Gene 482 (2011) 34–42

Contents lists available at ScienceDirect

Gene

j ourna l homepage: www.e lsev ie r.com/ locate /gene

Genome-wide identification of gibberellins metabolic enzyme genes and expressionprofiling analysis during seed germination in maize

Jian Song, Baojian Guo, Fangwei Song, Huiru Peng, Yingyin Yao, Yirong Zhang, Qixin Sun ⁎, Zhongfu Ni ⁎State Key Laboratory for Agrobiotechnology, Key Laboratory of Crop Heterosis Utilization (MOE), Key Laboratory of Crop Genomics and Genetic Improvement (MOA),Beijing Key Laboratory of Crop Genetic Improvement, China Agricultural University, Beijing, 100193, ChinaNational Plant Gene Research Centre (Beijing), Beijing 100193, China

Abbreviations: Bp, base pair(s); cDNA, DNA complesequence;CPS, ent-copalyldiphosphate synthesis; CT, cycle tacid; dNTP, deoxyribonucleoside triphosphate; EST, expressGA20ox, GA 20-oxidase; GA2ox, GA 2-oxidase; GA3ox,oxidase; KS, ent-kaurene synthase; PCR, polymerase chainPCR, reverse transcription PCR; UTR, untranslated region.⁎ Corresponding authors at: China Agricultural Univ

NO. 2, Haidian District, Beijing, 100193, China. Tel.: +62733426.

E-mail addresses: qxsun@cau.edu.cn (Q. Sun), nzhf2

0378-1119/$ – see front matter © 2011 Elsevier B.V. Adoi:10.1016/j.gene.2011.05.008

a b s t r a c t

a r t i c l e i n f o

Article history:Accepted 16 May 2011Available online 24 May 2011

Received by A.J. van Wijnen

Keywords:MaizeGibberellin metabolic gene familyGene expressionSeed germination

Gibberellin (GA) is an essential phytohormone that controlsmany aspects of plant development. To enhance ourunderstanding of GAmetabolism in maize, we intensively screened and identified 27 candidate genes encodingthe seven GAmetabolic enzymes including ent-copalyl diphosphate synthase (CPS), ent-kaurene synthase (KS),ent-kaurene oxidase (KO), ent-kaurenoic acid oxidase (KAO), GA 20-oxidase (GA20ox), GA 3-oxidase (GA3ox),and GA 2-oxidase (GA2ox), using all available public maize databases. The results indicate that maize genomecontains three CPS, four KS, two KO and one KAO genes, and most of them are arranged separately on the maizegenome,whichdiffers fromthat in rice. In addition, the enzymes catalyzing the later steps (ZmGA20ox, ZmGA3oxand ZmGA2ox) are also encoded by gene families in maize, but GA3ox enzyme is likely to be encoded by singlegene. Expression profiling analysis exhibited that transcripts of 15 GAmetabolic genes could be detected duringmaize seed germination, which provides further evidence for the notion that increased synthesis of active GA inthe embryo is required for triggering germination events. Moreover, a variety of temporal genes expressionpatterns of GAmetabolic genes were detected, which revealed the complexity of underlying mechanism for GAregulated seed germination.

mentary to RNA; CDS, codinghreshold;DNA,deoxyribonucleiced sequence tag; GA, gibberellin;GA 3-oxidase; KO, ent-kaurenereaction; PB, paclobutrazol; RT-

ersity, Yuanmingyuan Xi Road86 10 62731452; fax: +86 10

002@yahoo.com.cn (Z. Ni).

ll rights reserved.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Gibberellin (GA) is an essential phytohormone that controls manyaspects of plant development, including seed germination, leafexpansion, stem elongation, flowering, and seed development(Davies, 1995; Richards et al., 2001). Biochemical studies revealedthat the GAs form a large family of tetracyclic diterpenoid phytohor-mones, and the biosynthesis of GA in plants can be divided into sevensteps: (1) ent-CPS (ent-copalyl diphosphate synthesis) modifies GGDP(trans-geranylgeranyl diphosphate) to CDP (ent-copalyl diphos-phate); (2) KS (ent-kaurene synthase) modifies CDP to ent-kaurene;(3)KO (ent-kaurene oxidase) modifies ent-kaurene to ent-kaurenoicacid; (4) KAO (ent-kaurenoic acid oxidase) modifies ent-kaurenoicacid to GA12-aldehyde and then to GA12(to GA53 by GA13-oxidase); (5)GA 20-oxidase modifies GA12/GA53 to GA20 /GA9; (6)GA3-oxidase

modifies GA20/GA9 to GA1/GA4; (7) GA 2-oxidase modifies GA1/GA4 toGA8/GA34 (Lange, 1998; Kamiya and García-Martínez, 1999; Heddenand Phillips, 2000; Yamaguchi and Kamiya, 2000).

Recently, genome wide analysis revealed that enzymes catalyzingthe early steps in the GA biosynthetic pathway (i.e. CPS, KS and KO) areencodedby single genes inArabidopsis (Ogawa et al., 2003),while thosefor later steps (i.e. GA20ox, GA3ox, and GA2ox) are encoded by genefamilies. In contrast to the Arabidopsis genome,multiple CPS, KS, and KOgenes were identified in the rice genome (Sakamoto et al., 2004), mostof which are contiguously arranged. Some genes encoding seven GAmetabolic enzymes (CPS, KS, KO, KAO, GA20ox, GA3ox, and GA2ox) andtheir related mutants have been isolated from various plants (Heddenand Phillips, 2000). However, to our best knowledge, the maize GAmetabolic genes have not been analyzed in detail.

