This study was carried out in accordance with a protocol approved by the Institutional Review Board of Chang Gung Memorial Hospital (Document no. 100-4029B). All subjects provided written informed consent in accordance with the Declaration of Helsinki.

Heparinized venous blood was obtained from 5 healthy adult volunteers. PMNs were isolated by dextran sedimentation followed by density gradient separation as described previously (Berends et al., 1994). Neutrophils were suspended in HBSS, counted, and diluted to 1.5 × 10⁷ cells/ml. The purity of each preparation was assessed by 0.25% trypan blue staining followed by a microscopic analysis. The suspensions contained 95–98% PMNs. Neutrophils were incubated at 37°C in a water bath in an Eppendorf tube with Ham's medium for further analyses. In all cases, replicate tests were performed using PMNs from different donors.

KP-M1 (ST23; capsule serotype K1) was isolated from a patient with PLA. An acapsular K. pneumoniae mutant of DT-X was obtained by a subculture of strain DT-S (biotype edwardsii, capsule serotype K1, ST23; Lee et al., 2014). The lack of a capsule in DT-X was confirmed by staining with India ink. Bacteria were routinely cultured at 37°C in LB medium.

KP-M1 and DT-X were grown for 18 h in Trypticase soy broth (TSB, Difco, Lawrence, KS, USA), washed twice with a large volume of saline solution and adjusted to 1.5 × 10⁷ cells/mL in a Ham's medium. Bacteria were then opsonized (30 min, 37°C) in the presence of 5% fresh pooled human serum in Ham's medium and were added to the neutrophil cultures in pooled serum at a multiplicity of infection of 1 bacteria/neutrophil. Then, 0.3 mL of the mixed solution was incubated at 37°C for 30 min to allow infection to occur. This test was repeated five times with neutrophils isolated from five separate volunteers.

Bactericidal activity was measured using a standard assay method (Hampton et al., 1994). After infection of PMNs as described above, samples (50 μL) were drawn at 0, 2, 4, 6, 12, and 24 h post-infection and diluted with 2.45 mL of H₂O (pH 11.0) to lyse the neutrophils and sufficiently disperse the bacteria to permit evaluation by a colony assay. The plate was incubated overnight at 37°C and the number of colonies was counted. All tests were performed in triplicate to assure their reproducibility.

PMNs were infected with K. pneumoniae (as above) or left untreated. At various time points post-infection, the apoptosis of PMNs was analyzed by flow cytometry using the FITC Annexin V Apoptosis Detection Kit (BD Biosciences, Franklin Lakes, NJ, USA) according to the manufacturer's instructions. Briefly, the cells were suspended in annexin V-binding buffer at a concentration of 10⁶ cells/mL. This suspension (100 μL) was stained with annexin V-FITC and propidium iodide. The cells were incubated for 15 min at room temperature. After adding 400 μL of binding buffer to each tube, the cells were analyzed by flow cytometry (FACSCalibur; BD Biosciences) using CellQuest Pro (Version 5.1).

Culture supernatants were collected from PMNs infected with K. pneumoniae (see above) and untreated PMNs at various time points to determine for lactate dehydrogenase (LDH) activity. An LDH release assay was using a Cytotoxicity Detection Kit protocol (Clontech, Mountain View, CA, USA), following the manufacturer's protocol, with a few modifications. In brief, 200 μL of medium from each well was transferred to a new 96-well polystyrene plate and spun at 300 × g for 10 min. Next, 100 μL of supernatant was transferred to a new plate in which 100 μL of reaction mixture was added to each well, followed by incubation at room temperature in the dark for 30 min. The plate was read at 492 and 690 nm using a SpectraMax Plus 384 Plate Reader (Molecular Devices, Sunnyvale, CA, USA). The reference absorbance at 690 nm was subtracted from the absorbance at 492 nm. The readings from the untreated and test wells were subtracted from the reading from control wells containing medium alone. This yielded the corrected reading for the untreated or tests well. The percentage of cytotoxicity was calculated using the following formula:

(corrected reading from test well − corrected reading from untreated well)/

Following 30 min of infection with K. pneumoniae (as described above), PMNs (4.5 × 10⁶ cells) were collected after 0, 2, 4, and 6 h and then re-suspended in 50 μL of RIPA buffer (Cell Signaling Technology, Danvers, MA, USA) supplemented with a protease inhibitor at −80°C overnight and centrifuged at 16,100 × g for 20 min at 4°C to obtain cell lysates. After discarding the cell debris, the supernatant protein concentrations were determined using the Bio-Rad Protein Assay Dye Reagent Concentrate (Hercules, CA, USA). A western blot analysis was performed and a housekeeping protein, COX-IV, was used as a control for equal loading. Samples were re-suspended in a loading buffer (50 mM Tris–HCl pH 6.8, 2% SDS, 0.1% bromophenol blue, 10% glycerol, 100 mM dithiothreitol), and heated for 5 min at 90°C. Denatured proteins (20–70 μg/sample) were separated on a 12% denaturing polyacrylamide gel by SDS-PAGE and transferred to nitrocellulose membranes (Hybond C; Amersham Biosciences, Little Chalfont, UK). Membranes were blocked for 1 h at room temperature with a 5% (w/v) non-fat dry milk solution containing 10 mM Tris–HCl pH 7.5, 140 mM NaCl, and 0.1% Tween 20 (TBS–T) before incubation overnight at 4°C with the primary antibodies anti-Bax, anti-BCL-₂, anti-Mcl, anti-Caspase-3, or anti-COX-IV (all from Cell Signaling), anti-Bcl-xS/L, anti-FLIP (from Santa Cruz Biotechnology, Santa Cruz, CA, USA), anti-Caspase-8, and anti-X-chromosome-linked IAP (from BD Bioscience) diluted 1/100 to 1/1,000 with 1% BSA in TBS–T according to the manufacturers' instructions. After washing, the membranes were incubated for 1 h with a species-appropriate horseradish peroxidase-labeled secondary antibody (Santa Cruz Biotechnology) diluted 1/10,000 with 2.5% non-fat milk in TBS–T, and the labeled proteins were detected using Western Lightning Plus-ECL Detection Kit (Perkin–Elmer, Waltham, MA, USA). The signals were quantified by densitometry.

PMNs were infected with K. pneumoniae (as described above) or left untreated and DNA fragmentation was measured at various time points post-infection. Briefly, PMNs were incubated for 1 h at 37°C with TES lysis buffer [20 mM EDTA, 100 mM Tris, pH 8.0, 0.8% (w/v) SDS] containing 10 mg/mL RNase and then incubated overnight at 56°C with TES lysis buffer containing 20 mg/mL proteinase K for DNA preparation. The DNA was analyzed by electrophoresis on 1.6% agarose gels stained in Novel Green (GeneDireX, Atlanta, GA, USA) containing a TBE buffer for 30 min, following by de-staining in H₂O. DNA ladders were visualized using a UV light source.

PMNs were left untreated, infected with KP-M1 or DT-X (as described above), or treated with 500 ng/mL mouse anti-Fas IgM Ab (human, activating; Millipore, Temecula, CA, USA) in 5% pooled serum at 37°C for 30 min. The pellet was washed with HBSS, and the medium was then replaced with 5% pooled serum, followed by incubation at 37°C. PMNs caspase-3 activity was determined at various time points post-infection using ApoAlert Caspase Colorimetric Assay Kits (Clontech) according to the manufacturer's instructions. In each case, caspase-3 activity was assessed by quantifying the luminescence generated upon the cleavage of a caspase-specific proluminogenic substrate.

PMNs were infected with K. pneumoniae as described above. At various time points post-infection, PMNs were stained with a combination of SYTO-16 (Invitrogen, Carlsbad, CA, USA) and ethidium bromide (MDBio Inc., Taipei, Taiwan). Staining with 0.5 nmol SYTO-16 for 10 min at room temperature was used to visualize viable PMNs. This dye penetrates cell membranes and stains DNA. Consequently, the nuclei of viable cells appear green. Subsequent to SYTO-16 staining, the cells were stained with 0.5 nmol ethidium bromide for 5 min to reveal nonviable PMNs. Images were captured using an Axio Vision Digital Imaging System (Carl Zeiss Micro Imaging, Oberkochen, Germany).

