An Interview with Dennis Burton

Structure-function in HIV research

Dennis Burton, PhD, is one of the world's leading experts on antibodies. After gaining his PhD from the University of Lund, Sweden, in physical biochemistry, Burton returned to the UK and held faculty positions in biochemistry and molecular biology at the University of Sheffield. He then took a position at The Scripps Research Institute as Professor of Immunology and Molecular Biology, where he has been for the last 16 years.

Burton holds a number of scientific advisory positions, including at the University of California, San Diego, the HIV Vaccine Trials Network, the Foundation for AIDS Research (amfAR), and the Conference on Retroviruses and Opportunistic Infections, and has had a hand in organizing many prestigious scientific conferences, including this year's Keystone Symposium on HIV vaccines. He also sits on a number of grant review committees, including at the National Institutes of Health, amfAR, and the Wellcome Trust in the UK. Since 2001 he has been the Scientific Director of IAVI's Neutralizing Antibody Consortium (NAC).

At the end of the 1980's, Burton was a key member of the team that first demonstrated the expression of antibodies from a combinatorial phage display library, a technique that greatly facilitated the study of antibody. Soon after he turned his attention to human antibody responses against HIV and how they might be manipulated in a vaccine setting. Burton has called his approach retrovaccinology, the idea being that by fine mapping broadly-neutralizing antibodies one can work back from that point to design an immunogen that will effectively elicit such antibodies. His work in nonhuman primates has shown that neutralizing antibody can completely protect from infection with immunodeficiency virus, studies that contribute to the hope that an effective vaccine will one day be developed.

Burton is widely acknowledged by his peers as a front-runner in AIDS vaccine research, but still manages to be well known for his affable nature and for inspiring teamwork among colleagues. The success of the NAC is due in no small part to Burton's personal qualities that enable him to foster a spirit of cooperation and collaboration. He recently spoke with IAVI Report Editor Simon Noble about structures—from HIV's proteins to research consortia—and the functions they bring about.

How successful do you think IAVI's Neutralizing Antibody Consortium has been to date?

As far as I know it was the first research consortium of this type in AIDS vaccine research, and I think it's been amazingly successful, it has driven the field in lots of aspects and has been very, very influential. I think you'd expect it to be, because the individual scientists in the NAC are just outstanding. I've never worked with such an extremely talented group of individuals. Collectively they have defined what is, I think, the only sensible approach to coming up with an antibody component for an AIDS vaccine—a rational, molecular approach. The members of the consortium are right at the frontier of understanding at the molecular level the interaction between neutralizing antibodies and the HIV envelope, and that's where a vaccine is going to be developed.

How do you think the NAC could be improved?

The NAC started out as a group of investigators studying a relatively narrow area covering the structure and function of HIV envelope and antibodies. So it could always be improved in terms of the breadth of talent and, as we progress, then we'll need new expertise in many areas, for example, in B cell biology, in understanding immunogenicity, and in understanding better how to elicit antibody responses generally. So there's always great scope for improving the quality of the whole outfit.

A key organizational feature of the NAC has been 'enabling projects' that accelerate research by pooling certain resources—for example, we have common repositories of antibodies and envelope protein reagents—which is straightforward but has enormous impact because individual members can get hold of these standard reagents very quickly and easily.

The other organizational aspect that helps greatly is standardization of assays. We've put into place standard neutralization assays and immunization protocols, and run these as much as possible on a consortium-wide basis, freeing up the individual investigators to be innovative. That's really what we want the members to be free to do, to innovate, to take those tedious and labor-intensive tasks out of the research equation.

What about the intellectual openness that the NAC has fostered?

That's been very important in two aspects. One is in the sharing of information, ideas, reagents, and so on, and that has been quite exemplary, and you can clearly see that in the number of publications that contain multiple authors from the NAC. The second aspect has been in terms of intellectual property, where the NAC has pioneered the sharing of intellectual property between multiple laboratories so that discoveries made by investigators will favor them proportionately more, but would also benefit anyone within the consortium. That's a healthy and stimulating environment to do research in. The agreements also give IAVI the opportunity to develop vaccines so they'll be available where they're most needed, in developing countries.

What are your impressions of the other consortia in AIDS vaccine research?

The first organization that I would point out is the Vaccine Research Center (VRC), which I think has been a huge success. I think Gary Nabel's recruiting to the VRC has been inspirational. I usually visit the VRC several times a year and it's just a real pleasure to go because there are so many really, really great people there and the work they're doing is so good. I think that is a great institution for HIV vaccine research and its success certainly underlines what can be done when you bring people together working on a common problem.

