Distinct biological activities of C3 and ADP-ribosyltransferase-deficient C3-E174Q Astrid Rohrbeck, Tanja Kolbe, Sandra Hagemann, Harald Genth and Ingo Just Institute of Toxicology, Medizinische Hochschule Hannover, Hanover, Germany
Keywords ADP-ribosyltransferase; apoptosis; C3 exoenzyme; proliferation; RhoA Correspondence A. Rohrbeck, Department of Toxicology, Medizinische Hochschule Hannover, Carl-Neuberg-Straße 1, D-30625 Hanover, Germany Fax: +49 511 5322879 Tel: +49 511 5322807 E-mail:
[email protected] (Received 25 October 2011, revised 15 May 2012, accepted 16 May 2012) doi:10.1111/j.1742-4658.2012.08645.x
Low-molecular-weight GTP-binding proteins of the Rho family control the organization of the actin cytoskeleton in eukaryotic cells. Dramatic reorganization of the actin cytoskeleton is caused by the C3 exoenzyme derived from Clostridium botulinum (C3), based on ADP-ribosylation of RhoA ⁄ B ⁄ C. In addition, wild-type as well as ADP-ribosyltransferase-deficient C3-E174Q induce axonal outgrowth of primary murine hippocampal neurons and prevent growth cone collapse, indicating a non-enzymatic mode of action. In this study, we compared the effects of C3-E174Q and wild-type C3 in the murine hippocampal cell line HT22. Treatment of HT22 cells with C3 resulted in Rho ADP-ribosylation and cell rounding. The ADP-ribosyltransferase-deficient mutant C3-E174Q did not induce either Rho ADP-ribosylation or morphological changes. C3 as well as C3-E174Q treatment resulted in growth arrest, reduced expression of cyclin D levels, and increased expression of RhoB, a negative regulator of cell-cycle progression. Serum starvation induced apoptosis in HT22 cells, as determined on the basis of increased expression of caspase-9 and Bax. C3 but not C3-E174Q protected serum-starved HT22 cells from apoptosis. This is the first study separating ADP-ribosyltransferase-dependent from ADP-ribosyltransferase-independent effects of C3. While morphological changes and anti-apoptotic activity strictly depend on ADP-ribosyltransferase activity, the anti-proliferative effects are independent of ADP-ribosyltransferase activity. Structured digital abstract l Rhotekin physically interacts with RHOA by pull down (View interaction)
Introduction Low-molecular-weight GTP-binding proteins of the Rho family control the organization of the actin cytoskeleton in eukaryotic cells and are involved in the regulation of gene expression, cell proliferation, apoptosis and axonal growth [1,2]. The Rho family members RhoA, B and C exhibit significant amino acid sequence identity (approximately 85%) and are thought to interact with the same effectors. RhoB appears to possess unique functions compared to RhoA and RhoC. RhoB is exclusively localized to membranes, both plasma
membrane and endosomes, even when inactive [3]. RhoA is cytosolic within the GDI complex, and only moves to the plasma membrane upon activation and dissociation of RhoGDI. Interestingly, RhoGDI-1 does not bind to RhoB, consistent with the finding that RhoB is always found in association with membranes [4]. RhoB has a specialized function in intracellular trafficking of cytokine receptors such as epidermal growth factor receptor [5]. Additionally, RhoB protein is relatively short-lived in cells, and its expression can
Abbreviations ART, ADP-ribosyltransferase; C3, C3 exoenzyme derived from Clostridium botulinum; C3-E174Q, ADP-ribosyltransferase-deficient C3.
FEBS Journal 279 (2012) 2657–2671 ª 2012 The Authors Journal compilation ª 2012 FEBS
