Comparison of BCG prime-DNA booster and rBCG regimens for protection against tuberculosis

Abstract

Developing an effective adult prophylaxis vaccine is a high priority in the global control of tuberculosis (TB), because TB remains an important public health problem and the current widely used BCG vaccine provides effective protection only for children but variable protection against adult TB. BCG priming-heterologous vaccines booster and recombinant BCG technologies have been thought as two important regimens for inducing effective protection against adult TB. Obviously, defining the protective efficacy of the two regimens would benefit more rational design of the future adult TB vaccines. In this study, a recombinant BCG strain (rBCG::685A) expressing the fusion protein of ESAT-6 and Ag85A (r685A) of Mycobacterium tuberculosis was constructed successfully and the secretion of r685A protein from rBCG strain was confirmed by western blotting with anti-ESAT-6 and anti-Ag85A polyclonal antibodies, respectively. The immune responses and protective effects in rBCG::685A vaccinated C57BL/6 mice were compared with that of our previous reported BCG prime-pcD685A booster regimen. Boosting BCG with pcD685A DNA elicited higher level of r685A protein specific IFN-γ secreted by splenocytes and a more significant increase of both TNF-α and iNOS responses in the lung, thus providing better control of bacterial growth in both lung and spleen of immunized mice challenged with virulent M. tuberculosis, compared with mice vaccinated with rBCG::685A or BCG alone. Our results have implications for development of more effective adult TB vaccines for improved control of TB.

Keywords: :

• prime-boost
• recombinant BCG
• ESAT-6-Ag85A
• DNA vaccine
• tuberculosis

Introduction

Tuberculosis (TB) remains a serious threat to human health with 8.7 million new cases of TB and 1.4 million TB-related deaths worldwide.¹ Although Mycobacterium bovis Bacille Calmette-Guerin (BCG), currently the only vaccine available against TB, has effective protection against the severe forms of TB in children,² it has unstable protective effects for adult pulmonary TB,³ which accounts for more than 94% of all TB cases.¹ Therefore, more effective vaccines against adult TB are urgently needed.

Up to now, two regimens for developing adult TB vaccines have been generally accepted and achieved significant progress.⁴ One is prime-boost strategy⁵ where BCG priming followed by heterologous subunit vaccine booster has been a favored approach for providing protection against adult TB.^6,7 To take advantage of the proven efficacy of BCG in children, BCG might first be used for neonatal vaccination as in most countries worldwide.⁸ When the effects of BCG start to wane during adolescent, heterologous booster vaccines such as subunit vaccines,^9-11 DNA vaccines,^12-16 recombinant adenovirus,^17,18 and modified vaccinia virus,^19-22 could further be used to extend the protective efficacy of BCG priming. This strategy would benefit a significant proportion of the global population, because about 4 billion people have been immunized with BCG since 1921. The other strategy also relies on present BCG vaccination strategy and aims to replace BCG vaccine by recombinant BCG (rBCG) technology. rBCG strains could induce longer-lasting immune responses and provide stronger protection for adults than BCG after neonatal immunization, developing by different methods such as expressing immunodominant antigens of M. tuberculosis,^23,24 deletion antigens from BCG,²⁵ cytokines, or combination,^26,27 and enhanced CD8⁺ T-cell responses,²⁸ etc. Clearly, defining the protective efficacy of the two regimens would benefit more rational design of the future adult TB vaccines.

In our previous study, a candidate DNA vaccine pcD685A expressing the fusion protein of ESAT-6 and Ag85A (r685A) of M. tuberculosis was constructed and pcD685A could act as an effective BCG booster, which results in a significant reduction in bacterial load of both spleen and lung of BCG priming C57BL/6 mice challenged with virulent M. tuberculosis H37Rv, when compared with BCG or pcD685A alone.¹⁵ In the present study, a recombinant BCG strain expressing the fusion protein of ESAT-6 and Ag85A (rBCG::685A) was also constructed. The immune response and protective efficacy induced separately by rBCG::685A and BCG prime-pcD685A boost were evaluated and compared in vaccinated C57BL/6 mice.

