Abstract
The lengthy and complicated current regimen required to treat drug-susceptible tuberculosis (TB) reflects the ability of Mycobacterium tuberculosis (Mtb) to persist in host tissues. The stringent response pathway, governed by the dual (p)ppGpp synthetase/hydrolase, RelMtb, is a major mechanism underlying Mtb persistence and antibiotic tolerance. In the current study, we addressed the hypothesis that RelMtb is a “persistence antigen” presented during TB chemotherapy and that enhanced T-cell immunity to RelMtb can enhance the tuberculocidal activity of the first-line anti-TB drug, isoniazid, which has reduced efficacy against Mtb persisters, C57BL/6 mice and Hartley guinea pigs were aerosol-infected with Mycobacterium tuberculosis (Mtb) and, 4 weeks later, received either human-equivalent daily doses of isoniazid alone, or isoniazid in combination with a DNA vaccine targeting relMtb. After isoniazid treatment, the total number of Mtb antigen-specific CD4+ T cells remained stable in mouse lungs and spleens, as did the number of RelMtb-specific T cells, although there was a significant reduction in dominant antigen ESAT6-specific CD4+ or TB10.4-specific CD8+ T cells in the lungs and spleens of mice, Therapeutic vaccination enhanced the activity of isoniazid in Mtb-infected C57BL/6 mice and guinea pigs. When treatment with isoniazid was discontinued, mice immunized with the relMtb DNA vaccine showed a lower mean lung bacterial burden at relapse compared to the control group. Our work shows that antitubercular treatment shapes antigen presentation and antigen-specific T-cell responses, and that therapeutic vaccination targeting the Mtb stringent response may represent a novel approach to enhance immunity against Mtb persisters, with the ultimate goal of shortening curative TB treatment.
Introduction
Despite the high efficacy of the current 6-month “short-course” combination regimen, tuberculosis (TB) remains a global health emergency. Improper provision and supervision of treatment leads to excess morbidity and mortality, continued transmission, and emergence of drug resistance [1]. The prolonged duration of curative TB treatment is believed to reflect the ability of a subpopulation of Mycobacteria tuberculosis (Mtb) bacilli to remain in a non-replicating persistent state in the infected host [2]. These “persisters” exhibit reduced susceptibility to isoniazid [3], which inhibits mycolic acid synthesis in the cell wall [4, 5]. This phenomenon is referred to as “antibiotic tolerance” [6], in which slowly dividing or non-dividing bacteria become less susceptible to killing by antibiotics targeting actively multiplying organisms.
One of the key bacterial pathways implicated in antibiotic tolerance is the stringent response, which is triggered by the rapid accumulation of the key regulatory molecules hyperphosphorylated guanosine ((p)ppGpp) and inorganic polyphosphate (poly(P)) in response to nutrient starvation and other stresses [7]. Mtb has a dual-function enzyme, RelMtb, which is able to synthesize and hydrolyze (p)ppGpp [8, 9], as well as two polyphosphate kinases (PPK1, PPK2) and two exopolyphosphatases (PPX1, PPX2), which regulate intracellular poly(P) homeostasis [10-13]. The mycobacterial stringent response appears to be a positive feedback loop, as poly(P) phosphorylates and activates the two-component system MprAB, which induces expression of sigE and relMtb [14], leading to increased synthesis of (p)ppGpp, which inhibits the hydrolysis of poly(P) by the exopolyphosphatase, PPX2 [13]. Deletion of Rv2583/relMtb results in profound bacterial phenotypes, including defective growth at elevated temperatures, and reduced long-term survival in nutrient starvation and progressive hypoxia [15], as well as in mouse lungs [16]. RelMtb deficiency also results in defective Mtb survival in a mouse hypoxic granuloma model of latent TB infection [17] and in guinea pig lungs [18].
Although the Mtb stringent response contributes to the formation of persisters and antibiotic tolerance, it remains to be determined whether this pathway can be targeted therapeutically during chronic Mtb infection. We have shown previously that DNA vaccination with four key stringent response genes (relMtb, sigE, ppk2, and ppx1) generated antigen-specific CD4+ T-cell responses and vaccination prior to Mtb aerosol challenge augmented the tuberculocidal activity of isoniazid in mice [19]. However, very limited data are available regarding antigen presentation of stringent response factors during antitubercular treatment in vivo, as well as the optimal target for therapeutic DNA vaccination, which is an effective strategy for generating cellular and humoral immunity in various diseases [20]. In the current study, we first characterized the expression of relMtb in Mtb-infected macrophages treated with the first-line antitubercular drug, isoniazid, as well as the abundance of RelMtb-specific CD4+ and CD8+ T cells in the lungs and spleens of Mtb-infected C57BL/6 mice following daily oral treatment with human-equivalent doses of isoniazid. Next, we studied the efficacy of DNA vaccination targeting RelMtb as an adjunctive therapy to isoniazid in two different animal models of chronic TB infection.
Material and methods
Bacteria and growth conditions
Wild-type Mtb H37Rv was grown in Middlebrook 7H9 broth (Difco, Sparks, MD) supplemented with 10% oleic acid-albumin-dextrose-catalase (OADC, Difco), 0.1% glycerol, and 0.05% Tween-80 at 37°C in a roller bottle [21]. An Mtb H37Rv strain deficient in Rv2583c/relMtb (Δrel) and its isogenic wild-type strain [15, 18] were kindly provided by Dr. Valerie Mizrahi and grown as described above in order to validate the RelMtb antiserum.
IC-21 Macrophages infection and real-time PCR analysis
The C57B/6J macrophage cell line, IC-21 (ATCC, No. TIB-186), was grown in RPMI medium with 10% FBS and 1% penicillin/streptomycin. IC-21 cells were divided and plated in a multilayer culture flask (Millipore). At day 0, 106 macrophages were infected with 5×106 of logarithmically growing H37Rv bacilli. The cells were harvested 4 days after infection, RNA was extracted and qPCR analysis was performed, as described previously [19]. The primers were listed in Table 1.
Antigen preparation
The previously generated relMtb expression plasmid, pET15b[relMtb] [22], was used for expression and purification of RelMtb protein, as previously described [19]. Mtb peptides TB10.4 4 – 11 (IMYNYPAM) and ESAT6 1–15 (MTEQQWNFAGIEAAA)[23] were commercially synthesized (Genescript), dissolved in DMSO and stored at 20°C until use. Mtb whole cell lysate (10μg/ml, BEI) was used to measure Mtb-specific T cells during infection.
DNA vaccine
The plasmid pSectag2B encoding relMtbwas used as the relMtb DNA vaccine [19]. ESAT6 was cloned into pSectag2B using the restriction enzymes BamHI and HindIII by forward and reverse primer (F: GCGAAGCTTGTGGCCGAGGACCAGCTCAC, R: CGCGGATCCGCGAACATCCCAGTGACGT). The insertion was confirmed by sequencing. Each DNA vaccine was delivered as previously described [24, 25], and the original plasmid, pSectag2B, was used as control treatment. All procedures were performed according to protocols approved by the Johns Hopkins University Institutional Animal Care and Use Committee. Briefly, each plasmid was delivered by intramuscular injection into the quadriceps femoris muscle of mice (100μg) or guinea pigs (500μg) at the indicated time points.
Harvest of lung cells and splenocytes
At the indicated time points, the mice were sacrificed and peripheral blood and splenocytes were collected, as previously described [19, 25]. At necropsy, the lungs were perfused with 1ml normal saline by direct injection into the right ventricle of the heart. A random section of the lung was used for cytometry analysis and the tissue samples were incubated in 37° C for 1 hour with intermittent agitation in RPMI medium (Gibco) containing collagenase D (1mg/ml), DNase (0.25mg/ml) and hyaluronidase type V (1mg/ml). The cells were then filtered through a 70-μm nylon filter mesh to remove undigested tissue fragments and washed with complete RPMI medium.
Intracellular cytokine stain and flow cytometry analysis
To detect antigen-specific T-cell responses by IFN-γ intracellular staining, splenocytes or lung cells were stimulated individually with the purified recombinant proteins, RelMtb (10μg/ml), ESAT6 peptide (1 μg/ml) or TB10 peptide (1 μg/ml) for 24 hours at 37°C before the addition of GolgiPlug (BD Pharmingen, San Diego, CA). The concentration of protein and peptides used were according to previous reports [19, 23]. After incubation, the cells were washed once with FACScan buffer and then stained with PE-conjugated monoclonal rat anti-mouse CD4 (BD Pharmingen) and/or APC-conjugated monoclonal rat anti-mouse CD8 (eBioscience). Cells were permeabilized using the Cytofix/Cytoperm kit (BD Pharmingen, San Diego, CA). Intracellular IFN-γ was stained with FITC-conjugated rat anti-mouse IFN-γ and APC-conjugated rat anti-mouse TNFα (BD Pharmingen, San Diego, CA). Flow cytometry was performed by FACSCalibur and analyzed with FlowJo software.
Aerosol infection of mice with Mtb and therapeutic DNA vaccination
Female C57b/6J mice (6-8 week-old) were aerosol-infected with ∼100 bacilli of wild-type Mtb H37Rv. After 28 days of infection, the mice received isoniazid (10 mg/kg) by esophageal gavage once daily (5 days/week), and were randomized to receive DNA vaccine containing relMtb, ESAT6, or the empty vector (100 μg) by intramuscular injection. Five mice in each group were sacrificed on Days 30, 60 and 90 after aerosol challenge and the lungs were homogenized and plated for CFU [26]. A partial sample of each infected lung was fixed with 10% buffered formaldehyde, processed, and paraffin-embedded for histological staining with hematoxylin and eosin staining. Morphometric analysis of histology was performed as described previously [27].
Guinea pig infection and vaccination
Female outbred Hartley guinea pigs (250 to 300 g) were purchased from Charles River Labs (Wilmington, MA). Guinea pigs were aerosol-infected with Mtb H37Rv using a Madison chamber (University of Wisconsin, Madison, WI) to deliver ∼2 log10 CFU to the lungs [28]. After 28 days of infection, the guinea pigs received human-equivalent doses of isoniazid (60 mg/kg) daily (5 days/week) by esophageal gavage [29]. During isoniazid treatment, the guinea pigs were randomized to receive either relMtb DNA vaccine (500μg) or empty vector (pSectag2B, 500μg) by intramuscular injection once weekly for four times. 56 days after treatment initiation, three guinea pigs in each group were sacrificed and lung CFU were determined by plating on Middlebrook 7H11 selective agar (BD).
Enzyme-linked immunosorbent assay and immunoblots
Antigen-specific antibody responses were measured by ELISA as described previously [30], with minor modifications in coating and sera incubation. The 96-well microplate was coated with purified RelMtb (1μg/ml) overnight. After blocking, serum from individual vaccinated mice or guinea pigs was diluted 1:100 with PBS, added to the wells and incubated at room temperature for 2 hours. HRP rabbit anti-mouse antibody (Abcam) was used for ELISA with mice sera. For immunoblots, the serum from relMtb DNA-vaccinated mice was used as the primary antibody to detect RelMtb expression levels. Monoclonal Anti-Mtb HSP70 (Clone IT-40, BEI) was used to detect the housekeeping protein, HSP70 (DnaK) [19]. To detect anti-RelMtb IgG in guinea pigs, RelMtb protein was loaded in each lane and proved by probing with anti-His antibody. The serum from vaccinated guinea pigs was used as a primary antibody at 1:50 dilution. HRP-linked goat anti-guinea pig antibody (Novus) was used for ELISA and immunoblots.
Statistical analysis
Data from at least three biological replicates were used to calculate means and standard error (SEM) for graphing purposes. Pairwise comparisons of group mean values for log10 counts and flow cytometry data were made by using one-way analysis of variance and Bonferroni’s multiple comparison test posttest with GraphPad prism 7 (GraphPad, San Diego, Calif.), and a P value of <0.05 was considered significant.
Results
Antibiotic treatment shapes antigen presentation and T-cell responses
It has been shown that antitubercular treatment decreases Mtb-specific T cells [31, 32]. Using an intracellular cytokine-releasing assay after stimulation with ESAT6 and TB10.4 peptides, we found that ESAT6-specific CD4+ and TB10.4-specific CD8+ T cells were significantly reduced in the lungs and spleens of Mtb-infected mice following isoniazid treatment (Figure 1). To further determine if isoniazid treatment can affect antigen-specific T cells, we used Mtb lysate to stimulate splenocytes and lung-derived T cells. Intracellular cytokine staining revealed a significant reduction in Mtb-specific CD8+ T cells in the spleens, while Mtb-specific CD4+ T cells were increased in the lungs and unchanged in the spleens (Figures 2A and 2B).
In order to further characterize the isoniazid-induced changes in Mtb antigen presentation by lung APCs, we harvested lung cells from Mtb-infected C57B/6J mice that were either treated with isoniazid or left untreated (controls), and co-cultured these cells with splenocytes derived from a separate group of C57B/6J mice, which were aerosol-infected with Mtb one month previously. A significantly increased number of IFNγ-releasing CD4+ T cells were detected among splenocytes when lung APC were used from the isoniazid-treated group compared to lung APC from the untreated control group (Fig 2C, 2D). Taken together, our data show that isoniazid decreases ESAT6-specific CD4+ T cells and TB10-specific CD8+ T cells in the spleens and lungs of Mtb-infected mice, but Mtb antigen-specific CD4+ T cells were increased in the lungs after treatment, suggesting that isoniazid increases the diversity of antigens presented by APCs.
Isoniazid induces expression of Mtb stringent response genes during ex vivo and in vivo infection
We found previously that DNA vaccination targeting four genes of the Mtb stringent response showed synergy with isoniazid against Mtb in vivo [19]. To determine if stringent response genes were upregulated during isoniazid treatment, we used RT-PCR to study the expression of each of the four genes of our previously reported stringent-response vaccine (relMtb, sigE, ppk2, or ppx1) during Mtb H37Rv infection of IC21 mouse macrophages, which were exposed or unexposed to isoniazid 2 μg/ml for 4 days [33]. As shown in Figure 3A, relMtb was the most highly induced gene following isoniazid exposure (p<0.001).
In contrast, expression of the Mtb gene encoding the immunodominant antigen ESAT6 and the housekeeping gene rrs did not change significantly after isoniazid exposure of Mtb-infected IC21 mouse macrophages. In order to confirm our findings at the protein level, axenic cultures of Mtb H37Rv were exposed to different concentrations of isoniazid for 4 days. By densitometric analysis, isoniazid exposure (2 μg/ml) increased the expression of RelMtb by 2.5-fold relative to unexposed bacteria, following normalization with the housekeeping protein, HSP70 (Figure 3B). Therefore, the stringent response genes were upregulated during infection of macrophages exposed to isoniazid, and RelMtb was significantly upregulated following isoniazid exposure during ex vivo infection and in vitro.
RelMtb antigen is continuously presented during isoniazid treatment in mice
In order to determine the significance of our findings in the context of TB treatment in the mammalian host, we studied RelMtb-specific CD4+ T-cell responses in the lungs of Mtb-infected mice treated with isoniazid as a surrogate for RelMtb antigen presentation by APCs during TB treatment. We harvested spleen-derived and lung-derived T cells from Mtb-infected and untreated C57BL/6J mice or infected mice treated with human-equivalent doses of isoniazid for 4 weeks. Consistent with the lower lung bacillary burden in isoniazid-treated mice relative to untreated controls, the proportion of ESAT6-specific CD4+ T cells declined in the lungs of the former group relative to those in the latter group. In contrast, the proportion of RelMtb-specific CD4+ T cells was increased in the lungs and spleens of isoniazid-treated mice compared to untreated mice (Figure 3C), suggesting that RelMtb antigens continued to be presented during isoniazid treatment.
Therapeutic DNA vaccination targeting RelMtb augments the tuberculocidal activity of isoniazid in C57B/6J mice
Next, we tested whether DNA vaccination targeting the principal stringent response factor, RelMtb, could augment the bactericidal activity of isoniazid, as in the case of the 4-component stringent-response vaccine [19]. C57BL/6J mice were infected with the virulent Mtb strain H37Rv via aerosol. After four weeks of infection, all mice were treated with isoniazid 10 mg/kg once daily (5 days/week) for four weeks. The mice were vaccinated either with relMtb DNA vaccine (100 μg), or empty vector control (100 μg) once weekly for four weeks starting concurrently with isoniazid treatment. After four weeks of treatment, adjunctive therapy with relMtb DNA vaccine lowered the mean lung bacillary burden by 0.695 log10 relative to isoniazid alone (p=0.048, Figure 4A). To confirm the therapeutic specificity of the relMtb DNA vaccine, we repeated the mouse study with a longer duration of treatment and added another group of mice, which received isoniazid plus DNA vaccine targeting ESAT6. After 8 weeks of treatment, the group receiving relMtb DNA vaccine again showed a significant reduction in mean lung CFU compared to the empty vector group (p=0.045, Figure 4B). In contrast, the mean lung bacillary burden of mice vaccinated with esat6 DNA vaccine was not statistically different from that of control mice (Figure 4B). Based on these two independent studies, we conclude that immunization with relMtb DNA vaccine potentiates the tuberculocidal activity of isoniazid in mice.
relMtb DNA vaccination reduces Mtb regrowth upon discontinuation of isoniazid treatment
To further study if relMtb DNA vaccine can target persistent bacilli, C57BL/6J mice were aerosol-infected with Mtb and, 4 weeks later, were treated daily with isoniazid for 4 weeks. At the completion of isoniazid treatment, separate groups of mice received either relMtb DNA vaccine (100 μg) or empty vector (100 μg) by intramuscular injection once weekly for two weeks (Figure 5A). Four weeks after cessation of isoniazid, the mean lung bacillary burden of mice receiving the relMtb DNA was lower by 0.5 log10 compared to that of the empty vector vaccine group (p=0.02, Figure 5B), suggesting that relMtb DNA vaccine can restrict regrowth of Mtb after cessation of antibiotic therapy.
relMtb DNA vaccine enhances the tuberculocidal activity of isoniazid in guinea pigs
To determine if the adjunctive antitubercular activity of the relMtb DNA vaccine can be duplicated in different species, we infected guinea pigs with virulent Mtb strain H37Rv via aerosol, and 4 weeks later, they were treated once daily with human-equivalent doses of isoniazid (60 mg/kg) for a total of 4 weeks [29]. Separate groups of guinea pigs were vaccinated either with relMtb DNA vaccine (500 μg) or empty vector control (500 μg) once weekly for four weeks beginning with the initiation of isoniazid treatment (Figure 6A). To confirm the immunogenicity of the relMtb DNA vaccine, we used ELISA and immunoblots to demonstrate that this vaccine generated RelMtb-specific IgG responses (Figures 6B and 6C). At the time of treatment completion, the mean lung bacillary burden in the relMtb DNA vaccine group was lower by ∼0.5 log10 relative to the that in the isoniazid alone control group (p=0.05; Figure 6D), demonstrating that, as in mice, relMtb DNA vaccine potentiates the antitubercular activity of isoniazid in guinea pigs.
Discussion
Several potential reasons may explain the limited efficacy of immunodominant antigen-specific T cells in eliminating bacteria during chronic Mtb infection, including the reduced amount of immunodominant antigens presented, resulting in a reduction in the number of immunodominant antigen-specific CD4+ cells, as well as the continuous release of these antigens, leading to exhaustion of antigen-specific CD4+ T cells [34]. Therefore, subdominant Mtb antigens, including those induced during drug treatment, may represent better targets for therapeutic vaccines. In the current study, we show that treatment with the classical bactericidal drug isoniazid increases the abundance of CD4+ T cells specific to the Mtb stringent response factor RelMtb in the lungs of mice chronically infected with Mtb, but not those specific to the Mtb immunodominant antigen ESAT6. Furthermore, a DNA vaccine targeting RelMtb, but not one targeting ESAT6, potentiated the antitubercular activity of isoniazid in two different animal models of chronic TB infection.
Recently, there is significant interest in host-directed therapies to treat TB [35]. Although BCG vaccination does not enhance the bactericidal activity of chemotherapy in the murine model [36], a DNA vaccine expressing heat shock protein 65 has been shown to synergize with conventional antitubercular drugs, further reducing the bacterial burden in the lungs of Mtb-infected mice or non-human primates [36, 37]. A fragment whole cell lysate therapeutic vaccine, RUTI, has shown efficacy in generating protective immunity in pre-clinical studies [38, 39]. However, the primary factors responsible for TB immunity during antibiotic treatment remain unknown. Previously, we have shown that a DNA vaccine targeting key Mtb stringent response factors generates IgG and antigen-specific CD4+ T cells, and increased the antitubercular activity of isoniazid when given as a therapeutic vaccine in Mtb-infected mice [19]. The findings of the current study corroborate our previous findings and highlight the importance of immunity to Mtb stringent response factors during isoniazid treatment. One potential explanation for this phenomenon is that the population of persistent bacilli in untreated, chronically infected mice is relatively small [40], but exposure to isoniazid further drives the formation of drug-tolerant persisters [29, 41]. In favor of this hypothesis, isoniazid exposure induces Mtb expression of RelMtb, which is required for persister formation [7, 42, 43]. Further studies are required to determine the potential utility of the relMtb vaccine as an adjunctive therapeutic intervention in shortening the duration of treatment for active TB or latent TB infection.
Based on our co-culture results, we conclude that isoniazid treatment increased APC presentation of RelMtb in the lungs of mice (Figures 2C and 2D). One potential explanation for this finding is that isoniazid increased the amount of RelMtb antigen presented due to bacillary lysis. Alternatively, isoniazid treatment may have directly modified Mtb protein expression, leading to altered antigen processing and presentation. We found that ESAT6-specific CD4+ and CD8+ T cells were significantly decreased in the lungs, even though RelMtb-specific CD4+ T cells were increased after isoniazid treatment (Figure 1), suggesting that isoniazid treatment shapes the diversity of antigen presentation by APCs, enhancing the processing and presentation of subdominant antigens such as RelMtb.
The relMtb DNA vaccine potentiated the killing activity of isoniazid against chronic TB infection in C57B/6J mice, which develop cellular lung granulomas lacking central necrosis, as well as in guinea pigs, which develop a potent DTH response and more human-like granulomas with necrosis and tissue hypoxia. Interestingly, the adjunctive antitubercular activity of the relMtb DNA vaccine was more pronounced in the guinea pig model, in which the defective survival phenotype of a relMtb-deficient Mtb mutant was significantly accelerated relative to the C57B/6J mouse model [44, 45]. These findings suggest that tissue necrosis is not a prerequisite for vaccine efficacy, but its presence may further enhance the activity of isoniazid. Further studies are needed to test the relMtb DNA vaccine in other animal models with necrotic granulomas, including the C3Heb/FeJ mouse [27] and the nonhuman primate [46].
In conclusion, we have shown that isoniazid treatment shapes the diversity of antigen presentation during chronic TB infection in mice, leading to altered antigen-specific CD4+ T-cell profiles in the lungs. In addition, this is the first study to use the Mtb stringent response protein Relmtb as a therapeutic vaccine target. Additional studies are needed to test the therapeutic efficacy of this vaccine in other animal models, and to characterize the immunological basis for its efficacy in mice and guinea pigs.
Funding
National Institute of Allergy and Infectious Diseases of the National Institutes of Health grants: R21AI22922 (P.C.K.).
Author contributions
Y-M.C. contributed to the design of the study, performance of the experiments, data interpretation and the writing of the manuscript; N.K.D. contributed to the performance of the experiments; M.L.P. contributed to the performance of the experiments; C-F.H. contributed to the design of the study and data interpretation; and P.C.K. contributed to the design of the study, data interpretation and the writing of the manuscript.
Declaration of interests
The authors have no financial conflicts to declare.