ABSTRACT
Alcohol impairs the host immune system, rendering the host more vulnerable to infection. Therefore, alcoholics are at increased risk of acquiring serious bacterial infections caused by Streptococcus pneumoniae, including pneumonia. Nevertheless, how alcohol affects pneumococcal virulence remains unclear. Here, we showed that the S. pneumoniae type 2 D39 strain is ethanol tolerant and that alcohol upregulates alcohol dehydrogenase E (AdhE) and potentiates pneumolysin (Ply). Hemolytic activity, colonization, and virulence of S. pneumoniae, as well as host cell myeloperoxidase activity, proinflammatory cytokine secretion, and inflammation, were significantly attenuated in adhE mutant bacteria (ΔadhE strain) compared to D39 wild-type bacteria. Therefore, AdhE might act as a pneumococcal virulence factor. Moreover, in the presence of ethanol, S. pneumoniae AdhE produced acetaldehyde and NADH, which subsequently led Rex (redox-sensing transcriptional repressor) to dissociate from the adhE promoter. An increase in AdhE level under the ethanol condition conferred an increase in Ply and H2O2 levels. Consistently, S. pneumoniae D39 caused higher cytotoxicity to RAW 264.7 cells than the ΔadhE strain under the ethanol stress condition, and ethanol-fed mice (alcoholic mice) were more susceptible to infection with the D39 wild-type bacteria than with the ΔadhE strain. Taken together, these data indicate that AdhE increases Ply under the ethanol stress condition, thus potentiating pneumococcal virulence.
INTRODUCTION
Alcoholism is one of the most important risk factors for pulmonary infections involving Streptococcus pneumoniae (pneumococcus) or Klebsiella pneumoniae (1–4). Alcohol impairs alveolar macrophages, which are the first cellular line of defense against pathogens in the lower airways (5). Impaired alveolar macrophages produce fewer proinflammatory cytokines and chemokines in response to pulmonary infection, resulting in an elevated bacterial burden and increased mortality (6, 7). Alcoholics are particularly susceptible to pulmonary infection by pneumococcus (8, 9); current or former alcohol abuse patients have an increased risk of pneumococcal infection (10). Pneumococcus spread in the bloodstream of ethanol (EtOH)-fed rats more rapidly than it did in control rats (11). However, the role of bacterial gene(s) induced by alcohol in virulence remains unclear.
Alcohol dehydrogenase (Adh; EC 1.1.1.1), which is produced by many different types of organisms ranging from bacteria to mammals, is involved in ethanol oxidation. Adh facilitates the conversion of alcohol to aldehyde with the reduction of NAD+ to NADH (12).
Previously, we showed that among six Adh genes in the S. pneumoniae type 2 D39 strain, including SPD_0265, SPD_1126, SPD_1636, SPD_1834, SPD_1865, and SPD_1985, only SPD_1834 (104-kDa protein), which encodes a bifunctional two-domain protein (named AdhE) containing Adh and acetaldehyde-coenzyme A (CoA) dehydrogenase domains in the C terminus and the N terminus, respectively, was strongly induced after ethanol shock (13). Ethanol-induced production of AdhE constituted 23% of all induced proteins (13).
Most of the studies on Adh in bacteria are focused on metabolism (14, 15) and tolerance, including nitrosative stress (16) and ethanol tolerance (17, 18). There have been a few studies on the contribution of Adh to bacterial virulence. Although the Gram-positive Listeria monocytogenes alcohol acetaldehyde dehydrogenase (LAP) is secreted and binds to the host-cell receptor Hsp60 to promote adherence (19, 20), S. pneumoniae type 2 AdhE was not secreted and did not influence pneumococcal adherence (data not shown). Therefore, the role of Adh in bacterial virulence, in particular, the mechanism through which AdhE contributes to pneumococcal virulence under ethanol-tolerant conditions, needs further clarification.
Here, we demonstrated that S. pneumoniae D39 could tolerate ethanol stress, which subsequently increased pneumolysin (Ply) via upregulation of AdhE. Mice were more susceptible to infection with D39 wild-type (WT) bacteria than with adhE mutant bacteria. Moreover, ethanol-fed mice (alcoholic mice) were more susceptible to infection with D39 WT bacteria than normal control mice infected with D39 WT and adhE mutant bacteria, suggesting that AdhE can potentiate S. pneumoniae type 2 virulence under the ethanol stress condition.
MATERIALS AND METHODS
Ethics statement.Male CD-1 (ICR) mice, 4 to 6 weeks old, were obtained from OrientBio, Inc. (Seongnam, Kyonggi, Republic of Korea), and housed in a specific-pathogen-free barrier facility (12-h light/dark cycle, 22 ± 2°C, 50% ± 10% relative humidity) at the School of Pharmacy at Sungkyunkwan University (Suwon, Republic of Korea). The animals received water and sterilized food ad libitum. All animal experiments were approved by the Animal Care and Use Committee of the SungKyunKwan University (Suwon, Republic of Korea). Infection procedures followed protocol PH-530518-06, in accordance with the animal care guidelines of the Korean Academy of Medical Sciences, and all efforts were made to minimize animal suffering.
In vivo survival and colonization.For the in vivo survival test, mice were infected with S. pneumoniae via the intranasal route (i.n.). Mouse survival was recorded eight times during the first 4 days and four times until the end of this experiment (14 days) (21).
For the in vivo colonization test, mice were infected with 1.5 × 107 CFU of S. pneumoniae D39 or the ΔadhE strain in phosphate-buffered saline (PBS) i.n. Mice were anesthetized, and nasopharynx, lung wash, lung, and blood were collected and subjected to viable cell counting on Todd-Hewitt–yeast (THY) agar supplemented with 5% sheep blood after appropriate serial dilution using PBS. Blood agar contained 10 μg/ml gentamicin only (for D39) or both 10 μg/ml gentamicin and 2.5 μg/ml erythromycin (for the ΔadhE strain) (21).
