Correlation of Surface Toll-Like Receptor 9 Expression with IL-17 Production in Neutrophils during Septic Peritonitis in Mice Induced by E. coli

IL-17 is a proinflammatory cytokine produced by various immune cells. Polymorphonuclear neutrophils (PMNs) are the first line of defense in bacterial infection and express surface Toll-like receptor 9 (sTLR9). To study the relationship of sTLR9 and IL-17 in PMNs during bacterial infection, we infected mice with E. coli intraperitoneally to establish a septic peritonitis model for studying the PMNs response in peritoneal cavity. We found that PMNs and some of “giant cells” were massively accumulated in the peritoneal cavity of mice with fatal septic peritonitis induced by E. coli. Kinetically, the CD11b+ PMNs were increased from 20–40% at 18 hours to >80% at 72 hours after infection. After E. coli infection, sTLR9 expression on CD11b+ and CD11b− PMNs and macrophages in the PLCs were increased at early stage and deceased at late stage; IL-17 expression was also increased in CD11b+ PMNs, CD11b− PMNs, macrophages, and CD3+ T cells. Using experiments of in vitro blockage, qRT-PCR and cell sorting, we confirmed that PMNs in the PLCs did increase their IL-17 expression during E. coli infection. Interestingly, sTLR9−CD11b+Ly6G+ PMNs, not sTLR9+CD11b+Ly6G+ PMNs, were found to be able to increase their IL-17 expression. Together, the data may help understand novel roles of PMNs in septic peritonitis.


