Tornado Alley (PROG)

[fuagf]

Vaccines Will Not Produce Worse Variants

"A vaccine scientist’s discredited claims have bolstered a movement of misinformation"

The more unvaccinated people the virus gets into the more chance you have of a worse mutant
evolving. Those who take an anti-vaccination stance create that situation. Not a vaccine.

10 Sep 2021 By Derek Lowe With Comments

Let's jump back into the coronavirus world and cover a question that I (and many others) have been fielding - President Biden's announcement of tougher vaccine mandates in the US are bringing this one up again. The claim is that vaccination is somehow self-defeating, since all we would be doing is putting more selection pressure on the virus to escape vaccine-derived immunity. Why doesn't that just lead to disaster?

To address a key point that comes up right at the start: no, we have not actually been seeing more dangerous variants occurring since vaccination became more common. It's easy to see the rise of the Delta variant this year and jump to the "after, therefore because" fallacy. But it wasn't even "after" to start with: the Delta variant was first detected in India back in October of last year .. https://www.nature.com/articles/s41586-021-03777-9 . This is before anyone was getting vaccinated. The Delta variant is by far the dominant one in the world, crowding out all the others, and it did not come as a result of vaccination.

But that said, the idea of vaccines affecting coronavirus evolution is not a ridiculous question, and it's worth thinking about to understand more about viral evolution and the effect of our own immune systems. It's certainly true that if you want to induce resistant variations to some antiviral drug in the lab, you let them infect cells in culture while treating them with not-quite-adequate amounts of your proposed drug. You "passage" these into fresh cell cultures, often increasing the amount of the antiviral along the way as you do so, with the idea being that the viral particles that are infecting each new population of cells are the ones that have made it past the effects of the drug. This can go on for weeks or even months, depending on the organism, the drug, and how easy or hard it might be to stumble on effective resistance mutations, but at the end of it, you can produce a form of the virus that will laugh off what would be a killing dose against the original wild-type.

The same general idea is used to produce resistant bacteria, although you don't need to keep putting them in with other cells (you just grow them in some sort of nutrient medium, with the same sort of long-term exposure and/or gradual increase of antibiotic compound). Now, you may be asking why anyone would do such a thing: isn't producing resistant organisms sort of a bad idea? Well, for one thing, you want to use good technique and containment any time you're doing these things. And this sort of experiment is done well before any such investigational drug is put into trials, much less out on the market, and the resistant organisms you've produced are very likely to have traded off something else to get that way. In the absence of pressure from the drug, odds are they are in fact slightly less fit to compete against their wild-type kin and will be crowded out.

One big reason to do such an experiment, though, is when you have a new compound that worked in a phenotypic assay, but you don't know its mechanism. In these modern times of easy sequencing, you take your newly resistant organisms and compare their genomes with the originals to see what genes have changed under the selection pressure. There may well be several, but odds are good that one of them is the target of your drug candidate or at least in its pathway. And seeing what amino acids have changed in the corresponding protein(s) is usually very informative. Even if you have a good idea about the mechanism already, a resistance experiment might tell you some other things about exposure to the compound that you didn't know (and probably should). And in general, if you're coming up with a new antiviral or antibiotic, you want to know just how easy it will be to evade. If you try this induced-resistance experiment and the organisms dodge out of trouble in the first ten days, that's not a real good sign. On the other hand, if you blast away for weeks and months and still can't see useful resistance mutants popping up, that could be a really good feature.

OK, that's resistance as most of us understand it. Now, when you're talking about resistance to a small molecule drug, the sorts of mutations that show up can be things like an amino acid changing in an active site, in a way that doesn't keep the target enzyme from functioning, but does block your new enzyme inhibitor from getting and binding. Or it could be another amino acid changes somewhere else in the protein, altering the 3-D structure around the active site in the same way - not enough to kill its function, but enough to keep your new drug from getting to its site of action. When you're working with bacteria, you always have to look out for their habit of sharing plasmids, loops of DNA that swap new proteins (and their new functions) in and out. For example, a common way that bacteria became resistant to penicillin-type antibiotics was to pick up a plasmid from some other resistant organisms that coded for the beta-lactamase enzyme, a protein that would efficiently break the key ring structure in all of those drugs and inactivate them before they had a chance to do their work. Before the advent of such drugs, the only bacteria that carried such a thing around were the ones that frequently encountered the other organisms (molds, etc.) that secreted penicillin and related compounds. For others it would just be a useless burden, like walking around with a chain saw strapped to your back. But if the landscape suddenly fills up with downed trees blocking the roads and sidewalks, that instant-chainsaw gene becomes a hot commodity.

What about resistance after vaccination, then? Keep in mind the various sorts of immune attack that vaccination can enhance. A big one is the ability of specific antibodies to recognize the surface proteins of the pathogen (we'll stick with the virus example from here on). In the case of most of the vaccines available now, these are antibodies to the coronavirus Spike protein, since that's what the adenovirus vectors, mRNA sequences, or (in the case of Novavax) engineered proteins are all providing. These antibodies can bind so thoroughly and tightly to the virus particles that their Spike protein (especially its business end, the receptor-binding domain that targets the human ACE2 protein on the cell surface) is no longer able to function. Antibodies bound to a pathogen like this also trigger other immune cells to attack and clear the whole antibody/target complex; it's like slapping on a bright orange Waste Disposal tag.

And past antibodies, there's the whole T-cell system - those T cells are looking to recognize human cells that have been attacked by the virus, whereupon they move in to kill them off before they can break open like a piñata and release a big pile of new viral particles. They are primed for this task by having pieces of viral proteins presented to them by other immune cells, and these T cells become specifically sensitized to the appearance of these in the future - these pieces get taken to the surface of infected cells by the MHC glycoproteins where the T cells can detect them. So you can see that in all of these cases the key is protein surface recognition, which tells you how viruses can work their way around to evading such attacks. They have to change their surface proteins in such a way that they can still function, but that defeats that antibody/T-cell binding that the immune system has settled in on. 

That's not so easy, because (for one thing) there are an enormous number of different antibodies involved (and an enormous number of T-cell recognition proteins). There are any number of ways to bind to a given protein target, and the adaptive immune system's whole function is to be ready for all kinds of targets and to hit them in all kinds of ways. And there's that constraint mentioned above: the virus still has to be able to function! Losing the entire Spike protein or mutating it completely beyond recognition would definitely evade vaccine-induced immunity, but it would also definitely produce a coronavirus that couldn't infect human cells in the way it's completely evolved to do. Coming up with a completely new infection route is (mutationally) extremely costly and complex, and not something that can be done "on the fly". Various coronaviruses use different human cell surface proteins to do their attack, but these have gradually developed and diverged over evolutionary time (hundreds of thousands, or millions of years) through untold numbers of tiny steps.

But it can be done, in principle. And as with everything in evolution, if it gets done at all, it'll get done by similarly untold numbers of individual mutants, and mutants on top of those mutants, until something appears that can both avoid being inactivated by the immune response and still infect cells and reproduce. There is no guarantee that such a virus can exist, and there is no guarantee that it can't. Evaluating the number of possibilies is frankly beyond computation - we didn't, for example, see the details of the Delta variant coming, and if you'd given someone that exact sequence last year, there's no guarantee that they would have been able to predict how much more infectious it would be. 

The more chances you give the coronavirus to reproduce, the more mutations it will explore. Its proofreading system for reproduction is pretty good but not perfect, and that's where the mutations come from. It's a numbers game all the way. The virus is not thinking about how to evade vaccine-induced immunity; it's throwing stuff randomly against every available wall in every available direction, and whatever sticks gets a chance to go on throwing some more. Remember, an unvaccinated person is still mounting an antibody defense against the virus - they're just having to do it from scratch, rather than having a pre-primed leg up like someone who's been vaccinated. The longer these infections go on inside human bodies, the more bets the virus gets to put down on the table. The good news is that so far, there is not much evidence .. https://www.nature.com/articles/s41577-021-00544-9 .. that the virus is doing much evasion inside a given person during the course of normal infection. 

So one key way to cut down on the odds of a nasty mutant popping up is to just keep the virus from reproducing so much. Cut down on the number of people it infects. When it does infect people, cut down on the amount of time it spends reproducing inside the body. These countermeasures are exactly what a mass vaccination program does. Fewer people get infected in the first place, and when they do get infected, their disease course tends in the great majority of cases to be shorter and milder. A nasty variant is almost certainly going to come up by accident, so let's not have so many accidents going on constantly around the clock, around the world.

But back to the earlier discussion: what if the vaccines are still putting direct pressure on the virus? Aren't we selecting for exactly the things we fear the most? The answer to that is counterintuitive. Take a look, for example, at this preprint .. https://www.medrxiv.org/content/10.1101/2021.07.01.21259833v1 .. from July. The authors looked over 1.8 million coronavirus genomes from infections around the world, and compared that background data set to specific breakthrough infection sequences in vaccinated patients. What they find is that the genomic sequences from the breakthrough infection patients are significantly less diverse than what's seen in the wild. The authors believe that this shows that "COVID-19 vaccines are fundamentally restricting the evolutionary and antigenic escape pathways accessible to SARS-CoV-2", and that's the flip side of the above argument. You are putting pressure on the virus to escape the immune attack, but at the same time you are cutting sharply back on the pathways it can use to get there. Remember, a true vaccine-evading mutant is going to need a set of several mutations (off the existing variants) all at the same time. The vaccine-induced immune response looks like it's knocking down a lot of these intermediate-step mutations before they can keep on throwing off subsequent mutations on top of the first ones. These pathways are choked off before they can even get explored, and this "evolutionary smothering" is something that you don't see so dramatically when you're doing those in vitro experiments with specifically targeted small molecules mentioned at the top of this post. A broad antibody and T-cell response is a different thing altogether.

There is, then, every reason at both the population and individual level to expect that vaccination will strongly decrease the chances of a more dangerous coronavirus strain taking hold. If we'd had them earlier and were able to deploy them quickly and widely enough, we never would have seen the Delta variant in the first place. If we keep deploying them now, we will keep worse variants from even being able to form. Anyone who tells you that vaccines will make things worse is at best deeply misinformed and at worst lying to you for profit.

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He’s worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer’s, diabetes, osteoporosis and other diseases.

https://www.science.org/content/blog-post/vaccines-will-not-produce-worse-variants