Bacterial strains, culture, transformation, and cell culture conditions.The bacterial strains (Escherichia coli and S. pneumoniae) and plasmids used in this study are shown in Table 1. Encapsulated S. pneumoniae D39 (serotype 2; NCTC7466) (22) and WU2 (serotype 3) (23) strains and their isogenic ΔadhE (serotype 2 and 3), Δrex (serotype 2), and ΔspxB (serotype 2) mutants were used. The mutants were created using an ermB cassette inserted opposite to the target gene orientation to disrupt its expression (24). To create an insertion-deletion mutation in the S. pneumoniae adhE and rex genes, an 860-bp ermB cassette was amplified with primers prs3 and prs4 (24) and used to disrupt adhE and rex expression. Two arms were flanked by S. pneumoniae D39 genomic DNA, and the triple PCR product was integrated into S. pneumoniae D39 by homologous recombination. Transformants were selected by 2.5 μg/ml erythromycin and confirmed by PCR and Western blotting as well as by DNA sequencing.
Bacterial strains and plasmids used in this study
The complemented strain that contains a recombinant pMV158 plasmid with the functional adhE gene in the isogenic ΔadhE mutant background was used. The full adhE gene, including its promoter, was amplified with primers (available upon request), which incorporated HindIII and EcoRI restriction enzyme sites from the D39 genome. Tetracycline-resistant transformants containing the recombinant plasmid were screened (1 μg/ml) and confirmed by PCR and Western blotting as well as by DNA sequencing. All pneumococci were cultured in Todd-Hewitt broth containing 0.5% (wt/vol) yeast extract (THY broth).
Murine macrophage RAW 264.7 cells were obtained from the American Type Culture Collection. Cells were cultured at 37°C in 95% air–5% CO2. RAW 264.7 cells were cultured in Dulbecco's modified Eagle's medium (DMEM) with 4.5 g/liter glucose, 10% fetal bovine serum (Gibco BRL), 100 units/ml penicillin G, and 100 μg/ml streptomycin.
Cloning and purification of recombinant AdhE and Rex.The adhE (GenBank accession number ABJ54182.1) and rex (GenBank accession number ABJ53704.1) genes in S. pneumoniae D39 were amplified by PCR using D39 genomic DNA as a template with AdhE and Rex primers (available upon request) that incorporated HindIII and EcoRI (AdhE) or BamHI and XhoI (Rex) restriction enzyme sites. The PCR products were digested with the appropriate enzymes (New England BioLabs) and cloned into the corresponding restriction sites in pET30a(+) (AdhE) or pET32b(+) (Rex) (Novagen) plasmids that were used to transform E. coli BL21(DE3) to generate the strains pPSH01 and pKEH01 (Table 1), respectively. The nucleotide sequences in the recombinant strains were confirmed by DNA sequencing (Cosmo, Seoul, Republic of Korea). E. coli strains were cultured in Luria broth (LB), and kanamycin (50 μg/ml) or ampicillin (100 μg/ml) was used to select recombinant colonies. Plasmid DNA was extracted and purified from E. coli with a commercially available kit (Cosmo, Seoul, Republic of Korea). Proteins were then induced with 0.1 mM isopropyl-β-d-thiogalactopyranoside (IPTG) for 6 h at 37°C and purified using a nickel-nitrilotriacetic acid column (Ni-NTA; Probond, Invitrogen) in default buffer (50 mM Tris-Cl, pH 7.5, 150 mM NaCl) and imidazole, according to the manufacturer's instructions. The eluted fraction was dialyzed against 50 mM Tris-HCl buffer (pH 7.5).
Histological analysis of lung tissue.Whole lungs of control and infected mice (n = 2) were fixed in 10% formalin (Sigma). The samples were embedded in paraffin, sectioned (5 μm), and stained with hematoxylin-eosin. Histological changes were evaluated randomly in nonconsecutive fields using an AxioPlan instrument (Zeiss, Germany) at a magnification of ×200. Images were analyzed using AxioVision, version 3.1 (Zeiss, Germany).
AO-EB stain.RAW 264.7 cells were infected with S. pneumoniae D39 or the ΔadhE strain (multiplicity of infection [MOI] of 100) for 2 h. Cells were stained with acridine orange and ethidium bromide (AO-EB) for apoptosis detection (25). Samples were visualized immediately using an IX70 instrument (Olympus).
MPO assay.Myeloperoxidase (MPO) was quantified to evaluate neutrophil infiltration in lung tissue (26). Mice were asphyxiated, and lungs were collected and homogenized in 50 mM sodium phosphate buffer (pH 6.5). The crude homogenate was added to hexadecyltrimethylammonium bromide (Sigma) to a final concentration of 0.5% before samples were frozen at −70°C. Frozen samples were thawed, sonicated, and centrifuged (20,000 × g for 30 min at 4°C). The MPO assay was performed as described previously (26). A 10-μl sample was added to a 96-well plate, and subsequently 100 μl of 3,3′,5′,5-tetramethylbenzidine liquid substrate was added. The colorimetric assay was recorded at A650 using a microplate reader (Molecular Device, USA).
Hemolytic assay.Hemolytic activity was determined, as previously described (27), using sheep red blood cells (Hanil Komed, Seongnam, Republic of Korea). S. pneumoniae was incubated to log phase (A550 of 0.3) prior to harvest, and the pellet was resuspended in PBS. Sodium deoxycholate (Sigma) was added to the suspension to a final concentration of 0.1% for lysis. A 50-μl sample was added to each well of a 96-well plate with 50 μl of 1.5% sheep red blood cell solution. The hemolytic titer was recorded as equal to the dilution at which 50% of red blood cells were lysed at A540.
Acetaldehyde assay.To measure acetaldehyde concentration, the S. pneumoniae D39 and ΔadhE strains were cultured until they reached an A550 of 0.3. Acetaldehyde level was measured with an aldehyde quantification kit (Abcam, England), according to the manufacturer's instructions. THY broth was used as a blank control.
ADH activity.AdhE activity was determined, as described previously (12), based on NADH release. Fresh, recombinant AdhE (300 nM) was added to a solution that contained 33 mM sodium phosphate buffer, pH 7.5, 1 mM NAD+, and 5 mM ethanol. NADH release was measured at A340.
