Recently, the efficacy of Mycobacterium bovis bacillus Calmette–Guérin (BCG) vaccination is being reassessed in accordance with the achievements of clinical tuberculosis (TB) vaccine research. However, the mechanisms ultimately determining the success or failure of BCG vaccination to prevent pulmonary TB remain poorly understood. In this study, we analyzed the protective effects of intradermal BCG vaccination by using specific pathogen–free cynomolgus macaques of Asian origin that were intradermally vaccinated with BCG (Tokyo strain) followed by Mycobacterium tuberculosis (Erdman strain) infection. Intradermal BCG administration generated TB Ag-specific multifunctional CD4 T cell responses in peripheral blood and bronchoalveolar lavage and almost completely protected against the development of TB pathogenesis with aggravation of clinical parameters and high levels of bacterial burdens in extrapulmonary organs. However, interestingly, there were no differences in bacterial quantitation and pathology of extensive granulomas in the lungs between BCG-vaccinated monkeys and control animals. These results indicated that the changes in clinical parameters, immunological responses, and quantitative gross pathology that are used routinely to determine the efficacy of TB vaccines in nonhuman primate models might not correlate with the bacterial burden and histopathological score in the lung as measured in this study.

Tuberculosis (TB) is among the top 10 causes of death overall and is the leading cause of death due to infection with a single type of pathogen. In 2018, an estimated 10 million people worldwide had TB, and 1.6 million people died of the disease (1, 2). It is estimated that almost one quarter of the global population, between 2 billion and 3 billion people, has been infected with Mycobacterium tuberculosis and may be at risk for progression to reactive TB (2, 3).

The current incidence of TB indicates that intradermal (i.d.) immunization of Mycobacterium bovis bacillus Calmette–Guérin (BCG) fails to protect sufficiently against pulmonary TB, the major disease manifestation and source of dissemination (4). The protective efficacy of BCG is on average 50% but varies substantially, depending on the geographical location, and is poorer in individuals with previous exposure to mycobacteria. BCG can also cause adverse reactions in immunocompromised individuals (5). However, i.d. vaccination with BCG has contributed to a reduction in infant TB mortality by protecting against extrapulmonary TB (6). In addition, BCG vaccination has been shown to be associated with reduced general childhood mortality by stimulating immune responses (4). There have been recent breakthroughs in BCG vaccination in both preclinical and clinical TB vaccine studies, including the first proof-of-concept study showing that revaccination with BCG can protect adolescents from sustained M. tuberculosis infection (7). Furthermore, it has been shown that pulmonary mucosal BCG vaccination provided prevention from repeatedly exposed rhesus macaques with induction of polyfunctional Th17 cells, IL-10, and IgA in the lung (8). Remarkably, i.v. administration of BCG vaccine provided strong protection against the pathogenesis of TB (9, 10). Therefore, assessment of new strategies for BCG usage in preclinical models to select the best conditions of vaccination before clinical vaccine testing is necessary.

Nonhuman primate (NHP) models of TB might be better than other animal models of TB because NHP models show the whole spectrum of human lesions and because NHPs are the closest species to humans and would provide the opportunity to conduct parallel human and NHP trials. Among NHPs, cynomolgus macaques are an especially good model of TB, and the outcome of infection clinically and pathologically resembles that of human infection (11). The variable course of TB disease is shown in cynomolgus macaques, and they are the only animal model of natural latent TB similar to that in humans (12). Active TB disease and latent infection are observed with equal frequency in both humans and cynomolgus macaques, providing an opportunity to study both aspects of this infection. These infection outcomes, which have been clinically defined, were confirmed by statistically significant differences in gross pathology, bacterial burden, dissemination, and immunologic characteristics (13). The epidemic of HIV–TB coinfection is a major worldwide public health concern, and HIV infection dramatically increases the risk for development of active TB disease (14). We previously studied SIV infection and simian HIV (SHIV) infection in cynomolgus macaques as a model for HIV infection and AIDS (15). A feasible cynomolgus macaque model appears to have some distinct advantages over other animal models as well as providing the possibility for M. tuberculosis–SIV/SHIV coinfection models.

The crucial point is that observation of efficacy for vaccine candidates in an NHP model of M. tuberculosis infection is required to advance the candidates to clinical studies. Specific pathogen–free (SPF) cynomolgus monkey colonies that are suitable for infection/medical research have been maintained in Tsukuba Primate Research Center (TPRC) (16). The cynomolgus macaques in TPRC were obtained from Indonesia, Malaysia and the Philippines and have been bred as pure blood of each origin without interbreed crossing (17).

In the current study, SPF cynomolgus macaques housed in TPRC were used to assess immunogenicity, course of infection, and protection against M. tuberculosis challenge following BCG i.d. immunization. The primary goal of this study was to establish an ideal monkey model of M. tuberculosis infection that reflects the preventive effects of BCG vaccination observed in humans. In our NHP TB model, BCG vaccination could not prevent pulmonary TB, showing an increase of the bacterial burden in the lungs and aggravation of the quality of granulomas, although the vaccination provided complete protection against extrapulmonary dissemination of TB. Comparative studies of BCG-protective effects with cellular immune responses showed that the response magnitude but not coexpression profiles of cytokines by CD4 T cells was involved in the correlation with TB protection in our NHP model.

Twenty healthy, SPF cynomolgus macaques (4–7 y of age, 2.8–4.5 kg) were divided into two groups (BCG vaccination and nonvaccination groups). Randomization was conducted on the basis of sex, age, body weight, and bronchoalveolar lavage (BAL) T cell responses against purified protein derivative (PPD) (Table I). The animals were housed in animal biosafety level 2 (immunization phase) and animal biosafety level 3 (M. tuberculosis challenge phase) facilities at TPRC of National Institutes of Biomedical Innovation, Health and Nutrition (NIBIOHN) and were monitored throughout the study for physical health, food consumption, body weight, body temperature, and serum chemistry. The cynomolgus macaques in TPRC were obtained from Indonesia, Malaysia, and the Philippines (Table I). All of the monkeys in this study were negative for various kinds of viruses (SIV, simian type D retrovirus, simian T cell lymphotropic virus, simian foamy virus, EBV, CMV, and B virus), bacteria (Mycobacterium spp., Shigella, and Salmonella), and an intestinal helminth (16). This study was performed in TPRC, NIBIOHN, after approval by the Committee on the Ethics of Animal Experiments at NIBIOHN in accordance with the guidelines for animal experiments at NIBIOHN. The animals were handled under the supervision of the veterinarians in charge of the animal facility.

The animals were intradermally (i.d.) immunized with 2 × 106 CFU BCG (Tokyo strain; Japan BCG Laboratory, Tokyo, Japan) in a 100-μl volume of saline at week 0 in accordance with the manufacturer’s instructions for human use. Sterile saline was administered to animals in the control group. The skin of the animals was not cleaned with an antiseptic to avoid inactivation of the BCG vaccine. A sterile syringe and a sterile needle were used for each injection.

M. tuberculosis Erdman strain (35801; American Type Culture Collection) was cultured in 300 ml of 7H9 medium containing 0.05% Tween 80 for 3 wk at 37°C. The culture medium was filtrated through 5-μm mesh to remove debris and was divided into two 1-ml tubes (1 ml/tube) and stocked at −80°C. To evaluate CFUs, the M. tuberculosis stock sample was thawed and centrifuged at 3500 rpm for 15 min at 4°C and then washed with 0.05% Tween 80 in PBS. Ten-fold serial dilutions were then prepared on 7H10 Agar Plates. After cultivation for 3 wk at 37°C, titration of the M. tuberculosis stock sample was conducted, and the titer was determined to be 1.66 × 108 CFU/ml.

The M. tuberculosis stock sample (1.66 × 108 CFU/ml) was diluted with PBS containing 0.05% Tween 80 in a glass vial to obtain 40 CFU/ml reconstituted M. tuberculosis challenge solution for 20 monkeys as follows: 1.66 × 102 CFU/ml solution was generated from six 10-fold serial dilutions of an M. tuberculosis stock sample (a 0.9-ml portion of the M. tuberculosis stock sample was transferred to the next tube containing 8.1 ml of PBS with 0.05% Tween 80 and mixed well; this step was repeated six times). The above-prepared–diluted solution (1 ml) was added to 3.15 ml of PBS (plus 0.05% Tween 80) to obtain 40 CFU/ml challenge solution.

Intratracheal administration was performed in a biosafety cabinet with each animal under anesthesia (7.5 mg/kg ketamine HCL and 3 mg/kg xylazine). A tracheal tube was inserted in the trachea as a guide vessel and an indwelling feeding tube (catheter) was inserted along the tracheal tube. The M. tuberculosis challenge solution (40 CFU in 1-ml volume) was inoculated via the tracheal tube, followed by 1.5 ml of PBS for tube washing. The infection dose was estimated by plating an aliquot of inoculum on 7H11 Agar Plates and enumerating CFU after 4 wk of incubation. Infection doses for the animals in this study ranged from 35 to 50 CFU.

Whole blood (1.6 ml) was collected into a 3.8% sodium citrate vacuum blood collection tube (Terumo, Tokyo, Japan) and transferred to a capillary tube using an erythrocyte sedimentation rate (ESR) system stand (Terumo). The blood was settled under the influence of gravity, and the number of millimeters of clear plasma present at the top of the capillary after 1 h (millimeters per hour) was measured.

Whole blood was collected into a blood collection tube (plain, 5 ml) and was left to stand for at least 30 min at room temperature prior to centrifugation at 1200 × g for 10 min. The C-reactive protein (CRP) values of serum samples were determined using a calibrated Fuji DRI-CHEM system (FUJIFILM, Tokyo, Japan).

