d ism)

r ofyer.the

atereenlingthatthe

en-of

androl,maon-

inaticandrly,the

Scientia Horticulturae xxx (2008) xxx–xxx

1

2

3

45

67

89

101112131415161718192021222324252627

28293031323334353637383940414243444546474849

erm

its

and

tted

eric

red.

ssue

net-

esis

ase

ting

erm

ion.

B.V.

G Model

HORTI 2911 1–8

UN

CO

RR

EC

TED

PR

OO

FWounding of melon fruits as a model system to study rind netting

Natalie Gerchikov a,b, Alexandra Keren-Keiserman a, Rafael Perl-Traves b, Idit Ginzberg a,*a Department of Vegetable Research, Institute of Plant Sciences, ARO, The Volcani Center, P.O. Box 6, Bet Dagan 50250, Israelb Department of Life Sciences, Bar Ilan University, Ramat Gan 52900, Israel

1. Introduction

The rind of muskmelon (Cucumis melo L. var. reticulatus) fruitscontains a network of suberized tissue, referred to as the ‘net’. Innetted varieties, the fully developed net is aesthetically important,increasing the marketability of the fruit. Netting starts towards theend of the fruit-expansion stage (Keren-Keiserman et al., 2004a). Itusually appears first around the blossom scar, gradually coveringthe whole fruit within a few days. An immature net is a shallow,greenish, non-dried tissue that protrudes only slightly above thefruit surface. The mature netting tissue is fully suberized andappears as a dry, white material that extends above the fruitsurface.

Histological studies have shown that the net originates fromfine cracks that appear on the fruit’s surface. As the fruit enlarges,the cracks deepen and widen and disrupt both the cuticle and somesubtending epidermal and hypodermal cells. Below these fissures,periderm cells start multiplying, and produce masses of cells withsuberized walls that extend above the fruit surface (Webster andCraig, 1976; Keren-Keiserman et al., 2004a).

The periderm is a secondary tissue made up of three cell types:phellem, phellogen and phelloderm. Suberized phellem (or cork)

forms several layers at the outermost level of the periderm, anderived from the meristematic phellogen layer (or cork cambiuunderneath it. The phelloderm cells form the innermost layethe periderm, and are similarly derived from the phellogen laThe periderm has a protective role in wound-healing assuberized phellem protects the wounded tissue from excess wloss and penetration of pathogens. Suberization has bsuggested to be a fundamental process involved in wound-heain plants (Dean and Kolattukudy, 1976). Hence, the cracksappear in the melon rind can be viewed as wounds andresultant net tissue as a healing periderm.

The composition and structure of suberin have been extsively studied on the natural and wound-induced peridermpotato. Suberin macromolecules contain both polyaromaticpolyaliphatic domains, cross-linked by esterification of glyceand are located between the primary cell wall and the plasmembrane (Kolattukudy, 2001; Bernards, 2002). Suberizatiassociated peroxidase activity was suggested to be involvedpolymerizing phenolic monomers to generate the arommatrix of suberin (Cottle and Kolattukudy, 1982; EspelieKolattukudy, 1985; Roberts and Kolattukudy, 1989). Similaperoxidase activity associated with suberization processes of

A R T I C L E I N F O

Article history:

Received 13 December 2007

Received in revised form 19 March 2008

Accepted 19 March 2008

Keywords:

Cucumis melo

Fruit development

Fruit rind

Ethylene

Periderm

Wounding

A B S T R A C T

The rind of muskmelon (Cucumis melo L. var. reticulatus) fruits contains a network of suberized perid

tissue, referred as the ‘net’, which originates in response to cracking of the fruit surface during

enlargement. Shallow cuts were made on the surface of melons to mimic naturally occurring cracking

induce net-like periderm development. Histological analysis of wounded fruits of the climacteric ne

variety Krimka, and of two smooth melon varieties: the climacteric Momordica and the non-climact

Tamdew, indicated that smooth melon varieties can undergo netting when their rind is fissu

Furthermore, the results implied that the climacteric character is not essential for net-ti

development, even though most netted varieties are climacteric. The involvement of ethylene in

like periderm development was studied by analyzing the expression pattern of the ethylene-biosynth

genes 1-aminocyclopropane-1-carboxylate synthase 1 and 1-aminocyclopropane-1-carboxylate oxid

1 following wounding and during periderm development, and by applying the ethylene-genera

chemical, Ethrel, as a lanolin paste, on the fresh cuts. Results suggested ethylene involvement in perid

initiation, but continuous exposure may interfere with further tissue development and organizat

General implications of the current study on periderm development are further discussed.

� 2008 Published by Elsevier

Contents l is ts ava i lab le at ScienceDirec t

Scientia Horticulturae

journa l homepage: www.e lsev ier .com/ locate /sc ihor t i

meanden-

50515253

* Corresponding author. Tel.: +972 3 9683787; fax: +972 3 9669583.

E-mail address: iditgin@volcani.agri.gov.il (I. Ginzberg).

0304-4238/$ – see front matter � 2008 Published by Elsevier B.V.

doi:10.1016/j.scienta.2008.03.015

Please cite this article in press as: Gerchikov, N. et al., Wounding o(2008), doi:10.1016/j.scienta.2008.03.015

melon rind has been shown and the cDNA of the main isozycontributing to that activity was isolated, sequenced,designated NAPOD (netting associated peroxidase) (KerKeiserman et al., 2004b).

f melon fruits as a model system to study rind netting, Sci. Hortic.

