The Delta to Ba.4/Ba.5 COVID Pipeline

A working, preliminary model for the current Omicron phenomenon through the examination of antigens and immune escape.

With the Ba.4/Ba.5 wave hitting the US and fears brewing up again, I’ve been trying to figure out the current predicament and a way of explaining how we got from Ba.1 to the current COVID variant.

There is the current argument that this variant is being driven by the vaccinated, but I personally don’t find this to be a satisfactory argument, mostly because these arguments don’t take into account the sudden emergence of Omicron and it’s position in the COVID timeline.

Similar to gaps in fossil records that may span hundreds of thousands of years with a sudden emergence of new species, we ourselves have a evolutionary gap between Delta and Omicron which was never properly addressed.

My original posts on Omicron suggested a possible mouse origin, but as Brian Mowrey has continuously written (including a recent post), the evidence appears to point towards a lab origin for Omicron:

Omicron Origins: Summary / Spoiler

What a strange turn of events if Wuhan and it’s possible lab origins was uprooted by another virus of lab origin (same lab? different lab? no information in regards to either unfortunately).

But if this were true, it completely confounds the COVID timeline, creating the supposed Delta/Ba.1 Omicron gap that would have been, under these circumstances, artificially produced.

So in thinking of how to figure out our current predicament, it may be best not to think of it through an evolutionary lens- it certainly won’t explain the Delta → Ba.1 predicament.

But a virus is still gonna virus, and so we may figure out the trajectory through mapping of the virus’ spike and comparing how different they are to one another.

Omicron, everywhere, all at once

This heading probably isn’t the most accurate, but it gets pretty close to the matter. It’s important that we take into account the emergence of the Omicron family.

In thinking of the Omicron lineages, I considered them to have been all been related in some manner i.e. Ba.1 led to Ba.2 and so on. Of course, Brian Mowrey pointed out that Ba.1 and Ba.2 were actually distinct. In fact, in an era of uncertainty and ambiguity it appears that there’s a clear distinction between Ba.1 and Ba.2, such that neither emerged from the other.

This was substantiated by Willaim A. Haseltine in his Forbes article:

As well as in studies such as the Mykytyn, et. al.¹ study and the antigenic map they designed:

So where did both lineages derive from? Brian posits in his post above that a progenitor Omicron (he labeled Ba.0) likely went through different passages and tests that led to the eventual divergence (i.e. Ba.0 was used for different serial passage tests leading to distinctions between Ba.1, Ba.2, and Ba.3.)

That’s not for me to argue (one would expect the medical establishment to do that…), but this brings up a critical point.

Up to now other variants emerged from distinct parts of the world, such as Alpha emerging from the UK and Delta emerging from India. But here we have distinct strains of a virus all linked to similar geographical regions such as South Africa.

This includes the Ba.4/Ba.5 lineage, which was not detected until two months after Ba.1 and Ba.2 were detected based on the reports from South Africa²:

The Omicron lineage BA.4 was first detected from a specimen collected on 10 January 2022 in Limpopo. As of 29 April 2022, BA.4 has been detected in all provinces. The percentage of sequences designated BA.4 has grown from <1% in January 2022 to > 35% in April 2022 (3).

The Omicron lineage BA.5 was first detected from a specimen collected on 25 February 2022 in KwaZulu-Natal. As of 22 April 2022, BA.5 has been detected in Gauteng, Limpopo, Mpumalanga, KwaZulu-Natal, the North West, and the Western Cape. The percentage of sequences designated BA.5 has increased from <1% in January 2022 to 20% in April 2022 (3).

This too becomes extremely important for our working model, because it completely dampens the odds that Ba.1, Ba.2, and Ba.4/Ba.5 all began circulating the globe at different timepoints. The detections in South Africa eventually leading to detections everywhere for each subvariant seems nearly impossible, unless someone on business travel every other week happened to keep getting infected and carrying different subvariants all across the globe.

This also doesn’t coincide with Ba.1 and Ba.2 being divergent viruses yet appearing at the same time in South Africa, or that Ba.4 and Ba.5 were detected at different timepoints but are now circulating together worldwide.

The only way to explain all of this, is to argue that all of these Omicron subvariants had to have been circulating together as soon as they emerged from South Africa.

That means that, as soon as Ba.1 began hitting the US hard, it’s likely that Ba.2 and Ba.4/Ba.5 went undetected but were widely circulating among the population, waiting for their moment in the viral spotlight³.

