Contribution of the ADP-ribosylating and receptor-binding properties of cholera-like enterotoxins in modulating cytokine secretion by human intestinal epithelial cells

Marco Soriani1, Lorna Bailey1 and Timothy R. Hirst1

Department of Pathology and Microbiology, University of Bristol, Bristol BS8 1TD, UK1

Author for correspondence: Timothy R. Hirst. Tel: +44 117 928 7538. Fax: +44 117 930 0543. e-mail: t.r.hirst{at}bristol.ac.uk


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
When epithelial cells first encounter cholera toxin (Ctx) produced by Vibrio cholerae they secrete not only chloride ions responsible for causing diarrhoea, but also a number of cytokines that may contribute to the toxin’s potent immunomodulatory properties. Much less is known about the ability of the heat-labile enterotoxin of Escherichia coli (Etx), a close homologue of Ctx, to elicit cytokine secretion by epithelial cells. This study shows that treatment of human intestinal epithelial T84 cells with Etx induces expression of IL-6, IL-10, IL-1R antagonist, as well as IL-1{alpha} and IL-1ß and low levels of IL-8. Such induction was totally dependent on the intrinsic ADP-ribosylating activity of the toxin A-subunit, and could be mimicked by cAMP-elevating agents, such as forskolin and dibutyryl cAMP. By comparison, neither an enzymically inactive mutant of Etx nor EtxB was able to induce cytokine secretion. The behaviour of Ctx and CtxB was very similar to that of Etx and EtxB, respectively. The spectrum of cytokines released by Etx and Ctx indicates that the toxins may create a local microenvironment that strongly biases the immune response towards an anti-inflammatory and a polarized Th2 response.

Keywords: E. coli heat-labile enterotoxin, cholera toxin, cAMP, epithelial cells, cytokine expression

Abbreviations: Etx, E. coli heat-labile enterotoxin; Ctx, cholera toxin; dbcAMP, dibutyryl cAMP; PKA, cAMP-dependent protein kinase


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Cholera toxin (Ctx) and Escherichia coli heat-labile enterotoxin (Etx) are composed of an A-subunit that possesses ADP-ribosyltransferase activity and five B-subunits that bind with high affinity to GM1-ganglioside receptors (Merritt et al., 1994 ; Sixma et al., 1992 ). Whilst these toxins are well recognized as the principal virulence determinants of cholera and traveller’s diarrhoeal disease, respectively (Kaper et al., 1995 ; Nataro & Kaper, 1998 ), there is a growing appreciation that they possess remarkable immunological properties. Notably, Ctx and Etx can trigger exceptionally potent anti-toxin responses when administered at mucosal or systemic sites and can also act as adjuvants augmenting immune responses to admixed antigens (for reviews see Lycke, 1998 ; Rappuoli et al., 1999 ; Simmons et al., 2001 ; Williams et al., 1999 ). More recently, studies have sought to determine the relative contributions of the respective A- and B-subunits in stimulating such effects, and whether or not the two toxins are equally effective. This has led to an increasing consensus that both subunits possess immunomodulatory properties in their own right, and that a combination of both A- and B-subunits engenders the greatest adjuvant effect (Simmons et al., 2001 ). Studies by Rappuoli, Dougan and coworkers have found that holotoxin mutants of Etx and Ctx with ablated or reduced ADP-ribosyltransferase activity exert a greater adjuvant effect than the B-subunits alone (Douce et al., 1995 , 1997 ). Notwithstanding this, it is clear that if the B-subunit of E. coli enterotoxin (EtxB) is administered intranasally it triggers a high-titre immune response to admixed antigens (De Haan et al., 1998 ; Millar et al., 2001 ; Richards et al., 2001 ), and moreover, that this effect is dependent on the receptor-binding property of the molecule (De Haan et al., 1998 ; Guidry et al., 1997 ). Such binding triggers signalling events that lead to up-regulation of activation markers on B-cells, altered CD4+ T-cell differentiation, maturation of dendritic cells, preferential apoptosis of CD8+ T-cells and altered cytokine secretion by macrophages and monocytes (Francis et al., 1992 ; Gagliardi et al., 2000 ; Matousek et al., 1996 ;Nashar et al., 1996a , b , 1997 ; Truitt et al., 1998 ). Although cholera toxin B-subunit (CtxB) shares many of the features of EtxB, including a capacity to modulate immune cells in vitro and to induce a potent anti-B-subunit antibody response, it does not appear to be as good an adjuvant as EtxB (Millar et al., 2001 ; Richards et al., 2001 ). Thus, while most studies to date have taken the view that findings on Ctx would also hold for Etx, it is not evident that this is the case, and at present few studies on the immunomodulatory effects of Etx have been undertaken.

Given that epithelial cells are the initial targets of the toxins, both during infection and following experimental mucosal immunization, it is conceivable that the apparent spectrum of adjuvanticity of the intact holotoxins, attenuated derivatives and B-subunits results from their differential effects on modulating epithelial cell cytokine production. McGee et al. (1993) showed that Ctx was able to strongly promote the production of IL-6 by rat IEC-6 epithelial cells, whilst non-recombinant preparations of CtxB had no effect. Similarly, Bromander et al. (1993) found that Ctx-treatment of IEC-17 cells stimulated both IL-1 and IL-6 secretion. Here we show that recombinant preparations of fully intact and enzymically active Etx, but not an inactive mutant or EtxB, causes elevated cytokine secretion by human intestinal epithelial T84 cells. Such effects were found to dependent on activation of adenylate cyclase and protein kinase A. The nature of the cytokines activated, which included IL-10 and IL-1R{alpha}, is consistent with the toxins exerting an anti-inflammatory effect (Ryan et al., 2000 ), which may contribute to the initial survival and growth of toxinogenic pathogens in vivo (Triadafilopoulos et al., 1989 ). These findings provide a further insight into the enhanced immunomodulatory properties of the intact holotoxin as well as explaining how the toxin may alter the mucosal microenvironment in which immune responses occur.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Materials.
All reagents were purchased from Sigma unless otherwise stated. Recombinant preparations of Etx, Etx(E112D) (an enzymically inactive form of Etx), and Ctx were purified from culture supernatants of Vibrio cholerae O395NT harbouring pAM29, pAM220 and pRC9, respectively, as described previously (Rodighiero et al., 1999 ). Recombinant EtxB and CtxB were purified as described by Richards et al. (2001) .

