Recombinant Actinobacillus actinomycetemcomitans Cytolethal ...

INFECTION AND IMMUNITY, Sept. 2001, p. 5925–5930 0019-9567/01/$04.00⫹0 DOI: 10.1128/IAI.69.9.5925–5930.2001 Copyright © 2001, American Society for Microbiology. All Rights Reserved.

Vol. 69, No. 9

Recombinant Actinobacillus actinomycetemcomitans Cytolethal Distending Toxin Proteins Are Required To Interact To Inhibit Human Cell Cycle Progression and To Stimulate Human Leukocyte Cytokine Synthesis SUMIO AKIFUSA,1 STEPHEN POOLE,2 JO LEWTHWAITE,1 BRIAN HENDERSON,1* 1 AND SEAN P. NAIR Cellular Microbiology Research Group, Eastman Dental Institute, University College London, London WC1X 8LD,1 and Division of Endocrinology, National Institute for Biological Standards and Control, Potters Bar, Herts,2 United Kingdom Received 27 March 2001/Returned for modification 14 May 2001/Accepted 15 June 2001

It has recently been discovered that Actinobacillus actinomycetemcomitans, an oral bacterium causing periodontitis, produces cytolethal distending toxin (CDT), a cell cycle-modulating toxin that has three protein subunits: CdtA, CdtB, and CdtC. In this study, we have cloned and expressed each toxin gene from A. actinomycetemcomitans in Escherichia coli and purified the recombinant Cdt proteins to homogeneity. Individual Cdt proteins failed to induce cell cycle arrest of the human epithelial cell line HEp-2. The only combinations of toxin proteins causing cell cycle arrest were the presence of all three Cdt proteins and the combination of CdtB and CdtC. A similar experimental protocol was used to determine if recombinant Cdt proteins were able to induce human peripheral blood mononuclear cells (PBMCs) to produce cytokines. The individual Cdt proteins were able to induce the synthesis by PBMCs of interleukin-1␤ (IL-1␤), IL-6, and IL-8 but not of tumor necrosis factor alpha, IL-12, or granulocyte-macrophage colony-stimulating factor, with CdtC being the most potent and CdtB being the least potent cytokine inducer. There was evidence of synergism between these Cdt proteins in the stimulation of cytokine production, most markedly with gamma interferon, which required the minimum interaction of CdtB and -C to stimulate production. Actinobacillus actinomycetemcomitans, a gram-negative coccobacillus found in the human oral cavity, has been implicated in the pathogenesis of certain forms of periodontal disease and also in several systemic diseases, such as endocarditis, meningitis, and osteomyelitis (32, 38). This organism has the usual panoply of putative virulence factors, such as lipopolysaccharide (LPS) (6, 19, 43). It also secretes a number of unusual proteins, including a leukocyte-specific leukotoxin with proapoptotic activity (15), a chaperonin 60 with osteolytic activity (14, 34), and various inhibitors of cell cycle progression (21, 39, 40). There is growing evidence that microorganisms have evolved a range of mechanisms to evade both the innate and acquired host immune response (7). It has been proposed that virulence factors acting to impair host defense mechanisms play significant roles in the pathology of infections with A. actinomycetemcomitans (29). One recently discovered class of bacterial toxin, called cytolethal distending toxin (CDT), has been isolated from a range of pathogenic bacteria, including Escherichia coli, Campylobacter species, Shigella species, Haemophilus ducreyi, and Helicobacter species (1, 22, 23, 24, 28, 41, 45). This toxin was originally discovered as a factor which induced distension of CHO cells (reviewed in reference 24). However, it is now clear that the biological activities of CDT, such as

distension, actin rearrangement, and apoptosis, depend on the type of cell being studied (24). Thus, mouse Y-1 adrenal cells and NIH 3T3 fibroblasts are not affected by CDT treatment (2, 13). CDT is able to induce growth arrest at the G2/M phase in epithelial cells and apoptosis of cultured B-cell lines in an ataxia telangiectasia-mutated kinase-dependent manner (3). Recent studies suggest that degradation of chromosomal DNA by CdtB, which possesses DNase I motifs, is responsible for the blocking of the cell cycle-dependent dephosphorylation of Cdc2, the catalytic subunit of cyclin B, followed by arrest of sensitive cells in the G2/M phase (5, 17). A. actinomycetemcomitans produces CDT, which is encoded by three genes, cdtA, cdtB, and cdtC, located on the bacterial chromosome in tandem to form an operon with two other genes that have not yet been characterized (18, 31, 35). All Cdt proteins have signal peptides and can be isolated in the culture supernatant (18, 31, 35). Shenker and coworkers have reported that the A. actinomycetemcomitans CdtB is the active toxic component able to block the proliferation of T lymphocytes and thus induce immunosuppression (30, 31). In contrast, it has been reported that a CdtC-deficient mutant of H. ducreyi lacked cytotoxicity (33) and that a monoclonal antibody to H. ducreyi CdtC neutralized CDT cytotoxicity (1). Thus, the roles of the various Cdt proteins are still unclear. In addition to inhibiting cell cycle progression, it has recently been reported that Campylobacter jejuni CDT directly mediated the release of interleukin-8 (IL-8) from intestinal epithelial cells (11). This suggested the possibility that CDT may play a dual role and that different components of this toxin could