Introduction
Sepsis affects approximately 700,000 people annually and accounts for about 210,000 deaths per year in the US. Its incidence is rising at rates between 1.5% and 8% per year, despite continuous progress in the development of antimicrobial therapeutics and supportive cares [1]. Most of the sepsis is caused by bacteria and bacteria of abdominal origin contribute to the second major reason for sepsis. This type of sepsis is designated as abdominal sepsis or septic peritonitis. The septic peritonitis is the host's systemic inflammatory response to the bacteria, initiated by the pathogen associated molecule patterns (PAMPs) including lipopolysaccharide and lipid A from Gram-negative bacteria and lipoteichoic acid and peptidoglycan from Gram-positive bacteria [2]. The inflammatory initiation leads to the release of chemokines, such as interleukin-8 (IL-8), and proinflammatory cytokines, such as tumor necrosis factor-(TNF-), IL-1, IL-6, and IL-12, in a massive amount. The cytokines, if not appropriately controlled, may severely impair the functions of vital organs or systems, resulting in death [2]. Data from the CIAOW study (complicated intra-abdominal infections worldwide observational study) showed that the overall mortality rate was 10.5% in the patients with septic peritonitis in a worldwide context [3].
Septic peritonitis is characterized by a massive infiltration of neutrophils into the peritoneum in response to bacterial invasion, where they are activated and act as a first line of defense against the bacteria [4]. Morphologically, the mature neutrophils are unique among the other white blood cells by their lobulated nucleus, which inspired the renaming of them as polymorphonuclear neutrophils (PMNs) [5]. Based on cytokine production, macrophage activation, expression of Toll-like receptor (TLR), and surface antigen expression, murine PMNs have been classified into three types [6]: normal PMN (PMN-N), PMN-I, and PMN-II. PMN-I produces IL-12/CCL3, activates macrophages in a classical way, expresses TLR2/TLR4/TLR5/TLR8, and has a CD49d + CD11b − phenotype. PMN-II produces IL-10/CCL2, activates macrophages in an alternative manner, expresses TLR2/TLR4/ TLR7/TLR9, and possesses a CD49d − CD11b + phenotype. PMN-N rarely produces cytokine/chemokine, displays no effect on macrophage activation, and expresses TLR2/TLR4/ TLR9, with CD49d − CD11b − phenotypes. The PMN-N may convert to PMN-I or PMN-II in response to bacterial infections [6]. Recently, macrophages have been reported to cooperate with neutrophils by promoting extravasation and activation of PMNs, even their clearance of bacteria [7].
Accumulated evidence revealed that Toll-like receptors (TLRs) could play a fundamental role in induction of hyperinflammation and tissue damage in sepsis. TLRs are pattern recognition receptors (PRRs) that sense microbial invasion and initiate innate immune responses. Positioned at the cell surface, TLR4 is essential for sensing lipopolysaccharide of Gram-negative bacteria and TLR2 is involved in the recognition of a large panel of microbial ligands, while TLR5 recognizes flagellin. Endosomal TLR3, TLR7, TLR8, and TLR9 are specialized in the sensing of nucleic acids produced notably during viral infections or during bacterial infections [8]. Conventionally, TLR9 is thought to sense microbial DNA in endolysosomes and not at the cell surface. Recently, it has been found that TLR9 is expressed on the surface of human and mouse PMNs, that the surface TLR9 (sTLR9) senses DNA in PMNs, and that sTLR9 expressing PMNs (sTLR9 + PMNs) have roles in inducing rapid inflammation [9]. Human and mouse PMNs spontaneously express sTLR9, and the sTLR9 expression is upregulated in response to PMN stimulation [10]. In front of TLR9 ligands, sTLR9 + PMNs could be activated, leading to cytokine secretion and CD11b upregulation [10]. Notably, TLR9 stimulation was demonstrated to be detrimental in mice with bacterial sepsis. TLR9 −/− mice exhibited lower serum inflammatory cytokine levels, higher bacterial clearance, and greater survival after experimental peritonitis induced by cecal ligation and puncture (CLP). A single injection of TLR9 antagonist protected the wide type mice, even when administered as late as 12 hours after CLP [11]. It has been proposed that, during infection, sTLR9 + PMNs could be involved in the detrimental effect by releasing massive proinflammatory cytokines, including IL-6 and TNF-, and possibly other cytokines in response to PAMPs [12,13].
Interleukin-17 (IL-17) is a cytokine family that signatures T helper 17 (Th17) cell subset and contains six members (IL-17A to IL-17F), among which IL-17A is considered as one of the major proinflammatory cytokines mediating the innate and adaptive immune responses against bacterial infections [14]. In addition to Th17 cells, T cells, innate lymphoid cells (ILCs), mast cells, PMNs, and macrophages are also IL-17 producing cells [14]. Notably, innate immune cell-derived IL-17 constitutes a major element in the immune response against infectious agents by recruiting PMNs to the sites of infections and by inducing the production of antimicrobial peptides, CXC chemokines, and granulocyte colony stimulating factor (G-CSF) [15,16]. In bacterial infection, PMNs act to produce increased IL-17, which in turn recruits more PMNs to join the fighting against invaded bacteria, resulting in increased production of innate immune cell-derived IL-17 [16,17]. IL-17 was found to work in an IL-23-IL-17 axis which was critical for the survival of the host infected with bacteria [18]. In the axis, tissue-infiltrating PMNs could be the main source of IL-23 [19] which induces production of IL-17 from macrophages [20].
In the present study, we kinetically observed the infiltration of PMNs and expression of sTLR9 and IL-17 on/in the PMNs in the peritoneal lavage cells (PLCs) of mice intraperitoneally infected with various doses of Escherichia coli (E. coli), aiming to find the correlation of the expression of sTLR9 and IL-17 on/in the PMNs during the development of bacterial septic peritonitis. The achieved results could provide a basis for further investigation on the roles of sTLR9 expressing PMNs and IL-17 producing PMNs in bacterium caused septic peritonitis.

