Multivalent vaccines demonstrate immunogenicity and protect against Coxiella burnetii aerosol challenge

Abstract

Vaccines are among the most cost-effective public health measures for controlling infectious diseases. Coxiella burnetii is the etiological agent of Q fever, a disease with a wide clinical spectrum that ranges from mild symptoms, such as fever and fatigue, to more severe disease, such as pneumonia and endocarditis. The formalin-inactivated whole-cell vaccine Q-VAX^® contains hundreds of antigens and confers lifelong protection in humans, but prior sensitization from infection or vaccination can result in deleterious reactogenic responses to vaccination. Consequently, there is great interest in developing non-reactogenic alternatives based on adjuvanted recombinant proteins. In this study, we aimed to develop a multivalent vaccine that conferred protection with reduced reactogenicity. We hypothesized that a multivalent vaccine consisting of multiple antigens would be more immunogenic and protective than a monovalent vaccine owing to the large number of potential protective antigens in the C. burnetii proteome. To address this, we identified immunogenic T and B cell antigens, and selected proteins were purified to evaluate with a combination adjuvant (IVAX-1), with or without C. burnetii lipopolysaccharide (LPS) in immunogenicity studies in vivo in mice and in a Hartley guinea pig intratracheal aerosol challenge model using C. burnetii strain NMI RSA 493. The data showed that multivalent vaccines are more immunogenic than monovalent vaccines and more closely emulate the protection achieved by Q-VAX. Although six antigens were the most immunogenic, we also discovered that multiplexing beyond four antigens introduces detectable reactogenicity, indicating that there is an upper limit to the number of antigens that can be safely included in a multivalent Q-fever vaccine. C. burnetii LPS also demonstrates efficacy as a vaccine antigen in conferring protection in an otherwise monovalent vaccine formulation, suggesting that its addition in multivalent vaccines, as demonstrated by a quadrivalent formulation, would improve protective responses.

Keywords: Coxiella burnetii; adjuvant; aerosol challenge; guinea pig; hypersensitivity; multivalency; reactogenicity; subunit vaccine.

Figure 1

Antibody and T cell profiling after administration of Q-VAX^®. (A) Time course IgG profiles from plasma of C57BL/6 mice administered Q-VAX via subcutaneous and intramuscular routes (n = 5 mice per group). The top 5 Tier 1 antigens and the top 30 reactive antigens on d28 are included. The hashed line in panel A indicates when the boost was administered (d69); mice in the intramuscular group were not boosted. (B) In T cell recall experiments, C57BL/6 mice (n = 5) were immunized with Q-VAX^® intraperitoneally, and their splenocytes were harvested 10 days after the prime. The splenocytes were stimulated by purified protein from the downselected panel of lead antigens. Significance is compared to the no -antigen group. Statistics were performed with one-way ANOVA and Dunnett’s multiple comparison test. p-value ≤ 0.0001 (****).

Figure 2

Immunogenicity screen of candidate (C) burnetii antigens. Groups of C57BL/6 mice (n = 3 per group) were immunized with purified CBU1910-SorHis protein (pp1910) or His-Trap resin to which different (C) burnetii proteins expressed in IVTT reactions were bound via polyhistidine tags and formulated in AddaVAX™ for immunization. (A) Plasma was probed on microarrays displaying purified (C) burnetii proteins, and bound Abs were visualized with secondary Abs against IgG, IgG1, and IgG2c. Heat map shows signals on d28; red = high, yellow = intermediate, white = low; arrayed proteins were listed on the left and immunizing antigens were listed at the top. (B) Array signals (group mean ± SD) at different time points post-immunization. (C) T -cell immunogenicity screen of candidate (C) burnetii antigens. Groups of mice (n = 3) were immunized with IVTT-expressed proteins and boosted 8 weeks later with purified proteins indicated in each panel for the recall assay (IFNγ ELISpot). Numbers of spot-forming cells at different concentrations of antigen are expressed as a fold-over the number of spots at 0 mg/ml antigen. Colored bars = assay recall antigen corresponding to immunizing antigen.

