Research ArticlePlant biology

LEAFY Controls Auxin Response Pathways in Floral Primordium Formation

See allHide authors and affiliations

Science Signaling  09 Apr 2013:
Vol. 6, Issue 270, pp. ra23
DOI: 10.1126/scisignal.2003937

This article has a correction. Please see:

Abstract

The transcription factor LEAFY is a master regulator of flowering and of flower development. It acts as a component of a switch that mediates the transition from the vegetative to the reproductive phase of plant development. Auxin is a plant hormone with many different roles in plant growth, including the induction of new primordia of both leaves and flowers at the shoot apex. We report that LEAFY acts in part by controlling the auxin response pathway in new primordia. Therefore, regulation of flower development by transcriptional master regulators and hormonal control of morphogenesis appear to be interacting processes. We found that hormone perception not only controls but is also controlled by the transcriptional signals that create plant form.

Introduction

The aboveground development of plants is the result of the continued production of new leaf and flower primordia and new stem tissue, which form from a collection of stem cells at the tip of each shoot, called the shoot apical meristem (SAM). The formation of new leaves and flowers in the peripheral zone of the SAM is a result of locally increased concentration of the plant hormone auxin, the application of which has long been known to induce new primordia (1). This increased auxin concentration results, at least in part, from regulated auxin transport mediated by the asymmetric distribution within SAM epidermal cells of proteins of the PIN-FORMED family of auxin efflux carriers. The subcellular localization of these auxin efflux carriers is regulated by the antagonistic actions of the kinase PINOID (PID) and phosphatase PP2A (2). This auxin distribution pattern ultimately results from the cellular interpretation of the physical forces created by expanding cells on their borders (3, 4). Spatially and temporally regulated auxin biosynthesis also contributes to flower and leaf development (5, 6).

Auxin signal transduction is the result of a network of interactions between three types of cellular components (7). Auxin receptors interact with auxin and indole acetic acid (AUX/IAA) proteins such that binding of the hormone to the receptor triggers the degradation of the AUX/IAA proteins through a conserved degron. AUX/IAA proteins interact with auxin response factors (ARFs) to suppress their transcriptional regulation capability. Removal of AUX/IAA proteins, therefore, releases ARFs to either activate or suppress auxin-induced transcription. The AUX/IAA protein family of Arabidopsis has 29 members, and the ARF family has 23 members. These components exhibit different temporal and spatial patterns of expression, and specific AUX/IAA-ARF interaction pairs are major controls in tissue-specific auxin regulation pathways (8).

LEAFY (LFY) is a plant-specific transcription factor that integrates the environmental and internal signals that trigger the floral transition (9, 10). It functions by activating downstream meristem identity genes, such as APETALA1, that also serve as positive regulators of LFY expression, thereby creating a switch that irreversibly activates floral development (11). LFY is expressed in newly initiating floral primordia in their early developmental stages (12). LFY interacts with the auxin signaling pathway: LFY exhibits a genetic interaction with pinoid mutants, induces PINOID and auxin-related genes, and affects auxin response (1316). These results point to a role for LFY in the auxin-induced outgrowth of floral primordia, although lfy-null mutants do not cause severe defects in primordial formation or outgrowth. Thus, interactions between LFY-related processes and auxin signaling appear to exist, but their nature and importance remain unclear.

Results

LFY and PID interact genetically

To learn more about the genetic pathways that interact with the LFY network during floral induction and flower development, we performed a genetic modifier screen in which more than 5000 ethyl methanesulfonate–mutagenized weak LFY loss-of-function lfy-5 homozygous mutant seeds were grown and monitored for second-site enhancement or suppression of the lfy-5 phenotype (17). Homozygous lfy-5 plants have defective floral development, in which fewer than the normal number of petals and stamens are formed because of reduction in B-function homeotic gene activity, and in which increased numbers of secondary inflorescences develop, indicating a delay in specification of the SAM as an inflorescence, rather than a vegetative, meristem (17). One line of mutagenized lfy-5 plants showed pin-like shoot apices (Fig. 1A, pid-102 lfy-5, and fig. S1A). The second-site mutation, named pid-102, when crossed away from lfy-5, had a phenotype similar to reported weak alleles of pid (Fig. 1A, pid-102, and fig. S1B), and map-based cloning and sequencing showed that this mutant was an R204K change in the PID protein (fig. S1C). Wild-type plants of both Landsberg erecta (Ler-0) and Wassilewskija ecotypes produce continuous leaves and flowers around the stems following a spiral phyllotactic pattern. This pattern is unaffected in lfy-5, pid-102, or pid-8 plants (Fig. 1A) (18). A mutant phenotype similar to that of the pid-102 lfy-5 double mutant was observed in double mutants of either the weak lfy-5 or the strong lfy-6 mutant with pid-8 (fig. S1, D and E), a weak loss-of-function allele of PID (14). The pin-like apices of the double mutants retain auxin responsiveness because auxin application leads to lateral organs with floral characteristics (fig. S1F).