2657
Effect of C3 exoenzyme on hippocampal cells
A. Rohrbeck et al.
be elevated by a number of stimuli, including epidermal growth factor, transforming growth factor b and genotoxic stress [6,7]. Moreover, silencing of RhoA induced strong up-regulation of both total and active RhoB protein levels [8]. RhoA, RhoB and RhoC are substrates of bacterial C3-like exoenzymes, which inactivate RhoA, B and C by ADP-ribosylation at the amino acid asparagine at position 41 [9]. In cultured cells, Rho ADP-ribosylation causes dramatic changes in the actin cytoskeleton of most cell types, resulting in rounding of cultured fibroblasts or epithelial cells [10,11]. However, C3 from Clostridium botulinum (C3bot) possesses an additional axonotrophic activity independently of its inherent ADP-ribosyltransferase (ART) activity, as C3 induces axonal and dendritic growth, and, in addition, branching and synapse formation in cultured primary murine hippocampal neurons [12]. This axonotrophic activity is specific for C3bot, as C3 isoforms from other microbes do not exhibit axon and dendrite growthpromoting activity in hippocampal neurons [13]. Moreover, the axonotrophic activity of the ART-deficient mutant C3bot-E174Q is as effective as the enzymeactive C3bot. It is known that C3 stimulates growth cone formation in neuronal cells N1E-115 by inhibition of growth cone collapse through RhoA inactivation [14]. On the other hand, C3 induces axonal outgrowth from DRG (dorsal root ganglion) neurons, but formation of lamellipodia and filopodia in the growth cones of DRG is not detectable under these conditions [15]. However, in addition to axon growth-promoting activity, C3 causes inhibition of cell growth in cultured rat pheochromocytoma PC-12 cells and formation of neurites [16]. Growing evidence suggests that C3 influences cell proliferation by inactivation of RhoA. RhoA is known to be instrumental in the kinetics of cyclin D1 expression [17]. Moreover, others have observed inhibition of cyclin D1 expression upon C3 treatment [18]. RhoA also suppresses p21 levels in
multiple normal and transformed cell lines [19]. Degradation of G1 cyclin-dependent kinase inhibitor p27(KIP1) is inhibited by C3, further supporting the important role of RhoA in cell proliferation [20]. Interestingly, RhoB negatively regulates cell proliferation [21]. A negative role of RhoB in growth control would contrast with the positive effects of RhoA and RhoC [22]. Additionally, C3 also appears to play a role in survival signalling. It was shown that C3-catalyzed inactivation of RhoA induces apoptosis in haematopoietic cells [23] and causes adhesion-independent apoptosis [24]. Recently, it has been reported that C3 results in activation of caspase-3 and thus apoptosis of cardiomyocytes [25]. By contrast, inactivation of RhoA by C3 protects neurons from cell death [26]. In another study, it was observed that, after contusion injury, cellpermeable C3 decreased neuronal and glial apoptosis, as detected by TUNEL [27]. In this study, the biological effects of C3 and C3E174Q were analyzed in the mouse hippocampal HT22 cells, a cell line that exhibits sensitivity to C3. C3 and C3-E174Q both induced inhibition of cell proliferation and cyclin D1 down-regulation as well as RhoB expression. These effects thus appear to be independent of the ART activity of C3. In contrast, actin re-organization and anti-apoptotic activity were specifically observed in C3-treated cells, thus appeared to depend on the ART activity. This report strongly indicates that C3 possesses ART-independent activities.
Results ART dependency of the morphological changes induced by C3 HT22 cells exhibited spindle-shaped morphology (Fig. 1A). Upon C3 treatment for 72 h, a population of HT22 cells became round and exhibited pronounced
Fig. 1. Morphology of HT22 cells after treatment with C3. HT22 cells were treated for 72 h with 500 nM of C3 or C3-E174Q. Untreated cells served as a control. Phase-contrast microscopic images are shown. Scale bar = 50 lm.
2658
FEBS Journal 279 (2012) 2657–2671 ª 2012 The Authors Journal compilation ª 2012 FEBS
A. Rohrbeck et al.
retraction fibers (Fig. 1). In contrast, treatment with C3-E174Q did not cause any morphological changes (Fig. 1A), suggesting that the morphological changes are dependent on the ART activity of C3. C3bot ADP-ribosylated RhoA from HT22 cells in a timeand concentration-dependent manner, as evidenced by the apparent shift of ADP-ribosylated RhoA to higher molecular mass in SDS ⁄ PAGE (Fig. 2A) and sequential [32P]ADP-ribosylation of non-ADP-ribosylated Rho (Fig. 2B). In contrast, C3-E174Q did not ADP-ribosylate cellular RhoA (Fig. 2A–C). Furthermore, C3E174Q did not possess any ART activity in a cell-free system (Fig. S1). RhoA ADP-ribosylation by C3
Effect of C3 exoenzyme on hippocampal cells
resulted in a reduced level of RhoA-GTP, but treatment with C3-E174Q did not affect the level of RhoAGTP (Fig. 3). These observations were consistent with previous reports showing that C3-E174Q lacked ART activity [13] and that ADP-ribosylation resulted in RhoA inactivation [28,29]. Binding measurements for purified RhoA using microscale thermophoresis Next, we investigated binding of fluorescence-labelled RhoA to C3 exoenzyme using microscale thermophoresis. Binding of C3 to RhoA was readily observed as
Fig. 2. Sensitivity of HT22 cells towards C3 exoenzyme. (A) Exposure of HT22 cells to increasing concentrations of C3 causes a molecular mass shift of RhoA in SDS ⁄ PAGE and a decrease in RhoA level in a concentration- and time-dependent manner. Cells treated with the C3-E174Q did not show an RhoA shift. (B) HT22 cells were exposed to increasing concentrations of C3 for 8 or 24 h at 37 C. Cells were lysed and subjected to sequential [32P]ADP-ribosylation. Phosphorimages from representative experiments are shown. (C) Densitometric measurements for a representative experiment.