* Corresponding author. Mailing address: Cellular Microbiology Research Group, Eastman Dental Institute, University College London, 256 Gray’s Inn Rd., London WC1X 8LD, United Kingdom. Phone: 44 2079151190. Fax: 44 2079151190. E-mail: b.henderson @eastman.ucl.ac.uk. 5925

5926

NOTES

play different roles in inducing or inhibiting immune responses. In this report, we have expressed each A. actinomycetemcomitans cdt gene product independently using an E. coli expression system, purified each toxin “subunit,” and assessed its capacity to stimulate the release of cytokines from human peripheral blood mononuclear cells (PBMCs). Bacterial strains and growth conditions. A. actinomycetemcomitans Y4 (serotype b; ATCC 43718) was grown on brain heart infusion agar (Oxoid, Hampshire, United Kingdom) supplemented with 5% (vol/vol) horse blood at 37°C for 2 days in an atmosphere of 5% CO2, harvested from the plates with sterile saline, and centrifuged at 3,000 ⫻ g for 20 min (10). E. coli strains TOP10 (Invitrogen, Leek, The Netherlands) and HMS174(DE3) (Novagen, Nottingham, United Kingdom) were used in this study. E. coli was routinely grown in LuriaBertani (LB) broth. Cells and culture conditions. The human epithelial cell line HEp-2 was grown in Dulbecco’s minimal essential medium (GIBCO, Paisley, United Kingdom) containing L-glutamine, 10% fetal calf serum, streptomycin (100 ␮g/ml), and penicillin (100 IU/ml) in an atmosphere containing 5% CO2. PBMCs were prepared from buffy coat blood by density gradient centrifugation as previously described (36). Cloning of cdt genes into an N-terminal polyhistidine expression vector. The oligonucleotides 5⬘-GGATCCTGTTCG TCAAATCAACGA and 5⬘-CTGCAGTTAATTAACCGCTG TTGC were designed to amplify the 625-bp cdtA gene. The oligonucleotides 5⬘-GGATCCAACTTGAGTGATTTCAAA and 5⬘-CTGCAGTTAGCGATCATGAACAAA were designed to amplify the 785-bp cdtB gene. The oligonucleotides 5⬘-GG ATCCCATGCAGAATCAAATCCT and 5⬘-CTGCAGTTAG CTACCCTGATTTCT were designed to amplify the 506-bp cdtC gene. These primers were based on the sequence data for the cdt genes reported by Sugai and coworkers (35) and were designed to amplify each cdt gene without the DNA encoding the N-terminal signal peptide and also contained recognition sequences for restriction enzymes BamHI and PstI (underlined). Chromosomal DNA from A. actinomycetemcomitans was used as the template for PCR. The PCR fragments were initially cloned into pCR4-TOPO (Invitrogen) and transformed into E. coli TOP10. The cdt genes were cut from the pCR4-TOPO on BamHI-NotI fragments and ligated to BamHI-NotI-digested pET-28a(⫹) (Novagen). The ligation mixtures were transformed into E. coli HMS174(DE3), and transformants were selected by growing at 30°C on LB agar containing kanamycin (30 ␮g/ml). Expression of cdt genes and purification of recombinant proteins. For gene expression, positive clones were grown overnight in LB broth containing kanamycin (30 ␮g/ml) and rifampin (200 ␮g/ml), diluted 1:20 in fresh broth, and incubated for a further 2 h at 37°C. Gene expression was induced with 1 mM isopropyl-␤-D-thiogalactopyranoside (IPTG) for 6 h at 30°C. Cells were harvested by centrifugation at 6,000 ⫻ g for 20 min and were then resuspended and lysed for 10 min with bacterial permeabilizing reagent (B-PER) protein extraction reagent (Pierce & Warriner Ltd., Cheshire, United Kingdom). The expressed proteins were contained in inclusion bodies. Purification of these inclusion bodies was performed as described by the manufacturer of the B-PER reagent. Briefly, lysates were centrifuged and pellets were resuspended with the

INFECT. IMMUN.

