RESEARCH PAPER

Gene profiling of the red light signalling pathways in roots

Maria Lia Molas1, John Z. Kiss1,* and Melanie J. Correll2

1 Department of Botany, Miami University, Oxford, OH 45056, USA2 Agricultural and Biological Engineering Department, University of Florida, Gainesville, FL 32611–0570, USA

Received 2 February 2006; Accepted 19 June 2006

Abstract

Red light, acting through the phytochromes, controls

numerous aspects of plant development. Many of the

signal transduction elements downstream of the phy-

tochromes have been identified in the aerial portions of

the plant; however, very few elements in red-light sig-

nalling have been identified specifically for roots. Gene

profiling studies using microarrays and quantitative

Real-Time PCR were performed to characterize gene

expression changes in roots of Arabidopsis seedlings

exposed to 1 h of red light. Several factors acting

downstream of phytochromes in red-light signalling in

roots were identified. Some of the genes found to be

differentially expressed in this study have already been

characterized in the red-light-signalling pathway for

whole plants. For example, PHYTOCHROME KINASE 1

(PKS1), LONG HYPOCOTYL 5 (HY5), EARLY FLOWER-

ING 4 (ELF4), and GIGANTEA (GI) were all significantly

up-regulated in roots of seedlings exposed to 1 h of red

light. The up-regulation of SUPPRESSOR OF PHYTO-

CHROME A RESPONSES 1 (SPA1) and CONSTITUTIVE

PHOTOMORPHOGENIC 1-like (COP1-like) genes sug-

gests that the PHYA-mediated pathway was attenuated

by red light. In addition, genes involved in lateral root

and root hair formation, root plastid development,

phenylpropanoid metabolism, and hormone signalling

were also regulated by exposure to red light. Interest-

ingly, members of the RPT2/NPH3 (ROOT PHOTOTRO-

PIC 2/NON PHOTOTROPIC HYPOCOTYL 3) family,

which have been shown to mediate blue-light-induced

phototropism, were also differentially regulated in

roots in red light. Therefore, these results suggest

that red and blue light pathways interact in roots of

seedlings and that many elements involved in red-

light-signalling found in the aerial portions of the plant

are differentially expressed in roots within 1 h of red

light exposure.

Key words: Arabidopsis, gene profiling, microarray, photomor-

phogenesis, red light, roots.

Introduction

Acquiring information about the surrounding environmentis crucial to the survival of all living organisms. For plants,light is one of the most important environmental signals foroptimal growth and survival. Seed germination, hypocotylgrowth and inhibition, cotyledon expansion, chloroplastdevelopment, time-to-flowering, and plant architecture areall light-regulated processes (for a review see Chen et al.,2004). Light can also regulate many aspects of root growthand development such as gravitropism (Lu and Feldman,1997; Oyama et al., 1997; Kiss, 2000), root hair formation(Oyama et al., 1997; De Simone et al., 2000a, b), orienta-tion and growth of lateral roots (Bhalerao et al., 2002;Kiss et al., 2002), primary root elongation (Lariguetet al., 2003; Correll and Kiss, 2005), negative (Okadaand Shimura, 1992) and positive phototropism (Ruppelet al., 2001; Kiss et al., 2003), root greening (Oyama et al.,1997; Usami et al., 2004), and secondary metabolite pro-duction (Hemm et al., 2004).

To detect and respond to the varying fluence, wave-length, and direction of light, plants have evolved severaltypes of photoreceptors. These molecules include the blue/UVA photoreceptors, i.e. the cryptochromes and photo-tropins, and the red/far-red photoreceptors, i.e. the phyto-chromes (Sharrock and Quail, 1989; Briggs et al., 2001;Lin and Shalitin, 2003). The phytochromes, designatedPHYA to PHYE in Arabidopsis, play important roles inregulating many of the light-induced responses. For red-light-induced responses, PHYB is the primary photorecep-tor in transducing the signal with other phytochromesplaying more minor roles. In roots, both PHYA and PHYBare involved in regulating light-induced gravitropism,

* To whom correspondence should be addressed. E-mail: kissjz@muohio.edu

Journal of Experimental Botany, Vol. 57, No. 12, pp. 3217–3229, 2006

doi:10.1093/jxb/erl086 Advance Access publication 14 August, 2006

ª The Author [2006]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved.For Permissions, please e-mail: journals.permissions@oxfordjournals.org

by on 18 October 2009 http://jxb.oxfordjournals.orgDownloaded from

positive red-light-induced phototropism, chloroplastdevelopment, secondary metabolite production, and roothair development (Feldman and Briggs, 1987; Johnsonet al., 1994; De Simone et al., 2000b; Kiss et al., 2003;Hemm et al., 2004). The roles of the other phytochromes,PHYC to PHYE, in light-regulated processes in roots arelargely unknown. In addition, very few of the downstreammolecules of the photoreceptors in the light-signallingcascade have been characterized for roots. Most of whatis known about light-signal transduction comes fromstudies using hypocotyls, stems, or leaves, even thoughvery different gene expression profiles have been foundbetween upper portions (hypocotyls and leaves) and rootsexposed to different light treatments (Jiao et al., 2005; Maet al., 2005). Therefore, roots provide a unique system tocharacterize light signalling in the whole plant.

Many signal transduction elements have recently beenidentified in red-light-induced responses. For example,once the light signal is detected by phytochromes, thesephotoreceptors can interact in the cytosol with proteins likePKS1 (PHYTOCHROME KINASE SUBSTRATE 1), apossible negative regulator of phytochrome-based re-sponses (Fankhauser et al., 1999), or they can be trans-located into the nucleus where they interact with otherproteins (Ni et al., 1998). In addition to PKS1, PHYBcan bind with ARR4 (ARABIDOPSIS RESPONSE REG-ULATOR 4), PIF3 (PHYTOCHROME INTERACTINGFACTOR3), PIF4, and COP1 (CONSTITUTIVE PHOTO-MORPHOGENESIS 1) to transduce the red-light signal(Sweere et al., 2001; Yang et al., 2001; Huq and Quail,2002). ARR4 binds preferentially to the active form ofPHYB (Pfr) and appears to stabilize this form by acting asa positive regulator of red-light-induced responses (Sweereet al., 2001). Although originally thought to be a positiveregulator (Halliday et al., 1999; Martinez-Garcia et al.,2000), PIF3 now appears to be a negative regulator inPHYB signalling, similar to PIF4 (Huq and Quail, 2002;Bauer et al., 2004; Kim et al., 2004; Park et al., 2004).COP1 is involved in the ubiquitin-dependent degradation ofPHYA in the light. In addition, COP1 interacts with PHYBin a yeast two-hybrid assay (Yang et al., 2001). A positivecorrelation between COP1 abundance and PHYB-mediatedresponses has also been reported (Boccalandro et al.,2004), although the precise roles of COP1 in PHYB-mediated signalling and protein turnover are unclear.CONSTANT LIKE 3 (COL3) is a COP1-interacting pro-tein acting downstream of phytochromes and COP1 in thered-light-signalling pathway (Datta et al., 2006). It seemsthat COL3 is a positive regulator of photomorphogenesisand can promote lateral root development in red light(Datta et al., 2006).

HY5, a bZIP (BASIC REGION-LEUCINE ZIPPER)transcription factor, acts downstream of phytochromes inlight-signalling pathways and is a positive regulator ofphotomorphogenic responses. HY5 is targeted for degra-

dation to the COP9 signalosome through COP1 in both thedark and light (Bauer et al., 2004). In roots, HY5 isinvolved in the inhibition of both root hair and primary rootelongation, inhibition of lateral root development, and thepromotion of gravitropic orientation, secondary wall thick-ening, the phenylpropanoid pathway, and chloroplastdevelopment (Oyama et al., 1997; Hemm et al., 2004).Two other elements in root development and light-regulated responses include the ATP-binding cassetteproteins, MDR1 (MULTIDRUG RESISTANCE 1) andPGP1 (PLEIOTROPIC DRUG RESISTANCE 1). Theseproteins are involved in both photomorphogenesis andauxin-based responses in roots such as elongation andgravitropism (Lin and Wang, 2005). Recently, a gene in-volved in phytochrome-regulated gravitropism in hypoco-tyls was identified, GIL1 (GRAVITROPIC IN THE LIGHT1); however, the role that this plays in root responses isunknown (Allen et al., 2006). Interestingly, it appears thatMDR1 acts upstream of PHYA in attenuating theseresponses. Other light-signalling proteins have also beenidentified in roots such as PKS1, LAF6, (LONG AFTERFAR-RED 6; Lariguet et al., 2003; Moller et al., 2001),although the roles of these proteins in the light-regulatedprocesses of roots are largely unknown.