The other consortia, the NIH in the form of the CHAVI [Center for HIV/AIDS Vaccine Immunology] and the Gates Foundation, in the form of CAVD [Collaboration for AIDS Vaccine Discovery], I think that undoubtedly they took note of the success of the NAC. Right now it's too early to say how successful they're going to be.

I think the CHAVI, obviously, has some outstanding individuals involved, the likes of Joe Sodroski at Harvard and Andrew McMichael at Oxford, they are clearly world class AIDS researchers. It'll be interesting to see how it goes, they are trying to cover a lot of ground in the CHAVI and of course, given the amount of money involved, there are huge expectations. If you take one research area, primary acute infection, there's no doubt that prominent researchers outside the CHAVI have found it very difficult after the formation of the CHAVI. These guys are far too valuable to the overall research effort to lose.

The CAVD has been formed only recently so it's very difficult to make a judgment. Some components of the CAVD are following a closely parallel track to that of the NAC, which is flattering. I'm very impressed with Tachi Yamada, having met him he seems like a really terrific choice for that job.

The VRC is a real bricks-and-mortar, vaccine-devoted institution, existing under one roof. Do you think that's significant to its success?

I think that the neutralizing antibody problem, for example, could definitely benefit from being focused into one, two, or three centers, rather than as it is now, dispersed over many centers. It could also benefit from a re-thinking about mission-driven research. To take one example: grants. If you look at the RO1 [NIH funding] structure, the classical way to keep your RO1 grants flowing is to move quickly, seize on opportunities, write another grant, something else comes up, seize on that, raise another grant, and so on.

What we're trying to do in the HIV vaccine design field is somewhat different from that—it's more mission-oriented. We don't necessarily want people running to the next opportunity, to some degree you want to keep people on mission and that is not particularly well-suited to the RO1-style of funding.

I'm not criticizing RO1s or peer-reviewed research at all, I want to really emphasize that, I think that the NIH peer-reviewed process has been massively successful and has produced a huge amount of science that's been of great benefit for mankind. So I don't think that should be replaced at all. What I'm talking about here is an alternative, given a specific set of problems-HIV vaccine design. In this instance, one needs different structures.

What's required is not quite the funding that happens in biotechnology either, because this is much riskier than classical biotech. It's something in between classical biotech and classical RO1 investigator-initiated funding. So we need some sort of funding mechanism that recognizes the unique nature of this mission and seeks to deal with it accordingly.

I think an R&D vaccine institute, or institutes, could be very useful for this type of problem. This would also affect the career structures of not just the PIs, but of the individuals actually doing the bench research. Some of this research is not necessarily going to produce the same amount of publications that one typically has to accrue to make an academic career, so one needs to think about alternate career structures that can hold the best researchers into this mission-driven project.

To bring young people in to HIV vaccine research, they'll probably have to be guaranteed a longer than usual tenure to do certain research projects and their accomplishments recognized as being somewhat different from traditional academic research. This can be done, people have found ways to do this, for example, on the Human Genome Project. So it just takes some imagination and the political will to take a crack at it.

Obviously all of this requires a lot of money, and it would also involve bringing in some of the tools of biotechnology and project management and so on. I think that would be helpful.

But the Human Genome Project gave rise to a whole new career ladder in bioinformatics, which didn't really exist before that.

Right, and if HIV vaccine design can be cracked, then I think that will open up the whole field of vaccinology and we'll then have rational vaccine design for a whole slew of vaccines. There is no health measure that's done more for humankind than vaccines, and if we could actually get control of vaccines so that we could rationally design them, rather than be at the mercy of a slightly altered pathogen, then we'd be in control of a whole new area of human health.

You were involved in the very recent structural definition of an HIV neutralizing antibody, b12, binding to the CD4 binding site of gp120 (Nature 445, 732, 2007), a paper that has generated a fair bit of enthusiasm, even in the non-scientific press. How big a step forward do you think this represents?

I think it's a major step. We have known about this site of vulnerability since we discovered the antibody in the early '90s, but this report from Peter Kwong's lab provides the molecular level details of this HIV:antibody interaction. That's key because vaccine design now, I believe, has to be done at the molecular level. This new molecular understanding, hopefully, will transfer into molecular design of vaccine candidates.

