INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Plants can recognize certain pathogens and can activate defense mechanisms that restrict pathogen growth at the site of infection. These incompatible interactions are characterized by the synthesis of pathogenesis-related proteins that exhibit antimicrobial activities, thickening and hardening of cell walls, and accumulation of antimicrobial compounds called phytoalexins (Lamb et al., 1989; Dixon et al., 1994). In some incompatible interactions the plant defense response is accompanied by the death of plant cells surrounding the sites of pathogen infection within a few hours of pathogen contact (Dangl et al., 1996). This inducible cell death response known as the hypersensitive response (HR) prevents the spread of pathogens into healthy tissues (Goodman and Novacky, 1994). Often the HR is associated with increased resistance, throughout the plant, to subsequent infection by a broad spectrum of pathogens that would normally cause a susceptible interaction. This type of resistance is called systemic acquired resistance (Ryals et al., 1996).

Plant resistance requires a recognition process between components of the host and the pathogen, which is mediated by specific receptors and pathogen- or plant-derived signal molecules called elicitors (Ebel and Cosio, 1994). This interaction initiates the activation of complex signal transduction pathways that generate second messengers and trigger the inducible defense responses. Characteristic early events occur rapidly, including membrane potential changes, ion fluxes, and active oxygen species (AOS) production (Goodman and Novacky, 1994; Levine et al., 1994; Hammond-Kosack and Jones, 1996). These rapid responses are followed by later responses such as the production of phytoalexins and the transcriptional activation of so-called defense genes. For a number of incompatible interactions, changes in the distribution of plant microtubules (MTs) and microfilaments during fungal penetration processes have been reported (Kobayashi et al., 1992, 1994; Gross et al., 1993; Skalamera and Heath, 1998). In animal cells, changes in cytoskeleton dynamics and organization occur rapidly in response to the activation of signaling pathways and this may contribute to the transmission of signals to downstream targets. The evidence of signaling molecules that interact with cytoskeleton indicates also that the cytoskeleton is likely to be critical to the spatial organization of signal transduction. One of the cytoskeleton-associated proteins involved in signal transduction is dynein light chain that binds to IB, which is a negative regulator of the transcription factor, NFB. It was shown that MT depolymerization by drugs leads to IB destruction through a kinase-dependent mechanism, allowing NFB to bind DNA and stimulate transcription (Gundersen and Cook, 1999). In ML-1 human cells, MTs have been integrated in the signal transduction pathway that controls the antiapoptotic MCL1 gene expression. It has been shown that MT breakdown stimulated a mitogen-activated protein kinase- (MAPK) mediated pathway leading to the increase of MLC1 expression (Townsend et al., 1998). In human breast cancer cells MT damage activates a signal transduction pathway that ultimately leads to apoptosis. It was also suggested that MT damage induced loss of Bcl2 anti-apoptotic function by hyperphosphorylation through cAMP-dependent protein kinase (Srivastava et al., 1998).

In this study we have examined the microtubular cytoskeleton dynamics and the viability of tobacco (Nicotiana tabacum cv Xanthi) cells in response to two different elicitors, cryptogein, a protein secreted by Phytophthora cryptogea, and oligogalacturonides (OGs) derived from the plant cell wall. In tobacco plants cryptogein triggered a hypersensitive-like response and induced systemic acquired resistance against a broad spectrum of pathogens (Ricci et al., 1989; Bonnet et al., 1996), whereas OGs induced defense responses, but failed to trigger cell death (Darvill and Albersheim, 1984; Mathieu et al., 1991). The cryptogein transduction pathway has been previously investigated with tobacco cell suspensions. These studies showed that cryptogein specifically interacts with high-affinity binding sites on the plasma membrane (Wendehenne et al., 1995; Bourque et al., 1998, 1999) and induces early events including protein phosphorylation (Viard et al., 1994), a large calcium influx (Tavernier et al., 1995), extracellular medium alkalinization (Blein et al., 1991), chloride and potassium efflux, plasma membrane depolarization, activation of a plasma membrane NADPH oxidase (Pugin et al., 1997), activation of MAPKs (Lebrun-Garcia et al., 1998), and defense gene activation (Suty et al., 1995; Petitot et al., 1997). The same early effects (calcium influx, activation of MAPKs, extracellular medium alkalinization, and H₂O₂ production) were monitored in tobacco cells in response to OGs (Mathieu et al., 1991, 1996; Binet et al., 1998; Lebrun-Garcia et al., 1998). The main difference is a higher calcium influx and a stronger MAPK activation induced by cryptogein compared with OGs (Binet et al., 1998; Lebrun-Garcia et al., 1998).