Fig. 1 The lfy, pid, and double mutant phenotypes and imaging of auxin signaling and PIN localization using reporters in developing floral buds.

(A) Inflorescence apices of wild type (Ler-0) and the lfy-5 and pid-102 mutants show similar phyllotactic patterns. The double mutant pid-102 lfy-5 shows reduced apical dominance, reduced SAM activity, and increased branching. Scale bars, 1 mm. (B) The auxin efflux reporter PIN1-GFP (green) produced from pPIN1::PIN1-GFP and the DR5 auxin signaling reporter (red) produced from pDR5rev::3XVenus-N7 are shown in the plants of the indicated genotypes. See table S1 for statistical analysis of Venus-N7 intensity in wild-type and lfy-5 SAMs. (C) Effect of LFY overexpression on the Venus-N7 and PIN1-GFP signals. Shown are the shoot apices of pPIN1::PIN1-GFP pDR5rev::3XVenus-N7 35S::GVG-6XUAS::LFY plants. The first two images show the signals from the SAMs of a plant before and after 3 hours of mock (DMSO) treatment. The second pair of images shows the SAMs of a plant before and after 3 hours of dexamethasone (DEX) treatment. Images in (B) and (C) are representative of 30 or more observed SAMs. Scale bars, 50 μm (B and C).

LFY positively regulates auxin signaling

To better understand how loss of LFY function affects auxin pathways, we examined plants transgenic for both the PIN1 translational reporter pPIN1::PIN1-GFP and the synthetic auxin response reporter pDR5rev::3XVenus-N7 (12, 16, 19) in lfy-5 and lfy-5 pid-8 shoot apices. In wild-type plants, PIN1-GFP has a polar localization toward incipient primordia (and then subsequently reverses to aim away from primordia as they develop), and production of Venus-N7 from pDR5rev::3XVenus-N7 marks incipient and developing floral primordia (12). The polarity of PIN1-GFP appeared unaffected in lfy-5 SAMs, but the Venus-N7 signal was significantly reduced compared to wild type (Fig. 1B and table S1). In lfy-5 pid-8 double mutants, both PIN1-GFP and Venus-N7 signals were reduced, and PIN1 did not appear polarized (Fig. 1B). These observations and the observations in (16) indicated that reduction of LFY function resulted in reduced auxin signaling output in the shoot apex.

To explore this further, we induced LFY overproduction using a dexamethasone-inducible p35S::GVG-6XUAS::LFY transgene (20) and found that dexamethasone caused a phenotype of abnormal inflorescence structure, altered floral development, and altered phyllotaxis in transgenic plants (fig. S2) (11). In addition, LFY induction strongly increased the signal from the pDR5rev::3XVenus-N7 transgene at the sites of floral primordium formation (Fig. 1C). This activation can also occur ectopically because dexamethasone-treated p35S::GVG-6XUAS::LFY roots showed a significant increase in the pDR5rev::3XVenus-N7 signal (fig. S3). Consistent with these results, expression of the AUX/IAA genes IAA1, IAA17, and IAA29 was regulated by the LFY gene product (fig. S4, A and B). It is interesting to note that expression of IAA29 was significantly increased in both LFY-null (lfy) and LFY-overexpressing (p35S::GVG-6XUAS::LFY) plants. Genes that respond in a similar way to both LFY overexpression and lfy loss of function might be candidates to help elucidate why overproduction of LFY results in a phenotype similar to that of lfy loss-of-function mutants [(11), and this study].

