The Arabidopsis Pleiotropic Drug Resistance Transporters PEN3 and PDR12 Mediate Camalexin Secretion for Resistance to Botrytis cinerea

PEN3 and PDR12 function redundantly to mediate the secretion of camalexin for Arabidopsis resistance to the necrotrophic fungus Botrytis cinerea.

Abstract

Plant defense often depends on the synthesis and targeted delivery of antimicrobial metabolites at pathogen contact sites. The pleiotropic drug resistance (PDR) transporter PENETRATION3 (PEN3)/PDR8 in Arabidopsis (Arabidopsis thaliana) has been implicated in resistance to a variety of fungal pathogens. However, the antimicrobial metabolite(s) transported by PEN3 for extracellular defense remains unidentified. Here, we report that PEN3 functions redundantly with another PDR transporter (PDR12) to mediate the secretion of camalexin, the major phytoalexin in Arabidopsis. Consistent with this, the pen3 pdr12 double mutants exhibit dramatically enhanced susceptibility to the necrotrophic fungus Botrytis cinerea as well as severe hypersensitivity to exogenous camalexin. PEN3 and PDR12 are transcriptionally activated upon B. cinerea infection, and their expression is regulated by the mitogen-activated protein kinase 3 (MPK3) and MPK6, and their downstream WRKY33 transcription factor. Further genetic analysis indicated that PEN3 and PDR12 contribute to B. cinerea resistance through exporting not only camalexin but also other unidentified metabolite(s) derived from Trp metabolism, suggesting that PEN3 and PDR12 have multiple functions in Arabidopsis immunity via transport of distinct Trp metabolic products.

INTRODUCTION

Plant responses to an attempted pathogen infection include activation of various secondary metabolite pathways. Pathogen-triggered biosynthesis and metabolism of secondary metabolites play crucial roles in plant disease resistance (Bednarek, 2012; Piasecka et al., 2015). Many antimicrobial metabolites are secreted from the infected plant cells to prevent or restrict pathogen growth. Thus, specific transporters are often required to deliver antimicrobial metabolites to the sites of pathogen attack (Jasiński et al., 2001; Kang et al., 2011; Lu et al., 2015).

In Arabidopsis (Arabidopsis thaliana), the Trp-derived indolic metabolites, especially camalexin and indole glucosinolates (IGs), constitute one of the key components of the innate immune system (Glawischnig, 2007; Bednarek, 2012; Piasecka et al., 2015). The entry reaction into indolic metabolism in Arabidopsis is catalyzed by the two cytochrome P450 enzymes CYP79B2 and CYP79B3 that convert Trp into indole-3-acetaldoxime (IAOx; Zhao et al., 2002). From IAOx, several branches of indolic metabolism diverge, leading to the formation of camalexin, IGs, and a series of other low molecular weight indolics including the phytohormone indole-3-acetic acid (an auxin; Zhao et al., 2002; Nafisi et al., 2007; Sønderby et al., 2010).

Camalexin is the most prominent phytoalexin in Arabidopsis and accumulates to high levels in response to a variety of pathogens (Tsuji et al., 1992; Thomma et al., 1999; Bednarek et al., 2009; Schlaeppi et al., 2010; Stotz et al., 2011; Hiruma et al., 2013). The cytochrome P450 enzymes CYP71A13 and CYP71B15 (PHYTOALEXIN DEFICIENT3 [PAD3]) catalyze the rate-limiting steps of camalexin synthesis from IAOx (Zhou et al., 1999; Schuhegger et al., 2006; Nafisi et al., 2007; Böttcher et al., 2009). The pad3 mutant is defective in camalexin biosynthesis and exhibits enhanced susceptibility to a number of fungal and oomycete pathogens (Glazebrook and Ausubel, 1994; Thomma et al., 1999; Bednarek et al., 2009; Schlaeppi et al., 2010; Stotz et al., 2011; Hiruma et al., 2013), indicating the importance of camalexin in Arabidopsis resistance to these pathogens.

Camalexin biosynthetic genes, including CYP71A13 and PAD3, are expressed at very low levels in the absence of stress and are highly induced by pathogen infection to activate camalexin synthesis (Schuhegger et al., 2006; Nafisi et al., 2007). The involvement of a transcription factor, WRKY33, has been well characterized in regulation of camalexin synthesis (Qiu et al., 2008; Mao et al., 2011; Birkenbihl et al., 2012). Camalexin induction is largely blocked in wrky33 mutants after infection by the fungal pathogen Botrytis cinerea or the bacterial pathogen Pseudomonas syringae (Qiu et al., 2008; Mao et al., 2011). In vivo binding of WRKY33 to the promoters of CYP71A13 and PAD3 suggests that WRKY33 directly activates camalexin synthetic genes (Qiu et al., 2008; Mao et al., 2011; Birkenbihl et al., 2012). Previously, we showed that the pathogen-responsive mitogen-activated protein kinase 3 (MPK3) and MPK6 act upstream of WRKY33 in regulating camalexin biosynthesis (Mao et al., 2011). WRKY33 was demonstrated to be a substrate of MPK3/MPK6. Phosphorylation of WRKY33 by MPK3/MPK6 enhances its activity in promoting the expression of camalexin biosynthetic genes, which drives the metabolic flow to camalexin synthesis (Mao et al., 2011).

Another branch of Trp-derived indolic metabolism results in the formation of IGs, which also play significant roles in Arabidopsis immunity (Bednarek et al., 2009; Clay et al., 2009; Schlaeppi et al., 2010; Hiruma et al., 2013; Piasecka et al., 2015). The function of glucosinolates in plant immunity requires their activation by specialized thioglucosidases, also called myrosinases (Bednarek, 2012; Piasecka et al., 2015). Myrosinases are often stored in specialized plant cells, which can be released upon tissue damage and then convert glucosinolates into a suite of toxic compounds, such as the isothiocyanates (Bednarek, 2012; Piasecka et al., 2015). Interestingly, infection of Arabidopsis by fungal or bacterial pathogens has been shown to trigger tissue damage–independent metabolism of IGs, which is initiated by the peroxisome-associated myrosinase PENETRATION2 (PEN2; Bednarek et al., 2009; Clay et al., 2009). Genetic evidence suggests that the antimicrobial end products of the PEN2 myrosinase-mediated IG metabolism are exported to the apoplast at pathogen contact sites by the pleiotropic drug resistance (PDR) transporter PEN3/PDR8 (Stein et al., 2006; Bednarek et al., 2009; Lu et al., 2015). The PEN2/PEN3-dependent extracellular defense contributes to Arabidopsis resistance against a variety of fungal and oomycete pathogens (Lipka et al., 2005; Stein et al., 2006; Bednarek et al., 2009; Schlaeppi et al., 2010; Stotz et al., 2011; Hiruma et al., 2013). However, the identity of the defense-relevant PEN3 substrates remains unknown.

PDR transporters belong to the full-size ATP binding cassette (ABC) transporter protein subfamily G (ABCG) and are found exclusively in plants and fungi (Kang et al., 2011). The expression of plant PDR genes is often stimulated by microbial infection and defense hormones, such as salicylic acid and jasmonic acid (Campbell et al., 2003; Stukkens et al., 2005; Kang et al., 2011). In addition to PEN3/PDR8, several other PDR family transporters also have been implicated in plant immunity (Jasiński et al., 2001; Campbell et al., 2003; Stukkens et al., 2005; Krattinger et al., 2009; Kang et al., 2011). The wheat (Triticum aestivum) PDR transporter Lr34 confers durable resistance of wheat to leaf rust and powdery mildew pathogens (Krattinger et al., 2009), but the substrate transported by Lr34 is still unknown. The PDR1/ABC1 from Nicotiana plumbaginifolia contributes to the transport of sclareol, an antimicrobial diterpenoid compound secreted by Nicotiana species in response to pathogen attack (Jasiński et al., 2001; Stukkens et al., 2005). The Arabidopsis PDR12/ABCG40 encodes a putative NpPDR1 homolog, whose expression is regulated by plant defense signaling (Campbell et al., 2003). Loss of PDR12 in Arabidopsis leads to increased sensitivity to exogenous sclareol (Campbell et al., 2003), but sclareol is not synthesized in Arabidopsis. Thus, the natural defense-relevant substrate of PDR12 in Arabidopsis also remains elusive.