FEBS Journal 279 (2012) 2657–2671 ª 2012 The Authors Journal compilation ª 2012 FEBS
2659
Effect of C3 exoenzyme on hippocampal cells
A. Rohrbeck et al.
Fig. 3. GTP-RhoA levels were determined in lysates of HT22 cells using pull-down assay. HT22 cells were treated with either C3 or C3-E174Q for various times. (A) Cell lysates were subjected to Rho binding by pull-down assay followed by western blot analysis using antibodies against RhoA or b-actin. Western blot analyses from representative experiments are presented (n = 3). GTPyS = positive control. (B,C) Cellular levels of GTP-RhoA proteins at 24 h (B) and 72 h (C) were quantified using KODAK 1D image analysis software. All signal intensities were normalized to the intensity of the corresponding b-actin signal.
a change in the thermophoretic property of the fluorescently labelled RhoA upon complex formation. The C3–RhoA complex shows a stronger increase of normalized fluorescence than the unbound protein. The signal shown in Fig. 4A is a binding curve, and starts at an Fnorm of approximately 830 units. When the concentration of C3 is increased, the microscale thermophoresis signal increases to approximately 920 units. Thus, a significant amount of the RhoA is in complex with C3. From the data, we inferred a dissociation coefficient of 32.18 lM (Fig. 4A). We next assessed the interaction of RhoA with C3-E174Q. Figure 4B shows the resulting sigmoidal binding curve for the C3-E174Q–RhoA interaction, with significant differences between low and high C3-E174Q concentrations and a calculated Kd of 7.23 lM. These findings indicate a four- to fivefold higher affinity of RhoA for C3-E174Q compared to C3. Distinct kinetics of RhoB expression induced by C3 and C3-E174Q RhoA ADP-ribosylation by C3 has been reported to result in RhoB expression [30]. Consistently, RhoA ADP-ribosylation in C3-treated HT22 cells was 2660
accompanied by pronounced RhoB expression (Fig. 5A). Interestingly, RhoB expression was also observed in HT22 cells upon prolonged treatment with C3-E174Q for 2–6 days (Fig. 5B). Under these conditions, neither RhoA nor RhoB was ADP-ribosylated, as evidenced by the lack of shift of RhoA and RhoB to apparent higher molecular masses in SDS ⁄ PAGE (Fig. 5B). However, RhoB expression in C3-E174Qtreated cells correlated with a decreased cellular level of RhoA. C3-E174Q thus appeared to induce RhoB expression independently of RhoA ADP-ribosylation. Reduced cell proliferation upon treatment with both C3 and C3-E174Q RhoB has been implicated in the regulation of cell proliferation, as ectopic expression of constitutively active RhoB inhibits proliferation [31]. Next, the effects of C3 and C3-E174Q on HT22 proliferation was analysed over a 7-day period (Fig. 6). Untreated cells showed a reduced growth rate, probably due to density inhibition. The cells reach confluence and form a monolayer at day 5 (Fig. 6). Upon treatment with either C3 or C3-E174Q, exponential growth was observed up to 2 days, and then proliferation ceased.
FEBS Journal 279 (2012) 2657–2671 ª 2012 The Authors Journal compilation ª 2012 FEBS
A. Rohrbeck et al.
Effect of C3 exoenzyme on hippocampal cells
Fig. 4. Microscale thermophoresis binding analysis. The concentration of the fluorescently labelled RhoA is kept constant and the C3 exoenzyme is titrated. (A) Binding curve for labelled RhoA–C3. (B) Binding curve for labelled RhoA–C3-E174Q. The binding curve is sigmoidal and reaches a plateau. Binding curves from representative experiments are shown (n = 2).
