This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
An understanding of complex triggers and molecular players regulating induction, maintenance and release of dormancy in plants are crucial to resolving problems associated with crop production. Although mostly motivated by the economic importance of crops, it has been the subject of several research studies concurrently leading to numerous fundamental findings in plant science. For over a decade now, the rate of yield increase for staple food crops have been dwindling and a cause for widespread concern that we might be approaching the sustainable yield barriers once overcome by the green revolution (Huang et al. 2002, Jung and Müller 2009). The need for new technologies is not only necessitated by this food security challenge, but also by rising and competing demands for plant biomass as a source of renewable energy. For both generative and vegetative crops, the phenological developmental processes are critical to increasing crop yield. This is evidenced by effects of photoperiod and vernalization on timing and vigor of flowering. A recent study demonstrates that genes linked to the control of circadian-mediated physiological and metabolic pathways have a major influence on growth vigor and accumulation of plant biomass (Ni et al. 2009). This is of great essence to productivity as reflected in seed crops where floral transition is a key developmental switch that determines dry matter yield, in vegetative crops like fodder grasses where early bolting limits potentials for high yields and in perennials like fruit trees where breeding progress is vulnerable to inconsistent late onset of flowering and loss of flowers and/or immature fruits to spring frost.
Although this review centers on dormancy in floral and vegetative buds of perennial fruit trees, other studies on dormancy among several in other plant taxa will also be mentioned as they relate to perennial fruit trees. Furthermore, after about a century of research on dormancy, the field of study has evolved in different aspects such as site of dormancy (seeds, floral and vegetative buds, tubers, bulbs and stolons), photoperiod and other environmental induction of dormancy, chilling requirement and its effective temperature range, differences in buds, modification of chilling requirement by environmental factors and cultural practices, models for calculating chilling requirement and dormancy breaking chemicals and/or stress treatments (Samish 1954, Wareing 1956, 1969, Romberger 1963, Vegis 1964, Leike 1965, Perry 1971, Erez and Lavee 1974, Saunders 1978, Doorenbos 1985, Saure 1985, Nooden and Weber 1978, Lang 1987, 1994, Champagnat 1989, Rowland and Arora 1997, Arora et al. 2003, Horvath et al. 2003).
Recent advances lacking in previous reviews will be highlighted to keep pace with the developments made in this discipline. These will include mechanisms underlying bud dormancy from induction to release, gene pathways and signals; cell-to-cell crosstalk, physiological delineation of different stages of dormancy, separation of dormancy from other related biological processes like freezing and dehydration tolerance, hormonal physiology and the genetics of dormancy in woody plants, which includes identification of associated quantitative trait loci (QTLs), mapping of dormancy-related genes, gene action of dormancy-related genes and expression profiles of these genes.
Complexity of Bud Dormancy and Overlap with Related Biological Processes
In spite of extensive progress made in dormancy in general, gaps are still prevalent at all stages of bud dormancy due to the complexity and nature of the molecular pathways that also overlap with other distinct biological processes, some of which are difficult to dissociate processes central to dormancy. An example of such biological processes includes freezing and dehydration tolerance and the dilemma of distinguishing cause and effects between them and dormancy. The capability of temperate perennials to surviving freezing winter temperatures depends heavily on their adaptation, which involves evolved mechanisms for transitioning into a dormant state as well as cold hardiness, a measure of freezing and dehydration (Powell 1987). The same environmental cues (photoperiods and colder temperatures) that cause a shift from summer dormancy or correlative dormancy (paradormancy) to winter dormancy (endodormancy), concurrently induces cold acclimatization, while plant tissues become more cold hardy during winter dormancy (Fuchigami 1978). Consequently, the induction and release from dormancy in the annual growth cycles of woody perennials is superimposed on the acquisition and loss of cold hardiness, respectively (Fuchigami et al. 1982). To resolve the physiological and molecular events associated with the regulation of bud dormancy and that of cold hardiness, several strategies have been employed to study them independent of each other.
The first effort involved the use of genetically related peach (Prunus persica) genotypes that segregated for deciduous and evergreen habits. The lack of endodormancy in one genotype and cold hardiness in both facilitated their use for a comparative study of changes in protein content as it relates to seasonal changes and the degree of cold hardiness (Arora et al. 1992, Arora and Wisniewski 1994, Artlip et al. 1997). Another system in Vitis labruscana was originally explored by Fennell and Hoover (1991) since it was able to completely transition into an endodormant state in response to short photoperiod but without cold acclimatization. Salzman et al (1996) eventually utilized the system to characterize differential expression of proteins in buds exposed to only short photoperiods. A study in blueberry (Vaccinium section Cyanococcus) cultivars attempted resolving the problem by observing changes in bud proteins specifically associated with dehardening based on the premise that only temperatures between 0o and 7o C are effective towards contributing to chilling unit accumulation (Erez et al. 1979, Erez and Couvillon 1986). Cold acclimatized buds (50% chilling requirement acquired) were exposed to controlled-temperature regimes warm enough to cause dehardening without negating chill unit accumulation or releasing it from endodormancy i.e. not affecting the dormancy status of the buds (Arora et al. 1997). Based on the studies mentioned above, there was a consensus that the metabolism of certain dehydrins, a subgroup of late embryogenesis abundant (LEA) proteins referred to as the D-11 family (Close 1997), was more closely associated with cold hardiness transitions than bud dormancy (Rowland and Arora, 1997, 2003).
