Influences of Cross Pollination on Pollen Tube Growth and Fruit Set
in Zuili Plums (Prunus salicina)
Hui-Juan Jia1*, Feng-Jie He1, Cai-Zhen Xiong 2, Fu-Rong Zhu 3 and Goro Okamoto4
(1. Department of Horticulture, Zhejiang University, Hangzhou 310029, China;
2. Fores and Silkworm Statipon, Xiuzhou District, Jiaxing 314000, China;
3. Department of Agriculture and Economic,Tongxiang 314500,China;
4. Faculty of Agriculture, Okayama University, Okayama 700-8530, Japan)
Handling editor: Hong-Wen Huang
Received 29 Mar 2006 Accepted 27 Jun. 2006
Supported by the Science Project of Jiaxin，Zhejiang，China
*Author for correspondence.
Tel: +86 (0)571 8570 4223;
© 2006 Institute of Botany, the Chinese Academy of Sciences
Zuili plum trees usually set fruit poorly, although they produce high quality fruit. To elucidate the causes of the poor fruit set, pollen tube growth into pistils and fruit set percentage were investigated after cross-, self- and open-pollination. Ovule development in Zuili pistils was also investigated. Pollen tube penetration into the ovules via the obturator and micropyle was best when Zuili pistils were pollinated by cv. Black Amber (P. domestica L.) pollen grains, although cross-pollinations with Hongxinli and Miili (P. salicina, L.) pollen were more effective than self- and open-pollination. The fruit set percentage was also highest in pistils pollinated with Black Amber pollen grains. Morphological observation of Zuili pistils revealed that the trees produce ‘double pistils’, developing two ovaries from a basal pistil, at a rate as high as 28%. In such abnormal pistils, most ovules were lacking an embryo sac or were entirely degenerated. The percentage of normally developed ovules was 24.3% and 8.9% in normal and double pistils, respectively. From these results, we conclude that the main causes of poor fruit set of Zuili plums are a lack of effective cross-pollination and the production of high percentages of double pistils in which normally developed ovules are scarcely formed.
Key words: double pistils; pollen tube; pollination; poor fruit set; Zuili plum.
Jia HJ, He FJ, Xiong CZ, Zhu FR, Okamoto G (2006). Influences of cross pollination on pollen tube growth and fruit set in Zuili plums (Prunus salicina). J. Integr. Plant Biol. 48(x), xxx–xxx.
Zuili is a cultivar of plum (Prunus salicina L.) that is distributed in the Jiaxin district in Zhejiang Province, southern China. It has been planted for over 2000 years, probably because of its high fruit quality with abundant juiciness, high sugar (higher than 17%), low acidity (less than 0.8%) and its high acceptance by most consumers (Chen et al. 1987). The fruits are globular with a faint suture with red skin and yellow flesh, and covered with a thin powder. Despite the high quality of the fruit, cultivation of this plum in this area has been limited, mainly due to inadequate fruit set (Chen et al. 1987). Self-incompatibility might be one reason for the poor fruit set in Zuili. The trees produce large numbers of flowers but the fruit set percentage is usually very low, which has discouraged growers to cultivate this plum. There are few studies which have clearly confirmed the causes of the poor fruit set. This lack of cultural information is inhibiting the industrial development of Zuili production.
It is well established that fruit set and yield are strongly dependent on genotype and genotype interactions for many important fruit crops within the family of Rosaceae, such as apples, pears, cherries, apricots and almonds. Insufficient fruit set in plums may be due to a genetic predisposition for abnormal embryo sac development (Thompson and Liu 1973), and low temperature conditions during and after bloom time that result in poor pollen tube growth (Jaumien 1968). Poor fruit set also results from a shortened effective pollination period of the stigma as well as discrepancy between ovule longevity and the time required for the pollen tube to reach the egg cell (Williams 1965, 1970).
Improved and sustainable production of Zuili plums would be possible by interplanting proper pollinizer plants if cross-compatibility between Zuili is elucidated. In this study, we focused on the compatibility of several pollen parents for Zuili by studying pollen germinability and tube growth in vitro, pollen tube growth into the pistil, and ovule development. Finally, we investigated the effect of cross- and self-pollination on the fruit set of Zuili.