Gene expression profiling using microarrays has shownthat light regulates more than 30% of the genome inseedlings of Arabidopsis, including more than 26 cellularpathways (Ma et al., 2001). In whole seedlings and leaves,microarray analyses have identified many of the genesinvolved in red- and far-red-light signalling (Teppermanet al., 2001, 2004; Wang et al., 2002; Jiao et al., 2005; Maet al., 2005). For whole seedlings, as the duration of lighttreatment increased, a greater number of genes was dif-ferentially expressed. For example, only 1.7%of the genomewas differentially expressed in seedlings after 1 h of redlight treatment, but 11% of the genome was differentiallyexpressed after 24 h of red light treatment (Teppermanet al., 2004). Therefore, many more downstream elementsare regulated after a longer duration of light treatment.

Although there have been large numbers of microarraystudies comparing light-regulated gene expression patternsin whole seedlings, leaves, and hypocotyls, there are onlya few studies that have analysed the gene expressionchanges in roots with different light treatments (Ma et al.,2001, 2005; Tepperman et al., 2001, 2004; Wang et al.,2002). One study compared the gene expression profiles ofroots (18-d-old seedlings) that were exposed to 4 h of far-red light from dark-adapted (4 d dark-adapted) plants toroots from dark-adapted plants (Sato-Nara et al., 2004).Surprisingly, no genes (out of 7000) were significantlyregulated after the 4 h of far-red light treatment (Sato-Naraet al., 2004). However, these authors used microarrays withonly 7000 elements represented and had a conservative 3-fold threshold to define significant regulation (Sato-Naraet al., 2004). Therefore, using arrays with more elements or

3218 Molas et al.

by on 18 October 2009 http://jxb.oxfordjournals.orgDownloaded from

lower thresholds for significance may identify moresignificantly regulated genes in response to far-red light.

Another recent study compared the gene expressionbetween roots from dark- and white-light-grown seedlingsof both Arabidopsis and rice (Jiao et al., 2005; Ma et al.,2005). These authors found that approximately 40% of thegenome was expressed in Arabidopsis roots, and thatapproximately 3.5% of the total genome was differentiallyexpressed in roots from dark- versus light-grown seedlingsof Arabidopsis (>2-fold change; Jiao et al., 2005; Maet al., 2005). Of these genes, only a small portionoverlapped with light-regulated genes from hypocotylsand cotyledons, suggesting that roots have very differentlight-regulated processes (Ma et al., 2005).

Despite these recent studies, limited information aboutthe early light-signalling events in roots is available. In thispaper, the gene expression profiles of roots from dark-grown seedlings exposed to 1 h of red light were comparedwith controls. Through microarray studies with AffymetrixGenechips�, the differential expression of genes involvedin the phenylpropanoid pathway, root plastid development,auxin and ethylene signalling, lateral root development, andtranscription have been identified. In addition, the up-regulation of some of the elements in blue-light-signallingpathways was identified with the red-light treatment usedhere, suggesting an interaction of pathways induced by redand blue light.

Materials and methods

Plant material and growth condition

Seeds of Arabidopsis thaliana ecotype Landsberg erecta (Ler) weresterilized in 70% (v/v) ethanol for 5 min, two rinses of 90% (v/v)ethanol and four rinses in sterile ddH2O. Seeds were sown ontoa presterilized cellophane that was placed on top of an agar growthmedium (Kiss and Swatzell, 1996) consisting of ½ MS medium with1% (w/v) sucrose, 1.2% (w/v) agar in square (100 mm315 mm) Petridishes. Seeds were stratified for 2 d at 4 8C in the dark and thenexposed to white light for 2 h at room temperature to synchron-ize germination. Seedlings were grown for 7 d in the dark at 21 8Cbefore being transferred to either red light (1 h) or continueddarkness. Seedlings were collected in RNAlater� (Ambion), androots were excised and stored at �80 8C. Samples were pooledfrom several plates for each biological replicate. Three biological re-plicates were used for microarray analysis.Red light was obtained by passing light from fluorescent bulbs

through Plexiglas filters (Rhom and Hass No. 2423, Dayton Plastics,Columbus, OH). Fluence rate through the red filter was 12–14 lmolm�2 s�1 with a transmission maximum of 630 nm. The fluence ratewas measured with a Li-Cor LI-189 Quantum Radiometer Photo-meter equipped with a LI-190SA Quantum sensor.

Microarray procedures

Preparation of labelled copy RNA: Total RNA was extractedfrom each sample and prepared for hybridization according to theAffymetrix GeneChip� Expression Analysis Technical Manual.Briefly, RNA was extracted from frozen tissue using the RNeasy�

Mini Kit (Qiagen Inc, Valencia, Ca), and residual DNA was removed

by performing an on-column digestion using a DNA-free kit(Ambion).

Target labelling and array hybridization: Total RNA samples weresubmitted to the University of Florida’s Interdisciplinary Center forBiotechnology Research (ICBR) Gene Expression Core Facility(Gainesville, FL). The quality of each of the RNA samples wasdetermined by evaluating the relative amounts of 28S and 18Sribosomal peaks using a Bioanalyzer (Agilent Technologies, PaloAlto, CA). Five lg of total RNA was used as a template forcomplementary RNA (cRNA) synthesis with the GeneChip� One-Cycle Target Labeling Kit (Affymetrix, Santa Clara, CA). First strandsynthesis was primed with a T7-(dT)24 oligonucleotide primercontaining a T7 RNA polymerase promoter sequence on the 59end. Second strand products were cleaned and used as a template forin vitro transcription (IVT) with biotin-labelled nucleotides. IVTreactions were cleaned, and 20 lg of the product was heated at 94 8Cfor 35 min in fragmentation buffer provided with the labelling kit inorder to produce fragments that are 35–200 base pairs in length. A15 lg aliquot of fragmented cRNA was hybridized for 16 h at 45 8Cto an Affymetrix GeneChip� ATH1 genome array. After hybridiza-tion, each array was stained with a streptavidin–phycoerythrinconjugate, washed (Molecular Probes, Eugene, Oregon) and visual-ized with a GeneChip� Scanner 3000 (Affymetrix, Santa Clara, CA).Images were inspected visually for hybridization artefacts. Qualityassessment metrics were generated for each scanned image. Samplesthat did not pass quality assessment were eliminated from furtheranalyses. All expression data was submitted to GEO (GeneExpression Omnibus) database (http://www.ncbi.nlm.nih.gov/geo/)under the series accession number GSE4933.

Generation of expression values and data analysis: MicroarraySuite Version 5 software (Affymetrix, Santa Clara, CA), was used toconvert intensity data into quantitative estimates of gene expression.All expression values were globally scaled to 500. A probabilitystatistic associated with each gene’s presence or absence was alsogenerated. Genes not expressed in any of the samples were con-sidered absent. Absent genes were removed from the data set andnot included in further analyses. The natural log-transformed expres-sion values were subjected to an analysis of variance (ANOVA) fora complete block design (Steel and Torrie, 1960) where the differenttreatment times served as blocks. P-values were calculated usingTukey HSD test. Statistical analyses were performed with AnalyseItTools, software developed by ICBR.

Real time RT-PCR

Total RNA was extracted from roots as described above, and cDNAwas synthesized according to SuperScript II RNaseH� kit protocol(Invitrogen) on 500 ng of RNA and oligo dT oligonuclotide asa primer. PCR primers were designed using OligoAnalizer 3.00(http://biotools.idtdna/biotools) to create amplicons 100–180 bp inlength. Primers used had the following characteristics: meltingtemperature between 55–60 8C, 40–60% G/C content, 39 end contentoverall and matching only desired tentative consensus sequence inThe Arabidopsis Information Resource (TAIR). Actin8 was used ashousekeeping control and the primers were: F_ACTIN8 CTTTCC-GGTTACAGCGTTTG, R_GAAACGCGGATTAGTGCCT; F_ RPT2TGCCAAGTTCTGTTACGGTG, R_ACAACGGCGGACAATAC-CAA; F_SPA1 TTGTCCAGAGGAGATAAATG, R_AGGATGTA-GAAGCCAAGAC; F_PKS1 TGCGCCAAGTGAAGTAAGCGT, R_CGGTTGTGTCTTGTTCATGCTGC; F_COP1 FAMILY AT5g52250TACGACGTGGAGAAACAAGTGCC, R_AACGGCTATGGAA-GATCCACCG; F_GI AACCATCTTCTGTGGGGACT, R_AGAA-CCCTGCGAGTCTATCA.Real-Time PCR was performed using Rotor Gene RG 3000

(Corbett Research) and the experimental conditions were: activation

Gene profiling of the red light signalling pathways in roots 3219

by on 18 October 2009 http://jxb.oxfordjournals.orgDownloaded from

of Taq 95 8C for 15 min, denaturing 40 cycles at 95 8C for 10 s, andannealing and extension 56–60 8C for 1 min. The following reagentswere combined with each sample and control: 20 ll of SYBR GreenPCR master mix (Qiagen), 2.5 ll of primers (0.2 lM) and 3 ll oftemplate. Data were analysed using Rotor Gene 6.0 software andgene expression data were calculated using Standard Curve methods(Livak, 1997).