same volume of B-PER-containing lysozyme (100 ␮g/ml), and the lysates were then incubated for a further 5 min at room temperature. A 10⫻ volume of 1:10-diluted B-PER was added to the lysates, and the inclusion bodies were collected by centrifugation at 15,000 ⫻ g for 20 min. After being washed twice with the same volume of 1:10-diluted B-PER, the pellets were treated with 8 M urea containing 300 mM NaCl and 100 mM sodium phosphate buffer (pH 8.0). The recombinant proteins were purified using Ni-nitrilotriacetic acid-agarose columns under denaturing conditions as specified by the manufacturer (Qiagen Ltd.), except that after lysates were loaded onto the column, an additional column wash, consisting of 2.5 mg of polymyxin B/ml in wash buffer, was performed to remove contaminating LPS. The refolding of denatured proteins was performed as described by Takemura et al. (37). Analysis of cell cycle. To measure the cell cycle arrest induced by the Cdt proteins, HEp-2 cells at a density of 2 ⫻ 105 cells/ml were cultured in Dulbecco’s minimal essential medium with 40, 200, or 1,000 ng of recombinant Cdt (rCdt) protein/ml for 1 to 4 days. These concentrations were determined by initial dose-ranging experiments to give a good dose-response relationship between toxin concentration and cell cycle inhibition. At the end of the culture period the HEp-2 cells were washed and fixed for 60 min with 80% cold ethanol. After washing, the cells were stained in the dark at 4°C for 1 h with propidium iodide (10 ␮g/ml) in phosphate-buffered saline containing RNase (1 mg/ml). The data from 2 ⫻ 104 cells were collected on a FACScan flow cytometer (Becton Dickinson). Cell cycle analysis was performed using CellQuest. All experiments examining cell cycle inhibition and cytokine induction were repeated a minimum of three times and gave consistent results. Cytotoxic assay. The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) viable-cell assay was performed as previously described (20). Briefly, HEp-2 cells were plated into 96-well plates at a concentration of 2 ⫻ 104 cells/ well. Various concentrations of the rCdt proteins were added, and the HEp-2 cells were cultured for a further 3 to 6 days. At the end of the culture period a stock MTT solution (20 ␮g/ well) was added to the wells, and the plate contents were incubated for a final 4 h. Acid-isopropanol (100 ␮l of 0.04 N HCl in isopropanol) was added to cells to elute the red formazan, and the solution was mixed thoroughly. The concentration of formazan in solution in each well was determined on a Dynex plate reader by using a test wavelength of 570 nm and a reference wavelength of 620 nm. Assay of cytokine production by PBMCs. Human PBMCs were prepared from normal donor blood by density gradient centrifugation and differential adherence as described by Tabona et al. (36) and were plated at a density of 2.0 ⫻ 106 cells/ml in 24-well plates and stimulated with a graded concentration of rCdt proteins. Initial dose-ranging studies were done to define the most sensible toxin protein concentrations for obtaining good dose-response relationships. To control for LPS contamination in the recombinant proteins, polymyxin B (2 ␮g/ml) was added to each well, the contents of which were then incubated for 20 h. Cytokine synthesis was determined by two-site enzyme-linked immunosorbent assay. The coating and detection antibodies for IL-10, gamma interferon (IFN-␥), and granulocyte-macrophage colony-stimulating factor (GM-CSF)

VOL. 69, 2001

NOTES

5927

FIG. 1. Effect of multiple Cdt complexes on cultured HEp-2 cells. HEp-2 cells were incubated for 3 days in the presence of 1.0 (A), 0.2 (B), or 0.04 (C) ␮g of combined rCdtA, -B, and -C/ml and were heat treated (100°C, 20 min) with 1.0 ␮g of combined rCdtA, -B, and -C/ml (D). Likewise, cells were incubated with 1.0-␮g/ml rCdtA and -B (E), rCdtA and -C (F), rCdtB and -C (G), rCdtA alone (H), rCdtB alone (I), rCdtC alone (J), and medium alone (K). The cells were then analyzed for cell cycle distribution following staining with propidium iodide. The percentages of cells in G0, G1, S, and G2/M phases of the cell cycle are shown. At least 15,000 cells were analyzed per sample.

were from Pharmingen (Oxford, United Kingdom), and those for IL-12 were from BioSource (Watford, United Kingdom). Assays for IL-1␤, IL-6, IL-8, and tumor necrosis factor alpha (TNF-␣) and all cytokine standards were from the National Institute for Biological Standards and Control. Cell supernatants were assayed for the presence of all cytokines using the enzyme-linked immunosorbent assay as previously described (11, 36, 42). In some experiments, cells were stimulated with E. coli LPS (DIFCO) at a concentration of 10 ng/ml and the anti-CD14 antibody MY4 was used as to control for the role of CD14 in cell stimulation. To determine the sensitivity of the Cdt proteins to heat they were boiled for 20 min. Cloning and expression of cdt genes. The cdtA, cdtB, and cdtC genes encoding the mature proteins without signal peptides were amplified by PCR from chromosomal DNA of A. actinomycetemcomitans and were inserted into the pET28a(⫹) N-terminal polyhistidine-tagged fusion vector. These plasmid constructss were introduced into E. coli HMS174(DE3). All Cdt protein products were found in inclusion bodies and were refolded in arginine-containing buffer after being dissolved in 8 M urea. These proteins were then dialyzed against phosphate-buffered saline to remove low-molecular-weight re-