Tsmo(Pitrclimnetthasthe windustruinva

Eet aund(ACCconvconv199stepmel(YamACO

ThesearlThisdiviitsdeve

Tprocinvomenclimcontthecharvariclim

2. M

2.1.

Tstudexhiandtwoclimfilledaylnutr4:2.concwerFlowbearwou14 dnet.forchloIsraeconcmeninclupost

54555657585960616263646566676869707172737475767778798081828384858687888990

91

92

93949596979899

100101102103104105106107108109110111112113114115116

117

118119120121122123124125126127128

129130

131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167

168

169

N. Gerchikov et al. / Scientia Horticulturae xxx (2008) xxx–xxx2

G Model

HORTI 2911 1–8

Ple(2

UN

CO

RR

EC

TED

PR

OO

F

he netted melon varieties are climacteric fruits, whereas theoth varieties can be either climacteric or non-climactericat et al., 2000). Although net development initiates prior to theacteric evolution of ethylene, a role for this hormone in theing characteristic cannot be ruled out, particularly as ethylenebeen shown to be involved in plant stress responses, including

ound response (Abeles et al., 1992; O’Donnell et al., 1996), bycing defensive compounds directed toward reinforcing

ctural and chemical barriers against the threat of pathogension (Adie et al., 2007).thylene biosynthesis has been extensively studied (Arguesol., 2007). The last two steps of the biosynthetic pathway areer the control of 1-aminocyclopropane-1-carboxylic acid) synthase (ACS) and ACC oxidase (ACO). ACS catalyzes theersion of S-adenosylmethionine to ACC, which is thenerted to ethylene by ACO (Zarembinski and Theologis,

4); ACS catalysis is generally regarded as the rate-limitingin ethylene biosynthesis (Yang and Hoffman, 1984). In

on, three ACS genes, ME-ACS1, ME-ACS2, and ME-ACS3

amoto et al., 1995), and three ACO genes, CM-ACO1, CM-

2, and CM-ACO3 (Lasserre et al., 1996), have been isolated.e genes are expressed not only in ripening fruit but also in

y developing fruit at varying levels (Lasserre et al., 1997).and the involvement of ethylene in determining cell

sion and orientation (Kazama et al., 2004) further implypotential role in processes associated with peridermlopment.he current study approaches melon netting as a healingess and uses the melon rind system as a model to follow thelvement of ethylene in wound-induced periderm develop-t. The presented work includes netted versus smooth andacteric versus non-climacteric melon varieties to study theribution of endogenous or exogenously applied hormone todevelopment of the net-like periderm. For clarity, fruitacteristics are indicated in subscript form after the melonety’s name: N for netted rind, S for smooth rind and C foracteric fruit.

aterials and methods

Plant material and growth conditions

he following melon plants (C. melo L.) were used in the currenty: Krimka (KRYCN) of the variety reticulatus, the rind of whichbits a fine net, PI 414723 of the variety momordica (MOMCS),Tamdew (TADS, honeydew type) of the variety inodorus; thelatter varieties bear smooth-rind fruits. KRYCN and MOMCS areacteric varieties. Plants were grown in a greenhouse in pots

d with a volcanic ash-based soilless medium under naturalight. Irrigation was applied daily, together with a commercialient solution (Deshanim, Haifa, Israel) with an N:P:K ratio of5:6, an NO3:NH4 ratio of 9:1, and micronutrients. The nitrogenentration was 100 ppm throughout the experiment. Plants

e trimmed to a single stem, and supported by training threads.ers were pollinated manually, and the plants were pruned toone or two fruits each. The rinds of on-vine fruits were

nded by making shallow cuts using a scalpel blade at aroundays post-pollination (DPP), prior to development of the naturalFruits were harvested and rind samples were collected daily

2.2. Tissue embedding and histological staining

Rind samples (blocks of 3 mm � 3 mm � 5 mm) were fixed inFAA, dehydrated in an ethanol–xylene series and embedded inparaplast (Paraplast Plus, Oxford Labware, St. Louis, MO) accordingto standard methods (Ruzin, 1999). Tissue sections (20–25 mm)were stained with Safranin O/Fast green (Sigma Chemicals) forexamination of tissue morphology, or with Sudan IV (SigmaChemicals) for the identification of suberized cell walls (Johansen,1940). Sections were observed under UV or light microscope (LeicaDMLB, Bensheim, Germany) and images were displayed on amonitor through a CCD camera (Leica DC2000) using the LeicaIM1000 program.

2.3. RNA extraction and northern and semi-quantitative RT-PCR

analyses

Rind samples (three replicates) were collected by using ascalpel to peel very thin (about 1 mm) pieces of the fruit surface atthe wounding sites.

