Pneumolysin-Dependent and -Independent Gene Expression Identified by cDNA Microarray Analysis of THP-1 Human Mononuclear Cells Stimulated by Streptococcus pneumoniae

Abstract

Pneumolysin is an important virulence factor of Streptococcus pneumoniae, interacting with the membranes of host cells to elicit a multitude of inflammatory responses. We used cDNA microarrays to identify genes which are responsive to S. pneumoniae in a pneumolysin-dependent and -independent fashion. The THP-1 human monocytic cell line was coincubated for 3 h with medium alone, with the virulent type 2 S. pneumoniae strain D39, or with the isogenic strain PLN, which does not express pneumolysin. RNA was isolated from the monocytes and hybridized on cDNA microarrays. Of 4,133 genes evaluated, 142 were found to be responsive in a pneumolysin-dependent fashion, whereas 40 were found to be responsive independent of pneumolysin. Genes that were up-regulated in cells exposed to D39 relative to those exposed to PLN included genes encoding proteins such as mannose binding lectin 1, lysozyme, α-1 catenin, cadherin 17, caspases 4 and 6, macrophage inflammatory protein 1β (MIP-1β), interleukin 8 (IL-8), monocyte chemotactic protein 3 (MCP-3), IL-2 receptor β (IL-2Rβ), IL-15 receptor α (IL-15Rα), interferon receptor 2, and prostaglandin E synthase. Down-regulated genes included those encoding complement component receptor 2/CD21, platelet-activating factor acetylhydrolase, and oxidized low-density lipoprotein receptor 1 (OLR1). Pneumolysin-independent responses included down-regulation of the genes encoding CD68, CD53, CD24, transforming growth factor β2, and signal transducers and activators of transcription 1. These results demonstrate the striking effects of pneumolysin on the host cell upon exposure to S. pneumoniae.

The bacterium Streptococcus pneumoniae (pneumococcus) is a major cause of pneumonia, meningitis, bacteremia, otitis media, and sinusitis in humans. Clinical infection by this extracellular pathogen is preceded by colonization of the nasopharynx via adherence of the bacterium to the mucosal cells of the upper respiratory tract through interactions between bacterial surface adhesins and epithelial cell receptors (4, 53). Adherent bacteria are then internalized within mucosal epithelial cells via receptor-mediated endocytosis and transported across the cytoplasm, where they are released at the submucosa (54). Bacteremia may then occur by interaction with endothelial cells, allowing access to the bloodstream (49). S. pneumoniae evades phagocytosis by virtue of its capsule. Additionally, several pneumococcal proteins have been shown to influence the virulence of this organism, including pneumococcal surface protein A (PspA), pneumococcal surface adhesin A (PsaA), autolysin, neuraminidase, and pneumolysin (2, 37, 54, 56).

Pneumolysin is produced by all clinical isolates of S. pneumoniae (44, 56). It interacts with cholesterol in the cell membrane, inserting itself into the lipid bilayer and forming oligomeric pores that lead to cell lysis. Furthermore, pneumolysin activates the classical complement pathway in the absence of specific antibody by binding to the Fc region of human immunoglobulin G (8). Pneumococcal invasion of the pulmonary epithelium and endothelial cells has been shown to be pneumolysin dependent (50). Pneumolysin has numerous effects on human immune cells, including inhibition of neutrophil and monocyte function (50). In human neutrophils, pneumolysin increases intracellular calcium concentrations, phospholipase A2 activity, CD11b/CD18 expression, and superoxide production. Also, pneumolysin increases potassium efflux independently of calcium-ATPase or sodium-potassium-ATPase activity in these cells (16). Furthermore, pneumolysin causes calcium-dependent prostaglandin E2 (PGE₂) and leukotriene B4 production in human neutrophils and nitric oxide, interleukin-1 (IL-1), tumor necrosis factor α (TNF-α), IL-6, and cyclooxygenase 2 production in mononuclear phagocytes (3, 15, 16, 31). In mice, chronic bacteremia rather than rapidly progressive sepsis is observed with infection by pneumolysin mutants compared to isogenic wild-type isolates (6, 7). This difference may be due to greater host resistance in mice infected with the pneumolysin-deficient isolate, suggesting that pneumolysin may impair this process.