agents, including urea and arginine. All rCdt proteins (rCdtA, rCdtB and rCdtC) were homogenous, as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and migrated with molecular masses of about 26, 32, and 22 kDa, respectively. Ability of rCdt proteins to arrest cell cycle. To determine the contribution of each CDT component in generating biological activity, independently expressed and purified rCdt proteins were analyzed for their effects on cell cycle progression. Preliminary dose-ranging experiments found that exposure of cells to Cdt proteins in the range of 40 to 1,000 ng/ml produced a satisfactory dose-response relationship. As shown in Fig. 1, in control cultures of HEp-2 cells exposed to medium alone, 78% of the cells were in the G0/G1 phase of the cell cycle with a 2n DNA content. None of the rCdt proteins, when added alone, could induce cell cycle arrest in HEp-2 cells. Treatment of HEp-2 cells with a Cdt “complex” containing all three recombinant proteins resulted in 74% of the cells being present in the peak of propidium iodide fluorescence, consisting of cells in G2/M phase. Cultures exposed to Cdt complexes containing rCdtA and rCdtB or containing rCdtA and rCdtC did not demonstrate cell cycle arrest. However, treatment of cells with rCdtB and rCdtC did produce cell cycle arrest.

5928

NOTES

FIG. 2. Effect of multiple Cdt complexes on cell viability. HEp-2 cells were incubated for 3 days in the presence of 1.0 (f) and 10.0 (䡺) ␮g of the multiple complexes of the Cdt proteins per ml. The cytotoxicity of the toxins was then assessed by a colorimetric cytotoxicity assay as described in Materials and Methods. The data are the means and standard deviations of three replicate cultures.

We also examined the cytotoxic effect of CDT on the HEp-2 cells using the MTT cell viability assay (Fig. 2). A mixture containing all three rCdt proteins at 10 ␮g/ml, a concentration able to significantly inhibit cell cycle progression, reduced Hep-2 cell viability by 58.0% after 3 days of culture. Capacity of Cdts to stimulate cytokine production. To investigate the capacity of CDT to stimulate cytokine synthesis, various combinations of the recombinant proteins were incubated with human PBMCs and the release of IL-1␤, IL-6, IL-8, IL-10, IL-12, TNF-␣, IFN-␥, and GM-CSF was measured. To confirm that the effects seen were not due to LPS contamination, the rCdt proteins were heated to 100°C for 20 min or the anti-CD14 monoclonal antibody MY4 was added at a concentration known to block the action of nanogram levels of LPS as defined by Tabona et al. (36). Heat treatment completely abolished the cytokine-inducing activity of the CDTs, but the antiCD14 monoclonal antibody, while able to block LPS, was unable to inhibit the activity of individual or combined rCdt proteins. The effect of these treatments on the production of IL-8 is shown in Fig. 3. Similar results were also found when media supporting the CDT-stimulated cells were assayed for IL-1␤, IL-6, and IFN-␥ (not shown). PBMCs exposed to individual rCdt proteins or to combinations of rCdt proteins failed to produce TNF-␣, IL-10, IL-12, or GM-CSF. In contrast, all three proteins were able to induce IL-1␤, IL-6, and IL-8 production in a dose-dependent manner. CdtC was the most potent and CdtB was the least potent cytokine-inducing agonist. At a concentration of 40 ng/ml, CdtC was able to stimulate human PBMCs to produce nanogram quantities of IL-8. The maximum levels of IL-1␤ produced were on average 1,500 pg/ml, while PBMCs released up to 15,000 pg/ml of IL-6, with the maximal IL-8 production being even higher, reaching about 40,000 pg/ml. The maximum levels of IFN-␥ production were similar to those of IL-1␤ (Table 1). Comparison of the levels of cytokines produced by individual proteins versus those produced by combinations of the rCdt proteins revealed the existence of synergistic interactions. A combination of rCdtB and -C or of rCdtA, -B, and -C stimulated significantly more cytokine synthesis than the sum of the cytokine synthesis produced by cells exposed to the

INFECT. IMMUN.

FIG. 3. Effect of addition of MY4, an anti-human CD14 antibody (␣CD14), or of heat treatment on the capacity of rCdt to induce the secretion of IL-8 from human PBMCs. PBMCs were cultured with 10 ng of LPS/ml or 1 ␮g of CdtA, -B, and -C/ml in the presence or absence of anti-CD14 antibody. Heat treatment was performed at 100°C for 20 min. The results are expressed as the mean and standard deviation of three replicate cultures.