LFY inhibits auxin biosynthesis

Reduced auxin signaling output in lfy-5 could result from suppressed auxin signaling or from reduced auxin biosynthesis, or a combination of both. To discriminate between these possibilities, we measured the concentration of free IAA in wild-type and lfy-6 shoot apices from which all but the earliest floral primordia had been dissected. The apices from lfy-6 had a higher concentration (Fig. 2A), which is inconsistent with the hypothesis that lfy mutants have reduced auxin biosynthesis. Overexpression of LFY with the dexamethasone-inducible p35S::GVG-6XUAS::LFY reduced the concentration of IAA (Fig. 2B). The IAA concentration in the Ler-0 plants used for the analysis in Fig. 2A and that in the control Ler-0 plants used in Fig. 2B were different, and this difference might be due to the different batches of plants used for sampling.

Fig. 2 Free IAA amount and expression of auxin biosynthesis genes.

(A) Free IAA quantified from 3 to 8 mg (fresh weight) of inflorescence apices from Ler-0 and lfy-6 plants (n = 5 for each genotype). (B) Free IAA quantified from 8 to 15 mg (fresh weight) of inflorescence apices from Ler-0 and p35S::GVG-6XUAS::LFY transgenic plants in the absence of stimulation, in the presence of DMSO (mock), or 24 hours after the addition of dexamethasone (DEX) (n = 5 for each genotype and condition; P = 0.0079, Wilcoxon test). Data in (A) and (B) are expressed as averages and SEs. (C) Expression of auxin biosynthetic genes YUC1 and YUC4 in plants of the indicated genotypes. (D) Expression of auxin biosynthetic genes TAA1 and TAR2 in plants of the indicated genotypes. Data in (C) and (D) are normalized to the expression of ACTIN2 and shown as the averages and SEs of three experiments with 15 to 20 dissected shoot apices in each experiment. *P = 0.028, Wilcoxon test. (E) Relative amount of LFY transcript in p35S::GVG-6XUAS::LFY shoot apices exposed to 20 μM DEX or DMSO [mock treatment, 0.05% (v/v)] for 4, 12, 24, or 48 hours. (F) Expression of YUC1 in p35S::GVG-6XUAS::LFY shoot apices exposed to 20 μM DEX or DMSO (mock treatment) for 4 or 12 hours. (G) Expression of YUC4 in p35S::GVG-6XUAS::LFY shoot apices exposed to 20 μM DEX or DMSO (mock treatment) for 4 or 12 hours. Data in (D) to (G) are normalized to ACTIN2 expression and shown as the averages and SEs of three experiments with 15 to 20 shoot apices in each experiment. (H) Yeast one-hybrid analysis of the binding of LFY to the YUC4 gene. Genomic regions are indicated relative to the A in the start codon ATG, which is set at +1. Data shown are averages and SEs of three experiments. (I) Abundance of the endogenous LFY transcript p35S::GVG-6XUAS::LFY in shoot apices exposed to 20 μM DEX or DMSO (mock treatment) for 4 or 12 hours. Data are normalized to ACTIN2 expression and shown as the averages and SEs of three experiments with 15 to 20 shoot apices in each experiment.

Furthermore, we examined the regulation of genes in the auxin biosynthesis pathway to see whether LFY regulates their expression in a manner consistent with inhibition of auxin biosynthesis. Consistent with the IAA analysis, examination of the expression of genes that function in the YUCCA pathway of auxin biosynthesis, YUC1, YUC4, TAA1, and TAR2, also suggested that LFY may inhibit this biosynthetic pathway in the shoot apex. Among the four genes tested, the expression of YUC4 and TAR2 was significantly increased in lfy-5 shoot apices compared to their expression in the shoot apices of wild-type Ler-0 plants (Fig. 2, C and D). With a dexamethasone-inducible p35S::GVG-6XUAS::LFY transgenic line (Fig. 2E), we found that the expression of YUC1 and YUC4 was reduced more than twofold within 4 hours of dexamethasone treatment (Fig. 2, F and G). Four hours of LFY induction did not reduce TAA1 or TAR2 expression, but TAR2 expression was reduced to 56% of that of the mock treatment after 12 hours of induction (fig. S4, C and D). In addition, LFY bound to the YUC4 genomic region [+1749 to 1998 base pairs (bp), A in the start codon ATG as +1], which suggests a direct role for LFY in regulating YUC4 expression (Fig. 2H). In addition to inhibiting auxin biosynthesis genes, LFY overexpression reduced endogenous LFY promoter activity (Fig. 2I). The data, therefore, favor the hypothesis that auxin signaling in flower primordia is reduced in lfy mutants, even in the presence of increased auxin concentration and auxin biosynthesis.

