A

osdgOcrrg©

K

1

ilcasr1AacIbcia

f

0d

Postharvest Biology and Technology 45 (2007) 184–192

Multiple forms of �-expansin genes are expressed duringbanana fruit ripening and development

Asha, Vidhu A. Sane, Aniruddha P. Sane, Pravendra Nath ∗Plant Gene Expression Lab, National Botanical Research Institute, Lucknow 226001, India

Received 3 November 2006; accepted 7 March 2007

bstract

Fruit ripening and softening involve depolymerization of complex cell wall components. The complexity of fruit softening through involvementf several cell wall modifying genes has been well documented. More than one class of enzymes and proteins are involved in the process ofoftening and these may either act sequentially or synergistically. In this paper we provide further evidence of the complexity of fruit softening byemonstrating the simultaneous expression of multiple genes of the same family during softening. We report the identification of four �-expansinenes, MaEXPA2, MaEXPA3, MaEXPA4 and MaEXPA5 from banana fruit which express differentially during fruit development and ripening. TheRFs range from 750 to 774 bp in size. All four genes also possess two introns with variable sizes ranging from 75 to 343 nucleotides that are

onserved in position. All the four genes were expressed during the course of ripening. Of these, MaEXPA2 was fruit specific and showed a strong

ipening related and ethylene dependent expression in banana fruit. MaEXPA4 expressed both during fruit growth as well as ripening and might beelated to expansion. MaEXPA3 and MaEXPA5 were also expressed in tissues other than fruit. It is concluded that expression of multiple expansinenes might be required for softening not only in dicot fruit such as pear, apple and strawberry but also in monocot fruit such as banana. 2007 Elsevier B.V. All rights reserved.

ylene

Elpdebafe�awut

eywords: Expansin; Expansin gene family; Banana; Softening; Ripening; Eth

. Introduction

Fruit softening during ripening is associated with depolymer-zation and solubilization of cellulose microfibrils, hemicellu-ose and pectins in the cell wall. A wide range of hydrolases thatatalyze the process of softening are activated at transcriptionalnd translational levels. However, various molecular and genetictudies suggest that these hydrolases may not be individuallyesponsible for softening (Giovannoni et al., 1989; Smith et al.,988; Brummell et al., 1999a; Brummell and Harpster, 2001).

synergistic action appears to be more plausible for bringingbout softening in fruit tissue. The foremost action in this pro-ess may be started by a family of proteins known as expansins.t has been suggested that expansins disrupt the hydrogen bonds

etween cellulose microfibrils and matrix polysaccharides of theell wall thereby increasing the accessibility of cell wall mod-fying enzymes such as various hydrolases (McQueen-Masonnd Cosgrove, 1995; Rose and Bennett, 1999; Li et al., 2003).

∗ Corresponding author. Tel.: +91 522 2205841x331;ax: +91 522 2205836/2205839.

E-mail address: pravendranath@hotmail.com (P. Nath).

tret1otP

925-5214/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.postharvbio.2007.03.003

xpansins were first described in expanding/growing cells butater identified from a number of plant species for several otherrocesses. They play roles in many aspects of plant growth andevelopment such as cell enlargement, germination, fruit soft-ning, abscission, stress response and parasitism, etc. (reviewedy Cosgrove, 2000, 2005; Choi et al., 2006). These proteinsre encoded by a super family of genes that are divided intoour families namely �-expansin (EXPA), �-expansin (EXPB),xpansin like A (EXLA) and expansin like B (EXLB). Of these,- and �-expasins have the ability to extend cell walls (Li etl., 2002; Sampedro and Cosgrove, 2005; Choi et al., 2006)hereas the actual biological function of EXLA and EXLB is yetnknown. Fruit specific expansins have been characterized fromomato, pear, strawberry, banana and mango, etc. LeEXPA1 washe first fruit specific tomato expansin, which was related toipening (Rose et al., 1997). The expansin genes present in fruitxpress differentially with distinct expression patterns correlatedo either fruit size or ripening stage of fruit (Brummell et al.,

999a; Hiwasa et al., 2003). Transgenic studies have shown thatver-expression or suppression of specific expansin genes alteromato fruit texture and juice viscosity (Brummell et al., 1999b;owell et al., 2003).

and

esapreesa(bafwioc

2

2

Clt2f(Hw1Mtfalahaft(ba

2

fwt0mlk

(tcecRa(dgA

2

pdptTatfffcct(u(i(a(afispggsegdiAUfrwtT

Asha et al. / Postharvest Biology

We have been interested in the process of ripening and soft-ning in banana fruit and in employing genetic tools to extend itshelf life post ripening. Interference in the activity of genes thatre responsible exclusively for softening appears to be a goodossibility to extend shelf life as this may not affect the process ofipening as such. Since expansins have been shown to express inarly stages of softening in many fruit (Rose et al., 1997; Hiwasat al., 2003; Obenland et al., 2003; Sane et al., 2005), suppres-ion of their action is likely to be effective in reducing softenings has already been demonstrated for tomato by Brummell et al.1999b). No comprehensive study was available on expansins ofanana fruit. In this paper, we report the identification and char-cterization of four new members of the expansin gene familyrom banana fruit [Musa acuminata (AAA), cultivar Harichal],hich express differentially during fruit development and ripen-

ng. These genes along with one previously identified memberf this family (Trivedi and Nath, 2004) provide evidence foromplexity of the softening process in banana.

