Group A Streptococcal M1 Protein Provides Resistance against the Antimicrobial Activity of Histones

Abstract

Histones are essential elements of chromatin structure and gene regulation in eukaryotes. An unexpected attribute of these nuclear proteins is their antimicrobial activity. A framework for histone release and function in host defense in vivo was revealed with the discovery of neutrophil extracellular traps, a specialized cell death process in which DNA-based structures containing histones are extruded to ensnare and kill bacteria. Investigating the susceptibility of various Gram-positive pathogens to histones, we found high-level resistance by one leading human pathogen, group A Streptococcus (GAS). A screen of isogenic mutants revealed that the highly surface-expressed M1 protein, a classical GAS virulence factor, was required for high-level histone resistance. Biochemical and microscopic analyses revealed that the N-terminal domain of M1 protein binds and inactivates histones before they reach their cell wall target of action. This finding illustrates a new pathogenic function for this classic GAS virulence factor, and highlights a potential innate immune evasion strategy that may be employed by other bacterial pathogens.

Figure 1. Histones are released from neutrophils.

Released (A) DNA was quantified by staining with the cell-impermeable dye Sytox Green or (B) histones from neutrophils after 4 h stimulation with 25 nM PMA as compared to controls via IHC fluorescent measurements using primary rabbit anti-histone H2A antibody and secondary goat anti-rabbit Alexa 488 antibody. NETs were induced by stimulation with 25 nM PMA for 4 h as analyzed by immunofluorescence (IF) microscopy using primary (C) anti-neutrophil elastase (NE), (D) anti-histone H1-H4 and (E) anti-histone H2A antibody followed by secondary Alexa 594 antibody (red) and Sytox Green to stain DNA (green). (F) DNA release from neutrophils infected with a panel of six different Gram-positive bacteria at an MOI of 10 was determined by incubation with cell-impermeable Sytox Green at 4 h. (G) Histone release in response to bacteria at MOI of 10 was detected by IHC using Alexa 488 antibody against polyclonal anti-histone H2A antibody at a 4 h time point. For (C–E) randomized images with scale bars representing 10 μm are shown and are representative of at least three independent experiments. Experiments in (A,B and F,G) were combined from a minimum of three independent experiments in triplicates and normalized to untreated control at each experiment. Results shown represent average ± SEM values and were analyzed by Student’s t-test (**P < 0.01, ***P < 0.001).

Figure 2. Antimicrobial activity of histones against Gram-positive bacteria.

The resistance of multiple Gram-positive bacterial species was tested in MIC assays against (A) histone mixture and (B) histone H2A. (C) Resistance of representative GAS M serotypes 1, 2, 3, 4, 6, 12, 22, 28 and 49 against histone H2A in MIC assay were performed. (D) NET-mediated killing of GAS WT and MRSA WT was performed at an MOI of 0.1 at 15 min post-infection and bacterial survival was calculated by CFU enumeration vs. initial inoculum. (E) Representative confocal microscopy image of NET-mediated killing for GAS and MRSA at an MOI of 1 by Live/Dead BacLight staining with indication to identify dead (red arrows) and live (green arrows) bacteria. Results shown represent average ± SEM values and were analyzed by Student’s t-test (D) or Mann-Whitney test (A–C) (*P < 0.05, **P < 0.01, ***P < 0.001). Each dot in (A–D) represents one sample and all experiments have been performed at least three independent experiments in triplicates.

Figure 3. GAS M1 protein protects against histones.

