Membrane vesicles derived from Streptococcus suis serotype 2 induce cell pyroptosis in endothelial cells via the NLRP3/Caspase-1/GSDMD pathway

Keda Shi ,Yan Li ,Minsheng Xu 3,Kunli ZhangHongchao GouChunling Li2#,Shaolun Zhai2#

1 Institute of Animal Health,Guangdong Academy of Agricultural Sciences/Key Laboratory of Livestock Disease Prevention of Guangdong Province/Scientific Observation and Experiment Station of Veterinary Drugs and Diagnostic Techniques of Guangdong Province,Ministry of Agriculture and Rural Affairs,Guangzhou 510640,China

2 Maoming Branch,Guangdong Laboratory for Lingnan Modern Agricultural Science and Technology,Maoming 525000,China

3 College of Animal Science & Technology,Zhongkai University of Agriculture and Engineering,Guangzhou 510225,China

Abstract Streptococcus suis serotype 2 (S.suis 2) is a zoonotic pathogen that clinically causes severe swine and human infections (such as meningitis,endocarditis,and septicemia).In order to cause widespread diseases in different organs,S.suis 2 must colonize the host,break the blood barrier,and cause exaggerated inflammation.In the last few years,most studies have focused on a single virulence factor and its influences on the host.Membrane vesicles (MVs) can be actively secreted into the extracellular environment contributing to bacteria-host interactions.Gramnegative bacteria-derived outer membrane vesicles (OMVs) were recently shown to activate host Caspase-11-mediated non-canonical inflammasome pathway via deliverance of OMV-bound lipopolysaccharide (LPS),causing host cell pyroptosis.However,little is known about the effect of the MVs from S.suis 2 (Gram-positive bacteria without LPS) on cell pyroptosis.Thus,we investigated the molecular mechanism by which S.suis 2 MVs participate in endothelial cell pyroptosis.In this study,we used proteomics,electron scanning microscopy,fluorescence microscope,Western blotting,and bioassays,to investigate the MVs secreted by S.suis 2.First,we demonstrated that S.suis 2 secreted MVs with an average diameter of 72.04 nm,and 200 proteins in MVs were identified.Then,we showed that MVs were transported to cells via mainly dynamin-dependent endocytosis.The S.suis 2 MVs activated NLRP3/Caspase-1/GSDMD canonical inflammasome signaling pathway,resulting in cell pyroptosis,but it did not activate the Caspase-4/-5 pathway.More importantly,endothelial cells produce large amounts of reactiveoxygen species (ROS) and lost their mitochondrial membrane potential under induction by S.suis 2 MVs.The results in this study suggest for the first time that MVs from S.suis 2 were internalized by endothelial cells via mainly dynamin-dependent endocytosis and might promote NLRP3/Caspase-1/GSDMD pathway by mitochondrial damage,which produced mtDNA and ROS under induction,leading to the pyroptosis of endothelial cells.

Keywords: Streptococcus suis serotype 2,membrane vesicles,endocytosis,pyroptosis,NLRP3 inflammasomes,mitochondrial damage,endothelial cell

1.lntroduction

Streptococcus suisis a Gram-positive bacterium that causes meningitis,septicemia,endocarditis,and other infections in both swine and humans,which poses a severe threat to public health and the swine industry worldwide (Fenget al.2014;Goyette-Desjardinset al.2019).In particular,S.suisserotype 2 (S.suis2) is known as a zoonotic agent with the strongest virulence.It has been frequently isolated from infected pigs among the 35 serotypes which have been identified by capsular polysaccharide (CPS) antigens (Fittipaldiet al.2012).Several virulence factors ofS.suis2 are considered important for the pathogenesis ofS.suisinfection,among them contributing to colonize or invade,leading to severe systemic diseases in its host (Fittipaldiet al.2012;Fan 2017;Xiaet al.2019),such as CPS,Sly (suilysin),Mrp (muramidase-released protein),and Enolase (Eno).These virulence-associated factors are either secreted into the environment to damage host cells and stimulate innate and adaptive host immune responses or linked with the cell surface to promote colonization and proliferation (Fenget al.2009;Xuet al.2010;Wanget al.2015;Royet al.2018).However,the current studies of pathogenetic mechanisms ofS.suis2 are mostly based on a single virulence factor or gene.The complex mechanisms involved inS.suis2 infections and multiple organ injuries on host cells remain poorly characterized.

Membrane vesicles (MVs) can be actively secreted into the extracellular environment,about 20 to 400 nm in diameter,and enrich with surface-associated or extracellular virulent proteins,that contribute to both bacteria-bacteria and bacteria-host interactions (Kimet al.2015;Toyofukuet al.2019).MVs fromNeisseria gonorrhoeae,uropathogenicEscherichia coli,andPseudomonas aeruginosatransport toxins to mitochondria could disrupt organellar homeostasis and induce host macrophage apoptosis,and NLRP3 inflammasome activation (Deoet al.2020).MVs of enterohemorrhagicE.colicarry a cocktail of key virulence factors,including Shiga toxin 2a (Stx2a),cytolethal distending toxin V (CdtV),EHEC hemolysin,and flagellin,which are internalized by cellsviadynamindependent endocytosis (Bielaszewskaet al.2017).MVs fromS.suis2 are first isolated and the proteomic analysis of the MVs revealed that MVs contain 46 proteins,9 of which are considered as proven or suspected virulence factors (Haas and Grenier 2015).However,the factors carried by MVs fromS.suis2 and the complex interaction between MVs and host cells need to be better understood.

