Previous Article | Next Article 
Infection and Immunity, July 2003, p. 4178-4181, Vol. 71, No. 7
0019-9567/03/$08.00+0 DOI: 10.1128/IAI.71.7.4178-4181.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Cytotoxic Necrotizing Factor 1 Enhances Reactive Oxygen Species-Dependent Transcription and Secretion of Proinflammatory Cytokines in Human Uroepithelial Cells
Loredana Falzano,1 Maria Giovanna Quaranta,2 Sara Travaglione,1 Perla Filippini,1 Alessia Fabbri,1 Marina Viora,2 Gianfranco Donelli,1 and Carla Fiorentini1*
Department of Ultrastructures,1
Department of Immunology, Istituto Superiore di Sanità, 00161 Rome, Italy2
Received 16 December 2002/
Returned for modification 27 January 2003/
Accepted 28 March 2003

ABSTRACT
Uropathogenic
Escherichia coli strains frequently produce a
Rho-activating protein toxin named cytotoxic necrotizing factor
type 1 (CNF1). We herein report that CNF1 promotes transcription
and release of tumor necrosis factor alpha, gamma interferon,
interleukin-6 (IL-6), and IL-8 proinflammatory cytokines and
increases the production of reactive oxygen species (ROS) in
uroepithelial T24 cells. The antioxidant
N-acetyl-
L-cysteine
counteracts these phenomena, a fact which suggests a role for
ROS-mediated signaling in CNF1-induced proinflammatory cytokine
production.

TEXT
Uropathogenic
Escherichia coli strains, the most common bacterial
cause of urinary tract infections, produce a number of virulence-associated
factors, including aerobactin, hemolysin, P fimbriae, type 1
pili, and cytotoxic necrotizing factor type 1 (CNF1) (
7,
8,
24,
29). CNF1 is a chromosomally encoded toxin that activates
the small GTP-binding proteins of the Rho family (Rho, Rac,
and Cdc42) by catalyzing their deamidation at a specific glutamine
residue (
19,
28,
35). As very recently reported by Doye and
coworkers (
12), CNF1 principally activates the Rac GTPase, and
this activation is transient because the deamidated (activated)
form of Rac is more susceptible to ubiquitin- and proteasome-mediated
degradation. Degradation of the toxin-activated Rac switches
on the cellular responses to CNF1 (
12), which are largely dependent
on the cell type (
6). Recent in vivo studies reported that CNF1
plays an important role in establishing infection and promoting
inflammation in the mouse bladder (
32) and also contributes
to
E. coli virulence in a model of acute prostatitis (
2,
33).
This finding is in accordance with the generally accepted wisdom
that infection invariably results in inflammation (
31), which
triggers, in turn, a rapid up-regulation of the transcription,
synthesis, and release of proinflammatory cytokines (
25). Although
principally governed by specialized cells like macrophages,
an inflammatory reaction can also be promoted by epithelial
cells via mediators that transmit cellular signals to the immune
system (
22). The type of cytokine involved is rigorously defined
by the strategy employed by the bacterium to interact with the
host cell and by the nature of the two challengers (
3,
23,
36).
To investigate whether CNF1 contributes to the development of
E. coli-driven inflammation, we tested the ability of the toxin
to induce the transcription and secretion of proinflammatory
cytokines in T24 cells, an epithelial cell line derived from
human bladder tissue (American Type Culture Collection, Rockville,
Md.) T24 cells were cultured in McCoy's 5A medium supplemented
with 10% fetal calf serum, nonessential amino acids, 100 µg
of streptomycin per ml, and 100 U of penicillin per ml (all
from GIBCO-BRL, Gaithersburg, Md.). CNF1 was obtained from the
392 ISS strain (kindly provided by V. Falbo, Rome, Italy) and
purified as previously described (
11,
13). Importantly, the
CNF1 preparation was free of lipopolysaccharide (LPS), as determined
by the Limulus assay (i.e., <0.03 U of LPS per ml). For all
experiments, 2
x 10
4 cells/ml were seeded and 10
-10 M CNF1 was
added directly to the culture medium. Supernatant aliquots of
T24 cells exposed to CNF1 for 3, 6, and 18 h were collected,
and cytokines were measured by using commercially available
enzyme-linked immunoabsorbent assay kits (R&D Systems).
The cytokines investigated were those typically secreted by
epithelial cells, and a closer look was taken at the cytokines
stimulated by
E. coli strains (
1,
22). In particular, we analyzed
gamma interferon (IFN-