In most flowering plants, seed germination is the first and may bethe foremost growth stage in the plant's life cycle. It starts with wateruptake by the quiescent dry seed, terminates with extension of theembryonic axis 24–36 h post imbibition and is followed by radicleelongation (Koornneef and Karssen, 1994; Bewley, 1997; Koornneefet al., 2002). It has been known that GAs are required for enhancingseed germination, and further investigation showed that an increasein the synthesis of active GA, rather than a decrease in GAdeactivation, contributes to the promotion of this process (Hooley,1994; Debeaujon and Koornneef, 2000; Ogawa et al., 2003; Miransariand Smith, 2009). For example, the expression profile of AtGA3ox2 in

Table 1Primers used for real-time quantitative PCR.

Gene Primer sequences (5′–3′) Primer sequences(5′–3′)

ZmCPS1 ATGATGAGCCATGTCGATGA GAAAGGTCTGCCTTGTCTCGZmCPS2 ACGGCTCCTTCTTGTTCTCTC AGTCCATGCACTGCTCAATCTZmCPS3 TGGATTGGGAAGGTCCTGTA CTGCATCGTACGTCGACAAGZmKS1 AAAGCGTCGCAGATTCAGAT TGCTCTATGTGCCTCCTGTGZmKS2 CAAGCAACGCGATGACGA CCTCGTGCCTCTCTAACCAAZmKS3 CAAGGTGCTCCACCTGTTCT CCTCCGGTCAGCATTTGTATZmKS4 CACAGTGGTGGTTCCATGTC CTTCCAGAACAGCTCCTTGCZmKO1 CAATCTGTACGGGTGCAACA CCTCTAGGTGCGAGGTACACATZmKO2 GGAGGAAGGAGGGCCTGTG GCTCACAGACCAGAGACCAAZmKAO TCTTCCTCCACCACTTCCTC CAGCCGCTAGAGTTTCTGCTZmGA20ox1 GCGCTACTGCTCTGAGATGA CACGGCGGGTAGTAGTTGAGZmGA20ox2 GTCAACATCGGCGATCTGTT TAGGACGGGCACAGGAAGTAZmGA20ox3 GAGAGGTGTGGGTTCAGGTG CGAGCTTGCAATTACCCTGTZmGA20ox4 GAGAGGTTCTCCATGCCCTA AAGAAGTCGCCCCAGTTGTAZmGA20ox5 CACTCCCAGCCCCAGCATT CAGCAGTAGGGCCAGCGTZmGA3ox1 GACTAGCCGATTGCAAAAGC TAGCTGCGGAACGGAATTAGZmGA3ox2 ATTCGTGAATAAGGGCGTTG TCGACCGTAAAAATGGTGAAZmGA2ox1 TTCACCTTCGGGGAGTACAG GCATCCACCGTGAAGTTACAZmGA2ox2 CGCCTTCTTCATCAACGTC GGCGAAGTAGATGGTGGACAZmGA2ox3 CCGTAGTACACACGCACACC CACTCCCGAGAAGCTGTCCZmGA2ox4 GAGAAAGGTGCAGGAAGACG CTGCACTCTCCTCTCCATTGZmGA2ox5 GAGAGGTGTGGGTTCAGGTG CGAGCTTGCAATTACCCTGTZmGA2ox6 GACACGGAATCCTTCTTCGTC CCACGTGAACTCCCTGTAGCZmGA2ox7 ACTTCACATGGGGCGACTAC GTGTTGGGGCCGGTCTATZmGA2ox-8 GCGGGTGCACTAGACTACG GCATGGACCAGGGAGCAGZmGA2ox-9 CCTGCAGGTGTTGACGAAC TTGTGCTCCACGCTCTTGTAZmGA2ox10 CGTGGGGCGAGTACAAGAC CTCCCCAGCTGTACACATCA18s rRNA ACATGCGCCTAAGGAGAAATAG ACCTCCATGCTCACTGGTACTT

35J. Song et al. / Gene 482 (2011) 34–42

Arabidopsis is highly correlatedwith changes inGA4 contents, suggestinga direct role for this gene in increasing active GA levels to cause radicleemergence (Ogawa et al., 2003). Therefore, to understand the roles of GAin germination and radicle elongation, it is necessary to study how GAwas regulated and how seeds responded to GA, and one of the ways togain insight into the regulation of GA is to examine the expression ofgenes encoding enzymes involved in GA biosynthesis and catabolism.

In this study, genomic structures, chromosomal locations andsequence homology of seven GA metabolic enzymes (CPS, KS, KO,KAO, GA20ox, GA3ox and GA2ox) genes were identified and deter-mined. Moreover, their expression patterns during seed germinationwere analyzed, and signals of 15GAmetabolic genesweredetectedwithvarying temporal expression profiles after seed imbibitions, which willpave the way for elucidating the precise roles of these genes in maizeseed germination in the future.

2. Materials and methods

2.1. Plant materials

Seeds of maize inbred line B73, were used for this study. Seeds fromB73 were placed embryo side down on two pieces of Whatman No.1filter paper in a plastic Petri dish under dark at 30 °C for 72 h afterimbibitions. For expression analysis, embryos of the 10 maize inbredline seeds imbibed for 0 h, 24 h, 32 h, 40 h, 48 h, 56 h, 64 h and 72 hwere separated, frozen in liquid nitrogen and stored at−80 °C for SYBRGreen-based real-time quantitative PCR. In addition, special care wastaken to characterize these materials used and three biological sampleswere harvested.

Germination experiments were conducted in a growth chamber at30 °C in dark treatedwith water, paclobutrazol (50 μM; an inhibitor ofent-kaurene oxidase, early in the GA biosynthetic pathway) and GA3

(50 μM), separately. Seeds were considered germinated when radicalprotrusion was visible. Germination tests were performed at leasttwice. And embryos of the 10 maize inbred line seeds at 32 h wereexcised from seeds treated with water and paclobutrazol, frozen inliquid nitrogen and stored at −80 °C for SYBR Green-based real-timequantitative PCR.

