Number of cells in tomato fruit depending on fruit position and source-sink balance during plant development

Transcription

1 Plant Growth Regulation 00: 1 8, Kluwer Academic Publishers. Printed in the Netherlands. 1 Number of cells in tomato fruit depending on fruit position and source-sink balance during plant development Nadia Bertin*, Hélène Gautier and Cédric Roche Unité Plantes et Systèmes de culture Horticoles, INRA, Site Agroparc, F-84914, Avignon, Cedex 9, France; *Author for correspondence ( nadia.bertin@avignon.inra.fr; phone: +(33)-(0) ; fax: +(33)-(0) ) Received 19 March 2001; accepted in revised form 30 May 2001 Key words: Cell division, Floral development, Lycopersicon esculentum Mill., Potential growth, Sink strength, Tomato Abstract Fruit sink strength or its ability to attract assimilates depends both on sink activity and size. This study investigated one main component of sink size, that is the number of fruit cells during tomato plant development. Plants were grown in a controlled climate chamber under a limiting (LS, six fruits per truss) and non-limiting (NLS, two fruits per truss and CO 2 enrichment) supply of carbon assimilates. Under NLS conditions, fruit cell number was homogeneous among successive trusses, and fruits contained on average more cells than under LS conditions, though differences were not significant on the first truss which underwent the lowest competition. Under LS conditions, an ontogenetic increase in cell number was observed in proximal fruits of the upper trusses attributed to the enlargement of the apical meristem during plant development. The decrease of cell number from proximal to distal fruits within a truss, that was expected from the literature, was generally observed in the LS experiment, with an average significant difference of about cells between the first and fifth fruits. Nevertheless, whereas the gradient in cell number from proximal to distal fruits was steep in the upper trusses, it was not significant on the lower trusses indicating that this gradient largely depended on the level of competition during floral development. Thus, under low assimilate supply, cell division is a main limiting factor for fruit growth, although cell enlargement during further fruit development is also affected, but was not measured in this work. Introduction Regulation of fruit size is a major issue in higher plant development, and an economical factor for many horticultural crops, including fresh tomato. Under non-limiting light flux, yield is mainly restricted by the number or the size of fruits, which is determined by the partitioning of assimilates in the plant (Ho 1996). The sink strength or its ability to attract assimilates depends both on sink activity and size. Whereas sink activity is determined by many processes such as phloem transport, metabolism and compartmentation, sink size is determined by the number of cells (Ho 1992, 1996). The concept of potential growth or sink strength has been analysed by many authors (e.g. De Koning (1994); Heuvelink (1996)), but its experimental assessment is rather complex (De Koning 1994; Bertin et al. 1998), probably because only sink activity is considered. Although cell size makes the greatest contribution to the final fruit size, in many species differences in fruit size have been attributed to differences in cell number rather than cell size (Jackson and Coombe 1966; Cowan et al. 1997; Higashi et al. 1999). In tomato, the cell number is also one main component of differences in fruit size, for instance among proximal and distal fruits within a tomato truss (Bohner and Bangerth 1988a) or among cultivars differing in fruit size (Bohner and Bangerth 1988b), but little is known about the regulation of cell number during plant de- XPS (GROW) product element ICPC/Grafikon

2 2 velopment. Tomato fruit development proceeds through three distinct phases, cell division, cell expansion and ripening. The relatively short period of cell division ends about two weeks after anthesis (Mapelli et al. 1978), and is followed bya6to8week period of cell expansion. Many studies report an early control of cell division by competition for assimilates. For instance, insufficient light during flower development completely stops mitotic activity and provokes inflorescence abortion, whereas under favourable light conditions only the last distal flower buds abort (Kinet 1989; Kinet et al. 1985); this suggests that both the competition between vegetative and reproductive sinks and among flower buds of the same inflorescence are involved. Indeed within a tomato truss, proximal fruits which are bigger than distal ones, already contain more cells at anthesis (Bangerth and Ho 1983) and are assumed to exert a dominance on the subsequent fruits via competition for assimilates or via hormonal signals (Bohner and Bangerth 1988a, 