In C3-treated cells, the number of viable cell decreased by approximately 30%, probably due to C3-mediated cytotoxicity. The number of C3-E174Q-treated cells remained almost constant from 3 to 7 days, suggesting that C3-E174Q inhibited cell proliferation (Fig. 6). To determine whether the anti-proliferative activity of C3 and C3-E174Q was due to blocked G1–S transition, cellular levels of cyclin D1 mRNA and cyclin D1 protein were analysed by quantitative RT-PCR and western blot analysis, respectively. C3 and C3-E174Q comparably down-regulated both cyclin D1 mRNA (Fig. 7A) and cyclin D1 protein (Fig. 7B) levels in time-series experiments. As shown in Fig. 6B,C, untreated cells show a time-dependent decrease in cyclin D1, probably due to contact inhibition [32]. These observations strongly suggest that C3-induced inhibition of cell proliferation is independent of ART activity. ART dependency of the anti-apoptotic activity of C3 Depending on the cell line and the experimental conditions, C3 has been reported to exhibit either pro-apoptotic [33] or anti-apoptotic activity [34]. To determine whether C3 promoted or inhibited apoptosis of HT22
cells, starvation-induced apoptosis of HT22 cells was analysed in the presence and absence of C3. C3 treatment for 48 h resulted in a significant reduction of bax mRNA and caspase-3 mRNA (Fig. 8A). To elucidate the role of additional apoptotic factors, expression of caspase-8 and caspase-9 mRNA was assessed. Both caspases were strongly down-regulated by C3 (Fig. 8A). However, C3-E174Q did not influence expression of the genes studied (data not shown). As p53 is an important player in various signalling pathways including apoptosis, the protein level of p53 was examined. Wild-type C3 induced a decrease in p53, as shown by western blot analysis. In contrast, C3-E174Q did not alter p53 protein level (Fig. 8B). Furthermore, starvation-induced apoptosis of HT22 cells resulted in expression of Bax (Fig. 8B, lanes 1 and 2). To test for changes in the expression of Bax, HT22 cells were cultured for 24 h in serum-free medium to induced apoptosis, and then incubated with C3 or C3-E174Q in serum-free medium for further 24 or 48 h (Fig. 8B, lanes 4–7). We observed a reduced protein level of Bax 48 h after apoptosis induction if C3 was added after 24 h starvation-induced apoptosis (Fig. 8, Lane 4). C3-E174Q did not reduced Bax 48 h after apoptosis induction in comparison with 72 h
FEBS Journal 279 (2012) 2657–2671 ª 2012 The Authors Journal compilation ª 2012 FEBS
2661
Effect of C3 exoenzyme on hippocampal cells
A. Rohrbeck et al.
Fig. 5. RhoB up-regulation. (A) HT22 cells were exposed to 500 nM C3-E174Q for up to 6 days or 500 nM C3 for 24 h at 37 C. Cells were lysed and submitted to western blot analysis for RhoA, RhoB and b-actin (n = 3). (B) HT22 cells were exposed to 500 nM C3 or C3-E174Q for up to 6 days. Cellular levels of RhoB proteins were determined by western blot analysis and the signal intensity was measured densiometrically (n = 3). The signal intensity was normalized to the intensity of the corresponding b-actin signal. The differences in results were not statistically significant.
serum-deprived control cells (Fig. 8, lanes 6 and 2). Similar results were found for the pro-forms of caspase-3 and caspase-9 (Fig. 9). Next, we determined the enzymatic activity of caspase-3 using a caspase-3 ⁄ 7 ApoONE assay. Serum
starvation for 48 h induced was a threefold increase in caspase-3 ⁄ 7 activity compared with control cells (Fig. 10A). This finding strongly supports previous reports showing that serum starvation results in induction of cell death [35,36]. Consistent with the results of western blot analysis (Fig. 9), caspase-3 activity was reduced in C3- but not in C3-E174Q-treated cells at 48 h after apoptosis induction. To further provide evidence for the anti-apoptotic activity of C3, staurosporin-induced apoptosis of HT22 cells was analysed for its responsiveness to C3. Staurosporin treatment strongly activated caspases in HT22 cells. C3 (but not C3-E174Q) reduced staurosporin-induced caspase activation (Fig. 10B). Finally, staurosporine-induced loss of membrane integrity was analysed using acridine orange ⁄ ethidium bromide staining. Staurosporin-treated HT22 cells lost their spindle-like morphology, shrank, detached from the matrix and were stained orange, indicating loss of membrane integrity. Loss of membrane integrity was decreased in C3-treated cells but not in C3-E174Q-treated cells, further indicating that the anti-apoptotic activity of C3 depended on the ART activity.
Discussion Most cultured cell lines are almost insensitive to C3, unless C3 has been fused to ectopic cell entry domains. A previous study showing that C3 and C3-E174Q induced biological effects in primary hippocampal neurons [13] indicated that neuron-derived cells exhibit remarkable sensitivity to C3bot independently of fused cell entry domains. In this study, SV40-transformed hippocampal neuronal HT22 cells were used as a cell culture model to differentially study the biological effects of C3 and C3-E174Q [37].
Fig. 6. Anti-proliferative effects of C3 exoenzyme. Cell numbers were counted in a trypan blue exclusion assay using a Neubauer counting chamber. Viable cells are not detected by trypan blue but exclude the dye. Incubation of HT22 cells with 500 nM of wild-type C3 for 7 days caused a significant inhibition in cell number 72 h after C3 application compared to untreated cells. Results were statistically significant relative to the control at 72 h (with the exception of C3-E174Q at 5 days) (P £ 0.05). Experiments for each sample were performed in triplicate, and three independent experiments were performed (n = 9 for each value).