Quantity and quality of pollen grains
As shown in Table 1, different results were obtained for pollen viability when assessed by in vitro tests and fluorescent microscopic examination of the Zuili stigma surface (Figure 1). Among the four cultivars, the highest pollen germinability, a notably high value of over 37%, was recorded in Black Amber pollen, about 25% in Miili and Hongxinli, while only 13.2% in Zuili pollen. However, tube length and rate of tube growth were more preponderant in Miili pollen than those in other cultivars.
Black Amber pollen grains germinated well also on the Zuili stigma surface (Figure 1A). On the contrary, Zuili pollen, the self pollen, exhibited significantly poorer germination than other pollen sources. The most pollen tube tips swelled or were arrested to grow further (Figure 1B). A notable number of pollen grains were found on open-pollinated stigmas but pollen tubes that penetrated into the style were few (Figure 1C).
Pollen tubes, elongated into various parts of a pistil, were easily distinguishable by fluorescent microscopic observation of cross sections stained with aniline blue (Figure 2). The most rapid pollen tube growth into the style base and upper ovary was observed in pistils cross-pollinated by Miili pollen when compared 4 days after pollination (DAP) (Table 2). The trend was also found at 8 DAP, although similar pollen tube growth was also detected in pistils pollinated by Hongxinli pollen. In pistils of self- and open-pollination treatment, few pollen tubes grew into the ovary even at 8 DAP. In pistils of self-pollination, pollen tubes grew slowly and only a small number of pollen tubes finally reached the micropyle through the obturator. A similar rate of pollen tube growth was detected for open-pollination treatment; the average number of pollen tubes reaching the micropyle was below 0.1. The number of pollen tubes that finally penetrated into ovule was higher in pistils pollinated by Black Amber and Miili pollen than by Hongxinli pollen. The numbers of pollen tubes counted in various parts of ovaries at 8 DAP decreased when measured at 14 DAP in each pollination treatment.
Most of the fruit drop in all treatments occurred within 3 weeks after pollination. Fruit set was strongly affected by the pollen source (Table 3). Open-pollination resulted in a very low set percentage of 5.2%. Among cross-pollinations, percentages of fruit set in Hongxinli, Miili, and Black Amber were 8.8%, 10.6%, and 14.6%, respectively, significantly higher when Black Amber was selected as the pollen source. Self-pollination resulted in the lowest fruit set, 0.7%.
Ovule and embryo sac development
Zuili plum normally contains two ovules per pistil: the primary ovule within which an embryo sac develops and an abortive secondary ovule which normally degenerates. In some cases both primary and secondary ovules degenerate after pollination, in others both ovules remain viable (Figure 3). It is noted that there were two types of pistils in Zuili cultivar, single and double (Figure 4). Out of 50 pistils examined, 14 pistils were the double type, in which more than 90% of the ovaries were not viable. Degeneration of ovules in single type pistils was observed at 29.7% of the 74 ovules examined and 45.9% of the ovules contained no embryo sac (Table 4).
There was significant inosculate between the pollinations as measured by both pollen tube growth and final fruit set. In vitro tests of pollen germination and tube elongation revealed that Black Amber pollen had the highest germinability and that Black Amber and Miili pollen tubes could elongate most rapidly. Such high abilities of pollen germinability and tube growth were also detected in pollen tube penetration into pistils. In the pistil pollinated with Hongxinli and Miili pollen, lager numbers of pollen tubes were counted in the style base, upper ovary, and locule top than in those pollinated by Black Amber pollen when examined 8 days after pollination. However, the number of pollen tubes decreased significantly at the obturator and micropyle in pistils pollinated with Hongxinli and Miili pollen, while in pistils pollinated with Black Amber pollen the decrease was smaller. This fact might indicate the higher compatibility of Black Amber pollen with Zuili pistils compared to other pollen sources. The highest fruit set percentage obtained from pollination using Black Amber pollen also proved the highest compatibility of the combination. The decrease in the pollen tube numbers in each part of the pistils, when examined 14 days after pollination, may be caused by digestion of the pollen tube membrane by surrounding pistil tissues.
