Gram-Positive Bacteria Cell Wall Peptidoglycan Polymers Activate Human Dendritic Cells to Produce IL-23 and IL-1β and Promote TH17 Cell Differentiation

Gram-positive bacterial infections are a major cause of organ failure and mortality in sepsis. Cell wall peptidoglycan (PGN) is shed during bacterial replication, and Bacillus anthracis PGN promotes a sepsis-like pathology in baboons. Herein, we determined the ability of polymeric Bacillus anthracis PGN free from TLR ligands to shape human dendritic cell (DC) responses that are important for the initiation of T cell immunity. Monocyte-derived DCs from healthy donors were incubated with PGN polymers isolated from Bacillus anthracis and Staphylococcus aureus. PGN activated the human DCs, as judged by the increased expression of surface HLA-DR, CD83, the T cell costimulatory molecules CD40 and CD86, and the chemokine receptor CCR7. PGN elicited the DC production of IL-23, IL-6, and IL-1β but not IL-12p70. The PGN-stimulated DCs induced the differentiation of naïve allogeneic CD4+ T cells into T helper (TH) cells producing IL-17 and IL-21. Notably, the DCs from a subset of donors did not produce significant levels of IL-23 and IL-1β upon PGN stimulation, suggesting that common polymorphisms in immune response genes regulate the PGN response. In sum, purified PGN is a highly stimulatory cell wall component that activates human DCs to secrete proinflammatory cytokines and promote the differentiation of TH17 cells that are important for neutrophil recruitment in extracellular bacterial infections.


Introduction
Gram-positive bacteria such as Staphylococcus aureus comprise the majority of antibioticresistant strains encountered in U.S. hospitals and cause most human skin and soft tissue infections, which may be invasive and life-threatening [1]. Indeed, Gram-positive bacterial infections are present in nearly half of the sepsis patients in the U.S. and are a major cause of organ failure and mortality in sepsis [2]. Although rare, respiratory infection with the Grampositive Bacillus anthracis leads to extreme bacteremia and signs of sepsis [3]. Cutaneous anthrax infections are more common in rural regions where Bacillus anthracis is endemic in the soil [4]. Previous studies showed that highly purified cell wall peptidoglycan (PGN) polymers isolated from Gram-positive bacteria, including S. aureus and B. anthracis, are potent inducers of pro-inflammatory cytokines in human monocytes [5][6][7]. PGN is shed during bacterial replication and is present in vegetative bacteria. Therefore, circulating polymeric PGN may play a significant role as a pro-inflammatory agent that can induce innate immune pathways in the DCs and macrophages during either cutaneous or systemic infections.
Resources, NIAID, NIH. Briefly, the B. anthracis and S. aureus vegetative bacteria were collected after overnight culture and boiled in 8% SDS in water. The crude cell wall was digested with DNase and RNase, followed by digestion with proteinase K. Hydrofluoric acid was used to remove traces of teichoic acids and was unlikely to alter the sugar acetylation, as most of the PGN in B. anthracis is naturally de-acetylated [24]. The PGN purity and concentration were determined by amino acid analysis. As reported, the B. anthracis polymeric macromolecules had a median length of 0.268 µM and width of 0.205 µM [16]. The chemical composition of the muropeptides after PGN digestion was as reported: B. anthracis was identified as the DAP-type PGN that activates NOD1 and NOD2, and S. aureus was identified as the Lys-type PGN that can activate NOD2 [7,14]. Prior work showed that this purification protocol led B. anthracis PGN to be devoid of TLR2 ligands, while the S. aureus PGN contained contaminating lipopeptides with TLR2 activity when assayed on murine macrophages [14]. To generate HKB, bacteria were incubated for 1 h in a 70 • C water bath, followed by plating on solid medium to confirm that the cells were killed. Both the HKB and PGN were washed with endotoxin-free water and sonicated at 4 W for 10 min to yield a more uniform suspension prior to their addition to the cells. The amounts of HKB added to the cells, as reported in the figures, were based on calculations of the PGN equivalents in the whole bacteria. Typically, 10 µg/mL PGN corresponded to 10 7 cfu/mL HKB [25].