. Materials and methods

.1. Plant material, treatment and sampling of tissue

Hands of mature unripe banana (M. acuminata, Dwarfavendish, Genome AAA, var. Harichal) were obtained from a

ocal orchard. Ripening in fruit was initiated by exposing theseo 100 �L/L ethylene in an airtight 10 L container for 24 h at2 ◦C in the dark and then they were allowed to ripen in airor another 7 days without any exogenous ethylene. For 1-MCPethylene perception inhibitor, EthylBloc, Biotechnologies fororticulture Inc., Walterboro, SC, USA) treatment, banana fruitere sealed in an airtight container and exposed to 40 �L/L of-MCP. For this 280 mg EthylBloc powder (containing 0.43% 1-CP) was placed in a tube in the 10 L container, water was added

o release 1-MCP gas and the container was immediately sealedor 24 h at 22 ◦C, followed by ethylene treatment as describedbove. Fruit firmness was measured each day as described ear-ier (Lohani et al., 2004; Trivedi and Nath, 2004) and considereds the criteria for the progression of ripening. Pulp tissue wasarvested every 24 h after treatment, frozen in liquid nitrogennd stored at −70 ◦C till further use. Pulp was also harvestedrom control fruit kept in air without ethylene and/or 1-MCPreatment. Four stages of developing unripe fruit D1 (6 cm), D210 cm), D3 (14 cm), D4 (18 cm), spaced equally on the bananaunch were also included in the study along with leaf, flowernd stem, pseudostem and bract tissues for expression analysis.

.2. DNA extraction

Genomic DNA was isolated from young leaves of bananaollowing the method of G.D. May (personal communication)ith slight modifications. Briefly, 20 mL pre-warmed extrac-

ion buffer (0.2 M Tris–Cl pH 8.0, 1% sarcosyl, 0.8 M NaCl,

.02 M EDTA, 0.8%CTAB, 0.14 M mannitol and 0.14% �-ercaptoethanol) was added to 2 g leaf tissue pulverized in

iquid N2. An equal volume of chloroform was added to it andept for 20 min at 65 ◦C. After centrifugation of the mixture

Mtpw

Technology 45 (2007) 184–192 185

6000 × g, 5 min), an equal volume of isopropanol was addedo the aqueous phase. The pellet of nucleic acid obtained afterentrifugation (10,000 × g, 10 min, RT) was washed with 70%thanol. Nucleic acid pellet was suspended in 1 mL TE andontaminating RNA was removed by treating with DNase freeNase followed by phenolization. The DNA was dissolved inppropriate volumes of TE buffer. Aliquots of genomic DNA20 �g) were digested with EcoRI, EcoRV HindIII and XbaI andigested samples were electrophoresed on a 0.8% agarose-TAEel. DNA was transferred to a nylon membrane (Hybond-N,mersham Pharmacia Biotech, UK).

.3. Cloning and sequencing of banana expansins

Total RNA was extracted from 5 g pulp tissue (pooled sam-le) of fourth day post ethylene treated banana using CTAB asescribed by Asif et al. (2000). DNA free RNA was used torepare cDNA. All the primer sequences, which have been usedo isolate different expansins are given in Table 1. The Reverse-ranscriptase reaction was carried out using 3′APRACE primersnd MMuLV reverse transcriptase from Invitrogen according tohe manufacturer’s instructions. To amplify the expansin generagments from banana, degenerate forward and reverse primersor expansin (ExpF1 and ExpR1) were used. PCR amplifiedragments were cloned in pBluescript II KS. Sequencing wasarried out on an automated sequencer (ABI 373A) using theycle sequencing dye terminator kit (ABI). Based on the par-ial sequence, gene specific primers were designed for 3′RACEMaEXP2-F3) and this in combination with 3′AP primer wassed to obtain the 3′-end. For 5′RACE, gene specific primersMaEXP2-R3 and MaEXP2-R4) were designed and used tosolate the 5′RACE product in combination with 5′AP primersInvitrogen). Based on nucleotide sequence of these fragmentscomposite sequence was constructed. Two specific primers

F0 and R0) were synthesized out of composite sequence andfull-length cDNA fragment of 768 bp fragment was ampli-

ed, cloned and sequenced and designated as MaEXPA2. Theame primer pair was used with banana genomic DNA as tem-late, to get the information about introns in the MaEXPA2ene. MaEXPA3 was obtained using EXPF1 and EXPR1 andenomic DNA as template. The 700 bp long fragment wasequenced and specific primer (MaEXP3R) was designed forxtension towards 5′-end. Based on sequence of this fragment,ene specific primers (MaEXPA3Fo and MaEXPA3F2) wereesigned to obtain the genomic sequence by genome walkingn combination with genome walker adapter primers (AP1 andP2). For this a genome walker library was constructed usingniversal genome walker kit (Clontech) according to manu-

acturer’s protocol. Based on the genomic sequence, specificeverse primer MaEXPA3R0 (containing termination codon)as designed and used in combination with MaEXPA3F0 to get

he cDNA fragment containing the entire ORF for MaEXPA3.wo other closely related partial expansin genes MaEXPA4 and

aEXPA5 were obtained during extension of the upstream por-

ion of MaEXPA3. Based on their sequences, specific forwardrimers (MaEXPA4F1, F2 and MaEXPA5F1, F2 respectively)ere designed and used in combination with Adapter primers

186 Asha et al. / Postharvest Biology and Technology 45 (2007) 184–192

Table 1Nucleotide sequences of different primers used to isolate various banana expansins (details are given in Section 2)

