Cell wall modifications regulate flower development in dendrobium crumenatum

CELL WALL MODIFICATIONS REGULATE FLOWER DEVELOPMENT IN DENDROBIUM CRUMENATUM YAP YOU MIN [B. Sc. (Hons.), NUS] A THESIS SUBMITTED FOR THE DEGREE OF MASTER OF SCIENCE DEPARTMENT OF BIOLOGICAL SCIENCES NATIONAL UNIVERSITY OF SINGAPORE 2008 Acknowledgements I would like to express my sincere gratitude to my supervisors, Dr. Ong Bee Lian and A/P Loh Chiang Shiong, for their constant guidance and advice throughout the course of this research. I would also like to especially thank A/P Yeoh Hock Hin for providing the various enzymatic substrates, Dr. Carol Han and Mr. Heng Mok Wei Dennis for their invaluable help and advice regarding molecular work, Mr. Chong Ping Lee for his expertise in histology work, Mr. Yap Wee Peng, Mr. Hee Kim Hor Daryl, Mr. Koh Teng Seah, Miss Ng Seow Leng and Miss Lim Huiqin for their assistance in many ways. Big thanks also goes to Daphne, June, Cipto, David and Sinteck for their help in one way or another, and for providing much cheer and joy in the laboratory. Lastly, I would like to thank my family and Chuanling for their constant love and support. ii Contents Page no. Acknowledgements Abstract List of Tables List of Figures List of Abbreviations ii vi viii ix xi Chapter 1. Introduction 1 Chapter 2. Literature Review 2.1 Plant cell wall 2.1.1 Primary cell wall components 2.1.2 Bonds between cell wall components 4 4 4 6 2.2 Cell wall metabolism 2.2.1 Pectin demethylesterification 2.2.2 Pectin solubilisation 2.2.3 Cellulose metabolism 2.2.4 Hemicellulose metabolism 8 8 9 10 11 2.3 Cell wall hydrolases 2.3.1 Cellulase 2.3.2 Polygalacturonase 2.3.3 Pectin methylesterase 2.3.4 Exo-glycosidases 12 12 13 15 16 2.4 Flowers and their influences 18 2.5 Flowering physiology 2.5.1 Cell wall changes during flowering 2.5.2 Floral bud opening, flower longevity and their regulators 2.5.3 Flowering and the senescence programme 19 20 22 23 2.6 Orchids 2.6.1 Dendrobium crumenatum 24 25 Chapter 3. Materials and Methods 3.1 Plant material 27 27 3.2 Physical parameters analyses 27 iii 3.3 Microscopy 30 3.4 Electrolyte leakage 31 3.5 Cell wall compositional analyses 3.5.1 Preparation of EIR 3.5.2 Cellulose content 3.5.3 Total pectin content 3.5.4 Soluble pectin and hemicellulose contents 31 31 32 33 33 3.6 Cell wall enzyme analyses 3.6.1 Enzyme extraction 3.6.2 Cellulase assay 3.6.3 Polygalacturonase assay 3.6.4 Pectin methylesterase assay 3.6.5 Glycosidases assay 3.6.6 Soluble protein content 35 35 37 39 39 40 40 3.7 Gene expression profiling 3.7.1 Total RNA isolation 3.7.2 Estimation of RNA quality and quantity 3.7.3 Reverse transcription 3.7.4 PCR amplification 41 41 42 42 43 3.8 Controlling time of flower opening in D. crumenatum 3.8.1 General setup 3.8.2 Treatment 43 43 44 3.9 Statistical analysis 44 Chapter 4. Results 4.1 Growth changes 45 45 4.2 Changes in anatomy during development 45 4.3 Cell wall composition 4.3.1 EIR 4.3.2 Cellulose 4.3.3 Hemicelluloses 4.3.4 Total pectins 4.3.5 Soluble pectins 48 48 52 52 52 53 4.4 Activities of cell wall-based enzymes 4.4.1 Soluble proteins 4.4.2 Cellulase 56 56 58 iv 4.4.3 Polygalacturonase 4.4.4 Pectin methylesterase 4.4.5 β-galactosidase 4.4.6 β-glucosidase 4.4.7 β-mannosidase 4.4.8 β-xylosidase 61 61 63 65 65 68 4.5 Expression of cell wall-based enzyme gene transcripts 4.5.1 Total RNA integrity and quality 4.5.2 Optimization of PCR 4.5.3 Expression of β-GAL and PME during floral bud and flower development 68 70 70 74 4.6 Membrane stability 74 4.7 Control of flower opening 78 Chapter 5. Discussion 5.1 Cell wall changes related to D. crumenatum floral bud/ flower development 83 83 5.2 Model for cell changes accompanying D. crumenatum floral bud/ flower development 89 5.3 Species-specific variations in cell wall modifications associated with flowering 91 5.4 Relationship between flowering and senescence 95 5.5 Effects of growth regulators on the control of flower opening 96 5.6 Further works 100 Chapter 6. Conclusion 101 Chapter 7. References 102 Appendix A: 117 Publication in Scientia Horticulturae v Abstract The involvement of cell wall modifications, in particular, changes in cell wall components and activities of cell wall-based enzymes, in regulating flower development in a sympodial orchid, Dendrobium crumenatum, were investigated. Plants were subjected to cold treatments to release floral buds from dormancy, and various parameters were investigated from young floral bud stage till flower senescence. Anatomical studies demonstrated structural disorganization in sepals and petals in developing floral buds. The packing and arrangement of the cells were observed to become increasingly disorganized during flower opening and flower senescence. Subsequent analysis of cell wall composition showed that the cell walls of sepals and petals were modified extensively during floral bud development, flower opening and flower senescence, as observed by the changes in the amounts of cellulose, hemicelluloses and total pectins. Pectin solubilisation was also observed to commence during early floral bud development. Of the tested cell wall-based enzymes, β-glucosidase demonstrated the highest specific activity, followed by pectin methylesterase and βgalactosidase. Significant changes in the activities of the enzymes were also observed during floral bud and flower development. The results indicated that cell wall modifications began early in young floral buds, and regulated flower development. A model for cell wall modifications, which involved loosening of the cellulose/hemicellulose and pectin networks, in D. crumenatum was proposed. Furthermore, comparisons of the cell wall modifications in D. crumenatum floral buds/ flowers to those in other species suggested the presence of species-specific changes. Throughout the development of D. crumenatum floral buds up till flower opening, senescence hallmarks, such as the decrease in membrane stability, were observed. Attempts vi were made to generate floral buds that exhibited abnormal patterns of flowering for future studies. This would allow comparisons of cell wall modifications (and other physiological factors) between the normal and abnormal floral buds. Exogenous application of the plant growth regulator, benzyladenine, was found to suppress flower opening and caused floral buds to abort. vii List of Tables Page no. Table 1. Stages of floral bud development in D. crumenatum. 28 Table 2. Extraction procedures of cellulase and PG from floral buds of D. crumenatum. 36 Table 3. Total and soluble pectins in EIR derived from D. crumenatum sepals at various developmental stages. 54 Table 4. Total and soluble pectins in EIR derived from D. crumenatum petals at various developmental stages. 55 Table 5. Quality of RNA obtained from D. crumenatum samples at various floral bud/ flower developmental stages. 72 Table 6. Percentages of D. crumenatum floral buds that displayed full flower opening one day after treatment. 81 Table 7. Percentages of D. crumenatum floral buds that displayed dormancy after one day treatments, and percentages of dormany floral buds that subsequently aborted two days after treatment. 82 Table 8. Summary of modifications of cell wall polysaccharides and cell wall enzyme activities during floral bud development, flower opening and flower senescence of D. crumenatum, carnation, sandersonia and daylily. 94 viii List of Figures Page no. Fig. 1 Dendrobium crumenatum (pigeon orchid). 26 Fig. 2 Separation of floral parts in D. crumenatum floral buds and flowers. 29 Fig. 3 Sequential extraction of pectins and hemicelluloses from EIR of D. crumenatum floral buds. 34 Fig. 4 Characteristics of weight changes of sepals and petals of D. crumenatum. 46 Fig. 5 Changes in anatomical features of D. crumenatum during flower development. 47 Fig. 6 Anatomical changes during D. crumenatum floral bud development. 49 Fig. 7 D. crumenatum sepal and petal cell wall components, expressed on a per floral part basis, at each developmental stage. 50 Fig. 8 D. crumenatum sepal and petal cell wall components, expressed on a per gram fresh weight basis, at each developmental stage. 51 Fig. 9 D. crumenatum sepal and petal soluble protein contents at each developmental stage. 57 Fig. 10 Optimisation of cellulase gel diffusion assay using commercial cellulase. 59 Fig. 11 Gel diffusion assay for cellulase. 60 Fig. 12 Changes in pectin methylesterase activity in sepals and petals during development of D. crumenatum floral buds. 62 Fig. 13 Changes in β-galactosidase activity in sepals and petals during development of D. crumenatum floral buds. 64 ix Fig. 14 Changes in β-glucosidase activity in sepals and petals during development of D. crumenatum floral buds. 66 Fig. 15 Changes in β-mannosidase activity in sepals and petals during development of D. crumenatum floral buds. 67 Fig. 16 Changes in β-xylosidase activity in sepals and petals during development of D. crumenatum floral buds. 69 Fig. 17 Electrophoresis of total RNA from sepals and petals of D. crumenatum. 71 Fig. 18 Gradient PCR amplification of β-TUB, β-GAL and PME transcripts. 73 Fig. 19 PCR amplifications of β-TUB, β-GAL and PME transcripts over 24, 26, 28, 30 and 32 cycles. 75 Fig. 20 Expression of β-GAL and PME transcripts at various developmental stages of D. crumenatum floral buds/flowers. 76 Fig. 21 Membrane stability of sepals and petals of D. crumenatum at various floral bud/flower developmental stages. 77 Fig. 22 General physical features of day 9 D. crumenatum floral buds after treatments. 80 Fig. 23 Proposed model for cell wall modifications accompanying D. crumenatum floral bud and flower development. 90 Fig. 24 Comparison of flower development events in carnation, sandersonia, daylily and D. crumenatum. 92 x List of Abbreviations AOA Ara BA CDTA CMC CSP DNS EDTA EIR GA Gal GalA Gal-Fuc GOD MSI PAR PCD PG PGR PME Rha SSP TBA Xyl β-gal β-glu β-man β-xyl Aminooxyacetic acid α-L-arabinose Benzyladenine trans-1,2-cyclohexanediamine-N,N,N’,N’-tetraacetic acid Carboxymethylcellulose CDTA-soluble pectins Dinitrosalicylic acid Ethylenediaminetetraacetic acid Ethanol-insoluble residue Gibberellic acid α-D-galactose D-galacturonic acid β-D-galactosyl-α-L-fucose Glucose oxidase/о-dianisidine Membrane stability index Photosynthetically active radiation Programmed cell death Polygalacturonase Plant growth regulator Pectin methylesterase 1,2-α-L-rhamnose Na2CO3-soluble pectins Tertiary-butanol α-D-xylose β-galactosidase β-glucosidase β-mannosidase β-xylosidase xi Chapter 1. Introduction Flowering is the first step of sexual reproduction in plants, and is a highly controlled biological event in the life cycle of the angiosperms (Bernier et al. 1993; van Doorn and van Meeteren 2003). Flowering and the eventual senescence of the flowers are also events of commercial value, as they contribute to the visual quality and postharvest vase-life of the flowers (O’Donoghue et al. 2002; Nell 2007). The cut flower trade has become a globalized market, involving US$4.5 billion in international trade yearly, and with Singapore as one of the major players in orchid export (Hew and Yong 2004; O’Donoghue 2006). Besides, with the world’s increasing interest in ‘green buildings’ to aid energy efficiency and the accompanying issue of using flowers for aesthetic benefits (Spala et al. 2008), it is important to understand the biochemistry, physiology and genetics for flowering, longevity and senescence (Nell 2007). However, publications on the study of tropical flowers are limited and the few detailed studies focus mainly on flowers of temperate species. There has been some focus on the possible involvement of cell wall modifications and/or cell wall remodelling in the regulation of flowering (O’Donoghue 2006). Flowers that have been studied include alstroemeria (Alstroemeria peruviensis), carnation (Dianthus caryophyllus L.), daylily (Hemerocallis spp.) and sandersonia (Sandersonia aurantiaca Hook.) (de Vetten and Huber 1990; Panavas et al. 1998; O’Donoghue et al. 2002; Wagstaff et al. 2003). These studies also addressed the question on whether flowering could be a process regulated by the senescence programme of the plant. Some supporting evidence include the increase in oxidation of membrane components prior to flower opening in daylily (Panavas and Rubinstein 1998), and the increasing trend of