Hs

ID

a

ARRAA

KEOPSG

1

dbsetia

[tmiposIc

(

t

0h

Plant Science 197 (2012) 40– 49

Contents lists available at SciVerse ScienceDirect

Plant Science

jou rn al hom epa ge: www.elsev ier .com/ locate /p lantsc i

istochemical location of key enzyme activities involved in receptivity andelf-incompatibility in the olive tree (Olea europaea L.)

rene Serrano1, Adela Olmedilla ∗

epartment of Plant Biochemistry, Cell and Molecular Biology, Estación Experimental del Zaidín (CSIC), Profesor Albareda 1, 18008 Granada, Spain

r t i c l e i n f o

rticle history:eceived 27 March 2012eceived in revised form 13 July 2012ccepted 16 July 2012vailable online 25 August 2012

eywords:nzyme activity

a b s t r a c t

Stigma-surface and style enzymes are important for pollen reception, selection and germination. Thisreport deals with the histochemical location of the activity of four basic types of enzyme involved inthese processes in the olive (Olea europaea L.). The detection of peroxidase, esterase and acid-phosphataseactivities at the surface of the stigma provided evidence of early receptivity in olive pistils. The stigmamaintained its receptivity until the arrival of pollen. Acid-phosphatase activity appeared in the style at themoment of anthesis and continued until the fertilization of the ovule. RNase activity was detected in theextracellular matrix of the styles of flowers just before pollination and became especially evident in pistils

lea europaea L.ollen–pistil interactiontigma receptivitySI

after self-pollination. This activity gradually decreased until it practically disappeared in more advancedstages. RNase activity was also detected in pollen tubes growing in pollinated pistils and appeared afterin vitro germination in the presence of self-incompatible pistils. These findings suggest that RNases maywell be involved in intraspecific pollen rejection in olive flowers. To the best of our knowledge this is thefirst time that evidence of enzyme activity in stigma receptivity and pollen selection has been described

in this species.

. Introduction

Fertilization of the olive (Olea europaea L.) relies on sexual repro-uction, controlled by pollen–pistil interaction. Proteins presentoth at the stigma surface and in the extracellular matrix of thetyle play crucial roles in pollen–pistil communication, which isssential for the regulation of mating in flowering plants. Amonghese proteins, those that exert enzyme activity have been stud-ed with regards to stigma receptivity and/or pollination-barrierctivity [1–5].

The effective pollination period (EPP), first defined by Williams6] as the number of days that the flower remains in conditiono be successfully pollinated so that it gives a fruit, is one of the

ost important factors determining successful fertilization. EPPs determined by several factors: the maturity of the ovules, thehysical condition of the pistil and, particularly, the receptivityf the stigma [7,8]. Stigma receptivity refers to the ability of the

tigma to support the germination of viable, compatible pollen [9].n general the stigma is receptive at the time of anthesis and thisondition may last for one or several days [2,10–13]. Although there

∗ Corresponding author. Tel.: +34 958 181600x318/119; fax: +34 958 129600.E-mail addresses: iserrano@indiana.edu (I. Serrano), adela.olmedilla@eez.csic.es

A. Olmedilla).1 Present address: Department of Biology, Indiana University, 47405 Blooming-

on, IN, USA.

168-9452/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved.ttp://dx.doi.org/10.1016/j.plantsci.2012.07.007

© 2012 Elsevier Ireland Ltd. All rights reserved.

are no satisfactory rapid ways of determining stigma receptivity,the surface of a receptive stigma invariably shows the presenceof several enzyme activities. Within this context, the activities ofperoxidase, esterase and acid phosphatase have been related tostigma receptivity and studied in different species [2,14,15]. Inaddition, during the EPP the pistil might be able to actively rec-ognize and inhibit self-pollen, either on the stigma surface beforeits germination (sporophytic self-incompatibility, or SSI) or duringthe growth of the pollen tube through the pistil (gametophyticself-incompatibility, or GSI) before reaching the ovule.

Ribonuclease (RNase) activity in the style plays a key rolein intraspecific pollen rejection in the majority of species thatdisplay GSI studied to date [16]. Self-incompatible ribonucleases(S-RNases) have been located via immunological assays [17,18] andstylar-RNase activity has been detected on protein gels in severaldifferent species [19] but we have been able to find no reportsconcerning the location of RNase activity.

O. europaea is a species of outstanding socio-economic impor-tance in Mediterranean countries, where it is cultivated for itsedible fruit and their oil. A large number of olive cultivars aregrown in Spain but one of the most widespread is Picual, due toits ideal qualities for oil production [20]. Olive-tree flowers aremainly wind pollinated and, although there is general agreement as

to the benefits of cross-pollination for fruit production [21–27], lit-tle data concerning the factors that determine self-incompatibilityin the olive are available. Different methods, including fruit settingafter flower bagging or pollen-tube growth after hand pollination,

Plant S

hotNccposobttap[g

weasopoat

bfltmrfttidiotst

2

2

gmpe

bssoicl4iaa

I. Serrano, A. Olmedilla /

ave been used to determine self-incompatibility mechanisms inlive flowers, but contradictory results can be found in the litera-ure about self-incompatibility in the same cultivar [22,24,29–34].evertheless, paternity tests using microsatellite markers, whichan identify the true donor pollen, have shown that whilst theultivars Picual and Arbequina are self-incompatible they are com-atible with each other [28,35,36]. It has been suggested that thelive belongs to the group of species possessing gametophyticelf-incompatibility [22,37,38] and shows features characteristicf species belonging to this self-incompatibility group [39]: it hasicellular pollen, wet-papillate stigma and a solid style, and fur-hermore, although a large number of pollen grains germinate inhe stigma, it is rare to detect more than one in the style [40–42]. Inddition, we have recently shown that olive-pollen grains undergorogrammed cell death (PCD) as a consequence of the SI response42]. These features reinforce the idea that the olive belongs to theroup of plants displaying GSI.

