1: Some thoughts on gain of function (GoF) and virology research in general because I do feel some of it is being misrepresented, and as virologists we are all being tarred with a suspicion which is unfair.

2: firstly, viruses make people sick.
We work on them to understand how they do this, how to block them with drugs and vaccines, and in some cases as tools to understand fundamental biology or even as therapeutics
3: We have containment levels 1-4, the point of which is to reduce the risk of working on a given virus to levels where they mitigated. These are mandated by law and working with a live virus below its required containment is unacceptable.
4: GoF for the purposes here is the adaptation, mutation or augmentation of a virus to increase its growth, transmission or pathogenic potential. This can come in many forms.

5: Selecting HIV-1 in the lab to be resistant to a new drug and identify how resistance occurs is GoF.
We do it all the time because it’s necessary to know how this might happen in vivo. And we contain the virus at BSL2+ - BSL3 rules where the chance of a lab escape is negligible.

6: However, there are examples of where it is unjustified and downright dangerous
7: A classic example was this paper, where the authors added the gene for the immunoregulatory cytokine IL-4 into the genome of Myxoma Virus that causes myxomatosis in rabbits.
8: The resulting virus had enhanced pathogenesis, but crucially killed rabbits that were genetically resistant to regular myxomatosis. Poor Rabbits, right? But imagine if someone did the same thing to another virus of the same family – monkeypox.
Well then you have Smallpox 2.0

9: The other famous case is of course avian H5N1 influenza. 2 groups in 2012 combined natural mutations found in human infections and found that together they allowed efficient respiratory spread between ferrets (and by extension people).
10: They created a strain of highly pathogenic H5N1 Flu that if it had escaped could have caused a lethal pandemic

11: They did it from the best of intentions – to understand what we should be looking for in wild. And these guys weren’t cowboys.
But even the very producing of this virus brought a risk that the most thought unacceptable even if contained at BSL-4.

12: It didn’t need to be made to answer the question, and therein lies the point.
Some GoF studies are potentially acceptable if contained properly, others pose unacceptable risks irrespective. And as such all of us would agree these things need stronger international oversight.
13: Of course, the vast majority of virologists know exactly why GOF experimentation on a virus to enhance its pathogenicity and transmissibility in humans is wrong. And far better, I would argue, than many of the self-appointed armchair critics.
14: In fact we have many molecular tricks to side-step having to do GoF experimentation with live viruses by using surrogate systems that are usually easier and need far less containment. This is particularly true of viral glycoproteins like SARS CoV-2 S or Ebola GP.
15: All enveloped mammalian viruses need at least one glycoprotein embedded within the membrane to allow the virus to interact with a target cell via a receptor and then mediate fusion. In the case of S this is ACE-2 and associated factors.
16: In principle you can coat a different virus or replication defective viral vector with a heterologous glycoprotein of choice and use that to understand viral entry and cell and/or species tropism – this is called “pseudotyping”
17: The most common workhorse of such assays are vectors based on retroviruses like HIV-1.
Without going into their biology too much, on can delete all the protein coding genes in the HIV genome and replace them with a marker gene that turns infected cells green or makes them glow in the dark.
18: You then introduce this genome in a DNA form into 293T cells as well as a separate DNA encoding the proteins you’ve deleted that make up the core and enzymes of the virus. You supply the glycoprotein of choice as a 3rd DNA fragment.
19: 48h you take the supernatant and you have replication-defective HIV-1 particles coated with your glycoprotein that can infect target cells, turn their RNA genome in to DNA, integrate it in the cell’s DNA and express your marker to measure the amount of cells infected.
20: Totally safe and no chance of something that can recombine.

21: With such systems you can ask: what receptor does the virus use? Can it use the orthologous receptors from other species? What bits of the protein do what?
And of course, measure antibody responses in infected people and vaccinees – this is how the majority of S2 work is being done
22: The problems with retroviral pseudotypes are (1) not all glycoproteins work efficiently in HIV systems, and (2) because the virus doesn’t replicate you can’t do selection or in vivo experiments easily..
23: (3) you are limited to asking only about the glycoprotein, not the rest of the virus in question. And (4) none of this is applicable to non-enveloped viruses anyway.

