Engineering Mycobacteria for the Production of Self-Assembling Biopolyesters Displaying Mycobacterial Antigens for Use as a Tuberculosis Vaccine

Bacterial strains and cultivation conditions.Bacterial strains used in this study are listed in Table 2. E. coli strains were cultivated in Luria broth (Difco, Detroit, MI). For LB agar, 1.5% (wt/vol) agar was added to liquid. M. smegmatis was cultivated on BBL Middlebrook 7H9 broth or 7H10 agar (BD, Franklin Lakes, NJ, USA) supplemented with 0.2% (vol/vol) glycerol and 10% (vol/vol) OADC (BD). Additionally, 7H9 broth was supplemented with 0.05% Tween 80 (vol/vol).

If required, antibiotics were added to the medium at the following concentrations: ampicillin, 100 μg/ml (E. coli); kanamycin, 30 μg/ml (E. coli and M. smegmatis); and hygromycin, 200 μg/ml (E. coli) or 90 μg/ml (M. smegmatis). Unless stated otherwise, E. coli strains and M. smegmatis mc²155 were cultivated at 37°C with aeration at 200 rpm. When required for protein induction, tetracycline or isovaleronitrile was added at a final concentration of 30 ng/ml or 50 μm/ml, respectively.

Construction of plasmids for the production of PHB.All plasmids and oligonucleotides used can be found in Table 2. All synthesized genes have been codon usage optimized for M. tuberculosis. DNA sequencing was used to confirm all amplified and final plasmids. Several different plasmids were generated for the expression of PHA synthase alone (phaC) or as a fusion protein with antigens Ag85A-ESAT-6 utilizing either the nitrile-inducible (pMycVec1 and pMV261 derivatives) or the tetracycline-inducible (pMIND derivatives) promoter. A single plasmid was generated for the expression of PHB precursor genes (phaAB) regulated under a weak constitutive M. smegmatis promoter (Pwmyc).

A one-plasmid system was also created which carried all three genes, with phaC regulated under the nitrile promoter (pNit) and genes phaA and phaB under Pwmyc utilizing the pMV261 vector (31) for comparison with the two-plasmid system for protein and PHB production.

(i) Plasmid pMycVec1_pNit-phaC.This plasmid carries the gene for the class I PHA synthase (phaC) regulated under the strong nitrile-inducible promoter pNit (pNit-1::gfp was kindly provided by Chris Sassetti, University of Massachusetts, Worcester, MA). First, the nitrile promoter- and ext-gfp-carrying fragment was amplified from plasmid pNit-1::gfp (31) using primers fwd_BamHI_pNit and rev_pNit_XbaI. The resultant fragment was ligated into intermediate cloning vector pGEM-T Easy (Promega, Madison, WI, USA). Following confirmation, the pNit-gfp fragment was excised using BamHI and XbaI and successively ligated with vector pET-16b. NdeI and SwaI were used to replace the ext-gfp-carrying fragment with the M. tuberculosis codon-optimized gene phaC with complementary sites isolated from pUC57_phaC. The DNA fragment carrying pNit_phaC from plasmid pET-16b_pNit-phaC was then excised using BamHI and XbaI and subsequently ligated with E. coli-Mycobacterium shuttle plasmid pMycVec1.

(ii) Plasmid pMycVec2_Pwmyc-phaAB.The E. coli codon usage-optimized genes phaA and phaB required for precursor synthesis in the two-plasmid system were amplified from plasmid pMCS69 (50) using primers fwd_NheI_phaAB and rev_phaAB_PacI and subsequently ligated into pGEM-T easy. The amplified fragment containing phaA and phaB was excised using NheI and PacI following confirmation by DNA sequencing. A synthesized DNA fragment containing the weak constitutive promoter Pwmyc from M. smegmatis was excised from pGEM-T_Pwmyc using EcoRV and NheI. DNA fragments carrying precursor genes (phaA and phaB) and Pwmyc were used in a single ligation reaction with E. coli-Mycobacterium shuttle plasmid pMycVec2 linearized with EcoRV and PacI.

(iii) Plasmid pMIND_pTet-phaC.The DNA fragment carrying codon-optimized gene phaC was excised from plasmid pUC57_phaCAB using BamHI and HindIII. The fragment was subsequently ligated into the corresponding sites of plasmid pMIND (30) downstream of the tetracycline-inducible promoter (pTet).

(iv) Plasmid pMV261_pNit-phaC.Plasmid pNit-1::gfp was hydrolyzed with the NdeI and HindIII excising reporter ext-gfp to allow the ligation of codon-optimized gene phaC from plasmid pUC57_phaC into the corresponding sites.