E. coli Strain and
Mice. E. coli strain of JM109 was recovered from lyophilized powder by being suspended into LB medium and then cultured on LB agar plate at 37 ∘ C overnight. The single colonies of E. coli on the plate were used as a group of original seed E. coli.
Female ICR mice were purchased from the Experimental Animal Center, Medical College of Norman Bethune, Jilin University, and maintained in microisolator cages under specific pathogen-free conditions. All the mice were used at 6 to 8 weeks of age. The experimental manipulation of mice was undertaken in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals, with the approval of the Scientific Investigation Board of Science & Technology of Jilin Province.

Preparation of E. coli Culture.
To prepare E. coli culture, each single colony of E. coli was picked up and cultured in 5 mL LB medium at 37 ∘ C with shaking at 200 rpm. When the OD value (A 600 ) of the culture reached 1.4, E. coli was harvested from the culture by centrifugation and mixed with 20% glycerol solution. The resultant glycerol E. coli was stored at −20 ∘ C as working E. coli seeds. The working seed E. coli were seeded into LB medium and cultured at 37 ∘ C till the OD value reached 0.7 followed by harvesting the E. coli pellet after centrifugation at 4000 g for 5 min. The E. coli pellet was diluted in a serial tenfold, plated on an eosin methylene blue agar plate, and then cultured at 37 ∘ C for 14 hours. The colonyforming units (CFUs) of the E. coli culture were calculated and determined as approximately 0.8 × 10 8 CFUs/mL. For animal experiment use, the E. coli pellet was washed twice using sterile 0.9% NaCl (saline) and then resuspended in 1.0 mL saline containing 0.4, 0.8, 1.6, and 2.4 × 10 8 CFUs of E. coli, respectively, and was ready to use for infecting mice.

Induction of Bacterial Septic Peritonitis and Preparation of PLCs.
For inducing bacterial septic peritonitis, mice were injected intraperitoneally (i.p.) with 1.0 mL of E. coli preparations containing 0.4, 0.8, 1.6, and 2.4 × 10 8 CFUs/mL in sterile saline, respectively. The saline injection was used as control. The survival of the mice was monitored every 4 hours for 4 days.
To harvest the peritoneal lavage cells (PLCs) from mice either infected with E. coli or injected with saline, the mice ( ≥ 3 per group) were euthanized at defined time points with 50 mg/kg pentobarbital sodium followed by washing peritoneal cavities using 6 mL per mouse of ice-cold phosphatebuffered saline (PBS). The peritoneal lavage fluid was centrifuged at 750 g for 5 minutes at 4 ∘ C for harvesting the PLCs. The PLCs were resuspended in cold PBS for further use.

Cell
Culture and Treatment. PLCs were harvested from the naïve mice and maintained in RPMI 1640 supplemented with 10% (V/V) fetal bovine serum (FBS) (GIBCO) and antibiotics (100 IU of penicillin/mL and 100 IU of streptomycin/mL). In experiments using viable bacteria, antibiotics were not added to the culture medium during the isolation, washing, or subsequent culturing period. No bacterial contamination was observed in PLCs cultured in the absence of antibiotics. Cells were counted and then plated in 24-well cell culture plates (Costar, Cambridge, MA) at an approximate density of 1 × 10 6 cells/well. The PLCs were cocultured with E. coli at 1 × 10 5 or 1 × 10 6 CFUs/well or saline as a vehicle control for 14 hours, and then they were cultured with BFA for another 4 hours, in a 5% CO 2 humidified incubator at 37 ∘ C. The PLCs were collected and stained with FITC-labeled anti-IL-17 mAb, followed by flow cytometry analysis to detect the expression of IL-17.

Cell Counting.
To count the numbers of cells in each peritoneal lavage sample, the samples were spin down for harvesting the cell pellets. The cell pellets were smeared on slides and then stained using hematoxylin-eosin (HE) followed by counting the cell numbers on hemocytometer (Beckman Coulter, Fullerton, CA) and taking photos of the cells.