Even though dehydrins are ubiquitous hydrophilic proteins considered to guard cells against cellular dehydration (in this case, freeze-induced dessication) and are therefore expected to build up in cold hardened tissues, Faust et al. (1997) thinks they might not be exclusive to cold hardiness but also involved in bud endodormancy. This was based on MRI studies where bound to free water ratio increased during late fall or early winter (Faust et al. 1991), hence, proposing that dehydrins bind water after being induced by low temperatures and abscisic acid (ABA) and leading to freeze protection and simultaneous deepening of dormancy.
Another attempt to dissociate bud dormancy and cold hardiness explored endogenous ABA levels and its effect on endodormancy and/or cold hardiness. Numerous studies have implicated ABA as a stress-inducible hormone and growth inhibitor as well as mediating short-day-induced growth cessation and dormancy induction in buds (Barros and Neill 1989, Dumbroff et al. 1979, Iwasaki and Weaver 1977, Lenton et al. 1972). After manipulating endogenous ABA content of buds and using an ABA-deficient mutant of birch (Betula pubescens), the involvement of ABA in dormancy induction was questioned in birch. The wild-type expressed elevated levels of ABA before onset of cold acclimatization under short-day, followed by tissue desiccation and accumulation of dehydrin proteins (Welling et al. 1997, Rinne et al. 1998), while the ABA-deficient mutant had lower water loss, lower tolerance to low-temperature stress and lacked accumulation of dehydrins. Nevertheless, the mutant was still capable of inducing dormancy (Rinne et al. 1998), suggesting that ABA was necessary for dormancy induction or that there were other pathways that augment the ABA-induced dormancy response. Increasing ABA contents by spraying ABA on long-day exposed plants and water stress also led to increased cold hardiness but without dormancy induction in the wild type (Welling et al. 1997). These findings support ABA has being more directly involved with photoperiodic control of cold acclimatization rather than in bud dormancy induction, although influence of ABA in other developmental processes of dormancy (maintenance and release) were not investigated.
The role of ABA in inducing and imposing dormancy in seeds (Karssen et al. 1983, Groot et al. 1991) and its requirement for dormancy maintenance during imbibition (Debeaujon et al. 2000, Grappin et al. 2000) are strong hints of its involvement in bud dormancy though studies have not confirmed this. Furthermore, inhibitory effects of glucose on seedling growth using glucose insensitive mutants (Rolland et al. 2002) showed that the same mutants had ABA deficiency and ABA insensitivity i.e. gin1, gin5, and gin6 were allelic to aba2/sis4/isi4, aba3/los5, and abi4/sun6/sis5/isi3, respectively (Eckardt 2002). To further confound ABA experiments, ABA biosynthesis genes have been shown by several studies to underestimate the complexity of sugar, stress and hormone interactions (Eckardt 2002). The complexities of these hormones' functions, distribution, accumulation and metabolic states are now been more appreciated than before (Cutler 2005). For example, one interesting aspect of ABA metabolic states as glucose esters (ABA-GE) were assumed to be dead-end catabolites just like auxin-conjugates but mounting evidence now confirm that ABA-GE is a transported state of ABA from which free ABA can be released in ''target'' tissues (Cutler 2005, Davies et al. 2005). ABA-protein conjugates that could not enter cells yet were also validated to be biologically active in inducing ion channel activity and gene expression (Schultz and Quatrano 1997, Jeanennette et al. 1998), while analogues of ABA in germination and gene expression studies have allowed for the identification of plausible ABA receptors with diverse structural requirements for activity in various response pathways (Walker-Simmons et al. 1997, Kim et al. 1999 Rock 2000).
Besides the complexities associated with ABA, inferences made from previous ABA experiments are difficult to validate considering that the promotion of flowering by a primary factor (day length and vernalization) can be interfered with or even eliminated by other less predictable factors. This has been demonstrated in studies where flowering was suppressed in favorable photoperiodic conditions by water stress in both long-day (Lolium temulentum) and short-day plants (Xanthium strumarium and Pharbitis nil) or by excess nitrogen input (Georges and Claire 2005). Sometimes, these primary factors can be conditional in certain instances like in Calceolaria that requires vernalization at low irradiance even though vernalization is not required at high irradiance (Bernier 1988). These interactions corroborate the fact that plants are sessile opportunists that will employ alternative factors to control flowering.
Regulation of Growth Cycles and Dormancy in Woody Perennials
While angiosperms originated in humid tropical climates where temperature, day length and availability of water were fairly stable all year round, one of the key evolutionary forces differentiating plant species and ultimately temperate species was environmental changes (Okubo 2000). This has a profound impact on their growth habit and life cycle. In order to synchronize timing of flowering with ambient temperature that ensures fertilization and seed/fruit development, perception and transduction pathways (vernalization) that senses prolonged cold winter temperatures has been evolved to translate environmental cues into increased competence for flowering in spring or summer. Even though early studies have shown that shortened growth period of shoots caused by water stress promotes early induction of bud dormancy leading to reduced dormany duration chilling requirement, there still remains a poor understanding of the molecular and mechanistic processes involved (Muller-Thurgau 1885, Arora et al. 2003). This was also confirmed by Chandler and Tufts (1934) after an extended growth period of shoots delayed bud break during the following spring when chill accumulation was not sufficient.
Plants initially undergo a period of vegetative development, but in woody perennials, this vegetative/juvenile stage can last for several years before the switch to a flowering developmental state. Infact there are several variations to plant life cycle with reference to transition to flowering in woody perennials. Raspberries, having a biennial or perennial growth habit, mostly produce a juvenile vegetative primocanes in the first year and adult fruiting laterals in the following year(s). The primocane phenotype has been a desired habit because it allows for some berry production in the first year, although yields and fruit size are low (Keep 1988). Prunus species on the other hand don't flower and fruit until the 2-3 years of juvenile phase is fulfilled. Georges and Claire (2005) provide an extensive review of the major factors that influence this flowering habit. Even though woody perennials require an initial vegetative phase, the size rather than age has been confirmed to be specifically more important (Lacey 1986). In nature, some plants that don't flower until the third to fifth vegetative phase are known to flower during the second year under cultivation in resource rich conditions (Lacey 1986, Kinkhamer et al. 1987). Thus, it seems the best predictor of flowering time in perennials is a threshold size which varies across different species and ecotypes (Lacey 1986, Wesselingh et al. 1993). Size in turn is directly related to amount of resources accumulated, which in turn depends on ambient temperature, irradiance, water/mineral availability and presence/absence of neighbors (Georges and Claire 2005).
Considering the difficulty of identifying QTLs in perennials that cosegregate with mapped photo-receptor genes, there are speculations that there are other molecular players downstream of photo-receptors that regulate flowering and dormancy by transducing the light signal. Perhaps, what the plant measures during the vegetative phase of development or before onset of flowering is biomass accumulation, which is a function of light, rather than light signal itself. Sucrose and Cytokinin have been identified as potential long-distance signaling molecules by analysis of the phloem sap exported by leaves in response to floral induction (Georges and Claire 2005). The increased export of sucrose in Arabidopsis in response to long-day induction might be partially due to increased efficiency of sucrose loading (Corbesier et al. 1998). After loading sucrose into the shoot apical meristem, a number of cellular and molecular events are initiated (Bernier 1988) and the hydrolysis of sucrose by local invertase i.e. vacuolar (Koch 2004) and cell wall (Heyer et al. 2004) invertase. Cytokinins activate invertase activity and increase the rate of cell division, while hexoses are known to participate with GAs in the upregulation of LFY expression (Georges and Claire 2005).
In Arabidopsis, the flowering response to environmental cues involves several signaling pathways but they all converge towards the regulation of floral meristem identity genes (Mouradov et al. 2002). Downstream of this convergence are the LEAFY (LFY) and APETALA 1(AP1) genes that control flower morphogenesis. Genes acting upstream of this are the considered integrator genes and their mutations adversely delays flowering in different growing conditions. The integrator genes include FLOWERING LOCUS T (FT) and SUPPRESSOR OF OVEREXPRESSION OF CO 1 (SOC1). The primary environmental factors that initiate the pathways include photoperiod and temperature (Martínez-Zapater et al. 1994). Mutants that flowered late and delayed in lond-day were termed genes in the "LD pathway", while mutants responsive to photoperiod but impaired in their response to cold were considered genes in the "vernalization pathway" (Figure 1). Mutants that were sensitive to both photoperiod and cold temperatures were classified into the "autonomous flowering pathway" (Georges and Claire 2005).
The response to vernalization is facilitated by cascade of gene regulatory pathways, which is initiated during prolonged cold exposure by the induction and upregulation of the homeodomain finger gene VIN3 (VERNALIZATION INSENSITIVE 3) and results in the chromatin-based and mitotically stable repression of the FLC (FLOWERING LOCUS C) gene (Sung et al. 2004). FLC, a MADS-box transcription factor, in turn acts as a repressor of floral transition. In the following generation, FLC expression is reset around the time of early embryogenesis (Sheldon et al. 2008, Choi et al. 2009), thus ensuring a renewed requirement for vernalization. Although the FLC ortholog in other Brassica species are functionally related to the Arabidopsis FLC (Tadege et al. 2001, Kim et al. 2007), the extent of conservation outside the Brassicaceae family is still contentious (Jung and Muller, 2009). Expressed sequence tags of the gene in rosids, asterids and caryophyllids have been identified (Reeves et al. 2007) but proof for the functional conservation remains inadequate (Jung and Muller, 2009). Additionally, in temperate cereals, identification of key regulators of vernilzation requirement and response (wheat VRN1, VRN2 and VRN3, which are homologs of Arabidopsis VRN gene) does not include an FLC-like gene and revealed regulatory pathway whose components differ from the FLC-dependent vernalization pathway.
Currently, regulatory mechanisms underlying floral induction in perennial plants remain poorly characterized, although attempts are being made to test pathways (Figure 1) already characterized in model plants like Arabidopsis. Floral induction in woody perennials differ from that of annual and biennials plants in that they comprise a morphogenetic transition of cells in apical meristems as well as in lateral meristems. In perennials, above ground meristems are not induced by strong floral promoter and therefore remain vegetative, thus guaranteeing a long life span. Little is known about how perennials achieve this but silencing of genes probably via DNA methylation and inaccessibility of floral promoters or RNAi may be a promising approach. This may be achieved by the long time transcription of floral-repressing genes like FLC or similar homologs in perennials (Chen and Coleman 2006, Bangerth 2009).
Figure 1: Flowering time control in Arabidopsis (a) and cereals (b). Exogenous cold () and light () signals are indicated by symbols. Positive and negative regulatory actions are indicated by arrows and lines with bars, respectively. Dashed lines designate more speculative interactions. The dashed line with a single filled circle at the end indicates a regulatory but yet little understood effect of LHY and CCA1 on SVP protein accumulation (Fujiwara et al. 2008). Lines with filled circles at either end indicate protein-protein interactions. The green and yellow boxes designate genes shown to affect natural variation in flowering time in Arabidopsis and cereal accessions, respectively. The figure incorporates aspects from various previously published models (He and Amasino 2005, Trevaskis et al. 2007, Alonso-Blanco et al 2009, Distelfeld et al. 2009). (Adapted from Jung and Müler 2009)
Following dormancy bud break, active bud growth is associated with actions of hormones and increased cell division. During this dormancy release, gene expression changes are tightly coordinated with the cell cycle (Devitt and Stafstrom 1995, Cambell et al. 1996, Horvath et al. 2002, Freeman et al. 2003). Understanding cell cycle regulation has been shown to be fundamental to understanding the molecular basis growth cessation during dormancy. In most cases, buds seem to be arrested at the transition between the G1-phase (preparation for DNA replication) and before the S-phase (DNA replication) (Gutierrez et al. 2002). Release from dormancy is accompanied by upregulation of genes (Fig. 2) such as D-type cyclins (CYCD) and histones that act at the G1-S-interphase (Devitt and Stafstrom 1995, Horvath et al. 2002, Freeman et al. 2003). Growth inducing signals (cytokinin, brassinosteroids, gibberellic acid) are known to be initially transduced (D'Agostino and Kieber 1999, Hu et al. 2000, Riou-Khamlichi et al. 2000, Ogawa et al. 2003) through post-translational modification of a series of proteins resulting in the transcription of CYCD and histones (Sherr 1994, Sauter 1997, Horvath et al. 2002).
Once CYD is expressed, it binds to cyclin-dependent kinases (CDKs) and this complex phosphorylates the retinoblastoma protein (RB) (Healy et al. 2001), which in turn releases bound transcription factors, including E2F that induces genes needed for DNA synthesis. The CDKs require activation by CDK-activating kinase (CAK) before they can phosphorylate RB. The RB-E2F complexes have been evidenced to play a role in chromatin remodeling required for appropriate expression of genes involved in regulation of cell cyle (Mironov et al. 1999). Chromatin remodeling leading to global changes in DNA methylation and histone H4 acetylation when overlayed with transitions in bud morphological changes (Meijón et al. 2010) has been in
Figure 2: Model for G1-S and G2-M transitions in plants based on combined models by Stals and Inzo (2001), Anderson et al. (2002) and Gutierrez (2002) and on recent results obtained from plants and animals. Activation of G1 progression involves the expression of D-type cyclins (CYCD) and their catalytic subunit, cyclin-dependent kinase (CDKA), dissociation of CDK inhibitory protein (ICK1) from CDKA-CYCD complex and phosphorylation of the Thr160 residue (P highlighted in purple) of CDKA. CYCD and CDKA are upregulated by various growth regulators including auxin, cytokinin, brassinosteroids (BR), sugar and gibberellic (GA). ICK1 is induced by abscissic acid (ABA). Phosphorylation of CDKA is the activity of CDK-activating kinase (CAK), which is induced by GA (Riou-Khamlichi 2000). Active CDKA-CYCD complex hyperphosphorylates retinoblastoma protein (RB), which inhibits its binding to transcription factors (E2F) and the docking protein (DP), thus initiating chromatin remodeling, transcription activation, DNA replication and S-phase transition. The SCF (SKP1-Cullin-F-box-protein) complex mediates ubiquitination and proteolysis (scissors) of ICK1 (Henchoz et al. 1997) and CYCD that is necessary to trigger the G1-S-phase transition in yeast (and possibly plants, denoted by '?'). Initiation of the G2-M-phase transition requires induction of the A- and B-type cyclins (CYCA/B) and the activity of A- and B-type CDKs (CDKA/B). The plant hormones auxin, cytokinin and GA have all been implicated in CYCA/B and CDKA/B expression and/or stability. At G2, the Thr160 (P highlighted in purple) of CDKs is positively phosphorylated (þ) by CAK, and the Thr14 or Tyr15 (P highlighted in green) of CDKs are negatively phosphorylated (Devitt 1995) by a tyrosine kinase in the CDKA/B-CYCA/B complex. A cytokinin-regulated tyrosine phosphatase (CDC25) removes the inhibitory phosphate and allows the G2-M-phase transition to occur. Commitment to mitosis requires ubiquitin-dependent proteolysis of B-type cyclins. The anaphase-promoting complex (APC) regulates ubiquitination and proteolysis of CYCA/B. Auxin appears to be involved in the degradation of cyclins. Jasmonic acid (JA) inhibits CDK activity in both the G1-S-phase and the G2-M-phase transitions. (Adapted from Horvath 2003)
turn confirmed to delimit four basic phases in the development of azalea buds (Rhododendron sp.). This allowed for the identification of a stage of epigenetic reprogramming that led to a sharp decrease of whole DNA methylation similar to that defined in other developmental processes in plants and animals (Kouzarides 2007, Tessadori et al. 2007). Initiation of the G2-M-phase (after the S-phase) requires induction of the B-type cyclins (CYCB) and the CDKB gene, while auxin, cytokinin and GA have all been implicated in the CYCB and CDKB expression and/or stability (Francis and Sorrell 2001). The interaction between CYCB and CDKB initiate phosphorylation and activation of proteins and expression of genes required for cytokinesis (Mironov et al. 1999). Even though there is a lot of information on regulation of growth and development, it is only recently that information are been linked between these processes to signals regulating bud dormancy.
Prior to growth cessation during winter, paradormancy (apical dominance or correlative inhibition) marks the first stage towards bud dormancy, which allows for the plant to allocate resources for reproduction, control plant architecture and maximize light harvesting while allowing for regeneration should individual shoots become damaged. Hormones have been proposed to be the major culprits for the induction of bud dormancy and implicated as transducers of environmental cues (Hermberg 1949). Infact, the term dormin was later proposed for endogenous dormancy inducers (Eagles and Wareing 1963).
Although it's quite alluring to think of dormancy on the basis of hormonal control alone, it is quite obvious that dormancy is controlled by several integrated plant structures and functions; and even its path is a continuum and begins as early as during budbreak in spring (Simpson 1990, Crabbe 1994). ABA as also been implicated in both short-day and water stress-induced dormancy in Betula pubescens (Rinne et al 1994a, 1994b, Welling et al. 1997) and Vitis vinifera 'Merlot Noir' (Koussa et al. 1998) where evidence support a relationship between ABA and bud water content. Depth of bud dormancy has also been suggested to be related to endogenous ABA levels (Tamura et al. 1993). Faust et al. (1991) demonstrated that endodormant buds has less free water than ecodormant buds, implying that chilling requirement satisfaction is related to the conversion of water from a bound state to Free State. Viccinium cultivars with the deepest dormancy and highest chilling requirement reportedly possess the most bound water (Parmentier et al. 1998), while bound water is also shown to increase in endodormant and freeze tolerant peach buds in response to induction by either photoperiods or cold temperatures (Erez et al. 1998). Although, the studies above concluded that bound the water status was associated with cold temperatures stress tolerance than directly to dormancy itself, Fennell et al. (1996) revealed increasing amount bound water after 2 weeks of short-day photoperiod exposure in Vitis riparia, while Fernell and Line (2001) demonstrated increasing amount of bound water with endodormancy in both grape buds and cortex/gap tissue adjacent to the bud.
Several studies initially monitored endogenous levels of hormones in whole buds, leaves, stems, cambium and root tissue under fall and dormancy-inducing controlled-environment conditions (Samish 1954, Wareing 1956, Nitsch 1957, Phillips and Wareing 1958, Dennis and Edgerton 1961) but the experiment though amenable (measuring responses and application of hormones) had several pitfalls. These problems include degradation and differential responses of commercial (±)-ABA and natural (+)-ABA (Wilen et al. 1996), root uptake of ABA reduced by casparian strip formation in hypodermis (Freundl et al. 2000), loss of ABA to medium if more alkaline than root cortex and pH of root zone and ABA concentration may modify root-to-shoot signaling as they affect apoplastic transport of ABA (Arora et al 2003). Strauss et al. (2001) also demonstrated experimentally that exogenously applied ABA was distributed differentially from compartmentalized endogenous ABA within the cell. Proteins and other molecules that bind and/or modify ABA might exist in the cytosol and/or endoplasmic recticulum and prevent ABA distribution based on cellular pH gradient alone.
The problems around hormones are further complicated by findings showing that their levels vary from basal to apical parts of the plants. Other factors to consider for hormone studies include the use of lateral buds against terminal buds, distinguishing determinate and indeterminate growth patterns, use of whole buds against partitioned bud tissues, sampling buds at quantitatively established stages of dormancy and differential photoperiodic response of young and mature leaves. Other more recent studies have further deepened the complexity of ABA due to many other processes mediated by ABA particularly auxin- and ethylene-triggered ABA induction.
While the implication of basipetal transport of auxin as the primary signal regulating paradormancy is well documented (Horvath 2003), other signals have been proposed to have significant influence on shoot outgrowth based on grafting studies (Cline 1994, Beveridge et al. 2000). Although the inhibitory role of basipetal transport of auxin on growth is slightly complicated by concurrent production of auxin in growing buds and by the plant's requirement for auxin, the effect auxin produced from the distal meristem seems to be different from that inside the buds once dormancy is broken. Several studies confirm that auxin signaling alters cell cycle directly or through crosstalk in concert with other plant hormones. It inhibits the production or sensing of cytokinin, which is required to induce both CYCD3 and CDKB expression (Francis and Sorrell 2001).
Other plant hormones acting alongside auxin in paradormancy include ABA and GA, which inhibit and promote growth, respectively. ABA induces expression of an inhibitor (ICK1) of CDK action at the G1-S-phase transition (Wang et al. 1997), while GA induces S-phase progression but not full induction of CYCA, CYCB and CDKB (Sauter 1997). Auxins signaling pathways are thought to target degradation of specific proteins and regulation of cytokinin production in the stem segments adjacent to the axillary buds (Shimizu-Sato and Mori 2001, Stirnberg et al. 2002, Xiangdong and Harberd 2003). It has also been proposed that auxin might regulate ABA content through expression of a P450 mono-oxygenase gene (Shimizu-Sato and Mori 2001). Details of a pathway or an auxin controlled complex remain elusive. Besides hormones, sugars also play a complex role in paradormancy. They are required for expression of CYCD3 in Arabidopsis (Healy et al. 2001, Oakenfull et al. 2002) and probably the expression of ICK1. The role of sugars in determining the competence of a perennial plant for flowering during the vegetative juvenile stage and just before bud set has been mentioned earlier (Georges and Claire 2005).
Bud Dormancy Maintenance and Release
Endodormancy, resulting from physiological changes, in woody perennials follows paradormancy in the growth cycle at the onset of winter. This response is internal to the bud and prevents untimely growth during seasonal transitions, when temperatures often fluctuate between favorable warm and inhibitory cold temperatures. This stage of dormancy reflects the plants adaptive mechanisms to maintain buds in a physiologically dormant state until a stable return of favorable conditions. Compared to paradormancy after dormancy induction, the molecular components of endodormancy maintenance are even more poorly understood and seem to overlap and share similar aspects with cold acclimatization, hence, making it more intractable and less investigated than other stages. It has been studied in buds of poplar (Populus deltoids) and grape (Vitis vinifera) and in potato tuber buds.
Endodormancy is known to occur concurrently as plant senescence during fall in several plant systems (Fedoroff 2002), with ethylene and ABA been implicated in both processes. In potato microtubers, ethylene directly induces endodormancy (Suttle 1998), while the role of ABA is well established in several systems including growth cessation in potato tubers and inhibitory effects seed germination. Cases of phytochromes acting synergestically with both ethylene and ABA has been reported (Finlayson et al. 1998, Weatherwax et al. 1998). The signaling pathways for this molecular mechanism in woody perennials are developing and especially for ABA action but no concrete connections have been characterized.
One major challenge in the horticultural industry is the economic importance of chilling on temperate fruit trees in regions with cold and warmer winter temperatures. Warmer climates lack sufficient chilling required to overcome floral and vegetative bud dormancy, while fulfilled chilling requirements in colder regions are prone to spring frost damage. Breeding new varieties for high chill are required in cold climates, while cultivars with variable (for early and relatively late blooms) low chilling requirement are desirable in warm climates. Several studies have been published on regulation of dormancy bud break (Saure 1985, Erez et al 1971, Iwahori et al. 2002) as well s the use of chemicals to break dormancy and the regime of application for greater efficacy and reduced phytotoxicity (Erez et al. 1971, Erez 1987, Fernandez-Escobar and Martin 1987, Siller-Cepeda et al. 1992, Wood 1993). A proper understanding of pathways involving signaling molecules and target genes underlying bud dormancy release may aid development of markers for proper timing of breaking practices for marker-assisted breeding (Tamura et al. 1998).
Several strategies have been utilized to elucidate dormancy bud break and they include approaches based on regulation within the apical meristem by changes in cell-to-cell communication and plasmodesmatal connections (van der Schoot 1996, Jian et al. 1997, Rinne et al. 2001), control of the cell cycle (Rhode et al. 1997, MacDonald 2000), regulation of water with initial findings based on supercooling examined the vascular connections into the bud (Sakai 1979, Ashworth 1984, Quamme et al. 1995), the sequence and regulation of water uptake into the bud (de Fay et al. 2000), water stress and availability during dormancy (Faust et al. 1997), studying molecular events involved in the perception and transduction of dormancy-breaking signals during chemical-induced dormancy release (Or et al. 200, 2002) and mechanism of dormancy induction and release via a metabolic and communication block or permeability barrier between the bud and adjacent tissues (Crabbe and Barnola 1996, Faust et al. 1997, Champagnat 1989).
To reproduce the effect of chilling requirement on dormancy release, horticultural practices have successfully used chemicals such as hydrogen cyanamide (HC) for controlled dormancy release in grape buds (Henzell 1991). Transcript populations from HC-treated and control buds have been used to identify a sucrose non-fermenting (SNF)-like protein kinase that is upregulated during initial stages of dormancy release (Or et al. 2000, 2002). Although the mechanisms underlying dormancy release using the chemicals are unknown, there is mounting evidence that an SNF-like protein kinases plays a role in the signaling cascade. Since SNF-like protein kinases are known to be transcriptionally regulated by stress stimuli in plants (Anderberg and Walker-Simmons 1992, Hardie 1994), Or et al. (2002) suggests that they might be involved in perception of stress signal induced by HC and similar chemicals (e.g. azide, cyanide, thidiazuron) in grape. These chemicals are theorized to transiently disrupt the respiratory metabolism by inducing H2O2 via oxidative stress, an explanation supported by reduced catalase activity (a free radical scavenger) soon after HC application (Nir et al. 1986, Wang et al. 1991, Faust and Wang 1993, Pérez and Lira 2004). The inhibition of catalase by HC could be as a result of H2O2 production or the action H2O2 as a chemical signaling molecule inducing the expression and upregulation of genes related to endodormancy release (Desikan et al 2000, Neill et al. 2002).
However, studies in dormant apple buds indicate that dormancy release in buds coincides with the upregulation of the antioxidant system, reflected in increased peroxide scavenging enzymes (Wang et al. 1988, Rowland and Arora 1997). The antioxidant machinery is also kown to be upregulated for protection against freezing stress (Guy 1990). More recently, the Mitogen-activated protein kinase (MAPK) cascade has been implicated to play a role in transducing signals involving reactive oxygen species (ROS) like H2O2 (Dóczi et al. 2007; Pitzschke and Hirt 2009) and in turn a corresponding H2O2-induced dormancy release in grape (Pérez and Lira 2005) and raspberry (Mazzitelli et al. 2007). Several studies now show that the MAPK cascade is not only induced by ROS but can also regulate production of ROS (Pitzschke and Hirt 2009). MAPK, sometimes referred to as extracellular-signal-regulated kinases (ERKs), are one of the best studied signal transduction pathways that play a central role in signaling cells to progress past the G1/S boundary (Meskiene and Hirt 2000; Roberts et al. 2000). These growth factor signaling pathways are implicated in the upregulation of cyclin D1 and CKIs (Cook et al. 2000) and in activation of CAK (Chiariello et al. 2000). Recently, components of the MAPK signal cascade have also been associated with oxidative stress-induced cell cycle arrest at G2/M (Chien et al. 2000; Kurata 2000).
As mentioned earlier, along with changes in gene expression, there is also evidence for more general epigenetic changes associated with endodormancy induction and release. Major changes in DNA methylation have been observed during bud set, dormancy induction and release in potato (Law and Suttle 2003) and azalea floral buds (Meijón et al. 2010). Increased DNA methylation and Histone deacetylation act simultaneously and co-ordinately following induction suggests that chromatin remodeling play an important role in restructuring chromatin and regulating gene expression during bud dormancy. Interetingly, the previously mentioned SNF1-like protein kinase, activated in grape by HC, is similar to kown component of a DNA modifying protein complex SW1-SNF from yeast and animals (Fan et al. 2003). Other components of this complex interact with RB-E2F (Figure 2) in both plants and animals (Shen 2002).
Genetic Control of Endodormancy-Related Traits in Woody Perennials
In the past decades, little effort was made to understand regulation of dormancy from a holistic and genetic perspective because dormancy-related traits like many other polygenic traits were considered too complex. This was partly due to limited genomics resources and analytical tools now just been developed in the last 2 decades (Tanksley and Hewitt 1988, Tanksley et al. 1989). Other obstacles that prevented performing genetic studies in woody perennials include a long generation time, high ploidy levels in economically important crops, inbreeding depression, self- and cross-incompatibility (Janick and Moore 1975, Moore and Janick 1983). Early genetic studies on bud dormancy comprised estimating heritability and classic mendelian genetic analyses of a few traits, followed by genetic studies of hazelnut (Thompson et al. 1985) and peach (Rodriguez et al. 1994) on evergrowing mutants suggesting their lack of dormancy induction was due to a single recessive gene. Hansche (1990) reported high heritability estimates for leaf abscission during fall and spring bloom date in peach, hence, implying a strong genetic component for these traits. Studies in apple (Malus x domestica Borkh) also confirmed a strong genetic component for chilling requirement and evidence that low chill requirement is controlled by a major dominant gene, while minor genes modulate its effect (Hauagge and Cummins 1991).
Most dormancy-related trait are inherited in a quantitative manner and displaying a continuous distribution in progenies, hence, indicating a polygenic mode of inheritance (Farmer and Reinholt 1986, Billington and Pelham 1991, Bradshaw and Stettler 1995, Lawson et al. 1995, Howe et al. 1999, 2000). The first QTL analysis on dormancy-related traits was performed in woody perennials and where conducted in an F1 population (double pseudo-testcross) of apple (Lawson et al. 1995) and in an F2 population of poplar (Bradshaw and Stettler 1995). Two QTLs were detected on the apple map for bud flush; while 5 QTLs explaining 85% of the phenotypic variance was detected in the poplar map. Another study with a larger F1 populayion size (172) in apple for vegetative bud flush detected 8 QTLs on 6 linkage groups that explained 42 % of the phenotypic variance (Conner et al. 1998); however, none of these linkage groups was homologous to the linkage group with the QTLs from the initial study. In poplar the population size was also increased to 346 in an F2 population for fall bud set and spring bud flush (Frewen et al. 2000). With the intent of mapping possible candidate genes, 3 QTLs distributed over 3 linkage groups were associated with bud set and 6 QTLs were distributed over 6 linkage groups for bud flush. The 3 bud set QTLs co-localized with 3 of the QTLs for bud flush implying that a single QTL could have pleiotropic effects on both traits as a result of shared components in their biochemical pathways. After comparing the 2 poplar maps, 3 QTLs were found to be common in both studies and all 3 contained bud flush QTLs. Following the mapping of candidate genes involved in perception of photoperiod, PHYB1 and PHYB2, and genes involved in the signal transduction of ABA response signals, ABI1B, ABI1D and ABI, only PHYB2 and ABI1B were found to map near QTLs affecting both bud set and bud flush but were not inside or co-localized with the QTLs. The lack co-localization of the perceptors of photoperiod further supports light only as an indirect signal and probably sucrose as the direct signaling molecule.
Several other maps have been constructed for detection of QTLs controlling bud dormancy and related traits but with little success at identifying candidate genes. Some of these studies include studies in apple (Liebhard et al. 2003, Segura et al. 2006), sour cherry (Wang et al. 2000), raspberry (Graham et al. 2009) and Douglas fir (Pseudotsuga menziesii Franco var. menziesii) (Jermstad et al. 2001). Besides bud set and bud flush QTLs, efforts are also been direct towards identifying QTLs and mode of gene actions underlying chilling requirement in blueberry (Vaccinium section Cyanococcus) (Rowland et al. 1999). Chilling requirement was chosen as a choice phenotype because of interest focused at developing low-chill requirement cultivars for warmer winter (Hancock and Draper 1989, Hancock et al. 1995). The chilling requirement of a cultivar is known to have a broad impact on the timing of bud flush, preventing growth during transitory periods and synchronizing plant growth with exposure to stable favorable conditions and in selecting for cold hardiness.
Vast amounts of information has been uncovered from transcriptome analysis and expression studies approaches that propose a plethora of plausible candidate genes in grape (Mathiason et al. 2009, Ophir et al. 2009), raspberry (Mazzitelli et al. 2007) and poplar (Rohde et al. 2007), but the short-comings of such studies lie in their inability to identify cause and effect genes from the differential expressions. The expression study on dormancy release by Mathiason et al. (2009) reported differential expression several genes already characterized in vernalization pathways of model plants in relation to flowering time, indicating that some components of these pathways are conserved in woody perennials. These genes include FLOWERING TIME LOCUS T (FT), SUPPRESSOR OF OVER-EXPRESSION OF CONSTANS 1 (SOC1), LEAFY (LFY), FRIGIDA (FRI), FLOWERING LOCUS C (FLC), GIGANTEA (GI), CONSTANS (CO), VERNALIZATION INDEPENDENT3 (VIN3), VERNALIZATION 1 (VRN) and VERNALIZATION 2 (VRN2).
The most prominent genes from an amplified fragment length polymorphism-based (cDNA-AFLP) transcript profiling in poplar (Rohde et al. 2007) was revealed by differential gene expression after exposure to 24 short-days. Three regulatory genes, AP2/EREBP (APETALA 2/ ETHYLENE-RESPONSIVE ELEMENT BINDING FACTOR 13), ERF4 (ETHYLENE RESPONSIVE ELEMENT BINDING FACTOR 4), and WRKY11 (Calmodulin binding/ transcription factor), were linked with critical steps in dormancy induction (Rohde et al. 2007). Inference from the gene function in respective Arabidopsis homologues show that they act downstream of the ethylene and/or abscisic acid (ABA) signaling cascade, which is successively initiated during bud development (Ruttink et al., 2007). The closest homologue of the AP2/EREBP transcription factor in Arabidopsis (RAP2.6L), an AP2-like ABA repressor 1 gene (McGrath et al. 2005, Nakano et al. 2006), acts in a network that regulates shoot regeneration from root explants (Che et al. 2006). The Arabidopsis homologue for ERF4 (McGrath et al. 2005) is induced transcriptionally by ethylene, ABA and jasmonate and has been identified independently during short-day-induced bud set in poplar (Ruttink et al., 2007). It acts as a transcriptional repressor that modulates ethylene and ABA responses in Arabidopsis, while overexpression causes ethylene insensitivity and reduced ABA sensitivity (Yang et al., 2005). The role of sugars was verified by Mazzitelli et al. (2007) where a putative raspberry plasma membrane H+-ATPase gene was signicantly upregulated during dormancy release. Sugar influx have been suggested to occur through H+/sugar symports based on the pH gradient produced by a (Améglio et al. 2000). Gevaudant et al. (2001) also confirmed in dormancy release of peach buds that carbohydrate uptake capacity of buds increases concurrently with the upregulation and increased activity of the plasma membrane H+-ATPase.
Molecular biology is advancing and changing rapidly and with the aid of new high-throughput technologies, holistic studies on an -omics platform can be conducted on large scale and fast pace for generation of huge amounts of data. This has changed the playing field and made approaches that were not feasible decades ago plausible. The field of gene expression for gene-by-gene expression and functional genomics approach for gene interactions has opened up new frontiers in understanding regulation of bud dormancy. Ongoing research is shedding more light on differentiating sensing and signaling genes from those that maybe be regulatory or target genes. Genetic mapping studies are providing information about key genes containing polymorphisms that are crucial for regulating bud dormancy. This will help to define gene dormancy cause and effect relationships since differential regulation of several genes only implicates them in the pathway and not as candidates for differences in phenotypic variation and genotype-specific responses. This will lead to remarkable practical value for breeding programs and judicious utilization of existing germplasms. A reasonable next step beyond identification of QTLs controlling bud dormancy will be the use of tightly linked markers for marker-assisted breeding and map-based cloning of causative/candidate genes from BAC libraries using chromosome walking approaches in unsequenced genomes or mining the sequences in genomes that are already sequenced. This will provide a targeted approach for identifying and characterizing genes and their causative polymorphisms as well as their roles in signaling pathways. A detailed understanding of the molecular components and gene networks will be indispensable for targeted manipulations of bud dormancy in crops of economic importance.
Subsequent studies in this thesis aims to generate a map of the apricot genome using genetic linkage and linkage disequilibrium (Association) mapping approaches towards defining genomic regions (QTLs) responsible for chilling requirement and dormancy bud break. These studies along with other genomic resources (BAC libraries, peach genome, QTL maps and expression study data) available from the Rosaceae community will also be used to identify candidate genes within the QTLs. Fine mapping of the genomic regions saturated with newly developed SSR markers and a comparative mapping approach among peach, apricot and apple QTL maps will facilitate this approach as well as preliminary expression studies of buds sample throughout bud dormancy induction to release.
Reported in this thesis, is a high density apricot linkage map comprised of SSR and AFLP markers, identification of several QTLs controlling chilling requirement and dormancy bud break, fine mapping of 2 of the major QTLs for candidate gene approach to gene discovery using a linkage disequilibrium mapping, a comparative mapping approach to determine conserved QTLs within the Pruns genus and within the Rosaceae family (Prunus and Malus). Preliminary expression studies are also reported on some candidate gene to further validate their role in regulating bud dormancy.