24: A way round (2) is to use a fairly benign virus that you can pseudotype in a replication competent form.
One example of this is Vesicular Stomatitis Virus (or VSV), a relative of rabies virus, that can infect nearly any mammalian cell type. It’s a disease of livestock but in humans is very mild.
25: One can clone SARS CoV-2 S into the VSV genome in place of its own glycoprotein, and now you have a VSV than can only infect ACE2 +ve cells.
Where this system comes into its own is as a potential vaccine (an ebola virus vaccine made like this is a key frontline response)…

26....and also as a tool to understand how a virus might escape antibodies.
Because it replicates to very high titer with an error-prone polymerase you can rapidly select antibody escape variants in a few days. This is what @PaulBieniasz and @theodora_nyc did here:
27: Believe it or not, none of us want to be adding to the problem of variants of CoV-2. This group used the VSV(SARS2) system to select escape variants of spike from antibodies to predict what might arise in real life.
In mid 2020 they predicted many of the mutations we now see in the VoCs.

28: But you still can’t do experiments to understand pathogenesis of the nasty viruses with such systems. To do that you need the real thing, or something close to it.
What you really want is a molecular clone of a virus in which you can make targeted mutations and see what they do in cells or an animal model

29: The overwhelming majority of such experiments involve inactivating or dysregulating a given viral gene.
This is what is called Loss of Function and is reasonably predicted to either do nothing or weaken the virus by reducing its replication or abolishing its antagonism of host immunity.

30: LoF needs the same containment level as the real thing BUT is not the problem here
Then there is GoF research of concern (GOFROC) that you might want to do on a virus to answer a question. And this is where we get into contentious waters.
31: And I will say up front for GOFROC to be permitted it needs to be (1) essential to understanding a pressing need, (2) to be not doable in a heterologous system and (3) PERFORMED AT THE APPROPRIATE BIOLOGICAL CONTAINMENT – so the risk vs need is justified
32: For example, doing something to see if it is possible (for example selecting an airborne EBOV for arguments sake) is not justifiable even if one has a BSL4 lab.

33: And this is the grey area that the Bat CoV research sits in.
To understand the pandemic potential of a bat CoV you need to understand the biology of spike – all can be done in a heterologous systems described above, so not GOFROC.
34: However, you also might want to know if a Bat CoV spike renders a virus potentially transmissible in vivo, especially in the presence of Abs. And you don’t want to isolate all these different viruses (its not practical anyway despite what you may hear).
35: So one of the aims of the EcoHealth grant was to clone the spikes of bat CoVs into an already constructed bat CoV relative of SARS1 called WIV. This virus already infects using human ACE2 and causes disease in ACE2 Tg mice
36: So you are asking whether the new bat CoV spike REPLACES a function that WIV1 already has.

37: Now initially when this was done in vitro, these recombinant viruses had no growth advantage over the parental WIV1 – i.e. they had not gained function.
But it remains possible that the Spike might change the in vivo pathogenesis in Tg mice

38: The progress reports for the grant suggest this may have happened with one of the chimeric viruses.
But as I said in a different thread, the data presented makes me dubious this is really true. However, it has Gained Function until shown otherwise.

39: Could this have been predicted – yes it should have been, and therefore flagged as potentially GoF.
But knowing if a spike has more pathogenic potential above getting the virus into a cell is potentially v important.

40: particularly if you want to develop a vaccine that will protect against a whole bunch of these viruses – just as we are seeing in the CoV2 VOCs.
41: Is the research justified? For me, yes BUT before anyone quotes me out of context, none of it should ever have been done a BSL-2, but in a BSL-3 specifically equipped for respiratory pathogens
42: Lastly though, none of the studies documented in the progress report or proposed in the grant can be themselves be the source of SARS CoV-2. They would have had to isolate proximal ancestor first. And if that was there, the game changes.

• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with Stuart Neil

Stuart Neil Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @stuartjdneil

Apr 4
1: Read the article about SARS2 origins. Most of it is just the usual rehash of EHA/DARPA/WIV innuendo that we all know. Nothing new in that. Legitimate questions about GoF and safety interspersed with conjecture trying to link PREDICT to SARS2. vanityfair.com/news/2022/03/t…
2: as has been stated many times, the contentious chimeric virus research KNOWN to have been performed by WIV/EHA would not result in SARS-CoV-2. See my thoughts on the merits of GoF and its risks – they haven’t changed lest someone accuse me of apologia.
3: The rest boils down to EHA lobbying for funding for what anyone arguing in good faith would be laudable aims – predicting the risk of zoonotic overspill of bat viruses into humans. Amply illustrated: paperhttps://www.sciencedirect.com/science/article/pii/S0092867422001945
Read 29 tweets
Mar 5
Yuri here (despite his personal incredulity) raises the most important question.

