- Email: [email protected]
Cytolethal distending toxins and activation of DNA damage-dependent checkpoint responses
IJMM IJ
Int. J. Med. Microbiol. 291, 495-499 (2002) © Urban & Fischer Verlag http://www.urbanfischer.de/journals/ijmm
Cytolethal distending toxins and activation of DNA damage-dependent checkpoint responses Teresa Frisan, Ximena Cortes-Bratti, Monica Thelestam Microbiology and Tumorbiology Center, Karolinska Institutet, S-17177 Stockholm, Sweden
Abstract Cytolethal distending toxins (CDTs) are unique among bacterial protein toxins in their ability to cause DNA damage, due to their functional similarity to the mammalian deoxyribonuclease I (DNase I). The cellular response to CDT intoxication is characterised by activation of DNA damage-induced checkpoint responses, and the final outcome is cell type dependent. Cells of epithelial origin and normal keratinocytes are arrested in the G2 phase of the cell cycle, normal fibroblasts are also arrested in G1, while B cells die of apoptosis. CDTs are encoded by three linked genes (cdtA, cdtB and cdtC), and CdtB is the toxin subunit which possesses the DNase I-like activity. All the three genes have to be present in the bacterium in order to produce an active cytotoxin, however cytotoxic Haemophilus ducreyi CDT, purified from a CdtABC recombinant E. coli strain, contains the CdtB and CdtC subunits, suggesting that they constitute the holotoxin and that CdtC may be required for CdtB internalization. The role of the CdtA subunit is currently unknown, but it might modify and therefore activate CdtC. This review will focus on the cellular responses induced by CDTs in mammalian cells. Key words: cytolethal distending toxin – DNase – DNA damage – Mre11 DNA repair complex – holotoxin
Introduction The cytolethal distending toxins (CDTs) belong to a novel bacterial protein toxin family, with the unique ability to induce cell cycle arrest or apoptosis in a number of mammalian cells. Only recently we have begun to understand the mode of action and the cellular target of CDTs. The first CDT was detected in 1987 in strains of Escherichia coli. Intoxicated cells showed marked cell distension evident 96–120 h after treatment with culture supernatants, resulting in cell death (Johnson and Lior, 1987). To date, several gram-negative bacteria, such as Haemophilus ducreyi, Actinobacillus actinomycetemcomitans, Campylobacter sp., Shigella dysenteriae and Helicobacter hepaticus, have been shown to produce CDTs (Cope et al., 1997; Mayer et al., 1999;
Okuda et al., 1997; Pickett et al., 1996; Sugai et al., 1998; Young et al., 2000). The CDT activity is encoded by three linked genes, designated as cdtA, cdtB and cdtC. Expression of all the three genes is required to produce an active CDT (reviewed in (Pickett and Whitehouse, 1999)). The cdtB gene shows the highest homology among all the different species of CDT-producing bacteria, while the cdtA and cdtC present more variability. This suggested that CdtB might be the protein carrying the putative enzymatic activity required to damage the cellular target. CDT toxicity can be reconstituted by combining noncytotoxic cell sonicates from recombinant bacterial strains expressing individually the CdtA, CdtB and CdtC proteins (Elwell et al., 2001; Frisk et al., 2001). The importance of the CdtC subunit as component of the Haemophilus ducreyi CDT (HdCDT) holotoxin is
Corresponding author: Teresa Frisan, Microbiology and Tumorbiology Center, Karolinska Institutet, Box 280, S-17177 Stockholm, Sweden, Phone: +46 87 28 71 80, Fax: +4 68 34 26 51, E-mail: [email protected] 1438-4221/01/291/6-7-495 $ 15.00/0
496
T. Frisan et al.
demonstrated by the ability of a CdtC-specific monoclonal antibody to neutralise the cytotoxic activity of HdCDT (Cope et al., 1997). Studies on the A. actinomycetemcomitans CDT (AaCDT), however, suggested that AaCdtB alone might be sufficient to cause G2 arrest; in human T cells. Despite this observation the authors suggested that AaCdtC can also play a role in the cell cycle arrest and absence of AaCdtC was not confirmed by Western-blot analysis (Shenker et al., 1999, 2000). This review will focus on the effect of CDTs on mammalian cells. For more details on the CDT nomenclature and the involvement of CDTs in pathogenicity the reader is referred to (Cortes-Bratti et al., 2001b).
