Mutation of Arabidopsis CATALASE2 results in hyponastic leaves by changes of auxin levels

Authors


Abstract

Auxin and H2O2 play vital roles in plant development and environmental responses; however, it is unclear whether and how H2O2 modulates auxin levels. Here, we investigate this question using cat2-1 mutant, which exhibits reduced catalase activity and accumulates high levels of H2O2 under photorespiratory conditions. At a light intensity of 150 μmol m−2 s−1, the mutant exhibited up-curled leaves that have increased H2O2 contents and decreased auxin levels. At low light intensities (30 μmol m−2 s−1), the leaves of the mutant were normal, but exhibited reduced H2O2 contents and elevated auxin levels. These findings suggest that H2O2 modulates auxin levels. When auxin was directly applied to cat2-1 leaves, the up-curled leaves curled downwards. In addition, transformation of cat2-1 plants with pCAT2:iaaM, which increases auxin levels, rescued the hyponastic leaf phenotype. Using qRT-PCR, we demonstrated that the transcription of auxin synthesis-related genes and of genes that regulate leaf curvature is suppressed in cat2-1. Furthermore, application of glutathione rescued the up-curled leaves of cat2-1 and increased auxin levels, but did not change H2O2 levels. Thus, the hyponastic leaves of cat2-1 reveal crosstalk between H2O2 and auxin signalling that is mediated by changes in glutathione redox status.

Introduction

H2O2, a major reactive oxygen species (ROS), functions as an important signal molecule in various aspects of plant growth, such as seed germination, root elongation (Kwak et al. 2003), auxin-regulated root gravitropism (Joo, Bae & Lee 2001) and stomatal closure (Murata et al. 2001), and in biotic/abiotic stress responses (Dat et al. 2000; Apel & Hirt 2004). Photorespiratory H2O2 arising from photorespiration in peroxisomes can be a major source of H2O2 in C3 plants (Noctor et al. 2002; Karpinski et al. 2003). Although antioxidative enzymes such as superoxide dismutase (SOD) and ascorbate peroxidases (APX) are active in leaf peroxisomes, catalase is the major scavenging enzyme in the degradation of photorespiratory H2O2 (Willekens et al. 1997; Corpas, Barroso & Del 2001; Vandenabeele et al. 2004). Whereas low H2O2 levels are eliminated by APX and other peroxidases with the aid of various reductants [e.g. ascorbate and glutathione (GSH) ], catalase, which degrades H2O2 without any reducing power and cellular reductants, is mainly active at relatively high H2O2 contents (Dat, Inze & Van Breusegem 2001; Noctor et al. 2002; Mateo et al. 2004; Gechev et al. 2006; Foyer & Noctor 2013). Three catalase genes (CAT1, CAT2 and CAT3) have been reported in Arabidopsis; however, photorespiratory CAT2 cannot be entirely replaced by either of the other two catalase isoforms or by other antioxidative enzymes (Hu et al. 2010; Mhamdi et al. 2010b). Thus, the cat2-1 mutant exhibits the reduced catalase activity and accumulates high levels of H2O2 under photorespiratory conditions (Hu et al. 2010). The knockout mutant of Arabidopsis CAT2 provides a useful tool for investigating the effects of increases in intracellular H2O2, with the advantage that the excess H2O2 signal can be switched on and off by transferring plants between photorespiratory and non-photorespiratory conditions, respectively, such as high and low light (Dat et al. 2001; Mhamdi et al. 2010b). In cat2-1 mutant, oxidized GSH accumulates to high levels following redox perturbation triggered by oxidative stress (Queval et al. 2007; Mhamdi et al. 2010a,b). In addition, H2O2-triggered oxidative stress can result in cell death, increased salicylic acid levels and decreased myo-inositol (MI) contents in long day (LD) conditions (Queval et al. 2007; Chaouch & Noctor 2010; Chaouch et al. 2010).

As a growth-promoting hormone, auxin is crucial for plant growth and environmental responses, and its homeostasis is temporally and spatially controlled (Muday & DeLong 2001; Swarup & Bennett 2003; Dharmasiri, Dharmasiri & Estelle 2005; Sieberer & Leyser 2006) by various mechanisms, including biosynthesis, degradation, transport and conjugate formation (Scheres & Xu 2006; Ludwig-Muller 2011). For example, Arabidopsis leaf development was found to be regulated by auxin biosynthetic/signalling genes. Epinastic (i.e. downward bending) leaves are a feature of auxin overproduction mutants such as rooty (King et al. 1995), sur2 (superroot 2; Delarue et al. 1998) and iaaM (tryptophan-2-monooxygenase) overexpression lines (Romano et al. 1995), whereas hyponastic (i.e. upward curling) leaves similar to those of cat2-1 mutant emerge from loss-of-function mutations of NPH4/ARF7 (NON-PHOTOTROPHIC HYPOCOTYL4; Harper et al. 2000), SHY2/IAA3 (SHORT HYPOCOTYL 2; Tian & Reed 1999), Aux/IAA12 (INDOLE-3-ACETIC ACID INDUCIBLE 12; Hamann et al. 2002) and Aux/IAA17 (INDOLE-3-ACETIC ACID INDUCIBLE 17; Leyser et al. 1996). In addition, plants overexpressing IAMT1 (IAA CARBOXYLMETHYLTRANSFERASE 1) exhibited reduced levels of auxin and hyponastic leaf phenotypes, whereas knockdown of IAMT1 expression resulted in plants with higher auxin levels and epinastic leaves (Qin et al. 2005). A study using four mutants from the YUC (YUCCA) family (yuc1, yuc2, yuc4, yuc6), which have reduced levels of auxin synthesis, revealed that leaf curvature is inversely associated with the number of active YUC genes, indicating that auxin synthesis is involved in leaf formation (Cheng, Dai & Zhao 2006). Several Trp-dependent routes have been proposed to be involved in auxin synthesis, with CYTOCHROME P450 79B2/B3 (CYP79B2/B3), YUCCA (YUC) and TRYPTOPHAN AMINOTRANSFERASE OF ARABIDOPSIS1/TRYPTOPHAN AMINOTRANSFERASE RELATED (TAA1/TAR) acting in separate pathways (Zhao 2010). However, recent data suggest that YUC and TAA1/TAR may function in a single pathway in Arabidopsis thaliana (Mashiguchi et al. 2011; Stepanova et al. 2011; Won et al. 2011).