PMNs were infected with K. pneumoniae (see description above), and culture supernatants were collected at various time points post-infection. Supernatants were stored at −80°C until use for cytokine determination. The IL-8, IL-10, and TNF-α concentrations were determined by enzyme-linked immunosorbent assays (ELISA) using commercial kits (R&D Systems, Wiesbaden-Nordenstadt, Germany). In some tests, 1 μg of anti-IL-8 blocking monoclonal antibody (R&D Systems, Minneapolis, MN, USA) or isotype control IgG (Sigma-Aldrich) was added to 20 μL of supernatant to estimate IL-8 depletion.

Data obtained in experiments that included a control group and one experimental group were analyzed using unpaired Student's t-tests. Data obtained from studies containing multiple experimental groups were analyzed by one-way analysis of variance (ANOVA) followed by a Tukey's post-hoc tests. Comparative data obtained in the time course study are presented as normalized values with respect to the initial time of 1.0. Data were considered statistically significant when two-sided p-values were < 0.05.

The viable bacterial counts (log₁₀ cfu/mL) after neutrophils infected with K. pneumoniae for 30 min were 6.51 ± 0.43 for KP-M1 and 4.19 ± 0.37 for DT-X (Figure 1). The PMNs bactericidal activity against KP-M1 was significantly poor compared to the PMNs bactericidal activity against DT-X. After infection of PMNs, we withdraw the samples at 2, 4, 6, 12, 24 h post-infection to evaluate the colony counts. The viable counts of KP-M1 stabilized and increased after 6 h post-infection, whereas the viable counts of DT-X declined sharply to below the limit of detection after 12 h post-infection (Figure 1). It showed impaired KP-M1 clearance by PMNs.

LDH release assay showed increasing cell cytotoxicity on untreated PMNs in a time-dependent fashion (Figure 2A), results consistent with the induction of constitutive apoptosis in human PMNs over time (Figure 2B). In contrast, cell cytotoxicity on PMNs infected with KP-M1 increased more slowly than that of untreated PMNs or PMNs infected with DT-X (6 h post-infection; Figure 2A).

According to the flow cytometric diagram of annexin V-fluoresce in isothiocyanate (FITC)/ propidium iodide (PI) staining, this double-staining approach allows three subgroups of cells to be distinguished: viable, non-apoptotic cells are annexin V-negative and PI-negative (lower left quadrant); early apoptotic cells with intact membranes are annexin V-positive and PI-negative (lower right quadrant); and late apoptotic cells with compromised plasma membranes are annexin V-positive and PI-positive (upper right quadrant). Representative Fluorescence correlation spectroscopy (FCS) plots are shown in Figure 2B, and pooled data from 12 independent tests are also shown in Figure 2B. Our data demonstrate that PMNs infected with KP-M1 displayed a slow increase of PMNs with annexin V labeling but not more PI uptake than untreated PMNs, which are feature characteristics of delayed PMN apoptosis. PI staining of KP-M1-infected PMNs population was also increasing more slowly than untreated PMNs or DTX-infected PMNs, confirming that infection with KP-M1 maintained plasma membrane integrity of PMNs as indicated by the cell cytotoxicity data shown above.

