Chemokines kill bacteria without triggering antimicrobial resistance by binding anionic phospholipids

Abstract

Classically, chemokines coordinate leukocyte trafficking; however, many chemokines also have direct antibacterial activity. The bacterial killing mechanism of chemokines and the biochemical properties that define which members of the chemokine superfamily are antimicrobial remain poorly understood. We report that the antimicrobial activity of chemokines is defined by their ability to bind phosphatidylglycerol and cardiolipin, two anionic phospholipids commonly found in the bacterial plasma membrane. We show that only chemokines able to bind these two phospholipids kill bacteria and that they exert rapid bacteriostatic and bactericidal effects with a higher potency than the antimicrobial peptide β-defensin 3. Both biochemical and genetic interference with the chemokine-cardiolipin interaction impaired microbial growth arrest, bacterial killing, and membrane disruption by chemokines. Moreover, unlike conventional antibiotics, Escherichia coli failed to develop resistance when placed under increasing antimicrobial chemokine pressure in vitro. Thus, we have identified cardiolipin and phosphatidylglycerol as binding partners for chemokines responsible for chemokine antimicrobial action.

Fig. 1.. Antimicrobial chemokines bind CL and PG phospholipids.

(A) Chemokines are more potent antimicrobials than hBD3. Antimicrobial assay showing the number of surviving CFU after incubation (2 hours at 37°C) of 10⁺⁵ CFU of E. coli (strain W3110) with increasing doses of the proteins indicated in the legend in antimicrobial assay buffer (AAB). Data are represented as the means ± SEM CFU of three to five independent experiments analyzed in triplicate. Color-coded asterisks indicate statistically significant differences between each chemokine and hBD3 analyzed by two-way analysis of variance (ANOVA) with Bonferroni multiple comparison test (**P < 0.01; ****P < 0.0001). (B) Screening of antimicrobial activity against E. coli and S. aureus (as coded on the inset of the left-most panel) of increasing doses of the human chemokines indicated above each graph. Experiments were performed and analyzed as in (A). Data are represented as the means ± SD CFU of technical triplicates from one experiment representative of three independent experiments. (C) Direct chemokine binding to anionic phospholipid-containing liposomes analyzed by biolayer interferometry (BLI). Top row: BLI sensorgrams showing chemokine binding to liposomes replicating the phospholipid composition of E. coli (lpEC; magenta) or S. aureus (lpSA; blue). Bottom row: Chemokine binding to phosphatidylcholine (PC) liposomes containing 30% of the phospholipids indicated on the inset of the left graph. Data in (B) and (C) are from one experiment representative of three independent experiments.

Fig. 2.. Antimicrobial chemokines share common bacterial binding sites and localize to the cell poles of E. coli.

(A) Antimicrobial chemokines bind directly to bacteria. Representative images of the binding of the AZ647-labeled chemokines (0.3 μM) indicated above each image to E. coli (strain W3110). Panels CXCL8 and CXCL9 are separated from the other panels to indicate that these images were acquired with a different microscope (see Materials and Methods). (B and C) Binding of antimicrobial chemokines to bacteria can be competed with other antimicrobial chemokines. In (B), representative images show the binding of CXCL11-AZ647 (0.3 μM) to E. coli (strain W3110) preincubated with buffer (AAB) or the unlabeled chemokines indicated above each image column. In (C), quantification of the fluorescence intensity of CXCL11-AZ647 staining is shown for each treatment group. Each dot corresponds to one bacterium (n ≈ 100). Data are from one experiment representative of two independent experiments. Statistical differences between each chemokine group and the “buffer” treatment group were analyzed by ANOVA with Tukey’s test for multiple comparisons (n.s., not significant; ****P < 0.0001). In (A) and (B), the top and bottom image rows show the staining for chemokine alone (magenta) or merged with 4′,6-diamidino-2-phenylindole (DAPI) (green), respectively. A white scale bar (20 μm) is inserted in the bottom left image. a.u., arbitrary units. (D) Antimicrobial chemokines bind to the PG/CL-rich membrane domains at the bacterial cell poles. Representative Airyscan confocal images show CXCL9-AZ647 binding to E. coli (strain W3110) costained with NAO. Graph on the right shows the normalized fluorescence intensity profile along a bacterium, as indicated by the dashed line in the “merge” panel, of NAO-524 (green), NAO-630 (orange), and CXCL9-AZ647 (magenta).