Generation of Monocyte-Derived Dendritic Cells
Heparinized peripheral blood was obtained from healthy volunteers (male and female genders) with written informed consent, according to a protocol approved by the OMRF Institutional Review Board. Leukocyte buffy coats provided by anonymous donors were purchased from the Oklahoma Blood Institute. PBMC were isolated using Lymphocyte Separation Medium gradients (Mediatech Inc., Manassas, VA, USA). The CD14 + monocytes were isolated by negative selection using an EasySep human monocyte enrichment kit (Stem Cell Technologies, Vancouver, BC, Canada). The monocytes were cultured at 10 6 /mL in RPMI, including 10% FCS with 30 ng/mL GM-CSF and 20 ng/mL IL-4 (recombinant cytokines from Peprotech, Rocky Hill, NJ, USA), for 6 days to promote DC differentiation. The differentiated DCs were CD14 -CD11c + CD209 + HLA-DR + .

Assessment of DC Activation
On day 6 after differentiation was initiated, the DCs were harvested, plated at 5 × 10 5 /mL and left unstimulated or stimulated with purified and sonicated PGN (1-100 µg/mL), HKB (1-10 µg/mL PGN equivalents) or LPS (100 ng/mL) + IFNγ (2000 IU/mL) in the presence of 5% pooled human AB serum (Innovative Research Inc). In the preliminary experiments (not shown), we observed reduced DC responses when the human serum was omitted, consistent with our prior work showing the monocyte FcγR-mediated uptake of PGN bound to the anti-PGN antibodies present in human serum [15]. Therefore, pooled human serum that was shown to contain antibodies that bind both sources of PGN [15] was included in the experiments discussed here. After 18 h, the DCs were assessed for changes in cell surface markers using flow cytometry. For the measurement of the secreted cytokines at 24 h after stimulation, DC supernatants were collected from the duplicate wells, each containing 50,000 DCs.

Allogeneic T Cell Assays
Naïve CD4 + CD45RA + CD45RO -T cells were isolated using an EasySep human naïve T cell kit (Stem Cell Technologies) with a purity of 95%. The T cells (40,000 per well) were incubated with allogeneic DCs (3000 per well) in triplicate on round-bottomed 96-well plates for 7 days. The T cells remained viable and increased in number over the 7 days, as judged by the observation of the culture wells with an inverted microscope. On day 7, the T cells were stimulated for 16 h with 10 ng/mL phorbol 12,13-dibutyrate (PDBU) and 200 ng/mL ionomycin, after which the supernatants were collected for the measurement of the secreted cytokines.

Statistics
The statistical analyses were performed using Prism GraphPad software and are indicated in the figure legends. The data involving the surface markers on the multiple-donor DCs were analyzed using repeated measure one-way ANOVAs, followed by multiple comparison tests. The data involving the cytokine measurements of the responder individuals were log-transformed prior to Friedman's ANOVA analyses and Dunn's multiple comparison tests. The significance of the differences in the cytokine values obtained from responders and non-responders for each stimulus were evaluated using Mann-Whitney tests.

