Streptococcus anginosus Activates the NLRP3 Inflammasome to Promote Inflammatory Responses from Macrophages

Abstract

Chronic inflammation and oral dysbiosis are common features of oral squamous cell carcinoma (OSCC). The commensal streptococci, S. anginosus, is increased in oral diseases including OSCC. Our previous work revealed that S. anginosus promotes inflammatory responses from macrophage cell lines, however the molecular mechanism by which S. anginosus interacts with macrophages to instigate this response remains to be investigated. Here, we found S. anginosus activated primary bone marrow derived macrophages (BMMs), which presented increased NF-κB activation and downstream inflammatory cytokines TNF⍺, IL-6 and IL-1β at 24 hours post-infection. S. anginosus viability, TLR2, TLR4 and MyD88 were dispensable for NF-κB activation, but each promoted the induction of distinct downstream inflammatory mediators, with only MyD88 being necessary for NF-κB activation in response to heat-killed S. anginosus. S. anginosus replicated intracellularly within BMMs without causing cell death and induced expression of inflammasome sensors AIM2, NLRC4 and NLRP3. S. anginosus-infected BMMs lacking the inflammasome adapter protein ASC (Asc^−/−) or Caspase-1 (Caspase1^−/−) had significantly diminished IL-1β production compared to wild type BMMs, indicating that S. anginosus activated the inflammasome. S. anginosus primarily triggered the inflammasome through NLRP3 as S. anginosus-infected Nlrp3^−/− BMMs and NLRP3 inhibitor (MCC950)-treated wild type BMMs displayed diminished IL-1β production compared to wild type controls. Lastly, S. anginosus-infected Asc^−/− and to a lesser extent Nlrp3^−/− mice displayed reduced weight loss, reduced inflammatory cytokines, and increased bacterial burden compared to C57BL/6 mice. These findings indicate that S. anginosus replicates within macrophages and promotes a proinflammatory response in part through activating the NLRP3 inflammasome.

brief summary sentence:

S. anginosus replicates intracellularly within macrophages and is sensed by the NLRP3 inflammasome to promote proinflammatory response.

1. Background

Head and neck squamous cell carcinoma (HNSCC) is the seventh most common type of cancer worldwide [1] with 890,000 new cases and 450,000 deaths annually. HNSCC refers to a diverse group of malignancies arising from the anatomic sites involving the oral cavity, pharynx, hypopharynx, larynx, nasal cavity, and salivary glands that compose the upper aerodigestive tract [2]. Oral squamous cell carcinoma (OSCC) is the predominant form of HNSCC and is responsible for about 389,846 new cancer cases with an estimated 188,438 deaths in 2022 [3]. The number of newly diagnosed OSCC cases is predicted to increase by 62% in 2035 [4]. Despite advances in treatment, OSCC has a poor prognosis, with less than a 50% 5-year survival rate in many regions [5]. While tobacco and heavy alcohol use account for 74% of OSCC cases, 2–6% are linked to human papillomavirus (HPV) infection [6]. Notably, 15% of oral cancer cases occur in individuals without a history of tobacco or alcohol use, suggesting the potential role of other emerging risk factors for OSCC, including oral dysbiosis [7]. These challenges have urged research into new risk factors and prognostic markers, with increasing attention on the role of the oral microbiome in oral carcinogenesis [7].

Dysbiosis of the oral microbiome exacerbates disease largely through prolonged inflammatory signaling, although some bacteria are directly linked to other cancer hallmarks. An increase in pathogenic species such as Fusobacterium nucleatum and Porphyromonas gingivalis is linked to an increase in pro-inflammatory NF-κB signaling, elevated production of pro-inflammatory cytokines TNF⍺ and IL-1β, and an increase in OSCC proliferation and invasion [8],[9],[10]. Additionally, Streptococcus species such as S. mutans can increase inflammatory cytokines including IL-6 and promote metastasis by decreasing vascular endothelial cadherin [11]. Other Gram-positive streptococcus species are reportedly elevated in OSCC including S. constellatus, S. salivarius, S. gordonii and S. mitis, although their mechanism in disease progression remains to be investigated [12]. Similarly, S. anginosus is also increased in OSCC [13]. In vitro, the S. anginosus virulence factor Streptolysin S increases production of HCS-2 cell immediate early genes (IEGs), which are responsible for transcription of early growth factors (EGFs) [14]. Conversely, S. anginosus decreases proliferation, migration and invasion of the SCC-15 cell line [15]. While there is a scarcity of studies investigating the role of S. anginosus during OSCC in vivo, S. anginosus promotes proliferation, inhibits apoptosis and induces gastric tumorigenesis in mice [16].

The host innate immune system plays a crucial role in balancing mucosal immunity versus tolerance to bacteria in the oral cavity. Tissue macrophages represent important contributors of innate immunity through their direct interactions with oral microbes and subsequent responses that direct immune activation. Microbial recognition by macrophages is mediated by pattern recognition receptors (PRRs), which include surface PRRs, including most Toll-like receptors (TLRs) and C-Type Lectin receptors (CLRs), and cytosolic PRRs, including NOD-like receptors (NLRs), RIG like receptors (RLRs), and AIM2-like receptors (ALRs). PRRs are able to recognize a variety of pathogen-associated molecular patterns (PAMPs) and damage-associated molecular patterns (DAMPs) to signal an inflammatory response [17]. TLRs recognize many microbial surface ligands and nucleic acids within endosomes [18]. Upon recognition of its ligand, TLRs upregulate proinflammatory signaling pathways such as NF-κB and induce production of proinflammatory cytokines such as TNFα, IL-6 and type I interferons [18]. Additionally, TLRs can upregulate the expression of intracellular PRRs such as NLRs/ALRs and immature cytokines such as pro-IL-1β and pro-IL-18 [19]. Several intracellular sensors, including NLRP3 and AIM2, form an inflammasome by binding with the adaptor protein ASC and the inactive enzyme pro-caspase-1. Inflammasomes can be activated by a variety of PAMPS and DAMPS (e.g., extracellular ATP, flagellin and lipoteichoic acid), depending on the NLR sensor, while the AIM2 inflammasome is triggered by double-stranded DNA [20], [21]. Inflammasome assembly leads to pro-caspase-1 cleavage into its active form, capasase-1, which subsequently cleaves pro-IL-1β, pro-IL-18 and Gasdermin D. Mature IL-1β and IL-18 are then secreted through pores formed by Gasdermin D to initiate inflammatory responses, although this can also occur through Gasdermin D-independent mechanisms [22],[23]. During severe infections, the inflammasome can also mediate a Caspase-1-dependent form of inflammatory cell death called pyroptosis [20]. However, some infections and distinct inflammasome agonists can uncouple IL-1β release from pyroptosis, thus demonstrating divergent outcomes during inflammation activation [24], [25].

A variety of ligands from differing streptococcus species can both upregulate the expression of and activate TLRs, NLRs and other intracellular sensors. Components of the cell wall of S. pneumoniae upregulates numerous TLRs including TLR1, 2, 6 and 10 [26]. Additionally, pilus components of S. pyogenes, which are necessary for bacterial adhesion, are recognized by TLR2 [27]. Several intracellular sensors are crucial to innate control of streptococcus species found in the oral cavity as well. NLRP3 recognizes S. pneumonia virulence factors pneumolysin and bacterial-derived hydrogen peroxide [28]. S. pneumoniae and S. gordonii are recognized by NLRP6, although the microbial ligand sensed by NLRP6 was not identified in these studies [29], [30]. However, NLRP6 recognizes lipoteichoic acid found on other Gram-positive bacteria, including Listeria monocytogenes [31]. Additionally, AIM2 plays a crucial role in recognizing and mediating an inflammatory response against S. pneumoniae infection [32], thus highlighting the coordination of multiple inflammasome sensors during streptococcus recognition. Other intracellular sensors that recognize streptococcus species include NLRC4, which traditionally senses flagellin and type III secretion system components, although it also recognizes bacteria without flagellin and host-derived DAMPs [33] [34] [35]. It is unclear which intracellular sensors recognize S. anginosus infection to promote inflammation.

Our previous work using the RAW264.7 macrophage cell line indicates that infection by S. anginosus induces robust production of inflammatory cytokines TNFα, IL-1β, and IL-6, induction of inflammatory mediators including active NF-κB, iNOS, and COX2, and upregulation of the intracellular sensor NLRP3 [36]. Here, we demonstrate that S. anginosus induces a similar proinflammatory protein and cytokine profile in primary bone marrow derived macrophages (BMMs), including IL-1β production. Moreover, S. anginosus replicates within resting BMMs, and while heat-inactivated S. anginosus induces early proinflammatory signaling by NF-κB, it fails to induce IL-1β, indicating bacterial viability is crucial to mounting a fulminate inflammatory response. However, early NF-κB activation by live S. anginosus was not dependent on TLR2, TLR4 or MyD88 alone, but NF-κB induction in response to heat-killed S. anginosus did require MyD88, an adaptor protein required for multiple TLRs. Additionally, S. anginosus upregulates multiple intracellular sensors including NLRP3, NLRC4 and AIM2. Genetic ablation or pharmacological inhibition of NLRP3 in S. anginosus-infected BMMs lead to reduced IL-1β production. Thus, S. anginosus recognition and downstream IL-1β production is partially mediated by the NLRP3 inflammasome in vitro. Lastly, infection of Asc^−/− and to a lesser extent Nlrp3^−/− mice with S. anginosus results in diminished weight loss, increased bacterial burden and reduced production of proinflammatory mediators, indicating that the inflammasome facilitates the acute inflammatory response and clearance of S. anginosus in vivo.