CDT activates checkpoint responses One of the earliest effect of CDTs demonstrated on cell lines was the induction of cell cycle arrest in G2 and this was associated with the accumulation of the inactive hyperphosphorylated form of the cyclin-dependent kinase cdc2 (reviewed in (Pickett and Whitehouse, 1999)). When a broader panel of cell lines were tested, it became clear to us that the effects of the HdCDT were cell type dependent. Epithelial cell lines, such as HeLa, HEp-2, and normal keratinocytes were indeed arrested in the G2 phase of the cell cycle. However, primary human fibroblasts from foreskin and embryonic lung were arrested also in G1, while cell lines of B cell origin rapidly underwent apoptosis (Cortes-Bratti et al., 2001a). This observation suggested that HdCDT, and possibly the other CDTs, activates checkpoint responses induced by DNA damage. Indeed the type of cell cycle arrest induced by HdCDT was similar to that induced by ionizing radiation (IR) in HEp-2 cells and in normal fibroblasts. In fibroblasts, which were arrested both in G1 and G2, p53 was stabilised and phosphorylated on serine 15 with the same kinetics in intoxicated as in irradiated cells. This was associated with increased expression of the p53regulated cyclin-dependent kinase inhibitor p21 and with upregulation of p27. In HEp-2 cells, early modification of p53 and p21 levels was not detected. However, in this cell type the HdCDT-induced G2 arrest was associated with an accumulation of the phosphorylated form of the chk2 kinase, known to play a central role in activation of the G2 checkpoint response. Induction of different responses in a cell type-dependent manner suggested to us that HdCDT acted at the level of protein(s) involved in the control of checkpoint responses, which prevent progression through the cell cycle until the DNA damage has been repaired, thus avoiding genetic instability. The phosphatidylinositol (PI)-related kinase Ataxia Telangiectasia Mutated (ATM) and its homologue ATR are key molecules in sensing DNA damage. ATM and/or ATR activate
checkpoint responses at different stages of the cell cycle: the G1/S transition (G1 checkpoint), the S phase progression, and the G2/M boundary (G2/M checkpoint) (reviewed in (Shiloh, 2001)). The early molecular response to HdCDT was ATM dependent, since the induction of cell death upon intoxication was delayed in ATM-deficient B cell lines, compared to ATM wildtype cells, and the delay in cell death was associated with a delayed kinetics of p53 stabilisation (CortesBratti et al., 2001b). These findings limited the target candidates to DNA or molecules, acting upstream ATM and directly involved in activation of checkpoint responses. A schematic representation of the cellular responses activated by HdCDT is depicted in Figure 1.
DNase I-like activity of CdtB and DNA damage Recent data indicate that the human Mre11 complex, constituted by the Mre11, Rad50 and Nbs1 proteins, acts as possible sensor of DNA damage. It is well established that the Mre11 complex associates with damaged DNA and forms discrete nuclear foci after induction of double strand breaks (DSBs) following irradiation (reviewed in (Petrini, 1999)). We observed the same pattern of Mre11 re-localisation upon HdCDT treatment of B cell lines, HeLa cells, and foreskin fibroblasts (Frisan, T. et al., submitted, Figure 2), strongly suggesting that the primary toxin target is DNA. These data complement and expand those demonstrating that CdtB from C. jejuni, E. coli, and H. ducreyi presents structural and functional homology with the mammalian DNase I (Elwell and Dreyfus, 2000; Frisk et al., 2001; Lara-Tejero and Galan, 2000). Crude CdtB preparations from E. coli and H. ducreyi possessed DNase activity as detected by in vitro digestion of a DNA plasmid substrate (Elwell and Dreyfus, 2000; Frisk et al., 2001). Expression of CdtB either by transfection of the C. jejuni cdtB gene or by electroporation of the E. coli purified protein was sufficient to induce a slowly appearing nuclear fragmentation and a marked chromatin disruption (Elwell et al., 2001; Lara-Tejero and Galan, 2000). In both cases the DNase activity, as well as the induction of cell cycle arrest, was abolished by point mutations of conserved residues required for catalysis or for magnesium binding, indicating that the cytotoxic effects are related to the enzymatic activity. Upon microinjection, C. jejuni CdtB (CjCdtB) was able to produce the same effects as the CdtABC toxin, suggesting that CdtB alone, when present intracellularly can induce cell cycle arrest. However, CjCdtB alone added extracellularly did not induce a cell cycle block
Cytolethal distending toxins induce DNA damage
497
Fig. 1. Schematic representation of the mode of CDTs action. CDT-induced DNA damage causes: i) activation of the ATM kinase and downstream checkpoint responses resulting in G2 and G1/G2 arrest, or apoptosis depending on the cell type; ii) re-localization of the Mre11 complex.
in HeLa cells (Lara-Tejero and Galan, 2000). We observed similar results upon microinjection of a Histagged recombinant H. ducreyi CdtB subunit into HeLa cells (Frisan, T. et al., submitted).