Results

rBCG::685A expressing the secreted r685A fusion protein

The Mycobacterium-E.coli shuttle expression plasmid pM685A (Fig. 1) was successfully constructed and confirmed by double-digestion with the BamHI and EcoRI and DNA sequencing. The secretory expression of r685A fusion protein, with a molecular size about 38kDa, was confirmed only in the culture supernatants of rBCG::685A, whatever anti-Ag85A or anti-ESAT-6 antibodies were used (Fig. 1). In addition, Ag85A protein with a size of about 32kDa in the culture supernatant of both BCG and rBCG::685A was demonstrated by western blotting with anti-Ag85A antibody (Fig. 1). No expression of ESAT-6 protein in the culture supernatant of both BCG and rBCG::685A was demonstrated as expected^29,30 by western blotting with anti-ESAT-6 antibody (Fig. 1).

Figure 1. Recombinant Mycobacterium–E.coli shuttle plasmid pM685A and the secretory expression of r685A fusion protein from rBCG::685A. The plasmid pM685A containing the full-fragment of ESAT-6 and Ag85A was engineered to express the secreted r685A protein under the direction of Hsp60 promoter (PHsp60) and α-Ag signal peptide (α-AgSP). The expression of the r685A fusion protein in the culture supernatant from rBCG::685A was confirmed by western blotting with anti-ESAT-6 rabbit antibody or anti-Ag85A chicken polyclonal antibody. The cultural supernatant of BCG was used as negative control.

Higher levels of r685A specific IFN-γ secreted by splenocytes of mice vaccinated with BCG prime-pcD685A booster than rBCG::685A

To determine cell-mediated immune responses to vaccine candidates, splenocytes were isolated from immunized mice following stimulation with r685A and PPD, respectively. The level of IFN-γ production by splenocytes determined by ELISA was shown in Figure 2. Upon stimulation with PPD or r685A, all BCG vaccinated groups produced higher levels of IFN-γ than PBS control mice (P < 0.05) and higher levels of IFN-γ were induced in both rBCG::685A and BCG prime-pcD685A booster groups than BCG (P < 0.05). Most importantly, higher levels of r685A specific IFN-γ were secreted by splenocytes of mice vaccinated with BCG prime-pcD685A booster than rBCG::685A (P < 0.05).

Figure 2. The level of IFN-γ secreted by splenocytes of different group mice detected by ELISA (n = 5). Nine weeks after immunization, splenocytes were obtained from 5 mice in each group were prepared individually and 10⁶ cells were added into each well of 96-well microtiter plates. PPD (10 μg/ml), r685A protein (5 μg/ml) in complete RPMI-1640 medium were used as the stimulator, respectively. The results were shown as mean ± SD (pg/ml).

Higher mRNA levels of TNF-α and iNOS in the lung of mice vaccinated with BCG prime-pcD685A booster than rBCG::685A

As shown in Figure 3, the mRNA expression level of iNOS, cytokines IFN-γ, TNF-α, and IL-10 in the lung of vaccinated mice were detected by qRT-PCR. When compared with control mice, the expression of these molecules in the different vaccinated groups increased at different degrees. Especially, all BCG based vaccination strategies increased the expression of IFN-γ and IL-10, though there was no statistics significant between these different groups. Notably, the expression of both TNF-α and iNOS in the lung of BCG prime-pcD685A booster vaccination mice increased most significantly than BCG and rBCG::685A groups (P < 0.05).

Figure 3. mRNA levels of IFN-γ, IL-10, TNF-α, and iNOS in the lung of different vaccinated mice detected by qRT-PCR (n = 5). Tested cDNAs were normalized to the endogenous RNA levels of the internal control GAPDH. The fold increase of gene expression in vaccinated group was calculated using 2^−ΔΔCT method. The results were shown as mean ± SD.