2. Materials and Methods

2.1. Bacteria

Streptococcus anginosus was purchased from the American Type Culture Collection (ATCC 33397) and maintained in a 50% glycerol stock solution at −80°C until use. Streptococcus anginosus containing a TFP (TFP-S. anginosus) reporter and spectinomycin resistance was a kindly provided by Jeffery Banas (University of Iowa) [37] and maintained in 50% glycerol stock solution at −80 until use. For all described experiments, S. anginosus was grown on blood agar plates containing Tryptic Soy Agar (22091–500G Millipore) and 20% defibrinated donor sheep blood (610–500 Quad Five) at 37°C for 18 hrs. TFP-S. anginosus was grown on blood agar plates containing Tryptic Soy Agar (22091–500G Millipore), 20% defibrinated donor sheep blood (610–500 Quad Five) and 500 ug/mL of spectinomycin at 37°C for 18 hrs. On the day of infection, a loop of plated S. anginosus was inoculated into Brain Heart Infusion Broth (53286–500G Millipore) and grown to an OD 600 of 0.1 (5.4×10⁷ CFUs/mL) at 37°C [36].

2.2. Mice

All animal protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Arizona in accordance with the US National Institutes of Health Guide for the Care and Use of Laboratory Animals. Animal numbers were empirically determined to optimize numbers necessary for statistical significance based on our previous reports utilizing these disease models while adhering to IACUC polices on maximum numbers of animals used in research. Animals had access to water and food ad libitum. All experiments were performed under specific pathogen-free conditions using 6- to 8-week-old mice. Wild-type (WT) C57BL/6J mice and Nlrp3^−/− (B6.129S6-Nlrp3/J) mice were originally obtained from the Jackson Laboratories (Bar Harbor, ME) and maintained at University of Arizona Health Sciences Animal Facility. Asc^−/− mice were generously provided by Dr. Vishva Dixit (Genetech) and housed at the University of Arizona [38]. Nlrc4^−/− mouse bones were generously provided by Dr. Isabella Rauch (Oregon Health and Science University). Aim2^−/− mice were generated as previously described and housed at the University of Arizona [39]. Tlr2^−/− (B6.129-Tlr2^tm1Kir/J), Tlr4^−/− (B6(Cg)-Tlr4^tm1.1Karp/J), Myd88^−/− (B6.129P2(SJL)-Myd88/J) and Caspase-1^−/− (B6.Cg-Casp1^em1Vnce/J) mice were purchased from Jackson Laboratories.

2.3. Cell Culture

Primary bone marrow macrophages (BMMs) were generated by culturing bone marrow isolated from mouse femurs in 10 mLs of DMEM containing 4.5 g/L glucose, L-glutamine, sodium pyruvate (10–013-CV Corning), 20% heat inactivated fetal bovine serum (Gibco), 15% conditioned media from GM-CSF-producing CMG14–12 cells and 1X penicillin-streptomycin (HyClone Thermo) for 7 days. Adherent BMMs were removed from the plates using 0.25% Trypsin-EDTA (Corning) and plated overnight prior to infection.

2.4. In vitro infection

Primary BMMs were plated at a seeding density of 1.5×10⁵ for ELISA and 3×10⁵ or 2×10⁶ for Western Blot on 48 well, 12 well plates and 6 well plates, respectively, in triplicate per genotype. BMMs were infected with S. anginosus at a MOI of 10 for 24 hours at 37°C in 200 uL or 1 mL of media without antibiotics for ELISA or Western Blot, respectively. For S. anginosus viability studies, S. anginosus was heat killed by exposure to 80°C for 30 minutes prior to infecting BMMs. A subset of heat killed S. anginosus was plated on blood agar plates and kept at 37°C for 18 hours to ensure no live bacteria remained. WT and Nlrp3^−/− BMMs were treated with 5 μg/mL LPS for 3 hours followed by 7.5 μg/mL of Nigericin (InvivoGen) for 1 hour as a positive control for NLRP3 inflammasome activation. WT and Aim2^−/− BMMs were treated with 5 μg/mL of LPS for 3 hours followed by 1 μg/mL of poly(dA:dT) (InvivoGen) in lipofectamine for 6 hours as a positive control for AIM2 inflammasome activation. For infections also involving pharmacological inhibition of the NLRP3 inflammasome, 10 mM of MCC950 or 0.01% DMSO (vehicle control) was added to each well containing DMEM without FBS for 15 minutes prior to infection. Cells were washed with PBS, supplemented with DMEM containing FBS and infected with S. anginosus for 24 hours. After infection or treatment with inflammasome positive control ligands, supernatant or protein was collected and stored at −80°C before assessing cytokine production by ELISA or protein measurements by Western Blot.

2.5. In vivo Infection

Mice were injected intraperitoneally with 2×10⁹ CFUs of S. anginosus suspended in 200 μL of PBS. Animals were weighed at 0, 5, 24, 28 and 48 hours post-infection. After 8 or 16 hours post-infection, the mice were euthanized and given an i.p. injection of 4–5 mLs sterile PBS, which was removed and centrifuged at 450 × g for 5 minutes to pellet eukaryotic cells. The remaining supernatant containing bacteria was serially diluted and plated on blood agar plates to assess CFUs/mL of S. anginosus within the peritoneal cavity. Red blood cells within the cell pellet were lysed using ACK lysis buffer (Gibco A10492–01), the cells were washed once with sterile PBS and centrifuged again at 450 × g for 5 min. RNA was extracted from 1.5×10⁵ of the pelleted cells, protein was extracted from 2×10⁶ of the pelleted cells and FACS analysis was carried out on 5×10⁵ of the pelleted cells. Lastly, the spleens were removed from the mice, snap frozen in liquid nitrogen and stored at −80°C for RNA extraction.

2.6. Crystal Violet Staining and Imaging Quantification

S. anginosus-infected BMMs were washed with PBS and fixed with 4% paraformaldehyde for 15 minutes at room temperature. Fixed BMMs were stained with 0.25% crystal violet for 30 minutes at room temperature. Crystal violet was aspirated after the incubation. The cells were washed with water until all stain was removed and air dried for 15 minutes. BMMs were imaged at 20X using ECHO Revolve D2224. BMM areas were quantified using ImageJ software (NIH).

2.7. Protein Purification for Caspase-1 Western Blot

Protein was purified from supernatant as described in NLR Proteins Methods and Protocols by Springer Protocols, with some modifications [40]. A total of 2×10^6 BMMs were plated in a 6 well plate containing 1mL of DMEM with 10%FBS and infected with S. anginosus at a MOI of 50 for 6 hours. Supernatant was removed and centrifuged at 1500 × g for 5 minutes at room temperature to pellet bacteria. Supernatant was removed, and 100 μL of chloroform and 500 μL of methanol was added to 500 μL of supernatant. The samples were vortexed to mix, and were centrifuged for 3 minutes at 15,800 × g at room temperature. After centrifuging, a white disk was present at the interphase between the aqueous and organic phase. The upper phase was discarded without disrupting the white disk. 500 μL of chloroform was added to the lower phase, and vortexed to mix. Samples were centrifuged again at 15,800 × g for 3 minutes at room temperature. After centrifuging, a white pellet was present at the bottom of the tube. Supernatant was discarded, and the pellets were left to dry in a sterile hood for 30 minutes until they turned a transparent-yellowish color. Pellets were resuspended in 50 μL of 4X LD containing DTT (Life technologies). The back of the Eppendorf tubes were scratched with a pipette tip to fully resuspend any protein that remained on the side of the tube. Samples were boiled at 98°C and sonicated using a sonicating water bath on low for 12 one second pulses. The samples were then subjected to SDS-PAGE to determine caspase-1 expression by western blot as described below.

2.8. Western Blot

Cells were lysed in 100 uL of RIPA lysis buffer (Boston Bioproducts) containing PhosSTOP (Roche) and Complete protease inhibitor (Roche). Lysates were then centrifuged at 4°C at 15,000 × g for 10 minutes to pellet the insoluble fraction. Soluble lysate was removed, 4X LDS sample buffer (Life technologies) with 0.5M DL-dithiothreitol (R&D) was added, and the samples were stored at −80°C until assessment by western blot.