3′APRACE primer 5′-GGCCACGCGTCGACTAGTACTTTTTTTTTTTTTTTTT-3′EXPF1 5′-GCH TCD GGM ACH ATG GDT GGD GCD TGT GG-3′EXPR1 5′-CCR TTK AYB GTR AAY CKD ATH CCT CC-3′MaEXPA2-F2 5′-GTA CGG CAT CAA AAA CAC GGC ACT-3′5′APRACE primer 5′-GGCCACGCGTCGACTAGTACGGGIIGGGIIGGGIIG-3′MaEXPA2-R3 5′-GGC AGT AGT TGG TGG CGG TGA CGG-3′MaEXPA2-R4 5′-CCG TGT TTT TGA TGC CGT ACC CGT C-3′MaEXPA2-F0 5′-TCG AGC AAG GAT CCA TGG AGA AGG CTG CA-3′MaEXPA2-R0 5′-ACA GCG CAA GCT TCT AAA ACC GGA TGG AG-3′GWAP1 5′-GTAATACGACTCACTATAGGGC-3′GWAP2 5′-ACTATAGGGCACGCGTGGT-3′MaEXPA3R 5′-CAG AAG TTG GTA GCC GTG ATG AA-3′MaEXPA3F0 5′-ACC CTT GAT CGC GAG TGA GAG AGG AGC A-3′MaEXPA3F2 5′-ACA CTT CGA CCT CTC CAT GCC CAT G-3′MaEXPA3R0 5′-AGG CGG GAA GGA GTT TGA GCT TGG AGA-3′MaEXPA3R3 5′-AGCGGTGCTCAGCGCCGCCGTCTCC-3′MaEXPA4F1 5′-AGA TGG CCT CCT TCG TCC CTG TCG TCT-3′MaEXPA4F2 5′-GCG GGG CGC ATT CCC GGC GTC TAC A-3′MaEXPA4R0 5′-AGA GTG GAG CAG CAT TCG AGT TGC AG-3′MaEXPA5F1 5′-AGT CTC CGT CGC CCT TCT GGC TCT TC-3′MaEXPA5F2 5′-ACT GCG CGC ATT CCC GGT TTG TAC G-3′MaEXPA5R0 5′-AGT GGA GAA GTT CTC GCG GGA ATG GGT-3′MaEXPA2ProR 5′-GCA GTA GTG GAT CCT TCT CCA TCT C-3′

(Miac5

maac

2

fNa(oube

2

(aeR(

wbtwPf5c(d(apt3bawsaisfabtw

AP1 and AP2) to obtain the complete genomic sequences ofaEXPA4 and MaEXPA5. Specific reverse primers (contain-

ng termination codon) for MaEXPA4 and 5 were then designednd used in combination with respective forward primers to getDNA fragments containing the entire ORF for MaEXPA4 and, respectively.

The Genome walker library was also used to isolate pro-oter of MaEXPA2 using gene specific primer MaEXP2-R4

nd MaEXPA2proR and Genome walker adapter primer AP1nd Ap2. A 0.8 kbp long fragment containing the promoter wasloned in TOPO vector (Invitrogen) and sequenced.

.4. Sequence analysis

Homology searches and signal peptide predictions were per-ormed using the Basic Local Alignment Search Tool (BLAST,ational Center for Biotechnology Information, Bethesda, MD)

nd SIGNALP version 1.1 World Wide Web Prediction ServerCenter for Biological Sequence Analysis, Technical Universityf Denmark), respectively. Multiple alignments were carried outsing CLUSTAL W program. Phylogenetic analyses of putativeanana expansins were carried out using Phylip 3.5c packagemploying parsimony and bootstrap analysis (100 replicates).

.5. Northern and Southern analysis

Total RNA was isolated from banana pulp from controlethylene untreated), different days post ethylene treated fruit

nd 1-MCP treated fruit. RNA from fruit pulp from differ-nt developmental stages and vegetative tissues were isolated.NA gel blot analyses were carried out as previously described

Sane et al., 2005). The 40 �g total RNA from each sample

t5sM

as resolved on a 1.2% formaldehyde agarose gel as describedy Sambrook et al. (1989) and modified in the Qiagen Oligo-ex handbook (2002) protocol for RNA electrophoresis. RNAas transferred to Hybond-N nylon membrane (Amersham-harmacia, Biotech, UK). Blots were pre-hybridized at 42 ◦Cor 6–8 h in hybridization buffer (50% formamide, 6× SSC,× Denhardt’s solution, 0.1% SDS and 100 �g/mL denaturedalf thymus DNA). A 216 bp long MaEXPA2 cDNA fragmentcontaining 5′-region) was labeled with � 32P-dCTP using ran-om priming method and used as a probe for northern blotsSambrook et al., 1989). Hybridization was carried out overnightt 42 ◦C in fresh hybridization buffer containing the radiolabeledrobe. Other probes were prepared using the variable regions ofhe 5′-end of each gene: for MaEXPA3, 265 bp; for MaEXPA4,33 bp; for MaEXPA5 238 bp. These fragments were labeledy the random priming method as described by Sambrook etl. (1989). Cross-hybridization experiments using these probesere carried out to ensure that the probes identify specific tran-

cript/DNA fragment (data not shown). The same probes werelso used for Southern hybridization. Washings were performedn 2× SSC (saline sodium citrate), 0.1% SDS (sodium dodecylulphate) for 2 × 5 min, 0.5× SSC, 0.1%SDS for 2 × 20 min,ollowed by 0.1× SSC, 0.1%SDS for 2 × 5 min at room temper-ture. The autoradiograms were developed after exposing thelots to X-ray films (Fuji) at −70 ◦C for 3–4 days. Hybridiza-ion experiments for each gene were replicated at least thriceith different RNA samples.To study the expression pattern of MaEXPA3, semi quan-

itative PCR was carried out using cDNA prepared from00 ng DNA free RNA from different fruit and vegetative tis-ues. PCR was performed using the primers MaEXP3AF2 and

aEXPA3R3 and actin as internal control.