In this study, we showed that PEN3 and PDR12 function redundantly to mediate the secretion of camalexin. Expression of both PEN3 and PDR12 is highly induced by B. cinerea infection. PEN3 and PDR12 redundantly contribute to Arabidopsis resistance against B. cinerea. In response to B. cinerea infection, the pen3 pdr12 double mutants hyperaccumulate camalexin but secret much less camalexin out of plants. When treated with exogenous camalexin, pen3 pdr12 mutants exhibit hypersensitivity due to defective export of camalexin. These results indicate that PEN3 and PDR12 are the major transporters that mediate camalexin secretion in Arabidopsis. We also found that the B. cinerea–induced expression of PEN3 and PDR12 is regulated by MPK3, MPK6, and their downstream WRKY33 transcription factor, suggesting the coordinated regulation of camalexin synthesis and transport by the MPK3/MPK6-WRKY33 signaling module. Further genetic analysis indicated that PEN3 and PDR12 contribute to B. cinerea resistance through transporting not only camalexin but also other unidentified Trp metabolic product(s), suggesting that PEN3 and PDR12 have multiple functions in Arabidopsis immunity via transport of distinct Trp-derived metabolites.

RESULTS

Induction of PDR Gene Expression in Arabidopsis upon B. cinerea Infection

PDR transporters belong to the ABCG transporter subfamily (Verrier et al., 2008; Kang et al., 2011). There are 15 PDR transporters (PDR1 to PDR15) in total in the Arabidopsis genome, and they are designated accordingly as ABCG29 to ABCG43 (van den Brûle and Smart, 2002; Verrier et al., 2008). To identify PDR transporters involved in plant defense, we profiled the expression of all 15 PDR genes in Arabidopsis infected by the B. cinerea strain T4. As shown in Figure 1A, the expression of PDR6, PEN3/PDR8, PDR9, and PDR12 was significantly induced after inoculation with B. cinerea, whereas other PDR genes were not transcriptionally activated. Among these four pathogen-responsive PDR genes, the transcript levels of PEN3 and PDR12 after induction were much higher than those of PDR6 and PDR9 (Figure 1A). The expression of PDR12 increased more than 10,000-fold relative to its low basal expression level (Supplemental Figure 1B). The induction fold of PEN3 expression was relatively low (approximately eightfold; Supplemental Figure 1A), but its basal level was much higher (Figure 1A).

Figure 1.

Induction of PDR Gene Expression in B. cinerea–Infected Arabidopsis.

(A) Twelve-day-old Arabidopsis seedlings were inoculated with B. cinerea spores in half-strength MS liquid media. Expression of 15 PDR genes at indicated time points post inoculation was determined by qPCR and calculated as percentages of the EF1α transcript, which was used as a reference. Asterisks above the columns indicate the data that are statistically different from their respective controls without B. cinerea inoculation (P < 0.05), as determined by Student’s t test. Error bars indicate sd (n = 3). ND, not detectable.

(B) to (E) Four-week-old leaves of soil-grown P_PDR6:GUS (B), P_PEN3:GUS (C), P_PDR9:GUS (D) and P_PDR12:GUS (E) transgenic plants were inoculated with droplets of B. cinerea spore suspension. The infected transgenic leaves were stained for GUS activity at 48 h post inoculation. At least 20 individual lines for each transgenic construct were examined with similar results. The dotted circles on leaves indicate B. cinerea infection areas. Bars = 2 mm.

To confirm the pathogen-induced PDR gene expression, we performed a β-glucuronidase (GUS) reporter-aided analysis of the PDR promoter activities induced by B. cinerea infection. Histochemical staining of GUS activity in the PEN3 promoter (P_PEN3):GUS and P_PDR12:GUS transgenic lines revealed the B. cinerea–induced expression of PEN3 and PDR12 around the fungal infection sites in Arabidopsis leaves (Figures 1C and 1E). However, GUS reporter gene activation was not observed in the P_PDR6:GUS or P_PDR9:GUS lines (>20 lines for each construct) after inoculation with B. cinerea (Figures 1B and 1D), probably because the induction levels of PDR6 and PDR9 were low (Figure 1A). Without pathogen infection, both P_PEN3:GUS and P_PDR9:GUS were preferentially expressed in the roots of seedlings (Supplemental Figure 2), which is related to the involvement of PEN3 and PDR9 in regulating auxin homeostasis in roots (Strader and Bartel, 2009; Ruzicka et al., 2010). By contrast, P_PDR6:GUS showed defined expression in trichomes (Supplemental Figure 2), whereas P_PDR12:GUS was specifically expressed in the junction of the hypocotyl and cotyledons (Supplemental Figure 2). The functions of PDR6 and PDR12 relevant to their tissue-specific expression remain to be studied.

Redundant Contribution of PEN3 and PDR12 in Arabidopsis Resistance to B. cinerea

The high-level induction of PEN3 and PDR12 expression in response to B. cinerea infection prompted us to assess the contribution of these two PDR transporters in Arabidopsis resistance to B. cinerea. To quantify plant disease resistance, we inoculated the fully developed leaves of the wild-type plants and pdr knockout mutants (Supplemental Figure 3) with droplets of B. cinerea spore suspension and recorded the lesion size at 60 h post inoculation. As shown in Figures 2A and 2B, lesions in the pen3, but not pdr12, mutant alleles were significantly larger than that in the wild-type control plants, indicating that loss of PEN3, but not PDR12, enhanced Arabidopsis susceptibility to B. cinerea.

Figure 2.

PEN3 and PDR12 Redundantly Contribute to Arabidopsis Resistance against B. cinerea.

(A) to (C) Fully developed leaves of the 4-week-old wild-type (WT) plants and pdr mutants were inoculated with droplets of B. cinerea spore suspension. Leaf images were taken at 60 h post inoculation (A). Meanwhile, lesion size was quantified (B), and the growth of B. cinerea was determined by quantifying transcript levels of the B. cinerea housekeeping gene BcTUBb relative to the Arabidopsis EF1α in the infected leaves (C). Error bars in (B) and (C) indicate sd (for [B], n = 20; for [C], n = 3). Different letters above the columns in (B) and (C) indicate significant differences (P < 0.05), as determined by one-way ANOVA. Bar = 5 mm.

To reveal the potential functional redundancy, we further generated two different pen3 pdr12 double knockout mutants (pen3-3 pdr12-2 and pen3-4 pdr12-3) and quantified their disease resistance. Notably, lesions in the two pen3 pdr12 double mutants were ∼2.5-fold larger than that in pen3 single mutants (Figures 2A and 2B), indicating that additional mutation of PDR12 in the pen3 mutant background resulted in significantly enhanced susceptibility to B. cinerea. These results suggested that PEN3 and PDR12 transporters redundantly contribute to Arabidopsis resistance to B. cinerea. In addition, we quantified the B. cinerea growth by monitoring the transcript level of a B. cinerea housekeeping gene encoding β-tubulin (BcTUBb) in the infected Arabidopsis leaf tissues. Consistent with the plant disease symptom, the pen3-3, but not pdr12-2, single mutant accumulated a higher level of BcTUBb transcript compared with the wild-type plants, while the accumulation of BcTUBb transcript in the pen3-3 pdr12-2 double mutant was further increased to ∼3-fold of that in pen3-3 (Figure 2C). These data confirmed the hypersusceptibility of pen3-3 pdr12-2 to B. cinerea and the redundant contribution of PEN3 and PDR12 transporters in Arabidopsis resistance to B. cinerea. PEN3 likely contributes more than PDR12 in B. cinerea resistance since the pen3, but not pdr12, single mutants exhibited enhanced susceptibility to B. cinerea (Figure 2).