Holotoxin composition The HdCDT preparation used to investigate the cellular responses in mammalian cells contained as major components CdtB and CdtC as detected by Westernblot analysis. Traces of a shorter polypeptide (but not the full-length protein) reacting with the anti-CdtA monoclonal antibody could be detected only in overexposed membranes (Frisan, T. et al., submitted). Since the CdtB subunit alone is not sufficient to induce either cell cycle arrest or cell distention when provided extracellularly, we propose that CdtC may be required to promote internalization of the enzymatic component and that CDTs act as an AB toxin. However it is important to stress that all the three genes have to be present in the producing bacterium. The HdCDT holotoxin needs to be internalized by endocytosis via clathrin-coated pits and it requires an intact Golgi apparatus in order to intoxicate HEp-2 cells (Cortes-Bratti et al., 2000). Most probably the
Fig. 2. HdCDT causes re-localization of the Mre11 complex and induction of Mre11 foci. The lymphoblastoid B cell line SN-B1 (Cortes-Bratti et al., 2001a), the epithelial line HeLa and human foreskin fibroblasts were treated with HdCDT (2 mg/ml) for 8 h. Mre11 staining was performed as previously described (Maser et al., 1997).
498
T. Frisan et al.
toxin (or at least the CdtB component) is then transported from the Golgi to the ER, as demonstrated for many bacterial toxins targeting cytosolic molecules (Lord and Roberts, 1998), and therefrom delivered directly or indirectly to the nucleus (Figure 3). The role of CdtA still remains to be elucidated. Frisk et al. (2001) showed that that the isoelectric point (pI) of the CdtC component is 1.5 pH units higher in recombinant strains expressing all three components than in recombinant strains expressing the CdtC protein alone. A similar change of pI occurred after mixing the three individual components in vitro. Therefore, it is possible that CdtA is required to modify and activate CdtC, in analogy with the situation demonstrated for other bacterial toxins (reviewed in (Lally et al., 1999)).
Conclusions The past year has witnessed a significant development in the field of the cytolethal distending toxins. However, several aspects of CDT-mediated intoxication need to be further elucidated, such as the precise role of the CdtA subunit, and a more detailed definition of how CdtC participates in the toxin internalization. It is noteworthy that bacterial toxins have been extremely useful in the study of different aspects of cell biology (Aktories, 2000). Since CDTs are the only bacterial protein toxins described so far to act at the nuclear level (causing DNA damage), they may constitute a useful tool to study protein transport from the Golgi/ ER compartment to the nucleus. Furthermore, the CDT-mediated cell distension and promotion of the
Fig. 3. Proposed model for the HdCDT holotoxin and schematic representation of the HdCDT internalization pathway. Our hypothesis for the production of an active HdCDT holotoxin is the following: CdtC requires activation, possibly via CdtA, to CdtC*. CdtC* and CdtB may constitute an AB toxin, where CdtB is the enzymatic component and CdtC* could be involved in CdtB internalization. HdCDT is internalized via clathrin-dependent endocytosis and requires an intact Golgi apparatus in order to be active on the target cells. CDT translocation to the ER and its direct or indirect transport to the nucleus have still to be demonstrated; therefore, they are indicated as dotted arrows.
Cytolethal distending toxins induce DNA damage
actin cytoskeleton may be used to dissect the still poorly understood cross-talk between molecules that control the cytoskeleton and those that control cell cycle progression/arrest. Bacterial toxins can also be used for therapeutic purposes, e. g. the use of immunotoxins as an alternative or complement in cancer treatment, or the use of toxins acting as vehicles for the delivery of specific proteins into cells (Kreitman, 1999; Sandvig and Van Deurs, 2000). The interference of CDT with the cell cycle makes it a potentially good candidate for an anti-tumor agent, provided that the toxin can be selectively delivered to tumour cells. Acknowledgements. This work was supported by the Swedish Medical Research Council (05969), the Karolinska Institutet and the Stiftelsen Lars Hiertas Minne, Stockholm, Sweden. T. Frisan is supported by the Swedish Society for Medical Research.