Results

Overall effects of red light on gene expression in roots

A total of 661 genes were significantly differentiallyexpressed in roots of 7-d-old seedlings exposed to 1 h ofred light compared with dark-grown controls (P <0.05). Ofthese, 351 genes were regulated at least 2-fold (log2 ratioof red/dark signal>1) with 128 that were induced, and 223that were repressed (Fig. 1; see supplementary Table 1 atJXB online). This corresponds to ;1.4% of the genome asbeing differentially expressed within 1 h or red lighttreatment in roots of Arabidopsis seedlings. A preliminaryoverview using the Functional Catalog from the MunichInformation Center for Protein Sequences (MIPS) classifiedthe majority of genes regulated in response to red light inthe metabolism category (11 induced and 21 repressed,Fig. 2) with only a smaller portion in the transcription cate-gory (8 induced and 5 repressed). However, by groupinggenes in their putative functional categories, the expres-sion changes in 21 induced (>2-fold ) transcription factorsand 17 repressed transcription factors (see supplementaryTable 2 at JXB online) were found. The largest group oftranscription factors, with nine members, was from the zincfinger family where six were up-regulated and 3 threedown-regulated (see supplementary Table 2 at JXB online).

Other elements that were significantly differentiallyexpressed (>2-fold ) were classified in the followingcategories: photomorphogenesis (Table 1), transcriptionalregulation, protein turnover, phenylpropanoid metabol-ism, root growth, chloroplast and light harvesting, cell walldevelopment, hormone signalling, transporters, and trans-cription (see supplementary Table 2 at JXB online).In addition to these genes, 87 of the 351 differentiallyexpressed genes were classified as hypothetical or ex-pressed (see supplementary Table 1 at JXB online).

PHYA-mediated signal transduction pathway isrepressed and PHYB-mediated pathway is induced

The red-light signalling pathway in plants is primarilymediated by PHYB with other phytochromes playing lesserroles in regulating the response. In this study, we foundthat some genes involved in either PHYA-mediated (i.e.

Fig. 1. Number of genes defined as responding to 1 h of red light inroots of Arabidopsis seedlings. Venn diagrams show the percentage (%)and number of genes [] regulated by red light (left) and those genesthat are up- and down-regulated >2-fold (right).

Fig. 2. Overview of induced and repressed biological processes in roots of dark-grown seedlings exposed to 1 h of red light. Classification of genes(at least 2-fold differential regulation) was based on the Functional Catalogue of the Munich Information Center for Protein Sequences (MIPS). Genesof unknown function or classification are not shown. Numbers of genes found in each category are identified on the x-axis.

3220 Molas et al.

by on 18 October 2009 http://jxb.oxfordjournals.orgDownloaded from

SPA1, SUPPRESSOR OF PHYTOCHROME A 1) orPHYB-mediated-signalling (i.e. ELF4, EARLY FLOWER-ING 4 and GI, GIGANTEA) pathways were significantlyregulated as well as some genes involved in both path-ways (e.g. HY5, PKS1; Table 1).

SPA1 is a nuclear-localized repressor of PHYA-mediatedsignalling (Hoecker et al., 1999) and was significantly up-regulated in roots exposed to 1 h of red light. In addition,a member of the COP1 family (At5g52250; COP1-like) wasalso significantly induced. COP1 has been shown to beinvolved in suppressing PHYA-mediated responses (Saijoet al., 2003; Seo et al., 2004).

Elements in the PHYB-mediated pathway were alsoinduced in response to red light. ELF4 and GI are spe-cifically implicated in PHYB-regulated responses, and thegenes encoding these proteins were both induced (GI 2-fold in qRT-PCR treatment only). In addition, two light-inducible genes related to both PHYA- and PHYB-signaltrandsuction, PKS1 and HY5, were up-regulated.

Overlap in blue and red-light-signal transductionpathways

Blue light induces a wide range of physiological responsesincluding phototropism, stomatal opening, and chloroplastmovement. Recent molecular genetic studies have shownthat PHOT1 (phototropin 1) and PHOT2 function as pho-toreceptors for phototropism (Briggs et al., 2001), andRPT2 and NPH3 transduce signals downstream of photo-tropins to induce the phototropic response (Inada et al.,2004).

Interestingly, three elements involved in the phototropinor UVA/blue-light-signalling pathway were found signifi-cantly differentially expressed in roots following red-lighttreatment. These included RPT2/NPH3-family genes, twoof which were up-regulated (NPH3-like At5g48800 andRPT2) and the other down-regulated (another NPH3-likeAt1g03010). These genes belong to the novel NPH3/RPT2

family, which is intimately involved in blue-light signal-ling pathways, including phototropism (Motchoulski andLiscum, 1999; Liscum and Stowe-Evans, 2000).

Chloroplast genes are differentially regulated

Chloroplast development in roots exposed to light has beenwell-documented for a variety of species. In Arabidopsis,red light can promote, although less effectively than blueor white light, chloroplast development and greening inroot tissues (Usami et al., 2004). Eight genes were foundup-regulated (see supplementary Table 2 at JXB online)associated with the chloroplast or in light harvestingelements including HY5, CBL-INTERACTING PROTEINKINASE 13 (CIPK13); PHOTOSYSTEM II OXYGEN-EVOLVING COMPLEX 2 (PSBO2); PROTEIN SA(PSAB); PsaC subunit of photosystem I (PASC), and sixgenes down-regulated including a DIFFERENTIATIONAND GREENING (DAG)-similar (At1g72530), a Psb sub-unit of photosystem II (PSBP family, At4g15510) and anoxidoreductase (At4g10500). Some of these differentiallyexpressed elements, although not directly involved in lightharvesting or chloroplast development, are known to belocalized in the chloroplast (see supplementary Table 2 atJXB online).

CIPK13 is a protein located in the chloroplast andinvolved in protein amino acid phosphorylation and signaltransduction (Harmon et al., 2000; Sanders et al., 2002).PSBO2, a protein that localizes in the thylakoid, is anextrinsic subunit of photosystem II and has been proposedto play a central role in stabilization of the catalytic man-ganese cluster. PSAB, the gene of which is part of thechloroplast genome, encodes the D1 subunit of photosys-tem I and II reaction centres and it is involved in lightharvesting and photosynthesis (Klein et al., 1988). PSBPencodes a 23 kDa extrinsic protein that is part of photosys-tem II and participates in the regulation of oxygen evolu-tion. DAG is a plastid developmental protein required for

Table 1. Differentially regulated expression (>2-fold) of genes involved in photomorphogenesis in roots of 7-d-old dark-grownseedlings exposed to 1 h (;12 lmol m�2 s�1) of red light

AGI no.a Description Gene name Log2 ratiob

PhotomorphogenesisAt5g11260 bZIP transcription factor HY5 3.8At5g48800 BTB/POZ domain and coiled-coil domain NPH3-family 1.0At2g46340 Coiled-coil domain SPA1 2.7At2g02950 Soluble protein involved in red, far-red phototransduction PKS1 1.9At2g30520 BTB/POZ domain and coiled-coil domain RPT2 1.7At5g52250 WD-40 repeat family protein PnCOP1 similar 1.2At2g40080 Circadian rhythm regulator ELF4 2.3At1g03010 BTB/POZ domain and coiled-coil domain NPH3 family �1.0At3g46240 Similar to light repressor receptor protein kinase LRRRK �2.0At1g 22770c Gigantea protein GI 0.8

a Arabidopsis Genome Initiative number.b log2 ratio (red-light/dark-grown roots).c log2 ratio <1.0 but 1.0 for quantitative PCR (see Fig. 3).

Gene profiling of the red light signalling pathways in roots 3221

by on 18 October 2009 http://jxb.oxfordjournals.orgDownloaded from

chloroplast differentiation (Chatterjee et al., 1996). Overall,differential regulation of the aforementioned genes add tothe growing body of evidence reporting root greening andindicate that this process starts within 1 h of red light.

Phenylpropanoid metabolism is induced

In plants, large amounts of carbon from aromatic aminoacid metabolism are diverted into the biosynthesis ofnatural products based on a phenylpropane skeleton. Thesediverse phenylpropanoid compounds, which include flavo-noids, lignin, coumarins, and many small phenolic mole-cules, have a multiplicity of functions in structural support,pigmentation, defence, and signalling. The phenylpro-panoid metabolism was induced by red light in this study(see supplementary Table 2 at JXB online) with the up-regulation of key genes in phenylpropanoid metabolismi.e. PAL3 (PHENYLALANINE AMMONIA LYASE 3;At5g04230), two CAD class III genes (CINNAMYL AL-COHOL DEHYDROGENASE; At2g2170 and At2g21890), and COMT-like II genes (CAFFEIC ACID O-METHYLTRANSFERASEs, At3g53140/At5g37170).

ABC transporters are differentially regulated

Six ABC (ATP-Binding Cassette) transporters differen-tially regulated by red light were found (see supplementary

Table 2 at JXB online). Two genes that encode membersof the non-intrinsic ABC proteins (NAPs) family, NAP2(At5g44110) and NAP9 (At5g02270) and a gene encodinga white-brown complex homologue (WBCs) family mem-ber WBC4 (At4g25750) were up-regulated. In addition,one member of the multidrug resistance associated protein(MRP) family, MDR7 (At5g46540) and two pleiotropicdrug resistance (PDR) subfamily (PDR12 and PDR5-At1g15520 and At2g37280, respectively) genes, weredown-regulated (see supplementary Table 2 at JXB online).NAP2 and NAP9 were highly induced, with NAP2being among the highest induced genes in this study(NAP2/POP1).