1988b; Bünger-Kibler and Bangerth 1983). In indeterminate tomato, one truss develops at every 3 5 internodes and 5 flowers are initiated and developed sequentially on the inflorescence, with a lapse of about 5 7 days between the first proximal and the last distal flowers. Since the few studies on tomato fruit cell number have been done on plants with few fruiting trusses (e.g. Bohner and Bangerth (1988a, 1988b)), hence considering only limited source/sink relationships, the regulation of fruit cell division related to changes in the source-sink ratio during plant development is unknown. The present work was performed to analyse the variations in fruit cell number during plant development and after many fruiting trusses developed. Two contrasting situations of carbon assimilate supply were created through CO 2 enrichment and truss pruning in order to assess the impact of the early control of potential fruit size by competition among the sinks. Materials and methods Plant material and cultural conditions Two experiments were carried out successively in a 8.75 m 2 (21 m 3 ) growth climate chamber under controlled environmental conditions. Tomato seeds, cv Raïssa, were sown in sand, and twelve plants were pricked out at a developmental stage of about 6 visible leaves, into 10 dm 3 -pots filled with a balanced oxygenated nutrient solution (detailed in Le Bot et al. (1998)), whose composition was checked every week and readjusted when necessary. Sowing took place in the growth climate chamber itself under climatic conditions similar to those monitored after planting. From sowing to the end of measurements, a day-night air temperature of 20 ± 0.5 C was maintained continuously. A photoperiod from 0800 h to 2000 h was applied with a photon flux of about 500 mol m 2 s 1 PAR above the canopy. Air relative humidity was around 70%. In the first experiment, plants were pruned to six flowers per inflorescence (or truss) when necessary, and no CO 2 enrichment was applied. In the second experiment, inflorescences were pruned to 2 flowers and air was enriched to 600 l CO 2 l 1 after anthesis of the first truss. Flowers were pollinated twice a week with an electrical bee, and all side shoots were removed as they appeared. Thereafter the first and second experiments will be referred to as LS (limiting supply) and NLS (non-limiting supply), respectively. Developmental observations Plant developmental stage was recorded twice a week on the 12 individual plants by counting the numbers of visible trusses and leaves longer than 1cm and assessing the developmental stage of each individual flower bud on all trusses from appearance to fruit set (about 5 mm diameter). Anthesis time was considered as full flower opening. As fruits developed, the diameter of the first, third and fifth fruit on each truss in the first experiment, and the diameter of all fruits in the second experiment were measured. Diameter growth curves of individual fruits were fitted to a three parameter Gompertz function by non linear regression. Determination of fruit cell number The number of pericarp cells was measured on the oldest five trusses in the first, third and fifth fruit in the first experiment, and in the two single proximal fruits in the second experiment. At fruit sampling time, the plant developmental stage was anthesis of the tenth and eighth truss in the LS and NLS experiments, respectively. Measurements were made on four plants from each treatment, according to a method adapted from that of Bünger-Kibler and Bangerth (1983). Fruits smaller than 2.5 cm diameter, were picked and fixed in a solution of ethanol, form-

3 3 aldehyde and acetic acid (90:5:5). After washing in water, the whole fruit pericarp was isolated under a binocular microscope and its volume was measured by water displacement. It was then put into a maceration solution of 3.5%pectinase + 0.1M EDTA + 0.4M mannitol at a ph of 4 for 2 3 days at 32 C. Similarly pericarps of fruits bigger than 2.5 cm were isolated after picking and their volume and fresh weight were measured. They were then macerated for at least 24h at ambient temperature in a solution of 0.05M EDTA + 0.4M mannitol at a ph of After maceration, samples were stirred for 15 minutes on a magnetic stirrer and diluted with an 87% solution of glycerol. Cells of an aliquot of the suspension were counted under a microscope using counting chambers (Fuchs-Rosenthal chamber 0.2 mm depth for big fruits, and Malassez chamber 0.1 or 0.2 mm depth for small fruits). A minimum of five samples per fruit were counted, and more in case of high variability (variation coefficient > 10%). The total number of cells in the whole fruit pericarp was calculated from volumes of the counting chamber, of the maceration solution and of the pericarp. Data analysis The effects of fruit position and growth conditions on the number