Fig. 3.. Liposomes containing CL or PG neutralize microbial binding and killing by chemokines.

(A and B) Liposomal PG and CL block chemokine binding to bacteria. Representative images (A) and quantification (B) show the binding of CCL20-AZ647 to E. coli (strain W3110) in the presence of PC liposomes (100 μM) containing 30% PE, PG, or CL or buffer alone (AAB) as indicated above each column. Top and bottom image rows show the staining for the chemokine alone or merged with DAPI, respectively. A white scale bar (20 μm) is inserted in the bottom left image. In (B), each dot corresponds to one bacterium (n ≈ 100). All liposome-treated groups were compared to buffer by ANOVA with Tukey’s test for multiple comparisons (****P < 0.0001). (C and D) PG- and CL-containing liposomes protect bacteria from antimicrobial chemokines. Antimicrobial assays show the number of surviving CFU after incubation (2 hours at 37°C) of E. coli (C) or S. aureus (D) with the chemokines (5 μg/ml) indicated above each graph in the presence of increasing doses of liposomes containing 30% PE, PG, or CL (inset) in AAB. Data are represented as means ± SD CFU from technical triplicates of one experiment representative of three independent experiments. The top and bottom horizontal dotted lines indicate the number of CFU counted when bacteria were incubated with buffer alone or with chemokine in the absence of liposome, respectively.

Fig. 4.. CL-deficient bacteria are resistant to killing but susceptible to OM permeabilization by antimicrobial chemokines.

(A) TLC showing the phospholipid composition of E. coli strains W3110 and BKT12. (B and C) Binding of fluorescent chemokines (0.3 μM; AAB-85) to W3110 and BKT12. (B) Representative FACS histograms. (C) Quantification of means ± SD median fluorescence intensity (MFI) of biological triplicates from one experiment representative of three experiments. (D to F) Antimicrobial assays showing the survival of parental (W3110 and BW25113) or CL-deficient (BKT12 and JW1241) E. coli strains after incubation (37°C for 2 hours) with chemokines (1.2 μM; AAB-85) or buffer. In (D) and (F), strains (10⁺⁵ CFU) were treated separately. Bars represent means ± SEM CFU of three to six biological replicates (dots) analyzed in triplicate from two experiments relative (%) to the CFU number in buffer-treated samples. In (E), W3110 and BKT12, mixed 1:1 (10⁺⁵ total CFU), were treated and plated with and without kanamycin to calculate BKT12 CFU (kanamycin-resistant) and total CFU (W3110 + BKT12). Bars represent means ± SD CFU of biological triplicates analyzed in triplicate from one experiment representative of three experiments. (G) BLI sensorgrams showing the binding of the indicated proteins (1 μM) to LPS or Kdo2–lipid A liposomes. (H) OM permeabilization assays. NPN fluorescence in W3110 and BKT12 (2 × 10⁺⁷ CFU) was measured after addition (orange arrowheads) of the indicated proteins (4 μM; AAB-85) or buffer. Results are represented as means ± SEM of biological triplicates relative (%) to the maximum fluorescence obtained with PolB. Data are from one experiment representative of three experiments. Data were analyzed by two-way ANOVA with Bonferroni test (C, E, F, and H) or unpaired t test (D). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Fig. 5.. CL promotes rapid bacteriostatic and bactericidal action by antimicrobial chemokines.

(A and B) FACS analysis of the effect of chemokines/hBD3 (1.2 μM; AAB-85) on the growth (A) and killing (B) of W3110 and BKT12 E. coli strains over time. The growth ratio was calculated as number of live bacteria (SYTO24⁺ SYTOX⁻)/initial bacterial input. The percentage of killed bacteria (SYTO24⁺ SYTOX⁺) was calculated relative to the number of total bacteria (SYTO24⁺) at each time point. Data are means ± SEM from three to five experiments combined, with biological triplicates in each experiment. Data in (A) were log transformed before statistical analysis. In (A), statistical significances for buffer versus chemokine and W3110 versus BKT12 are color coded above the graphs or above the x axes, respectively. (C) Live (black lines) and dead (orange lines) W3110 and BKT12 bacteria 1 or 2 hpt with increasing doses of CXCL11 and CCL21. Live bacteria (left y axis) posttreatment are represented relative (%) to the number of live bacteria in buffer-treated samples. The percentage of killed bacteria (right y axis) was calculated as in (B). Data are means ± SD of biological triplicates from one experiment representative of three experiments. Statistical significances of the bacteriostatic and killing activity are color coded above the graphs. (D) Right: Quantification of the percentage of killed (SYTOX⁺) W3110 or BKT12 bacteria after treatment with the indicated proteins (4.8 μM; AAB-85; 20 min). Bars show the means ± SD of biological triplicates from one experiment representative of three experiments. Left: Representative FACS contour plots. Data were analyzed by two-way ANOVA with Tukey (A) or Bonferroni (B and C) test or by multiple t test (D). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. SSC, side scatter; Protam, protamine.