PEN3 and PDR12 Function Redundantly in Mediating B. cinerea–Induced Camalexin Secretion

The redundant contribution of PEN3 and PDR12 in disease resistance suggests that they may function in transport of the same or different defense-relevant metabolite(s). PEN3 has been implicated in exporting antimicrobial metabolites derived from the PEN2-mediated IG metabolism (Stein et al., 2006; Bednarek et al., 2009; Lu et al., 2015). However, the identity of the IG-derived PEN3 substrates remains elusive. Recently, we identified the thiocyanate ion as an extracellular derivative of the PEN3 substrate produced from the PEN2-dependent IG metabolic pathway (Xu et al., 2016).

To determine whether PEN3 and PDR12 redundantly contribute to B. cinerea resistance via transporting the same IG-derived metabolite, we quantified the B. cinerea–induced production of thiocyanate ion in pen3, pdr12, and pen3 pdr12 mutant plants. As shown in Figure 3A, mutation of PEN3 caused ∼65% reduction in Arabidopsis production of extracellular thiocyanate ion in response to B. cinerea infection, suggesting that PEN3 is the major transporter for its proposed substrate derived from the PEN2-dependent IG metabolism. However, PDR12 mutation did not affect B. cinerea–induced production of thiocyanate ion (Figure 3A). Additional mutation of PDR12 in the sensitized pen3 mutant background also did not further reduce the production of thiocyanate ion (Figure 3A). These results suggest that PDR12 does not function redundantly with PEN3 in transport of the PEN3 substrate produced from the PEN2-mediated IG metabolism. In addition, mutation of PEN2 did not enhance Arabidopsis susceptibility to B. cinerea (Figures 3B and 3C), suggesting that the PEN2-dependent IG metabolic products are not required for Arabidopsis resistance to B. cinerea. Together, these data indicated that the metabolite(s) transported by PEN3 and PDR12 for B. cinerea resistance is not derived from the PEN2-dependent IG metabolic pathway.

Figure 3.

The PEN2-Dependent IGs Metabolic Pathway Is Not Likely Involved in PEN3- and PDR12-Mediated Resistance to B. cinerea.

(A) PEN3, but not PDR12, contributes to Arabidopsis production of extracellular thiocyanate in response to B. cinerea infection. The 12-d-old wild-type (WT) and pdr mutant seedlings were inoculated with B. cinerea spores in liquid medium. Medium samples were collected at 18 h post inoculation, and the contents of thiocyanate ion were quantified. FW, fresh weight.

(B) and (C) Mutation of PEN2 did not enhance Arabidopsis susceptibility to B. cinerea. Mature leaves from the 4-week-old wild-type plants and the pen2-2 mutants were inoculated with droplets of B. cinerea spore suspension. At 60 h post inoculation, leaf images were taken (B) and lesion size was quantified (C). Bar = 5 mm.

Error bars in (A) and (C) indicate sd (for [A], n = 3; for [C], n = 20). Different letters above the columns in (A) indicate significant differences (P < 0.05), as determined by one-way ANOVA.

To further identify the defense-relevant substrates of PEN3 and PDR12, we profiled the metabolites secreted into the liquid media by Arabidopsis seedlings after inoculation with B. cinerea. Our HPLC analysis with fluorescence detection revealed a peak that was much less abundant in culture media of the pen3-3 pdr12-2 mutant compared with that of the wild-type plants (Figure 4A). The absence of this peak in the media of the camalexin-null mutant pad3-1 indicated that the peak represented camalexin, which was further confirmed by referring to the camalexin standard (Figure 4A).

Figure 4.

PEN3 and PDR12 Function Redundantly in Mediating B. cinerea–Induced Secretion of Camalexin.

(A) Reduced secretion of camalexin by the pen3-3 pdr12-2 double mutant upon B. cinerea infection. The 12-d-old wild-type (WT) and pdr mutant seedlings were inoculated with B. cinerea spores in liquid medium. Medium samples were collected at 18 h post inoculation, and the metabolites secreted by Arabidopsis seedlings into the media were analyzed by HPLC with fluorescence detection. The sample of the pad3-1 mutant was used as a control, and pure camalexin was used as the reference standard. Arrows indicate the peak of camalexin.

(B) Contents of camalexin secreted into the liquid media by the wild-type (WT) and pdr mutant seedlings at 18 h post inoculation with B. cinerea. FW, fresh weight.

(C) Enhanced accumulation of camalexin in pen3 and pen3 pdr12 mutant alleles after infection by B. cinerea. The wild-type and pdr mutant seedlings were collected at 18 h post inoculation with B. cinerea, and the camalexin accumulated in plant tissues was quantified. Error bars in (B) and (C) indicate sd (n = 3). Different letters above the columns in (B) and (C) indicate significant differences (P < 0.05), as determined by one-way ANOVA. FW, fresh weight.

Quantification analysis showed that the content of camalexin secreted into the media was reduced to approximately one-third in two pen3 pdr12 mutant alleles compared with the wild-type plants at 18 h post inoculation with B. cinerea (Figure 4B), suggesting that PEN3 and PDR12 are the major transporters that mediate camalexin secretion in Arabidopsis. Reduction of camalexin secretion was also observed in two pen3 single mutants, which was much less pronounced than that in the pen3 pdr12 double mutants (Figure 4B), indicating that PEN3 and PDR12 act redundantly in mediating camalexin secretion. PEN3 likely contributes more than PDR12 in B. cinerea–induced camalexin secretion since pen3, but not pdr12, single mutants exhibited defects in this process (Figure 4B).

Consistent with these inferences, camalexin accumulated to a higher level in pen3 single mutants, and the camalexin accumulation was further increased in pen3 pdr12 double mutants, to ∼2.5-fold of that in the wild-type plants (Figure 4C; Supplemental Figure 4). The reduction of camalexin secretion in pen3 and pen3 pdr12 mutants was closely correlated with their increased susceptibility to B. cinerea (Figure 2 and 4), suggesting that PEN3- and PDR12-mediated camalexin secretion plays an important role in Arabidopsis resistance to B. cinerea.

Mutation of PEN3 and PDR12 Leads to Hypersensitivity to Exogenous Camalexin

To further demonstrate the function of PEN3 and PDR12 in camalexin secretion, we tested the sensitivities of pdr mutants to exogenously applied camalexin. As shown in Figures 5A to 5C, the pen3-3 pdr12-2 double mutant seedlings exhibited dramatically enhanced sensitivity to different levels of camalexin compared with the wild-type plants. The growth of pen3-3 pdr12-2 seedlings was completely retarded in the presence of 20 µg/mL camalexin, whereas the growth of the wild-type plants was only moderately impaired. In addition, the pen3-3, but not pdr12-2, single mutant showed increased root growth sensitivity to the higher level of camalexin (Figures 5A and 5B). Apparently, the camalexin sensitivity of pen3-3 was much lower than that of pen3-3 pdr12-2. These results indicate the redundancy of PEN3 and PDR12 and the larger contribution of PEN3 in conferring Arabidopsis tolerance to exogenous camalexin, which is consistent with the unequal redundancy of these two transporters in mediating B. cinerea–induced camalexin secretion (Figure 4).

Figure 5.

Mutation of PEN3 and PDR12 Caused Hypersensitivity to Exogenous Camalexin Due to Defective Export of Camalexin.

(A) Hypersensitivity of the pen3-3 pdr12-2 double mutants to exogenous camalexin. The wild-type (WT) and pdr mutant seedlings were grown on half-strength MS media containing 0, 10, or 20 µg/mL camalexin for 12 d. Bar = 1 cm.

(B) and (C) Root length (B) and fresh weight (C) of the wild-type and pdr mutant seedlings grown on half-strength MS media containing different concentration of camalexin for 12 d.

(D) Plasma membrane localization of PEN3-GFP and PDR12-GFP in Arabidopsis transgenic plants. Confocal images were taken on abaxial cotyledon epidermis of 35S:PEN3-GFP and 35S:PDR12-GFP transgenic plants. The plasma membrane was stained with FM4-64. Bar = 50 µm.

(E) Kinetics of camalexin accumulation in the wild-type (WT) and pdr mutant seedlings after treatment with 5 µg/mL camalexin in liquid media. FW, fresh weight.

