Regulation of a Proteinaceous Elicitor-induced Ca2+ Influx and Production of Phytoalexins by a Putative Voltage-gated Cation Channel, OsTPC1, in Cultured Rice Cells*

Background: Molecular mechanisms for elicitor-induced changes in cytosolic Ca2+ concentration and its molecular link with regulation of phytoalexin biosynthesis in plant immunity remain mostly unknown. Results: TvX-induced Ca2+ influx and the phytoalexin accumulations were suppressed in Ostpc1 knock-out cells. Conclusion: OsTPC1 plays a role in TvX-induced Ca2+ influx consequently required for the regulation of phytoalexin biosynthesis. Significance: Voltage-dependent plasma membrane Ca2+-permeable channel activity of the plant TPC1 was shown for the first time. Pathogen/microbe- or plant-derived signaling molecules (PAMPs/MAMPs/DAMPs) or elicitors induce increases in the cytosolic concentration of free Ca2+ followed by a series of defense responses including biosynthesis of antimicrobial secondary metabolites called phytoalexins; however, the molecular links and regulatory mechanisms of the phytoalexin biosynthesis remains largely unknown. A putative voltage-gated cation channel, OsTPC1 has been shown to play a critical role in hypersensitive cell death induced by a fungal xylanase protein (TvX) in suspension-cultured rice cells. Here we show that TvX induced a prolonged increase in cytosolic Ca2+, mainly due to a Ca2+ influx through the plasma membrane. Membrane fractionation by two-phase partitioning and immunoblot analyses revealed that OsTPC1 is localized predominantly at the plasma membrane. In retrotransposon-insertional Ostpc1 knock-out cell lines harboring a Ca2+-sensitive photoprotein, aequorin, TvX-induced Ca2+ elevation was significantly impaired, which was restored by expression of OsTPC1. TvX-induced production of major diterpenoid phytoalexins and the expression of a series of diterpene cyclase genes involved in phytoalexin biosynthesis were also impaired in the Ostpc1 cells. Whole cell patch clamp analyses of OsTPC1 heterologously expressed in HEK293T cells showed its voltage-dependent Ca2+-permeability. These results suggest that OsTPC1 plays a crucial role in TvX-induced Ca2+ influx as a plasma membrane Ca2+-permeable channel consequently required for the regulation of phytoalexin biosynthesis in cultured rice cells.

substances, known as phytoalexins. In rice, fourteen diterpenoid phytoalexins have been identified and can be classified into four groups, based on the structure of their hydrocarbon precursors: phytocassanes A-E, oryzalexins A-F, momilactones A and B, and oryzalexin S. Biosynthesis of these phytoalexins are induced by various elicitors including chitin fragments, cerebrosides and xylanase protein from Trichoderma viride (TvX)/ethylene-inducing xylanase (EIX) in rice-cultured cells, along with a variety of defense responses (7)(8)(9). Ca 2ϩ channel blockers inhibit cerebroside-induced phytoalexin production (8), suggesting possible involvement of Ca 2ϩ -permeable channels in the regulation of elicitor-induced phytoalexin biosynthesis. However, the molecular identity of the Ca 2ϩ channel(s) involved remains unknown.
