The Institutional Animal Care and Use Committee (IACUC) of Augusta University (protocol # 2013-0586) approved all experimental procedures on C57BL/6 mice.

Generation of BMDCs was performed as previously described (Elashiry et al., 2020; Elsayed et al., 2020). Briefly, bone marrow was isolated from femurs and tibiae of young (2 months) and old (22-24 months) C57B6 mice. Red blood cells were lysed using ACK cell lysis buffer (Invitrogen, Thermofisher scientific, and Columbia, SC, USA). Cells were cultured in complete media (RPMI 1640 containing 10% FBS and 100 IU/mL penicillin/streptomycin) in the presence of murine GM-CSF (20 ng/ml) and IL-4(20 ng/ml) (Peprotech, Rocky Hill, NJ, USA). Culture media was changed every 2 days with the addition of fresh growth factors. Cells were re-incubated on day 6 in EXO depleted complete media to generate iDCs or in the presence of 1µg/ml LPS (Sigma, St. Louis, M.O., USA) to generate mature stimDCs for 48 hours. On day 8, cells were washed and re-incubated in EXO free media for 24 h and on day 9, cells were isolated and, culture supernatants were collected for EXO purification. In some of the experiments, bone marrow derived dendritic cells (DCs) were treated at day 6 of culture with Doxorubicin (Sigma Aldrich, St Louise, MO, USA) at a dose of 50nM, Ecoli LPS (Sigma Aldrich, St Louise, MO, USA) at 100ng/ml. P.gingivalis LPS (Invivogen Inc., San Diego, CA, USA) at 100ng/ml, Rapamycin 1µM(R8781)(Sigma Aldrich, St Louise, MO, USA) or PgDCexo, StimDCexo, iDCexo 108 particles/ml (Elashiry et al., 2020).

P.gingivalis 381 (Pg) was maintained anaerobically in (10%, H2, 10% CO2, and 80% N2) in a Coy lab vinyl anaerobic chamber (Coy Laboratory Products, Inc., Grass Lake, MI) at 37◦C in Wilkins-Chalgren anaerobe broth. Bacterial cells were maintained until mid-log phase. Bacterial CFU were calculated based on a spectrophotometer OD 660 reading of 0.11, previously determined to equal 5x10^7 CFU (El-Awady et al., 2015; Meghil et al., 2019). At day 6 of DC culture, cells were pulsed with P.gingivalis at a multiplicity of infection (MOI) of 10 for 48 hours. On day 8, cells were washed and incubated for 24 hrs and on day 9, cells were isolated and, culture supernatants were collected for EXO purification.

Exosome isolation was performed as previously described (Kim et al., 2005; Elashiry et al., 2020). Briefly, supernatants from DC cultures were subjected to differential centrifugation (successive centrifugations at 500 g for (5 min), 2000g for (20 min), and 10,000 g for (30 min) to eliminate cells and debris. This was followed by ultrafiltration 2x with 0.2 µm and 2x with 100 kDA filters (to remove microvesicles and free proteins) and ultracentrifugation for 90 min at 120,000 g. Then EXO pellets were washed with PBS and ultra-centrifuged 2x at 120,000 g for 90 min, and finally re-suspended in 100 µl of PBS, stored in -80 for further analysis and experiments.

For quantification of size and count of nanoparticles in suspension, Nano tracking analysis (NTA) was performed using Zeta view PMX 110 (Particle Metrix, Meerbusch, Germany) as previously described (Helwa et al., 2017; Elashiry et al., 2020). Briefly, 10 µl of the sample was diluted in 1xPBS and loaded into the sample chamber, then size and concentration of the sample were automatically calculated by the software (ZetaView 8.02.28).