References Aktories, K.: Bacterial protein toxins as tools in cell biology and pharmacology. In: Cellular microbiology (P. Cossart, P. Boquet, S. Normak, R. Rappuoli, eds.), pp. 221–237. ASM press (American Society for Microbiology), Washington DC 2000. Cope, L. D., Lumbley, S., Latimer, J. L., Stevens, M. K., Johnson, L. S., Purvén, M., Munson, R. S., Lagergård, T., Radolf, J., Hansen, E. J.: A diffusible cytotoxin of Haemophilus ducreyi. Proc. Natl. Acad. Sci. USA 94, 4056– 4061 (1997). Cortes-Bratti, X., Chaves-Olarte, E., Lagergård, T., Thelestam, M.: Cellular internalization of cytolethal distending toxin from Haemophilus ducreyi. Infect. Immun. 68, 6903–6911 (2000). Cortes-Bratti, X., Karlsson, C., Lagergård, T., Thelestam, M., Frisan, T.: The Haemophilus ducreyi cytolethal distending toxin induces cell cycle arrest and apoptosis via the DNA damage checkpoint pathways. J. Biol. Chem. 276, 5296–5302 (2001a). Cortes-Bratti, X., Frisan, T., Thelestam, M.: The cytolethal distending toxins induce DNA damage and cell cycle arrest. Toxicon 39, 1729–1736 (2001b). Elwell, C., Chao, K., Patel, K., Dreyfus, L.: Escherichia coli CdtB mediates cytolethal distending toxin cell cycle arrest. Infect. Immun. 69, 3418–3422 (2001). Elwell, C. A., Dreyfus, L. A.: DNAase I homologous residues in CdtB are critical for cytolethal distending toxinmediated cell cycle arrest. Mol. Microbiol. 37, 952–963 (2000). Frisk, A., Lebens, M., Johansson, C., Ahmed, H., Svensson, L., Ahlman, K., Lagergård, T.: The role of different protein components from the Haemophilus ducreyi cytolethal distending toxin in generation of cell toxicity. Microb. Pathog. 30, 313–324 (2001). Johnson, W. M., Lior, H.: Response of Chinese hamster ovary cells to a cytolethal distending toxin (CDT) of Escherichia coli and possible misinterpretation as heat-labile (LT) enterotoxin. FEMS Microbiol. Lett. 43, 19–23 (1987).
499
Kreitman, R. J.: Immunotoxins in cancer therapy. Curr. Opin. Immunol. 11, 570–578 (1999). Lally, E. T., Hill, R. B., Kieba, I. R., Korostoff, J.: The interaction between RTX toxins and target cells. Trends Microbiol. 7, 356–361 (1999). Lara-Tejero, M., Galan, J. E.: A bacterial toxin that controls cell cycle progression as a deoxyribonuclease I-like protein. Science 290, 354–357 (2000). Lord, M. J., Roberts, L. M.: Toxin entry: retrograde transport through the secretory pathway. J. Cell Biol. 140, 733–736 (1998). Maser, R. S., Monsen, K. J., Nelms, B. E., Petrini, J. H. J.: hMre11 and hRad50 nuclear foci are induced during normal cellular response to DNA double-strand breaks. Mol. Cell. Biol. 17, 6087–6096 (1997). Mayer, M. P., Bueno, L., Hansen, E. J., Di Rienzo, J. M.: Identification of a cytolethal distending toxin gene locus and features of a virulence-associated region in Actinobacillus actinomycetemcomitans. Infect. Immun. 67, 1227–1237 (1999). Okuda, J., Fukumoto, M., Takeda, Y., Nishibuchi, M.: Examination of diarrheagenicity of cytolethal distending toxin: suckling mouse response to the products of the cdt ABC genes of Shigella dysenteriae. Infect. Immun. 65, 428–433 (1997). Petrini, J. H. J.: DNA repair ‘99. The mammalian Mre11Rad50-Nbs1 protein complex: integration of functions in the cellular DNA-damage response. Am. J. Hum. Genet. 64, 1264–1269 (1999). Pickett, C. L., Pesci, E. C., Cottle, D. L., Russell, G., Erdem, A. N., Zeytin, H.: Prevalence of cytolethal distending toxin production in Campylobacter jejuni and relatedness of Campylobacter sp. cdtB gene. Infect. Immun. 64, 2070– 2078 (1996). Pickett, C. L., Whitehouse, C. A.: The cytolethal distending toxin family. Trends Microbiol. 7, 292–297 (1999). Sandvig, K., Van Deurs, B.: Entry of ricin and Shiga toxin into cells: molecular mechanisms and medical perspectives. EMBO J. 19, 5943–5950 (2000). Shenker, B. J., Hoffmaster, R. H., McKay, T. L., Demuth, D. R.: Expression of the cytolethal distending toxin (Cdt) operon in Actinobacillus actinomycetemcomitans: evidence that CdtB protein is responsible for G2 arrest of the cell cycle in human T cells. J. Immunol. 165, 2612–2618 (2000). Shenker, B. J., McKay, T., Datar, S., Miller, M., Chowhan, R., Demuth, D.: 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 (1999). Shiloh, Y.: ATM and ATR: networking cellular responses to DNA damage. Curr. Opin. Genet. Dev. 11, 71–77 (2001). Sugai, M., Kawamoto, T., Peres, S. Y., Ueno, Y., Komatsuzawa, H., Fujiwara, T., Kurihara, H., Suginaka, H., Oswald, E.: The cell cycle-specific growth-inhibitory factor produced by Actinobacillus actinomycetemcomitans is a cytolethal distending toxin. Infect. Immun. 66, 5008– 5019 (1998). Young, V. B., Knox, K. A., Schauer, D. B.: Cytolethal distending toxin sequence and activity in the enterohepatic pathogen Helicobacter hepaticus. Infect. Immun. 68, 184–191 (2000).