The NLRP3 (NOD-like receptor family,pyrin domain containing 3),known as multiprotein complex inflammasome,coordinates innate immune responses to various pathogenic or physiological stimuli (Elliott and Sutterwala 2015).The NLRP3 inflammasome activates the cells through the Caspase-1-dependent pyroptosis pathway in response to pathogen-associated molecular patterns (PAMPs) or danger-associated molecular patterns (DAMPs).NLRP3 oligomerizes and recruits ASC and pro-Caspase-1,which produce the activated Caspase-1 fragment and cause the maturation and release of proinflammatory cytokines,including IL-1β and IL-18.GSDMD-N,which is formed by inflammatory Caspase cleavage,mediates cell membrane pore formation and encourages the release of inflammatory factors,cell swelling,and pyroptosis (Jorgensen and Miao 2015;Jorgensenet al.2017).Lipoprotein,lipopolysaccharide (LPS),and some toxins from bacteria mediate cell pyroptosis (Kayagakiet al.2015;Wanget al.2020;Renet al.2021).Gram-negative bacteria secrete outer membrane vesicles that bind and deliver lipopolysaccharide (LPS) to the host cell cytosol,triggering Caspase-11 activation,which cleaves GSDMD,thereby mediating pyroptosis (Shiet al.2015;Vanajaet al.2016).Few studies have reported that MVs fromS.suis2 carried cargo protein and interacted with host cells.

In this study,we demonstrated that MVs fromS.suis2,carrying virulence factors,were transported to endothelial cells by cellsviamainly dynamin-dependent endocytosis,and triggered the activation of the NLRP3-mediated canonical inflammasome and induced cell pyroptosis.

2.Materials and methods

2.1.Bacterial strains and growth conditions

TheS.suis2 strain was isolated from the brain of a pig that suffered from meningitis (Institute of Animal Health,Guangdong Academy of Agricultural Sciences,China).TheS.suis2 strain was cultured on a blood agar plate (Huankai Microbial) at 37°C for 18 h and then single isolated colonies were inoculated into 5% bovine serum brain heart infusion broth (Huankai Microbial,China).Bacterial cultures were grown at 37°C to the exponential growth phase,in which optical density was measured as 0.4-0.6 at 600 nm (OD600).To increase the secretion of MVs,1.25 μg mL-1lysozyme (20,000 U mg-1,Sangon Biotech,China) was used for lysozyme treatment of the exponential growthS.suis2 culture for 3 h.

2.2.lsolation and quantification of MVs

TheS.suis2 strain in the logarithmic growth period was centrifuged for 10 min at 10,000×g at 4°C,and the supernatant was filtered with a 0.45-μm pore (Millipore,the United States).The filtrate was concentrated in a 100 kDa ultrafiltration cube (Millipore).The MVs in the concentrate were ultracentrifuged for 3 h at 150,000×g at 4°C by an ultracentrifuge (BECKMAN Optima L-100XP,the United States),and the precipitated MVs were resuspended in 50 mmol L-1HEPES (Hyclone,the United States).Then,MV preparations were obtained by density gradient centrifugation (BECKMAN Optima MAX-XP,the United States).The isolated MVs were transferred to the bottom layers of an ultra-clear centrifuge tube (BECKMAN Coulter,the United States) which was composed of different 1 mL OptiPrep/HEPES (Alere Technologies,United Kingdom) layers (50,40,30,and 20%,from the bottom-up).Gradient centrifugation was performed for 6 h at 268,000×g at 4°C in an MLS-50 rotor (BECKMAN,the United States).An equal volume of 1 mL solution or the visible band between the 20 and 50% gradients were collected.The MVs in the 20 and 30% OptiPrep/HEPES layers were used for the following assays.The concentration of MVs was quantified using the BCA Protein Assay Kit (Thermo Fisher Scientific,the United States) following the manufacturer’s protocol.An aliquot of 0.5 μmol L-13,3´-dioctadecyloxacarbocyanine perchlorate (Dio,Beyotime Biotechnology,China) was used to fluorescently label the MVs at 4°C overnight in order to investigate the transmembrane transport of the MVs.

2.3.SEM and TEM analysis of S.suis 2 and MVs

Streptococcus suis2 in the exponential phase (5×108CFU mL-1,OD600=0.8) after treatment with 1.25 μg mL-1lysozyme for 3 h was pelleted at 3,000×g for 5 min at 4°C.The supernatant was discarded and the precipitate was washed with PBS (phosphate buffer solution,pH 7.4) thrice for 5 min.Then,the bacterial precipitation was fixed in 2.5% glutaraldehyde for 1 h;dehydrated by a graded series of ethanol (30,50,70,90,and 100%) for 10 min;and centrifuged at 3,000×g for 5 min at 4°C after every step.Finally,theS.suis2 bacteria were resuspended in 100% ethanol.Before observation by a scanning electron microscope (SEM,Zeiss Sigma 300,Germany) operated at 10.0 kV,the bacteria in ethanol were placed into a silicon chip,and dried by 45°C warm air followed by nano carbon powder in a vacuum field.

Purified MVs ofS.suis2 (100 μL) were incorporated into glutaraldehyde at a 2.5% final concentration and fixed for 10 min.Then,after standing on nickel grids for 2 min,the items were washed twice in PBS.The grids were then dyed for 2 min with 3% (w/v) phosphotungstic acid,followed by two washes with PBS and one with distilled water.Finally,after absorption with filter paper,they were allowed to air dry for 10 min and observed under the transmission electron microscope (TEM,Hitachi H7650,Japan) operated at 80.0 kV.

2.4.Dynamic light scattering (DLS) analysis of S.suis 2 MVs

The size of theS.suis2 MVs was determined with DLS using a Malvern Zetasizer Nano (Malvern Instruments Ltd.,United Kingdom).Triplicate measurements were carried out using MVs samples diluted 1:3 in ddH2O.The size of the MVs was analyzed with the Zetasizer software.The average diameter of the MVs and the number (percent) were calculated by the cumulants.

2.5.Cells culture

Ea.hy926 (ATCC,CRL-2922,the United States) is a vascular endothelial cells line.Ea.hy926 cells were cultured in the complete culture medium,which consists chiefly of DMEM (Dulbecco’s Modified Eagle’s Medium,Gibco) supplemented with 10% heat-inactivated fetal bovine serum (FBS,Gibco,the United States) and 1% (v/v) penicillin-streptomycin (Hyclone,the United States).Cells were cultured at 37°C in a suitable incubator containing 5% CO2.