Alcohol administration and determination of in vivo ethanol level.Mice were fed alcohol directly in their drinking water. The alcohol or water and standard chow intake were monitored every day. The alcohol group drank 10% (vol/vol) alcohol for 2 days, 15% (vol/vol) alcohol for 5 days, and 18% (vol/vol) alcohol until the end of the study (28). Ethanol level in vivo was determined from 100 μl of mouse serum, nasopharynx (nasal tissue) homogenates, or lung wash using alcohol dehydrogenase, NAD+ (Sigma), and sodium phosphate buffer, pH 8.0, to check the release of NADH, measured at A340 for 5 min. Ethanol (Merck) was used to construct a standard curve of NADH release.
Antisera and Western blotting.To prepare antisera, 5-week-old male CD1 mice were immunized via the intraperitoneal route (i.p.) with 10 μg of purified AdhE or Rex protein mixed with 100 μg of aluminum adjuvant (Sigma) at 14-day intervals. Sera were collected (after mice were asphyxiated) 1 week after the third immunization.
To collect whole-cell lysates, S. pneumoniae was grown in THY broth and lysed in lysis buffer (50 mM Tris-Cl, pH 8.0, 1 mM dithiothreitol [DTT], 0.1% Triton X-100) (24). To collect released protein, the supernatant of S. pneumoniae culture was precipitated with 10% trichloroacetic acid (Sigma) (23). Protein samples were separated by 7.5 to 12.5% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) prior to transfer onto polyvinylidene difluoride (PVDF) membranes (Millipore, USA). Membranes were then blocked by 5% (wt/vol) skim milk (Merck) and probed with the appropriate primary antibody (AdhE, Rex, Ply, SpxB, and PsaA). Horseradish peroxidase (HRP)-conjugated anti-mouse IgG (Promega) was used as the secondary antibody.
RNA isolation and qRT-PCR.Bacterial RNA was isolated by the hot-phenol method (29), and eukaryotic RNA was isolated using TRIzol reagent (Invitrogen), as instructed by the manufacturer. All RNA samples were treated with DNase I (TaKaRa). Briefly, a 1-μg RNA sample was subjected to reverse transcription into cDNA using random primers (TaKaRa). Quantitative reverse transcription-PCR (qRT-PCR) was performed according to the manufacturer's instructions (Applied Biosystems), as described previously (24).
ELISA.Groups of mice (n = 5) were infected with S. pneumoniae D39 or the ΔadhE strain (1.5 × 107 CFU/mouse) i.n. for 4, 8, 12, 24, or 48 h. After mice were asphyxiated, lungs were collected and homogenized, and cytokine (tumor necrosis factor alpha [TNF-α], interleukin-1β [IL-1β], and IL-6) levels were measured by enzyme-linked immunosorbent assay (ELISA), according to the manufacturer's instructions (BD Biosciences).
Recombinant AdhE protein was time and dose dependently added to RAW 264.7 cell culture. Supernatant was collected and used to check cytokine levels by ELISA, as instructed by the manufacturer (BD Biosciences). Lipopolysaccharide (LPS), heat-inactivated AdhE, and VncR were used as controls.
Electrophoretic mobility shift assay (EMSA).To determine the binding ability of Rex to the adhE promoter, a band shift assay was carried out with minor modifications (30). Purified recombinant Rex or 65°C-denatured Rex was coincubated with 4 nM purified adhE promoter DNA in reaction buffer (20 mM Tris, pH 7.5, 2 mM EDTA, 3% [wt/vol] Ficoll 400, 0.5 mM DTT, 150 mM NaCl) for 15 min at 25°C. The reaction mixture was then analyzed in a 7.5% native acrylamide/bisacrylamide gel at 4°C. The shift bands were visualized under UV light after 45-min of Gel Red staining (Biotium, CA, USA).
To analyze the influence of NAD+ and NADH on the binding affinity of Rex to the adhE promoter, another similar experiment was performed but in the presence of NAD+ (Sigma) or NADH (Sigma). To prevent DNA-nicotinamide coprecipitation, HEPES buffer was used instead of Tris-HCl buffer, as above (30).
NAD+-NADH pool isolation and NAD cycling assay.To extract NAD+ and NADH, the S. pneumoniae D39 and ΔadhE strains were cultured to the log phase and incubated with 100 mM ethanol for 30 min. The dinucleotides were extracted as described previously, with some modifications (31). The cell pellet was immediately frozen in liquid nitrogen before being suspended in a final volume of 250 μl of either 0.2 M HCl (for NAD+ extraction) or 0.2 M NaOH (for NADH extraction). NAD+-NADH levels were determined with a cycling assay (31) measured at A570.
Hydrogen peroxide production assay.To measure the concentration of released H2O2, 5 ml of S. pneumoniae bacteria was cultured in a 15-ml glass tube at 37°C with 100-rpm agitation to supply oxygen. A total of 900 μl of bacterial culture supernatant was mixed with 100 μl of reaction solution (0.3% [wt/vol], 2,2′-azinobis [ABTS; 3-ethylbenzthiazolinesulfonic acid]; Sigma), and 0.0002% (wt/vol) horseradish peroxidase (HRP) (Sigma) in 0.1 M sodium phosphate buffer, pH 7.0. Before the A540 was measured, the mixture was incubated at 25°C for 3 min. The H2O2 concentration was determined with an H2O2 (Sigma) standard curve (32).
Cytotoxicity assay.RAW 264.7 cells were infected with S. pneumoniae D39 or the ΔadhE strain (MOI of 100) for 2 h. Cytotoxicity was determined using 3-(4,5-dimethylthiazole-2-yl)-2,5-dephenyl-tetrazolium bromide) (MTT), as described previously (33).