Int. J. Med. Microbiol. 291, 495-499 (2002) © Urban & Fischer Verlag http://www.urbanfischer.de/journals/ijmm
Cytolethal distending toxins and activation of DNA damage-dependent checkpoint responses Teresa Frisan, Ximena Cortes-Bratti, Monica Thelestam Microbiology and Tumorbiology Center, Karolinska Institutet, S-17177 Stockholm, Sweden
Abstract Cytolethal distending toxins (CDTs) are unique among bacterial protein toxins in their ability to cause DNA damage, due to their functional similarity to the mammalian deoxyribonuclease I (DNase I). The cellular response to CDT intoxication is characterised by activation of DNA damage-induced checkpoint responses, and the final outcome is cell type dependent. Cells of epithelial origin and normal keratinocytes are arrested in the G2 phase of the cell cycle, normal fibroblasts are also arrested in G1, while B cells die of apoptosis. CDTs are encoded by three linked genes (cdtA, cdtB and cdtC), and CdtB is the toxin subunit which possesses the DNase I-like activity. All the three genes have to be present in the bacterium in order to produce an active cytotoxin, however cytotoxic Haemophilus ducreyi CDT, purified from a CdtABC recombinant E. coli strain, contains the CdtB and CdtC subunits, suggesting that they constitute the holotoxin and that CdtC may be required for CdtB internalization. The role of the CdtA subunit is currently unknown, but it might modify and therefore activate CdtC. This review will focus on the cellular responses induced by CDTs in mammalian cells. Key words: cytolethal distending toxin – DNase – DNA damage – Mre11 DNA repair complex – holotoxin
Introduction The cytolethal distending toxins (CDTs) belong to a novel bacterial protein toxin family, with the unique ability to induce cell cycle arrest or apoptosis in a number of mammalian cells. Only recently we have begun to understand the mode of action and the cellular target of CDTs. The first CDT was detected in 1987 in strains of Escherichia coli. Intoxicated cells showed marked cell distension evident 96–120 h after treatment with culture supernatants, resulting in cell death (Johnson and Lior, 1987). To date, several gram-negative bacteria, such as Haemophilus ducreyi, Actinobacillus actinomycetemcomitans, Campylobacter sp., Shigella dysenteriae and Helicobacter hepaticus, have been shown to produce CDTs (Cope et al., 1997; Mayer et al., 1999;
Okuda et al., 1997; Pickett et al., 1996; Sugai et al., 1998; Young et al., 2000). The CDT activity is encoded by three linked genes, designated as cdtA, cdtB and cdtC. Expression of all the three genes is required to produce an active CDT (reviewed in (Pickett and Whitehouse, 1999)). The cdtB gene shows the highest homology among all the different species of CDT-producing bacteria, while the cdtA and cdtC present more variability. This suggested that CdtB might be the protein carrying the putative enzymatic activity required to damage the cellular target. CDT toxicity can be reconstituted by combining noncytotoxic cell sonicates from recombinant bacterial strains expressing individually the CdtA, CdtB and CdtC proteins (Elwell et al., 2001; Frisk et al., 2001). The importance of the CdtC subunit as component of the Haemophilus ducreyi CDT (HdCDT) holotoxin is
Corresponding author: Teresa Frisan, Microbiology and Tumorbiology Center, Karolinska Institutet, Box 280, S-17177 Stockholm, Sweden, Phone: +46 87 28 71 80, Fax: +4 68 34 26 51, E-mail: [email protected] 1438-4221/01/291/6-7-495 $ 15.00/0
496
T. Frisan et al.
demonstrated by the ability of a CdtC-specific monoclonal antibody to neutralise the cytotoxic activity of HdCDT (Cope et al., 1997). Studies on the A. actinomycetemcomitans CDT (AaCDT), however, suggested that AaCdtB alone might be sufficient to cause G2 arrest; in human T cells. Despite this observation the authors suggested that AaCdtC can also play a role in the cell cycle arrest and absence of AaCdtC was not confirmed by Western-blot analysis (Shenker et al., 1999, 2000). This review will focus on the effect of CDTs on mammalian cells. For more details on the CDT nomenclature and the involvement of CDTs in pathogenicity the reader is referred to (Cortes-Bratti et al., 2001b).