Along with complement and antibodies, phagocytosis is a critical component of the host immune response to S. pneumoniae. Upon the organism's entry into the bloodstream, this response is likely initiated through contact with naive monocytes and macrophages, resulting in an inflammatory response and activation of complement, opsonization of bacteria, phagocytosis, and intracellular killing of the bacterium.

In an effort to further elucidate the role of pneumolysin in the host response to S. pneumoniae, we used cDNA microarray analysis to identify genes in the human monocytic cell line THP-1 that are pneumolysin responsive. Furthermore, by comparing the gene expression profiles of THP-1 cells in response to both a wild-type and pneumolysin-deficient mutant with the profile of uninfected cells, we have identified genes that are responsive to S. pneumoniae independently of pneumolysin. The pneumolysin-dependent nature of the response of select marker genes and gene products was confirmed and further characterized through semiquantitative and real-time reverse transcriptase-PCR (RT-PCR) as well as enzyme-linked immunosorbent assay (ELISA). This study characterizes the global transcriptional response of human monocytes to contact with S. pneumoniae and demonstrates the utility of cDNA microarray analysis in elucidating the role of specific virulence factors in the host-pathogen interaction.

MATERIALS AND METHODS

Bacterial strains and growth conditions.

The pneumococcal strains used in this study included the serotype 2 strain S. pneumoniae D39 (38) and its isogenic derivative, PLN (9), which lacks pneumolysin. S. pneumoniae was grown in Todd-Hewitt broth containing 0.5% yeast extract, plated on blood agar plates (Difco Laboratories, Detroit, Mich.), and incubated at 37°C under reduced-oxygen conditions. Aliquots of exponentially growing bacteria were concentrated to 1/10 volume by centrifugation and stored at −80°C in Todd-Hewitt broth containing 0.5% yeast extract supplemented with 10% glycerol. The number of CFU per milliliter was determined by plate count on test aliquots for each strain. Aliquots were thawed and adjusted to the desired concentration prior to challenge of cell cultures.

Cell culture exposures.

Human acute monocytic leukemia THP-1 cells were maintained in RPMI 1640 culture medium containing 10% fetal bovine serum, penicillin-streptomycin, l-glutamine, and 5 × 10⁻⁵ M β-mercaptoethanol. Cells were suspended in 75-cm² tissue culture flasks at a concentration of 10⁶ cells per milliliter in antibiotic-free supplemented medium and coincubated with either medium alone, the virulent type 2 strain S. pneumoniae D39, or the isogenic mutant strain PLN. Coincubation was continued for 3 h for microarray and initial RT-PCR analysis and for up to 12 h for real-time PCR and ELISA analysis. Cells were incubated at 37°C in 5% CO₂ at a bacterium-to-cell ratio of 100:1. Two independent experiments were carried out for the microarray hybridization analysis. For some assays, where indicated, multiple cell culture exposure experiments were performed.

RNA isolation.

Total RNA was isolated by using the Trizol reagent (Invitrogen/GibcoBRL, Carlsbad, Calif.) in accordance with the manufacturer's instructions.

Probe preparation.

A 10-μg sample of total RNA in diethyl pyrocarbonate-treated H₂O was mixed with 2 μg of oligo(dT) primer (Invitrogen/ResGen), denatured at 70°C, and chilled on ice. This was added, in a total volume of 24 μl, to a mixture of 0.1 M dithiothreitol (Invitrogen/GibcoBRL); 20 mM (each) dATP, dGTP, and TTP (Amersham Pharmacia Biotech, Buckinghamshire, England); 300 units of Superscript II reverse transcriptase (Invitrogen/GibcoBRL); and 100 μCi [α-³³P]dCTP (Amersham Pharmacia Biotech). The mixture was then incubated at 37°C for 90 min, after which 70 μl of diethyl pyrocarbonate-treated H₂O was added and the mixture was placed on ice. Probes were then purified on chromatography columns and placed on ice.