DNA laddering throughout petal development in alstroemeria (Wagstaff et al. 2003). Studies on carnation and sandersonia 1 flowers demonstrated that the transitions of floral stages from opening floral bud to fully mature flower till senescence were accompanied by changes in the levels of various cell wall polymers, such as cellulose and pectins, and activities of cell wall-based enzymes (de Vetten and Huber 1990; O’Donoghue et al. 2002). These observations were similar to the loss of cell wall integrity in ripening fruits of carambola (Averrhoa carambola) and grapes (Vitis vinifera) (Chin et al. 1999; Deng et al. 2005). In daylily flowers, analyses of cell wall composition were not published, but reported changes in activities of cell wall-based enzymes during flower development suggested the involvement of cell wall metabolism in flowering (Panavas et al. 1998). While cellulase activity was detected in daylily flowers, it was reported to be absent in sandersonia flowers, indicating the possibility of a speciesspecific variation in cell wall metabolism that regulates flowering (Panavas et al. 1998; O’Donoghue et al. 2002). In the above-mentioned studies on the possible involvement of cell wall modifications and cell wall remodelling in regulating flowering, cell wall changes were compared only between stages of mature bud (just prior to opening), opening flower, mature flower, wilting flower and senesced flower (de Vetten and Huber 1990; O’Donoghue et al. 2002; Wagstaff et al. 2003). To fully understand if and/ or how flowering is regulated by a senescence programme that has already started, investigating the physiological, biochemical and molecular changes occurring throughout the development of a newly-induced young floral bud till flower senescence would be advantageous. Few studies on the physiology of flowering in tropical orchids have been conducted to date. As orchid cultivation continues to be a highly profitable commercial market (Hew and Yong 2004), characterization of tropical orchid flowering is of paramount importance. 2 Dendrobium crumenatum (Swartz), also known as the pigeon orchid, is a common native epiphytic orchid species of South-east Asia. The floral buds of D. crumenatum exhibit dormancy, and can be induced to resume growth and development by cold-induction, culminating into the opening of the flowers exactly nine days after (Holttum 1953; Corner 1988). Flower opening in D. crumenatum is a rapid and short process, taking about 4 hours from the onset of floral bud crack, achieving full flower opening before dawn (Yap 2006). The flowers are short-lived, lasting for only 24 h under natural conditions, before senescence sets in (Tan and Hew 1993). The synchronized and predictable pattern of floral bud development in D. crumenatum, together with its short life cycle, makes the pigeon orchid an ideal system to study the control of flowering. In this study, anatomical changes in sepals and petals were studied over various stages throughout the development of newly-induced floral buds till flower senescence in D. crumenatum. The levels of cell wall components such as cellulose, hemicellulose and pectins, activities of various cell wall-based enzymes, and their corresponding expression levels of the gene transcripts were also followed throughout development. We aim to use all these data to help us understand the involvement of cell wall modifications and remodelling in the regulation of flowering in tropical orchids. We also aimed to develop a system to obtain floral buds that exhibit abnormal flower opening patterns. Attempts to delay flower opening, using plant growth regulators, were made in mature D. crumenatum floral buds. Such system would be useful for future studies that may involve analysing and comparing cell wall modifications between floral buds showing normal and abnormal opening patterns. 3 Chapter 2. Literature Review 2.1 Plant cell wall Each plant cell comprises a specialised and complex cell wall that serves many functions (Cosgrove 1999). The cell walls provide structural support and maintain the shape of the cells, act as a protective barrier against water loss, pathogens, and other mechanical and environmental stresses, take part in cell-cell communication and interaction by carrying surface signalling molecules, and act as a storage organ for carbohydrates, proteins and various other materials (Cosgrove 1999; Carpita and McCann 2000). The high mechanical strength and rigidity of the cell walls allowed plants to become some of the largest organisms on Earth. Yet, the cell walls remain extensible, allowing plant cells to grow until the cells cease growth (Cosgrove 1999). The dynamic structure of the cell wall is important in regulating cell expansion and cell growth. For example, cell wall loosening is a pre-requisite for the incorporation of newly synthesized wall polymers during cell expansion and cell growth (Carpita and McCann 2000). Plant cell walls also have important roles in controlling fruit ripening. The degree of cell wall disassembly and/or cell wall weakening in fruits has been shown to regulate the time of ripening and the texture of a variety of fruits (Brummell 2006). Cell wall modifications in petals have also been linked to flowering (O’Donoghue 2006). 2.1.1 Primary cell wall components The primary cell wall is a complicated matrix, composed of various polymers made up of polysaccharides, proteins and some phenolics (Carpita and McCann 2000; Brummell and Harpster 2001; Brummell 2006; Liepman et al. 2007). A highly hydrated complex, the 4 primary cell wall also contains various aromatic substances, dissolved solutes and ions, and soluble proteins including enzymes (Brummell 2006). Cellulose is the most abundant plant polysaccharide and acts as the principal scaffold in plant cell walls (Carpita and McCann 2000). Cellulose microfibrils are composed of 1,4-βD-glucan chains assembled together by extensive hydrogen bonding, resulting in long, rigid, inextensible fibres. The cellulose microfibrils have a crystalline internal region that excludes water, and an amorphous outer layer that interacts with other matrix molecules (Pauly et al. 1999; Carpita and McCann 2000; Brummell 2006; Liepman et al. 2007). Another main component of the cell wall is hemicellulose (also known as crosslinking glycan), which can hydrogen bond to cellulose microfibrils, thus forming a network between various microfibrils (Carpita and McCann 2000). Predominantly made up of neutral sugars, hemicelluloses are neutral or weakly acidic. There are three major types of hemicellulose. The first, xyloglucan, is the most abundant. Similar to cellulose, xyloglucan comprises of a 1,4-β-D-glucan backbone, but has regularly spaced α-D-xylose (Xyl) side chains (on three consecutive glucose residues out of four). The xylose side chains may also be extended with β-D-galactosyl-α-L-fucose (Gal-Fuc) or α-L-arabinose (Ara). Another major hemicellulose is xylan, which has a backbone comprising of 1,4-β-linked xylopyranosyl units. Another form of xylan is arabinoxylan, which has a backbone consisting of 1,4-β-Dxylan, with occasional α-L-arabinose substitutions. The third major hemicellulose is (galacto)glucomannan, comprising of alternating regions of 1,4-β-D-glucan and 1,4-β-Dmannan in approximately equal amounts. Single units of terminal α-D-galactose (Gal) are also occasionally found on galactoglucomannan (Carpita and McCann 2000; Brummell and Harpster 2001; Brummell 2006; Minic and Jouanin 2006; Liepman et al. 2007). 5 The cell wall is also a pectin rich construct. Pectins belong to a class of polysaccharides that can be linear or branched, highly hydrated and rich in D-galacturonic acid (GalA) residues. Some functions of pectins include regulating wall porosity (Baron-Epel et al. 1988) and cell-cell adhesion at the middle lamella (Pena and Carpita 2004). Due to their ability to control wall porosity, pectins may also affect cell wall modifications by regulating the access of cell wall enzymes to their respective substrates in the matrix (Carpita and McCann 2000; Brummell 2006). One of the fundamental constituents of pectins is homogalacturonan, which has a backbone of 1,4-α-D-GalA. Pectins may exist as linear, unbranched homopolymers of 1,4-α-D-GalA, or exist as structurally modified homogalacturonans, such as xylogalacturonan, which has a homogalacturonan backbone with single Xyl side chains. Another modified homogalacturonan is rhamnogalacturonan II, which has a homogalacturonan backbone with highly conserved side chains, consisting of a diversity of neutral sugars. Pectins may also exist as rhamnogalacturonan I. Rhamnogalacturonan I has a backbone of alternating 1,2-α-L-rhamnose (Rha) and GalA disaccharide units, and may also possess linear or branched arabinan and galactan side chains (Carpita and Gibeaut 1993; Carpita and McCann 2000; Brummell 2006; Liepman et al. 2007). In brief, the primary cell wall consists of homogalacturonan, rhamnogalacturonan I, rhamnogalacturonan II, hemicellulose and cellulose, with an almost equal distribution of each component, whereas the middle lamella consists of mainly homogalacturonan and structural proteins (Brummell 2006). 2.1.2 Bonds between cell wall components The various cell wall components are linked by a variety of bonds (Brummell 2006). Hydrogen bonds occur extensively between cellulose microfibrils and also help to bind 6 xyloglucan to cellulose. Homogalacturonan molecules are attached to each other by ionic calcium bridges, and pectin molecules are bound to each other, or to other pectins, hemicelluloses or phenolic molecules via ester linkages (Fry 1986; Carpita and Gibeaut 1993; Iiyama et al. 1994; Rose and Bennett 1999; Carpita and McCann 2000; Liepman et al. 2007). Covalent linkages have also been reported between homogalacturonan and rhamnogalacturonan II (Vincken et al. 2003), between xyloglucan and arabinan or galactan side chains of rhamnogalacturonan I (Thompson and Fry 2000; Popper and Fry 2005), between rhamnogalacturonan I and extensin (Qi et al. 1995), and between various structural proteins and phenolics (Fry 1986). Besides being chemically bonded to one another, the various cell wall components may also be attached to each other by physical means, via entanglement, for instance (Brummell 2006). For example, rhamnogalacturonan I side chains are wound around cellulose microfibrils, creating a pectin network that is interlocked with the cellulose-hemicellulose network (Vincken et al. 2003; Zykwinska et al. 2005). Due to the various bonds present in the cell wall matrix, the study of cell wall composition requires specific chemical treatments to release the respective cell wall components (Brummell 2006). The chemical treatments include chelating agents such as trans-1,2-cyclohexanediamine-N,N,N’,N’-tetraacetic acid (CDTA) or ethylenediaminetetraacetic acid (EDTA) to extract ionically bound pectins, sodium carbonate to extract pectins held by ester bonds, weak alkali such as 1 M KOH to extract loosely attached hemicelluloses, and strong alkali such as 4 M KOH to extract hemicelluloses held tightly by hydrogen bonds (Brummell 2006). 7 2.2 Cell wall metabolism Various regulated cell wall architecture changes occur with the development of plants (Carpita and McCann 2000). Fruit ripening is one developmental event whereby many changes occur in the cell walls, resulting in the final texture of the fruit (Brummell 2006). Consequently, the majority of the information on cell wall metabolism is derived mainly from fruits. Some of the cell wall modifications involved include pectin demethylesterification, pectin solubilisation, and metabolism of cellulose and hemicellulose (Carpita and McCann 2000; Brummell and Harpster 2001; Brummell 2006). 