This also creates a new conundrum. Unlike prior variants which (we assume) went through selective pressure to eventually emerge as the dominant strain, all of the players of Omicron were already around.

There was no need for them to go through selective pressure- they just had to wait for the right moment to make themselves known.

In short, we know that the collection of Omicron subvariants are extremely distinct from Delta, and likely explains the immune escape that was seen with Ba.1 along with other subvariants. We also are working with the assumption that all subvariants of Omicron (Ba.1, Ba.2, Ba.4/Ba.5) were circulating at the same time, waiting for the right moment to fill the appropriate niche.

Now, with all the players all we need to do is rationalize the timeline, and that requires us to look at the immunity landscape for that answer.

Mapping it Out

The Wuhan → Delta Pipeline

In order to figure out the Omicron pipeline as we are seeing it today, we need to take into account the immunity landscape before the onset of Omicron.

Prior to Omicron, all variants and subvariants came from the Wuhan strain in some way, with selective pressure and antigenic drift slowly leading to mutations that either conferred greater virulence and transmissibility to the virus, or that helped aid in immune escape.

Generally speaking, variants with the above characteristics (that aid in immune evasion or greater infectivity) win out against other mutations- the story of variants is one told by the victors.

Thus, the immunity landscape- the type of immunity that the general population carries⁴- somewhat dictates which variant dominates. Variants that are “same-same, but different” may not cut it- they may be too similar to an already circulating variant, and prior immunity among the population may make it difficult for these new variants to find a niche. However, variants that are different enough can find a niche within the population by escaping prior immunity and begin to dominate.

In essence, in order to figure out which variant may dominate, we can turn to their spike antigen and see which ones are the most distinct/distant from one another.

Let’s consider the Wuhan pipeline of variants going from:

Wuhan (D614G) → Alpha → Beta → Delta

and use the antigenic map from the Mykytyn, et. al. study above.

Antigenic cartography (mapping) is a way of mapping antigens and their distance from one another. Generally, the more distant one antigen is from another one the more distinct they are, usually alluding to higher rates of mutations or greater loss of neutralizing antibodies between the two antigens.

Mapping utilizes results from neutralization assays, so there’s an inherent bias based on the testing procedures done⁵. Nonetheless, these assays help to discern which antibodies would work against which antigens, and in doing so they map which variants may have the greatest level of immune escape from one another.

So when looking at this map, understand that the greater the distance between the variants the greater the differences in their antigen.

My (crudely) drawn outline of the antigenic distance between different variants. Alpha was one of the early variants, and in a landscape that was either naïve to COVID or one that lacked other variants the shift from Wuhan/D614G spike to Alpha may not have needed to be a big leap. However, the move from Alpha to Beta, and then Beta to Delta, at least indicates the importance of the immune landscape in shaping which variant dominates. It’s interesting that studies that look at RBD and neutralizing antibodies usually show Beta as being distinct from responses to other strains such as Wuhan or Alpha, and in some ways it makes sense given its position on the map above.

Now, aside from my crudely drawn arrows you can see the general distance between the variants. The D614G/Wuhan variant did not differ much from Alpha, and it’s likely because not many variants were in circulation at the early onset of the pandemic. However, at the point of going from Beta to Delta you can see that they were quite different antigenically speaking. Remember that the map could be biased by mutations that greatly reduce neutralizing antibodies, so keep that in mind.

What this tells us is that the immunity landscape can dictate which variant would be favored. In a landscape of mostly circulating Beta, an antigenically distant Delta may come to dominate due to greater immune escape.

At the time of the Delta wave there was an immunity landscape mostly shaped by either Delta via natural infected or Wuhan via vaccination.⁶

Until, however, a brand new family of variants called the Omicrons emerged, and with it caused a huge shift in the mapping.

The Omicron Ba.1 → Ba.4/Ba.5 Pipeline

This point has been hammered down pretty well by now, but the coming of Omicron in an immunity landscape full of Delta and Wuhan immunity (among others) likely led to the widespread infection of Omicron because of how distant Omicron was from the prior circulating strains.

More crude lines, but note that if we consider an immunity landscape shaped by Delta, both Omicron Ba.1 and Ba.2 are highly distant relative to Delta. This explains the high number of immune escape and loss of neutralizing antibodies.

So we know why Omicron came to dominance. We just need to figure out why Ba.1 was the first to dominate, followed by Ba.2 then Ba.4/Ba.5.