To understand pollen–pistil interaction in the olive more clearlye conducted histochemical assays to locate the activity of the

nzymes involved in stigma receptivity (esterase, peroxidase andcid phosphatase) in this species. To this end we analyzed severaltages before and after pollination. We also developed a methodf locating RNase activity in vivo in cv. Picual pistils after self-ollination, free-pollination and hand cross-pollination with pollenf cv. Arbequina an in cv. Picual pollen germinated in vitro in thebsence and presence of compatible or incompatible pistils, in ordero test whether this enzyme might be involved in pollen rejection.

Our results suggest early receptivity in the olive pistil, the stigmaeing receptive even before the opening of the anthers of their ownower, which physically enhances cross-pollination, the promo-ion of pollen dispersal and the reduction of interference between

ale and female functions [43]. In addition, the stigma seems toemain receptive until the arrival of the pollen grains on its sur-ace. Our data concerning the location of RNase activity in bothhe stylar extracellular matrix and germinated pollen grains arehe first of their kind to suggest that RNases may well be involvedn intraspecific pollen rejection in O. europaea. Although moreata are needed to understand more fully the mechanisms of self-

ncompatibility involved in the pollination process in this fruit tree,ur results support the idea that in addition to a physical barriero self-pollination as a consequence of early stigma receptivity, theelf-incompatibility of this species could well be of a gametophyticype, controlled by RNases.

. Materials and methods

.1. Materials

Perfect flowers (with both pistillate and staminate parts) wereathered periodically before, during and after anthesis during theonths of May and June from O. europaea, cv. Picual, grown in the

rovinces of Granada, Jaén and Córdoba (S. Spain). Samples andxperiments were repeated yearly for 3 years.

Inflorescences of cv. Picual with pistils at different stagesefore and after free-pollination were collected. Before anthe-is we chose three stages: stage 0: swollen, green flower buds;tage 1: white-cream flower buds before anthesis (petals about topen) and stage 2: opening petals, turgid, yellow anthers enclos-ng the pistil, thus preventing its pollination. After anthesis wehose 4 stages: stage 3: completely open flower with turgid, yel-ow anthers leaving the pistil accessible to pollen landing; stage

: open anthers in a completely open flower; stage 5: brown-

sh anthers in a completely open flower; and stage 6: brownnthers in a completely open flower. Finally we studied two stagesfter petal abscission: stage 7: pistils with green stigmas and

cience 197 (2012) 40– 49 41

enlarged ovaries and stage 8: pistils with brown stigmas andenlarged ovaries. In this way we covered 3 important points in pis-til development: before pollination, during pollination and afterfertilization.

2.2. Self- and controlled pollination

Self-pollinated pistils cultivar Picual, were collected at differentstages before and after anthesis to evaluate possible differencesin RNase activity associated with self-incompatibility. For self-pollination 12 inflorescences of cv. Picual containing white flowerbuds were enclosed in bags. Flowers inside the bags were takenfor analysis when inflorescences from other branches on the sametrees showed flowers after anthesis.

For compatible-pollination, 12 inflorescences of cv. Picual con-taining white flower buds were emasculated, cross-pollinated withpollen of the compatible cultivar Arbequina and enclosed in bags.Flowers inside the bags were collected for enzyme activity analy-sis when non-bagged inflorescences on the same branch showedflowers after anthesis.

To check whether pollination could influence stigma receptivity,flowers buds of 12 inflorescences at stage 1 were emasculated andenclosed in bags to avoid pollen entrance. After 10 days 6 of themwere dissected and enzyme activities were analyzed. The other6 inflorescences were hand-pollinated with cv. Picual pollen andenzyme activity was analyzed after two hours. Experiments wereconducted over three years, with at least three repetitions eachyear.

2.3. Peroxidase activity

Peroxidase activity was detected essentially as described in theWorthington Enzyme Manual. Fresh, freely pollinated pistils fromflowers at different stages of development were incubated with0.33 M o-dianisidine (Fluka) and 3 mM hydrogen peroxide (Perhy-drol, Merck) in 0.2 M phosphate buffer for 1 h at 25–30 ◦C. Squashedpistils were mounted in glycerol and studied under a light micro-scope.

2.4. Esterase activity

Esterase activity was detected according to Higuchi et al. [44].Fresh, freely pollinated pistils from flowers at different stages ofdevelopment were incubated with a mixture of 0.1% (w/v) �-naphthyl acetate (Sigma) and 0.2% (w/v) Fast Blue B Salt (Fluka)in 0.1 M Tris buffer (pH 7.2) for 1 h. Squashed pistils were mountedin glycerol and studied under a light microscope.

2.5. Acid-phosphatase activity

Acid-phosphatase activity was detected following essentiallyGomori’s method [45]. Fresh pistils excised from self-pollinated orfreely pollinated olive flowers at different stages of developmentwere incubated with a mixture of 0.003 M lead nitrate and 0.01 M�-glycephosphate in 0.05 M acetate buffer (pH 5.0) for 4 h, washedin distilled water, immersed in 1% (v/v) acetic acid for 10 s, rinsedin distilled water and incubated for 2 min in water saturated withhydrogen sulphide. Squashed pistils were mounted in glycerol andobserved under a light microscope.

2.6. RNase activity

RNase activity was detected using a method derived fromGomori’s reaction for DNases [46] using RNA as a substrate insteadof DNA. In this method the polynucleotides produced by RNaseactivity were detected by the addition of an acid phosphatase to

42 I. Serrano, A. Olmedilla / Plant Science 197 (2012) 40– 49

F al: (As w area pc pap

tttppib5aifiwTpd

ig. 1. Peroxidase activity in squashed pistils of freely pollinated flowers of cv. Picuurface; (B) stigma after anthesis, where the positive reaction is confined to just a fereas of positive reaction (arrow); (D) negative control of a stigma before anthesis.

he assay, which produced phosphate groups that were precipi-ated in the shape of black granules of lead phosphate, formed byhe addition of lead nitrate as a substrate. Pistils excised from self-ollinated, cross-pollinated (with pollen of cv. Arbequina) or freelyollinated olive flowers at different developmental stages and fixed

n Carnoy’s solution (3/1, v/v, 70% ethanol/acetic acid) were incu-ated for 4 h at 37 ◦C in 0.2 M acetate buffer (pH 5.0) containing

mg (w/v) of yeast RNA (Sigma), 5 mg (w/v) of acid phosphatasend 0.02 M of aqueous lead nitrate. The samples were then rinsedn distilled water, placed in water saturated with hydrogen sul-de for 2 min and then washed in distilled water. Squashed pistils

ere mounted in glycerol and studied under a light microscope.

o estimate the number of pollen tubes showing RNase activity, 50ollen tubes were counted for each pollination in three indepen-ent experiments.

