Prototype of mucosal vaccine against hepatitis E virus resists gastrointestinal conditions in vitro

Abstract

The Hepatitis E virus causes hepatitis; its principal port of entry is the gastrointestinal mucosa. Its genome consists in ORF1 that encodes for a polyprotein needed for replication, ORF2 the viral capsid and ORF3 a multifunctional phosphoprotein. The LysM domains are ubiquitous small domains, which mediate attachment of enzymes to bacterial peptidoglycan or to fungal chitin. Bacterium-like-particles (BLP) are non-live bacteria treated with heat and acid, which conserve their shape but lose proteins and DNA. The aim of our study was to generate fusion proteins and displayed them on the surface of BLP, to generate a mucosal vaccine which combines carrier and adjuvant properties for oral administration. In this work we cloned and expressed a chimeric protein with 5 LysM domains (LysM5) and ORF2 most immunogenic domain(O2P2). We obtained the protein under native and denaturing conditions purificated by NiNTA chromatography. BLP derived from Lactiplantibacillus plantarum IBL027, previously reported to have adjuvant activity on mucosa, were used for evaluating antigen display on its surface. Briefly, we put in contact bacterial lysates or purified proteins and BLP for an hour in rotation at room temperature and then washed them with PBS to get LysM5O2P2-BLP027. These complexes were tested on solutions simulating saliva, gastric and intestinal juices and incubated in 37°C for 5 minutes, 1 and 3 hours, respectively. Protein integrity was checked by SDS-PAGE. We tested several conditions for optimal expression of LysM5O2P2 in E. coli. Surprisingly, after purification, the protein did not bind to BLP but when we tested crude supernatants (under native and denaturing conditions), it bound at different proportions. Then, we tested the resistance of complexes to gastrointestinal conditions; both were resistant to artificial saliva (pH:7.2, lysozyme 100ppm) and simulated gastric juice (pH:2.5) but BLP exposing the native protein was resistant to simulated intestinal juice (pH:7.2; pancreatin: protease>1900USP) and BLP exposing the denatured protein was not. SDSPAGE revealed LysM5O2P2 was split in two by pancreatin, which is consistent with the presence of a trypsin recognition site in the protein sequence. Capsid proteins of enteric viruses must resist gastrointestinal conditions to reach their target cells. We postulate that the proper folding of LysM5O2P2 protects it from digestion. Some authors reported that native O2P2 dimerization protects it against trypsin digestion. The next step will be to study oral immunization protocols with the prototype vaccine LysM5O2P2-BLP.