As Omicron mutates wildly the virus shows first signs of convergent evolution
Over the last couple of months researchers tracking emerging SARS-CoV-2 variants have started noticing something strange. No one new variant has looked like taking over but instead a variety of different subvariants seemed to be accumulating the same mutations.
These mutations all seemed to be converging in a way to evade our pre-existing immunity, and a striking study recently appeared speculating the virus has the potential to completely escape our current immune responses. As many people around the world return to normality, deeming the pandemic over, these new evolutionary signals suggest we may be done with thinking about COVID but SARS-CoV-2 is most certainly not done with us …
The Twitter Trackers
Twitter has been described by some as a town-square, analogous to a massive open space where anyone can offer their two cents on anything. But in practice it’s actually nothing like that. A more apt analogy would be Twitter resembles a massive apartment building filled with countless rooms devoted to conversations between people with shared interests or beliefs.
Inside one small room in the gargantuan Twitter high rise resides a diverse assortment of virologists, infectious disease researchers, epidemiologists, data visualization nerds, and ambitious armchair experts. They all focus on one particular subject – tracking the genetic mutations of SARS-CoV-2.
Wading into the world of Variant Tracking Twitter can be dizzying for the uninitiated. A whirlwind of dense terminology accompany tweets filled with graphs tracking coded mutations and increasingly complicated variant names. Pango lineages, GISAID data, Nextstrain clades.
In September a word started popping up frequently amongst the coded conversations citing RBD mutations in K444T and N460K. That word was "convergence."
The great convergence: https://t.co/cs1aqQZioD https://t.co/V0fBCFcUfI pic.twitter.com/GeMhn7oUis— Theo Sanderson (@theosanderson) September 21, 2022
For the last two years the evolution of SARS-CoV-2 has been strange, to say the least. Across 2020 experts frequently warned of the potential for this novel coronavirus to mutate, but initially, it remained remarkably unchanged until a trio of variants emerged late in the year.
Alpha, Beta and Gamma all suddenly popped up. Three different lineages, in three different parts of the world. All with relatively similar mutations.
The changes had begun and in 2021 we saw a series of infection waves cross the world, each one driven by a new variant. Alpha leading to Delta leading to Omicron. What was particularly unusual about these successive waves was each subsequent variant was different from the one that preceded it. This wasn’t the case of a single lineage slowly mutating and changing but big evolutionary leaps were taking place, with new lineages coming out of nowhere.
Since Omicron hit in late 2021 and early 2022 the path of the pandemic seems to have changed. Instead of big evolutionary leaps the Omicron lineage has splintered into scores of different subvariants.
Omicron first emerged with three relatively independent arms (BA.1, BA.2 and BA.3) but now the lineage features a head-spinning array of subtypes. And since the recent BA.5 wave no single subvariant seems to be dominating. Instead, researchers are watching all these different subvariants slowly start to take similar shape, with the same mutations appearing to be successful across multiple independent evolutionary pathways.
What we call (or don't call) a virus plays an important role in how we frame our communications and responses to it. Here is a long thread about how SARS-CoV-2 variants are named.🧵 pic.twitter.com/IjeZBC4GfJ— T. Ryan Gregory (@TRyanGregory) October 3, 2022
The Great Convergence
In the world of evolutionary biology the theory of convergent evolution explains how completely unrelated organisms can separately evolve incredibly similar traits. When faced with the same selection pressures distinctly different organisms can often end up with fascinating similarities.
One of the most commonly cited examples of convergent evolution is the striking similarity between shark and dolphin bodies. These two organisms share no common ancestor yet they evolved to look alike based on facing the same selection pressures.
In terms of SARS-CoV-2, many of us may be familiar with the hearing about the virus’s unique spike protein. It’s the highly identifiable protein that our successful vaccines use to generate protective immune responses. One part of the spike protein is known as the receptor binding domain (RBD). It sits at the top of the protein and helps the virus attach to certain receptors in human cells, a crucial process for infection and replication in our bodies.
