I was prompted to go on a bit of a rave – pardon, to expand on some long-held opinions – on Bluesky the other day (@edrybicki.bsky.social) after reading this post from Daniel Goldhill (@influenzal.bsky.social):
“I wrote a short piece about a new flu subtype (H19) which uses a surprising receptor. We’ve known about glycans terminating in sialic acid as a receptor for flu for 70 years but the virus still surprises us! Great work by Karakus et al. “
I quoted his comment – “Perhaps, the ancestors of all extant viruses had to be flexible in their ability to switch to new receptors.” by saying:
“And perhaps quite a few still are??”
So let me expand a bit on that: it’s a bit dismaying how many people are still wedded to the “one enzyme/protein, one function” paradigm – which was never true, and which basically was a convenient fiction to make biochemists happy. The fact is that proteins are happy to bind to ANYTHING you can think of, and probably quite a few that you can’t, by simple virtue of accidental molecular complementarity, somewhere in their structure – and given that classic virus envelope glycoproteins are pretty big AND multimerised, that’s a lot of structure. Meaning that, like any good scientist, envelope proteins can hold more than one receptor in mind, and some of their binding may be completely adventitious AT FIRST – however, if it is selected for (more virions enter cells, or do it faster), then suddenly it becomes a virtue, and – new tissue tropisms / higher infectivity result(s). I am spurred on here by the example of SARS-CoV-2 S protein, which has mutated in front of our morbidly fascinated eyes since early 2020 to increase its binding affinity to human ACE2 by orders of magnitude, which has resulted in greatly increased infectivity and a significant increase in R0 for virions.
Thus, expecting that any given given viral membrane glycoprotein will only bind ONE receptor is naive: it will bind one (or more) WELL, and others badly – until selection occurs, because the new property is a survival mechanism when faced with neutralising Abs, for example. Viruse genomes are plastic: they change quickly to adapt to adverse environments, and even small changes that confer only slight advantages, will be selected for, and then reselected and improved over time. That this could result in the HA of an influenzavirus being first, able to bind another molecule than the sialic acid-bearing glycoproteins that are the canonical receptors; second, improving that binding by selection of mutations that improve binding affinity; third, losing the ability to bind the original receptor while preserving the new interaction, should not come as a surprise to any serious virologist. Moreover, these sorts of receptor changes / additions are almost certainly still going on, rather than being an ancestral condition.
And yes, this is going into the new edition of Cann’s Principles of Molecular Virology, whenever that comes out 😁
ViroBlogy 