2662
FEBS Journal 279 (2012) 2657–2671 ª 2012 The Authors Journal compilation ª 2012 FEBS
A. Rohrbeck et al.
Effect of C3 exoenzyme on hippocampal cells
Fig. 7. Effect of C3 on cell-cycle regulation of HT22 cells. (A) Quantitative RT-PCR of C3bot-treated cells. Summary of differential expression of the cyclin D1 gene verified by quantitative RT-PCR in comparison with untreated control cells. Each experiment was performed in triplicate and the runs were repeated. Error bars indicate the standard deviation among the samples. Statistical differences between C3-treated and control cells were determined using a two-sided Student’s t test (*P £ 0.05; **P £ 0.01; ***P £ 0.001). (B) For western blot analysis, HT22 cells were exposed to 500 nM C3 or C3-E174Q for up to 6 days at 37 C. Cells were lysed and subjected to western blot analysis against cyclin D1 and b-actin. Western blots from representative experiments are shown (n = 3). (C) Densitometric measurements for cyclin D1 (means ± SE) from three independent experiments. The results were not statistically significant.
C3 induced morphological changes in HT22 cells (including cell rounding and pronounced formation of retraction fibres), RhoA ADP-ribosylation and RhoA degradation. Furthermore, C3 exhibited anti-proliferative and anti-apoptotic activity. These effects have been reported for cell-permeable versions of C3 in various cellular systems including other neuronal cell lines such as PC-12 cells. All these effects have been attributed to C3-catalysed ADP-ribosylation. Consistent with this view, we found that C3-induced cell rounding as well as the anti-apoptotic activity strictly depend on Rho ADP-ribosylation. In particular, apoptosis induced by serum starvation or staurosporin is specifically inhibited by C3 (but not by C3-E174Q). Recently, it was shown that inhibition of the Rho ⁄ ROCK (Rho associated kinase) pathway by C3 exoenzyme rescues transplanted cells from apoptosis both in vitro and in vivo [38]. Another study showed that inhibition of RhoA activity by C3 attenuated thrombin-induced cell
death in cultured neurons and astrocytes [39]. In agreement with these studies, we show down-regulation of caspases and pro-apoptotic Bax in C3-treated HT22 cells in which apoptosis was induced by serum starvation or staurosporin. In addition, we demonstrate inhibition of caspase-3 activity by C3. All these findings strongly suggest an anti-apoptotic effect of C3 in cultivated hippocampal HT22 cells. In animal models, beneficial neuro-protective and neuro-regenerative effects were observed with both C3 and C3-E174Q [40,41]. Possibly, anti-apoptotic or pro-apoptotic effects of C3 depends on tissue and cell types, but it appears that C3 is predominantly anti-apoptotic in neuronal cells. Whereas C3 specifically exhibited cell rounding and anti-apoptotic activity, both C3 and C3-E174Q reduced cell proliferation as well as cyclin D1 expression with comparable kinetics. Thus we conclude that C3-E174Q enters HT22 cells comparably to C3. C3-E174Q did not exhibit ART activity, as evidenced
FEBS Journal 279 (2012) 2657–2671 ª 2012 The Authors Journal compilation ª 2012 FEBS
2663
Effect of C3 exoenzyme on hippocampal cells
A. Rohrbeck et al.
Fig. 8. Effect of C3 on apoptosis of HT22 cells. (A) HT22 cells were cultured for 24 or 48 h in serum-free medium followed by incubation with 1000 nM C3 or C3-E174Q in serum-free medium for a further 24 or 48 h (total 72 h). Summary of differential expression of various apoptosis genes verified by quantitative RT-PCR in comparison with 72 h serum-free control cells. Each experiment was performed in triplicate and the runs were repeated. Error bars indicate standard deviation among the samples. Statistical differences between C3-treated and 72 h serum-free control cells were determined using a two-sided Student’s t test (*P £ 0.05; **P £ 0.01). (B) For western blot analysis, cells were cultured for 24 or 48 h in serum-free medium followed by incubation with 1000 nM C3 or C3-E174Q in serum-free medium for a further 24 or 48 h (total 72 h) at 37 C. Cells were lysed and subjected to western blot analysis for p53, Bax and b-actin. Western blots from representative experiments are shown (n = 3). (C) Densitometric measurements for Bax (means ± SE) from five independent experiments. Statistical differences between C3-treated and 72 h serum-free control cells were determined using a two-sided Student’s t test (*P £ 0.05).