Overlap in gene expression from red- andwhite-light-induced pathways

A comparison of these results from dark-grown rootsexposed to 1 h of red light (7-d-old seedlings) with studiesthat compared roots from dark- and light-grown seedlings(6-d-old seedlings) indicated that only 26 out of the 351differentially expressed genes (>2-fold ) from the 1 h red-light studies overlapped with 883 of differentially expre-ssed genes from roots grown in continuous white light(Table 2; Ma et al., 2005). The differences in gene re-gulation between these two studies may be associated with

Table 2. Comparison of gene expression profiles from roots grown in continuous white light with roots of seedlings exposed to 1 h ofred light relative to dark-grown roots

AGI no. Description Gene name Continuous white lighta 1 h red light

Up-regulated in bothAt5g11260 bZip transcription factor HY5 4.1b 3.8At5g52250 Transducin family PnCOP1-like 2.9 1.2At2g40080 Expressed (circadian rhythm associated) ELF4 2.3 2.3At5g48880 Acetyl-coA-acyl transferase PKT1 1.9 2.4At3g21560 UDP-glucosyltransferase 2.4 1.2At5g44110 ABC transporter POP1; NAP2 3.8 4.4At5g02270 ABC transporter NBD-like NAP9 (POP like) 2.6 2.3At3g22830 Heat Shock Protein Transcription Factor glycine rich Similar to GRP2 1.1 2.3At2g46340 Suppressor of phytochrome responses SPA1 1.7 2.7At1g43160 AP2 RAP2.6 2.0 1At5g54470 Zn Finger (B-box) 1.1 3.1At3g45160 Expressed 4.3 3.2

Down-regulated in bothAt4g27150 Seed storage protein albumin NWMU2-2s �2.4 �2.2At3g56400 WRKY transcription factor-Group III WRKY70 �1.6 �1.0At4g09610 Gibberellin related GASA2 �1.5 �4.1At5g04370 S-adenosyl L-methionine carboxyl methyl transferase SAMT-similar �2.0 �1.5At4g10500 Oxidoreductase 20G-Fe (II) oxygenase �1.7 �1.7At3g11340 UDP-glucoronosyl/UDP-glucosyltransferase �1.5 �1.4At5g43770 Proline-rich family extension domain �1.5 �2.5At5g45500 Similar to resistance complex protein Expressed �1.5 �5.7At3g07600 Heavy metal associated domain FP4-similar �1.3 �1.2At5g45070 Disease resistant protein TIR class PP2-A8 �1.97 �2.8

Differentially expressedAt1g58270 Meptrin TRAF similar to ubiquitin-specific protease ZW9 1.4 �2.4At4g13290 Cytochrome P450 CYP71A19 3.2 �1.4At5g38000 NADP oxidoreductase putative 2.5 �2.9At5g06980 Expressed �1.7 1.4

a Gene expression results are from seedlings grown in continuous white-light (150 lmol m�2 s�1) for 6 d compared to roots from seedlings grown inthe dark. Results are from 70 mer ;26 k oligo custom arrays (Ma et al., 2005).

b HY5 was marked as absent (Ma et al., 2005).

3222 Molas et al.

by on 18 October 2009 http://jxb.oxfordjournals.orgDownloaded from

the different ecotypes (Columbia versus Landsberg), dif-ferent plant ages (6- versus 7-d-old roots), different lightfluence rates (150 versus 14 lmol m�2 s�1) and quality(white versus red light), the duration of illumination(continuous versus 1 h), and the factors associated withmicroarray platforms (custom slide arrays versus Affyme-trix Genechips�). In addition, due to the sensitivity oflight-induced genes even to dim-green light, roots from thedark-grown seedlings were collected in absolute darknessusing RNAlater (Ambion). The method used to collect 6-d-old dark root samples from the previous study was notdescribed (Ma et al., 2005). Due to all of the differencesbetween red-light and continuous-white-light studies withroots, it is not surprising that there were only a few genesoverlapping in the expression profiles (Table 2).

Of the 26 overlapping genes from roots in the red-lightand continuous white light treatments, 12 were up-regulatedin both treatments, 10 were down-regulated in both treat-ments, and four were differentially regulated between thetwo treatments relative to dark controls (Table 2). HY5,PnCOP1-like, SPA1, RAP2.6 (RAN-BINDING 2.6) and twoABC transporter proteins (POP; POP1), all light-regulatedgenes in whole seedlings (Tepperman et al., 2004), wereinduced under both conditions in roots (Table 2). Becausethese genes were up-regulated after 6 d of white-lighttreatment, it appears that these genes may remain inducedin roots even after a long duration of light treatment.Repressed genes from the two light conditions includeda gibberellin-regulated protein, GA-STIMULATED TRAN-SCRIPT 2 (GASA2; At4g09610), a gene encoding a heavymetal associated protein (At3g07600), and a gene encod-ings a disease resistant protein (At5g45070).

Confirmation of the results with Real-Time PCR

Quantitative Real-Time PCR (qRT-PCR) was used toconfirm the level of expression of six genes associatedwith photomorphogenesis (HY5, SPA1, COP1-like-At5g52250, PKS1, RPT2, and GI) whose expression wasshown to be differentially regulated in red light usingmicroarrays (Fig. 3). ForGI, the gene expression was foundonly to be 1.7-fold increased (although significant P <0.05;Table 1) with the microarrays. Because this gene is knownto be involved in PHYB-mediated responses, it was decidedto check expression levels using qRT-PCR and found theexpression difference to be greater than 2-fold (Fig. 3).Using the Standard Curve Method (Livak, 1997), it wasfound that all six genes were induced within 1 h of red-lighttreatment (>2-fold ). Thus, in general, the qRT-PCR resultsconfirm the microarray results. Interestingly, the RT-PCRmethod showed a greater level of gene expression than wasfound with the microarray studies for almost all the genesanalysed, and this is probably due to the enhanced sensi-tivity of RT-PCR techniques. For example, HY5 showeda 14-fold increase using microarrays whereas qRT-PCR

detected 21-fold-induction in red light compared with thedark-grown controls. Similar trends, although to a lesserextent, were found for RPT2, PKS1, and SPA1 whereasCOP1-like showed little differences in expression usingboth methods. The qRT-PCR, along with microarrayresults, indicates that the red-light signalling cascade isbeing activated in roots of seedlings within 1 h of red-lighttreatment.

Discussion

In this study, we show that 1.4% of the genome issignificantly differentially regulated (>2-fold ; P <0.05)in roots of Arabidopsis seedlings within 1 h of red-lighttreatment compared with control seedlings grown in con-tinuous dark. Many of the genes identified to be differen-tially regulated are involved in photomorphogenesis, lateraland root hair development, phototropism, root greening,and phenylpropanoid metabolism (Table 1; see supplemen-tary Table 2 at JXB online). These results provide an insightinto the genetic mechanisms plants use to co-ordinate themany responses of roots to red light.

Red light suppresses PHYA-mediated signalling whileinducing both PHYB-mediated and blue/UVApathways

Suppression of PHYA-mediated signal transduction: In thisstudy, up-regulation of PKS1 was found in roots within 1 hof red-light exposure. This may act as a possible me-chanism to inhibit PHYA responses in red light. PKS1 actsas a negative regulator of phytochromes, particularly

Fig. 3. Relative transcript abundance changes after 1 h continuous redlight were analysed by microarray and real-time PCR using SYBR greenfor detection and the Standard Curve Method. Microarray valuesrepresent the mean ratio, and the qRT-PCR data are shown as an averageof three independent biological replicates. (The asterisk indicates 1.74-fold induction for microarray and not the >2-fold as found with theother genes).

Gene profiling of the red light signalling pathways in roots 3223

by on 18 October 2009 http://jxb.oxfordjournals.orgDownloaded from

PHYA, by preventing nuclear import of the phytochromeupon light activation (Fankhauser et al., 1999). Althoughnot found with gene expression data (Ma et al., 2005),PKS1 expression remained high in roots after several daysin the light and localized to the region of root elongation(Lariguet et al., 2003).

In addition to PKS1, SPA1 and COP1-like (At5g52250)genes were found to be significantly up-regulated in redlight. In the light, SPA1 interacts with COP1, and thiscomplex can target HY5 for degradation, thereby attenuat-ing PHYA-mediated signalling (Saijo et al., 2003). COP1,a member of the E3 ubiquitin ligase family, is involved intargeting a variety of proteins for degradation includingPHYA, HY5, and LAF1 (LONG AFTER FAR RED 1; fora review see Chen et al., 2004). The specific role of theCOP1-like gene that was up-regulated in these studies isunknown. However, it is known that the COP/DET/FUSgroup of proteins regulates HY5 activity at the level ofprotein stability (Oyama et al., 1997; Osterlund et al.,2000). Overall, up-regulation of PKS1, SPA1, and COP1-like genes suggests that the PHYA-mediated signallingpathway is repressed within 1 h of red-light exposure inroots of seedlings (Fig. 4).