For northern blotting, total RNA was extracted from the rindtissues (Chang et al., 1993). Samples of 10 mg total RNA, consistingof equal amounts from three different fruits, were fractionated byelectrophoresis on a formaldehyde agarose gel (1%) (Sambrooket al., 1989); the bands of RNA were transferred to Hybond Nmembrane (Amersham, Buckinghamshire, UK), and hybridizedwith gene-specific probes labeled with 32P (NEBlot Kit, NewEngland BioLabs Inc., Ipswich, MA). Hybridization was carried outat 65 8C and the membrane was washed twice with 2� SSC, 0.1%(w/v) SDS at room temperature for 10 min, and twice with 0.2�SSC, 0.1% SDS at 65 8C for 10 min. The blot was exposed to X-rayfilm (Bio-Max, Kodak) with an intensifying screen or to a Phosphor-Imager screen, using Fujix BAS 1500 program (Fuji, Tokyo, Japan)for radioactivity quantification.

For identification of NAPOD transcripts, a 250-bp PstI fragmentfrom the 30 region of the corresponding cDNA, GenBank accessionno. AY373372 (Keren-Keiserman et al., 2004b), was used as a probe.Probes specific for the melon ethylene-biosynthesis genes wereprepared by standard PCR protocol using a cDNA library of melonfruit (kindly donated by Dr. N. Katzir, Newe Yaar, Israel) as thetemplate and specific primers for ACS1 (GenBank accession no.AB032935): 50-TCCGATGAGGAGTTTGTTGAC-30 (direct) and 50-GTGACATCATGTGGGGAGACTT-30 (reverse), or for ACO1 (GenBankaccession no. X95551): 50-ACGCCATCTTCCTGAATCAAAC-30 (direct)and 50-TTCGCCTGAAACTTCACTCCTAC-30 (reverse). Melon 18S rRNAprobe was amplified similarly using the specific primers 50-CAACGAGTCTATAGCCTTGG-30 (direct) and 50-CCACCAACCCTCAAT-GATCC-30 (reverse).

Semi-quantitative RT-PCR was performed using Reverse-iTTM

One-Step RT-PCR Kit (ABgene, Surrey, UK) according to themanufacturer’s protocol, with RNA samples and primers asdescribed above. Reaction products were fractionated by electro-phoresis on a TAE-agarose gel (1.5%) (Sambrook et al., 1989),stained with ethidium bromide and photographed.

3. Results

3.1. Healing tissue development and its similarity to net tissue

170171172173174175176

8 days post-wounding.Ethylene-generating compound 2-roethylphosphonic acid (Ethrel, Sigma Chemicals, Rehovot,l) was applied to the cut surface in a lanolin paste at finalentrations of 3.3 � 10�3, 3.3 � 10�4 and 3.3 � 10�5 M. Experi-ts were repeated in three growing seasons, and each treatmentded two to four fruits. Rind samples were collected for 8 days

-wounding.

ase cite this article in press as: Gerchikov, N. et al., Wounding o008), doi:10.1016/j.scienta.2008.03.015

To mimic the natural cracking of the rind during fruit expansion,shallow cuts were made at the surface of the melon. The woundingwas performed in the non-netted region of young fruits of the varietyKrimka (KRYCN), when the native net tissue covered only the lowerpart of the fruit (around 14 days post-pollination, DPP). Thedevelopment of healing tissue was monitored and compared to thatof the natural developing net tissue.

f melon fruits as a model system to study rind netting, Sci. Hortic.

inithewothuldy of

ingPP.to

th-wasYCN

ed-the

inwastemsizeuits

ellreand-

177178179180181182183184185186187188189190191192193194195

196

197198199200201202203204205206207208209210

211

212213214215216217218219220221222

E–L)

ttom

with

mm.

N. Gerchikov et al. / Scientia Horticulturae xxx (2008) xxx–xxx 3

G Model

HORTI 2911 1–8

UN

CO

RR

EC

TED

PR

OO

FFor histological study of healing periderm development, rind

samples were collected for 8 days post-wounding. Microscopicexamination of cross sections revealed the morphology of thehealing tissue (Fig. 1E–L). A mass of cells accumulated under-neath the cut, resulting in wound widening (Fig. 1F). Later, thedeveloping periderm protruded above the fruit surface (Fig. 1Gand H). Faint autofluorescence of suberizing cells could be seenfirst at the outer layer of the developing wound-periderm(Fig. 1J); as the healing periderm continued to develop, more cellswere suberized and their autofluorescence increased (Fig. 1Kand L).

The morphology of the developing healing periderm was foundto be similar to that of the natural net periderm (Fig. 1A–D),although the healing periderm did not protrude as much above thefruit surface.

Further indication of the similarity between the wounding andnetting processes was obtained by monitoring the expression ofNAPOD, which was induced upon wounding and remained at ahigh level during healing-periderm development (Fig. 2).

3.2. Induced ‘‘netting’’ in smooth-rind melons

To examine whether net/healing periderm could developsmooth-rind melons, wounding experiments were conducted wtwo smooth melon varieties: a non-climacteric variety, Tamd(TADS), and a climacteric variety, Momordica (MOMCS). Bvarieties were included in the experiment so that we coexamine the ability to produce net-like periderm independentlthe climacteric trait.

The rinds of TADS and MOMCS fruits were wounded by makshallow cuts with a scalpel every day for 8 days, starting at 14 DOn day 8, rind samples were collected and subjectedhistological observation of the wound healing process. Smoorind fruits of both varieties developed healing periderm thatsimilar to the healing tissue and the native net tissue of the KRvariety (Fig. 3).