T84 cell cultures.
The human epithelial cell line T84 was obtained from the European Centre for Applied Microbiology and Research (ECACC, Salisbury, UK) and routinely cultured in Dulbecco’s modified minimal essential medium (D-MEM) supplemented with Hams F10, 10% (v/v) fetal bovine calf serum, glutamine, penicillin and streptomycin. Cells were passaged at 5–6 day intervals and were used between passages 3 and 12. T84 cell monolayers were grown by plating (5x105 cells per well) in 12-well tissue culture plates and cultured for 5–6 days until they were fully confluent and had the morphological appearance of polarized cells. Experiments were also performed using T84 cells grown for 3 weeks on 6-well polycarbonate plates (Costar) and with addition of fresh medium every 3 days. When cultures were treated with various agents the following concentrations were used unless otherwise stated: Etx 1 nM; Etx(E112D) 1 nM; EtxB 100 nM; V. cholerae lipopolysaccharide (LPS) 10 µg ml-1; forskolin 10 µM; dibutyryl cAMP (dbcAMP) 100 µM; H89 10 µM.

Immunofluorescence and confocal microscopy.
Immunofluorescence microscopy was performed on confluent adherent T84 cells grown for 5–6 days on collagen-coated glass coverslips as described by Casciola-Rosen et al. (1996) . In order to detect early apoptotic events in Etx-treated cells, we used Annexin-V staining of phosphatidylserine exposed on the plasma membrane. Control or Etx-treated cells were washed three times with ice-cold PBS and subsequently Annexin V-FITC was added for 20 min at 4 °C in the dark. Cover slips were then washed by dipping in PBS prior to fixation with 4% (w/v) paraformaldehyde and permeabilization with acetone. Cover slips were mounted on glass slides with Vectashield containing 4',6-diamidino-2-phenylindole dihydrochloride (DAPI) (Vector Laboratories) before viewing on a Leica TCS-SP2 confocal Ar plus UV laser scanning microscope attached to a Leica DM IRBE inverted phase-contrast/epifluorescence microscope (Leica Microsystems). An oil-immersion objective lens 63x, NA 1·32, was used, and imaging parameters were selected to optimize resolution.

cAMP detection assay.
To measure intracellular cAMP in T84 cells the Biotrak cAMP scintillation proximity assay (SPA) kit from Amersham was used. Confluent T84 cells cultured in a 96-well plate were treated with cAMP-inducing agents for 1 h. Cells were then lysed and the level of cAMP determined. Briefly, the assay is based on the competition between intracellular unlabelled and a fixed quantity of 125I-labelled cAMP for a limited number of binding sites on cAMP-specific antibody. The antibody-bound cAMP is then reacted with the SPA reagent, which contains anti-rabbit secondary antibody bound to fluoromicrospheres. 125I-labelled cAMP that is bound to the primary rabbit antibody is immobilized on the fluoromicrospheres and the radiations emitted measured by a ß-scintillation counter.

Analysis of secreted cytokines.
Cytokine levels in T84 supernatants were determined by cytokine-specific ELISA. All cytokine standards, capture antibodies and detection antibodies were obtained from Pharmingen. Maxisorp immunoplates were coated overnight with monoclonal anti-human IL-2, IL-4, IL-6, IL-8, IL-10 and TGF-ß1. After blocking, samples and serial twofold dilutions of standards were added to duplicate wells and incubated overnight at 4 °C. The plates were then washed and incubated with the respective biotinylated antibody for 1 h at room temperature. After incubation, extravidin peroxidase conjugate was added and the reaction developed with tetramethylbenzidine and hydrogen peroxide.

Analysis of cytokine gene expression by multi-probe RNase protection assay.
RNase protection assays were performed according to the RiboQuant manual for human cytokines (hCK-2 template array, Pharmingen). Briefly, total RNA from T84 cells was prepared by the LS TRI-reagent protocol and quantified both spectrometrically and electrophoretically (TAE 1%, w/v, agarose gel). The Pharmingen multi-probe set contained a series of cDNA templates for IL-12, IL-1{alpha}, IL-1ß, IL-1R{alpha}, IL-6, IFN-{gamma} and the housekeeping genes L32 and GAPDH to allow normalization of the samples. The multi-probe template set was transcribed as antisense RNA probes by T7 RNA polymerase in the presence of [{alpha}-32P]UTP (ICN). Labelled probes were hybridized overnight with 4–5 µg total RNA, after which free probe and other single-stranded RNA species were digested with RNase A and T1. The remaining ‘RNase-protected’ probes were purified and resolved by urea denaturing PAGE. Mouse control RNA and yeast tRNA were also run. The expressed genes were detected by autoradiography and the level of gene expression determined using the Scion Image program (Scion Corporation). To accurately establish the identity of each protected fragment, the migration of each fragment was plotted against a standard curve of the migration distance for undigested probe versus nucleotide length on a logarithmic scale.