Auxin stimulates LFY expression

We analyzed the dynamics of LFY expression and auxin signaling by live imaging of the expression of pDR5rev::3XVenus-N7 and of pLFY::GFP-ER in developing shoot apices (Fig. 3, A and B, and fig. S5). The appearance of the Venus-N7 signal preceded LFY reporter signal in developing flower primordia, suggesting that auxin signaling increases before the onset of LFY expression. The LFY reporter signal was consistently reduced in lfy-5 pid-8, pin1-1, and pid-4 mutant backgrounds (Fig. 3, C to E), making it possible that auxin signaling stimulates LFY expression. To verify the ability of auxin to stimulate LFY expression, we tested the response of pLFY::GFP-ER in roots, where LFY is normally not expressed (Fig. 3F), and found that 24 hours after the addition of the auxin NAA (1-naphthylacetic acid), the LFY reporter was visible in root epidermal and pith cells (Fig. 3G). This induction of the LFY reporter was suppressed by the inclusion of cycloheximide, indicating that de novo protein synthesis was required (Fig. 3H). These data suggested a role for auxin in promoting, or derepressing, LFY transcription.

Fig. 3 Live imaging of a LFY reporter and a reporter of auxin signaling in developing floral buds.

The LFY reporter is pLFY::GFP-ER (green) and the auxin signaling reporter is pDR5rev::3XVenus-N7 (red). (A) LFY and DR5 reporters in Ler-0 SAM before dexamethasone (DEX) application (0 hours). (B) LFY and DR5 reporters in the same shoot apices imaged in (A), but 6 hours later. The right panels in (A) and (B) are the overlay of the left and middle panels. Numbers in (A) and (B) indicate the sites of primordia, with the larger numbers representing older primordia. Number 0 in the top panels indicates the sites of incipient primordia with the background intensity of LFY-GFP signal. Numbers are positioned at the primordial regions unless arrowheads are present, in which case the designated primordial sites are indicated by the arrowheads. (C) The LFY and DR5 reporters in lfy-5 pid-8 double mutant plants. (D) The LFY and DR5 reporters in pin1-1 plants. The LFY reporter is barely detectable. (E) Distribution of the LFY and DR5 reporters in pid-4 plants, which is a strong loss-of-function allele of PID. The LFY reporter is undetectable. (F) The root of a plant with pLFY::GFP-ER under control buffer conditions. The LFY reporter is undetectable in the root epidermal and pith cells. (G) The root of a plant with pLFY::GFP-ER after 24 hours of growth on 10 μM NAA–supplemented medium. LFY signal is detectable in the root epidermal and pith cells. (H) The root of a plant with pLFY::GFP-ER pretreated with 10 μM cycloheximide and, after 24 hours, grown on 10 μM NAA. The LFY reporter is undetectable in the root epidermal and pith cells. In (F) to (H), the LFY reporter is green; chlorophyll autofluorescence is red. Shown are the roots of 7-day-old seedlings. (I) Distribution of the LFY and DR5 reporters in Ler-0 SAMs. The LFY reporter is localized in floral primordia and developing floral buds. (J) Distribution of the LFY and DR5 reporters in lfy-5 SAMs. (K) Distribution of the LFY and DR5 reporters in pid-8 SAMs. Scale bars, 50 μm.

