Group A Streptococcus M1T1 serotype M1T1 5448 was originally isolated from a patient with necrotizing fasciitis and toxic shock syndrome (Chatellier et al., 2000) and an animal passaged (AP) version of the M1 5448 GAS parent strain (5448AP) containing a single inactivating adenine insertion at the 877-bp position of covS were used (Aziz et al., 2004). The isogenic M1T1 5448 ΔSLO mutant were described previously (Timmer et al., 2009). GAS strains were propagated using Todd Hewitt broth (THB) or agar (THA) at 37°C.

Phlebotomy was performed on healthy donors with full informed consent under a protocol approved by the University of California San Diego (UCSD) Human Research Protections Program. RBCs were isolated for hemolysis assays or preparation of RBC nanosponges (see below). Neutrophils were isolated from freshly collected whole blood of healthy donors under a protocol approved by the, using PolyMorphPrep Kit (Fresenius Kabi) as previously described (Kristian et al., 2005).

Human keratinocyte cell line HaCaT, murine macrophage cell line J774 and human monocyte cell line THP1 were cultured in RPMI-1640 media (Invitrogen) + 10% heat-inactivated fetal bovine serum (FBS) at 37°C in humidified air with 5% CO₂. Primary bone marrow-derived macrophages (BMDM) were prepared as described (Hsu et al., 2004) with slight modification. Bone marrow cells were collected from mice and cultured in Dulbecco’s modified Eagle’s medium (high glucose) supplemented with 20% L-929 cell conditioned medium for 7 days. Adherent cells (BMDM) were then collected and cultured in Dulbecco’s modified Eagle’s medium (high glucose) with 10 ng/ml macrophage colony-stimulating factor (Pepro-Tech) overnight before bacterial infection.

RBC nanoparticles were prepared following published methods (Hu et al., 2011). Briefly, ∼100 nm PLGA polymeric cores were prepared using 0.67 dL/g of carboxy-terminated 50:50 poly-(D,L-lactide-co-glycolide) (LACTEL Absorbable Polymers) through a nanoprecipitation process. The PLGA polymer was first dissolved in acetone at a concentration of 10 mg/ml. One ml of the solution was then added rapidly to 3 ml of water, and the mixture placed in a vacuum for at least 3 h to evaporate the organic solvent. Human blood from healthy donors was washed with PBS + 1 mM EDTA × 3 by centrifugation at 500 × g for 10 min. RBC membrane vesicles were then prepared via hypotonic treatment and centrifugation at 10°C at 800 × g for 12 min. RBC membrane coating was completed by fusing RBC membrane vesicles with PLGA particles via sonication using an FS30D bath sonicator at frequency = 42 kHz and power = 100 W for 2 min.

Blood was drawn from healthy donors after informed consent, and 2 × 10⁵ CFU of bacteria at OD₆₀₀ = 0.4 were added to 400 μL heparinized whole blood in siliconized tubes. Tubes were placed on a rotator at 37°C for 1 h, then diluted and plated for colony forming unit (CFU) enumeration. Growth index was calculated as the ratio of surviving CFU after incubation vs. the initial inoculum.

J774 murine macrophages were seeded at ∼5 × 10⁵ cells in 350 μl of RPMI-2% FBS in a 24-well plate. Overnight bacterial cultures were diluted 1:10, subcultured for 3 h, resuspended and serially diluted in RPMI-2% FBS, and used to inoculate J774 cells at a multiplicity of infection (MOI) = 10 bacteria/cell. Freshly purified human neutrophils in serum-free RPMI were added to 96-well plates at a density of 1 × 10⁶ cells/well and infected with GAS at MOI = 1. Plates were centrifuged at 600 × g for 5 min to facilitate bacterial contact with cells. To assess total killing, macrophages were incubated at 37°C for 2 h and neutrophils for 15 min, washed three times with PBS, detached with 100 μl of 0.05% trypsin, and lysed with 900 μl of 0.025% Triton X-100 in PBS. Samples were serially diluted in PBS and plated on THA overnight for CFU enumeration.