The possibility that auxin and H2O2 interact has stimulated much interest. Previous studies reported that both gravistimulation and exogenous auxin application resulted in a transient increase in the intracellular ROS concentration (Joo et al. 2001; Tognetti, Muhlenbock & Van Breusegem 2010). In contrast, when plants were exposed to oxidative stresses, some auxin-related phenotypes, such as faster testa rupture, inhibition of root elongation, stimulation of lateral root formation and reduced cotyledon and leaf expansion, were observed (Pasternak et al. 2005). These data suggest that ROS, including H2O2, regulate auxin activity. However, further studies must be carried out to determine whether and how H2O2 modulates the auxin response.

Here, we show that H2O2 plays an important role in leaf development by modulating auxin levels. We found that the leaves of cat2-1 plants grown under photorespiratory conditions were up-curled and exhibited high levels of H2O2 and low levels of auxin. Elevating auxin levels by reducing H2O2 contents by shifting the plant to low light conditions, exogenously applying auxin or genetically manipulating iaaM expression rescued the up-curled leaf phenotype of cat2-1 mutant. Furthermore, we found that the expression of genes involved in auxin synthesis and leaf curvature regulation, including TCP3 (TEOSINTE BRANCHED 1, CYCLOIDEA AND PCF TRANSCRIPTION FACTOR 3), TCP4, TCP10 and HASTY, was down-regulated in cat2-1 mutant. Finally, we showed that the GSH redox status affects the H2O2-mediated changes in auxin level specifically in hyponastic leaves.

Materials and Methods

Plant materials, growth conditions and sampling

The mutant Arabidopsis line cat2-1 (salk_076998) in the Col-0 background was obtained from the Arabidopsis Biological Resource Centre (ABRC; Columbus, OH, USA). Plants were sown in vermiculite in pots under the following controlled environmental conditions: 23 ± 2 °C, moderate-intensity (150 μmol m−2 s−1) light or low-intensity (30 μmol m−2 s−1) light, a 16 h light/8 h dark photoperiod (LD) and irrigated with nutrient solution three times per week. For GSH treatment, 0.4 mm GSH was sprayed onto leaves and vermiculite once daily with water as a control (Bashandy et al. 2010).

Construction of the iaaM expression plasmid and transformation

A plasmid containing the iaaM gene was obtained from Harry Klee (University of Florida; Romano et al. 1995) and digested with XbaI to release the coding sequence of the iaaM gene. The promoter fragment of CAT2 was subcloned into the BamHI site of the binary vector pCAMBIA1300, while that of iaaM was subcloned into the XbaI sites. The resultant plasmid construct was introduced into Arabidopsis by Agrobacterium tumefaciens (GV3101)-mediated transformation using the floral dip method (Hu et al. 2010). After selection of transformed plants on medium containing 50 mg L−1 hygromycin, the transgenic lines were varied by PCR using a promoter-specific forward primer and a vector-specific reverse primer. Primer sequences are reported in our previous paper (Hu et al. 2010).

Detection of H2O2 in Arabidopsis leaves

Plants were grown in soil for 3 weeks and leaves were excised. H2O2 contents were determined by peroxidase (POD)-coupled assay protocols described previously (Veljovic-Jovanovic et al. 2001; Hu et al. 2010). About 0.1 g of Arabidopsis leaves was ground in liquid N2 and the powder was extracted in 2 mL of 1 m HClO4 in the presence of insoluble polyvinylpyrrolidone (5%). The homogenate was centrifuged at 12 000 g for 10 min, and the pH of the supernatant was adjusted with 5 m K2CO3 to pH 5.6 in the presence of 100 μL 0.3 m phosphate buffer, pH 5.6. The solution was centrifuged at 12 000 g for 1 min, and the sample was incubated for 10 min with one unit of ascorbate oxidase to oxidize ascorbate prior to the assay. The reaction mixture was composed of 0.1 m phosphate buffer, pH 6.5, 3.3 mm 3-(dimethylamino) benzoic acid, 0.07 mm 3-methyl-2-benzothiazoline hydrazine and 0.3 units peroxidase. The reaction was initiated by the addition of 200 μL of sample. The absorbance change at 590 nm was monitored at 25 °C.

Preparation of indole acetic acid (IAA) in lanolin

IAA was dissolved in ethanol, and the solution was combined with lanolin at a ratio of 8:92, yielding a final concentration of IAA of 25 mm L−1 (Beveridge, Symons & Turnbull 2000).

β- glucuronidase (GUS) staining

GUS staining was carried out as previously described (Hu et al. 2010). Leaves or whole plants were incubated at 37 °C overnight in staining solution {100 mm sodium phosphate buffer, pH 7.5, 10.0 mm EDTA, pH 8.0, 0.5 mm K3[Fe(CN)6], 0.5 mm K4[Fe(CN)6], 0.1% Triton X-100 and 1.0 mm 5-bromo-chloro-3-indolyl-β-D-glucuronide}. Chlorophyll was removed by rinsing the plant materials in 70% ethanol. More than 10 plants were used for GUS staining for each condition.