3.3. The involvement of ethylene in periderm development

Our results indicated that healing periderm develops in nettand smooth-rind varieties, regardless of the climacteric trait offruit. To determine whether wound-ethylene plays a roleestablishing periderm development a molecular approachapplied as direct measurement of ethylene evolution in this syswas problematic. Wounding is performed with almost full-fruits while they are still on the vine; younger and smaller frcrack open upon wounding, and detached fruits do not heal wand they rot. In addition wounding is local and on a small arelatively to the total fruit surface; hence daily changes in wouethylene may be undetectable.

Fig. 1. Development of net-like wound-induced periderm. Shallow cuts were made in the rind of 14-DPP KRYCN fruits and the development of wound-induced periderm (

was compared to native net development (A–D). Cross sections were stained with Safranin O/Fast Green and viewed with light (upper and middle panels) and UV (bo

panel) microscopes to examine tissue morphology and autofluorescence of suberized cells, respectively. (A) Natural rind cracking; (B and C) filling of the fissures

peridermal cells; (D) mature net; (E and I) fresh cut; (F and J) 2 days post-wounding; (G and K) 5 days post-wounding; (H and L) 9 days post-wounding. Bar = 500

eticysisent.nd-iki

erre

223224225226227228229

Fig. 2. . Induction of NAPOD following wounding of KRYCN rind. Northern blot

analysis was performed with total RNA (10 mg) extracted from rind tissues collected

prior to wounding (1), immediately after wounding (2), during the healing period

for 5 days (3–7), or from naturally developed netted rind (8). Each sample contained

RNA from three independent fruits. The membrane was hybridized with the 30 PstI

fragment of NAPOD cDNA. Toluidine Blue staining of the membrane indicates the

amount of RNA loaded (bottom panel).

Please cite this article in press as: Gerchikov, N. et al., Wounding o(2008), doi:10.1016/j.scienta.2008.03.015

The expression pattern of the melon ethylene-biosynthgenes ACO1 and ACS1 was monitored by northern blot analfollowing wounding and during healing-periderm developmBoth genes, ACO1 and ACS1, have been shown to be wouresponsive, opposing to other members of their gene families (Met al., 1995; Lasserre et al., 1996; Bouquin et al., 1997; Lasset al., 1997).

f melon fruits as a model system to study rind netting, Sci. Hortic.

BtransteapeakwhiAlthvarifolddetewousamof th

Testa

230231232233234235236237238239240241242243

244245246247248249250

Fig. 3woun

F and

initia

N. Gerchikov et al. / Scientia Horticulturae xxx (2008) xxx–xxx4

G Model

HORTI 2911 1–8

Ple(2

UN

CO

RR

EC

TED

PR

OO

F

oth climacteric varieties, KRYCN and MOMCS, exhibitedsient induction of ACO1 transcript for 24 h, with a maximumdy-state level at 2 h post-wounding (Fig. 4). A second smaller

of transcript level was detected 2 days post-wounding, afterch the steady-state transcript level returned to its basal value.ough the basal level of ACO1 transcript was similar in botheties, its wound-induced level in MOMCS rind was about 10-

1–8 days post wounding) and specific primers for ACO1 and ACS1

genes (Fig. 5). Monitoring the expression pattern of ACO1 inwounded rinds of the climacteric varieties was repeated in thisreaction system to evaluate the induction pattern of ACO1

expression in the non-climacteric TADS. Similarly to the northernblot results KRYCN and MOMCS varieties exhibited constitutiveexpression of ACO1 in their rinds prior to wounding, with a drop in

. Net-like periderm development in the smooth-rind varieties. Shallow cuts were made in the rind of 14-DPP MOMCS (A–F) and TADS (G–L) fruits and the development of

d-induced periderm was followed by microscopic observations. Cross sections were stained with Safranin O/Fast Green and viewed with light (A–C and G–I) and UV (D–

J–L) microscopes for examination of tissue morphology and autofluorescence of suberized cells, respectively. (A and D) and (G and J) Fresh cut; (B and E) and (H and K)

tion of wound-induced periderm; (C and F) and (I and L) mature wound-induced periderm. Bar = 500 mm.

251252253254255256257

higher than that in KRYCN. ACO1 transcript could not bected in a similar northern blot analysis of samples fromnded rind of the TADS variety (not shown). Furthermore, thee blots hybridized with ACS1 probe resulted in no signal for anye varieties, suggesting a low transcript level of this gene.o overcome these difficulties, semi-quantitative RT-PCR wasblished with the same RNA samples (unwounded rind, 4 h and

ase cite this article in press as: Gerchikov, N. et al., Wounding o008), doi:10.1016/j.scienta.2008.03.015

level below the initial values after 6 days. The expression level ofACO1 in the rind of the non-climacteric variety TADS was hardlydetectable, and clear signals were observed only on days 1–3 post-wounding.

ACS1 transcript was barely detectable in the non-woundedrinds of the three varieties, but gene expression was inducedfollowing wounding. The high expression level remained in the

f melon fruits as a model system to study rind netting, Sci. Hortic.

oflop-thered

ells,.asteteds aenet to

on

hasits

thethe

e is

eenfirstermthatandruiteenthedi-of

aveind

doruiteenlar/the

rcesal.,

tiesthatpeds off ae ofact

258259260261262263264265266267268269270271

272273274275276277278279280281282283284285286287

288

289

290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320

N. Gerchikov et al. / Scientia Horticulturae xxx (2008) xxx–xxx 5

G Model

HORTI 2911 1–8

UN

CO

RR

EC

TED

PR

OO

Frinds of KRYCN and TADS during the entire experiment; however,wounding-related expression of ACS1 in the rind of MOMCS lasted ashorter time, with delayed induction and rapid decrease.Histological studies of the corresponding wounded fruit rindsshowed initiation of periderm development on day 3 post-wounding and suberization of the phellem cells from day 5post-wounding, in all three varieties (data not shown). Takentogether with the ACO1/ACS1 expression profile, these resultssuggest that transiently high ethylene biosynthesis precedesperiderm initiation.

