Diamond Annual Review 2019/20

12 13 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 1 9 / 2 0 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 1 9 / 2 0 Antibodies were isolated from human volunteers vaccinated with RH5, and were assessed for the ability to inhibit parasite growth. In this case, there were two groups of inhibitory antibodies, represented by 004 and 016, which bind to different epitope sites. Data frombeamline I04 allowed us to understand the nature of their epitopes on RH5, revealing that 004-like antibodies bind to a site which directly blocks basigin binding, while 016-like antibodies bind close to the basigin-binding site, most likely sterically blocking the RH5-basigin interaction when both RH5 and basigin are membrane associated. In this case, antibody characterisation also gave an interesting surprise. A fascinating class of antibody represented by 011 showed no inhibitory activity alone. However, the presence of 011 increased the potency of inhibitory antibodies such as 004 and 016, allowing them to function at lower concentrations. The crystal structure of 011 revealed a novel epitope on the side of the RH5 molecule, while video microscopy of parasite invasion by Paul Gilson in Melbourne, showed that 011 slows the process of invasion. It only takes around 20 seconds for a malaria parasite to get inside a red blood cell and so we hypothesise that, by slowing invasion, 011 gives inhibitory antibodies more time to act. In summary, structural mapping of the epitopes of the most effective antibodies at Diamond by Protein Crystallography has proved an important part of the characterisation of the antibody response to malaria vaccination. It has allowed us to understand the nature of the epitopes of inhibitory antibodies, revealing where they bind and how they act. These studies are guiding the design of next-generation vaccine immunogens, in which we aim to produce effective neutralising antibodies without generating antibody responses which interfere with their action. These immunogens will be included in the malaria vaccines of the future. Macromolecular Crystallography Group Beamlines I03 and I04 On course for effective newvaccines to targetmalaria Related publications: RawlinsonT. A., Barber N. M., Mohring F., Cho J. S., KosaisaveeV., Gérard S. F., Alanine D. G.W., Labbé G. M., Elias S. C., Silk S. E., Quinkert D., Jin J., Marshall J. M., Payne R. O., Minassian A. M., Russell B., Rénia L., Nosten F. H., Moon R.W., Higgins M. K. & Draper S. J. Structural basis for inhibition of Plasmodiumvivax invasion by a broadly neutralizing vaccine-induced human antibody. Nat. Microbiol. 4 , 1497 (2019). DOI: 10.1038/s41564-019-0462-1 Alanine D. G.W., Quinkert D., Kumarasingha R., Mehmood S., Donnellan F. R., Minkah N. K., Dadonaite B., Diouf A., Galaway F., Silk S. E., Jamwal A., Marshall J. M., Miura K., Foquet L., Elias S. C., Labbé G. M., Douglas A. D., Jin J., Payne R. O., Illingworth J. J., Pattinson D. J., Pulido D.,Williams B. G., de JonghW. A.,Wright G. J., Kappe S. H. I., Robinson C.V., Long C. A., Crabb B. S., Gilson P. R., Higgins M. K. &Draper S. J. Human antibodies that slow erythrocyte invasion potentiatemalaria-neutralizing antibodies. Cell 178 , 216 (2019). DOI: 10.1016/j.cell.2019.05.025 Publication keywords: Malaria;Vaccine development; Erythrocyte invasion; Antibodies M alaria is themost deadlyparasitic disease toaffect humans, causinghundreds ofmillions of severe cases andhundreds of thousands of deaths each year. Developing effective vaccines to prevent the disease has been hugely challenging. To guide the design of next- generation vaccines, researchers set out to understand the human antibody responses to vaccinationwith two promising vaccines. When volunteers are vaccinated, they generate a broad antibody response. Some of these antibodies are effective at neutralising the parasite, while others are not. The research team used Macromolecular Crystallography (MX) beamlines I03 and I04 to collect data from crystals of the human antibodies bound to their malaria surface protein targets. Their aim was to characterise the structure of the most effective antibodies, allowing us to understand how they work. Their results showed that some antibodies directly interfere with the function of malaria surface proteins, preventing them frombinding to human red blood cell receptors. They also revealed that this inhibition can be indirect, most likely by stopping the parasite fromgetting close enough to the red blood cell to allow the interaction to happen. Understanding these molecular details allows us to design new vaccines. The goal is now to use structure-guided methods (‘structural vaccinology’) to developmore effective vaccines. Malaria is still the deadliest parasitic disease to affect humans, causing hundreds of thousands of deaths and hundreds of millions of cases each year. Two parasites cause the majority of human malaria, with Plasmodium falciparum responsible for most fatalities and Plasmodium vivax also causing widespread disease. There is a pressing