Transcription factor WRKY23 assists auxin distribution ... - PNAS

Transcription factor WRKY23 assists auxin distribution patterns during Arabidopsis root development through local control on flavonol biosynthesis Wim Grunewalda,b,c,1, Ive De Smeta,b,d, Daniel R. Lewise, Christian Löfkef, Leentje Jansena,b, Geert Goeminnea,b, Robin Vanden Bosschea,b, Mansour Karimia,b, Bert De Rybela,b,2, Bartel Vanholmea,b, Thomas Teichmannf, Wout Boerjana,b, Marc C. E. Van Montagub,1, Godelieve Gheysenc, Gloria K. Mudaye, Jirí Frimla,b, and Tom Beeckmana,b,1 a Department of Plant Systems Biology, VIB, 9052 Ghent, Belgium; bDepartment of Plant Biotechnology and Bioinformatics, Ghent University, 9052 Ghent, Belgium; cDepartment of Molecular Biotechnology, Faculty of Bioscience Engineering, Ghent University, 9000 Ghent, Belgium; eDepartment of Biology, Wake Forest University, Winston-Salem, NC 27109; fAlbrecht-von-Haller-Institut für Pflanzenwissenschaften, Georg-August-Universität Göttingen, 37073 Göttingen, Germany; and dPlant and Crop Sciences Division, School of Biosciences, University of Nottingham, Loughborough LE12 5RD, United Kingdom

Contributed by Marc C. E. Van Montagu, December 22, 2011 (sent for review August 26, 2011)

Gradients of the plant hormone auxin, which depend on its active intercellular transport, are crucial for the maintenance of root meristematic activity. This directional transport is largely orchestrated by a complex interaction of specific influx and efflux carriers that mediate the auxin flow into and out of cells, respectively. Besides these transport proteins, plant-specific polyphenolic compounds known as flavonols have been shown to act as endogenous regulators of auxin transport. However, only limited information is available on how flavonol synthesis is developmentally regulated. Using reduction-of-function and overexpression approaches in parallel, we demonstrate that the WRKY23 transcription factor is needed for proper root growth and development by stimulating the local biosynthesis of flavonols. The expression of WRKY23 itself is controlled by auxin through the AUXIN RESPONSE FACTOR 7 (ARF7) and ARF19 transcriptional response pathway. Our results suggest a model in which WRKY23 is part of a transcriptional feedback loop of auxin on its own transport through local regulation of flavonol biosynthesis. flavonoid

| lateral root | WRKY

P

lant growth and development are characterized by recurrent organogenesis and continuous adaptation to environmental conditions. These intriguing features rely on the ability to establish and maintain meristematic activity. Both de novo induction and maintenance of root meristematic activity are governed by gradients of the plant hormone auxin (1–3). Although several plant tissues are able to synthesize auxin (4, 5), installation and maintenance of auxin maxima are mediated mainly by polar auxin transport (6). Besides the well-known auxin import (AUXIN RESISTANT 1/LIKE AUX1) and export (PIN-FORMED [PIN] and ABCB/P-GLYCOPROTEIN/MDR) proteins (7–10), additional regulators mediate the flow of auxin throughout the plant. For example, flavonols (plant-specific polyphenolic compounds), have been proposed to act as endogenous auxin transport regulators based on their competition with the synthetic auxin transport inhibitor 1-N-naphthylphthalamic acid (11). Although the molecular targets of flavonol regulation remain unknown, genetic and pharmacologic evidence clearly demonstrate a role for these secondary metabolites as negative regulators of auxin transport (12– 18). Flavonol biosynthesis was recently shown to be induced by auxin through a TRANSPORT INHIBITOR RESPONSE 1 (TIR1) auxin receptor-dependent pathway (16). However, our understanding of how flavonol biosynthesis is fine-tuned during development and in response to internal and environmental signals is still limited. Here, we report on the functional characterization of a member of the WRKY family, a large, plant-specific class of transcription factors that has been associated with responses to pathogen attack, mechanical stress, and senescence (19). Our results suggest that proper expression of WRKY23 is essential for normal root development, and that misexpression of WRKY23 causes defects in meristem organization by interfering with auxin distribution. Ge1554–1559 | PNAS | January 31, 2012 | vol. 109 | no. 5

netic, transcriptomic, and biochemical data suggest that WRKY23 executes its function by stimulating the biosynthesis of flavonols. Results WRKY23 Dosage Controls Maintenance of the Root Stem Cell Niche.

WRKY23 has been identified as an auxin-inducible gene involved in plant–parasitic nematode interactions (20). The same gene also emerged from microarray experiments as a potential key component of transcriptional networks during root meristem formation (21–23), impelling a thorough investigation of its role in a developmental context. None of the publicly available T-DNA insertion lines shows obvious phenotypical differences from WT seedlings, probably because of a lack of WRKY23 transcript reductions (20). To overcome this limitation for functional analyses, we performed reduction-ofexpression (RNAi) and overexpression analyses in parallel. Independent WRKY23::WRKY23RNAi lines, which displayed 80% reduction in WRKY23 transcript levels (Fig. S1A), showed variable phenotypes, which we divided into three classes: seedlings with no obvious defects (class I; 70.8% of the progeny), seedlings with a shorter and/or agravitropic root (class II; 21.5% of the progeny), and seedlings with a severely impaired development (class III; 7.7% of the progeny) (Fig. 1 A–C). Closer inspection of the class II seedlings revealed a 50% reduction in lateral root density (Fig. S1B) and defects in root tip organization (Fig. 1 D–F and Fig. S1 D–F). The root stem cell niche was less organized and contained too many cells, especially in the lateral root cap. As a result, the WRKY23:: WRKY23RNAi root tips had a swollen appearance (Fig. S1E). To ensure that these developmental defects were solely due to a reduced WRKY23 transcript abundance, we first tested the expression of several close homologs of WRKY23 in WRKY23::WRKY23RNAi seedlings using quantitative (q)RT-PCR. This revealed that none of the tested WRKY23 homologs were significantly differentially expressed (Fig. S1A). Next, to confirm the reduction-of-expression phenotypes of the WRKY23::WRKY23RNAi lines, we used the SRDX repressor domain, a short 12-amino acid motif that converts transcription factors into dominant repressors (24). However, this implies that the transcription factor acts as an activator. To investigate this, we made use of a transient expression assay in tobacco

Author contributions: W.G., I.D.S., G. Gheysen, J.F., and T.B. designed research; W.G., I.D.S., D.R.L., C.L., L.J., G. Goeminne, R.V.B., B.D.R., and B.V. performed research; M.K., T.T., W.B., and G.K.M. contributed new reagents/analytic tools; W.G., I.D.S., M.C.E.V.M., and T.B. analyzed data; and W.G., I.D.S., and T.B. wrote the paper. The authors declare no conflict of interest. 1

To whom correspondence may be addressed. E-mail: [email protected], tobee@psb. ugent.be, or [email protected].

2

Present address: Laboratory of Biochemistry, Wageningen University, 6703 HA Wageningen, The Netherlands.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. 1073/pnas.1121134109/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1121134109

appearance of starch granules, a characteristic of differentiated columella cells (28), in cells at the position of the columella initial cells (Fig. 1 K and L). The absence of these defects at younger stages hints at an initially correct specification of the quiescent center (QC) cells (marked by QC184) (Fig. S1 K–M), but argues for the necessity of a balanced WRKY23 expression level to maintain root stem cell and meristem activity postembryonically. Taken together, our results demonstrate that an altered WRKY23 expression level impairs the maintenance of stem cell identity in the primary root meristem, and that WT-levels of WRKY23 expression are required for proper root growth and development.