ChIP assay.S. pneumoniae D39 or the ΔadhE strain at log phase was treated with 100 mM ethanol for 30 min, and the bacterial pellet was subjected to a chromatin immunoprecipitation assay (ChIP), as described previously (34). The bacterial pellets were fixed in 1% formalin for 15 min, and fixation was stopped by the addition of 125 mM glycine, pH 2.5. Bacterial cell lysate was immunoprecipitated overnight at 4°C with anti-Rex and protein A-agarose beads (Santa Cruz) with gentle agitation. The beads were rinsed with lysis buffer and Tris-EDTA (TE) buffer before elution with 50 μl of TE buffer containing 1% SDS. Samples were then treated with 100 μg of proteinase K. DNA from treated samples was precipitated by the cold-ethanol method before PCR with adhE promoter primers (WT; 300 bp) (available upon request). Input was determined using 16S RNA primers.
Statistical analysis.Most graphs were created and statistical analyses performed using SigmaPlot, version 11.0, software (Systat Software). In vivo survival (see Fig. 3C and 6E) and qRT-PCR (see Fig. 3A) analyses were performed using GraphPad Prism, version 5.02. Statistical analysis was calculated using one-way analysis of variance (ANOVA) (Holm-Sidak method), two-way ANOVA (Bonferroni method), a Mann-Whitney U test (nonparametric), or a log rank (Mantel-Cox) test, as appropriate. P values of ≤0.05 (denoted on the figures by *), P ≤ 0.01 (**), and P ≤ 0.001 (***) were considered significant. Data presented are the means and standard deviations (SD) of results of three to four independent experiments.
RESULTS
EtOH-induced S. pneumoniae type 2 AdhE promotes reducing conditions.Although AdhE in S. pneumoniae D39 (type 2) was identified by sequence homology (13) and is highly conserved among streptococci (92 to 99% identity with Streptococcus mitis, Streptococcus australis, Streptococcus. sanguinis, Streptococcus pyogenes, and Streptococcus agalactiae) and S. pneumoniae strains (99% identity with types 1, 3, 4, 6B, 19A, 19F, and 14), its catalytic activity has not been identified. To confirm whether S. pneumoniae AdhE can catalyze ethanol and NAD+ to form acetaldehyde and NADH, recombinant S. pneumoniae AdhE (Fig. 1A) was expressed in E. coli, and the activity was determined. The results demonstrate that S. pneumoniae AdhE produced more NADH (P < 0.001) than the control (Fig. 1A), indicating that AdhE has ADH activity.
Pneumococcal AdhE produces NADH and acetaldehyde. (A) ADH activity of recombinant AdhE was determined at A340 using 1 mM NAD+, 5 mM ethanol, and 0.3 mM AdhE. (B) S. pneumoniae D39 or the ΔadhE strain was cultured until an A550 of 0.3 was reached, and then AdhE was induced by supplementation with 100 mM ethanol for 30 min. Bacterial cultures were used to detect acetaldehyde. Data are representative of three independent experiments. Significant differences were analyzed by one-way ANOVA (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001). (−), blank.
To determine the role of AdhE in S. pneumoniae D39, an ΔadhE mutant was constructed with the triple-PCR method by inserting an ermB cassette to disrupt adhE gene expression. In the presence of ethanol, the acetaldehyde level in S. pneumoniae D39 WT was 4-fold higher (P < 0.001) than in the untreated sample (Fig. 1B). Ethanol treatment consistently increased the NADH/NAD+ ratio in S. pneumoniae by approximately 19-fold (Table 2). These results indicate that ethanol-induced AdhE produces both NADH and acetaldehyde.
Increase of NADH after ethanol exposure
S. pneumoniae growth might be affected by culture medium (THY broth in normal cultures or DMEM during host cell infection) or by ethanol. To investigate this possibility, S. pneumoniae growth in THY broth or DMEM or in the presence of ethanol was performed; however, these media do not significantly affect the growth of the WT or adhE mutant (data not shown). To confirm that ethanol induced AdhE expression, S. pneumoniae D39 WT was treated with ethanol. The results showed that ethanol induced AdhE (50 mM) time dependently (Fig. 2A and B). In addition, a complemented strain of the ΔadhE mutant, in which adhE expression is reconstituted by introducing a pMV158 vector harboring the adhE gene, had increased levels of AdhE after ethanol treatment (Fig. 2C). To verify AdhE induction at the mRNA level, mRNA expression was measured by RT-PCR. Consistently, ethanol treatment induced adhE mRNA production (Fig. 2E), demonstrating that ethanol induced adhE production at both the mRNA and protein levels.
Ethanol-induced AdhE increases H2O2 production. (A and B) S. pneumoniae D39 was cultured until an A550 of 0.3 was reached, and then ethanol was added. Cell lysates were used for Western blotting. AdhE induction was dose (A) and time (B) dependent. (C, D, and E) S. pneumoniae strains were cultured until an A550 of 0.3 was reached, and then AdhE was induced with 100 mM ethanol for 30 min. The cell lysate was used for Western blotting (C). Bacterial cultures were used to determine H2O2 concentration (D) or to isolate RNA and determine mRNA levels by RT-PCR (E). Data are representative of three independent experiments. Significant differences were analyzed by one-way ANOVA (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001).
Since NADH is known to generate superoxide anions, the immediate precursor of H2O2 (35), and since ethanol-induced AdhE increases NADH levels (Table 2), H2O2 levels were determined after ethanol treatment. After ethanol treatment, the H2O2 level in treated D39 WT bacteria increased 2-fold compared to the level in nontreated D39 WT bacteria. The complemented strain had a significantly higher H2O2 level than the ΔadhE mutant (Fig. 2D). In contrast, the H2O2 level in the ΔadhE mutant increased only 1.2-fold compared to the control. The H2O2 level of a pyruvate oxidase deletion (ΔspxB) mutant was nearly abrogated (Fig. 2D) because SpxB is the major protein that produces H2O2 (36), and the SpxB level did not vary between the D39 WT and ΔadhE mutant strains (Fig. 3A); thus, the ΔspxB mutant was used as a negative control. These data suggest that ethanol-induced AdhE increases H2O2 production.
