All of the experiments involving animals were approved and conducted by the Animal Ethical Committee of the Eighth Medical Center of the Chinese PLA General Hospital. Mouse care met the standards of the Experimental Animal Regulation Ordinances defined by the China National Science and Technology Commission.

Ninety-six 6-8 week age of the specific pathogen-free (SPF) female BALB/c mice were purchased from Beijing Vital River Laboratory Animal Technology Company Limited, China, maintained under infection barrier conditions in a negative pressure animal room in the Eighth Medical Center of the Chinese PLA General Hospital, Beijing, China. They fed a sterile commercial mouse diet (Beijing KeAoXieLi Company Limited, China).

MTB H₃₇Rv was provided by National Institutes for Food and Drug Control, Beijing, China.

The MTB H₃₇Rv strain was inoculated on a Lowenstein-Jensen medium (Baso Biotechnology Co., LTD., Zhuhai, Guangdong province, China) and incubated at 37°C for three weeks. Bacterial colonies were collected, weighed, and homogenized in 0.5% Tween 80-saline, then 0.1mg/ml bacterial suspension was prepared and stored at -20°C. Before use, the frozen bacterial suspension was thawed, diluted to 0.5mg/ml, and then further serially diluted 10-fold. 0.1 ml of each dilution was plated in duplicate onto Lowenstein-Jensen plates and incubated at 37°C for 4 weeks. The MTB colonies on each plate were enumerated. The 0.05mg/ml MTB H₃₇Rv suspension contained 5×10⁴ CFUs/ml, each mouse was injected with 0.4 ml through tail vein. In this study, one mouse was challenged with 2×10⁴ CFUs of MTB H₃₇Rv through tail vein injection.

The method of constructing MTB ag85a/b chimeric DNA was described previously (29). Briefly, the DNA encoding amino acids 125-282 in the Ag85B protein of MTB H₃₇Rv were amplified by PCR with specific oligonucleotide primers containing the site of endonuclease enzyme Acc I. PCR product was purified and then inserted into the endonuclease enzyme Acc I site of the ag85a gene that was cloned into eukaryotic expression vector pVAX1 (Invitrogen Life Technologies Corporation, Carlsbad, CA, USA). The length of gene sequence encoding for strong immunogen Ag85A was 888 bp, and the Ag85B was 474 bp. MTB ag85a/b chimeric DNA vaccine was thus constructed by inserting the sequence encoding amino acids 125-282 of Ag85B protein into nucleotide 430-435 (endonuclease enzyme Acc I site) of Ag85A DNA vaccine. In this study, the MTB ag85a/b chimeric DNA was produced and purified by Guangzhou Baiyunhan Baidi Biological Medicine Co.Ltd (Guangdong province, China).

The flow chart of DNA vaccine treatment on the mouse TB model is shown in Figure 1 . Ninety-six female BALB/c mice were challenged with 2×10⁴ colony-forming units (CFUs) of MTB H₃₇Rv through tail vein injection, randomly divided into nine groups as follow: (1) PBS as a negative control (100 μl); (2) 10 μg ag85a/b DNA intramuscular injection (IM) (10 μg in 100 μl PBS); (3) 50 μg ag85a/b DNA IM (50 μg in 100 μl PBS); (4) 100 μg ag85a/b DNA IM (100 μg in 100 μl PBS); (5) 200 μg ag85a/b DNA IM (200 μg in 100 μl PBS);(6) 10 μg ag85a/b DNA IM +EP (10 μg in 100 μl PBS); (7) 50 μg ag85a/b DNA IM +EP (50 μg in 100 μl PBS); (8) 100 μg ag85a/b DNA IM +EP (100 μg in 100 μl PBS); (9) 200 μg ag85a/b DNA IM +EP (200 μg in 100 μl PBS). On the third day after infection, the mice in groups 1 to 5 were immunized intramuscularly, and in groups 6 to 9 were medicated by EP three times at two-week intervals, respectively. EP was performed using a TERESA Gene Delivery Device (TERESA Health Technology Co., LTD, Shanghai, China), set at 36 Voltage and 25 Hz, and six pulses of 10 ms in 3mm depth of thigh muscle of the hind leg of a mouse.