2.2.1 Pectin demethylesterification During pectin synthesis, pectins are polymerised in the cis Golgi, and are subsequently methylesterified in the medial Golgi (Goldberg et al. 1996). The methylesterified pectins may also be substituted with side chains in the medial Golgi cisternae (Goldberg et al. 1996). Thus, pectins are secreted into plant cell walls in highly methylesterified forms (Carpita and McCann 2000; Micheli 2001). The degree of methylesterification, however, decreases with development and has been shown to be a crucial physiological change during fruit ripening (Brummell et al. 2004), microsporogenesis and pollen tube growth (Wakeley et al. 1998; Futamura et al. 2000), seed germination (Ren and Kermode 2000), and hypocotyl elongation (Bordenave and Goldberg 1993). Roy et al. (1992) demonstrated that the time of onset and area of the cell wall to undergo pectin demethylesterification were tightly regulated in tomato (Lycopersicon esculentum). Highly methylesterified pectins became increasingly less methylesterified as ripening of tomatoes progressed, and the demethylesterification process started in the middle lamella, spreading throughout the rest of the cell wall (Roy et al. 1992). 8 The removal of methylester groups from pectin results in negatively charged carboxylic groups (Grignon and Sentenac 1991). These charged surfaces may be involved in regulating pH and ion balance, in turn, affecting the activity of cell wall hydrolases (Chun and Huber 1998; Almeida and Huber 1999). The charged surfaces resulting from demethylesterification of pectins may also affect the movements of charged molecules, such as proteins, within the cell wall matrix (Grignon and Sentenac 1991). In the presence of calcium, the demethylesterified charged pectate molecules can aggregate and bind to one another via calcium cross-links, forming calcium-pectate gels which can increase stiffness of the cell wall (Jarvis 1984). 2.2.2 Pectin solubilisation One of the major changes in cell wall pectins with the development of plants is the increasing solubilisation of pectins, as commonly observed during the ripening of fruits (Brummell and Harpster 2001). Pectin solubilisation is usually measured as the increase in ease of extractability of pectins by various extractants, and can be extrapolated to bond changes within the cell wall matrix (Brummell and Harpster 2001). In watermelon (Citrullus lanatus), the amounts of water-soluble and chelator-soluble pectins increased at the expense of sodium carbonate-soluble pectins during ripening (Rose et al. 1998). In tomato and avocado (Persea americana), increases in amounts of water-soluble pectins were observed in conjunction with decreases in the amounts of sodium carbonate-soluble pectins, but changes in amounts of chelator-soluble pectins were absent (Carrington et al. 1993; Wakabayashi et al. 2000). The increases in water-soluble and/or chelator-soluble pectins were attributed to the increasing proportions of pectins that were more weakly attached to the cell wall matrix. 9 Pectin demethylesterification has been proposed as a possible cause for pectin solubilisation (Brummell 2006). The resultant regions of negatively-charged groups during demethylesterification could cause electrostatic repulsion between negatively-charged molecules, detaching the pectins that were weakly bound to the cell wall (Grignon and Sentenac 1991). The loss of arabinan and galactan side chains from rhamnogalacturonan I has also been suggested to cause pectin solubilisation (Brummell 2006). The arabinan and galactan side chains firmly bind pectins to the cell wall via covalent linkages, hydrogen bonds or physical entanglement (Popper and Fry 2005; Zykwinska et al. 2005). Thus the loss of the side chains could cause loosening of the pectins. In papaya (Carica papaya), the presence of the galactan degrading enzyme, β- galactosidase, also caused increased pectin solubilisation (Ali et al. 1998). Pectin solubilisation is believed to result in cell wall swelling (Redgwell et al. 1997). This would indirectly cause changes in the movement of cell wall enzymes through the wall matrix, increasing their accessibility to their respective substrates (Brummell 2006). 2.2.3 Cellulose metabolism Due to the insolubility of cellulose in standard solvents, and its highly susceptible nature to hydrolysis in harsh solvents, the quantification of cellulose changes during development of plants is a challenging procedure. Consequently, little has been published regarding cellulose degradation. As the main component of the cell wall matrix, cellulose is expected to undergo distinct alterations during cell wall changes (Fischer and Bennett 1991; Rose and Bennett 1999). Cell wall cellulose content decreased during ripening of grapes (Deng et al. 2005). However, during the ripening of pear (Pyrus communis), tomato and avocado, cellulose levels remained constant or even increased slightly (Gross and Wallner 1979; Ahmed and 10 Labavitch 1980; Maclachlan and Brady 1994; Sakurai and Nevins 1997). The resistance of cellulose towards enzymatic degradation in most fruits reflects the classical description of cellulose microfibrils as structurally stable and highly crystalline (Rose and Bennett 1999). It has also been suggested that cellulose metabolism is not a major feature of cell wall modifications during plant development, specifically, during fruit ripening (Brummell 2006). 2.2.4 Hemicellulose metabolism It has been widely accepted that cellulose microfibrils are coated with hemicelluloses, particularly xyloglucan, on their surfaces and are further cross-linked by xyloglucan chains, thus creating a three-dimensional cellulose-xyloglucan network (Fischer and Bennett 1991; Rose and Bennett 1999; Carpita and McCann 2000). Hemicellulose metabolism would, thus, significantly affect the extent of the cellulose-xyloglucan network and the primary cell wall structure (Fischer and Bennett 1999; Carpita and McCann 2000; Brummell 2006). Hemicellulose content has been shown to decrease during the ripening process in a variety of fruits such as tomato, strawberry (Fragaria ananassa), muskmelon (Cucumis melo), capsicum (Capsicum annum), pepino (Solanum muricatum), carambola, grapes, boysenberry (Rubus idaeus x Rubus ursinus) (Huber 1983a, 1984; McCollum et al. 1989; Sethu et al. 1996; O’Donoghue et al. 1997; Chin et al. 1999; Deng et al. 2005; Vicente et al. 2007). The breakdown of hemicelluloses has been suggested to be a major contributor to reduced cell wall turgidity, resulting in the softening of cell walls and possible cell wall expansion (Fischer and Bennett 1999; Brummell 2006). Relaxation of the cellulosexyloglucan network brought about by xyloglucan breakdown could also cause cell wall swelling (Brummell 2006), which affects pectin solubilisation as mentioned previously. 11 2.3 Cell wall hydrolases The various cell wall modifications are the results of the actions of a range of cell wallmodifying enzymes, the activities of which vary with development. The activities of cell wall hydrolases and the expression of their corresponding genes have thus been intensely studied to fully understand cell wall modifications (Fischer and Bennett 1991; Brummell and Harpster 2001; Minic and Jouanin 2006). To cope with the complex cell wall polymers, the activities of the cell wall localised or plasma membrane bound enzymes are very diverse (Minic and Jouanin 2006). The majority of the information on cell wall hydrolases in plants is derived from studies on Arabidopsis and various fruits. Various families of cell wall hydrolases, such as glycosidases (also known as glycoside hydrolases) and carbohydrate esterases, have been shown to participate in cell wall modifications (Minic and Jouanin 2006). Glycosidases are enzymes that catalyse the hydrolysis of the glycosidic linkages between two or more carbohydrates or between a sugar moiety and a non-sugar moiety (Davies and Henrissat 1995). Cellulase, polygalacturonase, βglucosidase, β-galactosidase, β-mannosidase and β-xylosidase belong to the glycosidase family (Minic and Jouanin 2006; Minic 2008). Carbohydrate esterases are enzymes that catalyse the removal of non-carbohydrate groups on substituted polysaccharides. For example, pectin methylesterase belongs to this family of enzymes, and catalyses the removal of methyl groups from polysaccharides (Micheli 2001; Minic and Jouanin 2006). 2.3.1 Cellulase Modifications in cell wall glucans occur due to the actions of endo-1,4-β-glucanase (EC 3.2.1.4), or more commonly referred to as cellulase (Wood and Bhat 1988; Brummell and Harpster 2001). Cellulase catalyses the hydrolysis of linkages of 1,4-β-D-glucan chains 12 adjacent to unsubstituted residues (Brummell and Harpster 2001). Substrates of plant cellulase include carboxymethyl cellulose, amorphous cellulose and xyloglucan, and to a lesser extent, crystalline cellulose (Ohmiya et al. 1995; Molhoj et al. 2001). Cellulase activity has been detected in a variety of fruits, although the amount and pattern of change vary considerably. During ripening of banana (Musa acuminata) and pawpaw (Asimina triloba), cellulase activity (units mg protein-1) increased (Lohani et al. 2004; Koslanund et al. 2005); in carambola, guava (Psidium guajava), grapes and boysenberry, cellulase activity (units gFW-1) also increased during ripening (Chin et al. 1999; Abu-Bakr et al. 2003; Deng et al. 2005; Vicente et al. 2007). There was however no changes in levels of cellulase activity (units mg protein-1) during ripening in capsicum (Sethu et al. 1996), and the level of cellulase activity (units gFW-1) decreased during ripening in apple (Malus domestica) (Goulao et al. 2007). The action of cellulase on xyloglucan may cause the breakdown of the cellulose-xyloglucan network, and has been suggested to be a mechanism for fruit softening (Rose and Bennett 1999). However, molecular studies demonstrated that suppressing expression of the mRNA of the cellulase genes, LeCel1 and LeCel2 in tomato, and overexpression of CaCel1 in pepper did not substantially affect the ripening processes (Brummell and Harpster 2001). 2.3.2 Polygalacturonase Polygalacturonases (PGs) are cell wall-based enzymes that catalyse the hydrolytic cleavage of galacturonide linkages in pectins (Fischer and Bennett 1991; Brummell and Harpster 2001). Both exo- and endo-acting types of PGs have been identified and characterized in fruits (Hadfield and Bennett 1998). Exo-PG (EC 3.2.1.67) removes GalA residues from the non-reducing ends of polygalacturonic acid, while endo-PG (EC 3.2.1.15) cleaves 13 polygalacturonic acid at random (Brummell and Harpster 2001). Endo-PG, rather than exoPG, has been correlated with cell wall modifications such as pectin degradation (Huber 1983b; Fischer and Bennett 1991; Hadfield and Bennett 1998). Consequently, the majority of studies on cell wall changes focus on endo-PG, and the enzyme will be referred to hereafter as PG. The main substrates for PGs are the homogalacturonans that are secreted into plant cell walls in highly methylesterified forms, which must be de-esterified before the enzyme can hydrolyse them (Jarvis 1984; Carpita and Gibeaut 1993; Minic and Jouanin 2006). PG has been suggested to be a key enzyme in cell wall modifications during fruit ripening and softening (Fischer and Bennett 1991; Brummell and Harpster 2001; Minic and Jouanin 2006). During ripening of papaya, tomato, carambola, guava and grapes, high levels of PG activity (units mg protein-1) were reported (Lazan et al. 1989; Chin et al. 1999; Abu-Bakr et al. 2003; Deng et al. 2005); increase in PG activity (units gFW-1) was also observed in bananas (Lohani et al. 2004). In other fruits such as strawberry and apple, PG activity were reported to be absent, but PG activity and/or mRNA expression were subsequently detected (Hadfield and Bennett 1998). Using transgenic methods, PG was shown to be a major player in regulation of pectin solubilisation in fruits. In ripening-impaired tomato fruits containing the rin mutation, reduced accumulation of PG mRNA was observed, in conjunction with reduced PG activity and pectin solubilisation (Della Penna et al. 1987, 1989; Seymour et al. 1987; Knapp et al. 1989). In the ‘rescue’ experiments, PG activity and the level of pectin solubilisation were almost similar to those of the wild-type (Della Penna et al. 1990). 