Here we report that cryptogein treatment triggers a loss of MT network, which depends on calcium influx and occurs independently of AOS production. In addition, cryptogein induces a calcium-dependent cell death. In comparison, OGs do not affect the MT network nor cell viability. The molecular basis of cryptogein-induced MT disruption and its potential significance with respect to cell death are discussed.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Microtubular Responses in Tobacco Cells Treated with Cryptogein or OGs

The organization of MTs in tobacco cells was examined using classical fluorescence techniques and confocal microscopy. Control cells displayed a typical randomly oriented intact microtubular network (Wymer et al., 1997) during the assay (Fig. 1A). In 25 nM cryptogein-treated cells, a progressive depolymerization of MTs occurred during the 1st h of treatment. At 15 min, a partial disappearance of the cortical MT array was observed (Fig. 1B). At 30 min, some visualized MTs had the appearance of beaded MT bundles and of fluorescent aggregates, which may correspond to remaining pieces of MT bundles attached to the plasma membrane (Fig. 1C). After 1 h of cryptogein treatment, a dramatic depolymerization of cortical MTs was observed. Only a few MT bundles were present within the cytoplasm, associated or not with the nucleus (Fig. 1D). In contrast, the microtubular network of tobacco cells was not affected during the three 1st h of OG treatment (Fig. 1E).

View larger version (58K): [in this window] [in a new window]   Figure 1.   Distribution of MTs in tobacco cells during treatment with cryptogein or oligogalacturonides. Cells were analyzed by indirect immunofluorescence microscopy using anti--tubulin antibodies as described in "Materials and Methods." Control cells after 1 h (A), 25 nM cryptogein-treated cells for 15 min (B), 30 min (C), and 1 h (D), OGs-treated cells (50 µg mL1) for 3 h (E), and 2.5 nM cryptogein-treated cells for 1 h (F). The number of sections and the resultant depth of cells are 18 and 9 µm (A), 12 and 4 µm (B), 10 and 2 µm (C), 12 and 12 µm (D), 14 and 7 µm (E), and 12 and 2 µm (F), respectively. The scale bar represents 25 µm.

Comparison and Interaction with Drugs Affecting MT Organization

The effect of oryzalin, a tubulin polymerization-inhibitor, was compared with the effect of cryptogein on the MT network in tobacco cells. In cells treated with 5 µM oryzalin for 30 min, cortical MTs are disassembled and are dispersed as short pieces throughout the cell cortex (Fig. 2A). By comparison, cryptogein induced disruption inside the MT bundles (beaded MT phenomena) in the cortical area (Fig. 1C). Moreover, cotreatments with oryzalin and cryptogein induced a complete disruption of cortical MT after 30 min (Fig. 2B). Thus oryzalin and cryptogein act differently and complementarily, leading to a fast and complete MT network disruption.

View larger version (48K): [in this window] [in a new window]   Figure 2.   Analysis of calcium and AOS involvement in cryptogein-induced MT disruption and effects of MT inhibitors. Cells were analyzed by indirect immunofluorescence microscopy using anti- -tubulin antibodies as described in "Materials and Methods." Cells treated with 5 µM oryzalin for 30 min (A), cells cotreated with 5 µM oryzalin and 25 nM cryptogein for 30 min (B), cells treated with 20 µM taxol for 30 min (C), cells cotreated with 20 µM taxol and 25 nM cryptogein for 30 min (D), cells treated with 2 mM EGTA for 1 h (E), cells cotreated with 25 nM cryptogein and 2 mM EGTA for 1 h (F), cells treated with 10 µM DPI for 30 min (G), and cells cotreated with 25 nM cryptogein and 10 µM DPI for 30 min (H). The number of sections and the resultant depth of cells are 12 and 6 µm (A), 5 and 5 µm (B), 13 and 13 µm (C),13 and 7 µm (D), 20 and 14 µm (E), 18 and 6 µm (F), 12 and 6 µm (G), and 12 and 4 µm (H), respectively. The scale bar represents 25 µm.