(v) Plasmid pMV261_pNit-A:E-phaC.The DNA fragment encoding fusion antigens Ag85A-ESAT-6 was amplified from existing plasmid pHAS-Ag85A-ESAT-6 (23) using primers fwd_A:E and rev_A:E. The resultant fragment was then ligated into vector pGEM-T. Subsequently, the DNA fragment encoding Ag85A-ESAT-6 was excised using NdeI and ligated into the corresponding sites in plasmid pMV261_pNit-phaC located downstream of the pNit promoter and upstream of codon-optimized phaC.

(vi) Plasmid pMV261_pNit-phaC-Pwmyc-phaAB.pMV261_pNit-phaC-Pwmyc-phaAB is a single vector system utilizing vector pMV261. First, the PacI site was introduced downstream of phaB in plasmid pUC57_phaCAB using a small synthesized DNA fragment from plasmid pUC57_3′phaB_PacI carrying a short segment of the 3′ end of phaB and the PacI site. This allowed the subcloning of the entire DNA fragment carrying codon-optimized phaC and precursor genes (phaA and phaB) regulated under a weak constitutive mycobacterial promoter Pwmyc. The fragment was excised from plasmid pUC57_3′phaB_PacI using XhoI and SpeI and inserted into the corresponding sites of plasmid pUC57_phaC-Pwmyc-phaAB. Consequently, the DNA fragment encoding PHA synthase and carrying the precursor genes was excised from pUC57_phaC-Pwmyc-phaAB_PacI using NdeI and PacI and ligated into the corresponding sites in vector pMV261_pNit-phaC downstream of the pNit promoter, resulting in the replacement of the existing phaC gene.

PHB-accumulating growth conditions.To assess the ability of M. smegmatis to accumulate PHB in vivo, strains were cultivated under PHB-accumulating conditions. Ten milliliters of 7H9 broth supplemented with 1% glucose (vol/vol) as the carbon source and antibiotics was inoculated with frozen glycerol stock cultures and cultivated for 20 h. Then 50 ml of fresh 7H9 broth supplemented with antibiotics was added to the preculture and cultivated for another 20 h. Large 1-liter cultures of 7H9 broth supplemented with antibiotics were inoculated with 3% (vol/vol) preculture and cultivated to an optical density at 600 nm (OD₆₀₀) of 0.3 to 0.4. Once the OD₆₀₀ was reached, protein expression was induced by the addition of isovaleronitrile at a final concentration of 50 μm and then further cultivation was performed for 72 h.

Isolation of MBB.Forty-eight-hour postinduction cultures were harvested by centrifugation at 9,000 × g for 20 min. Sediment was washed once and then suspended as a 25% slurry (wt/vol) with 1× PBS (pH 7.4). Then 1 ml of the 25% slurry was aliquoted into 2-ml screw-cap tubes containing 1/3 volume or 500 μl (vol/vol) acid-washed 0.1-mm glass beads and chilled on ice for 10 min prior to cell lysis. Lysis was achieved by using a Hybaid RiboLyser (FP120HY-230) with the following procedure: 6 m/s for 3 1-min intervals with 2-min cooling steps in between each interval. Lysates was separated from glass beads by centrifuging briefly at 3,000 × g for 10 s. Then 1 ml of the 25% lysate was loaded onto a glycerol gradient consisting of 44%/66%/90% (vol/vol) glycerol layers made in 1× PBS (pH 7.4). Separation was achieved by centrifugation at 100,000 × g for 1.5 h, and MBB were recovered from the 66% and 90% glycerol interfaces. Isolated MBB were washed two times using 1× PBS (pH 7.4) with centrifugation at 9,500 × g for 30 min and then formulated in 1× PBS (pH 7.4) as a 20% (wt/vol) slurry. To ensure sterile preparations, washed MBB were heat treated in sterile 2-ml screw-cap tubes at 80°C for 30 min. Sterility was confirmed by cell culture on antibiotic-free supplemented BBL Middlebrook 7H10 agar (BD).

PHB analysis by GC/MS.Lyophilized material (40 to 60 mg) was suspended in 2 ml of chloroform and subjected to methanolysis in 2 ml of methanol in the presence of 15% (vol/vol) sulfuric acid. Methanolysis was performed in a heated oil bath for 5 h at 100°C. After methanolysis, tubes were cooled to room temperature, 2 ml of distilled water was added, and the tubes were briefly vortexed. Samples were then left at room temperature for phase separation. The bottom phase, containing methyl esters of the corresponding fatty acid constituents, was recovered and analyzed by GC/MS for 3-hydroxyalkanoate methyl esters.