and Technology 45 (2007) 184–192 187

3

3b

coN(os

M(egCatmMcow

Table 2Sizes of various banana expansin ORFs and their predicted proteins (sizes ofphase I and II introns present in four banana expansin genes are also mentioned)

Gene ORF (bp) Intron 1(bp)

Intron 2(bp)

Protein(aa)

MaEXPA2 750 75 78 250MaEXPA3 750 85 343 250MM

Mtcfiaaci

FbMM

Asha et al. / Postharvest Biology

. Results

.1. Identification, sequencing and phylogenetic analysis ofanana expansins

Four banana �-expansins were identified using fruit pulpDNA and genomic DNA as templates and utilizing vari-us techniques like RT-PCR, RACE and genome walking.ucleotide sequence identity amongst these four and MaEXP1

identified earlier in our lab) was in the range of 70–90%ver the equivalent stretch of nucleotide sequence (data nothown).

The encoded putative proteins of MaEXPA2 (250 aa),aEXPA3 (250 aa), MaEXPA4 (257 aa) and MaEXPA5

258 aa) were all essentially conserved with distinct features ofxpansins such as HFD domain (catalytic site of the family 45-lycosyl hydrolases) and conserved tryptophan (W) residues at-terminal portions (Fig. 1). All the putative proteins containedn N-terminal signal sequence, which probably targets the pro-ein to cell walls and have been shown to be present in other

embers of �-expansin families in other plants. In addition,

aEXPA2, 4 and 5 also possessed eight Cys residues that are

onserved in other expansins. However, MaEXPA3 possessednly seven Cys-residues while the last cysteine was replacedith valine at amino acid 159.

7p

d

ig. 1. Comparison of the deduced full-length amino acid sequence of five banana elack box and grey box, respectively. Gaps are shown as dashes; asterisk (*) indicatultiple alignments was done by CLUSTAL-W and viewed with BOXSHADE progaEXPA4 (EF213102), MaEXPA5 (EF213103).

aEXPA4 774 78 99 258aEXPA5 771 99 102 257

The sizes of genomic clones for MaEXPA2, MaEXPA3,aEXPA4 and MaEXPA5 were 1.0, 1.2, 1.0 and 1.2 kb, respec-

ively. Sequence comparison between cDNA and genomiclones revealed the presence of two introns, which is a commoneature of many �-expansin genes. The positions of both thentrons were conserved in all the four genes and were all locatedt a nucleotide position corresponding to the equivalent aminocid in aligned protein sequence. The introns were flanked byonsensus intron–exon splice sites (Choi et al., 2006). The firstntron ranged from 75 to 99 nt and second intron varied from8 to 343 nt (Table 2). The largest phase II intron (343 nt) was

resent in MaEXPA3.

A phylogenetic tree was generated from the alignment of theeduced amino acid sequences of five banana expansins with

xpansins with each other. Identical and similar amino acids are represented byes conserved cysteine residues. Putative signal sequences are shown in italics.ram. MaEXP1 (AY083168), MaEXPA2 (AF539540), MaEXPA3 (AY222302),

188 Asha et al. / Postharvest Biology an

Fig. 2. Phylogenetic analysis of the deduced amino acid sequencesof multiple banana expansins with other �-expansins. Full-length pro-tein sequences were aligned using CLUSTAL-W and a phylogenetictree was constructed using PHYLIP with the PROTPARS program. APhleum pollen allergen (Phlp1), identified as �-expansin, was used asoutgroup. Sequences are GenBank accession numbers: At (Arabidopsisthaliana), AtEXPA2 (U30481), AtEXPA6 (U30480); Cs (Cucumis sativus),EXPA1 (U30382), EXPA2 (U30460), EXPA6 (AAL31477); Fa (Fragariaananassa), EXPA2 (AF159563), EXPA4 (ABA62612); Le (Lycopersiconesculentum), EXPA1 (U82123), EXPA4 (AAD13632), EXPA9 (AJ243340),EXP18 (CAA06271); Pce (Prunus cerasus) EXPA1 (AAK48845), EXPA2(AAK48846), EXPA4 (AAK48848), EXPA5 (AAL40354); Pc (Pyrus commu-nis), EXPA1 (BAC67188), EXPA2 (BAC67189), EXPA3 (BAC67190), EXPA6(BAC67193), EXPA7 (BAC67194); Pp (Prunus persica), EXPA2 (AB047518),EXPA3 (AB047519); Ma (Musa acuminata), EXPA1 (AY083168); Md (Malusdomestica), EXPA1 (AY083166); Mi (Mangifera indica), EXPA1 (AY600964);Os (Oryza sativa), EXPA1 (Y07782), OsEXPA (AA018447), EXPA2 (U30477);Pa (Prunus armeniaca), EXPA1 (U93167), EXPA2 (AF038815); Ps (Pisumsativum), EXPA (X85187); Phlp1 (Phleum pratense), pollen allergen (X78813).MaEXPA1, MaEXPA2 (AF539540, EF213104), MaEXPA3 (AY222302),Ms

o(to(

3r

sTfneo(ld5Nwpeg5Mtoeetofruit (Fig. 3A). Unlike MaEXPA4, transcript accumulation ofMaEXPA5 was restricted to fruit ripening. Transcripts beganaccumulating from day 3 post ethylene treatment and contin-