) stigma before anthesis showing strong positive reaction across the whole stigmaas (arrows); (C) stigma from a flower after petal abscission, showing very confinedillar cells; st style; sr subpapillar region. Bars 100 �m.

Controls for all enzymes analyzed were carried out by incu-bating pistils in the different buffers without the correspondingenzyme substrates. At least 3 samples were analyzed in 3 repeti-tions at each stage of development.

2.7. In vitro pollen germination

Pollen grains isolated from olive trees were germinated accord-ing to Serrano et al. [47] in a medium containing 20% (w/v)sucrose, 0.01% (w/v) CaCl2, 0.01% (w/v) H3BO3 and 0.01% (w/v)KNO3 in a humid chamber at 30 ◦C for 3 h both in the absence

and presence of fresh receptive pistils of the same cultivar (Picual,self-incompatible).

After removing the pistils, pollen grains and pollen tubes wereused to detect RNase activity as described in Section 2.6.

I. Serrano, A. Olmedilla / Plant Science 197 (2012) 40– 49 43

Fig. 2. Esterase activity in squashed pistils of freely pollinated flowers of cv. Picual; (A) young stigma before anthesis showing positive reaction to esterase activity on allpapillar cells; (B) in stigmas after anthesis the reaction is positive only in certain areas of the surface (arrowheads); (C) in stigmas after petal abscission the positive reactioni eir pop

2

tpflngAdta2

s visible only in the pollen grains that have landed on the stigma surface and in thapillar cells; st style; sr subpapillar region. Bars 100 �m.

.8. Electron microscopy

Samples were processed for electron microscopy accordingo Serrano et al. [55]. Excised pistils from self-pollinated, cross-ollinated (with pollen of cv. Arbequina) or freely pollinated oliveowers at different developmental stages and pollen grains germi-ated in vitro were fixed overnight in 4% formaldehyde and 0.25%lutaraldehyde with 0.1 M cacodylate buffer at pH 7.2 and 4 ◦C.fter fixing, all the samples were washed in cacodylate buffer,

ehydrated in ethanol series and embedded in Unicryl. Ultra-hin sections were cut on a Reichert-Jung Ultracut E microtomend subjected to RNase activity detection as described in Section.6.

llen tubes (insert, arrowheads); (D) negative control in a pistil before anthesis. pc

3. Results

We used various stages of pistil development, ranging fromswollen buds before anthesis to flowers after petal abscission(Supplementary Fig. 1), to study enzyme activities involved instigma receptivity and pollen–pistil interaction in the olive tree (O.europaea cv. Picual).

3.1. Location of enzyme activities

3.1.1. Peroxidase activityPeroxidase activity was detected as a yellowish-orange precipi-

tate in the papillar cells of stigmas of flowers before anthesis at the

44 I. Serrano, A. Olmedilla / Plant Science 197 (2012) 40– 49

Fig. 3. Acid-phosphatase activity in squashed pistils of freely pollinated flowers of cv. Picual: (A) no acid-phospatase activity is detectable on stigmas before anthesis; (B)positive reaction in the papillar cells of stigmas of flowers after anthesis; (C) slight acid-phosphatase activity is detectable on stigmas from flowers after petal abscission; (D)the activity is located in the wall of the papillar cells as well as in some pollen grains and pollen tubes (insert, arrows); (E) style from a young pistil showing a slight positiver ) pistii he styB

tidatsrnw

3

pepo

TT

eaction; (F) style from a pistil after anthesis, showing a strong positive reaction; (Gt can be seen that positive reaction is concentrated in the extracellular matrix of tars (A, B, C, E, F, G) 100 �m (D, H) 20 �m.

wo stages selected. No reaction was detected either in the subpap-llar cells or in the styles of these flowers (Fig. 1A). A considerableecrease in the intensity of the reaction could be seen in pistilsfter anthesis, in which the yellowish-orange color was confinedo small areas of the surface of the stigma. Pollen grains on thetigma did not show any reaction (Fig. 1B). After petal abscission,eaction was detected only in very confined areas (Fig. 1C). Wheno reaction substrate was added to the mixture no positive reactionas visible, even in stigmas before anthesis (Fig. 1D).

.1.2. Esterase activityEsterase activity was detected by the presence of a brown

recipitate uniformly spread across the surface of stigmas in pistilsxcised from flowers before anthesis (Fig. 2A). After anthesis,ositive esterase reaction was concentrated within certain areasf the surface of the stigma (Fig. 2B) and after petal abscission the

able 1ime course of the enzyme activities assayed in this study in self-pollinated and freely po

Stigma Before pollination During polli

Stage 0 Stage 1 Stage 2 Stage 3

Free Free Free Free

Peroxidase +++ +++ +++ ++

Esterase +++ +++ ++ ++

Acid phosphatase − − ++ +++

Style Before pollination During pollination

Stage 0 Stage 1 Stage 2 Stage 3 Stage 4

Free Free Free Free Self Free

RNase − − + − ++ −

Acid phospatase − + ++ +++ +++ +++

l after petal abscission, showing no detectable reaction; (H) at higher magnificationle. pc papillar cells; pg pollen grain; pt pollen tube; sr subpapillar region; st style.

reaction was only visible in pollen grains and pollen tubes, whichwere deeply stained (Fig. 2C and insert). No signal was detected inthe negative controls containing no enzyme substrate (Fig. 2D).

