Reproductive biology of Olive ( Olea europaea L.) DOP Umbria cultivars

11 pages
of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Olive trees have a plentiful bloom but a low percentage of normal fruit set. To improve fruit set, numerous investigations have sought to identify the obstacles that prevent full production. In this work, flower development in five DOP Umbria
  ORIGINAL ARTICLE L. Reale   C. Sgromo   T. BonofiglioF. Orlandi   M. Fornaciari   F. FerrantiB. Romano Reproductive biology of Olive ( Olea europaea   L.) DOP Umbria cultivars Received: 14 November 2005/ Accepted: 26 May 2006/Published online: 25 August 2006   Springer-Verlag 2006 Abstract  Olive trees have a plentiful bloom but a lowpercentage of normal fruit set. To improve fruit set,numerous investigations have sought to identify theobstacles that prevent full production. In this work,flower development in five DOP Umbria cultivars(Leccino, Frantoio, Moraiolo, Dolce Agogia and SanFelice) was studied throughout different developmentalphases, from before microsporogenesis and megaspo-rogenesis to post-anthesis, by morphological and cyto-histological observations. Dolce Agogia was the mostprecocious cultivar, while full flowering was simulta-neous in Leccino, Frantoio, Moraiolo and San Felice.Frantoio and Leccino were also good pollen producers,having the highest percentage of pollen viability andgerminability. Dolce Agogia can also be considered agood pollen producer in terms of the high quantity of released pollen, but it had the lowest levels of pollenviability and germinability and the highest percentage of aborted flowers and ovaries. Morphological and cyto-histological observations on the number of flowers perinflorescence and the number of aborted flowers andovaries suggest that fruit set was not influenced by thenumber of flowers per inflorescence, but rather by thenumber of inflorescences, which depends on the globalfruiting potential of the tree. Keywords  Olea europaea  L  Æ  Flower biology  Æ Microsporogenesis  Æ  Megasporogenesis  Æ Pollen viability and germinability Introduction Olive ( Olea europaea  L.), like many other hermaphro-ditic plants, is characterized by plentiful flowering fol-lowed by poor fruit set and consequently low yield(Stephenson 1981). Several hypotheses have been pre- sented to explain this common phenomenon, includingresource and pollen limitation (Sutherland 1986, 1987; Burd 1998). A plant is considered pollen-limited if additional pollen increases fruit or seed production(Schemske 1980; Willson and Schemske 1980; Bier- zychudek 1981; Burd 1994; Larson and Barrett 2000), whereas a plant receiving sufficient pollen is consideredresource-limited if the addition of resources increasesfruit or seed production (Zimmerman and Aide 1989). However, pollen and resource limitation are not neces-sarily mutually exclusive phenomena and both may leadto low reproductive output.The production of surplus female flowers and theirsubsequent abscission appears to provide plants withsome degree of control over the number and quality of their offspring. Plants can match fruit production withthe available resources over a wide range of environ-mental conditions even though the number of pollinatedflowers, the total resources, the distribution of resourcesamong fertile branches, and the amount of fruit/seedpredation may be unpredictable at anthesis. Surplusflower and fruit production may provide plants with achoice of which offspring should mature (Stephenson1981). Higher quality offspring should result if thematernal parent can select which fruits should maturebased on the quality or quantity of pollen deposited(Bertin 1990). It has been suggested that pollen intensityinfluences the quality of seeds produced, in that highpollen loads intensify competition among male gameto-phytes, with only the most vigorous gametophyteseffecting fertilization and producing zygotes vigorousenoughtoformintoseeds.Nevertheless,ifthequalityandsource of pollen can always limit the reproductive success(Tangmitcharoen and Owens 1997), the effect of pollen Communicated by S. RussellL. Reale ( & )  Æ  C. Sgromo  Æ  T. Bonofiglio  