Fig 1.
K. kingae binds FH and/or FHL-1 from human serum.
Binding of FH and/or FHL-1 to the bacterial surface of K. kingae strains KK01, KK01ΔcsaA, KK01Δpam, and KK01ΔcsaAΔpam and E. coli strain DH5α was determined using flow cytometry. Bacteria were incubated with 10% or 50% HI-NHS for 1 h or with PBS (0% HI-NHS) as a primary and secondary antibody-only control. Cells were stained with propidium iodine (PI) prior to analysis; 50,000 events per biological replicate were analyzed. A total of three biological replicates were performed (n = 3). (A) Representative histograms are shown. Histograms: KK01, red; KK01ΔcsaA, purple; KK01Δpam, green; KK01ΔcsaAΔpam, black; DH5α, blue. Cells were gated for PI+ and DyLight-488+ (DL-488+) cells. One representative experiment out of three performed is shown. (B) Quantified data from panel A. The bars from the 0% HI-NHS group are negligible in size due to low signal. The bars from all serum concentrations for strain DH5α are negligible in size due to low signal. Data are presented as means, and the error bars represent the standard error of the mean. Statistical significance was determined by 2-way analysis of variance (ANOVA) with Tukey’s correction for multiple comparisons. *, P < 0.05; ****, P < 0.0001.
Fig 2.
K. kingae binds human FH to resist rat complement-mediated killing.
K. kingae strains (A) KK01 and (B) KK01ΔcsaAΔpam (~103 CFU) were incubated with 1%, 2%, 5%, 10%, or 25% normal rat serum (NRS) or heat-inactivated NRS (HI-NRS) with either 0 µg/mL or 100 µg/mL of human factor H (hFH) for 1 h. (A – B) The survival ratio was calculated by dividing NRS CFU counts by the HI-NRS CFU counts. A total of 3 biological replicates were performed (n = 3). The survival ratios are negligible for strain KK01 incubated with 0 µg/mL hFH and rat serum concentrations above 2% and for strain KK01ΔcsaAΔpam incubated with 0 µg/mL hFH and rat serum concentrations above 1%. Data are presented as means, and the error bars represent the standard error of the mean. Statistical significance was determined by 2-way analysis of variance (ANOVA) with Sidak’s correction for multiple comparisons. ***, P < 0.0005; ****, P < 0.0001.
Fig 3.
Treatment with human FH results in enhanced K. kingae virulence in infant rats.
The graph plots Kaplan-Meier survival curves for five-day-old Sprague-Dawley rats inoculated via the intraperitoneal (i.p.) route with either 50 µg human FH (hFH) in 0.15mL PBS, 1 x 107 CFU of K. kingae strain KK01 in PBS, 1 x 107 CFU of KK01 with 50 µg hFH, 1 x 107 CFU of KK01ΔcsaAΔpam in PBS, or 1 x 107 CFU of KK01ΔcsaAΔpam with 50 µg hFH. Data are for 16 animals inoculated with hFH, 17 animals inoculated with KK01 or KK01 with 50 µg hFH, and 18 animals inoculated with KK01ΔcsaAΔpam or KK01ΔcsaAΔpam with hFH. There was no mortality in the following cohorts: KK01, KK01ΔcsaAΔpam, and hFH. Statistical significance was determined by Log-rank (Mantel-Cox) test: KK01 & KK01 + hFH (****, P < 0.0001); KK01ΔcsaAΔpam & KK01ΔcsaAΔpam + hFH (****, P < 0.0001); KK01 + hFH & KK01ΔcsaAΔpam + hFH (ns, P > 0.05); KK01 & hFH (ns, P > 0.05).
Fig 4.
A K. kingae outer membrane protein binds human FH.
Outer membrane preparations were isolated from K. kingae strains KK01 and KK01ΔcsaAΔpam and E. coli strain DH5⍺. These outer membrane preparations were separated on a 12.5% SDS-PAGE gel and transferred to a nitrocellulose membrane or stained with Coomassie blue. A far-western blot was performed by incubating the membrane with 100 μg/mL purified human FH, anti-human FH mAb (OX-24), and an HRP-coupled anti-mouse IgG. One band of ~37-kDa was present for KK01 and KK01ΔcsaAΔpam in the far-western blot. DH5α and bovine serum albumin (BSA) served as negative binding controls. The black arrow next to the Coomassie-stained gel represents the ~ 37-kDa protein band that binds human FH. Representative images are shown.
Fig 5.