3.1.3. Acid-phosphatase activityNo acid-phosphatase activity was found in the stigmas of stages

before anthesis (Fig. 3A) but it was detected immediately beforeand at the moment of anthesis (Fig. 3B) and remained visible untilthe last stage studied (Fig. 3C). This enzyme activity was concen-trated within the walls of papillar cells as well as in some pollengrains and pollen tubes (Fig. 3D). Unlike the activity of the othertwo enzymes just described, acid-phosphatase activity was also

present in the styles. It was visible in young flower buds beforeanthesis (Fig. 3E) and especially in flowers after anthesis (Fig. 3F)but was not detected in styles after petal abscission (Fig. 3G). Thisenzyme activity was located in the extracellular matrix of the styles

llinated pistils of cv. Picual.

nation After fertilization

Stage 4 Stage 5 Stage 6 Stage 7 Stage 8Free Free Free Free Free++ ++ + + −++ + + − −++ ++ + + +

After fertilization

Stage 5 Stage 6 Stage 7 Stage 8

Self Free Self Free Self Free Self Free Self+++ − ++ − ++ − − − −+++ + + + + − − − −

I. Serrano, A. Olmedilla / Plant Science 197 (2012) 40– 49 45

Fig. 4. Detection of peroxidase, esterase and acid-phosphatase activity in non-pollinated and hand-pollinated pistils: (A, B, C) the papillar cells of emasculated pistils of cv.P tivity

p a of eo

(w(

r

3

tdladtaspair

3

ftlsietp

w(

icual showing a positive reaction for peroxidase, esterase and acid-phosphatase aceroxidase, esterase and acid-phosphatase activity decrease drastically in the stigmf cv. Picual. pg pollen grain; sg stigma. Bars 50 �m.

Fig. 3H). The reaction was not visible either in the stigmas or styleshen there was no enzyme substrate added to the reaction mixture

data not shown).The time course of all the enzyme activities assayed is summa-

ized in Table 1.

.2. Stigma receptivity maintenance

It has been postulated that stigma receptivity decreases dras-ically after pollination. To check whether stigma receptivityepended on the arrival of pollen olive flowers were emascu-

ated and bagged to avoid pollen arrival prior to assays of enzymectivity. Peroxidase, esterase and acid-phosphatase activities wereetectable 10 days after emasculation (Fig. 4A–C), indicating thathe stigma was still receptive. To confirm that pollen arrival causes

decrease in the enzyme activity that controls stigma receptivityome emasculated flowers were hand pollinated with cv. Picualollen (self-incompatible pollination) and enzyme activity wasssayed after 2 h. A significant decrease in all the enzyme activ-ties tested was detected (Fig. 4E and F), indicating that stigmaeceptivity is maintained until the arrival of pollen.

.3. RNase activity

To test whether RNases were active in olive pistils, and there-ore if they would be involved in the self-incompatible reaction inhis species, we developed a new method derived from Gomori’sead-nitrate technique. RNase activity has been related to theelf-incompatible response in species with gametophytic self-ncompatibility (GSI), which is why we decided to study thisnzyme activity not only in the freely pollinated pistils, in whichhe previous enzyme activities had been tested, but also in self-

ollinated pistils.

RNase activity was detected by lead-phosphate precipitates,hich were not found in the pistils of flowers before anthesis

Fig. 5A) except during stage 2, where the petals were opening

respectively in the papillar cells of the stigmas 10 days after emasculation; (D, E, F)masculated pistils two hours after hand pollination with self-incompatible pollen

but the anthers were still enclosing the pistils, thus preventingpollination. In self-pollinated pistils black deposits were visiblein the extracellular matrix of the styles after anthesis; these wereespecially abundant in flowers at stage 4, when their own antherswere open (Fig. 5B), but were not present in styles after petalabscission (Fig. 5C). In freely pollinated flowers no RNase activitywas detected in the styles after anthesis (Fig. 5D). RNase activitywas located in pollen grains germinating on the stigmas after bothself- and free-pollination (Fig. 5E–G), especially when their pollentubes were long enough to reach the style (Fig. 5E). No positivereaction could be seen in the stigma cells. No enzyme activity wasdetected in the negative controls without enzyme substrate addedat any of the stages studied nor in the pistils cross-pollinated withcv. Arbequina.

3.3.1. RNase activity in pollen germinated in vitroTo check whether the RNase activity detected in pollen grains

was due to the self-incompatibility response, pollen grains weregerminated in vitro in the absence and presence of pistils of thesame cultivar (Picual, self-incompatible). When they were germi-nated in the presence of pistils 87% of the pollen tubes countedshowed a positive reaction to RNase activity (Fig. 6A and B), whilstno pollen tube showing positive reaction was observed when thepollen was germinated in the absence of pistils (Fig. 6C and D).

3.3.2. Electron microscope histochemical location of RNaseactivity

RNase activity was studied in greater detail by TEM. Electronmicrographs of longitudinal sections of styles of self-pollinatedpistils showed a positive reaction to RNase activity; precipitatescorresponding to a positive reaction were located predominantlywithin the vacuole of the stylar cells of self-pollinated samples

(Fig. 7A), but were not detectable in samples cross-pollinated withpollen from the compatible cultivar Arbequina (Fig. 7C). Simi-larly, precipitates corresponding to a positive RNase reaction weredetectable in the cytoplasm of pollen tubes germinated in vitro in

46 I. Serrano, A. Olmedilla / Plant Science 197 (2012) 40– 49

Fig. 5. Ribonuclease activity in squashed pistils of cv. Picual: (A) style from a pistil after flower anthesis, showing no positive reaction; (B) style from a self-pollinated pistilafter anthesis, showing a slight positive reaction; (C) style from a self-pollinated pistil after petal abscission, showing no positive reaction; (D) styles from freely pollinatedp sitivep in a fra stigma

tti(

4

ttlttsshc

cici

edfeaa

istils after anthesis, showing no positive reaction; (E) at higher magnification, a poollen grain showing positive reaction in the lower part of its pollen tube growingfter ribonuclease activity detection in a freely pollinated pistil. pg pollen grain, sg

he presence of pistils of the same cultivar (Fig. 7B) but no posi-ive reaction was visible when these pollen grains were germinatedn the presence of pistils of the compatible cultivar ArbequinaFig. 7D).