14 2.3.3 Pectin methylesterase Pectin methylesterase (PME; EC 3.1.1.11) is an enzyme that contributes to the degradation of pectins (Minic and Jouanin 2006). It acts by catalysing the removal of methyl groups from the C6 position of GalA residues of high molecular weight pectins (Fischer and Bennett 1991; Brummell and Harpster 2001; Micheli 2001). The demethylesterification of pectins releases acidic pectins and methanol as products, and causes changes in the pH and charge of cell walls (Carpita and Gibeaut 1993; Stephenson and Hawes 1994; Micheli 2001). PME has been suggested to be an important regulator for various cell wall modifications related to pectin demethylesterification and pectin solubilisation, which had been previously described. In carambola, guava, grapes and boysenberry fruits, activity of PME (units gFW-1) increased throughout fruit development (Chin et al. 1999; Abu-Bakr et al. 2003; Deng et al. 2005; Vicente et al. 2007); in capsicum, banana and pawpaw fruits, activity of PME (units mg protein-1) demonstrated decreases upon full ripening (Sethu et al. 1996; Lohani et al. 2004; Koslanund et al. 2005). The mode of action of PME was previously described to be dependent on the pH of the PMEs. Acidic PMEs resulted in random demethylesterification of pectins, and alkaline PMEs resulted in linear (along the chain) demethylesterification of pectins (Markovic and Kohn 1984). However, more recent studies demonstrated that the action pattern of PMEs is much more complicated, and may be regulated by many factors such as pH and the degree of methylesterification of the pectins (Catoire et al. 1998; Denes et al. 2000). PME has been studied in great detail in tomato, and has been shown to consist of at least four genes, some of which are highly homologous (Harriman et al. 1991; Hall et al. 1994; Turner et al. 1996; Gaffe et al. 1997). During the ripening of tomatoes, PME protein and activity increased throughout and then declined slightly upon full ripening (Harriman et 15 al. 1991; Tieman et al. 1992). However, the accumulation of PME mRNA demonstrated an opposite trend, decreasing as ripening progressed, and it was suggested that the nonsynchronised patterns were due to the quantification of a composite of PME proteins of two or more highly homologous genes (Harriman et al. 1991; Brummell and Harpster 2001). 2.3.4 Exo-glycosidases There is a variety of glycosidases because of the structural and functional diversity of the polysaccharides and oligosaccharides (Davies and Henrissat 1995; Minic 2008). Some examples of exo-glycosidases include β-galactosidase (β-gal; EC 3.2.1.23), β-glucosidase (βglu; EC 3.2.1.21), β-mannosidase (β–man; EC 3.2.1.25) and β-xylosidase (β-xyl; EC 3.2.1.37), and all have been detected during fruit development in mango (Mangifera indica), capsicum, asparagus (Asparagus officinalis), carambola and boysenberry (Ali et al. 1995; Sethu et al. 1996; O’Donoghue et al. 1998; Chin et al. 1999; Vicente et al. 2007, Minic 2008). β-gal, which catalyses the removal of β-D-galactosyl residues from β-D-galactosides has been relatively well-studied (Brummell and Harpster 2001). The loss of Gal in cell walls is a common feature during fruit ripening, and it has been suggested that β-gal acts by hydrolyzing the Gal residues from the sidechains of pectins, causing changes in the structure of the pectin network (O’Donoghue et al. 1998; Brummell and Harpster 2001). β-gal exists in at least three isoforms in tomato and mango (Pressey 1983; Ali et al. 1995). In tomato, βgal is encoded by at least seven genes TBG1 – TBG 7 (Smith and Gross 2000). Transcripts of the seven genes exhibited differential expression during fruit development, and only TBG4 mRNA was significantly reduced in ripening-impaired rin and nor mutants (Smith and Gross 2000). 16 β-glu completes the breakdown of glucans by catalysing the hydrolysis of oligosaccharides to release glucose (Wood and Bhat 1988; Hrmova and Fincher 2001). In barley (Hordeum vulgare), β-glu demonstrates broad substrate specificity, hydrolysing βglucans, β-oligoglucosides and xyloglucan (Hrmova and Fincher 1998). Most studies on βglu focus on β-glucans as substrates, and little is known as to how β-glu participates in remodelling of the cell wall structure (Hrmova and Fincher 2001). It has been suggested that the ‘real’ substrate for β-glu is xyloglucan, thus affecting the cellulose-xyloglucan network (Rose and Bennett 1999; Hrmova and Fincher 2001). β-man is involved in the degradation of galactoglucomannans, by catalysing the removal of Gal on the sidechains of manno-oligosaccharides (Minic and Jouanin 2006). The activity of β-man in the cell walls of germinating seeds of monocotyledons, such as Phoenyx dactylifera, has been investigated and it was suggested that the enzyme is involved in the mobilization of galactoglucomannan in seeds (Buckeridge et al. 2000). β-xyl takes part in the degradation of xylan and arabinoxylan (Minic and Jouanin 2006). The enzyme is identified as a key enzyme for the complete breakdown of xylan by catalysing the hydrolysis of xylo-oligosaccharides from the non-reducing ends, and releasing xylose (Minic and Jouanin 2006; Minic 2008). β-xyl has also been shown to be involved in the degradation of arabinan side chains during hydrolysis of rhamnogalacturonan I (Minic et al. 2004; Minic and Jouanin 2006). Compared to endo-glycosidases (cellulase and PG) that break load bearing cross-links in the cell wall, exo-glycosidases appear to cause much less significant effects as they catalyse the removal of single glycosyl residues from polysaccharide chains (Rose and Bennett 1999; Hrmova and Fincher 2001). However, it has been proposed that exo- 17 glycosidases may still participate in cell wall remodelling as side chain removal increases the accessibility and availability of substrate sites for endo-acting enzymes (Rose and Bennett 1999). Also, actions of exo-glycosidases on side chains of xyloglucans may affect the binding affinity of xyloglucan to cellulose microfibrils, hence disrupting the cellulosexyloglucan network (Rose and Bennett 1999). 2.4 Flowers and their influences Flowering is a critical event in the life-cycle of angiosperms, allowing for the reproduction of these plants. A complete flower consists of sepals and petals (together, they form the perianth of the flower), gynoecium and androecium, which are the essential organs for sexual reproduction in the higher plants (Bernier et al. 1993; Burger 2006; O’Donoghue 2006). Biologically, sepals and petals play important roles by protecting immature reproductive structures, then providing attraction and accessibility required for pollination to occur (O’Donoghue 2006). Flowers of various plants are also highly prized objects of beauty, and can be considered as commercially valuable and luxurious commodities incorporated into everyday lives (Hughes 2000; O’Donoghue 2006). The perishable nature of flowers, however, poses as a constant setback for the horticulture industry (O’Donoghue 2006; Nell 2007). To help prolong postharvest life of flowers, there are a variety of chemical preservative treatments, including silver thiosulfate, 8-hydroxyquinoline citrate and/ or sucrose (Redman et al. 2002). However, cold storage/ precooling/ refrigeration remains as the most common and recognised form of postharvest treatment, as it effectively lowers the rate of plant metabolic processes and the rate of microbial growth (Geertsen 1990; Sun and Brosnan 1999, 2001; 18 Redman et al. 2002; van Meeteren 2007). Disadvantages of precooling techniques include its contribution to global warming, high energy demands, high costs, and the intolerance of certain species of flowers towards cold (Redman et al. 2002; van Meeteren 2007; Kim and Infante Ferreira 2008). The importance of flowers has also been readdressed in recent years, in conjunction with the world’s increasing interest in sustainability (Bartlett 1997; Espinosa et al. 2008; Spala et al. 2008). For example, to help minimize energy consumption of buildings, ‘green buildings’ or ‘green roofs’ have been developed. These structures make use of green plants to provide shade for the buildings, control temperature and humidity, mitigate the greenhouse effect, filter pollutants and mask noise, thus creating buildings that are energy efficient (Spala et al. 2008). For aesthetic reasons, flowers are important components, and flower physiology and flowering seasons have become essential criteria during plant selection (Spala et al. 2008). 2.5 Flowering physiology Flowering is considered a multifactorial process (Bernier 1988), comprising of two distinct processes: floral initiation (floral induction), referring to the transition from vegetative to reproductive development, and subsequent floral development, referring to the development from floral bud to mature floral bud to mature flower and finally to senesced flower (Hew and Yong 2004). In this thesis, the term ‘flowering’ will be used to define the processes involved in floral development, from floral bud till senesced flower stages. There are great variations among the angiosperms in the manner, timing, and physiology of flowering (van Doorn and van Meeteren 2003; O’Donoghue 2006). In tulips 19 (Tulipa genesriana), flower opening is due to movements of the petal lamina as mesophyll cells expand with temperature (Wood 1953). On the other hand, flower opening in Ipomoea is due to movements of the midrib rather than the petal lamina (Kaihara and Takimoto 1981). Flower opening in Portulaca occurs in the day (Ichimura and Suto 1998), while flower opening in Oenothera lamarkiana occurs at night (Saito and Yamaki 1967). Physiologically, it had been shown that carbohydrate metabolism (particularly the breakdown of storage carbohydrates into soluble sugars) and water relations affecting cell wall turgor are involved in regulating flowering in many species of flowers, such as alstromeria (Alstromeria peregrine) (Collier 1997), rose (Rosa) (Evans and Reid 1988), daylily (Hemerocallis spp.) (Bieleski 1993), lily (Lilium hybrid) (Bieleski et al. 2000), Campanula rapunculoides (Vergauwen et al. 2000) and Capparis spinosa L. (Rhizopoulou et al. 2006). Hormonal regulation is another influential mechanism of flowering. For example, supplementing tulip flowers with gibberellin4+7 (GA4+7) plus benzyladenine (BA) helped to promote longevity of the flowers (Kim and Miller 2008), exogenous ethylene caused delayed flower opening in rose (Tan et al. 2006) and exogenous abscisic acid (ABA) resulted in premature senescing of daylily flowers (Panavas et al. 1998). Changes in petal cell walls have also been suggested to be involved in controlling flowering (van Doorn and van Meeteren 2003; O’Donoghue 2006). Information on this topic is, however, very much limited when compared to the other physiological processes mentioned above. 2.5.1 Cell wall changes during flowering Only a few species of flowers have been studied on the influence of cell wall changes in regulating flowering, and these include carnation (Dianthus caryophyllus L.), sandersonia (Sandersonia aurantiaca Hook.) and daylily. Cell wall changes that occur in the flower 20 petals include alterations in cell wall compositions, activities of cell wall enzymes, and expression of cell wall-related genes (O’Donoghue 2006). Modifications in petal cell walls are necessary to provide the flexibility required during dramatic floral bud and flower growth (van Doorn and van Meeteren 2003; O’Donoghue et al. 2005). Comparison between the flowers shows certain similarities and differences in cell wall changes associated with flowering. In carnation flowers, full flower opening was accompanied by increases in contents of cell wall cellulose, total pectins, chelator-soluble pectins, carbonate-soluble pectins and neutral sugars. Upon senescence of the flowers, all of these cell wall components, except chelator-soluble pectins, decreased in content (de Vetten and Huber 1990; de Vetten et al. 1991). While PG activity was absent in opened and senesced flowers, activities of β-glu and β-gal were detected in senescing flowers (de Vetten et al. 1991). As with carnation flowers, sandersonia flowers also exhibited increase in contents of cellulose, total pectins and neutral sugars during flower opening, and with the exception of neutral sugars, the contents increased further during senescence (O’Donoghue et al. 2002). Levels of chelator-soluble and carbonate-soluble pectins remained unchanged during flower opening, and subsequently decreased upon senescence of the flowers. Similar to carnation flowers, PG activity was also not detected in sandersonia flowers. Cellulase activity was also reported to be absent in sandersonia flowers. The level of β-gal activity was unchanged during flower opening in sandersonia flowers, and subsequently, increased during flower senescence. In the same study, PME activity was found to increase during flower opening and decreased upon flower senescence (O’Donoghue et al. 2002). Three genes that putatively encode β-gal (SaGAL1, SaGAL2, and SaGAL3) have also been reported, and it was found 21 that all three genes were expressed during the onset of flower senescence (O’Donoghue et al. 2005). Analysis of cell wall composition was not reported for daylily flowers, but analyses on cell wall enzyme activities of daylily flowers indicated the participation of cell wall changes during flower opening. Cellulase and PME activities increased during floral bud development, and decreased upon senescence (Panavas et al. 1998). Unlike carnation and sandersonia flowers, PG activity was detected in daylily flowers and was reported to increase during senescence (Panavas et al. 1998). In Arabidopsis flowers, although cell wall compositional changes were not reported, cell wall enzyme activity assays and gene expression studies supported that cell wall modifications were regular features of flower opening (O’Donoghue 2006). PME activities were detected in mature Arabidopsis flowers, and at least one PME gene was shown to be strongly expressed in the flowers (Micheli et al. 1998; Francis et al. 2006). Five PG genes and 14 PME genes were reported to be differentially expressed in floral bud clusters (Imoto et al. 2005). 