Fig. 1. WRKY23 controls maintenance of the root stem cell niche. (A–C) WRKY23::WRKY23RNAi plants showing a phenotypic variation from WTlooking plants (class I) (A) to mildly (class II) (B) and strongly (class III) (C) affected seedlings. (D) Schematic representation of the Arabidopsis root tip. Blue, QC; light pink, columella initials; pink, columella root cap; brown, lateral root cap. (E–G) Primary root tips of WT (E) (94.1%; n = 17), class II WRKY23::WRKY23RNAi (F) (91.9%; n = 62), and 35S::WRKY23-SRDX (G) (96.7%; n = 30) seedlings. White arrowheads indicate extra divisions, and asterisks mark the rows of columella cells. (H) Ectopic expression of WRKY23 reduces root growth (Left, WT; Right, 35S::WRKY23). (I–L) Q0680 (I and J) and QC184 (K and L) marker analysis in WT (I and K) and 35S::WRKY23 (J and L) root tips. Arrowheads point toward columella initials.

cells (25). Both C-terminal and N-terminal WRKY23-GAL4 fusions were able to activate a UAS::LUCIFERASE construct (Fig. S1C), demonstrating that WRKY23 acts as a transcriptional activator. Subsequently, transgenic plants were generated in which a WRKY23SRDX translational fusion was driven by the broadly expressed 35S promoter (26). The 35S::WRKY23-SRDX seedlings had fewer lateral roots, and their primary roots were agravitropic and exhibited a reduced growth rate (Fig. S1H). Similar to WRKY23::WRKY23RNAi, 35S::WRKY23-SRDX root tips were enlarged and showed defects in cellular patterning (Fig. 1G and Fig. S1G). Thus, two independent approaches show that WRKY23 is required for the organization of the primary root tip, lateral root development, and root gravitropic responses. To examine the effect of ectopic expression of WRKY23 on root development, we selected several lines with high WRKY23 transcript levels in which WRKY23 was driven by the global 35S promoter (35S::WRKY23) or by the RPS5A promoter, which is more restricted to sites of active cell division (27). The root length of all 35S::WRKY23 and RPS5A>>WRKY23 seedlings was reduced compared with WT (Fig. 1H and Fig. S1R). This phenotype is correlated with reduced cell division in the meristem, as shown by the B-type cyclin CycB1;1::GUS marker (Fig. S1 N and O). Moreover, in contrast to the regular and well-organized cellular pattern in WT root tips (Fig. 1 D and E), alterations in columella cell shape were observed in the 35S::WRKY23 and RPS5A>>WRKY23 root tips (Fig. 1 I and J and Fig. S1 I, J, S and T). This coincided with an altered expression domain of the Q0680 and Q1630 columella cell markers (Fig. 1 I and J and Fig. S1 P and Q), as well as with the Grunewald et al.

Root growth and meristematic activity largely depend on the correct functioning of both cell-autonomous (29–31) and non–cell-autonomous mechanisms (1, 3). Because WRKY23 is clearly involved in the maintenance of the root stem cell niche, we investigated whether the loss of stem cells in WRKY23-overexpressing roots was provoked by a local misregulation of WRKY23 expression or through a more distantly WRKY23-derived signal. Therefore WRKY23 was expressed in a cell type-specific context with the GAL4VP16-UAS transactivation system. The J2341 and Q0680 activator lines were used to drive expression in the columella initial cells and in the youngest columella layer, respectively (Fig. S2 C and D, Insets). None of these transactivation lines displayed reduced primary root growth, altered cell patterning in the root tip, or enhanced differentiation of the GAL4-expressing cells (Fig. S2 C and D) compared with the control (Fig. S2A). This is in contrast to seedlings expressing WRKY23 under the control of the 35S promoter (Fig. S1O) or the RPS5A promoter (Fig. S2B). The difference between global misregulation of WRKY23 levels and misregulation of WRKY23 levels within the columella initials suggests that WRKY23 is not a cell-autonomously acting differentiation factor. WRKY23 Focuses Auxin Response Maxima During Organogenesis and Meristem Maintenance. To examine the effect of WRKY23 mis-

regulation on auxin maxima, which are known to drive meristem organization, we investigated the auxin response maximum in roots of WRKY23 reduction-of-expression and overexpression lines using the DR5::GUS reporter (32). In WT primary root tips, the auxin maximum was confined to the QC and the central columella cells (Fig. 2A). However, in WRKY23::WRKY23RNAi and WRKY23SRDX root tips, the DR5 expression domain was expanded radially (Fig. 2 B and C). The same broadening of the auxin response maximum was observed during lateral root initiation (Fig. 2 D and E). Prior to lateral root initiation, auxin accumulates in two neighboring pericycle founder cells that subsequently divide asymmetrically (2), after which intense DR5::GUS expression can be observed in the small daughter cells (Fig. 2D) (33). However, in WRKY23-SRDX plants, DR5 expression was less confined to the founder cells and was even expanded to the cortex layer (Fig. 2E). Conversely, overexpression of WRKY23 resulted in a narrowing of the DR5-visualized auxin response domain in the primary root (Fig. 2 F and G). To investigate whether WRKY23 might be involved in the regulation of auxin transport, we performed quantitative measurements of auxin transport. These analyses revealed that the acropetal auxin transport (toward the root tip) in WRKY23:: WRKY23RNAi seedlings was higher than that in WT seedlings (Fig. 2H). Because the primary roots of the 35S and RPS5A-driven overexpression lines were too short to analyze, we constructed an inducible 35S::WRKY23-GR line that phenocopied both overexpression lines on dexamethasone (Dex) treatment (Fig. S3 A–C). Auxin transport measurements on Dex-treated 35S::WRKY23GR seedlings revealed that ectopic WRKY23 expression negatively influenced auxin transport (Fig. 2H), whereas Dex treatment of WT or DMSO treatment of the inducible lines had no effect on transport. PNAS | January 31, 2012 | vol. 109 | no. 5 | 1555

DEVELOPMENTAL BIOLOGY

WRKY23 Is Not a Cell-Autonomously Acting Differentiation Factor.

Fig. 2. WRKY23 is a negative regulator of auxin transport. (A–C) Auxin response in WT (A), WRKY23::WRKY23RNAi (B), and 35S::WRKY23-SRDX (C) root tips visualized by short staining of DR5::GUS. Arrowheads indicate QC cells. (D and E) DR5::GUS-visualized auxin response during WT (D) and 35S:: WRKY23-SRDX (E) lateral root initiation. Arrowheads denote the cell walls of asymmetrically divided pericycle cells. P, pericycle; En, endodermis; Co, cortex; Ep, epidermis. (F and G) Auxin response in WT (F) and 35S::WRKY23 (G) root tips visualized by long staining of DR5::GUS. Arrowheads indicate QC cells. (H) Auxin transport measurements showing an enhanced transport in WRKY23::WRKY23RNAi and a reduced acropetal transport in 35S::WRKY23-GR roots on Dex treatment compared with the WT and DMSO control, respectively. *P < 0.05; **P < 0.01, Student t test.

Based on the aforementioned data and the phenotypic observations of WRKY23 transgenic plants, we propose that WRKY23 is part of a transcriptional network intimately associated with the local control of auxin transport required for the maintenance of auxin maxima in the root tip and during lateral root initiation. Identification of Genes Regulated by WRKY23. On a transcriptional level, the transport of auxin can be regulated by the expression of the PIN transport regulators (6, 18). Given the possible involvement of WRKY23 in the control of auxin transport, we evaluated the expression of the PIN efflux carriers in the WRKY23 transgenic lines, but found no explicit changes (Fig. S4). This suggests that WRKY23 might control the expression of other or upstream regulators of auxin transport. To gain insight into the WRKY23-controlled transcriptional cascade, we analyzed the transcriptome of Col0 WT, WRKY23::WRKY23RNAi, and RPS5A>>WRKY23 roots using Affymetrix ATH1 arrays. After statistical analysis, the significantly differentially regulated genes were clustered according to their transcriptional behavior. A group of 86 genes was up-regulated in RPS5A>>WRKY23 and down-regulated in WRKY23::WRKY23RNAi, whereas 38 genes exhibited a converse transcriptional behavior. Retaining only those genes with more than 1.5-fold altered expression in RPS5A>>WRKY23, we reduced this list to a selection of 42 putative target genes (Table S1). Among these genes, only TRANSPARENT TESTA 7 (TT7) had been previously implicated in auxin transport regulation. TT7 encodes flavonoid 3′ hydroxylase, a flavonol biosynthetic enzyme that mediates the conversion of dihydrokaempferol to dihydroquercetin (Fig. S5), thereby contributing to the synthesis of quercetin, an endogenous negative regulator of auxin transport (16). Interestingly, TT7 was also found to be significantly up-regulated in a transcript profiling study of WRKY23overexpressing poplar (Populus sp.) plants (34), suggesting that WRKY23 regulates this branch of flavonoid metabolism in divergent lineages of higher plants. Moreover, flavonols (more precisely, quercetin derivatives) accumulate at infection sites of plant-parasitic nematodes (35), where WRKY23 is highly expressed (20), justifying further investigation of TT7 as a putative target of WRKY23. For this investigation, we first analyzed the expression of TT7 and other flavonol biosynthetic genes in WRKY23::WRKY23RNAi and WRKY23-overexpression lines. Although all four biosynthetic genes tested (TT4, TT5, TT6, and TT7) were up-regulated in WRKY23overexpressing roots (35S::WRKY23), only a strong induction of TT7 was observed in Dex-treated 35S::WRKY23-GR seedlings (Fig. 3A). Conversely, TT7 expression was significantly down-regulated in WRKY23::WRKY23RNAi lines (Fig. 3A). Next, to investigate the spatial aspect of this regulation, we compared the distribution of 1556 | www.pnas.org/cgi/doi/10.1073/pnas.1121134109