2.2. Identification of GA Metabolic Genes in Maize

With reported sequences of the GA metabolic enzymes in Arabi-dopsis and rice as queries (Ogawa et al., 2003; Sakamoto et al., 2004),the blast programs were used to search the maize genome databases,including the Maize Genetics and Genomics Datebase (http://www.maizegdb.org/), Maize Genome Browser (http://www.maizesequence.org) and NCBI (http://www.ncbi.nlm.nih.gov). The genomic contextand the chromosomal location were determined based on informationfrom the Maize Genetics and Genomics Datebase (http://www.maizegdb.org/). ESTs were studied individually in alignment withArabidopsis and rice sequences, to checkwhether they represented full-length cDNAs and to eliminate transcript redundancy.

2.3. Alignment, phylogenetic analysis and motif detection

The amino acid sequences of GA metabolic enzymes in Arabidopsis,rice andmaizewere alignedusing theClustalWtool (http://www.ebi.ac.uk/Tools/clustalw2/index.html). Themultiple alignments resulted in anunrooted distance tree using Neighbor-Joining algorithms of MEGAversion 4 (Kumar et al., 2008). And the reliability of the tree wasexamined by bootstrap analyses (1000 replicates).

2.4. Isolation of Total RNA

Total RNA was extracted using standard Trizol RNA isolationprotocol (Life Technologies, USA). The amount and quality of the total

RNA were checked through electrophoresis in 1% agarose gel. Theconcentration of RNA was measured by spectrophotometer NanoDrop,ND1000 (Nano Drop Technologies).

2.5. Reverse transcription

Equal amounts of 2 μg RNA each were used to synthesize cDNA in20 μl reactions containing 50 mM Tris–HCl (pH 8.3), 75 mM KCl, 3 mMMgCl2, 10 mM DTT, 50 μM dNTPs, 200 U MMLV reverse transcriptase(Promega) and 1 μg random anchor primers. Reverse transcription wasperformed for 60 min at 37 °Cwith a final denaturation step at 95 °C for5 min. Aliquots of 2 μl of the obtained cDNA eachwere subjected to RT-PCR analysis.

2.6. Quantitative real-time PCR

GA metabolic gene specific primers were used for quantitativereal-time PCR analysis, and 18 s rRNA gene was amplified as anendogenous control (Table 1). For quantitative real-time PCR, cDNAsfrom three biological samples were used for analysis and all thereactions run three times. PCR conditions were performed as follows:94 °C for 4 min, followed by 40 cycles: 94 °C for 30 s, 66 °C/58 °C 30 s,72 °C 30 s, and then 72 °C for 5 min. Quantification results wereexpressed in terms of the cycle threshold (CT) value determinedaccording to the manually adjusted baseline. Relative gene expres-sions in different samples were determined using the method asdescribed previously. Briefly, differences between the CT values oftarget gene and 18s rRNA were calculated as ΔCT=CT target−CT18srRNA, and expression levels of target genes relative to 18s rRNA weredetermined as 2−ΔCT. For each sample, PCR was repeated three times,and the average values of 2−ΔCT were used to determine difference inexpression.

36 J. Song et al. / Gene 482 (2011) 34–42

3. Results

3.1. Identification and phylogenetic analysis of putative GAmetabolic genesin maize

We extensively searched public and proprietary transcript andgenomic databases, by using all previously reported genes encodingGA metabolism enzymes in maize and other species as BLAST queries,and identified 27 maize genes that have complete sequences in theappropriate bacterial artificial chromosome (BAC) clones. The nearestgenetic markers for each gene were determined from maize BACcontigs and used to position the GA metabolic genes on maize geneticmap. Except for ZmKS4, the other 26 genes were mapped on 9 of the10 maize chromosomes. Five genes are present on chromosome 3 and6; four on chromosome 1; three on chromosomes 2 and 8; two onchromosomes 5 and 9; only one on chromosome 4 and 10 (Table 2,Fig. 1). In addition, the intron/exon structures of 26 GA metabolicenzyme genes in maize were determined through alignments of thecDNA and genomic sequences. It was shown that, except forZmGA20ox1, the coding sequences of 25 genes are disrupted byintrons, and the number of exons varied from 1 to 11 (Fig. 2).

3.1.1. CPS and KS genesThe first committed intermediate in GA biosynthesis, ent-kaurene,

is derived from a two-step cyclization of the universal diterpenoidprecursor (E,E,E)-geranylgeranyl diphosphate (GGPP), catalyzed bytwo separate enzymes; ent-copalyl diphosphate synthase (ent-CPS;formerly kaurene synthase A) and ent-kaurene synthase (ent-KS),respectively (Olszewski et al., 2002).

Four CPS genes were identified in the rice, whereas single gene(AtCPS/GA1) was in Arabidopsis (Ogawa et al., 2003; Sakamoto et al.,2004). In present study, three CPS genes were identified in maizegenome, including a pseudogene (ZmCPS3), whose transcript lacksthe coding sequence of exons 8–12 of ZmCPS2. Both ZmCPS1 andZmCPS2 were located on chromosome 1, and ZmCPS3 was mapped on

Table 2The information about GA metabolic genes in maize.