Our original vaccines, using a version of the 2020 Wuhan coronavirus spike, are still incredibly effective against current Omicron variants with many mutations because until now there have been very few changes to the RBD.
A recent study, available as a preprint and yet to be peer-reviewed or published in a journal, presented a striking set of data suggesting a number of new Omicron subvariants are all appearing with similar mutations on the receptor binding domain. In an email to New Atlas, corresponding author on the paper Yunlong Cao explained the similar mutations across multiple independent variants indicates SARS-CoV-2 is successfully finding ways to evade our current immunity.
“RBD convergent evolution, means that the RBD mutations evolved by the recently emerged SARS-CoV-2 Omicron lineages converge on the same sites (hotspots), including R346, K444, V445, G446, N450, L452, N460, F486, F490, and R493,” Cao explained. “Seeing this convergent evolution pattern would mean that SARS-CoV-2 would evolve immune-evasive mutations much more frequently than before, and the resulting new variants would be much more immune-evasive.”
Sharing our investigation on the unprecedented convergent RBD evolution of BA.2.75 and BA.5 on sites including 346, 356, 444-446, 450, 460, 486, which have generated highly concerning variants such as BA.2.75.2, BR.1, BJ.1, and BQ.1.1. (1/n)https://t.co/itJGuLfW3y— Yunlong Richard Cao (@yunlong_cao) September 16, 2022
Hunting the Cryptic Lineages
In mid-September the first draft of the study by Cao and his colleagues was published on a preprint server. At the same time, in a serendipitous case of research convergence, variant trackers on Twitter were simultaneously reporting similar findings.
One researcher dubbed it “The Great Convergence,” as variant trackers consistently reported finding similar mutations across drastically different subvariants. As an example of convergent evolution, it seemed SARS-CoV-2 was finding certain mutations universally successful.
Marc Johnson, a microbiologist from the University of Missouri, was among the first variant-tracking Twitter researchers to begin visually graphing these cases of convergent evolution. He is one of the few researchers to be completely unsurprised by the mutations starting to appear in subvariants around the world. In fact, he had seen exactly these same mutations over a year ago.
For the past 18 months Johnson has been obsessed with tracking what are called cryptic lineages. These are extremely rare, distinctive forms of SARS-CoV-2 with massive suites of mutations not seen in any circulating variant.
I did some book keeping today. We're up to 24 cryptic lineages, and hope to have #25 by the end of the week. I tabulated some of the numbers. If it is of interest to anyone, these are the RBD aa sites with the most frequent changes. pic.twitter.com/5ly2nozLdq— Marc Johnson (@SolidEvidence) September 14, 2022
The work starts with wastewater, tracking traces of virus variants detected in sewage samples around the world. The current hypothesis Johnson is working with is these cryptic lineages are the result of long-term COVID infections, the virus primarily persisting in a person’s gut.
“What we think is happening is that there's these patients that can't clear the infection,” Johnson explained in an interview with New Atlas. “And the virus, because there's no bottlenecks from spreading from person to person, it just hits the evolutionary fast forward button. And it's evolving way faster than the circulating lineages.”
Johnson is clear that he doesn’t believe these cryptic lineages are anything to be concerned about from a circulation perspective. For the most part, the viral fragments he is picking up in wastewater are inactive. So it's unlikely these cryptic lineages will seed the next major variant.
Instead, these cryptic lineages offer a kind of perfect mutational laboratory, allowing Johnson insights into the potentialities of SARS-CoV-2. His obsessive characterizations of cryptic lineages over the past year and a half has resulted in him being completely unsurprised every time a novel mutation appears in a circulating variant.
Not only had he previously seen the mutations that appeared late last year in Omicron, but he’d been waiting for these new convergent mutations currently discussed to finally appear in circulating variants.
“When Omicron arrived, I'm like, check, check, check, check, check, check, check,” Johnson said. “All 11 of them were mutations I had seen before, but then some of the ones that I had seen a lot before that were not in Omicron were 452, 460, 346 – the big ones now that it's finally picking up. So for about a year, I've been saying when is it going to pick up these other two mutations?”