GUS activity assays

GUS activity assays were carried out as described previously (Ruzicka et al. 2007). Leaves (∼100 mg per measurement) were collected in Eppendorf tubes (Axygen, NY, USA) and homogenized with steel beads in GUS extraction buffer (50 mm potassium phosphate buffer, pH 7.2, 1 mm EDTA and 0.1% Triton X-100; all chemicals from Roche, Basel, Switzerland). The extract was centrifuged at 18 000 g for 15 min at 4 °C and the obtained supernatant was used for measurements. Protein concentrations were normalized with Bradford reagent (Bio-Rad, Hercules, CA, USA). The fluorescence was measured in 96-well plates in total volumes of 250 μL with 455 μm 4-methylumbelliferyl-β-D-glucuronide hydrate (Sigma-Aldrich, MO, USA) on a Fluoroskan Ascent FL fluorometer (excitation of 365 nm and emission of 455 nm, 22 °C). Each extraction was performed twice. Measurements were read each 15 min, and the standard curve was fitted. Enzyme activity was calibrated using the calibration standard 4-methylumbelliferone (Sigma-Aldrich).

Quantification of IAA

For IAA quantification, rosette leaves (0.5 g fresh weight) were excised and immediately frozen in liquid nitrogen. The extraction and purification of endogenous IAA was performed using the method described by Ding and Cao (Ding et al. 2008). The purified samples were methylated by a stream of diazomethane gas, resuspended in 100 μL of ethyl acetate and analysed by gas chromatography-selected ion monitoring mass spectrometry (GC-SIM-MS). A Shimadzu GCMS-QP2010 Plus equipped with a HP-5MS column (30 m long, 0.25 mm i.d., 0.25 μm Film; Agilent Technologies, Palo Alto, CA, USA) was used to determine the level of IAA. The chromatographic parameters were as follows: an injection temperature of 28 °C and an initial oven temperature of 70 °C for 1 min, followed by a temperature programme of 150–240 °C. Standard IAA and D2-IAA were purchased from Sigma-Aldrich. The monitored ions were m/z 130 and 132 (quinolinium ions from native IAA and the D2-IAA internal standard, respectively), and m/z 77, 189 and 191 (molecular ion and m++6).

qRT-PCR analysis

RNA was extracted using the Plant RNA Purification Reagent (Invitrogen, Burlington, ON, Canada). Poly (dT) cDNA was prepared from 1 μg of total RNA using SuperscriptIIReverse Transcriptase (Invitrogen), as recommended by Invitrogen. To quantify the expression of genes related to auxin biosynthesis in leaves of cat2-1 mutant and wild-type Arabidopsis (Columbia), qPCR analysis was performed on a Bio-Rad CFX96 apparatus with the dye SYBR Green I (Invitrogen). PCR was carried out in 96-well plates heated for 3 min at 95 °C, followed by 40 cycles of denaturation for 20 s at 95 °C, annealing for 20 s at 58 °C and extension for 30 s at 72 °C. Target quantifications were performed with specific primer pairs designed using Beacon Designer 7.0 (Premier Biosoft International, Palo Alto, CA, USA). We chose PP2AA3 (PROTEIN PHOSPHATASE 2A SUBUNIT A3, AT1G13320) as the best reference gene for our conditions using GeNorm software (Czechowski et al. 2005). All experiments were performed with three independent biological replicates and three technical repetitions. Some of the primer sequences used to amplify auxin-related genes can be found in Sun et al. (2009), and the other primer sequences are presented in Supporting Information Table S1 online.

GSH and GSSG assays

GSH and GSSG were measured using the plate reader assay described in Queval & Noctor (2007). For every assay, the leaves were washed to remove possible residual GSH. A control was performed as follows. Wild-type leaves were treated with GSH for 5 min, washed and then assayed for GSH and GSSG content. The finding that treated and untreated plants had the same amounts of GSH and GSSG indicates that there was no exogenous GSH contamination in our assays.

Results

The cat2 mutant shows a hyponastic leaf phenotype and higher contents of H2O2

To obtain the knockout mutant of the Arabidopsis CAT2 gene, T-DNA insertion lines, grown from T3 seeds obtained from the ABRC, were selected by PCR using CAT2-specific primers. The T4 seeds of a homozygous T-DNA insertion line, cat2-1, were used for all further analyses. Detailed information about cat2-1 mutants can be found in our previous paper (Hu et al. 2010). The cat2-1 plants grown under the light of 150 μmol m−2 s−1 and LD show several morphological alterations, such as reduced size, pale green colour, spontaneous lesions and up-curled leaves (Fig. 1a,b). The complemented line (Hu et al. 2010) was used to determine if the observed hyponastic leaf phenotype was due to the T-DNA insertion. Our results showed that all phenotypic changes, including up-curling of leaves, were rescued in cat2-1 CAT2::CAT2 transformants (Fig. 1c), demonstrating that the altered phenotype resulted from CAT2 knockout.

Figure 1.

Characterization of cat2-1 mutant grown under the light of 150 μmol m−2 s−1 in long day (LD) conditions.

(a–c) Top, 3-week-old wild-type, cat2-1 and cat2-1 CAT2::CAT2 plants grown at 150 μmol m−2 s−1 light in LD. Bars = 5 mm. (a–c) Bottom, transverse sections through leaves of the corresponding plants. Bars = 1 mm. (d) H2O2 content in leaves of 3-week-old wild-type, cat2-1 and cat2-1 CAT2::CAT2 plants grown under the light of 150 μmol m−2 s−1 in LD conditions. The means and LD were calculated from three independent experiments. Significant differences to cat2-1 are indicated by *P < 0.05; **P < 0.01; ***P < 0.001. FW, fresh weight; SD, standard deviation.