Error bars in (B), (C), and (E) indicate sd (for [B] and [C], n = 20; for [E], n = 3). Different letters marked on columns or lines in (B), (C), and (E) indicate significant differences (P < 0.05) among the data from a single treatment (see [B] and [C]) or a single time point (E), as determined by one-way ANOVA.

Arabidopsis transgenic plants expressing green fluorescent protein (GFP)–fused PEN3 or PDR12 showed localization of both PEN3-GFP and PDR12-GFP at the plasma membrane (Figure 5D). To confirm whether PEN3 and PDR12 contribute to camalexin tolerance through mediating its export from plant cells, we quantified the kinetics of camalexin accumulation in Arabidopsis seedlings after treatment with exogenous camalexin in liquid medium. As shown in Figure 5E, camalexin accumulated to much higher levels in the pen3-3 pdr12-2 double mutant compared with the wild-type plants at different time points. The pen3-3, but not pdr12-2, single mutant also showed increased camalexin accumulation, but the increased levels were less pronounced than that in pen3-3 pdr12-2 (Figure 5E). Again, these data indicated the requirement for PEN3 and PDR12 and the larger contribution of PEN3 in export of camalexin from plant cells.

Regulation of PEN3 and PDR12 Expression by MPK3/MPK6 and the Downstream WRKY33 Transcription Factor

Previously, we showed that Arabidopsis MPK3 and MPK6 positively regulate camalexin biosynthesis through phosphorylation of the WRKY33 transcription factor, which directly activates the expression of camalexin biosynthetic genes upon pathogen infection (Mao et al., 2011). To investigate whether MPK3/MPK6 are also involved in regulation of camalexin transport by controlling the expression of PEN3 and PDR12, we generated a conditional gain-of-function transgenic plant (named Est:MKK5^DD) in which a constitutively active mutant of the upstream MKK5 (MKK5^T215D/S221D, abbreviated as MKK5^DD) was expressed under the control of an estradiol (Est)-inducible promoter. Upon treatment with Est, expression of the active MKK5^DD activated the endogenous MPK3 and MPK6 in the Est:MKK5^DD plants (Figure 6C). Meanwhile, the expression of both PEN3 and PDR12 were highly induced after Est treatment (Figures 6A and 6B). To further determine whether WRKY33 is involved in regulating PEN3 and PDR12 expression induced by MPK3/MPK6 activation, we crossed the Est:MKK5^DD transgene into the wrky33-2 mutant background. As shown in Figures 6A to 6C, the expression of both PEN3 and PDR12 induced by MPK3/MPK6 activation after Est treatment was dramatically compromised in the Est:MKK5^DD wrky33-2 double mutant. These results indicate that gain-of-function activation of MPK3/MPK6 is sufficient to activate PEN3 and PDR12 expression via the downstream WRKY33 transcription factor.

Figure 6.

PEN3 and PDR12 Expression Are Regulated by MPK3/MPK6 and the Downstream WRKY33 Transcription Factor.

(A) and (B) Activation of MPK3/MPK6 in the Est:MKK5^DD plants induced PEN3 (A) and PDR12 (B) expression. Mutation of WRKY33 compromised the active MPK3/MPK6-induced expression of PEN3 (A) and PDR12 (B) in the Est:MKK5^DD wrky33-2 plants. Twelve-day-old transgenic seedlings were treated with Est (10 µM). Expression of PEN3 and PDR12 at indicated time points was determined by qPCR and calculated as percentages of the EF1α transcript.

(C) Comparable MPK3/MPK6 activation induced by Est treatment in Est:MKK5^DD and Est:MKK5^DD wrky33-2 plants. The activated MPK3 and MPK6 were detected by immunoblot analysis using anti–phospho-Erk1/2 antibodies (top). Equal loading was confirmed by Ponceau S staining (bottom). Erk1 and Erk2, extracellular signal‐regulated kinases (two human mitogen-activated protein kinases); Rubisco, ribulose-1,5-bis-phosphate carboxylase/oxygenase.

(D) and (E) Loss of MPK3 and MPK6 compromised the B. cinerea–induced activation of PEN3 (D) and PDR12 (E). The wild-type (WT) and MPK3SR seedlings were pretreated with DMSO (solvent of NA-PP1 inhibitor) or NA-PP1 (2.5 µM) for 30 min before B. cinerea inoculation. Expression of PEN3 and PDR12 at indicated time points post inoculation was determined by qPCR and calculated as percentages of the EF1α transcript.

(F) and (G) Mutation of WRKY33 compromised the B. cinerea–induced expression of PEN3 (F) and PDR12 (G). Expression of PEN3 and PDR12 in the wild-type and wrky33-2 seedlings after inoculation with B. cinerea was determined by qPCR and calculated as percentages of the EF1α transcript. Different letters in (A), (B), (D), (E), (F), and (G) indicate significant differences (P < 0.05) among the data at a single time point, as determined by Student’s t test (see [A], [B], [F], and [G]) or one-way ANOVA (see [D] and [E]).

(H) Diagrams showing the W-boxes in the promoters of PEN3 and PDR12. Arrows indicate the primers used for ChIP-qPCR analysis.

(I) and (J) WRKY33 binds to the W-box-containing regions of PEN3 and PDR12 promoters in vivo. ChIP-qPCR was performed using the 35S:4myc-WRKY33 wrky33-2 transgenic seedlings. The 4myc-WRKY33 protein-chromatin complex was immunoprecipitated using an anti-myc antibody. A control reaction was processed side by side using mouse IgG. ChIP- and input-DNA samples were quantified by qPCR using primers specific for the W-box–containing regions. The ChIP results are presented as percentages of the input DNA. Differences in DNA abundance of anti-myc and IgG antibody samples were analyzed by Student’s t test (*P < 0.05). Error bars in (A), (B), (D), (E), (F), (G), (I), and (J) indicate sd (n = 3).

MPK3 and MPK6 play redundant roles in regulating many developmental processes and immune responses (Zhang et al., 2018). The mpk3 mpk6 double mutant is embryo lethal (Wang et al., 2007). To provide loss-of-function evidence to support the role of MPK3/MPK6 in B. cinerea–induced activation of PEN3 and PDR12, we used a conditional mpk3 mpk6 double mutant (named MPK3SR; genotype mpk3 mpk6 P_MPK3:MPK3^T144A), whose embryonic lethality was rescued by the transgene of the chemical 4-amino-1-tert-butyl-3-(1′-naphthyl)pyrazolo[3,4-d]pyrimidine (NA-PP1)–sensitized version of MPK3 (MPK3^T144A; Xu et al., 2016). The kinase activity of MPK3^T144A can be specifically inhibited by NA-PP1, a derivative of the PP1 (4-amino-1-tert-butyl-3-(4-methylphenyl)pyrazolo[3,4-d]pyrimidine) kinase inhibitor with a bulky side chain, which therefore cannot enter the ATP binding pocket of a normal kinase (Bishop et al., 2000). As shown in Figures 6D and 6E, the B. cinerea–induced expression of PEN3 and PDR12 was compromised in MPK3SR seedlings by pretreatment with NA-PP1, which supports the involvement of MPK3 and MPK6 in regulating B. cinerea–induced PEN3 and PDR12 activation. By contrast, pretreatment of the wild-type control seedlings with NA-PP1 enhanced the B. cinerea–induced expression of PEN3 and PDR12 (Figures 6D and 6E), suggesting that NA-PP1 treatment has other nonspecific but positive effects on PEN3 and PDR12 expression, which further emphasizes the compromised expression of PEN3 and PDR12 by loss of function of MPK3/MPK6 in MPK3SR plants.

