Selenomonas sputigena Interactions with Gingival Epithelial Cells That Promote Inflammation

ABSTRACT

Increased prevalence and abundance of Selenomonas sputigena have been associated with periodontitis, a chronic inflammatory disease of tooth-supporting tissues, for more than 50 years. Over the past decade, molecular surveys of periodontal disease using 16S and shotgun metagenomic sequencing approaches have confirmed the disease association of classically recognized periodontal pathogens such as Porphyromonas gingivalis, Treponema denticola, and Tannerella forsythia while highlighting previously underappreciated organisms such as Filifactor alocis and S. sputigena. Despite abundant clinical association between S. sputigena and periodontal disease, we have little to no understanding of its pathogenic potential, and virulence mechanisms have not been studied. In this study, we sought to characterize the response of gingival epithelial cells to infection with S. sputigena. Here, we show that S. sputigena attaches to gingival keratinocytes and induces expression and secretion of cytokines and chemokines associated with inflammation and leukocyte recruitment. We demonstrate that S. sputigena induces signaling through Toll-like receptor 2 (TLR2) and TLR4 but evades activation of TLR5. Cytokines released from S. sputigena-infected keratinocytes induced monocyte and neutrophil chemotaxis. These results show that S. sputigena-host interactions have the potential to contribute to bacterially driven inflammation and tissue destruction, the hallmark of periodontitis. Characterization of previously unstudied pathogens may provide novel approaches to develop therapeutics to treat or prevent periodontal disease.

INTRODUCTION

Periodontitis is a chronic inflammatory disease affecting gingival soft tissues and leading to the slow, progressive loss of periodontal ligament attachment and alveolar bone (1). Chronic periodontitis is one of the most common diseases in humans, affecting ~115 million middle-aged adults, with global direct treatment costs estimated at $298 billion in 2010 (2, 3). Gingivitis is caused by a localized inflammatory response to the oral polymicrobial biofilm and, if untreated, can progress into chronic periodontitis characterized by nonresolving, destructive inflammation that can lead to alveolar bone loss. The chronic inflammation observed during periodontal diseases is linked to other chronic inflammatory diseases such as cardiovascular disease, rheumatoid arthritis, and cancer (4–6). Periodontitis is thought to develop as a collective action of the polymicrobial community that leads to dysregulated inflammation and subsequent tissue damage that provides nutrients to sustain the dysbiotic community. This vicious cycle is maintained by inflammophilic pathobionts that increase in abundance as disease progresses and exacerbate inflammation.

Dental plaque is a complex polymicrobial biofilm community that develops on the surface of teeth and is a direct precursor to periodontal diseases (7). In the first step of the development of pathogenic microbial communities, physiologically compatible organisms coadhere and coordinate their activities through complex signaling mechanisms. Dental plaque development progresses spatially and temporally as physiologically compatible organisms continue to accumulate (8, 9). Recent advances in molecular techniques for the identification of bacteria have highlighted several previously unrecognized or underappreciated organisms associated with chronic periodontitis (10–13). Selenomonas sputigena is one such resurgent periodontal pathogen that has recently been found more frequently from patients with periodontitis than from healthy subjects (10, 12, 14–16).

Crescent-shaped bacteria, such as Selenomonas, have been described for dental plaque since first observed by Antonie van Leeuwenhoek in 1683 and were routinely isolated from dental samples in the 1950s (17). Indeed, gingival inflammation has been associated with increased abundance of S. sputigena for more than 4 decades (18, 19). Improved culture conditions led to the isolation of additional oral Selenomonas species (S. noxia, S. flueggei, S. dianae, S. artemidis, and S. infelix), and many of these species were found to associate with the established periodontal pathogen Fusobacterium nucleatum (19, 20). While the abundance of Selenomonas varies in epidemiological studies, the high prevalence in association with periodontal disease suggests that S. sputigena may be used as a marker for active disease (21). Very little is known about the pathogenic capabilities of S. sputigena, and host-pathogen interactions involving S. sputigena have never been examined. While rare, there are several reports of fatal septicemia caused by S. sputigena, indicating the organism’s invasive and pathogenic potential (17, 22, 23). Periodontitis-associated metatranscriptomes compared to those from healthy sites demonstrated that S. sputigena significantly contributes to enrichment of bacterial chemotaxis, flagellar assembly, and two-component signaling systems during disease, suggesting that S. sputigena not only is present during disease but also contributes to disease-associated processes (24).

A recent microscopic analysis of gingival biopsy specimens found Selenomonas at high levels throughout the collected biofilms. Selenomonads were found to be densely packed within the biofilm in association with both the tooth surface and the gingiva (25). We hypothesized that S. sputigena would orchestrate inflammatory gingival epithelial cell (GEC) responses consistent with periodontal pathogens. Thus, we investigated whether S. sputigena adheres to GECs and induces expression and secretion of proinflammatory cytokines while also leading to cell death. The results begin to establish the pathogenic potential of S. sputigena and demonstrate that it may contribute to the elevated proinflammatory state that is associated with the progression of periodontitis.

RESULTS

S. sputigena attaches to HIGKs.

We utilized widefield fluorescence microscopy and laser scanning confocal microscopy (LSCM) to visualize fluorescently labeled S. sputigena binding to the GEC surface (Fig. S1). We consistently observed S. sputigena binding to the surface of gingival keratinocytes at a low frequency with human immortalized gingival keratinocytes (HIGKs) but at a much higher frequency with telomerase immortalized gingival keratinocytes (TIGKs). As a complementary approach, we incubated fluorescently labeled S. sputigena with HIGKs and TIGKs for 4 h with increasing multiplicities of infection (MOIs) before washing and quantifying binding by flow cytometry (Fig. 1). We observed an average of 1.92% of HIGKs with bound S. sputigena at an MOI of 10, 4.05% at an MOI of 50, 7.14% at an MOI of 100, and 19.85% at an MOI of 200, compared to 24.74% of TIGKs with bound S. sputigena at an MOI of 10, 54.94% at an MOI of 50, 68.32% at an MOI of 100, and 81.98% at an MOI of 200 across three independent experiments with at least 50,000 cells per replicate. To validate the approach, we also observed robust binding of Porphyromonas gingivalis to both HIGKs and TIGKs (see Fig. S2 in the supplemental material). These results demonstrate that S. sputigena attaches to the surface of two gingival epithelial cell lines, albeit with significantly variable efficiencies. We then sought to determine if these interactions with GECs may influence toll-like receptor activation.

FIG 1.

S. sputigena adherence to TIGKs is more abundant than to HIGKs. S. sputigena was fluorescently labeled with CellBright-488 prior to incubation with HIGKs or TIGKs on glass coverslips at MOIs of 10, 50, 100, and 200 for 4 h. Attachment of S. sputigena was measured by flow cytometry. (A) Representative histograms are shown for S. sputigena binding to HIGKs and TIGKs. (B) The results are the averages of 3 biological replicates, each with >50,000 HIGKs or TIGKs, with standard deviations. Treated cells were compared to untreated controls by one-way analysis of variance (ANOVA) with Dunnett’s post hoc test (**, P < 0.01; ****, P < 0.0001; ns, not significant).

S. sputigena selectively activates NF-κB through TLR activation.

Next, we sought to characterize S. sputigena activation of Toll-like receptors (TLRs) using a human TLR-reporter system. We infected HEK 293 cells expressing a soluble alkaline phosphatase (SEAP) that is induced by NF-κB and AP-1 following activation of either human TLR2/1, TLR2/6, TLR4, or TLR5 (HEK-Blue hTLR2-hTLR1, HEK-Blue hTLR2-TLR6, HEK-Blue hTLR4, or HEK-Blue hTLR5 cells) with S. sputigena at MOIs of 1, 10, and 100 and then incubated the cells for 24 h before colorimetric quantification of SEAP by spectrophotometry (Fig. 2A to D). As controls, HEK-Blue hTLR2-hTLR1, HEK-Blue hTLR2-TLR6, HEK-Blue hTLR4, or HEK-Blue hTLR5 cells were left unstimulated or treated with purified PAM3CSK4, FSL-1, lipopolysaccharide (LPS) or flagellin, respectively. We observed a robust induction of SEAP when HEK-Blue hTLR2-TLR6 and HEK-Blue hTLR4 cells were treated with S. sputigena at all MOIs, while the HEK-Blue hTLR2-hTLR1 and HEK-Blue hTLR5 cells did not respond to challenge with S. sputigena. These data suggest that S. sputigena flagella may have evolved to evade recognition by TLR5. We compared the flagellin sequence from S. sputigena to the flagellin sequence from bacteria known to stimulate and evade TLR activation and determined that S. sputigena flagellin contains conserved amino acid substitutions at residues within the TLR5 recognition motif (Fig. 2E).

FIG 2.

S. sputigena stimulates TLR2/6 and TLR4 but evades TLR2/1 and TLR5 recognition. (A to D) HEK-Blue hTLR2-hTLR1 (A), HEK-Blue hTLR2-hTLR6 (B), HEK-Blue hTLR4 (C), or HEK-Blue hTLR5 (D) cells were treated with S. sputigena (MOIs of 1, 10, and 100) or the appropriate control (1 or 10 ng/mL). Following the treatment, secreted SEAP was quantified using the HEK-Blue detection reagent and measured colorimetrically at 655 nm. The results are the averages of 3 biological replicates with SEMs. Treated cells were compared to untreated controls by one-way ANOVA with Dunnett’s post hoc test (****, P < 0.0001). All data under the line were determined to be significantly different from the untreated control. (E) Flagellin sequences were retrieved from NCBI, aligned by Clustal Omega, and visualized by MView. Bacterial flagellin proteins reported to evade TLR5 are shown in the yellow box, and conserved residues key to recognition by TLR5 are shown in the red boxes.

Gingival keratinocyte expression of selected MMPs is induced by S. sputigena.

To determine if S. sputigena challenge of HIGKs and TIGKs influences the production of matrix metalloproteases (MMPs), cells were infected with S. sputigena at 100 for 4 h, washed, and incubated for another 20 h before quantification of secreted MMPs and tissue inhibitors of metalloproteases (TIMPs) in the culture supernatant using the 13-plex MMP/TIMP array by Eve Technologies (Fig. 3). We observed significantly elevated levels of MMP9 and -13 and TIMP1, while MMP8 and TIMP3 and -4 were undetected and MMP2, -3, -7, -10, and -12 and TIMP2 were not significantly different from uninfected cells using the HIGKs. S. sputigena infection of TIGKs led to significantly higher detection of MMP1, -9, and -10, with no significant differences observed in MMP2, -3, -7, or -12. MMP8 and -13 and TIMP2, -3, and -4 were undetected, while the abundance of TIMP1 was modestly but statistically reduced. To validate the array data, we also performed reverse transcription-quantitative PCR (qRT-PCR) to detect expression of MMPs/TIMPs 4 h postinfection in the HIGKs (Fig. S3). We observed an MOI-dependent increase in the expression of MMP1, -9, -10, and -13 that was consistent with the protein abundance determined in the multiplex array. Expression of MMP3 was significantly elevated; however, MMP3 protein secretion was unchanged in both the HIGKs and TIGKs. Transcript levels for TIMP1 to -4 were unchanged in the qRT-PCR studies.

FIG 3.

Matrix metalloprotease secretion is elevated in response to S. sputigena infection. Both GEC lines were infected with S. sputigena at an MOI of 100 for 4 h prior; the cells were washed and incubated in fresh medium for 20 h prior to quantification of MMP and TIMP abundance by Eve Technologies on a fee-for-service basis and comparison to uninfected controls. The results are the averages of 3 biological replicates with standard deviations and were analyzed by Student’s t test (*, P < 0.05; **, P < 0.01; ***, P < 0.001).

S. sputigena infection induces proinflammatory responses.

Next, we sought to determine the cytokine and chemokine responses of GECs to S. sputigena. Cytokine and chemokine levels in culture supernatants were determined by using a 48-plex cytokine array by Eve Technologies (Fig. 4). HIGKs and TIGKs were infected with S. sputigena at an MOI of 100 for 4 h, before being washed, and incubated for 20 h. Twenty-four hours after infection with S. sputigena, both gingival epithelial cell lines had elevated levels of interleukin 6 (IL-6), IL-8, granulocyte colony-stimulating factor (G-CSF), CXCL1 (GROα), tumor necrosis factor alpha (TNF-α), IL-1α, CXCL10 (IP10), and RANTES compared to those in uninfected controls. The HIGKs were found to produce significantly higher levels of vascular endothelial growth factor A (VEGF-A) in response to S. sputigena infection, while only the TIGKs produced elevated granulocyte-macrophage colony-stimulating factor (GM-CSF) and IL-1RA. Notably, soluble CD40L (sCD40L), epidermal growth factor (EGF), eotaxin, fibroblast growth factor 2 (FGF-2), FLT-3 ligand, fractalkine, alpha 2 interferon (IFN-α2), IFN-γ, IL-1β, IL-2, IL-3, IL-4, IL-5, IL-7, IL-9, IL-10, IL-12(p40), IL-12(p70), IL-13, IL-15, IL-17A, IL-17E/IL-25, IL-17F, IL-18, IL-22, IL-27, MCP-1, MCP-3, M-CSF, MDC, MIG/CXCL9, MIP-1α, platelet-derived growth factor AA (PDGF-AA), PDGF-AB/BB, and TNF-β either were secreted below the level of detection of the array or were not significantly induced by GECs in response to S. sputigena (data not shown). Again, we determined the gene expression of IL-6, IL-8, and IP-10 (CXCL10) by qRT-PCR 4 h postinfection in the HIGKs (Fig. S4). In agreement with the cytokine array, the expression of all cytokines examined was significantly induced in an MOI-dependent manner along with the expression of RelA (p65).

FIG 4.

S. sputigena induces robust proinflammatory cytokine secretion. HIGKs and TIGKs were infected with S. sputigena at an MOI of 100 for 4 h and incubated for an additional 20 h prior to collection and filtration of supernatants. As controls, GECs were left uninfected. Cytokines were quantified by Eve Technologies on a fee-for-service basis. The results are the averages of 3 biological replicates with standard deviations and were analyzed by Student’s t test (*, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001).

Chemokines produced during S. sputigena infection of GECs lead to robust chemotaxis of monocytes and neutrophils.

Our observation that several chemokines were significantly overexpressed and more abundant following challenge with S. sputigena suggested that interactions at the junctional epithelium may signal for the recruitment of immune cells, which are the dominant drivers of periodontal immunopathology. Finally, we infected HIGKs with S. sputigena at MOIs of 10, 50, and 100 for 4 h before collecting supernatants. This conditioned supernatant was cleared of cells and bacteria by centrifugation and filtering, respectively. Next, we examined the potential chemotactic activity of the conditioned supernatants toward naive monocytes and neutrophils. Figure 5 shows potent chemotactic activity toward monocytes and neutrophils from S. sputigena-conditioned supernatant. In contrast, conditioned supernatant collected from uninfected HIGKs showed no chemotactic activity toward either leukocyte. Furthermore, similar numbers of naive monocytes and neutrophils migrated toward S. sputigena-conditioned supernatant as toward the positive chemotactic controls, CCL1 and CXCL1. Interestingly, supernatants from HIGKs infected with as low an MOI as 10 led to a robust chemotaxis of both monocytes and neutrophils. These data suggest that S. sputigena interactions with gingival epithelial cells lead to robust recruitment of leukocytes that may promote destructive inflammation.

FIG 5.

Supernatants from S. sputigena-infected GECs induce monocyte and neutrophil chemotaxis. Monocytes and neutrophils were placed in the upper chamber of the Transwell system. Buffer, CCL3 or CXCL1, or S. sputigena conditioned HIGK supernatant was placed in the lower well, and after 120 min (A) or 30 min (B) of incubation, the membrane was stained with a HEMA 3 stain set kit. Chemotaxis was assessed by light microscopic examination (magnification, ×100). Data are expressed as mean numbers ± SEMs of migrated cells/insert from 3 independent experiments. Results for control conditions (CCL3 or CXCL1) were compared to those for buffer control samples and analyzed by Student’s t test (#, P < 0.001). S. sputigena conditioned samples were compared to supernatants from uninfected HIGKs and analyzed by one-way ANOVA with Dunnett’s post hoc test (**, P < 0.01; ***, P < 0.001).

DISCUSSION

The abundance of S. sputigena in subgingival plaque has been associated with periodontitis for decades. Recently, efforts to understand the changes to the oral microbiome during the transition from periodontal health to disease have reinforced the elevated prevalence and abundance of S. sputigena during gingivitis and periodontitis (10, 12, 15). Prior to this study, nothing was known regarding S. sputigena pathogenicity or possible virulence mechanisms.

In this study, we demonstrated that S. sputigena attaches to gingival epithelial cells, resulting in elevated secretion of MMPs and proinflammatory cytokines. We found using HEK-Blue human TLR reporter cells that S. sputigena strongly induces NF-κB signaling via activation of TLR4 and TLR2/6 but does not activate TLR2/1 or TLR5. An important limitation of this study is that the activation of TLRs was not studied in the gingival epithelial cells directly. S. sputigena activation and signaling via TLR4 are supported in the literature, as the chemical composition of purified LPS from S. sputigena was determined to be typical; however, testing in mice revealed S. sputigena to be lethal, with a pyrogenic reaction 100 times stronger than LPS purified from Eikenella corrodens and Escherichia coli (26). The same study also demonstrated that LPS purified from S. sputigena provoked a positive Schwartzman reaction in rabbits at 0.5 μg, while E. coli LPS elicited the same reaction at 100 μg (26). Collectively, these studies demonstrate that LPS from S. sputigena is a potent virulence factor that promotes inflammation during periodontitis.

Interestingly, S. sputigena evaded TLR5 stimulation, as its flagellar protein contains amino acid differences within the TLR5 recognition motif. Helicobacter pylori and Campylobacter jejuni require motility for colonization and infection of mammals and produce flagellar proteins with similar properties to evade TLR5 activation, suggesting that S. sputigena motility may be critical to its colonization and fitness in the oral cavity (27). Oral metatranscriptomics demonstrate that S. sputigena heavily contributes to in vivo expression of genes associated with motility, chemotaxis, and two-component signaling in periodontal disease pockets (24). S. sputigena heavily glycosylates its flagella using heterogenic O-glycans, yet the role of this unique flagellar glycosylation, and motility in general, in pathogenicity remains uncharacterized (28). Motility and chemotaxis are required for tissue penetration and cell invasion by the periodontal pathogen Treponema denticola (29, 30). To what extent Selenomonas motility, chemotaxis, and signaling systems that regulate motility participate in host-pathogen interactions and virulence will be addressed in future studies.

Currently, there are six Selenomonas spp. isolated from the oral cavity and several other Selenomonas spp. that are elevated during periodontitis by 16S rRNA sequencing but have yet to be cultivated (16). Collectively, selenomonads can constitute 10 to 30% of the total bacterial population in subgingival sites showing signs of clinical periodontitis (31). Many oral selenomonads are enriched in gingivitis and periodontitis sites; in particular, S. noxia and S. flueggei have the most consistent association with disease, along with S. sputigena (11). S. noxia was shown to increase during experimental gingivitis, and studies of early periodontal lesions found that S. noxia was among the principal species associated with sites that converted from periodontal health to disease (32–35). S. noxia also correlates with obesity, as 98.4% of overweight women could be predicted by the presence of S. noxia over 1% of the total salivary microbiome. Obesity and poor glycemic control are reciprocal risk factors or comorbidities with periodontitis, suggesting that the abundance of S. noxia correlates with overall health. While our current study highlights S. sputigena as a periodontal pathogen, future studies should also investigate the role of other oral Selenomonas spp. in periodontal disease.

The oral microbiome is a dynamic and complex community of hundreds of microbes that synergistically promote periodontal disease. Our current study focused on S. sputigena in isolation to better understand its pathogenicity; however, additional studies must characterize biofilm development and polymicrobial community dynamics involving S. sputigena. A study by Kolenbrander et al. demonstrated that S. sputigena and other oral selenomonads adhere to F. nucleatum and exhibit coaggregation (20). Co-occurrence studies suggest that S. noxia and S. sputigena form multiple interactions with oral microbes associated with both health and disease (36), and topographical characterization of subgingival biofilms found that S. sputigena localized throughout the biofilm and contributed to its architecture (25). These studies suggest that S. sputigena is involved in multiple polymicrobial interactions to colonize the subgingival community. Within the plaque biofilms, cell-cell signaling and metabolic interactions contribute to synergistic interactions that can increase the fitness and nososymbiocity, or the pathogenic capability of the whole community. The role of S. sputigena in polymicrobial community dynamics remains unstudied and is critical to fully understanding its pathogenic contributions.

A limitation of this study is that all experiments were in vitro or ex vivo, with no experimental model of periodontitis in animals. The only known niche for S. sputigena is the human oral cavity, and the genus Selenomonas has yet to be detected in the murine oral microbiome (37). Thus, our current studies focused exclusively on human cell culture-based approaches to understand the potential pathogenic mechanisms of S. sputigena. Rodents are routinely utilized to study oral pathogens, and future studies will employ a combinatorial approach of more complex human models, such as tissue models or organoids, and murine models of experimental periodontitis.

Here, we show S. sputigena interacts with gingival keratinocytes to induce proinflammatory cytokines and chemokines that can lead to phagocyte recruitment, which may ultimately be responsible for promoting alveolar bone loss. This is the first study to demonstrate that S. sputigena interactions with host cells promote cellular processes associated with the immunopathology of periodontitis. Future studies will seek to identify and characterize virulence factors of S. sputigena and demonstrate direct pathogenesis in experimental models of periodontitis. Current prevention and treatment strategies for periodontitis focus on a few oral bacteria with previously characterized pathogenicity. This study highlights the opportunity for new approaches to prevent disease caused by emerging or previously uncharacterized periodontal pathogens.

MATERIALS AND METHODS

Bacterial strains, eukaryotic cells, and growth conditions.

Selenomonas sputigena ATCC 35815 was purchased from ATCC and grown anaerobically at 37°C in TYGVS broth (38). P. gingivalis ATCC 33277 was cultured anaerobically at 37°C in Trypticase soy broth (TSB) supplemented with 1 mg/mL yeast extract, 5 μg/mL hemin, and 1 μg/mL menadione. Human immortalized gingival keratinocytes (HIGKs) (39) and telomerase immortalized gingival keratinocytes (TIGKs) (40) were maintained at 37°C and 5% CO₂ in DermaLife K serum-free complete medium (Lifeline Cell Technologies). Epithelial cells between 70 and 90% confluency were counted using a Countess FL2 cell counter (Thermo) and were stimulated with bacteria as described for the individual experiments.

Fluorescence microscopy.

For widefield microscopy and laser scanning confocal microscopy (LSCM) analyses of S. sputigena association, GECs were seeded at 1 × 10⁵ on glass coverslips in 12-well plates and grown until ~60% confluent. Cells were infected with BacLight green-labeled S. sputigena at MOIs of 10, 50, and 100 for 1 h. Coverslips were washed 3 times with phosphate-buffered saline (PBS) and fixed for 10 min in 4% paraformaldehyde. Following a 20 min block in 5% bovine serum albumin, actin was labeled using 1:400 Alexa Fluor 568-phalloidin (Invitrogen) for 1 h at room temperature. After 4 washes in PBS, coverslips were mounted using ProLong gold with 4′,6-diamidino-2-phenylindole (DAPI) mounting medium (Invitrogen). Images were acquired using either a Zeiss Axio Observer 7 or a Zeiss LSM700 instrument as part of the VCU Microscopy Core Facility.

Flow cytometry.

S. sputigena and P. gingivalis cultures were adjusted to 1 × 10⁹ cells/mL and labeled with CellBrite488 membrane stain (Biotium). HIGKs and TIGKs were seeded in 6-well plates and grown to 80 to 90% confluency before infection with either S. sputigena or P. gingivalis (MOIs of 10, 50, 100, and 200) for 4 h. Following infection, cells were washed twice with PBS, detached with trypsin, and resuspended in PBS. To measure extracellular adherence, samples were analyzed on a BD FACS Canto II cell sorter, HIGKs were gated based on forward and side scatter of uninfected cells, and 50,000 cells were analyzed per sample at 488 nm for measuring green fluorescence.

Human MMP/TIMP and cytokine/chemokine arrays.

HIGKs and TIGKs were infected with S. sputigena at an MOI of 100 for 4 h at 37°C and 5% CO₂. Following the infection, the cells were washed and fresh medium was added to each well for an additional 20 h. The supernatant was then removed, centrifuged at 5,000 × g for 5 min at room temperature, and then filtered with 0.22-μm syringe filters to generate bacteria and cell-free conditioned medium. The multiplexing analysis was performed using the Luminex 200 system (Luminex, Austin, TX, USA) by Eve Technologies Corp. (Calgary, Alberta) on a fee-for-service basis. For the cytokine/chemokine array, 48 markers were simultaneously measured in the samples using Eve Technologies’ Human Cytokine 48-Plex Discovery Assay. The 48-plex consists of sCD40L, EGF, eotaxin, FGF-2, FLT-3 ligand, fractalkine, G-CSF, GM-CSF, GROα, IFN-α2, IFN-γ, IL-1α, IL-1β, IL-1RA, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-12(p40), IL-12(p70), IL-13, IL-15, IL-17A, IL-17E/IL-25, IL-17F, IL-18, IL-22, IL-27, IP-10, MCP-1, MCP-3, M-CSF, MDC, MIG/CXCL9, MIP-1α, MIP-1β, PDGF-AA, PDGF-AB/BB, RANTES, transforming growth factor α (TGFα), TNF-α, TNF-β, and VEGF-A. For the MMP/TIMP array, 13 markers were simultaneously measured in the samples using Eve Technologies’ Human MMP and TIMP Discovery Assay. The 13-plex consists of MMP1, MMP2, MMP3, MMP7, MMP8, MMP9, MMP10, MMP12, MMP13, TIMP1, TIMP2, TIMP3, and TIMP4.

Gene expression by qRT-PCR.

GECs were seeded on 6-well plates at 5 × 10⁵ per well and grown to 60 to 80% confluency. GECs were infected with S. sputigena at MOIs of 10, 50, and 100 for 4 h and incubated at 37°C. Following the infection, the medium was removed and RNA was isolated using the RNeasy Plus minikit (Qiagen). RNA concentrations were quantified using a NanoDrop One spectrophotometer (Thermo Scientific), and cDNA was synthesized from total RNA (1 μg RNA per reaction) using a high-capacity cDNA reverse transcription kit (Applied Biosystems). Quantitative RT-PCR used TaqMan fast advanced PCR master mix and gene expression assays listed in Table S1 (Applied Biosystems). Real-time PCR was performed on an Applied Biosystems QuantStudio 3 system. The cycle threshold (C_T) values were determined, and mRNA expression levels were normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and expressed relative to noninfected controls following the 2^−ΔΔCT method.

TLR reporter cell line assays.

HEK-Blue-hTLR2-hTLR1, HEK-Blue-hTLR2-hTLR6, HEK-Blue-hTLR4, and HEK-Blue-hTLR5 cells were obtained (InvivoGen, San Diego, CA) and are human embryonic kidney cells (HEK293) cotransfected with the human TLR gene and an inducible secreted embryonic alkaline phosphatase (SEAP) reporter gene induced by NF-κB and AP-1 activation. Cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Thermo Fisher Scientific, Waltham, MA), with 50 U/mL penicillin, 50 μg/mL streptomycin, 100 μg/mL Normocin (InvivoGen), and 10% heat-inactivated FBS (Atlanta Biologicals Inc., Flowery Branch, GA) supplemented with 30 μg/mL blasticidin and 100 μg/mL Zeocin for HEK-Blue-hTLR2-hTLR1, HEK-Blue-hTLR2-hTLR6, and HEK-Blue-hTLR5 cells and supplemented with 1× HEK-Blue Selection (InvivoGen) for HEK-Blue-hTLR4 cells. HEK-Blue cells were plated at 0.4 × 10⁶/well on a 24-well plate and left untreated or stimulated with increasing concentrations of the agonist. As controls, HEK-Blue-hTLR2-hTLR1 cells were stimulated with PAM3CSK4 (1 or 10 ng/mL; InvivoGen), HEK-Blue-hTLR2-hTLR6 cells were stimulated with FSL-1 (1 or 10 ng/mL; InvivoGen), and HEK-Blue-hTLR4 cells were stimulated with purified E. coli LPS (1 or 10 ng/mL; InvivoGen), while HEK-Blue-hTLR5 cells were stimulated with purified recombinant flagellin (1 or 10 ng/mL, InvivoGen). Both HEK-Blue reporter cells were also infected with S. sputigena (MOIs of 1, 10, and 100) for 24 h, after which the supernatant was transferred to new wells and soluble SEAP was detected using the HEK-Blue detection reagent (InvivoGen) and measured spectrophotometrically (655 nm) after a 1 h incubation.

Human neutrophil and monocyte isolation.

Blood was drawn from healthy donors and neutrophils were purified using plasma-Percoll gradients in accordance with the guidelines approved by the institutional review board of the University of Louisville (41). The purity of the isolated cell fraction was determined by microscopic evaluation, and cell viability was confirmed by trypan blue exclusion. Peripheral blood monocyte fractions were obtained after the plasma-Percoll gradient was washed twice in Krebs+, counted, and plated into 6-well plates to allow the monocytes to attach to the surface for 2 h at 37°C and 5% CO₂ as our group previously described (42).

Neutrophil and monocyte chemotaxis assays.

Freshly isolated neutrophils or monocytes (4 × 10⁵) were added to the upper chamber of the Transwell inserts contained in 24-well plates (VWR, Corning). For neutrophil chemotaxis, a polycarbonate membrane with a pore size of 3 μm was used, and for monocyte chemotaxis, an 8-μm-pore-size membrane was used. Chemotaxis was initiated by adding 600 μL Krebs-Ringer phosphate buffer (pH 7.2) containing 0.2% dextrose (Krebs) or by adding different chemoattractants into the lower chamber. The chemoattractants used were CXCL1 (10 nM; Sigma) and CCL3 (100 ng/mL; Sigma), along with the bacterial-cell-free supernatants collected from unstimulated, S. sputigena-challenged HIGKs (MOIs of 10, 50, and 100; 4 h). Bacteria were removed from the collected supernatants by passing through a sterile 0.2-μm filter. After 30 or 120 min, the Transwell membranes were stained with a HEMA 3 stain set kit following the manufacturer’s instructions (Thermo Fisher Scientific). Chemotaxis was microscopically assessed by examination of the underside of the membrane. The average number of cells from a total of 10 fields was determined, and data were normalized by the area of membrane circle and field of view as previously described (43).