Microarray hybridization and analysis.

GF211 Known Genes Genefilter cDNA array (Invitrogen/ResGen) membranes, spotted with cDNA elements representing 4,133 human genes, were initially prewashed in boiling 0.5% sodium dodecyl sulfate (SDS) for 5 min with agitation. Prehybridizations were performed for 2 h at 42°C in Microhyb solution (Invitrogen/ResGen) with poly(A) (Invitrogen/ResGen) and denatured Cot-1 DNA as blocking reagents. Denatured probes were then added and incubated for 18 h at 42°C. After hybridization, the membranes were washed twice in 2× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate)-1% SDS for 20 min (each washing) at 50°C followed by a wash in 0.5× SSC-1% SDS for 15 min at 55°C. The membranes were then exposed to phosphor screens (Molecular Dynamics, Sunnyvale, Calif.) for 2 to 5 days. Images were acquired on a Storm 860 Phosphorimager (Molecular Dynamics) and analyzed by using Pathways version 3.0 software (Invitrogen/ResGen). Normalization was performed by using the average of all data points for each array. Changes in expression levels were calculated by using normalized intensities and were given as ratios. Only elements with average signal intensities of greater than twofold that of the background in at least one experimental condition and only genes found to have a greater-than-twofold change in expression in both experiments were considered for further analysis.

RT-PCR.

One microgram of total RNA from each sample was denatured in the presence of 1 μg of oligo(dT) primer at 70°C. Thereafter, the mixture was chilled on ice, and a master mix containing 50 mM Tris-HCl (pH 8.3); 75 mM KCl; 3 mM MgCl₂; dATP, dCTP, dGTP, and TTP at 1.25 mM each; and 25 U of Superscript II reverse transcriptase was added to each tube. The reaction mixture was incubated for 10 min at room temperature, followed by a 60-min incubation at 37°C and a 5-min incubation at 90°C. PCR was performed by using 1 μl of the appropriate dilution of cDNA (empirically determined to yield product in the linear range for the final PCR conditions). Primer sequences used for amplification of specific genes by RT-PCR are shown in Table 1. The 18S (60°C) Classic (Ambion, Austin, Tex.) primer set was used to amplify 18S rRNA.

TABLE 1.

DNA sequences and annealing temperatures of primers used in RT-PCR

Primer name	Sequence	Annealing temp (°C)
MIP-1β-F	5′-CCAAACCAAAAGAAGCAAGC-3′	55
MIP-1β-R	5′-ACAGTGGACCATCCCCATAG-3′
OLR1-F	5′-AAGGACCAGCCTGATGAGAA-3′	57
OLR1-R	5′-CGGACAAGGAGCTGAACAAT-3′
IL-8-F	5′-ATGACTTCCAAGCTGGCCGTG-3′	55
IL-8-R	5′-TTATTGAATTCTCAGCCCTCTTCAAAAACTTCTC-3′

Real-time RT-PCR.

Real-time PCR was used to measure gene expression over time. Reverse transcription reactions using oligo(dT) were performed in accordance with the manufacturer's instructions (Advantage RT-for-PCR; BD Biosciences, San Jose, Calif.). Real-time PCR was performed on an ABI Prism 7000 sequence detection system by using TaqMan universal PCR master mix and ABI predeveloped assay reagents (Applied Biosystems, Foster City, Calif.) for macrophage inflammatory protein 1β (MIP-1β), IL-8, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) in accordance with the manufacturer's protocols. Two independent experiments were performed. Samples were run in triplicate, and the data were pooled for each experiment. Levels of mRNAs were normalized to the levels of GAPDH mRNA.

IL-8 and MIP-1β ELISA.

IL-8 and MIP-1β concentrations were determined with ELISA kits (R&D Systems, Minneapolis, Minn.). After exposure to experimental conditions, particulates were removed from cell cultures by centrifugation. Supernatants were stored at −70°C until assayed. All experiments and assays were performed in duplicate. Optical densities were read at the appropriate wavelength on an MRX microplate reader (Dynex Technologies, Chantilly, Va.).

