(Click on image for a larger version)
I am fortunate in that many of the themes have already been covered and the problems that we face in developing vaccines against these really complex organisms have already been touched on. You have heard about HIV and TB; and malaria certainly fits in the category of great challenges that we face in developing vaccines.
Most of you in this audience, I am sure, would in one way or another have heard about this problem, a global health problem. There has been a reassessment of global figures concerning malaria infection and disease prevalence. Not so long ago it was believed that about 90 per cent of the disease burden was in Africa and 10 per cent in the rest of the world, but recent reworkings of the figures suggest there is a much greater burden in South East Asia and Oceania than had previously been recognised. There are about 600 million clinical cases per annum and about 2 million deaths – but these figures are rubbery, as ever – and, of course, a huge economic impact.
To take a leaf from Wayne Koff’s book, I will briefly touch on the sorts of problems we face in developing a vaccine against malaria. But I am not going to illustrate it with a lot of slides.
You have a highly complex and diversified life cycle where the sporozoite will inject the infective form, it is believed, into the skin, and within as little as two to three minutes the sporozoite has made its way to the liver, where it undergoes a massive cellular transformation, re-expresses a whole new different set of proteins, enters the cell and starts to replicate in the liver. It only lasts there for about 10 days, through a stage of replication, then it comes out and once again massively re-expresses a completely new set of proteins, and invades the red cell. It replicates very rapidly, in high numbers, and it has a whole bunch of different receptors that it uses to bind to different tissues in the body.
So you have an extremely complex cell biology, you have an enormous range of redundancy in the invasion pathways into all these different tissues – they use different receptors and different proteins, and on a population basis they have a number of different choices of how to get in. Within this redundancy of invasion pathways and getting in and out of cells there is a great deal of allelic diversity, genetic diversity, in the parasite population. And on top of the allelic diversity there is also antigenic variation.
So you can see the challenges faced in developing subunit vaccines, in particular, and going to live attenuated vaccines is also extremely difficult for an organism as large and as complex as a eukaryote. It is certainly a challenge.
I am going to take another leaf from Wayne’s book, in that he said there was a necessity for innovative and risk-taking approaches and I am going to depart from all the problems of malaria and propose to start to look at something a little bit more positive.
(Click on image for a larger version)
I would like to make reference to something that has really been a wonderful public health success story in the last few years: the emergence of successful polysaccharide conjugate vaccines. I am sure you know what I am talking about when I make reference to all the wonderfully commercially successful and public health successful vaccines listed here.
(Click on image for a larger version)
The basic concept behind these was to take T cell independent polysaccharide antigens from the surface of bacteria and simply conjugate them to the carrier protein, and thereby confer a T cell dependency in the response and immunological memory. Of course, this is not immunological memory in response to challenge, because the carrier proteins are by and large not necessarily drawn from the same organism. But what this sort of process does is to enable you to convert the low-affinity, IgM-dominated responses that are particularly poorly developed in children into high-titre IgG responses to a saccharide candidate antigen. The formulation is relatively straightforward and this is why, in the 1990s in particular, we have seen the emergence of such an impressive class of candidates.
(Click on image for a larger version)
The fact is that the predominant source for carbohydrate material in the marketplace in these vaccine formulations has been native material sourced from growing the bug in vitro, and there are some problems associated with that. So, despite the successes in polysaccharide conjugate vaccines, I believe there is still room to identify some roadblocks and some potential improvements in how we can make this kind of approach even more successful.
It comes down to, essentially, overcoming the problems with purification and application of native material, where you have variable production. In some cases you simply cannot grow; there isn’t a mechanism for growing an organism to sufficient scale that you could purify enough polysaccharide. And so you are limited in what you can do with this material by what nature provides and what in vitro growth can provide.
In contrast, there is at least potentially a capacity, using synthetic chemistry, to produce analogues of this material, but to date this has been extremely limited by the complexity of the chemistry involved. The rest of my talk is going to focus a little bit on overcoming some of those roadblocks in relation to a malaria polysaccharide candidate molecule.
In principle, synthetic chemistry, if it was capable, could deliver a pure compound – certainly an optimised compound – with unique features that would enable coupling et cetera. You can see how chemistry would potentially give improvements, provided you could do it at scale and at cost.
(Click on image for a larger version)
Here is an example of some of the semi-synthetic carbohydrate antigens which have been identified in the literature as desirable targets, from bugs such as the Haemophilus ones, which we already have, but also meningococcal, Leishmania and Candida. And added to this is a malaria candidate, which is the molecule I am going to talk about.