Better protective effect conferred by BCG prime-pcD685A booster than rBCG::685A

Compared with the control group, the bacterial load in both lung and spleen was significantly decreased in all BCG vaccinated groups. Among these groups, BCG prime-pcD685A booster vaccinated mice resulted in the most significant reduction of bacterial load in the lung and spleen (P < 0.05). In addition, the difference of bacterial load in both organs between rBCG::685A and BCG vaccinated groups had no statistics significance (P > 0.05) (Fig. 4). The hierarchy of protection based on the bacterial load in the lung was consistent with the observation under microscopy after tissue sections of the lung from different groups were stained by acid-fast staining (Fig. 5). Acid-fast bacilli (AFB) were present in the whole lung tissue section of control mice, and the most serious lung pathology with extensive fibrosis, perivasculitis, and alveolitis was also observed in this group. Much fewer AFB and less lung inflammation were observed in alveolar tissue from mice vaccination with BCG plus pcD685A boosting than those in rBCG::685A or BCG alone (Fig. 5).

Figure 4. Bacterial load per lung and spleen in C57BL/6 mice (n = 7). Nine weeks later, different vaccinated mice were challenged i.v. with 10⁶ CFU of M. tuberculosis H37Rv. Four weeks after challenge, lungs and spleens were homogenized and cultured for bacterial load. The results were shown as the mean (± SEM) per group.

Figure 5. Representative lung pathology of different vaccinated C57BL/6 mice after infection. Four weeks post-challenge, the right lung from different vaccinated mice was fixed and embedded for HE staining (10 × 4, left row) and acid-fast staining (10 × 40, right row). Arrows indicate acid-fast staining positive bacteria.

Discussion

As a high priority in global TB control, developing more effective vaccines against adult TB than BCG has been attracting a lot of attention during last two decades.⁴ Up to now, 14 vaccine candidates have been assessed in clinical phase and more than 200 vaccine candidates have been preclinically evaluated. However, there are no other candidates based on different platforms except BCG, which provide stronger protective efficacy than BCG so far. More than 90% of infants worldwide are vaccinated with BCG annually since the Expanded Programme on Immunization (EPI) was established by WHO in 1974⁸ and BCG provides over 80% efficacy against childhood TB with confirmed safety. Therefore, trying to substitute BCG with other regimens might not be reasonable and practical. Correspondingly, developing a vaccine as an effective booster of BCG vaccine or designing more effective rBCG to replace BCG vaccine may be ideal, regardless of current immunization strategy. In addition, both regimens might be more effective than BCG, as evidenced in recent studies in animal models, even in clinical experiments.^{7,16,20,21,28} Obviously, comparing the protective efficacy of two regimens and choosing the suitable type play an important role in the development of adult TB vaccines. In this study, immunogenicity and protective efficacy of two regimens were compared in vaccinated C57BL/6 mice. Our results demonstrated that the protective effects induced by BCG prime-DNA booster are better than that of the recombinant BCG expressing the same two antigens ESAT-6 and Ag85A, as evidenced by higher levels of r685A specific IFN-γ secreted by splenocytes, a more significant increase of TNF-α and iNOS levels in the lung and better control of M. tuberculosis growth in both lung and spleen of vaccinated mice. To our knowledge, this is the first report on the differential protective efficacy of both BCG prime-DNA booster and rBCG regimens based on the same antigens.

