Role of p38alpha/beta MAP Kinase in Cell Susceptibility to Clostridium sordellii Lethal Toxin and Clostridium difficile Toxin B

Lethal Toxin from Clostridium sordellii (TcsL), which is casually involved in the toxic shock syndrome and in gas gangrene, enters its target cells by receptor-mediated endocytosis. Inside the cell, TcsL mono-O-glucosylates and thereby inactivates Rac/Cdc42 and Ras subtype GTPases, resulting in actin reorganization and an activation of p38 MAP kinase. While a role of p38 MAP kinase in TcsL-induced cell death is well established, data on a role of p38 MAP kinase in TcsL-induced actin reorganization are not available. In this study, TcsL-induced Rac/Cdc42 glucosylation and actin reorganization are differentially analyzed in p38alpha−/− MSCV empty vector MEFs and the corresponding cell line with reconstituted p38alpha expression (p38alpha−/− MSCV p38alpha MEFs). Genetic deletion of p38alpha results in reduced susceptibility of cells to TcsL-induced Rac/Cdc42 glucosylation and actin reorganization. Furthermore, SB203580, a pyridinyl imidazole inhibitor of p38alpha/beta MAP kinase, also protects cells from TcsL-induced effects in both p38−/− MSCV empty vector MEFs and in p38alpha−/− MSCV p38alpha MEFs, suggesting that inhibition of p38beta contributes to the protective effect of SB203580. In contrast, the effects of the related C. difficile Toxin B are responsive neither to SB203580 treatment nor to p38alpha deletion. In conclusion, the protective effects of SB203580 and of p38alpha deletion are likely not based on inhibition of the toxins’ glucosyltransferase activity rather than on inhibited endocytic uptake of specifically TcsL into target cells.


Introduction
Toxin-producing strains of C. difficile and C. sordellii cause intestinal infections, including C. difficile-associated diarrhea (CDAD) in humans and horses, and C. sordellii-induced hemorrhagic enteritis and enterotoxemia in cattle, sheep, and other ruminants [1][2][3][4]. The major virulence factors involved in these infections are toxin A (TcdA) and toxin B (TcdB) of C. difficile, and lethal toxin (TcsL) and hemorrhagic toxin (TcsH) from C. sordellii. These single chained toxins exhibit an AB-like toxin structure with the C-terminal delivery domain mediating cell entry of the N-terminal glucosyltransferase domain by receptor-mediated endocytosis [5,6]. The endocytosed glucosyltransferase domain associates with membrane phosphatidylserine facilitating mono-O-glucosylation of small GTPases of the Rho and Ras subfamilies in a monovalent and divalent metal ion-dependent manner [7][8][9]. The acceptor amino acid of toxin-catalyzed mono-O-glucosylation is Thr-35 in Rac1, Cdc42, and (H/K/N)Ras and Thr-37 in Rho(A/B/C). Mono-O-glucosylated Rho/Ras GTPases are incapable of coupling to their regulatory and effector protein and thus are functionally inactive [10][11][12][13]. Treatment of cultured cells with the glucosylating

Prevention of TcsL-Induced Actin Re-Organization upon Inhibition of p38 alpha/beta
TcsL time-dependently induced actin reorganization in p38 alpha -proficient p38 alpha −/− MSCV p38 alpha MEFs ( Figure 1), with TcsL treatment for 4 h being sufficient for almost complete cell rounding ( Figure 1). In contrast, TcsL-induced rounding of p38 alpha -deficient p38 alpha −/− MSCV EV MEFs was clearly delayed, suggesting a reduced susceptibility of p38 alpha −/− MSCV EV MEFs to TcsL. Next, TcsL-induced cell rounding was analyzed in p38 alpha −/− MSCV p38 alpha MEFs treated with SB203580, a pyridinyl imidazole inhibitor of p38 alpha/beta MAP kinase [27]. SB203080 concentration-dependently reduced TcsL-induced cell rounding, with a SB203580 concentration of 10 µM being sufficient for almost complete prevention of TcsL-induced cell rounding (Figure 2A,B). A pronounced protective effect of SB203580 was also observed in a time-dependent experiment ( Figure 2C). SB203580 (10 µM) alone did not change fibroblast morphology ( Figure 2A). Finally, the protective effect of p38 alpha inhibition was analyzed in TcsL concentration-dependent experiments ( Figure 3). Either genetic deletion of p38 alpha or SB203580 treatment delayed TcsL-induced cell rounding. Interestingly, SB203580 treatment of p38 alpha -deficient p38 alpha −/− MSCV EV MEFs further delayed TcsL-induced cell rounding, suggesting a role of p38 beta inhibition in the reduced susceptibility to TcsL. TcdB is highly related to TcsL (identity of 75% at amino acid level), as both TcdB and TcsL enter their target cells by receptor-mediated endocytosis and both cause changed actin dynamics by mono-O-glucosylation of small GTPases. Interestingly, neither genetic deletion ( Figure 4A,B) nor treatment with the p38 alpha/beta inhibitor SB203580 ( Figure 4C,D) changed the kinetics of TcdB-induced actin reorganization. p38 alpha/beta inhibition thus mediates protection of fibroblasts from TcsL-(not TcdB-) induced actin reorganization. MEFs) were treated with TcsL (1 μg/mL) for the indicated times. Cells were then washed, fixed, permeabilized, and stained with rhodaminephalloidin and DAPI. Cell morphology was visualized using fluorescence microscopy (20× amplification). TcsL-induced changes of the morphology were time-dependently quantified in terms of the number of rounded per total cells. Six representative microscopic fields were chosen and 300 cells total were counted for characteristic cell rounding. Values are the mean ± SD from three independent experiments performed in triplicates. p < 0.01 indicates significant differences comparing p38alpha-proficient with p38alpha-deficient cells using Student's t-test. Cells were then washed, fixed, permeabilized, and stained with rhodamine-phalloidin and DAPI. Cell morphology was visualized using fluorescence microscopy (20× amplification). TcsL-induced changes of the morphology were time-dependently quantified in terms of the number of rounded per total cells. Six representative microscopic fields were chosen and 300 cells total were counted for characteristic cell rounding. Values are the mean ± SD from three independent experiments performed in triplicates. p < 0.01 indicates significant differences comparing p38 alpha -proficient with p38 alpha -deficient cells using Student's t-test.
the indicated times. The cellular levels of non-glucosylated Rac/Cdc42, total Rac1, pS144/141-PAK1/2, PAK2, pT222-MAPKAPK2, MAPKAPK2, and beta-actin were analyzed by immunoblotting using the indicated antibodies. Quantifications of immunoblots were performed using Kodak software and relative amounts of non-glucosylated Rac/Cdc42 versus the total levels of Rac1, respectively, are expressed as mean ± SD of three independent experiments. * indicates significant differences, p < 0.05, as analyzed using Student's t-test. The cellular levels of non-glucosylated Rac/Cdc42, total Rac1, pS144/141-PAK1/2, PAK2, and beta-actin were analyzed by immunoblotting using the indicated antibodies. Quantifications of immunoblots were performed using Kodak software and relative amounts of non-glucosylated Rac/Cdc42 versus the total levels of Rac1, respectively, are expressed as mean ± SD of three independent experiments. * indicates significant differences, p < 0.05, as analyzed using Student's t-test. The cellular levels of non-glucosylated Rac/Cdc42, total Rac1, pS144/141-PAK1/2, PAK2, and beta-actin were analyzed by immunoblotting using the indicated antibodies. Quantifications of immunoblots were performed using Kodak software and relative amounts of non-glucosylated Rac/Cdc42 versus the total levels of Rac1, respectively, are expressed as mean ± SD of three independent experiments. * indicates significant differences, p < 0.05, as analyzed using Student's t-test.
TcsL-induced Rac/Cdc42 glucosylation were next analyzed upon pharmacological inhibition of p38 alpha/beta . SB203580 treatment of either p38 alpha −/− MSCV p38 alpha and p38 alpha −/− MSCV empty vector MEFs resulted in an almost complete loss of pT222-MAPKAPK2, confirming effective p38 inhibition (Figures 5 and 6). TcsL-catalyzed Rac/Cdc42 glucosylation and PAK1/2 deactivation were responsive to SB203580 treatment in both p38 alpha −/− MSCV p38 alpha and p38 alpha −/− MSCV empty vector MEFs, both in time-( Figure 5) and concentration-dependent ( Figure 6) experiments. The observation that cell rounding and Rac/Cdc42 glucosylation in p38 alpha −/− MSCV empty vector MEFs are responsive to SB203580 suggests that the protective effects of SB203580 involve inhibition of both p38 alpha and p38 beta . Taken together, p38 alpha/beta inhibition-mediated protection of fibroblasts from TcsL-induced actin reorganization coincides with protection from TcsL-catalyzed Rac/Cdc42 glucosylation. The cellular levels of non-glucosylated Rac/Cdc42, total Rac1, pS144/141-PAK1/2, and beta-actin were analyzed by immunoblotting using the indicated antibodies. Quantifications of immunoblots were performed using Kodak software and relative amounts of non-glucosylated Rac/Cdc42 versus the total levels of Rac1, respectively, are expressed as mean ± SD of three experiments.

SB203580 Preserves TcsL-Induced Loss of Epithelial Barrier Function
C. sordellii-associated disease include necrotic and hemorrhagic enteritis, whereby TcsL has been shown to alter epithelial permeability [33,34]. To check if SB203580-mediated inhibition of TcsL might be useful with regard to disease treatment, TcsL-induced loss of epithelial barrier function was next analyzed in terms of the loss of the transepithelial resistance (TER) of a Madin-Darby canine kidney (MDCK-C7) monolayer [9,35]. TcsL treatment time-dependently decreased the TER of the MDCK-C7 monolayer ( Figure 8A). In the presence of SB203580, TcsL-induced loss of TER was markedly attenuated ( Figure 8A). SB203580 alone did not affect the TER ( Figure 8A). In contrast, TcdB-induced loss of the TER of the MDCK-C7 monolayer was not responsive to SB203580 treatment ( Figure 8B), consistent with above observations showing that TcdB-induced actin reorganization was not responsive to SB203580 treatment (Figure 4). SB203580 treatment thus might be useful in the light of treatment of the TcsL-induced loss of epithelial barrier function. MEFs were treated with TcdB (1 ng/mL) for the indicated times. The cellular levels of non-glucosylated Rac/Cdc42, total Rac1, pS144/141-PAK1/2, and beta-actin were analyzed by immunoblotting using the indicated antibodies. Quantifications of immunoblots were performed using Kodak software and relative amounts of non-glucosylated Rac/Cdc42 versus the total levels of Rac1, respectively, are expressed as mean ± SD of three experiments.

SB203580 Preserves TcsL-Induced Loss of Epithelial Barrier Function
C. sordellii-associated disease include necrotic and hemorrhagic enteritis, whereby TcsL has been shown to alter epithelial permeability [33,34]. To check if SB203580-mediated inhibition of TcsL might be useful with regard to disease treatment, TcsL-induced loss of epithelial barrier function was next analyzed in terms of the loss of the transepithelial resistance (TER) of a Madin-Darby canine kidney (MDCK-C7) monolayer [9,35]. TcsL treatment time-dependently decreased the TER of the MDCK-C7 monolayer ( Figure 8A). In the presence of SB203580, TcsL-induced loss of TER was markedly attenuated ( Figure 8A). SB203580 alone did not affect the TER ( Figure 8A). In contrast, TcdB-induced loss of the TER of the MDCK-C7 monolayer was not responsive to SB203580 treatment ( Figure 8B), consistent with above observations showing that TcdB-induced actin reorganization was not responsive to SB203580 treatment ( Figure 4). SB203580 treatment thus might be useful in the light of treatment of the TcsL-induced loss of epithelial barrier function. The cellular levels of non-glucosylated Rac/Cdc42, total Rac1, pS144/141-PAK1/2, and beta-actin were analyzed by immunoblotting using the indicated antibodies. Quantifications of immunoblots were performed using Kodak software and relative amounts of non-glucosylated Rac/Cdc42 versus the total levels of Rac1, respectively, are expressed as mean ± SD of three experiments.

SB203580 Preserves TcsL-Induced Loss of Epithelial Barrier Function
C. sordellii-associated disease include necrotic and hemorrhagic enteritis, whereby TcsL has been shown to alter epithelial permeability [33,34]. To check if SB203580-mediated inhibition of TcsL might be useful with regard to disease treatment, TcsL-induced loss of epithelial barrier function was next analyzed in terms of the loss of the transepithelial resistance (TER) of a Madin-Darby canine kidney (MDCK-C7) monolayer [9,35]. TcsL treatment time-dependently decreased the TER of the MDCK-C7 monolayer ( Figure 8A). In the presence of SB203580, TcsL-induced loss of TER was markedly attenuated ( Figure 8A). SB203580 alone did not affect the TER ( Figure 8A). In contrast, TcdB-induced loss of the TER of the MDCK-C7 monolayer was not responsive to SB203580 treatment ( Figure 8B), consistent with above observations showing that TcdB-induced actin reorganization was not responsive to SB203580 treatment ( Figure 4). SB203580 treatment thus might be useful in the light of treatment of the TcsL-induced loss of epithelial barrier function.

Discussion
In this study, SB203580, a pyridinyl imidazole inhibitor of p38 alpha/beta MAP kinase, has been presented to efficaciously prevent TcsL-induced loss of epithelial barrier function of MDCK-C7 monolayers and to prevent TcsL-induced cell rounding, Rac/Cdc42 glucosylation, and PAK deactivation in murine fibroblasts. Furthermore, genetic deletion of p38 alpha is also sufficient for preventing TcsL-induced cell rounding, Rac/Cdc42 glucosylation, and PAK deactivation in murine fibroblasts. Interestingly, p38 alpha −/− MSCV empty vector MEFs turn out to be still sensitive to SB203580, suggesting that the protective effects of SB203580 involves inhibition of both p38 alpha and p38 beta . How does p38 inhibition mediate the protective effect against TcsL? TcdB and TcsL both are mono-O-glucosyltransferases that modify small GTPases. Their N-terminal glucosyltransferase domains are structurally and functionally highly related, as they share Rac/Cdc42 as substrate GTPases (with threonine-35 being the acceptor amino acid) and UDP-glucose as a sugar donor [36]. From the observation that TcdB-induced Rac/Cdc42 glucosylation is insensitive to p38 inhibition, it can be concluded that SB203580 does not interfere with the intracellular glucosyltransferase activity of the toxins. This leads to the new hypothesis that p38 alpha/beta inhibition affects endocytic uptake of TcsL into target cells. In fact, members of the p38 MAP kinase family are important regulators of endocytosis, as they control endocytic trafficking via the GDI-Rab5 complex [37]. In particular, p38 MAP kinase regulates stress-induced internalization of the epidermal growth factor receptor (EGFR) and µ opioid receptor endocytosis [38][39][40]. Against this background, inhibition of p38 alpha/beta might prevent cell entry of specifically TcsL (not TcdB) by receptor-mediated endocytosis. This hypothetic model implies that TcsL and TcdB enter their target cells by exploiting distinct cell surface receptors, with the TcsL cell surface receptors being internalized in a p38 alpha/beta -dependent and the TcdB cell surface receptors being internalized in a p38 alpha/beta -independent manner.
In a former study, the responsiveness of TcsL-induced effects such as apoptotic cell death to inhibition by SB203580 has been interpreted in terms of an involvement of p38 alpha/beta in TcsL-induced cell death [24]. The observations of this study have led to the conclusion that the protective effect of SB203580 is based on rather inhibition of TcsL uptake than on a role of p38 in the cytopathic effects of TcsL. Against this background, the responsiveness of TcsL-induced cell death to inhibition by SB203580 must be re-interpreted in terms of reduced TcsL uptake and subsequently reduced GTPase substrate glucosylation as the cause of cell death inhibition. In autophagy research, the pyridinyl imidazole inhibitors SB203580 and SB202190 have been shown to interfere with the autophagic flux independently of p38 MAP kinase [41]. These observations have led to the recommendation that pyridinyl imidazole class inhibitors should not be used as pharmacological tools in the analysis of MAPK11-MAPK14/p38-dependence [42]. The latter recommendation seems to be applicable also in the field of protein toxins.
The protective effect of the pyridinyl imidazole SB203580 is most interesting with regard to the development of non-antibiotic treatment for diseases caused by toxigenic C. sordellii. Further research will address the characterization of those pathways mediating the protective effect against TcsL upon inhibition of p38 alpha/beta . Furthermore, a screening of additional pyridinyl imidazole compounds capable of inhibiting the effects of TcsL is under way in our laboratory.

•
Genetic deletion of p38 alpha or treatment with SB203580 protects MEFs from Rac/Cdc42 glucosylation and actin reorganization induced by TcsL (not by the related TcdB). • Treatment with SB203580 protects epithelial monolayer from loss of epithelial barrier function induced by TcsL (not by TcdB).

•
The protective effects of SB203580 treatment and of p38 alpha deletion are likely based on inhibition of endocytic uptake of TcsL rather than on inhibition of the toxins' glucosyltransferase activity.
Toxins: TcsL was prepared from C. sordellii IP82, which is the same strain as 6018, and TcdB from C. difficile VPI10463. Toxins were produced and purified yielding only one band on SDS-PAGE as previously described [43,44]. In brief, a dialysis bag containing 900 mL of 0.9% NaCl in a total volume of 4 liters of brain heart infusion (Difco, BD Life Sciences, Heidelberg, Germany) was inoculated with 100 mL of an overnight culture of C. sordellii or C. difficile. The culture was grown under microaerophilic conditions at 37 • C for 72 h. Bacteria were removed from the dialysis bag solution by centrifugation. Proteins from the culture supernatant from were precipitated by ammonium sulfate (Merck Millipore, Darmstadt, Germany) at 70% saturation. The precipitated proteins were dissolved in 50 mM Tris-HCl pH 7.5 buffer and extensively dialyzed against 50 mM Tris-HCl pH 7.5 buffer for 24 h. The protein solution was loaded onto an anion exchange column (MonoQ, GE Healthcare Europe, Freiburg, Germany). Either TcsL or TcdB were eluted with 50 mM Tris-HCl, pH 7.5, at 500-600 mM NaCl and were subsequently dialyzed against buffer (50 mM Tris-HCl pH 7.5, 15 mM NaCl). The absence of TcdA (which eluted at 150-200 mM NaCl) in TcdB preparations was checked by immunoblot analysis.

Cell Culture and Preparation of Lysates
p38 −/− MSCV empty vector MEFs and the corresponding p38 −/− MSCV p38 alpha MEFs (kindly provided by Dr. Angel Nebreda, Institute for Research in Biomedicine, Barcelona, Spain) were cultivated in Dulbecco's modified essential medium supplemented with 10% FCS, 100 µg/mL penicillin, 100 U/mL streptomycin and 1 mM sodium pyruvate at 37 • C and 5% CO 2 according to standard protocols [45]. Cells sub-confluently seeded in 3.5-cm dishes were treated with TcsL, TcdB, and SB203580 for different times and concentrations as noted in the figures. Thereby, cells were pretreated with 10 µM SB202580 dissolved in DMSO (final DMSO concentration in the medium 2%) for 20 min and subsequently treated with the toxins or buffer. Upon incubation time, the cells were rinsed with 5 mL of ice-cold phosphate-buffered saline and scraped off in 200 µL of Laemmli lysis buffer per dish. The cells were disrupted mechanically by sonification (five times on ice). The lysate were submitted to immunoblot analysis.

Transepithelial Resistance of Epithelial Monolayers
Madin-Darby canine kidney (MDCK-C7) cells were cultured under standard conditions (37 • C, 5% CO 2 ) as described [35]. Briefly, MDCK-C7 cells were cultured in minimum essential medium (MEM) enriched with Earle's salts, non-essential amino acids, glutamic acid and 10% fetal calf serum (Biochrom, Berlin, Germany) and split twice weekly using standard culture techniques. MDCK-C7 cells were seeded onto 12 well filter transwell inserts (pore size 0.4 µM, BD Life Sciences, Heidelberg, Germany). The transepithelial electrical resistance (TER) was determined by a Voltohmmeter equipped with Endom 24 chamber (EVOM, World Precision Instruments, Berlin, Germany). MDCK-C7 monolayers were cultivated up to an initial resistance of >2 kΩ·cm 2 . The toxins and SB203580 (final DMSO concentration in the medium 2%) were applied on the basolateral site of the monolayer and toxin-induced loss of TER was analyzed in a time-dependent manner.