AdhE potentiates virulence. (A) S. pneumoniae D39, the ΔadhE mutant, and complemented strains were cultured until an A550 of 0.3 was reached. Cell lysates were subjected to Western blotting. (B) The hemolytic activity of the S. pneumoniae D39, ΔadhE, and complemented strains was determined with a 1.5% (vol/vol) red blood cell solution followed by measurement of the A540. Data are representative of three independent experiments. (C) Mice (n = 10 to 15) were infected with a high (1.5 × 107 CFU/mouse) or low (1.5 × 106 CFU/mouse) dose of pneumococcal strain serotype 2 intranasally (i.n.), and survival time was recorded until 14 days postinfection. (D) Mice (n = 10 to 15) were infected with a high dose (4.5 × 107 CFU/mouse) of pneumococcal strain serotype 3 i.n., and survival time was determined. (E) Mice (n = 7) were infected i.n. with 1.5 × 107 CFU of the pneumococcal strains of serotype 2, and viable cells were counted. Significant differences were analyzed using one-way ANOVA (A and B), a log rank test (C and D), or a Mann-Whitney rank sum test (D) (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001).
Attenuation of virulence by adhE mutation.To determine whether adhE expression affects expression of S. pneumoniae virulence factors, pneumococcal lysates at the log phase were subjected to Western blotting. Among several pneumococcal virulence factors at an A550 of 0.3, only intracellular pneumolysin (Ply) levels were decreased (1.7-fold) in the ΔadhE mutant compared to those in D39 WT bacteria. No significant differences in choline binding protein A (CbpA), pneumococcal surface adhesin A (PsaA), or SpxB were found (Fig. 3A). The difference in H2O2 production between the S. pneumoniae D39 WT strain and the ΔadhE mutant (Fig. 2D) was relevant to NADH levels (Table 2) but not because of SpxB levels (Fig. 3A). These data demonstrate that AdhE affects the Ply level but not CbpA, PsaA, and SpxB.
Moreover, the intracellular Ply level in the ΔadhE mutant at an A550 of 0.6 significantly decreased (1.58-fold) compared to that in the D39 WT strain (data not shown). Additionally, the Ply level released from the ΔadhE mutant to the culture supernatant was 1.59-fold lower than that from the D39 WT strain (data not shown). To determine whether the released protein level was associated with autolysis, S. pneumoniae was cultured in THY broth with an autolysis inhibitor, choline chloride (23). Consistently, when the pneumococci were cultured in the presence of 2% choline chloride, the level of Ply released from the ΔadhE mutant to the culture supernatant significantly decreased compared to that from the D39 WT strain (data not shown). These results indicate that the Ply level is increased by AdhE, but this occurs autolysis independently.
In addition, the effect of AdhE on hemolytic activity, primarily caused by Ply, was determined. The D39 WT and complemented strains had significantly higher hemolytic activities than the ΔadhE mutant (P < 0.05) (Fig. 3B), suggesting that AdhE increases hemolysis. This result confirms that AdhE increases the Ply level of S. pneumoniae.
To further corroborate that the adhE mutation attenuated virulence in vivo, 4-week-old mice were infected i.n. with S. pneumoniae D39 and its isogenic ΔadhE mutant at a high dose (1.5 × 107 CFU/mouse) and low dose (1.5 × 106 CFU/mouse). Mice infected with the D39 WT or complemented strains showed higher mortality than those infected with the ΔadhE mutant (Fig. 3C). In addition, high-dose infection showed a significantly higher mortality than low-dose infection (Fig. 3C). Thus, high-dose infection was used for further studies.
To further confirm the role of AdhE on virulence, another S. pneumoniae strain, type 3 WU2, and its isogenic WU2 ΔadhE mutant were used to infect mice via the i.n. route at both a low dose (1.5 × 106 CFU/mouse) and high dose (4.5 × 107 CFU/mouse) (37). No mice died after infection with the WU2 ΔadhE mutant at either infection dose although mice infected with the high dose were more morbid than those infected with the low dose. Our results showed that WU2 infection at the high dose and low dose caused 46.7% (P = 0.0023) (Fig. 3D) and 13.3% (data not shown) mortality, respectively. The survival results with these mice demonstrate that the S. pneumoniae ΔadhE mutation attenuated virulence in serotypes 2 and 3 and that high-dose infection produced a more significant difference in mortality between WT and adhE mutant infections than low-dose infection.
Furthermore, significantly fewer viable cells colonized the lung, nasopharynx, nasal fluid, and blood time dependently after i.n. infection (n = 7/group) with the ΔadhE mutant than after infection with the D39 WT strain (Fig. 3E). Taken together, these results demonstrate that the ΔadhE mutation attenuated S. pneumoniae virulence in vitro and in vivo.
AdhE increases cytokine production, inflammation, and cell death.Although the ΔadhE mutant affected virulence in vitro and in vivo (Fig. 3), its adherence to various cell types, including macrophage RAW 264.7, lung A549, and nasopharynx Detroit 562 cells, was not significantly different from that of the D39 WT strain (data not shown). Thus, we investigated how AdhE modulates host gene expression by using a systems biology approach. RAW 264.7 cells were infected with S. pneumoniae, and host gene expression was analyzed by microarray and systems biology analyses. NF-κBIZ, Cxcl, and TNF-α expression after infection with D39 WT were significantly higher than after infection with the ΔadhE strain (data not shown). To confirm this result, RAW 264.7 cells were infected with S. pneumoniae, and mRNA levels were measured by qRT-PCR. mRNA levels of the ATF-3, NF-κB, Cxcl2, TNF-α, and IL-6 genes were significantly higher after infection with D39 WT than after infection with the control or ΔadhE mutant. Cxcl2 and TNF-α levels were increased 5- and 3-fold, respectively, by infection with D39 WT (Fig. 4A). To corroborate the effect of AdhE on cytokine production, mice (n = 5/group) were infected i.n. with 1.5 × 107 CFU of S. pneumoniae, and cytokine levels were measured. Infection with the D39 WT strain significantly increased cytokine (TNF-α, IL-1β, and IL-6) levels time dependently compared to infection with the mutant (Fig. 4B). Moreover, to identify whether AdhE could affect cytokine secretion indirectly or directly, a recombinant AdhE was added to RAW 264.7 cell culture, and the levels of TNF-α, IL-6, and IFN-γ were determined. Results showed that recombinant AdhE induced TNF-α time dependently (Fig. 4C) but did not induce IL-6 and IFN-γ (data not shown); however, the effect of recombinant AdhE on TNF-α induction was not as potent as that of recombinant Ply (Fig. 4C). Taken together, these results suggest that AdhE could directly affect TNF-α production but not as potently as Ply. Therefore, AdhE-induced Ply production might be a major origin of inflammatory response enhancement.