Electrophysiological studies have characterized the activity of Ca 2ϩ channels localized at the plasma membrane (PM) and vacuolar membrane (VM) in many plant species (10 -11). The two-pore channel (TPC) family, originally isolated from rat, is homologous to the ␣1 subunit of vertebrate voltage-dependent Ca 2ϩ channels (12). Human TPCs mediate nicotinic acid adenine dinucleotide phosphate-induced Ca 2ϩ release from acidic organelles in HEK293 cells (13). Plant TPC family members have been characterized in several plant species. Arabidopsis AtTPC1 has been reported to show a slow-activating vacuolar cation channel activity (14 -15) and be involved in the sucroseinduced Ca 2ϩ rise (16), abscisic acid-induced repression of germination (14), and the stomatal response to extracellular Ca 2ϩ changes (14,17). NtTPC1s have roles in increasing Ca 2ϩ concentrations, defense-related gene expression, and regulation of programmed cell death triggered by cryptogein, an elicitor from an oomycete, in tobacco BY-2 cells (18). Characterization of the retrotransposon-insertional knock-out mutant of rice OsTPC1 revealed that OsTPC1 affects the sensitivity to TvX and plays a role in the regulation of TvX-induced activation of a MAP kinase and hypersensitive cell death in cultured rice cells (4). TPC1 has been suggested to amplify the elicitor-induced [Ca 2ϩ ] cyt increase (19). In contrast, increase in Ca 2ϩ concentrations, ROS generation, and gene expression induced by two MAMPs, elf18 and flg22, in the T-DNA insertional mutant of AtTPC1, attpc1-2, were comparable to the wild-type (15). A physiological role of the TPC family in plant innate immunity remains undefined.
In the present study, we characterized OsTPC1 and its knock-out mutant in cultured rice cells. Evidence presented here suggests that OsTPC1 is predominantly localized at the plasma membrane and has a role in the regulation of TvXinduced increases in [Ca 2ϩ ] cyt as well as phytoalexin biosynthesis.  (21), and Ostpc1 cells (22) were suspension-cultured at 25°C in a liquid broth L medium (23) and AA medium (24) containing 2,4-D (0.5 mg l Ϫ1 ) and subcultured in fresh medium every 7 days. Cells at 5 days after subculture were used for experiments on defense responses. Xylanase from Trichoderma viride was obtained from Sigma. N-acetylchitooligosaccharides were provided by Prof. Naoto Shibuya (Meiji University).

EXPERIMENTAL PROCEDURES
Monitoring of Cytosolic Calcium Concentration-Measurement of changes in cytosolic Ca 2ϩ concentration was performed essentially as described by Kurusu et al. (21). Briefly, apoaequorin-expressing rice cells (7 days after subculture) were incubated with 1 M coelenterazine for at least 12 h at 25°C. The cell suspension was transferred to a culture tube, set in a luminometer (Lumicounter 2500, Microtech Nition) and aerated by rotation (18). Ca 2ϩ -dependent aequorin chemiluminescence was measured after incubation for 15  Complementation Analysis-We transformed transgenic Ostpc1 cell lines expressing wild-type OsTPC1 and GUS (control) cDNA (4). The apoaequorin cDNA was cloned into the Ti-based vector pSMAB704 (27) and Agrobacterium-mediated transformation of the transgenic rice calli was performed. Transformed calli were screened and used for the complementation analysis.
Subcellular Membrane Fractionation and Immunoblot Analyses-PM and VM were isolated from cultured rice cells using an aqueous two-phase partition method comprising PEG/dextran (28 -29) and the sucrose/sorbitol method (30 -31), respectively. Rabbit polyclonal anti-OsTPC1 antibody was generated as described previously (32). The coding region of the linker domain of OsTPC1 (I359-S403) was amplified using sequence specific primers: I359-S403F, 5Ј-CACCATTGA-TGCTACTGGTCAGGGTTATCT-3Ј and I359-S403R, 5Ј-TCAACTCTGATCAAGCTCGGCAAAAATTAA-3Ј. A fusion protein consisting of the domain fused to a histidine-tag in the pDEST17 vector (Invitrogen) and transformed into Escherichia coli BL21-AI (Invitrogen). Inclusion bodies with the recombinant protein were obtained after induction at 37°C for 6 h and resolved using a preparative 15% SDS-PAGE gel. A ground polyacrylamide gel slice of the fusion protein was checked using MS/MS analysis and used to immunize rabbits by intradermal injections.