flavonols, visualized by the flavonoid-specific dye diphenylboric acid 2-aminoethyl ester (DPBA), with the expression pattern of WRKY23, detected by mRNA in situ hybridization and GUS/GFP transcriptional reporter lines. Interestingly, we detected both flavonol accumulation and WRKY23 expression in root tips, in lateral root primordia, in hydathodes, and at the hypocotyl–root transition zone (Fig. S6 A–J). Moreover, in 35S::WRKY23 seedlings, ectopic flavonols accumulated throughout the entire root (Fig. 3 B and C), suggesting that WRKY23 can stimulate the biosynthesis of flavonols through transcriptional regulation of genes encoding pathway enzymes or, alternatively, of upstream factors that control pathway enzyme gene expression. To further examine the impact of WRKY23 misregulation on flavonoid production, we analyzed root extracts of WT, WRKY23::WRKY23RNAi, and 35S::WRKY23 seedlings by liquid chromatography-mass spectrometry (LC-MS). 35S:: WRKY23 roots exhibited an enhanced accumulation of quercitrin (quercetin-3-O-rhamnoside) compared with WT (Fig. 3 D and E), whereas WRKY23::WRKY23RNAi roots were characterized by a highly reduced level of quercetin-rhamnoside-glucoside and rhamnetin-Oneohesperidoside (Fig. 3 F and G). We conclude that WRKY23 is involved in the local production of flavonol derivatives in the root tip. In particular, our results argue for a role in regulating the conversion of kaempferol to quercetin by either directly or indirectly stimulating the transcription of TT7. Auxin Regulates TT7 Expression Through a WRKY23 Transcriptional Cascade. Previously, we demonstrated that WRKY23 expression

can be induced by auxin in a SOLITARY ROOT/INDOLE-3ACETIC ACID14 (SLR/IAA14)-dependent manner (20). SLR/ IAA14 regulates auxin signaling by inhibiting the action of the ARF7 and ARF19 transcription factors (36). Typically, upon auxin perception by TIR1, Aux/IAA proteins, such as SLR/IAA14, are degraded (37). Accordingly, the auxin inducibility of WRKY23 expression was abolished in both arf7arf19 mutant seedlings as in the tir1 single mutant (Fig. 4A). Given that WRKY23 stimulates the biosynthesis of flavonols, we wondered whether flavonols (in particular, quercetin) could be involved in the WRKY23-dependent feedback mechanism on auxin transport. Therefore we analyzed the effect of auxin on TT4 and TT7 transcript levels as well as on DPBA fluorescence. Both TT4 and TT7 were significantly up-regulated by auxin upon a 6-h α-naphthaleneacetic acid (NAA) treatment (Fig. 4B). Moreover, compared with mock-treated seedlings, a much stronger DPBA fluorescence was observed in auxin-treated seedlings (Fig. 4 C and D). Although DPBA fluorescence was present in the elongation zone of untreated seedlings while WRKY23::GUS was not (Fig. 4 C and E), WRKY23 Grunewald et al.

expression and DPBA staining perfectly coincided in this region after auxin treatment (Fig. 4 D and F). In addition, the auxininduced increase in DPBA fluorescence observed in WT was lost in the arf7arf19 background (Fig. 4 G and H), as well as in WRKY23::WRKY23RNAi roots (Fig. 4 I and J). This indicates that

the WRKY23-mediated feedback mechanism through flavonol biosynthesis is part of a SLR-ARF7-ARF19 canonical auxinsignaling pathway. As final confirmation of the regulation of flavonol biosynthesis by WRKY23, we investigated whether the change in flavonol accumu-

Fig. 4. Local flavonol production is stimulated by WRKY23 in an ARF7/ ARF19-dependent manner. (A) Auxininduced WRKY23 expression is abolished in tir1-3 and arf7arf19 mutants. Black, mock-treated seedlings; gray, NAAtreated seedlings. Error bars represent SD. (B) qRT-PCR data showing auxin-induced expression of TT4 and TT7. Error bars represent SD. (C and D) DPBA-visualized flavonol accumulation induced by auxin (NAA) (D) compared with mocktreated roots (C). (E and F) Upon auxin treatment, WRKY23 expression is induced in the basal meristem, an area of strong flavonoid accumulation. Note that E and F are composite images. (G–J) Auxin-induced flavonol accumulation is not observed in the arf7arf19 background (G and H) or in WRKY23:: WRKY23RNAi roots (I and J). (K–M) Early root meristem defects typical for 35S:: WRKY23 seedlings are rescued by the absence of flavonols in the tt4-8 mutant. (K) tt4-8. (L) 35S::WRKY23. (M) tt4-8 × 35S::WRKY23.

Grunewald et al.

PNAS | January 31, 2012 | vol. 109 | no. 5 | 1557


Fig. 3. Flavonol biosynthesis is regulated by WRKY23. (A) qRT-PCR expression analysis of TT4, TT5, TT6, and TT7 in WT, WRKY23::WRKY23RNAi, RPS5A>>WRKY23, and mock-treated and Dex-treated 35S::WRKY23-GR seedlings. Error bars represent SD. (B and C) DPBA-stained WT (B) and 35S::WRKY23 (C) roots. (D and E) Extracted ion chromatograms (m/z = 447.09) showing the increased quercetin-3-O-rhamnoside (Q-3-O-R, green arrow) in 35S::WRKY23 roots (E) compared with WT C24 (D). (F and G) Base peak intensity chromatograms demonstrating the reduction in Q-R-G (red arrow) and rhamnetin-Oneohesperidoside (Rh, blue arrow) in WRKY23::WRKY23RNAi roots compared with WT Col-0.

lation could be causally connected to the observed WRKY23 developmental defects. We analyzed three independent tt7 insertion mutants and found a variation in root growth, similar to what was observed for the WRKY23 reduction-of-function lines. Although the majority of the seedlings exhibited a WT phenotype, ∼30% had reduced gravitropic root growth, a short root with a reduced number of lateral roots, and occasionally even stronger growth retardations (Fig. S6 K and M). To validate whether the effects of ectopic WRKY23 expression were due to an excess of flavonols, we crossed the 35S::WRKY23 line with the flavonol-deficient tt4-8 mutant. Intriguingly, the strong reduction in primary root growth and the early meristem differentiation, both characteristics of 35S:: WRKY23 roots, could be rescued by introducing the tt4-8 mutation into the 35S::WRKY23 background (Fig. 4 K–M and Fig. S6 N and O) after verifying that these crosses contained the tt4 mutation and WRKY23 transgene. This finding demonstrates that overproduction of flavonol derivatives is causal to the early arrest of WRKY23overexpressing root meristems, and thus that WRKY23 might act as a local regulator of flavonol biosynthesis. Discussion WRKY23 Fine-Tunes Transport-Dependent Auxin Maxima During Root Development. Over the past decade, a large body of evidence has

accumulated implicating WRKY proteins in the transcriptional reprogramming during plant defense responses (see ref. 19 for a review). Also the expression of WRKY23 is activated upon infection of plants with prokaryotic and eukaryotic pathogens (20, 38). However, WRKY23 also has appeared in several transcript profiling experiments studying plant developmental processes (21–23, 39–42). This occurrence is in agreement with the idea that WRKY transcription factors might be also involved in specific developmental programs (43–45). Using a parallel reduction-ofexpression and overexpression approach, we have demonstrated that WRKY23 is of major importance for proper root growth and development. The effect of misregulation of WRKY23 expression on root development, auxin response marker localization and intensity, and auxin transport indicates that WRKY23 negatively influences auxin transport and its dependent physiological processes. WRKY23 is expressed in root tips, and although its expression is strongly responsive to changes in auxin levels, its expression domain does not coincide exactly with the DR5 expression region, but also includes cells surrounding the auxin response maximum, such as the outer columella and lateral root cap cells. By regulating directional auxin transport in these cells, WRKY23 could contribute to the maintenance of the established auxin gradient and hence root meristematic activity. Accordingly, a reduced WRKY23 expression level was found to result in less-focused auxin responses as visualized by DR5, compromising the maintenance of the root stem cell niche organization upon germination. WRKY23 might thus be part of a feedback mechanism through which auxin can induce the focusing of its own distribution in the stem cells of the root apex. The hypothesis that WRKY23 acts in an extra fine-tuning system independent of the auxin gradient-organizing programs (46) is in line with the observations that WRKY23 does not affect expression of the PIN genes. Transcriptional Cascade from Auxin to Flavonol Accumulation. The expression of WRKY23 is regulated by auxin in a SLR-ARF7/ ARF19-dependent manner (20). Genetic, transcriptomic, and biochemical data suggest that WRKY23 affects auxin distribution by local control of the biosynthesis of flavonols. Previously, it was shown that the accumulation of flavonols is also elevated in response to auxin (16, 46). We could show that auxin-induced flavonol production is ARF7/ARF19-dependent and put forward a transcriptional cascade in which flavonol production is controlled by auxin through the action of WRKY23. Consistent with this model, the auxin induction of flavonol synthesis is substantially reduced in the WRKY23::WRKY23RNAi line. However, 1558 | www.pnas.org/cgi/doi/10.1073/pnas.1121134109