. Discussion

The pistil and pollen act upon one another from the momenthe pollen lands on the surface of the stigma to the time the pollenube reaches the embryo sac. Proteins from the pistil’s extracellu-ar matrix play a crucial role in this communication, among whichhose that exert enzyme activity are of particular importance inhe control of pollination. We assessed the activity of enzymesuch as esterases, peroxidases and acid phosphatases to determinetigma receptivity. For pollination to occur successfully the stigmaas to be receptive and this essential stage of pistil development isharacterized by the presence of these enzyme activities [1–3,48].

Furthermore, for effective fertilization the pistil has to dis-riminate between self and non-self pollen. Within this context,t has been proposed that RNase activity is responsible for theontrol of intraspecific pollination, causing pollen rejection in self-ncompatible gamethophytic species [4,49–51].

The aim of our study was to detect the activities of all thesenzymes by applying histochemical methods to pistils at differentevelopmental stages in order to understand more fully the

unction of these proteins in pollen–pistil interaction. Peroxidase,sterase and acid-phosphatase activities have already been foundt the surface of receptive stigmas of many species [1,2,11,12,52,53]nd their presence on papillar cells may be due to their active

reaction can be seen within the pollen tubes on a stigma after self-pollination; (F)eely pollinated stigma; (G) pollen grains and their pollen tubes showing no signal, st style. Bars 20 �m.

secretory function. In olive pistils we found active peroxidasesand esterases on young stigmas from flower buds before anthesis.Acid-phosphatase activity was also detected on the stigma, butin later stages of flower development. According to Ref. [15] theexpression of peroxidases on the stigma reflects their function inpollen–pistil interaction, loosening the cell-wall components ofthe stigma to allow penetration of the pollen tubes and acting asa signal in pollen–pistil recognition through H2O2 metabolism.

We found esterase activity in the outer part of the papillar cellsand also discovered its presence in practically all the stages whereperoxidases were detected, although their activity decreased at anearlier stage. Esterase activity was also detectable in both pollengrains and pollen tubes. Although the presence of esterases hasbeen reported in pollen grains of Primula obconica [54], to ourknowledge there are no previous reports about the presence ofthis enzyme activity in germinating pollen grains. The role of thisenzyme in germination and fertilization could well be the controlof the phosphorylation state of proteins involved in the growth ofpollen tube through the pistil tissues.

As has been reported elsewhere [41,55], the olive presents a wetstigma, and although its exudate is not very copious it containsproteins that might correspond to the active enzymes detected inthis work. The stigma is generally receptive at the time of anthesisbut, depending upon the species, it may be receptive only a fewhours after anthesis, as in Tectona grandis [56], or a few days after

anthesis, as in Grevillea robusta [10]. Our results indicate that olive-flower papillar cells are receptive before anthesis and that theyremain receptive until the dehiscence of the anthers, which occursno later than two days after anthesis, even though this decrease in

I. Serrano, A. Olmedilla / Plant Science 197 (2012) 40– 49 47

Fpr

Fpowa

ig. 6. Ribonuclease activity in pollen germinated in vitro in the absence and presence oistils of the same cultivar (self-incompatible), showing a positive reaction for ribonucleibonuclease activity cannot be detected. pg pollen grain; pt pollen tube. Bars 50 �m.

ig. 7. Ribonuclease activity detected in ultrathin sections of self- and cross-pollinated sollinated pistil, cv. Picual, showing positive ribonuclease activity (arrows); (B) pollen tubf cv. Picual with positive reaction to ribonuclease activity (arrows); (C) longitudinal sectihich ribonuclease activity reaction cannot be detected; (D) pollen tube of a cv. Picual p

ctivity. ct cytoplasm; cw cell wall; sg starch granule; vc vacuole. Bars 2 �m.

f self-incompatible pistils of cv. Picual: (A, B) pollen germinated in the presence ofase activity (arrows); (C, D) pollen germinated in the absence of pistils, in which

tyles and pollen germinated in vitro: (A) electron micrograph of the style of a self-e of a pollen grain, cv. Picual, germinated in the presence of self-incompatible pistilson of a style of a cross-pollinated pistil of cv. Picual with pollen of cv. Arbequina, inollen grain germinated in the absence of a pistil, showing no positive ribonuclease

4 Plant S

ridaf

litptspbstwcptsRP[imsppapwtRopmbttApp

tbvgsmdRoco

ipptssotse

8 I. Serrano, A. Olmedilla /

eceptivity proved to depend on pollen arrival. Therefore the stigmas receptive when the anthers of its own flowers are still closed. Thiselay in development is favorable for allogamy because althoughutogamy is not entirely prevented, it could be a physical barrieravoring cross-pollination.

Although we cannot exclude the possibility of a different systemeading the self-incompatible response in the olive, RNase activ-ty was studied because O. europaea shows some characteristics ofhe self-incompatibility of plants belonging to the stylar gameto-hytic system, such as a wet stigma, bicellular pollen and the facthat the growth of its pollen-grain tubes is stopped in the style,howing PCD hallmarks [39–42,55]. Although little is known aboutistil determinants of self-incompatibility in the olive, in specieselonging to families showing gametophytic self-incompatibility,uch as Solanaceae, Scrophulariaceae and Rosaceae, where these pis-il determinants have been studied, it appears that they are proteinsith RNase activity [49]. We found RNase activity in the extra-

ellular matrix of the style of self-pollinated flowers and in theollen tubes of both self-pollinated and freely pollinated olive pis-ils of the Picual cultivar. The high number of pollen tubes thathowed this activity in self-pollinated pistils supports the idea thatNases are involved in the rejection of their own pollen in cv.icual, which has been described as a self-incompatible cultivar36]. The results showing that RNase activity was not detectablen the styles after anthesis in free-pollination, although surprising,

ay be explained by a possible correlation between the inten-ity of this activity and the amount of self-incompatible pollenresent during pollination. In self-pollinated pistils the highesteak of RNase activity occurs in the style at stage 4, when thenthers are open, which means that high quantities of their ownollen reach the style at this stage. At previous and later stages,hen lower quantities of self-incompatible pollen attain the style,

his activity decreases. All of these results seem to indicate thatNase activity occurs just before pollination and that the levelf this activity depends upon the amount of self-incompatibleollen landing on the stigma. Low RNase activity in the styleight not be detectable by the technique used, although it could

e detected in pollen tubes growing within the same pistil dueo the high accumulation of this enzyme activity in these veryhin tubes. After cross-pollination with the compatible cultivarrbequina, no RNase activity was found confirming that in com-atible pollen grains this self-incompatible reaction does not takelace.