individual toxins. These synergistic interactions occurred with IL-1␤ and IL-6 at toxin concentrations lower than those seen with IL-8. The most striking synergistic interaction was seen with IFN-␥, where the individual recombinant proteins had no capacity to stimulate synthesis of this cytokine. However, the combination of rCdtB and -C or of rCdtA, -B, and -C produced significant amounts of IFN-␥ (Table 1). In contrast to the PBMCs, addition of the various rCdt proteins either singly or in combination failed to stimulate HEp-2 cells to synthesize cytokines. There is still confusion surrounding the biological actions of the proteins that constitute the activity known as CDT. Although there are now two reports that suggest that the cell cycle-blocking activity of CDT is due to the DNase activity of CdtB (5, 16), there is still controversy as to which Cdt proteins possess biological activity. A number of publications, including that of Elwell and Dreyfus (5), have claimed that CdtA and/or CdtC is required for the biological activity of CDT (1, 33, 35). However, Shenker and coworkers claimed that the rCdtB of A. actinomycetemcomitans alone could induce G2/M arrest in human T lymphocytes (31). For the present study we cloned each of the cdt genes of A. actinomycetemcomitans and expressed each protein individually as a polyhistidine-tagged fusion protein. Testing each of these recombinant proteins individually revealed that they could not arrest the progression of the human epithelial cell line HEp-2 through the cell cycle and that at least two components, CdtB and CdtC, were required to block the cells at G2/M and to produce cytotoxicity. This confirms the findings of other studies that have not used recombinant proteins and in which it was not possible to confirm that the individual toxin proteins used were homogeneous. The discrepancy between our results and those of Shenker et al. (31) may relate to differences in the mechanism of uptake of toxins by lymphocytes and epithelial cells. The pathology of the periodontal diseases would seem to be

VOL. 69, 2001

NOTES

TABLE 1. Cytokine-stimulating activity of Cdt Proteins incubated with PBMCsa Activity (pg/ml) for each conc of Cdt protein(s) (␮g/ml)

Cdt protein(s) used

0.04

0.2

1

IL-1␤

A B C AB AC BC ABC

11 ⫾ 1 14 ⫾ 1 64 ⫾ 12 22 ⫾ 2 42 ⫾ 3 58 ⫾ 6 168 ⴞ 15

53 ⫾ 6 50 ⫾ 4 104 ⫾ 9 123 ⫾ 24 198 ⫾ 30 149 ⫾ 13 470 ⴞ 58

462 ⫾ 1 186 ⫾ 26 859 ⫾ 149 515 ⫾ 89 890 ⫾ 262 890 ⫾ 145 1,544 ⫾ 223

IL-6

A B C AB AC BC ABC

223 ⫾ 168 274 ⫾ 51 378 ⫾ 179 200 ⫾ 93 757 ⫾ 217 2,178 ⴞ 167 3,045 ⴞ 630

1,982 ⫾ 414 578 ⫾ 167 4,694 ⫾ 452 3,623 ⫾ 353 5,217 ⫾ 1,024 5,083 ⫾ 1,412 6,337 ⫾ 481

11,148 ⫾ 1,368 5,093 ⫾ 681 11,570 ⫾ 1,893 10,750 ⫾ 964 11,094 ⫾ 2,981 14,287 ⫾ 1,922 14,609 ⫾ 1,395

IL-8

A B C AB AC BC ABC

154 ⫾ 11 63 ⫾ 1 2,546 ⫾ 5 394 ⫾ 73 2,921 ⫾ 66 1,139 ⫾ 239 2,462 ⫾ 644

1,142 ⫾ 59 1,236 ⫾ 167 6,364 ⫾ 6 2,061 ⫾ 359 2,727 ⫾ 1,062 6,384 ⫾ 1,030 6,292 ⫾ 1,454

5,308 ⫾ 989 3,742 ⫾ 382 9,831 ⫾ 172 9,116 ⫾ 656 11,692 ⫾ 2,594 22,879 ⴞ 2,105 39,243 ⴞ 5,753

IFN-␥

A B C AB AC BC ABC

NDb ND ND ND ND ND ND

ND ND ND ND ND ND 244 ⴞ 28

ND ND ND ND ND 181 ⴞ 147 1,121 ⴞ 88

Cytokine assayed

a Results are expressed as mean ⫾ standard deviation of cytokine concentration in picograms per milliliter. Boldfaced numerals indicate synergistic responses between Cdt proteins. b ND, cytokine not detected by immunoassay.

driven by proteins secreted by periodontopathogens, such as A. actinomycetemcomitans, that can promote the stimulation of specific cytokine networks and thus produce a specific inflammatory pathology (4, 12, 43, 44). Previous studies of A. actinomycetemcomitans have revealed that it produces an LPS with weak cytokine-inducing activity (25) and a small peptide with the ability to stimulate human gingival fibroblasts to secrete IL-6 without promoting the synthesis of the key proinflammatory cytokines IL-1␤ and TNF-␣ (26, 27). In the present study, we have now demonstrated that the rCdt proteins of A. actinomycetemcomitans are able to induce human PBMCs to release a specific network of cytokines. The release of cytokines was not due to contaminating LPS in the rCdt proteins, which had been removed during purification of the proteins on nickel columns. This was confirmed by the finding that the activity of the Cdts could be nullified by heating but not by addition of an antibody that binds to CD14 and inhibits the cytokine-inducing activity of LPS (36). It was of interest to find that while the Cdts, either alone or in combination, could stimulate human PBMCs to synthesize IL-1␤, IL-6, IL-8, and IFN-␥, this toxin could not induce the synthesis of the proinflammatory cytokines TNF-␣, IL-12, and GM-CSF and the anti-inflammatory cytokine IL-10. This is an unusual mode of cytokine network