CDT activates checkpoint responses One of the earliest effect of CDTs demonstrated on cell lines was the induction of cell cycle arrest in G2 and this was associated with the accumulation of the inactive hyperphosphorylated form of the cyclin-dependent kinase cdc2 (reviewed in (Pickett and Whitehouse, 1999)). When a broader panel of cell lines were tested, it became clear to us that the effects of the HdCDT were cell type dependent. Epithelial cell lines, such as HeLa, HEp-2, and normal keratinocytes were indeed arrested in the G2 phase of the cell cycle. However, primary human fibroblasts from foreskin and embryonic lung were arrested also in G1, while cell lines of B cell origin rapidly underwent apoptosis (Cortes-Bratti et al., 2001a). This observation suggested that HdCDT, and possibly the other CDTs, activates checkpoint responses induced by DNA damage. Indeed the type of cell cycle arrest induced by HdCDT was similar to that induced by ionizing radiation (IR) in HEp-2 cells and in normal fibroblasts. In fibroblasts, which were arrested both in G1 and G2, p53 was stabilised and phosphorylated on serine 15 with the same kinetics in intoxicated as in irradiated cells. This was associated with increased expression of the p53regulated cyclin-dependent kinase inhibitor p21 and with upregulation of p27. In HEp-2 cells, early modification of p53 and p21 levels was not detected. However, in this cell type the HdCDT-induced G2 arrest was associated with an accumulation of the phosphorylated form of the chk2 kinase, known to play a central role in activation of the G2 checkpoint response. Induction of different responses in a cell type-dependent manner suggested to us that HdCDT acted at the level of protein(s) involved in the control of checkpoint responses, which prevent progression through the cell cycle until the DNA damage has been repaired, thus avoiding genetic instability. The phosphatidylinositol (PI)-related kinase Ataxia Telangiectasia Mutated (ATM) and its homologue ATR are key molecules in sensing DNA damage. ATM and/or ATR activate
checkpoint responses at different stages of the cell cycle: the G1/S transition (G1 checkpoint), the S phase progression, and the G2/M boundary (G2/M checkpoint) (reviewed in (Shiloh, 2001)). The early molecular response to HdCDT was ATM dependent, since the induction of cell death upon intoxication was delayed in ATM-deficient B cell lines, compared to ATM wildtype cells, and the delay in cell death was associated with a delayed kinetics of p53 stabilisation (CortesBratti et al., 2001b). These findings limited the target candidates to DNA or molecules, acting upstream ATM and directly involved in activation of checkpoint responses. A schematic representation of the cellular responses activated by HdCDT is depicted in Figure 1.
DNase I-like activity of CdtB and DNA damage Recent data indicate that the human Mre11 complex, constituted by the Mre11, Rad50 and Nbs1 proteins, acts as possible sensor of DNA damage. It is well established that the Mre11 complex associates with damaged DNA and forms discrete nuclear foci after induction of double strand breaks (DSBs) following irradiation (reviewed in (Petrini, 1999)). We observed the same pattern of Mre11 re-localisation upon HdCDT treatment of B cell lines, HeLa cells, and foreskin fibroblasts (Frisan, T. et al., submitted, Figure 2), strongly suggesting that the primary toxin target is DNA. These data complement and expand those demonstrating that CdtB from C. jejuni, E. coli, and H. ducreyi presents structural and functional homology with the mammalian DNase I (Elwell and Dreyfus, 2000; Frisk et al., 2001; Lara-Tejero and Galan, 2000). Crude CdtB preparations from E. coli and H. ducreyi possessed DNase activity as detected by in vitro digestion of a DNA plasmid substrate (Elwell and Dreyfus, 2000; Frisk et al., 2001). Expression of CdtB either by transfection of the C. jejuni cdtB gene or by electroporation of the E. coli purified protein was sufficient to induce a slowly appearing nuclear fragmentation and a marked chromatin disruption (Elwell et al., 2001; Lara-Tejero and Galan, 2000). In both cases the DNase activity, as well as the induction of cell cycle arrest, was abolished by point mutations of conserved residues required for catalysis or for magnesium binding, indicating that the cytotoxic effects are related to the enzymatic activity. Upon microinjection, C. jejuni CdtB (CjCdtB) was able to produce the same effects as the CdtABC toxin, suggesting that CdtB alone, when present intracellularly can induce cell cycle arrest. However, CjCdtB alone added extracellularly did not induce a cell cycle block
Cytolethal distending toxins induce DNA damage
497
Fig. 1. Schematic representation of the mode of CDTs action. CDT-induced DNA damage causes: i) activation of the ATM kinase and downstream checkpoint responses resulting in G2 and G1/G2 arrest, or apoptosis depending on the cell type; ii) re-localization of the Mre11 complex.
in HeLa cells (Lara-Tejero and Galan, 2000). We observed similar results upon microinjection of a Histagged recombinant H. ducreyi CdtB subunit into HeLa cells (Frisan, T. et al., submitted).