SPA1 and PKS1were also up-regulated in 1 h of red-lighttreatment in whole seedlings, and the expression levels ofboth genes decreased as the duration of light exposureincreased (Tepperman et al., 2004). Therefore, it appearsthat PKS1 and SPA1 are similarly regulated in both rootsand whole seedlings after 1 h of red-light exposure.Interestingly, these genes had only slightly reduced expres-sion levels in whole seedlings from phyB mutants com-

pared withWT seedlings, suggesting that other phytochromesare involved in regulating these genes in response to redlight (Tepperman et al., 2004). Studies comparing ex-pression levels of these genes in other phytochrome mu-tants may help to identify which specific phytochromesare involved in regulating the expression of these genes.

Induction of PHYB-mediated signal transduction: Severalelements in the PHYB-mediated signalling pathway wereup-regulated within 1 h of red-light exposure in roots,including HY5, ELF4, PKS1, and GI. Although the GIexpression level was less than 2-fold for our microarrayanalysis (1.74-fold; P=0.0037), it was greater than 2-foldfor our RT-PCR analysis (Fig. 3). Both ELF4 and GI actdownstream of PHYB and are involved in circadian clockregulation and photoperiodism (Huq et al., 2000; Khannaet al., 2003). Up-regulation of these circadian clock-associated genes may be a result of a response or processesthat occur distal to the roots. For example, genes involvedin flowering can be expressed throughout the whole plant,including the roots (Wilson et al., 2005).

In addition to its role in circadian rhythms, GI is involvedin oxidative stress tolerance, cold stress responses, andcarbohydrate metabolism (Eimert et al., 1995; Kurepaet al., 1998; Fowler et al., 1999; Cao et al., 2005), andELF4 is involved in seedling de-etiolaton (Khannaet al., 2003). Consequently, the possibility cannot beruled out that these genes may regulate other non-circadianfunctions in roots.

Interestingly, ELF4was up-regulated at the same level inboth 1 h of red-light treatment and in continuous white-lighttreatment in roots (Table 2). For whole seedlings, ELF4 andGI were early-induced genes in red light with expressionlevels returning approximately to the starting levels after24 h of red-light exposure (Tepperman et al., 2004). Thesetime-course studies indicated that ELF4 expression in WTseedlings had a brief minor peak 1 h after the dark-grownseedlings were exposed to continuous red light and abroader, second peak at approximately 9 h, returning tobasal levels at 24 h of red-light treatment (Kikis et al.,2005). Therefore, it appears that roots regulate ELF4expression in a manner similar to the regulation found inwhole seedlings for the 1 h time point. However, for longer-term light exposure, the expression of ELF4 appears tobe different between roots and whole seedlings. Not sur-prisingly, the expression levels of ELF4 and GI werereduced in phyB mutants compared with the wild type(Tepperman et al., 2004), suggesting that PHYB plays asignificant role in the expression of these genes.

As described previously, it was found that a COP1-likegene (At5g52250) was significantly induced in roots ofseedlings exposed to red light. In addition to the role ofCOP1 in attenuating PHYA-mediated signalling, COP1may also be involved in regulating PHYB-mediated signal-ling (Boccalandro et al., 2004). These authors suggest that

Fig. 4. Summary diagram of the elements founded in this study asregulated by red-light in Arabidopsis roots. Processes indicated by arrowsare based on the current literature. Red light converts Pr forms in Pfrforms. PfrA and PfrB interact with PKS1 in the cytoplasm. Upon lightactivation, PfrA-E migrates to the nucleus. In the light, COP1 interactswith SPA1 and targeted HY5 for degradation. PHYB activatestranscription of genes acting downstream such as ELF4 and GI.PHYB-E activates transcription of HY5 which in turn regulatetranscription of light-regulated genes.

3224 Molas et al.

by on 18 October 2009 http://jxb.oxfordjournals.orgDownloaded from

COP1 can act as a positive regulator of PHYB-controlledresponses. Thus, they hypothesize that COP1 is involvedin the degradation of negative regulators of photomor-phogenesis or in the transcriptional activation of PHYB(Boccalandro et al., 2004) causing, in either case, a posit-ive effect in PHYB-mediated downstream signalling.

Taken together, 1 h of red-light exposure in the roots ofseedlings increase the transcription level of a subset ofgenes (i.e. GI, ELF4, COP1-like, PKS1, and HY5) that playa role in PHYB-mediated responses.

Blue/UV-light-induced pathway and phototropic response inroots: In this study, the induction of an NPH3-like(At5g48800) gene and the RPT2 gene, along with therepression of another NPH3-like gene (At1g03010), werefound. These results strongly suggest that the red and bluelight signalling pathways overlap. Phototropin 1 (PHOT1)and PHOT2, which are blue light receptor kinases, functionin blue-light-induced phototropism with RPT2 (root pho-totropism 2) and NPH3 (nonphototropic hypocotyl 3)transducing the signal downstream of the blue light. Muta-tions in NPH3 and RPT2 genes of Arabidopsis apparentlydisrupt the function of proteins acting early in PHOT1- andPHOT2-phototropic signalling pathways. Loss-of-functionrpt2 mutants retain nearly normal phototropism in hypo-cotyls, whereas the same response is impaired in roots,indicating that genetic regulation pathways mediating pho-totropism are different in roots and shoots (Okada andShimura, 1992).

NPH3 is highly expressed in dark-grown seedlings andremains unaffected by light (Liscum, 2002; Sakai et al.,2000), while RPT2 mRNA is barely detectable in etiolatedseedlings, but increases dramatically with increasing lightexposure. Moreover, RPT2 mRNA levels are induced byblue, green, and red light (Sakai et al., 2000; Teppermanet al., 2004), which indicates that this regulator ofphototropism is a light-inducible gene and may be a com-mon element in light signalling between the red and bluelight pathways.

Whole seedlings also had elements involved in thephototropin-mediated and blue/UVA-signalling pathwayregulated within 1 h of red light. For example, RPT2, PHR2(PHOTOLYASE/BLUE LIGHT PHOTORECEPTOR 2),and PHOT2 were induced in red light while PHOT1expression was repressed (Tepperman et al., 2004). Thesegenes, as well as the RPT2/NPH3 family, are potentialtargets for identifying the interacting affects of red-lightenhancement of blue-light-induced phototropism or thered-light-induced phototropism found in roots.

Synergistic interactions between phytochromes andblue-light photoreceptors do occur during the phototropicresponse (Iino, 1990; Liscum and Stowe-Evans, 2000;Kumar and Kiss, 2006). For example, PHYA and PHYBare the predominant phytochromes regulating phototro-pic enhancement of hypocotyls pretreated with red light

(Parks et al., 1996; Janoudi et al., 1997). In addition, phy-tochromes A, B, and D modulate blue-light phototropism inthe absence of a red-light pretreatment (Whippo andHangarter, 2004). Recently, PKS proteins were identifiedas elements linking phytochrome and phototropic curvaturein hypocotyls (Lariguet et al., 2006). However, results fromstudies with Arabidopsis and maize suggest that phyto-chromes may alter the activity and/or abundance ofelements downstream to enhance phototropic curvature inhypocotyls (Liu and Iino, 1996; Janoudi et al., 1997; Parkset al., 1997). On the other hand, the enhancement of blue-light phototropism by red light has not been establishedin roots. Nevertheless, phytochromes might modify theactivity of intermediaries downstream of blue-light-induced phototropism. This hypothetical mode of actionfor photoreceptors is summarized in Fig. 5A.

Roots also have a red-light-induced positive phototropicresponse (Ruppel et al., 2001; Kiss et al., 2003), and thismay be another explanation for the regulation of RPT2/NPH3 genes in Arabidopsis roots after 1 h of red light.Although it is known that PHYA and PHYB are thephotoreceptors sensing the light signal (Kiss et al., 2003),elements operating downstream from the phytochromeshave not been characterized. A potential model is illustratedin Fig. 5B.

It is also possible that both NPH3 and RPT2 genes areinvolved in responses other than phototropism in roots.RPT2 has recently been implicated in stomatal opening;however, NPH3 was not necessary for stomatal opening orchloroplast relocation (Inada et al., 2004). Further analysis

Fig. 5. Working model for the relationship of some of the elements inblue-light induced phototropism (A) and red-light-induced phototropism(B) in roots.

Gene profiling of the red light signalling pathways in roots 3225

by on 18 October 2009 http://jxb.oxfordjournals.orgDownloaded from

on the role of these elements in root responses may identifywhat role they play in root development.

