Article Figures & Data
Figures
- FIG. 1.
Construction and characterization of LVS ΔcapB. (A) Construction of LVS ΔcapB. LVS ΔcapB was constructed by allelic exchange between LVS (top) and a plasmid carrying an exchange cassette containing truncated capB (FTL_1416/FTT0805) and its flanking nucleotide sequences, including the upstream region of FTL_1417 (FTL_1417′) and the downstream region of capC and capA (capCA′) (middle). The resultant LVS ΔcapB strain (bottom) retains 6 amino acids from the N terminus and 4 amino acids from the C terminus of CapB and contains no antibiotic resistance marker (unmarked). (B) Schematic representations of the structure of the capB locus and its upstream and downstream regions in the genomes of LVS (top) and LVS ΔcapB (bottom), the position of a probe spanning the EcoRV restriction site located in FTL_1417 and the EcoRV site located in capA, and the predicted EcoRV, HindIII, and AvaII restriction fragments (in kilobases) detected by the probe by Southern blot analysis. A, restriction enzyme AvaII; E, EcoRV; H, HindIII. (C) Southern blot analysis of LVS ΔcapB (Δ) and the parental LVS genomic DNA. The genomic DNAs of LVS (lanes 1, 3, and 5, indicated below the panels) and LVS ΔcapB (lanes 2, 4, and 6) were isolated from bacteria cultured on chocolate agar for 2 days. The DNA was digested with EcoRV, HindIII, or AvaII, as indicated at the top of each panel; electrophoresed with a DNA ladder (L); blotted; and hybridized with the biotin-labeled probe as shown in B and the biotin-labeled DNA ladder. The left of each panel shows a 1-kb Plus DNA ladder, and to the right of each panel are the fragments resulting from restriction enzyme digestion of the bacterial DNA detected by the probe. (D and E) Quantitative RT-PCR analysis of capB and its neighboring genes. Total RNA was isolated from LVS, LVS ΔcapB, and the capB-complemented strain LVS ΔcapB/CapB; reverse transcribed in the presence (+) or absence (−) of reverse transcriptase; and amplified by using primer pairs specific to capB, its upstream region of FTL_1417, and downstream regions of capC and capA. 16S rRNA (16S) was used as an internal control. Transcript levels are presented as the mean fold changes versus LVS ± SE. ***, P < 0.001 by two-way ANOVA. qRT-PCR products were analyzed on a 2% agarose gel. The predicted sizes of the qRT-PCR product for capA, capB, capC, FTL_1417, and 16S rRNA are 231, 231, 212, 209, and 156 bp, respectively. The sizes of the DNA in the DNA ladder are shown at the left of each panel in E. F. tularensis strains from which the RNA was isolated are shown at the right of each panel in E.
- FIG. 2.
In vitro characterization of LVS ΔcapB. (A) Analysis of LVS and LVS ΔcapB LPS and proteins. (Left) LPS was analyzed by immunoblotting using a monoclonal antibody to LPS (note the characteristic LPS ladder pattern). (Right) Proteins were evaluated by Coomassie blue staining to provide a loading control. (B) Complement sensitivity assay. Each bacterial strain was incubated with either fresh human AB serum (ABS) or heat-inactivated AB serum (HI-ABS) for 10 min and then assayed for CFU on chocolate agar. This experiment was performed three times, with similar results. (C) Competition for growth in human macrophage-like THP-1 cells. THP-1 cells were coinfected with LVS ΔcapB carrying a hygromycin resistance gene (ΔcapB-h) and with LVS carrying a kanamycin resistance gene (LVS-k) (solid circles, left vertical axis) or with the same two strains carrying the opposite resistance markers (solid squares, right vertical axis). At 0, 7, and 22 h postinfection, the cell monolayer was lysed, serial dilutions of the lysate were plated onto chocolate agar supplemented with either kanamycin or hygromycin, CFU were enumerated, and the ratio between the numbers of LVS ΔcapB and LVS CFU was calculated. The ratios of LVS ΔcapB to LVS CFU on the left and right vertical axes differ because of the variability among the four strains in the numbers of CFU recovered from THP-1 monolayers at the start of the experiment. For the competition data shown on the left axis, the initial number of LVS ΔcapB organisms was greater than the initial number of LVS organisms (initial ratio of ∼10:1), and for the competition data shown on the right axis, the initial number of LVS ΔcapB organisms was fewer than the initial number of LVS organisms (initial ratio of ∼0.4:1). In both cases, LVS ΔcapB was outcompeted by LVS, and therefore, the ratio of LVS ΔcapB CFU to LVS CFU declined with time in culture. This experiment was performed twice, with similar results.
