Lactobacillus plantarum displaying CCL3 chemokine in fusion with HIV-1 Gag derived antigen causes increased recruitment of T cells

Abstract

Background: Chemokines are attractive candidates for vaccine adjuvants due to their ability to recruit the immune cells. Lactic acid bacteria (LAB)-based delivery vehicles have potential to be used as a cheap and safe option for vaccination. Chemokine produced on the surface of LAB may potentially enhance the immune response to an antigen and this approach can be considered in development of future mucosal vaccines.

Results: We have constructed strains of Lactobacillus plantarum displaying a chemokine on their surface. L. plantarum was genetically engineered to express and anchor to the surface a protein called CCL3Gag. CCL3Gag is a fusion protein comprising of truncated HIV-1 Gag antigen and the murine chemokine CCL3, also known as MIP-1α. Various surface anchoring strategies were explored: (1) a lipobox-based covalent membrane anchor, (2) sortase-mediated covalent cell wall anchoring, (3) LysM-based non-covalent cell wall anchoring, and (4) an N-terminal signal peptide-based transmembrane anchor. Protein production and correct localization were confirmed using Western blotting, flow cytometry and immunofluorescence microscopy. Using a chemotaxis assay, we demonstrated that CCL3Gag-producing L. plantarum strains are able to recruit immune cells in vitro.

Conclusions: The results show the ability of engineered L. plantarum to produce a functional chemotactic protein immobilized on the bacterial surface. We observed that the activity of surface-displayed CCL3Gag differed depending on the type of anchor used. The chemokine which is a part of the bacteria-based vaccine may increase the recruitment of immune cells and, thereby, enhance the reaction of the immune system to the vaccine.

Fig. 1

The expression cassette for C-terminal (a) and N-terminal (b) anchoring of CCL3Gag in L. plantarum. All parts of the cassette are easily exchangeable using restriction sites: SalI between the signal peptide or the N-terminal anchor and CCL3Gag, EcoRI, at the end of the insert, and MluI between CCL3Gag and the C-terminal anchor. a C-terminal anchoring was accomplished by fusing the N-terminus of the CCL3Gag fragment to a signal peptide (SP; from Lp_0373) and the C-terminus to a cell wall anchor (Cwa) from Lp_2578 (Cwa comprises 194 residues of Lp_2578) [42]. b Four N-terminal anchoring motifs were used, all containing an SP. Two lipoanchors were generated using lipobox fragments from Lp_1261 (residues 1–75) and Lp_1452 (residues 1–142), one transmembrane anchor was generated by fusing CCL3Gag to C-terminally truncated Lp_1568 (complete protein with 7-residues truncation), which contains an SP without a predicted signal peptide cleavage site, and one LysM anchor was generated by fusing CCL3Gag to full length Lp_3014 (a 204 residue putative transglycosylase with an N-terminal LysM domain)

Fig. 2

Detection of CCL3Gag fusion protein produced in L. plantarum strains harboring various plasmids. Bacterial cell-free protein extracts were prepared 3 h after induction by use of peptide pheromone, and analyzed by Western blotting, using polyclonal goat anti-CCL3 antibody and polyclonal rabbit anti-goat HRP-conjugated IgG. A strain harboring the pEV plasmid [36], not containing the ccl3gag fragment, was used as a negative control. The following CCL3Gag producing strains were analyzed (expected mass of the CCL3Gag protein between parenthesis): Lp_1452CCL3Gag (39.9 kDa), Lp_1568CCL3Gag (95.3 kDa), Lp_3014CCL3Gag (45.0 kDa), Lp_1261CCL3Gag (31.9 kDa), Lp_CCL3Gag_Cwa (50.5 kDa). The arrows indicate the location of the fusion proteins. Lane M shows a molecular mass standard

Fig. 3

Growth of CCL3Gag-expressing L. plantarum cells. OD₆₀₀ values were measured at the induction point (black bars). Next, each culture was divided into a pheromone induced culture (white bars) and a non-induced culture (gray bars). OD₆₀₀ values were measured 3 h after the point of induction. The data are presented as the means from triplicates +SD

Fig. 4

Flow cytometry (a) and microscopy (b) analysis of surface localization of CCL3Gag. L. plantarum cells harboring plasmids designed for N- or C-terminal anchoring of CCL3Gag were probed with goat anti-CCL3 polyclonal antibody and, subsequently, Alexa Fluor^®488-conjugated rabbit anti-goat IgG antibodies. L. plantarum harboring pEV without the ccl3gag gene fragment was used as a negative control and is shown in all five histograms in panel a (black line). The data are presented as one representative experiment. Each experiment was performed at least three independent times and gave the similar results

Fig. 5

Chemotactic properties of L. plantarum harbouring various constructs. The graphs show migration of Esb-MP cells towards CCL3Gag-expressing strains compared to migration towards the negative control (Lp_Ev). The number of migrated cells was counted using flow cytometry and relative chemotaxis is shown as the average fold change relative to the negative control. a L. plantarum strain displaying CCL3Gag anchored C-terminally to the surface. The data presented are the means from 3 replicates +SEM. Statistically significant differences compared to the negative control (p < 0.01) are indicated by an asterisk (*). b L. plantarum strains displaying CCL3Gag anchored N-terminally to the surface. The data presented are derived from one representative experiment. The experiments were performed at least three times and these independent experiments showed the similar trends

Fig. 6

Chemotactic properties of L. plantarum strains surface displaying CCL3Gag, in presence of soluble CCL3 protein. Migration of Esb-MP cells towards CCL3Gag expressing strains in the presence of 3 ng/ml free CCL3 was compared to migration towards the negative control (Lp_Ev). The number of migrated cells was counted using flow cytometry and relative chemotaxis is shown as average fold change relative to the negative control. The data are presented as the means from at least 5 replicates +SEM. Statistically significant differences (p < 0.01) are indicated by an asterisk (*)