Phenylpropanoid metabolism signal-transduction isinduced

Phenylpropanoid metabolism is an important metabolicpathway responsible for the synthesis of both developmen-tally required secondary compounds (such as lignin andflavonoids) and for the synthesis of defensive compounds.In roots, phenylpropanoid content depends on plant age, thequality and intensity of light treatment, and the source ofroot material (basal portions versus root tips; Hemm et al.,2004). The greatest amount of phenylpropanoids (conife-rin, syringin, Rha-Glu-Quercetin) were found in roots thatwere grown in white light compared with roots from plantsgrown in the dark, red, or blue light (Hemm et al., 2004).Although the roots from seedlings grown in red light hadthe lowest phenylpropanoid content of all treatments, rootsfrom phyB had reduced amounts of phenylpropanoidscompared with wild-type plants, suggesting that PHYB isinvolved in regulating phenylpropanoid metabolism inroots (Hemm et al., 2004).

For roots, the expression profiles of the genes fromphenylpropanoid metabolism are just beginning to be ex-plored. For instance, using RT-PCR, Hemm et al. (2004)identified several genes in the phenylpropanoid metabolismpathway of roots that were induced (>2-fold ) within 5 hof white-light treatment. These include PAL1 (PHENYL-ALANINE AMMONIA LYASE 1), PAL2, PAL3, CH(CHALCONE SYNTHASE), C4H (TRANS-CINNAMATE4- HYDROXYLASE), 4CL3 (4-COUMARATE:CoA LI-GASE), C39H (p-COUMARATE 3-HYDROXYLASE),CAD-C (CINNAMYL ALCHOHOL DEHYDROGENASE),and HY5. Here, five genes associated with phenylpropa-noid metabolism were found that were significantly up-regulated in 1 h of red-light treatment (HY5, PAL3,COMT-like 11; COMT-like 12; and CAD7/CAD8; seesupplementary Table 2 at JXB online). PAL3 was notsignificantly up-regulated in roots from seedlings thathave been grown continuously in white light, so it appearsthat there is a transient gene expression of PAL3 in roots(Hemm et al., 2004; Ma et al., 2005). Interestingly,cotyledons and hypocotyls from light-grown seedlingsand roots with 5 h of white-light treatment had significantup-regulation of both PAL1 and PAL2 (Hemm et al.,2004; Ma et al., 2005), suggesting that these genes maybe regulated during longer or different wavelength andfluence rates of light treatments than those performedhere. In contrast to our results for roots where there wasno significant regulation of PAL1, for whole seedlings, thisgene was induced within 1 h of red light (Teppermanet al., 2004). Studies on the time-course regulation of thePAL genes in roots will improve our understanding ofphenylpropanoid metabolism in plants.

Chloroplast development is differentially regulated

Roots, which are normally not exposed directly to light,express photoreceptors and can respond to light by de-veloping chloroplasts in a process known as root greening.In Arabidopsis, root greening occurs most effectively inblue or white light although some greening can occur inred light (Usami et al., 2004). Phytochrome B is the mainphotoreceptor involved in regulating root greening in redlight with HY5 playing an important downstream role inregulating the process (Oyama et al., 1997; Usami et al.,2004). COP1 and DET1 act downstream of photorecep-tors and are negative regulators of photomorphogenesis(Hardtke and Deng, 2000). These two proteins are alsonegative regulators of the greening process in red light ofroots (Usami et al., 2004). For roots of dark-grownseedlings exposed to 1 h of red light, seven genes werefound significantly up-regulated and six down-regulatedthat are associated with chloroplasts or photosyntheticelements (see supplementary Table 2 at JXB online). Acomparison of the changes in gene expression in rootsexposed to blue light may indicate the differences associ-ated with chloroplast development of roots exposed todifferent qualities of light.

Root hair formation is induced

In this study, two genes were identified that were closelyinvolved in root hair cell differentiation CPC (CAPRICE)and root hair formation RHD3 (ROOT HAIR DEFECTIVE3). CPC, a MYB protein, is a positive regulator of hair celldifferentiation in Arabidopsis roots (Wada et al., 1997).Elevated levels of CPC transcript result in a high concen-tration of the CPC protein, which in turn repress WER(WEREWOLF) and GL2 (GLABRA2) gene transcription,thus permitting the hair cell type to differentiate (fora review see Birnbaum and Benfey, 2004).

The RHD3 gene encodes a putative GTP-binding proteinrequired for appropriate cell enlargement in Arabidopsis.RHD3 expression is required for normal root hair growth(Schiefelbein and Somerville, 1990). Specifically, it hasbeen established that RHD3 is critical at the very beginningof hair formation and during root hair elongation (Parkeret al., 2000). In a physiological context, up-regulation ofCPC and RHD3 seems plausible since the normal de-velopmental pathway for root hair formation (i.e. notinduced by external stimuli) starts as early as 4 d inseedlings. In addition, root hair formation is stimulated byred light and regulated by phytochromes A and B (DeSimone et al., 2000b). Further studies on the gene expres-sion of these genes in root tips should prove useful inunderstanding light-regulated root development.

Light-regulated genes in roots and seedlings

Previous studies have examined the differential expressionof genes in roots from light- and dark-grown seedlings

3226 Molas et al.

by on 18 October 2009 http://jxb.oxfordjournals.orgDownloaded from

(Sato-Nara et al., 2004; Ma et al., 2005). In this study, veryfew genes (26) were found to be significantly differentiallyexpressed in both 1 h of red-light treatment and continuouswhite-light treatment (Table 2). Despite the reduced num-ber of genes overlapping in both studies, several major com-ponents in light signalling have been identified (i.e. HY5,ELF4, SPA1, PnCOP1-like, RAP2.6), suggesting that lightsignalling might involve similar elements in the early andlater stages of transduction.

Gene expression studies of whole seedlings comparedthe time-course expression profiles from red-light treatedplants with dark controls (Tepperman et al., 2004). Thesestudies found that 138 genes were differentially expressedwithin 1 h of the red-light treatment (Tepperman et al.,2004). Of these genes, nine overlapping elements with theprofiles from roots in red-light treatments were found.These genes include HY5, SPA1, PKS1, RPT2, STO, ELF4,POP1, and RAP2.6, which were induced, and bHLH (zetagene) which was suppressed in the red-light treatments.RAP2.6 is a putative AP2 domain ethylene response factor(ERF) that is up-regulated later in whole seedlings exposedto red light, suggesting it is an indirect target in lightregulation (Tepperman et al., 2004). Also, RAP2.6 regula-tion appears not to be directly regulated by PHYB butperhaps by other phytochromes (Tepperman et al., 2004).RAP2.6 has also been shown to be regulated by coldtreatment, and this, along with GI, may be an interactingelement in circadian clock rhythms and temperatureregulation (Fowler and Thomashow, 2002). In addition,RAP2.6 is also induced by bacterial infection and to alesser extent by jasmonic acid (He et al., 2004). Inductionof this gene correlates with increased susceptibility todisease, but the role of RAP2.6 in red-light regulatedprocesses in roots is unknown.

Gene expression analysis of dark-grown whole seedlingsexposed to 1 h of red light has also been investigated byTKrestch and coworkers. This microarray data was depositedin the NASCArray database (Nottingham Arabidopsis StockCentre, http://arabidopsis.info/) with experiment referencenumber NASCAARRAYS-124. Comparison of our resultswith roots of seedlings exposed to 1 h of red light withresults with dark-grown seedlings (mainly hypocotyls andcotyledons) show 15 overlapping genes (14 up- regulatedand 1 down-regulated genes). Major components in lightsignal transduction pathways are included in this group ofgenes, such as HY5, PKS1, ELF4, COP1-like (At5g52250),and RPT2. In addition, two ABC transporters (i.e. POP1and NAP9), a heat shock transcription factor (i.e. similarto GRP2), and four expressed proteins with unknown func-tion were up-regulated, and a homeobox–leucine zipperfamily protein was down-regulated in both studies.

Interestingly, HY5, POP1, SPA1, RAP2.6, and ELF4were up-regulated in roots of light-grown seedling, in rootsof seedling exposed to 1 h red light, and in whole seed-lings exposed to 1 h red light. This evidence reinforces the

idea that elements involved in the first steps of light signaltransduction are conserved in roots.

Light regulates many processes in roots, although theextent that these processes are controlled by light in normal(i.e. underground) conditions is largely unknown. Thereare several ways in which roots can be exposed to light.For example, roots that are uncovered from the soil surfacewill be exposed directly to illumination, light can diffusethrough the soil to the roots (Tester and Morris, 1987;Mandoli et al., 1990), or light can be transmitted throughtissues from the aerial tissues to the roots (Lauter, 1996;Sun et al., 2003). Therefore, roots may respond directly tothe light signal even when they are buried in soil. Futureexperiments comparing expression profiles from roots ofseedlings grown in different light quality and quantity willhelp identify wavelength specific pathways and the rolesof these genes in light-regulated responses in roots.

Supplementary data

Supplementary data can be found at JXB online.

Acknowledgements

Financial support was provided by NASA (grant NCC2-1200), theOhio Plant Biotechnology Consortium, Miami University (ShouppAward), and the University of Florida. We thank Mr Chris Wood(Miami University), Dr Frederic Souret (Delaware BiotechnologyInstitute), and Dr Mick Popp (University of Florida) for experttechnical assistance.

