The fbpA/sapM double knock out strain of Mycobacterium tuberculosis is highly attenuated and immunogenic in macrophages

Abstract

Tuberculosis (TB), caused by Mycobacterium tuberculosis (Mtb), is the leading cause of death due to bacterial infections in mankind, and BCG, an attenuated strain of Mycobacterium bovis, is an approved vaccine. BCG sequesters in immature phagosomes of antigen presenting cells (APCs), which do not fuse with lysosomes, leading to decreased antigen processing and reduced Th1 responses. However, an Mtb derived ΔfbpA attenuated mutant underwent limited phagosome maturation, enhanced immunogenicity and was as effective as BCG in protecting mice against TB. To facilitate phagosome maturation of ΔfbpA, we disrupted an additional gene sapM, which encodes for an acid phosphatase. Compared to the wild type Mtb, the ΔfbpAΔsapM (double knock out; DKO) strain was attenuated for growth in mouse macrophages and PMA activated human THP1 macrophages. Attenuation correlated with increased oxidants in macrophages in response to DKO infection and enhanced labeling of lysosomal markers (CD63 and rab7) on DKO phagosomes. An in vitro Antigen 85B peptide presentation assay was used to determine antigen presentation to T cells by APCs infected with DKO or other mycobacterial strains. This revealed that DKO infected APCs showed the strongest ability to present Ag85B to T cells (>2500 pgs/mL in 4 hrs) as compared to APCs infected with wild type Mtb or ΔfbpA or ΔsapM strain (<1000 pgs/mL in 4 hrs), indicating that DKO strain has enhanced immunogenicity than other strains. The ability of DKO to undergo lysosomal fusion and vacuolar acidification correlated with antigen presentation since bafilomycin, that inhibits acidification in APCs, reduced antigen presentation. Finally, the DKO vaccine elicited a better Th1 response in mice after subcutaneous vaccination than either ΔfbpA or ΔsapM. Since ΔfbpA has been used in mice as a candidate vaccine and the DKO (ΔfbpAΔsapM) mutant is more immunogenic than ΔfbpA, we propose the DKO is a potential anti-tuberculosis vaccine.

Figure 1. Southern and PCR analyses of M. tuberculosis strains.

a). Southern analysis of genomic DNA of M. tuberculosis (Mtb) wild type (H37Rv), fbpA mutant (ΔfbpA) and fbpA/sapM double knock out (DKO) strains. Genomic DNA was digested with NdeI and BamHI, separated on 1% agarose gels and transferred to nitrocellulose membranes. Membranes were hybridized with [³²P]dCTP labeled 3.3 kb DNA fragment containing sapM region and signals captured by autoradiography. Arrows indicate the sizes of the signals. b–c). PCR analysis for sapM region in M. tuberculosis H37Rv, ΔfbpA and DKO strains. PCR was performed using standard protocols with genomic DNA from the above strains as templates. Primer pairs RV3310EX1 and Rv3310EX2 (b) and RV3310EX1 and RV3310RT2 (c) were used to amplify DNA. d. RT-PCR analysis for sapM expression in M. tuberculosis H37Rv, ΔfbpA and DKO strains. Total RNA was used to synthesize cDNA from these strains. RT+ and RT- indicate cDNA templates generated in the presence or absence of reverse transcriptase (Superscript II; Invitrogen). The products obtained from both reactions were used as templates in RT-PCR to prove the absence of DNA contamination in total RNAs used for reverse transcriptions. PCR was performed using primers RV3310RT1 and RV3310RT2. MW: molecular weight marker; arrow indicates the size of the band. PCR products were separated on 1% agarose gels.

Figure 2. The ΔfbpAΔsapM double knockout (DKO) strain is attenuated in macrophages and induces stronger oxidant responses that reduce its viability:

Macrophages from C57Bl/6 mouse bone marrow (BMs) and human THP1 macrophages (pre-activated with phorbol ester) were infected with mycobacteria (MOI 1∶1), washed, incubated, lysed and plated for viable colony counts (CFUs). a). The DKO strain is more attenuated compared to wild type Mtb in BMs. b-c) Intracellular reactive oxygen species (ROS) and nitric oxide (NO) were measured respectively using dihydro-dichloro-fluorescein acetate (DCFDA) fluorescent probe and Greiss reagent. DKO induced elevated NO responses (p value by t test; panel c) but not ROS (panel b). d-e) DKO was attenuated in THP1 macrophages compared to ΔfbpA, ΔsapM or wild type H37Rv in BMs (p<0.01) that correlated with increased ROS responses (panel e). Nitric oxide responses of THP1 were not detectable (not shown). f). Mycobacteria (10⁵ CFU/mL; baseline shown as dotted line) were exposed to the bactericidal action of the superoxide and NO donor 3-morpholinosydnonimine (10 mM; SIN-1) in 7H9 broth and viable counts determined at intervals (24 and 72 hr post treatment) by plating on 7H11 agar. DKO is markedly susceptible by 72 hrs in vitro to the oxidants released by SIN-1 (p value by t test).

Figure 3. The ΔfbpAΔsapM double knockout (DKO) strain shows enhanced lysosomal localization in mouse macrophages:

gfpMtb H37Rv or Oregon green stained mutant strains were phagocytosed into BMs, incubated, fixed 24 hrs later and stained with primary antibodies to lysosomal markers LAMP1 (IDB4), CD63 and rab7 followed by Texas red conjugated conjugates. Mycobacteria colocalizing with antibodies were scored using a Nikon fluorescence microscope and Metaview deconvolution software. a) Illustration that the DKO mutant colocalizes better with rab7 lysosomal marker. b) Percent colocalization was determined by counting 200 macrophages per well each with 1–3 mycobacteria and averaging counts from triplicate chambers (SD). One of three similar experiments is shown. Text below the bar diagram indicates the colocalization of each marker in relation to different strains (*p<0.01, t test).

Figure 4. The ΔfbpAΔsapM double knockout (DKO) strain is more immunogenic in mouse macrophages and elicits stronger immune responses in mice:

a) BMs and DCs from C57Bl/6 mice were infected with mycobacteria (MOI 1∶1), washed and overlaid with Antigen 85B specific BB7 hybridoma T cells (1∶20 ratio). After 4 hrs, the supernatants collected were tested for IL-2 using sandwich ELISA. DKO induces BMs and DCs to prime T cells to secrete larger amounts of IL-2, indicating a better processing of DKO for Ag85B (4 experiments, SEM, * <0.009 vs. ΔfbpA or ΔsapM; by t test). b). C57Bl/6 mice (3 per group) were vaccinated with mycobacterial strains at 10⁶ CFU per mouse given once subcutaneously. At time intervals, the spleen derived T cells were tested for Ag85B responsive T cells using IFN-γ coated plates and Elispot assay. DKO vaccination leads to a larger expansion of Ag85B specific T cells. All Elispot numbers represent Ag85B stimulated numbers subtracted from KLH protein stimulated T cells. T cells from naïve mice were stimulated with KLH alone (3 separate experiments, 3 mice per group per time point).