For western blot analysis, protein lysates was separated on a SDS-polyacrylamide gels (BioRad) and transferred onto nitrocellulose membranes (Cytiva). Membranes were blocked for 1 hour in 5% non-fat milk in Tris-Buffered Saline with Tween (TBST), with the exception of the caspase-1 blots, which were blocked in 10% non-fat milk in TBST. Membranes were incubated with primary antibodies diluted at 1:1,000, with exception of anti-Caspase-1 and anti-Gasdermin D, which were diluted at 1:200, in 5% non-fat milk in TBST overnight at 4°C. Primary antibodies were washed five times with TBST, and membranes were then incubated with secondary antibodies conjugated to horseradish peroxidase diluted at 1:10,000 in 5% non-fat milk in TBST for two hours at room temperature, with the exception of caspase-1 blots, which were incubated with secondary antibodies at a dilution of 1:5,000 in 10% non-fat milk in TBST for one hour at room temperature. The membranes were washed with TBST and exposed to SuperSignal West Pico PLUS Chemiluminescent Substrate (ThermoFisher Scientific) or SuperSignal West Femto Maximum Sensitivity Substrate (ThermoFisher Scientific) for visualization on Blue Devil Film, which was developed using an automatic film processor (OPTIMAX). Protein band intensity was quantified using densitometric analysis via ImageJ software. Primary antibodies used were rabbit mAb anti-COX2 (Cell Signaling 12282S), rabbit mAb anti-NLRC4 (Cell Signaling 12421T), rabbit mAb anti-NLRP3 (Cell Signaling 1510S), rabbit mAb phospho-p65 (Ser536) (Cell Signaling 3033T), rabbit mAb anti-p65 (Cell Signaling 8242S), rabbit mAb anti-iNOS (abcam ab283655), rabbit mAb anti-AIM2 E518Z (Cell Signaling 53491S), mouse mAb anti-Caspase-1 p20 (Adipogen AB-20B-0042_C100), rabbit mAb anti-Gasdermin D (Cell Signaling 39754S), rabbit mAb anti-p-SAPK/JNK (Thr183/Tyr185) (Cell Signaling 4668T), rabbit mAb anti-SAPK/JNK (Cell Signaling 9252T), rabbit mAb anti-p-Akt (Ser473) (Cell Signaling 4961L), rabbit mAb anti-Akt (pan) (Cell Signaling 4060L), rabbit mAb anti-p-p44/42 MAPK (Thr202/Tyr204) (Cell Signaling 4370S), mouse mAb anti-ERK 1/2 (Santa Cruz Biotechnology sc-514302) and anti-β-Actin-HRP (Santa Cruz Biotech A1723). Goat Anti-Rabbit IgG, HRP conjugate (Millipore 12–348) and Goat Anti- Mouse IgG HRP conjugate (Millipore 12–394) was used as a secondary antibody for all listed primary antibodies with the exception of β-Actin.

2.9. ELISA

Enzyme-linked immunosorbent assays (ELISA) were performed on macrophage supernatants per manufacturer’s instructions. Briefly, a 96 well ELISA plate was coated with capture antibody overnight at room temperature and washed three times with wash buffer. The plate was blocked with reagent diluent at room temperature for 1 hour and washed three times with wash buffer. Cytokine standards, cell supernatants and animal serum was added to the plate for a 2 hour incubation at room temperature. The plate was washed three times with wash buffer. Detection antibody was added to the plate for a 1 hour incubation at room temperature. The plate was washed three times with wash buffer, and Streptavidin-HRP was added for a 20 minute incubation at room temperature. The plate was washed three times with wash buffer, and substrate solution was added to the plate for a 20 minute incubation at room temperature. Stop solution was added to the plate, and optical density was measured using a microplate reader set to 450 nm. The ELISA kits used included Mouse IL-1 beta/IL-1F (R&D Systems DY401), Mouse TNF-alpha (R&D Systems DY410–05) and Mouse IL-6 (R&D Systems DY406–05).

2.10. Bacteria Killing Assay

Primary BMMs were plated in triplicate on 12 well plates or on a 30mm × 10mm Glass Bottom Tissue Culture Dish (Cell Treat 229632) at a seeding density of 2×10⁵ cells/well or per dish and allowed to adhere overnight at 37°C. BMMs were infected with S. anginosus or TFP-S. anginosus at a MOI of 10 for 30 minutes. BMMs were then washed with PBS to remove external bacteria and treated with 150 ug/mL of gentamycin for 30 minutes to neutralize any non-internalized bacteria before a final PBS wash. The BMMs were then incubated in antibiotic-free media for 2.5 or 24 hours. The BMMs were given a final PBS wash and lysed in 100 uL of sterile water with scraping or were fixed with 4% PFA for Microscopy. The lysates were serially diluted five times, plated on blood agar plates and incubated at 37°C overnight for CFU counts the following day. A subset of macrophages were washed and lysed or fixed with 4% PFA after the initial 30 min bacterial infection to determine the total number of internalized bacteria.

2.11. LDH Assay

S. anginosus-induced macrophage cell death was assessed using a Lactate dehydrogenase (LDH) assay according to manufacturer’s protocol (AbCam ab65393). In brief, primary BMMs were plated in triplicate on 48 well plates at a seeding density of 1.5×10⁵ cells/well and incubated overnight at 37°C. BMMs were infected with S. anginosus at a MOI of 1, 10, 25, 50, 75 and 100 for 24 hours, and supernatant was taken. A subset of uninfected BMMs were incubated with lysis buffer as a positive control. A total of 10 μl of supernatant was mixed with 100 μl of LDH reaction mix and incubated at room temperature for 30 mins. The samples were then read at OD 450 nm using a microplate reader. The %LDH release was calculated as the sample OD 450 nm divided by the positive control OD 450 nm.

2.12. RT-qPCR

Total RNA was extracted from peritoneal cells and spleens using PureLink RNA Mini Kit (ambion 12183025) or RNeasy Mini Kit (250) (Qiagen 74106), respectively. Briefly, cells or tissue were homogenized in 350 μL of lysis buffer containing 1% 0.5 BME and 0.5% Dx Reagent. Lysates were centrifuged through a Qia shredder column followed by a genomic-DNA elimination column. The flow though was washed with an equal volume of 70% Ethanol and centrifuged through a RNA easy column. The flow through was washed and centrifuged three more times with RW1 buffer followed by RPE buffer containing Ethanol. Columns were dried for 2 minutes and rehydrated with 30 μL of molecular grade water. RNA was eluted through the column, and concentration was assessed using IMPLEN NanoPhotometer NP80. cDNA was created using either 50ng or 400ng of RNA from cells or tissue, respectively, with 5x iScript RT Supermix (BioRad L0014048) and nuclease free water for a total volume of 20 μL. RT-qPCR was ran using 4.75 μL cDNA, 5.25 μL of Taqman 2X universal PCR Master Mix (Applied Biosystems) and 0.25 μL of Taqman primer probes per reaction. Relative expression was quantified using the ddCt method. Taqman assays were purchased from Thermo Fisher. Primers used were Gapdh (Mm99999915_g1), Il1b (mM00434228_m1), Tnf (Mm00443258_m1), Il6 (Mm0046190_m1), Nos2 (Mm00440502_m1), and Arg1 (Mm00475988_m1).

2.13. Caspase-Glo 1 Inflammasome Assay

A total of 2×10⁶ BMMs were infected with S. anginosus at a MOI of 50 for 6 hours in 1mL of DMEM containing 10% FBS. Caspase-1 activity was assessed as described in the Promega Caspase-Glo 1 Inflammasome Assay kit (Promega 72532). Briefly, 50 μL of supernatant or 50 μL of control media with 10% FBS was plated in a white walled 96-well plate for each biological replicate and mixed with caspase-1 glo reagent containing an inhibitor targeted against nonspecific activity, MG-123, with or without caspase-1 inhibitor, Ac-YVAD-CHO. The plate was mixed on an orbital shaker for 30 seconds at 300 rpm, then was incubated at RT for 60 minutes in the dark. The luminescence was read using a microplate reader. The final values representing caspase-1 activity were calculated by first subtracting the blank values quantified from the wells containing only DMEM and FBS from each sample and then by subtracting the luminescence quantified from wells containing Ac-YVAD-CHO to normalize against any non-caspase-1-specific activity.

2.14. Flow Cytometry

A total of 5×10⁵ peritoneal cells were analyzed by Fluorescence-Activated Cell Sorting (FACS). Cells were stained with zombie violet viability dye at a dilution of 1:500 in 100 μL for 30 minutes at 4°C. Cells were then fixed with 4% paraformaldehyde for 15 minutes at 4°C followed by washing twice in 200 μL FACs buffer containing 3% FBS and 0.25% sodium azide in PBS. Cells were stained with 100 μL appropriate antibody master mix for 30 minutes at 4°C. Cells were washed twice in 200μL FACs buffer, resuspended in a final volume of 300 uL of FACs buffer and kept at 4°C. Flow Cytometry was performed on FACScanto II Flow Cytometer and data was analyzed with FlowJo v11 (Becton Dickinson). Antibodies used were Ly6G (1A7, 127651 Biolegend), F4/80 (BM8, 123113 Biolegend) and CD45.2 (104 11-0454082 Invitrogen). Zombie violet viability dye (77477 Biolegend) was used to stain dying cells.

2.15. Statistical Analysis

The data were expressed as the mean +/− the standard error of the mean (SEM) of values obtained from experiments. Statistical analysis was performed using GraphPad Prism Software using unpaired t test for comparison of two means or ANOVA for comparison of 3 or more means. P values of <.05 were considered statistically significant.

3. Results

3.1. S. anginosus infection elicits an inflammatory response in bone marrow macrophages in vitro

Our previous work indicates that S. anginosus induces a robust proinflammatory response accompanied by distinct morphologic changes in RAW 264.7 cells after 6 hour infection [36]. In order to assess the inflammatory potential of S. anginosus in more physiologically relevant primary cells, we generated bone marrow-derived macrophages (BMMs) from C57BL/6 (wild type) mice and infected them with S. anginosus at a MOI of 10 for 6 and 24 hours. After 6 hours of S. anginosus infection, primary BMMs displayed an elongated phenotype with a significantly smaller area than uninfected cells (Fig 1a), similar to S. anginosus-infected RAW 264.7 cells [36]. However, following 24 hours post-infection, BMMs displayed an enlarged phenotype and lost their elongated phenotype (Fig 1a). These findings suggest that primary BMMs display morphological changes reminiscent of an activated state in response to S. anginosus infection. Therefore, we next examined proinflammatory signaling and downstream proteins in S. anginosus-infected BMMs. This included phosphorylated levels of the NF-κB family member p65 and expression of iNOS and COX2, which are responsible for production of proinflammatory effectors nitric oxide and prostaglandins, respectively. After 6 hours of infection with S. anginosus, BMMs displayed a significant increase phospho-p65, iNOS and COX2 (Fig 1b). These pro-inflammatory markers remained elevated at 24 hours post infection (Fig 1c). The NF-κB family of transcription factors are required for the production of proinflammatory cytokines. Thus, we infected BMMs with S. anginosus for 24 hours and measured the levels of proinflammatory cytokines TNFα, IL-1β and IL-6 in supernatants. TNFα, IL-1β and IL-6 were significantly upregulated after 24 hour S. anginosus infection (Fig 1d). These results indicate that S. anginosus induces a robust pro-inflammatory response in primary BMMs in vitro.