by analysing ART activity in the presence of high C3-E174Q concentrations in a cell-free system or upon prolonged treatment of HT22 with C3-E174Q for up to 5 days. Not even traces of Rho ADP-ribosylation were detected by the detection methods used (effector pull-down assay, gel shift assay, direct and sequential ADP-ribosylation). The biological activity of 2664
C3-E174Q thus appeared to be independent of ART activity. In addition to reduced cyclin D expression, C3-E174Q induced (late) RhoB expression. The kinetics of C3-E174Q-induced late RhoB expression clearly differed from the kinetics of early RhoB expression induced by C3. Notably, RhoB expressed in C3-E174Q-treated cells was not ADP-ribosylated,
FEBS Journal 279 (2012) 2657–2671 ª 2012 The Authors Journal compilation ª 2012 FEBS
A. Rohrbeck et al.
Fig. 9. Effect of C3 on regulation of caspases. HT22 cells were cultured for 24 or 48 h in serum-free medium followed by incubation with 1000 nM C3 or C3-E174Q in serum-free medium for a further 24 or 48 h (total 72 h). Cells were lysed and submitted to western blot analysis against caspase-3 (A), caspase-9 (B) and b-actin. Western blots from representative experiments are shown (n = 3).
further excluding the presence of ART activity in C3-E174Q-treated cells. In C3-treated cells, RhoA and RhoB were ADP-ribosylated and thus inactive, whereas in C3-E174Q-treated cells, newly expressed RhoB was not ADP-ribosylated and was thus active (Fig. S2). Inhibited proliferation of C3-treated cells may therefore be attributed to RhoA inactivation, while inhibited proliferation of C3-E174Q-treated cells instead appears to depend on active RhoB. RhoB expression results from ‘de-suppression’ of RhoA-mediated suppression of the rhoB promoter, and is thus intrinsically tied to RhoA inactivation [30]. Against this background, the observation that C3-E174Q induced RhoB expression suggests that C3-E174Q inhibits RhoA by a mechanism that is independent of ART activity. C3-E174Q-induced RhoB expression was less pronounced compared to
Effect of C3 exoenzyme on hippocampal cells
Fig. 10. Caspase-3 ⁄ 7 activity in cell lysates. (A) HT22 cells were cultured for 24 h in serum-free medium followed by incubation with 1000 nM C3 or C3-E174Q in serum-free medium for a further 48 h. Caspase-Apo-ONE reagent (100 lL) was added, and cells were further incubated for 30 min at room temperature. Fluorescence was measured on a spectrofluorometer with 485 nm excitation and 520 nm emission. Values are means ± SD (n = 3). Statistical differences between C3-treated and control cells were determined using a two-sided Student’s t test (*P £ 0.05; **P £ 0.01). (B) HT22 cells were cultivated in Dulbecco’s modified Eagle’s medium with 10% fetal bovine serum. Cells were incubated with staurosporin 1 h before addition of 1000 nM C3 or C3-E174Q, and further incubated for 48 h at 37 C. Caspase-Apo-ONE reagent (100 lL) was added, and fluorescence was measured after 30 min. Data are means of triplicate experiments. (C) Acridine orange ⁄ ethidium bromide staining was performed to show the viability of the cells (green: viable, orange: apoptosis). HT22 cells were cultivated in Dulbecco’s modified Eagle’s medium with 10% fetal bovine serum. Cells were incubated with staurosporin 1 h before addition of 1000 nM C3 or C3-E174Q, and further incubated for 48 h at 37 C. Apoptotic cells show a degraded nucleus and non-intact cell membrane, and were stained by both acridine orange and ethidium bromide. Magnification · 20.
FEBS Journal 279 (2012) 2657–2671 ª 2012 The Authors Journal compilation ª 2012 FEBS
2665
Effect of C3 exoenzyme on hippocampal cells
A. Rohrbeck et al.
C3-induced RhoB expression, allowing the conclusion that C3-E174Q inhibited RhoA less efficiently compared to C3. This view is consistent with the observation that C3 (but not C3-E174Q) reduced the cellular level of RhoA-GTP in HT22 cells. The mechanism by which C3-E174Q inhibits RhoA is most likely distinct from a covalent modification and may include RhoA sequestration. This is supported by the findings of microscale thermophoresis. In these experiments, RhoA binds with four- to fivefold higher affinity to C3-E174Q than to C3. It is conceivable that RhoA is retained in the cytosol through sequestration by C3-E174Q. In summary, we demonstrate that murine hippocampal HT22 cells are sensitive to both C3 and C3-E174Q. C3 induced cell rounding, inhibition of cell proliferation and anti-apoptotic effects. C3-E174Q induced RhoB expression and inhibition of cell proliferation independently of ART activity, with RhoB probably being involved in anti-proliferative effects.
