Chat message

Slack / Private Message Drop, p.23 [SLACK_000237] · slack_pm:msg:00234

Date: 2020-05-12 21:28
Type: chat message · slack
recipient: Robert F. Garry, Edward C. Holmes, Andrew Rambaut
speaker: Kristian G. Andersen
Topics: Furin cleavage site and molecular features

Recipients on this medium are inferred from channel membership, not per-message addressing.

The three main new things that bothers me - the first two shown in this post http://virological.org/t/identification-of-a-common-deletion-in-the-spike-protein-of-sars-cov-2/451: 1. The furin site is clearly active in tissue culture. We see the virus lose it under some culturing conditions, but this effect could be very cell-type specific. What bothers me here is that the virus is clearly fiddling with this site inculture. If it can lose it, likely it can gain it? Typically SARS-like CoVs are cultured in the presence of trypsin (to allow for cleavage of S1/S2), but if trypsin isn't added, there's a *huge* selection pressure for gaining a furin site. SARS-CoV-2 can grow without trypsin - other SARS-like CoVs can't.2. The example from bovine CoVs (post #6 on the thread on virological) is exactly a 12bp insertion seen in tissue culture creating a furin cleavage site.3. We have now seen multiple examples of SARS-CoV-2 de novo obtaining exactly 12bp insertions in different partsof the spike, but only in tissue culture. They're not furin sites, but there seems to be some affinity specifically for12bp insertions (in T/C) - and of course, the furin cleavage site is exactly 12bp. There could be some specific mechanism behind this. We're looking into this a little closer at the moment.

Links shared

preprint http://virological.org/t/identification-of-a-common-deletion-in-the-spike-protein-of-sars-cov-2/451

Extracted statements

Rule-extracted, not adjudicated. The grade says what may be done with each one; read the document above before relying on any of them.

Statement	Grade	Attribution	Stance
The furin site is clearly active in tissue culture.	needs context	speaker_own	asserts
Typically SARS-like CoVs are cultured in the presence of trypsin (to allow for cleavage of S1/S2), but if trypsin isn't added, there's a huge selection pressure for gaining a furin site.	own voice, substantive	speaker_own	asserts
The example from bovine CoVs (post #6 on the thread on virological) is exactly a 12bp insertion seen in tissue culture creating a furin cleavage site.3.	own voice, substantive	speaker_own	asserts
They're not furin sites, but there seems to be some affinity specifically for12bp insertions (in T/C) - and of course, the furin cleavage site is exactly 12bp.	needs context	uncertain explicit hedge	asserts

Statement

Grade

Attribution

Stance

The furin site is clearly active in tissue culture.

needs context

speaker_own

asserts

Typically SARS-like CoVs are cultured in the presence of trypsin (to allow for cleavage of S1/S2), but if trypsin isn't added, there's a *huge* selection pressure for gaining a furin site.

own voice, substantive

speaker_own

asserts

The example from bovine CoVs (post #6 on the thread on virological) is exactly a 12bp insertion seen in tissue culture creating a furin cleavage site.3.

own voice, substantive

speaker_own

asserts

They're not furin sites, but there seems to be some affinity specifically for12bp insertions (in T/C) - and of course, the furin cleavage site is exactly 12bp.

needs context

uncertain explicit hedge

asserts

In context

A single line often inverts meaning once you see what it answers, so neighbouring messages are always shown.

2020-05-12 21:20 Edward C. Holmes open

What do we now think about the o-linked glycans and the low dn/ds values? Can they be used to rule anything in/out?
2020-05-12 21:22 Kristian G. Andersen open

Initially we thought the O-linked glycans could somewhat rule out lab culture (or wouldn't be consistent with), but I'm much less sure now - I'm not convinced they actually form a glycan shield and instead may help regulate the site itself. The low dN/dS value - primarily in the spike - has always been peculiar to me and a tiny bit suggestive ofculturing.
2020-05-12 21:24 Edward C. Holmes open

Thanks. Very important.
2020-05-12 21:25 Edward C. Holmes open

My first look at the virus suggested that the pattern of distribution of sequence variation across the genome looked natural, but that may too broad brush.
2020-05-12 21:28 Kristian G. Andersen

The three main new things that bothers me - the first two shown in this post http://virological.org/t/identification-of-a-common-deletion-in-the-spike-protein-of-sars-cov-2/451: 1. The furin site is clearly active in tissue culture. We see the virus lose it under some culturing conditions, but this effect could be very cell-type specific. What bothers me here is that the virus is clearly fiddling with this site inculture. If it can lose it, likely it can gain it? Typically SARS-like CoVs are cultured in the presence of trypsin (to allow for cleavage of S1/S2), but if trypsin isn't added, there's a *huge* selection pressure for gaining a furin site. SARS-CoV-2 can grow without trypsin - other SARS-like CoVs can't.2. The example from bovine CoVs (post #6 on the thread on virological) is exactly a 12bp insertion seen in tissue culture creating a furin cleavage site.3. We have now seen multiple examples of SARS-CoV-2 de novo obtaining exactly 12bp insertions in different partsof the spike, but only in tissue culture. They're not furin sites, but there seems to be some affinity specifically for12bp insertions (in T/C) - and of course, the furin cleavage site is exactly 12bp. There could be some specific mechanism behind this. We're looking into this a little closer at the moment.
2020-05-12 21:28 Kristian G. Andersen open

> My first look at the virus suggested that the pattern of distribution of sequence variation across the genome looked natural, but that may too broad brushAny new evidence on this? I haven't looked at e.g., dN/dS for a while.
2020-05-12 21:29 Edward C. Holmes open

Got it. No new evidence.

Full conversation →