5929

stimulation, as most bacterial stimulators either induce both IL-1␤ and TNF-␣ (8) or, more rarely, fail to induce either of these so-called early-response cytokines (27). Our cytokine assays demonstrated that treatment of human PBMCs with CdtA or CdtC alone could stimulate secretion of cytokines but that CdtB was a weaker cytokine-stimulating agonist. However, while CdtB had only minimal cytokine-stimulating activity, it appeared to synergize with CdtA and CdtC to promote PBMC cytokine synthesis. Synergy was most marked with IFN-␥ synthesis, for which the individual toxin components were inactive, and only the combination of rCdtB and rCdtC or of all three proteins was able to induce cytokine synthesis. This suggests that CdtB may interact with CdtA and CdtC at the surface of the target cell and induce a greater intracellular signal. The nature of the receptors and of the signaling pathways utilized by CDT to stimulate human leukocytes to synthesize cytokines is unknown but, given the very unusual cytokine network induced by this toxin, is likely to be different from the nature of those involved in responses to known bacterial cytokine stimulants, such as LPS, peptidoglycan, and other bacterial toxins (9). One explanation for the unusual cytokine network produced may be the fact that some or all of the Cdt proteins undergo endocytosis by target cells (2). An obvious question is whether there is any relationship between cell cycle blockade and stimulation of cytokine synthesis. Given that the human PBMCs, which produce cytokines in response to CDT, are noncycling cells, while HEp-2 cells are blocked in G2 by CDT but do not produce cytokines, the answer would seem to be no. Thus, CDT is a toxin which can both inhibit the proliferation of cells and induce the production of cytokines from noncycling human leukocytes—activities which are likely to contribute to the pathogenesis of conditions associated with colonization by A. actinomycetemcomitans. The costs of this study were not supported by any funding agency. We are grateful to T. Nishihara and T. Koseki for the gift of A. actinomycetemcomitans strainY4. REFERENCES 1. Cope, L. D., S. Lumbley, J. L. Latimer, J. Klesney-Tait, M. K. Stevens, L. S. Johnson, M. Purven, R. S. Munson, Jr., T. Lagergard, J. D. Radolf, and E. J. Hansen. 1997. A diffusible cytotoxin of Haemophilus ducreyi. Proc. Natl. Acad. Sci. USA 94:4056–4061. 2. Cortes-Bratti, X., E. Chaves-Olarte, T. Lagergard, and M. Thelestam. 1999. The cytolethal distending toxin from the chancroid bacterium Haemophilus ducreyi induces cell-cycle arrest in the G2 phase. J. Clin. Investig. 103:107– 115. 3. Cortes-Bratti, X., C. Karlsson, T. Lagergard, M. Thelestam, and T. Frisan. 2001. The Haemophilus ducreyi cytolethal distending toxin induces cell cycle arrest and apoptosis via the DNA damage checkpoint pathways. J. Biol. Chem. 276:5296–5302. 4. Dongari-Bagtzoglou, A. I., and J. L. Ebersole. 1996. Production of inflammatory mediators and cytokines by human gingival fibroblasts following bacterial challenge. J. Periodontal Res. 31:90–98. 5. Elwell, C. A., and L. A. Dreyfus. 2000. DNase I homologous residues in CdtB are critical for cytolethal distending toxin-mediated cell cycle arrest. Mol. Microbiol. 37:952–963. 6. Fives-Taylor, P. M., D. H. Meyer, J. P. Mintz, and C. Brissette. 1999. Virulence factors of Actinobacillus actinomycetemcomitans. Periodontol. 2000 20:136–167. 7. Henderson, B. 2000. Therapeutic control of cytokines: lessons from microorganisms, p. 243–261. In G. A. Higgs and B. Henderson (ed.), Novel cytokine inhibitors. Birkhauser, Basel, Switzerland. 8. Henderson, B., M. Wilson, and J. Hyams. 1998. Cellular microbiology: cycling into the millennium. Trends Cell Biol. 8:384–387. 9. Henderson, B., S. Poole, and M. Wilson. 1998. Bacteria-cytokine interactions