Figure 1. S. anginosus infection elicits an inflammatory response in bone marrow macrophages in vitro:

a) 40x images of BMMs infected with S. anginosus for 6h and 24h at a MOI of 10 and uninfected, stained with crystal violet. Quantification of 2000–3000 cell areas per condition. b) western blot and densitometry of phospho-p65, p65, iNOS and COX2 from BMMs uninfected and infected for 6h at a MOI of 10. Quantification of three biological replicates. c) western blot and densitometry of phospho-p65, p65, iNOS and COX2 from BMMs uninfected and infected for 24h at a MOI of 10. Quantification of three biological replicates. d) ELISA of TNFα, IL-1β and IL-6 from BMM supernatant uninfected and infected for 24h at a MOI of 10. Quantification of three biological replicates. Data is presented as mean +/− SEM. Stats quantified using an unpaired T test through GraphPad.

3.2. S. anginosus viability is required to induce fulminate inflammatory responses in bone marrow macrophages

We next determined the viability requirements of S. anginosus to induce a proinflammatory response in BMMs. To test this, we inactivated S. anginosus by heat exposure at 80°C for 30 minutes in order to maintain the structural integrity of S. anginosus, which should retain the capacity to activate surface PRRs on BMMs without releasing additional factors resulting from bacterial lysis. To confirm inactivation, we plated a subset of heat inactivated S. anginosus on blood agar plates, which resulted in no measurable CFUs. We then exposed BMMs to either live or heat-killed (HK) S. anginosus for 24 hours and assessed BMM morphologic changes after crystal violet staining by microscopy. Both live and HK S. anginosus induced an enlarged phenotype in BMMs, however HK S. anginosus induced BMM enlargement at a significantly reduced capacity compared to live S. anginosus (Fig 2a). Moreover, live S. anginosus infection induced a greater elongated phenotype in BMMs compared to HK S. anginosus infection. These findings indicate that S. anginosus viability may be partially responsible for its ability to activate macrophages. We next investigated the role of cellular activation pathways that may be triggered during S. anginosus infection. We infected wild type BMMs with live and HK S. anginosus and assessed phosphorylation levels of the PI3K member Akt and the MAPK members ERK and JNK by western blot. We found a significant increase in phospho-ERK in wild type BMMs infected with live S. anginosus, which was reduced in BMMs infected with HK S. anginosus. Live S. anginosus failed to induce significant Akt and JNK activation (Supplemental Figure 1a). These findings indicate that S. anginosus activates ERK, and this is dependent of bacterial viability. Next, to determine if S. anginosus viability is required for its ability to induce proinflammatory pathways in macrophages, we assessed levels of phosphorylated p65, iNOS and COX2 in BMMs infected with live or HK S. anginosus. While HK S. anginosus induced similar levels of phospho-p65 as live S. anginosus, HK S. anginosus displayed a reduced capacity to induce iNOS and COX2 in BMMs (Fig 2b). Additionally, BMMs infected with heat killed S. anginosus produced significantly less proinflammatory cytokines such as IL-1β and TNF⍺ compared to BMMs given live S. anginosus (Fig 2c). These findings indicate that S. anginosus exposure leads to activation of the NF-κB pathway independent of bacterial viability. This is likely mediated by TLRs that recognize microbial products that are present on the surface of S. anginosus, irrespective if live or heat killed. However, infection with live S. anginosus was needed for a strong proinflammatory response and induction of important downstream proinflammatory proteins and cytokines. Overall, this indicates that S. anginosus viability is needed for the complete inflammatory response initiated by primary BMMs and the induction of proinflammatory proteins and cytokines.

Figure 2. S. anginosus viability is required to induce fulminate inflammatory response in bone marrow macrophages:

a) 40x images of BMMs uninfected or infected for 24h with live or heat killed S. anginosus at a MOI of 10, stained with crystal violet. Quantification of 4000–6000 cell areas per condition. b) western blot and densitometry of phospho-p65, p65, iNOS and COX2 from BMMs uninfected or infected for 24h with live or heat killed S. anginosus at a MOI of 10. Quantification of three biological replicates. c) ELISA of TNFα and IL-1β from BMM supernatant uninfected or infected for 24h with live or heat killed S. anginosus. Quantification of three biological replicates. Data is presented as mean +/− SEM.

3.3. Activation of Early NF-κB signaling by Heat-Killed S. anginosus is not dependent on TLR2 or TLR4 alone, but requires MyD88

Infection of BMMs with HK S. anginosus induced NF-κB signaling seen through increased phospho-p65 expression similar to what is seen with live S. anginosus infection. Next, we explored the molecular mechanism by which this occurs. As external bacterial sensing is largely carried out through Toll-like receptors (TLRs) on the surface of macrophages, we tested the requirement of TLR2 and TLR4 in sensing live versus HK S. anginosus by infecting BMMs lacking TLR2 (Tlr2^−/−) or TLR4 (Tlr4^−/−). TLR2 recognizes peptidoglycan molecules and lipoproteins found on the surface of gram-positive bacteria, such as lipoteichoic acid (LTA) [18]. Compared to wild type counterparts, we found no significant difference in phospho-p65 or TNFα secretion in Tlr2^−/− BMMs exposed to HK S. anginosus (Fig 3a, d). However, TNFα levels were reduced, but not ablated, in Tlr2^−/− BMMs infected with live S. anginosus and were reminiscent of levels detected in HK S. anginosus-infected wild type BMMs. This finding points towards a partial role for TLR2 during the detection of live S. anginosus by macrophages. We next tested the requirement of TLR4 during S. anginosus detection using Tlr4^−/− BMMs. TLR4 recognizes molecules found on gram negative bacteria such as LPS, although it also detects a variety of DAMPs (e.g., HMGBI) and plays a role in detecting ligands found in endosomal compartments [18]. Compared to wild type counterparts, we found no significant diminishment in phospho-p65 in Tlr4^−/− BMMs infected with live or HK S. anginosus (Fig 3b). Interestingly, TLR4 was dispensable for TNF⍺ production in response to live S. anginosus infection, but loss of TLR4 lead to recovery of HK S. anginosus-induced TNF⍺, possibly due to compensatory PRR signaling and/or heat-induced alterations in PAMP expression (Fig 3d). These findings indicate that TLR2 and TLR4 on their own are not required for S. anginosus-induced NF-κB signaling, but do not rule out potential redundancy between these TLRs or the contribution of other TLRs. To determine if multiple TLRs play a role in S. anginosus-induced NF-κB signaling, we performed S. anginosus infections with BMMs lacking Myd88 (Myd88^−/−), which is a common signaling adaptor protein needed for proinflammatory signaling by multiple TLRs, including TLR2 and partially TLR4 [41]. While no changes were seen in response to live S. anginosus infection, we found a significant diminishment in phospho-p65 in Myd88^−/− BMMs exposed to HK S. anginosus in comparison to wild type BMMs (Fig 3c). Additionally, we found significantly reduced TNF⍺ secretion in Myd88^−/− BMMs in response to both live and HK S. anginosus infection in comparison to wildtype BMMs (Fig 3d). Together this indicates that early inflammatory signaling by HK S. anginosus is likely mediated by multiple TLRs and requires MyD88. Overall, these experiments reveal partial and differential roles for TLR2 and TLR4 during S. anginosus recognition, with neither TLR being exclusively responsible for the potent NF-κB activation induced by live or HK S. anginosus.

Figure 3. Activation of Early NF-κB signaling by Heat-Killed S. anginosus is not dependent on TLR2 or TLR4 alone, but requires MyD88:

a) densitometry of phospho-p65 and p65, from lysate from WT or Tlr2^−/− BMMs infected with live or heat-killed S. anginosus at a MOI of 10 for 6 hours. (Left). Treatment with LTA for 1hr was used as a positive control. Quantification of three biological replicates. Western blot of densitometry (Right). b) densitometry of phospho-p65 and p65 from lysate from WT or Tlr4^−/− BMMs infected with live or heat-killed S. anginosus at a MOI of 10 for 6 hours (Left). 1h LPS prime followed by 3h nigericin treatment was used as a positive control. Quantification of three biological replicates. Western blot of densitometry (Right). c) densitometry of phospho-p65 and p65 from lysate from WT or Myd88^−/− BMMs infected with live or heat-killed S. anginosus at a MOI of 10 for 6 hours. (Left) 1hr LPS+LTA prime followed by 3hr nigericin treatment was used as a positive control. Quantification of three biological replicates. Western blot of densitometry (Right) d) ELISA of TNFα from WT, Tlr2^−/−, Tlr4^−/−, and Myd88^−/− BMM supernatant uninfected or infected with live or heat-killed S. anginosus at a MOI of 10 for 6 hours. Quantification of three biological replicates. Data is presented as mean +/− SEM.