Flow Cytometry.
For surface staining, the PLCs were stained with fluorescence-conjugated mAbs against CD45, CD11b, and sTLR9, respectively, for 30 minutes at room temperature in the dark followed by washing twice with PBS. For IL-17 intracellular staining, the PLCs were surface stained as described above and then fixed with 4% paraformaldehyde and permeabilized with 0.1% saponin followed by staining with FITC-conjugated anti-IL17A monoclonal antibody. All stained cells were analyzed by flow cytometer FACSCalibur (BD) and CytoFLEX (Beckman Coulter). Live cells were carefully gated by forward and side scattering. Data were analyzed with FlowJo software (FlowJo 7.6.1).

Cell
Sorting. The PLCs were harvested from mice ( ≥ 6 per group) either infected with E. coli or injected with saline. The PLCs were stained with fluorescence-conjugated mAbs against CD14, CD11b, and CD3 for 30 minutes at room temperature in the dark, followed by washing twice with PBS. The stained cells were sorted using BD FACS Aria II by the methods described in Figure 5(g). The neutrophils were sorted by selection for CD3 − CD14 − CD11b + cells, and T cells were sorted by selection for CD3 + cells, according to the manufacturer's recommendations. T cells (defined as CD3 + cells with a purity >83% of living cells) and PMNs (defined as CD3 − CD14 − CD11b + cells with a purity >96.7% of living cells) were obtained.
2.9. qRT-PCR. Total RNA was isolated from the PLCs with Trizol reagent and reverse-transcribed using cDNA Synthesis Kit. Quantitative real-time PCR (qRT-PCR) was performed using two-step SYBR green qPCR assays and the target mRNAs were identified by the specific primers as follows: IL-17A, forward: 5 -AAGGCAGCAGCGATCATCCCT; reverse: 3 -TCTTCATTGCGGTGGAGAGTCC; GAPDH, forward: 5 -ATCACCATCTTCCAGGAGCGA; reverse: 3 -TCTCGTGGTTCACACCCATCA. The data were acquired using the Step One TM real-time PCR system (Applied Biosystems). The procedure of the target mRNA amplification was as follows: 1 cycle at 95 ∘ C (30 seconds) followed by 40 cycles at 95 ∘ C (5 seconds) and 64 ∘ C (31 seconds). Each assay plate included negative controls with no template. The relative amount of gene expression was calculated according to the and Ct is the crossing threshold value returned by the PCR instrument for every gene amplification.

Statistical Analysis.
Comparisons between groups were conducted using analysis of Student's -tests. Survival curves of mice were estimated using the Kaplan-Meier method and compared using the log-rank test. We considered the resulting values of less than 0.05 (95% CI) to be statistically significant. Statistics were analyzed using GraphPad Prism 5.0 for Windows (San Diego, CA).