given that TT7 expression could be induced only after a 12-h Dex treatment of 35S::WRKY23-GR seedlings, WRKY23 might affect TT7 rather indirectly. Although we observed induced expression of four flavonol biosynthetic genes in 35S::WRKY23, when considering the results for all of the WRKY23 misexpression lines, WRKY23 seems to most strongly regulate the synthesis of quercetin and its derivatives through changes in TT7 transcription. Quercetin-rhamnoside accumulates in WRKY23 overexpression roots, whereas WRKY23 reduction-ofexpression roots lack quercetin-rhamnoside-glucoside (Q-R-G). Interestingly, these compounds are among the most abundant flavonol derivatives in WT roots (47), and in line with our results, Q-R-G is also strongly reduced in tt4 and tt7 mutant seedlings (47). Moreover, it has been shown that the ratio of quercetin to kaempferol in the Arabidopsis thaliana root tip increases after auxin treatment (16), and that quercetin is the most effective flavonol in competing with NPA for auxin transporter-binding sites (11). A recent report comparing the phenotypes of tt4 (which produces no flavonols) and the tt7 mutant (which accumulates kaempferol but not quercetin) posited that quercetin is the active flavonol that regulates basipetal auxin transport (toward the shoot–root transition zone) and the auxindependent physiological processes of root elongation and gravitropism (16). This suggests that the WRKY23-regulated production of quercetin derivatives through modulation of TT7 levels might be important for the control of auxin-mediated root growth and development. However, it remains to be investigated whether glycosylated flavonols can act as auxin transport inhibitors directly or whether the changes in flavonol derivatives reflect the available aglycone pool. In addition, because plants with a reduced WRKY23 function, generated either by RNAi or SRDX approaches, showed more severe phenotypes than the single flavonoid biosynthesis mutants, we anticipate that WRKY23 may affect other pathways as well. As mentioned earlier, WRKY23 was originally identified in a promoter-trap experiment as a plant-parasitic nematode–inducible gene (20). By means of gland secretions, plant-parasitic nematodes are able to modify gene expression in selected root cells of their host plant, ultimately establishing a nematode feeding site (for a review, see ref. 48). During this process, several developmental programs of the host are hijacked, which has led to the hypothesis that WRKY23 might be part of a captured developmental program rather than a pathogen-induced response (20). Interestingly, plant-parasitic nematodes are known to actively manipulate polar auxin transport (49) and flavonoids (more precisely, quercetin derivatives) accumulate at nematode infection sites (35). Therefore, the WRKY23quercetin regulation of auxin transport could be the original developmental mechanism that has been hijacked by plant pathogens during evolution. Materials and Methods Plant Material and Growth Conditions. A. thaliana (L.) Heynh. seeds were sterilized, germinated, and grown as described previously (20). The mutants and transgenic lines used in this study are described in SI Materials and Methods. Histological Analyses and Microscopy. GUS staining was done as described previously (20). Whole-mount in situ hybridization was performed as described previously (50) using a full-length WRKY23 RNA probe. For visualization of starch granules, roots were incubated for 5–10 min in Lugol solution [4 g of potassium iodide and 2 g of iodine crystals in 200 mL of MilliQ H2O (Millipore)], washed with MilliQ H2O, cleared with chloral hydrate, and analyzed immediately under differential interference contrast microscopy. Confocal microscopy was performed using a Zeiss LSM 510 confocal microscope; roots were briefly incubated in propidium iodide (3 mg/L), and then washed with and subsequently mounted in MilliQ H2O. Transient Expression Assays. Transient expression assays were performed as described previously (25). Details are presented in SI Materials and Methods. Auxin Transport Measurements. Eight-d-old seedlings were used for the transport assays. Acropetal auxin transport in the roots was measured as described previously (51). For the Dex-inducible lines, 35S::WRKY23-GR seedlings were

Grunewald et al.

kaempferol fluorescence (Fig. 3) or individual channels for quercetin and kaempferol (Fig. 4) were captured. Details of the LC-MS analyses are provided in SI Materials and Methods.

qRT-PCR and Microarray Analyses. RNA extraction and qRT-PCR were performed as described previously (52). All qRT-PCR values represent three biological replicates, each containing three technical replicates. A detailed description of how the qRT-PCR data shown in Fig. S6 were generated is provided in SI Materials and Methods along with the sequence of the primers used to quantify gene expression levels (Table S2). Details on the microarray setup and analysis are provided in SI Materials and Methods. Flavonol Accumulation Analyses. Flavonol accumulation in seedlings was visualized as described previously (16) using DPBA. Either combined quercetin and

ACKNOWLEDGMENTS. We thank Ben Scheres and Renze Heidstra for providing RPS5A::GAL4 seeds and a UAS- and a GR-containing plasmid. I.D.S. is supported by the Research Foundation Flanders and a Biotechnology and Biological Sciences Research Council David Phillips Fellowship. B.D.R. was financed by the Special Research Fund of Ghent University (predoctoral scholarship). W.B. thanks the Hercules Foundation for the Synapt Q-Tof for metabolomics (Grant AUGE/014) and C.L. thanks the Laboratory for Radioisotopes (LARI) Goettingen. D.R.L. and G.K.M. are supported by National Science Foundation Arabidopsis 2010 Program Grant IOB-0820717, and G. Gheysen is supported by Research Foundation Flanders Grant G.0031.08. W.G. is a postdoctoral fellow of the Research Foundation Flanders.