3.4. S. anginosus replicates within infected macrophages and promotes macrophages cell death at high MOI.

S. anginosus viability was required to promote production of downstream proinflammatory proteins and cytokines from macrophages. This lead us to postulate that live S. anginosus presents an insult to infected macrophages that necessitate a robust inflammatory response. For example, while bacteria are typically neutralized following phagocytosis, other oral streptococcus species including S. gordonii replicate intracellularly, which triggers production of the potent inflammatory cytokine IL-1β [30]. In order to better understand how macrophages respond to S. anginosus, we next assessed the ability of S. anginosus to persist within BMMs using a bacterial killing assay as done previously with S. gordonii [30]. Accordingly, we infected BMMs with S. anginosus for 30 min, then treated BMMs with gentamicin to neutralize any non-internalized bacteria and incubated the BMMs for an additional 2.5 or 24 hours. BMMs were lysed, and lysates were then plated on blood agar plates to determine CFU counts. A subset of BMMs were lysed after the initial 30 min infection to determine the total number of internalized bacteria. We observed an initial decrease in internalized S. anginosus at 2.5 hours post-infection relative to the 30 min baseline measurements (representing the total input of internalized bacteria). However, at 24 hours post-infection, S. anginosus CFUs were significantly increased compared to the initial 30 min time points (Fig 4b). These results suggest that macrophages fail to fully neutralize phagocytosed S. anginosus, resulting in replication of S. anginosus in infected BMMs. To confirm these findings using a second approach, we repeated the bacteria killing assay with a previously reported strain of S. anginosus containing a teal-fluorescent reporter (TFP-S. anginosus) and assessed bacterial growth through microscopy [37]. BMMs were infected for 30min, 2.5hr and 24hr as described in Fig. 4a and were fixed for imaging by microscopy (Fig 4c). We found at 30 mins post-infection, around 60% of BMMs were associated with TFP-S. anginosus, which diminished significantly after the 30 min gentamycin treatment and additional 2.5 hr incubation (Fig 4d), similar to the CFUs detected at these time points (Fig 4b). This suggests that a large portion of TFP-S. anginosus is associated with the surface of BMMs at 30 mins post-infection, even after washing with PBS. However, at 24 hours post-infection, these levels of TFP-S. anginosus returned to levels observed at the initial 30 min post-infection time point (Fig 4d). These experiments further support that S. anginosus is internalized and replicates within BMMs over a period of 24 hours. We next asked if S. anginosus persistence within BMMs in turn promotes macrophage cell death during infection. Our prior work found that S. anginosus does not induce any measurable cell death in the RAW264.7 macrophage cell line at an MOI 10, [36], however this has not been confirmed in primary macrophages. To address this, we measured LDH release from BMMs infected with S. anginosus at range of MOIs from 1–100. BMMs infected at MOIs of 1 and 10, the maximum MOI used in this study, had LDH release similar to that of uninfected cells, indicating S. anginosus induced a robust inflammatory response without causing measurable cell death in macrophages. However, exposure of BMMs with S. anginosus at MOIs of 25 and higher lead to a significant release of LDH, indicating S. anginosus was capable of inducing cell death, but only at high MOIs (Fig 4e). Because our previous experiments used infections at an MOI of 10 to assess S. anginosus-induced inflammatory responses in BMMs, we next determined if S. anginosus-induced production of proinflammatory cytokines (i.e., IL-1β) was responsive to increasing MOIs. BMMs infected with S. anginosus at MOIs of 10, 25 and 50 secreted a similar amounts of IL-1β (Fig 4e). Together, this suggests that macrophages fail to effectively neutralize S. anginosus following phagocytosis, and that S. anginosus is instead capable of replicating within BMMs. Moreover, despite replicating within macrophages, S. anginosus does so without promoting macrophage cell death unless at a high (>10) MOI.

Figure 4. S. anginosus replicates within macrophages and promotes macrophages cell death at high MOI:

a) schematic of the bacteria killing assay b) bacteria killing assay looking at internalized S. anginosus after 30 min, 2.5h and 24h infection of WT BMMs. Shown as an average of three biological replicates per time point. c) 40x immunofluorescence images of WT BMMs infected with TFP-S. anginosus at 30 mins, 2.5h and 24hrs post infection. d) Quantification of internalized TFP-S. anginosus from immunofluorescence imaging in c. Quantification of 100 cells per timepoint. e) LDH assay of WT BMMs uninfected and infected at a MOI of 1, 10, 25, 50, 75 and 100 with S. anginosus. Percent LDH released quantified as experimental value divided by the positive control. Quantification of three biological replicates (Left). ELISA of IL-1β from WT BMM supernatant uninfected or infected at a MOI of 10, 25 and 50 for 24 hours. Quantification of three biological replicates (Right).

3.5. S. anginosus activates the NLRP3 inflammasome

The innate immune system largely relies on intracellular PRRs to recognize and mount immune responses against pathogens that actively access host cells for their replication. Among the most notable intracellular PRR families are the inflammasome-producing NLRs, which lead to the production of IL-1β, a potent proinflammatory cytokine that is often produced by macrophages to combat intracellular infections [42]. Because S. anginosus replicated within macrophages and its viability was required for IL-1β production, we hypothesized that the inflammasome contributes to the recognition of live S. anginosus infection. Production of IL-1β is mediated by a 2-step process involving: 1) a priming step whereby NF-κB activation (often via TLR signaling) leads to transcription and production of inactive pro-IL-1β; and 2) activation and oligomerization of the multi-protein complex termed the inflammasome, which activates the enzyme Caspase-1 through proteolytic processing, leading to IL-1β activation and release from the cell [43]. Several inflammasome sensors are triggered by other Streptococcus species, including NLRP3, NLRC4 and AIM2 [28], [35], [32]. To begin exploring the involvement of the inflammasome, we infected BMMs with live and HK S. anginosus, then measured caspase-1 activation using two approaches; 1) probing for cleaved caspase-1 by western blot; and 2) measuring caspase-1 activity using a commercially available caspase-1 glo assay. In line with our observation that live S. anginosus induces IL-1β production (Fig. 2c), we found a significant increase in Caspase-1 cleavage and Caspase-1 glo luminescence in BMMs infected with live S. anginosus, which was reduced in response HK S. anginosus exposure (Figure 5a). Additionally, we infected BMMs lacking Caspase-1(Caspase-1^−/−), which facilitates IL-1β cleavage and secretion downstream of inflammasome complexing. We found a significant diminishment of IL-1β secretion in Caspase-1^−/− BMMs infected with S. anginosus in comparison to wildtype BMMS, indicating involvement of the canonical caspase-1 inflammasome during infection (Fig 5b). To further confirm that S. anginosus-induced IL-1β production was inflammasome dependent, we infected BMMs lacking the common inflammasome adapter protein ASC (Asc^−/−) with S. anginosus for 24 hours and assessed IL-1β secretion. S. anginosus-infected Asc^−/− BMMs produced normal levels of TNF⍺, indicating that NF-κB activation was intact in these BMMs (Fig. 5c). However, there was significant diminishment of IL-1β production indicating that S. anginosus-induced IL-1β production is inflammasome dependent (Fig 5c). Because the inflammasome can induce a form of cell death called pyroptosis through cleavage of gasdermin D, which forms pores in the cell membrane, we assessed if S. anginosus induces gasdermin D cleavage in macrophages during infection. BMMs infected with live or HK S. anginosus at a MOI of 10 displayed no significant induction of cleaved gasdermin D. This suggests that S. anginosus also fails to induce significant BMM pyroptosis at an MOI of 10 (Supplemental Figure 1b), which is in line with a lack of LDH release at similar MOI (Fig. 4e). Overall, these findings implicate the involvement of the inflammasome during the proinflammatory response induced by live S. anginosus infection.

Figure 5. S. anginosus activates the NLRP3 inflammasome:

a) western blot of Caspase-1 from WT BMMs infected with live or heat killed S. anginosus at a MOI of 50 for 6 hours (Left). Densitometry of cleaved caspase-1 (Middle). Caspase-Glo 1 Assay from supernatant of WT BMMs infected with live and heat killed S. anginosus at a MOI of 50 for 6 hours. b) ELISA of IL-1β from WT and Caspase-1^−/− BMM supernatant infected for 24h. 1h LPS prime followed by 3h nigericin treatment was used as a positive control. Quantification of four biological replicates, or two biological replicates, repeated, for WT and Caspase-1^−/−, respectively. c) ELISA of TNFα and IL-1β from WT and Asc^−/− BMM supernatant uninfected or infected for 24h. Quantification of three biological replicates (Left). Western blot of NLRP3, NLRC4 and AIM2 from BMMs uninfected or infected for 24h with S. anginosus at a MOI of 10 (Right). d) ELISA of TNFα and IL-1β from WT and Nlrc4^−/− BMM supernatant uninfected or infected for 24h with S. anginosus. Quantification of three biological replicates. e) ELISA of TNFα and IL-1β from WT and Aim2^−/− BMM supernatant infected for 24h. 1h LPS prime followed by 6h treatment of poly dA:dT was used as a positive control. Quantification of three biological replicates. f) ELISA of TNFα and IL-1β from WT and Nlrp3^−/− BMM supernatant infected for 24h. 1h LPS prime followed by 3h nigericin treatment was used as a positive control. Quantification of three biological replicates. g) ELISA of IL-1β from WT BMM supernatant infected for 24h with or without 10mM NLRP3 inhibitor, MCC950. Data is presented as mean +/− SEM.

