Antibody Specificity Following a Recent Bordetella pertussis Infection in Adolescence Is Correlated With the Pertussis Vaccine Received in Childhood

Abstract

Bordetella (B.) pertussis resurgence affects not only the unvaccinated, but also the vaccinated population. Different vaccines are available, however, it is currently unknown whether the type of childhood vaccination has an influence on antibody responses following a B. pertussis infection later in life. Therefore, the study aim was to profile serum antibody responses in young adults with suspected B. pertussis infections, immunized during childhood with either whole-cell (wPV) or monocomponent acellular pertussis (aPV) vaccines. Serum anti-pertussis toxin (PTx) IgG antibody levels served as an indicator for a recent B. pertussis infection. Leftover sera from a diagnostic laboratory from 36 Danish individuals were included and divided into four groups based on immunization background (aPV vs. wPV) and serum anti-PTx IgG levels (- vs. +). Pertussis-specific IgG/IgA antibody levels and antigen specificity were determined by using multiplex immunoassays (MIA), one- and two-dimensional immunoblotting (1 & 2DEWB), and mass spectrometry. Besides enhanced anti-PTx levels, wPV(+) and aPV(+) groups showed increased IgG and IgA levels against pertactin, filamentous hemagglutinin, fimbriae 2/3, and pertussis outer membrane vesicles (OMV). In the wPV(-) and aPV(-) groups, only low levels of anti-OMV antibodies were detected. 1DEWB demonstrated that antibody patterns differed between groups but also between individuals with the same immunization background and anti-PTx levels. 2DWB analysis for serum IgG revealed 133 immunogenic antigens of which 40 were significantly different between groups allowing to differentiate wPV(+) and aPV(+) groups. Similarly, for serum IgA, 7 of 47 immunogenic protein spots were significantly different. This study demonstrated that B. pertussis infection-induced antibody responses were distinct on antigen level between individuals with either wPV or aPV immunization background. Importantly, only 2DEWB and not MIA could detect these differences indicating the potential of this method. Moreover, in individuals immunized with an aPV containing only PTx in childhood, the infection-induced antibody responses were not limited to PTx alone.

Keywords: 2-dimensional electrophoresis (2DE); Bordetella pertussis; acellular pertussis vaccine; antibody specificity; infection-induced response; pertussis toxin (PTx); vaccination; whole-cell pertussis vaccine.

Figure 1

Serum IgG and IgA antibody responses. The (A) IgG and (B) IgA antibody levels against PTx, OMV, Prn, FHA, and Fim2/3 were determined in sera using a MIA. Four groups (n = 9) were included with a distinct immunization background (aPV, wPV) and low (–) or high (+) serum anti-PTx IgG levels. Results are depicted in fluorescence intensity (F.I.). Significant differences are indicated by ^*p < 0.05, ^**p < 0.01, ^***p < 0.001, and ^****p < 0.0001 obtained using a t-test after log-transformation of data.

Figure 2

IgG and IgA responses determined with 1-Dimensional electrophoresis and immunoblotting on a B. pertussis B1917 lysate. (A) IgG and (B) IgA antibody profiles were analyzed in sera of 36 individuals divided over four groups (n = 9) with a distinct immunization background (aPV, wPV) and low (–) or high (+) serum anti-PTx IgG levels. Each group contains a marker lane (left) and nine blots representing single individual.

Figure 3

IgG responses determined with 2-Dimensional electrophoresis and immunoblotting. The presence and intensity of IgG antibodies were determined in sera for 16 individuals divided over four groups (n = 4). (A) Total variance of component 1 (PC1) and component 2 (PC2) were calculated of the total IgG responses analyzed by 2DEWB for each of the 16 individuals in the four groups. (B) The intensity of IgG responses of 133 spots were compared (red = higher, white = equal, blue = lower) to a common reference (median of all intensities for each individual per spot). Euclidean clustering was performed to identify patterns in antibody profiles for both the 16 individuals and the 133 spots. (C) Total variance of PC1 and PC2 were calculated of the significant IgG spots analyzed by 2DEWB for each of the 16 individuals in the four groups. (D) The intensity of the 40 spots that were found significant in the IgG response of all 16 individuals divided over four groups (n = 4) were compared (red = higher, white = equal, blue = lower) to a common reference (median of all intensities for each individual per spot). Euclidean clustering was performed to identify patterns in antibody profiles for both the 16 individuals and the 40 spots leading to three clusters (I–III).

Figure 4

IgA responses determined with 2-Dimensional electrophoresis and immunoblotting. The presence and intensity of IgA antibodies were determined in sera for eight individuals divided over four groups (n = 4). (A) Total variance of component 1 (PC1) and component 2 (PC2) were calculated of the total IgA responses analyzed by 2DEWB for each of the eight individuals in the four groups. (B) The intensity of IgA responses of 47 spots were compared (red = higher, white = equal, blue = lower) to a common reference (median of all intensities for each individual per spot). Euclidean clustering was performed to identify patterns in antibody profiles for both the eight individuals and the 47 spots. (C) Total variance of PC1 and PC2 were calculated of the significant IgA spots analyzed by 2DEWB for each of the eight individuals in the four groups. (D) The intensity of the seven spots (including the unidentified proteins U23–U26) that were found significant in the IgG response of all eight individuals divided over four groups (n = 4) were compared (red = higher, white = equal, blue = lower) to a common reference (median of all intensities for each individual per spot). Euclidean clustering was performed to identify patterns in antibody profiles for both the 8 individuals and the seven spots leading to four clusters (I–IV).