For this, we can turn to another antigenic map from Tuekprakhon, et. al.⁷

Antigenic map which includes both Ba.3 and Ba.4 (Ba.5 shares the same spike anitgen mutations as Ba.4). Note that Ba.4/Ba.5 are the most antigenically distant, yet it likely did not gain dominance first because of the shared L452R mutation with Delta.

It’s a bit smaller, but still similar to the Mykytyn, et. al. map used above. The researchers provide this remark in regards to their map:

This shows, as expected, that the Omicron sub-lineages are clustered together but well separated from early pandemic virus and earlier VoC. Amongst the Omicron cluster BA.4/5 is the most distant from the pre-Omicron viruses.

Now, that last part is interesting- if Ba.4/Ba.5 were the most distant based on the above map, and all of the Omicron subvariants and sublineages were circulating together, why didn’t Ba.4/Ba.5 come to dominate first?

I’ve been trying to rationalize this part of the hypothesis, and my thinking points me to the L452R⁸ mutation seen in both Delta and in Ba.4/Ba.5.

As I mentioned in my prior post on Ba.4/Ba.5, the L452R mutation is considered to be one of the biggest drivers for infectivity and fusogenicity for Delta, and there are a few studies that suggest this to be the case⁹ ¹⁰:

From Motozono, et. al. The infograph is intended to show the benefits the L452R mutation confers onto the virus.

Considering Ba.4/Ba.5 carry this mutation, it also explains why this current wave may see greater illness- because it slightly mimics the Delta variant.

And it’s because of this shared mutation that Ba.4/Ba.5 likely did not spread first.

In a Delta immunity landscape, we may infer that many within the population may have antibodies that target a Delta-specific spike, including epitopes that share the L452R mutation seen in Delta.

That means that Ba.4/Ba.5, which also carry the L452R mutation, will also be targeted by these same antibodies through cross-reactive immunity, and by targeting this mutation the virus was likely to be heavily inhibited.

In essence, it’s the shared mutation between Delta and Ba.4/Ba.5 that caused Ba.4/Ba.5 to not be favored in a Delta landscape.

That leaves us with Ba.1 or Ba.2, and as it appears based on the above antigenic maps we may surmise that Ba.1 was more antigenically different than Ba.2 causing it to win out in the end as the first Omicron subvariant to emerge.

The Immunity Landscape Pipeline

This is all a working model, and it’s one based on very loose data with many liberties taken to form this hypothesis. However, I hope it provides some other alternative as to the current situation with the Ba.4/Ba.5 wave. It’s important that we recognize that Omicron’s emergence was not one that was predicted, and one that needs a proper explanation if we are to fit it into our current models of evolution.

This includes the idea that all sub-lineages of Omicron are likely to have been circulating together, with each one waiting for a niche to open up within the immunity landscape.

In a world in which a majority of the population’s immunity have been shaped by Delta or Wuhan, we can expect that a strain carrying several dozen mutations are likely to escape prior immunity.

That leaves us with the Omicron family of viruses, but even within all of these variations selective pressure may weed out one sub-lineage from gaining dominance over another.

In summary, I posit the following series of events leading up to the Ba.4/Ba.5 wave:

1. The immunity landscape of the world is dominated by Delta and Wuhan immunity, either through natural infection or vaccination.

2. Omicron emerges with the ability to escape prior immunity due to the high level of mutations.

3. Omicron is not one variant- Ba.1, Ba.2, Ba.3, Ba.4, and Ba.5 are all likely to have emerged around the same time and began circulating around the globe together.

4. Out of all of the possible sub-lineages, Ba.4 and Ba.5 are the most antigenically different, and should have been favored to gain dominance first. However, prior immunity from Delta against the L452R mutation may have conferred cross immunity that targeted this sub-lineage¹¹.

   1. This mutation is considered to be responsible for the virus’ targeting of the lungs, fusogenicity, and greater binding to ACEII receptors, and thus targeting this mutation is likely to cripple the virus and prevent it from spreading. The immunity landscape shaped by Delta is likely to be inhospitable to Ba.4/Ba.5 because of the shared mutation, and thus selective pressure weeded it out.

5. Ba.1 is the next antigenically distant Omicron sub-lineage, and it gains dominance to replace Delta. Ba.1 is shown to have a high degree of immune escape (think of monoclonals such as Bamlanivimab, Etesevimab, and Regeneron [Casirivimab and Imdevimab] who stopped working).