Because S. anginosus replicated within macrophages and its viability was required for IL-1β production, but not TLR2/4-mediated NF-κB activation, we next explored the role of NLRs during the recognition of S. anginosus infection. Indeed, the protein expression of NLRP3, NLRC4 and AIM2 were all upregulated in response to S. anginosus infection (Fig 5c), indicating S. anginosus induces the expression of multiple inflammasome sensors. Since prolonged production of IL-1β by inflammasome activation can exacerbate disease such as OSCC [44], and the inflammasome is crucial to the control of many intracellular bacteria, we next investigated the impact of S. anginosus on the activation of specific inflammasomes. To determine which inflammasome sensor(s) are triggered by S. anginosus to induce IL-1β production, we infected BMMs lacking NLRP3, NLRC4, or AIM2. Nlrp3^−/−, Nlrc4^−/− and Aim2^−/− BMMs still produced TNF⍺ (Fig 5d, e, f). In addition, while Nlrc4^−/− and Aim2^−/− BMMs produced similar levels of IL-1β compared to wild type BMMs, Nlrp3^−/− BMMs had significantly diminished, but not complete loss of IL-1β production in comparison to wild type BMMs (Fig 5d, e, f). To confirm that S. anginosus was acting through the NLRP3 inflammasome using a pharmacological approach, we infected wild type BMMs in the presence of DMSO vehicle or 10 mM of the NLRP3 inhibitor, MCC950, for 24 hours. IL-1β was significantly diminished in wild type BMMs treated with MCC950 in comparison to those treated with the vehicle control (Fig 5g). This implicates that S. anginosus activates the inflammasome in part, through triggering NLRP3 to induce IL-1β in vitro.

3.6. S. anginosus elicits an inflammasome-dependent acute inflammatory response in vivo

To determine if S. anginosus induces a proinflammatory response in vivo, we infected wild type mice with 2×10⁹ CFUs of S. anginosus via intraperitoneal injection (i.p.), an approach used to assess the inflammatory capacity of similar streptococcus species in vivo [45]. After 24 hours of infection, wild type mice displayed significant weight loss, which largely recovered by 48 hours, but remained significantly lower than the initial starting weights. To determine the role of the inflammasome during the host response to S. anginosus infection in vivo, we next infected Asc^−/− mice with 2×10⁹ CFUs of S. anginosus. Asc^−/− mice displayed significantly reduced weight loss at 24 hours compared to similarly infected wild type mice, which was almost completely resolved by 48 hours (Fig 6a). These findings indicate that S. anginosus induces an acute pro-inflammatory response in vivo that is resolved by 48 hours and is partially driven by the inflammasome. We next assessed the levels of bacteria present within the peritoneal cavity of these mice 8 hours after infection and found a significant increase in recovered CFUs of S. anginosus in Asc^−/− mice in comparison to wild type animals. (Fig 6b). Next, we assessed markers of tissue inflammation through RT-qPCR and found significantly diminished levels of Il1b, Tnfa, and Il6 expression in the spleen from Asc^−/− mice compared to wild type mice (Fig 6c). Additionally, we measured the expression of these genes in the infiltrating peritoneal cells and found a significant diminishment of proinflammatory Nos2, Il1b, Tnfa and Il6 along with an increase in Arg1 in the infected Asc^−/− mice (Fig 6d). While the infiltrating peritoneal cells from WT mice had readily detectable levels phospho-p65, COX2 iNOS and cleaved Caspase-1, these levels were all significantly reduced in peritoneal cells from Asc^−/− mice, with the exception of iNOS, which was trending, but non-significantly diminished (Fig 6e). Bacterial burdens along with both tissue and cellular inflammatory markers were comparable between Asc^−/− and wild type mice by 16 hours post-infection, with exception of cleaved caspase-1, which remained significantly reduced in Asc^−/− mice (Supplemental Fig 2a–d). To determine if these differences in proinflammatory markers was due to differences in the cell populations recruited to the peritoneal cavity in wild type and Asc^−/− mice, we also performed FACS analysis on the cells collected from the peritoneal wash. We used F4/80 to identify macrophages and Ly6G to identify neutrophils, both of which were present during infection. We found no differences in cell populations between wild type and Asc^−/− mice (Supplemental Fig 2e). Overall, these results indicate that the inflammasome promotes a robust proinflammatory response and clearance of S. anginosus in vivo.

Figure 6. S. anginosus elicits an inflammasome-dependent actue inflammatory response in vivo:

a) percent weight change of WT or Asc^−/− mice infected by IP injection with 2×10⁹ CFUs of S. anginosus for 48 hours. WT n=8, Asc^−/− n=9. 24 hour and 28 hour Asc^−/− stats are compared against WT at these time points. 48 hour WT stats are comparted against initial WT values b) CFUs/mL of S. anginosus recovered from supernatant of i.p. wash of WT or Asc^−/− mice 8 hours post infection. WT n=9, Asc^−/− n=8 c) RT-qPCR of Il1b, Tnfa, and Il6 from RNA taken from the spleen of WT or Asc^−/− mice 8 hours post infection. WT n=8, Asc^−/− n=7. d) RT-qPCR of Nos2, Il-1b, Tnfa, Il-6 and Arg-1 from RNA taken from cells collected from i.p. wash from WT or Asc^−/− mice 8 hours post infection. WT n=10–12, Asc^−/− n=11–12. e) densitometry of phospho-p65, iNOS, COX2 and cleaved caspase-1 from lysate of cells collected from i.p. wash from WT or Asc^−/− mice 8 hours post infection. WT n=6–9, Asc^−/− n=6–8 (Left). Representative western blot of phospho-p65, iNOS, COX2 and cleaved caspase-1 from lysate of cells collected from i.p. wash from WT or Asc^−/− mice 8 hours post infection (Right).

3.7. NLRP3 promotes an Inflammatory response to S. anginosus in vivo

Our results indicate that compared to wild type mice, Asc^−/− mice display reduced morbidity, bacterial clearance and inflammatory markers in response to S. anginosus infection in vivo, indicating that the inflammasome plays a role during the acute inflammatory response and control of S. anginosus in vivo (Fig. 6). Moreover, NLRP3 was found to be an important inflammasome sensor that responds to S. anginosus infection in vitro (Fig. 5). To verify if the NLRP3 inflammasome sensor is responsible for mediating an inflammatory response to S. anginosus in vivo, we infected wild type and Nlrp3^−/− mice with 2×10⁹ CFUs S. anginosus via i.p. injection. We also included Aim2^−/− mice as an additional inflammasome control. Compared to wild type mice, Nlrp3^−/− mice displayed significantly reduced weight loss at 24 hours post-infection, while Aim2^−/− mice were comparable to infected WT controls (Fig. 7a). Nlrp3^−/− mice also displayed a trending, but non-significant (p=0.055) increase in peritoneal CFUs (Fig. 7b), similar to our observations with Asc^−/− animals. We next measured inflammatory cytokine gene expression in the spleens of these mice and found a trending diminishment in proinflammatory Il1b, Tnfa and Il6 in Nlrp3^−/− mice, although this did not reach significance (Fig 7c). However, there was a significant decrease in Il1b and Tnfa expression in the recruited peritoneal cells as well as a trending decrease in Nos2 and Il6 (Fig 7d). To further assess proinflammatory signaling in the recruited peritoneal cells, we performed western blot on cells collected from the peritoneal wash and found a significant decrease in iNOS in Nlrp3^−/− derived cells (Fig 7e), but no measurable differences in phosho-p65, COX2 or cleaved caspase-1 at this time point (Fig7e). With the exception of COX2, which was significantly diminished, all other measured inflammatory markers were comparable in the recruited peritoneal cells isolated from wild type and Nlrp3^−/− mice at 16 hours post-infection (Supplemental Fig 3a–d). Overall, these results indicate that the NLRP3 inflammasome plays a partial, but incomplete, role during the acute inflammatory response and clearance of S. anginosus infection in vivo. This is in line with our in vitro observations, where we detected only partial diminishment of inflammasome activation in S. anginosus-infected Nlrp3^−/− BMMs and WT BMMs treated with the NLRP3 inhibitor, MCC950 (Fig 5f, g).

Figure 7. NLRP3 promotes an Inflammatory response to S. anginosus in vivo:

a) percent weight change of WT, Aim2^−/− or Nlrp3^−/− mice infected by IP injection with 2×10⁹ CFUs S. anginosus for 48 hours. WT n=8, Aim2^−/− n=7, Nlrp3^−/− n=8. 24 hour and 28 hour Nlrp3^−/− stats are compared against WT at these time points. 48 hour WT stats are comparted against initial WT values b) CFUs/mL of S. anginosus recovered from supernatant of i.p. wash of WT or Nlrp3^−/− mice 8 hours post infection. WT n=10, Nlrp3^−/− n=9 c) RT-qPCR of Il1b, Tnfa, and Il6 from RNA taken from the spleen of WT or Nlrp3^−/− mice 8 hours post infection. WT n=9–10, Nlrp3^−/− n=8–9. d) RT-qPCR of Nos2, Il-1b, Tnfa, Il-6 and Arg-1 from RNA taken from cells collected from i.p. wash from WT or Nlrp3^−/− mice 8 hours post infection. WT n=11–12, Nlrp3^−/− n= 10–12. e) densitometry of phospho-p65, iNOS, COX2 and cleaved caspase-1 from lysate of cells collected from i.p. wash from WT or Nlrp3^−/− mice 8 hours post infection. WT n=7–10, Nlrp3^−/− n=8–9 (Left). Representative western blot of phospho-p65, iNOS, COX2 and cleaved caspase-1 from lysate of cells collected from i.p. wash from WT or Nlrp3^−/− mice 8 hours post infection (Right).

