Human papillomavirus (HPV) is well recognized as a risk factor for several malignancies. Mucosal HPV could be transmitted by sexual practice and persistent infection of mucosal HPV leads to cervical intraepithelial neoplasia, possibly progressing into cervical cancer. Cutaneous HPV can cause various cutaneous warts and are also associated with non-melanoma skin cancer in patients with impaired immunity.

As a result of decades of scientific endeavors, several licensed virus-like particle-based HPV vaccines are commercially available showing high efficacy in inhibiting HPV infection. However, these vaccines exhibit a notable immunogenic limitation in that they are highly HPV type-specific. In addition, the complex logistics associated with the licensed vaccines and their considerable manufacturing costs have motivated scientists to develop new generations of HPV vaccines based on the minor capsid protein L2, focusing on a major cross-neutralization epitope present in the L2 amino terminus. In our group, we have generated two candidates, one for mucosal HPV types - PANHPVAX, and another for cutaneous HPV types - CUT-PANHPVAX. Both vaccines contain a so-called ‘polytope’, i.e. a string of L2 epitopes derived from different HPV types. The polytope is inserted into a hyper thermostable scaffold, derived from thioredoxin of the archaeon Pyrococcus furiosus. Another shared feature is the presence of a c-terminal heptamerization domain.

The investigation into developing the vaccine candidates has centered on several key aspects. Firstly, selecting the types of HPV L2 epitopes present in the polytope aims to streamline the number of epitopes required to achieve maximal cross-protection. Secondly, determining which portion of the L2 epitope within one HPV type to incorporate. Further, the selection of the scaffold for an optimal epitope presentation is essential to ensure efficient and targeted immune activation. In both vaccine antigens, the L2 epitopes are separated by a ‘GGP’ amino acid linker. However, upon detailed examination of our candidate’s immunogenicity, we noted that for certain HPV type neutralizing antibody titers are not sufficiently achieved. Examples are HPV4 in the case of CUT-PANHPVAX and HPV31 in the case of PANHPVAX.

In addition, we wished to gain more insights on the spacer/linker function in polytopic protein vaccinology. For those purposes, different spacers were designed to insert into the upstream spacer on the problematic epitope. The key findings of our paper are:

• Spacer modifications influence the B cell response induced by the antigen

The single modifications of the spacer within the recombinant protein did not alter the hydrophobicity, solubility or thermal stability of the studied protein vaccines. However, the single modification of one spacer in the polytopic protein vaccine could lead to the changes in immunogenicity. In either case of our vaccine candidates, spacer V4, twice the length as the original spacer (V6), significantly impair the immunogenicity of both antigens. Overall, the ‘GGP’ spacer (original one) remains the optimal spacer that we would choose for our vaccine candidates.

• Antigen binding affinity to the neutralizing monoclonal antibody could predict the antigen immunogenicity

Neutralizing monoclonal antibodies (nmabs) produced against certain HPV-type L2 epitopes could be shaping the critical ‘native’ structure of the L2 epitope. When one antigen candidate has better affinity nmab than others, it would indicate that certain epitopes within the antigen involved in binding would be more similar to the ‘native’ structure of the L2 epitope in the virus capsid. In our study, we found that linker V6 antigen variant had the highest affinity to the tested nmAbs which aligns with the observation that V6-inserted antigen variant also induced the higher antibody. To attain improved immunogenicity, our vaccine candidate should mimic the ‘native’ virus epitope structure as more as possible. Accordingly, by the means of measuring antibody affinity, we would be able to select optimal antigens before proceeding animal experiments.

It was our intention to draw attention of careful spacer/linker selection within recombinant protein vaccine antigens. Currently, we are investigating the detailed structure of selected L2 epitopes, in complex with neutralizing monoclonal antibodies using x-ray crystallography, in order to gain more insight into the role of the L2 amino terminus and L2-based HPV vaccines.