As described previously (Helwa et al., 2017; Elashiry et al., 2020) exosome sample was fixed in 4% paraformaldehyde in 0.1M cacodylate buffer ph 7.4 overnight. Twenty microliter (20ul) of suspended exosome preparation was applied to a carbon-Formvar coated 200 mesh nickel grid and allowed to stand 30 minutes. The excess sample was wicked off onto Whatman filter paper. Grids were floated exosome side down onto a 20 µl drop of 1M Ammonium Chloride for 30 minutes to quench aldehyde groups from the fixation step. Grids were floated on drops of blocking buffer (0.4% BSA in PBS) for 2 hours. Then rinsed 3 X 5 minutes each with PBS. Grids were set up as follows and allowed to incubate in blocking buffer or the primary antibody (anti CD63, anti CD81, anti Mfa1) for 1 hour. Grids were floated on drops of 1.4 nm secondary antibody nanogold (Nanoprobes, Inc.) diluted 1:1000 in blocking buffer for 1 hour. Grids were rinsed 3 X 5 minutes each with PBS. Grids were rinsed 3 X 5 minutes each with DI H₂O. For double labelling, grids were enhanced 12 minutes (for Mfa1 antibody) and 3 minutes (for CD81 antibody) in Gold Enhance EM (gold enhancement reagent, Nanoprobes, Inc.) and rinsed in ice cold DI H₂O to stop enhancement. The different times for enhancement allows for the gold particles labeling the first primary antibody to be enhanced twice producing larger size gold particles between 20 and 30 nm in size. The gold particles labeling the second primary antibody is enhanced only once and for a much shorter time to produce gold particles which are smaller in size, between 10 and 12 nm in size. Then grids were negatively stained in 2% aqueous Uranyl Acetate and wicked dry. Grids were allowed to air dry before being examined in a JEM 1230 transmission electron microscope (JEOL USA inc., Peabody MA) at 110 kV and imaged with an UltraScan 4000 CCD camera & First Light Digital Camera Controller (Gatan Inc., Pleasanton, CA.)

Total EXO RNA Kit (4478545), (Thermofisher Scientific, Waltham MA, USA) was used to isolate miRNAs from exosomes according to manufacturer’s instructions. The concentration of miRNA was measured using a NanoDrop spectrophotometer (Thermo Scientific) and analysis of the quality of miRNA was performed using an Agilent 2100 Bioanalyzer. Analysis of miRNAs was performed using an Affymetrix GeneChip^® miRNA 4.0 Array at the Integrated Genomics Core, Augusta University, GA. A P-value cut-off of 0.05 and the miRNAs with a fold change above 1.5 were considered differentially expressed. Bioinformatic analysis was performed using ClastVist software and TargetScan software was used to detect computationally predicted miRNA-mRNA target relationships.

DCs suspension were subjected to cytospins on a microscopic slide before fixation and staining. SA-β-gal histochemical staining kit (CS0030) (Sigma Aldrich, St Louise, MO, USA) was used according to the manufacturer’s instructions. Briefly, cells were fixed with fixation buffer supplied with the kit (20% formaldehyde, 2% glutaraldehyde, 70.4 mM Na2HPO4, 14.7 mM KH2PO4, 1.37 M NaCl, and 26.8 mM KCl), for 6-8 minutes then washed 3x with PBS. This was followed by adding staining solution containing X-Gal and incubated for 12 hours at 37°C with no CO2. Cells were then washed with PBS and visualized under EVOS cell imaging system (Evos FL Auto Imaging System, Thermofisher) and at least 5 random pictures were taken per sample with 60x objective lens.

Cell Event Senescence Green flow cytometry assay kit (Thermofisher Scientific, Waltham MA, USA) was used according to the manufacturer’s instruction. Briefly DCs were first stained for cell surface marker CD11c using regular staining protocol discussed in detail in the following sections. After the last wash, cells were fixed with 4% paraformaldehyde (PFA) for 10 minutes at room temperature followed by washing to remove the fixative solution. Cells were then incubated with Cell Event Senescence Green probe (Thermofisher Scientific, Waltham MA, USA) at 1:500 dilution for 2 hours at 37°C with no CO2. Cells were then washed with PBS, 2% FBS and data acquired by MACSQuant analyzer machine and MACSQuantify software (Miltenyi Biotech Auburn, CA, USA).

Analysis of exosome uptake consisted of labelling exosomes with Dil (D282, Thermofisher Scientific, Waltham MA, USA), then coculturing them with recipient DCs for 24h. Cells were subjected to cytospins, fixed and stained on glass slides with Alex flour 647 phalloidin (A22287) and DAPI (D1306) (Invitrogen, Thermofisher scientific West Columbia, SC, USA). Slides were then visualized by Zeiss upright confocal microscope (Carl Zeiss AG, Oberkochen, Germany) and images were taken with 40x objective lens.