5930

NOTES

in health and disease. Portland Press, London, United Kingdom. 10. Henderson, B., P. Tabona, S. Poole, and S. P. Nair. 2001. Cloning and expression of the Actinobacillus actinomycetemcomitans thioredoxin (trx) gene and assessment of cytokine inhibitory activity. Infect. Immun. 69:154– 158. 11. Hickey, T. E., A. L. McVeigh, D. A. Scott, R. E. Michielutti, A. Bixby, S. A. Carroll, A. L. Bourgeois, and P. Guerry. 2000. Campylobacter jejuni cytolethal distending toxin mediates release of interleukin-8 from intestinal epithelial cells. Infect. Immun. 68:6535–6541. 12. Jiang, Y., T. R. Russell, D. T. Graves, H. Cheng, S. H. Nong, and S. M. Levitz. 1996. Monocyte chemoattractant protein 1 and interleukin-8 production in mononuclear cells stimulated by oral microorganisms. Infect. Immun. 64:4450–4455. 13. Johnson, W. M., and H. Lior. 1988. A new heat-labile cytolethal distending toxin (CLDT) produced by Escherichia coli isolates from clinical material. Microb. Pathog. 4:103–113. 14. Kamin, S., W. Harvey, M. Wilson, and A. Scutt. 1986. Inhibition of fibroblast proliferation and collagen synthesis by capsular material from Actinobacillus actinomycetemcomitans. J. Med. Microbiol. 22:245–249. 15. Kirby, A. C., S. Meghji, S. P. Nair, P. A. White, K. Reddi, T. Nishihara, K. Nakashima, A. C. Willis, R. Sim, M. Wilson, and B. Henderson. 1995. The potent bone resorbing mediator of Actinobacillus actinomycetemcomitans is homologous to the molecular chaperone, groEL. J. Clin. Investig. 96:1185– 1194. 16. Korostoff, J., J. F. Wang, I. Kieba, M. Miller, B. J. Shenker, and E. T. Lally. 1998. Actinobacillus actinomycetemcomitans leukotoxin induces apoptosis in HL-60 cells. Infect. Immun. 66:4474–4483. 17. Lara-Tejero, M., and J. E. Galan. 2000. A bacterial toxin that controls cell cycle progression as a deoxyribonuclease I-like protein. Science 290:354–357. 18. Mayer, M. P., L. C. Bueno, E. J. Hansen, and J. M. DiRienzo. 1999. Identification of a cytolethal distending toxin gene locus and features of a virulence-associated region in Actinobacillus actinomycetemcomitans. Infect. Immun. 67:1227–1237. 19. Meyer, D., and P. M. Fives-Taylor. 1997. The role of Actinobacillus actinomycetemcomitans in the pathogenesis of periodontal disease. Trends Microbiol. 5:224–228. 20. Muro, M., T. Koseki, S. Akifusa, S. Kato, Y. Kowashi, Y. Ohsaki, K. Yamato, M. Nishijima, and T. Nishihara. 1997. Role of CD14 molecules in internalization of Actinobacillus actinomycetemcomitans by macrophages and subsequent induction of apoptosis. Infect. Immun. 65:1147–1151. 21. Oguchi, M., A. Ishisaki, N. Okahashi, M. Koide, T. Koseki, K. Yamato, T. Noguchi, and T. Nishihara. 1998. Actinobacillus actinomycetemcomitans toxin induces both cell cycle arrest in the G2/M phase and apoptosis. Infect. Immun. 66:5980–5987. 22. Okuda, J., M. Fukumoto, Y. Takeda, and M. Nishibuchi. 1997. Examination of diarrheagenicity of cytolethal distending toxin: suckling mouse response to the products of the cdtABC genes of Shigella dysenteriae. Infect. Immun. 65:428–433. 23. Pickett, C. L., D. L. Cottle, E. C. Pesci, and G. Bikah. 1994. Cloning, sequencing, and expression of the Escherichia coli cytolethal distending toxin genes. Infect. Immun. 62:1046–1051. 24. Pickett, C. L., and C. A. Whitehouse. 1999. The cytolethal distending toxin family. Trends Microbiol. 7:292–297. 25. Reddi, K., M. Wilson, S. Poole, S. Meghji, and B. Henderson. 1995. Relative cytokine-stimulating activities of surface components of the oral periodontopathic bacterium Actinobacillus actinomycetemcomitans. Cytokine 7:534– 541. 26. Reddi, K., M. Wilson, S. P. Nair, S. Poole, and B. Henderson. 1996. Comparison of the pro-inflammatory cytokine-stimulating activity of the surfaceassociated proteins of periodontopathic bacteria. J. Periodontal Res. 31:120– 130. 27. Reddi, K., S. P. Nair, P. A. White, S. Hodges, P. Tabona, S. Meghji, S. Poole, M. Wilson, and B. Henderson. 1996. Surface-associated material from the bacterium Actinobacillus actinomycetemcomitans contains a peptide which, in

Editor: J. T. Barbieri

INFECT. IMMUN.

28. 29. 30.

31.

32. 33.

34. 35.

36.

37.