Peptidoglycan Polymers Isolated from Gram-Positive B. anthracis Activate Human Dendritic Cells
We generated a preparation of highly purified PGN polymers from B. anthracis vegetative bacteria (∆Sterne strain). This PGN preparation lacks the TLR-stimulating molecules (teichoic and lipoteichoic acid, palmitoylated proteins, polysaccharides, nucleic acids) present in native bacteria. The preparation also lacks LPS, which may be introduced during PGN isolation procedures. Notably, the purified B. anthracis PGN polymers do not activate NF-κB signaling in TLR2-transfected HEK293 cells, nor do they stimulate murine TLR2 + macrophages, indicating the absence of a TLR2 ligand [14].
The exposure of human-monocyte-derived DCs to B. anthracis PGN (0.1-100 µg/mL) resulted in dose-dependent DC activation ( Figure 1A-C). The PGN-exposed DCs increased the surface display of CD83, an activation-induced molecule that fosters the stable surface expression of MHCII and CD86 ( Figure 1A), the T cell costimulatory molecule CD40 ( Figure 1B) and the chemokine receptor CCR7 that directs migration to the lymph nodes ( Figure 1C). PGN at 10 µg/mL induced a maximal response, without reducing the viability, and this concentration was used in our subsequent experiments. The DC activation elicited by the PGN was comparable to that elicited by the B. anthracis heat-killed bacteria (HKB) present at 10 µg/mL as a PGN equivalent ( Figure 1A-C). PGN was not as potent as E. coli LPS; the response to 10 µg/mL PGN was comparable to 0.1 µg/mL LPS ( Figure 1A-C).
The DCs from all the human donors tested were capable of responding to B. anthracis PGN and HKB by increasing the surface expression of CD40, CD83, CD86 and HLA-DR ( Figure 1E-H). This response was comparable to that triggered by LPS/IFNγ. Taken together, these data show that human DCs respond to the purified PGN from Gram-positive bacteria by increasing the surface expression of MHC class II and costimulatory molecules nodes ( Figure 1C). PGN at 10 µg/mL induced a maximal response, without reducing the viability, and this concentration was used in our subsequent experiments. The DC activation elicited by the PGN was comparable to that elicited by the B. anthracis heat-killed bacteria (HKB) present at 10 µg/mL as a PGN equivalent ( Figure 1A-C). PGN was not as potent as E. coli LPS; the response to 10 µg/mL PGN was comparable to 0.1 µg/mL LPS ( Figure 1A-C). The DCs were incubated with B. anthracis (Ba) PGN (10 µg/mL, circles) and HKB (10 µg/mL, squares) or LPS/IFNγ (triangles). For each stimulus, the symbols represent individual donors (n = 6-8), and the mean and SEM are indicated. The significance of the differences was evaluated using a repeated measure one-way ANOVA followed by a Dunnett's multiple comparison test to compare the mean of the unstimulated control with that of each stimulated condition. The p-values for each stimulated DC response, relative to the unstimulated DCs, are indicated by **, p < 0.01; ***, p < 0.001; ****p < 0.0001. HLA-DR on the DCs exposed to the stimuli indicated on the x-axis relative to the unstimulated DCs, for which the values were set to 1. The DCs were incubated with B. anthracis (Ba) PGN (10 µg/mL, circles) and HKB (10 µg/mL, squares) or LPS/IFNγ (triangles). For each stimulus, the symbols represent individual donors (n = 6-8), and the mean and SEM are indicated. The significance of the differences was evaluated using a repeated measure one-way ANOVA followed by a Dunnett's multiple comparison test to compare the mean of the unstimulated control with that of each stimulated condition. The p-values for each stimulated DC response, relative to the unstimulated DCs, are indicated by **, p < 0.01; ***, p < 0.001; ****p < 0.0001.