4. Discussion

Although S. anginosus is a common commensal of the oral cavity, it’s abundance is reportedly increased during OSCC and it can directly impact the epithelium to drive tumor formation and progression [46], [47]. S. anginosus also promotes gastric cancer progression, which is associated with elevated inflammatory cell infiltration and the induction of the chemokines CCL20 and CXCR2 in vivo [16], yet how S. anginosus interacts with the innate immune system to drive the inflammatory responses during tumor initiation and progression remains unclear. Previously, we found that S. anginosus induces the inflammatory cytokines IL-1β, TNF⍺ and IL-6 in the RAW 264.7 macrophage cell line [36]. Additionally, we found that S. anginosus induces the expression of the cytosolic sensor NLRP3 and causes distinct morphological changes in RAW264.7 cells reminiscent of cellular activation [36]. While our previous work established a cellular link between S. anginosus and innate immune activation, the molecular mechanism behind this relationship remained to be investigated. In this study, we expanded on our previous findings using primary BMMs for greater physiological relevance and first confirmed that S. anginosus promotes a robust pro-inflammatory response in primary macrophages. This includes elevated NF-κB signaling, increased proinflammatory cytokine production (i.e., TNFα, IL-6 and IL-1β) and increased protein expression of inflammatory mediators COX2 and iNOS in S. anginosus-infected BMMs. We next focused our attention on understanding the mechanism by which S. anginosus drives macrophage activation, starting with the question: is bacterial viability required for the observed pro-inflammatory responses? We found that S. anginosus viability was dispensable for the early events of inflammatory signaling as heat-killed S. anginosus induced macrophage p65 activation equally as well as live S. anginosus following 24 hours of infection. However, heat-killed S. anginosus failed to induce the production of the inflammasome-associated cytokine IL-1β.

Because heat-killed S. anginosus induced early proinflammatory signaling, we investigated the role of Toll-like receptors (TLRs) including TLR2 and TLR4, which recognize other streptococcus species [48], [49]. TLRs are responsible for detecting extracellular ligands and are needed for early proinflammatory signaling following infection or cell damage [18]. TLR2 is traditionally associated with recognition of molecules found on the surface of gram-positive bacteria, while TLR4 recognizes components of gram-negative bacteria and a variety of damage-associated molecular patterns (DAMPs) [18]. We found that TLR2 and TLR4 alone are dispensable for inducing early proinflammatory signaling by live S. anginosus, as there were no differences in expression of early proinflammatory NF-κB signaling in BMMs lacking either TLR2 or TLR4 in comparison to wild type counterparts. However, TLR2 promoted TNFα production in response to live S. anginosus. Similar preferential roles for TLR2 have been identified during S. pneumonia-induced production of TNFα [50]. TLR2 and TLR4 alone were also dispensable for the ability of HK S. anginosus to elicit inflammatory signaling in BMMs. Thus, it is likely that multiple TLRs play both overlapping and distinct roles during the initial and downstream inflammatory responses to S. anginosus. This is corroborated by infection of BMMs lacking Myd88, a common adaptor protein for many TLRs, where we found a significant diminishment of early proinflammatory signaling by HK S. anginosus. Conversely, the induction of early NF-κB signaling may be due to detection of bacterial ligands by a number of other receptors expressed by macrophages. Other PRRs such as C-Type Lectin Receptors (CLRs) recognize other species of Streptococcus and induce a variety of downstream signaling pathways such as NF-κB and MAPK, leading to induction of cytokines including inflammasome-dependent IL-1β [51]. For instance, DC-SIGN and Mincle recognize capsular polysaccharide on Streptococcus pneumoniae and Streptococcus pyogenes, which facilitate bacterial phagocytosis, upregulate proinflammatory cytokines (e.g., TNFα and IL-6) and induce the expression of iNOS and ROS [52],[53]. Additionally, annexin A2 expressed by gastric epithelial cells binds to TMPC on the surface of S. anginosus, leading to MAPK signaling [16]. Although this relationship has not been explored in macrophages, annexin A2 is expressed by these cells, and it can promote phagocytosis of S. aureus and M. bovis [54]. In addition, soluble annexin A2 within human serum is increased during infection and can induce the activation of signaling pathways (e.g., MAPK and NF-κB) and upregulate proinflammatory cytokines such as TNFα, IL-1β and lL-6 within human monocyte derived macrophages in vitro [55], [56].

The requirement for S. anginosus viability to induce several inflammatory mediators lead us to explore the possibility that macrophages may fail to effectively neutralize live S. anginosus following phagocytosis. Indeed, when we infected BMMs with S. anginosus and assessed either CFUs or the presence of fluorescent internalized bacteria 24 hours later, we found greater amounts of bacteria than the amount that were internalized. This finding indicated that S. anginosus persisted within macrophages. However, it is unclear if S. anginosus completely evades macrophage-mediated killing or if the rate of S. anginosus replication within macrophages outpaced the rate of bacterial killing. Nonetheless, these findings indicated that macrophages fail to effectively neutralize S. anginosus. Many traditional pathogens often drive more robust and/or persistent inflammatory response through a variety of actions including intracellular replication. Indeed, other Streptococcus species, including S. gordonii, as well as other bacteria including S. aureus and S. typhimurium, can survive within macrophages, which causes increased secretion inflammasome-dependent cytokines such as IL-1β and exacerbates bacterial dissemination [30] [57], [58]. While the mechanism of infiltration and survival within macrophages of these other Streptococcus species types have been elucidated, how S. anginosus accomplishes this remains to be investigated

The innate immune system often relies upon intracellular PPRs, including the inflammasome-forming NOD-like receptor (NLR) proteins, to recognize and mount appropriate responses to intracellular pathogens. The inflammasome is a multi-protein complex consisting of a sensor (e.g., NLRP3 and AIM2), the adapter protein ASC and the enzyme Caspase-1 that, upon activation by a variety of microbial- and damage-associated ligands, leads to IL-1β production and potentially pyroptosis [17]. Because S. anginosus persisted within BMMs and induced the inflammasome-dependent cytokine IL-1β, we decided to investigate the requirement of the inflammasome during S. anginosus recognition and if S. anginosus induces pyroptosis. Indeed, we found that S. anginosus failed to induce IL-1β production from macrophages lacking the common inflammasome adapter protein, ASC, confirming the involvement of the inflammasome during this process. Further, we found that there was a significant diminishment of IL-1β production from BMMs lacking Caspase-1, which is needed for cleavage of active IL-1β. Additionally, we saw that macrophages infected with S. anginosus at the MOI used throughout this study (i.e., 10) released similar levels of LDH and displayed similar Gasdermin D cleavage as uninfected BMMs, indicating that S. anginosus infection at this MOI doesn’t lead to significant cell death. This finding is in line with our previous report that S. anginosus fails to induce cell death in the RAW264.7 macrophage cell line [36]. Multiple bacterial species have developed mechanisms to activate the inflammasome without inducing pyroptotic cell death including S. pneumoniae and S. aureus [59], [60]. Moreover, the intracellular pathogens S. typhimurium and L. monocytogenes can induce IL-1β production in the absence of pyroptosis [24], [31]. Our findings point towards a similar uncoupling of inflammasome activation and pyroptosis by S. anginosus. Interestingly, infection with S. anginosus at MOIs at 25 and greater caused significant LDH release, while IL-1β secretion remained similar across all MOIs. This indicates that lower levels of S. anginosus are sufficient to maintain proinflammatory cytokine production from BMMs without inducing cell death. However, it remains to be determined if this reflects an active mechanism by which S. anginosus inhibits macrophage cell death or if it fails to induce cellular perturbations that couple pyroptosis to IL-1β secretion like mitochondrial depolarization [25].

Because IL-1β production from S. anginosus infection was inflammasome dependent, we investigated if S. anginosus was detected by one or multiple inflammasome sensors linked to the recognition of other Streptococcus species including NLRP3, AIM2 and NLRC4. NLRP3 can recognize and be activated by a wide variety of PAMPs and DAMPs including extracellular ATP, bacterial toxins like nigericin, reactive oxygen species and changes in ion influx [61]. AIM2 recognizes cytosolic dsDNA from microbes and mammalian cells [62]. NLRC4 traditionally recognizes flagellum, although it has been linked to the recognition of some bacteria lacking flagella, as well as other DAMPs and PAMPs such as lysophosphatidylcholine and retrotransposon RNAs [35] [33] [63]. Here we found BMMs with genetic deletion or pharmacological inhibition of NLRP3 had a significant reduction in IL-1β secretion. This is in line with a role for NLRP3 during the recognition of many other streptococcus species, including S. pneumoniae, S. suis and S. pluranimalium [64] [65], [66]. Moreover, S. anginosus induced ERK signaling, which has been linked to priming the NLRP3 inflammasome [67]. Although IL-1β was significantly reduced in S. anginosus-infected NLRP3 deficient BMMs, these levels were not as drastic as what was observed in ASC deficient BMMs. Therefore, we cannot rule out the involvement of other and/or multiple inflammasome sensors. For example, NLRP6 recognizes S. pneumoniae [29], and multiple inflammasomes (AIM2, NLRP3 and NLRC4) are activated by S. mutans in human THP-1 macrophages [68]. However, we can conclude that AIM2 and NLRC4 do not contribute to S. anginosus-driven IL-1β since AIM2- and NLRC4-deficient macrophages presented similar levels of IL-1β as their wild type counterparts. As S. anginosus has no known flagella, it is not entirely surprising that the NLRC4 inflammasome was not involved in this process. Moreover, NLRC4 can also form an inflammasome independently of ASC, yet ASC was required for S. anginosus induced-IL-1β. However, it was somewhat surprising that AIM2 was dispensable for S. anginosus-induced inflammasome activation. This suggests that S. anginosus DNA likely does not access the cytosol of infected BMMs during the first 24 hours of infection and therefore may indicate that S. anginosus replicates in endosomal compartments while evading killing by macrophages. Additionally, our results suggest that S. anginosus likely triggers inflammasome activation through the generation of NLRP3-activating DAMPS and/or through virulence factors. Future studies will focus on characterizing the localization of S. anginosus within intracellular compartments of BMMs and identifying how S. anginosus triggers NLRP3.