LFY inhibits its own expression

Consistent with the observation that LFY overexpression reduced endogenous LFY promoter activity (Fig. 2I), the LFY transcript was more abundant (fig. S4E) in the shoot apices of lfy-5 mutant plants. In addition, in comparison with that in wild-type plants (Fig. 3I), the intensity of the LFY reporter signal from pLFY::GFP-ER was greater in lfy-5 mutants (Fig. 3J), indicating a higher LFY promoter activity in a lfy loss-of-function background. Furthermore, ChIP-seq data indicate that LFY can bind to its own promoter region (13, 15). Thus, it appears that, during the floral transition, auxin may increase LFY transcription, and that LFY has a positive role in promoting auxin signaling, which may lead, in turn, to a further increase in LFY activity (Fig. 4). A counterbalancing effect is the negative influence of LFY activity on auxin biosynthesis through the YUCCA pathway and the repressive effect of LFY on its own promoter.

Fig. 4 A model of LFY involvement in controlling auxin signaling in SAMs.

Discussion

LFY is a master regulator that functions in switching on floral development. High-throughput techniques, including microarray, ChIP-seq, and ChIP-ChIP, have suggested the involvement of LFY in auxin regulation, a hormonal regulation system critical for multiple developmental programs (13, 15, 16). Here, we present genetic and live-imaging data indicating that LFY has a positive role in the auxin signaling pathway and may exert a negative function in the auxin biosynthesis pathway. Our data also indicated that during floral primordium formation, LFY expression occurred temporally after auxin signaling, and that LFY was activated by auxin. In addition, we found that LFY had a negative role in regulating its own expression. Together, these data suggest that LFY might be involved in one positive and two negative feedback loops in the control of the auxin biosynthesis and auxin signaling pathways (Fig. 4).

Additionally, we found that auxin signaling both controlled and was controlled by the transcriptional signals that regulate the appearance and early development of flowers. This interrelation of hormone function with a key transcriptional regulator raises the question of how frequently similar interactions may explain specific hormonal effects, and the question of to which degree the functions of transcriptional master regulators involve hormone systems. These studies also raise questions about the evolutionary history of this interaction between an ancient regulatory gene and an ancient hormone signaling pathway.

Materials and Methods

Plant materials

The weak lfy-5 allele used for genetic modifier screening was previously described (10). Other mutant alleles including lfy-6, pid-8, pid-4, and pin1-1 have also been reported (10, 18, 21). Marker lines including pPIN1::PIN1-GFP pDR5rev::3xVenus-N7 and pLFY::GFP-ER were previously reported (12). These markers carry antibiotic resistance, and plants were screened on half-Murashige and Skoog (MS) medium plates supplemented with kanamycin (a marker for pPIN1::PIN1-GFP; 50 mg/liter), Basta (for pDR5rev::3XVenus-N7; 25 mg/liter), or gentamicin (for pLFY::GFP-ER; 35 mg/liter). Genotyping with gene- or allele-specific primers (table S2) was performed when needed for construction or phenotypic analysis of multiple mutants. Plants were grown under constant light at 22°C.

Construction of vectors, transformation, and transformant analysis

For pLFY::LFY-eGFP construction, a 3.4-kb promoter fragment upstream of the start codon was amplified from genomic DNA and inserted into a pBJ36 shuttle vector. The LFY coding sequence was amplified from complementary DNA (cDNA) obtained from inflorescence shoots of wild-type Col-0 plants, and the enhanced green fluorescent protein (eGFP) coding sequence was added to its 3′ end by ligation polymerase chain reaction (PCR). The cLFY-eGFP fragment was then inserted into the pBJ36 shuttle vector downstream of the pLFY sequence (22). The finalized pLFY::cLFY-eGFP was then transferred into the pMLBART binary vector with Not I sites. Confirmed binary vector DNA was transformed into Ler-0 and lfy-6/+ by the floral dip method (23). Transformants were screened on MS plates supplemented with Basta (25 mg/liter), and lines that both showed expression patterns consistent with previously reported pLFY::GFP-ER or in situ hybridization results and that rescued the lfy-6 mutation were used for further experiments.

For p35S::GV-6XUAS::LFY construction, a LFY coding sequence with gateway-compatible ends was amplified from cDNA obtained from inflorescence shoots of wild-type Col-0 plants and then incorporated into the pTA7002 binary vector with LR reaction. Confirmed binary vector was transformed into wild-type Ler-0 plants and pPIN1::PIN1-GFP pDR5rev::3XVenus-N7 double marker lines. Transformants were screened with B5 medium supplemented with hygromycin (35 mg/liter).

