The heterotrimeric G protein α subunit acts upstream of the small GTPase Rac in disease resistance of rice

Abstract

We used rice dwarf1 (d1) mutants lacking a single-copy Gα gene and addressed Gα's role in disease resistance. d1 mutants exhibited a highly reduced hypersensitive response to infection by an avirulent race of rice blast. Activation of PR gene expression in the leaves of the mutants infected with rice blast was delayed for 24 h relative to the wild type. H₂O₂ production and PR gene expression induced by sphingolipid elicitors (SE) were strongly suppressed in d1 cell cultures. Expression of the constitutively active OsRac1, a small GTPase Rac of rice, in d1 mutants restored SE-dependent defense signaling and resistance to rice blast. Gα mRNA was induced by an avirulent race of rice blast and SE application on the leaf. These results indicated the role of Gα in R gene-mediated disease resistance of rice. We have proposed a model for the defense signaling of rice in which the heterotrimeric G protein functions upstream of the small GTPase OsRac1 in the early steps of signaling.

Heterotrimeric G proteins, a major group of signaling molecules involved in a variety of cellular activities in mammals, are mainly responsible for various cellular responses to external signals (1). In mammals, G proteins consist of α, β, and γ subunits, and at least 23 α, 6 β, and 12 γ genes are known (2). In plants, a number of pharmacological studies suggested that heterotrimeric G proteins are involved in a variety of signaling, including light reception (3), hormone signaling (4), and regulation of ion channels (5). However, direct evidence to support these observations has been obtained only recently (6). Analysis of mutations in a gene encoding the Gα subunit of rice termed dwarf1 (d1) showed that Gα is involved in stem elongation and the determination of seed shape in rice (7, 8) and influences gibberellin signal transduction (9). More recently, in Arabidopsis, mutants in the single-copy Gα subunit gene were shown to have reduced cell division in aerial tissues (10) and to lack regulation of the ion channel by the phytohormone abscisic acid in guard cells (11). Furthermore, involvement of Gα in phytochrome-mediated light signal transduction of Arabidopsis was also demonstrated by the study of transgenic plants overexpressing the Gα gene (12). Arabidopsis mutants lacking Gβ were recently shown to have effects in leaf, flower, and fruit development (13). Therefore, the importance and diverse functions of heterotrimeric G proteins in the signaling of plants are recently becoming clear; however, the molecular mechanisms of G protein signaling remain to be studied.

Many studies using inhibitors and agonists of heterotrimeric G proteins in several plant species have suggested that G proteins are involved in defense signaling (14–17). Particularly, changes in cytosolic Ca²⁺ concentrations, which are often observed in elicitor-treated plant cells, are assumed to be regulated by heterotrimeric G proteins (18). However, the roles of the heterotrimeric G protein in plant defense have not been directly tested by the use of G protein mutants.

In the present study, we addressed the role of the heterotrimeric G protein in the disease resistance of rice by using mutants of the single-copy Gα gene of rice. Furthermore, we also addressed the role of Gα in defense responses induced by elicitors of rice. We then examined a possible regulatory link between Gα and the small GTPase OsRac1, which has been shown to be an important intermediate in the defense signaling of rice. Results of these experiments led us to propose a model of defense signaling in rice in which Gα plays an important role upstream of the small GTPase OsRac1.

Methods

Plant Materials and Cell Cultures.

A japonica rice cv. Kinmaze as a wild type (WT) and four d1 mutants (248, 723, 1232, and 1361) induced by N-methyl-N-nitrosourea in the Kinmaze background (19) were used. Rice suspension cultures expressing the constitutively active and the dominant-negative OsRac1 have been previously described (20, 21).

Rice Transformation.

Details of the constitutively active OsRac1 construct and production of suspension cell cultures have been previously given (20). The Gα promoter∷gus fusion gene was constructed by fusing a 900-bp promoter region of the rice D1 gene with the gus reporter gene. The Agrobacterium-mediated transformation of rice was performed according to a published method (22).

Infection of Rice Plants with Rice Blast.

A strain of blast fungus (Magnophorthe grisea), TH 67–22 (race 031), was used in this study. Race 031 is avirulent with cv. Kinmaze. The growth conditions of the blast fungus and the method for the infection of leaf sheath cells have been previously described (21, 23). For the infection of leaf blades, a previously described method was used (24). For the infection of rice leaves with rice blast for Northern blot analysis, seedlings were soaked in conidial suspensions, placed in a growth chamber (relative humidity 100%, 23°C, dark) for 48 h, and then placed in a greenhouse. For the infection of transgenic rice plants expressing the Gα promoter∷gus fusion gene, 2 μl of the fungal suspension was inoculated on leaves of 3-week-old plants by using micropipette tips.

Quantification of H₂O₂.

The methods for the quantification of H₂O₂ have been described (25). Briefly, 0.4 g of cells was transferred to 2 ml of a fresh medium and maintained in suspension for 16 h at 30°C. Cells were transferred to 2 ml of a fresh medium containing sphingolipid elicitors (SE) (10 μg/ml) and incubated for various lengths of time after treatment. Aliquots at indicated times were collected and filtered through a 0.22-μm filter; 100-ml aliquots were placed in 24-well tissue culture plates containing a 1-ml xylenol orange buffer (0.25 mM FeSO₄/0.25 mM (NH₄)₂SO₄/25 mM H₂SO₄/10 mM sorbitol/12.5 mM xylenol orange) and incubated for 2 h at room temperature. H₂O₂ levels were quantified by a spectrophotometer (Beckman DU 640) at 560-nm absorbance.

Results

Gα mRNA Was Induced by Infection with an Avirulent Race of Rice Blast and by SE.

To examine whether Gα mRNA was induced by infection with pathogens, we infected rice plants with avirulent and virulent races of rice blast, and the induction of Gα mRNA was examined (Fig. 1A). When infected with an avirulent race of rice blast, the Gα mRNA abundance decreased quickly for 12 h, and from 24 h after infection, it began to increase until 3 d, whereas in mock-infected plants Gα mRNA levels remained essentially constant over 2 d. In leaf sheath cells infected with an avirulent race of rice blast, hypersensitive response (HR) started around 17 h after inoculation, and the highest frequency of cell death was observed at 31 h after inoculation (26). Therefore, the increase of Gα mRNA coincided with HR. In contrast, when rice plants were infected with a virulent race of the fungus, the Gα mRNA levels dropped after infection, and no induction of Gα mRNA levels at later stages was observed (Fig. 1B). These results showed that the virulent fungus may suppress transcription of Gα mRNA. Weak induction of PBZ1 mRNA indicated that leaves were indeed infected with rice blast. PBZ1 is a rice PR gene whose RNA is induced by infection by pathogens (23, 27). The reason for the initial decrease of Gα mRNA levels caused by an avirulent fungus remains to be studied.

Figure 1.

Induction of Gα mRNA by infection with rice blast and SE application on the leaf. (A) Induction of Gα mRNA by infection with an avirulent race of rice blast. Results of mock-infected plants are also shown. (B) Expression of Gα and PBZ1 mRNAs by infection with a virulent race of rice blast. (C) Induction of GUS activity by an avirulent race of rice blast in the leaf of transgenic rice expressing the Gα promoter∷gus fusion gene. Leaf samples were incubated in an 5-bromo-4-chloro-3-indolyl β-D-glucuronide (X-gluc) solution for 15 h at 37°C. Arrows indicate the region of infection. M, mock-infected; I, infected by rice blast. (Bar = 1 mm.) (D) No induction of GUS activity by a virulent race of rice blast in the leaf oftransgenic rice expressing the Gα promoter∷gus fusion gene. Arrows indicate the region of infection. M, mock-infected; I, infected by rice blast. (Bar = 1 mm.) (E) Induction of GUS activity by SE in the leaf of transgenic rice expressing the Gα promoter∷gus fusion gene. Leaf samples were incubated in an X-gluc solution for 15 h at 37°C. Arrows indicate the region of SE application. B, buffer. (F) Induction of Gα mRNA by SE treatment (200 μg/ml) on the leaf. Results of buffer-treated control plants are also shown.

To analyze whether the observed Gα mRNA induction was localized to the infected region, we infected leaves of transgenic rice plants expressing the Gα promoter∷gus fusion gene. When leaves of the transgenic plants were infected by an avirulent race of rice blast, β-glucuronidase (GUS) activity was detected 24 h after infection in the region where the fungus had been inoculated (Fig. 1C). When a virulent race was used for the experiments, GUS activity was not detected (Fig. 1D). These findings indicated that induction of Gα mRNA by rice blast was R-gene-dependent. In these experiments, we used second or third leaves of 2- to 3-week-old plants and detected very low background levels of GUS activity in entire leaf blades. In contrast, we detected higher levels of Gα mRNA in Northern blot analysis, because we collected all of the leaves of 1-month-old plants, including younger leaves, for RNA isolation.

SE were recently isolated from the membranes of rice blast fungus and were shown to cause accumulation of phytoalexins, cell death in excised leaf discs, increased resistance to infection by virulent pathogens, and induction of PR gene expression (28, 29). Therefore, we tested whether application of SE on the leaf could induce Gα mRNA. GUS activity was clearly detected at the site where SE were applied with pipette tips (Fig. 1E), and Gα mRNA expression was induced by SE at 16 and 18 h after treatment (Fig. 1F). The buffer treatment gave no induction of Gα mRNA (Fig. 1F).

Reduced Resistance of d1 Mutants to Rice Blast Infection.

To study the possible role of Gα in the disease resistance of rice, we used four d1 alleles with cv. Kinmaze background, because cv. Kinmaze carries the Pi-a gene, an R gene for rice blast, and has been previously used for studies in rice–blast interactions (21, 23). All four d1 mutations exhibited the dwarf phenotype (Fig. 2A). Northern blot analysis of RNAs isolated from suspension cultures showed that very low levels of d1 transcripts were present in the three alleles, 723, 1232, and 1361 (Fig. 2B). In contrast, 248 contained transcripts of two distinct sizes, which were larger and smaller than that of the WT (Fig. 2B). We sequenced cDNAs derived from the mutants and parts of the genomic sequences of the mutants. Results of the sequencing analysis clearly indicated that they all had defects in the transcripts of the Gα subunit gene of heterotrimeric G proteins (see Table 1, which is published as supporting information on the PNAS web site, www.pnas.org).

Figure 2.

Reduced resistance of d1 mutants to rice blast. (A) Plant phenotypes of WT (WT; cv. Kinmaze) and four alleles of d1 mutants, 248, 723, 1232, and 1361. (B) Northern blot analysis of the Gα gene in cell cultures of WT and four d1 mutants. WT transcripts were 1.7 kb in size, and ubiquitin mRNA was used as a loading control. The Gα probe used was a full-length cDNA of 1,400 bp, and the ubiquitin probe was 245 bp in the 3′ end of its cDNA. (C) Four levels of resistance to rice blast infection. HR, HR showing complete resistance accompanied with cell death; level 1, minimum level of hyphal development in infected cells; level 2, advanced level of infection with some developed hyphae; level 3, fully susceptible cells. Arrows indicate appressoria formed from fungal spores. (D) Quantitative analysis of resistance to rice blast in WT (cv. Kinmaze) and four d1 mutants. For each experiment, 100 infected cells were examined 72 h after infection, and 5 experiments were repeated. Results shown are the mean of the five experiments. [Bar = SEM, and Student's t test (P < 0.05) was used for statistical analysis of the results.] (E) Expression of two PR genes, PR1 and PBZ1, in leaves of rice plants infected with an avirulent race of rice blast. RNAs were sampled at indicated times after infection and used for Northern blot analysis.

To examine whether Gα is involved in the disease resistance of rice, four d1 mutants were infected by the avirulent race 031 of the rice blast fungus M. grisea, and infected cells were microscopically examined 72 h after infection. For infection, we used a leaf sheath assay (21, 26), because in this assay infection steps can be observed under a microscope and the degree of resistance can be quantitatively measured. When leaf sheaths of the WT and the mutants were challenged with the avirulent race 031 of rice blast, the resistance could be classified into four levels (Fig. 2C): HR, which showed browning cells with clear signs of cell death; level 1, which showed minimum levels of hyphal development in infected cells; level 2, which exhibited a somewhat advanced level of infection with some developed hyphae in infected cells; and level 3, which showed fully susceptible cells in which the hyphae of infecting fungi developed within entire infected cells and invaded surrounding cells. For the quantitative analysis of infection, 100 infected cells were examined for each experiment, and five independent experiments were performed for each allele (Fig. 2C). In WT Kinmaze, more than 90% of the infected cells exhibited HR, and ≈8% were level 1. In those cells exhibiting HR, granulation was clearly visible, as previously reported for rice leaf cells undergoing HR (26). In all d1 mutants analyzed, a greatly reduced number of infected cells showed HR (25–56%, depending on alleles); 30–40% were level 1; 10–30% were level 2; and fully susceptible cells of level 3 were found in three of four mutants analyzed. These results clearly indicated that d1 mutations caused reduced resistance to an avirulent race of rice blast, suggesting that Gα is involved in blast resistance mediated by the disease resistance (R) gene, Pi-a. These results are consistent with those on Gα mRNA induction by rice blast.

We also infected all four d1 mutants with the virulent race 007; however, under the conditions used, we were not able to detect clear differences in responses between the WT and four d1 mutants (data not shown). Therefore, the role of Gα in general resistance to blast infection remains to be studied.

PR Gene Expression Was Delayed in Leaves of d1 Mutants After Rice Blast Infection.

To test whether the expression of defense-related genes was altered by d1 mutations, we examined the expression of two PR genes, PR1 and PBZ1, which were previously shown to be involved in the defense of rice (23, 27, 28, 30, 31). Results of Northern blot analysis of RNAs isolated from the leaves of rice plants at various times after infection with the avirulent race 031 of rice blast indicated that the induction of both PR1 and PBZ1 gene expression was detected at 24 h after infection, which coincided with the induction of Gα mRNA (Fig. 2E). In the d1 mutants, the induction of both PR1 and PBZ1 gene expression was delayed for 24 h in three alleles, 248, 723, and 1232, and for 48 h in one allele, 1361, relative to the WT. HR caused by avirulent races of rice blast occurs at 17–31 h after infection (26), and the occurrence of HR coincided with PR gene induction in our experiments (Fig. 2E). Therefore, the 24- to 48-h delay in PR gene expression could be due to suppression of HR in the d1 mutants.

Activation of H₂O₂ Production and PBZ1 Expression by SE Was Suppressed in d1 Mutant Cell Cultures.

Because SE were shown to induce Gα mRNA expression (Fig. 1 E and F), we first measured the levels of H₂O₂ production in rice suspension cultures generated from embryo-derived calli of four d1 mutants. At 2 and 4 h after SE treatment, the increase in H₂O₂ levels slowed down considerably relative to that before the treatment in each of the four d1 cell cultures compared with those in the WT cell cultures (Fig. 3A).

Figure 3.

Effects of d1 mutations and OsRac1 on sphingolipid-induced PBZ1 gene expression and H₂O₂ production in cultured rice cells. (A) Suppression of H₂O₂ production in cell cultures of WT and four d1 mutants (248, 723, 1232, 1361) caused by SE. (Bar = SEM obtained from five measurements.) (B) Suppression of PBZ1 expression in cell cultures of WT and four d1 mutants (248, 723, 1232, 1361) caused by SE. (C) OsRac1-dependent H₂O₂ production induced by SE in untransformed WT cell cultures (WT), a transgenic rice cell culture expressing the constitutively active OsRac1 (KA), and a transgenic rice cell culture expressing the dominant-negative OsRac1 (KD). (Bar = SEM obtained from five measurements.) (D) OsRac1-dependent PBZ1 expression induced by SE in untransformed WT type cell cultures (WT), WT Kinmaze cell cultures expressing the constitutively active OsRac1 (KA), and a transgenic rice cell culture expressing the dominant-negative OsRac1 (KD).

We next examined whether elicitor-induced PR gene expression was affected in the d1 mutant cell cultures (Fig. 3B). In WT cell cultures (WT), PBZ1 expression was first detected at 4 h after SE treatment, peaking at 6 h and gradually decreasing from 6 to 12 h. In contrast, in the d1 mutant cell cultures, PBZ1 expression was not detected at any of the times examined. Even at 24 h after the elicitor treatment, no PBZ1 expression was detected in the mutants (data not shown). These results indicated that PBZ1 expression in d1 cell cultures was completely suppressed. Together with the results of H₂O₂ production, these results suggested that Gα is essential for SE signaling in rice cell cultures.

SE-Induced H₂O₂ Production and PBZ1 Expression Were Mediated by the Small GTPase OsRac1 in Rice Cell Cultures.

We have shown that a rice homolog of the small GTPase Rac, OsRac1, is an important regulator of reactive oxygen species production, cell death, and disease resistance in rice (20, 21). Therefore, to test whether OsRac1 is also involved in SE signaling, we examined H₂O₂ production and PBZ1 expression after SE treatment in rice cell cultures expressing the constitutively active OsRac1-G19V and dominant-negative OsRac1-T24N.

