1. Introduction
Tuberculosis (TB) is a leading global cause of death from a single infectious agent, causing over 1.3 million deaths in 2022 [
1]. The coronavirus disease 2019 (COVID-19) pandemic has negatively impacted TB diagnosis and care, resulting in a reversal of the progress made up to 2019 in combating TB [
2]. It is estimated that a quarter of the world’s population is latently infected with
Mycobacterium tuberculosis, allowing the bacteria to survive and enter a quiescent phase that can reactivate later in life.
Mycobacterium bovis Bacillus Calmette–Guérin (BCG) is the only licensed TB vaccine available for humans. Despite being able to provide good protection for newborns and children against disseminated forms of TB, BCG displays a high variation in its degrees of protection against pulmonary TB infection in adolescents and adults (ranging from 0–80%), the population that bears the highest burden of the disease [
3]. Therefore, improved vaccines that offer efficient and consistent protection against this disease are urgently needed.
Several novel vaccine candidates against TB are currently in preclinical and clinical trials, including recombinant BCG (rBCG) strains, live attenuated
M. tuberculosis strains, mycobacterial extracts, recombinant viral-vectored platforms, mRNA, and protein–adjuvant combinations (reviewed in [
4]). Subunit vaccines have the advantage of being safer compared to live vaccines, especially in immunocompromised individuals, and may be given as multiple-dose regimens. Many pre-clinical and clinical studies have shown the ability of protein-based subunit vaccines to induce long-lasting T-cell responses; therefore, they are able to provide long-term protection against
M. tuberculosis infection [
5,
6,
7,
8]. In a landmark study, the M72/AS01E candidate exhibited 49.7% efficacy in preventing active pulmonary tuberculosis disease among adults with
M. tuberculosis infection [
9]. This protective signature was characterised by a sustained elevation of M72-specific antibodies and CD4
+ T cells.
Central to the development of successful TB subunit vaccines is the careful selection of antigens that can elicit robust and protective immune responses in humans. This selection process is influenced by several key factors, including the immunogenicity of the antigens in the human population and their expression by the bacteria at different stages of infection. During infection, mycobacteria encompass a diverse range of bacterial populations, existing in distinct stages of metabolic activity and exhibiting varied antigen expression profiles. This includes both actively replicating bacteria and dormant forms, with the ability to switch between these metabolic states. Many vaccine studies include different antigens in their study, often in the form of fusion proteins [
10,
11]. Including multiple antigens from
M. tuberculosis as candidates for TB subunit vaccines can contribute to broadening immunogenicity and enhancing the protective efficacy across populations with different HLA compositions. Previous studies have identified a protein family comprising the
M. tuberculosis sulphate-assimilation pathway (SAP), whose members are highly upregulated during conditions of intracellular stress [
12]. Protein components of the SAP pathway are immunogenic and recognised by both murine and human T cells during the course of
M. tuberculosis infection, and when tested as vaccine antigens, these proteins were protective in murine challenge models [
13,
14]. The fusion protein vaccine CysVac2 (CysD fused with the mycobacterial antigen Ag85B), when delivered with MPLA/DDA, was highly immunogenic, inducing both CysD- and Ag85B-specific polyfunctional Th1- and Th17-type CD4
+ T-cell responses. CysVac2 was also highly protective in both pre- and post-exposure models of
M. tuberculosis infection in mice at the chronic stage of infection, and when given as a BCG booster, CysVac2 was able to improve the protection afforded by prior BCG vaccination [
14,
15]. When overexpressed by recombinant BCG, it also enhanced protection at the chronic stages of
M. tuberculosis infection [
16].
