Role of bacterial surface structures on the interaction of Klebsiella pneumoniae with phagocytes

Abstract

Phagocytosis is a key process of the immune system. The human pathogen Klebsiella pneumoniae is a well known example of a pathogen highly resistant to phagocytosis. A wealth of evidence demonstrates that the capsule polysaccharide (CPS) plays a crucial role in resistance to phagocytosis. The amoeba Dictyostelium discoideum shares with mammalian macrophages the ability to phagocytose and kill bacteria. The fact that K. pneumoniae is ubiquitous in nature and, therefore, should avoid predation by amoebae, poses the question whether K. pneumoniae employs similar means to counteract amoebae and mammalian phagocytes. Here we developed an assay to evaluate K. pneumoniae-D. discoideum interaction. The richness of the growth medium affected the threshold at which the cps mutant was permissive for Dictyostelium and only at lower nutrient concentrations the cps mutant was susceptible to predation by amoebae. Given the critical role of bacterial surface elements on host-pathogen interactions, we explored the possible contribution of the lipopolysaccharide (LPS) and outer membrane proteins (OMPs) to combat phagoyctosis by D. discoideum. We uncover that, in addition to the CPS, the LPS O-polysaccharide and the first core sugar participate in Klebsiella resistance to predation by D. discoideum. K. pneumoniae LPS lipid A decorations are also necessary to avoid predation by amoebae although PagP-dependent palmitoylation plays a more important role than the lipid A modification with aminoarabinose. Mutants lacking OMPs OmpA or OmpK36 were also permissive for D. discoideium growth. Except the LPS O-polysaccharide mutants, all mutants were more susceptible to phagocytosis by mouse alveolar macrophages. Finally, we found a correlation between virulence, using the pneumonia mouse model, and resistance to phagocytosis. Altogether, this work reveals novel K. pneumoniae determinants involved in resistance to phagocytosis and supports the notion that Dictyostelium amoebae might be useful as host model to measure K. pneumoniae virulence and not only phagocytosis.

Figure 1. Virulence of K. pneumoniae against D. discoideum can be modulated.

(A) The ability of Dictyostelium to grow on a bacterial lawn was assessed by depositing amoebae (from 10 to 10,000) on a lawn of bacteria grown on SM agar medium. A phagocytosis plaque was observed 5 days later when bacteria were permissive. Bacteria tested were: K. pneumoniae (Kp52145), cps mutant (52145-Δwca _K2; Δwca _K2), or control strain (K. aerogenes). (B) The ability of wild-type K. pneumoniae (Kp52145), cps mutant (52145-Δwca _K2), or control strain (K. aerogenes) to resist predation by D. discoideum was tested on HL5-agar, pure or diluted. 1,000 amoebae were deposited on the bacterial lawns and plaques were recorded 5 days later. (C) The ability of Dictyostelium to grow on a bacterial lawn was assessed by depositing amoebae (from 10 to 10,000) on a lawn of bacteria grown on HL5–5% agar medium. A phagocytosis plaque was observed 5 days later when bacteria were permissive (K. aerogenes and 52145-Δwca _K2; Δwca _K2).

Figure 2. Role of K. pneumoniae CPS on phagocytosis resistance.

(A) Dictyostelium cells were incubated with Klebsiella and the number of surviving bacteria (total or cell-associated) was determined at different times by killing the Dictyostelium and plating the bacteria on LB plates. Bars represent mean ± s.e.m (n = 4). *; P<0.05 (results are significantly different from the results for the wild-type strain [Kp52145]; two tailed t test). (B) Dictyostelium AX2/RFP cells were incubated in the presence of Klebsiella strains, K. pneumoniae (Kp52145), cps mutant (52145-Δwca _K2; Δwca _K2), containing plasmid pFPV25.1Cm for 30 min, and then fixed. Images are representative of three independent experiments. (C) Dictyostelium cells were incubated with Klebsiella and the number of intracellular bacteria was determined at different times by killing the Dictyostelium and plating the bacteria on LB plates. Each point represents the mean and standard deviation of twelve samples from four independent experiments. (D) Klebsiella phagocytois by MH-S mouse alveolar macrophages. Intracellular bacteria were determined by the gentamicin protection assay. Bars represent mean ± s.e.m (n = 4). *; P<0.05 (results are significantly different from the results for the wild-type strain [Kp52145]; two tailed t test). (E) Immunofluorescence confocal microscopy of MH-S mouse alveolar macrophages infected with Klebsiella strains Kp52145 or 52145-Δwca _K2 (Δwca _K2), containing plasmid pFPV25.1Cm. Actin cytoskeleton was stained with Phalloidin-RRX (red) and host cell nuclei were stained with Hoechst (blue). Images are representative of five independent experiments.