B. anthracis PGN-Activated DCs Produce IL-23 and IL-1β but Not IL-12p70
To determine the cytokines produced by the PGN-stimulated DCs, we measured the levels of IL-12p70, IL-23, IL-6 and IL-1β in the DC culture supernatants collected 24 h after stimulation with B. anthracis PGN (Figure 2). In our analyses of the DCs generated by the 14 donors, we found a striking individual variation in the magnitude of the cytokine response to PGN. The B. anthracis PGN-stimulated DCs from~60% of the donors (9/14) produced significant amounts of IL-23 (labeled as R, responders, based on the IL-23 levels, which were greater than the mean + 3 SD of the negligible IL-23 levels produced by the unstimulated DCs) (Figure 2A). In contrast, the B. anthracis PGN-stimulated DCs from the other donors (5/14) did not produce any detectable IL-23 compared to the unstimulated DCs (labeled as NR, non-responders) (Figure 2A). Similarly, while the R group produced significant amounts of IL-1β, the NR group did not produce IL-1β in levels above the unstimulated DCs ( Figure 2B). The PGN-exposed responder DCs did not produce IL-12p70 ( Figure 2C). This was in notable contrast with the DCs stimulated by LPS/IFNγ, which produced both IL-23 and IL-12p70 (Figure 2A,C). PGN also elicited the DC production of IL-6 ( Figure 2D). For the individuals not producing IL-23, the amount of PGN-induced IL-6 was~10-fold lower than that of the IL-23-producing responders, suggesting that the overall magnitude of the response to PGN was much lower. B. anthracis HKB also induced the production of IL-23, IL-6 and IL-1β and significantly less IL-12p70-than LPS/IFNγ-stimulated DCs (Figure 2A-D), and again, the donors fell into the same groups of IL-23-producing responders or non-responders, as defined for PGN.
To determine the cytokines produced by the PGN-stimulated DCs, we measured t levels of IL-12p70, IL-23, IL-6 and IL-1β in the DC culture supernatants collected 24 h aft stimulation with B. anthracis PGN (Figure 2). In our analyses of the DCs generated by t 14 donors, we found a striking individual variation in the magnitude of the cytokine r sponse to PGN. The B. anthracis PGN-stimulated DCs from ~60% of the donors (9/14) pr duced significant amounts of IL-23 (labeled as R, responders, based on the IL-23 leve which were greater than the mean + 3 SD of the negligible IL-23 levels produced by t unstimulated DCs) (Figure 2A). In contrast, the B. anthracis PGN-stimulated DCs from t other donors (5/14) did not produce any detectable IL-23 compared to the unstimulat DCs (labeled as NR, non-responders) (Figure 2A). Similarly, while the R group produc significant amounts of IL-1β, the NR group did not produce IL-1β in levels above the u stimulated DCs ( Figure 2B). The PGN-exposed responder DCs did not produce IL-12p ( Figure 2C). This was in notable contrast with the DCs stimulated by LPS/IFNγ, whi produced both IL-23 and IL-12p70 (Figure 2A,C). PGN also elicited the DC production IL-6 ( Figure 2D). For the individuals not producing IL-23, the amount of PGN-induced I 6 was ~10-fold lower than that of the IL-23-producing responders, suggesting that t overall magnitude of the response to PGN was much lower. B. anthracis HKB also induc the production of IL-23, IL-6 and IL-1β and significantly less IL-12p70-than LPS/IFN stimulated DCs (Figure 2A-D), and again, the donors fell into the same groups of IL-2 producing responders or non-responders, as defined for PGN.  The NR were defined as those whose Ba PGN-stimulated DCs produced IL-23 in amounts less than the mean + 3 SD (0.464 + 2.32 pg/mL) of their unstimulated DCs. The mean + SD of the 14 unstimulated samples (all <2 pg/mL) was 0.464 + 0.775; the mean + SD of the 5 NR samples was 0.46 + 0.862 (all <2 pg/mL); and the mean + SD of the 9 R samples was 2642 + 3098 (range 22-9405 pg/mL). Symbols represent the averaged values obtained by assaying duplicate wells for each stimulus of the DCs from individual donors (n = 14). The significance of the differences between the responders (excluding the Ba PGN-induced IL-23 measurement used to define the R and NR groups) was evaluated using a Friedman's ANOVA of log-transformed data, followed by Dunn's multiple comparison test to compare the mean of the unstimulated control with each stimulated condition. The significance of the differences between the responders and non-responders for each stimulus (excluding the Ba PGN-induced IL-23 measurement used to define the R and NR groups) was evaluated using a Mann-Whitney test (p-values are indicated below the x-axis). The p-values are indicated by *, p < 0.05; **, p < 0.01; ***, p < 0.001; **** p < 0.0001. ND, not determined.
Thus, PGN isolated from B. anthracis elicited the DC production of IL-23, IL-6 and IL-1β but not IL-12. Robust DC cytokine production in response to PGN occurred in~60% of the donors studied, indicating significant donor variation in the ability to respond to PGN. The response to LPS/IFNγ was also decreased among the PGN non-responders. The DCs from all the donors could respond to PGN by increasing the surface expression of the costimulatory molecules CD86, CD40 and CD83 (Figure 1), and no correlation was found between the magnitude of the marker surface expression and the ability to produce IL-23.

B. anthracis PGN-Activated DCs Induce Naïve Allogeneic CD4 + T Cells to Produce IL-17
The complement of the cytokines produced by activated DCs determines their ability to polarize particular subsets of T H cells. We therefore hypothesized that PGN-stimulated DCs secreting IL-23, IL-6 and IL-1β would induce the polarization of naïve T cells into T H 17 cells. To test this, we incubated the stimulated DCs with allogeneic naïve CD4 + CD45RA + T cells for 7 days. To determine whether the CD4 + T cells had become polarized, the T cells were stimulated with PDBU/Ionomycin, and the levels of the secreted cytokines (IFNγ, IL-17, IL-21) were measured. Here, we report on the T cell responses of three donors whose DCs produced IL-23 in response to PGN. The DCs exposed to B. anthracis PGN and HKB induced the naïve T cells to differentiate into T H cells producing IL-17 and IL-21 ( Figure 3A,B). The T cells incubated with unstimulated DCs produced IFNγ, and this was unchanged by the stimulation of the DCs ( Figure 3C). Unstimulated DCs may produce other factors that induce IFNγ production by naïve T cells, as reported in [21], and this was not altered by PGN or HKB stimulation. In the assays performed on donor DCs that did not produce IL-23, we identified neither the T cell production of IL-17 nor changes in baseline IFNγ production (data not shown). Thus, the DCs incubated with B. anthracis PGN or HKB induced the naïve CD4 + T cells to differentiate into T H cells producing IL-17 and IL-21.

S. aureus PGN Stimulates Human DC Production of IL-23 and IL-1β, Resulting in their Ability to Promote TH17 Differentiation
To determine whether other PGN archetypes can similarly stimulate DCs, we tested PGN macromolecules isolated from S. aureus (strain MN8), which harbor a lysine-containing stem peptide and contain lipopeptide anchors with TLR2 immunostimulatory activity [14,27]. S. aureus PGN and HKB activated DCs from multiple donors to increase surface  (n = 3). The significance of the differences was evaluated using a one-way ANOVA, followed by a multiple comparison test. The p-values for each T cell response to the stimulated DCs, relative to the T cell response to the unstimulated DCs, are indicated by *, p < 0.05.

S. aureus PGN Stimulates Human DC Production of IL-23 and IL-1β, Resulting in their Ability to Promote T H 17 Differentiation
To determine whether other PGN archetypes can similarly stimulate DCs, we tested PGN macromolecules isolated from S. aureus (strain MN8), which harbor a lysine-containing stem peptide and contain lipopeptide anchors with TLR2 immunostimulatory activity [14,27]. S. aureus PGN and HKB activated DCs from multiple donors to increase surface expression of CD83, CD40, CD86 and HLA-DR ( Figure 4A-D). S. aureus PGN induced DC production of IL-23, IL-6 and IL-1β but not IL-12p70 ( Figure 4E-H). We used the definition of the responder and non-responder groups in Figure 2A for B. anthracis PGN. The S. aureus PGNstimulated DCs from individuals in the responder group produced IL-23 in amounts greater than the unstimulated DCs. The IL-23 responders also produced IL-1β and IL-6. The nonresponder individuals showed low responses to both S. aureus and B. anthracis PGN. Unlike the B. anthracis HKB, the S. aureus HKB induced a substantial production of both IL-23 and IL-12p70 ( Figure 4E,G).  The significance of the differences between the responders was evaluated using a Friedman's . The significance of the differences was evaluated using a one-way ANOVA followed by a multiple comparison test to compare the mean of the response to the unstimulated DCs with the response to the stimulated DCs. In all the panels, the p-values are indicated by *, p < 0.05; **, p < 0.01; ***, p < 0.001. The DCs exposed to S. aureus PGN induced the naïve CD4 + CD45RA + T cells to produce IL-17 and IL-21 ( Figure 4I,J). A similar trend was observed for S. aureus HKB ( Figure 4I,J). The T cells incubated with unstimulated DCs produced IFNγ, and this was unchanged by the stimulation of the DCs, even when they were stimulated by S. aureus HKB, which induces IL-12p70 production ( Figure 4K).
In sum, despite the differences in the structure of the PGN stem peptides, both S. aureus and B. anthracis PGN and HKB stimulated the DCs to produce IL-23 and IL-1β, which led to their ability to induce the naive CD4 + T cells to differentiate into T H cells producing IL-17 and IL-21.