Open- and self-pollination resulted in only a few pollen tubes reaching the micropyle. Some pollen tubes swelled, bent back, and grew in the opposite direction on the stigma, leading to the slowest and lowest number of pollen tubes penetrating into ovules. From these results it is clear that Zuili stigmas cannot receive the sufficient numbers of potent pollen grains by open-pollination and Zuili plum has incomplete self-incompatibility. To improve the successful cross-pollination by potent pollen grains, interplanting with effective pollinizer plants, such as Black Amber, must be indispensable. Studies of Kaufmane and Rumpunen (2002) on Japanese quince showed that pollen tubes reached the base of the ovary in 2–6 days in a case of a compatible combination, but not until 6–7 days after in an incompatible combination. Yamashita et al. (1990) found that in pears, compatible pollen tubes reached the ovary 3 days after pollination, but incompatible pollen tubes were completely arrested at the stigma.
The importance of normal ovule development for high fruit set in stone fruits has been reported. Alburquerque et al. (2004) reported that apricot cultivars producing good yields had developed mature ovules earlier than those with low productivity, and pollen tubes also grew faster in higher productive cultivars. An anatomical investigation of Zuili ovaries at the petal fall stage revealed that 24.3% of ovules in normal pistils (not double-type pistils) had developed normally. It is generally known that in stone fruit trees, one of the two ovules formed in a pistil usually degenerates before the blooming time (Thompsom and Liu, 1973; Pimienta and Polito, 1982). The percentage of normally developed ovules counted in this investigation indicated that 29.8% of total ovules degenerated entirely and 45.9% of them were malformed ovules, mainly without embryo sacs in normal-type pistils. The lack of differentiation of the embryo sac has been described as a malformation in apricot (Eaton and Jamont, 1964; Lichou et al. 1995) and other species such as avocado (Tomer et al. 1976), litchi (Stern et al. 1996) and peach (Fuss et al. 1990). Such imperfection of ovule development causing poor berry sets has been reported for tetraploid grapes by Okamoto et al. (1984). They investigated the ovule development anatomically and found that 50% to 80% of ovules are abnormal or immature even at the full bloom stage in cv. Kyoho and Pione, usually resulting in a very poor berry set. In our anatomical investigation for Zuili pistils, the notable percentages of imperfection of ovules and embryo sacs are thought to be a cause of the poor fruit setting. On the other hand, poor cultural conditions, as reflected by reduced vigor of apple trees and flowers, or reduced N levels, cause suppression of embryo sac development and reduction of ovule longevity, as well as slower pollen tube growth and reduced periods of stigma receptivity (Williams 1963). Poor fruit set has been attributed to a lack of adequate pollination or wet and cold weather conditions during bloom in apricot (Eaton 1959) and prune (Thompson and Liu 1973). It has been shown that embryo sac degeneration can account for the poor fruit set in sweet cherry (Bradbury 1929; Child 1966).
A particularly interesting finding in this work is the high percentage of double-type pistils in ‘Zuili’. Unfortunately, in the double pistils, the percentage of normally developed ovules was as low as 8.9%, which indicates the poor fruit set potential of such abnormal pistils. As noted previously, the high rate of double pistils, as 40% of total flowers, must be one of the major reasons for the poor set of Zuili plums in Jiaxin.
Development of double-type pistils in cherry trees has been also reported. Beppu et al (1999 2000, 2001) noted that the high temperature conditions during a summer season, corresponding to the time of sepal initiation, may cause the abnormality. We already investigated the time of pistil development in Zuili flower buds and determined it to be around early autumn (unpublished data). For preventing the double-pistil development in ‘Zuili’ trees, it might be effective to shade fruiting twigs during summer and early autumn.
This is the first study involving pollen tube growth and compatibility relationships of the Zuili genotype. The results have important implications for plantation establishment for this crop. The mechanism of poor crop in Zuili has been elucidated with two hypotheses. One is the inhibition of pollen tube growth at the base of stigma or in the style, and the other is the developmental arrest of embryo sac or endosperm. As a result of our experiment, pollen germination on stigma and pollen tube growth in the style was inhibited by self-pollination compared with cross-pollination. Thus, we assumed that self-incompatibility occurred before the pollen tube reached the ovary. Many malformations and degenerations of ovules such as those without embryo sacs and double pistil development might cause big problems for Zuili production.
Materials and Methods
The field work was carried out on mature trees of Zuili grown in a commercial orchard located at Jiaxin district in Zhejiang Province, which is a major Zuili producing area in China. All the trees were grown under the same environmental conditions, as well as with the same doses of irrigation, fertilization and phytosanitary treatments.