Figure 3. Role of K. pneumoniae LPS polysaccharide section on phagocytosis resistance.

(A) K. pneumoniae 52145 (Kp52145) oligosaccharide structure based on a published study . Lines denote the truncation level for the different core biosynthetic gene mutations. Residues J and K could be hydrogen (H) or GalA. (B) Dictyostelium cells were incubated with Klebsiella strains (wild type [Kp52145], LPS OPS mutants [52O21] and 52145-ΔwaaL [ΔwaaL], CPS mutant 52145-Δwca _K2 [Δwca _K2], or strain lacking the CPS and the OPS 52145-Δwca _K2-ΔwaaL [Δwca _K2-ΔwaaL]) and the number of surviving bacteria (total or cell-associated) was determined at different times by killing the Dictyostelium and plating the bacteria on LB plates. Bars represent mean ± s.e.m (n = 4). *; P<0.05 (results are significantly different from the results for the wild-type strain [Kp52145]; two tailed t test). ▵; P<0.05 (results are significantly different from the results for the cps mutant [52145-Δwca _K2]; two tailed t test). (C) Dictyostelium cells were incubated with Klebsiella strains (wild type [Kp52145], LPS OPS mutant 52145-ΔwaaL [ΔwaaL], and LPS core mutants 52145-ΔwaaQ [ΔwaaQ], 52145-ΔwaaL-ΔwaaQ [ΔwaaL-ΔwaaQ], 52145-ΔwabG [ΔwabG] 52145-ΔwabM [ΔwabM], 52145-ΔwabH [ΔwabH], and 52145-ΔwabK [ΔwabK]) and the number of surviving bacteria was determined at different times by killing the Dictyostelium and plating the bacteria on LB plates. Bars represent mean ± s.e.m (n = 4). *; P<0.05 (results are significantly different from the results for the wild-type strain [Kp52145]; two tailed t test). ▵; P<0.05 (results are significantly different from the results for the waaL mutant [52145-ΔwaaL]; two tailed t test). (D) Dictyostelium cells were incubated with Klebsiella mutant lacking the CPS and the OPS 52145-Δwca _K2-ΔwaaL (Δwca _K2-ΔwaaL), or the OPS and the first, second or third sugar of the core (strains 52145-Δwca _K2-ΔwabM [Δwca _K2-ΔwabM], 52145-Δwca _K2-ΔwabH [Δwca _K2-ΔwabH], and 52145-Δwca _K2-ΔwabK [Δwca _K2-ΔwabK] respectively). The number of surviving bacteria was determined at different times by killing the Dictyostelium and plating the bacteria on LB plates. Bars represent mean ± s.e.m (n = 4). *; P<0.05 (results are significantly different from the results for 52145-Δwca _K2-ΔwaaL; two tailed t test). (E) Phagocytosis of LPS polysaccharide mutants by MH-S mouse alveolar macrophages. Intracellular bacteria were determined by the gentamicin protection assay. Kp52145 (wild type); OPS mutants 52O21 and ΔwaaL (52145-ΔwaaL); LPS core mutants ΔwaaQ (52145-ΔwaaQ), ΔwaaL-ΔwaaQ (52145-ΔwaaL-ΔwaaQ), ΔwabG (52145-ΔwabG), ΔwabM (52145-ΔwabM), ΔwabH (52145-ΔwabH), and ΔwabK (52145-ΔwabK). Bars represent mean ± s.e.m (n = 4). *; P<0.05 (results are significantly different from the results for the wild-type strain [Kp52145]; two tailed t test). (F) MH-S cells engulfment of Klebsiella cps mutant, strain 52145-Δwca _K2 (Δwca _K2), or strains lacking the CPS and the OPS 52145-Δwca _K2-ΔwaaL (Δwca _K2-ΔwaaL), or the OPS and the first, second or third sugar of the core (strains 52145-Δwca _K2-ΔwabM [Δwca _K2-ΔwabM], 52145-Δwca _K2-ΔwabH [Δwca _K2-ΔwabH], and 52145-Δwca _K2-ΔwabK [-Δwca _K2-ΔwabK] respectively). *; P<0.05 (results are significantly different from the results for the cps mutant [52145-Δwca _K2]; two tailed t test).

Figure 4. Role of K. pneumoniae lipid A decorations on phagocytosis resistance.

(A) Dictyostelium cells were incubated with Klebsiella strains (wild type [Kp52145], pagP mutant (ΔpagP, 52145-ΔpagPGB), pmrF mutant (ΔpmrF, 52145-ΔpmrF), or pagP-pmrF double mutant (ΔpagP-ΔpmrF, 52145-ΔpagPGB-ΔpmrF). The number of surviving bacteria (total or cell-associated) was determined at different times by killing the Dictyostelium and plating the bacteria on LB plates. Bars represent mean ± s.e.m (n = 4). *; P<0.05 (results are significantly different from the results for the wild-type strain [Kp52145]; two tailed t test). ▵; P<0.05 (results are significantly different from the results for 52145-ΔpagPGB; two tailed t test). (B) Dictyostelium cells were incubated with Klebsiella lipid A mutants constructed in the background of the cps mutant (52145-Δwca _K2 [Δwca _K2]). The number of surviving bacteria was determined at different times by killing the Dictyostelium and plating the bacteria on LB plates. Bars represent mean ± s.e.m (n = 4). *; P<0.05 (results are significantly different from the results for the cps mutant [52145-Δwca _K2]; two tailed t test). ▵; P<0.05 (results are significantly different from the results for 52145-Δwca _K2-ΔpagPGB; two tailed t test). (C) Phagocytosis of lipid A mutants by MH-S cells. Wild type [Kp52145], pagP mutant (ΔpagP, 52145-ΔpagPGB), pmrF mutant (ΔpmrF, 52145-ΔpmrF), or double pagP-pmrF mutant (ΔpagP-ΔpmrF, 52145-ΔpagPGB-ΔpmrF). Bars represent mean ± s.e.m (n = 4) *; P<0.05 (results are significantly different from the results for the wild-type strain [Kp52145]; two tailed t test). ▵; P<0.05 (results are significantly different from the results for 52145-ΔpagPGB; two tailed t test). (D) Phagocytosis of lipid A mutants constructed in the background of the cps mutant (52145-Δwca _K2 [Δwca _K2]) by MH-S cells. Bars represent mean ± s.e.m (n = 4). *; P<0.05 (results are significantly different from the results for the cps mutant [52145-Δwca _K2]; two tailed t test). ▵; P<0.05 (results are significantly different from the results for 52145-Δwca _K2-ΔpagPGB; two tailed t test).

Figure 5. Role of K. pneumoniae OMPs on phagocytosis resistance.

(A) Dictyostelium cells were incubated with Klebsiella strains (wild type [Kp52145], ompA mutant (ΔompA, 52OmpA2), ompK36 mutant (ΔompK36, 52OmpK36), and the corresponding complemented strains. The number of surviving bacteria (total or cell-associated) was determined at different times by killing the Dictyostelium and plating the bacteria on LB plates. Bars represent mean ± s.e.m (n = 4). *; P<0.05 (results are significantly different from the results for the wild-type strain [Kp52145]; two tailed t test). (B) Dictyostelium cells were incubated with Klebsiella OMPs mutants constructed in the background of the cps mutant (52145-Δwca _K2, [Δwca _K2]). The number of surviving bacteria (total or cell-associated) was determined at different times by killing the Dictyostelium and plating the bacteria on LB plates. Bars represent mean ± s.e.m (n = 4). *; P<0.05 (results are significantly different from the results for the cps mutant [52145-Δwca _K2]; two tailed t test). (C) Phagocytosis of OMPs mutants by MH-S cells. Wild type [Kp52145], ompA mutant (ΔompA, 52OmpA2), ompK36 mutant (ΔompK36, 52OmpK36), and the corresponding complemented strains. Bars represent mean ± s.e.m (n = 4) *; P<0.05 (results are significantly different from the results for the wild-type strain [Kp52145]; two tailed t test). (D) Phagocytosis of OMPs mutants constructed in the background of the cps mutant (52145-Δwca _K2) by MH-S cells. Bars represent mean ± s.e.m (n = 4). *; P<0.05 (results are significantly different from the results for the cps mutant [52145-Δwca _K2]; two tailed t test).

Figure 6. Virulence of K. pneumoniae ompK36 mutant.

Bacterial counts in mouse organs at 24 h post infection or 72 h post infection. Mice were infected intranasally with a bacterial mixture containing 5×10⁴ bacteria of wild type (Kp52145, •) or ompK36 mutant (ΔompK36, ○) Results were reported as log CFU per gram of tissue (Log CFU/g). *, results are significantly different (P<0.05; two-tailed t test) from the results for Kp52145. (A) Trachea; (B) Lung; (C) Spleen, (D) Liver.