As the major enzyme involved in H2O2 degradation, CAT2 inactivation should change H2O2 accumulation in cat2-1 mutant. To test this expectation, H2O2 contents were analysed in cat2-1 mutant grown under the light of 150 μmol m−2 s−1 in LD conditions. We found that the content of H2O2 in the leaves of cat2-1 mutant [i.e. 301.3 ± 9.9 nmol g−1 fresh weight (FW) ] was higher than that in the leaves of the complemented (148.1 ± 11.3 nmol g−1 FW) and wild-type plants (149.2 ± 10.2 nmol g−1 FW; Fig. 1d), suggesting that the upward curling of cat2-1 leaves is correlated to a greater accumulation of H2O2 in the leaves.

We next sought to determine whether the increased levels of H2O2 underlie up-curling of leaves in the mutant, using a combination of phenotype observation and manipulation of H2O2 contents. Firstly, we examined the leaf phenotype of cat2-1 mutant grown under low light (30 μmol m−2 s−1) and LD conditions. It was previously reported that growth under low-intensity light (below 50 μmol m−2 s−1) corrected the reduced plant size and pale green colour of the mutant (Mhamdi et al. 2010b). In addition, we found that low-intensity light flattened out the up-curled leaves of the mutant (Supporting Information Fig. S1a,b). Secondly, we assayed H2O2 contents. When grown under low light and LD conditions, both cat2-1 and wild-type plants had similar H2O2 levels (Supporting Information Fig. S1d). These results suggest that upward curling is correlated with increased H2O2 accumulation in cat2-1 mutant leaves.

This notion was further re-enforced by the experiments in which cat2-1 plants were grown for 3 weeks under low light in LD and subsequently transferred to moderate-intensity light (150 μmol m−2 s−1) in LD for 4 d with the examination of both phenotype and H2O2 level. We found that the leaves of cat2-1 mutant became up-curled gradually and the leaves of wild-type plants gradually became epinastic (Fig. 2a,b & Supporting Information Fig. S2). Consistent with this, H2O2 levels in the mutant leaves were accumulated to 230.2 ± 15.7, 225.2 ± 18.7, 265.2 ± 16.7 and 270.2 ± 14.7 nmol g−1 FW, at 1, 2, 3 and 4 d, respectively, after transferring (Fig. 2c & Supporting Information Fig. S2). If the plants were transferred to the light of 30 μmol m−2 s−1 from the light of 150 μmol m−2 s−1, the leaves of cat2-1 mutant gradually became epinastic and H2O2 contents in the leaves were decreased to 190.2 ± 27.2, 189.4 ± 12.2, 190.9 ± 17.2 and 180.2 ± 12.2 nmol g−1 FW, at 1, 2, 3 and 4 d, respectively, after transferring (Supporting Information Figs S3 & S4).

Figure 2.

Characterization of plants exposed to the light of 150 μmol m−2 s−1 after growth under low light.

(a, b) Top, 3-week-old wild-type and cat2-1 plants were grown under low-intensity light (30 μmol m−2 s−1) in long day (LD) conditions and subsequently transferred to moderate-intensity light (150 μmol m−2 s−1) in LD conditions for 4 d. Bars = 5 mm. (a, b) Bottom, transverse sections through leaves from the corresponding plants. Bars = 1 mm. (d) DR5::GUS and (e) cat2-1 DR5::GUS plants, grown under the same conditions as those described in (a) and (b), were stained for GUS activity. Top bars = 5 mm and bottom bars = 1 mm. H2O2 content (c) and IAA level (f) were assayed in leaves of plants grown under the same conditions as those described in (a) and (b). Means and SD were calculated from three independent experiments. Significant differences to the wild type are indicated by *P < 0.05; **P < 0.01; ***P < 0.001. FW, fresh weight; SD, standard deviation.

Reduced auxin levels are associated with up-curled leaves in cat2 mutant

It has been documented that loss-of-function mutations of YUC (Cheng et al. 2006), NPH4/ARF7 (Harper et al. 2000), SHY2/IAA3 (Tian & Reed 1999), Aux/IAA12 (Hamann et al. 2002) and Aux/IAA17 (Leyser et al. 1996) result in the hyponastic leaf phenotype. The similarity in phenotype between these auxin-related mutants and cat2-1 implies the existence of H2O2-auxin crosstalk. To explore this possibility, DR5::GUS, an auxin-responsive reporter gene line (Ulmasov et al. 1997), was crossed with cat2-1 mutant and the resultant progeny were subjected to GUS staining to determine auxin level. Compared with the DR5::GUS line in wild-type background, cat2-1 DR5::GUS showed a remarkable reduction in GUS expression in leaves when grown under the light of 150 μmol m−2 s−1 and LD conditions (Fig. 3a). This was further evidenced by GUS enzyme activity assays that revealed that GUS expression was significantly reduced in cat2-1 mutant grown under this light condition (Fig. 3b). These results indicate that either the auxin level or response has been modulated in cat2-1 mutant. Thus, the auxin levels in the leaves of both wild-type and cat2-1 mutants were measured directly by GC–MS. IAA content in cat2-1 leaves (i.e. 9.7 ng g−1 FW) was lower than that in wild-type leaves (15.5 ng g−1 FW) when the plants were grown under the light of 150 μmol m−2 s−1 and LD conditions (Fig. 3c), suggesting that the up-curled leaf phenotype is correlated to reduce IAA contents in cat2-1 mutant.

Figure 3.

Auxin levels in wild-type and cat2-1 plants grown under different irradiance conditions.

DR5::GUS and cat2-1 DR5::GUS plants were grown under moderate (150 μmol m−2 s−1) or low (30 μmol m−2 s−1) intensity light and stained for GUS activity (a). GUS activities (b) and IAA levels (c) were also measured in leaves of 3-week-old wild-type and cat2-1 plants grown at different irradiance conditions. Top bars = 5 mm and bottom bars = 1 mm. Means and SD were calculated from three independent experiments. Significant differences to the wild type are indicated by *P < 0.05; **P < 0.01; ***P < 0.001. FW, fresh weight; SD, standard deviation.