For quantification of cellular ATP as an indicator of metabolic activity and cell viability, opaque-walled 96-well plates with were prepared using THP-1 cells seeded at density 2 × 10⁴ cells/well, and infected with GAS WT or ΔSLO mutant strains at MOI = 25 for 2 h at 37°C. After incubation, 100 μl of CellTiter-Glo^® reagent was added to contents of wells, mixed for 2 min on an orbital shaker to induce cell lysis, and incubated at RT × 10 min to stabilize luminescent signal. Luminescence was recorded using SpectraMax plate reader and software. Live-Dead staining of HaCaT cells and murine BMDMs were performed by growing cells to 70% confluency in 96-well plates and infecting with WT GAS at MOI = 50 for 2 h, or 0.2 μg purified SLO for 30 min. Cells were then washed with PBS and treated with the viability assay mixture from the LIVE/DEAD Viability/Cytotoxicity Kit for mammalian cells (Molecular Probes, Invitrogen) for 30 min at 37°C, and imaged using an Olympus BX51 fluorescent microscope.

THP1 monocytes were plated at 1 × 10⁶ cells/well in 12-well plates and infected with GAS strains at MOI = 20:1. Plates were centrifuged at 2,000 rpm for 5 min to ensure bacterial contact with cells and then incubated at 37°C, 5% CO2. One hour after infection, penicillin (5 μg/ml) and gentamicin (100 μg/ml) were added to the media to kill residual extracellular bacteria. At 4 h after infection, cells were collected, fixed, and permeabilized for apo-bromodeoxyuridine TUNEL assay per manufacturer’s (BD Bioscience) instructions.

In a 6-well plate, 1 × 10⁶ J774 macrophages were seeded per well in 2 ml of RPMI + 2% FBS. Cells were infected with WT GAS at MOI = 10 for 2 and 4 h. To harvest whole-cell lysates, cells were washed three times with PBS and treated with radioimmunoprecipitation assay (RIPA) lysis buffer. Nuclear and cytoplasmic fractions were isolated using the NER-PER Extraction Kit (Pierse, Rockford, IL, United States) according to manufacturer’s protocol. Protein abundances were determined in cell fraction lysates with bicinchoninic acid assay (BCA) colorimetric assay. Aliquots containing 30 μg of protein were separated on 10% SDS-PAGE gels and transferred onto nitrocellulose membranes. Blots were probed using rabbit anti-Caspase 1 [Santa Cruz Technologies, diluted 1:500 in Tris-buffered saline–Tween 20 (TBST)]. Enhanced chemiluminescence reagent (PerkinElmer) was used for detection.

J774 macrophages were infected at MOI = 10 for 2 h, supernatant collected and replaced with fresh media, then supernatant collected again 24 h post-infection. Skin from infected mice were homogenized in 1 ml PBS and centrifuged at 4°C, and the resulting supernatant used in ELISA for for IL-1β release (R&D Systems), via absorbance at 450 nm on a SpectraMax M3 plate reader and SoftMax Pro software. Caspase-1 activation was determined by Fam-YVAD-FMK (ImmunoChemistry Technologies) staining of THP-1 macrophages infected in 96-well plates per manufacturer’s specifications. Caspase-dependent apoptosis was determined in J774 murine macrophages using the APO-Caspase 3/7 activity assay (Promega) following the manufacturer’s protocol. Briefly, 100 μL of J774 cells were cultured in a white opaque 96-well plate at semi-confluency 1 day prior to infection with GAS at MOI = 20:1. Penicillin (5 μg/ml) and gentamicin (100 μg/ml) were added to the media 1 h after infection to kill residual extracellular bacteria. At 4 h after infection, caspase activity was quantified by adding 100 μL of detecting reagent per well, shaking gently for 5 min and incubating at room temperature × 1 h. Fluorescence at 520 nm was red on a Molecular Devices SpectraMax M3 reader to detect the level of caspase 3/7.

