Characterization of Vibrio cholerae neuraminidase as an immunomodulator for novel formulation of oral allergy immunotherapy

Abstract

To improve current mucosal allergen immunotherapy Vibrio cholerae neuraminidase (NA) was evaluated as a novel epithelial targeting molecule for functionalization of allergen-loaded, poly(D,L-lactide-co-glycolide) (PLGA) microparticles (MPs) and compared to the previously described epithelial targeting lectins wheat germ agglutinin (WGA) and Aleuria aurantia lectin (AAL). All targeters revealed binding to Caco-2 cells, but only NA had high binding specificity to α-L fucose and monosialoganglioside-1. An increased transepithelial uptake was found for NA-MPs in a M-cell co-culture model. NA and NA-MPs induced high levels of IFN-γ and IL10 in naive mouse splenocytes and CCL20 expression in Caco-2. Repeated oral gavage of NA-MPs resulted in a modulated, allergen-specific immune response. In conclusion, NA has enhanced M-cell specificity compared to the other targeters. NA functionalized MPs induce a Th1 and T-regulatory driven immune response and avoid allergy effector cell activation. Therefore, it is a promising novel, orally applied formula for allergy therapy.

Keywords: AAL; M-cell; Neuraminidase; Oral application; PLGA-microparticles; WGA.

Fig. 1. Characterization of NA as functionalization substance.

NA, AAL and WGA remained stable up to 180 min in SGF experiments (A) and bound to Caco-2 cells in a dose-dependent manner (B; FITC-NA: 16 μg: blue line, 32 μg: orange line, 64 μg: light green line, 128 μg: dark green line versus MFI control (red line) left panel, WGA: light blue line, AAL: orange line versus MFI control (red line). AAL was inhibited by α-L fucose and WGA by TCT. GM1 and α-L fucose inhibited NA binding to Caco-2 cells in a dose-dependent manner (C, n = 3). Data are presented as mean ± standard error of mean (SEM). ^#P, statistical significant differences to the uninhibited control (0 μM); *P, significant differences between α-L fucose and TCT inhibitions at same concentrations. *P/^#P < 0.05, **P/^##P < 0.01, ^###P < 0.001, ^####P < 0.0001. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2. Safety evaluation and immunomodulatory properties of the targeting molecules NA, AAL and WGA.

NA induced significantly higher IFN-γ (A) and IL10 (C) production in naive spleen cells, while IL4 was not increased (B). Caco-2 cell stimulated with the targeting molecule NA revealed significantly higher induction of CCL20 compared to AAL and WGA with a maximum peak after 3 h (D). (n = 4), data are mean + SEM, *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 3. Efficient binding of NA-FS to Caco-2 cells and M-cells.

Binding of FS to Caco-2 cells was higher when the particle surface was functionalized with NA, AAL or WGA compared to the negative control (Plain-FS) (A). The transepithelial uptake of both, NA- and AAL-FS, was significantly higher when M-cells are present (co-culture vs. monoculture) (B). In the presence of M-cells (co-culture) NA-FS were more efficiently transported through the epithelium compared to AAL-FS and WGA-FS. (n = 3), data are mean + SEM, *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 4. Digestion evaluation of functionalized MPs.

Encapsulation of OVA in uncoated MPs (Plain-MPs) or functionalized MPs (NA-MPs, AAL-MPs, WGA-MPs) ensured protein integrity up to 120 min in SGF experiments in contrast to unprotected OVA.

Fig. 5. Safety and immunomodulatory properties of NA-MPs.

Spleen cells of naive mice were stimulated with the respective functionalized MPs preparation. NA-MPs induced significantly higher levels of IFN-γ and IL10 measured in spleen cell supernatants (A and C), but did not affect IL4 levels (B). Caco-2 cell experiments revealed an induction of CCL20 after 3 h of NA-MPs stimulation compared to the other MP preparations (D). (n = 4), data are mean + SEM, ***P < 0.001. ****P < 0.0001.

Fig. 6. Repeated gavages of NA-MPs do not induce IgE responses.

After 6 cycles of oral application of MPs preparations (see Supplementary Fig. 2), no induction of OVA-specific IgE in serum was found (A). OVA-specific IgA was significantly higher in the NA-MPs and AAL-MPs fed mice (B), with the latter MPs inducing also significantly reduced OVA-specific IgA in intestinal lavages compared to naive animals (C). No significant differences were observed for total IgA in intestinal lavage (D). (n = 8), data are mean + SEM, *P < 0.05, **P < 0.01.

Fig. 7. Systemic immunomodulation is observed after repeated gavages of NA-MPs.

After 6 cycles of oral application of MPs preparations to allergic mice (see Supplementary Fig. 3), no elevation of OVA-specific IgE (A) and IgA (B) serum levels was measured. Six cycles of oral feeding of OVA-loaded NA-MPs was associated with elevated levels of IFN-γ (C) and IL10 (D) production by mouse splenocytes upon OVA stimulation. (n = 8), data are mean + SEM, *P < 0.05, **P < 0.01, ***P < 0.001.