AdhE increases cytokine production, neutrophil infiltration, inflammation, and cell death. (A) RAW 264.7 cells were infected with S. pneumoniae D39 or the ΔadhE mutant (MOI 100). Total RNA was isolated and used for qRT-PCR. Data are representative of three independent experiments. (B) Mice (n = 5) were infected i.n. with 1.5 × 107 CFU of the pneumococcal strains, and lung homogenates were used to determine cytokine levels. (C) TNF-α was induced by recombinant AdhE time dependently in RAW 264.7 cells. Recombinant AdhE was added into RAW 264.7 cell culture, and the TNF-α level in the culture supernatant was determined. LPS, heat-inactivated AdhE, and VncR were used as controls. (D) RAW 264.7 cells were infected with the pneumococcal strains (MOI of 100) for 2 h and stained with acridine orange-ethidium bromide (AO-EB) to determine cell death. (E) Mice (n = 2) were infected i.n. with 1.5 × 107 CFU of the pneumococcal strains, and lung histology was examined 24 h after infection via hematoxylin and eosin (HE) staining (magnification, ×200). (F) Mice (n = 3) were infected i.n. with 1.5 × 107 CFU of the pneumococcal strains, and lung homogenates were used for an MPO assay. Significant differences were analyzed by two-way ANOVA (A) or one-way ANOVA (B, C, and F) (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001).
To confirm these results in vitro, RAW 264.7 cells were infected with the ΔadhE mutant and stained with acridine orange and ethidium bromide (AO-EB). Infection with the ΔadhE mutant induced less apoptosis than infection with the WT (Fig. 4D), indicating that cell death induced by AdhE was potentiated.
To corroborate inflammation after i.n. infection (n = 2/group), mouse lungs were stained with hematoxylin and eosin, and histopathological changes were examined. Mouse lungs infected with the ΔadhE mutant had less inflammation than lungs infected with the WT (Fig. 4E), suggesting that AdhE induced lung inflammation in mice.
Furthermore, myeloperoxidase (MPO) activity, which is a hallmark of neutrophil infiltration (38), was determined. MPO levels in the lungs of mice (n = 7/group) infected with D39 WT was significantly higher than those in mice infected with the ΔadhE mutant (Fig. 4F).
These results demonstrate that AdhE enhances inflammatory cytokine production, lung inflammation, virulence, and cell death and might be considered a potential virulence factor.
Rex binds to the adhE promoter and represses adhE expression in the absence of NADH.In Staphylococcus aureus, Rex represses adhE expression by binding to the adh promoter, and this binding is competitively modulated by NADH and NAD+. In the presence of NADH, Rex dissociates from the adhE promoter; thus, the NADH/NAD+ ratio could potentially signal AdhE induction (30). S. pneumoniae has a conserved putative Rex binding motif in the adhE promoter (data not shown), and after ethanol treatment, Rex was induced in the D39 WT strain (Fig. 5A). Therefore, Rex could affect adhE expression. We examined the relationship between AdhE and Rex after ethanol treatment. AdhE expression in the Δrex mutant at both the mRNA and protein levels was significantly increased compared to the levels in the control (Fig. 5B), demonstrating that Rex downregulates adhE expression.
Rex binds to the adhE promoter and represses adhE expression, but NADH reverses this repression. (A) Increase of Rex level in ethanol shock depends on AdhE. S. pneumoniae D39 and the ΔadhE strain were incubated (A550 of 0.3) and induced with 100 mM ethanol for 30 min. Cell lysates were collected and subjected to Western blotting. (B) Pneumococcal strains were cultured until an A550 of 0.3 was reached. Total RNA was isolated for RT-PCR, or cell lysates were used for Western blotting. (C) Native recombinant Rex binds to the adhE promoter. The adhE promoter (final concentration, 4 nM) was incubated with 2.5, 5, or 10 nM native or heat-denatured Rex for 10 min, and EMSA was performed. (D) S. pneumoniae D39 and Δrex strains were cultured until an A550 of 0.3 was reached, and ethanol was added for 30 min. The cell lysates were subjected to a ChIP assay. The Δrex mutant was used as a negative control. (E) A point mutation was introduced in the Rex binding motifs of the adhE promoter, and the mutated AdhE promoter (4 nM) was incubated with 20 nM Rex for 10 min. Rex binding was examined by EMSA. (F and G) To determine Rex binding to the adhE promoter, the adhE promoter was incubated with various concentrations of NADH and NAD+ (final concentration, 4 nM), and EMSA was performed. Data are representative of two (A, C, D, E, F, and G) or three (B) independent experiments. Significant differences were analyzed using one-way ANOVA (C) (***, P ≤ 0.001).
To investigate whether Rex binds to the adhE promoter, recombinant Rex was purified from E. coli (data not shown), and an electromobility shift assay (EMSA) was performed. Purified Rex bound to the adhE promoter dose dependently, but heat-denatured Rex did not (Fig. 5C). To examine the effect of ethanol on Rex binding to the adhE promoter, a chromatin immunoprecipitation (ChIP) assay was performed. Antibodies raised against Rex were used to precipitate the Rex-adhE promoter complex from D39 WT lysate. Rex-bound DNA fragments were purified and amplified by PCR. Rex binding to the adhE promoter was decreased after ethanol treatment (Fig. 5D), demonstrating that ethanol increases AdhE expression by diminishing Rex binding to the adhE promoter.