2.5.2 Floral bud opening, flower longevity and their regulators There are various endogenous regulators that control flower opening, and hormones are one example (van Doorn and van Meeteren 2003). Some commonly investigated hormones or plant growth regulators (PGRs) include the cytokinin, benzyladenine (BA), and gibberellins (GAs) (Bernier 1988). Although various studies have been conducted to determine the effects of BA and GA on flowering, generalizations of the effects of the PGRs cannot be applied across for all plant species because some of the PGRs are present in supra- or sub-optimal amounts (Cleland 1982). 22 Exogenous applications of BA have been shown to suppress flower opening, flower wilting and senescence in roses, petunia and Grevillea (Mor et al. 1983; Lukaszewska et al. 1994; Taverner et al. 1999; Setyadjit et al. 2004), while it promoted flower senescence in carnations (Woodson and Brandt 1991). The effects of cytokinins on plant development are often dependent on the presence or the absence of other PGRs (Bernier 1988). For example, GA enhanced the stimulatory effects of BA on inducing floral transition in Dendrobium hybrids (Goh 1979; Bernier 1988). GA alone has been shown to promote flower opening in Ipomoea nil and iris (Iris pseudacorus) (Raab and Koning 1987; Celikel and van Doorn 1995), and prolong longevity in alstromeria and carnation (Saks and van Staden 1993; Jordi et al. 1995). However, GA was shown to have no effects on longevity of Grevillea and even increased flower abscission (Setyadjit et al. 2006). Other chemicals that had been shown to control flower opening include aminooxyacetic acid (AOA), an inhibitor of ethylene synthesis (Rattanawisalanon et al. 2003). Treatment of Dendrobium ‘Jew Yuay Tew’ inflorescences with AOA suppressed bud drop, promoted bud opening, and delayed flower senescence (Rattanawisalanon et al. 2003). Longevity of Dendrobium ‘Heang Beauty’ flowers was also pro-longed upon treatment with AOA (Chandran et al. 2006). It was proposed that AOA could act as an anti-microbial agent, which inhibited bacterial growth and hence, allowing continuous uptake of water and sugars by the flowers (Rattanawisalanon et al. 2003; Chandran et al. 2006). 2.5.3 Flowering and the senescence programme There are suggestions that flowering might be regulated by a senescence programme of the plant and/ or floral organs (Rubinstein 2000; O’Donoghue et al. 2002; Wagstaff et al. 2003). Senescence refers to the terminal phase in the development of leaves and flowers and 23 is often accompanied by events such as protein remobilization, increased proteinase activities, DNA laddering, membrane degradation, cell wall alterations and nuclear shrinkage, all of which are also characteristics of programmed cell death (PCD) (Rubinstein 2000; Wagstaff et al. 2003). In Ipomoea, dynamic structural changes such as cell enlargement, modification of cell shape and reduction in cell wall thickness occurred in the inner epidermal cells even before flower opening (Phillips and Kende 1980). In sandersonia, intercellular air spaces and increasingly disorganized packing of parenchyma cells also occurred prior to flower opening (O’Donoghue et al. 2002). In alstromeria, DNA laddering, nuclear shrinkage and increase in expression of cysteine protease all commenced as early as two days before flower opening (Wagstaff et al. 2003). The occurrences of indicators of PCD before flower opening imply that flower senescence is a continuum from a senescence programme that had already started, and suggest that flower opening might somehow be a consequence of this senescence programme (O’Donoghue et al. 2002; Wagstaff et al. 2003). 2.6 Orchids Orchids originated from lily-like ancestors which have either evolved into orchids or become extinct (Seidenfaden and Wood 1992). The classification of orchids is as such: superorder Lilianae, order Orchidales, and family Orchidaceae (Seidenfaden and Wood 1992). The column and the lip (or labellum) are the hallmarks of the orchids (Teoh 2005). Due to their vividly coloured lips that are often embellished with crests, hair, ribs and other protuberances, orchids are very appealing and are among the highly demanded cut flowers, 24 appreciated for their beauty and fragrance (Hew and Yong 2004; Teoh 2005; Chandran et al. 2006). 2.6.1 Dendrobium crumenatum Dendrobium crumenatum (Swartz), also known as the pigeon orchid, is a common native epiphytic orchid species of South-east Asia, naturally occurring in Singapore and Malaysia (Fig. 1). It exhibits an interesting diversion of the normal flowering process: upon transition of the meristem from a vegetative to a reproductive phase, floral buds develop to a certain stage and then become ‘dormant’. These floral buds resume growth and development after cold-induction, such as after a heavy rainfall, and culminating into the opening of the flowers exactly nine days after (Holttum 1953; Corner 1988). Full flower opening is achieved before dawn, and the white flowers are small (40 mm width) with yellow ridges running from the centre of the lip to its base, providing the only colour in the flowers (Fig. 1E insets). The flowers are short-lived, lasting only for a day before the onset of senescence (Tan and Hew 1993). The synchronized flowering of the pigeon orchid is an impressive sight, and the cultivation of the orchid along roadside trees for aesthetic reasons has been attempted by the National Parks Board of Singapore (Boo et al. 2006). 25 Fig. 1. Dendrobium crumenatum (pigeon orchid). An inflorescence stalk bearing floral buds of day 6 (A), day 9 (B) and day 10, i.e. fully opened flowers (C), after cold induction. (D) Synchronous flowering in nature. 26 Chapter 3. Materials and Methods 3.1 Plant material Plants of Dendrobium crumenatum (Swartz) were maintained under cool and partially shaded conditions (PAR ranged from 100 – 250 µmol m-2 s-1; average air temperature ranged from 25 – 33°C) in a planthouse of the Department of Biological Sciences, National University of Singapore. Plants were watered daily, and fertilized weekly with a foliar fertilizer (N:P:K = 18:36:18). Pots of D. crumenatum with inducible inflorescences carrying dormant floral buds were acclimatized at 30°C for 24 h in temperature-controlled growth chambers. They were then subjected to a cold induction at 20°C for 24 h. Growth chambers were maintained on a 12 h day/ 12 h night cycle and illumination was provided by fluorescent tubes (PAR ranged from 10 – 20 μmol m-2 s-1). Plants of D. crumenatum exhibit crassulacean acid metabolism, demonstrating different carbon dioxide exchange patterns during different times of the day. Thus, all plants were moved into the growth chambers at 1600 h, to minimize the effects of any possible temporal variations in the plant physiology. Plants were also watered daily to minimize dehydration stress. Floral buds or flowers were selected according to their age and features (Table 1, Fig. 5). For cell wall composition, cell wall enzyme activities and molecular studies, sepals and petals of the harvested floral buds and flowers were separated (Fig. 2) and stored at -80°C until use. Fresh samples were used for all other analyses. 3.2 Physical parameters analyses Freshly harvested sepals and petals from each flower were weighed to obtain fresh weight, then wrapped in aluminium foil and left to dry in an 80°C oven for 1 week for dry weight 27 Table 1. Stages of floral bud development in D. crumenatum. Timing of events is reported in relation to the time during which floral buds were subjected to cold induction (denoted as day 0). Features Exposure of dormant floral buds to cold induction at 20°C for 24 h. Time (days after induction) 0 Green bud (ca. 1 cm long) with reddish brown tinges along ventral side, elongation of mentum, mentum reddish brown. 4 Light green bud (ca. 2.5 cm long), reddish brown tinges only at beginning and tip of mentum, further elongation of mentum, length of mentum almost half of length of whole bud. 7 White bud (ca. 3 cm long), no splitting of sepals, elongated mentum pointing downwards away from tip of bud. 9 Full flower opening, sepals and petals fully expanded, lip fully protruded with visible yellow ridges running down from midlobe to foot of column. 10 Sepals and petals shrivelled and brownish. 12 28 Fig. 2. Separation of floral parts in D. crumenatum floral buds and flowers. (A) Day 7 floral bud after dissection, (B) sepals of day 7 floral bud, (C) petals of day 7 floral bud, (D) column of day 7 floral bud, (E) opened flower on day 10, (F) sepals of opened flower, (G) petals of opened flower, (H) column of opened flowers. Scale bar = 1 cm. 29 (amount of dry matter) determination. Water content was obtained by subtracting the amount of dry matter from total fresh weight. 3.3 Microscopy Segments of approximately 1 cm × 0.5 cm were cut about halfway from the tips of freshly harvested sepals or petals, to include the central vein. The fresh tissues were fixed in 96 % ethanol : acetic acid : formalin : water (10 : 1 : 2: 7 by volume) for 24 h. Fixed samples were then subjected to dehydration through a graded tertiary-butanol (TBA) series (water : 95 % ethanol : TBA) as follows: Grade 1 (4 parts of water : 5 parts of 95 % ethanol : 1 part of TBA), 4 h at room temperature (22°C – 24°C) ; Grade 2 (3 : 5 : 2), 4 h at room temperature; Grade 3 (1.5 : 5 : 3.5), 24 h at room temperature; Grade 4 (0 : 5 : 5), 4 h at room temperature; Grade 5 (0 : 2.5 100 % ethanol : 7.5) with Orange-G (BDH) dye, 4 h at room temperature, repeated once for 24 h; Grade 6 (0 : 0 : 10), 4 h at 40°C, repeated once. Dehydrated samples were infiltrated with melted paraffin wax over 4 days at 60°C, then embedded in paraffin wax. Embedded tissues were sectioned with a microtome to a thickness of 10 µm and sections were mounted onto albumin-coated slides. The slides were dewaxed by Histoclear 1, 2 and 3 for 3 min each, and were hydrated through a series of decreasing concentration of ethanol: 100 %, 95 %, 90 %, 70 %, 50 % and 0 % for 2 min each. The hydrated sections were finally stained with toluidine blue (0.1 % in 0.1 M sodium phosphate buffer, pH 5.5) and examined using light microscopy. The thickness of sepals or petals was obtained by measuring the thickness of the area between the middle two vascular bundles of each section. 