To determine whether auxin is a downstream component of H2O2 regulation or a parallel effector of H2O2 during the development of the hyponastic leaf phenotype, we changed auxin levels in cat2-1 mutant by growing the plants under different light conditions and examined the resultant leaf phenotype. We found that growing cat2-1 mutant under low light (30 μmol m−2 s−1) and LD conditions resulted in a higher auxin level (12.2 ng g−1 FW) than did growing the mutant under the light of 150 μmol m−2 s−1 and LD conditions (9.7 ng g−1 FW; Fig. 3c). The auxin levels in the leaves of the mutant plants grown under low light and LD conditions were similar to those of the wild type grown under the same conditions (12.6 ng g−1 FW; Fig. 3c). Furthermore, the cat2-1 mutant had the same leaf phenotype as wild-type plants under low light conditions, in contrast to the hyponastic leaf phenotype of the mutant grown under the light of 150 μmol m−2 s−1 in LD (Fig. 1 & Supporting Information Fig. S1). When cat2-1 plants were transferred from low-intensity to moderate-intensity light conditions, the up-curled leaves appeared again in the mutant, and auxin levels declined, as evidenced by GUS staining analysis (Fig. 2b,e) and GC–MS assays (9.9 ng g−1 FW; Fig. 2f). Furthermore, this process could be reversed if cat2-1 plants were moved from the light of 150 μmol m−2 s−1 to low light conditions (Supporting Information Fig. S3b). The up-curled leaves were rescued in mutant plants that contained elevated levels of auxin (13.3 ng g−1 FW), as assayed by gas chromatography – mass spectrometry (GC–MS; Supporting Information Fig. S3f). Together, these results suggest that changes in auxin level of the mutant under different light intensities are correlated to the curvature of cat2-1 leaves.

Elevated auxin levels can rescue the hyponastic leaf phenotype in cat2-1 mutant

To test whether changing auxin levels affects the leaf phenotype of cat2-1 mutant, auxin (25 mm L−1 in lanolin) was directly applied to the leaf margins of both cat2-1 and wild-type plants grown under the light of 150 μmol m−2 s−1. Two days later, the leaves of cat2-1 mutant curled downward, and the leaves of wild-type plants were even more epinastic (Fig. 4c,d). Mock-treated cat2-1 leaves exposed to lanolin without auxin maintained their hyponastic leaf phenotype. Then, both cat2-1 and cat2-1 DR5::GUS plants were transformed with the pCAT2::iaaM construct to generate lines that express iaaM, which encodes a tryptophan-2-monooxygenase. This enzyme catalyses the conversion of tryptophan to indole-3-acetamide (IAM), which is then hydrolysed to release IAA (Kosuge, Heskett & Wilson 1966). The resultant cat2-1 CAT2::iaaM and cat2-1 CAT2::iaaM DR5::GUS plants were examined for both iaaM expression and leaf phenotypes. When the transgenic plants were grown under the light of 150 μmol m−2 s−1, the hyponastic leaf phenotype of cat2-1 mutant was rescued by the pCAT2::iaaM construct (Fig. 4i), and endogenous auxin levels were elevated, as evidenced by GUS staining (Supporting Information Fig. S5). The leaves of some of these lines exhibited even more severe epinasty than those of the wild type. In addition, transformation of wild-type plants with pCAT2::iaaM resulted in transgenic lines with more severe down-curling of leaves than wild-type control (Fig. 4i). All of these data demonstrate that the increased content of H2O2 resulting from mutation of CAT2 plays a role in leaf development by down-regulating the level of auxin in the mutant leaves.

Figure 4.

The rescue of up-curled leaves in cat2-1 plants by increasing auxin levels.

(a–d) Rosette leaves of wild-type and cat2-1 plants exposed to mock treatment (a, b) or exogenous IAA (c, d). GUS staining of DR5::GUS and cat2-1 DR5::GUS plants was carried out as an indicator for auxin level in leaves exposed to mock treatment (e, f) or exogenous IAA (g, h). (i) Whole-plant phenotypes and transverse sections through the leaves of 3-week-old plants of the transformant lines (CAT2::iaaM, cat2-1 CAT2::iaaM-6 and cat2-1 CAT2::iaaM-11) grown under the light of 150 μmol m−2 s−1. Top bars = 5 mm and bottom bars = 1 mm.