So in light of that paper, what are the prospects of now designing an immunogen that will induce similar neutralizing antibodies?

We absolutely should not underestimate the gargantuan size of that task. We have now defined the shape of the HIV surface that the antibody recognizes, the epitope that is part of the CD4 binding site. Now we have to recreate that molecular shape with some high degree of precision in some other protein environment where we can use it as a vaccine candidate. That, in essence, is the approach that people are going to try. This is a great starting point on the molecular scale. It will be tough, there's undoubtedly lots and lots of work to be done and it's really pushing at our understanding of structure and of immune responses, but I think it can be done.

Supposing this immunogen design problem is surmounted, what then would be necessary for an effective AIDS vaccine?

Even if we could elicit antibodies like this one, b12, very effectively, that wouldn't be sufficient for an AIDS vaccine. First, we will need to elicit multiple other broadly-neutralizing antibody specificities. Second, I think that an effective HIV vaccine is going to require two components; a component eliciting multiple broadly-neutralizing antibodies, and another eliciting a robust, rapid, potent cellular immune response. Research into both of those areas needs to be maximized, because I think only when those two components are brought together are we going to have real success.

Will the antibody response need to be mucosal?

I think that's a very, very open question. I'm not sure that HIV really is a mucosal pathogen—it goes in via the mucosae and it does replicate early on in the gut, but there are pathogens that just do all their damage in the mucosae, and you can show mucosal immunity. I think most important is going to be a systemic response to control the virus. So I think there is still very, very many question marks about the requirement of mucosal immunity. And also it's much more difficult to elicit long-lasting mucosal immunity than systemic immunity.

So you don't agree that there is this really short window of opportunity to snuff out the virus immediately after an infective event?

I'm not even sure about that, because in the experiments we've done, we can protect monkeys with systemic antibody; we can also protect them with mucosal antibody but we needed a very high concentration of mucosal antibody. It may be that virus infects only when the mucosae are damaged, and it's not completely clear to me that any level of mucosal immunity would necessarily protect you if the mucosae are broken, for example. If you look at the initial integration events, they may well be systemic; activated T cells can be wherever. I think there are a lot of question marks, but in the first instance, I'd rather have systemic immunity than anything else.

What do you think is the best we could hope for in the near term of a partially effective vaccine?

Probably the best that we can hope for in the medium term is that we get some indication of control from the T-cell vaccines, that the vaccinated patients do better than those who receive the placebo. We'll find out in the next two years or so how effective they are. I'll be absolutely delighted if T-cell responses alone can provide some clearly measurable benefit, but I'm not convinced that that's what we're going to see.

There's currently an extremely limited repertoire of well-characterized neutralizing antibodies. Do we need more novel neutralizing antibodies, and a better understanding of their original biology in their host?

Absolutely, and there are a lot of efforts going on there, including in the NAC. John Mascola and Rich Wyatt at the VRC have some very nice data showing that at least one donor has broadly-neutralizing antibodies directed to the CD4 binding site, a bit like b12. I'm sure a lot more will be identified—the NAC is also targeting such donors in Africa—and we'll understand more how they're neutralizing. We'll then have many more monoclonal antibodies that will be useful for molecular vaccine design.

What's the next important HIV structure that the field really needs?

The trimer-that's what everybody wants to see, the native infectious form of the trimer. It's going to be very difficult because it's such an unstable, transient structure, and that's part of the problem with making a vaccine. We have to chemically engineer the proteins to stabilize them and still retain native conformation, and as the technologies of structural genomics and robotics are applied, hopefully sooner or later a crystallizing, stable trimer will appear.

What other areas of HIV research, particularly vaccine research, do you find particularly interesting or promising?

I think the restriction factors are very interesting and are indicative of how, possibly, drugs could be developed. They're almost like natural drugs, in a sense. I also think the whole biochemistry of viral entry is very interesting, and there'll be more drug targets there.

The other area that I find quite interesting, which is vaccine-related, is trying to elicit better immune responses through understanding the interplay of innate and adaptive immunity. Everyone has jumped on TLRs [Toll-like receptors], but I'm in the same institute as Bruce Beutler and even though he is one of the discoverers of TLR function, he also believes there's a lot more to it than TLRs. I think that's going to be a huge developing area. I can foresee a future where cellular immunity is elicited not by viral vectors, as it is now, but by designed immunogens that elicit strong, directed cellular immune responses. That could be terrific for vaccinology, I think.