1. Sabatini S, et al. (1999) An auxin-dependent distal organizer of pattern and polarity in the Arabidopsis root. Cell 99:463–472. 2. Dubrovsky JG, et al. (2008) Auxin acts as a local morphogenetic trigger to specify lateral root founder cells. Proc Natl Acad Sci USA 105:8790–8794. 3. Friml J, et al. (2002) AtPIN4 mediates sink-driven auxin gradients and root patterning in Arabidopsis. Cell 108:661–673. 4. Ljung K, et al. (2005) Sites and regulation of auxin biosynthesis in Arabidopsis roots. Plant Cell 17:1090–1104. 5. Petersson SV, et al. (2009) An auxin gradient and maximum in the Arabidopsis root apex shown by high-resolution cell-specific analysis of IAA distribution and synthesis. Plant Cell 21:1659–1668. 6. Vieten A, Sauer M, Brewer PB, Friml J (2007) Molecular and cellular aspects of auxintransport–mediated development. Trends Plant Sci 12:160–168. 7. Bennett MJ, et al. (1996) Arabidopsis AUX1 gene: A permease-like regulator of root gravitropism. Science 273:948–950. 8. Geisler M, Murphy AS (2006) The ABC of auxin transport: The role of p-glycoproteins in plant development. FEBS Lett 580:1094–1102. 9. Grunewald W, Friml J (2010) The march of the PINs: Developmental plasticity by dynamic polar targeting in plant cells. EMBO J 29:2700–2714. 10. Swarup R, et al. (2001) Localization of the auxin permease AUX1 suggests two functionally distinct hormone transport pathways operate in the Arabidopsis root apex. Genes Dev 15:2648–2653. 11. Jacobs M, Rubery PH (1988) Naturally occurring auxin transport regulators. Science 241:346–349. 12. Bailly A, et al. (2008) Modulation of P-glycoproteins by auxin transport inhibitors is mediated by interaction with immunophilins. J Biol Chem 283:21817–21826. 13. Brown DE, et al. (2001) Flavonoids act as negative regulators of auxin transport in vivo in Arabidopsis. Plant Physiol 126:524–535. 14. Buer CS, Djordjevic MA (2009) Architectural phenotypes in the transparent testa mutants of Arabidopsis thaliana. J Exp Bot 60:751–763. 15. Buer CS, Muday GK (2004) The transparent testa4 mutation prevents flavonoid synthesis and alters auxin transport and the response of Arabidopsis roots to gravity and light. Plant Cell 16:1191–1205. 16. Lewis DR, et al. (2011) Auxin and ethylene induce flavonol accumulation through distinct transcriptional networks. Plant Physiol 156:144–164. 17. Murphy A, Peer WA, Taiz L (2000) Regulation of auxin transport by aminopeptidases and endogenous flavonoids. Planta 211:315–324. 18. Peer WA, et al. (2004) Variation in expression and protein localization of the PIN family of auxin efflux facilitator proteins in flavonoid mutants with altered auxin transport in Arabidopsis thaliana. Plant Cell 16:1898–1911. 19. Rushton PJ, Somssich IE, Ringler P, Shen QJ (2010) WRKY transcription factors. Trends Plant Sci 15:247–258. 20. Grunewald W, et al. (2008) A role for AtWRKY23 in feeding site establishment of plant-parasitic nematodes. Plant Physiol 148:358–368. 21. Brady SM, et al. (2007) A high-resolution root spatiotemporal map reveals dominant expression patterns. Science 318:801–806. 22. Okushima Y, et al. (2005) Functional genomic analysis of the AUXIN RESPONSE FACTOR gene family members in Arabidopsis thaliana: Unique and overlapping functions of ARF7 and ARF19. Plant Cell 17:444–463. 23. Vanneste S, et al. (2005) Cell cycle progression in the pericycle is not sufficient for SOLITARY ROOT/IAA14-mediated lateral root initiation in Arabidopsis thaliana. Plant Cell 17:3035–3050. 24. Hiratsu K, Matsui K, Koyama T, Ohme-Takagi M (2003) Dominant repression of target genes by chimeric repressors that include the EAR motif, a repression domain, in Arabidopsis. Plant J 34:733–739. 25. De Sutter V, et al. (2005) Exploration of jasmonate signalling via automated and standardized transient expression assays in tobacco cells. Plant J 44:1065–1076. 26. Benfey PN, Chua N-H (1990) The cauliflower mosaic virus 35S promoter: Combinatorial regulation of transcription in plants. Science 250:959–966.

27. Weijers D, et al. (2001) An Arabidopsis Minute-like phenotype caused by a semidominant mutation in a RIBOSOMAL PROTEIN S5 gene. Development 128:4289–4299. 28. van den Berg C, Willemsen V, Hendriks G, Weisbeek P, Scheres B (1997) Short-range control of cell differentiation in the Arabidopsis root meristem. Nature 390:287–289. 29. De Smet I, et al. (2008) Receptor-like kinase ACR4 restricts formative cell divisions in the Arabidopsis root. Science 322:594–597. 30. Sarkar AK, et al. (2007) Conserved factors regulate signalling in Arabidopsis thaliana shoot and root stem cell organizers. Nature 446:811–814. 31. Stahl Y, Wink RH, Ingram GC, Simon R (2009) A signaling module controlling the stem cell niche in Arabidopsis root meristems. Curr Biol 19:909–914. 32. Ulmasov T, Murfett J, Hagen G, Guilfoyle TJ (1997) Aux/IAA proteins repress expression of reporter genes containing natural and highly active synthetic auxin response elements. Plant Cell 9:1963–1971. 33. Benková E, et al. (2003) Local, efflux-dependent auxin gradients as a common module for plant organ formation. Cell 115:591–602. 34. Levée V, et al. (2009) Expression profiling and functional analysis of Populus WRKY23 reveals a regulatory role in defense. New Phytol 184:48–70. 35. Jones JT, Furlanetto C, Phillips MS (2007) The role of flavonoids produced in response to cyst nematode infection of Arabidopsis thaliana. Nematology 9:671–677. 36. Okushima Y, Fukaki H, Onoda M, Theologis A, Tasaka M (2007) ARF7 and ARF19 regulate lateral root formation via direct activation of LBD/ASL genes in Arabidopsis. Plant Cell 19:118–130. 37. Lau S, Jürgens G, De Smet I (2008) The evolving complexity of the auxin pathway. Plant Cell 20:1738–1746. 38. Dong JX, Chen CH, Chen ZX (2003) Expression profiles of the Arabidopsis WRKY gene superfamily during plant defense response. Plant Mol Biol 51:21–37. 39. Goda H, et al. (2004) Comprehensive comparison of auxin-regulated and brassinosteroid-regulated genes in Arabidopsis. Plant Physiol 134:1555–1573. 40. Lee DJ, et al. (2007) Genome-wide expression profiling of ARABIDOPSIS RESPONSE REGULATOR 7(ARR7) overexpression in cytokinin response. Mol Genet Genomics 277: 115–137. 41. Morant M, et al. (2010) Metabolomic, transcriptional, hormonal, and signaling crosstalk in superroot2. Mol Plant 3:192–211. 42. Hachez C, Ohashi-Ito K, Dong J, Bergmann DC (2011) Differentiation of Arabidopsis guard cells: Analysis of the networks incorporating the basic helix-loop-helix transcription factor, FAMA. Plant Physiol 155:1458–1472. 43. Johnson CS, Kolevski B, Smyth DR (2002) TRANSPARENT TESTA GLABRA2, a trichome and seed coat development gene of Arabidopsis, encodes a WRKY transcription factor. Plant Cell 14:1359–1375. 44. Luo M, Dennis ES, Berger F, Peacock WJ, Chaudhury A (2005) MINISEED3 (MINI3), a WRKY family gene, and HAIKU2 (IKU2), a leucine-rich repeat (LRR) KINASE gene, are regulators of seed size in Arabidopsis. Proc Natl Acad Sci USA 102:17531–17536. 45. Ueda M, Zhang Z, Laux T (2011) Transcriptional activation of Arabidopsis axis patterning genes WOX8/9 links zygote polarity to embryo development. Dev Cell 20: 264–270. 46. Blilou I, et al. (2005) The PIN auxin efflux facilitator network controls growth and patterning in Arabidopsis roots. Nature 433:39–44. 47. Stracke R, et al. (2007) Differential regulation of closely related R2R3-MYB transcription factors controls flavonol accumulation in different parts of the Arabidopsis thaliana seedling. Plant J 50:660–677. 48. Grunewald W, et al. (2009) Manipulation of auxin transport in plant roots during Rhizobium symbiosis and nematode parasitism. Plant Cell 21:2553–2562. 49. Grunewald W, Cannoot B, Friml J, Gheysen G (2009) Parasitic nematodes modulate PIN-mediated auxin transport to facilitate infection. PLoS Pathog 5:e1000266. 50. Hejátko J, et al. (2006) In situ hybridization technique for mRNA detection in whole mount Arabidopsis samples. Nat Protoc 1:1939–1946. 51. Kitakura S, et al. (2011) Clathrin mediates endocytosis and polar distribution of PIN auxin transporters in Arabidopsis. Plant Cell 23:1920–1931. 52. Grunewald W, et al. (2009) Expression of the Arabidopsis jasmonate signalling repressor JAZ1/TIFY10A is stimulated by auxin. EMBO Rep 10:923–928.

Grunewald et al.

PNAS | January 31, 2012 | vol. 109 | no. 5 | 1559


transferred on half-strength Murashige and Skoog medium supplemented with 10 μM Dex (Sigma-Aldrich) or an equal amount of solvent (DMSO) at 26 h before the start of the assay.