2.6.Levels of lL-1β mRNA in Ea.hy926 cells following treatment with S.suis 2 MVs

The MVs extracted fromS.suis2 were added to the cells of each group after hatching for 12 and 24 h,respectively.Subsequently,the total RNA of cells was extracted with a Cell Total RNA Extraction Kit (Axgen,the United States) following the manufacturer’s protocol.Total RNA was reverse transcribed to cDNA with RTSuperMix for qPCR (Vazyme,China) according to the manufacturer’s instructions.Quantitative real-time PCR (qRT-PCR),which was performed on Lightcycler 480 (Roche,Switzerland),was used with SYBR Green Master Mix (Vazyme),using 0.5 μmol L-1of specific primers (Sangon Biotech,China) and 20 ng of cDNA.The thermocycling cycling conditions were according to the manufacturer’s instructions.Human primers were purchased from Sangon Biotech: IL-1β forward,5´-ATGCACCTGTACGATCACTG-3´;reverse,5´-ACAAAGGACATGGAGAACACC-3´;GAPDH forward,5´-GCACCGTCAAGGCTGAGAAC-3´;reverse,5´-TGGTGAAGACGCCAGTGGA-3´.

2.7.MVs uptake and microscopy fluorescence staining

To determine how the MVs uptake by the cells and the kinetics of the transport,Ea.hy 926 cells were seeded in 96-well culture plates with a concentration of 5×103cells/well in complete culture medium,3 multiple holes in each group,and placed in 37°C,5% CO2cells incubator for 24 h.The Dio-labeled MVs were added to the medium and incubated for 4,8,12,24,48 and 72 h.Then the cells were fixed with 4% paraformaldehyde (Aladdin,China) for 15 min at room temperature and washed with 10 mmol L-1PBS three times.The cells were pretreated (overnight,37°C) with phalloidin-FITC and then washed with PBS three times (Beyotime Biotechnology,China).The cellular nuclei were stained by 0.1 μg mL-1DAPI (Thermo Fisher Scientific,the United States) in assay buffer for 20 min.Preparations were mounted on the stage and analyzed with an inverted fluorescence microscope (EVOS FL Auto,Thermo Fisher Scientific,the United States).

2.8.Proteomic analysis of MVs from S.suis 2

A total of 100 μg MVs were separated by 10% SDSPAGE and stained by coomassie blue,then cut the gel as 1-2 mm gel particles.The particles were decolourized using decolorizing solution (50 mmol L-1NH4HCO3: acetonitrile=1:1,v/v) and dried.Add 10 mmol L-1DTT (dithiothreitol) (1 mol L-1DTT:25 mmol L-1NH4HCO3=1:100) until the liquid covers the gel,then put in the water bath at 56°C for 1 h and add 55 mmol L-1IAM (iodoacetamide) (0.55 mol L-1IAM:25 mmol L-1NH4HCO3=1:10) for 45 min to block the DTT.The gel particles were carried out overnight at 37°C by adding 10 ng μL-1of trypsin.The peptide samples were separated and then were taken for freeze-dried.The dried peptide samples were reconstituted with mobile phase A (2% CAN (acetonitrile),0.1% FA (formic acid)).Separation was performed by Ultimate 3000 UHPLC (Thermo Fisher Scientific,the United States) with a selfpacked C18 column.The peptides separated by liquidphase chromatography were ionized by a nanoESI source and then passed to a tandem mass spectrometer Q-Exactive HF X (Thermo Fisher Scientific,the United States) for DDA (Data Dependent Acquisition) mode detection.The peptide identification process starts by converting raw MS data into a peak list and searching for matches in the UniProt protein database.

We carried out a GO (Gene Ontology,http://geneontol ogy.org/) functional annotation analysis for all identified proteins,which were matches in the UniProt protein database.For the GO entries involved in the three ontologies (cellular component,biological process,molecular function),the IDs and the number of all the corresponding proteins are listed,a statistical chart is made,and the GO entries without the corresponding proteins were excluded.

2.9.Expression of S.suis 2 enolase (Eno)

The total gene synthesis of theenogene was performed in the Beijing Tsingke Biotech Co.,Ltd.After total gene synthesis and direct sequencing,theenogene was inserted into pET28(+)via NcoI/BamHI and using the HiFi Cloning Kit (Tsingke Biotechnology,China),generating pET28-eno.Escherichia coliBL21 transformed with pET28-enowas induced at 16°C for 16 h with the addition of 0.5 mmol L-1isopropyl beta-D-1-thiogalactopyranoside (IPTG).After sonication,bacterial lysate was subjected to centrifugation for the removal of the insoluble pellets.The acquired supernatant was filtered through a 0.22 μm-pore filter (Millipore,the United States) and applied on a Histag Protein Purification Kit (BEAVER,China) following the manufacturer’s protocol.The recombinant Eno protein was eluted with elution buffer containing 100 mmol L-1imidazole and further purifiedviaa 10 kd ultrafiltration tube (Millipore,the United States) and the Eno protein solution was replaced with TBS buffer (25 mmol L-1Tris-HCl,pH 7.4) and then quantified using the BCA Protein Assay Kit (Thermo Fisher Scientific,the United States).