In order to confirm that the RNase activity detected was dueo the self-incompatibility response, we germinated pollen in vitrooth in the absence and presence of pistils from the same culti-ar. RNase activity was detected only when the pollen grains wereerminated in the presence of self-incompatible pistils. It has beenhown that the S-RNases (S from self-incompatible) are present inature pistils and that they enter into pollen grains, where they

egrade the RNA of self-incompatible pollen [57]. The fact thatNase activity was only detected in pollen grains in the presencef self-incompatible reinforces the idea that this enzymatic activityould well be responsible for the self-incompatible response in thelive.

RNase activity was also studied at the electron-microscopy leveln self- and freely pollinated pistils, as well as in the tubes ofollen germinated in the presence of self-incompatible and com-atible pistils. These results allowed us to locate more preciselyhe sites of the RNase activity. The positive reaction detected inelf-pollinated pistils inside the vacuole of the cells of the styleeems to indicate that active RNases would be kept in these

rganelles before secretion. This activity was detected in the pollenubes when the pollen was in vitro germinated in the presence ofelf-incompatible pistils, but not when germinated in the pres-nce of compatible pistils. RNase activity was detected mostly in

cience 197 (2012) 40– 49

the cytoplasm of the pollen tubes; this data coincides with pre-vious immunolocation of S-RNases in pollen grains germinatingin self-incompatible pistils [18]. This is also the place wherethe S-RNases would degrade pollen RNA [58], thus inhibiting itsgrowth.

Two different models have been proposed to explain therejection of self-pollen mediated by S-RNases in the style: thedegradation model, where S-RNases not recognized as “self”(not having the same S allele of the pollen) are degraded bythe pollen and a compartmentalization model, in which unrec-ognized S-RNases are sequestered into pollen vacuoles, thushindering their activity. In either case, when the S-RNases arerecognized as “self” by the pollen they are neither degradednor mobilized to the vacuole and remain capable of degradingpollen-tube RNA and inhibiting pollen growth [57]. Our resultsshow RNase activity in the cytoplasm of pollen grains germi-nated in the presence of self-incompatible pistils but not in thosegerminated with self-compatible pistils after cross-pollinationwith cv. Arbequina These results seem to be in accordance withthe degradation model, since RNase activity can be detectedonly after germination in the presence of a self-incompatiblepistil. Nevertheless, the compartmentalization model cannot beruled out since there is no data available about whether S-RNases remain active once they are compartmentalized in pollenvacuoles.

Although in some species acid-phosphatase activity has beenreported only in the stigma [13,59] we found this activity not onlyin the stigma but also in the style of olive flowers. Evidence of a rela-tionship between phosphatase activity and the self-incompatibleresponse in sporophytic self-incompatible systems was found inBrassica oleraceae by using a specific inhibitor of this activity,which made the plants unable to overcome the self-incompatibleresponse [5]. This idea was later extended to species with game-tophytic self-incompatible system [43], in which it was suggestedthat the inhibition of pollen-tube growth could be due to factorsthat cause the production of acid phosphatase. This could be dueto the fact that the phosphorylation state of proteins is importantfor cell to cell communication during pollen–pistil interaction [60],which may explain the presence of this enzyme activity through-out pollination time and not only when the stigma is receptive;i.e. the phosphorylation state of papillar-cell proteins is crucial tothe fate of pollen landing on the surface of the stigma [11]. Inpollen grains these enzymes could be involved in germination, earlypollen.

Our histochemical study leads to a further understanding ofthe mechanisms by which olive flowers are able to receive pollenand how they can reject self-incompatible pollen. Although morework is needed to identify the genetic determinants of the self-incompatibility processes in the olive, and we can not rule out adifferent mechanism controlling self-incompatibility in the olive,the present work supports the idea that RNase activity plays a keyrole in the inhibition of pollen-tube growth in the style, suggest-ing that O. europaea belongs to the group of species with a stylargamethophytic self-incompatible system. Moreover, evidence ofearly receptivity in olive pistils was found, which could act as aphysical self-pollination barrier to avoid inbreeding in addition tothe self-incompatible system.

Acknowledgements

This work was supported by the Spanish MEC ProjectBFU2006-09876/BFI. I. Serrano received a research fellowshipfrom the Ministry of Science and Innovation. We thank Mrs.R. Luque and Mr. F. Yerón (Ingesar S.L.) for their valuable

Plant S

tt

A

fj

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[59] S. Ghosh, K.R. Shivanna, Studies on pollen–pistil interaction in Lilium grandiflo-

I. Serrano, A. Olmedilla /

echnical assistance and Dr. J. Trout for revising our Englishext.

ppendix A. Supplementary data

Supplementary data associated with this article can beound, in the online version, at http://dx.doi.org/10.1016/.plantsci.2012.07.007.

eferences

[1] A. Dafni, M.M. Maués, A rapid and simple procedure to determine stigma recep-tivity, Sex. Plant Reprod. 11 (1998) 177–180.

[2] K.R. Shivanna, D.C. Sastri, Stigma-surface esterase activity and stigma recep-tivity in some taxa characterized by wet stigmas, Ann. Bot. 47 (1981)53–64.

[3] C. Galen, R.C. Plowright, Testing the accuracy of using peroxidase activity toindicate stigma receptivity, Can. J. Bot. 65 (1987) 107–111.