Similar to PEN3 and PDR12 activation induced by the gain-of-function MKK5^DD transgene, the induction of these two genes triggered by B. cinerea infection was also compromised by the mutation of WRKY33 (Figures 6F and 6G), further supporting an important role of WRKY33 in regulating the expression of these two genes. We further performed chromatin immunoprecipitation-quantitative PCR (ChIP-qPCR) analysis to determine whether PEN3 and PDR12 genes are direct targets of WRKY33 in vivo. As shown in Figures 6H to 6J, immunoprecipitation of the 4myc-tagged WRKY33 protein in Arabidopsis transgenic plants using an anti-myc antibody greatly enriched the PEN3 and PDR12 promoter regions containing the W-box, which is known as the binding site for WRKY transcription factors. By contrast, the IgG control antibody failed to enrich either gene promoter. This result demonstrates that WRKY33 directly binds to the promoters of PEN3 and PDR12 in vivo. Together, our data indicate that the B. cinerea–induced expression of PEN3 and PDR12 is directly regulated by the MPK3/MPK6-WRKY33 module, which also controls the pathogen-responsive expression of camalexin biosynthetic genes (Mao et al., 2011). In this way, upon pathogen infection, camalexin synthesis and transport can be coordinately regulated by the MPK3/MPK6-WRKY33 signaling module.

PDR6 and PDR9 Are Not Likely Involved in Camalexin Secretion

Simultaneous mutation of PEN3 and PDR12 caused more than 60% reduction in B. cinerea–induced camalexin secretion (Figure 4B), suggesting that PEN3 and PDR12 are the major transporters that mediate camalexin secretion, while other transporters may be responsible for the residual activity and play a minor role in camalexin secretion. In addition to PEN3 and PDR12, the expression of PDR6 and PDR9 was also induced by B. cinerea infection, although their induction levels were lower than that of PEN3 and PDR12 (Figure 1A). To determine whether PDR6 or PDR9 functions redundantly with PEN3 and PDR12 in camalexin secretion, we isolated the pdr6-2 and pdr9-2 knockout mutants (Supplemental Figure 3) and further crossed these mutations into the pen3-3 and pen3-3 pdr12-2 mutant backgrounds. As shown in Figure 7C, mutation of PDR6 or PDR9 did not affect Arabidopsis secretion of camalexin upon B. cinerea infection, even in the sensitized pen3-3 or pen3-3 pdr12-2 backgrounds. Consistent with this, loss of PDR6 or PDR9 in different backgrounds did not alter Arabidopsis resistance to B. cinerea (Figures 7A and 7B). These results indicate that neither PDR6 nor PDR9 is required for B. cinerea resistance and B. cinerea–induced camalexin secretion. In addition, Figure 8 shows that mutation of PDR6 or PDR9 also did not enhance Arabidopsis sensitivity to exogenous camalexin, even in the sensitized pen3-3 or pen3-3 pdr12-2 backgrounds. Taken together, these data indicate that PDR6 and PDR9 are not likely involved in camalexin secretion.

Figure 7.

PDR6 and PDR9 Are Not Required for B. cinerea Resistance and B. cinerea–Induced Camalexin Secretion.

(A) and (B) Resistance of the wild-type (WT) and pdr mutant plants to B. cinerea. Mature leaves of the 4-week-old wild-type plants and pdr mutants were inoculated with droplets of B. cinerea spore suspension. Leaf images were taken at 60 h post inoculation (A). Meanwhile, lesion size was quantified (B). Bar = 5 mm.

(C) B. cinerea–induced secretion of camalexin in the wild-type and pdr mutant plants. The 12-d-old wild-type and pdr mutant seedlings were inoculated with B. cinerea spores in liquid media. Medium samples were collected at 18 h post inoculation, and the contents of camalexin in liquid media were quantified. FW, fresh weight.

Error bars in (B) and (C) indicate sd (for [B], n = 20; for [C], n = 3). Different letters above the columns in (B) and (C) indicate significant differences (P < 0.05), as determined by one-way ANOVA.

Figure 8.

PDR6 and PDR9 Are Not Required for Arabidopsis Tolerance to Exogenous Camalexin.

(A) Sensitivities of the wild-type (WT) and pdr mutant seedlings to exogenous camalexin. Th wild-type and pdr mutant seedlings were grown on half-strength MS media containing 0, 10, or 20 µg/mL camalexin for 12 d. Bar = 1 cm.

(B) and (C) Root length (B) and fresh weight (C) of the wild-type and pdr mutant seedlings grown on half-strength MS media containing different concentration of camalexin for 12 d.

Error bars in (B) and (C) indicate sd (n = 20). Different letters marked on columns in (B) and (C) indicate significant differences (P < 0.05) among the data from a single treatment, as determined by one-way ANOVA.

Multiple Trp-Derived Metabolites Are Involved in PEN3- and PDR12-Mediated Resistance to B. cinerea

Previous studies suggest that PEN3 transports multiple metabolites for Arabidopsis disease resistance (Stein et al., 2006; Clay et al., 2009; Lu et al., 2015). Here, we identified the redundant function of PEN3 and PDR12 in camalexin secretion. To further access the contribution of camalexin in PEN3- and PDR12-mediated resistance to B. cinerea, we compared the disease resistance of pen3-3 pdr12-2 and the camalexin-deficient mutant pad3-1. As shown in Figures 9A and 9B, pad3-1 was more susceptible to B. cinerea than the wild-type plants, demonstrating the importance of camalexin in Arabidopsis resistance to B. cinerea. However, the susceptibility of pad3-1 to B. cinerea was less pronounced than that of pen3-3 pdr12-2, suggesting that PEN3 and PDR12 contribute to B. cinerea resistance via transporting not only camalexin but also other metabolites (Figure 10).

Figure 9.

Multiple Trp-Derived Metabolites Are Involved in PEN3- and PDR12-Mediated Resistance to B. cinerea.

(A) and (B) In addition to camalexin, other unknown metabolite(s) also contributes to PEN3- and PDR12-mediated resistance to B. cinerea.

(C) and (D) Additional Trp-derived metabolites are involved in PEN3- and PDR12-mediated resistance to B. cinerea.

Mature leaves of the 4-week-old wild-type (WT) and different mutant plants were inoculated with droplets of B. cinerea spore suspension. Leaf images were taken at 60 h post inoculation (see [A] and [C]), and lesion size was quantified (see [B] and [D]). Error bars in (B) and (D) indicate sd (n = 20). Different letters above the columns in (B) and (D) indicate significant differences (P < 0.05), as determined by one-way ANOVA. Bars in (A) and (C) = 5 mm.

Figure 10.

Model Depicting the PEN3- and PDR12-Mediated Fungal Resistance via Transport of Multiple Trp-Derived Metabolites.

The Trp-derived indolic metabolism in Arabidopsis is started by the conversion of Trp to IAOx, which is catalyzed by the redundant CYP79B2 and CYP79B3 enzymes. From IAOx, several branches of indolic metabolism diverge, leading to the formation of camalexin, IGs, and some unidentified antimicrobial metabolites. PAD3 catalyzes the final step of camalexin biosynthesis. The PEN2-catalyzed hydrolysis of IGs produces a suite of toxic compounds, among which the isothiocyanates are thought to be exported by PEN3 for fungal resistance. PEN3 also functions redundantly with PDR12 in export of camalexin and additional unidentified IAOx-derived antimicrobial metabolites for Arabidopsis resistance to fungal pathogens, such as B. cinerea.

To further confirm the involvement of camalexin in PEN3- and PDR12-mediated resistance to B. cinerea, we generated the pen3-3 pdr12-2 pad3-1 triple mutant and compared its disease resistance with that of pen3-3 pdr12-2. Figures 9A and 9B show that introduction of the pad3-1 mutation into pen3-3 pdr12-2 did not enhance its susceptibility to B. cinerea, which again suggests that camalexin serves as an antimicrobial substrate of PEN3 and PDR12 for Arabidopsis resistance to B. cinerea.

Several PEN3 substrates have been identified or proposed to be Trp-derived indole-type metabolites, such as camalexin, IG metabolic products, and the auxin precursor indole-3-butyric acid (Bednarek et al., 2009; Strader and Bartel, 2009; Ruzicka et al., 2010; Lu et al., 2015). Thus, it is possible that additional metabolite(s) transported by PEN3 and PDR12 for B. cinerea resistance could also be produced from the Trp metabolism. To assess the contribution of Trp-derived metabolites in Arabidopsis resistance to B. cinerea, we quantified disease resistance of the Arabidopsis double mutant cyp79B2 cyp79B3 (Zhao et al., 2002). This mutant lacks the cytochrome P450 enzymes required for the conversion of Trp to IAOx (Zhao et al., 2002) and as a result does not produce either camalexin or IGs. As shown in Figures 9C and 9D, cyp79B2 cyp79B3 exhibited a much higher susceptibility to B. cinerea than pad3-1, indicating that besides camalexin, other IAOx-derived metabolites also contribute to Arabidopsis resistance to B. cinerea (Figure 10).