2.10.Western blot analysis

Whole-cell proteins were collected using RIPA (Thermo Fisher Scientific,the United States) buffer containing protease inhibitors.All protein lysates were centrifuged at 14,000×g,4°C for 15 min.Then supernatants were isolated by chloroform/methanol extraction method and measured with a BCA Kit (Thermo Fisher Scientific,the United States).Proteins of culture medium supernatant were extracted by chloroform-methanol method and redissolved with PBS.All protein samples were mixed with loading buffer and denatured in boiling water for 5 min.All proteins were electrophoresed through 10% SDS-PAGE at the constant voltage of 100 V and then transferred to 0.45-μm pore-sized PVDF membranes (Millipore,the United States) at the constant current of 250 mA for 1.5 h.The membranes were blocked for 1 h with 5% non-fat milk in TBS-Tween 20 buffer (200 mmol L-1glycine,25 mmol L-1Tris,0.1% Tween 20).And then incubated with the primary antibodies at 4°C overnight,and secondary antibody incubation was performed at room temperature for 1 h.The protein band signal was detected using an ECL kit (Millipore,the United States) by a chemiluminescence gel imaging system (Tano,China).The primary antibodies: anti-cleaved Gasdermin D (#36425),anti-Gasdermin D (#97558),anticleaved Caspase-1 (#4199),anti-IL-1β (#12703),anticleaved IL-1β (#81386),anti-Caspase-4 (#4450),anti-Caspase-5 (#46680),anti-NLRP3 (#15101),anti-GAPDH (#5174),obtained from cell signaling technology;anti-Eno polyclonal antibody as a generous gift from Liancheng Lei,Jilin University,China.

2.11.Detection of LDH-released levels

Cells were treated either with MVs for 6,9,12 and 24 h or withS.suis2 for 1,2,3,4,6 and 9 h.Culture medium supernatant was collected and centrifuged for 10 min at 12,000×g and 4°C.Aloquots (100 μL) of supernatant were tested for lactate dehydrogenase (LDH) with the LDH Cytotoxicity Assay Kit (Beyotime Biotechnology,China) following the manufacturer’s instructions.

2.12.Mitochondrial DNA level in Ea.hy926 cells following S.suis 2 MVs treatment

Cells were treated byS.suis2 or MVs for 6 and 12 h.The cells were dispersed in Lysis buffer (10 mmol L-1Tris-HCl (pH 8.0),100 mmol L-1NaCl,1 mmol L-1EDTA,and 1% SDS).Cells were frozen and thawed repeatedly 3 times at -80°C to disrupt the cell membrane.Then the supernatant was centrifuged for 5 min at 1,000×g in a new tube.The supernatant was centrifuged for 10 min at 12,000×g,and the mitochondria were in the precipitation.Then,the total mitochondrial DNA was extracted by DNA extraction kit following the manufacturer’s instructions (Omega Bio-Tek,the United States).Human primers were purchased from Sangon Biotech: Mito-ND4 forward,5´-TCACCCACCACATTAACAAC-3´;reverse,5´-AAAACCCGGTAATGATGTCG-3´;Mito-16S rRNA forward,5´-CTTTAATTCAACATCGAGG-3´,reverse,5´-CTAAGGGATAACAGCGTA-3´;Tert forward,5´-GGACCCTGTGCTAAAGACCT-3´,reverse,5´-ACACGCCCACAAATAGCC-3´.

2.13.Detection of mitochondrial membrane potential marker and ROS levels

Ea.hy926 cells were treated withS.suis2 or MVs for 6 and 12 h.The mitochondrial membrane potential fluorescence level was stained by the MitoTracker Red CMXRos Assay Kit per the manufacturer’s instructions (Beyotime Biotechnology,China),and the nucleus was stained with the Hoechst reagent.Cells were treated either with MVs for 1,3,6,and 12 h or withS.suis2 for 1,2,4,6,9,and 12 h,with untreated cells as the control.The fluorescence level of reactive oxygen species (ROS) was stained by the ROS Assay Kit (Beyotime Biotechnology,China) following the manufacturer’s instructions.All samples were observed under a fluorescence microscope.The mean fluorescence intensity (MFI) of the ROS level in cells was calculated with ImageJ.

2.14.Statistical analysis

Analysis of the experimental data was performed in triplicate independent runs.The numeric data are expressed as means and standard deviations (SD),which were calculated using Prism 8 GraphPad software (USA).One-way ANOVA was used to assess the significant differences between multiple groups.TheP-values less than 0.05 were considered to reflect significant differences.

3.Results

3.1.lsolation and characterization of S.suis 2 secreted MVs

To examine whetherS.suis2 secretes membranous vesicles,we first observed the morphology ofS.suis2 by scanning electron microscopy.Intriguingly,we found putative MVs protrusions on the surface ofS.suis2 on a small number of bacteria (Fig.1-A).However,when the bacterial cell wall was destroyed by treatment with 750 μg mL-1Lysm for 6 h,this led to an explosive increase in the putative MVs protrusions on theS.suis2 bacterial surface (Fig.1-B).Then,we further isolated the MVs using a series of steps to physically separate the MVs from cellular cultures ofS.suis2.Bacteria and debris were removed by centrifugation and subsequent filtration through a 0.22 μm filter.The supernatant was then concentrated into small volumes with 100 kDa ultrafiltration cubes.The concentrated supernatant was then spun with an ultracentrifuge and density gradient centrifugation to pellet the MVs,leaving the soluble proteins in the supernatant.The isolated MVs were then observed by TEM (Fig.1-C) and DLS analysis (Fig.1-D).The results indicated that the isolated MVs fromS.suis2 had the characteristic lipid bilayer structure,and the average diameter was 72.04 nm.The MVs ranged in size from 50 to 300 nm.The SDS-PAGE profiles bands of MVs (5,10,20 and 30%,density gradient layers) andS.suis2 membrane proteins shared significant similarities.The Eno protein expression was seen in the 10,20 and 30% density gradient layers of MVs-isolated (10,20 and 30% density gradient layers obtained the highest level) andS.suis2 cytoplasm membrane (Fig.1-E).Abundant MVs existed at the 10 and 20% density gradient layers,which were used for the following experiments.

Fig.1 Characterization of Streptococcus suis serotype 2 (S.suis 2) secreted membrane vesicles (MVs).A and B,scanning electron microscope images of S.suis 2.A,untreated bacteria.B,S.suis 2 treated with 750 μg mL-1 Lysm for 6 h.The white bar is 250 nm.C,MVs isolated from cultures inoculated with S.suis 2 and visualized by TEM.The white bar is 100 nm.D,DLS analysis of S.suis 2 MVs.The y-axis displays the number (%) of MVs particles,and the x-axis displays the MVs size distribution (Z-Average).E,Coomasie-brilliant-blue stained SDS-PAGE showing the profiles of MVs on density gradient centrifugation and S.suis 2 cytoplasm,cellular supernatant,membrane proteins.The Eno protein expression level in the MVs density gradient and S.suis 2 cytoplasm,cellular supernatant and membrane.Gray level analysis summary of Eno protein Western blotting results.