Fig. 3. Transcript abundance of multiple expansins in banana after ethyleneand 1-MCP treatments. (A) MaEXPA2, 4 and 5 accumulation after differentdays of post ethylene treatment. E0: control fruit without ethylene treatment;1–7: days after ethylene treatment; M0, 3 and 6: days after 1-MCP treatment.(B) Transcript abundance of banana expansin genes during fruit growth and indifferent vegetative tissues. D1–IV: developing stages of banana fruit (in order

aEXPA4 (EF213102) and MaEXPA5 (EF213103, EF213105) have beenhown in bold.

ther expansin homologues using a pollen allergen as outgroup

Fig. 2). MaEXPA3, MaEXPA4 and MaEXPA5 aligned withinhe same sub-lineage group. Interestingly, MaEXPA2 alignedutside the principal groups defined by Link and Cosgrove1998).

ospbh

d Technology 45 (2007) 184–192

.2. Differential accumulation of banana expansins duringipening

Transcript accumulation of different expansins in banana wastudied during the course of ethylene induced ripening for 7 days.he ripening/softening was characterized by a sharp decrease in

ruit firmness. There was almost a 40% decrease in fruit firm-ess from days 0 to 1 which decreased to 90% on day 2 (postthylene treatment) and after that there was a slow but continu-us decrease in firmness till day seven as previously describedLohani et al., 2004; Trivedi and Nath, 2004). RNA was iso-ated from the pulp of post ethylene treated banana on differentays and probed with different expansin probes containing the′-UTR of the gene, which was variable in all expansin genes.o transcript was detected on day 0 in ethylene untreated fruithen the 216 bp partial fragment of MaEXPA2 was used as arobe. The transcript started accumulating from day 2 in postthylene treated fruit and increased substantially with the pro-ression of ripening with maximum transcript levels on daysand 6. On day 7 there was a decline in transcript levels ofaEXPA2 (Fig. 3A). No transcript was detected in 1-MCP pre-

reated fruit on days 3 and 6, consistent with its inhibitory effectn softening (Lohani et al., 2004). The results showed that thisxpansin was related to ethylene triggered ripening of fruit. Thexpression of the other three-expansin genes was low comparedo MaEXPA2. MaEXPA4 showed transcript accumulation notnly throughout ripening but also in unripe and 1-MCP treated

f increasing size) as described in Section 2. R, S, L, B, Ps and F represent root,tem, leaf, bract, pseudostem and flower tissues, respectively. E0, 3 and 6: daysost ethylene-treated sample. The ethidium bromide-stained ribosomal RNAands in both the panels are shown to indicate RNA loading. All the treatmentsave been described in Section 2.

and Technology 45 (2007) 184–192 189

ueTlf

efloteiTfsoetcaaeTbusi

3

pw

Fig. 4. Expression pattern of MaEXPA3 gene. Total DNA free RNA was used asa template for semiquantitative PCR analysis using MaEXPA3 specific primers.A 500 bp banana Actin fragment was amplified using same RT product as inter-nal control. Amplified products were separated on 1.5% Agarose gel and stainedwith ethidium bromide. (A) Transcript accumulation during different stages ofethylene induced ripening. E0: control fruit without ethylene treatment; 1–7:days after ethylene treatment; +C: cloned MaExpA3 fragment used as a positivecontrol for PCR. (B) mRNA abundance of MaEXPA3 in vegetative tissues andduring fruit growth. L, S, Br, F, represent leaf, stem, bract and flower, respec-td

rewaMg5b

F(H

Asha et al. / Postharvest Biology

ed till the completion of the seven day time course study. Noxpression was observed in control and 1-MCP treated fruit.hus MaEXPA5, like MaEXPA2, appeared to be related to ethy-

ene induced ripening though its transcript levels did not changerom days 3 to 7, unlike MaEXPA2 (Fig. 3A).

Besides fruit ripening, the mRNA abundance of thesexpansins was also studied in different tissues such as leaf,ower, stem, pseudostem, bract and root and in different devel-ping stages of the unripe fruit. As shown in Fig. 3B, MaEXPA2ranscript accumulation was not detected in any other tissuexcept for ripening fruit. Expression was also not observed dur-ng the developing/expanding phase of the unripe fruit (Fig. 3B).hese results indicated that the expression of MaEXPA2 gene is

ruit specific and ripening related. Unlike MaEXPA2, the tran-cript of MaEXPA4 appeared to be present during various stagesf fruit development and increased with increase in size. Faintxpression was also observed in leaves but not in other vegetativeissues (Fig. 3B). The expression of MaEXPA5 transcript was notonfined only to ripening tissue but mRNA accumulation waslso observed in other vegetative tissues like root, stem, bractnd flower except for leaf. Expression in these tissues was how-ver lower as compared to expression in ripening fruit (Fig. 3B).ranscript accumulation for MaEXPA3 could not be detectedy northern hybridization. However, semi quantitative RT PCR,sing MaEXPA3 specific primers, revealed its expression in latertages of ripening (Fig. 4A). The expression was also observedn leaf and stage II of fruit development (Fig. 4B).