6. In the new Ba.1 immunity landscape, Ba.2 is quick to gain dominance due to mutations such as the S371F, T376A, D405N, and R408S mutations that may have removed many of the cross-reactive antibodies at play from prior immunity (think Sotrovimab, which held up against Ba.1 but not Ba.2 and targeted a conserved epitope of the spike protein).

7. Lastly, as the immunity landscape quickly shifts towards Ba.2 the sub-lineages Ba.4/Ba.5 can take hold, with enough distance allowing the L452R mutation to become effective once again, leading to greater lung infection, fusogenicity, and a possibility of more severe illness (compared to Ba.1 and Ba.2).

   1. Also, as Brian Mowrey noted via this study by Kurhade, et. al.¹² The F486V mutation may actually lead to loss of neutralizing antibodies. So the mutation may not be so innocent as I originally remarked, and likely removed whatever immunity was likely to be left.

And so the story goes- at least based on this assumption. Many questions would be left unanswered by this hypothesis, such as why immunity from pre-Omicron variants may not have carried over to a substantial degree. Or why Ba.1 or Ba.2 might not provide robust immunity against Ba.4/Ba.5.

Partially, I believe that it’s the unique combination of mutations in these Omicron sub-lineages that are at fault, as well as the timing. Remember that it took us over a year and a half to get to Delta from Wuhan, yet we went from Ba.1 → Ba.4/Ba.5 in under half a year. All of the variants were already available, and as soon as a niche opened up the next Omicron came to fill it in. It’s likely this jump from each Omicron sub-lineage that is also playing a factor in the immune response we are seeing, and hopefully more evidence can come to light in the future.

But that’s my preliminary working model as I have it. Please let me know what you all think. Hopefully if I dive into this topic further I may be able to substantiate/rebut it.

---

If you enjoyed this post and other works please consider supporting me through a paid Substack subscription or through my Ko-fi. Any bit helps, and it encourages independent creators and journalists outside the mainstream.

Click on the image above or visit the website: https://ko-fi.com/moderndiscontent

1. Omicron BA.1 and BA.2 are antigenically distinct SARS-CoV-2 variants

Anna Z. Mykytyn, Melanie Rissmann, Adinda Kok, Miruna E. Rosu, Debby Schipper, Tim I. Breugem, Petra B. van den Doel, Felicity Chandler, Theo Bestebroer, Maurice de Wit, Martin E. van Royen, Richard Molenkamp, Bas B. Oude Munnink, Rory D. de Vries, Corine GeurtsvanKessel, Derek J. Smith, Marion P. G. Koopmans, Barry Rockx, Mart M. Lamers, Ron Fouchier, Bart L. Haagmans

bioRxiv 2022.02.23.481644; doi: https://doi.org/10.1101/2022.02.23.481644

2. https://www.nicd.ac.za/omicron-lineages-ba-4-and-ba-5-faq/

3. If you’re wondering why I have not included Ba.3, it’s because Ba.3 (for the time being) appears to be a nothing burger at the moment. Ba.3 appears to be one of the siblings to have emerged from a Ba.0 progenitor along with Ba.1 and Ba.2, but unlike the other siblings Ba.3 doesn’t have its own unique mutations, but shares mutations between Ba.1 and Ba.2.

So it could be very likely that infections with either Ba.1 and Ba.2 could be coinfections with Ba.3 that went undetected due to shared mutations, or maybe Ba.3 couldn’t find an antigenic niche in a landscape that was already accustomed to either Ba.1 or Ba.2 mutations.

A Venn diagram showing the overlap in mutations between Ba.1, Ba.2, and Ba.3. Note that Ba.3 has no unique mutations of its own, but shares mutations between the other siblings, and it may be this lack of unique mutations that led to the other siblings overshadowing Ba.3.

Desingu, P.A., Nagarajan, K. and Dhama, K. (2022), Emergence of Omicron third lineage BA.3 and its importance. J Med Virol, 94: 1808-1810. https://doi.org/10.1002/jmv.27601

4. I’m not sure if this term is actually codified in the scientific literature, but I am using it rather loosely to describe the general type of immunity seen in a given population.

5. Because they rely on antibody binding, antigenic maps can be heavily biased by mutations that lead to high levels of antibody loss, such that one mutation that leads to a high degree of immune escape may show a greater antigenic distance than a variant that has 3 mutations but not the same loss of antibodies.