When tobacco cells were treated with taxol, an MT depolymerization inhibitor (20 µM), during 30 min, MTs appeared to be arranged in dense bundles (Fig. 2C). In 1-h taxol-pretreated cells, cryptogein treatment for 30 min triggered the depolymerization of MTs, indicating that the taxol does not prevent MT disruption induced by cryptogein (Fig. 2D).

Cell Death

In tobacco plants, cryptogein triggers a hyper-sensitive-like response corresponding to cell death (Ricci et al., 1989), whereas OGs induce defense responses, but fail to trigger cell death (Darvill and Albersheim, 1984; Mathieu et al., 1991). Using neutral red as vital dye, we examined the effects of cryptogein and OGs on the viability of tobacco cells after 1-h (time corresponding to the disruption of MT network) and 24-h treatments. The cells were still viable after a 1-h treatment with 50 µg mL¹ of OGs or 0.25 to 250 nM cryptogein (data not shown). After 24 h, cryptogein induced the death of tobacco cells in a dose-dependent manner (Fig. 3). In presence of 25 nM cryptogein (saturating concentration for other events; Binet et al., 1998), about 60% of the cells were dead. By contrast, cells were still viable after the 24-h treatment with OGs (Fig. 3).

View larger version (34K): [in this window] [in a new window]   Figure 3.   Effects of different concentrations of cryptogein (0.25 to 250 nM) and OGs (50 µg mL1) on tobacco cell viability. The percentage of dead cells was determined after 24 h of treatment by staining with neutral red. At least 500 cells were examined for each experiment and five independent experiments were performed for each treatment. Results are represented as the means of five independent experiments. SE bars are shown.

Effects of a Ca²⁺ Channel Blocker and a Ca²⁺ Chelator on MT Depolymerization and Cell Death in Cryptogein-Treated Cells

Calcium influx is one of the first required steps in the cryptogein transduction pathway (Tavernier et al., 1995; Pugin et al., 1997; Binet et al., 1998). To examine the possible involvement of Ca²⁺ in the cryptogein-induced MT depolymerization and cell death we added EGTA, a Ca²⁺ chelator or La³⁺, a Ca²⁺ channel blocker, just before addition of cryptogein. EGTA and La³⁺ have been shown to suppress cryptogein-induced responses such as extracellular alkalinization, AOS and phytoalexin production, plasma membrane depolarization, and anion efflux (Tavernier et al., 1995; Pugin et al., 1997). In the presence of 2 mM EGTA, a 1-h treatment with 25 nM cryptogein did not induce any destabilization of the microtubular network (Fig. 2F). In this condition the MT network was comparable with that observed in control cells (Fig. 1A) or cells treated with EGTA (Fig. 2E). In a similar manner, the MT network was not affected when cryptogein treatments were performed in presence of La³⁺ (data not shown). Moreover, La³⁺ reduced cryptogein-induced cell death in a dose-dependent manner (Fig. 4). In presence of 1 mM La³⁺, the percentage of cell death induced by a treatment with 25 nM cryptogein after 24 h of treatment decreased from 60% to 10%. Taken together these results indicate that calcium influx triggers MT depolymerization and cell death and occurs upstream of these two events.

View larger version (29K): [in this window] [in a new window]   Figure 4.   Effects of the calcium channel blocker, lanthanum, on cryptogein-induced cell death. The percentage of dead cells was determined after 24 h of treatment by staining with neutral red. At least 500 cells were examined for each experiment and five independent experiments were performed for each treatment. Results are represented as the means of five independent experiments. SE bars are shown.