The involvement of ethylene in periderm development wasfurther tested by application of Ethrel, an ethylene-generatingchemical, as a lanolin paste, on freshly cut wounds on the fruitsurface. Rind samples were collected 8 days later and subjected to

microscopic observation (Fig. 6). Interestingly, applicationlanolin paste with or without Ethrel inhibited periderm devement in the rinds of KRYCN and TADS, but not MOMCS fruits. Inlatter, suberized periderm developed, but with an altemorphology, i.e. expanded periderm tissue with enlarged ccompared to non-treated healing periderm (Fig. 6, top panel)

The observation that wound treatment with plain lanolin presults in altered tissue morphology similar to Ethrel-treatissue suggested that sealing the wound with paste producelocal accumulation of wound-induced endogenous ethyl(Eklund and Little, 1996, 2000). At this point, it is importannote that attempts to apply the hormones as a liquid solutionwounded or native net tissues were ultimately unsuccessful.

In summary, the results suggested that after periderminitiated, a high level of ethylene may interfere withdevelopment and organization.

4. Discussion

4.1. Native net is a wound-induced periderm

One of the factors determining muskmelon marketability isfruit’s appearance, i.e. complete netting of the rind withvariety-characteristic patterns of that tissue. However, littlknown about the factors that control netting.

Studies of early stages in net tissue development have bproblematic, as it was difficult to detect the micro-cracks thatappear on the surface of the fruit and initiate periddevelopment. The central hypothesis of this study waswounding of the melon rind would mimic natural crackinginduce net-like tissue. Making small cuts on the surface of the fenabled us to determine similarities in tissue morphology betwhealing periderm and net tissue (Fig. 1); induction of NAPOD,netting-associated peroxidase, following wounding, lent adtional support to the observed similarity (Fig. 2). The usewounding further showed that smooth-rind melon varieties hthe potential to produce net-like tissue upon fissuring of their r(Fig. 3).

It has been suggested that the smooth-rind melon varietiesnot develop netting because their rind does not crack during fexpansion (Keren-Keiserman et al., 2004a). The difference betwnetted and smooth varieties might be due to some cuticuepidermal characteristics involved in weakening the rind offormer and reducing their ability to resist the tensile fogenerated during fruit enlargement (Keren-Keiserman et2004a). The cuts made on the surface of the smooth-rind varieTADS and MOMCS induced the development of healing tissuewas similar to the native net and the healing tissue that develoon the surface of the wounded rind of KRYCN fruits. Columnphellem cells that are characteristic to the morphology operiderm tissue could be observed in the wound-healing tissuthe three melon varieties, but their arrangement was very comp

Fig. 4. Expression profile of the ACO1 gene following wounding and during

periderm development in wounded rinds of the climacteric varieties KRYCN and

MOMCS. Total RNA was extracted from wounded rind of both climacteric varieties at

various time points post-wounding. Samples (10 mg) consisting of a mixture of RNA

from three independently wounded fruits were analyzed by northern blot and

hybridized with 32P-PCR-amplified fragments of melon ACO1 or 18S rDNA (top

panels). Hybridization signals were quantified using Phosphor-Imager program and

ACO1 levels were calculated relative to 18S rRNA levels in each sample (bottom

graphs). R: Non-wounded rind.

Fig. 5. Expression profile of melon ACO1 and ACS1 genes in wounded rind of all three melon varieties. RNA samples described in Fig. 6 were applied for semi-quantitative RT-

PCR. The expression levels of ACO1 and ACS1 were monitored relative to 18S rRNA levels in each sample. R: Non-wounded rind.

Please cite this article in press as: Gerchikov, N. et al., Wounding of melon fruits as a model system to study rind netting, Sci. Hortic.(2008), doi:10.1016/j.scienta.2008.03.015

andnatuL; Fiperiinduinflu

MstagblosforKRYhavedeve

321322323324325326327328329330331332333

334335336337338

Fig. 6was a

Fast G

Figur

treat

N. Gerchikov et al. / Scientia Horticulturae xxx (2008) xxx–xxx6

G Model

HORTI 2911 1–8

Ple(2

UN

CO

RR

EC

TED

PR

OO

F

barely extended over the fruit’s surface, in comparison to therally occurring net tissue (compare Fig. 1C and D to Fig. 1H andg. 3C and F and I and L). That difference in the appearance of thederm tissue, despite the similar morphology and similarction, suggests that the native net periderm is part of, andenced by developmental processes of melon fruit.

rind responds to wounding in the same way as the smooth rinds ofTADS and MOMCS and the periderm appears as a compactlyarranged tissue. Nevertheless, it is possible that the cuts we madeinduced signaling that differs from that involved in the naturallyoccurring rind cracking.