After establishing a mechanistic role for NLRP3 inflammasome during S. anginosus infection in vitro, we next explored the inflammatory response to S. anginosus using an acute peritonitis infection model in vivo, similar to that used with other Streptococcus strains [45], [69]. We observed significant weight loss in wild type animals at 24 hours post-infection. By 48 hours post-infection, the wild type animals largely recovered but still displayed significant weight loss compared to the initial values observed prior to infection. In addition, we observed significantly increased levels of S. anginosus within the peritoneal cavity of Asc^−/− mice 8 hours post-infection, which was cleared by 24 hours. These findings indicate that S. anginosus induces an acute, but self-limiting proinflammatory response that clears the infection during a peritonitis infection model. We established an important role for the inflammasome during this inflammatory response as inflammasome-deficient Asc^−/− mice displayed significantly less weight loss at the 24 hour peak of disease and greater bacterial burdens in the peritoneal cavity 8 hours post-infection. Moreover, the spleens and recruited immune cells isolated from infected Asc^−/− mice displayed a decrease in proinflammatory cytokines and mediators, indicating that these mice have defective acute inflammatory responses, which may partially explain their diminished weight loss. This failure to mount an appropriate immune response likely results in the decreased bacterial clearance in the inflammasome-deficient mice. This finding is in line with a critical role for IL-1β during macrophage and neutrophil-mediated clearance of S. pneumonia in the nasopharynx [70]. Interestingly, we also found an increase in Arginase-1 expression in the recruited peritoneal cells recovered from Asc^−/− mice in comparison to wild type. Arginase-1 is typically associated with anti-inflammatory M2-like macrophages or other immunosuppressive cell types such as MDSCs [71]. This suggests that the inflammasome may also play an important role in limiting the immunosuppressive response against S. anginosus. S. pyogenes also promotes Arginase-1 during macrophage infection [72], however the involvement of the inflammasome during this response has not been explored. Together, this indicates that the inflammasome promotes an acute and limiting proinflammatory response during S. anginosus infection, which needed for rapid bacterial clearance.

To determine which inflammasome sensor was involved in this acute response to S. anginosus infection in vivo, we infected Aim2^−/− and Nlrp3^−/− mice and tracked disease progression over 48 hours. Nlrp3^−/− mice displayed a decrease in weight loss in comparison to wild type mice, similar to Asc^−/− mice. This indicates that the NLRP3 inflammasome plays a role in innate immune response against S. anginosus in vivo, similar to what we observed during macrophage infection in vitro. No mortality was observed during S. anginosus infection of wild type or NLRP3-deficient animals, which was in stark contrast to the in vivo response to S. pyogenes when delivered via the same infectious route [45]. Interestingly, while infected Nlrp3^−/− mice displayed similar morbidity as Asc^−/− mice, Aim2^−/− mice displayed more severe weight loss than wild type mice. Because Aim2^−/− mice presented a phenotypically distinct response to S. anginosus infection compared to Asc^−/− mice, these findings point towards an inflammasome-independent role for AIM2 during S. anginosus infection in vivo. This is in line with our findings that AIM2 was dispensable for S. anginosus-induced IL-1β production in macrophages. Indeed, AIM2 has inflammasome-distinct functions in non-immune cells (e.g., the epithelium) [39]. Conversely, Nlrp3^−/− mice presented less severe weight loss compared to wild type mice, similar to Asc^−/− mice. Moreover, the infected Nlrp3^−/− mice also presented a trending increase in peritoneal S. anginosus CFUs in comparison to wildtype, although this was not as pronounced as the levels seen in Asc^−/− mice. Consistent with this finding, there was a diminishment of proinflammatory Il1b and Tnfa expression along iNOS protein levels in the recruited peritoneal cells collected from Nlrp3^−/− mice in comparison to wild type controls. These findings indicate that NLRP3 plays a partial role in mounting the acute proinflammatory response driven by the inflammasome and clearing S. anginosus in vivo. This is not entirely surprising given NLRP3 was not fully responsible for complete inflammasome activation in response to S. anginosus in vitro.

Although there are multiple ways S. anginosus can expand and/or become dysbiotic in the oral cavity, a major question that remains is how and why this species is elevated during OSCC. In normal epithelium, oral streptococcus such as S. mitis, S. sanguinis and S. gordonii are pioneer colonizers, which form the structural architecture of the oral biofilm. This involves interactions with other streptococci species, promotion of microbial succession and fostering of diverse polymicrobial interactions in the oral cavity [73], [74]. These species restrict the colonization of other oral microbes through their production of alkali, bacteriocins, and hydrogen peroxide. Diet plays a large role in maintenance of early colonizing species and a high sugar diet along with poor oral hygiene can create an acidic environment that is less hospitable to these early commensal colonizers, which are not well-adapted to acidic conditions. This allows for the enrichment of acid-tolerant (aciduric) taxa such as S. mutans and favors cariogenic inflammatory environment leading to dysbiosis [75]. Additionally, the oral microbiome may become dysbiotic as a result of compromised immunity due to aging, as there is a positive correlation between S. anginous and age [76], much like OSCC. Other cancer-associated factors like smoking along with tissue injury or antibiotic use during infection can lead to dysbiosis of the oral microbiome as well [76]. While interspecies interactions of oral streptococci play an important role in limiting dysbiosis, the connection between S. anginosus and the immune system during localized and systemic inflammatory responses warrants more investigation, especially during the development and progression of cancer.

Our experiments largely explored the consequence of S. anginosus-macrophage interactions, but it is important to consider how these interactions would occur in healthy tissue versus oral pathologies like OSCC. While this is still an area of question for the oral cavity, S. anginosus infection of gastric tissue decreases expression of epithelial tight junction proteins claudin18, occluden and zonula occluden, leading to increased gastric epithelial permeability [16]. Additionally, S. anginosus increases expression of gastric chemokines CCL20 and CXCR2, which promote immune cell infiltration and increases both proinflammatory and immunosuppressive immune cell types, including Th17 CD4 T cells and tumor-associated macrophages, respectively [16]. S. anginosus is also associated with the recruitment of immune cells within human esophageal cancer as evidenced by an increase in chemo attractants IL-8 and GRO⍺ in esophageal tissue sections containing S. anginosus [77]. Overall, these findings highlight multiple mechanisms by which S. anginosus may promote interactions with immune cells like tissue-resident and recruited macrophages, but there is still much to learn about how S. anginosus interacts with different immune cells to impact proinflammatory responses. Here we show that S. anginosus replicates within macrophages, and its viability is essential to fully induce many downstream inflammatory pathways that require NF-κB, including COX2, iNOS, TNFα and inflammasome activation. The ability of S. anginosus to replicate intracellularly may be key to its observed sustained activation of the NF-κB pathway and inflammation within the epithelium and the tumor immune microenvironment. Additionally, intracellular persistence of S. anginosus may aid its dissemination during OSCC. Macrophages have been shown to act as a ‘Trojan horse’ for Group A streptococcus in tissue infections, where intracellular persistence of streptococcus within macrophages allows it to evade killing by antibiotics and causes prolonged infection and increased bacterial burden [78]. Our findings suggest S. anginosus may drive M1 differentiation of recruited and tumor associated macrophages (TAMs), which produce a variety of proinflammatory proteins and cytokines through the NF-κB pathway including COX2, iNOS, and TNFα. Increased COX2 and iNOS expression in TAMs may lead to increased VEGF and reactive oxygen species, respectively, which can exacerbate cancer by inducing angiogenesis and introducing new mutations through DNA oxidation [79]. TNFα is also increased in multiple cancer types and is associated with angiogenesis, tumor growth and DNA damage [80]. Both NF-κB and inflammasome activation are required for active IL-1β secretion, which is highly implicated during cancer [81]. Elevated IL-1β is present in a variety of cancers including OSCC [82], [83]. While acute production of inflammatory cytokines like IL-1β is essential to the clearance of many pathogens, prolonged IL-1β production and unchecked chronic inflammation is detrimental and may drive tumor progression through multiple mechanisms [42]. IL-1β may directly impact the epithelium to drive tumor growth and epithelial to mesenchymal transition through induction of MYC and downregulation of E-cadherin, respectively [84], [44]. Additionally, IL-1β can drive production of other pro-inflammatory molecules, which can cause cell death and release of DAMPs, which further contribute to disease progression [85]. Activation of the inflammasome for IL-1β secretion can occur by recognition of a variety of DAMPs and PAMPs. Here we found that S. anginosus is sensed by the NLRP3 inflammasome. NLRP3 and downstream IL-1β is largely implicated in several chronic inflammatory conditions and the development of OSCC [86]. Moreover, elevated NLRP3 in head and neck cancers including OSCC is associated with increased tumor invasion, survival and drug resistance [81]. However, the contribution of oral microbes like S. anginosus during NLRP3 overexpression and activation during cancer remains unclear. Dysbiosis and/or overgrowth of S. anginosus within the oral microbiome likely increases the presence of multiple damage- and pathogen-associated factors that trigger the NLRP3 inflammasome and could drive disease outcomes. As S. anginosus clearly impacts the TME in multiple ways, understanding its interactions with both epithelial cells and infiltrating immune cells will be key in understanding its impact on OSCC.

Data Availability:

Data will be made available upon reasonable request.