Involvement of the Calcium Influx Rate in MT Depolymerization and Cell Death Induced by Cryptogein

To study the role of the intensity of the calcium influx in MT destabilization and cell death we used different concentrations of cryptogein or 50 µg mL¹ OGs, which induce different rates of calcium influx. The calcium influx measured after 1 h of treatment increased with increasing concentrations of cryptogein (Table I), as previously reported (Tavernier et al., 1995). Comparison of cryptogein- or OG-treated cells revealed that the calcium influx in OG-treated cells was equal to that reached in 1 nM cryptogein-treated cells (0.450 and 0.415 µmol ⁴⁵Ca²⁺ g¹ fresh weight, respectively; Table I; 50 µg mL¹ OGs being the saturating concentration for all the monitored events; Binet et al., 1998). It is interesting that MT destabilization did not occur in these conditions. Moreover, cell death was low in cells treated with 1 nM cryptogein after 24 h (Fig. 3). But when a higher calcium influx was reached with higher cryptogein concentrations (2.5-25 nM), MT destabilization and cell death occurred. For example, in tobacco cells treated with 2.5 nM cryptogein, MT destabilization was observed during the 1st h of treatment (Fig. 1F) and 40% of cells were dead after 24 h (Fig. 3).

View this table: [in this window] [in a new window]   Table I.   Effects of OGs (50 µg mL1) and increasing concentrations of cryptogein (0.25-25 nM) on 45Ca2+ uptake into tobacco cells The uptake was determined after 1 h of treatment. The data are the mean value of two assays from one representative experiment taken from three independent experiments.

Relationship between MT Depolymerization and AOS Production

The transient AOS production is another calcium-dependent response induced by cryptogein (Tavernier et al., 1995). To determine whether MT depolymerization and cell death depend on AOS production, the cryptogein-induced oxidative burst was inhibited by diphenylene iodonium (DPI), an inhibitor of the mammalian neutrophil NADPH oxidase (Cross and Jones, 1986) that was reported to inhibit the cryptogein-induced AOS production without affecting calcium influx (Pugin et al., 1997; Simon-Plas et al., 1997). The MT disassembly induced by cryptogein was not modified in presence of 10 µM DPI (Fig. 2H). Cells treated with 10 µM DPI (Fig. 2G) had a MT network comparable with that observed in control cells (Fig. 1A). In a similar manner, DPI did not inhibit cryptogein-induced cell death measured after 3 h of treatment. Longer treatment with cryptogein and DPI were not possible because DPI alone had a toxic effect. After a 24-h incubation of tobacco cells with 10 µM DPI, all the cells were dead (data not shown). Taken together our results indicate that AOS production was not involved in MT depolymerization and cell death, and occurs independently to MT depolymerization.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

In the present study we have examined the microtubular cytoskeleton in tobacco cells treated with two elicitors of defense reactions differing in their chemical nature and biological properties. Cryptogein is a proteinaceous elicitor that triggers HR-like responses in tobacco leaves (Ricci et al., 1989), whereas OGs have no necrotic activity in plant tissues (Mathieu et al., 1994). In tobacco cells both elicitors induced similar early events, e.g. protein phosphorylation (Viard et al., 1994; Mathieu et al., 1996), calcium influx (Tavernier et al., 1995; Binet et al., 1998), K⁺ and Cl efflux (Pugin et al., 1997), MAPK activation (Lebrun-Garcia et al., 1998), AOS production (Tavernier et al., 1995; Binet et al., 1998), cytosol acidification, and plasma membrane depolarization (Mathieu et al., 1996; Pugin et al., 1997). The main differences include the higher calcium influx and higher MAPK activation level in cryptogein-treated tobacco cells compared with OGs (Binet et al., 1998; Lebrun-Garcia et al., 1998). The comparison of cryptogein and OG effects on microtubular cytoskeleton in tobacco cells indicate that the integrity of MT network appears unaffected in OG-treated cells, whereas cryptogein treatment causes a rapid and severe disruption of microtubular network. In addition to MT destabilization, cryptogein induced the death of tobacco cells. Using lanthanum, a calcium channel blocker, our results demonstrate that cryptogein-induced MT destabilization and cell death depends on calcium influx as reported for other cryptogein-induced events: AOS production (Tavernier et al., 1995), MAPK activation (Lebrun-Garcia et al., 1998), plasma membrane depolarization, cytosol acidification, and chloride efflux (Pugin et al., 1997). We verified the connection between the Ca²⁺-dependent AOS production and MT depolymerization using DPI, an inhibitor of mammalian and plant NADPH oxidases (Cross and Jones, 1986; Pugin et al., 1997). AOS suppression by DPI did not prevent the MT depolymerization, suggesting that the MT depolymerization occurs independently of the AOS burst. On the other hand, MT dynamics should not be involved in AOS production. Oryzalin or taxol did not induce AOS production nor modify the cryptogein-induced AOS production (data not shown).