. Development of wound-induced periderm following localized treatment with ethylene generating substance, Ethrel. Lanolin paste containing Ethrel (3.3 T 10S3 M)

pplied on shallow cuts made in the rinds of KRYCN, MOMCS, and TADS fruits. Cross sections of rind samples collected 8 days post-wounding were stained with Safranin O/

reen and viewed with light (top panel) and UV (bottom panel) microscopes for examination of tissue morphology and autofluorescence of suberized cells, respectively.

es represent results of three experiments covering three growing seasons, three replicates per treatment. Similar tissue morphologies were obtained for all Ethrel

ments, as described in Section 2. Bar = 500 mm.

339340

341342343344

elons were wounded around 14 DPP. At that developmentale of the fruit, the native net tissue of KRYCN encircles only itssom scar and non-netted regions of smooth rind are availablewounding. It is possible that these smooth-rind regions of

CN are not yet ‘‘programmed’’ to undergo netting and do notthe potential to perceive fissured epidermis as a signal for thelopment of net-like periderm. Thus, the immature non-netted

ase cite this article in press as: Gerchikov, N. et al., Wounding o008), doi:10.1016/j.scienta.2008.03.015

4.2. Ethylene may be involved with periderm initiation but interferes

with its organization

The fact that fruits of the non-climacteric variety TADS

responded to wounding in the same way as those of KRYCN andMOMCS suggested that the ability to produce healing/net peridermis independent of the climacteric character. Indeed, wound-

f melon fruits as a model system to study rind netting, Sci. Hortic.

ousCO1

the

cellion,ato

ermnceforon,

ferel toand

CS,ter-ilarthe

ula-undd insedere

that, as

suend-tiesg ofingay

atableple,msses

stryper

ress,

nses

ribo-

lant

andtwo

029–

ating

345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410

411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448

449

450451452

453

454455456457458459460461462463464465466467468

N. Gerchikov et al. / Scientia Horticulturae xxx (2008) xxx–xxx 7

G Model

HORTI 2911 1–8

UN

CO

RR

EC

TED

PR

OO

F

induced signaling has been shown to proceed independently ofethylene-induced signaling (Bouquin et al., 1997), and non-climacteric melon fruit can synthesize high amounts of ethyleneafter wounding that match those of the climacteric fruit (Perinet al., 2002). Furthermore, although all netted melon varieties arealso climacteric (Pitrat et al., 2000), the natural netting of thereticulatus fruits occurs prior to the ripening-associated ethylenecrisis; thus the role of ethylene in periderm development requiresfurther clarification.

The melon ethylene-biosynthetic genes ACO1 and ACS1 havebeen shown to be wound-response genes (Miki et al., 1995;Lasserre et al., 1996, 1997; Bouquin et al., 1997). Therefore, theirexpression pattern was monitored immediately after wounding ofthe rind to follow their wound-induced expression, and duringwound-induced periderm development to evaluate the associationof ethylene with these processes. Transcript levels of ACS1 are verylow in the rinds of both climacteric and non-climacteric varieties,and can be detected only by RT-PCR (Ishiki et al., 2000). In all threevarieties tested here, ACS1 expression was induced followingwounding of the rind and remained at relatively elevated levelsduring wound-induced periderm development.

While monitoring ACO1 gene expression, it was interesting tonote that the presence or absence of detectable levels of itstranscripts in the non-wounded rind samples corresponded to theclimacteric character of the melon varieties, i.e. relatively highconstitutive expression in KRYCN and MOMCS, and low/nilexpression in TADS (Figs. 4 and 5). Moreover, although all thevarieties exhibited wound induction of ACO1 transcript (Figs. 4 and5), the induction level in the rind of non-climacteric TADS wassignificantly lower. It has been suggested that in cases of the sameor closely related species with examples of both climacteric andnon-climacteric types, the non-climacteric phenotypes may reflectmutations in ethylene synthesis or signaling (Giovannoni, 2004).The delayed and low level of wound-induced ACO1 in TADS is inagreement with that suggestion, although alteration in ethyleneperception has previously been suggested for another non-climacteric melon type (Perin et al., 2002).

ACO1 expression in the rinds of the climacteric varieties KRYCN

and MOMCS indicated its immediate, but transient induction, witha peak at 2 h post-wounding (Fig. 4); yet another peak, albeit lowerin value, was observed on day 2 post-wounding, prior to thedevelopment of periderm tissue a day later (as observed byhistological analysis). These results are in agreement with a reporton wounding of melon leaves which induced CM-ACO1 transcriptaccumulation within 30 min, reaching a maximum at 2 h, afterwhich the level of ACO transcript rapidly decayed (Bouquin et al.,1997). Thus, if the first ACO1 peak in KRYCN and MOMCS samples isan early response to wounding, what is the role of the second peak,just prior to periderm development? Upon wounding, theestablishment of a systemic response is a priority for the plant,while local responses, such as periderm development, areestablished later (Delessert et al., 2004). Likewise, we suggestthat the second and lower ACO1 peak in the melon rind followingthe primary and high wound-response peak is related to the localresponse, including initiation of periderm tissue. Nonetheless,direct measurements of ethylene evolution during periderminitiation and development are required to clarify its role.

Moreover, it seems that continuous and elevated expression ofACO1 is not required for periderm development as its transcript

the biosynthesis (Argueso et al., 2007). Likewise, continuexpression of induced ACS1 and transient expression of A

suggest that the latter determines ethylene production inwounded melon rind.

Short-term ethylene exposure has been shown to stimulatedivision while continuous ethylene treatment suppresses divisas well as mitotic DNA synthesis (Apelbaum and Burg, 1972; Set al., 1976; Dan et al., 2003; Kazama et al., 2004). As periddevelopment involves extensive cell proliferation, the evidedescribed here could likewise imply that ethylene is requiredperiderm initiation in wounded melon rind, but latercontinuous high concentrations of the hormone may interwith tissue development. Indeed, external application of Ethrethe wounded rinds inhibited periderm development in KRYCN

TADS, and altered its characteristic morphology in MOMinducing enlarged and irregular cells, in contrast to the characistic columns of flattened peridermal cells (Fig. 6). The simresults obtained with plain lanolin suggest that sealingwounded area with paste may result in increased local accumtion of wound-induced ethylene, as described elsewhere (Ekland Little, 1996, 2000). Although no periderm was developewounded rinds of KRYCN and TADS in the presence of increaethylene levels, the cell walls at the margins of the wounds wsuberized (Fig. 6, lanolin and Ethrel treatments), implyingethylene may not play a role in wound-induced suberizationhad been suggested previously (Lulai and Suttle, 2004).

In conclusion, we show similarities between melon net tisthat develops following growth-induced rind fissures, and wouinduced periderm, and demonstrate that smooth-rind variealso have the potential to develop net-like tissue upon fracturintheir fruit surfaces. Ethylene is transiently induced followwounding; however, a constant high level of the hormone mimpede periderm cell division and organization. The dpresented here may contribute to the control of undesirarusseting of other agronomically important crops such as appear, tomato and potato, as well as of other peridermal systesuch as cork in woody species, root phellem, and of plant responto pathogens and herbivores.

Acknowledgements

This work was supported by the Chief Scientist of the Miniof Agriculture and Rural Development, Israel. Institution panumber 103/2007.

References

Abeles, F., Morgan, P., Saltveit, M., 1992. Ethylene in Plant Biology. Academic PLondon, UK.

Adie, B., Chico, J.M., Rubio-Somoza, I., Solano, R., 2007. Modulation of plant defeby ethylene. J. Plant Growth Regul. 26, 160–177.

Apelbaum, A., Burg, S.P., 1972. Effect of ethylene on cell division and deoxynucleic acid synthesis in Pisum sativum. Plant Physiol. 50, 117–124.

Argueso, C., Hansen, M., Kieber, J., 2007. Regulation of ethylene biosynthesis. J. PGrowth Regul. 26, 92–105.

Bernards, M.A., 2002. Demystifying suberin. Can. J. Bot. 80, 227–240.Bouquin, T., Lasserre, E., Pradier, J., Pech, J.-C., Balague, C., 1997. Wound

ethylene induction of the ACC oxidase melon gene CM-ACO1 occurs viadirect and independent transduction pathways. Plant Mol. Biol. Rep. 35, 11035.

Chang, S., Puryear, J., Cairney, J., 1993. A simple and efficient method for isolRNA from pine trees. Plant Mol. Biol. Rep. 11, 113–116.

tion:iated

latesrmis.

g in

469470471472473474475476

level immediately falls to basal values (Figs. 4 and 5). This is furtherexemplified in the wounded rind of TADS, where the inducedexpression of ACO1 was evident only from days 1 to 3 post-wounding, and a well-organized and suberized wound-inducedperiderm developed from day 3 on. Although ACS activity isaccepted as the rate limiting step in the biosynthesis, in conditionsof high ethylene production ACO often is the rate limiting step in

Please cite this article in press as: Gerchikov, N. et al., Wounding o(2008), doi:10.1016/j.scienta.2008.03.015

Cottle, W., Kolattukudy, P.E., 1982. Abscisic acid stimulation of suberizainduction of enzymes and deposition of polymeric components and assocwaxes in tissue cultures of potato tuber. Plant Physiol. 70, 775–780.

Dan, H., Imaseki, H., Wasteneys, G.O., Kazama, H., 2003. Ethylene stimuendoreduplication but inhibits cytokinesis in cucumber hypocotyl epidePlant Physiol. 133, 1726–1731.

Dean, B.B., Kolattukudy, P.E., 1976. Synthesis of suberin during wound-healinjade leaves, tomato fruits, and bean pods. Plant Physiol. 58, 411–416.

f melon fruits as a model system to study rind netting, Sci. Hortic.