Gene family Gene Accession no. UniGene Bin

CPS ZmCPS1 AY562491 Zm.19364 1.04ZmCPS2 L37750 Zm.228 1.08ZmCPS3 CF634281 – 10.03

KS ZmKS1 BT067769 Zm.86733 1.08ZmKS2 BT042236 Zm.31417 2.02ZmKS3 BT054450 Zm.85686 3.04ZmKS4 BU499812 Zm.6316 –

KO ZmKO1 BT042205 Zm.8089 4.09ZmKO2 EZ104608 Zm.119245 9.03

KAO ZmKAO NM_001112116 Zm.390 9.03GA20 oxidase ZmGA20ox1 NM_001112453 Zm.3184 1.12

ZmGA20ox2 NM_001143386 Zm.86468 2.09ZmGA20ox3 – – 3.06ZmGA20ox4 NM_001156071 Zm.66651 5.03ZmGA20ox5 ACN25832 – 8.06

GA3 oxidase ZmGA3ox1 – – 3.03ZmGA3ox2 EU974202 Zm.37898 6.05

GA2 oxidase ZmGA2ox1 NM_001154780 Zm.103202 2.03ZmGA2ox2 – – 3.04ZmGA2ox3 FL100324 – 3.07ZmGA2ox4 EC858802 – 5.05ZmGA2ox5 NM_001158585 Zm.23033 6.05ZmGA2ox6 NM_001153400 Zm.23305 6.05ZmGA2ox7 NM_001154796 Zm.33978 6.06ZmGA2ox-8 NM_001137734 Zm.30825 6.07ZmGA2ox-9 NM_001157265 Zm.25436 8.03ZmGA2ox10 - - 8.09

chromosome 10. The amino acid sequences of Arabidopsis AtCPS/GA1showed high homology with ZmCPS1 and 2 and relatively lowidentity with ZmCPS3 (Fig. 3). An Asp-rich box, DXDDTA, which isshared inmany terpene cyclases (Sakamoto et al., 2004), was conservedin ZmCPS2 and ZmCPS3; however, thefirst Aspwas replacedwith Glu inZmCPS1. ASAYDTAWVA motif present in CPS and KS proteins was alsoconserved in ZmCPS2, whereas VSAYDTAWVGwas observed in ZmCPS1and ISAYDTAWVA in ZmCPS3 (Supplemental Fig. 1). Phylogeneticanalysis of the CPS (Fig. 3) revealed that ZmCPS2 was most closelyrelated to OsCPS1, whose mutant showed severe dwarf phenotypewithout flower or seed development (Sakamoto et al., 2004). Thephenotype of the maize ZmCPS2 (an1/dwarf5) mutant includes semidwarfed height, shorter broader leaves and abnormal development ofreproductive structures (Bensen et al., 1995),which could be restored byintroducing ent-kaurene (Katsumi, 1964). Moreover, an1 (ZmCPS2)homozygous deletion mutant accumulated ent-kaurene only 20% of thewild-type level, suggesting that ZmCPS2was themajor functional geneofZmCPS gene family in maize (Bensen et al., 1995).

Four KS genes were identified in maize, nine KS genes in rice,whereas single gene (AtKS/GA2) in Arabidopsis. ZmKS1, ZmKS2 andZmKS3 were presented on chromosomes 1, 2 and 3, respectively.Phylogenetic analysis of KS proteins in Arabidopsis, rice and maizerevealed thatZmKS2was closely related to theOsKS1gene (Fig. 3)whichshowedseveredwarf phenotypewithoutfloweror seeddevelopment inmutants (osks1-1, osks1-2, and osks1-3; Sakamoto et al., 2004).The Asp-rich sequence, DDXXD involved in binding to the diphosphate group ofthe substrate (Sakamoto et al., 2004), was highly conserved in all ZmKSproteins. In addition, PSPYDTAWVA, PSSYDTAWAA, and PSAYDTAWVAmotifs shared among KS proteins were also present in ZmKS1, ZmKS3and ZmKS4, except for ZmKS2 (Supplemental Fig. 2).

3.1.2. KO and KAO genesIn the second stage of GA biosynthesis pathway, GA12 is produced by

stepwise oxidation following ring contraction catalyzed by ent-kaureneoxidase (KO) and ent-kaurenoic acid oxidase (KAO), which can be

BAC No. of intron No. of exon Length of CDS (bp)

AC196188 9 10 2477AC195814 11 12 2480AC215205 7 8 1266AC203832 9 10 2449AC209430 9 10 1773AC214360 6 7 1773– – – 2523AC183926 7 8 1527AC225885 6 7 1206AC189774 5 6 1559AC203246 0 1 1116AC201873 1 2 1107AC191666 2 3 1161AC209682 2 3 1049AC209074 2 3 1179AC204749 2 3 1146AC204513 2 3 1116AC183307 2 3 1062AC213610 1 2 726AC203362 3 4 996AC184051 2 3 1086AC198080 2 3 1101AC200263 2 3 999AC194246 2 3 1005AC210731 2 3 1122AC186515 3 4 1011AC194097 2 3 1038

Fig.1. Genomic locations of 27 nonredundant BACs that contained GA biosynthesis and catabolism genes.

37J. Song et al. / Gene 482 (2011) 34–42

converted further to GA53 by 13-hydroxylation (Olszewski et al., 2002).TwoKO genes (ZmKO1and ZmKO2)were identified inmaize and locatedon chromosomes 4 and 9, respectively. Whereas we only found oneZmKAO gene which was homologous to the reported KAO genes in riceand Arabidopsis,and mapped it on chromosome 9. The mutation ofZmKAO (d3) induced the severely dwarfed phenotype, further confirm-ing the function nonredundancy of ZmKAO gene (Winkler andHelentjaris, 1995).

3.1.3. GA20-oxidase and GA3-oxidase genesIn the third stage of GA biosynthesis pathway, GA12 and GA53 are

converted to various GA intermediates and bioactive GAs, includingGA1 and GA4 which were mainly biologically active in higher plants,by a series of oxidation steps catalyzed by 2-oxoglutarate-dependent

Fig. 2. Intron–exon structures o

dioxygenases, GA 20-oxidases (GA20ox), and GA 3-oxidases (GA3ox)(Olszewski et al., 2002).