Supporting Information Grunewald et al. 10.1073/pnas.1121134109 SI Materials and Methods Plant Material and Growth Conditions. Q0680, Q1630, tt7-5

(SALK_124157, Col-8), and tt7-6 (SALK_039417, Col-8) were obtained from the Nottingham Arabidopsis Stock Center. The RPS5A::GAL4 activator line has been described previously (1), as have arf7arf19 (2), tir1-3 (3), DR5::GUS (4), CycB1;1::GUS (5), QC184 (6), WRKY23::GUS and WRKY23::GFP (7), tt4-8 (Ws), and tt7-4 (Ws) (8). For WRKY23::WRKY23RNAi, 35S::WRKY23-SRDX, and 35S::WRKY23 plants, homozygous T3 plants were analyzed. For RPS5A>>WRKY23 analyses, the RPS5A::GAL4 × UAS:: WRKY23 F1 cross was analyzed and compared with UAS:: WRKY23 × WT Col-0 and RPS5A::GAL4 × WT Col-0 F1 crosses. For the analysis of tt4-8 (Ws) × 35S::WRKY23 (C24), internal crossing controls were used to cope with the different ecotype backgrounds. Next to the tt4-8 × 35S::WRKY23 plants, the tt4-8 single mutant and the 35S::WRKY23 overexpression line were reselected in the F2 tt4-8 × 35S::WRKY23 generation and used as respective controls. Construction of Transgenic Lines. DNA constructs were created with the Gateway cloning technology (9). For the WRKY23:: WRKY23RNAi construct, a 109-bp-long WRKY23-specific fragment (5′-TCCACCGTCAGAGCAATTAGTGACGTCAAAGGTGGAGTCTTTGTG TTCGGATCATTTGTTGATAAACCCACCGGCGACTCCTAACTCGTCATCGATTTCGTCTCTGCTT-3′) was amplified with primer W23GS5 (5′-TCCACCGTCAGAGCAATTAGT-3′) and primer W23GS3 (5′-AAGCAGACGAAATCGATGA-3′) and recombined with pDONR221 to yield pEN-L1-WGST-R2. The 35S promoter of the Gateway vector pK7GWIWG2(II) was replaced by a linker containing several unique cloning sites (pK7GWIWG2-L). The 3,195-bp WRKY23 promoter (7) was introduced into this linker (pK7GWIWG2-L/ WRKY23). The latter vector was recombined with pEN-L1WGST-R2 by Gateway LR cloning. For the 35S::WRKY23-SRDX construct, the genomic WRKY23 was amplified with WRKY23_ ATTB1_FOR (5′-GGGGACAAGTTTGTACAAAAAAGCAGGCTTCAT GGAGTTTACAGATTTCTCAAAGACGAG-3′) and WRKY23_SRDX_ATTB2_REV (5′-GGGGACCACTTTGTACAAGAAAGCTGGGTGTCAAGCGAAACCCAAACGGAGTTCTAGATCCAGATCGAGCTCTTCCTTCAACATATG-3′) containing the SRDX transcriptional repressor motif (underlined) (10), and then recombined with pDONR221 to yield pEN-L1-WRKY23SRDX-R2. The latter vector was used in a Gateway LR cloning with pK7WG2. For the 35S::WRKY23, genomic WRKY23 was amplified with WRKY23_ATTB1_FOR and WRKY23_ATTB2_REV (5′-GGGGACCACTTTGTACAAGAAAGCTGGGTGCTACTCTTC CTTCAACATATG-3′) and then recombined with pDONR221 to yield pEN-L1-WRKY23R2. The latter vector was used in a Gateway LR cloning with pK7WG2. For construction of the UAS::WRKY23 line, the UAS promoter (5′-GAATTCGATATGAAGCTTGATATCGG GTGACAGCCCTCCGAGCGGGTGACAGCCCTCCGAAGCGGGTGACAGCCCTCCGACGTCGGGTGACAGCCCTCCGAGCGGGTGACAGCCCTCCGAGCGGGTGACAGCCCTCCGCTGCAGCAAGACCCTTCCTCTATATAAGGAAGTTCATTTCATTTGGAGAGGACAG-3′) was cloned from a UAScontaining pGreen plasmid (a gift from Dr. Renze Heidstra, Utrecht University, Utrecht, The Netherlands) with the primers attB1_UAS_forward (5′-CAACTTTGTATAGAAAAGTTGGAATTCGATATGAAGCTTGATAT-3′) and attB2_ UAS_reverse (5′-GGGGACAAGTTTGTACAAACTTGCTGTCCTCTCCAAATGAA ATG-3′), and then cloned in the Grunewald et al. www.pnas.org/cgi/content/short/1121134109

pDONRP4P1R to yield pEN_L4_UAS_R1. The latter vector was used together with pEN-L1-WRKY23-R2 and pK7m24GW.3 in a multisite Gateway LR reaction. For construction of the 35S:: WRKY23-GR-inducible line, the GR sequence (a gift from Dr. Renze Heidstra) was cloned in pDONRP2P3 to yield pENL2-GR-R3 and then used in a multisite Gateway LR reaction together with pEN-L4-35S-R1, pEN-L1-WRKY23(-stop)-R2, and pK7m34GW. The obtained vectors were transferred to Agrobacterium tumefaciens strain C58C1 (pMP90), which was used in a floral dip transformation of Arabidopsis thaliana (L.) Heyhn Columbia-0 (Col-0), except for the 35S::WRKY23 construct, which was transformed in Arabidopsis C24. Transient Expression Assays. Protoplasts were prepared from a Bright Yellow-2 tobacco cell culture and cotransfected with a reporter plasmid containing the fLUC reporter gene driven by a promoter containing five GAL4-binding sites, a normalization construct expressing Renilla luciferase (rLUC) under control of the 35S promoter, and effector constructs. Effector constructs with GAL4 fused to the C terminus of WRKY23 were created by combining pEN-L4-2-R1 (35S promoter), pEN-L1-WRKY23 (-stop)-R2, and pEN-L2-GAL4-R3 (11), with pm43GW7 as destination vector, in a MultiSite Gateway LR reaction. For N-terminal fusions, the p2GAL4GW6 destination vector (11) was used in a LR reaction with pEN-L1-WRKY23-R2. Each experiment used 2 mg of each plasmid. Eight biological repeats were performed. After transfection, protoplasts were incubated overnight and then lysed; fLUC and rLUC activities were determined with the Promega Dual-Luciferase Reporter Assay System. Variations in transfection efficiency and technical errors were corrected by normalization of fLUC by rLUC activities. The data were statistically analyzed using SPSS version 12.1, with the Levene test for homogeneity of variance and the independent sample t test to compare means. Auxin Treatments for qRT-PCR. For the experiments shown in Fig. 4 A and B, plants were germinated on a nylon screen (03-100/32; Sefar Filtration) as described previously (12), with ∼100 seedlings per plate. At 6 d after being transferred to light, the nylon screen and plants were transferred to a control plate or to medium supplemented with 1 μM α-naphthaleneacetic acid for the indicated time. At the end of the treatment period, roots were excised, and the samples were promptly frozen in liquid nitrogen, ground, and used for RNA isolation. RNA Extraction, cDNA Synthesis, and Quantitative RT-PCR Analysis.

For RNA extraction, plants were ground, and total RNA was isolated with TRIzol (Invitrogen) in accordance with the manufacturer’s instructions. Poly(dT) cDNA was prepared from 2 μg of total RNA with SuperScript III reverse transcriptase (Invitrogen) and quantified on an LightCycler 480 apparatus (Roche Diagnostics) with the SYBR Green I Master kit (Roche Diagnostics) in accordance with the manufacturer’s instructions. For the experiment shown in Fig. 3A and the quantitative RTPCR (qRT-PCR) results presented here, target quantifications were performed with specific primer pairs designed with Beacon Designer 4.0 (Premier Biosoft). All individual reactions were done in triplicate. Data were analyzed with qBase (13). Expression levels were normalized to those of EEF1α4 and CDKA, which showed no clear systematic changes in Ct value. For the data shown in Fig. 4 A and B, primers were designed with Primer Express software (Applied Biosystems) or Primer3 primer design 1 of 8

analysis, roots of 4-d-old WT Col-0, WRKY23::WRKY23RNAi (50H

and 43A), WT C24, and 35S::WRKY23 seedlings were harvested and ground in liquid nitrogen. Approximately 100 roots were pooled for each extraction, and six biological repeats were performed for each line. Flavonols were extracted by adding 1 mL of HPLC-grade methanol. After centrifugation, the supernatant was dried, and the remnant was resuspended in 1:1 cyclohexane:MilliQ H2O (Millipore). After centrifugation, the water phase was collected, and a 5-μL aliquot was subjected to LC-MS analysis with an Acquity ultra-performance liquid chromatography system (Waters) connected to a Synapt HDMS quadrupole time-of-flight mass spectrometer (Micromass). Chromatographic separation was done on a Waters Acquity BEH C18 column (2.1 mm × 150 mm; 1.7 μm) with a gradient elution. Mobile phases were composed of (A) water containing 1% acetonitrile (ACN) and 0.1% formic acid and (B) ACN containing 1% water and 0.1% formic acid. The column temperature was maintained at 40 °C, and the autosampler temperature was maintained at 10 °C. A flow rate of 350 μL/min was applied during the gradient elution, with initialization at time 0 min 5% (B), 30 min 50% (B), and 33 min 100% (B). The eluant was directed to the mass spectrometer equipped with an electrospray ionization source and lockspray interface for accurate mass measurements. The MS source parameters were capillary voltage, 1.5 kV; sampling cone, 40 V; extraction cone, 4 V; source temperature, 120 °C; desolvation temperature, 350 °C; cone gas flow, 50 L/h; and desolvation gas flow, 550 L/h. The collision energy for the trap and transfer cells was set at 6 V and 4 V, respectively. For data acquisition, the dynamic range enhancement mode was activated. Fullscan data were recorded in negative centroid V-mode; the mass range was set between m/z 100 and 1,000, with a scan speed of 0.2 s/ scan, with Masslynx software (Waters). Leu-enkephalin [400 pg/μL solubilized in water/ACN 1:1 (vol/vol), acidified with 0.1% formic acid] was used for the lock mass calibration, with scanning every 10 s with a scan time of 0.5 s. The values from three scans were averaged. For MS/MS purposes, the same settings were applied, except that the trap collision energy was ramped from 10 to 45 V. All solvents used were ULC/MS grade (Biosolve), and water was produced by a DirectQ-UV water purification system (Millipore). The identity of quercitrin, quercetin-rhamnoside-glucoside, and rhamnetin-O-neohesperidoside was confirmed by accurate mass measurements and fragmentation patterns. For quercitrin, coelution with a standard was performed as well.