- FIG. 3.
LVS ΔcapB is more attenuated than parental LVS and complemented LVS ΔcapB/CapB in mice. Groups of four BALB/c mice were nonimmunized, immunized with PBS (sham), or immunized i.n. or i.d. with LVS, LVS ΔcapB, or LVS ΔcapB/CapB at the indicated doses (CFU) and monitored for survival, weight change, local lesions (after i.d. immunization), and other signs of illness for 2 or 3 weeks. (A1 and A2) Mice were weighed daily from days 2 to 9 postimmunization and monitored for survival for 3 weeks. (B1 to B3) Mice were weighed daily from day 0 (the day of immunization) to day 7 postimmunization and monitored for survival for 2 weeks. Values are means ± SE. RF, ruffled fur; HB, hunched back; SE, swollen eyes.
- FIG. 4.
Dissemination and clearance of LVS ΔcapB in mice. Groups of four BALB/c mice were immunized i.n. with 112 CFU LVS or 1 × 105 CFU LVS ΔcapB or i.d. with 1 × 105 CFU LVS or 1 × 106 CFU LVS ΔcapB. At the indicated times postimmunization, mice were euthanized, and numbers of CFU in the spleen, liver, lung, and skin at the site of injection (after i.d. immunization) were assayed. LVS ΔcapB is cleared faster than LVS in spleen, liver, and lung after i.n. immunization and at the site of injection in the skin after i.d. immunization. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (by ANOVA).
- FIG. 5.
Immunization and challenge protocols. Groups of BALB/c mice were immunized i.n. or i.d. with LVS ΔcapB or LVS. Mice immunized with PBS (sham) or LVS served as controls. At the indicated times postimmunization, mice were either euthanized for immunology studies (A), challenged with approximately 4,000 CFU LVS (>5× the LD50) by the i.n. route (B and C), or challenged with 10× the LD50 of type A F. tularensis strain SchuS4 by the aerosol route (D). For assessments of organ bacterial burden, four or eight mice per group challenged with LVS i.n. were euthanized at 5 days postchallenge, and the organs were homogenized and assayed for CFU of LVS (B). For assessments of survival, eight mice per group challenged with LVS i.n. or SchuS4 by aerosol were monitored for signs of illness and death for 3 weeks after challenge (C and D). LPA, lymphocyte proliferation assay.
- FIG. 6.
Immunization with LVS ΔcapB induces cell-mediated immune responses comparable to those induced by LVS. Groups of four BALB/c mice were immunized i.n. with 112 CFU LVS or 1 × 105 CFU LVS ΔcapB or i.d. with 1 × 105 CFU LVS or 1 × 106 CFU LVS ΔcapB. At 4 weeks postimmunization, mice were euthanized, and single-cell suspensions of splenocytes were prepared. A total of 1 × 106 splenocytes were incubated with 2 × 105 or 2 × 106 CFU heat-inactivated (HI) LVS as indicated, and splenic lymphocyte proliferation was assayed. The lines above the bars indicate a statistical comparison between the bar beneath the left end of the line and the bar beneath the right end of the line. Only comparisons where differences are statistically significant are shown. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (by two-way ANOVA).
- FIG. 7.