To corroborate the Rex binding motif in the adhE promoter, point mutations were introduced into the putative Rex binding site (Fig. 5E), and EMSA was performed. A single point mutation decreased Rex binding to the adhE promoter, and a double mutation nearly abolished Rex binding. These results suggest that Rex binds to the adhE promoter and represses adhE expression (Fig. 5E).
Rex plays an important role in redox balance by sensing the pyridine nucleotide NAD(H) (39). Moreover, NADH inhibits Rex binding to DNA in S. aureus (30). Therefore, in the presence of ethanol, higher NADH levels (Fig. 1A and Table 2) could cause a dissociation of Rex from the adhE promoter. To test this hypothesis, EMSA was performed in vitro. When the reaction solution was supplemented with NADH, Rex binding to the adhE promoter was decreased (Fig. 5F). Supplementing the reaction mixture with NAD+, however, did not significantly affect Rex binding to the adhE promoter (Fig. 5G).
These results indicate that Rex binding might depend on the relative NAD+ and NADH concentrations and that Rex could repress adhE expression by binding to the adhE promoter. Under ethanol shock conditions, however, increased NADH decreased Rex binding to the adhE promoter, resulting in AdhE induction. These data demonstrate that Rex is a redox-dependent repressor for adhE expression.
Alcohol potentiates S. pneumoniae type 2 virulence by upregulating AdhE.To determine the effect of ethanol on S. pneumoniae virulence in vitro, the cytotoxicity of S. pneumoniae on RAW 264.7 cells was measured. After ethanol treatment, D39 WT cytotoxicity increased significantly compared to that of the non-ethanol-treated control (P < 0.01) (Fig. 6A). Cytotoxicity of the ΔadhE mutant did not change after ethanol treatment (Fig. 6A). The complemented strain had a level of cytotoxicity similar to that of the WT after ethanol treatment (Fig. 6A), possibly due to constitutive high expression of AdhE. In addition, the D39 WT and complemented strains had higher hemolytic activity than the ΔadhE mutant (P < 0.001), and ethanol treatment slightly increased the hemolytic activity of the WT and complemented strains (Fig. 6B). Moreover, ethanol treatment increased the Ply level in the D39 WT strain compared to that in the non-ethanol-treated control (P < 0.01) but not in the ΔadhE mutant (Fig. 6C).
Pneumococcal virulence increased after exposure to ethanol due to AdhE induction. (A) Ethanol-treated cells were more susceptible to infection with D39 than with the ΔadhE mutant. Ethanol-treated RAW 264.7 cells were infected with pneumococcal strains (MOI of 100) for 2 h, and cytotoxicity was determined by MTT assay. (B) Hemolytic activity of the pneumococcal strains after ethanol treatment for 30 min was determined. (C) The cell lysates from the experiment shown in panel B were collected and subjected to Western blotting. (D) Mice (n = 5) were fed ethanol for 10 days, and the ethanol levels of the nasopharynx, lung wash, and blood were determined. (E) Mice (n = 20) were fed ethanol for 10 days and infected i.n. with 1.5 × 107 CFU of S. pneumoniae D39 or the ΔadhE strain. Survival times were determined. Data are representative of three independent experiments (A to C). Significant differences were analyzed by one-way ANOVA (A to C), a Mann-Whitney rank sum test (D), or a log rank test (E) (*, P ≤ 0.05).
When ethanol was administered to mice according to the ethanol feeding protocol (28), alcoholic mice did not exhibit any abnormal behavior. Moreover, mice fed up to 20% ethanol for 5 weeks were still considered healthy enough to tolerate experiments (40, 41) prior to infection with S. pneumoniae (41) or Mycobacterium tuberculosis (40). Therefore, this ethanol feeding procedure is suitable for experiments evaluating the effects of alcohol. To further identify the effect of AdhE on virulence in vivo, mice were fed ethanol for 10 days (alcoholic mice), and ethanol levels in the nasopharynx (nasal tissue), lung wash, and blood were measured by an alcohol dehydrogenase assay. The ethanol levels of alcoholic mice were significantly higher than those in the water-fed control group, and the highest ethanol concentration was detected in lung wash (Fig. 6D).
When alcoholic mice were infected i.n. with 1.5 × 107 CFU of S. pneumoniae and mortality rate was examined, alcoholic mice infected with D39 WT became morbid earlier than the other mice and were more susceptible to D39 WT than the control group (Fig. 6E) (P = 0.045). Alcoholic mice infected with the ΔadhE mutant, however, did not show any significant difference in susceptibility compared with nonalcoholic mice infected with the ΔadhE strain (Fig. 6E). Moreover, the control group infected with D39 showed 59% mortality, whereas the group infected with the ΔadhE mutant exhibited approximately 34% mortality (P = 0.03). These results indicate that AdhE potentiates S. pneumoniae virulence under an ethanol stress condition.
DISCUSSION
Alcohol abuse is a serious health problem in the United States and in most developed countries, especially among young people (3, 10). Approximately 20 million Americans meet the diagnostic criteria for alcoholism, and 20% to 40% of patients admitted to hospitals present with disease exacerbated by alcohol abuse (3). The most widely recognized infectious disease associated with ethanol abuse is bacterial pneumonia (3). Thus, a virulence factor that contributes to pathogenicity during infection might be a candidate target for pneumococcal treatment in alcoholics.
Bacterial metabolism is maintained by balancing the redox state of all cellular components (42). Increased oxidative stress can break this balance and damage essential cellular components. Therefore, bacteria have evolved specialized sensors to monitor redox signals and adapt to altered redox environments. One bacterial redox sensor, Rex, senses the NADH/NAD+ redox state and modulates respiratory gene expression as a redox-regulated repressor (39). Most Gram-positive bacteria, including several important human and animal pathogens, such as S. aureus, L. monocytogenes, and S. pneumoniae, encode Rex homologues (39). The adhE-binding activity of Rex is modulated specifically by the ratio of NADH to NAD+. Under aerobic conditions, the NAD+ concentration is higher than the NADH concentration, and Rex has a high affinity for the adhE promoter. Under anaerobic conditions, Rex binds to NADH and loses its affinity for the adhE promoter, leading to the transcription of genes involved in respiration and metabolism (39), such as fermentative enzymes comprising AdhE (30, 43, 44). In this study, we identified AdhE as a virulence factor that might be regulated by Rex under an alcoholic condition.