(A) Screening of virulence-associated genes with GAS M1 background of isogenic mutants Δsic, ΔdltA, ΔgacI, ΔhasA, Δscl1, Δpil and Δemm1 mutants compared to GAS M1 WT. (B) Histone H2A MIC assays with GAS WT, Δemm1 mutant and complemented mutant (Δemm1+pM1) as well as LL and heterologous expression of M1 protein in LL (LL+pM1) are shown. (C) Effect of histone treatment on GAS WT and Δemm1 mutant bacteria were visualized using the cell-impermeable DNA dye Sytox Green (green), the cell-permeable DNA dye DAPI (blue) and the membrane dye FM4-64 (red) by confocal microscopy after 3 h of incubation in the presence or absence of histone at 62.5 μg/mL or 150 μg/mL histone H2A and (D) the Sytox Green-positive population in random view fields was quantified. (E) Time-dependency of permeability was assessed using the cell-impermeable dye Sytox Green, which yielded into fluorescent signal upon interaction of dye with bacterial DNA, and was monitored with 1000 μg/mL histone for GAS WT and Δemm1 mutant and bacterial survival was simultaneously calculated via CFU enumeration relative to initial inoculum over a 3 h time course. (F) Bacterial permeability was investigated by treatment with increasing concentrations of histones ranging from 7.8 μg/mL to 1000 μg/mL histones and simultaneous CFU enumeration of bacterial survival relative to initial inoculum after 3 h incubation. Results shown represent average ± SEM values and were analyzed by Student’s t-test in (D–H) or Mann-Whitney test in (A,B) (**P < 0.01, ***P < 0.001). Each dot represents one sample. All experiments have been carried out a minimum of three times run in triplicates, except for (C) which was repeated three times in monoplicates.

Figure 4. M1 protein protects against histones in NETs.

2 × 10⁵ neutrophils were stimulated for 4 h with 25 nM PMA to induce NETs and infected with GAS WT, Δemm1 mutant and complemented strain (Δemm1+pM1) at an MOI of 0.1. Surviving CFUs were calculated relative to initial inoculum after 15 min of infection in (A) untreated cells, (B) cells treated with 100 U/mL DNase 1 to degrade the DNA backbone, or (C) cells treated with blocking anti-histone H2A antibodies. (D) Survival in NETs of GAS WT naïve or anti-M1 antiserum at MOI of 0.1 after 15 min of infection. All results are showing the percentage of bacterial survival relative to the bacterial inoculum. Each dot represents one sample from three independent experiments in triplicates and neutrophils were obtained from three different donors. Results shown represent average ± SEM values and were analyzed by Student’s t-test (n.s., not significant and *P < 0.05, ***P < 0.001).

Figure 5. M1 protein binds to and sequesters histones.

(A) Binding of histone H2A at 31.25 μg/mL to whole, live GAS M1 WT and Δemm1 mutant bacteria was determined via IHC using primary anti-histone H2A antibody followed by secondary Alexa 488 antibody by shift in fluorescence intensity by flow cytometry and quantified using the geometric mean fluorescence intensity (gMFI). (B) Binding of histone H2A was analyzed for whole, live LL WT and LL+pM1 bacteria by flow cytometry and quantified by gMFI. Visualization of histone binding to the most surface-exposed protein from GAS, the M1 protein (hair-like structures) of GAS M1 WT and Δemm1 mutant (C) using primary rabbit anti-histone H2A antibodies followed by secondary anti-rabbit immunogold-labeled antibodies visualizing 15 nm gold particle with white arrows by transmission electron microscopy (TEM). (D) Qualitative binding of histones as determined by identification of immune-gold positive bacteria and quantification of immune-gold particle number per bacterium were determined from > 20 bacteria in random view fields at a 49,000x magnification. Results in (A–C) show representative results or TEM images from at least three independent experiments. Quantification of results obtained by flow cytometry in (A,B,E) represent average ± SEM values and were analyzed by Student’s t-test (***P < 0.001).

Figure 6. N-terminal portion of M1 protein binds to histones and mediates resistance against histones.

(A) Schematic of M1 protein (2OTO) model highlighting the N-terminal hyper-variable region (HVR), and A and B repeat and the start of the C and D repeat which terminate at the C-terminal cell wall anchor. (B) Full length, recombinant M1 protein (rM1full-length) was used as bait to analyze interaction with histone H2A as determined by pull-down analysis and developed by western blot. (C) Binding of recombinant, truncated rM1 fragments HVR+A, A+B or B+C part of M1 protein to histone H2A by pull-down was visualized by western blot. (D) GAS M1 WT, Δemm1 mutant, Δemm1+pM1 and a complemented strain lacking the NT region (Δemm1+pM1ΔNT) were tested for resistance to histone H2A by MIC as well as (E) LL WT, LL+pM1 and LL+pM1ΔNT. (F) Effect of exogenous 10 μM NT fragment on histone killing for GAS Δemm1 mutant in MIC testing. Results shown were obtained from at least three independent experiments. Results shown represent average ± SEM values and were analyzed by Mann-Whitney test in (D–F) (*P < 0.05, ***P < 0.001).