[4] B.A. McClure, V. Haring, P.R. Ebert, M.A. Anderson, R.J. Simpson, F. Sakiyama, A.E.Clarke, Style self-incompatibility products of Nicotiana alata are ribonucleases,Nature 342 (1989) 955–957.

[5] C.P. Scutt, A.P. Fordham-Skelton, R.R.D. Croy, Okadaic acid causes breakdown ofself-incompatibility in Brassica oleracea: evidence for the involvement of pro-tein phosphatases in the incompatibility response, Sex. Plant Reprod. 6 (1993)282–285.

[6] R.R. Williams, Techniques used in fruit set experiments, in: R.R. Williams, D.Wilson (Eds.), Towards Regulating Cropping, Grower Books, London, 1970, pp.57–61.

[7] F. Orlandi, B. Romano, M. Formaciari, Effective pollination period estimation inolive (Olea europaea L.): a pollen monitoring application, Sci. Hortic. 105 (2005)313–318.

[8] J. Cuevas, V. Pinillos, V.S. Polito, Effective pollination period for Manzanillo andPicual olive tres, J. Hortic. Sci. Biotechnol. 84 (2009) 370–374.

[9] K.R. Shivanna, V.K. Sawhney, Pollen Biotechnology for Crop Production andImprovement, Cambridge University Press, Cambridge, UK, 1997.

10] A. Kalinganire, C.E. Harwood, M.U. Slee, A.J. Simons, Floral structure, stigmareceptivity and pollen viability in relation to protandry and self-incompatibilityin silky oak (Grevillea robusta A. Cunn), Ann. Bot. 86 (2000) 133–148.

11] M.K. Kandasamy, M.K. Thorness, S.J. Rundle, M.L. Goldberg, J.B. Nasral-lah, M.E. Nasrallah, Ablation of papillar cells function in Brassica flowersresults in the loss of stigma receptivity to pollination, Plant Cell 5 (1993)263–275.

12] M. Stpiczynska, Stigma receptivity during the life span of Platanthera chloranthaCuster (Rchb) flowers, Acta Biol. Cracov. Ser. Bot. 45 (2003) 37–41.

13] M. Herrero, H.G. Dickinson, Pollen–pistil incompatibility in Petunia hybrida:changes in the pistil following compatible and incompatible intraspecificcrosses, J. Cell Sci. 36 (1979) 1–18.

14] Y. Heslop-Harrison, K.R. Shivanna, The receptive surface of the angiospermstigma, Ann. Bot. 41 (1977) 1233–1258.

15] S.M. McInnis, D.C. Emery, R. Porter, R. Desikan, J.T. Hancock, S.J. Hiscock, The roleof stigma peroxidases in flowering plants: insights from further characteriza-tion of a stigma-specific peroxidases (SSP) from Senecio squalidus (Asteraceae),J. Exp. Bot. 57 (2006) 1835–1846.

16] B.A. McClure, New views of S-RNase-based self-incompatibility, Curr. Opin.Plant Biol. 9 (2006) 639–646.

17] A.C. Certal, A.M. Sanchez, H. Kokko, W. Broothaerts, M.M. Oliveira, J.A. Feijó, S-RNases in apple are expressed in the pistil along the pollen tube growth path,Sex. Plant Reprod. 12 (1999) 94–98.

18] D. Luu, X. Qin, D. Morse, M. Cappadocia, S-RNase uptake by compatible pollentubes in gametophytic self-incompatibility, Nature 407 (2000) 649–651.

19] T.R. Boskovic, K.R. Tobutt, H. Duval, I. Batlle, F. Dicenta, F.J. Vargas, A stylarribonuclease assay to detect self-compatible seedlings on almond progenies,Theor. Appl. Genet. 99 (1999) 800–810.

20] D. Barranco, Variedades y patrones, in: D. Barrando, R. Fernández Escobar, L.Rallo L (Eds.), El Cultivo del Olivo, Ediciones Mundi-Prensa y Junta de Andalucía,Madrid, 2007, pp. 63–88.

21] S. Lavee, Z. Datt, The necessity of cross-pollination for fruit set of Manzanilloolives, J. Hortic. Sci. 53 (1978) 261–266.

22] J. Cuevas, V.S. Polito, Compatibility relationships in Manzanillo olive, Hortic.Sci. 32 (1997) 1056–1058.

23] S. Lavee, J. Taryan, J. Levin, A. Haskal, The significance of cross-pollination forvarious olive cultivars under irrigated intensive growing conditions, Olivae 91(2006) 25–36.

24] N. Moutier, Self-fertility and inter-compatibilities of sixteen olive varieties,Acta Hortic. 586 (2002) 209–212.

25] N. Ghrisi, B. Boulouha, M. Benichou, S. Hilali, Agro-physiological evaluation ofthe phenomenon of pollen compatibility in olive. Case of the Mediterranean

collection at the Menara station, Marrakech, Olivae 79 (1999) 51–59.

26] J. Cuevas, A.J. Diaz Hermoso, D. Galian, J.J. Hueso, V. Pinillos, M. Prieto, D. Sola,V.S. Polito, Response to cross pollination and choice of pollinisers for the olivecultivars (Olea europaea L.) ‘Manzanilla de Sevilla’, ‘Hojiblanca’ and ‘Picual’,Olivae 85 (2001) 26–32.

[

cience 197 (2012) 40– 49 49

27] J. Guerin, M. Sedgley, Cross-pollination in Olive Cultivars, RIRDC PublicationNo. 07/169.

28] A. Díaz, P. Rallo, R. De la Rosa, Self- and cross-incompatibility mechanisms: astrategy to ensure a great variability in olive (Olea europaea L.) populations,Olea 25 (2006) 29–33.

29] A. Díaz, A. Martin, P. Rallo, R. De la Rosa, Cross-compatibility of the parents asthe main factor for successful olive (Olea europaea L.) breeding crosses, J. Am.Hortic. Sci. 132 (2007) 1–6.

30] S.G. Sibbett, M. Freeman, L. Ferguson, V.S. Polito, Effect of topically applied‘Sevillano’ pollen on normal seeded and parthenocarpic shotberry fruit set of‘Manzanillo’ olive, Hortic. Technol. 2 (1992) 228–230.