The Establishment of E. coli Induced Septic Peritonitis in
Mice. To understand how neutrophils contribute to the development of sepsis, we first established a murine model of septic peritonitis induced by Escherichia coli (E. coli). The female ICR mice, 10 in each group, were intraperitoneally injected with 2.4 × 10 8 , 1.6 × 10 8 , 0.8 × 10 8 , and 0.4 × 10 8 colony forming units (CFUs) of E. coli, respectively, and then monitored for their physical conditions and survival. Around 6 hours after the infection, all of the 10 mice received 2.4 × 10 8 CFUs of E. coli, 8 out of the 10 mice received 1.6 × 10 8 CFUs of E. coli, and 5 out of the 10 mice that received 0.8 × 10 8 CFUs of E. coli began to exhibit multiple neurological symptoms, including dispirited behavior, staggered gait, and trembling, but these phenomena were not present in all of the 10 mice that received 0.4 × 10 8 CFUs of E. coli or saline. Around 18 hours after the infection, the infected mice began to die. Around 48 hours, the infected mice underwent a massive death, especially in those that received high infectious dose of E. coli. By 72 hours after the infection, 80%, 60%, and 40% of the mice infected with 2.4, 1.6, and 0.8 × 10 8 CFUs of E. coli, respectively, died and yet 100% of the mice which were infected with 0.4 × 10 8 CFUs of E. coli or received saline survived ( Figure 1). The results indicated that the infection with E. coli at 1.6 × 10 8 CFUs and 2.4 × 10 8 CFUs could induce septic peritonitis in ICR mice. When counting the cells, we found that at 18, 48, and 72 hours, PMNs constituted 50%-80% and macrophages constituted 20%-40% of the PLCs from the mice infected with 4 doses of E. coli, respectively, compared with nearly 30% for PMNs and 60% for macrophages in the PLCs from the mice that received saline. At 18 hours, interestingly, there were about 20% of the giant cells in the PLCs of the mice infected with 1.6 and 2.4 × 10 8 CFUs of E. coli (Figure 2  Noticeably, when the mice died of the infection, we collected the PLCs immediately and observed and found that plenty of bacteria coexisted with the PMNs in the PLCs of the mouse infected with 1.6 × 10 8 CFUs of E. coli for 31 hours (Figure 2(c)). These results indicate that neutrophils can move to infected peritoneal cavity by infiltration and change their morphologies after engulfing bacteria possibly. At 72 hours, the numbers of PLCs were decreased to 0.5 × 10 6 cells/mL of PLF in control mice and decreased to nearly 0.5 × 10 6 cells/mL of PLF in the mice infected with three doses of E. coli (Figure 3(a)). In the PLCs from the infected mice, most of the cells were CD45 + nucleated leukocytes, representing PMNs, macrophages, and lymphocytes, respectively. At 18

The Correlation of Neutrophil Numbers in PLCs with
Mediators   hours, the PMNs, macrophages, and lymphocytes were displayed on the left, middle, and right of the histogram, respectively, sequentially based on their nuclear structure and fluorescence intensity. At 48 and 72 hours, the PMNs were found to be shifted to the right upper quadrant, indicating that the PMNs were with more ovulated nuclei and expressed elevated CD45. As shown in Figure 3( (Figure 3(c)). The results indicated that the peritoneal infection of E. coli induced massive infiltration of CD45 + PMNs. Among the CD45 + PMNs, CD11b + PMNs were increased with the progression of infection. In parallel, the CD11b − PMNs were decreased. At 18 and 48 hours after infection, the ratios of CD11b + PMNs or CD11b − PMNs were correlated with the doses of E. coli negatively or positively.