1. Aida M, et al. (2004) The PLETHORA genes mediate patterning of the Arabidopsis root stem cell niche. Cell 119:109–120. 2. Okushima Y, Fukaki H, Onoda M, Theologis A, Tasaka M (2007) ARF7 and ARF19 regulate lateral root formation via direct activation of LBD/ASL genes in Arabidopsis. Plant Cell 19:118–130. 3. Ruegger M, et al. (1998) The TIR1 protein of Arabidopsis functions in auxin response and is related to human SKP2 and yeast grr1p. Genes Dev 12:198–207. 4. Ulmasov T, Murfett J, Hagen G, Guilfoyle TJ (1997) Aux/IAA proteins repress expression of reporter genes containing natural and highly active synthetic auxin response elements. Plant Cell 9:1963–1971. 5. Colón-Carmona A, You R, Haimovitch-Gal T, Doerner P (1999) Technical advance: Spatio-temporal analysis of mitotic activity with a labile cyclin-GUS fusion protein. Plant J 20:503–508. 6. Sabatini S, Heidstra R, Wildwater M, Scheres B (2003) SCARECROW is involved in positioning the stem cell niche in the Arabidopsis root meristem. Genes Dev 17: 354–358. 7. Grunewald W, et al. (2008) A role for AtWRKY23 in feeding site establishment of plant-parasitic nematodes. Plant Physiol 148:358–368. 8. Routaboul J-M, et al. (2006) Flavonoid diversity and biosynthesis in seed of Arabidopsis thaliana. Planta 224:96–107.

9. Karimi M, Depicker A, Hilson P (2007) Recombinational cloning with plant gateway vectors. Plant Physiol 145:1144–1154. 10. Hiratsu K, Matsui K, Koyama T, Ohme-Takagi M (2003) Dominant repression of target genes by chimeric repressors that include the EAR motif, a repression domain, in Arabidopsis. Plant J 34:733–739. 11. Pauwels L, et al. (2010) NINJA connects the co-repressor TOPLESS to jasmonate signalling. Nature 464:788–791. 12. Levesque MP, et al. (2006) Whole-genome analysis of the SHORT-ROOT developmental pathway in Arabidopsis. PLoS Biol 4:e143. 13. Hellemans J, Mortier G, De Paepe A, Speleman F, Vandesompele J (2007) qBase relative quantification framework and software for management and automated analysis of real-time quantitative PCR data. Genome Biol 8:R19. 14. Rozen S, Skaletsky H (2000) Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol 132:365–386. 15. Pfaffl MW (2001) A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 29:e45. 16. Irizarry RA, et al. (2003) Summaries of Affymetrix GeneChip probe level data. Nucleic Acids Res 31:e15. 17. Saeed AI, et al. (2003) TM4: A free, open-source system for microarray data management and analysis. Biotechniques 34:374–378.

software (14) using default parameters. The efficiency coefficient E (15) was calculated for all primer sets individually by plotting the relationship between the Ct value and log[cDNA]. A standard curve was prepared, and for all primer sets used, the linear relationship was verified over a 1,000-fold concentration range using the standard curve method in accordance with the manufacturer’s instructions (Applied Biosystems). The resultant primer efficiencies were used in an efficiency-corrected ΔΔCt formula. E values were 1.9 for ACTIN2, 2.0 for CHS, 2.1 for F3′ H, and 2.1 for WRKY23. Primer-specific master mixes containing SYBR Green, water, and the gene-specific primers were prespared and dispensed to a 96-well plate. Then 2 μL of the cDNA samples was added to each well with an electronic repeater pipette (Rainin). The transcript levels of the reference gene, ACTIN, did not vary by more than 1 Ct between treatments, suggesting that this gene is an appropriately stable transcript for normalization. All qRT-PCR values represent three biological replicates, each containing three technical replicates. These biological replicates represent samples grown at three separate times but under very carefully matched growth conditions, including media, light, and time of day. The reference and target Ct numbers were entered into a Microsoft Excel spreadsheet to calculate the efficiency-corrected relative expression measured in fold increase over untreated control with an efficiency-corrected ΔΔCt (15). Microarray Analysis. Roots of 3-d-old Col-0 WT, WRKY23 RNAi (line 50H, classes I and II), and RPS5A>>WRKY23 plants were collected for RNA isolation. All points were sampled in three independent experiments. Total RNA (200 μg per array) was used to hybridize ATH1 Affymetrix Arabidopsis arrays in accordance with standard procedures. The expression values were RMA-normalized (16) with R (www.r-project.org) and the Bioconductor package affylmGUI (http://bioinf.wehi.edu.au/affylmGUI/). Genes with the same or contrasting WRKY23 expression profiles were selected by Pavlidis template matching in TMeV 4.0 (TIGR) (17). Finally, genes with a significant P value (>WRKY23 (Right) seedlings. (S and T) Propidium iodide-stained root tips of WT (S) and RPS5A>>WRKY23 (T) seedlings. (U and V) Primary root length (U) and lateral root density (V) of RPS5A>>WRKY23 and WT Col-0.

Fig. S2. WRKY23 is not a cell-autonomous differentiation factor. Differential interference contrast (DIC) microscopic analysis of UAS::WRKY23 (A), RPS5A>>WRKY23 (B), J2341>>WRKY23 (C), and Q0680>>WRKY23 (D) roots. (Insets) Respective GAL4 activation domains.

Fig. S3. Phenotypic analysis of dexamethasone (Dex)-inducible overexpression lines. (A) WT Col-0 (Left) and 35S::WRKY23-GR (Right) seedlings germinated on MS medium without Dex and transferred to MS medium with 10 μM Dex. The white line illustrates the transfer. (B) 35S::WRKY23-GR seedlings grown in the absence of Dex exhibit no morphological aberrations. (C) DIC image of a differentiated Dex-induced 35S::WRKY23-GR root tip. (D and E) Upon transfer to MS medium supplemented with 10 μM Dex, columella initial cells of 35S::WRKY23-GR roots accumulate starch granules (E, at 4 d after transfer), identical to what is observed in 35S::WRKY23 roots, and in contrast to what occurs in Dex-treated WT roots (D). Asterisks indicate QC cells; the arrowhead points to columella initial cells.

Grunewald et al. www.pnas.org/cgi/content/short/1121134109

4 of 8

Fig. S4. Expression analysis of PIN genes. PIN1 (black). PIN2 (gray). PIN3 (light gray). PIN4 (white), and PIN7 (dark gray) in WT Col-0, WRKY23::WRKY23RNAi, WT C24, and 35S::WRKY23 seedlings. Error bars represent SD.

Fig. S5. Schematic presentation of the flavonoid biosynthesis pathway. CHS, chalcone synthase; CHI, chalcone isomerase; F3H, flavanone 3-hydroxylase; F3′H, flavonoid 3′-hydroxylase; FLS, flavonol synthase; DFR, dihydroflavonol 4-reductase; TT, transparent testa. Double arrows represent multiple enzymatic steps.