Fig. 6.. Antimicrobial chemokines lyse phospholipid bilayers in an anionic phospholipid-dependent manner.

(A) Calcein-leakage assay showing the release of calcein from PE/PG/CL liposomes upon injection (arrowhead) of chemokines/protamine (1.2 μM). Curves represent the percentage of calcein released relative to the maximum release observed with 0.1% Triton X-100. Solid lines represent the mean of biological triplicates. Colored shaded area represents the SD. Data are from one experiment representative of three experiments. (B) Radially integrated SAXS spectra of chemokines interacting with 20/80 PG/PE liposomes at increasing (1/6 to 3/2) peptide-to-lipid (P/L) charge ratios. Bragg structure peaks were indexed for the observed phases: cubic (Q) (arrowed indices), lamellar (L), and inverted hexagonal phases (H). (C) Linear fits of the peak position for the cubic phases, Pn3m, Im3m, and Ia3d (illustrated next to symbol key), indexed in (B). Estimation of mean NGC, <k>, from fits is displayed next to each plot. Colors represent each P/L ratio as noted in (B). (D) Calcein-leakage assays showing the percentage of calcein released from three types of liposomes (inset of CCL19 panel) after injection (arrowheads) of the indicated proteins (1.2 μM). Data were analyzed as in (A) and correspond to biological triplicates from one experiment representative of three experiments. (E) Quantification of the percentage of calcein released from different liposomes (inset of CXCL9 panel) 20 min after addition of increasing doses of the indicated proteins. Data are means ± SD of biological triplicates from one experiment representative of three experiments. Data were analyzed by two-way ANOVA with Tukey test. Statistical differences (PE/PG/CL versus PE/PG) are indicated above the graphs. *P < 0.05.

Fig. 7.. CCL20 kills parental and antibiotic-resistant E. coli without triggering chemokine resistance.

(A) Coomassie Blue–stained gel for commercially available CCL20 (Peprotech) and in-house produced CCL20-His. (B) BLI assays showing the binding of CCL20-His to liposomes containing 30% of the indicated phospholipids (inset). (C) Determination of the MIC of CCL20-His. E. coli (W3110 strain) bacteria were incubated (18 hours at 37°C) with increasing doses of CCL20-His in MHB, and the percentage of viable bacteria relative to bacteria treated with buffer alone was calculated using the BacTiter-Glo Kit. Graph shows means ± SEM of bacterial viability (%) from four experiments combined analyzed in duplicate. (D) Bacteria develop resistance against conventional antibiotics but not against CCL20-His. W3110 bacterial cultures were maintained in MHB for 2 weeks in the presence of a sublethal dose (0.5 × MIC) of tetracycline (Tet), ampicillin (Amp), or CCL20-His, as indicated on the inset. On selected days, MIC for each antimicrobial agent with the corresponding culture was recalculated, and bacteria were subcultured adjusting the agent dose to the new MIC. Graphs show the MIC fold change relative to the initial MIC of two independent bacterial cultures, represented by circle and triangle symbols, for each agent. (E) CCL20-His kills Tet- and Amp-resistant bacterial strains. Parental W3110 bacteria (5 × 10⁺⁵ CFU/ml) or the conditioned bacterial strains generated in (D), W3110-Amp, W3110-Tet, or W3110-CCL20-His, were incubated (18 hours at 37°C) in MHB with Amp, Tet, or CCL20-His at a concentration equivalent to their original MIC. Bacterial viability in each sample was calculated as in (C). Bars represent means ± SEM of bacterial viability (%) of biological triplicates from one experiment representative of three experiments.