Experimental procedures Cell culture Murine hippocampal HT22 cells were cultivated in Dulbecco’s modified essential medium (Biochrom AG, Berlin, Germany, +10% fetal bovine serum, 100 gÆmL)1 penicillin, 100 unitsÆmL)1 streptomycin and 1 mM sodium pyruvate) at 37 C and 5% CO2. Upon confluence, cells were passaged. For the growth kinetics experiments, cells were seeded onto 3.5 cm plates at a concentration of 50 000 cellsÆmL)1. Cells were grown for 24 h before treatment, and then C3 exoenzyme from Clostridium botulinum was added to the medium at a concentration of 500 nM. After 12, 24 and 48 h, the medium was removed and the cells were washed with NaCl ⁄ Pi. After the indicated time points, cells were detached with 0.25% trypsin solution. A 1 mL volume of culture medium was added to neutralize the trypsin action. Cells were counted every day by the trypan blue exclusion assay using a Neubauer counting chamber (Carl Roth GmbH, Karlsruhe, Germany) to determine the number of viable cells. For apoptosis experiments, cells were seeded onto 3.5 cm plates at a concentration of 100 000 cellsÆmL)1. After 24 h, the medium was changed to serum-free medium, and 48 h later, cells were treated with 1000 nM C3 in serum-free medium. The medium was replaced every 48 h by new serum-free protein-containing medium.
Cell lysis for western blot analysis After toxin treatment, cells were washed and scraped into Laemmli sample buffer. The obtained suspension was sha-
2666
ken at 37 C for 10 min. Ultrasonic disruption was performed using a cycle of 10 · 5 s, 5 · 10% sonic energy with a sonotrode (Bandelin Electronic, Berlin, Germany). The lysate was then incubated at 95 C for 10 min and subjected to SDS ⁄ PAGE.
Western blot analysis Complete lysate proteins were separated using SDS ⁄ PAGE and subsequently transferred onto nitrocellulose membranes by a tank blot system. The membranes were blocked using 5% w ⁄ v non-fat dried milk for 60 min; incubation with primary antibody was performed overnight at 4 C, and treatment with the secondary antibody was performed at room temperature for 1 h. For immunoblotting, the following primary antibodies were used: RhoA was identified using a mouse monoclonal IgG from Santa Cruz Biotechnologies (Santa Cruz, CA, USA); antibodies against RhoB (rabbit polyclonal IgG) were obtained from Bethyl Laboratories (Montgomery, TX, USA); Bax (p-19) monoclonal antibody was purchased from Santa Cruz Biotechnologies, pAkt (Ser473), caspase-3, p53 and cyclin D1 antibodies were purchased from Cell Signaling Technologies, and caspase-9 rabbit polyclonal antibody antibody was purchased from Biomol GmbH (Hamburg, Germany); Identification of C3bot was achieved using an affinity-purified rabbit polyclonal antibody that was raised against the full-length exoenzyme C3bot (accession number CAA41767). Actin (Sigma-Aldrich, Munich, Germany) was used as a loading control. For the chemiluminescence reaction, ECL Femto (Pierce ⁄ Thermo Fisher Scientific Inc., Rockford, IL, USA) or Immobilon (Millipore, Schwalbach, Germany) was used. All signals were analysed densitometrically using KODAK 1D software (Kodak GmbH, Stuttgart, Germany) and normalized to b-actin signals.
Expression and purification of recombinant C3 proteins C3 wild-type and C3-E174Q were expressed as recombinant glutathione S-transferase (GST) fusion proteins in Escherichia coli TG1 harbouring the respective DNA fragment in plasmid pGEX-2T (Amersham Life Sciences, Arlington Heights, IL, USA). Bacteria were grown at 37 C in 1 L of LB medium containing 100 lgÆmL)1 ampicillin to an attenuance at 595 nm of 0.7. Isopropyl-thio-b-D-galactoside was added to a final concentration of 0.2 mM, and the cultures were incubated at 37 C for another 3 h. Bacteria were sedimented at 7700 g (15 min, 4 C) and resuspended in 20 mL lysis buffer (50 mM Tris, pH 8, 150 mM NaCl, 2 mM MgCl2, 2 mM dithiothreitol), complemented with EDTAfree protease inhibitor cocktail (Roche Pharma AG, Grenzach-Wyhlen, Germany). Bacteria were lysed by means of a French pressure cell press system (SIM Aminco Spectronic
FEBS Journal 279 (2012) 2657–2671 ª 2012 The Authors Journal compilation ª 2012 FEBS
A. Rohrbeck et al.
Instruments, New York, USA), and the bacterial debris was sedimented at 12 000 g (15 min, 4 C). The supernatant was added to 4 mL of a 50% slurry of glutathione– Sepharose 4B beads in lysis buffer, and incubated for 4 h at 4 C while shaking. The slurry was poured into a disposable Econo-Pac chromatography column (Bio-Rad Laboratories GmbH, Mu¨nchen, Germany) and washed five times with 10 bed volumes of lysis buffer. For cleavage of the fusion proteins from glutathione S-transferase, the beads were incubated with five ‘‘NIH’’ units of thrombin for 12 h at 4 C. Thereafter, the protein was eluted with 12 mL lysis buffer. Thrombin was removed by precipitation with benzamidine–Sepharose beads (AP Biosciences, New York, USA). After centrifugation at 500 g (10 min, room temperature), the supernatant was exchanged with 20 mM HEPES pH 7.5 using PD-10 columns (GE healthcare, Mu¨nchen, Germany), sterile-filtered (0.22 lm) and used for cell culture experiments as indicated. ART activity was measured by an in vitro ADP-ribosylation assay.