), a multifunctional cytokine first regarded
as an antiviral factor; tumor necrosis factor alpha (TNF-

),
a major mediator of inflammation; interleukin-6 (IL-6), a multifunctional
cytokine with immunoregulatory and inflammatory effects; and
IL-8, a potent chemoattractant for polymorphonuclear leukocytes.
As shown in Fig. 1a through d, although slight amounts of TNF-
, IFN-
, IL-6, and IL-8 were appreciable even in the supernatant of untreated T24 cells, exposure to CNF1 significantly amplified their secretion. Specifically, the release of IFN-
increased significantly in a time frame of toxin treatment ranging from 3 to 18 h (Fig. 1a), with a rapid decrease thereafter (data not shown). CNF1 also provoked a rapid increase in TNF-
secretion that peaked after 3 h of challenge with the toxin and then returned to the control cell value (Fig. 1b). IL-6 release increased with a constant trend along with the time of exposure to CNF1 (Fig. 1c), whereas the level of IL-8 secretion became significant after 6 h of treatment and reached a nearly fourfold higher value within 18 h (Fig. 1d). Moreover, to further rule out the possibility that the observed cytokine secretion could be due to LPS contamination (4, 34), a sample of CNF1 was heat inactivated by exposure of the protein toxin preparation to 98°C for 10 min; this procedure destroys protein but not LPS activity. No cytokine release was induced by the heat-inactivated CNF1 in T24 cells (data not shown). It is worth noting that the peak of TNF-
secretion (at 3 h) preceded those of IFN-
(at 6 h) and IL-8 (at 18 h), thus confirming the role of TNF-
as a pluripotent activator of inflammation, capable of inducing a proinflammatory cytokine cascade (10). TNF-
, in fact, regulates many aspects of a host cell's defense mechanisms against pathogenic microbes, and its major effect is the activation of gene expression. Therefore, to analyze whether the augmented cytokine release correlated with an increase in mRNA transcription, T24 cells were treated with CNF1 for 3, 6, and 18 h and total cellular RNA was extracted by using the RNeasy kit (Qiagen). Reverse transcription-PCR (RT-PCR) was performed as previously described (30) with the Access RT-PCR system (Promega) according to the manufacturer's instructions. A set of primers for IFN-
, TNF-
, IL-6, and IL-8 was purchased from Clontech, while GAPDH (glyceraldehyde-3-phosphate dehydrogenase) sequences have been previously published (30). The reaction products were electrophoresed through a 1.8% agarose gel (Bio-Rad). Ethidium bromide-stained gels were scanned with a densitometer (Biomed Instruments), and cytokine mRNA levels were normalized to GAPDH mRNA levels (21). IFN-
, TNF-
, IL-6, and IL-8 mRNA transcription was up-regulated within 3 h of CNF1 exposure (Fig. 1e through h). Notably, whereas IFN-
, IL-6, and IL-8 mRNA transcription remained sustained until at least 18 h after CNF1 treatment, the levels of mRNA coding for TNF-
showed complete extinction after 18 h of treatment; these results are consistent with the cytokine secretion data reported above.
Increasing evidence suggests that reactive oxygen species (ROS),
such as superoxide anions and hydrogen peroxide, can act as
signaling intermediates for cytokine induction (
15,
16). In
order to investigate whether CNF1 could promote the production
of cytosolic ROS, T24 cells exposed to the toxin for 30 min
and for 1,2, 3, and 6 h were stained for 15 min at 37°C
with 5 µM 2'-7' dichlorodihydrofluorescein diacetate (DCF-DA;
Sigma, St. Louis, Mo.), a fluorophore that measures intracellular
ROS production (
16). Both control and treated cells were immediately
analyzed on a FACScan flow cytometer (Becton-Dickinson, Mountain
View, Calif.) equipped with a 15-mW, 488-nm-wavelength, air-cooled
argon ion laser. In order to analyze only living cells, 5 µM
propidium iodide was added to all samples a few minutes before
the acquisition of data. As shown in Fig.
2a, CNF1 increased
cytosolic ROS in uroepithelial cells, which occurred in a time
frame ranging from 1 to 3 h. It is worth noting that heat-inactivated
CNF1 failed to augment the cytosolic ROS level in T24 cells
(data not shown). To clarify whether this redox imbalance was
responsible for the CNF1-mediated effects on cytokine production,
additional experiments were carried out by using as an antioxidant
the powerful thiol supplier
N-acetyl-
L-cysteine (NAC; Zambon
Group, Milan, Italy) (
9). NAC (10 mM) was directly dissolved
in the culture medium and added to T24 cells for 2 h before
the cells were exposed to CNF1. At this dose, the scavenger
could counteract the toxin-raised ROS production (Fig.
2a),
as described previously (
17). NAC was also able to diminish
CNF1-induced up-regulation of TNF-