Æ  F. OrlandiM. Fornaciari  Æ  F. Ferranti  Æ  B. RomanoDepartment of Plant Biology and Zootechnicaland Agroenvironmental Biotechnology,University of Perugia, Borgo XX Giugno 74,06121 Perugia, ItalyE-mail: citolabo@unipg.itTel.: +39-075-5856407Fax: +39-075-5856069Sex Plant Reprod (2006) 19: 151–161DOI 10.1007/s00497-006-0032-6  quantity has only been demonstrated in a few species of noncultivated plants including  Campsis radicans  (Bertin1990). Such effects have been seen, however, in severalcultivated species, including  Zea mays  (Ottaviano et al.1983) and  Cucurbita pepo  (Stephenson et al. 1986). Studies carried out in different Mediterranean coun-tries, on olive and grape, have demonstrated a correla-tion between the quantity of pollen released duringflowering and fruit production (Besselat 1992; Fornaciari et al. 1998; Minero et al. 1998). In particular, the con- centration of pollen during the second ten days of theflowering period is directly correlated with future fruitproduction (Orlandi et al. 2005). This finding demon-strates the need to determine the quantity of pollen re-leased, the physiological characteristics of the pollen andthe exact period in which pollination takes place.In olive and in hermaphroditic plants, in general, thepresence of staminate or imperfect flowers is frequentlyobserved; such flowers have a normal-sized perianth andandroecium but abnormal pistils, which are shriveledand smaller than normal (Uriu 1959; Griggs et al. 1975; Cuevas and Polito 2004). Staminate flowers may en-hance plant fitness in at least two ways: (1) they canaugment the attractiveness of plant pollinators, thusensuring that its flowers receive enough pollen and (2)they can enhance pollen donation to fertilize otherflowers. In olive, the most obvious advantage providedby staminate flower production is the increase in thenumber of pollen grains available to achieve fertilization(Cuevas and Polito 2004). This is particularly important under anemophily and, as pointed out by Stephensonand Bertin (1983), in species experiencing intense male competition in which the likelihood of paternity isinfluenced by the quantity of pollen produced.The large variation observed in olive in the propor-tion of staminate flowers among years, trees, branches,shoots and even inflorescences on the same shoot(Brooks 1948) may be part of an overall reproductivestrategy that adjusts maternal investment in sex expres-sion in response to available resources and environ-mental conditions. Variations in fertility and pistilabortion are related to different factors such as light,temperature, humidity, moisture availability, nutritionalconditions and leaf-to-bud ratio (Uriu 1959; Ferrara et al. 1991). High leaf/flower ratios and nitrogen fertil-ization promote hermaphroditic flower production. Inthe Santulhana cultivar, the alterations leading to pistilabortion started in epidermal cells and proceeded fromthere into internal tissues. The condition of the pistils atthe onset of lesions and the similarity of their responsewith the response of resistant forms to plasmodialorganisms suggests that biotic factors may be involvedin pistil damage (Medeira et al. 2002). While thesestudies were carried out on pistil development, mostresearch concerning fruit set and production potential of olive have dealt with the post-anthesis period. In mostcases, the level of normal fruit set in olive is based onheavily blooming trees in their ‘‘ON’’ year (Lavee et al.1996). This raises the question whether in years withfewer inflorescences and flowers, the fruit set and normalfruit retained on the tree until maturation will be at thesame relative level. A study conducted on  Styrax obassia reported that the initial and mature fruit set ratios werenegatively affected by inflorescence number per tree(Kato and Hiura 1999). Similarly Andersson (1988) found that the proportion of flowers that set fruit de-creased with inflorescence number over the whole size in Anchusa officinalis . Inflorescence size, as well as inflo-rescence numbers, seem to be negatively correlated withfruit ratios (Kato and Hiura 1999). Willson and Price (1977) reported that