The paper in Nature Microbiology is here: https://go.nature.com/2Q8EDMp
Pathogenic arenaviruses are a diverse group of viruses capable of causing severe diseases in humans. A brief and highly fatal outbreak of a new arenavirus species, Lujo virus (LUJV), took place in Southern Africa in 2008. Genomic data indicated that LUJV could not be classified into the two canonical ‘Old World’ and ‘New World’ groups of arenaviruses. Almost a decade later, the main cellular receptor for LUJV GP1, Neuropilin-2 (NRP2), was identified in a genetic screen at the lab of Sean Whelan. This was a surprising discovery since all other arenaviruses that were investigated to date use either transferrin receptor-1 or α-dystroglycan as cell entry receptors. Therefore, the unique use of NRP2 was very intriguing, raising many questions regarding the molecular mechanisms this virus is utilizing to engage with its receptor as well as the molecular differences between this virus and other arenaviruses.
To gain an understanding of this newly emerged, highly pathogenic virus, and to provide the basis for the development of anti-viral therapeutics, we decided to solve the structure of the LUJV cell-attachment machinery (GP1) in complex with its receptor NRP2 using X-ray crystallography. We based the protein design for expressing LUJV-GP1 on our former solved structure of GP1 from the pathogenic Lassa virus. This approach paid off and LUJV-GP1 was well produced along with the first CUB domain of NRP2. Using high throughput crystallography screens, we were able to get crystals of the JUJV-GP1/NRP2 protein complex. The crystals we have obtained had a porcupine appearance and were very thin, which could have hindered structure determination efforts.
However, we then found additives that slightly improved the thickness of the crystals. Having these improved crystals and utilizing a micro-focus X-ray beamline at the European Synchrotron Radiation Facility allowed us to record adequate X-ray diffraction data for this complex and subsequently solve the crystal structure.
The structure that we recently published provided important insights regarding the LUJV mechanism of interaction with NRP2 and its evolutionary relatedness to other arenaviruses. Structural analysis indicates a resemblance of LUJV to Old World arenaviruses, both in the GP1 structure itself as well as in the overall location of the receptor binding surfaces on GP1. The structure elucidated the interaction interface between LUJV-GP1 and NRP2, which was subsequently validated experimentally. An interesting discovery was the involvement of a Ca2+ ion in the formation and upholding of the binding site on NRP2. Another prominent observation is that the complete binding site for NRP2 is formed between two adjacent GP1 subunits, in the context of the trimer, an insight that will be important for devising strategies to combat LUJV.
The information gained from our study makes an important step toward better understanding the intricate cell entry mechanisms that arenaviruses utilize, and it may advance the development of novel therapeutic applications against LUJV. However, there are still many open questions regarding these mechanisms that LUJV in particular, and other arenaviruses in general utilize in order to enter cells. Specifically, it looks like LUJV needs to dissociate from NRP2 prior to membrane fusion at the endocytic pathways and a molecule termed CD63 is somehow involved in facilitating membrane fusion. This is reminiscent of the receptor switching mechanism of the Old World Lassa arenavirus that dissociates from α-dystroglycan and binds LAMP1 in order to trigger its own spike complex for membrane fusion. Whether LUJV is also utilizing such a triggering mechanism, and the exact molecular details of such processes are currently unknown. Such open questions will need to be addressed in the future.