Each animal was placed under anesthesia (7.5 mg/kg ketamine HCl and 3 mg/kg xylazine) and positioned on a slider table through an x-ray detector ring. Computed tomography (CT) and x-ray images were obtained (80 kV; 400 CT μA; field of view: 220 mm; scan time: standard 18 s or respiratory-gated 8 min). All CT scans were performed in a biosafety level 3 facility using a three-dimensional micro-CT scanner system (Cosmo Scan CT AX; Rigaku, Tokyo, Japan). Cosmo Scan CT Viewer was used for analysis, and the initial step is to extract the lung from the chest CT volume. Manual segmentation of the lungs was performed by trained veterinarians. Slices of the third, sixth, and ninth thoracic vertebrae (including the upper, middle, and lower lung areas, respectively) were selected. These sample slices, taken in the axial plane, were digitized and quantified to pixels as white spot lesions using ImageJ software, which is a public domain Java image processing program inspired by NIH Image. The increase in pixels of lesions in each axial section was observed at baseline and every 2 wk after TB infection (Supplemental Fig. 1). The average number of pixels in each axial CT image before TB infection was set to 1, and the deterioration of the lesion was recorded (Fig. 1C). For the calculation of chest CT volume of air-filled structures, such as healthy parenchyma, automatic lung segmentation and three-dimensional rendering of the normal lung structures were performed using VolumeAnalysis (Rigaku) software (Supplemental Fig. 2).

Immune assays of BAL were conducted using freshly harvested samples. BAL was obtained by collecting two consecutive 20-ml washes with room temperature saline using a feeding tube. Samples were resuspended in warm R10 medium (RPMI 1640 with 2 mM l-glutamine, 100 U/ml penicillin, 100 g/ml streptomycin, and 10% heat-inactivated FBS) containing 2 U/ml DNase I (Roche, Basel, Switzerland). BAL cells were rested before overnight (18 h) stimulation with tuberculin PPD (Japan BCG Laboratory). PBMCs were batch-analyzed from cryopreserved samples at the completion of the study. PBMCs were isolated from heparin-anticoagulated whole blood with lymphocyte separation medium using standard methods. For immune analysis, PBMCs were thawed in a 37°C water bath and rested in R10 medium at 37°C overnight before 6-h stimulation with PPD. For in vitro T cell cytokine assays, cells were cultured with PPD at 10 μg/ml or overlapping peptide pools (15-mer peptides overlapping by 11 aa) for Ag85B or with early secreted antigenic target of 6 kDa (ESAT-6) or mycobacterial DNA-binding protein 1 (MDP1) or 10-kDa culture filtrate protein (CFP10) at 2 μg/ml. Brefeldin A (BD Biosciences) was added 2 h after stimulation with Ags at a final concentration of 10 μg/ml. For flow cytometric analysis, cells were first incubated with LIVE/DEAD Fixable Aqua ECD Dead Cell Stain (Life Technologies, Carlsbad, CA) and then immunostained for cell surface markers (CD3, Alexa Fluor 700 [BD Biosciences, Franklin Lakes, NJ]; CD4, FITC; and CD8, and PE-Cy7 [BioLegend, San Diego, CA]). Intracellular staining of fixed and permeabilized (BD Cytofix/Cytoperm; BD Biosciences) samples was performed using the following Abs: IFN-γ, PE; IL-2, PerCP-Cy5.5 (BD Biosciences); and TNF, allophycocyanin, IL-17, allophycocyanin–H7, Foxp3, PE (BioLegend). Flow cytometric data were acquired using a BD FACSCanto II flow cytometer and analyzed using FlowJo software (BD Biosciences). All Ag-specific cytokine frequencies were reported after background subtraction of the frequency of an identically gated population of cells from the same samples incubated without Ag stimulation. Background responses in BAL were typically greater than those observed in PBMCs. In all analyses, gating on the lymphocyte population was followed by the separation of viable CD3+ T cell subset and progressive gating on the CD4+ and CD8+ T cell subsets. Ag-responding cells in both CD4+ and CD8+ T cell populations were determined by their intracellular expression of cytokines described above. To determine innate immune cells in BAL, viable BAL cells were followed by immunostaining for cell surface markers (CD3, Alexa Fluor 700 [BD Biosciences]; CD4, FITC or allophycocyanin; CD8, PE-Cy7; γδTCR, PerCP-Cy5.5; TCR Vα7.2, PE-Cy7 [BioLegend]; or CD159a (NKG2A), FITC [Miltenyi Biotec]). Intracellular staining of fixed and permeabilized samples was performed using the following Abs: Granzyme B, FITC [BioLegend].

PBMC ELISPOT (U-CyTech biosciences, Utrecht, the Netherlands) for Mycobacterium-specific IFN-γ was performed on the experimental monkeys. In monkeys given BCG, the frequency of PPD-specific IFN-γ–producing T cells was determined at baseline and 3, 9, 12, and 14 wk after BCG administration. Each ELISPOT also included a positive control (Con A, 5 μg/ml) or a negative control (no Ag stimulation, medium only). The media background was subtracted from the Ag-specific response at each time point. Investigation of BCG-induced cytokine responses was undertaken using cryopreserved PBMCs.

At the scheduled end point, necropsy that included gross pathologic scoring as well as histologic analysis and bacterial quantitation was performed. The necropsies were done by trained pathologists. Gross pathology scoring, a quantitative measure of TB disease, is based on careful analysis of the numbers and sizes of nodules or other pathological examinations in each lung lobe (on the surface and parenchyma), mediastinal lymph node (LN), and extrapulmonary sites (Supplemental Fig. 3). In the histopathological analysis, selected samples with gross lesions from whole tissue sections of the lung and other organs were microscopically examined. Microscopic histopathologic assessment was reviewed by three certified pathologists with a specific focus on granuloma characteristics, including overall architectural appearance, number and type of granuloma (nonnecrotizing, caseous, coalescent caseous) and tissue/cellular composition between and within monkey groups (Table II). A 0–4 scoring system for each of these criteria was applied (Supplemental Fig. 4). A histopathology scoring system was implemented using a semiquantitative grading system in which granulomas were assigned a numeric point value based on prevalence: (0, no change [no granuloma]; 1, minimal [1 or two nonnecrotizing granulomas present]; 2, mild [more than three nonnecrotizing granulomas present]; 3, moderate [caseous granulomas present]; 4, severe [coalescent caseous granulomas present]).

Bacterial burden was quantified at necropsy. Whole lungs, liver, spleen, and kidneys were sampled for the presence of viable M. tuberculosis post mortem, and the weights were determined. To collect fragments of the same sizes from the same places of the organs of each animal, a template containing evenly spaced 2-cm square holes was randomly placed over the whole organ. Tissue samples were selected randomly and collected through the template using a biopsy punch instrument (5 mm; Kaijirushi, Tokyo, Japan). The weights of all samples were determined. Tissue samples were dissociated and serially diluted in distilled water. Ten-fold serial dilutions were plated in duplicate on 7H10 Agar Plates and incubated at 37°C, and CFU were counted 3 and 5 wk later. These methods were based on previous studies (18, 19).

At necropsy, nodules in the lung were harvested, and the size of each sample was measured. To determine bacterial burden and sterility of granulomas, 8–10 nodules of various sizes were randomly selected from each lung lobe of an animal. In total, 93 (control) and 98 (BCG) granulomas were analyzed. Samples were homogenized, plated on 7H10 Agar Plates, and incubated for 6 wk. Sterile lesions were defined as lesions with no growth after 6 wk, as previously described (13).

Statistical analyses were conducted in GraphPad Prism 7 (GraphPad Software, San Diego, CA). For cytokine frequencies, statistical significance was determined using a two-tailed Wilcoxon rank-sum test using Prism software. For pie charts, significance was determined by two-tailed Student t test using SPICE software. Clinical parameters, bacterial burdens, and ratios of cytokine-producing cells were evaluated by the Mann–Whitney U test using Prism.

Animal experiments were performed in TPRC, NIBIOHN in accordance with the Guidelines for Animal Use and Experimentation to minimize animal pain and suffering, as set out by NIBIOHN. The protocol was approved by the Animal Welfare and Animal Care Committee of NIBIOHN (Permit Number DS23-8R2).

Animals were vaccinated at week 14 with BCG (n = 10) or saline (n = 10), and all of the 20 animals were infected with M. tuberculosis at week 0. A flare at the injection site was detected in BCG-inoculated monkeys but not in the saline group (Tables I, II). A diagnosis of M. tuberculosis infection was made in all monkeys by the IFN-γ release assay (QuantiFERON TB Gold In-Tube; QIAGEN, Hilden, Germany) (20) and Eiken TB-LAMP assay (Eiken Chemical, Tokyo, Japan) (21). The experimental schedule is summarized in Fig. 1A. To determine whether the BCG vaccine-elicited responses conferred protection against challenge, M. tuberculosis Erdman strain was instilled intratracheally, and clinical parameters were monitored. BCG inoculation suppressed progressive body weight loss compared with that in the nonvaccinated group (Fig. 1B). CT follow-up of indeterminate pulmonary nodules was carried out for each animal to determine changes in granulomas and/or inflammatory areas as the common strategy used in clinical settings. The animals received CT scans at 2 wk intervals after TB infection. The positive areas of granulomas and inflammation in animals that received BCG at 10 wk postinfection were smaller than those in nonvaccinated animals in the middle lung area (Fig. 1C). In addition, to improve the accuracy of CT analysis, changes in air-like voxels, such as healthy parenchyma, after TB infection were calculated using automatic lung segmentation as shown in Supplemental Fig. 2A. The volume of normal area in the lung from control animals was on a declining trend, whereas a decrease in the healthy lung region seemed to be prevented in BCG-vaccinated monkeys (Supplemental Fig. 2B). With respect to other disease-related clinical parameters, ESR and CRP were analyzed as markers for inflammation. With the exception of ESR in two animals and CRP in three animals, ESR and CRP levels in the unvaccinated group became elevated as clinical disease advanced at week 4 (Fig. 1D, Table III). BCG-vaccinated animals showed a clear effect, with only one ESR- and CRP-positive animal (number 071) (Fig. 1D, Table III). The results showed that BCG vaccination significantly reduced TB-associated inflammation as determined by measurements of ESR and CRP levels, although there was no significant difference in weight loss or CT value between BCG-vaccinated animals and unvaccinated animals.