3.2.Proteomic identification and GO analysis of MVs component from S.suis 2

After identification by the search engine (UniProt protein database,www.uniprot.org),557 spectra were matched,and 200 proteins and 474 peptides were identified in theS.suis2-MVs (Appendix A).Gene Ontology (GO) was used to analyze the functional properties of the identified proteins from the MVs.GO includes a total of three ontologies describing the biological process (BP),cellular component (CC),and molecular function (MF) of proteins.The identified proteins from MVs were mainly involved in the cellular and metabolic processes of the biological process and in catalytic activity and binding of the molecular function (Fig.2-A).Cellular compartments for the MV contents mainly included the cytoplasm and membrane.Furthermore,we checked the protein locations and obtained similar results (Fig.2-B).In addition,these identified proteins ofS.suis2 MVs included five bacterial virulence factors,which have been reported by other studies (Fig.2-C).Among them,Enolase is the most abundant protein with the highest score (i.e.,the amount of total unique peptide evidence related to a given protein) and the highest iBAQ (where the total peak area in each protein group is divided by the number of theoretical peptides).S.suis2 enolase has been shown to initiate host-cell apoptosis and increase blood-brain barrier permeability facilitating bacterial invasion.More importantly,enolase can stimulate humoral immune responses and produce antibodies.Similarly,several other virulence factors,e.g.,Lipoprotein,6-phosphogluconate dehydrogenase (6-PDG),NADPdependent glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and glutamine synthetase I alpha (Gln A),were also identified inS.suis2 MVs by proteomics.

Fig.2 Identification of the protein contents of Streptococcus suis serotype 2 (S.suis 2) membrane vesicles (MVs).A,GO enrichment analysis for the identified proteins from MVs by biological process (BP),cellular component (CC),and molecular function (MF).B,location of the identified proteins of MVs in S.suis 2.C,virulence proteins contained in the MVs.

3.3.Streptococcus suis 2 MVs were transported by vascular endothelial cells via mainly dynamin-dependent endocytosis

Dio is DiOC18(3) or called 3´,3´-dioctadecyloxacarbonine perchlorate,as a lipophilic stain labeled withS.suis2 MVs.S.suis2 MVs-Dio were taken up by endothelial cells in a time-dependent manner.Cellular MVs-Dio fluorescence intensity increased with time,peaking at 48 h (Fig.3-B and C).Streptococcus suis2 MVs uptake was significantly reduced by dynasore,an inhibitor of dynamin (Maciaet al.2006),in a dose-dependent manner (P＜0.001).Chlorpromazine (CPZ),an inhibitor of clathrinmediated endocytosis (Wanget al.1993),also could cause the dose-dependent reduction ofS.suis2 MVs uptake.Amiloride,an inhibitor of micropinocytosis (Wadiaet al.2004),had no effects on MVs uptake (Fig.3-A and B).In this part,the results demonstrated thatS.suis2 MVs are internalized by endothelial cellsviapartially clathrin-mediated and mainly dynamin-dependent endocytosis,but not by micropinocytosis.

Fig.3 Streptococcus suis serotype 2 (S.suis 2) membrane vesicles (MVs) were internalized in a time-dependent manner by vascular endothelial cells via mainly dynamin-dependent endocytosis.A,20 μg mL-1 S.suis 2 MVs-Dio (where S.suis 2 MVs had been labelled with the lipophilic stain,Dio (green)) were added to the vascular endothelial cells for 24 h,which were pretreated with the indicated endocytosis inhibitors.All cells were fixed in paraformaldehyde and then dyed with DAPI (blue) and F-actin (red).Scale bars are 100 μm.B,time-dependent MVs uptake study in endothelial cells.The fluorescence values of cells incubated with the indicated inhibitors and MVs-Dio were normalized to the fluorescence of Dio to cells (net fluorescence intensity).＊＊＊P＜0.001 and ns means no significant difference compared to inhibitor-untreated cells (one-way ANOVA).C,brightfield and fluorescence of vascular endothelial cells treated with S.suis 2 MVs-Dio for 48 h.Data are expressed as the mean±SD.

3.4.Streptococcus suis 2 MVs stimulated the canonical NLRP3/Caspase-1/GSDMD pathway and caused pyroptosis in vascular endothelial cells

