Slack / Private Message Drop — page 339

[2020-12-22 09:21:55] [Robert Garry] The pattern of prolines in SC1 and SC2 RBDs are distinct - look at PO1 - which creates a very different interaction with ACE2 - but we're seeing in real time the adaptation of SC2 to receptors in humans, minks and mice pretty much as you'd draw it up. The twist that it may be immune selection driving better receptor binding is not expected at least by me. Tuning up receptor binding affinity happened in SARS1 per the JVI paper above, but more subtle and [probably?] not an immune component there in 2002 and 2005. [2020-12-22 09:38:33] [Kristian Andersen] Interesting difference between SC1 and SC2 is that it was pretty darn quick in the former - essentially happened right after the jump from intermediate hosts. Slower for SC2, but it's probably because it was pretty darn good at binding already. [2020-12-22 10:33:47] [Kristian Andersen] [thread - ID: 2020-12-22 10:33:47] Yeah, so this "We present evidence that the Spike H69/V70 is a fitness enhancing change that may be stabilising other S gene mutations" from Gupta's recent preprint is what worries me the most and I'm wondering if E484K (or N417K) could play a similar role in the SA lineage that H69/V70 is playing in the UK lineage. Add Nelly on top of that, and you might end up with what we're observing now. [2020-12-22 10:33:57] [Kristian Andersen] https://www.biorxiv.org/content/10.1101/2020.12.14.422555v3.full.pdf [2020-12-22 12:32:34] [Robert Garry] Compiling some of the RBD mutations in the mice, minks and humans and comparing to the "contact" residues in ACE2 of the three species. I need to double-check that I have all the mutations -there is also some disagreement or maybe it's just confusion I need to resolve on my part on the contact residues. For example: N501Y/T could interact with Y41 or K/H353 or both. Does not seem likely that the species adaptations falling onto the analogous residues is by chance, particularly when you superimpose the SARS1 civet to human story detailed in the JVI paper above. Bottom line is most but not all of the mutations are in residues that contact ACE2 and all are likely in epitopes, most but maybe not all neutralizing. [2020-12-22 12:39:43] [Robert Garry] [thread - ID: 2020-12-22 10:33:47] @Kristian Andersen I think that the P681H mutation potentially could also be "fitness enhancing" for the reasons discussed above. I'm thinking that E417N, E484K and perhaps S494P (but proline) are probably pure immune escape. [2020-12-22 12:43:28] [Kristian Andersen] [thread - ID: 2020-12-22 10:33:47] @Kristian Andersen But how the heck does the SA lineage get its 'transmission' phenotype then? If it has it, that is... It's N501Y, E484K, N417K [2020-12-22 14:12:15] [Robert Garry] [thread - ID: 2020-12-22 10:33:47] @Kristian Andersen D614G and N501Y may be enough? [2020-12-22 14:14:12] [Robert Garry]