven by the immune system.

Yes - I didn't think long enough - an egg hasn't developed enough to have an immune system.

.

 

[AlaskaDenise]

Re: Don't Blame Birds for 1918 Flu

Rather, in this case everything has to do with efficient replication

If they put a "K" in position 627 of the PB2 and set the oven for 98.6F would it speed up vaccine production?

.

 

[super_flu]

Re: Don't Blame Birds for 1918 Flu

Various sources say that severe swine flu wasn't known until 1918. I'm not sure about the correctness of that statement, but given the above excerpt from the new PNAS paper, would it make sense that a low path swine flu would circulate for 2 to 15 years and only become noticeably-severe at the same time as a severe human influenza?

.

Swine flu wasn't isolated until 1930. But the data suggests that the precursor genes for the 1918/classic swine/seasonal H1N1 were present in some mammalian host. Based on what we know about the current outbreak I would suspect that host may have been swine.

The data also suggests that 2 variants, seasonal and the 1918 virus both co-circulated. This might help explain the less lethal spring wave followed by the more lethal second wave. It's possible they were caused by two different viruses.

 

[Evelyn]

Re: Don't Blame Birds for 1918 Flu

Cut my quote from above a little short...

...substitution rate changes that may result from inter-species transmission events.

It is generally believed that after an interspecies transmission event, such as transmission of avian influenza to humans, that nucleotide substitutions will accumulate at a higher rate due to viral adaptation to a novel host.

This phenomenon is observed throughout the genome, even though selection only takes place on the antigenic sites. I believe that codon usage bias is the reason we see increased rates on the internal genes as well. Mutations accumulate to look more like the novel hosts own genome - viral host mimicry. mRNA's are often expressed on the surface of a cell. If a cell is infected with a virus it will express those mRNAs. If the mRNAs look similar to host sequences (ie. similar codon usage to that of the host) then the host will not recognize the infection.

So, there are two reasons we often see substitution rates increase after an inter-species transmission event.

1. Antigenic escape
2. Codon usage bias

I'm sure someone will correct me if I've forgotten something.

Super_flu, thank you for that clear and extremely helpful explanation. I really appreciate it.

 

[KC]

Re: Don't Blame Birds for 1918 Flu

So,

It could be possible that both a novel H1N1 virus (swine) and a novel H5N1 (avian) virus are circulating around the globe, and the common mixing vessels (swine or humans) are contributing to re-assortment and/or recombination? At the same time, normal seasonal influenza was/is also spreading.

Influenza is influenza.... it evolves... sometimes rapidly regardless of where the final pandemic origin spreads from.

My big question... and I've asked it many times here before and been shunned.... What might the effects of other endemic or epidemic non-influenza viruses that have similar structures have in the equation?

Could co-infections between an influenza virus and a non-influenza virus cause a problem if they share similar molecular traits?

 

[AlaskaDenise]

Re: Don't Blame Birds for 1918 Flu

......mRNA's are often expressed on the surface of a cell. If a cell is infected with a virus it will express those mRNAs. If the mRNAs look similar to host sequences (ie. similar codon usage to that of the host) then the host will not recognize the infection...........

That was a great explanation of the increased mutation rate.

In thinking about this more, now I'm a little confused.

In "If a cell is infected with a virus it will express those mRNAs." - is "it" the new virons that will now have the host mRNAs on their (viron) surfaces?

Assuming the answer is yes, then....

In "If the mRNAs look similar to host sequences ..... then the host will not recognize the infection" - if the "baby" virons are disguished (using mRNA) to look like the host so they can escape detection, how does the immune system ever fight them off? (When I asked my self this question, I realized I might have misunderstood the explanation, hence this question)

.

 

[gsgs]

Re: Don't Blame Birds for 1918 Flu

they (edit---the mRNAs) look just a tiny bit different, giving a slighly increased
probability to survive - the process goes over decades.

But nucleotides are not recognized by the immune system
as I understand, only amino-acids.
Nor are they expressed on the cell-surface (?)

mammalean flu acquires more A,T nucleotides over the
years,decades,centuries - but this is presumably another process


epitopes in the cell are transported to the surface, so the killer cells
can see which cells to kill

 

[HenryN]

Re: Don't Blame Birds for 1918 Flu

they look just a tiny bit different, giving a slighly increased
probability to survive - the process goes over decades.

But nucleotides are not recognized by the immune system
as I understand, only amino-acids.
Nor are they expressed on the cell-surface (?)

mammalean flu acquires more A,T nucleotides over the
years,decades,centuries - but this is presumably another process


epitopes in the cell are transported to the surface, so the killer cells
can see which cells to kill

Interferon

 

[gsgs]

Re: Don't Blame Birds for 1918 Flu

Brevig/1918 is more distant in PB2 to all other early human H1N1 than in PB1 or PA.
This indicates that it is not the ancestor of all these but rather
a reassortant.
But the same is true for Wilson-Smith/1933 , so there should be other reasons

Hickox/1940 seems to be from ~1945 rather than from 1940
Swine/Jamesburg/1942 seems to be from ~1932 rather than 1942

-----edit-------------
otoh, this (more mutations in PB2) is not seen in the swine linage
but the swine segments don't show any similarity with the human ones,
so there probably was no mixing human-swine after 1918

where should this new PB2 have come from ?

 

[AlaskaDenise]

Re: Don't Blame Birds for 1918 Flu

they ........

??? What is the "they"?

What post are you referencing?

.

 

[Emily]

Re: Don't Blame Birds for 1918 Flu

So,

It could be possible that both a novel H1N1 virus (swine) and a novel H5N1 (avian) virus are circulating around the globe, and the common mixing vessels (swine or humans) are contributing to re-assortment and/or recombination? At the same time, normal seasonal influenza was/is also spreading.

Influenza is influenza.... it evolves... sometimes rapidly regardless of where the final pandemic origin spreads from.

My big question... and I've asked it many times here before and been shunned.... What might the effects of other endemic or epidemic non-influenza viruses that have similar structures have in the equation?

Could co-infections between an influenza virus and a non-influenza virus cause a problem if they share similar molecular traits?

Is this along the lines of what you are wondering about?

http://pathmicro.med.sc.edu/mhunt/genet.htm
"Recombination enables a virus to pick up genetic information from viruses of the same type and occasionally from unrelated viruses or even the host genome

 

[kiwibird]

Re: Don't Blame Birds for 1918 Flu

Really interesting Emily. Is this part of the answer to KC's question?

 

[gsgs]

Re: Don't Blame Birds for 1918 Flu

> seasonal and classic swine H1N1 viruses were not derived directly
> from BM/1918, but their precursors co-circulated during the pandemic.

> the 1918 H1N1 pandemic virus most likely was generated by
> reassortment between mammalian viruses and a previous human
> strain and was not a pure avian virus.

then it would have been much different and we would see it in BM/18

> the BM/1918 virus H1, N1, PB1, PA, and NP genes clustered with human
> H1N1 influenza A viruses, whereas its PB2, M, and NS genes
> clustered with swine.

differences in promille in the 8 segments:
BM/18-WS/33 : 48,32,37,81,42,68,27,29
BM/18-SW/31: 44,46,51,65,44,51,39,45

so I'd say there were more mutations in humans in HA and NA (because of immunity)
and more mutations in swine in the other 6 segments

> TMRCA:1903,1914,1914,1916,1909,1913,1896?,1908

reassortments are rare, so multiple reassortments in different years are even rarer

AT-score in the 8 segments: (the longer in mammals, the higher the score)

BM/18:5568,5665,5620,5614,5310,5751,5152,5465
WS/33:5621,5772,5720,5878,5354,5630,5340,5563
Ma/54:5740,5856,5784,5791,5368,5848,5266,5622
Sw/57:5620,5772,5754,5942,5456,5874,5246,5649

> (Table 1) suggest those genes (1,5,8)
> have circulated in humans since the 1889 H3 influenza pandemic

H3N8 ? The same virus which is still in horses,dogs. That virusis rare in birds,
did it go from birds to humans in 1889 in all segments ?
Still 19 years to gain in AT-score, which apparantly hasn't happened (?)

> However, our results suggest that the same NP gene lineage has
> been circulating in human influenza A viruses since the 19th
> century, consistent with the report by Gammelin et al. (19). [from 1990]


> Taken together, our results indicate that it is unlikely that the
> BM/1918 virus could have resulted from adaptation of an entire
> avian virus introduced directly into humans shortly before the
> pandemic. More likely, it was generated by reassortment between
> previously circulating swine and human strains and introduced
> avian viruses over a period of years.


> Phylogenetic relationships between BM/1918 and classic swine
> H1N1 virus PB2, M, and NS genes also indicate that classic swine
> H1N1 is a reassortant between BM/1918 and an unknown virus.

which segments from BM/1918 ?

> As such, classic swine H1N1 is derived partially from BM/1918
> and is not a precursor of the 1918 pandemic virus (Fig. 1 and
> Figs. S1–S8) (9).


I don't believe it (yet).
I haven't looked at the phylo-graphs or the 1957,1968 data yet.
But this looks even more doubtful to me, that 1957,1968
should have been created years ago


http://www.setbb.com/fluwiki2/viewtopic.php?p=1229

 

[super_flu]

Re: Don't Blame Birds for 1918 Flu

That was a great explanation of the increased mutation rate.

In thinking about this more, now I'm a little confused.

In "If a cell is infected with a virus it will express those mRNAs." - is "it" the new virons that will now have the host mRNAs on their (viron) surfaces?

Assuming the answer is yes, then....

In "If the mRNAs look similar to host sequences ..... then the host will not recognize the infection" - if the "baby" virons are disguished (using mRNA) to look like the host so they can escape detection, how does the immune system ever fight them off? (When I asked my self this question, I realized I might have misunderstood the explanation, hence this question)

.

The answer is no.

The host cell will express mRNAs on its surface. Regardless of whether viral or not. It's a mechanism for host cells to monitor its cells are behaving correctly. During infection viral genetic material and proteins are produced at high levels. Some of these RNAs are placed on the surface of the host cell. Other cells may recognize these RNAs as not self and kill the cell. If viruses accumulate mutations to account for codon usage bias this will make the viral genome appear more like its host. This does not necessarily result in new phenotypes. In other words, these are silent mutations that do not result in new amino acids. They just use codes for the same amino acid that are similar to the host species.