Statistical analysis.

Results from ELISA analysis of the three conditions tested were analyzed by using analysis of variance (ANOVA) with Tukey's correction for multiple comparisons (P = 0.001).

RESULTS

Microarray analysis of THP-1 cells exposed to S. pneumoniae.

We used cDNA microarray analysis to characterize the transcriptional response of the human monocytic cell line THP-1, used to represent naive peripheral monocytes, to the D39 and PLN strains of S. pneumoniae. Monocyte viability, as assessed by erythrosin B exclusion, was ≥90% prior to experimental use and ≥80% after exposure to experimental conditions. No appreciable difference between pre- and postexperimental bacterial cell counts was observed for either isolate in either of the two independent experiments performed. Of the differentially expressed genes identified, the expression of 142 (3.4%) were found to be responsive in a pneumolysin-dependent fashion, whereas 40 (0.92%) were found to be responsive independently of pneumolysin. Genes were annotated by using cellular roles assigned by the Proteome Public HumanPSD Database (29).

Pneumolysin-dependent changes in THP-1 cell gene expression.

Changes in gene expression were considered dependent on pneumolysin if there was a greater-than-twofold difference in transcription between cells exposed to D39 and those exposed to PLN and a corresponding change between either cells exposed to D39 or cells exposed to PLN and those exposed to medium alone. Of the genes evaluated, 116 were found to be up-regulated and 26 were found to be down-regulated in response to the presence of pneumolysin. These are listed in Table 2.

TABLE 2.

Human monocyte genes that are pneumolysin dependent and differentially expressed in response to S. pneumoniae^a

Function group and accession no.	Locus	Gene name	Molecular function	Change in expression (fold) with exposure to:
				D39 vs. medium	D39 vs. PLN
Antipathogen response
T72877	IL1RN	Interleukin 1 receptor antagonist		1.3	3.5
AA446103	LMAN1	Lectin, mannose-binding, 1	Chaperone; mannose-binding lectin	2.4	3.3
N63943	LYZ	Lysozyme (renal amyloidosis)	Lysozyme; amyloid protein; lysine	3.0	2.9
AA521362	CR2/CD21	Complement component receptor 2		1.2	0.3
Cell adhesion
T65118	CTNNA1	Catenin alpha 1	Cadherin-associated protein	16.0	14.2
AA088861	CDH17	Cadherin 17	Calcium-dependent cell adhesion	2.0	2.1
Cell cycle control: AA448659	CDC25B	Cell division cycle 25B	Protein tyrosine phosphatase	5.7	3.7
Cell death and apoptosis
H44953	CASP4	Caspase 4	Apoptosis-related Cys protease	3.1	2.6
W45688	CASP6	Caspase 6	Apoptosis-related Cys protease	3.2	2.2
Cell-to-cell signaling
H62864	SCYA4	Small inducible cytokine A4 (MIP-1β)	Protein kinase; chemokine	5.9	4.9
AA040170	SCYA7	Small inducible cytokine A7 (MCP3)	Chemokine	3.9	3.1
AA102526	IL8	Interleukin 8	Cytokine	2.7	2.4
Lipid, fatty-acid, and sterol metabolism
AA436163	PTGES	Prostaglandin E synthase	Antimicrobial humoral response	4.0	3.9
AA464238	PAFAH1B3	Platelet-activating factor acetylhydrolase	Lipid metabolism; neurogenesis	0.9	0.4
Protein degradation: H94487	CTSE	Cathepsin E	Proteolysis and peptidolysis	3.5	2.1
Protein folding
R98019	APCS	Amyloid P component, serum	Acute-phase response protein	1.5	4.0
AA676404	PPIC	Cyclophilin C	Peptidylprolyl isomerase	4.8	3.1
Protein translocation: AA682386	OLR1	Oxidized low-density lipoprotein receptor 1	LDL receptor	0.93	0.14
Signal transduction
AA057204	IL2RB	Interleukin 2 receptor, beta	Signal transduction	2.2	5.2
AA054754	IL15RA	Interleukin 15 receptor, alpha	Signal transduction	2.3	4.5
AA485426	IFNAR2	Interferon (α, β, and ο) receptor 2	Antiviral response protein	3.5	2.2
AA431967	TXN	Thioredoxin	Antimicrobial humoral response	2.3	2.1

^aData from all arrays used in this study are available at http://dcp.utmem.edu/drogers/.