Ca²⁺ is now firmly established as an intracellular second messenger that couples a wide range of extracellular stimuli to specific responses in plant cells (Malho et al., 1998; McAinsh and Hetherington, 1998). Calcium is involved in signal transduction pathways of cryptogein and OGs. The differences of the calcium influx rate in cryptogein- or OG-treated cells could explain their efficiency or inefficiency respectively to trigger MT destabilization and cell death. Using different concentrations of cryptogein, which triggered different rates of calcium influx and a saturating concentration of OGs inducing a low rate of calcium influx, our results suggest that a critical threshold level of intracellular Ca²⁺ concentration may be essential to trigger MT depolymerization and cell death in cryptogein pathway. In a further step, MT disintegration should contribute to signal amplification by opening additional plasma membrane calcium channels whose activity depends on the MT state as reported by Thion et al. (1996). This could be one explanation for the prolonged increase of intracellular Ca²⁺ induced by cryptogein (Binet et al., 1998). In a similar manner, the activation of MAPKs induced by cryptogein may also be the result of MT destabilization as reported in ML-1 human cells (Townsend et al., 1998).

Ca²⁺-dependent MT depolymerization has been described in animals and plants (Keith et al., 1983; Fisher et al., 1996). In several reports, Ca²⁺-calmodulin complexes were shown to modulate the activity of MT-associated proteins (MAPs). For example, MAPs such as stable tubulin only polypeptides and the higher plant homolog of the elongation factor-1 lose their ability to stabilize neuronal MTs (Bosc et al., 1996) and to bundle carrot MTs (Durso and Cyr, 1994), respectively, in response to Ca²⁺-activated calmodulin. In carrot protoplasts, cortical MT destabilization was reported to be due to an increase in free intracellular Ca²⁺ and was mediated by calmodulin (Fisher et al., 1996). During the interaction between the cowpea-resistant cultivar and the cowpea rust fungus, an increase in the cytoplasmic calcium concentration preceded the disappearance of cortical MTs in epidermal cells (Xu and Heath, 1998).

Other data relate changes in MT stability with defense responses in incompatible interactions. The attempt of fungal hyphae (Phytophthora infestans) to penetrate parsley cells resulted in a local depolymerization of MTs at the penetration site (Gross et al., 1993). In a similar manner, disruption of MTs was observed in flax and cowpea cells undergoing the HR during the interaction with an incompatible race of the flax rust fungus (Kobayashi et al., 1994) and the cowpea rust fungus, respectively (Skalamera and Heath, 1998). Although the direct relationship between MT dynamics and HR remains to be elucidated, these authors strongly suggested the possibility that disruption of MTs might be a specific and early sign of a HR. In mammalian cells it has been reported that reorganization of the cytoskeleton probably contributes to dramatic changes in cells undergoing cell death. Loss of microtubular structure occurs in apoptotic HL-60 cells (Martin et al., 1994). Moreover, binding of type 1 human immunodeficiency virus on intestinal epithelial cells triggered MT disruption (Delézay et al., 1997). In our study the causal link between the MT destabilization and the cell death is not verified because taxol, the only drug available up to now to abolish MT dynamics (Vallee and Collins, 1986), did not suppress the effects of cryptogein on MT destabilization.