Mouse infections with GAS were performed based on modifications to a previously described model (Nizet et al., 2001) under a protocol approved by the UCSD Institutional Animal Care and Use Committee (IACUC). Twenty-four hours prior to infection, the backs of eight C57Bl6 mice were shaved and hair was removed by chemical depilation (Nair). Mice were infected subcutaneously with 50 μL of a sublethal dose of log-phase GAS (1 × 10⁷ CFU) in PBS, and 15 min later subsequently treated with vehicle only (10% sucrose) or vehicle plus 50 mg/kg of nanosponges in an area proximal to the site of the infection. Digital photographs of skin lesions were taken and lesion size was measured using NIH Imager software. Lesions were biopsied on day 3 post-infection. Excised lesions were placed into 2 ml screw cap tubes containing 1 ml PBS + 1 mm silica/zirconia beads (Biospec Products). Tissue was homogenized by shaking twice with the mini-beadbeater-8 (Biospec Products) at full for speed for 1 min, placing on ice in between. The homogenate was serially diluted in sterile PBS and plated on THA for enumeration. Dilutions were plated on THA agar and cultured overnight at 37°C for enumeration of CFU. Hematoxylin-and-eosin (H&E) staining was performed by the UCSD Histology Core Facility. Images were obtained using an Olympus BX41 microscope.

Neutrophils (2 × 10⁵) were plated in 96 wells and infected with WT GAS and ΔSLO mutant at MOI = 1 in 37°C/5% CO₂ for 4 h. Cells were fixed with 4% paraformaldehyde and stained with anti-myeloperoxidase (MPO) antibody (1:300 dilution, Calbiochem) in PBS + 2% bovine serum albumin (BSA, Sigma) at room temperature for 1 h, followed by incubation with goat anti-rabbit

Alexa 488 antibody (1:500 dilution, Life Technologies). Cells were counterstained with ProlongGold + 4’,6’-diamidino-2-phenylindole (DAPI, Invitrogen) and imaged on a fluorescent microscope. Representative, randomized images (n = 3) were taken for each condition and individualized experiment. Ratio of NET-releasing cells to non-NET releasing cells was determined as % of total cells. Elastase release from neutrophils infected with GAS at MOI = 5 for 30 min into the supernatant was indirectly determined using 20 μM peptide substrate N(Methoxysuccinyl)-Ala-Ala-Pro-Val 4-nitroanilide (Sigma) for 20 min at RT and absorbance 405 nm (SpectraMax M3 plate reader/SoftMax Pro). Statistical Analysis Experiments were performed in triplicate and repeated at least twice. Error data represent standard errors of the means (SEM) of the results from experimental duplicates, triplicates, or quadruplets. Statistical analysis was performed using Student’s unpaired two-tailed t-test. Comparisons among three or more samples were evaluated using one-way analysis of variance (ANOVA) followed by the non-parametric Tukey’s post-test. Comparisons of multiple samples were evaluated using ANOVA followed by Dunnett’s or Tukey’s test (Graph Pad Prism).

Creation of RBC-camouflaged nanoparticles involves two main steps: membrane vesicle derivation from RBCs followed by fusion of the membrane vesicle with the polymeric nanoparticle core (Figure 1A), as previously reported in detail (Desilets et al., 2001; Cheng et al., 2007; Hu et al., 2011; Pang et al., 2015). The RBC-derived nanoparticles, hereafter termed “nanosponges,” have an approximate particle size of 80 nm, right side out orientation, and other physical characteristics and physiochemical properties that lead to favorable pharmacokinetics and biodistribution (Hu et al., 2011). We determined that RBC nanosponges (50 or 500 μg/ml) could significantly inhibit the hemolytic action of purified SLO in a dose-dependent manner to an SLO concentration of 0.1 μg/ml (Figure 1B). Dot blot analysis (see Supplementary Methods) confirmed sequestration of SLO from the media to the nanosponges (Supplementary Figure S1). These findings mirrored prior observations with S. aureus α-toxin, in which RBC nanosponges absorbed α-toxin to limit its interaction with subsequent cellular targets, a finding that was absent when using just RBC membranes or polymeric cores individually (Hu et al., 2013). We then tested the ability of RBC-derived nanosponges to block hemolysis induced by coincubation of freshly isolated human RBC with a high level SLO-producing WT GAS strain (Supplementary Figure S2) compared to its isogenic ΔSLO mutant as a control. Paralleling results with the purified toxin, 500 μg/ml nanosponges produced a significant (∼50%) reduction of hemolysis induced by the WT strain, but had no effect on the low level of hemolysis seen in the ΔSLO mutant control (Figure 1C). SLO can also trigger membrane damage and cytolytic cell death in keratinocytes, which may play a role in the severe tissue injury of invasive GAS necrotizing skin infection (Ruiz et al., 1998; Sierig et al., 2003). As measured by LIVE/DEAD immunofluorescent staining, we found that administration of RBC nanosponges allowed a significant increase in survival of keratinocytes upon coincubation of the cells with the WT SLO-producing GAS strain at MOI = 25 (50% survival vs. 20% untreated, Figure 1D).