.3. Southern blot analysis

Southern blot analysis was carried out to determine the com-lexity of the expansin gene family in banana. DNA blot analysesith different probes, under high stringency washing, revealed

eatr

ig. 5. Southern blot analysis of MaEXPA2, 3, 4 and 5. Genomic DNA (20 �g per laH) and XbaI (X). Gel blots were hybridized with � 32P-dCTP-labelled probes prepybridization and washing conditions are described in Section 2. Molecular weight m

ively. I–IV: developing stages of banana fruit (in order of increasing size) asescribed in Section 2.

estriction fragments of different sizes for all the four bananaxpansins (Fig. 5). The Southern blot of banana genomic DNAhen hybridized with MaEXPA2 probe (containing the 5′ vari-

ble region), showed a single major band. This indicated thataEXPA2 is present in the banana genome as a single copy

ene. Hybridization of the genomic blot with MaEXPA3, 4 andspecific probes also resulted in only one or two major bands

ut also a number of fainter bands. This suggested that although

ach of these genes was present as a single copy, there were prob-bly other closely related genes. Sequence comparison showedhat MaEXPA3, 4 and 5 are closely related and confirmed theesults of Southern hybridization.

ne) was digested with the restriction enzymes EcoRI (E), EcoRV (EV), HindIIIared from the 5′ regions of each expansin cDNA by random priming method.

arkers are indicated in kilobases.

1 gy an

3

otwttn(pG2a

4

siBftsgaaAiIeis2ase

iMeSrcec2ttdpiethnt

tfh(tpeoatr3wo(a(dhl5dl

MriiblaiopsatsosMbctbetdiiwsd

90 Asha et al. / Postharvest Biolo

.4. MaEXPA2 promoter

Using Genome walker library, an 800 bp promoter fragmentf ripening related and fruit specific MaEXPA2 was isolated. Theranscription start site was located approximately 37 bp upstreamith respect to the start codon. The sequence analysis suggested

hat the TATA box was present at position 33 with respect to theranscription start site (data not shown). Interestingly, we couldot find any homology with the ethylene responsive elementERE—ATTTCAAA) in the MaEXPA2 promoter. However, autative GCC box (AGCCGGC) that differed from the classicalCC box (AGCCGCC) at one variable position was observed08 nt upstream with respect to the transcription start site withchange of one nucleotide.

. Discussion

The complexity of fruit softening through involvement ofeveral cell wall modifying genes has been well documentedn previous studies (Giovannoni, 2001; Harrison et al., 2001;ennett, 2002; Hiwasa et al., 2003). The present study provides

urther evidence of this complexity by demonstrating the simul-aneous expression of multiple genes of the same family duringoftening. We have identified five members of the �-expansinene family MaEXPA1-MaEXPA5 (present study and Trivedind Nath, 2004) from banana fruit that express during ripeningnd their sequences shared characteristic features of expansins.nalysis of genomic sequences revealed the presence of two

ntrons with conserved sites (Choi et al., 2006). Phase I and phaseI introns in all the expansins were in the range of 80–100 bpxcept for MaEXPA3, which had a relatively larger phase IIntron of 343 nt. Large introns have also been reported in otherpecies such Arabidopsis, cotton, etc. (Sampedro and Cosgrove,005). An N-terminal signal sequence (ranging from 20 to 24mino acids) was present in all the members identified in thistudy and indicated that they are targeted to cell walls like otherxpansins.

Although essential features of expansins were conservedn the five banana expansins, the putative protein encoded by

aEXPA3 possessed only seven cysteine residues instead ofight that are present in other �-expansins (Cosgrove, 2000).ubstitution of the conserved Cys residues with Arg and Gly,espectively in GhEXP2 and GhEXP6 has been described inotton (Harmer et al., 2002) while the encoded protein of �-xpansin gene of Striga asiatica, lacks the fourth and fifthonserved cysteine though it is transcribed (O’Malley and Lynn,000). Whether the absence of one Cys residue has some func-ional significance is yet not known. Of all the five expansins,he expression of MaEXPA3 was the lowest and could only beetected by RT-PCR. Further studies related to stability of eachrotein and their extension abilities might provide some insightnto the significance of conserved Cys residues. The five bananaxpansins were grouped in different clades of the phylogenetic

ree. Out of these, MaEXPA3, MaEXPA4 and MaEXPA5 shareigh sequence homology not only at protein level but also atucleotide level and were clustered together in the phylogeneticree. Nevertheless, they differ in their expression patterns. In

twPo

d Technology 45 (2007) 184–192

omato fruit at least six expansin genes are expressed duringruit development. Of these, three (LeEXPA1, 4 and 6) wereighly phylogenetically related whereas the other three genesLeEXPA3, 5 and 7) were closely related to each other. Yet allhese tomato expansins showed a different and characteristicattern of mRNA expression (Brummell et al., 1999a). Harrisont al. (2001) made a similar observation in strawberry where twout of six expansins, two FaEXP2 and FaEXP7 showing 96%mino acid identity amongst themselves, expressed at differentimes during fruit development. Six expansins have also beeneported during apple fruit ontogeny. Three of these, MdEXPA2,, 4, were present in ripening fruit, however, only MdEXPA2as a ripening related expansin. On the other hand expressionf MdEXPA5 was restricted to only the early growth periodWakasa et al., 2003). In pear, out of seven expansins, PcEXP4nd PcEXP5 showed the highest homology at the protein level98% amino acid sequence identity), yet these genes showedifferent patterns of expression during fruit development. Theigh level of sequence identity (80–90% identity at nucleotideevel) between these three banana expansins (MaEXPA3, 4 and) suggested allelic forms of the same gene although they showifferences in intron size and their respective sequences. It isikely that these may have arisen due to gene duplication events.

mRNA abundance studies revealed that MaEXPA2, likeaEXP1 (Trivedi and Nath, 2004) is a fruit specific and ripening

elated expansin. Both the genes are expressed abundantly onlyn mature fruit during ethylene induced ripening and repressedn fruit treated with 1-MCP. The simultaneous expression ofoth MaEXPA2 and MaEXP1 may be essential for cell walloosening during banana ripening and these two expansins mayct synergistically to enhance softening. The other possibilitys that these two expansins may act on different componentsf cell wall (Im et al., 2000). Surprisingly, comparison of theromoter sequences of these two genes did not indicate anyimilarity or presence of any common conserved cis motif. Thislso indicates that different cis elements may be responsive tohe same ripening related signals. It would be interesting totudy if these two genes are activated differentially by hormonesther than ethylene (Cho and Cosgrove, 2004). Other expansinshowed varied transcription patterns. Both MaEXPA3 andaEXPA5 were expressed during ethylene induced ripening

ut unlike MaEXPA2 and MaEXP1, transcript accumulationould also be observed in some vegetative tissues althoughhe level of expression was very low. MaEXPA4 seemed toe different from other banana expansins in that it showedxpression throughout ripening as well as in control and 1-MCPreated fruit. It also showed transcript accumulation during fruitevelopment. It is likely to have a role during fruit growth ands not ethylene regulated. Similar observations have been maden other plants. During ripening of peach fruit, three expansinsere detected. All three expansins were fruit specific, but each

howed differential patterns of mRNA accumulation during fruitevelopment. The PpEXP2 mRNA was constitutively expressed

hroughout fruit development but was abundant in stage III,hen the fruit expands exponentially and then matures. ThepEXPA1 and PpEXPA3 mRNAs were up-regulated at the onsetf ripening. Studies on the expression pattern of these three

and

eAmoss6Po2gcrorerfp

Mletct(b

metfsasicsbs

A

oRf

R

A

B

B

B

B

C

C

C

C

G

G

H

H

H

H

H

H

I

L

L

L

L

M

O

O

P

Asha et al. / Postharvest Biology

xpansins in two cultivars of peach (Manami – stony hard flesh;katsuki – melting flesh) have clearly shown that PpEXPA3 isore likely to play a role in peach fruit softening than PpEXPA1

r PpEXPA2 and ethylene directly regulates the transcription ofoftening-related PpEXPA3 (Hayama et al., 2003, 2006). Out ofeven pear expansin genes, expression of four (PcEXP2, 3, 5 and) was closely correlated with a loss of flesh firmness however,cEXP4 showed expression from an early growing stage to anver-ripe stage and was unaffected by 1-MCP (Hiwasa et al.,003). PcEXP7 was expressed exclusively during early fruitrowth. Similarly, in strawberry fruit, FaEXP4 was expressedonstitutively throughout fruit development, but with a smalleduction in expression at the large green stage. In tomato,ut of six fruit expansins, only one LeEXPA1 was ripeningelated but others like LeEXPA 3, 4 and 5 were present in greenxpanding and maturing fruit with declining levels duringipening (Brummell et al., 1999b). Multiple fruit expansinsrom banana and other species, thus, have complex expressionatterns, the significance of which is not yet known.

Promoter analysis of the fruit specific and ripening relatedaEXPA2 reveals a GCC-box that has been found in many ethy-

ene responsive genes involved in defense and abiotic stress (Haot al., 1998). The GCC-box in MaExp2A promoter is identicalo a classical GCC box sequence except for a single nucleotidehange C → G (AGCCGGCA). It has been shown that the muta-ion in this position does not affect the specificity of this boxHao et al., 1998). To date GCC boxes have not been shown toe present in promoters of ripening related genes.

In conclusion, this study reveals the presence of multipleembers of the expansin gene family with staggered and differ-

ntial expression during ripening in banana fruit. It also appearshat expression of multiple expansin genes might be requiredor softening in monocot fruit like banana as has been demon-trated for various dicot fruit like apple, peach, pear, tomatond strawberry. Of the four expansins identified, MaEXPA2 ispecifically expressed in fruit and is abundant during ethylenenduced ripening. Thus, MaEXPA2 appears to be an attractiveandidate gene for manipulating shelf life of banana. Furthertudies on the knockouts of individual genes and the regulationy promoters might reveal fine regulation of each gene and theirpecific role in softening in fruit.

cknowledgements

Financial support for the work, provided by the Departmentf Science and Technology, Government of India and the Junioresearch Fellowship provided to Asha by CSIR, India, are grate-

ully acknowledged.

eferences

sif, M.H., Dhawan, P., Nath, P., 2000. A simple procedure for the isolationof high quality RNA from ripening banana fruit. Plant Mol. Biol. Rep. 18,109–115.

ennett, A.B., 2002. Biochemical and genetic determinants of cell wall disas-sembly in ripening fruit: a general model. Hortic. Sci. 37, 447–450.

rummell, D.A., Harpster, M.H., Dunsmuir, P., 1999a. Differential expressionof expansin gene family members during growth and ripening of tomatofruit. Plant Mol. Biol. 39, 161–169.

R

Technology 45 (2007) 184–192 191

rummell, D.A., Harpster, M.H., Civello, P.M., Palys, J.M., Bennett, A.B., Dun-smuir, P., 1999b. Modification of expansin protein abundance in tomato fruitalters softening and cell wall polymer metabolism during ripening. PlantCell 11, 2203–2216.

rummell, D.A., Harpster, M.H., 2001. Cell wall metabolism in fruit softeningand quality and its manipulation in transgenic plants. Plant Mol. Biol. 47,311–340.

ho, H.-T., Cosgrove, D.J., 2004. Expansins as agents in hormone action. In:Davies, P.J. (Ed.), Plant Hormones, 3rd ed. Kluwer Academic, Dordrecht,The Netherlands, pp. 262–281.

hoi, D., Lee, Y., Cho, H.-T., Lee, Y., 2006. Expansins: expanding importancein plant growth and development. Physiol. Plant 126, 511–518.

osgrove, D.J., 2000. Loosening of plant cell walls by expansins. Nature 407,321–326.

osgrove, D.J., 2005. Growth of the plant cell wall. Nat. Rev./Mol. Cell. Biol.6, 85–861.

iovannoni, J., DellaPenna, D., Bennett, A.B., Fischer, R.L., 1989. Expressionof a chimeric polygalacturonase gene in transgenic rin (ripening inhibitor)tomato fruit results in polyuronide degradation but not fruit softening. PlantCell 1, 53–63.

iovannoni, J., 2001. Molecular Biology of fruit maturation and ripening. Annu.Rev. Plant Physiol., Plant Mol. Biol. 52, 725–749.

ao, D., Ohme-Takagi, M., Sarai, A., 1998. Unique mode of GCC box recog-nition by the DNA-binding domain of ethylene-responsive element-bindingfactor (ERF domain) in plant. J. Biol. Chem. 273, 26857–26861.

armer, S.E., Orford, S.J., Timmis, J.N., 2002. Characterisation of six �-expansin genes in Gossypium hirsutum (upland cotton). Mol. Genet. Genom.268, 1–9.

arrison, E.P., McQueen-Mason, S.J., Manning, K., 2001. Expression of sixexpansin genes in relation to extension activity in developing strawberryfruit. J. Exp. Bot. 52, 1437–1446.

ayama, H., Ito, A., Moriguchi, T., Kashimura, Y., 2003. Identification of a newexpansin gene closely associated with peach fruit softening. PostharvestBiol. Technol. 29, 1–10.

ayama, H., Shimada, T., Fujii, H., Ito, A., Kashimura, Y., 2006. Ethylene-regulation of fruit softening and softening-related genes in peach. J. Exp.Bot. 57, 4071–4077.

iwasa, K., Rose, J.K.C., Nakano, R., Inaba, A., Kubo, Y., 2003. Differentialexpression of seven �-expansin genes during growth and ripening of pearfruit. Physiol. Plant 117, 564–572.

m, K.H., Cosgrove, D.J., Jones, A.m., 2000. Subcellular localization of expansinmRNA in xylem cells. Plant Physiol. 123, 463–470.

i, Y., Darley, C.P., Ongaro, V., Fleming, A., Schipper, O., Baldauf, S.L.,McQueen-Mason, S.J., 2002. Plant expansins are a complex multigene fam-ily with an ancient evolutionary origin. Plant Physiol. 128, 854–864.

i, Y., Jones, L., McQueen-Mason, S.J., 2003. Expansins and cell growth. Curr.Opin. Plant Biol. 6, 1–8.

ink, B.M., Cosgrove, D.J., 1998. Acid–growth response and alpha expansinsin suspension cultures of Bright Yellow2 tobacco. Plant Physiol. 118, 907–916.

ohani, S., Trivedi, P.K., Nath, P., 2004. Changes in activities of cell wall hydro-lases during ethylene induced ripening in banana: effect of 1-MCP, ABA andIAA. Postharvest Biol. Technol. 31, 119–126.

cQueen-Mason, S.J., Cosgrove, D.J., 1995. Expansin mode of action on cellwalls: analysis of wall hydrolysis, stress-relaxation, and binding. Plant Phys-iol. 107, 87–100.

benland, D.M., Crisosto, C.H., Rose, J.K.C., 2003. Expansin protein levelsdecline with the development of mealiness in peaches. Postharvest Biol.Technol. 29, 11–18.

’Malley, R.C., Lynn, D.G., 2000. Expansin message regulation in parasiticangiosperms: marking time in development. Plant Cell 12, 1455–1465.

owell, A.L.T., Kalamaki, M.S., Kurien, P.A., Gurrieri, S., Bennet, A.B., 2003.Simultaneous transgenic suppression of LePG and LeExp1 influences fruit

texture and juice viscosity in a fresh market tomato variety. J. Agric. FoodChem. 51, 7450–7455.

ose, J.K.C., Lee, H.H., Bennett, A.B., 1997. Expression of a divergent expansingene is fruit-specific and ripening-regulated. Proc. Natl. Acad. Sci. USA 94,5955–5960.

1 gy an

R

S

S

S

S

92 Asha et al. / Postharvest Biolo

ose, J.K.C., Bennett, A.B., 1999. Cooperative disassembly of the cellulose-xyloglucan network of plant cell walls: parallels between cell expansion andfruit ripening. Trends Plant Sci. 4, 176–183.

ane, V.A., Chourasia, A., Nath, P., 2005. Softening in mango (Mangifera indicavar Dashehari) is correlated with the expression of an early ethylene respon-

sive, ripening related expansin gene, MiExpA1. Postharvest Biol. Technol.38, 223–230.

ambrook, T., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A LaboratoryManual. Cold Spring Harbour Laboratory Press, Cold Spring Harbour, NewYork, USA.

T

W

d Technology 45 (2007) 184–192

ampedro, J., Cosgrove, D.J., 2005. The expansin superfamily. Genome Biol.6, 242.

mith, C.J., Watson, C.F., Ray, J., Bird, C.R., Morris, P.C., Schuch, W., Grierson,D., 1988. Antisense RNA inhibition of galacturonase gene expression intransgenic tomatoes. Nature 334, 724–726.

rivedi, P.K., Nath, P., 2004. MaExp1, an ethylene-induced expansin from ripen-ing banana fruit. Plant Sci. 167, 1351–1358.

akasa, Y., Hatsuyama, Y., Takahashi, A., Sato, T., Niizeki, M., Harada,T., 2003. Divergent expression of six expansin genes during apple fruitontogeny. Eur. J. Hortic. Sci. 68, 253–259.