ESAT-6 is a highly dominant T cell antigen with 6kDa molecular weight,³¹ which is encoded by RD1 region of M. tuberculosis and is missing in all BCG substrains distributed around the world.³² Antigen Ag85A (32kDa) has mycolyl-transferase activity and plays an important role in the biogenesis of cell wall of mycobacteria.³³ As promising vaccine candidates for TB, both antigens were designed as the fusion protein r685A in our previous work¹⁵ and we also confirmed that the whole blood cells from TST-positive healthy household contacts stimulated with the protein resulted in higher level of antigen-specific IFN-γ release than TST-negative population.¹⁵ We constructed a DNA vaccine pcD685A expressing the fusion r685A protein and confirmed its protective effect as an effective booster vaccine of BCG.¹⁵ In this study, r685A protein with the same size as the product purified from recombinant E.coli, was confirmed only in the culture supernatant of rBCG::685A strain as demonstrated by western blotting, which indicated that Hsp60 promoter and α-AgSP correctly directed the expression of secreted r685A protein from the rBCG strain. Although splenocytes from rBCG::685A immunized mice secreted higher level of PPD or r685A protein specific IFN-γ than those of BCG, rBCG::685A did not provide any enhanced protection against M. tuberculosis challenge in the both lung and spleen, when compared with BCG vaccine. Several recombinant BCG strains expressing 38kDa protein³⁴ or ESAT-6 protein³⁰ were reported no improving protection when vaccinated mice were challenged M. tuberculosis for 4 weeks. rBCG expressing the fusion protein of ESAT-6 and one member Ag85A complex, Ag85B, also did not improve the short-time protection efficacy against TB in a mouse model.^35,36 Meanwhile, DNA vaccine pE6/85B which consisted of ESAT-6 and Ag85B fusion gene with tpa signal sequence could provide the protection against M. tuberculosis infection in vaccinated mice and be used as an effective booster after BCG immunization,¹⁶ which provided further evidence to support our conclusion. Moreover, there are some rBCG strains, such as rBCG30,²³ rBCG::AB,²⁴ rBCG::X,³⁷ rBCG::85B-Rv3425,³⁸ rBCG UreC: Hly,²⁸ and AERAS-rBCG,³⁹ etc, providing stronger and/or longer-lasting protective efficacy than BCG in animal models or in humans. All of these studies indicate that developing more recombinant BCG vaccine candidates mainly depend on the choice of suitable antigens. Moreover, the route and efficacy of antigen presentation could determine immune responses and protection effect induced by vaccination. BCG prime-pcD685A DNA boosted mice produced the higher levels of r685 protein specific IFN-γ secreted by splenocytes and higher mRNA expression of TNF-α and iNOS in the lung of vaccinated mice and provided the stronger protection against the challenge infection of M. tuberculosis than rBCG::685A. Though pcD685A DNA and rBCG were separately used to express the same protein r685A, differential degrees of cell-mediated immune responses were induced between rBCG and BCG prime-DNA booster vaccinated mice. These might account for the different immune protective mechanisms of DNA vaccine, rBCG, and prime-boost strategies, although exact mechanism remains unknown.

Upon aerosol exposure with M. tuberculosis, alveolar macrophages and lung DCs phagocytose M. tuberculosis. Some components of M. tuberculosis such as ManLAM would inhibit infected macrophages to produce TNF-α, which might delay the influx of APCs in the infected site and the migration of these M. tuberculosis-loaded APCs to the local lymph node.⁴⁰ In addition, ManLAM could promote infected DCs to secret IL-10,⁴¹ which might participate in the process of immune evasion by suppressing T-cell priming in the local lymph node. Further, priming T lymphocytes were delayed to recruit to infected local site and protective immune responses were difficult to establish under this situation. However, if priming T lymphocytes are recruited to the lung, they could produce both TNF-α that might have an antagonistic effect of ManLAM, and IFN-γ that might act on new recruitment macrophages to trigger the expression of antimicrobial effectors, including iNOS, which could generate the production of nitric oxide (NO) and inhibit the growth of intracellular M. tuberculosis.⁴² Therefore, based on the host-pathogen interaction, the expression level of the aforementioned molecules in the lung of vaccinated mice could play a vital role in determining the outcome of M. tuberculosis infection as evidenced by recent studies.^43,44 In our study, the highest levels of r685 protein specific IFN-γ secreting by splenocytes and the most significant expression of TNF-α and iNOS in the lung were induced in BCG prime-pcD685A DNA boosted mice, which also provided the strongest protection against M. tuberculosis challenge infection among all vaccinated groups.