The expression of auxin synthesis-related genes is down-regulated by H2O2

As shown earlier, the cat2-1 plants grown under 150 μmol m−2 s−1 light and LD conditions accumulated higher levels of H2O2 (301.3 ± 9.9 nmol g−1 FW) in the leaves, which resulted in lower levels of auxin (9.7 ng g−1 FW; Figs 1d & 3c). In contrast, growth under low light conditions or transfer from moderate (150 μmol m−2 s−1) to low light conditions resulted in a lower H2O2 (125.8 ± 9.8 nmol g−1 FW and 180.2 ± 12.2 nmol g−1 FW; Supporting Information Figs S1d & S3c) and higher auxin (12.2 ng g−1 FW and 13.3 ng g−1 FW) levels (Fig. 3c & Supporting Information Fig. S3f), revealing that the auxin level in the mutant plant leaves was regulated by H2O2. So far, several pathways have been documented for IAA biosynthesis. These include one tryptophan (Trp)-independent and four Trp-dependent pathways, including the IAM, indole-3-acetaldoxime (IAOx), tryptamine and indole-3-pyruvic acid (IPA) pathways (Supporting Information Fig. S6). To test whether the expression of these auxin synthesis-related genes was modulated in cat2-1 along with changes in H2O2 contents under different light conditions, the expression of these genes was analysed by quantitative RT-PCR. Under the light of 150 μmol m−2 s−1 and LD conditions, the transcript levels of the Trp biosynthetic genes [i.e. ASA1 (ANTHRANILATE SYNTHASE ALPHA SUBUNIT 1), ASB1 (ANTHRANILATE SYNTHASE BETA SUBUNIT 1), PAT1 (PHOSPHORIBOSYLANTHRANILATE TRANSFERASE 1), PAI1 (PHOSPHORIBOSYLANTHRANILATE ISOMERASE 1), IGPS1, TSA1 (TRYPTOPHAN SYNTHASE ALPHA CHAIN) and TSB1 (TRYPTOPHAN SYNTHASE BETA-SUBUNIT 1)] were decreased in cat2-1 (Fig. 5a), and those of genes believed to be involved in Trp-dependent IAA biosynthesis [i.e. CYP79B2, YUCCA1, YUCCA6, TAA1, TAR1 (TRYPTOPHAN AMINOTRANSFERASE RELATED 1), TAR2 and NIT3 (NITRILASE 3)], were down-regulated (Fig. 5b). The expression of SUR1 (SUPERROOT 1) and SUR2, which are involved in indole glucosinolate (IG) biosynthesis, was also down-regulated in cat2-1 plants grown under these conditions (Fig. 5b). In contrast, under low-intensity light conditions (30 μmol m−2 s−1; LD), the expression of the genes listed above was similar in cat2-1 and wild-type plants (Supporting Information Fig. S7a,b). These data suggest that the lower level of auxin in the mutant was mainly due to transcriptional suppression of the auxin biosynthetic genes.

Figure 5.

qRT-PCR analysis of auxin biosynthesis-related genes.

The expression of auxin biosynthesis-related genes was analysed by qPCR assays in the leaves of 3-week-old cat2-1 and wild-type plants grown under the light of 150 μmol m−2 s−1 and long day (LD) conditions (a, b). The transcript levels of target genes were normalized to the expression of PP2AA3 (PROTEIN PHOSPHATASE 2A SUBUNIT A3, AT1G13320). All experiments were performed with three independent biological replicates and three technical repetitions. Error bars indicate the standard error of mean (SEM; N = 3). Significant differences to the wild type are indicated by *P < 0.05; **P < 0.01; ***P < 0.001. (a) Transcript levels of the Trp biosynthetic genes and (b) of genes involved in Trp-dependent IAA biosynthesis.

GSH redox state is involved in H2O2-modulated auxin level changes

We next investigated how the higher levels of H2O2 in the leaves of cat2-1 plants affect auxin homeostasis. A recent study reported links between thiol components and auxin signalling in root meristem development, based on an analysis of ntra ntrb cad2 mutants that showed that GSH regulates auxin homeostasis during meristem development (Bashandy et al. 2010). Several studies have shown that high levels of oxidized GSH (GSSG) accumulate in cat2 mutant, resulting in a very low GSH:GSSG ratio (Queval et al. 2007; Mhamdi et al. 2010a, b). These findings suggest that GSH may be involved in the catalase-mediated change in auxin in cat2 mutant. Thus, we also assayed GSH redox state by measuring GSH and GSSG in wild-type, cat2-1 and cat2-1 CAT2::iaaM plants under different light conditions (Supporting Information Fig. S8). Our results showed that the GSH reduction states (GSH/GSH + 2GSSG) were similar in the leaves of wild-type (92.3%), cat2-1 (94.4%) and cat2-1 CAT2::iaaM plants (92.7%) under the light of 30 μmol m−2 s−1. However, if the plants were grown under the light of 150 μmol m−2 s−1, the GSH reduction state was decreased to 32.9% in cat2-1 mutant and 28.6% in cat2-1 CAT2::iaaM plants as compared with 93.6% in wild-type plants. Combined with our earlier data about changes of auxin level and leaf phenotype in cat2-1 and cat2-1 CAT2::iaaM under different light conditions, these data suggest that auxin is a downstream factor of GSH redox status for hyponastic leaves because the hyponastic leaves of cat2-1 mutant with high H2O2/low GSH reduction state can be rescued in cat2-1 CAT2::iaaM plants with high auxin level under moderate-intensity light.

This hypothesis was tested by directly applying GSH (reduced GSH) to cat2-1 mutant with the examination of leaf phenotype of cat2-1 grown under the light of 150 μmol m−2 s−1. Our results indicated that GSH, but not the mock water treatment, rescued the hyponastic leaves of the mutant (Fig. 6a). Furthermore, H2O2 levels were high in both GSH and water-treated cat2-1 plants (Fig. 6b), indicating that the effect of GSH treatment on leaf curling is not due to modulation of H2O2 contents. In addition, the transcript levels of two oxidative stress markers, APX1 (ASCORBATE PEROXIDASE 1) and GSTU24 (GLUTATHIONE S-TRANSFERASE TAU 24; Davletova et al. 2005; Vanderauwera et al. 2005), were analysed by quantitative RT-PCR. The expression of these two genes was similar in GSH-treated and water-treated cat2-1 (Fig. 6c). Thus, GSH treatment did not affect the H2O2-triggered oxidative stress in cat2-1 plants grown under moderate-intensity light and LD conditions. These results demonstrate the importance of GSH in the H2O2-modulated hyponastic leaf phenotype in cat2-1. Then, the GSH redox state was assayed by measuring GSH and GSSG (Fig. 6d). The GSH reduction state was significantly higher in GSH-treated than water-treated cat2-1 plants (Fig. 6e), indicating that the change in GSH redox status following the application of exogenous GSH could restore normal curvature to the leaves of cat2-1. If the decreased auxin level in cat2-1 underlies the hyponastic leaf phenotype, the auxin level should be affected by application of exogenous GSH. Indeed, cat2-1 DR5::GUS showed stronger GUS expression in the leaves of the plants treated with exogenous GSH than in water-treated cat2-1 DR5::GUS (Fig. 7a). These changes in auxin levels were further verified by direct auxin measurement using GC–MS (Fig. 7b). The auxin level in GSH-treated cat2-1 leaves (13.9 ng g−1 FW) was greater than that in water-treated cat2-1 (9.4 ng g−1 FW). In addition, the expression of most auxin synthesis-related genes assayed was also up-regulated in cat2-1 mutant upon treatment with GSH (Fig. 7c,d). Thus, our results demonstrate that the increased levels of H2O2 in cat2-1 decreased auxin accumulation by changing the GSH redox status.

