An improved customized destination vector for heterologous expression of lysm fusionl proteins and antigen display on bacterium like particles

Abstract

The LysM (lysin motif) domain is a small globular domain of 42-65 amino acids long that is widely distributed in nature, it can be found in procaryotes and eucaryotes in more than 4000 proteins. One to 12 LysM domains bind to N-acetylglucosamine residues of bacterial peptidoglycan (PG) in a non-covalently way. The binding between proteins with LysM domains and PG is strong and stable; it can only be separated under harsh reducing conditions. This can be useful for antigen display on the surface of bacterial PG for immunization purposes. It has been reported that the number of LysM motifs in proteins affects the efficiency of the binding of foreign proteins to the PG. Proteins with LysM domains can be difficult to express in a heterologous system like E. coli because of their size. It is known that proteins with LysM domains tend to aggregate and form inclusion bodies (IB). Previously our laboratory constructed a customized expression vector with 5 LysM domains from a protein (Acglu) of Limosilactobacillus fermentum, which has not been described before. We cloned ORF68, the main antigenic glycoprotein from the Varicella-Zoster Virus (VZV) without the transmembrane domain, into this vector but the fusion protein did not express in E. coli, possibly because of its size (94kDa) or its insoluble nature. Therefore, we decided to construct a new vector with only 2 of the 5 LysM domains from Acglu of L. fermentum. We constructed the expression vector pET-NHis-LysM2 [rfB] and checked it by Next Generation Sequencing. The vector has 2 LysM domains as a N-terminal tag for binding to bacterial PG and is a so-called destination vector compatible with Gateway® cloning technology. It also has the tag RGS-His, which allows protein purification. We cloned VZV ORF68 into this new vector using the Gateway® LR reaction. The fusion protein LysM2-ORF68 was expressed in soluble form, although most of the protein aggregates and forms IB. We optimized the protein expression trying different conditions, even though it always formed IB. Using chaotropic agents, like urea, we could solubilize the aggregated protein and purify it in successive steps. The LysM2-VZVORF68 protein both soluble and recovered from IB, binds to the PG of Gram-positive bacteria. To enhance binding, we exposed the PG shield by treating lactobacilli with acid and heat. The structure of a LysM domain consists of a pair of antiparallel beta strands separated by a pair of short alpha helices. Considering that the binding of LysM domains to PG depends on the native folding of the protein, we can infer that the fusion protein retains its normal folding even after the treatment with chaotropic agents. Further studies are necessary regarding stability of the binding, but we can speculate this new expression vector is promising for the heterologous expression and purification of viral proteins as well as for antigen display on immunomodulatory lactobacilli without generating genetically modified organisms.