The Expression of Surface TLR9 (sTLR9) on Neutrophils and Macrophages in PLCs with Development of Septic Peritonitis in Mice
Infected with E. coli. Since primary human and mouse blood neutrophils were reported to be able to express a functional sTLR9 [10], we tried to detect whether the expression of sTLR9 was different on CD11b + PMNs and CD11b − PMNs and correlates with development of septic peritonitis in mice. The PLCs collected from the mice which were infected with E. coli or received saline were gated for detecting the expression of sTLR9 on both CD11b + and CD11b − PMNs (Figure 4(a)). Results showed that, at 18 hours after infection, the ratios of sTLR9 + CD11b + PMNs were increased to 57%, 51%, and 70% of the CD11b + PMNs in the mice infected with 0.8, 1.6, and 2.4 × 10 8 CFUs of E. coli, respectively ( < 0.05), compared to 12% in the mice that received saline control (control mice). At 48 hours, the ratios of sTLR9 + CD11b + PMNs were decreased to 2.3%, 2.3%, ( < 0.05) and 8% in the infected mice, respectively, and less than 13% as in control mice. At 72 hours, the ratios of sTLR9 + CD11b + PMNs were 7%, 5.5%, and 9% in the infected mice, respectively, compared with 18% in control mice (Figure 4(b)). The ratios of sTLR9 + CD11b − PMNs constituted 52%, 53%, and 68.5% of the CD11b − PMNs in the mice infected with 0.8, 1.6, and 2.4 × 10 8 CFUs of E. coli, respectively, for 18 hours ( < 0.05), compared with 3.5% in control mice, sharply decreased to 9%, 6%, and 9.5% in the mice infected with 0.8, 1.6, and 2.4 × 10 8 CFUs of E. coli, respectively, for 48 hours, compared with 8% in control mice, and reached 10%, 6%, and 10% in the mice infected with 0.8, 1.6, and 2.4 × 10 8 CFUs of E. coli, respectively, for 72 hours, compared with 8% in control mice (Figure 4(c)). These results indicated that E. coli could significantly induce the occurrence of both of the sTLR9 + CD11b + PMNs and sTLR9 + CD11b − PMNs in the peritoneal cavity at early stage of infection. Considering the fact that macrophages are the inflammatory cells which overlapped with the PMNs in the inflammatory sites, we also observed sTLR9 expression on macrophages in the PLCs. The results showed that the ratios of sTLR9 + macrophages constituted 5%, 6%, and 6.7% ( < 0.05) of the macrophages in the mice infected with 0.8, 1.6, and 2.4 × 10 8 CFUs of E. coli, respectively, for 18 hours, compared with 3% in control mice, decreased to 0.9%   ( < 0.05), 0.45% ( < 0.05), and 3% in the mice infected with 0.8, 1.6, and 2.4 × 10 8 CFUs of E. coli, respectively, for 48 hours, compared with 5% in control mice, and reached 4.8%, 2% ( < 0.05), and 2.8% in the mice infected with 0.8, 1.6, and 2.4 × 10 8 CFUs of E. coli, respectively, for 72 hours, compared with 5.5% in control mice (Figure 4(d)). These results indicated that, unlike PMNs, the macrophages barely express increased sTLR9 during septic peritonitis.

Expression of Interleukin-17 (IL-17) in Neutrophils in PLCs from Mice with Septic Peritonitis Induced by E. coli.
In recent years, primary human and mouse neutrophils have been found to be able to display autocrine IL-17 activity that probably contributes to the etiology of microbial and inflammatory diseases [17]. To find whether the different subtype neutrophil derived IL-17 was involved in the development of septic peritonitis, we detected IL-17 expression in the CD11b + PMNs and CD11b − PMNs ( Figure 5(a)) of the PLCs collected from the mice which were infected with E. coli or received saline (control mice  Figure 5(b)). Noticeably, the saline injection seemed to stimulate neutrophils to express IL-17. To confirm this, the PLCs were collected from the mice injected with saline at 18, 48, or 72 hours after the injection, respectively, and stained with fluorescence-labeled mAb against CD45, CD11b, and IL-17, followed by detection using flow cytometry. The control mice in this experiment were not injected with saline. We found that saline stimulation tended to induce IL-17 expression (not statistically significant) with a big range of variation individually (data not shown).  Figure 5(c)). The results indicated that both of the CD11b + PMNs and the CD11b − PMNs could not increase IL-17 expression during the development of septic peritonitis in mice induced by E. coli. Also, as shown in Figure 5   the PLCs to express IL-17. To clarify this, we set up an experiment to detect IL-17 mRNA expression in the PLCs from the mice infected with 0.8, 1.6, and 2.4 × 10 8 CFUs of E. coli for 18 hours, respectively, or from the control mice and found that the IL-17 mRNA levels in the mice infected with 3 doses of E. coli were significantly elevated in a dose dependent manner ( Figure 5(f)). Next we sorted CD3 + cells and CD3 − CD14 − CD11b + cells from the PLCs of the mice which were infected with E. coli or received saline for 18 hours, detected their IL-17 mRNA expression by qRT-PCR, and found that IL-17 mRNA levels were significantly increased in both of the CD3 + T cells ( = 0.0006) and the CD3 − CD14 − CD11b + cells ( = 0.0434) in the PLCs from the E. coli infected mice. Comparatively, the IL-17 mRNA levels in the CD3 + T cells were much higher than those in the CD3 − CD14 − CD11b + neutrophils. The result indicated that both of T cells and PMNs were the IL-17 producers in the peritoneal cavity during the development of septic peritonitis ( Figure 5(g)). The result suggests that the PLCs from the infected mice did express IL-17 and IL-17 in the PMNs might be secreted out during the development of septic peritonitis.