Discussion
Humans are typically exposed to Bacillus anthracis spores by the cutaneous or pulmonary route. The spores germinate to vegetative bacteria, which then replicate rapidly and disseminate throughout the host, often resulting in a high density (10 8 CFU/mL) in the blood [28]. In peripheral tissues or lymph nodes, resident DCs, as well as DCs derived from infiltrating inflammatory monocytes, are exposed to vegetative bacteria germinating from spores. The B. anthracis toxins, the lethal toxin and edema toxin, inhibit DC activation and immune responses [29,30]. However, the infection of animals with the ∆Sterne strain of B. anthracis (lacking toxins and capsule) leads to sepsis [31], indicating that components of the vegetative bacteria can induce significant inflammation. Indeed, the infusion of B. anthracis PGN alone induces a sepsis-like pathophysiology, including disseminated intravascular coagulation and multiple organ failure in baboons [32]. PGN is a strong inducer of inflammation and is shed during bacterial replication and accumulates to reach significant levels in the blood and tissue. PGN-derived muropeptides may also be released during growth and cell wall recycling [19]. In earlier studies, we showed that B. anthracis PGN polymers activate human monocytes and neutrophils to produce TNFα and IL-8 [6] and induce human platelets to aggregate and express prothrombinase activity [33].
Herein, we interrogated the innate responses initiated by human DCs upon exposure to B. anthracis and S. aureus PGN polymers and the parental heat-killed bacteria. The highly purified PGN polymers activated the DCs, leading to the upregulation of the T cell costimulatory molecules CD40 and CD86, as well as CCR7, HLA-DR and CD83. The PGN-activated DCs produced IL-23, IL-1β and IL-6 but not IL-12p70. The PGN-exposed DCs producing IL-23 and IL-1β promoted the differentiation of naïve CD4 + T cells into T H cells producing IL-17 and IL-21. In contrast, prior work showing that MDP alone is not stimulatory for human DCs, and that costimulation with a TLR2 ligand is necessary to induce IL-23 production [21,34]. This suggests that the free muropeptides generated during bacterial growth will not stimulate DCs independently. Here, we showed that B. anthracis PGN polymers lacking TLR2 agonist activity stimulated DCs, indicating that TLR2 priming is not necessary for PGN to induce DC responses capable of directing naive T cell differentiation into T H 17 cells.
In addition to high levels of circulating PGN, the DC-mediated T cell response is driven by the complement of the pathogen-associated molecular patterns (PAMPs) present in the entire bacteria. Reports show that additive signaling through the MDP/NOD2 axis and either TLR4 or TLR2 increases the DC production of IL-12 and T cell priming capacity [35][36][37]. Thus, the DC production of IL-12 elicited by the heat-killed bacteria in our study is likely due other PAMPs present in the whole bacterium. Similarly, NOD activation by free muropeptides may be optimized by concurrent PAMP signaling in vivo. Interestingly, the donors with poor responses to the purified PGN also showed low responses to the heat-killed bacteria, suggesting that the DC response to heat-killed bacteria is dominated by PGN, despite the presence of other PAMP ligands for innate sensors, such as TLRs.
Our work is consistent with reports showing that human DCs exposed to germinating B. anthracis spores produce IL-23 and induce T H 17 activation [38]. Similarly, our findings complement reports showing that commercially available S. aureus PGN promotes IL-23 production, leading to T H 17 differentiation [22,23]. Taken together, our data and those of other authors indicate that PGN-exposed DCs promote the differentiation of T H 17 cells that are important for neutrophil recruitment in Gram-positive bacterial infections. Interestingly, neutrophil infiltrates characterize cutaneous [39] but not inhalational anthrax [40], and Th17 responses have been reported in natural cutaneous anthrax infections in humans [41]. In addition, the DC production of pro-inflammatory IL-6 and IL-1β upon PGN stimulation could exacerbate the inflammatory sepsis pathology in instances when the bacteria and PGN reach high levels in the blood or tissues.
Monocytes and DCs generated by healthy human donors exhibit significant variability in the magnitude of cytokine production and innate immune signaling in response to the stimulation of pattern recognition receptors [42][43][44]. Common polymorphisms in genes regulating innate immune responses (e.g., IRF5, NOD2, CARD9) are present in populations with significant allele frequency. These common genetic variants regulate the extent of gene expression during pathogen-sensing innate immune responses [42][43][44]. For example, common IRF5 polymorphisms contribute to the individual variability in the magnitude of cytokine production induced by the NOD2 and TLR ligands in human monocyte-derived DCs, and IRF5 alleles associated with autoimmunity lead to increased cytokine secretion [42]. DCs generated by Crohn's Disease patients with homozygous NOD2 mutations have a reduced capacity to produce IL-23 and induce T H 17 when primed with TLR2 ligands in addition to MDP [21]. In our study, the individual donors with low responses to PGN isolated from B. anthracis or S. aureus may harbor one or more alleles of the innate immune genes that are associated with the reduced production of cytokines. Individual donors may also harbor polymorphic alleles that impact the pathways of PGN internalization [16]. The individual variation in cytokine synthesis, but not costimulatory molecule upregulation, documented in our study may also indicate that a higher threshold of stimulation is required for signaling pathways that are important for cytokine gene synthesis. Variation in the magnitude of the innate responses of myeloid cells to PGN likely governs the course of immunity during Gram-positive bacterial infections.

Institutional Review Board Statement:
The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Review Board of the Oklahoma Medical Research Foundation (protocol number 11-52, 12/7/2011).
Informed Consent Statement: Written informed consent was obtained from all subjects involved in the study.