30 3.4 Electrolyte leakage Electrolyte leakage was determined by the method described by Tuna et al. (2007) with some modifications. Twenty discs (5 mm diameter) from freshly harvested sepals or petals were obtained using a cork borer, and were equilibrated in 20 ml of water at room temperature for 1 h. Subsequently, electrical conductivity of the solution (C0) was measured using a conductance meter (WTW Cond 315i/SET). Samples were boiled for 30 min, and electrical conductivity of the solutions (C1) were measured after they cooled to room temperature. The percentage of electrolyte leakage of sepals/ petals was calculated as follows: Electrolyte leakage (%) = (C0/ C1) × 100 Membrane stability index (MSI) of sepals/ petals was calculated as follows: MSI = [1 - (C0/ C1)] × 100 3.5 Cell wall compositional analyses 3.5.1 Preparation of EIR Cell wall materials of sepals or petals were prepared by using a modified method described by Huber (1992). Frozen tissues were homogenized in 95 % ethanol at 4°C and then chilled at -20°C for 24 h. The homogenates were centrifuged at 8,000 g for 10 min at 4°C. Trisbuffered phenol (5 ml per g FW of sepal/ petal tissues) was added to the residues and allowed to incubate at room temperature for 45 min. The suspensions were centrifuged as described above, and the residues were re-suspended in 80 % ethanol at -20°C for 2 h, followed by centrifugation. The remaining residues were washed once with 80 % ethanol, followed by 80 % acetone, and then chloroform : methanol (1 : 1) mixture. All organic washings were conducted at room temperature. The final residues were recovered by filtration, washed 31 thrice with acetone until total whitening, yielding the crude cell wall material (ethanolinsoluble residue, EIR). The EIR was air-dried, weighed, and then stored at -80°C until use. The amount of EIR was expressed as mg per gram fresh weight and mg per floral part (sepal or petal). 3.5.2 Cellulose content Cellulose was extracted from 5 mg of EIR following the procedure described by Updegraff (1969). Initial hydrolysis was carried out in 5 ml of acetic-nitric acid reagent (10 volumes of 80 % acetic acid : 1 volume of 16N nitric acid) at 100°C for 30 min. Samples were centrifuged at 4,000 g for 5 min at room temperature, and the supernatants discarded. The remaining residues were washed with 10 ml of water, centrifuged as described above, and the supernatants were discarded. The remaining residues were hydrolysed in 10 ml of 67 % sulphuric acid for 1 h at room temperature. The resultant solution was diluted 10 times with water and was analysed for cellulose content. Cellulose contents were assayed by the anthrone method (Scott and Melvin 1953). Four ml of water were added to 1 ml of sample in an ice-water bath and then 10 ml of chilled anthrone reagent (Sigma) were added. The sample was mixed, and then incubated at 100°C for 15 min. The reaction was stopped by immediately returning the sample to an ice-water bath. The bluish-green colour of the mixture was allowed to develop at room temperature for 10 min and the absorbance of the mixture at 620 nm was measured. Cellulose (0 – 0.2 mg ml1 , Sigma) was used as the standard in the assays. Cellulose concentration was expressed as mg per gram fresh weight and μg per floral part (sepal or petal). 32 3.5.3 Total pectin content Total pectins in EIR (5 mg) were extracted as described by Ahmed and Labavitch (1977). Two ml of chilled, concentrated sulphuric acid were added to the samples and allowed to incubate for 5 min. This was followed by the addition of 0.5 ml of water to the mixture and incubated for 5 min. Another 0.5 ml of water were added until dissolution of the materials was completed. The whole extraction procedure was conducted in an ice-water bath with constant gentle swirling. The hydrolysed sample was made up to a total of 10 ml with water and the resultant solution was analysed for total pectin content. Pectin contents were expressed as uronic acid equivalents and were determined by the m-hydroxydiphenyl method (Blumenkrantz and Asboe-Hansen 1973). Six ml of chilled 0.0125 M sodium tetraborate-sulfuric acid solution (Sigma) were added to 1 ml of sample on an ice-water bath. The mixture was mixed well, and then incubated at 100°C for 6 min. The reaction was terminated by immediately returning the mixture to an ice-water bath. This was followed by the addition of 0.1 ml of m-hydroxydiphenyl (0.15 % w/v, Fluka) to the sample. The pink colour of the mixture was allowed to develop for 15 min at room temperature and the absorbance at 520 nm was measured. Galacturonic acid (0 – 0.2 mg ml-1, Fluka) was used as the standard in the assays. Total pectin concentration was expressed mg per gram fresh weight and μg per floral part (sepal or petal). 3.5.4 Soluble pectin and hemicellulose contents Soluble pectins and hemicelluloses were sequentially extracted from EIR (Fig. 3). The extractions were carried out in water (water-soluble pectins), 50 mM sodium acetate buffer (pH 6.0) containing 50 mM CDTA (CDTA-soluble pectins), 50 mM Na2CO3 containing 20 mM NaBH4 (Na2CO3-soluble pectins), and finally 6 N NaOH containing 0.13 mM NaBH4 33 Fig. 3. Sequential extraction of pectins and hemicelluloses from EIR of D. crumenatum floral buds. 34 (hemicelluloses). The proportion of EIR to extractant was 1 mg : 0.5 ml. Each extraction was conducted at room temperature for 2 h with constant shaking, and centrifuged at 10,000 g for 15 min at 4°C (O’Donoghue et al. 2002; Deng et al. 2005). The various supernatants were analysed for their respective contents. Hemicellulose contents were assayed by the anthrone method (Scott and Melvin 1953) described above and pectin contents were assayed by the the m-hydroxydiphenyl method (Blumenkrantz and Asboe-Hansen 1973) described above. Hemicellulose and soluble pectin concentrations were expressed as mg per gram fresh weight and μg per floral part (sepal or petal). 3.6 Cell wall enzyme analyses 3.6.1 Enzyme extraction Sepals and petals were ground to fine powder in liquid nitrogen with a mortar and pestle. The extraction buffer was then added and the materials were homogenized further. All enzyme extractions were carried out at 4°C. The homogenates were centrifuged at 12,000 g for 20 min at 4°C, and the supernatants were used to determine enzymatic activities. For cellulase (EC 3.2.1.4) and polygalacturonase (PG; EC 3.2.1.15), two extraction methods (Method A and B) were attempted on both fresh and frozen plant materials, as summarized in Table 2. Method A was a 1-step procedure extracting salt-soluble cellulase and PG with 20 mM citrate buffer (pH 5.1) containing 1 M NaCl, or 20 mM sodium phosphate buffer (pH 6.1 or pH 7) containing 1 M NaCl (Panavas et al. 1998). The extractions were conducted over 30 min. Method B was a 4-step procedure extracting both buffer-soluble and salt-soluble cellulase and PG (Ferrari and Arnison 1974; Huberman et al. 35 Table 2. Extraction procedures of cellulase and PG from floral buds of D. crumenatum. Method A was a 1-step procedure (Panavas et al. 1998), while Method B was a 4-step procedure (Ferrari and Arnison 1974; Huberman et al. 1975). The type of supernatants collected from each step of extraction is included. Method Extraction procedure Supernatant A 1. Tissue homogenized in either - 20 mM citrate buffer (pH 5.1) + 1 M NaCl or - 20 mM sodium phosphate buffer (pH 6.1) + 1 M NaCl or - 20 mM sodium phosphate buffer (pH 7) + 1 M NaCl Salt-soluble cellulase and PG B 1. Tissue homogenized in either - 20 mM citrate buffer (pH 5.1) or - 20 mM sodium phosphate buffer (pH 6.1) or - 20 mM sodium phosphate buffer (pH 7) Buffersoluble cellulase and PG/ 2. Residue washed with respective homogenizing buffer. 3. Residue washed with respective homogenizing buffer. 4. Residue resuspended in respective homogenizing buffer + 1 M NaCl for 30 min. Salt-soluble cellulase and PG 36 1975). The materials were homogenized for 30 min in 20 mM citrate buffer (pH 5.1) or 20 mM sodium phosphate buffer (pH 6.1 or pH 7). The homogenates were centrifuged to obtain the supernatants (buffer-soluble enzyme) and the remaining residues were washed twice with the respective homogenizing buffers and, after centrifugation, the supernatants were discarded. The final remaining residues were resuspended in the respective buffers with 1 M NaCl for 30 min, and centrifuged to obtain the supernatants (salt-soluble enzyme) (Ferrari and Arnison 1974; Huberman et al. 1975). The ratio of plant material to extractant ranged between 1g : 10 ml to 1g : 1 ml. Aliquots of the various supernatants were desalted on Sephadex G-25 columns (1 × 10 cm). All extracts (desalted and crude) were tested for activities of both cellulase and PG. For the extractions of pectin methylesterase (PME; EC 3.1.1.11), β-galactosidase (βgal; EC 3.2.1.23), β-glucosidase (β-glu; EC 3.2.1.21), β-mannosidase (β –man; EC 3.2.1.25) and β-xylosidase (β –xyl; EC 3.2.1.37), frozen plant materials were used. Plant materials were homogenized in 20 mM sodium phosphate (pH 7.5) containing 1.5 M NaCl for 30 min. The ratio of plant material to extractant was 1 g : 10 ml. The supernatants were recovered by centrifugation and the crude extracts were used for the various enzymatic assays. 3.6.2 Cellulase assay A gel diffusion assay (Wood et al. 1988) was adopted to screen for cellulase activity. Gels, consisting of 1.7 % agarose and 0.5 % carboxymethylcellulose (CMC; medium viscosity, Sigma) in 0.2 M citrate buffer (pH 5.1) or 0.2 M sodium phosphate buffer (pH 6.1 or 7), were prepared in Petri-dishes (diameter = 9 cm) to a thickness of approximately 4 mm. Wells of 5 mm diameter were cut using a cork borer, and the cellulase extracts (25 μl) were pipetted into the wells. The Petri-dishes were sealed with parafilm and were incubated at room 37 temperature or 37°C for 2 h or 24 h. The gels were then stained by covering the plates with 1 mg ml-1congo red solution (BDH) for 15 min, followed by 1 M NaCl for 10 min. The NaCl wash was repeated twice. Stained plates were preserved for later observations by flooding with 5 % acetic acid. Commercial cellulase (0.5 mg ml-1, Yakult Honsha) and boiled enzyme extracts were included as experimental controls. Cellulase activity was also assayed by spectrophotometric methods adapted from Ghose (1987). Using CMC as the enzyme substrate, the reaction mixture contained 0.4 ml of 0.05 % CMC, 0.2 ml of 0.1 M sodium acetate buffer (pH 5) and 400 μl of enzyme extract. Using filter paper as the enzyme substrate, the reaction mixture contained a filter paper strip (1 cm × 1 cm, Whatman No. 1), 0.6 ml of 0.1 M sodium acetate buffer (pH 5) and 400 μl of enzyme extract. Incubation was carried out at 37°C for 1 h or 24 h. In control tubes, boiled enzyme extracts were used. The amounts of reducing groups released from CMC and filter paper strips were determined by the dinitrosalicylic acid (DNS) and glucose oxidase/оdianisidine (GOD) methods (Ghose 1987). In the DNS method, 2 ml of DNS reagent, consisting of 0.02 M DNS (Sigma), 2 N NaOH and 1 M Rochelle salt (Sigma), were added to the reaction mixture, and then heated in a boiling water bath for 5 min. The mixtures were immediately placed in a ice-water bath for 5 min after the heat treatment, and 9 ml of water were added to each tube. The absorbance at 540 nm was measured. In the GOD method, 2 ml of assay reagent (glucose oxidase/о-dianisidine, Sigma) were added to 0.5 ml of reaction mixture, and incubated at 37°C for 30 min. The reactions were stopped by the addition of 2 ml of 12 N H2SO4. The absorbance at 540 nm was measured. Glucose (0 – 2 mg ml-1 for the DNS assay, 0 – 0.16 mg ml-1 for the GOD assay) was used as the standard in these assays. One unit of cellulase activity represented 1 μmol of reducing groups liberated per h. 38 Cellulase activity was expressed as units per mg protein, units per floral part (sepal or petal) and units per g fresh weight. 3.6.3 Polygalacturonase assay Polygalacturonase (PG) activity was determined by an adapted method described by AbuBakr et al. (2003). The reaction mixture, containing 0.4 ml of 0.1 % polygalacturonic acid solution (Fluka), 0.2 ml of 0.1 M sodium acetate buffer (pH 5) and 400 μl of enzyme extract, was incubated at 37°C for 1 h or 24 h. In control tubes, boiled enzyme extracts were used. The amount of reducing groups released from polygalacturonic acid was determined by the DNS and GOD methods as described above. Galacturonic acid (0 – 2 mg ml-1 for the DNS assay, 0 – 0.16 mg ml-1 for the GOD assay) was used as the standard in these assays. One unit of PG activity represented 1 μmol of reducing groups liberated per h. PG activity was expressed as units per mg protein, units per floral part (sepal or petal) and units per gram fresh weight. 