inflorescence size was commonlylimited phenotypically by the availability of energy ornutrients. In olive, fruiting potential is also determinedby its growing conditions and vigor, but the relationshipof both inflorescence number and flowers per inflores-cence to fruit set distribution is not clear. Another aspectthat has not been sufficiently defined is the degree of variability in flower quality within each inflorescenceand its relationship to potential fruit set and survival.Furthermore, understanding compatibility betweencultivars could reduce the many problems of self sterilitythat have been observed in olives. A heterogeneouslytimed pollen emission could lead to lower reproductivesuccess, since self-sterile cultivar may not be receptivewhen pollen is released from other cultivars or couldperhaps have passed their optimum receptivity, whichwould likely result in reduced fertilization. Greaterhomogeneity would instead result in pollen emissionthroughout pistil receptivity and may result in morefrequent cross-pollination between different self-sterilecultivars, resulting in earlier and higher levels of fertil-ization (Cuevas 1992) and consequently increasing totalproduction values (Orlandi et al. 2005a, b). Differences between various olive cultivars were observed regardingthe onset and the start and length of anthesis period(Pannelli et al. 2000). These differences are strictly cor-related to differences in micro- and megagametogenesis.Synchrony and asynchrony of micro- and megagameto-genesis are cultivar-specific characteristics (Bini 1981).Understanding the correlation between flower devel-opment, flower number and fruit production, shouldprove useful in optimizing cultivars to be planted in newolive orchards to increase olive production. In this studywe examine the timing of micro- and megasporogenesisand flowering in five DOP Umbria cultivars (Frantoio,Leccino, Moraiolo, Dolce Agogia and San Felice) usingmorphological and cyto-histological observations andexamine timing, pollen viability and germinability atanthesis and fruit set among these cultivars with the goalof optimizing pollination as a means to enhance fruit set. Materials and methods Plant materials and growth conditionField observations and experiments were carried out,over a 2-year period, in an olive grove of the ‘‘Istituto 152  Sperimentale di Olivicoltura ‘‘ of Spoleto, located inCastel Ritaldi, a village near Spoleto in southern Um-bria (42  49 ¢ N, 12  40 ¢ W) at 300 m above sea level. Themean relative humidity (RH) in this area is 82%, withaverage annual rainfall of 937 mm. In this olive grove,trees belonging to cultivars Frantoio, Leccino, Morai-olo, Dolce Agogia and San Felice were selected. Mete-orological data concerning this location were obtainedfrom the Regional Hydrogeographic Service for 2004and 2005.Morphological observationsThe number of flowers was determined at the start of flowering and at the end of anthesis on 40 labeled infl-orescences from 4 plants of each cultivar. At the end of anthesis, the number of aborted flowers (closed anddried flowers or open flowers with dried ovaries) andaborted ovaries (with dried style–stigma and shrunkenovary immediately after petal abscission) were countedon the same inflorescences. About 2 weeks after anthe-sis, the number of drupes was counted for each inflo-rescence. To observe floral phenology, in 2005, thenumber of open flowers per inflorescence was deter-mined every 2 days in 20 labeled inflorescences from 4plants of each cultivar, during the period from buddingto the end of anthesis (3–15 June).Cyto-histological observationsCyto-histological observations were carried out at dif-ferent times of flower development using flowers col-lected from each cultivar. Flowers were fixed in 3% (w/v) glutaraldehyde in 0.075 M cacodylate buffer, pH 7.2,for 5 h. Samples were then washed four times for 15 mineach in 0.075 M cacodylate buffer, pH 7.2, and werepost-fixed in 1% (w/v) OsO 4 . At this stage, samples weredehydrated in increasing concentrations of ethanol andwere embedded in resin (Epon, 2-dodecenylsuccinicanhydride, and methylnadic anhydride mixture) (Realeet al. 