TEM.Transmission electron microscopy (TEM) analysis was used to assess isolated PHA biobead material produced in M. smegmatis. Samples were processed for analysis as described previously (51).

Protein analysis.Proteins were separated using SDS-PAGE and visualized by staining with Coomassie brilliant blue. Immunoblotting was used to confirm PhaC or Ag85A-ESAT-6-PhaC fusion protein on MBB or the production of GFP. Protein bands separated by SDS-PAGE were transferred to nitrocellulose membranes using the iBlot system (Invitrogen, Carlsbad, CA). Membranes were blocked with 1% skim milk in phosphate-buffered saline with Tween 20 (PBST) for 1 h. Following washing with PBST, primary antibodies were diluted in 1% bovine serum albumin (BSA) and used accordingly: for detection of PhaC, 1:20,000 rabbit polyclonal (GenScript, NJ); ESAT-6, 0.1 μg/ml rabbit polyclonal (Abcam, Cambridge, United Kingdom); and GFP, 0.75 μg/ml rabbit polyclonal (A01388; GenScript, NJ). Following incubation for 1 h, the membrane was washed three times with PBST for 5 min. Secondary antibody anti-rabbit horseradish peroxidase (HRP) at 1:25,000 (Ab6721; Abcam, UK) was diluted in 1% BSA, added, and incubated for 1 h. Following three PBST washes, development was carried out using SuperSignal West Pico chemiluminescent substrate (Thermo Fisher, Waltham, MA).

The BCA protein assay kit (23227; Thermo Scientific, IL) was used according to the manufacturer's instructions to quantify the total protein in the isolated MBB material.

Mouse immunization trial.To evaluate the efficacy of MBB as a vaccine for tuberculosis a mouse immunology trial was performed. All animal experiments were approved by the Grasslands Animal Ethics committee (approval number 13100; Palmerston North, New Zealand).

Animals.C57BL/6 female mice, aged 6 to 8 weeks at the start of the experiment, were obtained from the AgResearch Ruakura Small Animal Unit (Hamilton, New Zealand). The animals were assigned to one of five vaccination groups (12 animals per group) and housed at AgResearch's PC2 Ulyatt-Reid Small Animal Facility (Palmerston North, New Zealand). The animals were kept in a separate room of this PC2 facility but not under specific-pathogen-free conditions.

Vaccination.One control vaccination group received three biweekly, subcutaneous injections of PBS. Another control group received three biweekly, subcutaneous injections of M. smegmatis vector control (MVC) lysate, which did not contain any MBB. Two further groups were vaccinated with three biweekly, subcutaneous injections of either MBB or MBB displaying Ag85A-ESAT-6 (A:E-MBB). Each of these vaccine preparations was applied at 300 μg of total protein in 200 μl of PBS. Finally, mice in another control group (BCG) received a single dose of 10⁶ CFU of BCG Pasteur strain 1173P2 (kindly provided by the Malaghan Institute of Medical Research, Wellington, New Zealand).

Cell-mediated immune response.Seven weeks after the first immunization, seven mice from each group were euthanized and splenocytes were prepared for further analysis. Splenocytes were stimulated in vitro using various antigen preparations at a final concentration of 5 μg/ml: (i) PBS (as an unstimulated negative control), (ii) concanavalin A (ConA) (C-0412 [Sigma-Aldrich], a mitogen, used as a positive control [data not shown]), (iii) purified protein derivative from M. bovis (PPD-B) (Prionics AG, Switzerland), (iv) MBB, (v) peptides from Ag85A (amino acids [aa] 99 to 118, aa 145 to 152) and ESAT-6 (aa 1 to 16, aa 9 to 24, aa 17 to 32, aa 57 to 72, and aa 80 to 95), or (vi) PhaC peptides (aa 110 to 118 and aa 118 to 126). Supernatants from these cultures were harvested, and the amounts of secreted IFN-γ and IL-17 were determined by ELISA.

Statistical analysis.For the analysis of cytokine measurements from culture supernatants of restimulated splenocytes, one-way analysis of variance (ANOVA) was carried out with 5 treatments (vaccine levels) separately for IL-17 and IFN-γ and separately for in vitro restimulation with PPD-B, MBB, and peptide subsets. The ANOVA assumptions (normality, homogeneity of variances, etc.) were examined using model residuals and fitted values via diagnostic graphs and Shapiro's test for normality and Levene's test for homogeneity of variances. Transformations were required, and the log (natural) transformations gave reasonably satisfactory ANOVA assumptions. All analyses were carried out using R software, version 3.3.1 (52).