Holotoxin composition The HdCDT preparation used to investigate the cellular responses in mammalian cells contained as major components CdtB and CdtC as detected by Westernblot analysis. Traces of a shorter polypeptide (but not the full-length protein) reacting with the anti-CdtA monoclonal antibody could be detected only in overexposed membranes (Frisan, T. et al., submitted). Since the CdtB subunit alone is not sufficient to induce either cell cycle arrest or cell distention when provided extracellularly, we propose that CdtC may be required to promote internalization of the enzymatic component and that CDTs act as an AB toxin. However it is important to stress that all the three genes have to be present in the producing bacterium. The HdCDT holotoxin needs to be internalized by endocytosis via clathrin-coated pits and it requires an intact Golgi apparatus in order to intoxicate HEp-2 cells (Cortes-Bratti et al., 2000). Most probably the
Fig. 2. HdCDT causes re-localization of the Mre11 complex and induction of Mre11 foci. The lymphoblastoid B cell line SN-B1 (Cortes-Bratti et al., 2001a), the epithelial line HeLa and human foreskin fibroblasts were treated with HdCDT (2 mg/ml) for 8 h. Mre11 staining was performed as previously described (Maser et al., 1997).
498
T. Frisan et al.
toxin (or at least the CdtB component) is then transported from the Golgi to the ER, as demonstrated for many bacterial toxins targeting cytosolic molecules (Lord and Roberts, 1998), and therefrom delivered directly or indirectly to the nucleus (Figure 3). The role of CdtA still remains to be elucidated. Frisk et al. (2001) showed that that the isoelectric point (pI) of the CdtC component is 1.5 pH units higher in recombinant strains expressing all three components than in recombinant strains expressing the CdtC protein alone. A similar change of pI occurred after mixing the three individual components in vitro. Therefore, it is possible that CdtA is required to modify and activate CdtC, in analogy with the situation demonstrated for other bacterial toxins (reviewed in (Lally et al., 1999)).
Conclusions The past year has witnessed a significant development in the field of the cytolethal distending toxins. However, several aspects of CDT-mediated intoxication need to be further elucidated, such as the precise role of the CdtA subunit, and a more detailed definition of how CdtC participates in the toxin internalization. It is noteworthy that bacterial toxins have been extremely useful in the study of different aspects of cell biology (Aktories, 2000). Since CDTs are the only bacterial protein toxins described so far to act at the nuclear level (causing DNA damage), they may constitute a useful tool to study protein transport from the Golgi/ ER compartment to the nucleus. Furthermore, the CDT-mediated cell distension and promotion of the
Fig. 3. Proposed model for the HdCDT holotoxin and schematic representation of the HdCDT internalization pathway. Our hypothesis for the production of an active HdCDT holotoxin is the following: CdtC requires activation, possibly via CdtA, to CdtC*. CdtC* and CdtB may constitute an AB toxin, where CdtB is the enzymatic component and CdtC* could be involved in CdtB internalization. HdCDT is internalized via clathrin-dependent endocytosis and requires an intact Golgi apparatus in order to be active on the target cells. CDT translocation to the ER and its direct or indirect transport to the nucleus have still to be demonstrated; therefore, they are indicated as dotted arrows.
Cytolethal distending toxins induce DNA damage
actin cytoskeleton may be used to dissect the still poorly understood cross-talk between molecules that control the cytoskeleton and those that control cell cycle progression/arrest. Bacterial toxins can also be used for therapeutic purposes, e. g. the use of immunotoxins as an alternative or complement in cancer treatment, or the use of toxins acting as vehicles for the delivery of specific proteins into cells (Kreitman, 1999; Sandvig and Van Deurs, 2000). The interference of CDT with the cell cycle makes it a potentially good candidate for an anti-tumor agent, provided that the toxin can be selectively delivered to tumour cells. Acknowledgements. This work was supported by the Swedish Medical Research Council (05969), the Karolinska Institutet and the Stiftelsen Lars Hiertas Minne, Stockholm, Sweden. T. Frisan is supported by the Swedish Society for Medical Research.