Self-pollination and cross-pollination of Zuili were carried out artificially by hand. For cross-pollination, pre-bloom flowers were collected from two Chinese cultivars, Hongxinli and Miili (P. salicina L.), and one American cultivar, Black Amber (P. domestica L.), and used as pollen sources. As shown in Table 5, the flowering period in these cultivars was similar so that pollen grains of each cultivar were available mutually. Open pollination was designed to be natural, in which no flowers were emasculated and bagged. Branches with 350–400 flowers, developed in a cluster containing 15 to 16 flower buds, were selected at the balloon stage, ready to open, and the flowers were emasculated using forceps after removing both late flower buds and open flowers. The clusters were bagged with paraffin paper bags to avoid pollination by bees. The bags were removed for artificial pollination in the morning, around 09.00 hours, of each full bloom stage. Immediately after pollination, clusters were re-covered with paper bags. Fruit set percentages correspond to the mean of ten branches (a total of 100 to 150 flowers pollinated) chosen randomly from two or more trees. The number of developing fruitlets was counted in each pollination treatment one month after pollination.
Pollen collection and germination tests
To study pollen germinability in vitro, anthers were sampled from above three cultivars grown at the same orchard as Zuili at the late balloon stage in 2005. Collected anthers were dried in an incubator at 20–22 ºC until dehiscence, which usually took about 24 h. The germination tests were then immediately carried out. The germination medium, consisting of 1% agar and 15% sucrose, was prepared by dissolving them in boiling water at pH 5.6. The mixture was poured into 90nm Petri dishes, and pollen was scattered onto the medium surface after being cooled and incubated for 6 h at 25 ºC. The percentage of pollen germination was determined under a light transmission microscope. Pollen grains with elongated tubes equal to or longer than the diameter of the pollen grain were recorded as germinated ones. Five microscopic fields, containing 200–300 pollen grains per field, were examined for counting. After counting of pollen germinability, each microscopic field was photographed and the tube length of germinated pollen (20–25 tubes per field) was measured using a software photo measure.
Pollen tube growth into the style and ovary tissues
For observation of the pollen tube growth into pistils, Zuili flowers at the balloon stage were emasculated. They were pollinated with each pollen source and then bagged. Twenty pistils for each treatment were sampled 0, 4, 8 and 14 DAP, and fixed with FAA solution (100% formalin: 100% acetic acid: 50% ethanol, in proportions of 5: 5: 90). For each treatment, ten styles were hydrated, softened, and stained with 0.1% aniline blue in 0.1 N-PO4K3, then covered with a cover glass and gently squashed. Pollen grains attached to the stigma were observed under a fluorescence microscope (Olympus BH2-RFL). For pollen tube counting in various parts in a pistil, dehydrated pistils were embedded into paraffin blocks and sliced using a rotary microtome into 18μm-thick cross sections. After staining with aniline blue solution, numbers of pollen tubes were counted under the fluorescence microscope.
Twenty fruitlets were collected at 4, 8, and 14 DAP and fixed with FAA solution. The ovaries were then dehydrated and embedded into paraffin blocks. They were sliced to 16μm-thick longitudinal sections using a rotary microtome. The sections were stained with Schiff reagent and alcian blue. In those ovaries, embryo and endosperm development were examined under a light microscope.
All analyses were performed with the SAS statistical software package, except Table 4 in which data were analyzed by 2 test.
Alburque NL Burgos, Sedgley M, Egea J (2004). Contributing to the knowledge of the fertilization process in four apricot cultivars. Scientia Hortic 102, 387–396.
Beppu K, Kataoka I (1999). High temperature rather than drought stress is responsible for the occurrence of double pistils in ‘Satohnishiki’ sweet cherry. Sci Hortic 81,125-134.
Beppu K, Kataoka I (2000). Artificial shading reduces the occurrence of double pistils in ‘Satohnishiki’ sweet cherry. Sci Hortic 83, 241–247.
Beppu K, Takayuki I, Kataoka I (2001). Effect of high temperature exposure time during flower bud formation on the occurrence of double pistils in ‘Satohnishiki’ sweet cherry. Sci Hortic 87, 77–84.
Bradbury, Dorothy (1929). A comparative study of the developing and aborting fruits of Prunus cerasus. Am J Bot 525–542.
Chen LR, Sheng DZ, Xu AK, Weng YL (1981). A historical investigation of the plance of origin of Zui-plum. J Zhejiang Agri Univ 7, 125–130.