Boosting BCG with BCG has no proven benefit in many clinical trials and experimental animals,⁶ and BCG revaccination had been ceased by WHO since 1995.⁶ While MVA85A/AERAS-485 in a Phase 2b study was evaluated as a boost to BCG and was not effective, it does not negate the potential for a boost to BCG, either in infancy or adolescence/adulthood.⁴⁵ In the future, the rational design of adult TB vaccines will need to confirm the protective effect of combination BCG or effective enhanced rBCG strains as prime vaccine and heterogolous boost strategy to identify the best candidate for clinical testing.

Materials and Methods

Bacterial strains and culture conditions

M. tuberculosis H37Rv, M. bovis BCG China, and rBCG strains were separately grown in either Middlebrook 7H9 medium (Difco Laboratories) or on Middlebrook 7H11 agar (Difco Laboratories), supplemented with 10% ADC, 0.5% glycerol, and 0.05% Tween 80. Escherichia coli strains DH5α and BL21 (DE3) were grown in Luria-Bertani medium and used for cloning and expression in E. coli, respectively. When required, a final concentration of 25 µg/ml kanamycin or 100 µg/ml ampicillin was used for agar plate or liquid culture medium.

Construction of recombinant BCG strain

The gene fragment from the Mycobacterium-E.coli shuttle plasmid pDE22 digested with XbaI and PstI, containing the DNA sequences of Hsp60 promoter (PHsp60), α-Ag signal peptide of BCG (α-AgSP), and multi-clone sites (MCS), was first ligated into the corresponding sites of pMV261 to construct the pMS261. Then, the full-length gene encoding the fusion protein of ESAT-6 and Ag85A of M. tuberculosis was isolated from pcD685A by digesting the plasmid with BamHI and EcoRI and ligated into BamHI/EcoRI-digested pMS261 vector, yielding the recombinant plasmid pM685A. The anticipated sequence of the insert was confirmed by DNA sequencing and enzyme digestion. Finally, pM685A was introduced into BCG China strain by electroporation and the recombinant BCG strain (rBCG::685A) was screened on kanamycin resistance plates. The recombinant plasmid construct rBCG::685A was engineered to express the recombinant fusion protein r685A under the direction of PHsp60 and α-AgSP.

Culture supernatants of rBCG::685A and BCG from 7H9 broth were prepared as previously described.²⁴ About 40 µg protein of each sample was separated by SDS-PAGE on a 12% gel and then analyzed by western blotting. The expression of r685A protein was detected using anti-ESAT-6 rabbit antibody (diluted 1/2000; ab45073, Abcam) or anti-Ag85A chicken polyclonal antibody (diluted 1/2000; ab14073, Abcam). The immunoblots were developed with a pro-light ECL kit according to the manufacture’s protocol (Tiangen) and were imaged by Bio-Rad Universal Hood II Gel Imager.

Mice and immunization protocol

Specific pathogen-free, female C57BL/6 mice at 6 weeks of age (Beijing Vital River) were used in this study. Mice were fed commercial mouse chow and water ad libitum. Mice were immunized s.c. once at the base of the tail with 1 × 10⁶ CFU of either BCG or rBCG::685A in a volume of 100 μl of PBS. pcD685A was used to boost thrice as described previously¹⁵ and the protocol was performed as shown in Figure 6. A control group was treated with 100 μl of PBS. Animal experiments were repeated twice. Animal experiments were performed in accordance with the guidelines of Chinese Council on Animal Care. The research protocol was approved by Tongji Medical School Committee on Biosafety.