References Aktories, K.: Bacterial protein toxins as tools in cell biology and pharmacology. In: Cellular microbiology (P. Cossart, P. Boquet, S. Normak, R. Rappuoli, eds.), pp. 221–237. ASM press (American Society for Microbiology), Washington DC 2000. Cope, L. D., Lumbley, S., Latimer, J. L., Stevens, M. K., Johnson, L. S., Purvén, M., Munson, R. S., Lagergård, T., Radolf, J., Hansen, E. J.: A diffusible cytotoxin of Haemophilus ducreyi. Proc. Natl. Acad. Sci. USA 94, 4056– 4061 (1997). Cortes-Bratti, X., Chaves-Olarte, E., Lagergård, T., Thelestam, M.: Cellular internalization of cytolethal distending toxin from Haemophilus ducreyi. Infect. Immun. 68, 6903–6911 (2000). Cortes-Bratti, X., Karlsson, C., Lagergård, T., Thelestam, M., Frisan, T.: The Haemophilus ducreyi cytolethal distending toxin induces cell cycle arrest and apoptosis via the DNA damage checkpoint pathways. J. Biol. Chem. 276, 5296–5302 (2001a). Cortes-Bratti, X., Frisan, T., Thelestam, M.: The cytolethal distending toxins induce DNA damage and cell cycle arrest. Toxicon 39, 1729–1736 (2001b). Elwell, C., Chao, K., Patel, K., Dreyfus, L.: Escherichia coli CdtB mediates cytolethal distending toxin cell cycle arrest. Infect. Immun. 69, 3418–3422 (2001). Elwell, C. A., Dreyfus, L. A.: DNAase I homologous residues in CdtB are critical for cytolethal distending toxinmediated cell cycle arrest. Mol. Microbiol. 37, 952–963 (2000). Frisk, A., Lebens, M., Johansson, C., Ahmed, H., Svensson, L., Ahlman, K., Lagergård, T.: The role of different protein components from the Haemophilus ducreyi cytolethal distending toxin in generation of cell toxicity. Microb. Pathog. 30, 313–324 (2001). Johnson, W. M., Lior, H.: Response of Chinese hamster ovary cells to a cytolethal distending toxin (CDT) of Escherichia coli and possible misinterpretation as heat-labile (LT) enterotoxin. FEMS Microbiol. Lett. 43, 19–23 (1987).
499
Kreitman, R. J.: Immunotoxins in cancer therapy. Curr. Opin. Immunol. 11, 570–578 (1999). Lally, E. T., Hill, R. B., Kieba, I. R., Korostoff, J.: The interaction between RTX toxins and target cells. Trends Microbiol. 7, 356–361 (1999). Lara-Tejero, M., Galan, J. E.: A bacterial toxin that controls cell cycle progression as a deoxyribonuclease I-like protein. Science 290, 354–357 (2000). Lord, M. J., Roberts, L. M.: Toxin entry: retrograde transport through the secretory pathway. J. Cell Biol. 140, 733–736 (1998). Maser, R. S., Monsen, K. J., Nelms, B. E., Petrini, J. H. J.: hMre11 and hRad50 nuclear foci are induced during normal cellular response to DNA double-strand breaks. Mol. Cell. Biol. 17, 6087–6096 (1997). Mayer, M. P., Bueno, L., Hansen, E. J., Di Rienzo, J. M.: Identification of a cytolethal distending toxin gene locus and features of a virulence-associated region in Actinobacillus actinomycetemcomitans. Infect. Immun. 67, 1227–1237 (1999). Okuda, J., Fukumoto, M., Takeda, Y., Nishibuchi, M.: Examination of diarrheagenicity of cytolethal distending toxin: suckling mouse response to the products of the cdt ABC genes of Shigella dysenteriae. Infect. Immun. 65, 428–433 (1997). Petrini, J. H. J.: DNA repair ‘99. The mammalian Mre11Rad50-Nbs1 protein complex: integration of functions in the cellular DNA-damage response. Am. J. Hum. Genet. 64, 1264–1269 (1999). Pickett, C. L., Pesci, E. C., Cottle, D. L., Russell, G., Erdem, A. N., Zeytin, H.: Prevalence of cytolethal distending toxin production in Campylobacter jejuni and relatedness of Campylobacter sp. cdtB gene. Infect. Immun. 64, 2070– 2078 (1996). Pickett, C. L., Whitehouse, C. A.: The cytolethal distending toxin family. Trends Microbiol. 7, 292–297 (1999). Sandvig, K., Van Deurs, B.: Entry of ricin and Shiga toxin into cells: molecular mechanisms and medical perspectives. EMBO J. 19, 5943–5950 (2000). Shenker, B. J., Hoffmaster, R. H., McKay, T. L., Demuth, D. R.: Expression of the cytolethal distending toxin (Cdt) operon in Actinobacillus actinomycetemcomitans: evidence that CdtB protein is responsible for G2 arrest of the cell cycle in human T cells. J. Immunol. 165, 2612–2618 (2000). Shenker, B. J., McKay, T., Datar, S., Miller, M., Chowhan, R., Demuth, D.: 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 (1999). Shiloh, Y.: ATM and ATR: networking cellular responses to DNA damage. Curr. Opin. Genet. Dev. 11, 71–77 (2001). Sugai, M., Kawamoto, T., Peres, S. Y., Ueno, Y., Komatsuzawa, H., Fujiwara, T., Kurihara, H., Suginaka, H., Oswald, E.: The cell cycle-specific growth-inhibitory factor produced by Actinobacillus actinomycetemcomitans is a cytolethal distending toxin. Infect. Immun. 66, 5008– 5019 (1998). Young, V. B., Knox, K. A., Schauer, D. B.: Cytolethal distending toxin sequence and activity in the enterohepatic pathogen Helicobacter hepaticus. Infect. Immun. 68, 184–191 (2000).