Statistical analysis.
Differences in cytokine production between treated and untreated T84 cells were analysed for statistical significance using unpaired Student’s t-test.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Prior to investigating the effect of cholera-like enterotoxins on the modulation of cytokine production, we assessed whether our recombinant preparations of Etx or EtxB exerted any morphological or cytotoxic effects on T84 cells. When T84 cells were treated with 1 nM Etx or 10 µM forskolin for 6 h, phase-contrast microscopy revealed that the cells had somewhat enlarged nuclei and that there was a reduction in contrast at the cellular junctions (data not shown). Given the recent findings that the B-subunits of cholera-like enterotoxins trigger apoptosis of certain lymphocyte subsets (Nashar et al., 1996b ; Truitt et al., 1998 ), it was decided to determine whether the morphological changes observed in T84 cells were also accompanied by induction of apoptosis. An early apoptotic event is the exposure of phosphatidylserine on the plasma membrane, which can be detected by FITC-Annexin-V. Treatment of T84 cells with 1 nM Etx/Ctx or 100 nM EtxB/CtxB for 6 h and 24 h failed to cause any detectable phosphatidylserine appearance on the cell surface or nuclear condensation (data not shown). Incubation of T84 cells with hydrogen peroxide clearly revealed Annexin V binding to phosphatidylserine (Asada et al., 2001 ). We therefore conclude that the morphological changes triggered by treatment of T84 cells with Etx are not associated with induction of apoptosis.

Etx induces cytokine gene expression in T84 cells
It has been previously reported that Ctx triggers cytokine activation in epithelial cells (Bromander et al., 1993 ; McGee et al., 1993 ), while no information is available on Etx-mediated effects. We assessed the multiple expression of cytokine mRNA transcripts in Etx-treated T84 cells using a highly sensitive and specific human cytokine multi-probe RNase protection assay system, which allows the simultaneous detection of expression of IL-12, IL-1{alpha}, IL-1ß, IL-1R{alpha}, IL-6 and IFN-{gamma}. Fig. 1(a) shows that 1 nM Etx was able to induce the early expression of IL-1{alpha}, IL-1ß, IL-6 and IL-1R{alpha}. The data from three independent experiments were normalized against the mRNA levels of the housekeeping gene GAPDH and expressed as fold induction with respect to PBS control, and are shown in Fig. 1(c–j). We noted while the basal level of IL-1{alpha} and ß cytokine expression in the absence of treatment (PBS control) appeared to vary, the fold increase in expression of these cytokines following addition of Etx and other modulators of cAMP levels was remarkably consistent. The expression of IL-6, IL-1{alpha}, IL-1ß genes, which peaked at 6 h post-treatment (see Fig. 1c, g and i, respectively), decreased at 24 h (see Fig. 1d, h and j). By contrast, mRNA levels for IL-1R{alpha} remained elevated at 24 h post-Etx treatment (see Fig. 1e and f). Protected probes for IL-12 and IFN-{gamma} were barely detectable in samples derived from either PBS- or Etx-treated T84 cells, while their transcripts were present in a control human RNA sample.




View larger version (105K):
[in this window]
[in a new window]
 
Fig. 1. Etx activates IL-1{alpha}, IL-1ß, IL-1R{alpha} and IL-6 gene expression in T84 cells by a cAMP-dependent mechanism. (a) Fully confluent T84 cells were incubated for 6 and 24 h in the presence of PBS, 10 µg LPS ml-1, 1 nM Etx, 1 nM Etx(E112D) or 100 nM EtxB, or pre-incubated for 1 h with 10 µM H89 and then treated with 1 nM Etx (H89). Total RNA was extracted and hybridized with a labelled human cytokine multi-probe template set as described in Methods. Unprotected probes (P), mouse control RNA (C) and yeast tRNA (Y) were also run simultaneously with the samples. Identical results were obtained on three independent occasions. (b) As in (a) except for treatments with 10 µM forskolin (Fors) and 100 µM dbcAMP. (c–j) Quantification of cytokine gene expression in T84 cells treated as in (a): IL-6 (c and d), IL-1R{alpha} (e and f), IL-1ß (g and h) and IL-1{alpha} (i and j). Quantification was obtained by normalizing cytokine band intensities in comparison with the GAPDH housekeeping gene. Mean values±SD of three independent experiments are shown.

 
Etx-mediated cytokine expression in T84 cells is dependent on an enzymically active A-subunit
In order to assess if the ADP-ribosylating activity of Etx was essential in triggering activation of cytokine expression, identical experiments to those above were carried out on T84 cells treated with Etx(E112D) (an enzymically inactive mutant of Etx), or a 100-fold molar excess of recombinant B subunits (EtxB). Neither Etx(E112D) nor EtxB caused elevation in mRNA levels of any of the cytokines tested (Fig. 1a). Addition of 10 µM forskolin (an adenylate cyclase activator) or 100 µM dbcAMP led to similar cytokine patterns to those observed following Etx treatment (Fig. 1b). As shown in Table 1, T84 cells were actively stimulated by Etx and forskolin to produce cAMP, while Etx(E112D) or EtxB failed to elicit a significant cAMP increase. These findings suggest that Etx-mediated ADP-ribosylation of adenylate cyclase and the resultant increase in cAMP levels are responsible for triggering altered cytokine expression. To further corroborate this, H89, a specific inhibitor of the cAMP-dependent protein kinase (PKA), was added 60 min prior to addition of Etx. This led to a complete loss of Etx-mediated elevation in cytokine expression (see Fig. 1a). As a control, 10 µg V. cholerae LPS ml-1 was added to T84 cells. This failed to increase mRNA levels of any of the cytokines tested above basal levels (Fig. 1a). We therefore conclude that the ability of Etx to increase cellular cAMP levels leads to transcriptional activation of IL-1{alpha}, IL-1ß, IL-1R{alpha} and IL-6. Addition of 1 nM Ctx gave identical results (data not shown).