2002). Semi-thin sections (1–2  l m) were cut withan ultramicrotome (OmU2, Reichert, Heidelberg)equipped with a glass blade, stained with toluidine blueand observed using a light microscope (DMLB, Leica,Wetzlar, Germany).Pollen viabilityTo check pollen viability, flowers of the investigatedcultivars (Frantoio, Leccino, Moraiolo, Dolce Agogiaand San Felice) were collected immediately before andduring anthesis and pollen was labeled with fluoresceindiacetate (FDA) according to Heslop-Harrison andHeslop-Harrison (1970) and observed using epifluores-cence microscopy.In vitro pollen germinationMature pollen grains of the cultivars Frantoio, Leccino,Moraiolo, Dolce Agogia and San Felice were germi-nated in Brewbaker and Kwack medium (1963) at 25  Cin the dark. Pollen grains were placed on the surface of 10 ml of medium in plastic Petri dishes of 60 mmdiameter (Majewska-Sawka et al. 2002). After 24 h of incubation without shaking, the pollen was observedusing a light microscope (DMLB, Leica, Wetzlar, Ger-many), and the germinated grains were counted by LeicaIM 1000 image analysis software.Statistical analysisCollected data on differences between cultivars, collec-tion years, flowering phenology, number of flowers perinflorescences, percentage of aborted flowers and ova-ries, pollen viability and germinability, and number of drupes per inflorescence were examined using analysis of variance (ANOVA). Source cultivars producing signifi-cant results were subjected to a Fisher probable least-squares difference (PLSD) test. The overall probabilityof rejecting the hypothesis was examined using multiplepairwise comparisons by a modified Bonferroni methodproposed by Holm (1979). Results Flower biologyThe DOP Umbria cultivars Frantoio, Leccino, Morai-olo, Dolce Agogia and San Felice were selected in theolive grove of the ‘‘Istituto Sperimentale di Olivicoltura’’of Spoleto. Flower development was observed in thesecultivars to correlate fruit set with pollen production,quality and timing.In the first year of the investigation (2004), flowersfrom different cultivars were collected at different stagesof development, from flower bud (second week of May)to the end of anthesis (fourth week of June). Micro-scopic observation of semi-thin sections showed that inthe second week of May the anther epidermis (exothe-cium), primary parietal cells and primary sporogenouscells were evident in cvs. Dolce Agogia and Frantoio(Fig. 1a). At the same time, the androecium andgynoecium were still not completely differentiated in theMoraiolo flower (Fig. 1b). In the anthers of the Leccinoand San Felice flowers, only the primary sporogenouscells had differentiated, but not the primary parietalcells.During the last week of May, microsporogenesis tookplace in the anthers of Dolce Agogia and Frantoio andtetrads were enveloped by a thick callose special wall(Fig. 1c). Microspores displayed a simultaneous-typearrangement. Cultivars Leccino and San Felice, 153  Fig. 1  Cyto-histological observations of micro- and megasporo-genesis:  a  semithin section of Dolce Agogia anther with epidermis,primary parietal cell and sporogeneous tissue differentiated;  b longitudinal section of very young flower of Moraiolo cv. withdeveloping petal and anther;  c  tetrads enveloped by callose wall of Frantoio cv. with large cells of tapetum at bottom of micrograph;  d transverse section of anther with anther wall completely differen-tiated (epidermis, transition layers, tapetum) and microsporemother cells;  e  cross section of ovule with one of two nucleus of functional dyad evident;  f   longitudinal section of ovule with funicle,integument, nucleus and megaspore mother cell;  g  section of antherlocule containing young microspores and strongly vacuolatedtapetum cells.  a  Anther;  ep  epidermis,  f   funicle,  fd   one nucleus of functional dyad,  i   integument,  m  microspore,  memc  megasporemother cell,  mimc  microspore mother cell,  n  nucellus,  ppc  primaryparietal cell,  pe  petal,  st  sporogeneous tissue,  t  tetrad,  ta  tapetumcell,  tl   transition layer.  