Outer membrane protein KK02920 is responsible for K. kingae binding of human FH.
(A) Binding of factor H to the bacterial surface of K. kingae strains KK01, KK01Δ02010, KK01Δ02920, KK01Δ02920(02920), KK01ΔcsaAΔpam, KK01ΔcsaAΔpamΔ02010, KK01ΔcsaAΔpamΔ02920, and KK01ΔcsaAΔpamΔ02920(02920) and E. coli strain DH5α was determined using flow cytometry. Bacteria were incubated with 50% HI-NHS for 1 h or with PBS (0% HI-NHS) as a primary and secondary antibody-only control. Cells were stained with propidium iodine (PI) prior to analysis; 50,000 events per biological replicate were analyzed. The gating strategy demonstrated in Fig 1 is the exact gating strategy used here. Representative histograms are shown: KK01 and KK01ΔcsaAΔpam, purple; KK01Δ02010 and KK01ΔcsaAΔpamΔ02010, green; KK01Δ02920 and KK01ΔcsaAΔpamΔ02920, orange; KK01Δ02920(02920) and KK01ΔcsaAΔpamΔ02920(02920), black; DH5α, blue. Top left histogram: strain KK01 background treated with 0% HI-NHS; bottom left histogram: strain KK01ΔcsaAΔpam background treated with 0% HI-NHS; top right histogram: strain KK01 background treated with 50% HI-NHS; bottom right histogram: strain KK01ΔcsaAΔpam background treated with 50% HI-NHS. Cells were gated for PI+ and DyLight-488+ (DL-488+) cells. One representative experiment out of three performed is shown (n = 3). (B) Quantified data from panel A. The bars from the 0% HI-NHS group are negligible in size due to low signal. The bars from both serum concentrations for strain DH5α are negligible in size due to low signal.The percentages represent events that registered as DyLight 488 positive (DL-488+). A total of three biological replicates were performed (n = 3). The bars from the 0% HI-NHS group are negligible in size due to low signal. Data are presented as means, and the error bars represent the standard error of the mean. Statistical significance was determined by 2-way analysis of variance (ANOVA) with Tukey’s correction for multiple comparisons. ***, P < 0.001; ****, P < 0.0001. (C) Outer membrane preparations were isolated from K. kingae strains KK01ΔcsaAΔpam, KK01ΔcsaAΔpamΔ02010, KK01ΔcsaAΔpamΔ02920, and KK01ΔcsaAΔpamΔ02920(02920) and E. coli strain DH5⍺. These outer membrane preparations were separated on a 12.5% SDS-PAGE gel and transferred to a nitrocellulose membrane or stained with Coomassie blue. A far-western blot was performed by incubating the membrane in 10% HI-NHS, anti-human FH mAb (OX-24), and an HRP-coupled anti-mouse IgG. The ~ 37-kDa band represents the outer membrane protein that binds human FH. The ~ 100-kDa band represents a loading control. A representative image is shown.
Fig 6.
Enhanced K. kingae virulence in the presence of human FH is dependent on KK02920.
The graph plots Kaplan-Meier survival curves of five-day-old Sprague-Dawley rats inoculated via the intraperitoneal (i.p.) route with either 1 x 107 CFU of K. kingae strain KK01ΔcsaAΔpam in PBS, 1 x 107 CFU of KK01ΔcsaAΔpam with 50 µg human factor H (hFH), 1 x 107 CFU of KK01ΔcsaAΔpamΔ02920 in PBS, 1 x 107 CFU of KK01ΔcsaAΔpamΔ02920 with 50 µg hFH, 1 x 107 CFU of KK01ΔcsaAΔpamΔ02920(02920) in PBS, or 1 x 107 CFU of KK01ΔcsaAΔpamΔ02920(02920) with 50 µg hFH. Data are for 16 animals inoculated with KK01ΔcsaAΔpam hFH or 18 animals for all other cohorts. Statistical significance was determined by Log-rank (Mantel-Cox) test: KK01ΔcsaAΔpam & KK01ΔcsaAΔpam + hFH (****, P < 0.0001); KK01ΔcsaAΔpamΔ02920 & KK01ΔcsaAΔpamΔ02920 + hFH (not significant, P > 0.05); KK01ΔcsaAΔpamΔ02920(02920) & KK01ΔcsaAΔpamΔ02920(02920) + hFH (**, P < 0.005); KK01ΔcsaAΔpam + hFH & KK01ΔcsaAΔpamΔ02920 + hFH (****, P < 0.0001); KK01ΔcsaAΔpamΔ02920 + hFH & KK01ΔcsaAΔpamΔ02920(02920) (**, P < 0.005); KK01ΔcsaAΔpam + hFH & KK01ΔcsaAΔpamΔ02920(02920) + hFH (*, P < 0.05).