Figure 6. Immunization scheme for 2 regimens. For prime-boost regimen, C57BL/6 mice were first immunized s.c. with a dose of 10⁶ CFU of BCG, then 100 μg pcD685A DNA was injected intramuscularly 2 weeks later and repeated twice with 2-week intervals. For recombinant BCG regimen, mice were vaccinated once with 10⁶ CFU of rBCG::685A. PBS was used as negative control. Nine weeks after the immunization, 7 mice were challenged with 10⁶ CFU M. tuberculosis H37Rv and 5 mice were used for the immunological assays. All experiments were repeated twice.

Antigen specific IFN-γ secreted by splenocytes detected by ELISA

Nine weeks after C57BL/6 mice were vaccinated, spleen cells from five mice per group were prepared individually and resuspended in complete RPMI-1640 medium (Invitrogen) with 10% fetal calf serum. The concentration of cell suspension was adjusted to 10⁷ cells /ml and 100 μl of cells were added into each well in 96-well microtiter plates. One hundred microlitres of PPD (10 μg/ml, Statens Serum Institute) and r685A protein (5 μg/ml) in complete RPMI-1640 medium were used as the stimulator, respectively. RPMI-1640 medium was used as negative control. The plates were incubated at 37 °C in a humidified CO₂ incubator for 72h. Supernatants were collected from each well and the concentration of IFN-γ in the supernatants was measured in duplicate by the mouse IFN-γ ELISA kit (Dakewei Biotech) with the detection limit of 8 pg/ml. The results were shown as mean ± SD (pg/ml) of five mice per group.

Cytokines and iNOS mRNA expression in the lung detected by qRT-PCR

DNA-free RNA samples from lung tissue of five vaccinated mice were extracted with TRIzol reagent (Invitrogen), respectively. qRT-PCR analysis was performed using SYBR Green kit (Roche) and Applied Rotor-Gene 3000 Real-Time system (Gene) with different specific primers listed in Table 1, including IFN-γ, IL-10, TNF-α, and iNOS. Tested cDNAs were normalized to the endogenous RNA levels of the internal control GAPDH. The fold increase of gene expression in vaccinated group was calculated using 2^−ΔΔCT method.

Table 1. Primers and qRT-PCR cycling parameters for murine cytokines and iNOS

Primer Name	Primer sequence	Conditions
iNOS	F:AACAGGAGAA GGGGACGAAC T R:CATTGGAAGT GAAGCGTTTC G	95 °C for 45s, 55 °C for 45s and 72 °C 45s, 40 cycles
IL-10	F:GGTTGCCAAG CCTTATCGGA R:ACCTGCTCCA CTGCCTTGCT
TNF-α	F:AAGGGAGAGT GGTCAGGTTG CC R:CCTCAGGGAA GAGTCTGGAA AGG
IFN-γ	F:GGCTGTCCCT GAAAGAAAGC R:GAGCGAGTTA TTTGTCATTC GG

Efficacy comparison and histopathological analysis

Nine weeks after immunization, mice in different groups were challenged with a dose of 10⁶ CFU M. tuberculosis in 100 μl PBS via tail vein. Four weeks post-challenge, 7 C57BL/6 mice per group were sacrificed for efficacy comparison and histopathological analysis. The spleens and lungs were removed aseptically, homogenized, and cultured for CFU of M. tuberculosis on Middlebrook 7H11 agar containing 2 μg/ml of 2-thiophenecarboxylic acid hydrazide (TCH, Beijing Luqiao Corp.) to selectively inhibit the growth of the residual BCG. Right lung lobes from different vaccinated C57BL/6 mice were fixed in 10% formalin in PBS and embedded in paraffin. Both Hematoxylin and Eosin (H&E) staining and acid-fast staining were used to record the pathological changes in the lung.

Data analysis

The Student t test was used to compare the difference between groups and the difference was considered statistically significant when P < 0.05.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

This work was supported by grants of the National Mega-Projects of Science Research for the 12th Five-Year Plan (2012ZX10003008-005) and 863 program (2012AA02A401).

10.4161/hv.26969