SE-induced H₂O₂ production was strongly enhanced in rice cell cultures expressing the constitutively active OsRac1, whereas it was almost completely suppressed in cell cultures expressing the dominant-negative OsRac1 (Fig. 3C). PBZ1 expression was similarly affected by OsRac1; its induction was greatly enhanced by the constitutively active OsRac1, whereas its expression was completely suppressed by the dominant-negative OsRac1 (Fig. 3D). The kinetics of PBZ1 induction was similar to that in the WT, except that it was constitutively activated without elicitor treatment in transgenic cells expressing the constitutively active OsRac1. In contrast, a very low level of expression was observed at all time points examined in cell cultures expressing the dominant-negative OsRac1. These results clearly indicate that the small GTPase OsRac1 is also an important intermediate in SE signaling in rice cell cultures.

Expression of the Constitutively Active OsRac1 Can Restore H₂O₂ Production and PBZ1 Expression in d1 Mutant Cell Cultures.

To determine the relative positions of Gα and OsRac1 in the SE signaling pathway of rice cells, we generated three independent d1 transgenic cell lines derived from the 1361 mutant expressing the constitutively active OsRac1 and examined H₂O₂ production and PBZ1 expression induced by SE in these transgenic cell cultures. Levels of H₂O₂ production in the three d1 cell cultures expressing the constitutively active OsRac1 were restored to a level close to that of the WT cell culture expressing the constitutively active OsRac1 (Fig. 4A).

Figure 4.

Expression of the constitutively active OsRac1 can restore H₂O₂ production and PBZ1 expression in d1 mutant cell cultures and blast resistance in transgenic d1 plants. (A) SE-induced H₂O₂ production in d1 mutant cells expressing the constitutively active OsRac1. WT, WT; 1361, d1 cell cultures; A1, A25, and A26, independent transgenic d1 (1361) cell cultures expressing the constitutively active OsRac1; KA, WT cell cultures expressing the constitutively active OsRac1. (Bar = SEM obtained from five measurements.) (B) SE-induced PBZ1 expression in d1 cell cultures expressing the constitutively active OsRac1. WT, WT; 1361, d1 cell cultures; A1, A15, and A25, independent transgenic d1 (1361) cell cultures expressing the constitutively active OsRac1; KA, WT cell cultures expressing the constitutively active OsRac1. A gene-specific OsRac1 probe of 396 bp prepared from the 3′ untranslated region was used for these experiments. Methods were the same as those used for other experiments, whose results are shown in Fig. 2 A and B. (C) Quantitative analysis of resistance to rice blast in WT, d1 mutant (1361), and two independent transgenic d1 plants (A15 and A26). [Bar = SEM, and the Student's t test (P < 0.05) was used for statistical analysis of the results.]

PBZ1 expression was constitutively induced in the absence of the elicitor (Fig. 4B). However, strong PBZ1 induction observed in the WT cells expressing the constitutively active OsRac1 was not detected in any of the three transgenic cell cultures examined. The lack of strong induction of PBZ1 expression by the constitutively active OsRac1 suggested that Gα in addition to OsRac1 was required for the strong induction of PBZ1 expression by SE.

Interestingly, induction of endogenous OsRac1 expression, which was detected by a gene-specific probe in the WT cell cultures, was suppressed in the d1 cell cultures (Fig. 4B; compare WT with 1361). This observation indicated that the OsRac1 mRNA accumulation was induced by SE and that this mRNA accumulation required Gα protein in rice cells, suggesting the presence of a regulatory link between the heterotrimeric G protein and OsRac1. Taken together, these results strongly suggest that OsRac1 acts downstream of Gα in SE signaling in rice cell cultures.

Recovery of Blast Resistance in d1 Mutants by the Constitutively Active OsRac1.

To see whether the introduction of the constitutively active OsRac1 restores disease resistance of d1 mutants, we produced transgenic d1 plants expressing the constitutively active OsRac1 and examined their resistance to rice blast. For the quantitative analysis of rice blast infection, 100 infected cells were evaluated for each experiment, and three independent experiments were performed for each plant (Fig. 4C). The results of the experiments clearly indicated that transgenic d1 plants expressing the constitutively active OsRac1 were fully resistant to an avirulent race of rice blast. These results clearly indicated that disease resistance was recovered in the d1 transgenic plants expressing the constitutively active OsRac1 and that the presence of the constitutively active OsRac1 was sufficient to make rice plants resistant to rice blast.

The phenotype of the d1 transgenic plants expressing the constitutively active OsRac1 was dwarf, and the seeds set in these transgenic plants were round, as were the original d1 mutant seeds (results not shown). These results showed that the d1 mutant phenotypes that were shown to be caused by defects in GA signaling (7–9) were not complemented by the constitutively active OsRac1. Therefore, it is likely that the defense signaling of rice is independent of GA signaling.

Discussion

Role of Gα in Disease Resistance of Rice.

Our conclusion that Gα is involved in disease resistance of rice is based on the following observations: (i) There was a clear reduction of resistance to rice blast in the d1 mutants. The reduction of HR by an avirulent race of rice blast suggests that the Gα is involved in R-gene-mediated disease resistance in rice, at least in rice–blast interactions. (ii) There was a delayed expression of PR genes in the leaf of d1 mutants after rice blast infection. (iii) There was a strong suppression of H₂O₂ production and PBZ1 expression by SE in the d1 mutant cell cultures. (iv) Gα mRNA accumulation was induced by an avirulent race of rice blast in leaf and by SE in leaf and cell cultures. (v) Gα mRNA induction by the rice blast infection was specifically localized to the infected region of the leaf.

Previous studies indicate that one possible target of the heterotrimeric G protein in defense signaling is an increase in cytosolic Ca²⁺ in the cell (17, 18, 32). The Ca²⁺ influx then leads to the activation of downstream signaling pathways in plant cells (33–35). Possible targets of the Ca²⁺ influx are Ca²⁺-dependent protein kinase (36) and calmodulins (37), both of which have been suggested to play important roles in the defense signaling of plants. Furthermore, recent findings that plant NADPH oxidase seems to be directly activated by Ca²⁺ also suggest the importance of the Ca²⁺ influx in plant defense (38). It will be of interest to examine the effects of d1 mutations on elicitor-induced Ca²⁺ signaling in rice cell cultures.

Model for the Defense Signaling of Rice Involving Two Different GTPases.

Results of this study led us to suggest a model for the defense signaling of rice involving two different GTPases, the heterotrimeric G protein and the small GTPase OsRac1 (Fig. 5). We have previously demonstrated that the constitutively active OsRac1 can induce H₂O₂ production and defense-gene expression in the absence of pathogen or elicitor, and that transgenic rice plants expressing the constitutively active OsRac1 became resistant to pathogen infection (20, 21). Restoration of H₂O₂ production and PR gene expression by the constitutively active OsRac1 in d1 mutants after SE treatment places OsRac1 downstream of Gα. The recovery of resistance to rice blast by the constitutively active OsRac1 in d1 mutants is consistent with this model. Signals from pathogens such as rice blast are likely to be received by unknown receptors, and the signals may then be transmitted to Gα. Simultaneously, Gα mRNA accumulation is likely to be induced by the signals. In the case of rice blast, transmission of signals is R gene-dependent. The signals are then transmitted to OsRac1 from Gα. The functional Gα is required for full accumulation of PBZ1 mRNA. Therefore, although they all depend on OsRac1 function, the pathways for H₂O₂ production and PBZ1 gene activation appear to be separated downstream of OsRac1.

Figure 5.

Model of defense signaling in rice. Signals from pathogens or elicitors such as SE are recognized by unknown receptors and transmitted to the heterotrimeric G protein Gα subunit. Gα mRNA accumulation is induced by signals from receptors, and Gα is required for accumulation of OsRac1 mRNA and the strong induction of PBZ1 mRNA expression. Gα transmits the signal to OsRac1, which has been shown to be a key molecular switch for the production of reactive oxygen species (ROS), defense-gene expression, and disease resistance (20, 21). Disease resistance is achieved by activation of multiple signaling pathways downstream of OsRac1. Dotted line indicates transcriptional activation.

In mammals, several cellular activities such as the regulation of the Na⁺-H⁺ antiporter, cell transformation, and formation of stress fiber are activated by Gα proteins, and their activations are mediated by the small GTPase Rho, which belongs to the same small GTPase family as Rac (39–41). In one case, it was shown that a guanine nucleotide exchange factor (GEF) for Rho, p115RhoGEF, is directly regulated by the activated Gα (42, 43) linking Gα and Rho. It is possible that the defense signaling of rice is regulated in a similar fashion.

In this study, we have shown that Gα increases OsRac1 mRNA abundance in the presence of SE (Fig. 4B). This may be one of the mechanisms to link the upstream G protein and the downstream OsRac1 in the defense signaling of rice.

Abbreviations

SE

sphingolipid elicitors

HR

hypersensitive response

GUS

β-glucuronidase