Bars  20  l m154  however, displayed pollen mother cells surrounded byan early callose wall (Fig. 1d), and Moraiolo had iso-diametric archesporial cells.At the same time, the completion of megasporogen-esis and the functional dyad formation were observed inthe pistils of cvs. Dolce Agogia and Frantoio (Fig. 1e).Leccino and San Felice had megaspore mother cellsenveloped by callose, while young megaspore mothercells (Fig. 1f) or primary sporogenous cells were ob-served in the Moraiolo ovules.In the first week of June, microsporogenesis wascompleted in all the cultivars, and microspores wereobserved (Fig. 1g). In the same period, megasporogen-esis was completed in cv. Moraiolo. In the second weekof June, anthesis was first observed in cultivars DolceAgogia and Frantoio. Full anthesis in all of the cultivarswas registered on 17 June; at which time the percentageof open flowers was higher than 45% in Dolce Agogiaand Leccino (47.6 ± 5.4 and 47.6 ± 7.6, respectively)while in Frantoio and San Felice, the values were35.8 ± 4.8 and 39.1 ± 7.4, respectively. A very lowpercentage of open flowers was observed in cv. Moraiolo(18.9 ± 5.6). The latest anthesis (23 June) was observedin cv Leccino.In the second year of the investigation (2005), flowerdevelopmental phenology was studied in the same plantsby counting the number of open flowers every 2–3 day,starting in the first week of June to the end of floralanthesis during the third week of June (Fig. 2).On 3 June, only Dolce Agogia showed the first signsof anthesis. Anthesis began on 6 June in Frantoio,Leccino, Moraiolo and San Felice, with the highestnumber of open flowers (full anthesis) observed on 10June. Full anthesis was observed in Dolce Agogia on 8June (Fig. 2). In all cultivars, the number of openflowers decreased markedly by 13 June, with the latestanthesis observed on 15 June. A two-way ANOVAstatistical test indicated that interaction effects betweencultivar and time were very highly significant( F  16,525  = 4.738,  P  = 4.93e-009).At full anthesis, Dolce Agogia displayed the highestpercentage of open flowers whereas the lowest value wasobserved in Moraiolo. ANOVA tests indicate that per-centages of open flowers at full anthesis for cultivarsDolce Agogia and Leccino were statistically differentfrom the values registered in Moraiolo; the other culti-vars had intermediate values ( F  4,84  = 3.428, P  = 0.012).Temperature trends were also noted during the studyperiod. In February and March of 2005 the daily meantemperature values (ranged from   2.2 to 12.2  C inFebruary and from  6.3 to 20.8  C in March) were lowerthan those observed in the same period in 2004 (from 1.2to 18.8  C in February and from 2.1 to 17.4  C in March).In contrast, the mean temperatures in the last week of March and particularly in May were higher in 2005 thanin 2004 (ranged from 16.8 to 30.6  C in 2005 and from12.2 to 23.7 in 2004).Flower number per inflorescenceBefore and at the end of anthesis, in both years of study,thenumberofflowersperinflorescencewascountedon40labeled inflorescences for each cultivar (Fig. 3). Thenumber of flowers observed before anthesis (Fig. 3a) wassignificantly higher for all cultivars in 2004 compared to2005 (Frantoio:  F   = 12.43; Leccino:  F   = 17.38; Morai-olo:  F   = 5.405; Dolce Agogia:  F   = 18.051; San Felice: F   = 35.922;forall df  (1,17)and P  £  0.05).InbothyearsMoraiolo had the most flowers per inflorescence, whereasSan Felice had the minimum value. In fact no significantinteraction effects were detected by a two-way ANOVAanalysis between cultivar and year ( F  4,170  = 1.26287, Fig. 2  Flowering trends in cultivars Frantoio, Leccino, Moraiolo,Dolce Agogia and San Felice in 2005. The  y -axis is the percentage(%) of open flowers, the  x -axis gives the sampling dates. For eachcultivar percentages of open flowers at the different dates werecompared (Frantoio:  F  4,68  = 10.824,  P  = 0.0001; Leccino: F  4,68  = 16.532,  P  = 0.0001; Moraiolo:  F  4,68  = 9.528, P  = 0.0001; Dolce Agogia:  F  4,76  = 20.178,  P  = 0.0001; SanFelice:  F  4,76  = 11.233,  P  = 0.0001).  Different letters  indicatestatistically different values ( P  £  0.005 after Bonferroni correc-tion)155
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks