Where have all the subtypes gone, long time passing?

Illustration from Roche et al (2014) on an individual-based transmission model for influenza A viruses in waterfowl. The drawing was made by John Megahan, and is used under CC-BY 4.0 license.

Illustration from Roche et al (2014) on an individual-based transmission model for influenza A viruses in waterfowl. The drawing was made by John Megahan, and is used under CC-BY 4.0 license.

By Jonas Waldenström

Ducks and men differ in some obvious ways. Whereas a Mallard is a fluffy, winged animal that dabbles, a man is an elongated, mostly fur-free mammal that intoxicates itself with beverages and too much screen time. There are, however, also shared traits and even shared pathogens, such as the influenza A virus.

But where you and I will experience only a single, or a couple of different influenza virus subtypes in our lives – for instance the H1N1 or the H3N2 that are touring the world today – an average Mallard can walk through up to 4 or 5 over the course of a few weeks in autumn. The diversity of influenza viruses in Mallards and other waterfowl is simply astounding.

For example, in our studies we have isolated 74 subtypes from Mallards caught in single little pond at the edge of a lagoon opening to the Baltic Sea. This is more than any other sampling site on the planet, and a large chunk of the global diversity for this virus.

There are some patterns to this mess. Some subtypes are actually fairly common at all times in autumn, such as H1 and H4 viruses, other subtypes are detected each autumn, but show more outbreak-like patterns. However, there are some subtypes that just pop up seemingly randomly in the population, sometimes years apart. So where do all subtypes go? A single animal is rarely excreting viruses for more than a couple of days, so the viruses have to go somewhere.

Of course there are hypotheses, and opinions. Perhaps some subtypes reside in other hosts, spilling over occasionally to Mallards? Perhaps the viruses can be maintained in the environment, in the water or in the sediment, or even in ice? Perhaps Mallards do not have long-term efficient immune memory to influenza? Perhaps the population size and annual recruitment of juveniles are large enough to maintain transmission of even rare subtypes? Perhaps they are seeded in from aliens?

A lot of hypotheses in the air, but not too much synthesis. Until now. A recent paper published by American researchers in PLOS Biology examines the existing evidence and model how well data fit different hypotheses. The method they used is phylodynamics. And the results are pretty cool.

Some possible epidemiological pathways are depicted in the figure at the start of this post (except the alien one, strangely enough). Viruses can be transmitted directly between individuals or move via the environment, including possible time delay. Both the environment and the host populations experience seasonal changes, some of which we can quantify and use as parameters in epidemiological models. The effect of these parameters can then be used to illustrate how the relationships between strains of the virus would do in simulated phylogenetic trees over time. If you change a parameter, say herd immunity, how will the resulting trees and stats look like after simulations? Will it look like the data we observe in natural populations, or will it look different?

In the left hand panels in the figure below you see how a tree based on human influenza typically looks like. There is little diversity, the antigenetic landscape (the part the immune system reacts to) changes with small steps, with accumulating mutations (A), and the phylogenetic tree has few, but long branches that stretches forward (G). A typical avian-based set, as seen in the panels to the right, show a mix of colors of different co-occurring lineages with large antigenic diversity (C), and the tree is more a thorny bush (I). In the middle panel you have an example of avian viruses based on direct transmission only. This model isn’t overly well matched with what is observed in nature, and predicts fewer circulating viruses (B), and a different tree topology (H). By doing this sort of analyses over and over again, one is left with the most plausible models. (I know this description is over simplistic, but this is a blog piece after all)

Figure from Roche et al (2014) used under CC-BY 4.0 license. Original caption: (A–C) Time series of influenza prevalence in humans, avian system with only direct transmission, and avian system with mixed transmission, respectively. Basic reproduction ratio, R0, is set to 1.5 for direct transmission, and environmental durability is set at 20 d when this transmission route is included. Colors represent antigenic distance between the introduced strain and the dominant variant at time t. (D–F) The black line represents antigenic diversity through time (i.e., number of antigenic strains), whereas the grey line demonstrates temporal changes in the antigenic distance of the dominant strain to the introduced strain. Time is expressed in years. (G–I) Associated reconstructed phylogenies. (J–L) Co-infection patterns for the situations depicted previously.

Figure from Roche et al (2014) used under CC-BY 4.0 license. Original caption: (A–C) Time series of influenza prevalence in humans, avian system with only direct transmission, and avian system with mixed transmission, respectively. Basic reproduction ratio, R0, is set to 1.5 for direct transmission, and environmental durability is set at 20 d when this transmission route is included. Colors represent antigenic distance between the introduced strain and the dominant variant at time t. (D–F) The black line represents antigenic diversity through time (i.e., number of antigenic strains), whereas the grey line demonstrates temporal changes in the antigenic distance of the dominant strain to the introduced strain. Time is expressed in years. (G–I) Associated reconstructed phylogenies. (J–L) Co-infection patterns for the situations depicted previously.

By tweaking the models and conducting lots of statistics, the research team could show that the two main factors explaining the diversity of co-existing avian influenza hemagglutinin subtypes were “subtype-specific differences in host immune selective pressure and the ecology of transmission (in particular, the durability of subtypes in aquatic environments).” Host immunity is interesting, and something we have worked with ourselves (see here, for instance). The most notable finding was however that they could show the importance of the environmental “storage effect”. Viruses could reenter the population of ducks from the environment – hidden in water or in sediment, perhaps – and start new chains of infection.

The question now is how to prove this, experimentally. Several studies have shown that virus can retain infectiousness for long time if stored in water in the laboratory, and some epidemiological work has suggested that reinfection from the aquatic environment can happen from one breeding season to the next. On the other hand, in our duck trap we see very little evidence for this being a frequent mode of transmission. During autumn, we have a massive deposition of fecal matter from infected ducks in and outside the trap, but when we put in naïve ducks in the trap again in spring, they rarely become infected until the next autumn.

However, perhaps even rare re-introductions of subtypes may be sufficient to re-establish locally extinct subtypes, cause new events of short-lived infection cycles, and then disappear again? Or maybe these virus transmission events are more prone to take place in certain environments than in others? In our case, the duck trap regularly freezes over, sometimes all the way to the bottom, and that may be too hard for our RNA virus friends/foes?

Anyway, always exiting with new well-written articles. Go read it yourself, and don’t forget the 22 supplementary files and appendices.

On a final note, let’s sing:

Where have all the subtypes gone, long time passing?
Where have all the subtypes gone, long time ago?
Where have all the subtypes gone?
Water and mud swallowed them everyone?
Oh, when will we ever learn?
Oh, when will we ever learn?

 

Link to the article:

Roche, B., Drake, J.M., Brown, J., Stallknecht, D.E., Bedford, T. & Rohani, P. 2014. Adaptive Evolution and Environmental Durability Jointly Structure Phylodynamic Patterns in Avian Influenza Viruses. PLOS Biology, 10.1371/journal.pbio.1001931

*******************************************************************************************************************

If you enjoyed this post, or other posts on this blog, why not follow the blog via email, Feedly or get updates via Twitter by following @DrSnygg?

Advertisements

5 thoughts on “Where have all the subtypes gone, long time passing?

  1. if it survives for long time in the environment and then transmits again, (and again…) , then we would
    often see pairs of viruses with fewer genetic differences than the usual ~35 mutations per year.
    While that does happen, e.g. Hong Kong, segment 1, 1975–>2001, it’s rare and I doubt that you
    can find examples from Sweden ?!

  2. it’s complicated and I don’t quite understand it, but I get the impression that they only consider HA and completely ignore the other segments.
    And they only consider North America, where we have no domestic ducks

    • Yes, they only look at HA, and it would have been interesting to see results for other segments too. I have not been sold on the environmental reservoir hypothesis before (especially not the ice tenacity story), but think that they have made a good case and am happy to convert with good evidence. As the famous quote goes: All models are wrong, but some models are useful. The models in this paper are certainly useful, and it will be interesting to see whether there will be some follow-ups on this from the research community in the next coming years.

      As I have alluded to in previous posts, we are finally in position to dig deep in our sequences and hope to have some fresh analyses finalized shortly. The sequence homology you referred to above is worth investigating. I doubt that it is a mislabeling. Also, in any labs there is a risk for cross-contamination. We’ll check.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s