IF the animals, which we now know were there in late Nov/early Dec, are the source of SARS2 in Wuhan, where did they all go by the time the market closed Jan 1 2020?
And he is absolutely correct, they would have had to pick it up outside Wuhan, somewhere in the supply chain or the farm/wildlife holding facility they came from. And don't forget that the people who brought hem there could well have been exposed too - just like sars1
We know that civets from Hubei were sold in Forshan in 2002/2003, and that civets on those farms were found to be sarbecovirus infected - presumably because the farms are located near the Karst formations where vast colonies of bast live.
Read 5 tweets
Feb 27
1: Yesterday 3 important preprints dropped. 2 make a powerful case that the origin of SARS CoV-2 in Wuhan resulted from at least 2 (possibly more) zoonotic spillovers associated with live animal trading in the Huanan market.
2: The other, while attempting to come to the opposite conclusion, inadvertently makes the case for the other two stronger (more later).
3. I won’t review the papers in detail. Read the authors @MichaelWorobey,
@K_G_Andersen, @arambaut threads on the manuscripts. Everything is fleshed out there. But there are a couple of things i’d draw attention to.They are all preprints, they have yet to be peer reviewed
Read 32 tweets
Feb 17
@PrometheusCHT @InWuchang @gdemaneuf @R_H_Ebright @AndrewButchart1 @Rossana38510044 @macroliter @Louis_San @thackerpd @MicrobiomDigest @marcendeweld @ugauthier @VirginieVilar @zeynep @arambaut @robertson_lab @SpyrosLytras So let me get this straight.

The ‘parsimonious’ explanation you are proposing is that:

1: using RaTG13 and a large set of hypothetical viruses (that must be by definition more closely related to SC2), virus jockeys at WIV designed a consensus virus genome in secret
@PrometheusCHT @InWuchang @gdemaneuf @R_H_Ebright @AndrewButchart1 @Rossana38510044 @macroliter @Louis_San @thackerpd @MicrobiomDigest @marcendeweld @ugauthier @VirginieVilar @zeynep @arambaut @robertson_lab @SpyrosLytras 2: they then clone into it the RBD of a virus from Laos that they had already characterized as a super ACE2 binder, which just so happens to the same as others sampled at a different time
@PrometheusCHT @InWuchang @gdemaneuf @R_H_Ebright @AndrewButchart1 @Rossana38510044 @macroliter @Louis_San @thackerpd @MicrobiomDigest @marcendeweld @ugauthier @VirginieVilar @zeynep @arambaut @robertson_lab @SpyrosLytras 3. The consensus virus then looks entirely natural in a phylogenetic tree when compared to viruses found AFTER the start of the pandemic, sharing an MRCA 40yrs + ago with RaTG13
Read 7 tweets
Dec 16, 2021
1: Interesting preprint, marred by some disappointing commentary from some who should know better clapped on by the usual phocine suspects. Luckily, we were spared others chiming in with their “expertise”. A shame because I looked forward to the insight.
2: What most people are not that aware of is that SARS-like CoVs are not just found in horseshoe bats in China and SE Asia, but all over centra asia, Africa and Europe. These are more distantly related to SARS1 and SARS2.
3: But when you look at the S1/S2 boundary of the spikes these viruses, you find sequences that are only one mutation away from a polybasic (furin-like) cleavage site.
Read 18 tweets
Dec 4, 2021
1.There is unsurprisingly a lot of speculation about where omicron has come from. There are several hypotheses – personally I lean towards evolution as an immune escape variant in an immunocompromised host.
2.Over the pandemic we have now found many examples of persistent, long-term infection of people who have compromised immune systems.
3.That might be that because of leukemia, inherited genetic disorders, or treatment with immunosuppressive agents (i.e. in cancer or transplant patients) or because of HIV infection.
Read 27 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!


0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy


3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!