References

Allen T, Ingles PJ, Praekelt U, Smith H, Whitelam G. 2006.Phytochrome-mediated agravitropism in Arabidopsis hypocotylsrequires GIL1 and confers a fitness advantage. The Plant Journal46, 641–648.

Bauer D, Viczian A, Kircher S, et al. 2004. Constitutivephotomorphogenesis 1 and multiple photoreceptors control de-gradation of phytochrome interacting factor 3, a transcriptionfactor required for light signaling in Arabidopsis. The PlantCell 16, 1433–1445.

Bhalerao RP, Eklof J, Ljung K, Marchant A, Bennett M,Sandberg G. 2002. Shoot-derived auxin is essential for earlylateral root emergence in Arabidopsis seedlings. The Plant Journal29, 325–332.

Birnbaum K, Benfey PN. 2004. Network building: transcriptionalcircuits in the root. Current Opinion in Plant Biology 7, 582–588.

Boccalandro HE, Rossi MC, Saijo Y, Deng XW, Casal JJ.2004. Promotion of photomorphogenesis by COP1. PlantMolecular Biology 56, 905–915.

Briggs WR, Beck CF, Cashmore AR, et al. 2001. The phototro-pin family of photoreceptors. The Plant Cell 13, 993–997.

Cao SQ, Ye M, Jiang ST. 2005. Involvement of GIGANTEA genein the regulation of the cold stress response in Arabidopsis. PlantCell Reports 24, 683–690.

Chatterjee M, Sparvoli S, Edmunds C, Garosi P, Findlay K,Martin C. 1996. DAG, a gene required for chloroplast differen-tiation and palisade development in Antirrhinum majus. EMBOJournal 15, 4194–4207.

Gene profiling of the red light signalling pathways in roots 3227

by on 18 October 2009 http://jxb.oxfordjournals.orgDownloaded from

Chen M, Chory J, Fankhauser C. 2004. Light signal transduc-tion in higher plants. Annual Review of Genetics 38, 87–117.

Correll MJ, Kiss JZ. 2005. The roles of phytochromes in elong-ation and gravitropism of roots. Plant and Cell Physiology 46,317–323.

Datta S, Hettiarachchi G, Deng XW, Holm M. 2006. ArabidopsisCONSTANS-LIKE3 is a positive regulator of red light signalingand root growth. The Plant Cell 18, 70–84.

De Simone S, Oka Y, Inoue Y. 2000a. Effect of light on roothair formation in Arabidopsis thaliana phytochrome-deficientmutants. Journal of Plant Research 113, 63–69.

De Simone S, Oka Y, Nishioka N, Tadano S, Inoue Y. 2000b.Evidence of phytochrome mediation in the low-pH-induced roothair formation process in lettuce (Lactuca sativa L. cv. GrandRapids) seedlings. Journal of Plant Research 113, 45–53.

Eimert K, Wang SM, Lue WL, Chen JC. 1995. Monogenicrecessive mutations causing both late floral initiation and excessstarch accumulation in Arabidopsis. The Plant Cell 7, 1703–1712.

Fankhauser C, YehKC, Lagarias JC, ZhangH, Elich TD, Chory J.1999. PKS1, a substrate phosphorylated by phytochrome thatmodulates light signaling in Arabidopsis. Science 284, 1539–1541.

Feldman LJ, Briggs WR. 1987. Light-regulated gravitropism inseedling roots of maize. Plant Physiology 83, 241–243.

Fowler S, Thomashow MF. 2002. Arabidopsis transcriptome pro-filing indicates that multiple pathways are activated during coldacclimation in addition to the CBF cold response pathway. ThePlant Cell 14, 1675–1690.

Fowler S, Lee K, Onouchi H, Samach A, Richardson K,Coupland G, Putterill J. 1999. GIGANTEA: a circadianclock-controlled gene that regulates photoperiodic floweringin Arabidopsis and encodes a protein with several possiblemembrane-spanning domains. EMBO Journal 18, 4679–4688.

Halliday KJ, Hudson M, Ni M, Qin MM, Quail PH. 1999. poc1:an Arabidopsis mutant perturbed in phytochrome signalingbecause of a T-DNA insertion in the promoter of PIF3, a geneencoding a phytochrome-interacting bHLH protein. Proceedingsof the National Academy of Sciences, USA 96, 5832–5837.

Hardtke CS, Deng XW. 2000. The cell biology of the COP/DET/FUS proteins. Regulating proteolysis in photomorphogenesis andbeyond? Plant Physiology 124, 1548–1557.

Harmon AC, Gribskov M, Harper JF. 2000. CDPKs: a kinasefor every Ca2+ signal? Trends in Plant Science 5, 154–159.

He P, Chintamanani S, Chen Z, Zhu L, Kunkel BN, Alfano JR,Tang X, Zhou JM. 2004. Activation of COL1-dependent path-way in Arabidopsis by Pseudomonas syringae type III effectorsand coronatine. The Plant Journal 37, 589–602.

Hemm MR, Rider SD, Ogas J, Murry DJ, Chapple C. 2004.Light induces phenylpropanoid metabolism in Arabidopsis roots.The Plant Journal 38, 765–778.

Hoecker U, Tepperman JM, Quail PH. 1999. SPA1, a WD-repeatprotein specific to phytochrome A signal transduction. Science284, 496–499.

Huq E, Quail PH. 2002. PIF4, a phytochrome-interacting bHLHfactor, functions as a negative regulator of phytochrome Bsignaling in Arabidopsis. EMBO Journal 21, 2441–2450.

Huq E, Tepperman JM, Quail PH. 2000. GIGANTEA is a nuclearprotein involved in phytochrome signaling in Arabidopsis. Pro-ceedings of the National Academy of Sciences, USA 97,9789–9794.

Iino M. 1990. Phototropism: mechanisms and ecological implica-tions. Plant, Cell and Environment 13, 633–650.

Inada S, Ohgishi M, Mayama T, Okada K, Sakai T. 2004. RPT2is a signal transducer involved in phototropic response andstomatal opening by association with phototropin 1 in Arabidopsisthaliana. The Plant Cell 16, 887–896.

Janoudi AK, Konjevic R, Whitelam G, Gordon W, Poff KL.1997. Both phytochrome A and phytochrome B are required forthe normal expression of phototropism in Arabidopsis thalianaseedlings. Physiologia Plantarum 101, 278–282.

Jiao YL, Ma LG, Strickland E, Deng XW. 2005. Conservationand divergence of light-regulated genome expression patternsduring seedling development in rice and Arabidopsis. The PlantCell 17, 3239–3256.

Johnson E, Bradley M, Harberd NP, Whitelam GC. 1994.Photoresponses of light-grown PhyA mutants of Arabidopsis:phytochrome A is required for the perception of daylengthextensions. Plant Physiology 105, 141–149.

Khanna R, Kikis EA, Quail PH. 2003. EARLY FLOWERING4 functions in phytochrome B-regulated seedling de-etiolation.Plant Physiology 133, 1530–1538.

Kikis EA, Khanna R, Quail PH. 2005. ELF4 is a phytochrome-regulated component of a negative feedback loop involving thecentral oscillator components CCA1 and LHY. The Plant Journal44, 300–313.

Kim J, Shen Y, Han YJ, Park JE, Kirchenbauer D, Soh MS,Nagy F, Schafer E, Song PS. 2004. Phytochrome phosphorylationmodulates light signaling by influencing the protein–proteininteraction. The Plant Cell 16, 2629–2640.

Kiss JZ. 2000. Mechanisms of the early phases of plant gravitropism.Critical Reviews in Plant Sciences 19, 551–573.

Kiss JZ, Miller KM, Ogden LA, Roth KK. 2002. Phototropismand gravitropism in lateral roots of Arabidopsis. Plant andCell Physiology 43, 35–43.

Kiss JZ, Mullen JL, Correll MJ, Hangarter RP. 2003. Phyto-chromes A and B mediate red-light-induced positive phototropismin roots. Plant Physiology 131, 1411–1417.

Kiss JZ, Swatzell LJ. 1996. Development of the gametophyte ofthe fern Schizaea pusilla. Journal of Microscopy-Oxford 181,213–221.

Klein RR, Mason HS, Mullet JE. 1988. Light-regulated translationof chloroplast proteins. I. Transcripts of Psaa-Psab, Psba, and Rbclare associated with polysomes in dark-grown and illuminatedbarley seedlings. Journal of Cell Biology 106, 289–301.

Kumar P, Kiss JZ. 2006. Modulation of phototropism by phyto-chrome E and attenuation of gravitropism by phytochromes B andE in inflorescence stems. Physiologia Plantarum 27, 304–311.

Kurepa J, Smalle J, Van Montagu M, Inze D. 1998. Effects ofsucrose supply on growth and paraquat tolerance of the late-flowering gi-3 mutant. Plant Growth Regulation 26, 91–96.

Lariguet P, Boccalandro HE, Alonso JM, Ecker JR, Chory J,Casal JJ, Fankhauser C. 2003. A growth regulatory loop thatprovides homeostasis to phytochrome A signaling. The PlantCell 15, 2966–2978.

Lariguet P, Schepens I, Hodgson D, et al. 2006. PHYTOCHROMEKINASE SUBSTRATE 1 is a phototropin1 binding proteinrequired for phototropism. Proceedings of the National Academyof Sciences, USA 103, 10134–10139.

Lauter FR. 1996. Root-specific expression of the LeRse-1 genein tomato is induced by exposure of the shoot to light. Molecu-lar and General Genetics 252, 751–754.

Lin CT, Shalitin D. 2003. Cryptochrome structure and signaltransduction. Annual Review of Plant Biology 54, 469–496.

Lin RC, Wang HY. 2005. Two homologous ATP-binding cassettetransporter proteins, AtMDR1 and AtPGP1, regulate Arabidopsisphotomorphogenesis and root development by mediating polarauxin transport. Plant Physiology 138, 949–964.

Liscum E. 2002. Phototropism: mechanisms and outcomes. In:Somerville CR, Meyerowitz EM, eds. The Arabidopsis Book.American Society of Plant Biologists, Rockville, MD, doi:10.1199/tab.0009, www.aspb.org/publications/arabidopsis/

3228 Molas et al.

by on 18 October 2009 http://jxb.oxfordjournals.orgDownloaded from

LiscumE, Stowe-Evans EL. 2000. Phototropism: A ‘simple’ physio-logical response modulated by multiple interacting photosensory-response pathways. Photochemistry and Photobiology 72,273–282.

Liu YJ, Iino M. 1996. Phytochrome is required for the occurrenceof time-dependent phototropism in maize coleoptiles. Plant, Celland Environment 19, 1379–1388.

Livak K. 1997. ABI Prism 7700 sequence detection system. UserBulletin 2. Foster City, CA: Applied Biosystem.

Lu YT, Feldman LJ. 1997. Light-regulated root gravitropism:a role for, and characterization of, a calcium/calmodulin-dependentprotein kinase homolog. Planta 203, S91–S97.

Ma LG, Li JM, Qu LJ, Hager J, Chen ZL, Zhao HY, Deng XW.2001. Light control of Arabidopsis development entails coordi-nated regulation of genome expression and cellular pathways. ThePlant Cell 13, 2589–2607.

Ma LG, Sun N, Liu XG, Jiao YL, Zhao HY, Deng XW.2005. Organ-specific expression of Arabidopsis genome duringdevelopment. Plant Physiology 138, 80–91.

Mandoli DF, Ford GA, Waldron LJ, Nemson JA, Briggs WR.1990. Some spectral properties of several soil types: implicationsfor photomorphogenesis. Plant,Cell and Environment 13, 287–294.

Martinez-Garcia JF, Huq E, Quail PH. 2000. Direct targetingof light signals to a promoter element-bound transcriptionfactor. Science 288, 859–863.

Moller SG, Kunkel T, Chua NH. 2001. A plastidic ABC proteininvolved in intercompartmental communication of light signaling.Genes and Development 15, 90–103.

Motchoulski A, Liscum E. 1999. Arabidopsis NPH3: a NPH1photoreceptor-interacting protein essential for phototropism.Science 286, 961–964.

Ni M, Tepperman JM, Quail PH. 1998. PIF3, a phytochrome-interacting factor necessary for normal photoinduced signal trans-duction, is a novel basic helix–loop–helix protein. Cell 95,657–667.

Okada K, Shimura Y. 1992. Aspects of recent developments inmutational studies of plant signaling pathways. Cell 70, 369–372.

Osterlund MT, Wei N, Deng XW. 2000. The roles of photoreceptorsystems and the COP1-targeted destabilization of HY5 in lightcontrol of Arabidopsis seedling development Plant Physiology124, 1520–1524.

Oyama T, Shimura Y, Okada K. 1997. The Arabidopsis HY5 geneencodes a bZIP protein that regulates stimulus-induced develop-ment of root and hypocotyl. Genes and Development 11,2983–2995.

Park E, Kim J, Lee Y, Shin J, Oh E, Chung WI, Liu JR,Choi G. 2004. Degradation of phytochrome interacting factor 3in phytochrome-mediated light signaling. Plant and Cell Phy-siology 45, 968–975.

Parker JS, Cavell AC, Dolan L, Roberts K, Grierson CS.2000. Genetic interactions during root hair morphogenesis inArabidopsis. The Plant Cell 12, 1961–1974.

Parks BM, Quail PH, Hangarter RP. 1996. Phytochrome Aregulates red-light induction of phototropic enhancement inArabidopsis. Plant Physiology 110, 155–162.

Parks BM, Sharrock RA, Spalding EP. 1997. Rapid growthinhibition by red light: a genetic approach using photomorpho-genic mutants. Plant Physiology 114, 1472–1472.

Ruppel NJ, Hangarter RP, Kiss JZ. 2001. Red-light-inducedpositive phototropism in Arabidopsis roots. Planta 212,424–430.

Saijo Y, Sullivan JA, Wang HY, Yang JP, Shen YP, Rubio V,Ma LG, Hoecker U, Deng XW. 2003. The COP1-SPA1 inter-

action defines a critical step in phytochrome A-mediated regula-tion of HY5 activity. Genes and Development 17, 2642–2647.

Sakai T, Wada T, Ishiguro S, Okada K. 2000. RPT2: a signaltransducer of the phototropic response in Arabidopsis. The PlantCell 12, 225–236.

Sanders D, Pelloux J, Brownlee C, Harper JF. 2002. Calcium atthe crossroads of signaling. The Plant Cell 14, S401–S417.

Sato-Nara K, Nagasaka A, Yamashita H, Ishida J, Enju A,Seki M, Shinozaki K, Suzuki H. 2004. Identification ofgenes regulated by dark adaptation and far-red light illuminationin roots of Arabidopsis thaliana. Plant, Cell and Environment27, 1387–1394.

Schiefelbein JW, Somerville C. 1990. Genetic control of roothair development in Arabidopsis thaliana. The Plant Cell 2,235–243.

Seo HS, Watanabe E, Tokutomi S, Nagatani A, Chua NH.2004. Photoreceptor ubiquination by COP1 E3 ligase desensitizesphytochrome A signaling. Genes and Development 18, 617–622.

Sharrock RA, Quail PH. 1989. Novel phytochrome sequences inArabidopsis thaliana structure, evolution, and differential expres-sion of a plant regulatory photoreceptor family. Genes andDevelopment 3, 1745–1757.

Steel RG, Torrie JH. 1960. Principles and procedures of statis-tics. New York: McGraw Hill Book Company.

Sun Q, Yoda K, Suzuki M, Suzuki H. 2003. Vascular tissue inthe stem and roots of woody plants can conduct light. Journal ofExperimental Botany 54, 1627–1635.

Sweere U, Eichenberg K, Lohrmann J, Mira-Rodado V,Baurle I, Kudla J, Nagy F, Schafer E, Harter K. 2001. Inter-action of the response regulator ARR4 with phytochrome B inmodulating red light signaling. Science 294, 1108–1111.

Tepperman JM, Hudson ME, Khanna R, Zhu T, Chang SH,Wang X, Quail PH. 2004. Expression profiling of phyB mutantdemonstrates substantial contribution of other phytochromes tored-light-regulated gene expression during seedling de-etiolation.The Plant Journal 38, 725–739.

Tepperman JM, Zhu T, Chang HS, Wang X, Quail PH. 2001.Multiple transcription factor genes are early targets of phyto-chrome A signaling. Proceedings of the National Academy ofSciences, USA 98, 9437–9442.

Tester M, Morris C. 1987. The penetration of light through soil.Plant, Cell and Environment 10, 281–286.

Usami T, Mochizuki N, Kondo M, Nishimura M, Nagatani A.2004. Cryptochromes and phytochromes synergistically regulateArabidopsis root greening under blue light. Plant and CellPhysiology 45, 1798–1808.

Wada T, Tachibana T, Shimura Y, Okada K. 1997. Epidermal celldifferentiation in Arabidopsis determined by a Myb homolog,CPC. Science 277, 1113–1116.

Wang HY, Ma LG, Habashi J, Li JM, Zhao HY, Deng XW.2002. Analysis of far-red light-regulated genome expressionprofiles of phytochrome A pathway mutants in Arabidopsis.The Plant Journal 32, 723–733.

Whippo CW, Hangarter RP. 2004. Phytochrome modulationof blue-light-induced phototropism. Plant, Cell and Environment27, 1223–1228.

Wilson IW, Kennedy GC, Peacock JW, Dennis ES. 2005.Microarray analysis reveals vegetative molecular phenotypesof Arabidopsis flowering-time mutants. Plant and Cell Phy-siology 46, 1190–1201.

Yang HQ, Tang RH, Cashmore AR. 2001. The signaling me-chanism of Arabidopsis CRY1 involves direct interactionwith COP1. The Plant Cell 13, 2573–2587.

Gene profiling of the red light signalling pathways in roots 3229

by on 18 October 2009 http://jxb.oxfordjournals.orgDownloaded from