Figure 3

IVAX-1 (MPLA, CpG1018, AddaVAX) is a potent combination adjuvant for enhancing IgG2c responses and increasing proinflammatory cytokine production. (A) Timeline of events. Two groups with the same immunizations composed of immunodominant protein CBU1910 and adjuvant combinations were run simultaneously. One group (n = 5) was used to study the longevity of the IgG response, and another (n = 3) was used for T -cell assays. (B) Plasma from both experimental groups was probed on a C. burnetii protein microarray looking at the response to immunizing antigen CBU1910. Q-VAX and PBS were used as positive and negative controls. (C) Plasma from day 42 was used to assess IgG1 and IgG2c responses on the protein microarray. (D) Animals from the T -cell recall group were terminated on day 9, and their splenocytes were subjected to stimulation for 18 h with CBU1910. Anti-IFNγ capture antibodies were used to determine spot counts in an IFNγ ELISpot. Statistics were performed with one-way ANOVA and Dunnett’s multiple comparison test. (E) Supernatants from the 18-h stimulation were assessed for Th1/Th2 cytokines using a cytokine bead assay. Extrapolated values for the PBS control group were subtracted from the other groups. Statistics were performed with two-way ANOVA and Tukey’s multiple comparison test. p-value ≤ 0.05 (*), ≤ 0.01 (**), or ≤ 0.0001 (****).

Figure 4

Guinea pig vaccine formulations were evaluated for reactogenicity. All animals were sensitized with Q-VAX and rested for 14 days, then intradermally administered either the six vaccine candidates (n = 4) or Q-VAX and PBS intradermally (n = 7) on shaved skin sections. (A) Representative histopathological hematoxylin and eosin (H&E)-stained skin sections of experimental groups at 4× magnification with a 200 -µm scale bar. Arrowheads border areas of immune cell infiltrate/inflammation, and asterisks indicate areas of degenerate neutrophils/abscess formation. (B) Mean histopathological scores for experimental groups separated into different morphologic categories. Significance is compared to the PBS group. Statistics were performed with one-way ANOVA and Dunnett’s multiple comparison test. p-value ≤ 0.05 (*) or ≤ 0.0001 (****).

Figure 5

Multiple antigens induce IgG responses in Hartley guinea pigs. (A) Timeline of events. Animals were rested for 7 weeks after initial immunization prior to challenge. A challenge study with Coxiella burnetii strain NMI RSA 493 was performed in Hartley guinea pigs (n = 5). Formulations with four antigens include CBU1910, CBU0891, CBU0612, and CBU0545. CBU1398 and CBU0307 were included in formulations involving six antigens. NMI LPS used as an immunogen in candidate formulations was extracted from formalin-inactivated (C) burnetii NMI RSA 493. TLR7 is 2Bxy and part of the top formulation from a previous challenge study serving as a baseline. IVAX-1 includes MPLA, CpG1018, and AddaVAX. (B) Plasma was collected at intervals on days 10, 21, 28, 35, and 42 post-prime and assessed for IgG production using the protein microarray platform containing (C) burnetii antigens. Significance looks at plasma from day 42 compared to WCV and performed with one-way ANOVA and Dunnett’s multiple comparison test. (C) Plasma from day 42 was assessed for the production of IgG1 and IgG2 on the (C) burnetii protein microarray. Statistics were performed with two-way ANOVA and Dunnett’s multiple comparison test. p-value ≤ 0.05 (*), ≤ 0.01 (**), ≤ 0.001 (***), or ≤ 0.0001 (****).

Figure 6

An intratracheal aerosol challenge study with Coxiella burnetii strain NMI RSA 493 was performed in Hartley guinea pigs (n = 5). (A) Changes in body weight in calculated percentages were recorded for 14 days after infection. Significance is compared to the WCV group and performed with two-way ANOVA and Dunnett’s multiple comparison test. (B) Changes in temperature in calculated percentages were recorded for 14 days after infection. Fever is denoted as an increase in temperature greater than 2%. Significance is compared to the WCV group and performed with two-way ANOVA and Dunnett’s multiple comparison test. (C) Splenomegaly was determined after termination by comparing to the PBS group. Statistics were performed with one-way ANOVA and Dunnett’s multiple comparison test. p-value ≤ 0.05 (*), ≤ 0.01 (**), ≤ 0.001 (***), or ≤ 0.0001 (****).