View this table:
[in this window]
[in a new window]
 
Table 1. Modulation of cAMP levels in T84 cells

 
Effect of cholera-like enterotoxins on cytokine secretion
The multiple activation of cytokine gene expression observed after Etx treatment led us to further investigate whether such effects were associated with an increase in cytokine secretion by T84 epithelial cells. For comparison, experiments were performed with both Ctx and Etx. We tested for the presence of TGF-ß1, IL-2, IL-4, IL-6, IL-8 and IL-10 in the supernatant of Etx- and Ctx-treated T84 cells by ELISA. It was found that neither 1 nM Etx, 1 nM Ctx, 100 nM EtxB nor 100 nM CtxB triggered increased secretion of IL-2, IL-4 or TGF-ß1 throughout the 24 h treatment period (data not shown). By contrast, treatment with 1 nM Etx or 1 nM Ctx led to an increase in IL-6 levels, which peaked between 16 h and 24 h post-treatment (Fig. 2a, b) and decreased after 48 h incubation (data not shown). No effect was observed using an identical concentration of Etx(E112D) (Fig. 2b) or 100 nM EtxB/CtxB (Fig. 2a, b). Addition of the cAMP-elevating agents forskolin or dbcAMP led to an increase in IL-6 secretion to levels comparable to those observed using Etx or Ctx (Fig. 2c). Inhibition of PKA activity by H89 completely blocked the capacity of Etx to trigger IL-6 production, while LPS treatment had no effect on IL-6 levels (Fig. 2c).



View larger version (13K):
[in this window]
[in a new window]
 
Fig. 2. Ctx and Etx induce IL-6 secretion by cultured T84 cells. (a) Cells were treated for 24 h with 1 nM Etx, 1 nM Ctx and 100 nM of their respective recombinant B-subunits. Medium supernatant was then collected and tested for IL-6 levels by ELISA as described in Methods. (b) Time course for IL-6 production by Etx in T84 cells. Cells were cultured as in (a) and treated with Etx, Etx(E112D) or EtxB for 6, 16 and 24 h. After each time-point an aliquot of medium was flash-frozen using liquid nitrogen and stored at -80 °C. ELISA was then performed as in (a). (c) Confluent T84 cells were treated for 24 h with 1 nM Etx, 10 µM forskolin, 100 µM dbcAMP or 10 µg LPS ml-1, or pre-incubated for 1 h with 10 µM H89 prior to Etx treatment (H89). IL-6 levels were assayed in the respective supernatants as in (a). The results are mean values±SD of three independent experiments. *P<0·05 vs PBS control.

 
cAMP-elevating agents had previously been shown to activate IL-10 expression in macrophages (Procopio et al., 1999 ) and T-cells (Benbernou et al., 1997 ). When the levels of IL-10 were tested in the supernatant of T84 cells after 1 nM Etx or 1 nM Ctx treatment, we found a late up-regulation of IL-10 production peaking at 24 h (Fig. 3a, b) and decreasing at 48 h (data not shown). As with IL-6 secretion, neither EtxB and CtxB (Fig. 3a), Etx(E112D) (Fig. 2b) nor LPS (Fig. 3c) triggered elevated IL-10 production. Moreover, in agreement with the concept that Etx-mediated up-regulation in IL-10 secretion depends on cAMP elevation, it was found that both forskolin and dbcAMP also increased IL-10 secretion by T84 cells (Fig. 3c), but that a 60 min pre-incubation with H89 blocked the ability of Etx to elevate IL-10 levels. We therefore conclude that the ADP-ribosylating activity of both Etx and Ctx is responsible for activating production of IL-6 and IL-10 in epithelial cells.



View larger version (15K):
[in this window]
[in a new window]
 
Fig. 3. Ctx and Etx induce IL-10 secretion by T84 epithelial cells. Cells were cultured and treated as in Fig. 2. Supernatants, stored at -80 °C after being flash-frozen in liquid nitrogen, were tested for IL-10 levels by ELISA. The results are mean values±SD of three independent experiments. *P<0·05 vs PBS control.

 
Effect of cholera-like toxins on IL-8 production by T84 cells
In addition to an elevation in IL-6 and IL-10 levels, it was found that 1 nM Etx or 100 µM dbcAMP (Fig. 4) or 1 nM Ctx (data not shown) slightly increased IL-8 levels (approx. 1·5-fold) in the supernatant of T84 cells after 24 h treatment. By contrast, EtxB, CtxB or Etx(E112D) failed to induce significant increase of IL-8 secretion (Fig. 4). Whilst these findings are consistent with cAMP triggering activation of IL-8 production in epithelial cells, we noted that addition of LPS alone also modulated IL-8 expression (Fig. 4).



View larger version (12K):
[in this window]
[in a new window]
 
Fig. 4. Etx, dbcAMP and LPS slightly up-regulate IL-8 levels in the supernatant of T84 cells. Fully confluent cells were treated for 24 h with 1 nM Etx, 1 nM Etx(E112D), 100 nM EtxB, 100 µM dbcAMP or 10 µg LPS ml-1. Medium was then collected and tested for IL-8 levels by ELISA as described in Methods. The results are mean values±SD of three independent experiments. *P<0·05 vs PBS control.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Enzymically active cholera-like holotoxins and their B-subunits are able to trigger cAMP-mediated and GM1-receptor-mediated signalling events, respectively (Peterson & Whipp, 1995 ; Soriani et al., 2001 ). However, the relative contributions of these events to creating a local microenvironment leading to immunoresponsiveness, and indeed, which cell types are involved, have remained unclear. Here we provide unequivocal evidence that the holotoxins of both Etx and Ctx can trigger cAMP-dependent up-regulation of a number of cytokines by epithelial cells, and we found no obvious difference in behaviour between the two toxins. During the course of infection released toxins will first encounter epithelial cells and be taken up by a process of endocytosis, which results in delivery of the toxic A-subunits into the cytosolic compartment and to subsequent activation of adenylate cyclase. The B-subunits, by contrast, do not enter the cytosol, and a proportion appear to be transcytosed and released from the basolateral membrane (Lencer et al., 1995 ). Thus, it is conceivable that the principal activities of the A- and B-subunits are segregated and function on different cell types, with the A-subunits primarily mediating their effect on epithelial cells while transcytosed B-subunits interact with immune cells in the lamina propria and Peyer’s patches. Consequently the A-subunit could be responsible for triggering induction of anti-inflammatory cytokines or factors that counter the activity of inflammatory cytokines in epithelial cells, while the B-subunits may exert a similar anti-inflammatory effect on monocytes, macrophages and dendritic cells (Williams et al., 1999 ). The sum of all of these effects is that the toxins may create a local microenvironment that strongly biases the immune response towards an anti-inflammatory and a polarized Th2 response.