Macrophages contribute directly to bacterial clearance and produce cytokines and other inflammatory signals that can orchestrate downstream immune responses. As measured by live-dead staining, we found that RBC nanosponges significantly reduced cell death of murine BMDM following exposure to purified SLO toxin (Figure 2A) or WT GAS (Figure 2B). Similarly, RBC nanosponges increased viability of human THP-1 macrophages following incubation with GAS in a SLO-dependent manner (Figure 2C). RBC nanosponges also enhanced murine macrophage killing of WT GAS to the level of killing observed with the GAS ΔSLO mutant (Figure 2D). Treatment with nanosponges reduced WT GAS-induced macrophage apoptosis as measured by caspase-3 cleavage (Figure 3A), caspase-3 activity assay (Figure 3B), and TUNEL assay (Figure 3C). Nanosponge treatment also blocked GAS SLO-induced inflammasome activation (measured by caspase-1 activity, Figure 3D) and IL-1β secretion (Figure 3E). In sum, RBC nanosponges improved macrophage viability in the face of GAS SLO activation, enhancing their antibacterial function against the pathogen.

Group A Streptococcus production of SLO has been shown to impair neutrophil microbicidal activity, degranulation and release of DNA-based neutrophil extracellular traps, or NETs (Sierig et al., 2003; Ato et al., 2008; Uchiyama et al., 2015). We found that addition of 500 μg/ml of RBC nanosponges to freshly isolated human whole blood or purified human neutrophils significantly increased killing of the WT GAS strain expressing SLO (Figure 4A). Nanosponge treatment also increased neutrophil degranulation as measured by release of neutrophil elastase (Figure 4B), as well as the production of antibacterial NETs as detected by immunofluorescent staining and DNA quantification (Figure 4C). Coupled with the macrophage studies above, our in vitro analyses suggest that nanosponge sequestration of the GAS toxin can ameliorate damaging effects on innate immune phagocytes, allowing improved clearance of the pathogen.

Group A Streptococcus production of SLO has been shown to contribute to the severity of necrotizing skin lesions in murine experimental infection models (Limbago et al., 2000; Fontaine et al., 2003; Zhu et al., 2017). We injected the invasive WT SLO-producing GAS strain subcutaneously into the flanks of WT C57bl6 mice and then subsequently treated the mice proximally with either vehicle only control or vehicle + 50 mg/kg of RBC nanosponges at 15 min post-infection, a dose known to be well tolerated by mice in studies of purified α-toxin neutralization (Hu et al., 2013). At 72 h post-infection, necrotic skin lesions in the nanosponge-treated group were significantly smaller than those in the control group (Figure 5A), and quantitative bacterial cultures showed that nanosponge-treated mice had significantly fewer GAS CFU recovered from the wound tissue (Figure 5B). Histopathologic analysis of lesion biopsies showed wholescale hemorrhagic necrosis and tissue destruction of the dermal and subcutaneous tissues in control (vehicle-treated) mice; in contrast, nanosponge treatment preserved tissue architecture with markedly reduced necrotic changes and diminished neutrophil infiltration (Figure 5C). Furthermore, levels of proinflammatory cytokine IL-1β associated with inflammasome activation and pyroptotic cell death were reduced in the infected mice upon nanosponge administration (Figure 5D).