MPT83 (Rv2873) is a surface-expressing lipoglycoprotein of
M. tuberculosis and an analogue to MPB83 in
M. bovis. MPT83 was upregulated after
M. tuberculosis infection of mice and functions as a highly immunogenic antigen [
17]. In addition, MPT83 was recognised by the immune system of patients with active TB. Subcutaneous immunisation with MPT83 protein vaccines induced significant protection against aerosol
M. tuberculosis challenge in mice [
18], and the protein has successfully been used as a component of protective TB fusion antigens [
11]. MPT83 is a
M. tuberculosis proapoptotic protein, which could contribute to its ability to stimulate strong immunity [
19]. In this study, we investigated the effect of modifying the CysVac2 fusion protein by the addition of the MPT83 antigen, resulting in the CysVac5 vaccine. Here, we examined the capability of this fusion protein to provide broad immune responses and protection against
M. tuberculosis challenge in two different strains of mice.
2. Materials and Methods
Bacterial strains and growth conditions. E. coli DH5α and BL21 (DE3) were grown in Luria–Bertani (LB) broth or agar (Difco BD, Franklin Lakes, NJ, USA). M. bovis BCG Pasteur (ATCC35734) and M. tuberculosis H37Rv (ATCC27294; BEI Resources, NIAID, NIH, NR-13648) were cultured in Middlebrook 7H9 (Difco BD) broth supplemented with albumin-dextrose-catalase (ADC; 10% v/v), Tween-80 (0.05% v/v) and glycerol (0.2% v/v). To enumerate the bacterial numbers, cultures were plated onto Middlebrook 7H11 (Difco BD) agar with oleic-acid-albumin-dextrose catalase (OADC; 10% v/v) and glycerol (0.5% v/v). All cultures were grown at 37 °C with or without shaking. The antibiotic kanamycin was added to the media when required at 50 μg/mL.
Recombinant CysVac5 expression and purification. The recombinant prokaryotic expression plasmid pET28a-CysVac5 was constructed by fusing the gene encoding the secreted antigen FbpB (Ag85B; Rv1886C) with downstream insertion of
cysD (Rv1285) and
mpt83 (Rv2873) genes from
M. tuberculosis genomic DNA. Plasmids were transformed into
E. coli BL21 (DE3), and selected transformants were grown until the mid-log phase (OD
600 of 0.5–0.6). CysVac5 expression was induced by 0.5 M of isopropyl β-D-1-thiogalactopyranoside (IPTG) shaking overnight at 37 °C. The bacterial culture was pelleted by centrifugation, resuspended in lysis buffer (50 mM KPO
4, 4 mM KCl, 5 mM b-ME, lysozyme 0.1 mg/mL, pH 8) and sonicated with a probe sonicator (Branson, Milford, CT, USA). The soluble and insoluble fractions were separated at 12,000 rpm for 1 h at 4 °C (Eppendorf, Hamburg, Germany). The insoluble fraction was resuspended overnight at 4 °C in urea buffer (8 M urea, pH 8) and then passed through a His-tag affinity chromatography column (Clontech, Mountain View, CA, USA). After several washes with urea buffer with increasing concentrations of imidazole, the His-tagged protein was eluted from the column in urea buffer containing 200 mM imidazole and then dialysed overnight in a refolding buffer (25 mM HEPES, 0.1 M KCl, 1.5 mM DTT, 5% glycerol, pH 7.5). The protein concentration was measured by the Bradford assay, and the protein expression was confirmed by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blot, as described previously [
14].
Preparation of live mycobacterial and MPLA/DDA liposomal adjuvanted vaccine. M. bovis BCG Pasteur strain was grown in 7H9 broth media with appropriate antibiotics to the mid-log phase (OD600 of 0.5–0.6). Bacterial cultures were then transferred into a 50 mL falcon tube and pelleted by centrifugation (3500 rpm, 10 min, room temperature). The supernatant was removed, and the bacterial cell pellet was resuspended in 1 mL of sterile PBS. Prior to infection, the cells were briefly sonicated (Branson) to disperse aggregated mycobacteria. Dimethyl dioctadecyl ammoniumbromide (DDA; 5 mg/mL; Sigma-Aldrich, St. Louis, MO, USA) was prepared in sterile water and heated to 80 °C with regular stirring for 20 min, whereas monophosphoryl lipid A (MPLA; InvivoGen, San Diego, CA, USA) was resuspended in 1 mL of DMSO and vortexed to allow complete solubilisation. The formulation of MPLA/DDA adjuvant was prepared in a ratio of 1:10 (w/w; 25 µg of MPLA with 250 µg of DDA).
Mice and immunisation procedures. Female C57BL/6 or BALB/c mice (6–8 weeks old) were obtained from the Animal Resources Centre (Perth, Australia). Mice were maintained in a specific pathogen-free condition at the Centenary Institute animal facility. All murine experiments were performed in strict accordance with the approvals granted by the University of Sydney Animal Ethics Committee (K75/9-2012/3/5846). Experiments were not blinded, and animals were randomly assigned to experimental groups. For subcutaneous (s.c.) administration, mice were anaesthetised with gaseous isofluorane (4%) and injected at the base of the tail with 3 μg of recombinant protein (CysVac2 or CysVac5) formulated in MPLA/DDA (25 μg/250 μg per dose, respectively; 200 μL total volume). Mice received three vaccinations 2 weeks apart and proceeded to the immunogenicity study at 2 and 6 weeks post-last vaccination, or M. tuberculosis challenge. Mice receiving BCG were vaccinated s.c. once with 5 × 105 CFU 10 weeks prior to M. tuberculosis challenge.
M. tuberculosis infection. Six weeks following the final vaccination, mice were challenged with M. tuberculosis H37Rv in an inhalation exposure system using a Middlebrook airborne infection apparatus (Glas-Col, Terre Haute, IN, USA) with an infective dose of approximately 100 CFU. At 4 and 20 weeks after challenge, serial dilutions of lungs and spleen homogenates were plated on supplemented Middlebrook 711 agar (Difco BD) to enumerate the bacterial loads.
Organ collection and processing. Mice were held in a restraint, and a scalpel blade was used to puncture the tail lateral vein. Approximately 200 μL of blood was collected from mice in PBS-heparin (20 U/mL; Sigma-Aldrich), and the suspension was stratified on Histopaque 1083 (Sigma-Aldrich) and centrifuged. The peripheral blood mononuclear cells (PBMCs) were collected and washed twice with RPMI medium (Thermo Fisher Scientific, Waltham, MA, USA). Whole lungs from euthanised mice were perfused post-mortem with cold PBS-heparin. For isolation of lung leukocytes, lung tissue in complete RPMI medium, consisting of L-glutamine and 25 mM Hepes (Invitrogen, Waltham, CA, USA), FCS (10% v/v), 2-mercaptoethanol (50 μM; Sigma-Aldrich) and PenStrep (100 U/mL; Invitrogen), was digested with collagenase IV (50 U/mL; Sigma-Aldrich) and DNAse I (13 μg/mL; Sigma-Aldrich) at 37 °C for 45 min prior to homogenisation (using a GentleMACS Dissociator, Miltenyi Biotec Australia, Macquarie Park, Australia) and multiple filtration steps. Spleens were homogenised by passing through a 40 μm cell strainer (BD Biosciences, San Jose, CA, USA) in RPMI medium and then pelleted by centrifugation. Erythrocytes were removed using ACK lysis buffer, and single-cell suspensions were prepared and counted by Trypan Blue (0.04%) exclusion followed by dilution to the desired concentration.
Intracellular cytokine staining and flow cytometry. Antigen-specific cytokine production was measured by intracellular immunostaining (ICS) and flow cytometry. The antigens utilised were single components of the recombinant protein vaccine (Ag85B, CysD and MPT83), as well as fusion protein CysVac2 and CysVac5. Up to 5 × 10
6 cells/mL of leukocytes were stimulated for 3–4 h (37 °C, 5% CO
2) with appropriate antigens (10 μg/mL). Cells were further incubated for 16 h with Brefeldin A (10 μg/mL; Sigma-Aldrich) to allow cytokine accumulation prior to immunostaining. Cell suspensions were incubated with 1.25 μg/mL anti-mouse CD16/CD32 (2.4G2; BD Biosciences) in FACS wash buffer (PBS with 2% FCS) to block Fc receptors, then washed and incubated for 30 min with antibody mix to label surface markers and Fixable Blue Dead Cell stain (Thermo Fisher Scientific) to differentiate dead cells. Cells were permeabilised using BD Cytofix/Cytoperm kit (BD Biosciences). Intracellular staining was performed using appropriate antibodies, as shown in
Supplementary Table S1. Immunostained cells or beads were fixed in 10% buffered formalin prior to the data acquisition using an LSR-Fortessa analyser (BD Biosciences). Acquired data was analysed using FlowJo analysis software version 10.10 (Treestar Inc., Ashland, OR, USA). A Boolean gate combination was used to calculate the frequency of single- or multiple-cytokine-positive cell subsets (the gating strategy is described in
Supplementary Figure S1).
Murine IFN-γ T-cell responses. Membranes of a 96-well ELISPOT filter plate (Millipore, Darmstadt, Germany) were coated (overnight, 4 °C) with anti-IFN-γ antibody (10 μg/mL) in PBS, then washed and blocked with complete RPMI medium (Thermo Fisher Scientific) for 2 h at 37 °C. Splenocyte suspensions were plated at 2 × 105 cells per well. Purified vaccine peptides or proteins (10 μg/mL), Concanavalin A (3 μg/mL) or medium alone were added and incubated for 20 h at 37 °C in 5% CO2. Plates were washed 6 times with PBST (0.1% Tween 20), and captured IFN-γ was detected with biotinylated anti-IFN-γ antibody (XMG1.2-biotin; 5 μg/mL in PBS/0.5% BSA, incubated overnight, 4 °C). Plates were washed, followed by the addition of avidin alkaline phosphatase (Sigma-Aldrich; 1:1000 v/v in PBS/0.5% BSA, 45 min, RT). The presence of IFN-γ-producing cells was visualised by the addition of alkaline phosphatase substrate solution (in NPP buffer; 0.01% Substrate A, 0.01% Substrate B; AP conjugate substrate kit, Bio-Rad, Hercules, CA, USA). Developed spots were then counted using an automated ELISPOT reader (AID EliSpot Reader software v 6.0; AID GmbH, Strassberg, Germany) and reported as frequency per million cells.
Histopathology. For histopathological analysis, the middle right lobe of each infected murine lung was harvested and then perfused with a 10% neutral buffered formalin solution. Tissue samples were embedded in paraffin, and 5 μm thick tissue sections were collected and stained with hematoxylin and eosin (H&E). Tissue slides were then visualised and observed using a Leica DM6000B microscope (Leica Microsystem©, North Ryde, NSW, Australia) with a magnification of 10× and acquired as mosaics. The average relative area of the granulomatous tissue in the lung was quantified using Adobe Photoshop CC 2015.15 [
15] (Adobe Inc., San Jose, CA, USA). The percentages of granulomatous tissue were calculated as a percent of the total lung section area, as described previously by Nusbaum et al., 2016 [
20].
Statistical analysis. Statistical analysis was performed using GraphPad Prism 10.3.1 software (GraphPad Software, La Jolla, CA, USA). The significance of differences between experimental groups was evaluated by one- or two-way analysis of variance (ANOVA), with a pairwise comparison of multi-grouped data sets achieved using Tukey’s multiple comparisons test and was considered significant when the p values were ≤0.05.
4. Discussion
The selection of novel and optimal antigens for inclusion in subunit vaccines is a critical step in TB vaccine design and development. However, the choice of antigens used in
M. tuberculosis vaccine candidates that are now in clinical trials has been primarily focusing on a small subset of immunodominant proteins, such as antigen 85 complex, ESAT6 or TB10.4 [
22,
23]. In this study, we characterised the CD4
+ T cell immunogenicity and the protective efficacy of a fusion protein consisting of three different protein components, namely Ag85B, CysD and MPT83. The fusion protein, termed CysVac5, was based on CysVac2, the first generation of the CysVac multistage vaccine family whose components are expressed at different stages of the
M. tuberculosis life cycle and provided significant protection against
M. tuberculosis infection in several murine models [
14,
15,
24]. In this study, the antigen MPT83 was included in the CysVac2 recombinant protein to form a novel fusion protein termed CysVac5. MPT83 (Rv2873) is a secreted mycobacterial lipoprotein expressed on the surface of
M. tuberculosis. Although its function is not yet well characterised, this antigen is recognised by T cells after
M. tuberculosis infection in both mice and humans [
18]. When mice were vaccinated with either protein or DNA vaccine, MPT83 conferred significant levels of protection in both lungs and spleen against aerosol
M. tuberculosis infection, with strong IFN-γ CD4
+ and CD8
+ T-cell responses [
18]. With the addition of another immunogenic antigen, MPT83, to the fusion vaccine, it was hypothesised that CysVac5 could provide an improved immune response and protective efficacy against TB infection compared to CysVac2.
The formulation of CysVac5 in a specific liposome construction MPLA/DDA adjuvant resulted in a vaccine that induced polyfunctional CD4
+ T cells secreting multiple cytokines in mice both pre- and post-
M. tuberculosis challenge and a reduced bacterial number in the lungs and spleen. Subcutaneous injection of CysVac5 formulated in MPLA/DDA adjuvant in C57BL/6 mice resulted in the generation of a mix of Th1 and Th17 responses both before and after
M. tuberculosis infection, indicated by high frequencies of CD4
+ T cells producing multiple cytokines (IFN-γ, IL-2, IL-17 and TNF). The induction of T cells producing IFN-γ, IL-2 and TNF (Th1 response) observed in CysVac5-vaccinated mice (
Figure 2) correlates with vaccine-induced Th1-mediated protection, as numerous studies have demonstrated the correlation of these responses with protective immunity in animal models, and evidence is accumulating on the important contribution of IL-17 to anti-TB immunity [
15,
25,
26]. In humans, it has been shown that stronger mycobacteria-specific polyfunctional CD4
+ T cells were found in adults with smear-negative sputum for acid-fast bacilli (AFB) than those with AFB smear-positive TB. In addition, successful TB treatments that rapidly reduce the bacterial load are associated with an increased proportion of this polyfunctional subset. The increased population of PPD-specific CD4
+ T cells was also reported to correlate with the significant inhibition of mycobacterial growth in BCG-vaccinated infants [
27]. On the other hand, several studies have demonstrated that polyfunctional CD4
+ did not correlate with protection and suggested that other functional attributes, such as tissue homing potential and additional effector functions, may play a bigger role in promoting protection [
28]. Thus, further study is required to define the true role of polyfunctional CD4
+ T cells in anti-TB immunity. Further, we did not measure the induction of vaccine-specific antibodies in this study, which may play an accessory role in protection against
M. tuberculosis infection [
29].
Ag85B is a secreted immunogenic antigen that has been used widely in developing subunit protein vaccines for TB. This antigen contains numerous well-characterised epitopes recognised in
M. tuberculosis infection models, as well as in TB patients [
30]. Several studies have shown that Ag85B displays the most robust antigenicity, with an epitope that binds to 13 major histocompatibility complexes [
31]. This might explain the high percentage of polyfunctional CD4
+ T cells observed in both CysVac5- and CysVac2-vaccinated mice following stimulation with Ag85B protein antigen (
Figure 3). Following recall to CysD or MPT83, the immune responses were not as strong as those seen when stimulated with Ag85B. Various factors are to be considered as to why CysD/MPT83 appeared to be less immunogenic, such as how major epitopes/antigenic determinants of these antigens relate to the issue of antigenic competition, optimal antigen presentation and the cellular environment for facilitating antigen-specific T cells [
32]. There are also linked effects of host genetic background and mycobacterial pathogen on infection susceptibility [
33]. Several epidemiological studies have shown that genetic factors have significant contributions to TB disease in humans. Multiple series and extensive experimental studies in various animal models have also established the importance of host genetic background in assessing the outcomes of
M. tuberculosis infection [
34]. Indeed, in this study, CysVac5 vaccination could provide protection against TB infection in mice with different genetic backgrounds. Similar findings were observed in a previous study determining the capacity of CysVac2 in inducing protection against aerosol
M. tuberculosis infection in C57BL/6, BALB/c and outbred mice [
14]. In BALB/c mice, CysVac5 was able to elicit strong mixed Th1/Th17 responses both before and after challenge. This was shown by the increased populations of multifunctional CD4
+ T cells producing IFN-γ, IL-2 and TNF, as well as IL-17, in the PBMCs or lungs of vaccinated mice (
Figure 5). This immunity correlated with protection against
M. tuberculosis infection in both the lungs and the spleens of mice
(Figure 5).