The PEN2-dependent IG metabolic products have been implicated in Arabidopsis resistance to several oomycetic or fungal pathogens (Lipka et al., 2005; Bednarek et al., 2009; Schlaeppi et al., 2010; Stotz et al., 2011; Hiruma et al., 2013). However, the pen2-2 knockout mutant did not show enhanced susceptibility to B. cinerea (Figures 3B and 3C; Figures 9C and 9D). Moreover, as seen in the pad3-1 pen2-2 double mutant, loss of PEN2 in the sensitized pad3-1 mutant background also did not enhance its susceptibility to B. cinerea (Figures 9C and 9D). These results suggest that the PEN2-mediated IG metabolic products are not involved in Arabidopsis resistance to B. cinerea. To determine whether additional unknown metabolite(s) derived from IAOx is involved in PEN3- and PDR12-mediated resistance to B. cinerea, we further generated the cyp79B2 cyp79B3 pen3-3 pdr12-2 quadruple mutant and compared its disease resistance with that of cyp79B2 cyp79B3. As shown in Figures 9C and 9D and Supplemental Figure 5, simultaneous mutation of PEN3 and PDR12 in the cyp79B2 cyp79B3 mutant background did not further enhance its susceptibility to B. cinerea at different time points post inoculation. This result suggests that besides camalexin, additional unidentified metabolite(s) transported by PEN3 and PDR12 for B. cinerea resistance is also derived from IAOx (Figure 10). In addition, the susceptibility of cyp79B2 cyp79B3 to B. cinerea was higher than that of pen3-3 pdr12-2 (Figures 9C and 9D; Supplemental Figure 5), suggesting that additional transporters may act redundantly with PEN3 and PDR12 in transport of IAOx-derived metabolites for B. cinerea resistance. Another possibility is that besides the substrates of PEN3 and PDR12, there may be additional IAOx-derived metabolite(s) involved in Arabidopsis resistance to B. cinerea.

DISCUSSION

PEN3 and PDR12 Are the Major Transporters That Mediate Camalexin Secretion in Arabidopsis

An essential defense pathway contributing to broad-spectrum fungal resistance in Arabidopsis involves the synthesis and targeted delivery of the Trp-derived metabolites camalexin and IGs (Glawischnig, 2007; Bednarek, 2012; Piasecka et al., 2015). Although the synthesis and function of camalexin and IGs in Arabidopsis disease resistance are well known (Schuhegger et al., 2006; Glawischnig, 2007; Nafisi et al., 2007; Böttcher et al., 2009; Sønderby et al., 2010), how these compounds are secreted still remains elusive. In this study, we identified that PEN3 and PDR12 function redundantly to mediate camalexin secretion in Arabidopsis upon B. cinerea infection. The expression of PEN3 and PDR12 was highly induced around the B. cinerea infection sites in Arabidopsis leaves (Figures 1C and 1E). Simultaneous mutation of PEN3 and PDR12 led to more than 60% reduction in B. cinerea–induced camalexin secretion (Figure 4B), consistent with the hypersusceptibility of the pen3 pdr12 double mutants to B. cinerea (Figure 2). Moreover, pen3 pdr12 mutants exhibited hypersensitivity to exogenous camalexin due to defective export of camalexin (Figure 5). These data indicated that PEN3 and PDR12 are the major transporters that mediate camalexin secretion in Arabidopsis.

In search of other transporters involved in camalexin secretion, our genetic analyses indicate that neither PDR6 nor PDR9 is required for Arabidopsis secretion of camalexin (Figures 7and 8). However, a recent report claimed that PDR6/ABCG34 is involved in camalexin secretion (Khare et al., 2017). In this report, Khare et al. used 4-week-old Arabidopsis rosette leaves for quantification of camalexin secretion after inoculation with the necrotrophic fungus Alternaria brassicicola. They found that the pdr6-1 and pdr6-2 single mutants secreted less camalexin to the leaf surface than the wild-type plants upon A. brassicicola infection. But we did not observe any effect of PDR6 mutation on B. cinerea–induced camalexin secretion, even in the sensitized pen3-3 or pen3-3 pdr12-2 backgrounds (Figure 7), when 12-d-old seedlings were used for the secretion assay. To test whether this inconsistency was resulted from different stages of Arabidopsis being used for the analyses, we further quantified the content of camalexin secreted from mature rosette leaves of the 4-week-old wild-type and pdr mutant plants after inoculation with B. cinerea. As shown in Supplemental Figure 6 and Figure 7, mature rosette leaves secreted much more camalexin than 12-d-old seedlings upon B. cinerea infection. Consistent with the results obtained from Arabidopsis seedlings (Figure 7), we observed very significant reduction of B. cinerea–induced camalexin secretion in mature rosette leaves of pen3-3 pdr12-2, and less reduction in pen3-3, but not in pdr6-2 (Supplemental Figure 6). Further mutation of PDR6 in the sensitized pen3-3 or pen3-3 pdr12-2 backgrounds also did not affect camalexin secretion from mature rosette leaves upon B. cinerea infection (Supplemental Figure 6). Again, these results do not support the involvement of PDR6 in B. cinerea–induced camalexin secretion, but it is possible that different PDR transporters may be involved in camalexin secretion in response to different fungal pathogens. Further study is needed to clarify whether PDR6 is involved in camalexin secretion.

PEN3 and PDR12 Are Multifunctional Transporters

In Arabidopsis, the Trp metabolism produces a series of indolics including camalexin, IGs, and the auxin indole-3-acetic acid. This study, together with previous work, implicates that PEN3 transports multiple Trp-derived indole-type metabolites, including camalexin and the unidentified PEN2-dependent IG metabolic products for plant disease resistance (Stein et al., 2006; Bednarek et al., 2009; Lu et al., 2015), and the auxin precursor indole-3-butyric acid for auxin homeostasis in roots (Strader and Bartel, 2009; Ruzicka et al., 2010). In addition to these structurally related indolic metabolites, PEN3 is also involved in export of cadmium for Arabidopsis tolerance to this toxic heavy metal (Kim et al., 2007). Therefore, PEN3 can transport multiple functionally unrelated substrates for plant adaptation to diverse biotic and abiotic stresses. Similarly, in addition to camalexin reported here, PDR12 is also implicated in transport of sclareol, abscisic acid, and the heavy metal lead (Campbell et al., 2003; Lee et al., 2005; Kang et al., 2010), indicating that PDR12 is also a multifunctional transporter. PDR-type ABC transporters can have multiple substrates, a property reflected in the name of the transporter family (Kang et al., 2011). Thus, the existence of multiple substrates of PEN3 and PDR12 in Arabidopsis is not unexpected.

Coordinated Regulation of Camalexin Synthesis and Transport by MPK3/MPK6 and the Downstream WRKY33

Previously, we reported that the pathogen-responsive MPK3 and MPK6 positively regulate camalexin synthesis through phosphorylation of the WRKY33 transcription factor, which directly activates the expression of camalexin biosynthetic genes (Mao et al., 2011). In this study, we found that MPK3/MPK6 also positively regulate the expression of PEN3 and PDR12. Gain-of-function activation of MPK3/MPK6 is sufficient to activate PEN3 and PDR12 expression (Figures 6A and 6B), while loss of MPK3 and MPK6 compromises the B. cinerea–induced expression of these two genes (Figures 6D and 6E).