For immunoblotting analyses, protein samples were separated by 7.5% SDS-PAGE gel and blotted on to a PVDF membrane. The membrane was blocked in 1ϫ TTBS buffer (10 mM Tris-HCl, 150 mM NaCl, 0.05% Tween-20, pH 7.5) with 5% fat free milk overnight at 4°C. Blots were incubated with the affinity-purified anti-OsTPC1 and then with HRP-linked anti-rabbit IgG (GE Healthcare). Bands were detected using the chemiluminescent HRP substrate (Millipore) and a chemiluminescent analyzer, LAS3000 (GE Healthcare).
RT-PCR Analysis-First-strand cDNA was synthesized from 3 g of total RNA. PCR amplification was performed with genespecific primers (supplemental Table S1). Actin was used as a quantitative control. Aliquots of individual PCR products were resolved by agarose gel electrophoresis and visualized with ethidium bromide under UV light.
Real-time RT-PCR Quantification-First-strand cDNA was synthesized from 3 g of total RNA. Real-time PCR was performed using an ABI PRISM 7300 sequence detection system (Applied Biosystems Instruments) with SYBR Green real-time PCR Master Mix (TOYOBO) and gene-specific primers (supplemental Table S1). Relative mRNA abundances were calculated using the standard curve method and normalized to the corresponding OsActin1 mRNA levels. Standard samples of known template amounts were used to quantify the PCR products.
Phytoalexin Measurements-Phytoalexins were extracted from suspension-cultured rice cells after elicitation and analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS) as described previously (33).
Functional expression of membrane currents was investigated using the whole-cell patch clamp technique. Whole-cell currents were recorded with a CEZ-2200 patch clamp amplifier (NIHON KOHDEN). The resulting values were not corrected for liquid junction potential and leak currents were not subtracted. We used pCLAMP 8.2 software (Molecular Devices) for data analysis. The pipette solution contained 120 mM CsCl, 3 mM MgATP, 10 mM EGTA, and 10 mM HEPES (pH 7.1). The bath solution contained 40 mM BaCl 2 , 80 mM CsCl, and 10 mM HEPES (pH 7.4) (35). A ramp voltage protocol from 0 to Ϫ130 mV (holding potential, 0 mV; ramp speed, 130 mV s Ϫ1 ) was used.
Statistical Analysis-Statistical significance was determined using an unpaired Student's t test, with a maximum p value of Ͻ 0.05 required for significance. *, p Ͻ 0.05; **, p Ͻ 0.01; ***, p Ͻ 0.001. (4, 36 -38). In rice cell culture, external Ca 2ϩ was required for TvX-induced hypersensitive cell death, suggesting that Ca 2ϩ influx through the PM is indispensable for TvX-induced defense responses (4). However, the effect of TvX on changes in [Ca 2ϩ ] cyt has never been characterized. To analyze the involvement of Ca 2ϩ flux in TvX-induced defense responses, we measured TvX-induced changes in [Ca 2ϩ ] cyt using transgenic rice cell lines expressing apoaequorin, a Ca 2ϩsensitive photoprotein (21) (supplemental Fig. S1). As shown in Fig. 1A, TvX triggered a prolonged increase in [Ca 2ϩ ] cyt in a dose-dependent manner.

TvX Triggered a Prolonged Increase in Cytosolic Ca 2ϩ in Suspension-cultured Rice Cells-TvX triggers a variety of defense responses in plants
To elucidate the origin of Ca 2ϩ flux, we first performed pharmacological analyses. The addition of a Ca 2ϩ chelator, BAPTA, into the extracellular medium or the substitution of Ca 2ϩ free medium inhibited the TvX-induced increase in [Ca 2ϩ ] cyt ( Fig.  1B and supplemental Fig. S2). Similarly, a Ca 2ϩ channel blocker, La 3ϩ or Gd 3ϩ , and a voltage-dependent Ca 2ϩ channel inhibitor, nifedipine (11), suppressed the TvX-induced increase in [Ca 2ϩ ] cyt ( Fig. 1B and supplemental Fig. S2). A phospholipase C inhibitor, neomycin, and a potential endomembrane Ca 2ϩ -permeable channel inhibitor, ruthenium red, has been reported to suppress Ca 2ϩ release from intracellular Ca 2ϩ stores (39 -40). In contrast to BAPTA and La 3ϩ , neomycin, and ruthenium red scarcely inhibited the TvX-induced increase in [Ca 2ϩ ] cyt (Fig. 1B), suggesting that TvX-induced increase in [Ca 2ϩ ] cyt is predominantly due to the influx of extracellular Ca 2ϩ through voltage-dependent Ca 2ϩ -permeable channels.
Involvement of OsTPC1 in the TvX-induced Changes in Cytosolic Ca 2ϩ Concentration-A putative voltage-dependent cation channel, OsTPC1 has been suggested to be involved in TvXinduced defense responses, including activation of a MAPK and hypersensitive cell death (4). To test the possible involvement of OsTPC1 in the regulation of TvX-induced Ca 2ϩ rise, we generated Ostpc1 knock-out cell lines harboring apoaequorin. We confirmed the expression of apoaequorin mRNA in transgenic rice cells by RT-PCR. Several lines expressing similar levels of apoaequorin mRNA were selected for further experiments (supplemental Fig. S1).
As shown in Fig. 1C, the TvX-induced increases in [Ca 2ϩ ] cyt were partially but significantly suppressed in Ostpc1 knock-out cell lines in comparison with that of the wild-type. To confirm that this phenotype was due to the functional knock-out of OsTPC1, we performed a complementation analysis using Ostpc1 cells expressing both wild-type OsTPC1 and apoaequorin cDNA (supplemental Fig. S1). Transformation of the mutant with a control vector that carried GUS had no effect the TvX-induced Ca 2ϩ increase (Fig. 1D). In contrast, expression of OsTPC1 recovered the TvX-induced Ca 2ϩ increase (Fig. 1D), indicating that the observed mutant phenotype was attributable to OsTPC1. These results suggest that OsTPC1 participates in TvX-induced Ca 2ϩ rise in cultured rice cells.
Intracellular Localization of OsTPC1-Molecular and electrophysiological studies have shown that Arabidopsis TPC1 was mainly localized in the VM (14,41). In contrast, TPCs in monocots including OsTPC1 have been suggested to be localized in the PM (4,32,42). To confirm the intracellular localization of OsTPC1, we prepared an affinity-purified specific rabbit anti-OsTPC1 antibody, performed membrane fractionation using an aqueous two-phase partitioning method, and analyzed the intracellular localization of OsTPC1. Immunoblot analyses of crude extracts from suspension-cultured rice cells using the anti-OsTPC1 antibody detected a protein migrating with an apparent molecular mass of 87 kDa ( Fig. 2A). This band was absent when the antibodies were incubated with the recombinant antigen as a competitor, but not the recombinant GUS fused histidine tag ( Fig. 2A), indicating that the affinitypurified anti-OsTPC1 could detect OsTPC1 protein, specifically.
We fractionated total protein extracts into PM and VM fractions. Each fraction was obtained with little cross-contamination, as determined by immunoblot analyses using specific marker proteins (Fig. 2B). As shown in Fig. 2B, we detected OsTPC1 predominantly in the PM. This result is consistent with the previous reports and confirm that OsTPC1 is predominantly localized at the PM in cultured rice cells (32).
Involvement of OsTPC1 in the Regulation of Phytoalexin Biosynthesis-Pharmacological evidence suggests that the Ca 2ϩ influx in elicitor-induced phytoalexin biosynthesis is important in rice (8,43). In rice, ent-copalyl diphosphate synthase 4 (OsCPS4) and ent-kaurene synthase like-4 (OsKSL4); and ent-copalyl diphosphate synthase 2 (OsCPS2) and entkaurene synthase-like 7 (OsKSL7) are responsible for the biosynthesis of momilactones and phytocassanes, respectively (44 -46). The expression of all these cyclase genes was induced by TvX, which was significantly suppressed by pre-treatment with BAPTA or La 3ϩ , suggesting that the influx of extracellular Ca 2ϩ through the Ca 2ϩ -permeable channels is required for the expression of these cyclase genes upon recognition of the elicitor ( Fig. 3A and supplemental Fig. S3). Similarly, TvX-induced induction of these genes was suppressed in Ostpc1 knock-out cells and the suppression was restored after complementation of Ostpc1 (Fig. 3B and supplemental Fig. S4). Quantitative HPLC-tandem mass spectrometry analyses revealed that TvX induces the accumulation of momilactones and phytocassanes, which are major phytoalexins in rice (9). The level of momilactones continued to increase from the first measurement at 24 h, through to the final measurement at 120 h, and that of phytocassanes reached a maximum at 72 h and gradually decreased thereafter in response to TvX (Fig. 4, A and B).
In Ostpc1 knock-out cells, the accumulation of both momilactones and phytocassanes was significantly suppressed (Fig.  5A). The suppression in Ostpc1 knock-out cells was restored by introducing the OsTPC1 gene (Fig. 5B), suggesting that OsTPC1 plays a role in regulation of TvX-induced phytoalexin biosynthesis. The close correlation between TvX-induced changes in cytosolic Ca 2ϩ concentration and phytoalexin biosynthesis in Ostpc1 knock-out cells suggest the significance of the signaling role of Ca 2ϩ regulated by OsTPC1 at least in part in TvX-induced phytoalexin biosynthesis.
Ca 2ϩ -permeability of OsTPC1-Possible involvement of OsTPC1 in the TvX-induced [Ca 2ϩ ] cyt increase (Fig. 1C) and its localization at the PM (Fig. 2B) led us to hypothesize that OsTPC1 may function as a Ca 2ϩ -permeable channel at the PM in rice cells. However, no electrophysiological characterization has so far been reported for OsTPC1. We thus recorded the whole-cell current in HEK293T cells expressing GFP or GFP-OsTPC1 cDNA. Membrane fractionation analysis revealed that heterologously expressed OsTPC1 was localized to the PM at least partially in HEK293T cells (supplemental Fig. S5A). Volt-age-dependent currents carried by Ba 2ϩ instead of Ca 2ϩ (Ϫ230 Ϯ 57 pA at Ϫ130 mV, n ϭ 5) was observed in GFP-OsTPC1-expressing cells (Fig. 6, A and B and supplemental Fig.  S5B). On the other hand, we detected no such currents in GFPexpressing cells (Ϫ59 Ϯ 7 pA at Ϫ130 mV, n ϭ 5). Treatment of La 3ϩ significantly suppressed the voltage-activated Ba 2ϩ currents in GFP-OsTPC1-expressing cells, but not neomycin (supplemental Fig. S5C), indicating that OsTPC1 functions as a voltage-activated Ca 2ϩ -permeable channel.  (1,3,6,43).

DISCUSSION
Here, we showed that TvX triggered a prolonged change in [Ca 2ϩ ] cyt , mainly due to Ca 2ϩ influx from the extracellular space (Fig. 1, A and B). A sustained increase in [Ca 2ϩ ] cyt is postulated to have a key role in the induction of HR cell death (47)(48). Ca 2ϩ influx is required for TvX-induced HR cell death (4). Hence, a prolonged change in [Ca 2ϩ ] cyt due to TvX may be a prerequisite to activate HR cell death in rice cultured cells.
We generated apoaequorin-expressing Ostpc1 cells and found that TvX-induced [Ca 2ϩ ] cyt changes were partially suppressed in Ostpc1 cells (Fig. 1C). We thus postulate that multiple Ca 2ϩ -permeable channels are involved in the elicitor-triggered [Ca 2ϩ ] cyt changes. OsTPC1 heterologously expressed in HEK293T cells showed voltage-dependent Ca 2ϩ permeability through the PM (Fig. 6), suggesting that OsTPC1 functions as a Ca 2ϩ -permeable channel. The present results suggest that OsTPC1 plays a role as one of the multiple Ca 2ϩ -permeable channels activated by TvX and is involved in the elicitor-induced [Ca 2ϩ ] cyt change.
It has been reported that AtTPC1 is not involved in [Ca 2ϩ ] cyt changes triggered by two MAMPs, elf18 and flg22 (15). Interestingly, [Ca 2ϩ ] cyt changes induced by chitin fragments (N-acetylchitooligosaccharides) in Ostpc1 cells was almost comparable to the wild-type cells (supplemental Fig. S6). This apparent discrepancy may be explained by the differences in signaling pathways among elicitors. In fact, the temporal patterns of increased [Ca 2ϩ ] cyt are significantly different between TvX and chitin fragments: [Ca 2ϩ ] cyt increase triggered by chitin fragments is large and transient (supplemental Fig. S6) (21), while that induced by TvX is much smaller but sustained (Fig. 1C). The differences in the temporal pattern of [Ca 2ϩ ] cyt changes correlates with the induction of programmed cell death; almost no cell death is induced by chitin fragments, while TvX triggers programmed cell death, which is also affected by OsTPC1 (4). These results suggest that OsTPC1 may be one of the multiple Ca 2ϩ permeable channels activated by some specific elicitors but not all to trigger sustained increase in [Ca 2ϩ ] cyt and HR cell death.
The primary structure of AtTPC1 and OsTPC1 is similar. According to WoLF PSORT, a protein subcellular localization prediction program (49), both proteins are predicted to be localized at the PM. However, AtTPC1 and OsTPC1 are predominantly localized at the VM and PM, respectively (4, 14, 32). A FIGURE 2. Subcellular localization of OsTPC1. A, protein sample of crude membrane fraction (20 g per lane) was subjected to SDS-PAGE and immunoblotting analyses. OsTPC1 was detected using an affinity-purified anti-OsTPC1 antibody (lane 1). As a competitor, 0.5 mg ml Ϫ1 histidine-tagged OsTPC1 (I359-S403) was added at 1:1000 (lane 2) or 0.5 mg ml Ϫ1 histidinetagged GUS was added at 1:1000 dilution (lane 3). B, subcellular fractions of suspension-cultured rice cells were prepared using an aqueous two-phase system and a sucrose/sorbitol system. Protein samples (10 g per lane) of each fraction were subjected to SDS-PAGE and immunoblotting analyses. OsTPC1 was detected using an affinity-purified anti-OsTPC1 antibody. The purified fractions were assessed using a polyclonal antibody of plasma membrane H ϩ -ATPase and vacuolar H ϩ -ATPase subunit A (COSMO BIO).
human TPC2 mutant lacking a di-leucine motif in its N-terminal has recently been shown to be localized to the PM instead of the lysosome, suggesting that this motif is required for its localization to intracellular acidic organelles (50). Plant TPC family members have a similar motif in their N-terminal tail, suggesting that they may be localized to the VM. However, previous studies (4,32), as well as data presented here, show that OsTPC1 is mainly localized at the PM in cultured rice cells as well as in HEK293T cells (Fig. 2B and supplemental Fig. S5A). TaTPC1 from wheat has also been reported to be localized at the PM (42,51). Other unknown components that interact with OsTPC1 may regulate the intracellular localization of OsTPC1. Intracellular localization of the plant TPC family is an emerging subject that warrants further analysis.
The production of major diterpenoid phytoalexins, momilactones and phytocassanes, is triggered in rice upon recogni-  tion of various elicitors (7)(8)33). OsTGAP1, a basic leucine zipper (bZIP) transcription factor induced by chitin fragments, has recently been shown to be involved in the expression of biosynthetic genes of the diterpenoid phytoalexins including the upstream MEP pathway genes, and that overexpression of OsTGAP1 exhibited enhanced expression of those phytoalexin biosynthetic genes leading to hyperaccumulation of the diterpenoid phytoalexins (7). TvX also induced the accumulation of momilactones and phytocassanes in rice cells (Fig. 4) (9). However, the time course of this accumulation appeared to be more prolonged in comparison with chitin fragments (33), and expression of OsTGAP1 was not induced by TvX treatment (data not shown). These results suggest that phytoalexin biosynthesis is regulated by multiple pathways, and the time course, as well as the regulatory pathways, are different, at least in part, between chitin fragments and TvX.
Pharmacological analyses using Ca 2ϩ channel blockers, LaCl 3 and GdCl 3 , indicate that Ca 2ϩ influx via Ca 2ϩ channels is associated with production of phytoalexins (8). This is consistent with our findings that a Ca 2ϩ chelator, BAPTA or a Ca 2ϩ channel blocker, La 3ϩ , suppressed the TvX-induced expression of diterpene cyclase required for phytoalexin biosynthesis ( Fig.  3A and supplemental Fig. S3). Both phytoalexin accumulation and the expression of diterpene cyclase genes were partially suppressed in Ostpc1 cells and complemented by expression of wild-type OsTPC1 (Figs. 3B and 5). These results indicate that OsTPC1 is involved in regulation of TvX-induced sustained increase in [Ca 2ϩ ] cyt , and consequently of phytoalexin biosynthesis in rice-cultured cells.
Only the treatment with a Ca 2ϩ ionophore, ionomycin, did not induce the expression of diterpene cyclase genes (supplemental Fig. S7). This is consistent with our previous observation that though ionomycin trigger a rise in [Ca 2ϩ ] cyt , it does not necessarily trigger Ca 2ϩ -dependent downstream events such as ROS production in rice cells (21). The regulation of the  phytoalexin biosynthetic pathway may require not only Ca 2ϩ entry from the extracellular space, but also other signaling events triggered by TvX. These results also reinforce the concept that Ca 2ϩ flux and sustained [Ca 2ϩ ] cyt increase triggered by TvX and mediated by OsTPC1 at least in part may have a specific role in defense signaling.
Little is known of the signaling components connecting PAMPs/MAMPs-induced [Ca 2ϩ ] cyt changes and downstream defense responses. OsTPC1 plays a crucial role in TvX-induced activation of a MAP kinase, OsMPK6 (4). Chitin fragment-induced synthesis of diterpenoid phytoalexins has recently been shown to involve the OsMKK4-OsMPK6 MAP kinase cascade (52). OsMPK6 whose activation is regulated by OsTPC1 may also participate in the induction of phytoalexin biosynthesis induced by TvX. OsCIPK14/15 activated by binding of calcineurin B-like Ca 2ϩ sensor proteins are involved in various layers of TvX-induced defense responses including the expression of the same diterpene cyclase genes and phytoalexin production (9,53). The phenotypes of knockdown cell lines of the CIPKs (9) are similar with those of OsTPC1 (Fig. 5), suggesting the Ca 2ϩ -permeable channel OsTPC1 may act upstream of the Ca 2ϩ -regulated protein kinases, OsCIPK14/15.
In summary, the present results indicate that OsTPC1 has a role in the regulation of TvX-induced sustained increase in [Ca 2ϩ ] cyt leading to phytoalexin biosynthesis in rice cultured cells. Considering that TvX-induced increases in [Ca 2ϩ ] cyt and phytoalexin biosynthesis were impaired only partially in Ostpc1 knock-out cells, multiple Ca 2ϩ -permeable channels may act redundantly to bypass OsTPC1 to regulate TvX-induced defense responses. Cyclic nucleotide-gated Ca 2ϩ -permeable channels (CNGCs) may have recently been implicated in a variety of plant immune responses (54 -57). CNGCs may therefore be candidates for such Ca 2ϩ -permeable channels.