, IL-6, and IL-8 gene expression
(Fig.
2b), whereas IFN-

seemed to be constitutively regulated
by the redox state of the cell. Since Rac is one of the main
promoters of cytosolic ROS production (
5,
15), we can hypothesize
that the observed transcription of the cytokines follows the
trend of Rac and ROS activation, increasing within 3 h of challenge
with CNF1 and decreasing thereafter, when Rac is degraded (
12).
NAC is effective in impairing the CNF1-induced transcription
of TNF-

, IL-6, and IL-8 at 3 h and, for IL-6 and IL-8, also
at 6 h. This result can speculatively be explained by the ability
of TNF-

to activate Rac once it is secreted, with Rac, in turn,
possibly reactivating transcription independently of ROS (
27).
Consequently, NAC failed to diminish TNF-

transcription at 6
h. The secretion of TNF-

conceivably mirrors the trend of transcription,
being appreciable at 6 h but not at 3 h. The decrease induced
by NAC at 6 h is most probably due to the hindrance of transcription
that occurs earlier (at 3 h) rather than to blocked secretion.
These data suggest that the molecular mechanisms utilized by
ROS to transduce their message might involve oxidation of signal
transduction molecules such as protein kinases and transcription
factors. In order to verify whether the onset of intracellular
ROS activity could also influence cytokine secretion, supernatant
aliquots of T24 cells preincubated with NAC and then challenged
with CNF1 for 3 and 6 h were analyzed. As illustrated in Fig.
2c, NAC had little effect on IL-6 production and the drop in
IFN-

secretion was independent of the toxin activity. In contrast,
the antioxidant diminished TNF-

secretion (to 64%) and nearly
abolished that of IL-8 (to 4%) after 6 h of CNF1 exposure. Of
interest, IL-8 is constitutively produced in T24 cells (
26),
but its secretion occurs only after infection or transformation
(
20). From this result, we can hypothesize that the secretion
of IL-8 strictly follows the trend of activation and degradation
of Rac (
12), whose proteosomal degradation switches on the cellular
responses (
12) and, most probably in this case, secretion. Thus,
secretion of IL-8 starts to be evident at 6 h although its transcription
(and most probably its presence inside the cell) is already
detectable after 3 h (Fig.1h).
All in all, these findings point toward the ability of CNF1-activated
T24 cells to increase in a time-dependent manner the transcription
and release of proinflammatory cytokines, such as TNF-