Table I.
Monkey list
Animal No.Monkey IdentifierAgeWeight (kg)CD4+IFN-γ+ (%)aOrigin
Control      
 003 1321010074 5 y 9 mo (69 mo) 3.545 0.6 Mixb 
 027 1321010073 5 y9 mo (69 mo) 3.77 2.26 Malaysia 
 063 1321207028 4 y (48 mo) 2.78 0.92 Mixc 
 064 1421103028 5 y 4 mo (64 mo) 3.085 0.76 Mixd 
 066 1321204016 4 y 3 mo (51 mo) 2.995 0.7 Malaysia 
 068 1521107004 5 y (60 mo) 3.61 0.28 Mixd 
 072 1421206050 4 y 1 mo (49 mo) 3.995 0.74 The Philippines 
 073 1421105041 5 y 2 mo (62 mo) 4.04 0.39 Indonesia 
 074 1321108043 4 y 11 mo (59 mo) 4.545 0.87 Indonesia 
 076 1421102009 5 y 5 mo (65 mo) 3.975 0.73 Malaysia 
 Average  59.6 mo 3.63  0.83  
BCG      
 001 1321011084 5 y 8 mo (68 mo) 3.475 0.48 Mixe 
 002 1321010071 5 y 9 mo (69 mo) 4.29 1.26 Malaysia 
 004 1321009062 5 y 10 mo (70 mo) 3.425 3.57 Mixc 
 062 1421204023 4 y 3 mo (51 mo) 2.915 0.48 Mixd 
 065 1320912114 6 y 7 mo (79 mo) 4.315 0.42 Malaysia 
 067 1421206052 4 y 1 mo (49 mo) 3.71 0.44 Mixd 
 069 1421206040 4 y 1 mo (49 mo) 3.555 0.4 The Philippines 
 070 1421112101 4 y 7 mo (55 mo) 3.575 0.46 Indonesia 
 071 1421201006 4 y 6 mo (54 mo) 3.9 0.77 Indonesia 
 075 1421107060 5 y (60 mo) 3.375 0.7 Malaysia 
 Average  60.4 mo 3.65 0.9  
Animal No.Monkey IdentifierAgeWeight (kg)CD4+IFN-γ+ (%)aOrigin
Control      
 003 1321010074 5 y 9 mo (69 mo) 3.545 0.6 Mixb 
 027 1321010073 5 y9 mo (69 mo) 3.77 2.26 Malaysia 
 063 1321207028 4 y (48 mo) 2.78 0.92 Mixc 
 064 1421103028 5 y 4 mo (64 mo) 3.085 0.76 Mixd 
 066 1321204016 4 y 3 mo (51 mo) 2.995 0.7 Malaysia 
 068 1521107004 5 y (60 mo) 3.61 0.28 Mixd 
 072 1421206050 4 y 1 mo (49 mo) 3.995 0.74 The Philippines 
 073 1421105041 5 y 2 mo (62 mo) 4.04 0.39 Indonesia 
 074 1321108043 4 y 11 mo (59 mo) 4.545 0.87 Indonesia 
 076 1421102009 5 y 5 mo (65 mo) 3.975 0.73 Malaysia 
 Average  59.6 mo 3.63  0.83  
BCG      
 001 1321011084 5 y 8 mo (68 mo) 3.475 0.48 Mixe 
 002 1321010071 5 y 9 mo (69 mo) 4.29 1.26 Malaysia 
 004 1321009062 5 y 10 mo (70 mo) 3.425 3.57 Mixc 
 062 1421204023 4 y 3 mo (51 mo) 2.915 0.48 Mixd 
 065 1320912114 6 y 7 mo (79 mo) 4.315 0.42 Malaysia 
 067 1421206052 4 y 1 mo (49 mo) 3.71 0.44 Mixd 
 069 1421206040 4 y 1 mo (49 mo) 3.555 0.4 The Philippines 
 070 1421112101 4 y 7 mo (55 mo) 3.575 0.46 Indonesia 
 071 1421201006 4 y 6 mo (54 mo) 3.9 0.77 Indonesia 
 075 1421107060 5 y (60 mo) 3.375 0.7 Malaysia 
 Average  60.4 mo 3.65 0.9  
a

The percentage of IFN-γ–producing CD4 T cells induced by PPD stimulation in BAL from preimmunized animals.

b

Mixed with Indonesia, Malaysia, and the Philippines.

c

Mixed with Malaysia and the Philippines.

d

Mixed with Indonesia and Malaysia.

e

Mixed with Indonesia and the Philippines.

Table II.
The number and histopathological score of granulomas
Control Animal No.
Mean ScoreBCG Animal No.
Mean Scorep Value
003027063064066068072073074076001002004062065067069070071075
Lung                        
 Nonnecrotizing >10 >10 >10 >10 >20 >10 >10  >10   
 Caseous granuloma 19 40 15 16 22 16 
 Coalescent caseous 
 Score 3.9 3.4 0.1192 
Hilar LN                        
 Nonnecrotizing >10 >10 >10 >10 >10 >10  >10 >10 >10   
 Caseous granuloma 15 11 
 Coalescent caseous 
 Score 2.2a 0.0001 
Hilar region                        
 Nonnecrotizing    
 Caseous granuloma 
 Coalescent caseous 
 Score 1.6 0.2a 0.0341 
Spleen                        
 Nonnecrotizing >10 >10 >10 >10 >10    
 Caseous granuloma 16 
 Coalescent caseous 
 Score 1.7 0.0031 
Liver                        
 Nonnecrotizing >10 >20  >10   
 Caseous granuloma 19 
 Coalescent caseous 
 Score 1.5 0.3 0.0573 
Kidney                        
 Nonnecrotizing >20    
 Caseous granuloma 
 Coalescent caseous 
 Score 1.3 0.1a 0.0325 
Control Animal No.
Mean ScoreBCG Animal No.
Mean Scorep Value
003027063064066068072073074076001002004062065067069070071075
Lung                        
 Nonnecrotizing >10 >10 >10 >10 >20 >10 >10  >10   
 Caseous granuloma 19 40 15 16 22 16 
 Coalescent caseous 
 Score 3.9 3.4 0.1192 
Hilar LN                        
 Nonnecrotizing >10 >10 >10 >10 >10 >10  >10 >10 >10   
 Caseous granuloma 15 11 
 Coalescent caseous 
 Score 2.2a 0.0001 
Hilar region                        
 Nonnecrotizing    
 Caseous granuloma 
 Coalescent caseous 
 Score 1.6 0.2a 0.0341 
Spleen                        
 Nonnecrotizing >10 >10 >10 >10 >10    
 Caseous granuloma 16 
 Coalescent caseous 
 Score 1.7 0.0031 
Liver                        
 Nonnecrotizing >10 >20  >10   
 Caseous granuloma 19 
 Coalescent caseous 
 Score 1.5 0.3 0.0573 
Kidney                        
 Nonnecrotizing >20    
 Caseous granuloma 
 Coalescent caseous 
 Score 1.3 0.1a 0.0325 

The number of each type of granulomas (nonnecrotizing, caseous, coalescent caseous granulomas). Histopathology scoring system: granulomas were scored as described in Supplemental Fig. 4.

a

p < 0.05 compared with control group. The p values were determined from Mann–Whitney U test.

FIGURE 1.

Clinical outcome after M. tuberculosis challenge. (A) Overview of the experimental design. Blood collection and clinical assessments were performed at the indicated intervals. Animals were followed for up to 12 wk postinfection. Monkeys were infected intratracheally with M. tuberculosis, 40 CFU (range, 35–50 CFU), 14 wk after BCG vaccination. (B) Average body weight change for each group with SD at the indicated weeks postchallenge. (C) CT analysis in BCG-vaccinated and nonvaccinated monkeys after TB challenge. The CT x-ray photography conditions were x-ray 80 kV, 400 μA, field of view 220 mm, and breathing/cardiac synchronized. From each axial slice, images located in the third, sixth, and ninth thoracic vertebrae; the average of lesion pixels from each slice was quantified by ImageJ. (D) ESR was determined in whole blood from individual animals up to 12 wk postinfection. Each line represents serial ESR results of an individual monkey in the nonvaccinated control group (D, left) or BCG group (D, right).

FIGURE 1.

Clinical outcome after M. tuberculosis challenge. (A) Overview of the experimental design. Blood collection and clinical assessments were performed at the indicated intervals. Animals were followed for up to 12 wk postinfection. Monkeys were infected intratracheally with M. tuberculosis, 40 CFU (range, 35–50 CFU), 14 wk after BCG vaccination. (B) Average body weight change for each group with SD at the indicated weeks postchallenge. (C) CT analysis in BCG-vaccinated and nonvaccinated monkeys after TB challenge. The CT x-ray photography conditions were x-ray 80 kV, 400 μA, field of view 220 mm, and breathing/cardiac synchronized. From each axial slice, images located in the third, sixth, and ninth thoracic vertebrae; the average of lesion pixels from each slice was quantified by ImageJ. (D) ESR was determined in whole blood from individual animals up to 12 wk postinfection. Each line represents serial ESR results of an individual monkey in the nonvaccinated control group (D, left) or BCG group (D, right).

Close modal
Table III.
CRP
Animal No.Weeks Post–M. tuberculosis Infection
0246812
Control       
 003 0.05 0.06 3.96 2.01 3.23 3.63 
 027 0.06 0.06 0.12 0.07 0.07 0.09 
 063 0.11 0.07 2.91 0.07 0.12 2.69 
 064 0.06 0.86 0.16 0.12 0.25 0.4 
 066 0.08 0.11 1.41 0.08 0.13 0.14 
 068 0.06 0.54 2.77 2.83 3.79 3.64 
 072 0.05 0.06 1.25 0.04 0.04 0.05 
 073 0.06 0.06 0.47 0.36 0.06 0.06 
 074 0.06 0.06 0.1 1.15 3.18 3.67 
 076 0.13 0.11 0.28 0.15 0.25 0.22 
BCG       
 001 0.05 0.07 0.06 0.03 0.05 0.07 
 002 0.06 0.07 0.07 0.04 0.04 0.06 
 004 0.06 0.06 0.07 0.05 0.07 0.06 
 062 0.07 0.09 0.1 0.07 0.07 0.17 
 065 0.07 0.09 0.07 0.05 0.05 0.06 
 067 0.07 0.08 0.09 0.04 0.05 0.06 
 069 0.07 0.07 0.07 0.05 0.05 0.06 
 070 0.06 0.06 0.07 0.04 0.04 0.06 
 071 0.17 0.19 0.77 2.11 3.07 3.98 
 075 0.07 0.07 0.08 0.11 0.08 0.17 
Animal No.Weeks Post–M. tuberculosis Infection
0246812
Control       
 003 0.05 0.06 3.96 2.01 3.23 3.63 
 027 0.06 0.06 0.12 0.07 0.07 0.09 
 063 0.11 0.07 2.91 0.07 0.12 2.69 
 064 0.06 0.86 0.16 0.12 0.25 0.4 
 066 0.08 0.11 1.41 0.08 0.13 0.14 
 068 0.06 0.54 2.77 2.83 3.79 3.64 
 072 0.05 0.06 1.25 0.04 0.04 0.05 
 073 0.06 0.06 0.47 0.36 0.06 0.06 
 074 0.06 0.06 0.1 1.15 3.18 3.67 
 076 0.13 0.11 0.28 0.15 0.25 0.22 
BCG       
 001 0.05 0.07 0.06 0.03 0.05 0.07 
 002 0.06 0.07 0.07 0.04 0.04 0.06 
 004 0.06 0.06 0.07 0.05 0.07 0.06 
 062 0.07 0.09 0.1 0.07 0.07 0.17 
 065 0.07 0.09 0.07 0.05 0.05 0.06 
 067 0.07 0.08 0.09 0.04 0.05 0.06 
 069 0.07 0.07 0.07 0.05 0.05 0.06 
 070 0.06 0.06 0.07 0.04 0.04 0.06 
 071 0.17 0.19 0.77 2.11 3.07 3.98 
 075 0.07 0.07 0.08 0.11 0.08 0.17 

CRP: milligrams per deciliter.

The main indicator for testing the potency of TB vaccines in animal models is generally reduction of the bacterial burden in organs at the acute phase the infection. To determine whether the BCG-elicited responses conferred protection against pulmonary TB and extrapulmonary dissemination, we analyzed the mean pathology score and bacterial burden in the lungs and other organs. At necropsy, nonvaccinated monkeys showed worse disease pathology (Table IV) and higher total bacterial burdens in the lungs, spleens, livers, kidneys, and mediastinal LN (Fig. 2A). From gross pathology data, BCG significantly prevented exacerbation of TB pathogenesis (Table IV). Although there was a clear protective effect of BCG in bacterial CFU cultured at necropsy from extrapulmonary organs, the pulmonary disease burden of BCG-vaccinated monkeys was not improved compared with that of unvaccinated controls (Fig. 2A). To determine whether BCG vaccination impacted the formation, severity, and sterilization of granulomas, we next assessed the histopathology and bacterial burden in individual granulomas at 12 wk postinfection. Although BCG vaccination showed strong suppression of the formation and severity of granulomas in extrapulmonary and hilar regions, the histopathological score (Table II) and bacterial burden of individual granulomas in the lung (Fig. 2B) showed no significant differences between BCG-vaccinated and control monkeys. Interestingly, despite the fact that BCG vaccination had a suppressive effect on the formation of white nodules (granulomas) in the lung, there was no apparent difference in granuloma severity compared with that in unvaccinated animals (Table II). Although correlation tests between bacterial burden in lung and data from gross and histopathological analyses were performed to search for correlates of protection, no factors correlated with lung CFU (Fig. 2C–F). In contrast, BCG vaccination resulted in significant prevention of the formation of abdominal granulomas as well as almost complete prevention of curatable bacteria in extrapulmonary organs (Fig. 2A, Table II).

Table IV.
Gross pathology score
Control Animal No.
003027063064066068072073074076Mean
Lung            
 Organ/body weight ratio 0.73% 1.10% 1.15% 1.06% 0.81% 3.46% 0.70% 0.87% 2.42% 0.57% 1.29% 
 White nodule 3.2 
Spleen          
 Organ/body weight ratio 0.14% 0.14% 0.14% 0.11% 0.13% 0.67% 0.13% 0.12% 0.19% 0.07% 0.18% 
 White nodule 2.4 
Liver           
 Organ/body weight ratio 2.96% 1.93% 2.29% 2.31% 1.93% 6.10% 1.66% 1.65% 2.46% 1.82% 2.51% 
 White nodule 2.6 
Kidney           
 Organ/body weight ratio 0.36% 0.40% 0.42% 0.46% 0.37% 1.09% 0.36% 0.36% 0.49% 0.46% 0.48% 
 White nodule 1.5 
Heart           
 Organ/body weight ratio 0.61% 0.70% 0.66% 0.75% 0.62% 0.51% 0.62% 0.71% 0.81% 0.57% 0.66% 
 White nodule 
Control Animal No.
003027063064066068072073074076Mean
Lung            
 Organ/body weight ratio 0.73% 1.10% 1.15% 1.06% 0.81% 3.46% 0.70% 0.87% 2.42% 0.57% 1.29% 
 White nodule 3.2 
Spleen          
 Organ/body weight ratio 0.14% 0.14% 0.14% 0.11% 0.13% 0.67% 0.13% 0.12% 0.19% 0.07% 0.18% 
 White nodule 2.4 
Liver           
 Organ/body weight ratio 2.96% 1.93% 2.29% 2.31% 1.93% 6.10% 1.66% 1.65% 2.46% 1.82% 2.51% 
 White nodule 2.6 
Kidney           
 Organ/body weight ratio 0.36% 0.40% 0.42% 0.46% 0.37% 1.09% 0.36% 0.36% 0.49% 0.46% 0.48% 
 White nodule 1.5 
Heart           
 Organ/body weight ratio 0.61% 0.70% 0.66% 0.75% 0.62% 0.51% 0.62% 0.71% 0.81% 0.57% 0.66% 
 White nodule 
BCG Animal No.
001002004062065067069070071075Meanp Value
Lung             
 Organ/body weight ratio 0.53% 0.50% 0.71% 0.64% 0.56% 0.48% 0.49% 0.59% 1.75% 0.70% 0.70%a 0.0042 
 White nodule 1.9a 0.0044 
Spleen             
 Organ/body weight ratio 0.09% 0.09% 0.11% 0.07% 0.07% 0.12% 0.17% 0.10% 0.18% 0.11% 0.11% 0.0642 
 White nodule 0.2a 0.0007 
Liver             
 Organ/body weight ratio 1.67% 1.61% 2.31% 1.50% 1.69% 1.62% 1.75% 1.81% 1.89% 1.82% 1.77%a 0.0219 
 White nodule 0.3a 0.0011 
Kidney             
 Organ/body weight ratio 0.38% 0.34% 0.41% 0.32% 0.33% 0.31% 0.34% 0.33% 0.39% 0.34% 0.35%a 0.0035 
 White nodule 0.3a 0.0094 
Heart             
 Organ/body weight ratio 0.65% 0.67% 0.66% 0.59% 1.12% 0.67% 0.79% 0.55% 0.67% 0.67% 0.70% 0.6438 
 White nodule >0.9999 
BCG Animal No.
001002004062065067069070071075Meanp Value
Lung             
 Organ/body weight ratio 0.53% 0.50% 0.71% 0.64% 0.56% 0.48% 0.49% 0.59% 1.75% 0.70% 0.70%a 0.0042 
 White nodule 1.9a 0.0044 
Spleen             
 Organ/body weight ratio 0.09% 0.09% 0.11% 0.07% 0.07% 0.12% 0.17% 0.10% 0.18% 0.11% 0.11% 0.0642 
 White nodule 0.2a 0.0007 
Liver             
 Organ/body weight ratio 1.67% 1.61% 2.31% 1.50% 1.69% 1.62% 1.75% 1.81% 1.89% 1.82% 1.77%a 0.0219 
 White nodule 0.3a 0.0011 
Kidney             
 Organ/body weight ratio 0.38% 0.34% 0.41% 0.32% 0.33% 0.31% 0.34% 0.33% 0.39% 0.34% 0.35%a 0.0035 
 White nodule 0.3a 0.0094 
Heart             
 Organ/body weight ratio 0.65% 0.67% 0.66% 0.59% 1.12% 0.67% 0.79% 0.55% 0.67% 0.67% 0.70% 0.6438 
 White nodule >0.9999 

White nodule scoring system: the numbers and types of white nodules were scored as described in Supplemental Fig. 2.

a

p < 0.05 compared with control group. The p values were determined from Mann–Whitney U test.

FIGURE 2.

Bacterial burden data. (A) M. tuberculosis CFU in lungs, spleens, livers, kidneys, and mediastinal LN for each animal at the time of necropsy. Median (black bar) is indicated for each group. Circles and crosses represent individual monkeys. (B, left) Determination of sterility of granulomas from the lungs in BCG-treated monkeys. Each point represents the M. tuberculosis in CFU per one granuloma in the lung. Ten granulomas were collected from the lungs of each monkey. Bars represent median. (B, right) The number of granulomas in BCG-vaccinated animals (n = 10 macaques, 98 granulomas) is shown with the proportion of sterile granulomas shown in red compared with those in nonvaccinated animals (n = 10 macaques, 93 granulomas). The proportion of nonsterile granulomas is indicated in blue. Up to 10 granulomas (randomized) were represented per animal to reduce bias. (CF) No significant correlation between protection and pathology in lung. The pathology: (C) gross pathology score, (D) histopathological score, (E) the number of caseous granuloma, (F) organ/body weight ratio are plotted against bacterial burden in lung for all individual monkeys (n = 20). Red circles indicate monkey 071, which was an ESR- and CRP-positive animal.

FIGURE 2.

Bacterial burden data. (A) M. tuberculosis CFU in lungs, spleens, livers, kidneys, and mediastinal LN for each animal at the time of necropsy. Median (black bar) is indicated for each group. Circles and crosses represent individual monkeys. (B, left) Determination of sterility of granulomas from the lungs in BCG-treated monkeys. Each point represents the M. tuberculosis in CFU per one granuloma in the lung. Ten granulomas were collected from the lungs of each monkey. Bars represent median. (B, right) The number of granulomas in BCG-vaccinated animals (n = 10 macaques, 98 granulomas) is shown with the proportion of sterile granulomas shown in red compared with those in nonvaccinated animals (n = 10 macaques, 93 granulomas). The proportion of nonsterile granulomas is indicated in blue. Up to 10 granulomas (randomized) were represented per animal to reduce bias. (CF) No significant correlation between protection and pathology in lung. The pathology: (C) gross pathology score, (D) histopathological score, (E) the number of caseous granuloma, (F) organ/body weight ratio are plotted against bacterial burden in lung for all individual monkeys (n = 20). Red circles indicate monkey 071, which was an ESR- and CRP-positive animal.