After ethanol ingestion, the in vitro antipneumococcal activity of surfactants and pneumococcal clearance were inhibited (10). Moreover, in ethanol-fed rats, more S. pneumoniae type 3 bacteria spread into the bloodstream (11). Although alcoholics are highly susceptible to S. pneumoniae infection (11), how alcohol contributes to pneumococcal virulence has not been resolved. Among six alcohol dehydrogenases in S. pneumoniae type 2 D39, only SPD_1834 (AdhE) was strongly induced by ethanol (15). AdhE has 98% identity with Streptococcus gordonii AdhE (AcdH), which uses only ethanol as its substrate (45). Moreover, a BLAST search revealed that SPD_0265 has 93% identity with S. gordonii AdhA, which preferentially catalyzes 1-propanol (45). SPD_1126 (GenBank) seems to function as a threonine dehydrogenase in amino acid transport and metabolism. SPD_1636 has 85% identity with S. gordonii AdhB, which preferentially catalyzes 1-butanol (45). On the other hand, SPD_1865 (GenBank) seems to function in propanoate metabolism and as a formaldehyde dehydrogenase-like ADH. Furthermore, SPD_1895 (GenBank) has 100% identity with the NADH-dependent methanol dehydrogenase in S. pneumoniae 70585. Therefore, in the presence of ethanol, AdhE could preferentially play a major role in catalysis of ethanol.
Here, we demonstrated that ethanol shock increases acetaldehyde and NADH concentrations via AdhE (Fig. 1 and Table 2). A high NADH level restores a reduced condition, which triggers superoxide or H2O2 generation (46). Thus, under ethanol shock, acetaldehyde and H2O2 accumulation could cause a toxic response in host cells during pneumococcal infection (Fig. 1B, 4D and E, and 6A and E). In particular, acetaldehyde, which inhibits cell proliferation (47), is a carcinogen in rats (48) and humans (49) and inhibits DNA repair enzymes (50). H2O2 is a common form of reactive oxygen species (ROS) that damages cell membranes via lipid oxidation (51). H2O2 in S. pneumoniae is a toxin for alveolar epithelial cells (52). S. pneumoniae tolerates up to 10 mM H2O2 (53); therefore, S. pneumoniae can survive in H2O2 concentrations that damage host cells during infection. The ethanol stress condition also increases Ply levels (both intracellular and released) (Fig. 3A and 6C and data not shown). Because Ply is a pore-forming cytolysin that has hemolytic activity, Ply and H2O2 in combination can impair human ciliated epithelium cells additively (54). An increase in Ply or H2O2 alone or in combination with ethanol shock could contribute to S. pneumoniae pathogenesis involving respiratory tract infection, thus enhancing pneumococcal virulence.
Colonization of the lungs is a key stage in a local pneumococcal infection of the nasopharynx as it converts into a systemic infection (55). This conversion is linked to the activation of alveolar macrophages and, subsequently, heavy neutrophil recruitment (56). A pneumococcal virulence factor, Ply, has been shown to activate nitric oxide (NO) production in neutrophils (57). Moreover, ply and spxB mRNAs were abundant in the nasopharynx following intranasal challenge with the D39 WT strain (58), suggesting that Ply and SpxB are important for the oxidative stress response. Alcoholics are susceptible to infectious disease, particularly bacterial pneumonia (59). Pneumonia is an inflammation of the alveoli that causes them to fill with excess fluid, and the resulting inhibition of O2 and CO2 exchange can allow pneumococci to proliferate. Therefore, the mechanism causing bacterial pneumonia should be elucidated. Since AdhE was expressed under reducing conditions, it can enhance pneumococcal survival and virulence. Importantly, Ply expression and H2O2 levels increased in WT bacteria after ethanol exposure. Thus, AdhE might contribute to the ability of pneumococci to colonize the upper respiratory tract and subsequently infect the lower respiratory tract.
S. pneumoniae type 2 AdhE increases inflammatory cytokines in vivo and in vitro (Fig. 4). Recombinant AdhE could increase TNF-α secretion but is not as potent as Ply (Fig. 4C). In addition, AdhE was not released into the culture supernatant (data not shown). Therefore, AdhE increases the Ply level, resulting in an increase of cytokine production. How AdhE stimulates TNF-α production should be studied further.
In conclusion, S. pneumoniae type 2 AdhE enhances hemolytic activity, neutrophil infiltration, inflammatory cytokine production, and inflammation (Fig. 3 and 4). Ethanol-induced AdhE produces NADH (Fig. 1 and Table 2), which disrupts the Rex-adhE promoter association (Fig. 5) and results in increased acetaldehyde (Fig. 1B) and H2O2 levels (Fig. 2D) and increased virulence (Fig. 6). Therefore, AdhE contributes to enhanced pneumococcal virulence under alcoholic conditions. These results suggest that S. pneumoniae D39 is tolerant to ethanol stress, which subsequently increases virulence through AdhE. Thus, AdhE might be a useful target in the treatment of pneumococcal disease in alcoholics.
ACKNOWLEDGMENTS
This work was supported by the Basic Science Research Program, through the National Research Foundation of Korea, funded by the Ministry of Education, Science, and Technology (2011-0024794) to D.-K.R. and also partly by the Mid-Career Researcher Program through the National Research Foundation of Korea (2012-016803) to S.C.
The funding agencies had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
FOOTNOTES
- Received 5 August 2014.
- Returned for modification 14 September 2014.
- Accepted 4 October 2014.
- Accepted manuscript posted online 13 October 2014.
- Copyright © 2015, American Society for Microbiology. All Rights Reserved.