of cells were analysed by two-way ANOVA (Jandel scientific Sigmastat) and F tests were used to determine the statistical significance. When significant effects were detected, a Tukey test was applied for all pairwise comparisons of mean responses. Other data are presented as means of four plants ± SE. Results Treatment effects on plant development and fruit growth Two experimental conditions were used to provide limiting (LS) or non limiting (NLS) carbon assimilate supply to fruits. Figure 1A illustrates this contrast via the ratio of fruit set to leaves, which is a good indicator of the sink/source ratio: under LS conditions this was more than twice that under NLS conditions. Under both conditions, the ratio increased by a factor of four from setting of the first truss to setting of the fifth one, whereas the plant development rate was hardly affected. Indeed the leaf appearance rate was similar in LS and NLS conditions (about 0.4 leaf day 1 ). The difference in assimilate supply to fruits was visible in the growth curves of the first proximal fruits under both conditions (Figure 1B): for instance on the first and third trusses, fruit growth was already more rapid under NLS than LS conditions one week after anthesis, which led to a 20 mm difference in fruit diameter after 35 days. Under LS conditions the mean fruit growth on the third truss was somewhat overestimated since some fruit abortions occurred, but this was not taken into account. The average number of fruit set on each truss of the LS experiment is given in Table 1. It decreased on average from 5.7 to 4 on the six first trusses, because of flower or fruit abortion usually at distal positions, except on the third truss on which only proximal flowers aborted. In the NLS experiment the two proximal fruit set on each truss. Effects of assimilate competition on fruit cell number The number of cells in the first proximal fruits were compared for the fifth oldest trusses of each experiment (Figure 2). Under NLS conditions, no significant difference was observed between the two proximal fruits and data were pooled. Under LS conditions numerous abortions were observed, which imply the cessation of cell division during flower development. Thus the question arose, whether these organs containing few cells (about 1 to cells from measurements on a few of them) should be considered to calculate the average number of cells at different fruit positions? To avoid a bias aborted organs were excluded. Under NLS conditions the number of cells was rather uniform, with a slight, but not significant (P = 0.115) decrease on the fourth and fifth trusses. Under LS conditions, the number of cells was significantly dependent on the truss position: the first fruit contained more cells on the fifth truss than on the first three trusses, and more cells on the fourth truss compared with the third. When cells were counted, the fifth truss was about 30 days old in the LS experiment, but only 15 days old in the NLS experiment. Since cell divisions might not be finished in the second case, a two way ANOVA was performed on the first four trusses to analyse the combined effect of growth conditions and truss position: on average fruits contained about more cells under NLS than LS conditions (P = 0.04), but interaction between the two factors was significant

4 4 position along the stem (Figure 3). On the fifth truss, the distal fruit aborted on all plants. A significant decrease of cell number from proximal to distal fruits was generally observed, with average differences of about cells between the first and fifth fruits (P = 0.014), and cells between the first and third fruits (P = 0.014). Nevertheless this pattern was not uniform among trusses, as no significant effect of fruit position was indicated on the first, second and third trusses, and only the first and fifth fruits were significantly different on the fourth (P < 0.01). Relation between fruit size and cell number at different positions on the plant Figure 1. (A) Change with time in the ratio of fruit set (diameter 5 mm) to leaves (longer than 1 cm) under LS (closed symbols) and NLS (open symbols) carbon assimilate supply. Small arrows indicate the fruit setting dates of the first, third and fifth trusses for each experiment. (B) Increase in diameter of proximal fruits on the first (circle, full lines) and third (diamonds, broken lines) trusses under LS (closed symbols) and NLS (open symbols) conditions. Data are means of 12 plants and vertical bars represent standard errors. (P = 0.01), due to the lack of a significant difference between NLS and LS conditions on the first truss, and an inverse relation on the fourth. Variations in fruit cell number according to truss and fruit position under LS conditions Fruit cell number is known to depend on fruit position within the truss, but it is unclear if