Chen LR, Zhang ZM, Xu AK (1982). Study on the pollination of Zui-plum. Fruit of China 2, 11–13.
Child RD (1966). Pollen tube growth in relation to temperature and ovule longevity in the cider apple Michelin. Ann Rpt Agr Hort Res Sta Long Ashton Bristol 115–120.
Eaton GW (1959). A study of the megagametophyte in Prunus avium and its relation to fruit setting. Can J Plant Sci 39, 466–476.
Eaton GW, Jamont AM (1964). Embryo sac development in the apricot, Prunus armenica L. cv. ‘Constant’. J Am Soc Hortic Sci 86, 95–101.
Fuss AM, Burne PM, Coombe BG, Sedgley M (1990). Culrural manipulation for out-of season peach production under glass. Sci Hortic 43, 15–27.
Kaufmane E, Rumpunen K (2002). Pollination, pollen tube growth and fertilization in Chaenomeles Japonica (Japaneses quince). Sci Hortic 94, 257–271.
Lichou J, Audubert A, Jay M, Costes E (1995). Influence of floral fertility and pollination on fruit drop and productivity of apricot (Prunus armerniaca L.). Acta Hortic 384, 333–337.
Jaumien F (1968). The causes of poor bearing of pear trees of the variety ‘Doyenne du Comice’. Acta Agrobotanica 21, 110–114.
Okamoto Goro, Kyoko Yamamoto, Kazuo Shimamura (1984). A study on the difference in the occurrence of seedless berries among the tetraploid grapes related to Kyoho variety. J Japan Soc Hort Sci 53, 251–258.
Pimienta E, Polito VS (1982). Ovule abortion in ‘Nonpareil’ almond (Prunus dulcis [Mill.] D. A. Webb). Am J Bot 69, 913–920.
Stern RA, Eisenstein D, Voet H, Gazit S (1996). Anatomical structure of two day old litchi ovules in relation to fruit set and yield. J Hotric Sci 71, 661-671.
Thompson MM, Liu LJ (1973). Temperature, fruit set and embryo sac development in ‘Italian’ prune. J Am Soc Hortic Sci 55, 193–197.
Tomer E, Cottreich M, Gazit S (1976). Defective ovules in avocado cultivars. J Am Soc Hortic Sci 101, 620–623.
Williams RR (1963). The effect of nitrogen on the self-fruitfulness of certain varieties of cider apples. J Hortic Sci 38, 52–60.
Williams RR (1965). The effective pollination period for some apple and pear varieties. Rpt Long Ashton Res Sta 65, 136-138.
Williams RR (1970). Factors affecting pollination in fruit trees.In: Luckville LC and Cutting CV eds. Physiology of Tree Crops. Academic, London. pp. 193-207.
Yamashita K, Saita H, Hashimoto N (1990). Pollen-stigma interaction which might be critical to the gametophytic incompatibility of Japanese pear. J Jpn Soc Hort Sci 59, 83-89.
Table 4 Ovule development in normal and double Zuili pistils at the petal fall stage†
Type of pistil‡
Without embryo sac
Total no.of ovules
† P＜0.005, χ2=10.3711
‡ Refer to Figure 1
§ Type of ovules (distribution %).
Figure 1. Pollen germination at the stigma of Zuili pistils.
(A) Cross pollination with Black Amber pollen.
(B) Self pollination.
(C) Open pollination.
Pistils were sampled 1 week after pollination. A 100; B and C 200.
Figure 2. Pollen tubes growing in various parts of Zuili pistils 10 days after pollination with
Black Amber pollen.
(A) Basal style.
(B) Upper ovary.
(C) Locule top.
Arrows indicate pollen tubes. A, B and C 100; D and E 200.
Figure 3. Ovule development in Zuili pistils at the petal fall stage.
(A) Normally developed ovule with an embryo sac and egg cell;
(B) Undeveloped ovule without embryo sac (right). The left ovulehad is degenerated and shrunken;
(C) Double pistils with a normally developed ovary in each.
Pistils were sampled 1 week after bloom. A 200; B and C 100.
Figure 4. “Double-pistil” flowers (A, B) and “Twin” fruit (C) developed in Zuili plum.
Double pistil flowers have two styles and upper ovaries generate from a common basal ovary. Arrows indicate double-pistils.