6. Although many people are likely to have immunity to the other variants, this comment is in regards to the circulating variant at the time, and also takes into account the massive vaccination against the Wuhan strain of the virus.

7. Further antibody escape by Omicron BA.4 and BA.5 from vaccine and BA.1 serum

Aekkachai Tuekprakhon, Jiandong Huo, Rungtiwa Nutalai, Aiste Dijokaite-Guraliuc, Daming Zhou, Helen M. Ginn, Muneeswaran Selvaraj, Chang Liu, Alexander J. Mentzer, Piyada Supasa, Helen M.E. Duyvesteyn, Raksha Das, Donal Skelly, Thomas G. Ritter, Ali Amini, Sagida Bibi, Sandra Adele, Sile Ann Johnson, Bede Constantinides, Hermione Webster, Nigel Temperton, Paul Klenerman, Eleanor Barnes, Susanna J. Dunachie, Derrick Crook, Andrew J Pollard, Teresa Lambe, Philip Goulder, OPTIC consortium, ISARIC4C consortium, Elizabeth E. Fry, Juthathip Mongkolsapaya, Jingshan Ren, David I. Stuart, Gavin R Screaton

bioRxiv 2022.05.21.492554; doi: https://doi.org/10.1101/2022.05.21.492554

8. Although this is one mutation, remember that all mutations may be critical, but some mutations may be more critical than others. Virus Farm may be an interesting book to construct in regards to viruses.

9. Motozono C, Toyoda M, Zahradnik J, Saito A, Nasser H, Tan TS, Ngare I, Kimura I, Uriu K, Kosugi Y, Yue Y, Shimizu R, Ito J, Torii S, Yonekawa A, Shimono N, Nagasaki Y, Minami R, Toya T, Sekiya N, Fukuhara T, Matsuura Y, Schreiber G; Genotype to Phenotype Japan (G2P-Japan) Consortium, Ikeda T, Nakagawa S, Ueno T, Sato K. SARS-CoV-2 spike L452R variant evades cellular immunity and increases infectivity. Cell Host Microbe. 2021 Jul 14;29(7):1124-1136.e11. doi: 10.1016/j.chom.2021.06.006. Epub 2021 Jun 15. PMID: 34171266; PMCID: PMC8205251.

10. Zhang Y, Zhang T, Fang Y, Liu J, Ye Q, Ding L. SARS-CoV-2 spike L452R mutation increases Omicron variant fusogenicity and infectivity as well as host glycolysis. Signal Transduct Target Ther. 2022 Mar 9;7(1):76. doi: 10.1038/s41392-022-00941-z. PMID: 35264568; PMCID: PMC8905570.

11. I didn’t quite form an analogy here, but I take the L452R mutation to be akin to a comedian with a huge punchline. Comedian Delta comes around with a joke- the L452R joke- and it becomes a big hit. Everyone laughs at the joke (at least everyone who gets it), and it spreads all around the world. Then comedian duo Ba.4/Ba.5 come onto the comedy scene, but they both have that same L452R punchline. Well, they can’t make that joke- everyone who knows of the joke will just laugh Ba.4/Ba.5 offstage. Which comedian would want to tell the same joke right after another comedian has made it the biggest hit globally? Like Amy Schumer and a Netflix special, you may have to give it time before you reuse a joke. And so Ba.4/Ba.5 disappear into the local comedy circuit while Ba.1 makes it big. However, after a few rounds of Ba.1 and Ba.2, it’s now time for Ba.4/Ba.5 to come back. At this point, maybe enough people have forgotten the joke, or maybe it’s changed enough that those who heard it the first time won’t recognize that it’s just a rehashing. And so off it goes, with the joke landing once again and spreading globally for another time.

Brian Mowrey has given a similar comment likening all of these sub-lineages to movies and re-release of director’s cuts, so apologies to Brian if my analogy is too close to your own!

12. Kurhade, C., Zou, J., Xia, H., Liu, M., Yang, Q., Cutler, M., Cooper, D., Muik, A., Sahin, U., Jansen, K. U., Ren, P., Xie, X., Swanson, K. A., & Shi, P. Y. (2022). Neutralization of Omicron sublineages and Deltacron SARS-CoV-2 by 3 doses of BNT162b2 vaccine or BA.1 infection. Emerging microbes & infections, 1–18. Advance online publication. https://doi.org/10.1080/22221751.2022.2099305