DelesSle

EklunrT

Eklunb

Espelan

GiovaC

IshikSye

Kazaee

Kereniss

Kerenam

KolatG

LasseSm

Lasse1d2

477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511

512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546

547

N. Gerchikov et al. / Scientia Horticulturae xxx (2008) xxx–xxx8

G Model

HORTI 2911 1–8

Ple(2

UN

CO

RR

EC

TED

PR

OO

F

sert, C., Wilson, I.W., Van der Straeten, D., Dennis, E.S., Dolferus, R., 2004.patial and temporal analysis of the local response to wounding in Arabidopsisaves. Plant Mol. Biol. 55, 165–181.d, L., Little, C.H.A., 1996. Laterally applied Ethrel causes local increases in

adial growth and indole-3-acetic acid concentration in Abies balsamea shoots.ree Physiol. 16, 509–513.d, L., Little, C.H.A., 2000. Transport of [1-C-14]-indole-3-acetic acid in Abies

alsamea shoots ringed with Ethrel. Trees 15, 58–62.ie, K.E., Kolattukudy, P.E., 1985. Purification and characterization of an abscisiccid-inducible anionic peroxidase associated with suberization in potato (Sola-um tuberosum). Arch. Biochem. Biophys. 240, 539–545.nnoni, J.J., 2004. Genetic regulation of fruit development and ripening. Plantell 16, S170–S180.i, Y., Oda, A., Yaegashi, Y., Orihara, Y., Arai, T., Hirabayashi, T., Nakagawa, H.,ato, T., 2000. Cloning of an auxin-responsive 1-aminocyclopropane-1-carbox-late synthase gene (CMe-ACS2) from melon and the expression of ACS genes intiolated melon seedlings and melon fruits. Plant Sci. 159, 173–181.ma, H., Dan, H., Imaseki, H., Wasteneys, G.O., 2004. Transient exposure tothylene stimulates cell division and alters the fate and polarity of hypocotylpidermal cells. Plant Physiol. 134, 1614–1623.-Keiserman, A., Tanami, Z., Shoseyov, O., Ginzberg, I., 2004a. Rind character-tics associated with melon (Cucumis melo) netting; comparative study with

moothed-rind varieties. J. Hort. Sci. Biotechnol. 79, 107–113.-Keiserman, A., Tanami, Z., Shoseyov, O., Ginzberg, I., 2004b. Peroxidase

ctivity associated with suberization processes of the muskmelon (Cucumiselo) rind. Physiol. Plant 121, 141–148.

tukudy, P.E., 2001. Polyesters in Higher Plants. Springer-Verlag, Heidelberg,ermany, pp. 1–49.rre, E., Bouquin, T., Hernandez, J.A., Bull, J., Pech, J.C., Balague, C., 1996.tructure and expression of three genes encoding ACC oxidase homologs fromelon (Cucumis melo L.). Mol. Gen. Genet. 251, 81–90.

rre, E., Godard, F., Bouquin, T., Hernandez, J.A., Pech, J.C., Roby, D., Balague, C.,997. Differential activation of two ACC oxidase gene promoters from melonuring plant development and in response to pathogen attack. Mol. Gen. Genet.56, 211–222.

Lulai, E.C., Suttle, J.C., 2004. The involvement of ethylene in wound-induced sub-erization of potato tuber (Solanum tuberosum L.): a critical assessment. Post-harvest Biol. Technol. 34, 105–112.

Miki, T., Yamamoto, M., Nakagawa, H., Ogura, N., Mori, H., Imaseki, H., Sato, T., 1995.Nucleotide sequence of a cDNA for 1-aminocyclopropane-1-carboxylatesynthase from melon fruits. Plant Physiol. 107, 297–298.

O’Donnell, P.J., Calvert, C., Atzorn, R., Wasternack, C., Leyser, H.M.O., Bowles, D.J.,1996. Ethylene as a signal mediating the wound response of tomato plants.Science 274, 1914–1917.

Perin, C., Gomez-Jimenez, M., Hagen, L., Dogimont, C., Latche, A., Pitrat, M., Lelievre,J.-M., 2002. Molecular and genetic characterization of a non-climacteric phe-notype in melon reveals two loci conferring altered ethylene response in fruit.Plant Physiol. 129, 300–309.

Pitrat, M., Hanelt, P., Hammer, K., 2000. Some comments on infraspecific classifica-tion of cultivars of melon. Proceedings of Cucurbitaceae 2000, Acta. Hort. 510,29–36.

Roberts, E., Kolattukudy, P.E., 1989. Molecular cloning, nucleotide sequence, andabscisic acid induction of a suberization-associated highly anionic peroxidase.Mol. Gen. Genet. 217, 223–232.

Ruzin, S.E., 1999. Plant Microtechnique and Microscopy. Oxford University Press,New York.

Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A LaboratoryManual. Cold Spring Harbor Press, Cold Spring Harbor.

Sato, T., Watanabe, A., Imaseki, H., 1976. Effect of ethylene on DNA synthesis inpotato tuber discs. Plant Cell Physiol. 17, 1255–1262.

Webster, B.D., Craig, M.E., 1976. Net morphogenesis and characteristics of thesurface of muskmelon fruit. J. Am. Soc. Hort. Sci. 101, 412–415.

Yamamoto, M., Miki, T., Ishiki, Y., Fujinami, K., Yanagisawa, Y., Nakagawa, H.,Ogura, N., Hirabayashi, T., Sato, T., 1995. The synthesis of ethylene inmelon fruit during the early-stage of ripening. Plant Cell Physiol. 36,591–596.

Yang, S.F., Hoffman, N.E., 1984. Ethylene biosynthesis and its regulation in higherplants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 35, 155–189.

Zarembinski, T.I., Theologis, A., 1994. Ethylene biosynthesis and action—a case ofconservation. Plant Mol. Biol. 26, 1579–1597.

ase cite this article in press as: Gerchikov, N. et al., Wounding of melon fruits as a model system to study rind netting, Sci. Hortic.008), doi:10.1016/j.scienta.2008.03.015