To investigate whetherS.suis2 and MVs induced endothelial cell pyroptosis,the result of Western blotting analysis (Fig.4-A and B) showed that bothS.suis2 andS.suis2-MVs could strongly stimulate the NLRP3 receptor.As the treatment concentration (20,40,or 80 μg mL-1) ofS.suis2 MVs in the vascular endothelial cells increased for 24 h,the NLRP3 protein grey level increased markedly.Similarly,S.suis2 could also cause NLRP3 receptor activation after incubation for 3,6,and 9 h.After inflammasome activation,triggering the activation of Caspase-1,the presence of cleaved Caspase-1 (p20) also increased obviously.The mature cleaved IL-1β,as the product of pyrolysis from pro-IL-1β cleaved by Caspase-1,could be detected in both culture supernatants (Sup) and cellular lysates (Lys) after treatment withS.suis2 orS.suis2-MVs at different concentrations or for different lengths of time.GSDMD plays an important role in pyroptosis.GSDMD-N (cleaved GSDMD),as the predominant active fragment,was significantly increased in the vascular endothelial cells after stimulation by theS.suis2 orS.suis2-MVs.The total GSDMD showed no evident change.However,[K+] efflux is a necessary upstream event in NLRP3 activation and cell pyroptosis happens (Kelleyet al.2019).Therefore,we detected whether theS.suis2 and MVs could induce endothelial cell pyroptosis,which depended on [K+] efflux.As the treatment of 80 μg mL-1MVs andS.suis2 (MOI=10),the specified [K+] was added to the culture medium.The endothelial cells were primed with 500 ng mL-1LPS for 3 h and followed by a 0.5-h dose of 10 μmol L-1nigericin therapy served as the positive control to stimulate the classical NLRP3-inflammasome and induce cell pyroptosis (Karmakaret al.2020).The result showed that the high extracellular concentration of [K+] inhibited NLRP3 receptor,cleaved Caspase-1 and GSDMD-N protein level which was induced byS.suis2 and secreted-MVs.Especially,Eno as the most abundant virulence protein in the MVs also could activate the NLRP3 receptor (Fig.4-C).However,the activation effect from the MVs and Eno was reduced by the ENOBlock which is the nonsubstrate analogue that directly binds to Eno and inhibits its activity (Junget al.2013).Moreover,we further found that 80 μg mL-1S.suis2-MVs significantly increased the cellular supernatant LDH level after treatment for more than 6 h (based on the fold change relative to mock).Similarly,the LDH level was also significantly increased byS.suis2 (MOI=10) treatment for more than 4 h (Fig.4-E).Cellular IL-1β mRNA expression was examined by qRTPCR (Fig.4-F).S.suis2 andS.suis2-MVs stimulated the IL-1β expression.The rate of cell death after a bacterial infection orS.suis2-MVs treatment (Fig.4-D) showed that after bacterial infections for 12 h,96.7% of the cells had died,and 34.6% of the cells died when cells were exposed to 80 μg mL-1MVs for an additional 12 h.In the bright field,we could observe that the cell body was like an over-inflated balloon and indicated with yellow arrows in Fig.4-G when pyroptosis occurred.This appearance was similar to the descriptions in other reports (Chenet al.2016).

Fig.4 Membrane vesicles (MVs) from Streptococcus suis serotype 2 (S.suis 2) stimulated the canonical NLRP3/Caspase-1/GSDMD pathway and caused the pyroptosis of endothelial cells.A and B,Ea.hy 926 cells were treated with S.suis 2 (MOI=10) for 3,6,and 9 h or MVs from S.suis 2 (20,40 or 80 μg mL-1) for 24 h.The specified concentration of [K+] was added to the culture medium at the same time.NLRP3,Caspase-1,Cleave Caspase-1,GSDMD,GSDMD-N (cleaved GSDMD),GAPDH,and cleaved IL-1β (Lys) proteins expressions in cell lysates were determined by Western blotting.Mature secreted IL-1β was determined in supernatants (Sup).A total of 500 ng mL-1 LPS and 10 μmol L-1 nigerian treatment served as the positive control.C,NLRP3 proteins expressions level of endothelial cells treated with the 0.5 μmol L-1 Eno or 80 μg mL-1 MVs,which had been incubated with 0.5 μmol L-1 ENOBlock for 1 h.D,endothelial cells were stimulated with S.suis 2 (MOI=10) for or 80 μg mL-1 MVs for 48 h and cell death was evaluated by the CCK8 assay.E,LDH release levels were determined in culture medium supernatants.F,IL-1β mRNA levels in Ea.hy 926 cells with indicated times of S.suis 2 (MOI=10) or MVs (80 μg mL-1).G,cell pyroptosis-like morphological changes were observed (indicated with yellow arrows,scale bar=25 μm).Data are expressed as the mean±SD.＊P＜0.05,＊＊P＜0.01,and ＊＊＊P＜0.001 indicate significant differences compared to the mock group.

3.5.Streptococcus suis 2 MVs caused pyroptosis of endothelial cells by activating NLRP3 classical pathways rather than non-classical pathways

For clarity,we determined whether the endothelial cell pyroptosis caused by the MVs fromS.suis2 was dependent on the NLRP3 classical pathway.SiNLRP3,as the siRNA of NLRP3,reduced the expression of NLRP3.The siNLRP3 was pre-transferred using the transfection reagent to endothelial cells for 8 h,and 80 μg mL-1MVs was added.The NLRP3,GSDMD,and GSDMD-N proteins levels were determined by Western blotting analysis.The results showed that the NLRP3 expression was successfully reduced by siRNA when the MVs stimulated it.However,GSDMD-N,the pyroptosis performer,also decreased significantly (Fig.5-A).Additionally,after NLRP3 was silenced,the cell death induced byS.suis2 infections or MVs treatment was significantly reduced (Fig.5-B).Similarly,the same results were obtained by the inhibitor of MVs transport.NLRP3 was silenced by the inhibitor when theS.suis2 or MVs stimulated the endothelial cells (Fig.5-D).Cell death caused by eitherS.suis2 infections or MVs treatment was significantly reduced after CPZ or dynasore treatment for 1 h (Fig.5-E).In addition,we determined whether the MVs could activate Caspase-4/-5 non-classical pathways to induce the pyroptosis of endothelial cells.We detected Caspase-4/-5 protein expression afterS.suis2 and MVs treatments,using the 500 ng mL-1LPS treatment as the positive control.The results show that Caspase-4/-5 protein levels were not significantly changed after theS.suis2 or MVs treatment (Fig.5-C).

Fig.5 Streptococcus suis serotype 2 (S.suis 2) membrane vesicles (MVs) induced pyroptosis through the classical NLRP3/Caspase-1 pathway in endothelial cells,but not the Caspase-4/-5 pathway.A,the protein levels of NLRP3 and GSDMD were detected by WB analysis after being transfected with or without siNLRP3 and treated with 80 μg mL-1 MVs or mock for 24 h.B,endothelial cells were stimulated with S.suis 2 (MOI=10) for 3 or 6 h,or with 80 μg mL-1 MVs for 24 h after transfection with NLRP3 siRNA for 24 h.C,WB analysis of Caspase-4/-5 from S.suis 2 (MOI=10) incubated for 6 or 9 h or treated with 20 or 80 μg mL-1 MVs for 24 h in endothelial cells.D,changes in NLRP3 levels after pretreatment with the indicated endocytosis inhibitors.E,endothelial cells were stimulated with S.suis 2 (MOI=10) for 3 h or 80 μg mL-1 MVs for 24 h after their treatment with the inhibitor for 1 h.Cell death was evaluated by the CCK8 assay.Data are expressed as the mean±SD.＊P＜0.05,＊＊P＜0.01,and ＊＊＊P＜0.001 indicate significant differences compared to the siNLRP3-untreated or inhibitor-untreated cells.