P.gingivalis was labeled with 10 μM carboxyfluoresceine succinimidyl ester (CFSE ebioscience, Invitrogen, Thermofisher Scientific West Columbia, SC, USA) for 1 hour at 37 degree with shaking, then residual stain removed by successive washings and bacterium suspended at 10 MOI for uptake experiments, with/without Cytochalasin D 1uM (Sigma Aldrich, St Louise, MO, USA) for 48 hours, after which cells were harvested and analyzed by flow cytometry and SA-B-gal staining. Cells were harvested, thoroughly washed and then fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100 in PBS and stained with actin ActinRed 555 ReadyProbes Reagent and nuclear counterstaining using DAPI mounting medium; ProLong Gold Antifade Mountant (Invitrogen, Thermofisher scientific, Waltham MA, USA), then images were taken using Zeiss 780 upright confocal microscope (Carl Zeiss AG, Oberkochen, Germany).

FACS Staining Buffer (Thermofisher Scientific, Waltham MA, USA) was used to stain cells on ice. FC receptors (FcR) were blocked using mouse FcR blocking reagent (Miltenyi Biotec, Auburn, CA, USA) for 15 minutes protected from light followed by incubation with conjugated antibodies on ice for 30 minutes. Cells were then washed and re-suspended in FACS buffer and data was acquired using MACSQuant analyzer machine and MACSQuantify software (Miltenyi Biotech Auburn, CA, USA). Antibodies used: CD11c APC; clone N418 (Affymetrix, eBioscience, Thermofisher Scientific, Waltham MA, USA)., CD86 (B7-2) PE; clone GL1(Affymetrix, eBioscience, Thermofisher Scientific, Waltham MA, USA)., CD80 FITC; clone 16-10A1(Invitrogen, Thermofisher Scientific, Waltham MA, USA), CD4 Vioblue; clone GK1.5 (Invitrogen, Thermofisher Scientific, Waltham MA, USA), MHCII Viogreen; clone M5/114.15.2, (Miltenyi Biotech Auburn, CA, USA).

Total RNA was isolated using QIAGEN RNeasy mini kit (Qiagen, Inc., Valencia, CA, and USA). RNA purity and concentration were measured using Nanodrop (NanoDrop 1000 UV-VIS Spectrophotometer Software Ver.3.8.1, Thermofisher Scientific). Ratio of 260/280 of 2.0 was considered adequate for analysis. Reverse transcription to cDNA was performed using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystem, Thermofisher Scientific, Waltham MA, USA) in total reaction of 20 μL. Quantitative real-time PCR was performed using TaqMan fast advanced master mix (Applied Biosystem, Thermofisher Scientific, Waltham MA, USA) and Taq-Man Gene Expression assay (Applied Biosystem, Foster City, CA, USA) specific for: IL6 (Mm00446190_m1), TNF (Mm00443258_m1) and IL1B (Mm00434228_m1), internal control Glyceraldehyde-3-phosphate dehydrogenase (GAPDH Mm99999915_m1). RT-PCR was run in StepOnePlus Real-Time PCR System. Calculation of relative mRNA expression was performed using delta-delta CT and presented as relative fold-change.

Cells or exosomes were lysed using RIPA buffer with the addition of protease/phosphatase inhibitor cocktail and incubated for 20 minutes on ice, then protein lysates stored at -80° till further use. After denaturation of proteins lysates, equal protein concentrations and volumes were loaded and separated by 10-15% Mini-PROTEAN TGX Precast Protein Gel (Bio-Rad Laboratories, Hercules, CA), and transferred onto PVDF membranes (Sigma-Aldrich). Membranes were blocked with 5% nonfat dry milk in TBST, then incubated with primary antibodies at 4° overnight. After washing with TBST, membranes were incubated with HRP-conjugated secondary antibodies for 1 h at room temperature. Membranes were then washed and developed by ECL kit and imaged with ChemiDoc MP Imaging Gel (Bio-Rad Laboratories, Hercules, CA). Antibodies used: anti-Beta-actin (8H10D10) as loading control (Cell Signaling Technology, Danvers, MA, USA). Anti-mouse anti p21^Waf1/Clip1 (#64016) (Cell Signaling Technology, Danvers, MA, USA). Anti-mouse anti p16 ^INK4A PA1030670 (Thermofisher Scientific, Waltham MA, USA) and anti p53(#2524) (Cell Signaling Technology, Danvers, MA, USA), anti-TSG101 (MA1-23296) (Thermofisher Scientific, Waltham MA, USA), anti-Alix (MA1-83977) (Thermofisher Scientific, Waltham MA, USA), anti CD81(#10037) (Cell Signaling Technology, Danvers, MA, USA) Anti IL6(12912) (Cell Signaling Technology, Danvers, MA, USA), anti cleaved-IL-1ββ (#52718) (Cell Signaling Technology, Danvers, MA, USA), anti TNFa (#11948) (Cell Signaling Technology, Danvers, MA, USA), anti-Mfa1 (mAb 89.15 against the native minor fimbriae (Zeituni et al., 2010), generated at the Cell Culture/Hybridoma Facility at Stony Brook University, as reported (Carrion et al., 2012)), secondary antibodies: anti-mouse IgG HRP-linked (#7076) (Cell Signaling Technology, Danvers, MA, USA), anti-rabbit IgG HRP-linked (# 7074) (Cell Signaling Technology, Danvers, MA, USA).

OVA antigen-specific T cells were isolated from OT-II transgenic mice (B6.Cg-Tg(TcraTcrb)425Cbn/J; Jackson Laboratory) using negative selection with Mouse T-cell Enrichment Kit (MagniSort; Thermofisher Scientific, West Columbia, SC, USA). T cell purity was assessed by flow cytometry analysis, using CD4 antibody. OT-II T cells were stained with 0.5 μM CFSE (ebioscience, Invitrogen, Thermofisher Scientific West Columbia, SC, USA) for 15 minutes at 37°C, after which cells were washed 2 times. DCs pretreated with PgDCexo, iDCexo, or StimDCexo, or infected with P.gingivalis (10 MOI) ± Rapamycin were harvested, washed thoroughly and pulsed with 3 μg/mL OVA323-339 peptide (Sigma Aldrich, St Louise, MO, USA) for 24 hours. Cells were then harvested, thoroughly washed and co-cultured with T cells at 1:10 DC: T cell ratio in 96-well round- bottom plates with complete RPMI 1640 (10% fetal bovine serum FBS,1% Pen Strep, 1x non-essential amino acids, 0.1% b-mercaptoetanol). After a 72-hours incubation at 37°C with 5% CO2, the proliferation of CFSE-labeled CD4+ T cells was then assessed by flow cytometry.

PgDCEXO were incubated with microbeads that recognize CD81, CD9 and CD63 positive exosomes (pan EXO) (Miltenyi Biotech Auburn, CA, USA). Magnetic separation of labeled EXO was performed by loading it onto a µ column placed in a magnetic field of µMACS separator (Miltenyi Biotech Auburn, CA, USA). The magnetically labeled EXO are retained within the column and the unlabeled vesicles run through the column. The positively selected retained EXO are then eluted from the column. The number of particles in the run through (passage1) of unlabeled extracellular vesicles (EV) fraction was then counted using NTA. Magnetic separation of the run through was repeated (passage 2) to completely deplete pan EXO from the run through and then particles counted in the final run through of unlabeled EV and a percentage to the total number of EV was calculated. The eluate (CD81, CD9 and CD63 positive exosomes) and the run through were then analyzed using WB.

Data was analyzed using GraphPad Prism 6 (GraphPad Software, La Jolla, CA). Data analysis was performed by two-way or one ANOVA with significance defined as P < 0.05, and confidence level of 95% confidence interval followed by Tukey’s multiple-comparisons test. Values are expressed as mean ± standard deviation (SD) and experiments were repeated 3 times.

DCs are resident sentinel immune cells in oral mucosa and the bloodstream, previously shown to contain P. gingivalis microbial signatures in periodontitis (PD) patients (Carrion et al., 2012; El-Awady et al., 2019; Rajendran et al., 2019). Here, bone marrow derived DCs (DCs) from young (yDCs) or old (oDCs) mice were cocultured with P. gingivalis or doxorubicin and subjected to CS-profiling. Both P. gingivalis and doxorubicin stimulated an increase in p16 ^INK4A, p53 and p21^Waf1/Clip1 in yDCs ( Figures 1A, B ). oDCs had a trend towards constitutively higher levels of senescence markers than yDCs shown by immunoblot analysis of induced p16 ^INK4A, p53 and p21^Waf1/Clip1,however, it only reached significance level in P53 upregulation. Chromogenic staining ( Figure 1C ) and FACS analysis ( Figure 1D ) confirmed SA-β-Gal induction in yDCs and oDCs in response to P. gingivalis or doxorubicin. Senolytic agent rapamycin (Sargiacomo et al., 2020) reversed P. gingivalis-induced p16 ^INK4A, p53 and p21^Waf1/Clip1 and P.gingivalis-induced SA-β-Gal in in yDCs and oDCs. P.gingivalis LPS induced increased expression of p53 and p21^Waf1/Clip1, but not p16 ^INK4A and SA-β-Gal in yDCs compared to control yDCs with no infection. IL-1β, TNFa, and IL6 are known to be major constituents of the SASP. Here we show that P. gingivalis was the most potent inducer of IL-1b, and TNFa transcripts in yDCs and oDCs while increasing IL-6 relative mRNA expression only in yDCs. Rapamycin ablated TNFa, and reduced IL-1b but did not influence IL-6 induction by P. gingivalis in yDCs ( Figure 1E ).