The operation procedure was described previously (35). 500 μl of whole blood was stimulated by adding phorbol 12-myristate 13-acetate (PMA, Cat. No. 8139, BD Biosciences, San Jose, Ca, USA) and ionomycin (Cas.10634, Sigma Chemical Co., St.Louis, Mo, USA) as a positive control or by adding 10μl of 1mg/ml Ag85AB protein and then incubating with 5% CO₂ at 37°C for 4 h in the presence of inhibitor brefeldin A (Cat. No. 347688, BD Biosciences, San Jose, Ca, USA). Samples incubated with brefeldin A alone were used as non-stimulated controls (as a negative control). 200μl of blood was then pipetted directly into a 12 × 75 mm fluorescence-activated cell sorting (FACS) tube containing FITC Hamster Anti-Mouse CD3e (Cat. No. 553061, Clone: 145-2C11, BD Biosciences, San Jose, Ca, USA) and PerCP-Cy™5.5 Rat Anti-Mouse CD8a (Cat. No. 551162, Clone: 53-6.7, BD Biosciences, San Jose, Ca, USA), and incubated at room temperature in the dark for 15 min. Then 60 µl of 1 × permeabilizing solution A (BD IntraSure kit, Cat. No.641776, BD Biosciences, San Jose, Ca, USA) was added into the tube and incubated at room temperature in the dark for 5 min. The erythrocytes were then lysed using 2 ml of 1 × FACS lysing solution (Cat. No. 349202, BD Biosciences, San Jose, Ca, USA) for 12 min before centrifugation at 1500 rpm for 5 min. The supernatant was aspirated and 30 µl of 1 × permeabilizing solution B (BD IntraSure kit, Cat. No.641776, BD Biosciences, San Jose, Ca, USA) and 5 µl of APC Rat Anti-Mouse IFN-γ (Cat. No. 554413, Clone: XMG1.2, BD Biosciences, San Jose, Ca, USA) 5 µl of PE Rat Anti-Mouse IL-4 (Cat. No. 554435, Clone: 11B11, BD Biosciences, San Jose, Ca, USA) were added into the sample tube, 5 µl of APC Rat IgG1 κ isotype control (Cat. No. 554886, Clone: R3-34, BD Biosciences, San Jose, Ca, USA), 5 µl of PE Rat IgG1 κisotype control (Cat. No. 554885, Clone: R3-34, BD Biosciences, San Jose, Ca, USA), respectively, and incubated at room temperature in the dark for 20 min. 1 ml of PBS containing 2% fetal bovine serum (FBS) was added before centrifugation at 1500 rpm for 5 min. The supernatant was aspirated, 500 µl PBS was added, and the cell suspension was transferred into 12 × 75 mm FACS tubes and analyzed within 1 h by flow cytometry. CD4⁺IFN-γ⁺ T cell, CD4⁺IL-4⁺ T cell, CD8⁺IFN-γ⁺ T cell, and CD8⁺IL-4⁺ T cell responding to recombinant Ag85AB proteins were calculated. Cells expressing IFN-γ and IL-4 were presented as a percentage of the total population of CD3⁺ cells. The data were collected using the FACS Calibur flow cytometer (BD Pharmingen) and analyzed using CellQuest software. The representing FACS figures with gating method and representing flow cytometry dot plot were shown in Supplementary Figure 1 .