this scheme may change under limiting assimilate supply and during plant development. To answer this question, cell number in the first, third and fifth fruits grown under LS conditions were analysed according to the truss To compare the size and cell number of fruits sampled at different age, the average growth curve was estimated by non linear regression made on all collected data (first, third and fifth fruits in the LS experiment and two proximal fruits in the NLS experiment). A three parameter Gompertz function was adjusted with determination coefficient of 0.94 and 0.95 for the LS and NLS experiments, respectively (Figure 4). Individual proximal (first) and distal (fifth) fruits identified by their age and size at sampling time were reported on the curve. Generally a good agreement was observed for different fruit positions on the plant, between the number of cells (Figures 2 and 3) and the fruit size relative to the average growth curve (Figure 4). In the LS experiment proximal fruits were generally above the average curve except for the third truss. In contrast distal fruits were on the average curve or below it especially for the second and fourth trusses. Within each truss, comparison of proximal and distal fruits indicated that a large difference in cell number was accompanied by a difference in size early in fruit development: this is clearly the case on the fourth truss (Figures 3 and 4). In the NLS experiment fruits grew more and faster than in the LS experiment. They were uniformly distributed on the average curve. Even fruits from the fourth and fifth trusses matched the average curve despite their lower cell number (Figure 2). It might be that cell division was not completed at sampling time, or that low competition among sinks allowed to compensate for the low cell number.

5 Table 1. Number of fruit set per truss and position of aborted fruits within the truss under LS conditions. Truss 1 is the oldest truss on the stem. Means and 95% confidence interval (I95%) were calculated on 12 plants. Truss 1 Truss 2 Truss 3 Truss 4 Truss 5 Truss 6 Fruit set Mean I95% Position of aborted fruits distal distal proximal distal both both 5 Figure 2. Number of cells in the pericarp of the first two proximal fruits under NLS conditions (open bars) and of the first proximal fruit under LS conditions (closed bars) on the first five trusses of tomato plants (1 indicates the oldest truss). Data are means of eight (NLS) and four (LS) fruits and vertical bars represent standard errors. Discussion The aim of the present work was to analyse the variations in fruit cell number during plant development, and to assess the influence of assimilate competition among various sinks, since the number of cells exerts an early control on final tomato size (Bangerth and Ho 1983; Bohner and Bangerth 1988a, 1988b). Fruit size is also largely determined after cell division by the partitioning of assimilates in the plant (Ho 1996), which may be manipulated by temperature (Pearce et al. 1993; Walker and Ho 1977), shoot density or fruit number (Cockshull and Ho 1995; Heuvelink 1996). Nevertheless, final sizes of fruits grown under a nonlimiting supply of assimilate and low fruit number per plant are quite variable (De Koning 1994; Bertin et al. 1998), and part of this variability probably occurs early during cell division. Our results indicate large variations in the number of pericarp cells due to the Figure 3. Number of cells in the pericarp of the first (closed), third (striped) and fifth (open) fruits measured on the first five trusses of tomato plants grown under LS conditions. Data are means of four fruits and vertical bars represent standard errors. inflorescence position on the stem, to the fruit position within the inflorescence and to the level of competition for assimilates. In this work a direct method was used to count cells, and thus the number of repetitions (four plants) had to be reduced, but it was of the same order as those reported in similar studies (Bohner and Bangerth 1988a, 1988b; Bünger-Kibler and Bangerth 1983). Moreover various types and sizes of counting chambers were used that were suitable for counting cells in a great range of sizes. Variations in cell number among successive trusses on the stem have not been previously reported in the literature. Nevertheless, the increase in cell number observed on the fourth and fifth trusses under LS conditions (Figure 2) agrees with the ontogenetic increase in fruit potential weight on successive tomato trusses (De Koning 1994). In the NLS experiment, no significant variations in cell number were observed among trusses, but it may be that cell division was not totally completed in the fifth truss. In poplar and sunflower leaf size increases with plant development due to apex enlargement accompanied by a larger cell