, IFN-

,
IL-6, and IL-8. This process involves cellular redox imbalance,
with the transcription counteracted by the antioxidant NAC.
Interestingly, NAC can also significantly impair the CNF1-induced
release of IL-8, a master of the acute inflammatory response
in the uroepithelium (
32). Whereas high levels of ROS are normally
produced by phagocytic cells as an essential mechanism of defense
against invading microorganisms and are also responsible for
injuries to host tissues, lower amounts are used by other cell
types, such as epithelial cells, for signal transduction (
16).
Very recently, we have discovered that CNF1 may bring about
a Rac-dependent superoxide anion release by epithelial cells,
suggesting a novel role for epithelial cells in sharing with
professional phagocytes the task of eliminating unwanted pathogens
(
14). In the present work, however, we have dealt with cytosolic
ROS that may, at the low amounts detected, act as intracellular
second messengers (
15). Thus, CNF1 can be viewed as a crucial
player in the urovirulence triggered by uropathogenic
Escherichia coli strains, in accordance with in vivo studies reporting the
contribution of the toxin to the inflammation and infection
processes of uroepithelial tissues (
2,
33). The ability of CNF1
to promote phagocyte-like behavior (
13,
18) as well as an inflammatory
reaction in epithelial cells may favor the spreading of bacteria
through the host organism.

ACKNOWLEDGMENTS
We are grateful to W. Malorni for critical reading of the manuscript
and useful suggestions.

FOOTNOTES
* Corresponding author. Mailing address: Department of Ultrastructures, Viale Regina Elena 299, 00161 Rome, Italy. Phone: 39 06 49903006. Fax: 39 06 49387140. E-mail:
carla.fiorentini{at}iss.it.