Our platform, you might say, is to try and generate synthetic oligosaccharide antigens of these types – in particular talking about malaria – conjugated to a carrier, to see if it can be of some efficacy.
(Click on image for a larger version)
So what is the candidate polysaccharide antigen of malaria origin that I am talking about? It is the GPI toxin. This molecule has been shown over the years, by myself and a few other people, to play a crucial role in malaria pathogenesis.
You can think of the GPI as being a Toll agonist. In actual fact, it turns out to be somewhat more complicated than that; there seem to be both Toll and possibly non-Toll pathways by which this molecule activates the immune system of the host, but certainly it has biological activities that are reminiscent of polysaccharide or TLR2 or TLR4 agonists. It has got a tri-mannose backbone with an additional mannose residue, a non-acetylated glucosamine and inositol, and three acyl chains. So when you take this purified molecule and inoculate it into mice, for instance, in a proof-of-concept experiment in sufficiency, it kills the mice with malaria-like symptoms. They get toxic shock and keel over in the D-galactosamine toxicity model.
(Click on image for a larger version)
Where does this molecule fit in? Well, these are the main targets of malaria vaccines developed around the world: sporozoite, invasive sporozoite, liver stage antigens, merozoite antigens and schizont antigens. These are the ones I am referring to that are subject to the inherent problem of diversity, variation and redundancy.
(Click on image for a larger version)
With the anti-toxic approach that targets this molecule, I should point out that this molecule is 100 per cent conserved across all malaria isolates that have been examined to date, and it is absolutely essential. If you try and disrupt it in any way, the organism falls apart and can’t survive. So it is not possible to use knockouts to explore its activity.
(Click on image for a larger version)
We certainly feel it is a conserved target, and a conserved carbohydrate target as such.
(Click on image for a larger version)
Here is a cartoon of a parasite that is sequestering in the brain. You see here the vascular endothelium. There is only one parasite, but you can imagine lots of them. It is sticking to the brain and developing there. It looks pretty poisonous! And at this point when it ruptures it releases toxin every 48 hours. The toxin has the effect of inflaming the vascular endothelium and of causing the release of a whole variety of inflammatory agents.
(Click on image for a larger version)
We know that in macrophages the GPI induces interleukin-1, interleukin-6, TNF, MIP-1, MIP-2 and iNOS; and, in the vascular endothelium, ICAM, VCAM and NOS expression. And added to that now is the very important molecule IP-10, which is an important chemokine for recruitment of activated leucocytes.
The GPI, by virtue of these activities, promotes parasite sequestration to the vascular endothelium in specific organs. As I said earlier, when you take a few micrograms of this and put it into D-galactosamine sensitised mice, they get a shock-like syndrome and get disseminated intravascular coagulation.
So this molecule is the key pro-inflammatory agent of malaria parasite origin.
(Click on image for a larger version)
This is the structure of GPI. We did some structure activity studies. We took off the lipids and found that the conserved glycan alone was not biologically active, in terms of pathogenesis – you need the lipids for it to be pathogenic – and this gave the fundamental rationale for developing a saccharide vaccine approach against this target.
(Click on image for a larger version)
To test that, we were not able to explore the biology by using knockout but by approaching it from a vaccine perspective, by intervening against this molecule, we could reduce pathogenesis in preclinical models.
So the structure was isolated from P. falciparum; the synthesis was undertaken by Peter Seeberger, at the Massachusetts Institute of Technology, based on the structure that we provided; a fully validated antigen was developed and we conjugated it to a nominal protein carrier in the first instance, keyhole limpet haemocyanin (KLH). And this was used to immunise mice.
(Click on image for a larger version)
It proved to be very efficacious in protecting mice against a severe disease type syndrome. The P. berghei ANKA model that I am referring to here recapitulates those features of cerebral malaria that you see in humans – namely, strong involvement of an inflammatory cascade.
I should point out that in academic circles there is debate about the exact contribution of inflammation to disease endpoints in malaria, but the most compelling evidence that severe disease is a predominantly inflammatory and toxin-driven condition comes from human genetic disease association studies, which show very strong associations for the molecules I alluded to earlier as being key determinants of pathogenesis. So in case control studies, when you look at frequencies of promoter polymorphisms that control levels of INS expression, ICAM expression and other NF-kappaB dependent endpoints, you see a strong disease association for severe disease in malaria. So it is really very compelling that there is an inflammatory disease process. And this would appear to be a molecule that is key in initiating that event.
This being the case, whereas mice that are sham immunised with KLH alone die very rapidly of a cerebral disease, vaccinated animals are protected when they have a synthetic core glycan vaccine candidate.
(Click on image for a larger version)
This slide shows a couple of samples, not very well labelled. At the right we have the occlusion of the cerebral vasculature, with parasites and leucocytes in the cerebral, in the sham immunised animals; and in the centre, taken at the same time point, you have vessels from the immunised animals and no such occlusion. Clearly the GPI seems to be involved in driving this occlusion event that is so characteristic of the cerebral syndrome.
Based on this sort of evidence, I am now going to talk about what I think is also an innovative message, something that has happened over the last few years, in relation to this kind of approach. That is, it is made very clear to us that we can’t really talk about developing an anti-carbohydrate vaccine with this type of synthetic approach unless we are able to achieve synthesis to a certain scale, to a certain degree of purity and to a certain cost. So, essentially, we set out to found a biotech company called Ancora, whose mission it was to do precisely that, and to use such approaches as proof of principle from a development prospect for the other indications of using synthetic chemistry to complement polysaccharide conjugate vaccine approaches.
The key person here is Bill Christ, who is a process R&D chemist who has worked for 15 years with Eisai Research, and who is personally responsible for establishing a GMP synthesis for carbohydrate chemistry to kilogram scale. He set up a research facility involving a large number of people, and at a certain late stage in his career he decided he wanted a lifestyle change and came to join Ancora.
The challenges that face us in making this candidate to scale are a core oligomannose backbone component; you need versatile mannose building blocks to fit them all in; the glucosamine component has to go into place; you have chiro-inositol building blocks to build the unit; and you would like to have differential phosphorylation and conjugation sites for coupling to your protein. And you need to be able to do this as a reproducible, scalable process, with flexible introduction of functionality and a robust final purification stage – which from a chemistry point of view are non-trivial issues, believe me.
I am not a chemist, so I am giving you just an overview of what Ancora has achieved in this area of late. I have to say it is looking extremely good.
So where does that leave us? Essentially, the initial challenges have been addressed: we have the optimisation and identification of the key biological lead. It has not actually been made to GMP but it is certainly a GMP-scalable process.
There are obviously cost-effectiveness and comparability issues that need to be addressed. We can only give very early estimates as to what a dose of this material might cost, but we certainly believe it should be possible to make this at a competitive price.
So the biotech has been started, with WEHI and MIT IP positions behind it; there is development funding from NIH SBIR; there has been a synthetic pathway developed, with potential for a large number of doses; a formulation process has been developed; and there are ongoing preclinical trials – all of which suggest, indeed, that we may be on the way to developing an anti-toxic vaccine. Provided this can overcome those further regulatory hurdles and achieve more investment, we may have an anti-toxic vaccine to trial.
(Click on image for a larger version)
I would like to acknowledge the funding behind this. The development program is largely funded by the NIH but we have had a lot of discovery research funding from various organisations over the years.
Discussion
Question: Really wonderful work. I am just wondering – I perhaps missed it – what is the carrier antigen? What is your T cell help?
Louis Schofield: There are several. For in-house discovery and biology research we use keyhole limpet haemocyanin. For onward development, that is a question of partnering with the pharmaceutical industry and other commercial carriers, and we are talking to a number of people. So there are a number of options there.
Question: Basically, that lethal dose protection is the final stage. Does your vaccine protect against the cytokines, chemokines, in the blood, or have you measured those as a toxin?
Louis Schofield: Certainly it protects against the inflammatory cascade. It does reduce the inflammatory cascade that is the precursor stage for the fatality – yes, absolutely.
Question (cont.): It does reduce those levels?
Louis Schofield: Yes, absolutely. I didn’t specify that.
Question: There are whole families of glycans of the various pathogens. Is there any cross-reactivity between this and others?
Louis Schofield: That is a good question. I can say that there is no cross-reactivity between this and self, so it doesn’t react with self-GPIs, which I think is a very important point from a development perspective. Whether these cross-react with bacterial saccharides, I don’t know. I haven’t made reference to it, but there would be the aim ultimately of extending this into a whole variety of other saccharide targets from different organisms. And so I guess down the track those sorts of questions can be answered. Right now I don’t have any data on it.Visit these websites
The Sir Mark Oliphant Conference Series is supported by the Australian Government under the International Science Linkages Program.
© 2008 Sir Mark Oliphant Conferences.
Website design acidgreen