6 6 Figure 4. Increase in fruit diameter during development in LS (upper) and NLS (lower) experiments. Average growth curves (lines) were fitted by a three parameter Gompertz function on all fruits in each experiment. Symbols identify individual fruits by their age and size at sampling time, for proximal and distal fruits on the first five trusses (first and fifth fruits in the LS experiment, and the two first fruits pooled in the NLS experiment). number initiating a primordium (reviewed by De Koning (1994)). In tomato the cell number of the internal layer in the shoot apical meristem affects the floral meristem size and the carpel number of tomato (Gyllaspy et al. 1993). Thus the ontogenetic increase in fruit potential weight on the successive trusses would possibly result from an increase in fruit cell number due to the enlargement of the apical meristem during plant development. In the present work a general significant decrease in cell number was observed from proximal to distal fruits in the LS experiment, which is in accordance with the literature. Differences in fruit size as affected by fruit position within the same truss are well known on unthinned trusses developed under limited assimilate supply (Bertin et al. 1998; Bohner and Bangerth 1988a; Ho 1980), and are attributed to larger number of cells in proximal fruits already at anthesis (Bangerth and Ho 1983). Nevertheless considering individual fruit trusses in the LS experiment showed that variations in cell number according to fruit position largely depended on truss position and that no significant effect of fruit position was observed on the first three trusses (Figure 3). In the fifth truss, abortion of distal flowers probably indicated the interruption of cell division under the high competition for assimilates during flower development (Kinet 1989; Kinet et al. 1985). Although hormones possibly play an important role in the regulation of cell divisions (Bohner and Bangerth 1988a, 1988b), the level of competition for assimilates during flower development seemed to be also involved in the regulation of cell number from proximal to distal fruits of the same inflorescence. In the present experiment, the first trusses developed under limited competition due to few growing sinks and thus the number of cells was rather independent of fruit position within the truss. In contrast the fourth and fifth trusses developed under relatively high competition due to the increasing number of growing sinks on the plant, and a clear negative gradient from proximal to distal fruits was observed. This is in accordance with a higher reduction of cell number in distal than in proximal fruits after increasing the competition for assimilates by plant defoliation (Bohner and Bangerth 1988a). The pollination sequence within the truss has been demonstrated to be the main cause of differences in the number of cells and final size of the successive fruits. Indeed synchronization of flower pollination reduced these differences among fruits within a truss (Bohner and Bangerth 1988a). This explains how differences in potential weight between proximal and distal fruits measured on trusses pruned to one or two flowers are largely reduced compared with intact trusses (Ho 1980), at least for long life and round tomato cultivars (Bertin et al. 1998; De Koning 1994). Comparison of the LS and NLS experiments indicates a consistent significant reduction of cell number by the competition for assimilates, depending on truss position. The absence of a competitive effect on the first truss may be explained by the low number of sinks when this truss developed under both conditions, but the inverse effect on the fourth truss remains unexplained. Reduction of tomato cell number in both proximal and distal fruits in response to plant defoliation has been reported in the literature (Bohner and Bangerth 1988a). It was also demonstrated that negative effects of low light intensity on inflorescence development and cell division activity may be coun-

7 7 teracted by low temperature or CO 2 enrichment (Kinet 1982). Thus, in the LS experiment, the low light flux and CO 2 concentration and the relatively high number of fruits per plant were responsible for the general reduction in cell number. As expected fruit growth was more enhanced in the NLS than in the LS experiment (Figure 4). In the LS experiment the positions of different fruits in relation to the average fruit growth curve were rather consistent with observations made on the number of cells (Figure 3). In the NLS experiment both fruit cell number and growth were uniform but only proximal fruits were considered. Thus under limiting assimilate supply, which affects cell division, cell number is a main limiting factor for fruit size, but not an absolute indicator since cell enlargement is also limited. For instance, in our experiment the