Correlation of sTLR9 Expression with IL-17 Production in Neutrophils at Early Stage of E. coli Induced Septic Peritonitis in Mice.
To find the possible correlation between the production of IL-17 and the expression of sTLR9 in/on the PMNs and macrophages in peritoneal cavity of the mice during septic peritonitis, we harvested the PLCs from the mice infected with 0.8, 1.6, and 2.4 × 10 8 CFUs of E. coli, respectively, at early stage of 8 hours after infection and detected their expression of IL-17 and sTLR9. The saline injected mice were as controls. The expression was represented by mean fluorescence intensity (MFI) to emphasize the increased expression of IL-17 and sTLR9 per cell. As shown in Figure 6, IL-17 expression   In parallel, we also tested the expression of sTLR9 on PMNs and macrophages and found that sTLR9 expression was significantly downregulated in the CD11b − PMNs from the infected mice and the lowest downregulation happened in the CD11b − PMNs from the mice infected with 2.4 × 10 8 CFUs of E. coli (Figure 6(d)). In contrast, sTLR9 expression was significantly upregulated in the CD11b + PMNs from the infected mice (Figure 6(e)). Interestingly, the significantly upregulated expression of sTLR9 was observed also in the macrophages from the mice infected with 2.4 × 10 8 CFUs of E. coli (Figure 6(f)). To validate whether other types of cells in the PLCs also expressed IL-17 and sTLR9, the PLCs from mice infected with E. coli for 8 hours were stained with fluorescence-labeled mAbs against Ly6G, CD11b, IL-17, or TLR9 and then analyzed by flow cytometry. The results showed that E. coli infection significantly increased IL-17 expressing CD11b + Ly6G + neutrophils (Figure 6(h)) as well as sTLR9 expressing CD11b + Ly6G + neutrophils ( Figure 6(g)) in a dose dependent manner but could not increase IL-17 expression in the sTLR9 + CD11b + Ly6G + neutrophils ( Figure 6(i)) at the time point. In parallel, the PLCs were stained with fluorescence-labeled mAbs against CD3, CD11b, and IL-17 and analyzed. It was found that both CD3 + T cells (Figure 6(j)) in the PLCs and CD3 − CD11b + PMNs ( Figure 6(k)) increased their IL-17 expression in response to the infection with E. coli for 8 hours. This result indicates that, at early stage, sTLR9 may be mainly expressed on CD11b + PMNs in PLCs of the mice infected with E. coli, and with the infection progression, macrophages may become successors to replace the CD11b + PMNs as the major sTLR9 + cells.