In view of CS profiling results, we examined the phenotype and maturation functions of oDCs and yDCs before and after co-culture with P. gingivalis or LPS by FACS analysis ( Figure 2A ). Constitutive expression of CD86 and MHCII was lower in oDCs than yDCs. E. coli LPS elicited a strong DC maturation response, relative to PgLPS in yDCs, but oDCs were less responsive to either stimulant. P.gingivalis did not induce maturation of yDCs or oDCs, relative to control and levels were significantly lower than that induced by E.coli LPS, and CD86 and MHCII significantly lower than that induced by Pg LPS. Interestingly, P.gingivalis significantly reduced constitutive CD86, and CD80 levels in oDCs (51.06%, 47.8%) compared to control oDCs with no infection (27.6% and 33.9%) respectively ( Figures 2B, C ). Rapamycin reversed P.gingivalis-induced deficit in maturation as shown by a significant increase in expression of MHCII, CD86 and CD80 in yDCs and oDCs.

We hypothesized that active invasion of DCs by P. gingivalis (Zeituni et al., 2009; Zeituni et al., 2010; Zeituni et al., 2010; Acosta et al., 2013) was a critical factor in robust immune senescence. Accordingly, yDCs were cocultured with CFSE-labeled P. gingivalis in the presence or absence of cytochalasin D. Invasion was imaged by confocal microscopy, with SA-β-Gal, MHCII, CD80 and CD86 expression assessed by flow cytometry. Images show internalized P. gingivalis ( Figure 3A ), associated with increased SA-β-Gal expression on yDCs, and unresponsiveness in MHCII, CD80 and CD86 expression ( Figures 3B, C ). Pre-treatment of yDCs with cytochalasin D reduced P. gingivalis invasion, reversed SA-β-Gal response and restored the MHCII, CD86 and CD80 responsiveness ( Figures 3B, C ). E coli LPS control established that yDCs were fully capable of maturation.

CS is a complex cellular inflammatory process in non-immune (Naylor et al., 2013), and immune cells (Mondal et al., 2013; Salminen, 2020); most notable feature being induction of the SASP. Exosomes have emerged as an important characteristic of the SASP (Acosta et al., 2013). Here, exosomes released by P.gingivalis infected yDCs (PgDCexo), relative to uninfected control DCs were isolated, characterized and quantitated. Correct size distribution (30-150 nm) was confirmed by nanotracking analysis ( Figure 4A ). Enumeration of secreted exosomes indicate 2-fold greater number of exosomes from P. gingivalis-infected yDCs (=4.2 x 10⁸ per 10⁶ DCs) than control DCs (ctl DC) (=2 x 10⁸). This increase was ablated by senolytic agent rapamycin (=2.2 x 10⁸) ( Figure 4B ). Correct phenotype of typical DC exosomes was confirmed by immunogold TEM for CD63 ( Figure 4C ) and SEM showing characteristic cup-shaped morphology ( Figure 4D ). Western blotting analysis ( Figure 4F ) establishes typical exosomal markers CD81, Alix and TSG101 on PgDCexo and control iDC exo and stim DCexo (Elashiry et al., 2020). PgDCexo also express the minor Mfa1 fimbrial protein, an adhesin used by P.gingivalis to invade DCs in vitro (Zeituni et al., 2010) and in vivo (Carrion et al., 2012; El-Awady et al., 2015; El-Awady et al., 2019). Co-localization of Mfa1 fimbrial protein with the tetraspanin CD81, a eukaryotic exosomal marker was confirmed using TEM immuno-gold double labeling ( Figure 4E ). Passage of PgDC exo over a Pan exo magnetic bead column 1x and 2x removed 99.9% and 99.99% of particles, respectively. Column eluate was enriched in Mfa1, CD81, Alix, TSG101 and CD9, consistent with bona fide exosomes, while run-through contained trace levels of Mfa1, Alix, TSG101 and CD9 ( Supplementary Figure 1 ). PgDCexo also contain IL-6 and mature IL-1β, consistent with inflammasome activation and the SASP (Acosta et al., 2013). StimDCexo, from donor yDCs treated with LPS, also expressed IL-6 and IL-1β, as previously reported (Elashiry et al., 2020), albeit at lower levels than PgDCexo ( Figure 4F ). Bioinformatics analysis of miRNA content of exosomes ( Figure 4G ) show miRNAs upregulated (red) and down regulated (blue) in PgDCexo, relative to control exosomes from imDCs or LPS-matured DCs (stimDC). Distinct pattern of blue/red miRNAs in PgDCexo versus controls is particularly striking. Of particular note are miRNAs involved in disruption of immune homeostasis, through anti-apoptosis and anti-autophagy functions, including miR106b, miR-17-5p, miR378a-3p, miR-324-5p, miR-132-3p. MiR17-5p was 1.71-fold upregulated in PgDCexo vs. imDCexo, and 1.2-fold downregulated in mDCexo vs imDCexo ( Figure 4G ). MiRNA 132-3p was 4-fold upregulated (p<0.05) in Pg-induced exo, consistent with that previously observed in aged BMDCs (Park et al., 2013).

We hypothesized PgDCexo would be potent inducers of immune senescence in non-senescent recipient yDCs. Co-culture of Dil-labeled PgDCexo with recipient yDCs confirmed their uptake as shown by confocal microscopy ( Figure 5A ). CS profiling of recipient yDCs indicate activation of SA-β-Gal, by chromogenic staining ( Figure 5B ) and confirmed by flow cytometry using fluorescent SA-β-Gal staining ( Figure 5C ). Moreover, this increase was significant compared to control yDCs and yDCs treated with exosomes from immature DCs (iDCexo). A significant increase in SA-β-Gal⁺ DCs was also induced directly by doxorubicin and Pg LPS, while stimDCexo induced a non-significant trend for increased SA-β-Gal⁺ DCs ( Figure 5C ). Western blot analysis of PgDCexo recipient DCs ( Figures 5D, E ) showed a significant upregulated expression of p16 ^INK4A, p53 and p21^Waf1/Clip1 relative to non-treated DCs and similar to Dox-treated yDCs. No effect on p16 ^INK4A, p53 and p21^Waf1/Clip1 expression was observed with iDCexo treatment, while stimDCexo induced upregulation which was not statistically significant. Pg LPS showed an upregulation in p16 ^INK4A with no significant effect on p53 and p21^Waf1/Clip1 expression. Rapamycin reversed PgDCexo-induced senescence ( Figures 5C, E ).

The maturation and antigen presentation functions of yDCs cocultured with PgDCexo were examined. FACS analysis revealed ablation of maturation function of PgDCexo recipient yDCs as shown by a significant decrease in maturation markers MHCII, CD86 and CD80 ( Figures 6A, B ), which was restored by rapamycin treatment. Positive control LPS and stimDCexo, but not iDCexo, promoted yDC maturation,. Ability of yDCs treated with PgDCexo to engage in antigen presentation was assessed by coculture of DCs with Ova peptide and CFSE labeled splenic CD4+ T cells from OTII mice. T cell proliferation was assessed by percentage loss of CFSE with each T cell generation, with gating shown in Figure 7A ). Negative controls for proliferation include T cells only, yDC/T cells only, Ova/T cells, and Rap/T cells, none of which supported proliferation ( Figure 7C ). Robust proliferation of T cells (=84.4%) was induced by yDCs pulsed with Ova. Coculture of yDCs with PgDCexo ablated T cell proliferation by 51%. This was reversed with rapamycin. Direct infection of yDCs with P. gingivalis inhibited T cell proliferation to Ova by 30%, which was restored with rapamycin. Control iDC exo from immature yDCs did not alter T cell proliferation ( Figure 7B ). Summary graphs are shown in Figure 7D .