31] E. Antognozzi, A. Standardi, Studio della biologia fiorale negli olive Gentile diChieti e Dritta di Moscufo, Riv. Ortoflorofrut. Ital. 62 (1978) 461–469.

32] I.I. Androulakis, M.H. Loupassak, Studies on self-fertility of some olive cultivarsin the area of Crete, Acta Hortic. 286 (1990) 159–162.

33] R.P. Singh, P.L. Kar, Compatibility studies in some olive cultivars, Prog. Hortic.12 (1980) 9–15.

34] S. Bartolini, R. Guerriero, Self-compatibility in several clones of oil olive cv.Leccino, Adv. Hortic. Sci. 9 (1995) 71–74.

35] S. Moorkerjee, J. Guerin, G. Collins, C. Ford, M. Sedgley, Paternity test analysisusing microsatellite markers to identify pollen donors in an olive grove, Theor.Appl. Genet. 111 (2005) 1174–1182.

36] A. Díaz, R. De la Rosa, A. Martín, P. Rallo, Development, characterization andinheritance of new microsatellites in olive (Olea europaea L.) and evaluation oftheir usefulness in cultivar identification and genetic relationship studies, TreeGenet. Genomes 2 (2006) 165–175.

37] A.F. Ateyyeh, R. Stosser, M. Qrunfleh, Reproductive biology of the olive (Oleaeuropaea L.) cv. Nabali Baladi, J. Appl. Bot. 74 (2000) 255–270.

38] S.B. Wu, G. Collins, M. Sedgley, Sexual compatibility within and between olivecultivars, J. Hortic. Sci. Biotechnol. 77 (2002) 665–673.

39] D. De Nettancourt D, Incompatibility in Angiosperms, Sex. Plant Reprod. 10(1997) 185–199.

40] M.M. Altamura Betti, G. Pascua, G. Mazzolani, Development of the female game-tophyte in Olea europaea L., Ann. Bot. (Roma) XL (1982) 111–117.

41] F. Ciampolini, M. Cresti, R.N. Kapil, Fine structural and cytochemical character-istics of style and stigma in olive, Caryologia 36 (1983) 211–230.

42] I. Serrano, S. Pelliccione, A. Olmedilla, Programmed cell death hallmarks in thepollen and stigmatic papillar cells of Olea europaea L. under free pollination,Plant Cell Rep. 29 (2010) 561–572.

43] D. De Nettancourt, Incompatibility and Incongruity in Wild and CultivatedPlants, Springer-Verlag, Berlin/Heidelberg/New York, 2001.

44] S. Higuchi, M. Suga, A.M. Wannenberg, B.H. Schofield, Histochemical demon-stration of enzyme activities in plastic and paraffin embedded tissue sections,Stain Technol. 54 (1979) 5–12.

45] G. Gomori, Microscopic Histochemistry, Principles and Practice, University ofChicago Press, Chicago, 1952.

46] P.B. Gahan, Plant Histochemistry and Cytochemistry: An Introduction, Aca-demic Press, London, 1984.

47] I. Serrano, M.C. Romero-Puertas, M. Rodríguez-Serrano, L.M. Sandalio, A.Olmedilla P, Peroxynitrite mediates programmed cell death both in papillarcells and in self- incompatible pollen in the olive (Olea europaea L.), J. Exp. Bot.63 (2012) 1479–1493.

48] R.B. Knox, pollen–pistil interactions, in: H.F. Linskens, J. Heslop-Harrison(Eds.), Cellular Interactions (Encyclopedia of Plant Physiology, New Series 17),Springer, New York/Berlin/Heidelberg, 1984, pp. 508–608.

49] T.H. Kao, T. Tsukamoto, The molecular and genetic bases of S-RNase-based self-incompatibility, Plant Cell 16 (2004) S72–S83.

50] B.A. McClure, V.E. Franklin-Tong, Gametophytic self-incompatibility: under-standing the cellular mechanism involved in self pollen tube inhibition, Planta224 (2006) 233–245.

51] T.H. Kao, T. Tsukamoto, The molecular and genetic bases of S-RNase-based self-incompatibility, Plant Cell 16 (2004) S72–S83.

52] R. Seymour, A.J. Blaylock, Stigma peroxidase activity in association with ther-mogenesis in Nelumbo nucifera, Aquat. Bot. 67 (2000) 155–159.

53] I. Dupuis, C. Dumas, Biochemical markers of female receptivity in maize (Zea-mays L.) assessed using in vitro fertilization, Plant Sci. 70 (1990) 11–19.

54] V.A.M. Stevens, B.G. Murray, Studies on heteromorphic self-incompatibilitysystems: the cytochemistry and ultrastructure of the tapetum of Primula obcon-ica, J. Cell Sci. 50 (1981) 419–431.

55] I. Serrano, C. Suárez, A. Olmedilla, H.F. Rapoport, M.I. Rodríguez-García, Struc-tural organization and cytochemical features of the pistil in Olive (Olea europaeaL.) cv. Picual at anthesis, Sex. Plant Reprod. 21 (2008) 99–111.

56] S. Tangmitcharoen, J.N. Owens, Pollen viability and pollen tube growth follow-ing controlled pollination and their relation to low fruit production in teak(Tectona grandis Linn F.), Ann. Bot. 80 (1997) 401–410.

57] B. McClure, Darwin’s foundation for investigating self-incompatibility and theprogress toward a physiological model for S-RNase-based SI, J. Exp. Bot. 60(2009) 1069–1081.

58] B. McClure, F. Cruz-Garcia, C. Romero, Compatibility and incompatibility in S-RNase-based systems, Ann. Bot. (2011), http://dx.doi.org/10.1093/aob/mcr179.

rum, Phytomorphology 32 (1982) 385–395.60] S.J. Rundle, J.B. Nasrallah, Molecular characterization of a type 1 ser-

ine/threonine phosphatase from Brassica oleracea, Plant Mol. Biol. 20 (1992)367–375.