Human papillomavirus

Human papillomavirus

SUPERVISORS: Arvind Varsani and Trevor Sewell

Determination of surface exposure of the cross neutralisation epitope of HPV-16 L2 in HPV-16 L1/L2 virus-like-particles (VLPs)

Cervical cancer and Human papillomavirus

An estimated 400 000 women develop cervical cancer each year accounting for almost 12% of all cancers in women (10422688). Cervical cancer is a preventable disease, however, it is a major burden on public health resources in Africa, where less than 5% of women are screened in the very setting where more than 80% of these cases occur due to the lack of screening (10.1002/ijc.2910630229). It is estimated that 529 000 cases of invasive cervical cancer would develop amongst HIV infected individuals during their life expectancy (10.1016/S0168-1702(02)00189-2).

The most promising means of controlling and reducing the incidence of cervical cancer would be by checking Human papillomavirus (HPV) infection, which is the major causative agent of cervical cancer. HPV DNA has been detected in more than 90% of all tumours of the uterine cervix (10.1002/(SICI)1096-9896(199909)189:1<12::AID-PATH431>3.0.CO;2-F). HPV types 16 and 18 can be detected in nearly 70% of squamous cell carcinomas of the cervix, with HPV-16 being predominant. However, in developing countries, the most appropriate intervention to this disease burden would be by immunisation against HPV infection by a prophylactic or therapeutic vaccine.

HPV vaccines

Papillomaviruses encode two viral structural proteins, the major capsid protein (L1) and the minor capsid protein (L2). The virions of papillomaviruses are highly immunogenic, inducing high titres of neutralising antibodies when systemically inoculated. Papillomavirus like-particles (VLPs), made from the major capsid protein L1 alone, or by the co-expression of L1 and the minor L2 capsid protein, have been proven to induce protective immunity in animal models and S. cerevisiae - produced HPV-16 VLPs used in a recent clinical trial resulted in a 99.7% seroconversion in those patients that received the vaccine (10.1056/NEJMoa020586).

Neutralising epitopes are present in L1 and L2 capsid proteins. However, the neutralising antibodies are highly type specific, despite the significant sequence conservation in the L1 genes of different genotypes. Monoclonal antibodies (Mabs) raised to HPV-16 VLPs have aided in the determination of the both conformational and linear epitopes. Mab H16:V5 was found to block serological reactivity of human sera with HPV-16 capsids; out of the 352 human serum samples tested for the presence of IgG against HPV-16, more than 75% of the reactive sera were completely blocked by this Mab (9292008[/pubmed]. Mab H16:9A was raised to against hybrid HPV-16/11 VLPs where the backbone was genetically made up of HPV-11 and the residues 1-173 were from HPV-16. Very limited data is available on the conformational epitopes that H16:V5 and H16:9A Mabs bind which are of great importance in vaccine design. A common-neutralising epitope for Human papillomavirus types 6 and 16 is present between aa 108-120 (LVEETSFIDAGAP) of the HPV-16 minor capsid protein, L2 [cite]10.1006/viro.1998.9168). Sera of mice immunised with this peptide cross-reacted with L1/L2 capsids of HPV types 6, 11 and 18.

Chimaeric vaccines

There are a number of studies describing the use of papillomavirus-like particles to deliver / display foreign epitopes. L1 C-terminal fusion chimaeras have worked well in presenting the epitopes of the HPV 16 E7 protein, HIV-1 p18 and HIV-1 gp120 CTL epitopes and HIV-1 gp41 epitopes, in HPV types 11, 16 and BPV-1. Attempts have been made to insert epitopes into the core sequences of the PV L1 with an end result of capsomers rather than VLPs being formed. There is sufficient data in the literature to suggest that capsomers are immunogenic and capable of eliciting a neutralising immune response; capsomers have moreover been shown to bind neutralising antibodies raised to intact VLPs (10.1016/S1097-2765(00)80449-9). Our research team has previously investigated the presentation of the cross-neutralising L2 epitope on various region of L1 and found that chimera ChiDF-L2 (substitution into the h4 helix region; 414-426) was the most effective at presenting the L2 epitope while maintaining the "neutralising" Mab H16:V5 and H16:E70 binding (10.1128/JVI.77.15.8386-8393.2003). Work on this continues in Prof Rybicki’s lab as part of the NRF Innovation fund project.

Basic project outline

Virus-like particles resulting from the co-expression of HPV-16 L1 and L2 capsid protein will be decorated with Fab fragments of Mab H16:L2epi, which binds to the cross-neutralising HPV-16 and HPV-18 L2 protein epitope contained in the sequence LVEETSDIGAP (aa 396-439 in L2). Electron density maps of decorated particles and non-decorated articles will be generated by 3- dimensional reconstruction of cryo-EM images of the particles embedded in vitrified ice. Atomic models of HPV-16 L1 and generic Fab fragments will be docked into these electron density maps to characterise the epitopes. Based on the quality of the 3D cryo-EM reconstruction of the L1 and L1/L2 VLPs, an attempt will be made to homology model the L2 protein and dock the modelled L2 capsid protein into L2 density.