3.6.4 Pectin methylesterase assay Pectin methylesterase (PME) activity was assayed by a continuous spectrophotometric method as described by Hagerman and Austin (1986) with some modifications. The reaction mixture contained 2 ml of 0.5 % citrus pectin solution (degree of esterification = ~60 %, Sigma), 0.15 ml of 0.01 % bromothymol blue (Sigma) solution in 3 mM potassium phosphate buffer (pH 7.5), 0.75 ml of water and 100 μl of enzyme extract. The pHs of the solutions were adjusted to pH 7.5 each time before use. After adding the enzyme extract, the reaction mixture was mixed well and allowed to stabilize for 1 min at room temperature before measuring the absorbance at 620 nm. The absorbance was again determined after 21 min of incubation at room temperature. The difference in absorbance was the measure of 39 PME activity and was calibrated against a galacturonic acid standard curve (0 – 0.17 mg ml1 ). One unit of PME activity represented 1 µmol of methylester liberated per h. Boiled enzyme extracts were included as controls. PME activity was expressed as units per mg protein, units per floral part (sepal or petal) and units per gram fresh weight. 3.6.5 Glycosidases assay β-galactosidase (β-gal) , β-glucosidase (β-glu), β-mannosidase (β –man) and β-xylosidase (β –xyl) activities were assayed by an adapted method described by Chin et al. (1999). The reaction mixtures consisted of 0.5 ml 5 mM p-nitrophenyl derivatives of β-Dgalactopyranoside, β-D-glucopyranoside, β-D-mannopyronoside or β-D-xylopyranoside (Sigma) as substrate, 50 mM sodium acetate buffer (pH 4.5) and 50 μl of enzyme extract in a total volume of 2 ml. After incubation at 37°C for 30 min, the reactions were stopped by adding 1 ml of 0.2 M Na2CO3, and the amount of p-nitrophenol formed was determined spectrophotometrically at 415 nm. One unit of glycosidase activity represented 1 μmol of pnitrophenol released per h. Boiled enzyme extracts were included as controls. Glycosidase activity was expressed as units per mg protein, units per floral part (sepal or petal) and units per gram fresh weight. 3.6.6 Soluble protein content Soluble protein concentration was determined by the Bradford method (1976) with bovine serum albumin (0 – 0.9 mg ml-1) as the standard. Three ml of 20 % (v/v) Bio-Rad reagent were added to 60 μl of extract. The reaction mixture was incubated at room temperature for 20 min. The absorbance at 595 nm was then measured. Soluble protein concentration was expressed as mg per gram fresh weight and μg per floral part (sepal or petal). 40 3.7 Gene expression profiling All reagents, centrifuge tubes and pipette tips were autoclaved at 121°C for 60 min and mortars/ pestles were baked at 180°C for at least 16 h to ensure that they were RNase free. 3.7.1 Total RNA isolation Total RNA was extracted from frozen sepals and petals. Samples were ground in a mortar and a pestle in liquid nitrogen and Trizol reagent (Invitrogen) was added (1 ml per 0.1 g fresh weight of plant materials). The samples were further ground into fine powder in liquid nitrogen and allowed to thaw at room temperature. The samples were then centrifuged at 12,000 g for 10 min at 4°C. The resultant supernatant was transferred to a new centrifuge tube and chloroform (0.2 ml per 1 ml Trizol) was added. This was followed by vortexing at high speed for 15 s, and incubation at room temperature for 4 min. The samples were centrifuged at 14,000 g for 15 min at 4°C, and the colourless aqueous phases were transferred to new centrifuge tubes. The chloroform extraction was repeated once. To the final aqueous phase, isopropanol and 0.8 M sodium citrate/1.2 M NaCl were added (0.5 ml each per 1 ml of aqueous phase), and mixed by gentle inversion. The samples were incubated on ice for 20 min, and then centrifuged at 14,000 g for 20 min at 4°C. The supernatants were discarded, and pre-cooled 75 % ethanol (same volume at Trizol) was added. Samples were vortexed briefly then centrifuged again at 7800 g for 10 min at 4°C. The supernatants were discarded and the resulting pellets were dried under vacuum for 5 min. The dried pellets were solubilized in RNase-free water and an equal volume of 8 M lithium chloride was added. The samples were mixed, and then incubated at -80°C overnight. The frozen samples were allowed to thaw on ice, and then centrifuged at 12,000 g for 20 min at 4°C. The supernatants were discarded and the remaining residues were washed 41 with pre-cooled 75 % ethanol (same volume as Trizol). Samples were vortexed briefly, and then centrifuged at 7800 g for 10 min at 4°C. The supernatants were discarded and the resulting pellets were dried under vacuum for 5 min. The dried pellets were solubilized in RNase-free water and then stored at -80°C. 3.7.2 Estimation of RNA quality and quantity The amount of RNA (ng/μl) was quantified spectrophotometrically. Purity of RNA was determined by the absorbance ratios A260/280 and A260/230, which are measures of contamination by proteins and polyphenols/ carbohydrates respectively. The integrity of RNA was also verified by analyzing approximately 1 μg RNA sample on 0.8 % agarose gels. 3.7.3 Reverse transcription First-strand cDNA synthesis was conducted using the SuperScript First-Strand Synthesis System for RT-PCR kit (Invitrogen). Approximately 100 ng of RNA were added to 1 μl of 10 mM dNTP, 1 μl of oligodT (0.5 ng/μl) and DEPC-treated water to a total volume of 10 μl. The samples were incubated at 65°C for 5 min, then incubated at 4°C for 2 min. This was followed by the addition of 2 μl of 10x RT buffer, 4 μl of 25 mM MgCl2, 2 μl of 0.1 M DTT and 1 μl of RNaseOUT Recombincat RNase Inhibitor. The mixtures were mixed, and then incubated at 42°C for 2 min. One μl (50 units) of SuperScript II RT was then added to each sample, mixed, and incubated at 42°C for 50 min. The reactions were terminated at 70°C for 15 min, and then chilled at 4°C. The final cDNA obtained was stored at -20°C. 42 3.7.4 PCR amplification The primers used for PCR are as follows: β-TUB (β-tubF, 5’-CGTAAGGAAGCTGAGAACTGTGATTGC-3’, and β-tubR, 5’- GCAAGAAAGCTTTACGCCTGAACATAG-3’); β-GAL (β-galF, 5’- CCTATGTGTTCTGGAACGGGC-3’, and β-galR, 5’-CATCTTCCTTGCACATGACCCATGG-3’); PME (PMEF, 5’-GCACCGTCGACTTCATCTTC-3’, and PMER, 5’GGCATATACCCCTCAGG-3’). PCR amplification was performed using the GoTAQ Flexi DNA Polymerase (Promega). Gradient PCR was performed at various temperatures to determine the optimum annealing temperatures for the respective pairs of primers. PCR amplification was also conducted over 24, 26, 28, 30 and 32 cycles to determine the plateau phase for each gene. The final optimized PCR conditions were as follows: 95°C for 5 min, 30 cycles of 94°C for 30 s, 52°C for 1 min, 72°C for 1 min and 72°C for 5 min. The amplified samples were analysed by gel electrophoresis on 1.2 % agarose gel containing GelRed stain (8 μl per 100 ml agarose, Biotium Inc.). 3.8 Controlling time of flower opening in D. crumenatum 3.8.1 General setup Inflorescences of D. crumenatum bearing mature floral buds were harvested, and each floral bud was then cut such that it included the stalk of the bud and a short section of the inflorescence stem. Individual buds were placed in 2 ml centrifuge tubes containing the test solution. The centrifuge tubes holding the floral buds were then placed into GA7 containers, and were sealed with parafilm. Holes were made in the parafilm to facilitate transpiration, and to prevent condensation in the GA7 containers. The GA7 containers were then placed on a culture rack under a 16 h light/ 8 h dark cycle (PAR = 32 μmol m-2 s-1) at room 43 temperature. The floral buds were examined for signs of flower opening over the next few days. 3.8.2 Treatment The excised floral buds were treated with different concentrations of benzyladenine (BA, Sigma) (10-6 M, 10-8 M, 10-10 M), gibberellic acid (GA3, Sigma) (10-6 M, 10-8 M, 10-10 M), a combination of BA/ GA3 (10-6 M, 10-8 M, 10-10 M) and aminooxyacetic acid (AOA, Sigma) (10-6 M, 10-8 M, in the presence and absence of 0.1 M glucose). Water was included as a control in the study. 3.9 Statistical analysis The results were presented as mean ± standard error. The results were analysed for statistical significance by multifactor ANOVA. Where ANOVA detected significant differences, Fisher’s least significance difference (LSD) test was used at a 5 % level of significance. 44 Chapter 4. Results 4.1 Growth changes Sepals and petals of D. crumenatum demonstrated similar patterns of weight changes (Fig. 4). Total fresh weight (Fig. 4A) increased during floral bud development (Phase I); it remained constant during flower opening (Phase II) and decreased upon flower senescence (Phase III). The increases in water content during floral bud development (Phase I) followed an exponential pattern, with a dramatic increase occurring during the late stages of floral bud development on day 9 (Fig. 4B), whereas the increases in dry matter content during floral bud development were relatively constant (Fig. 4C). During flower opening (Phase II), while no changes were observed for water content, dry matter content decreased slightly (Fig. 4B, C). Upon flower senescence (Phase III), dry matter content remained constant (Fig. 4C), while water content decreased significantly (P < 0.05), accounting for the significant decreases (P < 0.05) in total fresh weight and FW : DW ratio (Fig. 4A, B, C). 4.2 Changes in anatomy during development Cross-sections of sepals and petals from young D. crumenatum floral buds (day 4) showed round parenchyma cells packed tightly and neatly between the upper and lower epidermes (Fig. 5A). During floral bud development (Phase I), the parenchyma cells were less tightly packed together, exhibiting some intercellular spaces (Fig. 5B, C). This layer became progressively disorganized during flower opening (Phase II), with larger intercellular spaces and membrane disintegration becoming apparent (Fig. 5D). During flower senescence (Phase 45 Phase I II III Fig. 4. Characteristics of weight changes of sepals and petals of D. crumenatum. (A) Fresh weight, (B) Water content, (C) Dry matter content and FW : DW ratio. Phases of development – I: floral bud development, II: flower opening, III: flower senescence. Each value is the mean ± SE (n = 5), and values with different alphabet(s) (sepals: a – d; petals: a’ – d’) are significantly different according to Fisher’s LSD test (P < 0.05). 46 Stage Sepal Petal I Phase II III Fig. 5. Changes in anatomical features of D. crumenatum during flower development. Developmental stages of floral buds/ flowers after cold induction: A, day 4; B, day 7; C, day 9; D, day 10; E, day 12 floral buds/ flowers. Features of development were as described in Table 1. Phases of development – I: floral bud development, II: flower opening, III: flower senescence. Transverse sections of sepals and petals were stained with toluidine blue. Arrows point to areas with parenchyma cell disruption and /or intercellular spaces. Anatomical studies were repeated four times on different floral buds or flowers. 47 III) on day 12, both sepals and petals showed almost complete absence of intact parenchyma cells, although the epidermal layers remained rather intact (Fig. 5E). The development of D. crumenatum floral buds was accompanied by marked changes in cell size (Fig. 6). The thickness of sepals and petals increased significantly (P < 0.05) during floral bud development (Phase I), reaching a maximum on day 9; it subsequently decreased during flower opening (Phase II) and flower senescence (Phase III) (Fig. 6A). The pattern of changes in sepals/ petal thickness corresponded with the changes in individual cell length. The lengths of parenchyma cells in sepals and petals were maximum during late floral bud development (day 9); it then decreased significantly (P < 0.05) upon full flower opening and senescence (Fig. 6C). Cell width of parenchyma cells, on the other hand, increased throughout development (Fig. 6B). 4.3 Cell wall composition 4.3.1 EIR Sepals and petals demonstrated similar patterns of changes in cellulose content throughout development. On a per floral part (sepal or petal) basis, the amount of EIR increased during floral bud development (Phase I) and flower opening (Phase II), from approximately 0.2 mg (floral part)-1 to 0.8 mg (floral part)-1 (Fig. 7A). A significant loss in EIR was observed, decreasing to approximately 0.4 mg (floral part)-1, during flower senescence (Phase III). On a per gram fresh weight basis, EIR content decreased significantly (P < 0.05) during floral bud development (Phase I); it then remained at low and constant levels throughout flower opening (Phase II) and flower senescence (Phase III) (Fig. 8A). 