ADP-ribosylation of Rho in HT22 cells To verify the effectiveness of ADP-ribosylation of Rho by C. botulinum exoenzyme C3, the HT22 cells were either left untreated or were incubated with increasing concentrations of the exotoxin (10–1000 nM) for 8–24 h. The cells were then washed with NaCl ⁄ Pi and scraped into 100 lL of lysis buffer (20 mM Tris ⁄ HCl (pH 7.4), 1% Triton X-100, 10 mM NaCl, 5 mM MgCl2, 1 mM phenylmethanesulfonyl fluoride, 5 mM dithiothreitol). The obtained suspension was shaken at 37 C for 10 min. Ultrasonic disruption was performed using a cycle of 10 · 5 s, 5 · 10% sonic energy with a sonotrode (Bandelin Electronic). Protein concentrations were measured by the Bradford method. Cell lysates containing equal amounts of protein were incubated with 1000 nM recombinant C. botulinum exoenzyme C3 and 1 lCi [32P]NAD (Amersham Life Sciences, Arlington Heights, IL, USA) in 20 lL of 4· buffer containing 50 mM HEPES (pH 7.3), 10 mM MgCl2, 10 mM dithiothreitol, 10 mM thymidine and 10 lM NAD at 37 C for 20 min. The reaction was terminated by addition of Laemmli sample buffer, and then incubated at 95 C for 10 min. Samples were resolved by SDS ⁄ PAGE on 15% gels, and the ADP-ribosylated Rho was analyzed by phosphorimaging (Cyclone, Packard American Instrument, MA, USA).
Pull-down assay RhoA activity was measured by pull-down assay using a GST fusion protein of the Rho binding domain. HT22 cells were homogenized in lysis buffer [20 mM Tris ⁄ HCl (to a total of 25 lL) pH 7.4], 1% Triton X-100, 10 mM NaCl, 5 mM MgCl2, 1 mM phenylmethanesulfonyl fluoride, 5 mM dithiothreitol). Cleared lysates were incubated for 60 min at
Effect of C3 exoenzyme on hippocampal cells
4 C with GST–Rho binding domain-coupled beads (20– 30 lg per sample). Precipitates were washed three times with the binding buffer and suspended in the SDS sample buffer. Proteins were separated by 15% SDS ⁄ PAGE and transferred to a nitrocellulose membrane. GTP-bound RhoA and total RhoA were detected by western blot analysis using a monoclonal RhoA antibody (Santa Cruz Biotechnology).
RNA isolation RNA extraction was performed using an RNeasy Mini Kit (Qiagen, Santa Clarita, CA, USA) according to the manufacturer’s instructions. The RNA concentration of the samples was then measured using a NanoDrop ND-1000 spectrophotometer (PeqLab Biotechnology GmbH, Erlangen, Germany).
RT-PCR RNA (2 lg) from each sample was mixed with an appropriate quantity of water (to a total of 25 lL) and denatured for 5 min at 65 C by using an Eppendorf Thermomixer (Eppendorf AG, Hamburg, Germany). The reaction mix for the reverse transcription comprised 2 lL 10· reverse transcription buffer (Omniscript RT-Kit, Qiagen, Hilden, Germany), 2 lL dNTPs, 2 lL random hexamers (primers), 0.25 lL RNasin and 1 lL reverse transcriptase (Omniscript RT Kit, Qiagen, Hilden, Germany), and was added previously cooled samples. The composition of the mix is appropriate for an RNA concentration between 50 ng and 2 lg. The samples were subsequently incubated for 1 h at 37 C. During this step, the RNA is converted to cDNA. The enzymes were then denaturated by incubation for 5 min at 95 C in the Eppendorf Thermomixer. Finally the samples were cooled and frozen at )20 C. Real-time RT-PCR measurements were performed using an ABI PRISM 7500 sequence detection system instrument (Applied Biosystems, Carlsbad, CA, USA). PCR reactions were performed according to the manufacturer’s instructions using GoTaq qPCR Master Mix (Promega GmbH, Mannheim, Germany). Gene expression levels were normalized to those of b2-microglobulin, which was found to be stably expressed. The specificity of primers was confirmed by agarose-gel electrophoresis of PCR products. Each experiment was performed in triplicate and the runs were repeated. Mean CT values were calculated using the ABI PRISM software, and relative gene expression levels were expressed as the difference between mean CT values for the target gene and the control gene b2-microglobulin. A negative control was always included in RT-PCR experiments. Additionally, a paired two-tailed Student’s t test was performed, and the results were considered significant when the P value was