Among those up-regulated were genes for lysozyme, prostaglandin E synthase, mannose binding lectin 1, IL-1 receptor antagonist (IL-1Ra), α-catenin, cadherin 17, cell division cycle 25B, caspases 4 and 6, monocyte chemotactic protein 3 (MCP-3), IL-8, MIP-1β, IL-2 receptor β (IL-2Rβ), IL-15 receptor α (IL-15Rα), and interferon receptor 2. Those that were down-regulated included genes for complement component receptor 2/CD21 and oxidized low-density lipoprotein receptor 1 (OLR1).

Pneumolysin-independent changes in THP-1 cell gene expression.

Changes in gene expression were independent of pneumolysin if there were greater-than-twofold average differences in transcription between cells exposed to PLN and those exposed to medium alone. Of the genes evaluated, one was found to be up-regulated and 38 were found to be down-regulated independently of the presence of pneumolysin. These are listed in Table 3.

TABLE 3.

Human monocyte genes that are pneumolysin independent and differentially expressed in response to S. pneumoniae

Function group and accession no.	Locus	Gene name	Molecular function	Change in expression (fold) with exposure to:
				PLN vs medium	D39 vs medium
Anti-pathogen response
AA421296	CD68	CD68 antigen		0.5	0.5
AA132090	CD53	CD53 antigen	Antimicrobial humoral response	0.4	0.3
H59916	CD24	CD24 antigen		0.1	0.1
Cell-to-cell signalling: AA233738	TGFB2	Transforming growth factor β2		0.4	0.5
Signal transduction
AA488075	STAT1	Signal transducer/activator of transcription 1	Transcription activating factor	0.4	0.4
R82176	MADH7	MAD homolog 7	Receptor Ser/Thr kinase	0.4	0.4
AA446928	ERBB2	v-erb-b2 oncogene homolog 2	Protein Tyr kinase	0.3	0.4
Other: AA450003	DYRK4	Tyr phosphorylation-regulated kinase 4	Protein kinase	3.2	2.0

The only gene observed to be up-regulated independently of pneumolysin was that for dual-specificity tyrosine-(Y)-phosphorylation-regulated kinase 4. Among those down-regulated were genes including those for CD53, CD68, CD24, transforming growth factor β2 (TGF-β2), and signal transducers and activators of transcription 1 (STAT1).

Analysis of mRNA by using RT-PCR.

Using RT-PCR, we confirmed the differential expression of three pneumolysin-dependent genes (those for IL-8, MIP-1β, and OLR1) in experiments with RNA obtained independently of that used in microarray hybridizations. Figure 1 shows that mRNA levels of IL-8 were increased in response to both PLN and, to a greater extent, D39, whereas mRNA levels of MIP-1β were increased in response to D39 only. Conversely, mRNA levels of OLR1 were increased in response to PLN but decreased in response to D39, compared to those in medium alone. These data are consistent with those obtained in the microarray expression studies.

FIG. 1.

RT-PCR analysis of mRNA expression of selected genes in THP-1 cells exposed for 3 h to medium alone, to PLN, or to D39. Samples were analyzed by agarose gel electrophoresis to visualize the bands. The strains or medium to which THP-1 cells were exposed are indicated across the top. Along the side is the gene for which the mRNA level was assayed. The 18S rRNA is the control gene for which expression does not change.

Further studies were undertaken to characterize the gene expression profiles of IL-8 and MIP-1β over time by using real-time RT-PCR. Figure 2A and B show that after exposure to D39 and PLN for 3, 6, 9, and 12 h, mRNA levels of IL-8 are increased compared to levels after exposure to medium alone. Additionally, this difference is greater for cells exposed to D39 than for those exposed to PLN. In contrast, while mRNA levels of MIP-1β are marginally increased in response to PLN at 6, 9, and 12 h postexposure, there is a marked increase in response to D39 for all time points studied. Data shown are from one representative experiment of four RT-PCR experiments in total.