Discussion
In this study, we tried to investigate the correlation of TLR9 expression with IL-17 production in PMNs during septic peritonitis and found that both sTLR9 and IL-7 could be expressed in the PMNs infiltrated into the peritoneal cavity of the mice infected with E. coli. At early stage of the infection, sTLR9 was increasingly expressed in the infiltrated CD11b + PMNs, and IL-17 was increasingly expressed in both of the CD11b − PMNs and CD11b + PMNs. IL-17 expression in CD11b − PMNs was positively correlated with the doses of E. coli. When infected with the highest dose of E. coli (2.4 × 10 8 CFUs), IL-17 was increasingly expressed and sTLR9 was decreasingly expressed in/on the CD11b − PMN. When infected with the lowest dose of E. coli (0.8 × 10 8 CFUs), both IL-17 and sTLR9 were increasingly expressed in CD11b + PMNs. Furthermore, we stained the PLCs with both CD11b mAb and Ly6G mAb and confirmed that the CD11b + Ly6G + PMNs in the PLCs could increase their expression of both IL-17 and sTLR9 in response to E. coli infection. Taken together, with the fact that highest dose of E. coli (2.4 × 10 8 CFUs) was the most deadly and the lowest dose of E. coli was the least deadly to the mice with septic peritonitis, we may deduce that increased expression of both sTLR9 and IL-17 in CD11b + PMNs might benefit the survival of the mice with septic peritonitis and that the decreased expression of sTLR9 and increased expression of IL-17 in CD11b − PMNs may be detrimental to the mice.
In recent years, PMNs have been found to be able to express sTLR9 which engage PAMPs, such as bacterial DNA generated during infection, and damage-associated molecular patterns (DAMPs), such as mitochondrial DNA released from necrotic cells during sterile inflammation. The engagement, if intense, can prime PMNs to release massively produced cytokines [22], leading to sepsis or systemic inflammatory response syndrome (SIRS) [12]. In the present work, we found that the ratios of the sTLR9 + CD11b + PMNs and sTLR9 + CD11b − PMNs were significantly increased in the infiltrated CD11b + PMNs and CD11b − PMNs of the mice infected with E. coli for 18 hours, respectively. At this stage, we observed that a plenty of E. coli coexisted with the PMNs. When observed at 48 hours and 72 hours after infection, the ratios of both of the sTLR9 + CD11b + PMNs and sTLR9 + CD11b − PMNs were sharply decreased to the levels as those in saline control. Accompanying with the decrease, E. coli underwent disappearing. Possibly, the increased TLR9 + CD11b + PMNs and sTLR9 + CD11b − PMNs present a rapid innate immune response of the PMNs to bacterial invasion at the early stage of infection. PMNs armed with sTLR9 were demonstrated to be able to sense extracellular ligands and consequently initiate TLR9 mediated signaling in an intracellular TLR9 independent way. The response could offer a rescue mechanism for PMN activation when pathogen derived TLR9 ligands cannot reach the endosome in the early stage of infection [10]. Practically, the increased expression of sTLR9 on PMNs may be a proinflammatory activation marker [10] and sTLR9 + PMN should benefit antibacterial defense during infection, evidenced by the following: (1) TLR9 agonists, if they exist, could activate sTLR9 + PMN, not TLR9-deficient PMNs, to upregulate CD11b and secrete MIP-2 and IL-8 (CXCL8) [10]; (2) sTLR9 + PMNs are involved in inducing the rapid inflammation which is needed in the initial phase of bacterial infection by secreting antimicrobial peptides and elastases [9]; (3) encountering microbial DNA, sTLR9 signaling is able to activate PMNs, and the activated PMN increases sTLR9 expression [9]. However, it is worthy to note that sTLR9 + PMNs could be culprit for causing more severe pathological consequences. For instance, bacterial DNA or formulated peptides released following sepsis were reported to activate p38 MAP kinase through binding sTLR9 on PMNs, leading to acute lung injury which is characterized by protein-rich pulmonary oedema (swelling) and accumulation of large numbers of PMNs in the lungs [13].
It is well established that PMNs, as the firstly recruited inflammatory cells in infection sites, are IL-17 producing cells. IL-17 secreted from PMNs induces the release of proinflammatory factors from mesenchymal and myeloid cells, recruiting additional PMNs [15]. In addition to PMNs, macrophages located in the epithelial barriers are also important sources of IL-17. In this study, we initially found that the average ratios of the IL17 + CD11b + PMNs and IL17 + macrophages were significantly decreased in the PLCs at 18, 48, and 72 hours in the mice infected with 3 doses of E. coli. Seemingly, the CD11b + PMNs and macrophages, in response to the peritoneal infection, could not produce increased IL-17. However, when checking mRNA expression, we found significantly upregulated IL-17 mRNA in the PLCs from the mice infected with E. coli for 18 hours (Figure 5(e)). The data imply that the infiltrated cells in peritoneal cavity of the mice infected with E. coli