5 of 8

Fig. S6. WRKY23 is a positive regulator of flavonoid biosynthesis. (A–J) Comparison of diphenylboric acid 2-aminoethyl ester-visualized flavonoid accumulation (A, C, D, and E) and WRKY23::GUS staining patterns (F, H, I, and J). WRKY23::GFP (B) and WRKY23 mRNA in situ hybridization (G) in primary root tips (A, B, F, and G), lateral root primordia (C and H), cotyledon hydathodes (D and I), and the hypocotyl–root transition zone (E and J). (K–M) Phenotypic variation in tt7-4 (K), tt7-5 (L), and tt7-6 (M) mutant seedlings. (N) Comparison of root lengths in tt4-8, 35S::WRKY23, and tt4-8 × 35S::WRKY23 seedlings.*P < 0.05; ***P < 0.001, Student t test. (O) Early root meristem differentiation typical for 35S::WRKY23 seedlings is partially rescued by the lack of flavonoids in the tt4-8 mutant.


6 of 8

Table S1. Expression levels (fold changes) of selected transcripts differentially regulated in WRKY23::WRKY23RNAi (RNAi) and RPS5A>>WRKY23 (OE) AGI code

Annotation

Up-regulated in RPS5A>>WRKY23OE and down-regulated in RNAi AT2G47260 WRKY23 (WRKY DNA-binding protein 23); transcription factor AT2G05380 GRP3S (GLYCINE-RICH PROTEIN 3 SHORT ISOFORM) AT5G44400 FAD-binding domain-containing protein AT5G64120 Peroxidase; putative AT5G25610 RD22 (RESPONSIVE TO DESSICATION 22) AT3G45160 Unknown protein AT2G38530 LTP2 (LIPID TRANSFER PROTEIN 2) AT5G59320 LTP3 (LIPID TRANSFER PROTEIN 3) AT3G23570 Dienelactone hydrolase family protein AT2G38540 LP1 (nonspecific lipid transfer protein 1) AT2G31380 STH (salt tolerance homolog) AT3G20470 GLYCINE-RICH PROTEIN 5 AT5G07990 TT7 (TRANSPARENT TESTA 7); flavonoid 3′-monooxygenase AT4G21620 Glycine-rich protein AT1G23740 Oxidoreductase; zinc-binding dehydrogenase family protein AT3G16530 Legume lectin family protein AT4G04750 Carbohydrate transporter/sugar porter AT1G18100 E12A11; phosphatidylethanolamine binding AT4G37150 Esterase; putative AT2G38860 YLS5 (yellow-leaf-specific gene 5) AT1G77760 NIA1 (NITRATE REDUCTASE 1) AT3G02910 Unknown protein AT3G61990 O-methyltransferase family 3 protein AT4G11190 Disease resistance-responsive family protein AT1G51340 MATE efflux family protein AT3G61220 Short-chain dehydrogenase/reductase (SDR) family protein AT1G64500 Glutaredoxin family protein AT1G02930/20 Glutathione transferase; AT1G02930 (ATGSTF6); AT1G02920 (ATGSTF7) AT3G19710 BCAT4 (BRANCHED-CHAIN AMINOTRANSFERASE4) Down-regulated in RPS5A>>WRKY23OE and up-regulated in RNAi AT3G02170 LNG2 (LONGIFOLIA2) AT3G28750 Unknown protein AT1G11190 BFN1 (BIFUNCTIONAL NUCLEASE I) AT4G28410 Aminotransferase-related AT3G14770 Nodulin MtN3 family protein AT5G51810 ATGA20OX2/GA20OX2 (Gibberellin 20 oxidase 2) AT2G45220 Pectinesterase family protein AT1G69526 UbiE/COQ5 methyltransferase family protein AT1G12090 ELP (EXTENSIN-LIKE PROTEIN) AT5G61440 Thioredoxin family protein AT3G20340 Unknown protein AT5G04150 Basic helix–loop–helix (bHLH) family protein AT5G35940 Jacalin lectin family protein AT5G03860 Malate synthase; putative AT2G38600 Acid phosphatase class B family protein


OE

RNAi

18.08 5.39 3.28 2.78 2.78 2.59 2.58 2.46 2.26 2.25 2.24 2.21 1.98 1.96 1.91 1.81 1.76 1.70 1.67 1.65 1.64 1.62 1.62 1.58 1.57 1.55 1.53 1.53 1.52

0.41 0.53 0.85 0.83 0.82 0.67 0.45 0.52 0.78 0.66 0.61 0.74 0.11 0.69 0.90 0.59 0.82 0.58 0.81 0.86 0.85 0.78 0.85 0.69 0.56 0.82 0.71 0.62 0.48

0.74 0.72 0.72 0.70 0.63 0.62 0.61 0.61 0.58 0.57 0.52 0.47 0.37 0.36 0.30

1.32 1.24 1.29 1.32 1.34 1.30 1.24 1.66 1.21 1.29 1.23 1.44 1.40 1.53 1.39

7 of 8

Table S2. Primer sequences Primer name WRKY23_Q forward WRKY23_Q reverse EEF forward EEF reverse CDKA forward CDKA reverse TT4 forward TT4 reverse TT5 forward TT5 reverse TT6 forward TT6 reverse TT7 forward TT7 reverse ACTIN forward ACTIN reverse CHS forward CHS reverse MYB12 forward MYB12 reverse WRKY28 forward WRKY28 reverse WRKY48 forward WRKY48 reverse WRKY50 forward WRKY50 reverse WRKY51 forward WRKY51 reverse WRKY57 forward WRKY57 reverse WRKY59 forward WRKY59 reverse WRKY71 forward WRKY71 reverse WRKY8 forward WRKY8 reverse PIN1 forward PIN1 reverse PIN2 forward PIN2 reverse PIN3 forward PIN3 reverse PIN4 forward PIN4 reverse PIN7 forward PIN7 reverse


Sequence 5′-AGTCTCGGTAATGGTTGCTTTGG-3′ 5′-TGTTGCTGCTGTTGGTGATGG-3′ 5′-CTGGAGGTTTTGAGGCTGGTAT-3′ 5′-CCAAGGGTGAAAGCAAGAAGA-3′ 5′-ATTGCGTATTGCCACTCTCATAGG-3′ 5′-TCCTGACAGGGATACCGAATGC-3′ 5′-ATGGTGATGGCTGGTGCTTC-3′ 5′-CCTTGAGGTCGGTCATGTGTTC-3′ 5′-GGAGGCGGTTCTGGAATCTATC-3′ 5′-TCGTCCTTGTTCTTCATCATTAGC-3′ 5′-TCGTCTCTAGTCACCTCCAG-3′ 5′-TCACTTTCACCCAACCTTCC-3′ 5′-TAGCCGACCACCAAACTC-3′ 5′-AGCGTTCCAACCTCTTCC-3′ 5′-TGAGAGATTCAGATGCCCAGAA-3′ 5′-GCAGCTTCCATTCCCACAA-3′ 5′-CGTGTTGAGCGAGTATGGAAAC-3′ 5′-TGACTTCCTCCTCATCTCGTCTAGT-3′ 5′-AATCCAACGGTGAAGGTTCTTG-3′ 5′-TCCACGCTTGAGGTCTGATC-3′ 5′-GTTCTCCAGCAGCGTATGAATCTC-3′ 5′-AGTAACATCACGGTTCGGTTCTTG-3′ 5′-TTTACGGATTTGCCCTTACCTC-3′ 5′-CTCTTGTTGTTGATCTCCTTCTTC-3′ 5′-CAATAAGAAATGTTGTTCCCTACC-3′ 5′-GAGAGTTGCGTTCAAGACAC-3′ 5′-AGTGAAGGTTGCTCGGTGAAG-3′ 5′-GAGACTCTCATGGTTATGGACTCC-3′ 5′-CACCAAGTAAGCCTTCCTCTG-3′ 5′-GCACACGACCCACATAGC-3′ 5′-AAGTCGTCTGAATCACAATAAAC-3′ 5′-CAGGCATTATCACAAGTGTTC-3′ 5′-TCGACGTTGAGAGGAACC-3′ 5′-AGATCCCACCGATTGATAATTG-3′ 5′-CTCACTCTCCTGTTGATGAAATCC-3′ 5′-TGATCTCTTCCGTGTGCCATAC-3′ 5′-TACTCCGAGACCTTCCAACTACG-3′ 5′-TCCACCGCCACCACTTCC-3′ 5′-CCTCGCCGCACTCTTTCTTTGG-3′ 5′-CCGTACATCGCCCTAAGCAATGG-3′ 5′-GAGGGAGAAGGAAGAAAGGGAAAC-3′ 5′-CTTGGCTTGTAATGTTGGCATCAG-3′ 5′-GATGCTGGTCTTGGAATGG-3′ 5′-CCTGAACGATGGCTATACG-3′ 5′-CTTGGTATGGCAATGTTCAG-3′ 5′-CACACGCAATAGGTCTCC-3′

8 of 8