Transcript of interview with Professor Pamela Bjorkman* by Dr Lara Marks, 21 June 2021

(this transcript has been edited for clarity and brevity)

*David Baltimore Professor of Biology and Bioengineering, Merkin Institute Professor, California Institute of Technology

Lara

I understand that you are developing an all-in-one coronavirus vaccine and I would like to know, first of all, when you started working on the project and what kind of platform you're using to do that.

Pamela

We've been interested in trying to make vaccines against various viruses for a while. My lab really does structural biology, which means that we look at the 3-D structures of the targets of the immune system, which are usually spikes that come out of the virus. So coronaviruses have the famous spikes, and so does HIV and flu.

One of the things we've been trying to do [for a vaccine] is to make a nanoparticle, which is a small, little thing that looks like a miniature soccer ball. And attach pieces of the spike to that using a very easy technology that was developed at Oxford University.

Lara

Who helped develop it at Oxford?

Pamela

Mark Howarth. What he did was he started with a little protein nanoparticle which has several attachment sites to which he fused a protein from a bacterium that's called 'SpyCatcher'. Whatever you want to attach to that nanoparticle, you add what Howarth’s lab calls a ‘SpyTag’, which is only 13 amino acid residues. Now when you mix those two together they bind covalently to each other. In other words, you can just take one of these SpyCatcher nanoparticles, incubate it with whatever SpyTagged protein you want, and it's bound irreversibly, covalently. In our case the nanoparticle has 60 attachment sites.

Lara

When did Mark develop that platform?

Pamela

At least four years ago (Brune et al). But, he's been improving upon it over the years and he's got a company called Spy Biotech. Oxford has licensed the platform. It's called a ‘plug and display’ approach. This means you can plug in any antigen taken from a virus, or whatever you want, to add to it. You just SpyTag it and it binds. It's very, very powerful to be able to do this.

We've been using the platform to make a vaccine for the influenza virus and also for HIV. In those cases, instead of making what we call a homotypic nanoparticle, where all of the SpyTagged spikes are exactly the same, we've been creating what's called a ‘mosaic nanoparticle’ which has lots of different antigens. The advantage with the Howarth technology is that it's just as easy to make mosaic nanoparticles as to make a homotypic particle. Because all you need to do is make eight, or whatever number you choose, different [antigens and link Spytags to them] and then they all glom onto the nanoparticle.

Lara

How did you start collaborating with Mark?

Pamela

We found his papers and I think I got in touch with Mark to ask him some technical questions when we created a particle for an HIV vaccine which we published in 2019 (Escolano et al). Mark was very, very helpful and has been just great. He's not a co-author on our HIV paper because we used published technology. But he is an author on our paper for coronaviruses because we used a different type of nanoparticle that he gave us before it was published (Cohen et al).

Lara

When did you start working on [the Spycatcher platform] for HIV and influenza?

Pamela

The first thing we published was for HIV. We've been working on this for quite a while for which we have a lot of unpublished stuff. HIV is a great example of a virus that has multiple strains that we worry about. There's no such thing as one HIV. If somebody is infected, they carry a viral swarm of variants. Now, it's not that bad for coronavirus because it would be very unlikely that the person would have a bunch of variants. They would probably only carry one variant. But the problem for coronaviruses is that there's so many circulating viruses in animals. The fear is that new ones will emerge zoonotically, that is spill over from animal reservoirs into humans and we'll have no immunity against them.

Lara

When did you start working on the coronavirus?

Pamela

The pandemic became an issue in the US in March of 2020. And the first author of our [coronavirus vaccine] paper, Alex Cohen, was a graduate student at that time. He was making mosaic nanoparticles for the influenza virus using different strains of influenza virus and he just switched over immediately to make a mosaic nanoparticle with just the receptor binding domain (RBD) of the SARS-CoV-2 spike protein. [First] he made homotypic nanoparticles that way. But then he also looked through the literature and found other coronaviruses thought by coronavirus experts to represent threats of spillover into humans and very carefully chose [to target] sarbecoviruses, that means SARS-like beta coronaviruses. Based on this he created a promising pan-sarbecovirus vaccine which [covers] Alpha coronaviruses, Delta coronaviruses, and so on. MERS, they're too distant. So you could do what we've done and make a pan-alpha coronavirus vaccine or maybe make a pan-merbecoronavirus vaccine. But we don't think the cross reactivity for our vaccine candidate would extend to coronaviruses other than sarbecoviruses.