FIG. 2.

Shown is real-time RT-PCR analysis of mRNA expression of THP-1 cells exposed to medium alone, to PLN, or to D39 for IL-8 (A) and MIP-1β (B). Also shown is ELISA analysis of secreted protein at the corresponding time points for IL-8 (C) and MIP-1β (D). Data are presented as means ± standard errors of the mean. Statistical analysis was performed by using ANOVA with Tukey's correction for multiple comparisons (P ≤ 0.001). *, significantly different from exposure to medium alone; **, significantly different from exposure to PLN and medium alone.

Analysis of IL-8 and MIP-1β by ELISA.

To confirm that changes observed in mRNA levels corresponded with changes in protein expression, we measured IL-8 and MIP-1β in supernatants of THP-1 cells under these experimental conditions. Figure 2C and D show that while both IL-8 and MIP-1β are responsive to both PLN and D39, the response to PLN is less than that to D39. These differences were evident as early as 6 h for IL-8 and 3 h for MIP-1β. These data corresponded well with mRNA levels as measured by real-time RT-PCR.

Statistical analysis.

ANOVA with Tukey's correction for multiple comparisons was used to determine the significance of values among experimental groups in ELISA experiments (significance was defined as P ≤ 0.001).

DISCUSSION

Pneumolysin-dependent gene expression.

We applied cDNA microarray analysis to further clarify the role of pneumolysin in the host-monocyte response to exposure to S. pneumoniae. While certain genes exhibit similar responses to both isolates, a number of genes were clearly responsive to the presence of pneumolysin. IL-1Ra was down-regulated in response to PLN but not D39. D39, but not PLN, was able to induce the expression of the genes encoding mannose-binding lectin-1 and lysozyme, while expression of the gene encoding CD21 was up-regulated by PLN but not D39.

IL-1Ra is a soluble receptor antagonist produced by a multitude of cells, including monocytes and macrophages (21). Its release is believed to attenuate the deleterious effects of IL-1. While IL-1 has been shown to be expressed in human monocytes in response to S. pneumoniae and pneumolysin, it is observed at a later time point than that studied here (15, 31). No changes in IL-1α or IL-1β mRNA levels were observed at the 3-h time point used for cDNA array experiments. Early increases in expression of this molecule could have protective effects against a later IL-1-mediated inflammatory response or could benefit the bacterium by blunting this host response.

Mannose-binding lectin affects phagocytosis and cytokine production in phagocytes and activates complement in an antibody-independent fashion. This molecule also shows affinity for S. pneumoniae and has been shown to increase the association between Neisseria meningitides and monocytes, macrophages, and neutrophils by binding to that bacterium (33, 41). Increased production of this molecule in response to the presence of pneumolysin could serve to enhance phagocytosis but may also be a principle mediator of the inflammatory response to pneumococcus.

Lysozyme is a relatively ubiquitous antimicrobial enzyme that digests bacterial cell walls by cleaving peptidoglycan (32). It has been shown to enhance the phagocytic activity of neutrophils, stimulate monocytes, and induce autolysis in S. pneumoniae (17, 32). CD21 serves as the receptor for C3d, a degradation fragment of complement component C3, and for gp350 from Epstein-Barr virus (26). Although originally described in B-lymphocytes, CD21 has also been found to be expressed in other cell types, including monocytes. Binding of gp350 increases IL-1 and TNF-α production in human monocytes (19, 20). Since purified pneumococcal polysaccharides can activate C3 via the alternative pathway and bind to C3d in the absence of specific antibody (25), it is possible that S. pneumoniae-induced IL-1 and TNF-α production is mediated through this pathway as well. Down-regulation of CD21 would suggest an altered inflammatory response to D39 compared to that to PLN. While we observed no difference in IL-1 mRNA expression in response to D39 or PLN, others have observed differences at later time points (15). Unfortunately, the gene encoding TNF-α is not represented on the array used in the present study.

