Immunological fingerprint of 4CMenB recombinant antigens via protein microarray reveals key immunosignatures correlating with bactericidal activity

Abstract

Serogroup B meningococcus (MenB) is a leading cause of meningitis and sepsis across the world and vaccination is the most effective way to protect against this disease. 4CMenB is a multi-component vaccine against MenB, which is now licensed for use in subjects >2 months of age in several countries. In this study, we describe the development and use of an ad hoc protein microarray to study the immune response induced by the three major 4CMenB antigenic components (fHbp, NHBA and NadA) in individual sera from vaccinated infants, adolescents and adults. The resulting 4CMenB protein antigen fingerprinting allowed the identification of specific human antibody repertoire correlating with the bactericidal response elicited in each subject. This work represents an example of epitope mapping of the immune response induced by a multicomponent vaccine in different age groups with the identification of protective signatures. It shows the high flexibility of this microarray based methodology in terms of high-throughput information and minimal volume of biological samples needed.

Trial registration: ClinicalTrials.gov NCT00560313 NCT00661713 NCT00944034.

Conflict of interest statement

This study was sponsored by Novartis Vaccines, now acquired by the GSK group of companies. All authors have declared the following interests: E. Ba, E. Bo, A.M., M.G., S.I., R.P., M.M., S.B., S.M., C. Br, C.D., G. To, S.G., R.R., F.C., G.D.G., V.M., M.P. and D.M. were permanent employees of Novartis Vaccines at the time of the study. M.B. was holding a PhD studentship and collaborated with GSK at the time of the study as part of his PhD training. Following the acquisition of Novartis Vaccines by the GSK group of companies in March, 2015, E. Ba, E. Bo, A.M., M.G., S.I., R.P., S.B., S.M., C. Br, G. To, S.G., R.R., G.D.G., V.M., M.P. and D.M. are now permanent employees of the GSK group of companies. M.B. is now permanent employee of Polo GGB. C.D. is now permanent employee of Fondazione Edmund Mach, San Michele all’Adige. C. Be and G. Te are Professors, and M.D. was holding a PhD studentship at the time of the study, at the University of Messina. F.F. is Professor at the University of Molise. F.C. is now permanent employee of US Department of Health & Human services. G.D.G. own shares and stock options of the GSK group of companies.

Figures

Fig. 1. Phage display selection frequency of aminoacids along antigen primary structures.

Sera were analysed as pool by age group (40 infant, 12 adolescent and 29 adult samples). Aminoacid position in the sequence of reference proteins from N-term to C-term is reported on horizontal axis, while vertical axis shows the relative frequency of occurrence for the corresponding aminoacid in the in-frame affinity selected fragments. Frequency refers to the total in-frame fragments analysed in each experiment. Light grey curves are produced by considering only short fragments spanning less than 100 residues.

Fig. 2. Protein microarray analysis unveils the reactivity profile of 4CMenB antigens.

Each horizontal light grey bar represents a single protein or protein fragment spotted in the microarray for GNA2091-fHbp (A-01 to A-28), NadA (B-01 to B-41) and NHBA-GNA1030 (C-01 to C-37). The corresponding coloured squares on their right show the median antibody binding reactivity against that fragment in all tested sera from human vaccines, grouped by age: green squares for adult subject sera, blue squares for sera from adolescents, and pink squares for sera from infants. Stars indicate the full-length or whole fusion forms of each antigen.

Fig. 3. Clustering analysis of protein microarray data and corresponding SBA titres.

Clustering analysis of the mean fluorescence intensity (MFI) data obtained from all antigen fragments (rows) tested against single subject sera (columns). Colour scale of signal intensity is reported on top-left of the heatmap. Subjects are coloured according to age group: in green adult subject sera, in blue sera from adolescents, and in pink sera from infants. Rectangles enclose the four major clusters composed of similar sera reactivity profiles. Roman numbers on the top and bottom of the figure identify the four sera clusters considered for subsequent analysis. Individual SBA titres of subjects against fHbp and NadA reference strains are reported below the heatmap according to a grey scale code, SBA data for ten adolescents and one adult were not available as indicated by an ‘*’ symbol in the same box. No NHBA reference strain was available when the studies were conducted.

Fig. 4. Details of major clusters sera reactivity and functional activity.

Panels a and b respectively represents violin plots of MFIs of all specific antigen fragments and reference strain SBA titres available (vertical axes) in each of the four major clusters (horizontal axis). Black dots represent the mean of the distributions, while horizontal lines indicate quartiles. The wider sections of the violin plots represent a higher density in their data distributions, while the skinnier sections represent a lower density. § since the SBA data were not available for 11 subjects, the panel b violin plot for cluster III represents 7 infants, 12 adolescents and 23 adults; the plot for cluster IV represents 41 infants, 20 adolescents and 6 adults.

Fig. 5. Correlation of bactericidal activity with specific or general protein array reactivity.

Panels a and b respectively represent the single fHbp and NadA fragments contribution to the PLS regression model of the SBA outcome through MFI fragment signals. The fragment importance score corresponds to a weighted sum of the absolute regression coefficients, values > 75% suggest the relevant predictors of the model. Scatterplots of the absolute value of global MFI first PLS component scores against the log2 of SBA titres for respectively fHbp (panel c) and NadA (panel d) antigens in all age groups. Source data are provided as a Source Data file.