Five GA20ox genes were identified in maize, four in rice and five inArabidopsis. The five ZmGA20ox genes (ZmGA20ox1-5) weremapped to5 of the 10maize chromosomes (1, 2, 3, 5 and 8) with one chromosomeharboring one ZmGA20ox gene. The consensus sequence NYYPXCXXP of2ODDs for binding the common cosubstrate, 2-oxoglutarate (Sakamotoet al., 2004), was conserved in ZmGA20ox1, ZmGA20ox3 andZmGA20ox5 proteins, and changed to NHYPXCXXP in ZmGA20ox2and ZmGA20ox4 proteins. A sequence LPWKET, which is considered tobe responsible for the binding of the GA substrates (Sakamoto et al.,2004), was also conserved in ZmGA20ox proteins, except forZmGA20ox4 (Supplemental Fig. 2). Phylogenetic analysis of GA20oxrevealed that ZmGA20ox3 and ZmGA20ox5 shared homology with

f the GA metabolic genes.

Fig. 3. Phylogenetic analysis of Arabidopsis, rice and maize GA metabolic enzyme proteins by using MEGA neighbor-joining. The protein sequences of Arabidopsis and rice wereassigned a number based on their previous reports (Ogawa et al., 2003; Sakamoto et al., 2004).

38 J. Song et al. / Gene 482 (2011) 34–42

OsGA20ox2 (Fig. 3), which was identical to the rice Green Revolutiongene, Semi-Dwarf1 (SD1; Sasaki et al., 2002).

Two GA3ox genes were annotated in maize and rice, respectively,whereas four GA3ox genes in Arabidopsis. ZmGA3ox1 and ZmGA3ox2genes were mapped on chromosome 6 and 3, respectively. The aminoacid sequences of ZmGA3ox1 and ZmGA3ox2 showed high similaritywith those of OsGA3ox2 and OsGA3ox1 in rice, respectively (Fig. 3). Itwas reported that the mutation of OsGA3ox2 (d18) resulted in asemidwarf phenotype (Sakamoto et al., 2004). The maize mutantdwarf1, blocking the metabolism of GA20 to GA1 and exhibiting a dwarfphenotype, appeared to be the mutant of ZmGA3ox gene, but there wasno information about the DWARF1 gene sequence (Spray et al., 1996).

3.1.4. GA2-oxidase genesThe pool of active GA1, GA4 and its precursor GA20 could be

depeleted by the action of GA 2-oxidase, which converts those GAsinto inactive forms (Busov et al., 2003). In maize, a total of 10 genes(ZmGA2ox1-10) encoding ZmGA2ox were identified, and mapped onchromosomes 2, 3, 5, 6 and 8, respectively. The resulting phylogenetic

tree of GA2ox proteins in Arabidopsis, rice and maize implified thatZmGA2ox3 and ZmGA2ox10 showed high homology with OsGA2ox3(Fig. 3), which was speculated functioning in control of bioactive GAlevels in rice (Sakamoto et al., 2004).

3.2. The relationship of GA metabolism with maize seed germination

Previous studies have shown that GA plays an essential role inArabidopsis seed germination (Koornneef and van der Veen, 1980;Ogawa et al., 2003). In this study, seeds germination experiments wereconducted with treatments of water, paclobutrazol (50 μM) and GA3

(50 μM), with the purpose to investigate the effect of GA in seedgermination of maize. Growth patterns of germinating seeds weregraphed in Fig. 4. The results revealed that 12% seeds showed visibleradicle emergence after 24 h of imbibition in GA, but the percentage ofgerminated seeds treated with water and paclobutrazol were only 2%and 0%, respectively (Fig. 4). This superiority of GA treatment wasmaintained even after 56 hwhen all seeds have fully germinated. Thesegermination data indicated that exogenous GA is effective in enhancing

Fig. 4. The effect of GA3 and paclobutrazol on germination (%). Seeds were treated withwater (CK), solutions of paclobutrazol (PB; 50 μM) and GA3 (GA; 50 μM) at 30 h, 34 h,38 h, 42 h, 46 h, 52 h and 56 h after imbibitions.

Fig. 5. Expression analysis by real-time PCR of the maize GA meta

39J. Song et al. / Gene 482 (2011) 34–42

germination of seeds and reducing the de novo GA biosynthesis andplays an important role in inhibiting the germination of seeds.

To provide further insight into the relationship of de novo GAbiosynthesis with seed germination, the spatial and temporal expres-sion patterns of 27 maize GA metabolic genes were investigated byusing real-time PCR (Fig. 5). Transcripts of 15 GAmetabolic genes weredetected and 11 genes among them (ZmCPS1, ZmKS1, ZmKS4, ZmKO1,ZmKO2, ZmKAO, ZmGA20ox4, ZmGA3ox1, ZmGA2ox1, ZmGA2ox3 andZmGA2ox10) showed peak expression level at 72 h after imbibitions.Whereas the other 4 genes peaked at 24 h (ZmCPS2), 32 h (ZmKS2 andZmGA20ox5) and 40 h (ZmGA20ox4) after imbibitions, respectively.Moreover, the expression patterns of different members within thesame gene family also varied considerably. For example, the expressionlevel of ZmCPS1 genewasmuch higher than ZmCPS2, and the time pointof peak expression level was also different between ZmCPS1 andZmCPS2. It was worthy to note that the transcripts of 6 GA metabolicgenes (ZmCPS1, ZmCPS2, ZmKS1, ZmKS4, ZmKO1 and ZmGA20ox4) were

bolic genes during germination at 0 h–72 h after imbibitions.

Fig. 6. Expression analysis by real-time PCR of the maize GA metabolic genes during germination at 32 h after imbibitions treated with water (CK) and paclobutrazol (PB; 50 μM).

40J.Song

etal./

Gene

482(2011)

34–42

41J. Song et al. / Gene 482 (2011) 34–42

also detected in dry seed embryo, especially for ZmKS4 and ZmGA20ox4,which declined at the early stages of seed imbibitions, then peaked at72 h after imbibition.

Since paclobutrazol, as an inhibitor of GA biosynthesis (Nambaraet al., 1991; Jacobsen and Olszewski, 1993), restrained the maizegermination, the effect of paclobutrazol on the transcript level of GAmetabolic genes during seed germination was also examined. Asshown in Fig. 6, the relative mRNA abundance of ZmKAO, ZmGA20ox1,ZmGA20ox5 and ZmGA3ox1 genes increased in seed embryos at 32 hafter treatment with paclobutrazol, whereas transcripts of ZmCPS1,ZmCPS2, ZmKS1, ZmKS2, ZmKS4, ZmKO1 and ZmKO2 genes decreased.

4. Discussion

4.1. The evolution of GA metabolic gene families in maize

In the present study, we have isolated 27 candidate genes for sevenGA metabolic enzymes with an attempt to clarify the mechanism ofseed germination in maize. Although there was only one KAO gene inmaize genome, two to ten candidate genes were identified for theother six enzymes. Recently, four CPS, nine KS, and five KO genes wereidentified in rice genome, whereas the Arabidopsis genome containsonly one gene encoding each of CPS, KS or KO (Ogawa et al., 2003;Sakamoto et al., 2004; Sun and Kamiya, 1994). Similarly, maizegenome contains three CPS, four KS, and two KO genes, but theirarrangement differs from that of the OsCPS, OsKS and OsKO genes, andmost of them were arranged separately on the maize genome, withexceptions for ZmCPS2 and ZmKS1. The multiple copies of these maizegenes should be the GA biosynthetic genes, because the GAbiosynthetic genes are widely shared by monocot and dicot plants(Sakamoto et al., 2004). Therefore, the member expansion of maizeZmCPS, ZmKS, and ZmKO gene families reflected a succession of maizegenomic expansions due to extensive duplication and diversificationthat frequently occurred in the course of evolution as compared toArabidopsis (Wang et al., 2007).

The enzymes catalyzing the later steps are also encoded by genefamilies in maize, and the members for ZmGA20ox, ZmGA3ox, andZmGA2ox were 5, 2 and 10, respectively. In maize, dwarf-1 mutantblocks the conversion of GA20 to GA1, indicating that DWARF1 geneencoding that ZmGA3ox enzyme. Further analysis revealed that thelevel of GA in dwarf-1was less than 2% of that in normal shoots, whileGA20 and GA29 accumulated to levels over 10 times those in normals(Spray et al., 1996), suggesting the major enzyme (ZmGA3ox)involved in later step of GA biosynthesis in maize is likely to beencoded by single gene. On the contrary, to our best knowledge, noany genetic mutant with abnormal phenotype was identified forZmGA20ox and ZmGA2ox genes, indicating that their functions areredundant in the control of bioactive GA levels. Therefore, furthergenetic and biochemical studies on these genes would enhance ourcurrent understanding of the GA metabolic pathway in maize.

4.2. The regulation of gibberellins metabolic enzyme genes in maize

Up to date, most of genes encoding enzymes in all steps of the GAbiosynthetic and catabolic pathways havebeen identified, andmembersof these gene families are expressed differently, as determined bytranscript accumulation (Olszewski et al., 2002; Sakamoto et al., 2004).For example, two KAO genes are present in Arabidopsis, and both areexpressed in all tissues examined (Helliwell et al., 2001). In contrast, thecytosol-localizedGA20ox, GA3ox, andGA2ox each is encoded by a smallgene family, and each gene also shows tissue-specific expressionpatterns (Hedden and Phillips, 2000; Olszewski et al., 2002). In rice,OsGA3ox2 and OsGA20ox1 were broadly expressed in all of the organsthat were tested, but OsGA3ox1 and OsGA20ox3 was preferentiallyexpressed in the panicles (Sakamoto et al., 2004). In present study, theexpression patterns of 27 maize GA metabolic genes during seed

germination were investigated, but only transcripts of 15 GAmetabolicgenes were detected, suggesting that expression of the other genesmight be under tight developmental control. Taken together, thesestudies revealed that the complexmechanismsbywhich the amounts ofbioactive GAs are modulated in different tissues to regulate plantgrowth and development (Olszewski et al., 2002).

As an inhibitor of GA biosynthesis, paclobutrazol was thought tointeract directly with the ent-kaurene oxidase (KO) enzyme therebyreducing its activity (Sugavanam, 1984; Izumi et al., 1985). Recently,transgenic Arabidopsis with increased levels of one or both of the firsttwoGAbiosynthesis enzymes (CPS and KS), required for the conversionof geranyl geranyl diphosphate (GGDP) to the KO substrate ent-kaurene, have been characterized in terms of their response topaclobutrazol. Some lines contained levels of ent-kaurene 1000 timeshigher than WT plants, and these high KO substrate levels presumablyare responsible for the reduced effect of paclobutrazol. This explanationis consistentwith the observation that paclobutrazolmay competewithent-kaurene for access to the KO enzyme (Fleet et al., 2003). In addition,overexpression of KO gene in Arabidopsis also decreased paclobutrazolsensitivity resulting from increased KO protein levels (Swain et al.,2005). Interestingly, we found that the relative mRNA abundance ofZmCPS1, ZmCPS2, ZmKS1, ZmKS2, ZMKS4, ZmKO1 and ZmKO2 genesdecreased in seed embryos at 32 h after treatment with paclobutrazol.Although the underlyingmechanism is not known, in combination withthe observation that increased CPS, KS and KO levels decrease thesensitivity to KO inhibitors (Fleet et al., 2003; Swain et al., 2005 ), thedecreased expression of these genes may increase the sensitivity ofmaize seed embryo to paclobutrazol, and inhibit seed germination inturn. In addition, somedownstreamgenes ofZmKO(ZmKAO,ZmGA20ox1,ZmGA20ox5 and ZmGA3ox1) are up-regulated by paclobutrazol, whichmay be attributed to the feedback regulation of GA biosynthesis, asobserved in different GA response mutants (Bethke and Jones, 1998;Phillips, 1998; Hedden and Phillips, 2000).

4.3. The deduced role of GA metabolism in maize seed germination

Seed germination involves complex physiological processes thatare controlled by a range of developmental and external cues (Koornneefet al., 2002; Gallardo et al., 2001, 2002). In this study, we demonstratedthat the inhibitory effect of GA biosynthesis inhibitor, paclobutrazol,decreased maize seed germination, whereas exogenous GA treat-ment enhanced it, which provided further evidence for the notionthat increased synthesis of active GAs in the embryowas required fortriggering germination events (Ogawa et al., 2003).

In Arabidopsis, key GA biosynthesis transcripts such as AtGA3ox1 andAtGA3ox2 have been localized in the cortex and endodermis of theembryonic axis in germinating Arabidopsis seeds, suggesting active GAsynthesis in the embryo (Ogawa et al., 2003). Recently, Sreenivasulu et al.(2008) reported that the prominent peak of most of the barley GAbiosynthesis genes (KAO1, GA20ox2, and GA2ox1) in the embryo duringearly seed germination leads to the production of active GA necessary forgermination as measured by Jacobsen et al. (2002). In this study, wedemonstrated that the transcripts of 15 GA metabolic genes could bedetectedduring seedgermination,which consistentwith the idea thatGAplays an important regulatory role in seed germination (Sakamoto et al.,2004). Most notably, the transcripts of 6 GA metabolic genes (ZmCPS1,ZmCPS2, ZmKS1, ZmKS4, ZmKO1 and ZmGA20ox4) were detected in dryseed embryo, which could be regarded as stored mRNA. The presence ofthese GA biosynthetic transcripts in the dry embryo might contribute toconverting stored GA precursors into active GA at the early stages of seedgermination.

In Arabidopsis and barley, dynamic expression profile was detectedfor some key GA biosynthesis gene families during seed germination(Ogawa et al., 2003; Sreenivasulu et al., 2008). We also demonstratedthat a variety of temporal expression patterns of GAmetabolic genes. Forexample, expression level of ZmCPS1 genewasmuch higher than that of

42 J. Song et al. / Gene 482 (2011) 34–42

ZmCPS2, and the time point of peak expression was also differentbetween ZmCPS1 and ZmCPS2. Among the three expressed ZmGA20oxgenes during seed germination, the peak time for ZmGA20ox1,ZmGA20ox4 and ZmGA20ox5 was 32 h, 72 h and 40 h after imbibitions,respectively. Therefore, it appeared that subsets of these genesmight beregulated differently by dissimilar transcription factors, which revealedpart of the complexmechanismbywhichGAactivates seedgermination(Ogawa et al., 2003). Therefore, further investigation is to elucidate howGA signals are transmitted to a series of different cell types, thus affectseed germination in turn.

Supplementarymaterials related to this article can be found onlineat doi:10.1016/j.gene.2011.05.008.

Acknowledgements

This work was financially supported by the National Basic ResearchProgram of China (2007CB109000), the National Science Found forDistinguished Young Scholars (30925023), theNational Natural ScienceFoundation of China (30671297 and 30871577) and the 863 Project ofChina.

References

Bensen, R.J., Johal, G.S., Crane, V.C., Tossberg, J.T., Schnable, P.S., Meeley, R.B., Briggs, S.P.,1995. Cloning and characterization of the maize An1 gene. Plant Cell. 7, 75–84.

Bethke, P.C., Jones, R.L., 1998. Gibberellin signaling. Curr. Opin. Plant Biol. 1, 440–446.Bewley, J.D., 1997. Seed germination and dormancy. Plant Cell 9, 1055–1066.Busov, V.B., Meilan, R., Pearce, D.W., Ma, C., Rood, S.B., Strauss, S.H., 2003. Activation

tagging of a dominant gibberellin catabolism gene (GA 2-oxidase) from poplar thatregulates tree stature. Plant Physiol. 132, 1283–1291.

Davies, P.J., 1995. The plant hormones: their nature, occurrence, and functions. Planthormones. Kluwer Academic Publishers, Dordrecht, pp. 1–12.

Debeaujon, I., Koornneef, M., 2000. Gibberellin requirement for Arabidopsis seedgermination is determined both by testa characteristics and embryonic abscisicacid. Plant Physiol. 122, 415–424.

Fleet, C.M., et al., 2003. Overexpression of AtCPS and AtKS in Arabidopsis confersincreased ent-kaurene production but no increase in bioactive gibberellins. PlantPhysiol. 132, 830–839.

Gallardo, K., et al., 2001. Proteomic analysis of Arabidopsis seed germination andpriming. Plant Physiol. 126, 835–848.

Gallardo, K., Job, C., Groot, S.P.C., Puype, M., Demol, H., Vandekerckhove, J., Job, D., 2002.Proteomics analysis of Arabidopsis seed germination. A comparative study of wild-type and GA-deficient seeds. Plant Physiol. 129, 823–837.

Hedden, P., Phillips, A.L., 2000. Gibberellin metabolism: new insights revealed by thegenes. Trends Plant Sci. 5, 523–530.

Helliwell, C.A., Chandler, P.M., Poole, A., Olive, M.R., Dennis, E.S., Peacock,W.J., 2001. TheCYP88A cytochrome P450, ent-kaurenoic acid oxidase, catalyzes three steps of thegibberellin biosynthesis pathway. Proc. Natl Acad. Sci. U. S. A. 98, 2065–2070.

Hooley, R., 1994. Gibberellins: perception, transduction and responses. Plant Mol. Biol.26, 1529–1555.

Izumi, K., Kamiya, Y., Sakurai, A., Oshio, H., Takahashi, N., 1985. Studies of sites ofaction of a new plant growth retardant (E)-1-(4-chlorophenyl)-4, 4-dimethyl-2-

(1, 2, 4-trizol-1-yl)-penten-3-ol (S-3307) and comparative effects of its stereisomersin a cell free system of Cucurbits maxima. Plant Cell Physiol. 29, 821–827.

Jacobsen, S.E., Olszewski, N.E., 1993. Mutations at the SPINDLY locus of Arabidopsis altergibberellin signal transduction. Plant Cell 5, 887–896.

Jacobsen, J.V., Pearce, D.W., Poole, A.T., Pharis, R.P., Mander, L.N., 2002. Abscisic acid,phaseic acid and gibberellin contents associatedwith dormancy and germination inbarley. Physiol. Plant. 115, 428–441.

Kamiya, Y., García-Martínez, J., 1999. Regulation of gibberellin biosynthesis by light.Curr. Opin. Plant Biol. 2, 398–403.

Katsumi, M., 1964. Gibberellin-like Activities of Certain Auxins and Diterpenes. PhDDissertation. University of California, Los Angeles.

Koornneef, M., Karssen, C.M., 1994. Seed dormancy and germination. In: Meyerowitz, E.M., Somerville, C.R. (Eds.), Arabidopsis. Cold Spring Harbor Laboratory Press, NewYork, USA, pp. 313–334.

Koornneef, M., van der Veen, J.H., 1980. Induction and analysis of gibberellin sensitivemutants in Arabidopsis thaliana (L.) Heynh. Theor. Appl. Genet. 58, 257–263.

Koornneef, M., Bentsink, L., Hilhorst, H., 2002. Seed dormancy and germination. Curr.Opin. Plant Biol. 5, 33–36.

Kumar, S., Nei, M., Dudley, J., Tamura, K., 2008. MEGA: a biologistcentric software forevolutionary analysis of DNA and protein sequences. Brief. Bioinform. 9,299–306.

Lange, T., 1998. Molecular biology of gibberellin synthesis. Planta 204, 409–419.Miransari, M., Smith, D., 2009. Rhizobial lipo-chitooligosaccharides and gibberellins

enhance barley (Hordeum vulgare L.) seed germination. Biotechnology 8,270–275.

Nambara, E., Akazawa, T., McCourt, P., 1991. Effects of the gibberellin biosyntheticinhibitor uniconazol on mutants of Arabidopsis. Plant Physiol. 97, 736–738.

Ogawa, M., Hanada, A., Yamauchi, Y., Kuwahara, Y., Kamiya, Y., Yamaguchi, S., 2003.Gibberellin biosynthesis and response during Arabidopsis seed germination. PlantCell 15, 1591–1604.

Olszewski, N., Sun, T.-p., Gubler, F., 2002. Gibberellin signaling: biosynthesis, catabolismand response pathways. Plant Cell 14, 61–81.

Phillips, A., 1998. Gibberellins in Arabidopsis. Plant Physiol. Biochem. 36, 115–124.Richards, D.E., King, K.E., Aitali, T., Harberd, N.P., 2001. How gibberellin regulates plant

growth and development: a molecular genetic analysis of gibberellin signaling.Annu. Rev. Plant Physiol. Plant Mol. Biol. 52, 67–88.

Sakamoto, T.K., et al., 2004. An overview of gibberellin metabolism enzyme genes andtheir related mutants in rice. Plant Physiol. 134, 1642–1653.

Sasaki, A., et al., 2002. A mutant gibberellin-synthesis gene in rice. Nature 416,701–702.

Spray, C.R., Kobayashi, M., Suzuki, Y., Phinney, B.O., Gaskin, P., MacMillan, J., 1996. Thedwarf-1 (d1) mutant of Zea mays blocks three steps in the gibberellin-biosyntheticpathway. Proc. Natl Acad. Sci. U. S. A. 93, 10515–10518.

Sreenivasulu, N., et al., 2008. Barley grain maturation and germination: metabolicpathway and regulatory network commonalities and differences highlighted bynew MapMan/PageMan profiling tools. Plant Physiol. 146, 1734–1761.

Sugavanam, B., 1984. Diastereoisomers and enantiomers of paclobutrazol: theirpreparation and biological activity. Pest. Sci. 15, 296–302.

Sun, T.-p., Kamiya, Y., 1994. The Arabidopsis GA1 locus encodes the cyclase ent-kaurenesynthetase A of gibberellin biosynthesis. Plant Cell 6, 1509–1518.

Swain, S.M., Singh, D.P., Helliwell, C.A., Poole, A.T., 2005. Plants with increasedexpression of ent-kaurene oxidase are resistant to chemical inhibitors of thisgibberellin biosynthesis enzyme. Plant Cell Physiol. 46, 284–291.

Wang, D., et al., 2007. Genome-wide analysis of the auxin response factors (ARF) genefamily in rice (Oryza sativa). Gene 394, 13–24.

Winkler, R.G., Helentjaris, T., 1995. Themaize Dwarf3 gene encodes a cytochrome P450-mediated early step in Gibberellin biosynthesis. Plant Cell 7, 1307–1317.

Yamaguchi, S., Kamiya, Y., 2000. Gibberellin biosynthesis: its regulation by endogenousand environmental signals. Plant Cell Physiol. 41, 251–257.