Ctx is known to elicit cytokine secretion by epithelial cells as well as by bone-marrow-derived macrophages (Bromander et al., 1991 , 1993 ; Cong et al., 2001 ; Elson & Cong, 1995 ; Klimpel et al., 1995 ; McGee et al., 1993 ) and to inhibit the secretion of pro-inflammatory cytokines such as IL-12, TNF-{alpha}, RANTES and IL-1 by LPS-stimulated human dendritic cells (Braun et al., 1999 ; Gagliardi et al., 2000 ). Because of the similarity in structure and function of Ctx and Etx, it has been generally accepted that results obtained from studies on Ctx will also hold for Etx. Nonetheless, very few studies have been undertaken on Etx, and recent observations have highlighted some striking differences in the immunological properties of their respective B-subunit moieties (Millar & Hirst, 2001 ). Thus, while CtxB appears to be a poor adjuvant, EtxB can trigger immune responses to co-administered antigens that approach levels seen using the full holotoxins of Ctx or Etx as adjuvants. In order to clarify the mechanism of action of such toxins, we investigated the effects of cholera-like enterotoxins on cytokine expression by epithelial cells and whether such modulation depended on (a) the identity of the toxins, (b) their ability to bind to GM1 receptors or (c) their intrinsic ADP-ribosylating activity. Interestingly, Ctx and Etx behaved identically with respect to their effects on cytokine expression by T84 epithelial cells. Moreover, these effects depended on the elevation of intracellular cAMP levels, since the non-toxic B-subunits alone or a mutant form of Etx, lacking enzymic activity, failed to increase any of the cytokines activated by Ctx/Etx or cAMP-elevating agents, such as forskolin and dbcAMP (Fig. 1). By using the highly sensitive RNase protection assay, we were able to screen for transcriptional activation of a series of pro- and anti-inflammatory cytokines. Ctx and Etx, as well as forskolin and dbcAMP, triggered a large increase in IL-6 and IL-1R antagonist mRNA levels, while the expression of Th1-type cytokines like IL-12 was unaffected. The early activation of IL-1{alpha} and IL-1ß observed is in agreement with previous findings showing that Ctx enhances IL-1 production by macrophages (Bromander et al., 1991 ) and epithelial cells (Bromander et al., 1993 ). This potentiates antigen presentation by macrophages and modulates T-cell responses. The elevation in cAMP levels by cholera-like enterotoxins will lead to activation of cAMP-dependent protein kinase (PKA) and the translocation of PKA catalytic subunits to the nucleus (Zidek, 1999 ). The subsequent phosphorylation of cAMP-responsive element binding proteins would be expected to activate transcription of specific genes carrying cAMP-regulatory elements in their promoter region (Seternes et al., 1999 ), such as IL-6 (Krueger et al., 1991 ) and IL-1ß genes (Chandra et al., 1995 ). Our finding that H89, an inhibitor of PKA, abrogates the induction of IL-1{alpha}, IL-1ß, IL-6 and IL-1R{alpha} further corroborates the hypothesis of a direct effect of cAMP on Ctx/Etx-induced cytokine expression.

IL-1R{alpha} is an important anti-inflammatory molecule that acts by blocking IL-1-induced inflammatory reactions (Dinarello, 1998 ). Our finding that the expression of IL-1R antagonist is up-regulated in T84 cells by both Ctx and Etx provides evidence that these toxins might create an anti-inflammatory environment in the intestine by blocking IL-1 binding to its own receptor. Recently, Feng et al. (2000) reported that Ctx and 8-bromo-cAMP up-regulated expression of IL-1R{alpha}, as well as IL-10, in LPS-stimulated macrophages.

IL-6 and IL-10 are believed to play an important role in modulating mucosal immune responses (Klimpel et al., 1995 ). Both Etx and Ctx up-regulated IL-6 levels in the supernatant of T84 human epithelial cells, while H89 completely blocked such effects (Fig. 2). By contrast, CtxB, EtxB and the non-toxic mutant Etx(E112D) failed to induce the production of IL-6 in T84 cells. This is in agreement with previous findings indicating that Ctx triggers IL-6 production in rat epithelial cells by a mechanism involving cAMP (McGee et al., 1993 ). While IL-6 is produced in response to certain inflammatory stimuli, it has been shown that this cytokine can exert anti-inflammatory effects in several systems by inhibiting Th1 responses (Barton, 1997 ). In addition, IL-6 is known to act as a growth factor for activated T-cells, boosting the magnitude of the Th2 response, and promoting immunoglobulin production by B-cells, especially secretory IgA. The high-level production of IL-6 upon treatment of epithelial cells with Etx may contribute to both the polarization of the Th response and the apparent adjuvanticity of the toxin. IL-10 plays a pivotal role in determining the immunosuppressive environment resulting from the action of cholera-like toxins (Wiedermann et al., 1999 ). This concept is strengthened by the fact that putative cAMP-responsive elements were localized within the promoter/enhancer region of the IL-10 gene (Platzer et al., 1999 ). IL-10 levels were up-regulated by a cAMP-dependent pathway in the supernatants of Ctx/Etx-treated T84 cells. Whilst IL-10 secretion was clearly dependent on the ADP-ribosylating activity of Ctx/Etx, its delayed onset in production suggests that indirect autocrine mechanisms may be involved. In the intestinal mucosa IL-10 is released primarily by the Th2 subset of CD4+ T helper cells in response to common environmental antigens (Panja et al., 1995 ). However, the fact that IL-10 is also produced by isolated epithelial cells suggests that it may play a role in the regulation of the intestinal cell function. Indeed, it has been shown that IL-10 enhances sodium and chloride absorption in rat small intestine, limits forskolin-induced chloride secretion and prevents disruption of monolayer barrier integrity in T84 epithelial cells (Madsen et al., 1997 ).