need to tackle this scourge and an effective vaccine would be of huge benefit 1 . Malaria parasites live and divide within human blood cells and if can we prevent blood cell invasion we will stop the symptoms of the disease and prevent its transmission. Immunisation of human volunteers with components of the machinery used by parasites to invade blood cells generates an antibody response. However, the quality of these antibodies is variable. Some are highly effective at preventing invasion, while others have no effect. Some antibodies can even interfere with the function of effective inhibitory antibodies. How do we design a vaccine which only generates the most effective, protective antibodies? In two studies published this year we characterised the human antibody responses to blood-stage malaria vaccines 2,3 . Human volunteers in Oxford were immunised with two vaccines as they took part in early-phase clinical trials and the antibodies that they generated were assessed for their protective potential. The beamlines at Diamond Light Source were used to visualise the epitopes of the most effective antibodies, revealing details which will guide future vaccine design. The first study focused on the PvDBP molecule from Plasmodium vivax . PvDBP is used by the parasite to interact with the human receptor, DARC, in a process essential for blood cell invasion 4 . People in Africa with lower levels of DARC receptor are protected from vivax malaria. Antibodies targeting PvDBP would also be effective but variation in PvDBP across different strains of Plasmodium vivax makes this a challenge. Could we track down an antibody with broadly inhibitory effects? A panel of antibodies was isolated from vaccinated volunteers and these were screened for efficacy using three different assays. One antibody in particular, DB9, stood out as it was effective in all three assays. By inhibiting blood cell invasion by a panel of parasites with different PvDBP sequences found in naturally infected Thai hospital patients, DB9 showed the desired broadly neutralising effects. However, the presence of other antibodies from the panel reduced the efficacy of DB9, interfering with its neutralising activity. How can we design a vaccine which generates DB9-like antibodies but not those which antagonise its function? To guide this design process, we generated crystals of the antigen-binding fragment of DB9 bound to the DARC-binding domain of PvDBP and, using data collected at beamline I03, determined the structure. The epitope for DB9 was found in a region of PvDBP known as subdomain 3. This was surprising as the binding site on PvDBP for DARC has beenmapped to a different region of PvDBP. However, we hypothesise, based on the structure, that DB9 will sterically block the parasite from getting close enough to the blood cell membrane for PvDBP to bind DARC. It was also interesting to find that the antibodies which interfere with the function of DB9 do not bind to subdomain 3 of DARC. This opens the way to design of future vaccines which induce antibodies resembling DB9 without generating antibodies that interfere with their function. A similar approach was used to study antibody responses against the RH5 protein of Plasmodium falciparum . RH5 makes an interaction with basigin on the surface of red blood cells, which is essential for the parasite to invade 5 . References 1. Draper S. J. et al. Malaria vaccines: recent advances and new horizons. Cell Host Microbe. 24 , 43 (2018). DOI: 30001524 2. Rawlinson T. A. et al. Structural basis for inhibition of Plasmodium vivax invasion by a broadly neutralising vaccine-induced human antibody. Nat. Microbiol. 4 , 1497 (2019). DOI: 31133755 3. Alanine D. G. W. et al. Human antibodies that slow erythrocyte invasion potentiate malaria-neutralising antibodies. Cell 178 , 216 (2019). DOI: 31204103 4. Chitnis C. E. et al . Targeting the Plasmodium vivax Duffy-binding protein. Trends in Parasitology 24 , 29 (2007). DOI: 18023618 5. Crosnier C. et al . Basigin is a receptor essential for erythrocyte invasion by Plasmodium falciparum. Nature 480 , 7378 (2011). DOI: 22080952 Funding acknowledgement: SJD is aWellcome Trust Senior Research Fellow (106917/Z/15/Z) while MKH is aWellcome Trust Investigator (101020/Z/13/Z). Corresponding authors: Prof Simon J. Draper, University of Oxford, simon.draper@ndm.ox.ac.uk and Prof Matthew K. Higgins, University of Oxford, matthew.higgins@bioch.ox.ac.uk Structural insights into human antibodies which prevent blood cell invasion by the malaria parasite. The central panel illustrates the strategy of isolating human antibodies from vaccinated human volunteers. The left-hand panel shows the RH5 protein, essential for blood cell invasion by Plasmodium falciparum (yellow) bound to growth inhibitory antibodies 004 (blue) and 016 (red) and to potentiating antibody 011 (green). The right-hand panel shows a domain from the PvDBP protein (pink) required for blood cell invasion by Plasmodium vivax bound to part of the DARC receptor (orange) and growth inhibitory antibody DB9 (blue).