proximal fruit on the fourth truss contained more cells but was smaller under the LS than the NLS experiment. Conclusion Fruit cell number depended on plant development, source-sink relationship, and fruit position, but the gradient in cell number and thus in potential weight from proximal to distal fruits within the same truss largely depended on the level of competition during floral development. Under low competition, cell number hardly differed among fruits within the same truss and thus their initial fruit sink strength after cell division should not differ. When competition increased, cell number declined more in distal than in proximal fruits; this would explain their lower sink strength during further development, and then fruit size differences are expected to increase until ripening since proximal fruits grow faster and then exert an increasing competition against distal fruits. Acknowledgements This work was realised with the fruitful assistance of B. Brunel and JC. L Hôtel. References Bangerth F. and Ho L.C Fruit position and fruit set sequence in a truss as factors determining final size of tomato fruits. Annals. of Botany 53: Bertin N., Gary C., Tchamitchian M. and Vaissiere B.E Influence of cultivar, fruit position and seed content on tomato fruit weight during a crop cycle under low and high competition for assimilates. Journal of Horticultural Science and Biotechnology 73: Bohner J. and Bangerth F. 1988a. Effects of fruit set sequence and defoliation on cell number, cell size and hormone levels of tomato fruits (Lycopersicon esculentum Mill. within a truss. Plant Growth Regulation 7: Bohner J. and Bangerth F. 1988b. Cell number, cell size and hormone levels in semi-isogenic mutants of Lycopersicon pimpinellifolium differing in fruit size. Physiologia Plantarum 72: Bünger-Kibler S. and Bangerth F Relationship between cell number, cell size and fruit size of seeded fruits of tomato (Lycopersicon esculentum Mill., and those induced parthenocarpically by the application of plant growth regulators. Plant Growth Regulation 1: Cowan A.K., Moore-Gordon C.S., Bertling I. and Wolstenholme B.N Metabolic control of avocado fruit growth. Plant Physiology 114: Cockshull K. and Ho L.C Regulation of tomato fruit size by plant density and truss thinning. Journal of Horticultural Science 70: De Koning A.N.M Development and dry matter distribution in glasshouse tomato: a quantitative approach. PhD Dissertation, Wageningen Agricultural University, Wageningen. Gyllaspy G., Ben-David H. and Gruissem W Fruits: a developmental perspective. The Plant Cell 5: Heuvelink E Dry matter partitioning in tomato: validation of a dynamic simulation model. Annals of Botany 77: Higashi K., Hosoya K. and Ezura H Histological analysis of fruit development between two melon (Cucumis melo L. reticulatus) genotypes setting a different size of fruit. Journal of Experimental Botany 50: Ho L.C Control of import into tomato fruits. Berichte der Deutschen Botanischen Gesellschaft 93: Ho L.C Fruit growth and sink strength. In: Marshall C. and Grace J. (eds), Fruit and seed production: aspects of development, environmental physiology and ecology. SEB Seminar Series 47, Cambridge, pp Ho L.C Tomato. In: Zamki E. and Shaffer A.A. (eds), Photoassimilate distribution in plant and crops. Marcel Dekker, Inc, New York, pp Jackson D.I. and Coombe B.G The growth of apricot fruit. I. Morphological changes during development and the effects of various tree factors. Australian Journal of Agricultural Research 17: Kinet J.M Un abaissement de la température et une élévation de la teneur en CO 2 de l atmosphère réduisent l avortement des inflorescences de tomates cultivées en conditions d éclairement hivernal. Revue Agricole 35:

8 8 Kinet J.M Environmental and chemical controls of flower development. In: Lord E. and Bernier G. (eds), Plant reproduction: from floral induction to pollination. Vol. I. The American Society of Plant Physiologist Symposium Series, pp Kinet J.M., Zune V., Linotte C., Jacqmard A. and Bernier G Resumption of cellular activity induced by cytokinin and gibberellin treatments in tomato flowers targeted for abortion in unfavorable light conditions. Physiologia Plantarum 64: Le Bot J., Adamowicz S., Robin P., Andriolo J.L. and Gary C Modelling nitrate uptake by greenhouse tomato crops at the short and long time scales. Acta Horticulturae 456: Mapelli S., Frova C., Torti G. and Soressi G.P Relationship between set, development and activities of growth regulators in tomato fruits. Plant and Cell Physiology 19: Pearce B.D., Grange R.I. and Hardwick K The growth of young tomato fruit. I. Effects of temperature and irradiance on fruit growth in controlled environment. Journal of Horticultural Science 68: Walker A.J. and Ho L.C Carbon translocation in the tomato: carbon import and fruit growth. Annals of Botany 41: