Introduction
Generally, the chromosome set of each somatic cell consists of two sets of chromosomes, and this normal condition is defined as diploid (indicated with 2n), while with polyploidy (or genome duplication) the number of chromosomes within the nucleus of a cell is greater than 2n (Dar and Rehman, Reference Dar and Rehman2017). Polyploidy is known to play an important role in plant evolution and ecology (Hegarty and Hiscock, Reference Hegarty and Hiscock2008; Soltis et al., Reference Soltis, Visger and Soltis2014; Ruiz et al., Reference Ruiz, Oustric, Santini and Morillon2020), and it has been defined as a relevant phenomenon also for some genera belonging to the Oleaceae family (Taylor, Reference Taylor1945). Polyploidization is a process that can occur naturally, or it is induced (Islam et al., Reference Islam, Deepo, Nasif, Siddique, Hassan, Siddique and Paul2022; Liu et al., Reference Liu, Gao, Jin, Wang, Jia, Ma, Zhang, Zhang, Qi and Xu2022). In plants, polyploidy is a pathway that allows to obtain new cultivars (Abdolinejad et al., Reference Abdolinejad, Shekafandeh and Jowkar2021). Artificial polyploidization is regularly used in plant breeding programmes, as genome doubling often results in improved functionality of some tissues and desired horticultural properties (De Baerdemaeker et al., Reference De Baerdemaeker, Hias, Van den Bulcke, Keulemans and Steppe2018).
At the phenotypic level, with the polyploidy important modifications of particular scientific interest are manifested. Variations in ploidy levels can induce changes in plant cells that affect morphological traits, anatomical structure, physiological responses and chemical composition (Doyle and Coate, Reference Doyle and Coate2019; Ruiz et al., Reference Ruiz, Oustric, Santini and Morillon2020; Trojak-Goluch et al., Reference Trojak-Goluch, Kawka-Lipińska, Wielgusz and Praczyk2021). The most frequently observed morphological and anatomical changes due to ploidy variation in plants are: size of stem, leaf, flower, stomata, stomatal density, leaf length/width ratio and shape, plant height, etc. (Trojak-Goluch et al., Reference Trojak-Goluch, Kawka-Lipińska, Wielgusz and Praczyk2021). Some studies report that polyploid plants are more resistant to environmental stresses (Ruiz et al., Reference Ruiz, Oustric, Santini and Morillon2020; Islam et al., Reference Islam, Deepo, Nasif, Siddique, Hassan, Siddique and Paul2022) or to biotic stresses as found in the trifoliate orange (Wei et al., Reference Wei, Wang and Liu2020), in apple (Podwyszyńska et al., Reference Podwyszyńska, Markiewicz, Broniarek-Niemiec, Matysiak and Marasek-Ciołakowska2021), and also in olive, with some tetraploid genotypes having greater resistance to the olive leaf spot, due to the fungal pathogen Venturia oleaginea (Pannelli et al., Reference Pannelli, Famiani and Rugini1992; Rugini et al., Reference Rugini, Pannelli, Ceccarelli and Muganu1996).
The cultivated olive (Olea europaea subsp. europaea) belongs to a genus that includes diploid species, with a chromosome number 2n = 46 (Rugini et al., Reference Rugini, Pannelli, Ceccarelli and Muganu1996; Besnard et al., Reference Besnard, Garcia-Verdugo, Rubio De Casas, Treier, Galland and Vargas2008). However, cases of tetraploidy (subsp. cerasiformis) and hexaploidy (subsp. maroccana) have been found in some subspecies (Besnard et al., Reference Besnard, Garcia-Verdugo, Rubio De Casas, Treier, Galland and Vargas2008). Polyploids are easily achieved with the use of chemical products (colchicine, oryzalin, etc.) or with physical means such as ionizing radiation (Rugini and Gutiérrez-Pesce, Reference Rugini and Gutiérrez-Pesce2006). Polyploidy in olive has been studied with the aim to obtain compact genotypes suitable for high density systems, or to induce resistance to biotic and abiotic stress, and at the same time characterized by a high yield potential (Pannelli et al., Reference Pannelli, Famiani and Rugini1992). By irradiation of self-rooted olive trees of the cultivars Frantoio and Leccino, two mutants with a compact habit were obtained, distinguished by the abbreviations FC and LC, respectively for Frantoio Compact and Leccino Compact (Pannelli et al., Reference Pannelli, Famiani, Rugini, Bignami and Natali1990). These mutated olive trees have several morphological and physiological differences compared to the mother trees, such as shoots with larger diameters and shorter internodes, larger and thicker leaves, higher CO2 assimilation rate and tolerance to drought (Pannelli et al., Reference Pannelli, Famiani, Rugini, Bignami and Natali1990; Rugini et al., Reference Rugini, Cristofori and Silvestri2016a). These tetraploid trees also differ anatomically, with larger cell size in leaf, stem and root tissues (Pannelli et al., Reference Pannelli, Famiani and Rugini1992). Furthermore, tetraploidy increased the size of flowers, ovaries and ovary cells, but not fruit size (Caporali et al., Reference Caporali, Hammami, Moreno-Alías, Rapoport, Chiancone, Germanà and Rosati2014).
LC has been considered a low-vigour genotype (therefore characterized by slow growth and relatively small canopy), that could be used in crop intensification or as a rootstock (Pannelli et al., Reference Pannelli, Famiani and Rugini1992; Rugini et al., Reference Rugini, Silvestri, Mousavi, Baldoni and Mariotti2020). So far, however, the mechanisms that explain the reduced size of LC have not been identified. This is important, because if the slow growth and compact habit depends on characteristics of the aerial part of the tree, using it as a rootstock would not reduce canopy growth and size in grafted trees (Paoletti et al., Reference Paoletti, Cinosi, Lodolini, Famiani and Rosati2023). The ideal genotype for intensive systems is one that grows slowly only as a consequence of early fruiting and thus greater biomass partitioning into fruit and lower partitioning into vegetative growth (Rosati et al., Reference Rosati, Paoletti, Pannelli and Famiani2017, Reference Rosati, Paoletti, Al Hariri, Morelli and Famiani2018a, Reference Rosati, Paoletti, Al Hariri, Morelli and Famiani2018b; Paoletti et al., Reference Paoletti, Rosati and Famiani2021). If a genotype grows slowly for other reasons, the slow growth may delay fruiting as well, making the genotype unsuitable for intensive systems. The tetraploid Leccino has had little success as a cultivar, because it associates the slower growth with reduced and delayed yield, compared to the diploid Leccino. We hypothesize that its slower growth and reduced size are related to changes in assimilate partitioning among vegetative components (i.e. leaves and stems), due to the tetraploidy, making this genotype not ideal for intensive systems, or indeed for any system. At present, there is some information on the morphological characteristics of the 1-year-old shoots in LC, but the data are still scarce and incomplete. However, there is evidence that its leaves and stems are thicker (Pannelli et al., Reference Pannelli, Famiani, Rugini, Bignami and Natali1990, Reference Pannelli, Famiani and Rugini1992). This might imply a reduced leaf area per unit of whole-shoot biomass. This could entail slower growth compared to the diploid form due to reduced light interception and photosynthesis per unit of shoot biomass. Therefore, the aim of this work was to have a complete evaluation of how tetraploidy affects the morphology of the shoots in tetraploid Leccino, compared to its diploid form, and investigate whether tetraploidy brings about differences in partitioning, especially in leaf area per unit of shoot biomass, which could explain the slower growth of the tetraploid.
Materials and methods
Experimental description
Two olive genotypes were selected and compared in this study: Leccino (L) the diploid control, and Leccino Compact (LC) a tetraploid mutant of L considered as a slow growing genotype. Three adult trees of similar size and age (about 30 years) of each olive genotype were chosen. The trees of this two genotypes were grown under the same management and were cultivated in the same orchard, located in the Umbria region, in central Italy (Lat. 42°46′22″ N, Long. 12°51′26″ E, Alt. 450 m a.s.l.). Trees were trained to a vase and they were spaced 10 by 7 m. Trees were cultivated according to traditional local standards, and no irrigation was applied. No chemical treatments against diseases were applied, but no visible signs of diseases were apparent at the time of sampling.
Plant material and data collection
On 30 January 2023, five 1-year-old shoots from each of three different olive trees per each genotype were collected (15 shoots per genotype). The shoots were collected from each tree, around the whole periphery of the canopy. Shoots were chosen of different length in order to represent the whole spectrum of shoot length present in the canopy. Length, number of nodes and number of leaves were evaluated for each shoot. For each shoot, two perpendicular diameters were measured with a digital calliper at the base of the stem. Stem diameter was the mean of these two measurements. Average internode length was calculated by dividing shoot length by the number of nodes per shoot. For each shoot, a scanned image of all leaves was taken to determine the total leaf area. The total leaf area was determined from the pixel-area calculation through pixel values using Photopea, an open-source programme for image processing. The individual leaf area was calculated by dividing the total leaf area by the number of leaves per shoot. For each shoot, the length (excluding petiole) and the maximal width of each leaf were measured. These two measurements were used to calculate the length-to-width leaf ratio. From three leaves selected from the basal, median and apical portion of each shoot, three disks of known area were cut, using a paper puncher, avoiding the midrib of the leaf. Immediately after cutting, leaf thickness was measured on each disk, using a digital calliper. Disks were dried in a ventilated oven at 35°C until constant weight, then weighed. Specific leaf area (SLA) was calculated by dividing the disc area by its dry weight after drying. The fresh stem volume of each shoot was measured by immersing the individual stems in a graduated cylinder. Stems were then dried with the same procedure as for the leaf disks. Stem dry mass was then determined, and the density of the dry stem was calculated from the dry mass and the fresh stem volume. The same method was used to determine leaf density, using only the three disks, instead of the whole leaf, to avoid including the midrib and the petiole, which would bias the measurements of the leaf lamina density and create noise in the data. The remaining leaves and leaf portions were dried as described for disks and stems. Leaf mass ratio (LMR) was calculated for each shoot, as the ratio of the leaf dry weight to total shoot dry weight (leaf mass + stem mass). Leaf area ratio (LAR) was calculated for each shoot, as the ratio of the total shoot leaf area to total shoot dry weight (leaf mass + stem mass). Stem-to-leaf dry mass ratio was obtained for each shoot, as the ratio of the stem dry weight to leaf dry weight.
Statistical analysis
The effects of the ploidy level on each parameter were statistically analysed by a one-way ANOVA or by covariance analysis (ANCOVA) in cases of covariation with stem length of the analysed parameter (e.g. diameter, dry mass, volume and node length). When the genotype effects were significant post hoc tests were performed and averages were compared using the Tukey HSD test (P < 0.05, P < 0.01 and P < 0.001). Relationships between parameters were evaluated by calculating the coefficients of determination (R 2) and the statistical significance of the fits.
Results
Effect of tetraploidy on stem parameters
Stem diameter increased linearly with its length and for both genotypes stem length ranged from about 7 to 32 cm (Fig. 1a). LC had significantly larger diameters than L (+46% on average) (Table 1).
Relationship between stem diameter (a), stem dry mass (b), stem volume (c), and stem length for 1-year-old shoots of the two olive genotypes Leccino (L, diploid) and Leccino Compact (LC, tetraploid). Each point represents a single measured value. The genotype had a statistically significant effect as reported in Table 1.
Comparison of the morphological characteristics between leccino (diploid control) and its tetraploid LC
Data are averages ± S.E. Of 15 shoots, of varying length, per genotype. Within each row, statistically significant differences are indicated as: n.s. not significant;
** P < 0.01; ***P < 0.001. (1) denotes that the significance was determined by ANOVA, (2) denotes that the significance was determined by ANCOVA with stem length as the covariate.
Internode length also increased linearly with stem length (data not shown), but no significant difference was found between the two genotypes (Table 1). Stem dry mass increased exponentially with length, in both genotypes (Fig. 1b). At any stem length, LC had a significantly greater (about double) stem dry mass than L (Fig. 1b and Table 1). Stem volume also increased exponentially with length (Fig. 1c). Also in this case, LC had a volume significantly greater than L for the same length, about double on average (Table 1). There were no differences in stem density between the two genotypes (Table 1). Stem volume was linearly and positively correlated with stem dry mass with nearly identical regressions between genotypes (Fig. 2).
Relationship between stem dry mass and stem volume for 1-year-old shoots of the two olive genotypes Leccino (L, diploid) and Leccino Compact (LC, tetraploid). Each point represents a single measured value.
Effect of tetraploidy on leaf parameters
Total leaf area (i.e. all leaves of the shoot) was positively correlated with stem length (Fig. 3a). Total leaf area in LC was significantly higher than in L (about 44% on average, Table 1). Total leaf dry mass also increased with stem length, and was significantly higher by almost 66% in LC (Fig. 3b and Table 1). The same was found for total leaf volume, which was nearly 76% higher in LC (Fig. 3c and Table 1).
Relationship between total leaf area (i.e. all leaves of the shoot) (a), total leaf dry mass (b), total leaf volume (c), and stem length for 1-year-old shoots of the two olive genotypes Leccino (L, diploid) and Leccino Compact (LC, tetraploid). Each point represents a single measured value. The genotype had a statistically significant effect as reported in Table 1.
LC had a significantly lower (about 7%) leaf density than L (Table 1). Total leaf volume was linearly and positively correlated with total leaf dry mass (Fig. 4).
Relationship between total leaf dry mass and total leaf volume for 1-year-old shoots of the two olive genotypes Leccino (L, diploid) and Leccino Compact (LC, tetraploid). Each point represents a single measured value.
Leaves in LC were significantly longer (about 10%) and wider (about 34%) than in L (Table 1). The length-to-width leaf ratio was significantly lower in LC than in L by about 17% (Table 1). Furthermore, in LC the leaves were significantly thicker (+ 30%) than in L. LC had darker green leaf colour than L (Fig. 5a).
(a) Leaves of Leccino Compact (LC, tetraploid) (top) and Leccino (L, diploid) (bottom). LC leaves are evidently darker and of a different green colour than leaves of L. (b) 1-year-old shoot of L (left), and LC (right). The greater thickness of LC stem, at equal length, should be noted. The thicker stem in LC implies greater stem-to-leaf biomass ratio. This, added to thicker leaves, implies lower leaf area per unit of shoot (stem + leaves) biomass ratio.
LC had significantly greater individual leaf area, individual leaf dry mass and individual leaf volume (+43%, +66% and +73% respectively, Table 1).
SLA was significantly lower (by about 17%) in LC than in L (Table 1).
Effect of tetraploidy on the wood-to-leaf biomass ratio
The leaf area ratio (LAR) decreased linearly with increasing stem length (Fig. 6a). LAR was significantly lower in LC, approximately 18% less than in L (Table 1). Leaf mass ratio (LMR) also decreased with increasing stem length (Fig. 6b).
Relationship between leaf area ratio (a), leaf mass ratio (b), stem-to-leaf dry mass (c), and stem length for 1-year-old shoots of the two olive genotypes Leccino (L, diploid) and Leccino Compact (LC, tetraploid). Each point represents a single measured value. The genotype had a statistically significant effect as reported in Table 1.
LMR was significantly lower in LC, although only by about 4% (Table 1). Stem-to-leaf dry mass increased with increasing stem length, was significantly higher in LC (by about 20% Table 1). The different thickness of the stem and the different level of stem-to-leaf ratio can also be appreciated in Fig. 5b.
Discussion
Although some data on LC already existed in the literature, this study provides a much more complete and detailed picture of the effects of tetraploidy on leaf and stem morphology. We analysed for the first time a set of 1-year-old shoots with a wide range of length, thus providing information on trait variability with shoot length. Furthermore, we considered a wide series of traits and their ratios, (such as the LMR, the LAR and the stem-to-leaf dry mass), which had been never considered before for the characterization of polyploid genotypes.
Pannelli et al. (Reference Pannelli, Famiani, Rugini, Bignami and Natali1990) found that, for the same length of the 1-year-old shoot, LC had a larger diameter (about 38%) than L, in agreement with our results (Table 1). Also, in the common fig, a greater (+48 to 59%) basal diameter of the shoots has been observed for tetraploid plants compared to diploid ones (Abdolinejad et al., Reference Abdolinejad, Shekafandeh and Jowkar2021). An in vitro experiment with pear also indicated larger diameters for tetraploid shoots, compared to diploid ones (Sun et al., Reference Sun, Sun, Li and Bell2009). Thicker stems for the same length, with tetraploidy, was also found in other plants such as Plumbago auriculata (Jiang et al., Reference Jiang, Liu, Hu, He, Liu, Chen, Lei, Li, Yang, Li, Hu, Li and Gao2020) and in two Citrus species (Jokari et al., Reference Jokari, Shekafandeh and Jowkar2022).
In the present study, LC had slightly shorter (though not significantly) internodes (i.e. both the node number and shoot length were similar, Table 1), while Pannelli et al. (Reference Pannelli, Famiani, Rugini, Bignami and Natali1990) found significantly shorter internodes in LC than in L.
Although it has been previously stated that LC has larger leaves than L (Pannelli et al., Reference Pannelli, Famiani and Rugini1992; Rugini and Gutiérrez-Pesce, Reference Rugini and Gutiérrez-Pesce2006; Rugini et al., Reference Rugini, Silvestri, Mousavi, Baldoni and Mariotti2020), no data were available. The present data confirm this statement and further show that LC leaves are both larger and longer (Table 1). Larger leaves with tetraploidy have been reported also in several other species such as in Ligustrum japonicum (also in the Oleaceae family) (Fetouh et al., Reference Fetouh, Kareem, Knox, Wilson and Deng2016), in clones of Pyrus communis (Sun et al., Reference Sun, Sun, Bell, Li, Zhou, Xin and Wei2015), in Salix viminalis (Dudits et al., Reference Dudits, Török, Cseri, Paul, Nagy, Nagy, Sass, Ferenc, Vankova, Dobrev, Vass and Ayaydin2016), Morus multicaulis (Xi-Ling et al., Reference Xi-Ling, Jin-Xing, Mao-De, Zhen-Gang, Xiao-Yun and Qi-You2011) and Ziziphus jujuba (Shi et al., Reference Shi, Liu, Liu, Wang and Xu2015; Wang et al., Reference Wang, Luo, Wang, Deng, Wei, Liu and Liu2019).
The leaf length-to-width ratio here found for L (3.71, Table 1) was consistent with what has been observed in a previous study with the same cultivar (Petruccelli et al., Reference Petruccelli, Beghè, Ganino, Bartolini, Ciaccheri, Bernardi and Durante2020). The Leccino leaf is defined as elliptical (Saqib et al., Reference Saqib, Khalid, Ahmad, Anjum and Hussain2019), and since the length-to-width leaf ratio for both genotypes (L and LC) was lower than 4, their leaves can be considered elliptical (Mikhail, Reference Mikhail2015; Touati et al., Reference Touati, Ayadi, Bouajila, Acila, Rahmani, Bouajila and Debouba2022). However, LC had a markedly more elliptical leaf than L, in fact the length-to-width leaf ratio was significantly lower (3.07, Table 1). This ratio decreased significantly in tetraploids compared to the corresponding diploids, also in London plane (Liu et al., Reference Liu, Li and Bao2007), black locust (Li et al., Reference Li, Zhang, Ren, Feng, Guo, Dong, Sun and Li2021), Chinese jujube (Shi et al., Reference Shi, Liu, Liu, Wang and Xu2015; Cui et al., Reference Cui, Hou, Li, Huang, Pang and Li2017), mulberry (Xi-Ling et al., Reference Xi-Ling, Jin-Xing, Mao-De, Zhen-Gang, Xiao-Yun and Qi-You2011) and apple (Podwyszyńska et al., Reference Podwyszyńska, Markiewicz, Broniarek-Niemiec, Matysiak and Marasek-Ciołakowska2021; Wójcik et al., Reference Wójcik, Marat, Marasek-Ciołakowska, Klamkowski, Buler, Podwyszyńska, Tomczyk, Wójcik, Treder and Filipczak2022).
Pannelli et al. (Reference Pannelli, Famiani, Rugini, Bignami and Natali1990) found that LC had a thicker (about 30%) leaves than the diploid. The mean values obtained from Pannelli et al. (Reference Pannelli, Famiani, Rugini, Bignami and Natali1990) were 0.517 mm for L and 0.670 for LC, therefore much higher than those found in this study (0.366 and 0.475 respectively for L and LC, Table 1), probably due to different sampling methodologies. In fact, in this study, thickness was measured on leaf disks, thus excluding the midrib. In Pannelli et al. (Reference Pannelli, Famiani and Rugini1992) on the other hand, the thickness was consistent with our observations for both genotypes. Increased leaf thickness with tetraploidy has been observed also in other species such as Ligustrum japonicum (Fetouh et al., Reference Fetouh, Kareem, Knox, Wilson and Deng2016), Platanus acerifolia (Liu et al., Reference Liu, Li and Bao2007), Robinia pseudoacacia (Li et al., Reference Li, Zhang, Ren, Feng, Guo, Dong, Sun and Li2021), Morus multicaulis (Xi-Ling et al., Reference Xi-Ling, Jin-Xing, Mao-De, Zhen-Gang, Xiao-Yun and Qi-You2011), Malus domestica (Hias et al., Reference Hias, Leus, Davey, Vanderzande, Van Huylenbroeck and Keulemans2017) and Pyrus communis (Sun et al., Reference Sun, Sun, Bell, Li, Zhou, Xin and Wei2015).
Another marked difference between the tetraploid leaves and the diploid ones was the colour, which was markedly darker green in LC (Fig. 5a). This characteristic had been also noted in other tetraploid plants such as London plane (Liu et al., Reference Liu, Li and Bao2007), black locust (Li et al., Reference Li, Zhang, Ren, Feng, Guo, Dong, Sun and Li2021), Rangpur lime (Allario et al., Reference Allario, Brumos, Colmenero-Flores, Tadeo, Froelicher, Talon, Navarro, Ollitrault and Morillon2011), Chinese jujube (Shi et al., Reference Shi, Liu, Liu, Wang and Xu2015; Cui et al., Reference Cui, Hou, Li, Huang, Pang and Li2017; Wang et al., Reference Wang, Luo, Wang, Deng, Wei, Liu and Liu2019), crape myrtle (Ye et al., Reference Ye, Tong, Shi, Yuan and Li2010), pear (Sun et al., Reference Sun, Sun, Li and Bell2009) and apple (Hias et al., Reference Hias, Leus, Davey, Vanderzande, Van Huylenbroeck and Keulemans2017). The more intense green colour was often accompanied by a higher chlorophyll content in the leaves of tetraploid plants (Li et al., Reference Li, Zhang, Ren, Feng, Guo, Dong, Sun and Li2021), suggesting that there is a close relationship between these two characteristics, as found in Chinese jujube (Wang et al., Reference Wang, Luo, Wang, Deng, Wei, Liu and Liu2019) and in apple (Hias et al., Reference Hias, Leus, Davey, Vanderzande, Van Huylenbroeck and Keulemans2017).
There is little information in the literature regarding SLA in olive cultivars and no data on Leccino. The value here found in L (50.5 cm2 g−1, Table 1) was slightly lower than previously observed in some Portuguese olive cultivars (ranging from 51.7 to 58.2 cm2 g−1) (Bacelar et al., Reference Bacelar, Santos, Moutinho-Pereira, Gonçalves, Ferreira and Correia Carlos2006). LC had significantly lower SLA than L (Table 1) due to the greater leaf thickness (Witkowski and Lamont, Reference Witkowski and Lamont1991). In the tetraploid leaves of Eucalyptus polybractea the specific leaf area (SLA) was also lower than in the diploid (Fernando et al., Reference Fernando, Goodger, Chew, Cohen and Woodrow2019), while in poplar clones Alía et al. (Reference Alía, Lüttschwager and Ewald2015) found no significant differences. Consistent with our result, Sun et al. (Reference Sun, Sun, Li and Bell2009) found that the leaves of tetraploid pear trees had a higher specific leaf mass (SLM, which is the inverse of SLA) than in diploid plants.
LC is considered a slow growing genotype with reduced tree size, and therefore it is hypothesized that it could be used as a rootstock to reduce vigour in grafted cultivars (Pannelli et al., Reference Pannelli, Famiani and Rugini1992; Rugini et al., Reference Rugini, Silvestri, Mousavi, Baldoni and Mariotti2020). However, there is no information why it grows more slowly. We hypothesized that tetraploidy might induce differences in partitioning between woody parts and leaves, reducing leaf area per unit of total biomass. This reduces light interception, canopy photosynthesis and thus resources for growth and/or fruiting (Rosati et al., Reference Rosati, Paoletti, Al Hariri and Famiani2018c). The present results support this hypothesis, at least at the shoot level. The leaf mass ratio was lower for LC (Fig. 6b). Combining this with thicker leaves and lower SLA (Table 1), the leaf area ratio was even more reduced (about 18%) in LC, compared to L (Fig. 6a and Table 1). This means that for a total amount of biomass invested in shoots, LC has 18% less leaf area. In a small tree with little self-shading, this can significantly reduce tree carbon assimilation and thus tree growth. This effect adds exponentially, year after year, leading to increasingly larger differences in tree size year after year (Norby et al., Reference Norby, Todd, Fults and Johnson2021). Thus, the observed shoot morphological differences here reported are sufficient alone to explain differences in tree growth rate.
However, the differences in partitioning here observed in the shoots are likely to extend to the rest of the woody structures of the tree. In fact, Umeda-Hara et al. (Reference Umeda-Hara, Iwakawa, Ohtani, Demura, Matsumoto, Kikuchi, Murata and Umeda2022) studying poplar, observed that tetraploidy promotes basal growth of the stem predominantly in the radial direction, while growth is retarded in the longitudinal direction. In olive, Pannelli et al. (Reference Pannelli, Famiani and Rugini1992) found that the ratio of whole plant dry matter to leaf area (‘Plant dw/Leaf area’) was significantly higher in LC than in L, implying a lower leaf area ratio at the whole-tree level. Additionally, graphically re-elaborating data from Rugini et al. (Reference Rugini, Silvestri, Ceccarelli, Muleo and Cristofori2016b), we found larger TCSA (trunk cross-sectional area) in tetraploid olive trees than in diploid ones for the same plant height (Fig. 7), implying greater biomass partitioning into woody structures compared to the diploid plant.
Relationship between plant height and trunk-cross-sectional area (TCSA) in four genotypes: Leccino (diploid and tetraploid) and Frantoio (diploid and tetraploid). Data graphically re-elaborating from Rugini et al. (Reference Rugini, Silvestri, Ceccarelli, Muleo and Cristofori2016b).
This would add to the same mechanism described at the shoot level, further slowing down tree growth. It appears, therefore, that the slower growth of tetraploids might be simply explained with increased partitioning into woody structures and thus lower leaf area ratios, without resorting to the alternative hypothesis of a water stress mechanism (Fernando et al., Reference Fernando, Goodger, Chew, Cohen and Woodrow2019).
Although a low vigour is desirable for high and very high-density olive groves, when growth is slowed down by greater proportional investments in woody structure and lower investments in leaf area, this is not advantageous, because slower growth due to lower resource availability also reduces and retards fruit production. The ideal tree grows quickly until reaching the desired canopy size, then slows down as a consequence of fruiting, i.e. partitioning more biomass into fruit, rather than partitioning more into wood (Famiani et al., Reference Famiani, Cinosi, Paoletti, Farinelli, Rosati and Lodolini2022; Paoletti et al., Reference Paoletti, Cinosi, Lodolini, Famiani and Rosati2023). Greater partitioning into wood slows down growth from the beginning, and then continues to reduce resource availability also later on, reducing yield. Therefore, tetraploid genotypes that invest proportionally more into woody structures might not be desirable genotypes. Nor the ability of slowing down growth would necessarily be transferred to the scion when using tetraploids as rootstock: the scion is likely to maintain its own partitioning habitus and grow normally.
While it is likely that tetraploidy will induce similar effects on other olive genotypes and/or in other environments, possible differences in the genotypic response to tetraploidy cannot be ruled out, nor possible genotype–environment interactions. Further experiments across different locations and genotypes are needed to investigate whether the present results can be generalized in olive.
Conclusions
This study allowed us to thoroughly characterize the effects of tetraploidy on the morphology of 1-year-old shoots. Tetraploidy induced significant changes in most of the observed parameters. As has been the case in other plants, all the changes found in this research due to tetraploidy level may be exploited to induce desirable agronomic characteristics. Generally, in addition to the morphological changes, polyploidy involves anatomical and physiological changes, and alterations of resistance to abiotic and biotic stresses, which in the case of LC, have yet to be thoroughly studied.
The results also suggest that LC (and tetraploid trees in general) grows slower than L because tetraploidy increases biomass partitioning into woody structures and reduces it in leaf mass and thus leaf area. For a total amount of biomass invested in shoots, LC has 18% less leaf area. At the whole tree level, the lower leaf area per unit of biomass is likely to be much further decreased by tetraploidy, given that not only the shoot stems are thicker, but also the branches and trunk. However, no data are available for the whole tree LAR. Future researches should explore this different partitioning of the biomass in woody structures at the whole tree level.
Author contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by A.P.. The first draft of the manuscript was written by A.P. and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Funding statement
This work was supported by the Italian Ministry of Education, Universities and Research (Decreet No. 761-02/04/2021, project BIODIVERSIFY), within the PRIMA 2019, section 2.
Competing interests
The authors declare that they have neither financial nor non-financial conflicts of interest relevant for the content of the article.
Open access