Lara

And the one at the moment that's causing us the problem with new variants, what type is that?

Pamela

SARS-CoV-2 is a type of sarbecovirus. The major concern right now among most people is the SARS-CoV-2 variants. But when we started nobody was concerned about those variants. Our main concern was other sarbecoviruses that constantly feed into the human population from probably bats, maybe rodents. There's evidence that people get infected from these zoonotic animal viruses all the time. Most of the time nothing happens because they don’t pass easily between humans. But if you keep rolling the dice like that, eventually you'll get a SARS-CoV-2 like virus that spreads like crazy among humans. And that's what's happened. We also need to be prepared for the next one, because unfortunately I don't see any reason that there's not going to be another one.

Lara

How quickly do you think you'll have one for the new variants for SARS-CoV-2?

Pamela

It already works against the new variants.

Lara

Why does it work against the variants?

Pamela

The reason is that where SARS-CoV-2 varies is in the same part of the RBD that varies between sarbecoviruses. So far, what we've done, even though we didn't know it at the time, works fine against the variants.

Lara

Have you already tested it out in animals?

Pamela

We're trying very, very hard. Let me back up and say a few things. So in March 2020, Alex started making a variety of RBDs based on the literature, and he made eight to put on the particle. And then he made another four that he wasn't going to put on the particle which were for tests.

So he made both mosaic and homotypic particles. The homotypic have only SARS-CoV-2 RBD, the mosaic have either four or eight different RBDs. And then he injected [those into] mice and they made immune responses. And when he [checked the immune responses he found that] the ones from the mosaic injected animals worked better against the RBDs than the homotypic one (Cohen et al). That implied that if those four [or eight coronaviruses] happened to cause new epidemics this vaccine would work against them. We can't predict what's going to come up as a new epidemic. All we can do is take the ones we know about and ask would our vaccine make immune responses to those and the answer is yes. The homotypic vaccine also does, but to a lesser extent.

Lara

What is the homotypic vaccine - does that mean against one strain?

Pamela

Yes, it means one strain, but in our case it means only SARS-CoV-2.

Lara

Okay.

Pamela

So, in the meantime, Mark Howarth and Alain Townsend [in Oxford], were making a homotypic vaccine, exactly like what we made for our homotypic control. And they published that (Tan et al) and contacted us because we know each other from years ago. We're now working together because Alain wants to get their homotypic particle into human clinical trials and we want to get our mosaic nanoparticle into humans. But the first one that should be trialled should be the homotypic one because manufacturing that is much easier because you only have to make one RBD instead of eight.

Lara

Who's funding this work?

Pamela

It is not easy to get funding for this work. At the beginning of the pandemic I did not have funding to work on coronaviruses. But I did have funding to work on HIV from Wellcome LEAP, which is an organisation linked to the Wellcome Trust. This funding enabled us to do the following study. You inject eight non-human primates with our mosaic nanoparticle and eight you don't inject. And then you challenge a group of the immunised animals with SARS CoV-2 and you challenge another four with SARS-CoV-1, conventional SARS, and then you do that in the control group as well. The point about SARS is that we deliberately did not include the SARS-CoV-1 RBD on our particle, so it could be a control and represent an unknown virus that might spill over into humans.

But I don't have funding to do challenge studies in animals. These are not experiments I can do at Caltech and universities can't do these experiments. What you want to do is to inject non-human primates or small animal models or something with your vaccine, and then challenge them with a virus and then ask are they protected? We didn't have the funding for that.

Lara

How did you get involved initially with all of this work?

Pamela

Well, I've been interested in immune recognition of viral antigens my whole career, the structural immunology really. About 10 or 15 years ago now I started working on HIV. In that case the hope is you could raise broadly neutralising antibodies [to target] all strains instead of just some strains. I figured we could solve the structure of a bunch of HIV spikes [by binding them] with antibodies and figure out how to raise those antibodies. That's what a lot of people have been doing for HIV for a long time. But it's a very difficult problem.