3.6.Mitochondrial depolarization and cellular ROS increased in endothelial cells induced by S.suis 2 MVs

We next sought to elucidate the molecular mechanisms responsible for uncoupling NLRP3/Caspase-1-dependent inflammasome,triggering in endothelial cells induced by MVs fromS.suis2.The ROS level of the endothelial cells increased gradually and strongly after MVs orS.suis2 infections,and the maximum was reached at 6 h after stimulation (Fig.6-A).Interestingly,we then asked whether the ROS came from mitochondrial homeostasis disruption or dysfunction.The mitochondrial membrane potential was detected by Mito-Tracker red probes (Fig.6-B) and mtDNA (16S rRNA and ND4) expression in qPCR (Fig.6-C).These data demonstrate that mitochondria injury induced by MVs fromS.suis2 led to ROS production and mtDNA leakage into the cytoplasm.

The shoemaker s piety14 is stressed again and again and signals that he is deserving of the reward given to him and also protected against the pagan spirits who help him out by discharging his chores (Tatar, Annotated15 Grimms, 184).

Fig.6 Cellular reactive oxygen species (ROS) increase and mitochondrial membrane potential loss in endothelial cells were induced by Streptococcus suis serotype 2 (S.suis 2) and membrane vesicles (MVs).A,quantitative fluorescence of ROS from endothelial cells treated with MVs for 1,3,6,and 12 h,or with S.suis 2 for 1,2,4,6,9 and 12 h,with untreated as the control.MFI,mean fluorescence intensity.ROS fluorescence images at 6 h are shown.Scale bar=200 μm.B,fluorescent images of the indicated endothelial cells stained with Hoechst (blue) and mito-tracker (red,mitochondria).Scale bar=200 μm.C,qPCR of mtDNA (16S rRNA and ND4) from mitochondrial fractions quantified relative to total nuclear DNA (Tert).Data are expressed as the mean±SD (n=3).

4.Discussion

Almost all Gram-negative and Gram-positive bacteria release nano-sized MVs into the extracellular environment.These bacterial MVs are enriched with lipids,nucleic acids,bioactive proteins,and virulence factors (Brownet al.2015;Dhitalet al.2021).Gram-negative bacteria have been known to produce extracellular vesicles for nearly six decades,but Gram-positive bacteria have only recently been shown to naturally produce MVs in the extracellular milieu through proteomic analyses and transmission electron microscopic observations (Knoxet al.1966;Leeet al.2009).In previous study,Haaset al.(2015) first isolated MVs fromS.suis2 and performed proteomic analysis of the MVs and revealed 9 virulence factors (virulence factors: Mrp and Eno also were identified in our study).However,different methods of purification,bacterial culture medium and stress conditions,possibly cause the different protein types in MVs which are in proteomic analysis (Honget al.2019;Klimentovaet al.2019).In addition,MVs-isolated in the previous study did not include standard density-gradient ultracentrifugation which is necessary to distinguish the contents of MVs from other factors (soluble secreted proteins,flagella,pili etc.) that pellet by simple ultracentrifugation (Klimentová and Stulík 2015;Leeet al.2016).In this study,we isolated and characterized MVs from pathogenicS.suis2 by the standard density-gradient ultracentrifugation and analyzed the purity of MVs.The MVs could be visible on the surface ofS.suis2,constantly secreting MVs into the extracellular environment.This phenomenon was even more pronounced when lysozyme was used.This result means that MVs production would increase dramatically when the cell wall was broken down,like other reports (Toyofukuet al.2014;Toyofukuet al.2017).The diameters of the MVs ranged from 20 to 400 nm,and the average diameter was 72.04 nm.Moreover,we found that other serotypes ofS.suis2,such as 9 and 14,also produced MVs with the same characteristics (data not shown).To investigate whether the shedding of MVs was a common phenomenon inS.suis2,a proteomic analysis was performed for the MVs,which aimed to demonstrate that the pathogenicity and functions ofS.suis2 primarily act through the secretion of MVs.A total of 200 proteins were identified inS.suis2-released MVs,most of which came from the cell cytoplasm and membrane.These identified proteins included several well-known toxin proteins ofS.suis2,such as Enolase,Membrane-anchored lipoprotein,6-PDG,and Gln A.These toxin proteins are involved in bacterial invasion,adhesion,and colonization (Brassard 2004;Fittipaldiet al.2012;Xiaet al.2019).Remarkably,the abundance and score of Eno were much higher than those of the other toxin proteins,illustrating the potential contribution of Eno in MVs in mediating pathogenesis in the host.Eno could involve bacterial adhesion,induce hostendothelial cells apoptosis and increase blood-brain barrier permeability for facilitating bacterial invasion (Sunet al.2016;Liuet al.2021).

In this study,we explored the interactions of endothelial cells andS.suis2-secreted MVs,including cellular uptake and innate immunity mechanisms.TheS.suis2-secreted MVs were transported by the endothelial cellsviamainly dynamin-dependent and partially clathrinmediated endocytosis.This is similar to the MVs from EnterohemorrhagicE.coliandStaphylococcus aureus(Bielaszewskaet al.2017;Wanget al.2020),but differs from some other bacteria,such asPseudomonas aeruginosaandAggregatibacter actinomycetemcomitans,where internalization mostly depends on caveolin and cholesterol-rich lipid rafts (Bombergeret al.2009;Thayet al.2014).Notably,the MVs fromS.suis2 are internalized by target cells in a time-dependent manner,becoming stable in 48 h,but this system shows the fastest transport rate within 12 h.This result suggests that cells may be able to respond to MVs stimulation in a short time.