Editor: F. C. Fang

REFERENCES
1 - Agace, W., S. Hedges, U. Andersson, J. Andersson, M. Ceska, and C. Svanborg. 1993. Selective cytokine production by epithelial cells following exposure to Escherichia coli. Infect. Immun. 61:602-609.[Abstract/Free Full Text]
2 - Andreu, A., A. E. Stapleton, C. Fennell, H. A. Lockman, M. Xercavins, F. Fernandez, and W. E. Stamm. 1997. Urovirulence determinants in Escherichia coli strains causing prostatitis. J. Infect. Dis. 176: 464-469.[Medline]
3 - Ausiello, C. M., and P. Mastrantonio. 2000. Modification of cytokine networks induced in the host by intestinal bacteria producing exotoxins. Microb. Ecol. Health Dis. Suppl. 2:157-164.
4 - Backhed, F., S. Normak, and A. Richter-Dahlfors. 2002. TLR4-dependent lipopolysaccharide signalling in epithelial cells is independent of extracellular protease activity. Cell. Microbiol. 4:297-303.[CrossRef][Medline]
5 - Bokoch, G. M., and B. A. Diebold. 2002. Current molecular models for NADPH oxidase regulation by Rac GTPase. Blood 100:2692-2696.[Abstract/Free Full Text]
6 - Boquet, P., and C. Fiorentini. 2000. The cytotoxic necrotizing factor 1 from Escherichia coli. Handb. Exp. Pharmacol. 145:361-379.
7 - Caprioli, A., V. Falbo, F. M. Ruggeri, L. Baldassarri, R. Bisicchia, G. Ippolito, E. Romoli, and G. Donelli. 1987. Cytotoxin necrotizing factor production by hemolytic strains of Escherichia coli causing extraintestinal infections. J. Clin. Microbiol. 25:146-149.[Abstract/Free Full Text]
8 - Caprioli, A., V. Falbo, F. Minelli, M. Famiglietti, and G. Donelli. 1989. Production of hemolysin and cytotoxic necrotizing factor by uropathogenic Escherichia coli belonging to different serogroups. Microecol. Ther. 18:335-338.
9 - De Flora, S., C. F. Cesarone, R. M. Balansky, A. Albini, F. D'Agostini, C. Bennicelli, M. Bagnasco, A. Camoirano, L. Scatolini, and A. Rovida. 1995. Chemopreventive properties and mechanisms of N-acetylcysteine. The experimental background. J. Cell. Biochem. Suppl. 22:33-41.[Medline]
10 - Dinarello, C. A. 2000. Proinflammatory cytokines. Chest 118:503-508.
11 - Donelli, G., C. Fiorentini, L. Falzano, M. Pouchelet, E. Oswald, and P. Boquet. 1994. Effects induced by cytotoxic necrotizing factor 1 (CNF1) from pathogenic E. coli on cultured epithelial cells. Zentbl. Bakteriol. Suppl. 24: 60-71.
12 - Doye, A., A. Mettouchi, G. Bossis, R. Clement, C. Buisson-Touati, G. Flatau, L. Gagnoux, M. Piechaczyk, P. Boquet, and E. Lemichez. 2002. CNF1 exploits the ubiquitin-proteasome machinery to restrict Rho GTPase activation for bacterial host cell invasion. Cell 111:553-564.[CrossRef][Medline]
13 - Falzano, L., C. Fiorentini, G. Donelli, E. Michel, C. Kocks, P. Cossart, L. Cabanié, E. Oswald, and P. Boquet. 1993. Induction of phagocytic behaviour in human epithelial cells by E. coli cytotoxic necrotizing factor type 1. Mol. Microbiol. 9:1247-1254.[Medline]
14 - Falzano, L., R. Rivabene, M. T. Santini, A. Fabbri, and C. Fiorentini. 2001. An Escherichia coli cytotoxin increases superoxide anion generation via Rac in epithelial cells. Biochem. Biophys. Res. Commun. 283:1026-1030.[CrossRef][Medline]
15 - Finkel, T. 1998. Oxygen radicals and signaling. Curr. Opin. Cell Biol. 10:248-253.[CrossRef][Medline]
16 - Finkel, T. 2001. Reactive oxygen species and signal transduction. IUBMB Life 52:3-6.[Medline]
17 - Fiorentini, C., L. Falzano, R. Rivabene, A. Fabbri, and W. Malorni. 1999. N-acetylcysteine protects epithelial cells against the oxidative imbalance due to Clostridium difficile toxins. FEBS Lett. 453:124-128.[CrossRef][Medline]
18 - Fiorentini, C., L. Falzano, A. Fabbri, A. Stringaro, A. Logozzi, S. Travaglione, S. Contamin, G. Arancia, W. Malorni, and S. Fais. 2001. Activation of Rho GTPases by cytotoxic necrotizing factor 1 induces macropinocytosis and scavenging activity in epithelial cells. Mol. Biol. Cell 12:2061-2073.[Abstract/Free Full Text]
19 - Flatau, G., E. Lemichez, M. Gauthier, P. Chardin, S. Paris, C. Fiorentini, and P. Boquet. 1997. Rho GTPase activation by bacterial toxin-induced glutamine deamidation. Nature 387:729-733.[CrossRef][Medline]
20 - Hang, L., B. Wullt, Z. Shen, D. Karpman, and C. Svanborg. 1998. Cytokine repertoire of epithelial cells lining the urinary tract. J. Urol. 159:2185-2192.[CrossRef][Medline]
21 - Haraguchi, S., R. A. Good, M. James-Yarish, and G. J. Cianciolo. 1995. Differential modulation of Th1- and Th2-related cytokine mRNA expression by a synthetic peptide homologous to a conserved domain within retroviral envelope protein. Proc. Natl. Acad. Sci. USA 92:3611-3636.[Abstract/Free Full Text]
22 - Hedges, S. R., W. W. Agace, and C. Svanborg. 1995. Epithelial cytokine responses and mucosal cytokine networks. Trends Microbiol. 3:266-270.[CrossRef][Medline]
23 - Henderson, B., M. Wilson, and B. Wren. 1997. Are bacterial exotoxins cytokine network regulators? Trends Microbiol. 5:454-458.[CrossRef][Medline]
24 - Johnson, J. R. 1991. Virulence factors in Escherichia coli urinary tract infection. Clin. Microbiol. Rev. 4:80-128.[Abstract/Free Full Text]
25 - Knaus, U. G. 2000. Rho GTPase signaling in inflammation and transformation. Immunol. Res. 3:103-109.
26 - Lang, K., B. Niggemann, K. S. Zanker, and F. Entschladen. 2002. Signal processing in migrating bladder carcinoma cells: role of the autocrine interleukin-8 loop. Int. J. Cancer 99:673-680.[CrossRef][Medline]
27 - Larrick, J. W., and S. C. Wright. 1990. Cytotoxic mechanism of tumor necrosis factor-
. FASEB J. 4:3215-3223.[Abstract]
28 - Lerm, M., J. Selzer, A. Hoffmeyer, U. R. Rapp, K. Aktories, and G. Schmidt. 1999. Deamidation of Cdc42 and Rac by Escherichia coli cytotoxic necrotizing factor 1: activation of c-Jun N-terminal kinase in HeLa cells. Infect. Immun. 67:496-503.[Abstract/Free Full Text]
29 - Martinez, J. J., and S. J. Hultgren. 2002. Requirement of Rho-family GTPases in the invasion of type 1-piliated uropathogenic Escherichia coli. Cell. Microbiol. 4:19-28.[CrossRef][Medline]
30 - Quaranta, M. G., E. Tritarelli, L. Giordani, and M. Viora. 2002. HIV-1 Nef induces dentritic cell differentiation: a possible mechanism of uninfected CD4+ T cell activation. Exp. Cell Res. 275:243-254.[CrossRef][Medline]
31 - Rappuoli, R. 2000. Pushing the limits of cellular microbiology: microarrays to study bacteria-host cell intimate contacts. Proc. Natl. Acad. Sci. USA 97:13467-13469.[Free Full Text]
32 - Rippere-Lampe, K. E., A. D. O'Brien, R. Conran, and H. A. Lockman. 2001. Mutation of the gene encoding cytotoxic necrotizing factor type 1 (cnf1) attenuates the virulence of uropathogenic Escherichia coli. Infect. Immun. 69:3954-3964.[Abstract/Free Full Text]
33 - Rippere-Lampe, K. E., M. Lang, H. Ceri, M. Olson, H. A. Lockman, and A. D. O'Brien. 2001. Cytotoxic necrotizing factor type 1-positive Escherichia coli causes increased inflammation and tissue damage to the prostate in a rat prostatitis model. Infect. Immun. 69:6515-6519.[Abstract/Free Full Text]
34 - Schilling, J. D., M. A. Mulvey, C. D. Vincent, R. G. Lorenz, and S. J. Hultgren. 2001. Bacterial invasion augments epithelial cytokine responses to Escherichia coli through a lipopolysaccharide-dependent mechanism. J. Immunol. 166:1148-1155.[Abstract/Free Full Text]
35 - Schmidt, G., P. Sehr, M. Wilm, J. Selzer, M. Mann, and K. Aktories. 1997. Gln 63 of Rho is deamidated by Escherichia coli cytotoxic necrotizing factor-1. Nature 387:725-729.[CrossRef][Medline]
36 - Wilson, M., R. Seymour, and B. Henderson. 1998. Bacterial perturbation of cytokine networks. Infect. Immun. 66:2401-2409.[Free Full Text]
Infection and Immunity, July 2003, p. 4178-4181, Vol. 71, No. 7
0019-9567/03/$08.00+0 DOI: 10.1128/IAI.71.7.4178-4181.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
This article has been cited by other articles:
-
Miraglia, A. G., Travaglione, S., Meschini, S., Falzano, L., Matarrese, P., Quaranta, M. G., Viora, M., Fiorentini, C., Fabbri, A.
(2007). Cytotoxic Necrotizing Factor 1 Prevents Apoptosis via the Akt/I{kappa}B Kinase Pathway: Role of Nuclear Factor-{kappa}B and Bcl-2. Mol. Biol. Cell
18: 2735-2744
[Abstract]
[Full Text]
-
Falzano, L., Filippini, P., Travaglione, S., Miraglia, A. G., Fabbri, A., Fiorentini, C.
(2006). Escherichia coli Cytotoxic Necrotizing Factor 1 Blocks Cell Cycle G2/M Transition in Uroepithelial Cells. Infect. Immun.
74: 3765-3772
[Abstract]
[Full Text]
-
Brest, P., Turchi, L., Le'Negrate, G., Berto, F., Moreilhon, C., Mari, B., Ponzio, G., Hofman, P.
(2004). Escherichia coli Cytotoxic Necrotizing Factor 1 Inhibits Intestinal Epithelial Wound Healing In Vitro after Mechanical Injury. Infect. Immun.
72: 5733-5740
[Abstract]
[Full Text]