Thread by @DrEricDing

📍Mutation “feat never seen before”...

UPDATE on UK 🇬🇧 strain of #SARSCoV2 & how it evolved so fast:

Variant B.1.1.7 (aka VUI-202012-01) has 23 mutations—but acquired suddenly 17 mutations “all at once”—which experts say is *a feat never seen before*.🧵
sciencemag.org/news/2020/12/m…

2) First, today’s 🧵 further elaborates on yesterday’s 31-post 🧵.

Notably, we previously discussed the possibility of antibody therapy (convalescent plasma and monoclonal antibody drugs) in maybe exacerbating the mutation in someone chronically infected. (Post #18 and #23+)

https://twitter.com/drericding/status/1340487062903001090

3) But let’s first talk about the mutations.

One reason to be concerned is that among the 17 are eight mutations in the gene that encodes the spike protein on the viral surface, two of which are particularly worrisome...

4) “One [mutation], called N501Y, has previously been shown to increase how tightly the protein binds to the ACE2 receptor, its entry point into human cells.”

➡️this one is also found in other troubling South Africa 🇿🇦 strain... that is also worrisome.

https://twitter.com/drericding/status/1340517608643317761?s=21

5) “The other, named 69-70del, leads to the loss of two amino acids in the spike protein and has been found in viruses that **eluded the immune response** in some immunocompromised patients.”

This is a bad deletion mutation.

6) But we also got lucky w/ 69-70 deletion—fortunate fate would have it—One of the PCR tests used widely in UK, called TaqPath, normally detects pieces of 3 genes. But viruses w/ 69-70del lead to a negative signal for the gene encoding the spike; instead only 2 genes show up”!!!

7) “That means PCR tests, which the U.K. conducts by the hundreds of thousands daily and which are far quicker and cheaper than sequencing the entire virus, can help keep track of B.1.1.7.”

Basically UK’s Taqpath PCR can find the new variant via a shortcut! (w/out slow 🧬 lab)

8) This is partly why we know that around 20 September, variant accounted for about 26% of cases in mid-November. “By the week commencing the 9th of December, London, over 60% of all the cases were the new variant.”

9) 70% faster? Maybe, maybe not.

Boris Johnson added that the slew of mutations may have increased the virus’s transmissibility by 70%.
Christian Drosten, a virologist in Berlin, says that was premature. “There are too many unknowns to say something like that,”

10) Because it shares N501Y with S African variant, there is worry is that UK B.1.1.7 could cause more severe disease, which anecdotal evidence support in young people and otherwise healthy, says director of Africa CDC. “It’s concerning, but we really need more data to be sure.”

11) The African Task Force for Coronavirus will convene an emergency meeting to discuss the issue on Monday.

We need to be more careful though. With another strain, scientists initially thought it had 50% higher mortality, but that turned out to be “purely messy, biased data”

12) “I think that is a very strong reminder that we always have to be really careful with early data.” Says @firefoxx66.

“In the case of N501Y (mutation shared by 🇬🇧 and 🇿🇦 strains), more young people may be getting sick because many more are getting infected.”

13) Back to that 69-70 deletion... “The 69-70del mutation appeared together with another mutation named D796H in the virus of a patient who was infected for several months and was given convalescent plasma to treat the disease. (The patient eventually died.)”

14) “In the lab, Gupta’s group found that virus carrying the two mutations was less susceptible to convalescent plasma from several donors than the wildtype (common strain) virus. That suggests **it can evade antibodies** targeting the wildtype virus”!! 👀

15) “Gupta also engineered a lentivirus to express mutated versions of the spike protein and found that the (69-70) deletion alone **made that virus twice as infectious**.” 👀

16) “scientists say B.1.1.7 may already be much more widespread. Dutch researchers have found it in a sample from one patient taken in early December, Dutch health minister Hugo de Jonge wrote in a letter to Parliament today”. They will try to trace more.

17) “Other countries may well have the variant as well—the United Kingdom may just have picked it up first because that country has the most sophisticated SARS-CoV-2 genomic monitoring in the world”. **Many countries have little or no sequencing**

18) “With vaccines being rolled out, the selective pressure on the virus is going to change, meaning variants that help the virus thrive could be selected for, says Kristian Andersen @K_G_Andersen. The important thing in the coming months will be picking up such events.”

19) “Whatever enabled the B.1.1.7 lineage to emerge is likely going on in other parts of the world”, Anderson says. "Will we be able to actually detect it and then follow up on it? That, to me is one of the critical things.”

20) “Researchers have watched SARS-CoV-2 evolve in real time more closely than any other virus in history. So far, it has accumulated mutations at a rate of about 1 to 2 changes per month...”

21) “That means many of the genomes sequenced today differ at around 20 points from the earliest genomes sequenced in China in Jan. “Because we have very dense surveillance of genomes, you can almost see every step,” Loman says.”

22) “But scientists **have never seen the virus acquire more than a dozen mutations seemingly at once**...

23) “They think it happened during a **long infection of a single patient that allowed SARS-CoV-2 to go through an extended period of fast evolution**, with multiple variants competing for advantage.”

24) Also *where* on the spike protein of the virus matters. Not all parts are equal, some more easily accessible—Receptor binding motif/domains (RBDs) that latch human ACE2 are among most accessible to antibodies.

so now guess where the N501Y mutation is? In the RBD! Not good.

25) more specifically, the RECEPTOR BINDING DOMAIN (ie the “docking sites” on the spike to human receptor)... has 4 sites—4 classes. And again, you can see for yourself in 3D image where the damn N501Y mutation is—right *inside one of the 4 docking sites*! cogconsortium.uk/wp-content/upl…

26) alternate top 3D view of spike. The 501 spot on receptor binding domains are localised in the regions targeted by some types **monoclonal antibody (MAB) drugs**. And yellow = class 2: it is ACE2 blocking, & can bind open and closed RBD.

Hence, possibly functional mutation.

27) Also what about the 69-79 deletion mutation?

📌 69-70 is **near a region targeted by other mAbs** and deletion might alter the structure of neighbouring amino acids

➡️ (translation: that could affect the efficacy of antibodies / MAB drug targeting).

28) this is likely why the Gupta study above found the 69-70del maybe involved in eluding immune response.

These antibody binding based findings may or may not affect vaccine-based antibody response too. We don’t know yet.

But antibody MAB can also drive evolution of virus...

29) This can also maybe relate to convalescent plasma.. “Convalescent plasma is often given when patient viral loads are high, and Kemp et al. (2020) report that intra-patient **virus genetic diversity increased after plasma treatment was given**.” 👀 medrxiv.org/content/10.110…

30) The UK authors also point out that “the selection (for new UK variant) arising from antibody therapy may be strong due to high antibody concentrations...”

https://twitter.com/drericding/status/1340545731724255232?s=21

31) “if antibody therapy is administered after many weeks of chronic infection, the virus population may be unusually large and genetically diverse at the time that antibody-mediated selective pressure is applied...” virological.org/t/preliminary-…

32) ...”creating suitable circumstances for the rapid fixation of multiple virus genetic changes through direct selection and genetic hitchhiking.
These considerations lead us to hypothesise that the unusual genetic divergence of lineage B.1.1.7...”

33) “may have resulted, at least in part, from virus evolution with a chronically-infected individual. Although such infections are rare, and onward transmission from them presumably even rarer, they are not improbable given the ongoing large number of new infections.”

34) So now you know some of WHAT mutations there are, WHY they may be functional, and HOW they maybe evolved suddenly in this 23 mutation feat.

Now we just need to wait on precise studies on transmission, vaccine response, and how they affect CP/MAB therapies. We need to wait.

35) BOTTOMLINE: we need to stop this damn #COVID19 pandemic. We can’t let this damn virus linger in the wild any further. Vaccines do work and will likely work on this strain (likely still uncommon worldwide). Sooner we eradicate, sooner we can stop fretting these damn mutations!

36) Why do I write these long threads🧵 on the coronavirus?

Besides being an epidemiologist, it’s because I have been wracked with guilt since January. Background on what happened....

nymag.com/intelligencer/…

37) updated thread with more info.

https://twitter.com/drericding/status/1341231738249486337?s=21