The blood (100 µl) with Heparin sodium anticoagulant from seven mice per group was pipetted directly into second 12×75 mm FACS tube containing 2 µl of FITC Rat Anti-Mouse CD4 (Cat. No. 553046, Clone: RM4.5, BD Biosciences, San Jose, Ca, USA) and 5 µl of APC Rat Anti-Mouse CD25 (Cat. No. 557192, Clone: PC61, BD Biosciences, San Jose, Ca, USA) and incubated at room temperature in the dark for 15 minutes. Red blood cells were lysed using 1 ml of 1×FACS lysing solution (Cat. No. 349202, BD Biosciences, San Jose, Ca, USA) for 30 minutes before centrifugation at 1200 rpm for 5 minutes. 1 ml of PBS was added into the tube before centrifugation at 1200 rpm for 5 minutes, the supernatant was aspirated, and 1×fixation/permeabilization concentrate (Foxp3/Transcription factor staining buffer set Cat. No. 00-5523, eBioscience, San Diego, Ca, USA) was added into the tube and incubated for 15 minutes at 4°Cin the dark. 1 ml of PBS was added into the tube before centrifugation at 1200 rpm for 5 minutes, the supernatant was aspirated, and 2 ml of 1× permeabilization buffer (FoxP3/Transcription factor staining buffer set Cat. No. 00-5523, eBioscience, San Diego, Ca, USA)was added into the tube before centrifugation at 1200 rpm for 5 minutes. The supernatant was aspirated, and 2 μl normal rat serum was added into the tube and incubated for 15 minutes at room temperature in the dark. 5 µl of PE Anti-Mouse/Rat FoxP3 (Cat. No. 12-5773, Clone: FJK-16s, eBioscience, San Diego, Ca, USA), 2.5 µl of APC Rat IgG1 λ Isotype control (Cat. No. 550884, BD Biosciences, San Jose, Ca, USA) 2.5 µl of PE Rat IgG2a κ isotype (Cat. No. 12-4321, eBiosciences, San Diego, Ca, USA) were added into the tube and incubated for 40 minutes at 4°C in the dark. 2 ml of permeabilization buffer was added into the tube before centrifugation at 1200 rpm for 5 minutes. The supernatant was aspirated, 300 µl of PBS was added into the tube, and the cell suspension was transferred into 12×75 mm FACS tubes and analyzed in an hour by FACS Calibur flow cytometry (BD Pharmingen) and analyzed using CellQuest software. The FACS figures with gating method and representing flow cytometry dot plot were shown in Supplementary Figure 2 .

The mice were sacrificed by cervical dislocation under anesthetic with 5ml dethylether (Beijing Chemical Reagents Company, Beijing, China) at 17-18 days after the third immunization. The tissue suspensions of mouse lungs and spleens were serially diluted 10-fold, and 100 μL suspension dilution was inoculated in duplicate on Lowenstein-Jensen medium plates and cultured at 37°C for four weeks. MTB colonies on the medium were counted, and the results were shown as CFUs per organ.

The mouse lungs tissues paraffin-embedded were sliced into 3-μm thick tissue sections, which were dyed with hematoxylin and eosin, and then examined by a certified and veteran pathologist.

Data are shown as means and standard deviations. Statistical analyses were performed using one-way ANOVA followed by Dunnett’s multiple comparison test and two factors factorial design ANOVA followed by compared with t-test using SAS 9.1 software, and a P-value of <0.05 was considered to be statistically significant.

The results of immunological evaluation on mouse TB model immunized with different doses of MTB ag85a/b DNA vaccine by IM and EP were shown in Figures 2 , 3 . The proportion of CD4⁺ T cells expressing IFN-γ in response to recombinant Ag85AB chimeric protein in the whole blood from all DNA groups were higher than that in the PBS group by flow cytometry, but there was a significant difference among those only in the 200 μg DNA IM group and 10 μg, 50 μg, 100 μg, 200 μg DNA EP groups (P<0.05, P<0.0001, Figure 2A ). There was no significant difference among those in the four doses of DNA IM groups or the DNA EP group (P>0.05). The proportion of CD4⁺ T cells expressing IL-4 in the blood only from the 50μg DNA IM group was significantly higher than that of the PBS group (P<0.05, Figure 2B ).

The proportion of CD8⁺ T cells expressing IFN-γ in response to recombinant Ag85AB chimeric protein in the whole blood from all DNA groups were higher than that of the PBS group, but only the 200 μg DNA EP group was significantly higher than that of the PBS group (P<0.01, Figure 2C ). The proportion of CD8⁺IFN-γ⁺ T cells in the 10μg, or 100 μg or 200 μg DNA EP group was significantly higher than that in the 10 μg or 100 μg or 200 μg DNA IM group (P<0.05), the proportion of CD8⁺IFN-γ⁺ T cells in the 50μg DNA EP group was higher than that in the 50 μg DNA IM group, but there was no significant difference among them (P>0.05). The proportion of CD8⁺ T cells expressing IL-4 in the blood from the 50 μg, 100 μg DNA IM groups and 100 μg, 200 μg DNA EP groups were significantly higher than that in the PBS group (P<0.01, P<0.0001, Figure 2D ). The proportion of CD8⁺IL-4⁺ T cells in the 50μg and 100 μg DNA IM group was significantly higher than those in the 10 μg DNA IM group and 10 μg DNA EP groups (P<0.01, P<0.0001). The proportion of CD8⁺IL-4⁺ T cells in the 100 μg and 200 μg DNA EP group was higher than that in the 10 μg,and 200 μg DNA IM, 10μg and 50μg DNA EP groups (P<0.01, P<0.0001), but there was no significant difference between them (P>0.05).