Immunization with LVS ΔcapB induces potent antibody responses. Groups of four BALB/c mice were immunized i.n. with 112 CFU LVS or 1 × 105 CFU LVS ΔcapB or i.d. with 1 × 105 CFU LVS or 1 × 106 CFU LVS ΔcapB. At the indicated times postimmunization, mice were euthanized, and serum was collected and analyzed for IgG, IgA, IgM, IgG1, and IgG2 antibodies specific to heat-inactivated LVS. The antibody level was calculated as the log10 of the reciprocal of the endpoint dilution of the test serum. Data represent means ± SE. The lines above the bars indicate a statistical comparison between the bar beneath the left end of the line and the bar beneath the right end of the line. Only comparisons where differences were statistically significant between LVS and LVS ΔcapB by the same route are shown. *, P < 0.05; ***, P < 0.001 (by two-way ANOVA).
- FIG. 8.
Immunization with LVS ΔcapB induces protective immunity against F. tularensis LVS i.n. challenge in a dose-dependent manner. BALB/c mice (four mice per group) were immunized i.n. (left) or i.d. (right) with LVS ΔcapB at the doses indicated on the horizontal axes. Mice immunized with PBS (sham) (eight per group) or LVS (four per group) served as controls. Four weeks later, the mice were challenged i.n. with 4,000 CFU LVS. At 5 days postchallenge, the spleen (A and B), liver (C and D), and lung (E and F) were removed and assayed for bacterial burden. Symbols represent each animal in a group. CFU data were compared by one-way ANOVA (Prism 5.01 software) with Bonferroni's posttest. Dashed lines indicate the limit of detection. Differences in CFU between the sham group and each of the groups vaccinated with either LVS or LVS ΔcapB were statistically significant (P < 0.001). The lines above the bars indicate a statistical comparison between the bar beneath the left end of the line and the bar beneath the right end of the line. Only comparisons between the LVS-immunized group and the LVS ΔcapB-immunized group where differences were statistically significant are shown.
- FIG. 9.
Immunization with LVS ΔcapB induces protective immunity against i.n. F. tularensis LVS challenge comparable to that of the parental LVS. Groups of 12 mice were sham immunized, immunized i.d. with 1 × 105 CFU LVS or 1 × 106 CFU LVS ΔcapB, or immunized i.n. with 200 CFU LVS or 1 × 105 CFU LVS ΔcapB. Four weeks (A and C) or 8 weeks (B and D) later, mice were challenged i.n. with 4,000 CFU LVS. (A and B) At 5 days postchallenge, four mice per group were euthanized, and the spleen, liver, and lung were assayed for bacterial burden. CFU values are shown as means ± SE. The dashed line indicates the limit of detection. The mean CFU in each organ among different groups was analyzed by grouped ANOVA with Bonferroni's posttest (Prism 5.01 software). The lines above the bars indicate a statistical comparison between the bar beneath the left end of the line and the bar beneath the right end of the line. Only comparisons where differences were statistically significant are shown. **, P < 0.01; ***, P < 0.001. (C and D) The remaining mice in each group were monitored for survival and signs of illness for 3 weeks. The differences in survival between the mice in the vaccinated groups and the mice in the sham-vaccinated group were evaluated by using a log-rank (Mantel-Cox) test (Prism 5.01). ***, P < 0.0001 versus Sham. a, 3 of 12 mice (25%) died after i.n. immunization with LVS; of the remaining 9 mice, 4 were studied for the data shown in A, and 5 were studied for the data shown in C.
- FIG. 10.
Immunization with LVS ΔcapB induces protective immunity against F. tularensis SchuS4 aerosol challenge. Groups of eight mice were sham immunized, immunized i.d. (A and C) with 1 × 105 CFU LVS or 1 × 106 CFU LVS ΔcapB, or immunized i.n. (B and D) with 200 CFU LVS or 1 × 105 CFU LVS ΔcapB. Six weeks later, all mice were challenged by the aerosol route with 10× the LD50 of F. tularensis strain SchuS4. Mice were monitored for weight change (A and B) and survival (C and D) for 3 weeks. Weights at each time point are shown as means ± SE. Mean survival time was calculated by dividing the sum of the survival times of all mice in a group by the total number of mice challenged, with animals surviving until the end of the experiment given a survival time of 21 days, when the experiment was terminated. The differences in survival between the mice in the vaccinated groups and mice in the sham-vaccinated group were evaluated by using a log-rank (Mantel-Cox) test (Prism 5.01 software).