Catenins and cadherins play critical roles in regulating cytoskeletal rearrangements. Interestingly, the most up-regulated pneumolysin-dependent gene in this study, that for α-1 catenin, has been implicated in the uptake of Listeria monocytogenes in human epithelial cells (35). Furthermore, several bacterial toxins have been shown to target actin, such as the enterotoxin of Bacteroides fragilis, which is a protease specific for E-cadherin (48). Pneumolysin-dependent up-regulation of the genes encoding α-1 catenin and cadherin 17 may represent a mechanism by which this factor facilitates bacterial uptake.

A number of cell cycle control genes were found to be up-regulated in a pneumolysin-dependent fashion in this study. Relative to adherence in the G₀-G₁ and S phases, the adherence of S. pneumoniae to the human epithelial lung cell line A549 has been shown to increase during the G₂ phase of the cell cycle due to increased cell size (11). The CDC25B phosphatase plays a critical role in the control of G₂-M progression (39). It is possible that changes in expression of cell cycle control genes represent a defensive host response to S. pneumoniae to reduce bacterial adherence to host cells.

Apoptosis of macrophages has been suggested to be both a mechanism whereby bacteria can avoid immune-mediated killing and a normal host immune response. A number of bacterial species have been shown to induce macrophage apoptosis, including S. pneumoniae (22). Apoptosis of macrophages in response to opsonized serotype 1 S. pneumoniae was associated with successful phagocytosis and bacterial killing. The finding of pneumolysin-dependent up-regulation of both caspases 4 and 6 in the present study suggests that apoptosis associated with the host immune response to S. pneumoniae may be a caspase-mediated event. Up-regulation of the gene encoding cyclophilin C is consistent with this hypothesis, as this enzyme has been implicated in genome degradation during apoptosis (40).

Several genes encoding chemokines were up-regulated in response to the presence of pneumolysin, including the genes for MIP-1β, MCP-3, and IL-8. These gene products are mediators of inflammation, chemotaxis, and cell adhesion (1, 12). In the present study, time course analysis of IL-8 and MIP-1β gene and protein expression in THP-1 cells by real-time RT-PCR and ELISA, respectively, revealed differing responses for D39 and PLN. Without pneumolysin, the pneumococcus can elicit a moderate induction of IL-8 and MIP-1β. However, the presence of this virulence factor is associated with a substantial increase in these chemokines.

Also up-regulated in a pneumolysin-dependent fashion was the gene encoding prostaglandin E synthase. Prostaglandin endoperoxide H2 is formed from arachidonic acid by cyclooxygenases 1 and 2, and it is subsequently converted to PGE₂ by prostaglandin E synthase. Interestingly, pneumolysin-dependent up-regulation of cyclooxygenase-2 has been demonstrated in murine macrophages, but only in the presence of gamma interferon (13). Taken together, these data suggest a coordinated increase in production of PGE₂ in response to the presence of pneumolysin in activated monocytes and/or macrophages.

Platelet-activating factor (PAF) acetylhydrolase is the enzyme responsible for degradation of PAF, a phospholipid which mediates intercellular interactions and inflammation (45). Decreased PAF acetylhydrolase production has been associated with the sepsis syndrome (52). The PAF receptor has been implicated as a site of attachment for virulent S. pneumoniae (18). The gene encoding the cytoplasmic γ-form of PAF acetylhydrolase was up-regulated in response to PLN but not in response to D39. Of interest, the gene encoding the plasma form of this enzyme was down-regulated by both PLN and D39, although not to the extent to meet the criteria used in this study. Pneumolysin-mediated impairment of PAF acetylhydrolase expression could contribute to the host inflammatory response through increased circulating PAF. This also raises the possibility that increased PAF might compete with S. pneumoniae and thus impair it from binding to the PAF receptor.