In addition, Etx, as well as dbcAMP, slightly increased IL-8 levels and failed to induce expression of IL-12 in T84 cells. IL-8 is a potent neutrophil, T-cell and basophil chemoattractant involved in acute inflammatory responses (Harada et al., 1994 ) and it has been postulated that during intestinal invasion by luminal bacteria IL-8 can function as a signal for mucosal barrier penetration (Eckmann et al., 1993 ). However, it has been shown that Ctx inhibits in vitro IL-8-induced lymphocyte migration (Bacon & Camp, 1990 ) and that in a rabbit intestine model Ctx did not cause neutrophil infiltration in the lamina propria or necrosis of enterocytes (Triadafilopoulos et al., 1989 ). Thus, while cAMP elevation by Etx or dbcAMP results in some IL-8 secretion by epithelial cells, the toxins appear to be able to couteract this, possibly by direct effects on leukocytes. Such findings are consistent with the clinical picture of disease in which diarrhoea ensues without any evident inflammatory response.

Taken together, these findings suggest that Etx and Ctx elevate expression of anti-inflammatory molecules such as IL-10 and IL-1R{alpha}, and are unable to trigger the potent secretion IL-8 and IL-12. This is consistent with the view that activation of cAMP-mediated pathways in epithelial cells promotes an anti-inflammatory response, as well as a cytokine microenvironment that may favour induction of a polarized Th2 immune response.

The observation that neither an enzymically inactive mutant of Etx nor EtxB was able to induce cytokine secretion may in part explain why these molecules are not such potent mucosal adjuvants as their respective holotoxins. Moreover, the differential adjuvanticity being reported for EtxB, compared to CtxB, does not appear to be due to a differential induction of cytokine expression by epithelial cells, since both B-subunits behaved identically.


   ACKNOWLEDGEMENTS
 
We thank Dr M. Kenny and T. Jones for technical help, Dr M. Jepson for advice and assistance in confocal microscopy and Dr L. de Haan for critically reading the manuscript. This work was supported in part by grants from the Medical Research Council of Great Britain and Oratol Ltd.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Asada, S., Fukuda, K., Nishisaka, F., Matsukawa, M. & Hamanisi, C. (2001). Hydrogen peroxide induces apoptosis of chondrocytes; involvement of calcium ion and extracellular signal-regulated protein kinase. Inflamm Res 50, 19-23.[Medline]

Bacon, K. B. & Camp, R. D. (1990). Interleukin (IL)-8-induced in vitro human lymphocyte migration is inhibited by cholera and pertussis toxins and inhibitors of protein kinase C. Biochem Biophys Res Commun 169, 1099-1104.[Medline]

Barton, B. E. (1997). IL-6: insights into novel biological activities. Clin Immunol Immunopathol 85, 16-20.[Medline]

Benbernou, N., Esnault, S., Shin, H. C., Fekkar, H. & Guenounou, M. (1997). Differential regulation of IFN-gamma, IL-10 and inducible nitric oxide synthase in human T cells by cyclic AMP-dependent signal transduction pathway. Immunology 91, 361-368.[Medline]

Braun, M. C., He, J., Wu, C. Y. & Kelsall, B. L. (1999). Cholera toxin suppresses interleukin (IL)-12 production and IL-12 receptor beta1 and beta2 chain expression. J Exp Med 189, 541-552.[Abstract/Free Full Text]

Bromander, A., Holmgren, J. & Lycke, N. (1991). Cholera toxin stimulates IL-1 production and enhances antigen presentation by macrophages in vitro. J Immunol 146, 2908-2914.[Abstract/Free Full Text]

Bromander, A. K., Kjerrulf, M., Holmgren, J. & Lycke, N. (1993). Cholera toxin enhances alloantigen presentation by cultured intestinal epithelial cells. Scand J Immunol 37, 452-458.[Medline]

Casciola-Rosen, L., Rosen, A., Petri, M. & Schlissel, M. (1996). Surface blebs on apoptotic cells are sites of enhanced procoagulant activity: implications for coagulation events and antigenic spread in systemic lupus erythematosus. Proc Natl Acad Sci USA 93, 1624-1629.[Abstract/Free Full Text]

Chandra, G., Cogswell, J. P., Miller, L. R., Godlevski, M. M., Stinnett, S. W., Noel, S. L., Kadwell, S. H., Kost, T. A. & Gray, J. G. (1995). Cyclic AMP signaling pathways are important in IL-1 beta transcriptional regulation. J Immunol 155, 4535-4543.[Abstract]

Cong, Y., Oliver, A. O. & Elson, C. O. (2001). Effects of cholera toxin on macrophage production of co-stimulatory cytokines. Eur J Immunol 31, 64-71.[Medline]

De Haan, L., Verweij, W. R., Agsteribbe, E. & Wilschut, J. (1998). The role of ADP-ribosylation and GM1-binding activity in the mucosal immunogenicity and adjuvanticity of the Escherichia coli heat-labile enterotoxin and Vibrio cholerae cholera toxin. Immunol Cell Biol 76, 270-279.[Medline]

Dinarello, C. A. (1998). Interleukin-1, interleukin-1 receptors and interleukin-1 receptor antagonist. Int Rev Immunol 16, 457-499.[Medline]

Douce, G., Turcotte, C., Cropley, I., Roberts, M., Pizza, M., Domenighini, M., Rappuoli, R. & Dougan, G. (1995). Mutants of Escherichia coli heat-labile toxin lacking ADP-ribosyltransferase activity act as nontoxic, mucosal adjuvants. Proc Natl Acad Sci USA 92, 1644-1648.[Abstract]

Douce, G., Fontana, M., Pizza, M., Rappuoli, R. & Dougan, G. (1997). Intranasal immunogenicity and adjuvanticity of site-directed mutant derivatives of cholera toxin. Infect Immun 65, 2821-2828.[Abstract]

Eckmann, L., Jung, H. C., Schurer-Maly, C., Panja, A., Morzycka-Wroblewska, E. & Kagnoff, M. F. (1993). Differential cytokine expression by human intestinal epithelial cell lines: regulated expression of interleukin 8. Gastroenterology 105, 1689-1697.[Medline]

Elson, C. O. & Cong, Y. (1995). The effect of cholera toxin on intestinal cells and their cytokines In Cytokines, Cholera, and the Gut , pp. 83-87. Edited by G. T. Keusch & M. Kawakami. Amsterdam:IOS Press.

Feng, W., Wang, Y., Zhang, J., Wang, X., Li, C. & Chang, Z. (2000). Effects of CTx and 8-bromo-cAMP on LPS-induced gene expression of cytokines in murine peritoneal macrophages. Biochem Biophys Res Commun 269, 570-573.[Medline]

Francis, M. L., Ryan, J., Jobling, M. G., Holmes, R. K., Moss, J. & Mond, J. J. (1992). Cyclic AMP-independent effects of cholera toxin on B cell activation. II. Binding of GM1 induces B cell activation. J Immunol 148, 1999-2005.[Abstract/Free Full Text]

Gagliardi, M. C., Sallusto, F., Marinaro, M., Langenkamp, A., Lanzavecchia, A. & De Magistris, M. T. (2000). Cholera toxin induces maturation of human dendritic cells and licences them for Th2 priming. Eur J Immunol 30, 2394-2403.[Medline]

Guidry, J. J., Cardenas, L., Cheng, E. & Clements, J. D. (1997). Role of receptor binding in toxicity, immunogenicity, and adjuvanticity of Escherichia coli heat-labile enterotoxin. Infect Immun 65, 4943-4950.[Abstract]

Harada, A., Sekido, N., Akahoshi, T., Wada, T., Mukaida, N. & Matsushima, K. (1994). Essential involvement of interleukin-8 (IL-8) in acute inflammation. J Leukoc Biol 56, 559-564.[Abstract]

Kaper, J. B., Morris, J. G.Jr & Levine, M. M. (1995). Cholera. Clin Microbiol Rev 8, 48-86.[Abstract]

Klimpel, G. R., Asuncion, M., Haithcoat, J. & Niesel, D. W. (1995). Cholera toxin and Salmonella typhimurium induce different cytokine profiles in the gastrointestinal tract. Infect Immun 63, 1134-1137.[Abstract]

Krueger, J., Ray, A., Tamm, I. & Sehgal, P. B. (1991). Expression and function of interleukin-6 in epithelial cells. J Cell Biochem 45, 327-334.[Medline]

Lencer, W. I., Moe, S., Rufo, P. A. & Madara, J. L. (1995). Transcytosis of cholera toxin subunits across model human intestinal epithelia. Proc Natl Acad Sci USA 92, 10094-10098.[Abstract]

Lycke, N. (1998). The mechanism of cholera toxin adjuvanticity. Res Immunol 148, 504-520.

Madsen, K. L., Lewis, S. A., Tavernini, M. M., Hibbard, J. & Fedorak, R. N. (1997). Interleukin 10 prevents cytokine-induced disruption of T84 monolayer barrier integrity and limits chloride secretion. Gastroenterology 113, 151-159.[Medline]

Matousek, M. P., Nedrud, J. G. & Harding, C. V. (1996). Distinct effects of recombinant cholera toxin B subunit and holotoxin on different stages of class II MHC antigen processing and presentation by macrophages. J Immunol 156, 4137-4145.[Abstract]

McGee, D. W., Elson, C. O. & McGhee, J. R. (1993). Enhancing effect of cholera toxin on interleukin-6 secretion by IEC-6 intestinal epithelial cells: mode of action and augmenting effect of inflammatory cytokines. Infect Immun 61, 4637-4644.[Abstract]

Merritt, E. A., Sarfaty, S., VandenAkker, F., L’Hoir, C., Martial, J. A. & Hol, W. G. J. (1994). Crystal structure of cholera toxin B-pentamer bound to receptor G(M1) pentasaccharide. Protein Sci 3, 166-175.[Abstract/Free Full Text]

Millar, D. G. & Hirst, T. R. (2001). Cholera toxin and Escherichia coli enterotoxin B-subunits inhibit macrophage-mediated antigen processing and presentation: evidence for antigen persistence in non-acidic recycling endosomal compartments. Cell Microbiol 3, 311-329.[Medline]

Millar, D. G., Hirst, T. R. & Snider, D. P. (2001). Escherichia coli heat-labile enterotoxin B subunit is a more potent mucosal adjuvant than its closely related homologue, the B subunit of cholera toxin. Infect Immun 69, 3476-3482.[Abstract/Free Full Text]

Nashar, T. O., Webb, H. M., Eaglestone, S., Williams, N. A. & Hirst, T. R. (1996a). Potent immunogenicity of the B subunits of Escherichia coli heat-labile enterotoxin: receptor binding is essential and induces differential modulation of lymphocyte subsets. Proc Natl Acad Sci USA 93, 226-230.[Abstract/Free Full Text]

Nashar, T. O., Williams, N. A. & Hirst, T. R. (1996b). Cross-linking of cell surface ganglioside GM1 induces the selective apoptosis of mature CD8+ T lymphocytes. Int Immunol 8, 731-736.[Abstract]

Nashar, T. O., Hirst, T. R. & Williams, N. A. (1997). Modulation of B-cell activation by the B subunit of Escherichia coli enterotoxin: receptor interaction up-regulates MHC class II, B7, CD40, CD25 and ICAM-1. Immunology 91, 572-578.[Medline]

Nataro, J. P. & Kaper, J. B. (1998). Diarrheagenic Escherichia coli. Clin Microbiol Rev 11, 142-201.[Abstract/Free Full Text]

Panja, A., Siden, E. & Mayer, L. (1995). Synthesis and regulation of accessory/proinflammatory cytokines by intestinal epithelial cells. Clin Exp Immunol 100, 298-305.[Medline]

Peterson, J. W. & Whipp, S. C. (1995). Comparison of the mechanisms of action of cholera toxin and the heat-stable enterotoxins of Escherichia coli. Infect Immun 63, 1452-1461.[Abstract]

Platzer, C., Fritsch, E., Elsner, T., Lehmann, M. H., Volk, H. D. & Prosch, S. (1999). Cyclic adenosine monophosphate-responsive elements are involved in the transcriptional activation of the human IL-10 gene in monocytic cells. Eur J Immunol 29, 3098-3104.[Medline]

Procopio, D. O., Teixeira, M. M., Camargo, M. M., Travassos, L. R., Ferguson, M. A., Almeida, I. C. & Gazzinelli, R. T. (1999). Differential inhibitory mechanism of cyclic AMP on TNF-alpha and IL-12 synthesis by macrophages exposed to microbial stimuli. Br J Pharmacol 127, 1195-1205.[Abstract/Free Full Text]

Rappuoli, R., Pizza, M., Douce, G. & Dougan, G. (1999). Structure and mucosal adjuvanticity of cholera and Escherichia coli heat-labile enterotoxins. Immunol Today 20, 493-500.[Medline]

Richards, C. M., Aman, A. T., Hirst, T. R., Hill, T. J. & Williams, N. A. (2001). Protective mucosal immunity to ocular herpes simplex virus type 1 infection in mice by using Escherichia coli heat-labile enterotoxin B subunit as an adjuvant. J Virol 75, 1664-1671.[Abstract/Free Full Text]

Rodighiero, C., Aman, A. T., Kenny, M. J., Moss, J., Lencer, W. I. & Hirst, T. R. (1999). Structural basis for the differential toxicity of cholera toxin and Escherichia coli heat-labile enterotoxin. Construction of hybrid toxins identifies the A2-domain as the determinant of differential toxicity. J Biol Chem 274, 3962-3969.[Abstract/Free Full Text]

Ryan, E. J., McNeela, E., Pizza, M., Rappuoli, R., O’Neill, L. & Mills, K. H. (2000). Modulation of innate and acquired immune responses by Escherichia coli heat-labile toxin: distinct pro- and anti-inflammatory effects of the nontoxic AB complex and the enzyme activity. J Immunol 165, 5750-5759.[Abstract/Free Full Text]

Seternes, O. M., Sorensen, R., Johansen, B. & Moens, U. (1999). Activation of protein kinase A by dibutyryl cAMP treatment of NIH 3T3 cells inhibits proliferation but fails to induce Ser-133 phosphorylation and transcriptional activation of CREB. Cell Signal 11, 211-219.[Medline]

Simmons, C. P., Ghaem-Magami, M., Petrovska, L., Lopes, L., Chain, B. M., Williams, N. A. & Dougan, G. (2001). Immunomodulation using bacterial enterotoxins. Scand J Immunol 53, 218-226.[Medline]

Sixma, T. K., Terwisscha van Scheltinga, A. C., Kalk, K. H., Zhou, K., Wartna, E. S. & Hol, W. G. J. (1992). X-ray studies reveal lanthanide binding sites at the A/B5 interface of E. coli heat-labile enterotoxin. FEBS Lett 297, 179-182.[Medline]

Soriani, M., Williams, N. A. & Hirst, T. R. (2001). E. coli enterotoxin B subunit triggers apoptosis of CD8+ T-cells by activating the transcription factor c-myc. Infect Immun 69, 4923-4930.[Abstract/Free Full Text]

Triadafilopoulos, G., Pothoulakis, C., Weiss, R., Giampaolo, C. & Lamont, J. T. (1989). Comparative study of Clostridium difficile toxin A and cholera toxin in rabbit ileum. Gastroenterology 97, 1186-1192.[Medline]

Truitt, R. L., Hanke, C., Radke, J., Mueller, R. & Barbieri, J. T. (1998). Glycosphingolipids as novel targets for T-cell suppression by the B subunit of recombinant heat-labile enterotoxin. Infect Immun 66, 1299-1308.[Abstract/Free Full Text]

Wiedermann, U., Jahn-Schmid, B., Lindblad, M., Rask, C., Holmgren, J., Kraft, D. & Ebner, C. (1999). Suppressive versus stimulatory effects of allergen/cholera toxoid (CTB) conjugates depending on the nature of the allergen in a murine model of type I allergy. Int Immunol 11, 1131-1138.[Abstract/Free Full Text]

Williams, N. A., Hirst, T. R. & Nashar, T. O. (1999). Immune modulation by the cholera-like enterotoxins: from adjuvant to therapeutic. Immunol Today 20, 95-101.[Medline]

Zidek, Z. (1999). Adenosine-cyclic AMP pathways and cytokine expression. Eur Cytokine Netw 10, 319-328.[Medline]

Received 14 September 2001; revised 26 October 2001; accepted 7 November 2001.