Figure 6.

GSH application rescued the hyponastic leaves of cat2 mutant without altering the level of H2O2.

(a) Top, 3-week-old cat2-1 mutants grown under moderate-intensity light (150 μmol m−2 s−1) and long day (LD) conditions were treated with water alone (+H2O) or with exogenous glutathione (+GSH; 0.4 mm). Bars = 5 mm. Bottom, transverse sections through the leaves of the corresponding plants. Bars = 1 mm. (b) H2O2 content was assayed in the leaves of plants grown under the same conditions as described in (a). Means and SD were calculated from three independent experiments. FW, fresh weight; SD, standard deviation. (c) The expression of two oxidative stress markers (APX1 and GSTU24) was analysed by qPCR assays in the leaves of 3-week-old cat2-1 mutants grown under the same conditions described in (a). The transcript levels of target genes were normalized to the expression of PP2AA3 (PROTEIN PHOSPHATASE 2A SUBUNIT A3, AT1G13320). All experiments were performed with three independent biological replicates and three technical repetitions. Error bars indicate the standard error of mean (SEM; N = 3). Significant differences to the wild type are indicated by *P < 0.05; **P < 0.01; ***P < 0.001. (d) Leaf glutathione levels were assayed in plants grown under the same conditions as described in (a). (e) The glutathione reduction state [calculated as 100 GSH / (GSH + 2GSSG)] is shown for leaves of plants grown under the same conditions as described in (a). The means and SD were calculated from three independent experiments. Significant differences to the water-treated control are indicated by *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 7.

The auxin level and the expression of auxin biosynthesis-related genes in the GSH-treated cat2-1 mutant.

(a) Three-week-old cat2-1 DR5::GUS plants grown under moderate-intensity light (150 μmol m−2 s−1) and LD conditions were treated with water alone (+H2O) or with exogenous glutathione (+GSH; 0.4 mm) and stained for GUS activity. Top bars = 5 mm and bottom bars = 1 mm. IAA level (b) was assayed in the leaves of cat2-1 mutants grown under the same conditions as described in (a). The means and SD were calculated from three independent experiments. Significant differences to the water-treated control are indicated by *P < 0.05; **P < 0.01; ***P < 0.001. FW, fresh weight. (c, d) The expression of auxin biosynthesis-related genes was analysed by qPCR assays in the leaves of 3-week-old cat2-1 plants grown under the same conditions as described in (a). The transcript levels of the target genes were normalized to the expression of PP2AA3 (PROTEIN PHOSPHATASE 2A SUBUNIT A3, AT1G13320). All experiments were performed with three independent biological replicates and three technical repetitions. Error bars indicate the standard error of mean (SEM; N = 3). Significant differences to the water-treated control are indicated by *P < 0.05; **P < 0.01; ***P < 0.001. (c) The transcript levels of the Trp biosynthetic genes and (d) the expression of genes involved in Trp-dependent IAA biosynthesis.

Several genes involved in leaf curvature are repressed in cat2-1 plants grown under moderate light intensity

Several genes, including members of the TCP family and HASTY, are known to be involved in regulating leaf curvature development (Qin et al. 2005). TCP genes are homologous to CINCINNATA, a curvature regulation gene in snapdragon (Nath et al. 2003). Our results indicated that the expression of three TCP genes (TCP3, TCP4 and TCP10) is lower in cat2-1 under moderate-intensity light (150 μmol m−2 s−1) than under low-intensity light (30 μmol m−2 s−1). The expression of other TCP gene members assayed, that is, TCP17 and TCP24, showed similar expression levels in cat2-1 under both light conditions tested. HASTY, another gene that results in a curly leaf phenotype when mutated (Telfer & Poethig 1998), was also down-regulated in cat2-1 under moderate intensity (Fig. 8).

Figure 8.

Expression analysis of leaf curvature-related genes.

Leaf expression of curvature-related genes, including TCP3, TCP4, TCP10, TCP17, TCP24 and HASTY, was assayed by qPCR in cat2-1 plants grown under moderate intensity (150 μmol m−2 s−1) or low intensity (30 μmol m−2 s−1) light and LD conditions. The transcript levels of the target genes were normalized to the expression of PP2AA3 (PROTEIN PHOSPHATASE 2A SUBUNIT A3, AT1G13320). All experiments were performed with three independent biological replicates and three technical repetitions. Error bars indicate the standard error of mean (SEM; N = 3). Significant differences to the 30 μmol m−2 s−1 control are indicated by *P < 0.05; **P < 0.01; ***P < 0.001.

Discussion

Both ROS and auxin have been reported to be involved in the plant's response to environmental cues; however, it is unclear whether ROS modulates auxin activity and, if so, how it does this. In this paper, we exploited an Arabidopsis thaliana mutant, cat2-1, that displayed dramatic hyponastic leaf phenotypes because of increased levels of H2O2. We showed that the reduction in auxin levels in cat2-1 leaves, which resulted in up-curled leaves, is due to the accumulation of H2O2, indicating that ROS plays a novel role in leaf development. Further evidence suggested that the down-regulation of auxin synthesis-related genes could underlie the lower auxin level in the mutant leaves, indicating that increased levels of H2O2 may affect auxin synthesis by suppressing the expression of the genes involved in auxin synthesis in the mutant. It also reports that the direct oxidation of IAA to OxIAA occurs in several plant species (Woodward & Bartel 2005a; Tognetti et al. 2012). Whether IAA oxidation is involved in low IAA level in cat2-1 mutant under moderate-intensity light conditions needs to be experimentally analysed. Recent studies have also reported that the application of jasmonate and cytokinin leads to an increase in auxin biosynthesis by increasing the abundance of transcripts for several putative auxin biosynthetic genes (Sun et al. 2009; Jones et al. 2010). However, it is unclear how jasmonate and cytokinin regulate the expression of auxin synthesis-related genes.