Using Western blot techniques (Figure 3A) and quantified densitometry (Figures 3B–E), we followed the time course of apoptosis processing in untreated PMNs and KP-M1-infected PMNs. We found untreated PMNs were undergoing constitutive apoptosis; traces of the processing of procaspase-3 to its mature form were detected at 6 h post-infection. There is evidence that the Bax-to-Bcl-2 ratio appears to control the relative sensitivity or resistance of many cell types to apoptotic stimuli (Salakou et al., 2007). In comparison with the immunoblot band intensity of untreated PMNs, increasing amounts of Mcl-1 (Figure 3B) and decreasing expression ratio of Bax/Bcl-2 (Figure 3C), and less cleaved caspase-3/ procaspase-3 were detected in PMNs infected with KP-M1 in a time-dependent manner (Figure 3E). The FADD-like IL-1ß converting enzyme inhibitory protein (FLIP) is the main anti-apoptotic mechanism in the extrinsic pathway (Hotchkiss et al., 2009). It prevents activation of caspase-8 following ligation of Fas and the TNF-related apoptosis inducing lig and receptors and subsequent cleavage of effectors caspase. Another group comprises the family of inhibitors of apoptosis (IAPs) protein (Berends et al., 1994). We found no difference of FLIP and IAPs expression between untreated PMNs and KP-M1-infected PMNs (data not shown). As expected, there was no statistically significant difference in the ratio of cleaved caspase-8/procaspase-8 between untreated PMNs and KP-M1-infected PMNs at all time points post-infection (Figure 3D). These data demonstrate that, relative to untreated PMNs, activation of anti-apoptotic protein of Bcl-2, Mcl-1 was found in KP-M1-infected PMNs, suggesting that this organism acts on the Bcl-2 family protein in the apoptotic cascade to curtail activation of executioner caspase-3 and extend PMN lifespan. During the late stages of apoptosis, caspase-activated DNases cleaved nuclear DNA. The DNA fragmentation assay demonstrates that relative to untreated PMNs, DNA fragmentation of KP-M1 infected PMNs was significantly delayed at all time points examined between 12 and 24 h post-infection (see in Supplementary Material).

Anti-Fas Ab promoted caspase-3 activation in PMNs was demonstrated as previously reported (Faouzi et al., 2001). Corresponding to the prompted increase in caspase-3 activity in PMNs infected with DT-X, PMNs treated by anti-Fas Ab, or untreated PMNs, KP-M1-infected PMNs activated less caspase-3 activity (p < 0.01 at 2, 4, and 6 h post-infection; Figure 4).

A two-color immune-fluorescent staining (live/dead staining) was used to investigate PMNs survival after ingesting K. pneumoniae strains. After staining with SYTO 16 (green) and ethidium bromide (red), viable cells showed green staining, whereas dead cells with compromised membranes stained red. This staining demonstrated that death of PMNs increased in a time-dependent manner (Figure 5). The results showed that 24 h post-infection, more than 40% PMNs infected with KP-M1 were viable (green staining; Figure 5). In contrast, less than 10% untreated PMNs or those infected with KP-DT-X were viable (green staining; Figure 5).

Chemokine production by PMNs is thought to affect the inflammatory process by recruiting various activated leukocyte populations. As the primary target cells of IL-8 are PMNs, the production of this chemokine by inflammatory PMNs appears to serve as an amplification loop, attracting more PMNs to the site of inflammation. IL-8 was measured in the supernatants of PMN cultures with or without infection with KP-M1 or DT-X at various time points post-infection. It showed KP-M1 or DT-X infected PMNs produced IL-8 in a time-dependent fashion and KP-M1 induces higher levels of IL-8 than DT-X (Figure 6A). KP-M1 infected PMNs did not induce the production of IL-10 and TNF-_α (data not shown). IL-8 has been reported to inhibit the constitutive apoptosis of PMNs (Dunican et al., 2000). To investigate whether IL-8 was responsible for anti-apoptotic effect of the supernatants, IL-8 was depleted from the supernatants taken from PMNs-KP-M1 co-cultures at 6 h after infection. Using immunoprecipitation with anti-IL-8 or isotype control IgG, the IL-8 concentration in the supernatants was reduced from 1398 ± 101.8 to 22.4 ± 2.2 μg/mL (n = 5; Figure 6B). Co-incubation of these samples with PMNs demonstrated that depletion of IL-8 from the supernatant restored the KP-M1 infected PMNs constitutive apoptosis (+Annexin V/-PI flow cytometry) from 36.4 ± 3.3 to 52.0 ± 1.4% (p < 0.01; Figure 6C).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.