The proportion of CD4⁺CD25⁺ Treg cells in the whole blood from all DNA groups was significantly lower than that of the PBS group by flow cytometry (P<0.01, P<0.0001, Figure 3A ), but there was no significant difference in the proportion of CD4⁺CD25⁺ FoxP3⁺ Treg of all groups (P>0.05, Figure 3B ). The proportion of CD4⁺CD25⁺ Treg cells in the 10 μg DNA EP group was higher than that in the 10 μg DNA IM group (P<0.001); the proportion of CD4⁺CD25⁺ Treg cells in the 100 μg and 200 μg DNA EP group were significantly lower than those in the same doses of DNA IM group (P<0.05, P<0.0001). The proportion of CD4⁺CD25⁺ Treg cells in the 100 μg DNA IM group was significantly higher than those in the 10 μg DNA IM group and 50 μg, 100 μg, 200 μg DNA EP groups (P<0.05, P<0.001, P<0.0001), but was lower than that in the 200 μg DNA IM group (P>0.05). The proportion of CD4⁺CD25⁺ Treg cells in the 50 μg DNA EP group was significantly lower than those in the 10 μg DNA EP group and 100 μg, 200 μg DNA IM groups (P<0.05, P<0.001), but was significantly higher than that in the 200 μg DNA EP group (P<0.05).

The live bacteria in mouse lungs and spleens were determined at 17-18 days after the third immunotherapy (shown in Figure 4 ). Compared with the PBS group, except the 10 μg DNA IM. group, the CFUs of mouse lungs and spleens in other doses of DNA vaccine groups delivered by IM and EP decreased significantly (P<0.0001, P<0.01, P<0.05). The CFUs of mouse lungs and spleens in10 μg and 50 μg DNA EP groups were significantly lower than those in the same doses of the DNA IM group (P<0.0001, P<0.01, P<0.05). However, the CFUs of mouse lungs and spleens in the 100μg DNA EP group was higher than that in the 100 μg DNA IM. group (P< 0.01, P< 0.0001), and in the 200μg DNA EP group were slightly higher than that in the 200 μg DNA IM group, but there was no significant difference (P>0.05). There was no significant difference in lung CFUs between the 100 μg DNA IM group and the 50μg DNA EP group (P>0.05). These two groups were significantly lower than those in other groups (P<0.01, P<0.0001). The spleen CFU in the 50μg DNA EP group was significantly higher than that in the 100 μg DNA IM group (P<0.01), but lower than other groups, especially significantly lower than the 10 μg and 50 μg DNA IM group (P<0.05, P<0.0001). As a result, the 100 μg DNA IM group and 50 μg DNA EP group had a better therapeutic effect on the mouse TB model.

In the PBS group, the alveolar wall was severely thickened, a large number of lymphocytes infiltrated under the bronchioles mucosa, and large granuloma nodules were formed locally, its percentage of lung lesion area was (65 ± 23)%. In the 10 μ g DNA IM group, the lesions were similar to that in the PBS group, and more granuloma nodules were formed locally, its percentage of lung lesion area was (58 ± 28)%. In the 50 μg, 200 μg DNA IM group, and in the 10 μ g, 100 μg, 200 μg DNA EP groups, the lesions were lighter, the alveolar wall was slightly thickened, small granulomatous nodules were formed locally, with a small amount of lymphocyte infiltration, and the area of the lesions was reduced at different agree, their percentages of lung lesion area were (54 ± 27)%, (52 ± 25)%, (51 ± 27)%, (52 ± 23)% and (55 ± 27)%, respectively. The lesions in the 100 μg DNA IM group and 50 μg DNA EP group were the lightest, and most of the alveolar was relatively clear and had typical structures, their percentage of lung lesion areas were perspectively (39 ± 25)% and (39 ± 33)%, which were significantly lower than that in the PBS group (P<0.05). Representative histopathological changes of nine groups were shown in Figure 5 .