In yeast and animal cells, changes in MT dynamics are regulated by several MT effectors that modulate MT assembly or sever MTs during interphase and mitosis (Cassimeris, 1999). These proteins interact with tubulin and/or MTs and are potential targets of signal transduction pathways. They can be divided into two main classes: proteins capable of stabilizing MTs (the MAPs) and proteins that destabilize MTs. Several MT destabilizers have recently been described, notably the yeast Kar3p and the Xenopus XCMK1 that destabilize MTs during the cell cycle by increasing plus- or minus-end MT catastrophes, respectively (Endow et al., 1994; Walczak et al., 1996). Stathmin, an ubiquitous cytosolic phosphoprotein, inhibits MT growth by sequestering tubulin (Belmont and Mitchison, 1996) and finally, MT-severing proteins such as katanin, which cut MTs at internal sites and so generate an increase of free ends (Ahmad et al., 1999). Stathmin and katanin can be targeted for destruction of MT arrays by phosphorylation (Melander-Gradin et al., 1997). The phosphorylation-dependent inactivation of MT stabilizing activity of MAPs during the cell cycle or cell morphogenesis has been also reported. This process may be mediated by MAPKs, which were shown to negatively regulate MAP2 and MAP4 MT-stabilizing activity (Hoshi et al., 1992). In plant cells, regulators of MT dynamics are not yet characterized. Therefore, one may speculate that the cryptogein-induced destabilization of cortical MT arrays we observed may be the consequence of the inactivation of MAPs or the activation of destabilizing effectors. The disruption of MT bundles in cryptogein-treated cells is localized all along the MT bundles and may suggest that it could be triggered by MT-destabilizing factors such as katanin, as described in animal cells. In addition, protein phosphorylation and MAPK activation are two early events induced by cryptogein, which may be involved in the phosphorylation of MT stabilizers or destabilizers. Thus investigation on the mode of action of cryptogein presents the opportunity to identify regulatory proteins involved in microtubular cytoskeleton dynamics in response to extracellular signals in plants and to study cell death signaling pathways. The characterization of components involved in the Ca²⁺-regulated MT depolymerization induced by cryptogein is in progress. Furthermore, the MT and actin cytoskeletons are coordinated for many cellular processes and it will be important to study if the cryptogein signaling pathway may integrate the responses of the two systems.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Plant Material and Elicitors

Tobacco (Nicotiana tabacum cv Xanthi) cells were cultivated as previously described (Tavernier et al., 1995). Cells were maintained in the exponential phase and subcultured 1 d prior to utilization. Cryptogein was purified according to Bonnet et al. (1996) and was a gift of M. Ponchet (Institut National de la Recherche Agronomique, Antibes). Purified OGs were a gift of M.A. Rouet-Mayer (Centre National de la Recherche Scientifique, Gif-sur-Yvette, France) and were used as a mixture of oligomers with degrees of polymerization of 7-20.

Chemicals and Radiochemical

Taxol and DPI (Sigma, St. Louis) were dissolved in dimethyl sulfoxide at 50 and 10 mM, respectively. Oryzalin (Dow AgroSciences, UK) was taken up in ethanol at 20 mM. ⁴⁵CaCl₂ (1.82 GBq mg¹) was from Amersham (Buckinghamshire, UK).

Elicitor and Chemical Treatments

Cells were collected during the exponential growth phase and washed by filtration in a suspension buffer containing 175 mM mannitol, 0.5 mM CaCl₂, 0.5 mM K₂SO₄, and 2 mM HEPES [4-(2-hydroxyethyl)-1-piperazineethane-sulfonic acid] adjusted to pH 5.75 with KOH. Cells were resuspended at 0.1 g fresh weight mL¹ with suspension buffer, equilibrated for 2 h on a rotary shaker (150 rpm, 24°C), and then treated with 25 nM cryptogein or 50 µg mL¹ OGs. EGTA (2 mM), La³⁺ (1 mM), and DPI (10 µM) were added to cell suspensions 30 s before the addition of cryptogein. Taxol (20 µM) and oryzalin (5 µM) were added 30 min before addition of cryptogein. Appropriate controls, with and without La³⁺, EGTA, DPI, and MT inhibitors, were included.

Detection of Cell Death

The vital dye neutral red was used to test for cell death. Accumulation of the dye within the vacuole was observed by light microscopy. Cells that lost membrane integrity and did not stain neutral red were considered dying (Naton et al., 1996). A stock solution of 1 mg mL¹ in water was diluted for staining to a final concentration of 0.01% (w/v) in suspension buffer, pH 7.5. At least 500 cells were examined for each experiment and five independent experiments were performed for each treatment.

Ca²⁺ Influx Measurements

Ca²⁺ uptake measurements were carried out as previously described (Tavernier et al., 1995). Calcium ⁴⁵Ca²⁺ (0.033 MBq g¹ fresh weight of cells) was added 5 min prior to treatment with cryptogein. After various treatment times (0-60 min), duplicate samples of 2 mL were withdrawn by filtration and washed once for 1 min and twice for 20 s on GF/A glass-microfiber filters (Whatman, Clifton, NJ) with 10 mL of 2 mM LaCl₃ in suspension buffer without Ca²⁺ to remove extracellular ⁴⁵Ca²⁺. Cells were scraped off, placed in scintillation vials, and weighed. Ten milliliters of Ready Safe Coktail (Beckman, Fullerton, CA) was added to the vials and the vials were gently shaken overnight before counting in a scintillation counter (LS 600 TA, Beckman).

Staining for MTs

Treated or control cells (0.5 g) were filtered and rinsed twice in suspension buffer. Cells were immediately fixed for 1 h in MT-stabilizing buffer (MSB; 50 mM PIPES [1,4-piperazinediethanesulfonic acid], pH 6.9, 10 mM EGTA, and 10 mM MgSO₄) containing 4% (w/v) freshly prepared paraformaldehyde and 0.2 M mannitol. Then, cells were washed four times for 30 min in 0.2 M mannitol in MSB and treated for 1 h at 25°C with an enzyme solution (2% [w/v]cellulase, 2% [w/v] pectolyase [Seishin Pharmaceuticals, Japan], and 0.2 M mannitol in MSB). Cell wall digestion was stopped by enzyme dilution in MSB containing 0.2 M mannitol. Cells were then attached to 1 mg mL¹ poly-L-Lys-coated coverslips and permeabilized for 20 min with 0.5% (w/v) Triton X-100 in MSB containing 0.2 M mannitol. After two washings for 5 min in MSB, cells were treated with 5% (w/v) bovine serum albumin (BSA) in MSB overnight at 4°C to block non-specific binding sites of antibodies. Then, cells were incubated for 3 h in the primary mouse monoclonal anti -tubulin antibody (N357, Amersham, 1:100 dilution in MSB containing 0.1% [w/v] BSA) and rinsed four times (15 min each) in MSB containing 0.1% (w/v) BSA. Cells were then incubated for 1 h in the secondary Cy 3-conjugated donkey antimouse antibody (Jackson, 1:250 dilution in MSB containing 0.1% [w/v] BSA). Cy 3 is a fluorescent dye analogous to rhodamine. After four washes in MSB (15 min each), cells were mounted in 50% (w/v) glycerol in phosphate-buffered saline (137 mM NaCl, 1.5 mM KH₂PO₄, 7 mM Na₂HPO₄, and 2.7 mM KCl, pH 7).

Confocal Microscopy

At least 40 cells were examined for each experiment and five independent experiments were performed for each treatment. Observations were performed with a confocal microscope (TCS 4D, Leica Microsystems, Wetzlar, Germany) equipped with an argon and krypton laser and epifluorescence attachments (excitation 568 nm, emission LP590). The ×63/1.4 oil-immersion objective was used for most images. Each image corresponds to the projection of optical sections taken from the cell cortex to the nucleus (i.e. one-half of the cell). The resultant depth (Z) of each projection is between 2 and 14 µm, depending of the thickness of the cells. The optical section number of each projection was between 5 and 20.