- Top of page
- Materials and Methods
- Supporting Information
Branching is an integral factor in determining the architecture and productivity of a plant. The huge range of growth forms observed in plants is largely dependent on developmental decisions made at two major checkpoints: firstly, the formation of axillary meristems in the leaf axils; and, secondly, the outgrowth of these meristems. Axillary meristem outgrowth is controlled by multiple interacting environmental and genetic signals. Although the phenomenon of apical dominance is one point of control for meristem outgrowth (Sachs & Thimann, 1964; Cline, 1994), investigations of mutant plants from annual species exhibiting altered branching phenotypes have also revealed the presence of a root-derived branching signal. These mutants include the decreased apical dominance (dad) mutants of petunia (Petunia hybrida) (Napoli, 1996; Napoli & Ruehle, 1996), the ramosus (rms) mutants of pea (Pisum sativum) (Beveridge, 2000; Morris et al., 2001; Rameau et al., 2002), the more axillary growth (max) mutants of Arabidopsis (Arabidopsis thaliana) (Booker et al., 1999; Stirnberg et al., 2002; Sorefan et al., 2003) and the high-tillering dwarf (htd) and dwarf (d) mutants of rice (Oryza sativa) (Zou et al., 2005; Arite et al., 2007). Identification of the CAROTENOID CLEAVAGE DIOXYGENASE (CCD) genes CCD7 and CCD8 that are affected in a group of these branching mutants (Sorefan et al., 2003; Booker et al., 2004; Snowden et al., 2005; Johnson et al., 2006; Zou et al., 2006; Arite et al., 2007; Drummond et al., 2009) has demonstrated that there are conserved pathways acting in diverse plants involving a carotenoid-derived signal. DAD1, MAX4, RMS1 and D10 are orthologous genes encoding CCD8, while DAD3, MAX3, RMS5 and D17 are orthologous genes encoding CCD7. The CCD family includes proteins in animals and plants that act on carotenoid substrates. The first plant member to be identified was Vp14 from maize, which encodes a 9-cis-epoxycarotenoid dioxygenase (NCED) (Schwartz et al., 1997). A number of other CCD genes have been shown to have carotenoid cleavage activity (Schwartz et al., 2004; Auldridge et al., 2006; Huang et al., 2009).
Studies in pea and rice have revealed the involvement of strigolactones in the branch signal (Gomez-Roldan et al., 2008; Umehara et al., 2008). These investigations have shown that branching mutants lacking CCD7 or CCD8 have reduced concentrations of strigolactone and applications of synthetic strigolactones can restore the wild-type branching phenotype to the mutant. Strigolactones are a group of terpenoid lactones initially characterized as a plant-derived signal promoting germination of the parasitic plants Striga and Orobanche (Bouwmeester et al., 2007) and have also been shown to be involved in symbiotic interaction with arbuscular mycorrhizal fungi (Akiyama et al., 2005).
Although annual plants have been the focus for investigating the control of branch development, perennial plants, such as kiwifruit and fruit trees, have the potential for many additional points of regulation. The architectural framework of a perennial plant is dependent on growth in the previous year, any pruning of that growth, and the organogenesis of the new season’s meristems. Exogenous signals from the environment, including extremes in temperatures and day length, are integrated over months by the whole plant and influence the progress of axillary meristems through the stages of dormancy induction and release (Arora et al., 2003). To determine the role of the CCD pathway in the control of branch development in the more complex system of a perennial plant, we investigated branch development in kiwifruit (Actinidia chinensis).
Kiwifruit, a deciduous woody perennial, as grown in commercial orchards commonly consists of a single trunk with two horizontal branches (leaders) trained in opposite directions along a support. Shoots that have developed from the leaders are tied down in winter and the next season’s shoots arise from axillary buds formed during the previous season (Snowball, 1997; Walton et al., 1997). The architecture of the kiwifruit plant is important for obtaining good yields of fruit and allowing adequate light for flower initiation in the following season (Sale & Lyford, 1990). Currently this is managed through manual pruning. The development of shoots and the potential to produce flowers and fruit are influenced by low winter temperatures, which increase the proportion of buds that break dormancy (McPherson et al., 1994, 1995). In kiwifruit, both the shoot architecture and yield can also be influenced by the rootstock (Wang et al., 1994; Clearwater et al., 2004, 2006), suggesting that root-derived signals, such as strigolactones, may be important for controlling kiwifruit branching.
In this paper we show that kiwifruit contain CCD7 and CCD8 genes that are predominantly expressed in the roots. Furthermore, we report that reducing the expression of AcCCD8 alters branch development and delays leaf senescence. Therefore, we conclude that the control of branch development observed in model annuals is conserved in a woody perennial plant.
- Top of page
- Materials and Methods
- Supporting Information
Mutants in the CCD7 and CCD8 genes resulting in an increased branching phenotype have been identified in petunia, Arabidopsis, pea and rice, indicating that there is a conserved pathway that controls branch development in both monocots and eudicots. Recently, the reduction of CCD7 expression in tomato has been shown to increase branching (Vogel et al., 2010) and the tomato mutant Sl-ORT1, which is deficient in strigolactones, has been shown to have reduced CCD7 expression and increased branching (Koltai et al., 2010). Genome sequencing from evolutionary diverse species such as the moss P. patens, M. truncatula, black cottonwood P. trichocarpa, and grape V. vinifera has revealed the presence of predicted CCD7 and CCD8 genes. However, proof of function of these genes in the control of branch development is yet to be determined in these species.
We have isolated the CCD7 and CCD8 genes from A. chinensis, a deciduous woody perennial, and have demonstrated that a reduction in expression of AcCCD8 correlates with an increase in branch development. The AcCCD7 and AcCCD8 genes, under the control of the CaMV 35S promoter, were also sufficient to restore the branching of the corresponding Arabidopsis mutants to a wild-type phenotype. This provides functional evidence that the CCD pathway controlling branch development is conserved in a woody perennial even though kiwifruit exhibits quite a different plant form to previously studied species.
The AcCCD8 gene has been mapped to the linkage group 3 using a microsatellite marker (Fraser et al., 2009). In the future it would be worthwhile determining if architectural traits, such as branch number, branch length and bud burst, co-segregate with the marker for AcCCD8.
The expression of the AcCCD7 and AcCCD8 genes predominantly in roots of A. chinensis is consistent with the patterns of expression observed in petunia (Snowden et al., 2005; Drummond et al., 2009) and Arabidopsis (Sorefan et al., 2003; Booker et al., 2004). The significantly lower level of expression detected for AcCCD7 than AcCCD8 in kiwifruit roots is also similar to that found for PhCCD7 in petunia (Drummond et al., 2009). Reciprocal grafting experiments in petunia with the dad1 and dad3 mutants and wild-type plants have shown that both CCD7 and CCD8 are required in the roots for the generation of a signal controlling branching (Simons et al., 2007). Therefore, the co-localization of expression of these genes in kiwifruit roots is reasonable. In rice and pea, the expression of CCD7 and CCD8 is required for the production of strigolactones (Gomez-Roldan et al., 2008; Umehara et al., 2008), which have been shown to be involved in the inhibition of branch development. In tomato plants with reduced CCD7 expression, there was also a reduction in the concentration of strigolactones detected in root exudates (Koltai et al., 2010; Vogel et al., 2010). The isolation and identification of strigolactones have been carried out in a wide range of plants and have revealed a diverse range and mixtures of known and unknown strigolactones (Yoneyama et al., 2008). Strigolactones have been shown to be involved in a range of activities, such as interaction with parasitic plants and arbuscular mycorrhizal fungi, nutrient acquisition and the control of branch development. However, it is yet to be determined if the different strigolactones isolated vary in their specificity (or efficacy) in these different activities.
The expression of the AcCCD7 and AcCCD8 genes in the young fruit and seeds of wild-type A. chinensis suggests that these genes may have a role in fruit and seed development. The effect of reducing AcCCD8 expression on fruit development and flavour will be determined when the transgenic plants reach reproductive maturity.
The reduction of AcCCD8 expression in transgenic A. chinensis plants is correlated with an increase in the total number of branches produced by the plants over two growing seasons. The increase in the total number of branches reflected the large increase in branches < 5 cm in length. Increasing the number of short branches could increase the number of flowers produced on kiwifruit vines by increasing the proportion of nodes with the potential to flower (normally nodes one to nine) (Snowball & Considine, 1986). It will be interesting to determine if there is an increase in the number of flowers and fruit produced on the AcCCD8 RNAi plants when they reach flowering maturity. Different Actinidia rootstocks, such as Actinidia hemsleyana, have been shown to increase the percentage bud burst and the number of flowers produced per shoot on the scion (Wang et al., 1994). Low-vigour rootstocks (Actinidia polygama and Actinidia kolomikta) have also increased the proportion of short shoots on the scion and increased the fruit load relative to total leaf area compared with the scions on high-vigour rootstocks (Clearwater et al., 2004, 2006). The mechanism of this effect is not clear, but a recent investigation in cherry has revealed differential expression of genes involved in brassinosteroid signalling, flavonoid metabolism and cell wall biosynthesis between the scions grafted to dwarfing or semivigorous rootstock (Prassinos et al., 2009).
The delayed leaf senescence observed in RNAi plants with reduced levels of CCD8 expression is similar to that seen in the dad1 (ccd8) mutant of petunia (Snowden et al., 2005), showing a conserved link between CCD8 expression, branch development and leaf senescence in both perennial and annual species. Another Arabidopsis branching mutant max2 (Stirnberg et al., 2002) was originally isolated as the ore9 mutant through a screen for delayed onset of age-dependent leaf senescence (Oh et al., 1997) and was also shown to have delayed hormone-induced leaf senescence (Woo et al., 2001). The ORE9 gene was shown to be a member of the F-box leucine-rich repeat gene family and involved in ubiquitin-dependent proteolysis (Woo et al., 2001). A similar relationship between leaf senescence and branch development was observed in the rice tillering dwarf mutant d3, an orthologue of the Arabidopsis mutant max2/ore9 (Yan et al., 2007). However, the pea orthologue rms4 does not display delayed leaf senescence (Johnson et al., 2006), suggesting that the interaction between leaf senescence and branch development is not conserved in all plant species.
An inherent difficulty in the study of the regulation of branching in woody perennial plants is the length of time required for plants to develop branches. Consequently the length of time for screening transgenic plants for altered phenotypes is also an extended process. The first time point (4 months after transfer) selected for screening the transgenic kiwifruit plants for the expression of CCD8 provided an indication that there had been a reduction in the expression in the RNAi lines compared with the control lines; however, it was not until the two later time points that a significant reduction could be shown. An understanding of the timing of expression of CCD8 over the growth of a perennial plant and the relationship between the timing of CCD8 expression and the visible growth of a branch would assist in selecting the best time for screening RNAi transgenic plants as well as providing further insights into the role of CCD8 in the control of branch development.
We have demonstrated that the CCD pathway controlling branch development and leaf senescence observed in model plant systems is also conserved in a woody perennial plant. The effect of reducing root expression of AcCCD8 on fruit production is yet to be determined. However, the ability of low-vigour rootstocks (A. polygama and A. kolomikta) to increase the number of short branches on the scion, and consequently increase the fruit load (Clearwater et al., 2004, 2006), suggests that this aspect of plant development will be worth investigating further when the plants reach reproductive maturity.