@CorneliusRoemer @PeacockFlu @siamosolocani @LongDesertTrain @JosetteSchoenma— Marc Johnson (@SolidEvidence) September 21, 2022
I'm trying to keep these convergent lineages straight in my head. Is this accurate? I limited it to lineages with L452R+N460K+(K444* or R346T) to keep it "simple", so BM and XBB were left off. pic.twitter.com/iSx5C8EOUV
Around the same time Cao published his preprint study, Johnson posted a graph on Twitter trying to map out the most common convergent lineages. Within days a relatively simple visual depiction of convergent lineages had become a kaleidoscopic mess of connections. Resembling one of those pinboards you see amateur detectives in movies use to connect clues in a crime, the convergent lineages being followed by variant trackers on Twitter were moving fast. Every day new lines were being connected as different sub variants converged on the same mutations.
Surprisingly, the novel 'level 6' BA.4.6 derivative that I noted last week was not detected this week.— Marc Johnson (@SolidEvidence) October 7, 2022
Last week that lineage only matched a single sequence in the database (from Australia).
Now there are 5 matches from 4 countries (including US) and 3 continents. pic.twitter.com/3HLo8IpL4B
Original Antigenic Sin
So it’s clear a number of different SARS-CoV-2 subvariants seem to be converging on the same mutations, but what does that mean? If these particular mutations are offering the virus growth advantages, helping it slip past our immune defenses, how is that happening?
This was the question Yunlong Cao and colleagues wanted to answer. In a preprint described as a “tour-de-force set of deep mutational scanning experiments,” the researchers probed thousands of antibodies isolated from vaccinated people who experienced breakthrough COVID infections.
The goal was to understand what kind of immunity exists in a person who had been vaccinated and subsequently infected with Omicron subvariants BA.2 or BA.5. The million-dollar-question: As the virus drifts away from its original 2020 form, are we concurrently developing immunity to its newer iterations?
Cao’s study discusses a 60-year-old theory known as original antigenic sin. Back in 1960 an epidemiologist named Thomas Francis was studying the historical ebbs and flows of influenza epidemics.
Francis suggested the body’s first exposure to a pathogen can leave a permanent immune imprint, or memory. This "original sin" can hinder our ability to fight that same pathogen if that pathogen begins to change its shape and become less recognizable.
Subsequent research since the 1960s has somewhat validated the idea of original antigenic sin. When we are young and our immune system faces a new pathogen for the first time our naive B cells learn to produce the right seek-and-destroy antibodies.
That initial encounter creates memory B cells. These immune cells can exist for decades, patrolling the body on the hunt for that same pathogen. This is how our immune system can quickly respond to future infections.
But the problem with this technique is that the immune system tends to lean on these memory B cells and not learn how to recognize newer versions of pathogens as they slowly evolve. For the immune system, near-enough is good-enough as long as the original imprint still somewhat recognizes newer shapes of pathogenic invaders.
One of the crucial findings in Cao’s study was that when a vaccinated person experiences a breakthrough Omicron infection they primarily recruit memory B cells to produce antibodies. In fact, according to the study, 80% of the B cell response to Omicron breakthrough infections are existing vaccine-induced memory cells.
4⃣ They show (as also demonstrated by other studies) that boosting (in this case, breakthrough infection) with Omicron ends up ~80% recruiting existing Wuhan-Hu-1 specific B cells and ~20% eliciting new B cells. pic.twitter.com/M7SLvCtnvV— Bloom Lab (@jbloom_lab) September 17, 2022
Generally this is a good thing. It is why the original vaccine with the 2020 spike protein still offers good protection from Omicron. It's also why many vaccinated people who have experienced an Omicron breakthrough infection likely faced a relatively mild disease. But crucially it means our immune systems may not be learning how to recognize newer Omicron subvariants.
“The advantage of this [memory B cell response], and the reason why our body evolved to behave like this, is that when infected with Omicron, our body could rapidly generate antibodies against the new antigen through memory B cell recall, not depending on eliciting antibodies through naive B cell maturation which is slow,” Cao explained to New Atlas. “However, the bad part is that infection with Omicron would not very effectively broaden the breadth of our neutralizing antibody repertoire since our immune system majorly reuses the memory B cells elicited by [original] vaccination.”
The new wave is more like a rising tide
The next part of Cao’s study has perhaps been the most contentious amongst COVID researchers, as more hyperbolically inclined commentators have inferred apocalyptic outcomes. Here, the scientists looked at the specific mutations that many subvariants seem to be converging upon right now. These mutation hotspots seem to be converging on the RBD of the coronavirus spike. And these are the spots our previous memory B cell antibodies were targeting to protect us from Omicron infections.
This is how the virus is mutating to evade our immune defenses.
So the scientists set out to model what future mutational pathways the virus could hypothetically take to evade our current antibodies. Adding just six specific mutations to BA.5 was found to be enough to evade most current RBD antibodies.
And most significantly, several subvariants have already been detected with some of these mutations. Subvariants BQ.1.1 and XBB in particular were found to be the most immune evasive, escaping even Omicron-specific antibodies.
A couple of weeks ago BA.2.75.2 was shown to be the most immune evasive variant seen to date and replicated. That's now been surpassed by XBB, which is predicted here may challenge BA.5 bivalent vaccine protection https://t.co/qWJ4AGtiur by @yunlong_cao and colleagues https://t.co/HOoz6pkmEm pic.twitter.com/rfh2bX37y4— Eric Topol (@EricTopol) October 4, 2022
What happens from here is decidedly unclear. Immunologist Menno van Zelm has studied ideas of immune memory and original antigenic sin. He said it’s still an open question whether immune imprinting is a lifelong thing or whether immune responses can evolve alongside a virus.
“We've been very happy to show that immune memory after effects can last for a long time,” explained van Helm. “But it could bite us now that after years, you still have that memory and that memory could still outcompete any new responses. We honestly don't know. And it might be a balance, right? Because as soon as you don't have enough response anymore, it means that you generate completely new responses, and then your memory will not outcompete.”
Van Helm is also clear to stress our immune responses are complex and multi-faceted. Cao’s research focuses solely on memory B cells and the antibody responses generated. These immune responses do govern basic levels of transmission and infection but plenty of other factors influence how severely sick we become from an infection.
“We have our T cells,” van Helm added, referring to the main immune cells that kill and clear most pathogens. “Many people have, since they've been vaccinated, also been infected. So they have responses of T cells and B cells to other parts of the virus. So we're not completely naive, as we were back in 2020. And I think that is part of why the disease severity is much lower now.”
Cao is a little more wary of the potential increase in disease severity from these new converging subvariants. He speculates it is possible T-cell responses are maintained in the face of these new RBD mutations, as those immune cells can target more conserved parts of the virus that have yet to shape-shift. But Cao also speculated there's potential for a future iteration of the virus that can completely evade antibody neutralization, cause high rates of infection, and generate severe disease in some people.
“T-cell response could make some contribution to preventing severe symptoms and deaths,” Cao said. “However, they do not efficiently prevent infection and transmission; otherwise, there would not be the BA.5 wave after BA.2. [And] it is possible that [new variants] cause more severe symptoms, as vaccines and the previous infection would not provide humoral immunity protection, both from the memory B cell level and sera neutralization.”
Both Cao and van Helm do agree that current variant-specific vaccines may not be the most optimal path forward but they are certainly the best solution we currently have. Despite evidence of original antigenic sin there is also strong evidence that boosters are beneficial in lowering risk of serious disease.
Nevertheless, the pandemic certainly isn’t over and scientists are still racing to keep vaccines and therapeutics ahead of where the virus is going. One suggestion from van Helm to potentially bypass the problem of original antigenic sin is to create a new vaccine designed to evade current memory B cell responses and generate a whole new immune imprint.
“It might be worth by taking only the parts of the protein that are new, and vaccinating only with that, so that you only generate a new response,” van Helm speculated. “And that will then be in addition to the original memory that you have towards the wild type virus.”
And while those scientists work on the next generation of therapeutics, the variant trackers are still obsessively following the rapid viral mutations in granular detail.
The Twitter Trackers Pivot
In the past, new pandemic waves had been easy to identify. A single variant emerges and genomic testing around the globe quickly shows it growing in prevalence. But now, with the dramatic fragmentation of Omicron into dozens of subvariants, new trends are harder to detect.
A number of variant trackers have recently shifted their surveillance tactics to try and get a better grasp on emerging waves. Instead of setting their systems to follow specific variants some are now grouping together variants with convergent mutations. By rolling variants with shared mutations into similar “buckets” it quickly becomes evident that these new SARS-CoV-2 iterations are becoming prevalent.
With 3 more days of data, here's a look at how far various collections of immune escape variants have progressed so far.— Cornelius Roemer (@CorneliusRoemer) October 6, 2022
I now count S:493 as a key mutation, so what used to be >=5 (Pentagon) is now >=6.
The more RBD mutations, the faster the growth.https://t.co/RJVwphrqXV 1/ pic.twitter.com/tLD1FvIai2
Cornelius Roemer, a viral evolution researcher based in Europe, has been at the forefront of chronicling the global growth of these new convergent variants. His monitoring has shown the more key RBD mutations in a subvariant, the faster its growth.
Roemer’s tracking suggests subvariants with four key RBD mutations may account for more than 30% of new viral samples up to the end of September. But add another two to three key RBD mutations and you see some extraordinary new subvariants (XBB, BQ.1.1 or BA.2.3.20) emerging and growing at rapid rates.
Where this is all going is anyone’s guess. Will one of these subvariants ultimately dominate as in previous waves? Or will we face what some have referred to as a "variant soup," with several similar subvariants becoming prevalent in different parts of the world?
Marc Johnson describes the current emerging wave as more like a rising tide, and as all these subvariants converge on the same mutations we will see rising caseloads. And he says the rate of the virus mutating is “ridiculously fast.”
From his work with cryptic lineages Johnson suggests the virus still has a lot of room to move in terms of potential space to mutate. Whether that means it becomes more or less severe is unknown, but with such a massive volume of ongoing infections allowing new mutations to flourish Johnson expects to see plenty more novel lineages appearing in the future.
“I mean, people say when’s it going to reach the optimum, like it's a moving target. It depends what immunity it’s up against. It will just keep adapting to it. It might go full circle one day, who knows?”
Please keep comments to less than 150 words. No abusive material or spam will be published.
Lockdowns did nothing, "social distancing" did nothing, the vaccines are about as effective as flu vaccines, as everyone credible predicted, and despite all that even the Antarctic research stations had multiple outbreaks, with 100% control of everything going in and out.
Which is what we expected.
Nice article Rich, in USA the medical community is wary of the "escape" variants. If the virus cannot thrive in the vaccinated - it would either mutate in the wild where it once again could infect and produce trillions of copies - or it would take every opportunity in a vaccinated human to "test" a degree of mutation survive-ability. The fact that these viruses mutate so quickly give the various strains and edge against our vaccines - and against our preliminary B-cell antibodies.
SARS-CoV-2 can and will mutate into a less deadly virus - there are a few varieties that are mild like a common cold. But it can mutate just as easily into a more virulent and deadly strain but if it also escapes the vaccine induced antibody recognition, we will have another Spring 2020 and worse! Lets hope that the vaccine researchers can target a few mutations in the RBD domain before it takes off and proves me right!
Remember that the Army doctors geno-typed every molecule of the coronavirus and set out to understand what each portion of the virus does and how it does it. This is why it took a longer time to develop it.
I look forward to the opportunity to get the last shot I will need for covid prevention.