Furthermore, WRKY33 is required for both the active MPK3/MPK6- and the B. cinerea–induced expression of PEN3 and PDR12 (Figures 6A, 6B, 6F, and 6G). In vivo binding of WRKY33 to the promoters of PEN3 and PDR12 indicates that WRKY33 directly activates the expression of these two genes (Figures 6I and 6J). Therefore, together with camalexin biosynthetic genes, PEN3 and PDR12 expression is also directly regulated by the MPK3/MPK6-WRKY33 module, resulting in the coordinated regulation of camalexin synthesis and transport by this signaling module (Figure 11). In this way, after being synthesized upon pathogen infection, camalexin can be rapidly extruded into the apoplast where it restricts pathogen growth.

Figure 11.

Model Depicting the Coordinated Regulation of Camalexin Synthesis and Transport by MPK3/MPK6 and the Downstream WRKY33 Transcription Factor in Arabidopsis.

In response to pathogen infection, the MPK3/MPK6 cascade positively regulates camalexin synthesis through phosphorylation of the WRKY33 transcription factor, which directly activates the expression of camalexin biosynthetic genes, including CYP71A13 and PAD3. We report here that MPK3/MPK6 also promote camalexin export through activation of PEN3 and PDR12 expression via the downstream WRKY33 that directly regulates the expression of these two genes. In this way, upon pathogen infection, camalexin synthesis and transport can be coordinately regulated by the MPK3/MPK6-WRKY33 signaling module.

PEN3 and PDR12 Transport Multiple Trp-Derived Metabolites for Arabidopsis Resistance to B. cinerea

Blocking camalexin synthesis by mutation of PAD3 enhanced Arabidopsis susceptibility to B. cinerea, but PAD3 mutation in the pen3-3 pdr12-2 mutant background did not enhance its susceptibility to B. cinerea (Figures 9A and 9B), further suggesting that camalexin serves as an antimicrobial substrate of PEN3 and PDR12 for B. cinerea resistance. However, the fact that pen3-3 pdr12-2 was more susceptible than pad3-1 indicates that, in addition to camalexin, PEN3 and PDR12 also transport other metabolite(s) for B. cinerea resistance (Figure 10).

PEN3 has also been implicated in transport of the PEN2-mediated IG hydrolysis products that are involved in Arabidopsis resistance against several fungal and oomycete pathogens (Lipka et al., 2005; Stein et al., 2006; Bednarek et al., 2009; Lu et al., 2015). Unexpectedly, we found the resistance to B. cinerea was unaffected in the pen2-2 mutant (Figures 3B, 3C,, 9C, and 9D). Previously, the products of PEN2-dependent IG metabolism and camalexin were shown to function sequentially during pre- and postinvasion defenses to establish resistance against the powdery mildew fungus Erysiphe pisi and the oomycete Phytophthora brassicae (Bednarek et al., 2009; Schlaeppi et al., 2010). Combination of the pen2 and pad3 mutations significantly enhanced the susceptibility of either parental single mutant to E. pisi and P. brassicae (Bednarek et al., 2009; Schlaeppi et al., 2010). However, we also did not observe enhanced susceptibility of the pad3-1 pen2-2 double mutant to B. cinerea in comparison to pad3-1 (Figures 9C and 9D). These results suggest that the PEN2-dependent IG metabolic products are not likely to be involved in Arabidopsis resistance to B. cinerea, which may have evolved strategies to evade the PEN2-mediated immunity in Arabidopsis.

Simultaneous mutation of PEN3 and PDR12 in the cyp79B2 cyp79B3 mutant background did not enhance susceptibility to B. cinerea (Figures 9C and 9D; Supplemental Figure 5), suggesting that the additional unidentified metabolite(s) transported by PEN3 and PDR12 for B. cinerea resistance is also produced from CYP79B2- and CYP79B3-mediated Trp metabolism (Figure 10). In addition to camalexin and the PEN2-dependent IG derivatives, other unidentified Trp metabolic products were also proposed to function in Arabidopsis resistance to P. brassicae and the necrotrophic fungus Plectosphaerella cucumerina, based on the lower susceptibility to these pathogens in pen2 pad3 mutants compared with cyp79B2 cyp79B3 (Sanchez-Vallet et al., 2010; Schlaeppi et al., 2010). One candidate group of compounds comprises indole-3-carboxylic acid and its derivatives, which accumulate in response to pathogens and are synthesized via IAOx (Hagemeier et al., 2001; Bednarek et al., 2005; Böttcher et al., 2009; Sanchez-Vallet et al., 2010). Unfortunately, the possible contribution of these compounds to disease resistance is difficult to assess since mutants specifically impaired in the synthesis of indole-3-carboxylic acid have not yet been identified. Further studies are needed to identify additional metabolite(s) transported by PEN3 and PDR12 for fungal resistance.

METHODS

Plant Materials

Arabidopsis (Arabidopsis thaliana) Columbia ecotype (Col-0) was used as the wild-type control. All the mutants and transgenic plants used in this study are in Col-0 background. Growth conditions are described with the appropriate experiments. The knockout mutants pdr6-2 (SALK_036087; Khare et al., 2017), pen3-3 (SALK_110926; Stein et al., 2006), pen3-4 (SALK_000578; Stein et al., 2006), pdr9-2 (SALK_050885; Ruzicka et al., 2010), pdr12-2 (SALK_005635; Campbell et al., 2003), pen2-2 (GABI_134C04; Lipka et al., 2005), pad3-1 (Glazebrook and Ausubel, 1994), cyp79B2 cyp79B3 (Zhao et al., 2002), and wrky33-2 (GABI_324B11; Mao et al., 2011) were described previously. The pdr12-3 (SALK_148565) was obtained from the Arabidopsis Biological Resource Center. The conditional MPK3SR mutant (Xu et al., 2016) and the 35S:4myc-WRKY33 wrky33-2 transgenic plant (Mao et al., 2011) were generated previously. The double, triple, and quadruple mutants were generated by genetic cross, and homozygous lines were used for experiments.

Generation of Transgenic Plants

The PDR promoters were amplified by PCR from Arabidopsis genomic DNA and used to substitute the 35S promoter in the pBI121 vector to generate the P_PDR:GUS constructs. To make the 35S:PEN3-GFP and 35S:PDR12-GFP constructs, the PCR products of PEN3 and PDR12 cDNAs were fused to the GFP coding sequence in a modified pCAMBIA1300 vector driven by the 35S promoter. To make the Est:MKK5^DD construct, the PCR product of MKK5 cDNA was cloned into the pUC18-T vector (Takara), in which the T215D and S221D mutations were introduced into MKK5 by site-directed mutagenesis. The MKK5^DD was then subcloned into a modified pER8 vector and fused with a hemagglutinin (HA) tag to generate the Est:MKK5^DD construct. Arabidopsis transgenic plants were generated using the Agrobacterium-mediated floral-dip transformation method (Clough and Bent, 1998). For all transgenic plants, 20 to 50 T1 plants per construct were screened for transgene expression using GUS staining or immunoblotting, and T2 (for P_PDR:GUS, 35S:PEN3-GFP and 35S:PDR12-GFP) or T3 (for Est:MKK5^DD) lines with a single transgene insertion were used for phenotypic characterization. The primers used for generation of the constructs are listed in Supplemental Table 1.

Gene Expression Analysis

To avoid the effect of light/dark cycle on pathogen-induced gene expression, Arabidopsis seedlings were grown on half-strength Murashige and Skoog (MS) plates for 6 d in a growth chamber at 22°C with continuous light and then transferred to 6 mL of half-strength MS liquid medium in 20-mL gas chromatography (GC) vials (10 seedlings per vial) and cultured for another 6 d under continuous light. Thereafter, B. cinerea spores (8 × 10⁴ spores/mL) were inoculated into GC vials, and seedlings were collected at the indicated times post inoculation. Total RNA was extracted from seedling samples using TRIzol reagent (Invitrogen). After DNase treatment, 1 μg of total RNA was used for RT. Expression of PDR genes at different time points after inoculation with B. cinerea was quantified by qPCR using a real-time qPCR machine (CFX Connect, Bio-Rad). The PDR full-length transcripts in the wild-type plants and pdr mutant lines were analyzed by PCR. The transcript of elongation factor 1α (EF1α) was used as a reference. The levels of PDR expression were calculated as percentages of the EF1α transcript or fold induction relative to the basal levels before treatment. The primer pairs used for RT-qPCR and RT-PCR are listed in Supplemental Table 1.