It is quite interesting that up-curled leaves are not a feature of the apx1 mutant, which also has increased levels of H2O2. One possible explanation for this is that the content of H2O2 in the apx1 mutant is not as high as that in cat2-1. It was reported that Apx1-knockout plants have ∼60% more H2O2 than wild-type plants (Pnueli et al. 2003), whereas we found that the H2O2 content of cat2 mutant grown under moderate-intensity light (150 μmol m−2 s−1) was much higher than that of wild-type plants. Thus, a relatively high level of H2O2 may be needed to regulate auxin synthesis in the pathway that governs leaf development. In fact, several studies have reported that, while low H2O2 levels are eliminated by APX and other peroxidases with the aid of various reductants, such as ascorbate and GSH, catalase is mainly active at relatively high contents of H2O2 (Dat et al. 2001; Noctor et al. 2002; Mateo et al. 2004; Gechev et al. 2006). Additionally, APX1 and CAT2 have different cellular locations. Whereas APX1 is reported to be a cytosolic protein (Pnueli et al. 2003), CAT2 is localized to peroxisomes (Kamada et al. 2003). Peroxisomes are small and single membrane-delimited organelles that execute numerous metabolic reactions and have pivotal roles in plant growth and development. Plant peroxisomes are known to participate in numerous metabolic functions, including indole-3-butyric acid (IBA) metabolism (Zolman, Yoder & Bartel 2000; Titorenko & Terlecky 2011). IBA has been suggested to act as a reservoir of auxin by being metabolized into IAA (Woodward & Bartel 2005b; Zolman et al. 2008). These differences in the localization and function of APX1 and CAT2 may be responsible for the differences between apx1 and cat2-1 leaves. In addition, several reports have assayed H2O2 contents in both cat2 RNAi lines and cat2 mutant under different growth conditions (Mhamdi et al. 2010b). An early paper with cat2 RNAi lines showed increased diaminobenzidine (DAB) staining for H2O2 when the transgenic lines were exposed to either high light or ozone (Vandenabeele et al. 2004). Then, this increase in H2O2 level based on DAB staining was evidenced in cat2 mutant either grown under the light of 60 μmol m−2 s−1 or infected with avirulent bacteria (Bueso et al. 2007; Simon et al. 2010). Our previous paper with the POD-coupled assay for H2O2 also indicated about twofold increase in extractable H2O2 in moderate light-exposed cat2-1 mutant compared with wild-type plants (Hu et al. 2010). However, both DAB staining or assays of extractable H2O2 with luminol did not detect increased H2O2 in the cat2 exposed to moderate levels of irradiance (Chaouch et al. 2010; Mhamdi et al. 2010a), whereas the changed GSH redox state was assayed in the cat2 mutant under these conditions in these papers and our results. The differences among these results for H2O2 level could be due to the use of different techniques and growth conditions (Queval et al. 2008).

It is possible that the redox perturbation in cat2-1 caused by increased availability of H2O2 could underlie the decreased level of auxin. To examine the role of GSH redox status in oxidative stress, a GSH reductase 1 (GR1) knockout, which has increased levels of GSSG, was used by Mhamdi et al. (2010a). The cat2 gr1 double mutant had very high levels of GSSG and exhibited a dwarf rosette phenotype that was more severe than that of cat2 (Mhamdi et al. 2010a). Exogenous application of GSH was previously used to partially reverse the pin-like phenotype of ntra ntrb cad2 by increasing the availability of GSH (Bashandy et al. 2010). In our study, GSH treatment rescued the up-curled leaf phenotype of cat2-1 by modulating the GSH redox status and auxin level, but without changing the H2O2 content, indicating that the GSH redox status plays an essential role in changing the levels of auxin that is mediated by high levels of H2O2 in cat2 mutant. Recently, it has been reported that some auxin response genes and auxin-related genes are repressed in cat2-1 under different growth conditions (Bueso et al. 2007; Mhamdi et al. 2010a; Queval et al. 2012), consistent with our finding that the auxin level was decreased in cat2-1. Moreover, UDP-glucosyltransferase UGT74E2, which is involved in IBA glycosylation, is shown to be up-regulated by exogenous H2O2 application or in the cat2 mutant (Tognetti et al. 2010). Overexpression of this gene resulted in increased levels of both free IBA and IBA-Glc in the transgenic lines (Tognetti et al. 2010). Additionally, Han et al. have indicated that oxidative stress modulates other phytohormone-dependent signalling in the cat2 mutant in a GSH-dependent manner (Han et al. 2013a,b). All of these data could contribute to our understanding of how the H2O2 signal is transmitted.

In summary, our data indicate that H2O2, as one of the major ROS generated by cells, plays an important role in determining leaf morphology. H2O2 regulates the level of auxin by altering the expression of auxin synthesis-related genes through modulating the GSH redox status in cat2-1 mutant. Furthermore, the expression of genes involved in leaf curvature regulation, including TCP3, TCP4, TCP10 and HASTY, is down-regulated in cat2-1 mutant.

Acknowledgments

We thank Harry Klee (University of Florida) for kindly providing the plasmid containing iaaM gene. This work was supported by the National Natural Science Foundation of China (#90917001) and the Key Project of Chinese Ministry of Education (#311026) to YT Lu.

Ancillary