PMAP-36 reduces the innate immune response induced by Bordetella bronchiseptica-derived outer membrane vesicles

Abstract

Host defense peptides (HDPs), such as cathelicidins, are small, cationic, amphipathic peptides and represent an important part of the innate immune system. Most cathelicidins, including the porcine PMAP-36, are membrane active and disrupt the bacterial membrane. For example, a chicken cathelicidin, CATH-2, has been previously shown to disrupt both Escherichia coli membranes and to release, at sub-lethal concentrations, outer membrane vesicles (OMVs). Since OMVs are considered promising vaccine candidates, we sought to investigate the effect of sub-bactericidal concentrations of PMAP-36 on both OMV release by a porcine strain of Bordetella bronchiseptica and on the modulation of immune responses to OMVs. PMAP-36 treatment of bacteria resulted in a slight increase in OMV release. The characteristics of PMAP-36-induced OMVs were compared with those of spontaneously released OMVs and OMVs induced by heat treatment. The stability of both PMAP-36- and heat-induced OMVs was decreased compared to spontaneous OMVs, as shown by dynamic light scattering. Furthermore, treatment of bacteria with PMAP-36 or heat resulted in an increase in negatively charged phospholipids in the resulting OMVs. A large increase in lysophospholipid content was observed in heat-induced OMVs, which was at least partially due to the activity of the outer-membrane phospholipase A (OMPLA). Although PMAP-36 was detected in OMVs isolated from PMAP-36-treated bacteria, the immune response of porcine bone-marrow-derived macrophages to these OMVs was similar as those against spontaneous or heat-induced OMVs. Therefore, the effect of PMAP-36 addition after OMV isolation was investigated. This did decrease cytokine expression of OMV-stimulated macrophages. These results indicate that PMAP-36 is a promising molecule to attenuate undesirable immune responses, for instance in vaccines.

Keywords: Bordetella bronchiseptica; Cathelicidins; Host defense peptides; Outer membrane vesicles; PMAP-36; Vaccine development.

Graphical abstract

Fig. 1

Effect of PMAP-36 on OMV release of B. bronchiseptica. (a) Increasing concentrations of PMAP-36 were supplemented during OMV induction and isolated OMVs were analyzed with SDS-PAGE. Black arrow points to PMAP-36 dimers present in isolated OMVs. (b) Protein concentrations of isolated OMVs were quantified using the BCA assay. Shown is the mean of three measurements with SD. Significant differences are indicated by ****p<0.0001, obtained using a one-way ANOVA with a post-hoc Dunnett test, compared to 0 µM PMAP-36. (c) Size of isolated OMVs was determined using DLS. Shown is the mean of three measurements with SD. (d) Isolated OMVs were visually inspected using EM. Bars represent 200 nm. Shown is a representative image of three individual experiments.

Fig. 2

Protein and lipid quantification of induced OMVs. (a-b) Protein content and concentrations of isolated OMVs were determined using (a) SDS-PAGE and quantified using (b) BCA assay (n=5). (c) LPS content of isolated OMVs was visualized using silver staining on SDS-PAGE and (d) total lipid concentration was quantified using FM4-64 lipid dye (n=8). s = sOMVs, p = pOMVs, h = hOMVs. Significant differences compared to sOMVs are indicated by **p<0.01, ****p<0.0001, obtained using a one-way ANOVA with a post-hoc Dunnett test. Gels are representative of three individual experiments.

Fig. 3

Morphology and stability of induced OMVs. (a) Size of isolated OMVs was measured with DLS (n=5). (b) Isolated OMVs visualized using EM. Insert of pOMVs is four times enlarged. Shown is a representative image of three individual experiments. Bars represent 200 nm. (c) Stability of OMVs was assessed by size (left panel) and count (right panel) variations over a temperature gradient (n=3). Shown is the mean with SEM. Statistic differences compared to sOMVs were calculated using a linear mixed-model analysis with a post-hoc Dunnett test.

Fig. 4

Lipidomic analysis of isolated OMVs. Isolated OMVs were subjected to mass spectrometry to determine lipid composition (n=3). (a) Total lipid composition of OMVs and (b) of whole cells. (c) Four most affected lipid species. s = sOMVs, p = pOMVs, h = hOMVs, WC = Whole Cells. Significant differences compared to sOMVs are indicated by **p<0.01, ***p<0.001, ****p<0.0001, obtained using a paired two-way ANOVA with a post-hoc Dunnett test.

Fig. 5

pBMDM1 response to induced OMVs. pBMDM1 macrophages were stimulated with isolated OMVs and activation was measured by determining (a) cell-surface markers using FACS and (b) cytokines using ELISA. MFI = mean fluorescence intensity. - = no stimulation, s = sOMVs, p = pOMVs, h = hOMVs (n=7-12). Significant differences compared to the non-stimulated control are indicated by *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, obtained using a repeated measures one-way ANOVA with Geisser-Greenhouse correction and post-hoc Dunnett test.

Fig. 6

PMAP-36 modulation of pBMDM1 cytokine expression in response to OMVs. pBMDM1 macrophages were stimulated with (a) sOMVs or (b) hOMVs in the presence of different concentrations of PMAP-36. Cytokine secretion was determined by ELISA. Values were converted to fold change with respect to the results obtained with OMVs not supplemented with PMAP-36. Shown is the mean with SD. Significant differences are indicated by *p<0.05, **p<0.01, obtained by using a linear mixed-model analysis on the raw data with a post-hoc Dunnett test (n=3-7).