38. 39.

40. 41. 42.

43. 44. 45.

contrast to lipopolysaccharide, directly stimulates fibroblast interleukin-6 gene transcription. Eur. J. Biochem. 236:871–876. Scott, D. A., and J. B. Kaper. 1994. Cloning and sequencing of the genes encoding Escherichia coli cytolethal distending toxin. Infect. Immun. 62:244– 251. Shenker, B. J. 1987. Immunologic dysfunction in the pathogenesis of periodontal diseases. J. Clin. Periodontol. 14:489–498. Shenker, B. J., T. McKay, S. Datar, M. Miller, R. Chowhan, and D. Demuth. 1999. Actinobacillus actinomycetemcomitans immunosuppressive protein is a member of the family of cytolethal distending toxins capable of causing a G2 arrest in human T cells. J. Immunol. 162:4773–4780. Shenker, B. J., R. H. Hoffmaster, T. L. McKay, and D. R. Demuth. 2000. Expression of the cytolethal distending toxin (Cdt) operon in Actinobacillus actinomycetemcomitans: evidence that the CdtB protein is responsible for G2 arrest of the cell cycle in human T cells. J. Immunol. 165:2612–2618. Slots, J., and M. Ting. 1999. Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis in human periodontal disease: occurrence and treatment. Periodontol. 2000 20:82–121. Stevens, M. K., J. L. Latimer, S. R. Lumbley, C. K. Ward, L. D. Cope, T. Lagergard, and E. J. Hansen. 1999. Characterization of a Haemophilus ducreyi mutant deficient in expression of cytolethal distending toxin. Infect. Immun. 67:3900–3908. Suda, T., I. Nakamura, E. Jimi, and N. Takahashi. 1997. Regulation of osteoclast function. J. Bone Miner. Res. 12:869–879. Sugai, M., T. Kawamoto, S. Y. Peres, Y. Ueno, H. Komatsuzawa, T. Fujiwara, H. Kurihara, H. Suginaka, and E. Oswald. 1998. The cell cycle-specific growth-inhibitory factor produced by Actinobacillus actinomycetemcomitans is a cytolethal distending toxin. Infect. Immun. 66:5008–5019. Tabona, P., K. Reddi, S. Khan, S. P. Nair, S. J. Crean, S. Meghji, M. Wilson, M. Preuss, A. D. Miller, S. Poole, S. Carne, and B. Henderson. 1998. Homogeneous Escherichia coli chaperonin 60 induces IL-1 beta and IL-6 gene expression in human monocytes by a mechanism independent of protein conformation. J. Immunol. 161:1414–1421. Takemura, S., R. Asano, K. Tsumoto, S. Ebara, N. Sakurai, Y. Katayose, H. Kodama, H. Yoshida, M. Suzuki, K. Imai, S. Matsuno, T. Kudo, and I. Kumagai. 2000. Construction of a diabody (small recombinant bispecific antibody) using a refolding system. Protein Eng. 13:583–588. van Winkelhoff, A. J., and J. Slots. 1999. Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis in nonoral infections. Periodontol. 2000 20:122–135. White, P. A., M. Wilson, S. P. Nair, A. C. Kirby, K. Reddi, and B. Henderson. 1995. Characterization of an antiproliferative surface-associated protein from Actinobacillus actinomycetemcomitans which can be neutralized by sera from a proportion of patients with localized juvenile periodontitis. Infect. Immun. 63:2612–2618. White, P. A., M. Patel, S. P. Nair, J. Ashmore, P. Galgut, M. Wilson, B. Henderson, and I. Olsen. 1998. Control of the human cell cycle by a bacterial protein, gapstatin. Eur. J. Cell Biol. 77:228–238. Whitehouse, C. A., P. B. Balbo, E. C. Pesci, D. L. Cottle, P. M. Mirabito, and C. L. Pickett. 1999. Campylobacter jejuni cytolethal distending toxin causes a G2-phase cell cycle block. Infect. Immun. 66:1934–1940. Williams, R. J., J. M. Ward, B. Henderson, S. Poole, B. P. O’Hara, M. Wilson, and S. P. Nair. 2000. Identification of a novel gene cluster encoding staphylococcal exotoxin-like proteins: characterization of the prototypic gene and its protein product, SET1. Infect. Immun. 68:4407–4415. Wilson, M., and B. Henderson. 1995. Virulence factors of Actinobacillus actinomycetemcomitans relevant to the pathogenesis of inflammatory periodontal diseases. FEMS Microbiol. Rev. 17:365–379. Wilson, M., K. Reddi, and B. Henderson. 1996. Cytokine-inducing components of periodontopathogenic bacteria. J. Periodontal Res. 31:393–407. Young, V. B., K. A. Knox, and D. B. Schauer. 2000. Cytolethal distending toxin sequence and activity in the enterohepatic pathogen Helicobacter hepaticus. Infect. Immun. 68:184–191.