Testing S. sonnei GMMA with and without Aluminium Salt-Based Adjuvants in Animal Models

Abstract

Shigellosis is one of the leading causes of diarrheal disease in low- and middle-income countries, particularly in young children, and is more often associated with antimicrobial resistance. Therefore, a preventive vaccine against shigellosis is an urgent medical need. We have proposed Generalised Modules for Membrane Antigens (GMMA) as an innovative delivery system for Shigella sonnei O-antigen, and an Alhydrogel formulation (1790GAHB) has been extensively tested in preclinical and clinical studies. Alhydrogel has been used as an adsorbent agent with the main purpose of reducing potential GMMA systemic reactogenicity. However, the immunogenicity and systemic reactogenicity of this GMMA-based vaccine formulated with or without Alhydrogel have never been compared. In this work, we investigated the potential adjuvant effect of aluminium salt-based adjuvants (Alhydrogel and AS37) on S. sonnei GMMA immunogenicity in mice and rabbits, and we found that S. sonnei GMMA alone resulted to be strongly immunogenic. The addition of neither Alhydrogel nor AS37 improved the magnitude or the functionality of vaccine-elicited antibodies. Interestingly, rabbits injected with either S. sonnei GMMA adsorbed on Alhydrogel or S. sonnei GMMA alone showed a limited and transient body temperature increase, returning to baseline values within 24 h after each vaccination. Overall, immunisation with unadsorbed GMMA did not raise any concern for animal health. We believe that these data support the clinical testing of GMMA formulated without Alhydrogel, which would allow for further simplification of GMMA-based vaccine manufacturing.

Keywords: GMMA-based vaccine; OMV; adjuvant; immune response; unadsorbed GMMA.

Figure 1

S. sonnei-specific humoral responses elicited upon vaccination with GMMA either alone or with Alhydrogel or AS37. Five-week-old CD1 mice were immunised intramuscularly (50 µL, 25 µL/leg) on study days 0 and 28. Mice were bled on study days −1, 27, and 42. LPS-specific (A) and S. sonnei GMMA protein-specific (B) total IgG responses were quantified by ELISA. Antibody complement-mediated functionality in killing S. sonnei (C) was assessed by SBA. The GMMA dose (quantified as total OAg) received by animals in each immunisation group is reported below each graph. Individual IgG levels are reported in the graphs (symbols) together with geometric means (bars) and 95% CI. Light grey bars represent the geometric means after the first immunisation, whereas dark grey bars represent the geometric means after the second immunisation. Dotted lines represent the level of antibodies (A,B) or the bactericidal activity (C) of preimmune sera.

Figure 2

Comparison of S. sonnei-specific humoral responses elicited upon vaccination with GMMA either in the absence or presence of the two tested adjuvants. Five-week-old CD1 mice were immunised intramuscularly (50 µL, 25 µL/leg) on study days 0 and 28. Mice were bled on study days-1, 27, and 42. LPS-specific (A,B,E,F) and S. sonnei GMMA protein-specific (C,D,G,H) total IgG responses and antibody complement-mediated functionality in killing S. sonnei (I,J) observed in mice immunised with unadsorbed S. sonnei GMMA (black lines) were compared to those elicited by S. sonnei GMMA adsorbed on Alhydrogel (red lines) or S. sonnei GMMA adsorbed on AS37 (blue lines) by the parallel line approach with the possibility of changing the dose range (PLAS). Each curve is represented with the log-transformed dose on the abscissa and the log-transformed ELISA units or SBA titres on the ordinate (natural log was used for both transformations). For each formulation type, three out of four consecutive doses were used by selecting the dose range that provided a higher slope for the linear curve; no weight function was used for the regression. For each formulation type, the chosen dose range in the comparison could be different (PLAS). The calculated relative potency with a 95% confidence interval is reported in the graph when applicable. Relative potency was not calculated when the dose–response was evaluated as nonlinear.

Figure 3

S. sonnei-specific IgG subclasses and IgM elicited upon two vaccinations with GMMA either alone or adsorbed on Alhydrogel or AS37. Five-week-old CD1 mice were immunised intramuscularly (50 µL, 25 µL/leg) on study days 0 and 28 with 36 ng of S. sonnei GMMA (quantified as OAg). Mice were bled on study days −1, 27, and 42. LPS-specific (A–C) and S. sonnei GMMA protein-specific (B–D) IgM and IgG subclasses were quantified by ELISA. Geometric means of each IgG subclass level are reported in graphs (A,B) (bars), whereas individual IgM levels are reported in graphs (C,D) (symbols) together with geometric means (bars) and 95% CI. * p < 0.05, ** p < 0.01; Mann–Whitney test.

Figure 4

S. sonnei-specific total IgG and IgM levels elicited upon two vaccinations with S. sonnei GMMA either unadsorbed or adsorbed on Alhydrogel. Female New Zealand White rabbits were vaccinated intramuscularly (0.5 mL) on study days 0 and 28 with 15 µg of S. sonnei GMMA (quantified as total OAg). LPS-specific (A,B) and S. sonnei GMMA protein-specific (C,D) total IgG and IgM were quantified by ELISA. Antibody complement-mediated functionality in killing S. sonnei (E) was assessed by SBA. Individual IgG and IgM levels are reported together with geometric means (bars) and 95% CI. * p < 0.05, ** p < 0.01, *** p < 0.001; Mann–Whitney test. A paired t-test was performed between different time points, and an unpaired t-test was performed between different immunisation groups. Light grey bars represent the geometric means after the first immunisation, whereas dark grey bars represent the geometric means after the second immunisation. Dotted lines represent the level of antibodies (A–D) or the bactericidal activity (E) of preimmune sera.

Figure 5

Rabbit body temperature and weight changes after vaccination with S. sonnei GMMA either unadsorbed or adsorbed on Alhydrogel. Female New Zealand White rabbits were vaccinated intramuscularly (0.5 mL) on study days 0 and 28 with 15 µg of S. sonnei GMMA (quantified as OAg). Rabbit body temperature (A–C) was monitored before and 6 and 24 h after each immunisation. Body weight (D) was measured before immunisation and 24, 36, and 72 h after each vaccination. Mean and standard deviation for all groups and individual values of ∆T (°C) are reported in graphs (A,B) (∆T > 1.5 in red). Changes in T (°C) and weight (Kg) during the time after vaccination are reported in box and whisker graphs (plotting min to max values, (C,D)). * p < 0.05, ** p < 0.01; A paired parametric t-test was performed between different time points, and an unpaired parametric t-test was performed between different immunisation groups.