Fig 7.
KK02920 plays a major role in inhibiting complement-mediated serum killing of K. kingae by resisting the alternative complement pathway.
(A) K. kingae strains KK01, KK01Δ02920, KK01 Δ02920(02920), KK01ΔcsaAΔpam, KK01ΔcsaAΔpamΔ02920, KK01ΔcsaAΔpamΔ02920(02920), and E. coli strain DH5α (~106 CFU) were incubated with 50% NHS for 1 h. (B) K. kingae strains KK01, KK01Δ02920, KK01Δ02920(02920), KK01ΔcsaAΔpam, KK01ΔcsaAΔpamΔ02920, and KK01ΔcsaAΔpamΔ02920(02920) (~106 CFU) were incubated with 50% NHS plus 5mM EGTA and 9mM Mg2+ for 1 h. The percent (%) survival was calculated by dividing the NHS CFU counts by the HI-NHS CFU counts and multiplying by 100. A total of 3 biological replicates were performed (n = 3). Data are presented as means, and the error bars represent the standard error of the mean. Statistical significance was determined by 1-way analysis of variance (ANOVA) with Tukey’s correction for multiple comparisons. *, P < 0.05; ***, P < 0.001; ****, P < 0.0001.
Fig 8.
FH retains cofactor activity when bound to K. kingae, and this activity is dependent on KK02920.
K. kingae strains KK01ΔcsaAΔpam and KK01ΔcsaAΔpamΔ02920 at an inoculum of 1 x 106 CFU were preincubated with 10 µg/mL FH for 1 h. Bacteria were subsequently incubated with 10 µg/mL C3b and 2 µg/mL factor I (FI) for 1 h. Samples without FH or FI were included as controls. Presence (+) and absence (-) of each component is listed above the image. Samples were separated on a 10% SDS-PAGE gel and transferred to a nitrocellulose membrane. A western blot was performed to detect the degradation of C3b by incubating the membrane with an anti-human C3 pAb and an HRP-coupled anti-goat IgG. The arrows indicate the position of the two intact C3b chains (α′ and β) and the degradation fragments (α′ 43-kDa and α′ 41-kDa). A representative image is shown.
Fig 9.
A collection of K. kingae clinical isolates bind human FH to resist rat complement-mediated killing.
K. kingae strains KK01, KK01∆02920, and clinical isolates were evaluated for (A) FH binding via far-western blot analysis and for (B) resistance to rat complement-mediated killing via FH binding. Abbreviations: 93, K. kingae clinical isolate PYKK93; 46, K. kingae clinical isolate KK146; 58, K. kingae clinical isolate PYKK58; 30, K. kingae clinical isolate ATCC 23330; 98, K. kingae clinical isolate PYKK98; 39, K. kingae clinical isolate E3339; 60, K. kingae clinical isolate PYKK60. 70, K. kingae clinical isolate BB270. (A) Outer membrane preparations were isolated and separated on a 12.5% SDS-PAGE gel and transferred to a nitrocellulose membrane or stained with Coomassie blue. A far-western blot was performed by incubating the membrane in 10% HI-NHS, anti-human FH mAb (OX-24), and an HRP-coupled anti-mouse IgG. The band between 35 and 40 kDa for each strain represents the outer membrane protein that binds human FH. The ~ 100-kDa band represents a loading control. A representative image is shown. (B) K. kingae strains (~103 CFU) were incubated with 10% normal rat serum (NRS) or heat-inactivated NRS (HI-NRS) with either 0 µg/mL or 20 µg/mL of human factor H (hFH) for 1 h. The survival ratio was calculated by dividing NRS CFU counts by the HI-NRS CFU counts. A total of 3 biological replicates were performed (n = 3). The survival ratios are negligible for bacteria incubated with 0 µg/mL hFH and for strain KK∆02920 incubated with 20 µg/mL hFH. Data are presented as means, and the error bars represent the standard error of the mean. Statistical significance was determined by 2-way analysis of variance (ANOVA) with Sidak’s correction for multiple comparisons. **, P < 0.005; ***, P < 0.0005; ****, P < 0.0001.
Table 1.
Strains and plasmids used in this study.
Table 2.
Primers used in this study.