Lara

Okay.

Pamela

Everything we've done for coronavirus has been done in year and a half and a lot of it's working, at least so far. But HIV is just a whole other ball game.

Lara

Do you think what you will learn from [your work on coronaviruses] will help you with HIV?

Pamela

It will. It shows that it can in theory work. The cross reactivity we got was incredible and really encouraging. The great thing was that in our mouse experiments, the immune responses they raised to the mosaic nanoparticle worked just as well against SARS CoV-2 as the immune responses against the homotypic particle. What that would mean is right now you could either get the homotypic vaccine, if Alain can get it out there, or in another year or so maybe you'd have the option for the mosaic vaccine. So if we still have concerns about SARS-CoV-2, both would protect equivalently. But the mosaic vaccine would give you the possibility that if something else spills over, you'd be protected against that.

Lara

Amazing. So it's quite hopeful, although obviously you've still got a lot of clinical trials and all that.

Pamela

Yeah, I'm taking this one step at a time. I don't have any money for clinical trials. It's been difficult because the attitude has been we have the vaccines and we don't need anything [more] because we have things that work. But I can tell you this, it's not practical. The mRNA vaccines are fantastic, but cold chain storage is not going to work for the whole world, not even for parts of South Central Los Angeles near where I live.

The great thing about the particles that the Townsend lab and our lab are making is you can lyophilise them, meaning you can turn them into a powder and mail them at room temperature anywhere you want. And they'd be very, very cheap to make. So we think this is more of a possibility for global distribution. I also think that the more vaccine technologies are out there the better. But that being said, it's a manufacturing issue to try to make the mosaic particle.

Lara

What do you think have been the advances in science that allowed you to get to this point to develop this technology?

Pamela

There was a lot of coronavirus research by virologists in the past. Even before SARS and MERS there were people working on coronaviruses because they cause common colds and then there was a lot of coronavirus research out there on SARS and MERS. So they knew what the spike proteins of coronavirus looked like and that they had RBDs. And they knew how to make the RBDs alone. Then we had the SpyCatcher-SpyTag technology. All of this was in place to work really, really fast.

So there was a lot known about coronavirus biology. But more has been learned in the past year or year and a half than I ever thought would have been possible. It's been very remarkable in that sense. And it's going to really help people. Also I don't want to downplay the mRNA vaccines because getting that technology out there and showing that it works means it is going to be applied to all kinds of different things.

Lara

But the mosaic nanoparticles sound to me like that's another platform and unique.

Pamela

Yes, it could be. There might be ways to make mosaic mRNA vaccines, we're trying to work on that. But we haven't published anything on that yet. I'm sure companies are working on that too.

Lara

Do you think these mosaic nanoparticles are going to revolutionise what we can do in vaccines?

Pamela

It might. The reason we want to put it all on one particle is that we think you'd get better cross-reactive immunity if the different spikes are on the same particle. This is because what we're trying to do is activate [a cross-reactive] response.

When you make antibodies, you first have to stimulate B lymphocytes. The B lymphocytes have antibodies hooked to their surface. And to start the whole B cell response, that will eventually make secreted antibodies that will go around in your blood, you have to first activate the B cell that has the cell surface antibody. The antibody has two arms, it's [shaped] like a ‘Y’. But it seems to us that you'd get better cross-reactivity if everything's on one particle because you get better binding if both [antibody] arms bind.

Now imagine you've got this particle and that all the nearest neighbours are different from each other. In order for the thing to bind with both arms it has to bind to two things that are next to each other that are different. What it's going to do is to find a conserved site on those two things, that it's two identical arms can bind to and it's going to ignore the [other] parts. Because if it binds to the parts that differ between the neighbours on the particle it can only bind with one arm, not two, just one. And that doesn't bind as tightly. The analogy I always give is that if you're hanging on to a bar with two arms, you can hang on longer, because if one arm gets tired you can drop it or drop the other. But if you're hanging on with one arm and your arm gets tired, and you drop it, it's going to fall off.

Lara

And that's what you've managed to do with the mosaic nanoparticle.

Pamela

Now we can't prove that. But that's the theory we're going on. We're trying to actually isolate the antibodies and study them as single antibodies to look at that. That's part of what we're trying to do. But it's going to take a long time to study exactly what these immune responses are. This will be done over the years, I imagine, by many different labs. But we're trying to do this and it's going to take a while and a lot of resources.

Lara

So where do you think you are with the all in one coronavirus vaccine?

Pamela

Well, there's a number of approaches that people are taking where you try to target the intact spike [protein]. You don't just use the RBDs which are at the top [of the spike].The spike happens to have three copies of itself [which is called] a trimer. But the part of the trimer that doesn't vary as much between different coronaviruses, and are more conserved, is at the bottom. But that part is difficult [to target] because its a bit inaccessible to the immune system. If you could invent a way to get at those, and people are trying, maybe you could make a pan-coronavirus. But I'm not so confident [about that approach]. The reason is that all of us have had common cold virus infections and we've made antibodies against those but do not make antibodies that protect us against SARS CoV-2, or SARS or MERS. So everyone that's ever gotten SARS, MERS, or SARS-CoV-2 had many infections with common cold coronaviruses. I mean, we're working on that [approach], and so are a lot of other people. I just think it might take [a long time].

Maybe I'm wrong about this, I hope so. But maybe what we have to settle for is an approach that would work against the Alpha coronaviruses, an approach that would work against MERS-like viruses and an approach that would work against Delta coronaviruses. Maybe that's what's necessary.

Lara

But, you're quite hopeful for the one that you've just been developing with the help of the Oxford technology.

Pamela

Yes. First of all, I think that the homotypic particle which we made, and also the Oxford team made, really needs to be out there as one of the vaccines, because it will work like all the others. But it has some advantages that the others don't have. It has some disadvantages as well. mRNA vaccines are faster to change from one variant to another, but I think our one can be made quickly. Oxford just looked into making it cheaply, and it could be a cheap alternative that could be easily stored. Then we have to go into all the manufacturing issues to try to make the mosaic particle which is eight times as hard because we want to put eight RBDs on there instead of one.

Lara

So is that where the challenge is, making the nanoparticle?

Pamela

We have no problem making it in the lab, but scaling up to make something that could go into a human, using the good manufacturing protocol, is a big [challenge]. We can make enough to go into non-human primates no problem. But we're not a biotech company that scales things up. It's got to be done by a company that has experience with making things for vaccines.

Lara

How quickly can you do it in the lab?

Pamela

A week maybe. It's very, very versatile.

Lara

What does it involve?

Pamela

Assuming you already have the RBDs expressed and the particles [stockpiled] it can be ready in a day. If we don't have those on hand, it would take maybe a week to make them. Once we've made them, we incubate them overnight to bind and then just separate the unbound stuff and we're good to go. Francis Collins, the NIH Director, wrote a blog about [our technology] when our paper came out (Collins). Caltech also wrote something about it (Caltech).

References

Brune, KD, et al (19 Jan 2016) 'Plug-and-Display: decoration of virus-like particles via isopeptide bonds for modular immunization', Scientific Reports, 6/19234. Back

Caltech (12 Jan 2021) 'Nanoparticle immunization technology could protect against many strains of coronaviruses'.Back

Cohen, AA, et al (12 Feb 2021) 'Mosaic nanoparticles elicit cross-reactive immune responses to zoonotic coronaviruses in mice', Science, 371/6530, 735-41.Back

Collins, F (22 Jan 2021) 'Nanoparticle Technology Holds Promise for Protecting Against Many Coronavirus Strains at Once', NIH Director’s Blog.Back

Escalano, HB et al (29 May 2019) 'Immunization expands B cells specific to HIV-1 V3 glycan in mice and macaques', Nature.Back

Tan, TK et al (22 Jan 2021) 'A COVID-19 vaccine candidate using SpyCatcher multimerization of the SARS-CoV-2 spike protein receptor-binding domain induces potent neutralising antibody responses', Nature Communications.Back