Close modal

An ESR- and CRP-positive monkey (071) showed complete protection from miliary TB despite having the highest bacterial burden in the lung (Fig. 2A). From gross pathology observations, increases in the organ/body weight ratio of the lung and the severity of granulomas in hilar LNs were confirmed in monkey 071 despite the low organ/body weight ratio of the lung and the low degree of severity of granulomas in other BCG-immunized animals. At the lesion level, there was a broad and overlapping distribution of CFU per granuloma from monkey 071 as in other BCG-vaccinated animals (Fig. 2B). These results suggested that i.d. immunization of BCG was not involved in the basic remedy for pulmonary TB induced by a high-titer M. tuberculosis challenge and the induction of sterilization of granulomas. From the data for BCG-vaccinated animals, granuloma severity of hilar LNs might be involved in deterioration of TB pathogenesis.

To determine whether BCG vaccination elicits sufficient and protective TB-specific T cell cytokine responses in the lung, we investigated the magnitude and quality of Ag-specific T cell responses in PBMCs from BCG-vaccinated monkeys. The magnitude of Ag-specific memory T cell responses was assessed as the total percentage of IFN-γ–, IL-2–, or TNF-producing cells. These cytokines were chosen on the basis of their importance in controlling TB infection (IFN-γ, TNF) and ability to augment T cell expansion (IL-2). In PBMCs, the ratio of PPD-specific CD4+ T cells gradually increased after BCG inoculation (Fig. 3A, 3B). The frequency of PPD-specific CD8+ T cells was relatively low compared with that of CD4 T cells in PBMCs following BCG vaccination and was significantly different between the BCG group and unvaccinated animals at 9 wk post-BCG vaccination (Fig. 3C). It has been suggested that the quality (cytokine expression profile) of the BCG-elicited T cell response has implications for disease outcome in TB and other infections that require T cells for protection. Thus, we assessed the quality of CD4+ T cell responses in PBMCs prior to challenge (Fig. 3D). Approximately 75% of the CD4+ T cell cytokine responses in the BCG group produced IFN-γ, IL-2, and TNF simultaneously, whereas CD8+ T cell responses were far less multifunctional, with 50% of responses producing only IFN-γ (data not shown). Interestingly, the expression levels of PPD-specific cytokines by CD4+ T cells but not those by CD8+ T cells in blood from an ESR- and CRP-positive animal (071) were higher than those in the other BCG-vaccinated monkeys, but there was no difference in the quality of T cell responses. In addition, the PPD-specific cellular immune response observed in immunized monkeys was exhibited in PBMCs from monkeys vaccinated with BCG but not in those that received saline, and a robust immune response was detected in PBMCs from monkey 071 (Fig. 3E). These results suggested that the clinical parameters in blood after TB challenge reflected the quantity but not the quality of CD4+ T cell responses against TB Ags in blood. Although BCG expresses Ag85B and MDP1, CD4+ and CD8+ T cell responses were not significantly different between the BCG-vaccinated and nonimmunized groups of monkeys after stimulation with overlapping peptides of Ag85B or MDP1 (Fig. 3F).

FIGURE 3.

Prechallenge cellular immune responses in PBMCs. Percentages of CD4 (A and C) and CD8 (B) T cells in PBMCs producing IFN-γ, IL-2, or TNF in response to PPD (A–D), Ag85B peptides or MDP1 peptides (F) were determined by flow cytometry at a preimmune (0 wk) time point and at 3 and 9 wk after BCG vaccination. The black line indicates the median of individual responses. (D) Quality of the T cell responses in (A) 9 wk after BCG vaccination. Pie charts show the fraction of total cytokine response comprising any combination of IFN-γ, IL-2, or TNF. Pies (average of n = 10) are shown for groups with measurable responses. (E) In vitro ELISPOT analysis of IFN-γ–releasing cells in PBMCs was performed at 9 wk after BCG immunization. The percentage of BCG-induced IFN-γ cells was determined after 24 h of PPD stimulation.

FIGURE 3.

Prechallenge cellular immune responses in PBMCs. Percentages of CD4 (A and C) and CD8 (B) T cells in PBMCs producing IFN-γ, IL-2, or TNF in response to PPD (A–D), Ag85B peptides or MDP1 peptides (F) were determined by flow cytometry at a preimmune (0 wk) time point and at 3 and 9 wk after BCG vaccination. The black line indicates the median of individual responses. (D) Quality of the T cell responses in (A) 9 wk after BCG vaccination. Pie charts show the fraction of total cytokine response comprising any combination of IFN-γ, IL-2, or TNF. Pies (average of n = 10) are shown for groups with measurable responses. (E) In vitro ELISPOT analysis of IFN-γ–releasing cells in PBMCs was performed at 9 wk after BCG immunization. The percentage of BCG-induced IFN-γ cells was determined after 24 h of PPD stimulation.

Close modal

T cell responses in BAL were assessed before immunization and at 3 and 9 wk after BCG vaccination. In contrast to the weak systemic PBMC responses, CD4+ and CD8+ T cell responses were relatively strong in BAL of monkeys that had been i.d. immunized with BCG. Median percentages of Ag-specific CD4+ T cells 9 wk after BCG immunization (ranging between 3 and 13% to PPD) (Fig. 4A, 4B) were higher than the highest CD4+ T cell responses (∼4%) in PBMCs (Fig. 3A). The percentages of PPD-specific CD8+ T cells in BAL from all BCG-immunized animals were higher than those in the control group of monkeys, but the differences were NS (Fig. 4C). We also examined the quality of T cell responses in BAL prior to challenge for analysis of the protection against TB in BCG-immunized monkeys and we found no significant difference between monkey 071, which was ESR- and CRP-positive and had the highest bacterial burden in the lung, and other monkeys in the cytokine profile elicited by BCG vaccination (Fig. 4D). Almost all of the Ag-specific CD4+ T cells in BAL produced TNF and, compared with PBMCs, were comprised of fewer IFN-γ, IL-2, and TNF triple-positive cells and a higher proportion of IL-2 and TNF double-positive and IL-2 single-positive cells in BCG-immunized monkeys, including animal 071. As noted, CD8+ T cell responses in BAL were comprised of 50% single-positive cells at 9 wk after BCG vaccination (data not shown). Although we analyzed the immune responses in animal 071, which did not show any BCG vaccine effects, to investigate potential immune correlates between the cellular immune response before the challenge and clinical outcome after the TB challenge, animal 071 showed neither magnitude nor quality of PPD-specific T cell responses in BAL (Fig. 4).

FIGURE 4.

Induction of PPD-specific cellular immune responses in BAL from BCG-immunized monkeys. For PPD-stimulated BAL cells, cytokines were determined by intracellular cytokine staining FACS analysis following in vitro stimulation with PPD for 18 h. The percentages of cytokine-producing cells (IFN-γ, IL-2, TNF) in the CD4 (A and C) and CD8 (B) T cell populations were determined at the preimmune (0 wk) time point and at 3 and 9 wk after BCG inoculation. (D) The quality of T cell responses in (A) was determined at 9 wk after BCG inoculation, as described in the legend of Fig. 3. The relative contribution of each of these subpopulations to the responding T cell population was determined. (E) The frequency of CD4 T cells producing IL-17 in response to PPD, CD4 regulatory T cells (Treg), γδ T cells, granzyme B+ CD8 T cells, NKG2D+ CD8 T cells, and MAIT-like T cells was determined at 9 wk after BCG vaccination. The frequency of γδ T cells, granzyme B+ CD8 T cells, NKG2D+ CD8 T cells, and MAIT-like T cells was analyzed using BAL cells from nine animals in control group (003, 027, 063, 064, 066, 068, 072, 073, 074) and five animals in BCG group (002, 062, 065, 069, 071). Fewer animals were used in this experiment because of unavailability of enough BAL cells.

FIGURE 4.

Induction of PPD-specific cellular immune responses in BAL from BCG-immunized monkeys. For PPD-stimulated BAL cells, cytokines were determined by intracellular cytokine staining FACS analysis following in vitro stimulation with PPD for 18 h. The percentages of cytokine-producing cells (IFN-γ, IL-2, TNF) in the CD4 (A and C) and CD8 (B) T cell populations were determined at the preimmune (0 wk) time point and at 3 and 9 wk after BCG inoculation. (D) The quality of T cell responses in (A) was determined at 9 wk after BCG inoculation, as described in the legend of Fig. 3. The relative contribution of each of these subpopulations to the responding T cell population was determined. (E) The frequency of CD4 T cells producing IL-17 in response to PPD, CD4 regulatory T cells (Treg), γδ T cells, granzyme B+ CD8 T cells, NKG2D+ CD8 T cells, and MAIT-like T cells was determined at 9 wk after BCG vaccination. The frequency of γδ T cells, granzyme B+ CD8 T cells, NKG2D+ CD8 T cells, and MAIT-like T cells was analyzed using BAL cells from nine animals in control group (003, 027, 063, 064, 066, 068, 072, 073, 074) and five animals in BCG group (002, 062, 065, 069, 071). Fewer animals were used in this experiment because of unavailability of enough BAL cells.

Close modal

Recently, the contribution of various immune cells, including innate immune cells to mycobacterial immunity, has been discussed in detail. Therefore, we next assessed the percentages of resident Th17 cells (CD4+, IL-17+), regulatory T cells (CD4+, CD25+, Foxp3+), γδ T cells (CD3+, γδ TCR+), granzyme B–expressing CD8 T cells (CD8+, granzyme B+), nonconventional CD8 T cells restricted by HLA-E (CD8+, NKG2D+), and mucosal-associated invariant T (MAIT) cells (CD3+, TCR Vα7.2+) in BAL. The i.d. BCG vaccination did not affect the infiltration of these immune cells to the lung, suggesting that the protective effect induced by BCG vaccination was not involved in the establishment of local immunity before the M. tuberculosis challenge (Fig. 4E).

A critical aspect that may impact protective efficacy is whether the Ag-specific vaccine-elicited responses are rapidly expanded after M. tuberculosis infection. Therefore, we assessed the kinetics of T cell responses in PBMCs to BCG and/or M. tuberculosis Ags following TB challenge (Fig. 5).

FIGURE 5.

Postchallenge cellular immune responses in PBMCs. The percentages of CD4 (AE) and CD8 (FJ) T cells in PBMCs producing IFN-γ, IL-2, or TNF in response to PPD (A and F), Ag85B peptides (B and G), MDP1 (C and H), ESAT-6 (D and I), or CFP10 (E and J) were determined at prechallenge (0 wk) and at 2, 4, 6, and 8 wk after M. tuberculosis challenge. The black line indicates the median for 10 monkeys per group.

FIGURE 5.

Postchallenge cellular immune responses in PBMCs. The percentages of CD4 (AE) and CD8 (FJ) T cells in PBMCs producing IFN-γ, IL-2, or TNF in response to PPD (A and F), Ag85B peptides (B and G), MDP1 (C and H), ESAT-6 (D and I), or CFP10 (E and J) were determined at prechallenge (0 wk) and at 2, 4, 6, and 8 wk after M. tuberculosis challenge. The black line indicates the median for 10 monkeys per group.

Close modal

By 4 wk postinfection, nearly all of the control group animals had high and comparable percentages (∼2%) of PPD-specific CD4+ T cells that exceeded prechallenge levels (Fig. 5A). In striking contrast, CD4+ T cell responses to M. tuberculosis/BCG Ags Ag85B and MDP1 in PBMCs were remarkably lower than responses to PPD. Low de novo CD4+ T cell responses to Ag85B and MDP1 were generated in control animals by 4 wk postchallenge (Fig. 5B, 5C). Of note, BCG-vaccinated animals showed neither an accelerated response nor enhanced magnitude of BCG Ag-specific CD4+ T cell response compared with that in unvaccinated control animals at any time point postchallenge (Fig. 5A–C). Thus, the percentage of BCG Ag-specific CD4+ T cells in BCG-vaccinated animals was not significantly increased from prechallenge levels (Fig. 5A–C). In contrast, BCG Ag-specific CD8+ T cells were neither generated de novo nor boosted in the blood after challenge (Fig. 5F–H). Taken together, the results indicated that the low-level pre-existing BCG Ag-specific CD4+ and CD8+ T cells generated by BCG immunization in PBMCs were not appreciably boosted by M. tuberculosis infection.

Finally, we measured postchallenge CD4+ and CD8+ T cell responses to the M. tuberculosis–specific Ags ESAT-6 and CFP10, which are not expressed by BCG. Between 4 and 6 wk after challenge, nearly all of the animals in the control group and BCG-immunized groups generated significant CD4+ T cell responses against both M. tuberculosis Ags (Fig. 5D, 5E). Although M. tuberculosis Ag-specific CD8+ T cells were generated in both groups postinfection, greater median percentages were found only in ESAT-–stimulated PBMCs from control group animals at 6 wk postchallenge (Fig. 5I, 5J). In addition, the percentage of IL-17–producing CD4+ T cells responding to M. tuberculosis Ags in PBMCs was remarkably low, and no significant differences were found between the BCG-vaccinated group and control animals (data not shown). The percentages of PPD- and MDP1-specific CD4+ T cells in monkey 071, which did not show any BCG vaccine effects, tended to be higher but were not significantly different from those in the other BCG-immunized monkeys (Fig. 5). Similarly, PPD-, Ag85B-, ESAT-6–, and CFP10-specific CD8+ T cells were generated in monkey 071 at later time points. From comparative studies of monkey 071 with other BCG-immunized monkeys, no significant differences were found between postchallenge cellular immune responses and outcome (Fig. 5).

Despite the widespread use of BCG vaccination, M. tuberculosis infection remains the leading global cause of death from a single infectious disease (1, 2). The complicated dynamics of infection and disease make clinical trials to test new TB vaccines extremely complex. Fixing the conditions of a preclinical study using an animal model is therefore essential (19, 22, 23). NHP TB models are more attractive models because they show the whole spectrum of human lesions and because NHPs are the closest species to humans and provide the opportunity to conduct parallel human and NHP trials. To assess the relative merits of cynomolgus monkeys as screens for TB vaccines, we analyzed the efficacy of i.d. BCG vaccination and the course of infection.

The outcomes of preventive effects of BCG vaccination in TB-infected cynomolgus macaques could be affected by the monkey origin, M. tuberculosis and BCG strains, infection route, and size of the challenge dose (2426), although these can be taken into consideration. To overcome these problems, we used cynomolgus macaques of Asian origin that have been bred as pure blood of each origin housed in TPRC (16). Such pure-blood macaques are important for comparison of various genetic factors in biological studies aimed at TB vaccine development. The challenge condition, 40 CFU (range, 35–50 CFU), is the minimum dose that is sufficient to induce a consistent and measurable disease burden and for survival until the fixed end point in all unvaccinated control animals for evaluation of the protective effects of TB vaccines.

Infection typically remains latent, which is the most common manifestation of human M. tuberculosis infection, but it can be reactivated in the setting of immune suppression, including HIV coinfection (27) and other forms of immune suppression such as TNF neutralization in humans (12, 28). The cynomolgus macaque is the only animal model that is able to establish latent TB infection (27). Active TB disease occurs in ∼45% of infected macaques and is defined by clinical, microbiologic, and immunologic signs, whereas the remaining infected animals are clinically asymptomatic when using cynomolgus macaques that have been infected via bronchoscopic instillation of <25 CFU M. tuberculosis (Erdman strain) (11, 13, 29). In our conditions, TB pathogenesis was not observed in two (20%) of 10 unvaccinated control monkeys, based on clinical assessments. Furthermore, a TB reactivation model by TNF neutralization using latently infected cynomolgus macaques has already been established (11, 13, 30). BCG vaccination is contraindicated for a person infected with HIV because the BCG vaccine is an attenuated live vaccine. Thus, there is very strong connection between TB, BCG vaccination, and HIV infection, and an animal model that allows analysis of the whole spectrum of M. tuberculosis and HIV infection and is very similar to that of humans is essential for obtaining a better understanding of TB. In this study, we developed a TB model using cynomolgus macaques that resulted in progression to AIDS through SIVmac239 and SHIV89.6P infection (15). The model appears to have some distinct advantages over other models as well as providing the possibility for an M. tuberculosis–SIV coinfection model.

Conventional CD4+ and CD8+ T cells producing proinflammatory cytokines (IFN-γ, IL-2, IL-17, and TNF) and regulatory T cells play key roles in protective immunity to M. tuberculosis and are targeted by current vaccine and immunodiagnostic strategies (31, 32). In addition, evidence for other correlates of protective immunity arose with the discovery of polyfunctional CD4+ T cells that simultaneously produce multiple cytokines (usually TNF, IFN-γ, IL-17, and IL-2). The contribution of these cells to mycobacterial immunity has been discussed in detail elsewhere (33), but the importance of polyfunctional T cells in protective immunity is controversial (34). Consistent with results of previous studies, BCG vaccination induced multifunctional CD4+ T cells in PBMCs and BAL cells. However, one BCG-vaccinated monkey (071) displayed the highest bacterial burden in the lung and highest disease level after the challenge despite showing robust immune responses, including multifunctional CD4+ T cells against PPD that were induced in both BAL cells and PBMCs 9 wk after BCG i.d. administration. Remarkably, monkey 071 showed the highest cellular immune responses in PBMCs prior to M. tuberculosis challenge despite being the only monkey showing no protective effects of BCG. A previous study showed that monkeys in which active disease developed had higher IFN-γ responses to specific mycobacterial Ags in both PBMCs and airways (35). The impact of viability, integrity, and persistence of BCG on its efficacy needs to be addressed.

TB infection and vaccination with BCG result in a localized and self-limiting infection that exposes the immune system to a complex Ag repertoire, including mycobacterial proteins, lipids, and glycolipids, and induces both conventional MHC-restricted and unconventional T cell responses as well as Ab responses and trained immunity (4, 36, 37). γδ T cells produce IL-17, which facilitates optimal activation of myeloid cells and efficient recall responses (38, 39). During this process, loosely aggregated innate granulomas are already formed (40). Innate-like MR1 (nonclassical MHC class Ib)–restricted MAIT cells have also been shown to be important players in mycobacterial immunity by potentially acting as early sentinels to M. tuberculosis infection (41, 42). Increasing evidence shows that nonconventional CD8+ T cells restricted by MHC class Ib molecules can recognize distinct microbial Ags and contribute to protection against M. tuberculosis infection (43, 44). In our experimental setting, we could not observe these innate immune cells and granzyme B–expressing CD8+ T cells in the lungs from BCG-vaccinated monkeys before the challenge.

It is well known that the protective efficacy of BCG vaccination against pulmonary TB in adults is variable and partial (45). In contrast, this vaccine affords ∼80% protection against TB meningitis and miliary TB in infants and young children (46). Moreover, one cohort study in Kazakhstan showed that estimates of prevention effectiveness levels based on laboratory-confirmed TB cases were 92% for the BCG Tokyo strain (47). In our NHP model, BCG vaccination did not result in significant decreases of bacterial burden and granuloma severity in the lungs, but it provided complete protection against the development of extrapulmonary dissemination of TB, and the frequency of BCG-protective effectiveness based on clinical and immunological assessments was 90%. Our NHP model might be a potential TB model to recapitulate human TB pathogenesis and the protective effectiveness of BCG vaccination in humans.

In this study, clinical parameters (CT analysis, ESR, CRP) and gross pathology observations did not translate into bacterial burden in the lungs and quality of granulomas. Therefore, identifying reliable clinical and immunological parameters that correlate with pulmonary TB protection is critical for assessing vaccine efficacy in NHPs as well as humans.

In our study, i.d. vaccination with BCG resulted in significantly reduced M. tuberculosis burden in the mediastinal LNs compared with unvaccinated control animals, although complete M. tuberculosis growth inhibition was not observed in 5 out of 10 BCG-vaccinated monkeys (Fig. 2A). From histopathological analyses for BCG-vaccinated monkeys, the formation of coalescent caseous granuloma in hilar LNs appeared to be relevant to aggravation of TB pathogenesis (Table II). Lung-draining LNs are infected with M. tuberculosis, along with the lungs, and serve as crucial sites for M. tuberculosis growth and persistence (48). Therefore, it might be necessary for vaccine development to examine efficacy of preventing M. tuberculosis infection in LNs and not just lungs.

In a phase 2 trial, it was shown that revaccination with BCG can protect adolescents from sustained M. tuberculosis infection (7). It has also been shown by using rhesus macaques that BCG vaccination via aerosol or i.v. routes provided strong prevention against TB infection (8, 9). Accordingly, the dimensions of BCG vaccination should be considered carefully. In this study, we evaluated the preventive effectiveness of i.d. vaccination with BCG and TB pathogenesis in the early phase of M. tuberculosis infection using cynomolgus macaques. The use of a high-dose challenge for assessment of efficacy in NHP models has the risk of vaccines with the potential to be efficacious against a natural challenge appearing ineffective and being disregarded. Thus, there is a need to develop a challenge regimen that is more relevant to natural human infection for future studies on TB latent infection, TB–SIV coinfection, and TB vaccine development (49, 50).

We thank all members of Prof. Yasutomi’s laboratory for valuable comments and assistance. Special thanks go to the members and veterinary staff of HAMRI Co., Ltd., particularly J. Sawata, T. Horikawa, M. Kogure, S. Yokota, and H. Ayukawa, for technical expertise and assistance with animal care, sample processing, and study design. We also thank the members of the Corporation for Production and Research of Laboratory Primates, especially K. Ohto, for excellent technical help.

This work was supported by Health Science Research grants from the Ministry of Health, Labor and Welfare of Japan (17H04079), the Ministry of Education, Culture, Sports, Science and Technology of Japan (19K08968), and Research on Development of New Drugs, Research Program on Emerging and Re-emerging Infectious Diseases from the Japan Agency for Medical Research and Development (18fk0108007h0003, 19ak0101047h0004). The funders had no role in the study design, data collection, interpretation, and decision to submit this work for publication.

The online version of this article contains supplemental material.

Abbreviations used in this article:

BAL

bronchoalveolar lavage

BCG

bacillus Calmette–Guérin

CFP10

10-kDa culture filtrate protein

CRP

C-reactive protein

CT

computed tomography

ESAT-6

early secreted antigenic target of 6 kDa

ESR

erythrocyte sedimentation rate

i.d.

intradermal(ly)

LN

lymph node

MAIT

mucosal-associated invariant T

MDP1

mycobacterial DNA-binding protein 1

NHP

nonhuman primate

NIBIOHN

National Institutes of Biomedical Innovation, Health and Nutrition

PPD

purified protein derivative

SHIV

simian HIV

SPF

specific pathogen–free

TB

tuberculosis

TPRC

Tsukuba Primate Research Center.

1
World Health Organization
. Global tuberculosis report 2018. Available at: https://apps.who.int/iris/handle/10665/274453.
2
Andersen
,
P.
,
T. J.
Scriba
.
2019
.
Moving tuberculosis vaccines from theory to practice.
Nat. Rev. Immunol.
19
:
550
562
.
3
Houben
,
R. M. G. J.
,
P. J.
Dodd
.
2016
.
The global burden of latent tuberculosis infection: a re-estimation using mathematical modelling.
PLoS Med.
13
: e1002152.
4
Nieuwenhuizen
,
N. E.
,
S. H. E.
Kaufmann
.
2018
.
Next-generation vaccines based on bacille Calmette-Guérin.
Front. Immunol.
9
:
121
.
5
Talbot
,
E. A.
,
M. D.
Perkins
,
S. F. M.
Silva
,
R.
Frothingham
.
1997
.
Disseminated bacille Calmette-Guérin disease after vaccination: case report and review.
Clin. Infect. Dis.
24
:
1139
1146
.
6
Rodrigues
,
L. C.
,
V. K.
Diwan
,
J. G.
Wheeler
.
1993
.
Protective effect of BCG against tuberculous meningitis and miliary tuberculosis: a meta-analysis.
Int. J. Epidemiol.
22
:
1154
1158
.
7
Nemes
,
E.
,
H.
Geldenhuys
,
V.
Rozot
,
K. T.
Rutkowski
,
F.
Ratangee
,
N.
Bilek
,
S.
Mabwe
,
L.
Makhethe
,
M.
Erasmus
,
A.
Toefy
, et al
C-040-404 Study Team
.
2018
.
Prevention of M. tuberculosis infection with H4:IC31 vaccine or BCG revaccination.
N. Engl. J. Med.
379
:
138
149
.
8
Van Der Meeren
,
O.
,
M.
Hatherill
,
V.
Nduba
,
R. J.
Wilkinson
,
M.
Muyoyeta
,
E.
Van Brakel
,
H. M.
Ayles
,
G.
Henostroza
,
F.
Thienemann
,
T. J.
Scriba
, et al
.
2018
.
Phase 2b controlled trial of M72/AS01E vaccine to prevent tuberculosis.
N. Engl. J. Med.
379
:
1621
1634
.
9
Dijkman
,
K.
,
C. C.
Sombroek
,
R. A. W.
Vervenne
,
S. O.
Hofman
,
C.
Boot
,
E. J.
Remarque
,
C. H. M.
Kocken
,
T. H. M.
Ottenhoff
,
I.
Kondova
,
M. A.
Khayum
, et al
.
2019
.
Prevention of tuberculosis infection and disease by local BCG in repeatedly exposed rhesus macaques.
Nat. Med.
25
:
255
262
.
10
Darrah
,
P. A.
,
J. J.
Zeppa
,
P.
Maiello
,
J. A.
Hackney
,
M. H.
Wadsworth
II
,
T. K.
Hughes
,
S.
Pokkali
,
P. A.
Swanson
II
,
N. L.
Grant
,
M. A.
Rodgers
, et al
.
2020
.
Prevention of tuberculosis in macaques after intravenous BCG immunization.
Nature
577
:
95
102
.
11
Lin
,
P. L.
,
J.
Dietrich
,
E.
Tan
,
R. M.
Abalos
,
J.
Burgos
,
C.
Bigbee
,
M.
Bigbee
,
L.
Milk
,
H. P.
Gideon
,
M.
Rodgers
, et al
.
2012
.
The multistage vaccine H56 boosts the effects of BCG to protect cynomolgus macaques against active tuberculosis and reactivation of latent Mycobacterium tuberculosis infection.
J. Clin. Invest.
122
:
303
314
.
12
Lin
,
P. L.
,
M.
Rodgers
,
L.
Smith
,
M.
Bigbee
,
A.
Myers
,
C.
Bigbee
,
I.
Chiosea
,
S. V.
Capuano
,
C.
Fuhrman
,
E.
Klein
,
J. L.
Flynn
.
2009
.
Quantitative comparison of active and latent tuberculosis in the cynomolgus macaque model.
Infect. Immun.
77
:
4631
4642
.
13
Lin
,
P. L.
,
C. B.
Ford
,
M. T.
Coleman
,
A. J.
Myers
,
R.
Gawande
,
T.
Ioerger
,
J.
Sacchettini
,
S. M.
Fortune
,
J. L.
Flynn
.
2014
.
Sterilization of granulomas is common in active and latent tuberculosis despite within-host variability in bacterial killing.
Nat. Med.
20
:
75
79
.
14
McShane
,
H.
2005
.
Co-infection with HIV and TB: double trouble.
Int. J. STD AIDS
16
:
95
100, quiz 101
.
15
Okamura
,
T.
,
Y.
Tsujimura
,
S.
Soma
,
I.
Takahashi
,
K.
Matsuo
,
Y.
Yasutomi
.
2016
.
Simian immunodeficiency virus SIVmac239 infection and simian human immunodeficiency virus SHIV89.6P infection result in progression to AIDS in cynomolgus macaques of Asian origin.
J. Gen. Virol.
97
:
3413
3426
.
16
Yasutomi
,
Y.
2010
.
Establishment of specific pathogen-free macaque colonies in Tsukuba Primate Research Center of Japan for AIDS research.
Vaccine
28
(
Suppl. 2
):
B75
B77
.
17
Honjo
,
S.
,
F.
Cho
,
K.
Terao
.
1984
.
Establishing the cynomolgus monkey as a laboratory animal.
Adv. Vet. Sci. Comp. Med.
28
:
51
80
.
18
Luciw
,
P. A.
,
K. L.
Oslund
,
X. W.
Yang
,
L.
Adamson
,
R.
Ravindran
,
D. R.
Canfield
,
R.
Tarara
,
L.
Hirst
,
M.
Christensen
,
N. W.
Lerche
, et al
.
2011
.
Stereological analysis of bacterial load and lung lesions in nonhuman primates (rhesus macaques) experimentally infected with Mycobacterium tuberculosis.
Am. J. Physiol. Lung Cell. Mol. Physiol.
301
:
L731
L738
.
19
Darrah
,
P. A.
,
D. L.
Bolton
,
A. A.
Lackner
,
D.
Kaushal
,
P. P.
Aye
,
S.
Mehra
,
J. L.
Blanchard
,
P. J.
Didier
,
C. J.
Roy
,
S. S.
Rao
, et al
.
2014
.
Aerosol vaccination with AERAS-402 elicits robust cellular immune responses in the lungs of rhesus macaques but fails to protect against high-dose Mycobacterium tuberculosis challenge.
J. Immunol.
193
:
1799
1811
.
20
Diel
,
R.
,
R.
Loddenkemper
,
A.
Nienhaus
.
2010
.
Evidence-based comparison of commercial interferon-gamma release assays for detecting active TB: a metaanalysis.
Chest
137
:
952
968
.
21
Mitarai
,
S.
,
M.
Okumura
,
E.
Toyota
,
T.
Yoshiyama
,
A.
Aono
,
A.
Sejimo
,
Y.
Azuma
,
K.
Sugahara
,
T.
Nagasawa
,
N.
Nagayama
, et al
.
2011
.
Evaluation of a simple loop-mediated isothermal amplification test kit for the diagnosis of tuberculosis.
Int. J. Tuberc. Lung Dis.
15
:
1211
1217
.
22
Darrah
,
P. A.
,
R. M.
DiFazio
,
P.
Maiello
,
H. P.
Gideon
,
A. J.
Myers
,
M. A.
Rodgers
,
J. A.
Hackney
,
T.
Lindenstrom
,
T.
Evans
,
C. A.
Scanga
, et al
.
2019
.
Boosting BCG with proteins or rAd5 does not enhance protection against tuberculosis in rhesus macaques.
NPJ Vaccines
4
:
21
.
23
Hansen
,
S. G.
,
D. E.
Zak
,
G.
Xu
,
J. C.
Ford
,
E. E.
Marshall
,
D.
Malouli
,
R. M.
Gilbride
,
C. M.
Hughes
,
A. B.
Ventura
,
E.
Ainslie
, et al
.
2018
.
Prevention of tuberculosis in rhesus macaques by a cytomegalovirus-based vaccine.
Nat. Med.
24
:
130
143
.
24
Langermans
,
J. A. M.
,
P.
Andersen
,
D.
van Soolingen
,
R. A. W.
Vervenne
,
P. A.
Frost
,
T.
van der Laan
,
L. A. H.
van Pinxteren
,
J.
van den Hombergh
,
S.
Kroon
,
I.
Peekel
, et al
.
2001
.
Divergent effect of bacillus Calmette-Guérin (BCG) vaccination on Mycobacterium tuberculosis infection in highly related macaque species: implications for primate models in tuberculosis vaccine research.
Proc. Natl. Acad. Sci. USA
98
:
11497
11502
.
25
Peña
,
J. C.
,
W.-Z.
Ho
.
2015
.
Monkey models of tuberculosis: lessons learned.
Infect. Immun.
83
:
852
862
.
26
Kaushal
,
D.
,
S.
Mehra
,
P. J.
Didier
,
A. A.
Lackner
.
2012
.
The non-human primate model of tuberculosis.
J. Med. Primatol.
41
:
191
201
.
27
Selwyn
,
P. A.
,
P.
Alcabes
,
D.
Hartel
,
D.
Buono
,
E. E.
Schoenbaum
,
R. S.
Klein
,
K.
Davenny
,
G. H.
Friedland
.
1992
.
Clinical manifestations and predictors of disease progression in drug users with human immunodeficiency virus infection.
N. Engl. J. Med.
327
:
1697
1703
.
28
Keane
,
J.
,
S.
Gershon
,
R. P.
Wise
,
E.
Mirabile-Levens
,
J.
Kasznica
,
W. D.
Schwieterman
,
J. N.
Siegel
,
M. M.
Braun
.
2001
.
Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent.
N. Engl. J. Med.
345
:
1098
1104
.
29
Lin
,
P. L.
,
S.
Pawar
,
A.
Myers
,
A.
Pegu
,
C.
Fuhrman
,
T. A.
Reinhart
,
S. V.
Capuano
,
E.
Klein
,
J. L.
Flynn
.
2006
.
Early events in Mycobacterium tuberculosis infection in cynomolgus macaques.
Infect. Immun.
74
:
3790
3803
.
30
Lin
,
P. L.
,
A.
Myers
,
L.
Smith
,
C.
Bigbee
,
M.
Bigbee
,
C.
Fuhrman
,
H.
Grieser
,
I.
Chiosea
,
N. N.
Voitenek
,
S. V.
Capuano
, et al
.
2010
.
Tumor necrosis factor neutralization results in disseminated disease in acute and latent Mycobacterium tuberculosis infection with normal granuloma structure in a cynomolgus macaque model.
Arthritis Rheum.
62
:
340
350
.
31
Kaufmann
,
S. H.
,
S.
Baumann
,
A.
Nasser Eddine
.
2006
.
Exploiting immunology and molecular genetics for rational vaccine design against tuberculosis.
Int. J. Tuberc. Lung Dis.
10
:
1068
1079
.
32
Prezzemolo
,
T.
,
G.
Guggino
,
M. P.
La Manna
,
D.
Di Liberto
,
F.
Dieli
,
N.
Caccamo
.
2014
.
Functional signatures of human CD4 and CD8 T cell responses to Mycobacterium tuberculosis.
Front. Immunol.
5
:
180
.
33
Bhatt
,
K.
,
S.
Verma
,
J. J.
Ellner
,
P.
Salgame
.
2015
.
Quest for correlates of protection against tuberculosis.
Clin. Vaccine Immunol.
22
:
258
266
.
34
Hawkridge
,
T.
,
T. J.
Scriba
,
S.
Gelderbloem
,
E.
Smit
,
M.
Tameris
,
S.
Moyo
,
T.
Lang
,
A.
Veldsman
,
M.
Hatherill
,
L.
Merwe
, et al
.
2008
.
Safety and immunogenicity of a new tuberculosis vaccine, MVA85A, in healthy adults in South Africa.
J. Infect. Dis.
198
:
544
552
.
35
Mohan
,
A. K.
,
T. R.
Coté
,
J. A.
Block
,
A. M.
Manadan
,
J. N.
Siegel
,
M. M.
Braun
.
2004
.
Tuberculosis following the use of etanercept, a tumor necrosis factor inhibitor.
Clin. Infect. Dis.
39
:
295
299
.
36
Scriba
,
T. J.
,
S. H.
Kaufmann
,
P.
Henri Lambert
,
M.
Sanicas
,
C.
Martin
,
O.
Neyrolles
.
2016
.
Vaccination against tuberculosis with whole-cell mycobacterial vaccines.
J. Infect. Dis.
214
:
659
664
.
37
Kaufmann
,
E.
,
J.
Sanz
,
J. L.
Dunn
,
N.
Khan
,
L. E.
Mendonça
,
A.
Pacis
,
F.
Tzelepis
,
E.
Pernet
,
A.
Dumaine
,
J. C.
Grenier
, et al
.
2018
.
BCG educates hematopoietic stem cells to generate protective innate immunity against tuberculosis.
Cell
172
:
176
190.e19
.
38
Yao
,
S.
,
D.
Huang
,
C. Y.
Chen
,
L.
Halliday
,
G.
Zeng
,
R. C.
Wang
,
Z. W.
Chen
.
2010
.
Differentiation, distribution and gammadelta T cell-driven regulation of IL-22-producing T cells in tuberculosis.
PLoS Pathog.
6
: e1000789.
39
Lockhart
,
E.
,
A. M.
Green
,
J. L.
Flynn
.
2006
.
IL-17 production is dominated by gammadelta T cells rather than CD4 T cells during Mycobacterium tuberculosis infection.
J. Immunol.
177
:
4662
4669
.
40
Shaler
,
C. R.
,
C. N.
Horvath
,
M.
Jeyanathan
,
Z.
Xing
.
2013
.
Within the enemy’s camp: contribution of the granuloma to the dissemination, persistence and transmission of Mycobacterium tuberculosis.
Front. Immunol.
4
:
30
.
41
Harriff
,
M. J.
,
E.
Karamooz
,
A.
Burr
,
W. F.
Grant
,
E. T.
Canfield
,
M. L.
Sorensen
,
L. F.
Moita
,
D. M.
Lewinsohn
.
2016
.
Endosomal MR1 trafficking plays a key role in presentation of Mycobacterium tuberculosis ligands to MAIT cells.
PLoS Pathog.
12
: e1005524.
42
Greene
,
J. M.
,
P.
Dash
,
S.
Roy
,
C.
McMurtrey
,
W.
Awad
,
J. S.
Reed
,
K. B.
Hammond
,
S.
Abdulhaqq
,
H. L.
Wu
,
B. J.
Burwitz
, et al
.
2017
.
MR1-restricted mucosal-associated invariant T (MAIT) cells respond to mycobacterial vaccination and infection in nonhuman primates.
Mucosal Immunol.
10
:
802
813
.
43
Bian
,
Y.
,
S.
Shang
,
S.
Siddiqui
,
J.
Zhao
,
S. A.
Joosten
,
T. H. M.
Ottenhoff
,
H.
Cantor
,
C. R.
Wang
.
2017
.
MHC Ib molecule Qa-1 presents Mycobacterium tuberculosis peptide antigens to CD8+ T cells and contributes to protection against infection.
PLoS Pathog.
13
: e1006384.
44
Lewinsohn
,
D. M.
,
A. L.
Briden
,
S. G.
Reed
,
K. H.
Grabstein
,
M. R.
Alderson
.
2000
.
Mycobacterium tuberculosis-reactive CD8+ T lymphocytes: the relative contribution of classical versus nonclassical HLA restriction.
J. Immunol.
165
:
925
930
.
45
Mittrücker
,
H. W.
,
U.
Steinhoff
,
A.
Köhler
,
M.
Krause
,
D.
Lazar
,
P.
Mex
,
D.
Miekley
,
S. H.
Kaufmann
.
2007
.
Poor correlation between BCG vaccination-induced T cell responses and protection against tuberculosis.
Proc. Natl. Acad. Sci. USA
104
:
12434
12439
.
46
Fine
,
P. E.
1995
.
Variation in protection by BCG: implications of and for heterologous immunity.
Lancet
346
:
1339
1345
.
47
Favorov
,
M.
,
M.
Ali
,
A.
Tursunbayeva
,
I.
Aitmagambetova
,
P.
Kilgore
,
S.
Ismailov
,
T.
Chorba
.
2012
.
Comparative tuberculosis (TB) prevention effectiveness in children of Bacillus Calmette-Guérin (BCG) vaccines from different sources, Kazakhstan.
PLoS One
7
: e32567.
48
Ganchua
,
S. K. C.
,
A. M.
Cadena
,
P.
Maiello
,
H. P.
Gideon
,
A. J.
Myers
,
B. F.
Junecko
,
E. C.
Klein
,
P. L.
Lin
,
J. T.
Mattila
,
J. L.
Flynn
.
2018
.
Lymph nodes are sites of prolonged bacterial persistence during Mycobacterium tuberculosis infection in macaques.
PLoS Pathog.
14
: e1007337.
49
Soma
,
S.
,
S.
Kawai
,
H.
Inada
,
K.
Watanabe
,
S.
Mizuno
,
S.
Kato
,
K.
Matsuo
,
Y.
Yasutomi
.
2018
.
Primary role of suppressor of cytokine signaling 1 in Mycobacterium bovis BCG infection.
Infect. Immun.
86
: e00376-18.
50
Mizuno
,
S.
,
S.
Soma
,
H.
Inada
,
T.
Kanuma
,
K.
Matsuo
,
Y.
Yasutomi
.
2019
.
SOCS1 antagonist-expressing recombinant Bacillus Calmette-Guérin enhances antituberculosis protection in a mouse model.
J. Immunol.
203
:
188
197
.

The authors have no financial conflicts of interest.

Supplementary data