B. cinerea Resistance Assay

Arabidopsis seeds were sown in soil and grown under a 12-h-light/12-h-dark cycle in a growth chamber at 22°C and 65% humidity for 4 weeks. To quantify disease resistance, fully developed rosette leaves were detached and inoculated with 5-μL drops of B. cinerea (strain T4) spore suspension (1.5 × 10⁵ spores/mL). The inoculated leaves were kept in Petri dishes on wet filter paper. The lesion size was measured at indicated time points after inoculation with B. cinerea.

Metabolic Analysis

To avoid the effect of light/dark cycle on pathogen-induced secondary metabolism, Arabidopsis seedlings were grown on half-strength MS plates for 12 d under continuous light and then transferred to 3 mL of half-strength MS liquid medium in GC vials (20 seedlings per vial); thereafter, they were inoculated with B. cinerea spores (8 × 10⁴ spores/mL) or treated with exogenous camalexin (5 µg/mL, Sigma-Aldrich). The samples of liquid media and seedlings were collected at indicated time points after treatment for metabolic analyses.

The liquid medium samples were diluted with methanol and then subjected to metabolic analyses directly. Seedlings were frozen in liquid nitrogen, homogenized with plastic pestles, and extracted with 80% (v/v) methanol (10 µL/mg tissue) at 60°C for 20 min. The samples of liquid media and tissue extractions were subjected to HPLC analyses on a LC-6AD system with a RF-10AXL fluorescence detector (Shimadzu). Samples were analyzed using an Inertsil ODS-SP column (250 × 4.6 mm, 5 µm, GL Sciences) with 5% (v/v) methanol as solvent A and 100% methanol as solvent B at a flow rate of 0.7 mL/min (gradient of solvent B: 60% for 2 min followed by a 14-min linear gradient to 100% and then ended after 4 min 100%). The camalexin peak was identified by referring to the standard (Sigma-Aldrich). The camalexin concentrations were calculated based on the comparison of peak areas (excitation at 310 nm, emission at 390 nm) in tested samples with those of known amounts of the standard. The contents of thiocyanate ion in culture media were quantified using the isonicotinic-barbituric acid method as described previously (Xu et al., 2016).

To analyze the B. cinerea–induced camalexin secretion in mature plants, fully developed rosette leaves from 4-week-old plants were detached and floated on 20 mL of half-strength MS liquid medium in Petri dishes (20 leaves per dish) and then inoculated with B. cinerea spores (8 × 10⁴ spores/mL). The amounts of camalexin secreted into the liquid media were quantified at 24 h post inoculation.

GUS Staining

To investigate the B. cinerea–induced expression of PDR genes, 4-week-old leaves of soil-grown P_PDR:GUS transgenic plants (≥20 lines for each construct) were detached and inoculated with 5-μL drops of B. cinerea spore suspension (1.5 × 10⁵ spores/mL). The inoculated leaves were kept in Petri dishes on wet filter paper. At 48 h post inoculation, transgenic leaves were incubated in GUS staining buffer (10 mM EDTA, 0.1% Triton X-100, 2 mM potassium ferricyanide, 2 mM potassium ferrocyanide, and 1 mg/mL X-Gluc in 50 mM sodium phosphate buffer, pH 7.0) for 2 h (P_PEN3:GUS and P_PDR12:GUS lines) or 24 h (P_PDR6:GUS and P_PDR9:GUS lines) at 37°C. To observe the tissue-specific expression pattern of PDR genes, 12-d-old P_PDR:GUS transgenic seedlings grown on half-strength MS plates were stained for GUS activity in the same way.

Camalexin Sensitivity Assay

To analyze Arabidopsis sensitivity to exogenous camalexin, the wild-type and pdr mutant seeds were sown on half-strength MS plates in the absence or presence of the indicated concentrations of camalexin and grown under a 12-h-light/12-h-dark cycle in a growth chamber at 22°C. The root length and fresh weight of seedlings were measured 12 d after planting.

ChIP-qPCR Analysis

ChIP-qPCR assay was performed as described previously (Mao et al., 2011; Xu et al., 2016). Briefly, chromatin was isolated from 1 g of 12-d-old 35S:4myc-WRKY33 wrky33-2 seedlings. Immunoprecipitation was performed by incubating the sheared chromatin samples with 2 μg of anti-myc antibody (Millipore) or mouse IgG (negative control) for 1 h at 4°C. The protein-chromatin immunocomplexes were captured using the Protein G-Dynal magnetic beads (Invitrogen). After Proteinase K digestion, the immunoprecipitated and input DNA samples were analyzed by qPCR using primers specific for the W-box–containing regions of PEN3 and PDR12 promoters (Supplemental Table 1). The ChIP-qPCR results are presented as percentages of the input DNA.

Statistical Analyses

At least three independent repetitions were performed for the experiments in this study. Results from one of the independent repeats that gave similar results are shown. Student’s t test was used to determine whether the difference between two groups of data is statistically signi?cant. Asterisks above the columns indicate differences that are statistically signi?cant (P < 0.05). When more than two groups of data at a speci?c time point or from a speci?c treatment are compared, one-way analysis of variance (ANOVA) with Tukey’s post hoc test was performed (P < 0.05). Different letters above the data points are used to indicate differences that are statistically signi?cant. See Supplemental Data Set 1 for a statistical report of t test and ANOVA results for all figures.

Accession Numbers

Sequence data from this article can be found in the Arabidopsis Genome Initiative or GenBank/EMBL databases under the following accession numbers: PDR1 (AT3G16340), PDR2 (AT4G15230), PDR3 (AT2G29940), PDR4 (AT2G26910), PDR5 (AT2G37280), PDR6 (AT2G36380), PDR7 (AT1G15210), PEN3/PDR8 (At1G59870), PDR9 (AT3G53480), PDR10 (AT3G30842), PDR11 (AT1G66950), PDR12 (AT1G15520), PDR13 (AT4G15215), PDR14 (AT4G15233), PDR15 (AT4G15236), PEN2 (At2G44490), PAD3 (AT3G26830), MKK5 (AT3G21220), MPK3 (At3G45640), MPK6 (At2G43790), WRKY33 (AT2G38470), CYP79B2 (At4G39950), CYP79B3 (At2G22330), EF1α (At5G60390).

Supplemental Data

• Supplemental Figure 1. Fold-induction of PEN3 and PDR12 transcripts in Arabidopsis upon B. cinerea infection.

• Supplemental Figure 2. Tissue-specific expression patterns of PDR genes in Arabidopsis seedlings.

• Supplemental Figure 3. Identification of Arabidopsis pdr knockout mutants.

• Supplemental Figure 4. Hyperaccumulation of camalexin in pen3-3 pdr12-2 upon B. cinerea infection.

• Supplemental Figure 5. Simultaneous mutation of PEN3 and PDR12 in cyp79B2 cyp79B3 did not enhance its susceptibility to B. cinerea.

• Supplemental Figure 6. PDR6 and PDR9 are not required for camalexin secretion in mature leaves upon B. cinerea infection.

• Supplemental Table 1. Primers used in this study.

• Supplemental Data Set 1. Statistical report of t tests and ANOVAs results for the data presented in each figure.

Dive Curated Terms

The following phenotypic, genotypic, and functional terms are of significance to the work described in this paper:

• PEN3 Gramene: AT1G59870

• PEN3 Araport: AT1G59870

• pdr12 Gramene: AT1G15520

• pdr12 Araport: AT1G15520

• pad3 Gramene: AT3G26830

• pad3 Araport: AT3G26830

• CYP79B2 Gramene: AT4G39950

• CYP79B2 Araport: AT4G39950

• CYP79B3 Gramene: AT2G22330

• CYP79B3 Araport: AT2G22330

• pen2 Gramene: AT2G44490

• pen2 Araport: AT2G44490

• WRKY33 Gramene: AT2G38470

• WRKY33 Araport: AT2G38470

• camalexin CHEBI: CHEBI:22990