Cathepsin E was up-regulated in a pneumolysin-dependent fashion. This enzyme is responsible for proteolytic degradation and major histocompatiblity complex class II processing for presentation of antigen on antigen-presenting cells (5). Up-regulation of this gene may enhance the presentation of pneumococcal antigens. The gene encoding serum amyloid P component (SAP) was down-regulated by PLN, but this was abolished by D39. SAP binds to DNA in chromatin to protect it from degradation and also possesses immunomodulatory properties. SAP has been shown to bind to certain bacteria, such as Streptococcus pyogenes and N. meningitidis, but not S. pneumoniae, and to elicit an antiopsonic effect, impairing phagocytosis and bacterial killing (28). SAP^−/− mice have been shown to be resistant to lethal infection by organisms to which SAP binds but to be more susceptible to those which SAP does not bind, suggesting that increased SAP might be beneficial to the host during pneumococcal infection (42). Interestingly, SAP has recently been shown to be the Escherichia coli Shiga toxin 2-neutralizing factor of human plasma (34).

OLR1 was up-regulated in response to PLN but not in response to D39. OLR1 is a type II membrane glycoprotein which serves as an endocytosis receptor for oxidized low-density lipoprotein and is highly expressed in a multitude of cells, including macrophages. OLR1 has been shown to be induced by TNF-α, TGF-β, and lipopolysaccharide. Recently OLR1 was shown to act as a cell surface receptor for Staphylococcus aureus and E. coli (51).

The genes encoding IL-2Rβ and IL-15Rα were both down-regulated in response to PLN but up-regulated in response to D39, while the genes encoding thioredoxin and interferon receptor 2 were both up-regulated in response to D39. Thioredoxin is a redox enzyme that exhibits chemokine-like activities, including induction of the IL-2 receptor (10). While it is interesting to speculate on the effects of pneumolysin on the response to IL-2 and IL-15 in these cells, it is important to note that no change was observed in the expression of the gamma subunit of the IL-2 receptor.

Pneumolysin-independent gene expression.

Several genes were differentially expressed in a manner independent of pneumolysin. These pneumolysin-independent responsive genes included those involved in anti-pathogen response, cell-to-cell signaling, RNA polymerase II transcription, and signal transduction.

CD68 is a membrane antigen highly expressed on monocytes and macrophages that plays a role in phagocytosis of pathogens (30). It is also involved in homing of macrophages to particular sites (47) and interacts with membrane lectins and selectins in cell-cell interactions (14). CD53 is also an integral membrane protein. It is expressed on a broad range of different hematopoietic cell types, including monocytes and macrophages (43), and is involved in associating with integrins and protein kinase C to facilitate their interaction and thus their engagement in growth regulation of these cells (58). Down-regulation of CD68 and CD53 would lead to severe curtailment of the monocyte defense against pneumococcus. On the other hand, CD24 is a monocyte antigen whose expression is repressed during differentiation of monocytes (24). Its down-regulation may indicate that the monocyte has initiated the differentiation process.

TGF-β2, a cell-signaling molecule, has many immunomodulatory roles. Previously, it has been shown to up-regulate C3 in monocytes (23). Potentially, down-regulation of this molecule would lead to less C3, the key component in both the classical and alternative complement cascades. In addition, TGF-β2, shown to be produced by monocytes as well as other cells, has been shown to down-regulate gamma interferon and nitric oxide production in activated monocytes and macrophages (55). STAT1 is a signal transduction molecule required for gamma interferon signaling (36). MADH7, another signal transduction molecule, is a member of the Smad family of proteins. It has been shown to suppress transcription of TGF-β responsive genes (27) and is up-regulated by TGF-β itself (57). Therefore, down-regulation of TGF-β in light of MADH7 down-regulation is not surprising. The ErbB2 signal transduction molecule is a tyrosine kinase receptor essential for IL-6 signaling, since down-regulation of ErbB2 has been shown to abolish activation of IL-6-responsive genes (46).

In this study, we have identified genes that are responsive to S. pneumoniae in both a pneumolysin-dependent and -independent manner. The results indicate an intricate interaction between host and pneumococcus; however, these responses demonstrate that the interaction between host and pathogen are more complex, relying on many factors in the pathogen to elicit such responses. Similar experiments with pneumococcal strains deficient in other known or putative virulence factors should be performed to determine the contribution of such factors to the pathogenesis of infection.