Gram-negative bacteria secrete outer membrane vesicles that bind and deliver LPS to the host cell cytosol,triggering Caspase-11 activation,which directly cleaves gasdermin D,thereby mediating pyroptosis and indirectly activating the NLRP3 inflammasome (Vanajaet al.2016).In contrast,Gram-positive bacteria lack LPS endotoxin,so the secreted-MVs do not bind LPS.The absence of export cues in cytoplasmic proteins makes the MVs a potential secretory pathway for Gram-positive bacteria (Deatherage and Cookson 2012;Wanget al.2020).Thus,the MVs fromS.suis2 may have different interaction mechanisms with cells than the Gram-positivereleased MVs which do not bind LPS.Our study showed that the cargo ofS.suis2 MVs includes cell wall-anchored proteins and membrane proteins,but cytoplasmic proteins represent the most abundant component of the MVs.Furthermore,we demonstrate thatS.suis2 secreted MVs could activate the NLRP3 inflammasome,a major signaling pathway of the innate immune system and initiator of cell pyroptosis (Kelleyet al.2019;Huanget al.2021).Our data indicate that pyroptosis was induced upon the incubation of endothelial cells withS.suis2 MVs.The endothelial cells were inflated like balloons and died,which indicated the occurrence of cell pyroptosis.MV-induced NLRP3 inflammasome activation occurred in a concentration-dependent manner.In the current study,the NLRP3 inflammasome activates the Caspase-1 precursor and mediates GSDMD cleavage and the maturation of IL-1β and IL-18 during pyroptosis (Heet al.2015;Wanget al.2021).Our data demonstrated that the Caspase-1 precursor was activated,GSDMD cleavage was mediated,and IL-1 matured.The LDH and cleaved IL-1 were released into the cell supernatant.However,the knockdown of NLRP3 did not fully abolish the either the MVs-induced GSDMD cleavage or cell death,but it was effective in early infections.The apoptosis-programmed cell death pathway could also be induced byS.suisEno and occur duringS.suis2 infections (Liuet al.2021;Wuet al.2022).The MVs-induced cytotoxicity of endothelial cells caused by MVs likely involves multiple cell death pathways.Interestingly,we found that both shrunkenshaped and balloon-like cells could be observed in theS.suisand MVs infections.In addition,MVs endocytosis inhibition resulted in an almost complete decrease in NLRP3 expression and a partial decrease in cell death.We also asked whether the Caspase-4/-5 (Caspase-11 mouse) non-classical pathway of pyroptosis was activated byS.suis2 or MVs,but it was obvious that Caspase-4/-5 protein expression levels did not change in our present results (Bakeret al.2015;Downset al.2020).These results confirmed thatS.suis2 causes the pyroptosis of endothelial cells primarilyviathe secreted MVs by activating the classical NLRP3/Caspase-1 inflammasome pathways.

Recently,many studies have focused on the molecular mechanism of NLRP3/Caspase-1 inflammasome activation.The NLRP3 inflammasome recognizes a range of stimuli,including PAMPs and DAMPs,in order to activate pro-Caspase-1,which then matures into active Caspase-1 and leads to the maturation and secretion of IL-1β and IL-18 (Xueet al.2019).Diverse PAMPs and DAMPs,such as lipopolysaccharide (LPS),asbestos,ATP,and uric acid,lead to NLRP3 activation through an increase in ROS (Cruzet al.2007;Dostertet al.2008).The role of ROS in driving inflammasome activation is well understood and has been identified as an important mechanism of NLRP3 inflammasome activation (Linet al.2019).Studies have shown that damage to mitochondria and the respiratory chain in mitochondria is the main cause of reactive oxygen species (Nolfi-Doneganet al.2020).Contrast media (iohexol) causes mitochondrial damage to epithelial cells,which induces an increase in mitochondrial ROS and activation of the NLRP3 inflammasome (Linet al.2019).In our study,S.suisreleased MVs induced mitochondrial damage and the loss of mitochondrial membrane potential in endothelial cells,and then led to the production and release of mtDNA and ROS.Remarkably,Eno was the most abundant toxin carried by MVs,induced a time-dependent decrease in the mitochondrial membrane potential and led to cell apoptosis (Wuet al.2022).The stress of cytoplasm ROS and mtDNA caused the NLRP3 inflammasome activation (Swansonet al.2019).Our study also indicated that the Eno virulence factor played an important role in the MVs for the MVs-stimulated NLRP3 activation.

5.Conclusion

In summary,the potential mechanisms ofS.suis2 in bacteria-host interactions are based on membrane vesicles (MVs) released fromS.suis2.The MVs carry a variety of toxins and can be transported to endothelial cellsviamainly dynamin-dependent endocytosis,and activate the NLRP3/Caspase-1/GSDMD canonical inflammasome signaling pathway,resulting in cell pyroptosis by mitochondrial damage,which produces mtDNA and ROS (Fig.7).The MVs thus serve as a complex toxin cocktail ofS.suis2 and are quite likely to be involved in theS.suis2 infections and pathogenesis of the associated diseases.

Fig.7 Proposed pathway of pyroptosis induction by Streptococcus suis serotype 2 (S.suis 2)-released membrane vesicles (MVs) in endothelial cells.Streptococcus suis 2-released MVs are internalized by cells via dynamin-dependent endocytosis,which might promote the NLRP3/Caspase-1/GSDMD pathway by inducing mitochondrial damage and causing dissipation of the mitochondrial membrane potential (Δψm).This produces a high cytoplasm ROS level,leading to the pyroptosis of the endothelial cells.

Acknowledgements

Declaration of competing interest

The authors declare that they have no conflict of interest.

Ethical approval

The authors declared that all the experimental procedures did not involve the ethical issues of animal or human research.All experimental procedures were performed strictly in a biosafety-level 2 facility and in accordance with the guidelines of the Institute of Animal Health,Guangdong Academy of Agricultural Sciences,China.

Appendixassociated with this paper is available on https://doi.org/10.1016/j.jia.2023.09.022