Microscopy

Confocal microscopy was performed with a Zeiss LSM 510 microscope. For live imaging, developing floral buds were removed from plants and were grown in a wet box, with images taken periodically using a water-dipping lens (12). For observation of the effect of dexamethasone treatment, 20 μl of 20 μM dexamethasone in 0.05% dimethyl sulfoxide (DMSO) was applied to the shoot apex, and imaging was done at scheduled time points with the exactly same settings at each point. For mock treatment, 20 μl of 0.05% DMSO was applied. For observing fluorescence markers, shoot apices freshly detached from plants were used as previously described (24). When needed, FM4-64 staining to visualize cell membranes was performed by applying 20 μl of FM4-64 (5 mg/liter) in water onto the shoot apex with developing floral buds removed, with unbound dye washed away with distilled water after 5 min of staining at room temperature. Images were processed with Zeiss LSM 510 software. Dissecting microscopy was performed with a Zeiss Stemi SV11 dissecting microscope. Sizes were measured by imaging a standard ruler at the same setting as used for imaging plant organs.

For root experiments, seeds homozygous for the pLFY::GFP-ER (12) or pLFY::cLFY-eGFP transgene were sown on half-MS medium, and 7-day-old seedlings were transferred to half-MS supplemented with either a mock treatment solution (0.01% DMSO), 10 μM NAA, or 10 μM NAA + 10 μM cycloheximide. For cycloheximide treatment, seedlings were pretreated by flushing briefly with 10 μM cycloheximide and then left for 2 hours before being transferred onto plates. Imaging was performed after 24 hours of treatment.

Quantification of the Venus-N7 signals in representative images of the lfy-5 and wild-type SAMs was performed with ImageJ (http://rsbweb.nih.gov/ij/). Each raw data file of a stack of images was subjected to a Z-project function, with “MAX intensity” selected for image projection. Only the incipient primordial sites in wild-type and mutant SAMs were used for quantification. The primordial sites were enlarged with the “Square” then “Image-Zoom-To selection” functions. To measure the signal intensity, the inner portion of the nuclei, where Venus-N7 signal is located, was marked with the “circle” function, and the mean value of the signal intensity was obtained with the “Analyze-Histogram” function. The final readings of the mean signal intensity range from a minimum of 0 to a maximum of 4095. The background was subtracted with the Process-Subtract background function. Results from signal intensity quantification of five images of wild-type SAMs and six images of lfy-5 SAMs were analyzed. All images were taken with the same setting.

Quantitative reverse transcription PCR

Unless otherwise noted, plant materials used for quantitative reverse transcription PCR were young inflorescence apices with developing floral buds up to stage 4. Plant RNA extraction was performed with Qiagen RNeasy Kit following the manufacturer’s instructions. For reverse transcription, 1 μg of RNA was pretreated with Invitrogen RNA-free deoxyribonuclease and then subjected to reverse transcription with SuperScript II kit. Quantitative PCR was performed with a LightCycler 480 following the instruction manual. See table S2 for a list of primers used for quantitative PCR. For most experiments excepting those noted below, at least three biological replicates were included. For each biological replicate, three technical replicates were performed for quantitative PCR. Results were analyzed with an absolute quantitative approach with LightCycler 480 software. A standard curve test was performed for each gene-specific primer pair using series dilutions of control cDNA from Ler-0 shoot apices, and the concentration of samples was derived from the standard curve. The derived concentrations were normalized to the expression of ACTIN2 as an internal control. Results were subjected to Wilcoxon rank-sum test for equal medians with Matlab R2011B (MathWorks).

Free IAA quantification

For determining free IAA concentrations in shoot apices, shoot tips with young floral buds up to stage 3, from inflorescence shoots with length ranging between 2 and 5 cm, were used. Older floral buds and stem tissues were removed with a spring scissors (Fine Science Tools, cat. no. 91500-09). Five biological replicates were prepared for each of the Ler-0 and lfy-6 genotypes. For testing the effect LFY overexpression on auxin biosynthesis, the third generation of transgenic plants homozygous to the p35S::GVG-6XUAS::LFY transgene was used. Both Ler-0 and transgenic plants grown under the same conditions were subjected to no treatment, or either mock treatments [0.05% (v/v) DMSO], or dexamethasone [20 μM in 0.05% (v/v) DMSO] for 24 hours with a floral dipping approach. Sampling was similar to that previously described, and five independent replicates were collected for free IAA quantification. For each replicate, 3 to 15 mg of fresh tissue were weighed, frozen in liquid nitrogen, and stored at −80°C. For each sample, 20 μl of homogenization buffer (35% of 0.2 M imidazole and 65% isopropanol, pH 7.0) containing 0.2 ng of [13C6]IAA was added before homogenization. The amount of free IAA was analyzed by microscale solid-phase extraction followed by gas chromatography–selected reaction monitoring–mass spectrometry on a Thermo Trace GC Ultra coupled to a TSQ Vantage triple quadrupole MS system (Thermo Scientific) as previously described (25, 26).

Yeast one-hybrid assay

For yeast one-hybrid experiments, the YUC4 genomic regions used were selected on the basis of the presence of putative LFY binding sites according to the consensus LFY binding sequences as suggested by Winter et al. (13) and Moyroud et al. (15). These genomic regions (about 250 bp in length) were PCR-amplified and cloned into pLacZi bait vectors containing a LacZ reporter gene, and LFY cDNA was fused to the GAL4 activation domain (GAL4-AD) in a pDEST22 vector (Invitrogen). The two vectors were transformed into the yeast strain YM4271 (Clontech), and the DNA-protein interaction was determined by the quantification of β-galactosidase activity in triplicate experiments. In a parallel experiment, empty GAL4-AD vector combined with YUC4 promoter fragments in pLacZ was included as a negative control.

Supplementary Materials

www.sciencesignaling.org/cgi/content/full/6/270/ra23/DC1

Fig. S1. Phenotypic analysis of the pid-102 allele, lfy pid double mutant plants, and response of lfy-5 pid-8 shoot apex pin structures to exogenous auxin treatment.

Fig. S2. Effect of inducible overexpression of LFY on flower development.

Fig. S3. Effect of LFY overexpression on the Venus-N7 reporter of auxin signaling in roots.

Fig. S4. Relative expression of LFY, PID, IAA1, IAA17, IAA29, TAA1, and TAR2 genes in lfy-5 and dexamethasone-treated p35S::GVG-6XUAS::LFY plants.

Fig. S5. Live imaging of the LFY reporter and the auxin signaling reporter DR5 in SAMs of wild-type and lfy-5 plants.

Table S1. Venus-N7 signal intensity in primordial cell nuclei in wild-type and lfy-5 plants.

Table S2. List of primers used in this study.

References and Notes

Acknowledgments: We thank the members of the Meyerowitz laboratory for comments and discussion. We thank A. Roeder for technical assistance with mapping, Z. Nimchuk for technical assistance with construct preparation, V. Chickarmane for assistance in statistical analysis, M. Heisler for technical assistance with imaging, and A. Garda for plant care. We thank Y. Watanabe for the pTA7002 binary vector and The Arabidopsis Information Resource (TAIR) for essential genome information. Funding: Work at the California Institute of Technology was supported by the Department of Energy Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences grant DE-FG02-88ER13873 (E.M.M.). Work at the University of Minnesota was supported by the NSF (grants MCB-0725149, IOS-PGRP-0923960, IOS-PGRP-1238812, and MCB-1203438), the Minnesota Agricultural Experiment Station, and the Gordon and Margaret Bailey Endowment for Environmental Horticulture. Author contributions: W.L. and E.M.M. conceived the project; W.L., Y.Z., X.L., and P.Y. performed the experiments; and W.L. and E.M.M. wrote the manuscript. X.L. and J.D.C. developed the method for auxin analysis, and J.D.C. reviewed the mass spectral data and interpreted the analytical results. Competing interests: The authors declare that they have no competing interests. Data and materials availability: Requests for materials should be addressed to E.M.M.

Stay Connected to Science Signaling

Navigate This Article