48 I Phase II III Fig. 6. Anatomical changes during D. crumenatum floral bud development. (A) Sepal and petal thickness measured between two middle vascular bundles, (B) width of parenchyma cells, (C) height of parenchyma cells. Phases of development – I: floral bud development, II: flower opening, III: flower senescence. Each value is mean ± SE (n = 5), and values with different alphabet(s) (sepals: a – c; petals: a’ – c’) are significantly different according to Fisher’s LSD test (P < 0.05). 49 I Phase II III I I Phase II Phase II III III Fig. 7. D. crumenatum sepal and petal cell wall components, expressed on a per floral part (sepal/ petal) basis, at each developmental stage. (A) EIR, (B) cellulose, (C hemicelluloses, (D pectins, (E) water-soluble pectins, (F) CDTA-soluble pectins, (G) Na2CO3-soluble pectins. Phases of development – I: floral bud development, II: flower opening, III: flower senescence. Each value is mean ± SE (n = 5) and values with different alphabet(s) (sepals: a – d; petals: a’ – d’) are significantly different according to Fisher’s LSD test (P < 0.05). 50 I Phase II III I I Phase II Phase II III III Fig. 8. D. crumenatum sepal and petal cell wall components, expressed on a per gram fresh weight (gFW-1) basis, at each developmental stage. (A) EIR, (B) cellulose, (C) hemicellulose, (D) pectin, (E) water-soluble pectin, (F) CDTA-soluble pectin, (G) Na2CO3-soluble pectin. Phases of development – I: floral bud development, II: flower opening, III: flower senescence. Each value is mean ± SE (n = 5) and values with different alphabet(s) (sepals: a – c; petals: a’ – c’) are significantly different according to Fisher’s LSD test (P < 0.05). 51 4.3.2 Cellulose Sepals and petals demonstrated similar patterns of changes in cellulose content throughout development. On a per floral part (sepal or petal) basis, cellulose content increased significantly (P < 0.05) during floral bud development (Phase I) from approximately 5 μg (floral part)-1 to 20 μg (floral part)-1; it continued to increase during flower opening (Phase II) to approximately 30 μg (floral part)-1 (Fig. 7B). Upon flower senescence (Phase III), cellulose content decreased significantly, reducing to approximately 18 μg (floral part)-1. On a per gram fresh weight basis, a general increase in cellulose content was observed during floral bud development (Phase I), flower opening (Phase II) and flower senescence (Phase III) (Fig. 8B). 4.3.3 Hemicelluloses Sepals and petals demonstrated similar patterns of changes in hemicellulose content throughout development. On a per floral part (sepal or petal) basis, the amount of hemicelluloses decreased constantly throughout floral bud development (Phase I), flower opening (Phase II) and flower senescence (Phase III), decreasing from approximately 30 μg (floral part)-1 on day 4 (Phase I) to 2 μg (floral part)-1 on day 12 (Phase III) (Fig. 7C). On a per gram fresh weight basis, hemicellulose content also decreased constantly throughout floral bud development till flower senescence (Fig. 8C). 4.3.4 Total pectins Sepals and petals demonstrated similar patterns of changes in total pectin content throughout development. On a per floral part basis, the amount of total pectins increased during floral bud development (Phase I), increasing from approximately 30 μg (floral part)-1 on day 4 to 52 80 μg (floral part)-1 on day 9. No changes were observed in total pectin content during flower opening (Phase II) and flower senescence (Phase III) (Fig. 7D). On a per gram fresh weight basis, a decrease in total pectin content was observed during early stages of floral bud development (Phase I); it remained constant till flower opening (Phase II) and increased upon flower senescence (Phase III) (Fig. 8D). 4.3.5 Soluble pectins Sepals and petals demonstrated similar patterns of changes in soluble pectin content throughout development. On a per floral part basis, the quantities of water-soluble, CDTAsoluble and Na2CO3-soluble pectins increased during floral bud development (Phase I) (Fig. 7E – G). While the increases in CDTA-soluble and Na2CO3-soluble pectins during this developmental period were relatively constant, that of water-soluble pectins was very drastic, with an approximately three fold increase in content during the early stages of floral bud development (day 7). The amount of CDTA-soluble pectins remained at high levels throughout flower opening (Phase II) and flower senescence (Phase III). On the other hand, the quantities of water-soluble and Na2CO3-soluble pectins decreased upon flower opening and flower senescence respectively. When expressed as a percentage of total pectins, all soluble pectins demonstrated an increase during floral bud development (Phase I), and subsequently decreased during flower senescence (Phase III) (Table 3, 4). In sepals, watersoluble pectins increased from 5.5 % to 9.1 % during floral bud development, then decreased to 4.6 % upon senescence; CDTA-soluble pectins increased from 13.5 % to 24.5 % during floral bud development, then decreased slightly to 22.8 % upon senescence; Na2CO3-soluble pectins increased from 36.9 % to 63.0 % during floral bud development, then decreased to 50.1 % upon senescence (Table 3). The corresponding changes in petals were 5.1 % to 6.7 % 53 Table 3. Total and soluble pectins in EIR derived from D. crumenatum sepals at various developmental stages. Numbers in parentheses represent solubilized pectins as a percentage of total pectins. Stage of development (days after induction) Total pectins [μg (floral part)-1] 4 25.2 1.4 (5.5) 3.4 7 70.5 6.4 (9.1) 14.2 (20.1) 42.1 (59.7) 9 90.5 6.6 (7.3) 22.2 (24.5) 57.0 (63.0) II 10 95.4 4.9 (5.1) 21.2 (22.2) 55.9 (58.6) III 12 90.6 4.2 (4.6) 20.7 (22.8) 45.4 (50.1) Phase of development I Soluble pectins [μg (floral part)-1] Watersoluble CDTAsoluble (13.5) Na2CO3soluble 9.3 (36.9) 54 Table 4. Total and soluble pectins in EIR derived from D. crumenatum petals at various developmental stages. Numbers in parentheses represent solubilized pectins as a percentage of total pectins. Stage of development (days after induction) Total pectins [μg (floral part)-1] Watersoluble CDTAsoluble 4 23.4 1.2 (5.1) 1.3 (5.6) 6.3 7 45.1 3.0 (6.7) 7.4 (16.4) 27.4 (60.8) 9 60.3 3.2 (5.1) 8.5 (14.1) 38.9 (64.5) II 10 63.5 3.3 (5.2) 9.4 (14.8) 35.4 (55.7) III 12 61.0 1.5 (2.5) 7.8 (12.8) 28.5 (46.7) Phase of development I Soluble pectins [μg (floral part)-1] Na2CO3soluble (26.9) 55 to 2.5 % for water-soluble pectins, 5.6 % to 16.4 % to 12.8 % for CDTA-soluble pectins, and 26.9 % to 64.5 % to 46.7 % for Na2CO3-soluble pectins (Table 4). Maximum percentages of water-soluble, CDTA-soluble and Na2CO3-soluble pectins were all observed during the intermediate (day 7) or late (day 9) stages of floral bud development (Phase I), indicating a maximum level of pectin solubility during this developmental period. On a per gram fresh weight basis, a constant decrease in water-soluble pectin content was observed throughout floral bud development (Phase I), while decreases in contents of CDTA-soluble and Na2CO3-soluble pectins were observed only during the later stages of floral bud development (Fig. 8E – G). During flower opening (Phase II), decreases in amounts of CDTA-soluble and Na2CO3-soluble pectins were also observed. Upon flower senescence (Phase III), while content of water-soluble pectins remained at low and constant levels, contents of CDTA-soluble and Na2CO3-soluble pectins increased (Fig. 8E – G). 4.4 Activities of cell wall-based enzymes 4.4.1 Soluble proteins Sepals and petals demonstrated similar patterns of changes. On a per floral part basis, soluble protein content increased steadily during floral bud development (Phase I) and flower opening (Phase II) (Fig. 9A). Between young floral buds (day 4) and newly opened flowers (day 10), soluble protein content increased by approximately two folds. A drastic and significant (P < 0.05) decrease in soluble protein content of approximately eight folds was observed during flower senescence (Phase III). 56 Phase I II III Fig. 9. D. crumenatum sepal and petal soluble protein content at each developmental stage. (A) Expressed on a per floral part basis; (B) expressed on a per gram fresh weight (gFW-1) basis. Phases of development – I: floral bud development, II: flower opening, III: flower senescence. Each value is mean ± SE (n = 5) and values with different alphabet(s) (sepals: a – e; petals: a’ – e’) are significantly different according to Fisher’s LSD test (P < 0.05). 57 On a per gram fresh weight basis, soluble protein levels decreased during floral bud development (Phase I), and remained at low and relatively constant levels throughout flower opening (Phase II) and flower senescence (Phase III) (Fig. 9B). 4.4.2 Cellulase Optimization of the cellulase gel diffusion assay using commercially available cellulase demonstrated that the assay, when applied on acidic gel, provided the most sensitive results (Fig. 10). Increasing contrast from the congo red staining was observed with increasing acidity of the gel, aiding the measurement of the diffusion zone (Fig. 10A, B, C). The pH of the gel only affected the clarity of the diffusion zones and did not affect cellulase activity (Fig. 10D – R). Cellulase activity increased with higher incubation temperature and longer incubation duration (Fig. 10D – R). Due to the different pH values of the gel and cellulase solutions, a pH gradient would have existed upon the diffusion of the enzyme into the gel. The comparison of commercially available cellulase activity in such a setup was required as the extraction of cellulase from the sepals and petals of D. crumenatum was conducted over various pH as well (refer to Section 3.6.1 under Materials and Methods). In the presence or absence of a pH gradient, the extent of zone clearing at each gel pH was observed to be unaffected by the pH of the cellulase solutions (Fig. 10D – R). Subsequently, the gel diffusion assay for D. crumenatum extracts was conducted on pH 5.7 gel. When extracted at pH 5.1, desalted extracts from fresh D. crumenatum sepals and petals of all developmental stages did not exhibit any zone clearing. Zone clearance was however observed using desalted extracts from fresh D. crumenatum column tissues extracted at pH 5.1 (Fig. 11). Similar results were obtained with crude extracts from fresh 58 59 2h 22°C 24h 2h 37°C 24h 24 h 37°C Fig. 10. Optimisation of cellulase gel diffusion assay using commercial cellulase (0.5 mg ml-1). Gels (1.7 % agarose) contained 0.5 % CMC as substrates. Zone clearing for assays conducted at gel pH 5.7, pH 6.1 and pH 7.0, at 22°C for 2 h (D – F), 22°C for 24 h (G – H), 37°C for 2 h (J – L), 37°C for 24 h (A – C, M – O). Gels without CMC were included as controls (P – R). Positioning of wells: 1) cellulase (pH 5.1); 2) boiled cellulase (pH 5.1); 3) cellulase (pH 7.0); 4) boiled cellulase (pH 7.0). 2h 37°C Fig. 11. Gel diffusion assay for cellulase. Assays were carried out at gel pH 5.7 at 37°C for a 24 h incubation period; 0.5% CMC as substrate. (A) Day 4, (B) Day 7, (C) Day 9, (D) Day 10, (E) Day 12 after cold induction, (F) position of wells and desalted samples extracted at pH 5.1. Positioning of wells: 1) sepals, 2) boiled sepals, 3) petals, 4) boiled petals, 5) columns, 6) boiled columns, 7) commercial cellulase (0.5 mg ml-1), 8) boiled commercial cellulase. Similar results were obtained from samples extracted and assayed at pH 6.1 and 7. Lower concentrations of CMC (0.05% and 0.1%) also showed similar results. 60 tissues, and when tissues were extracted at pH 6.1 and 7.0. No cellulase activities in all sepal and petal extracts were observed in tests with lower concentrations of CMC (0.05 % and 0.1 %). All extracts from frozen tissues exhibited no cellulase activity. Colorimetric analyses using crude or desalted extracts from fresh sepal or petal tissues, regardless of extraction pH, demonstrated that the extracts did not contain components to hydrolyze CMC and filter paper. Crude or desalted extracts from column tissues, however, demonstrated cellulase activity. It was concluded that sepals and petals of D. crumenatum had no or very insignificant levels of cellulase activity and no further studies were conducted. 4.4.3 Polygalacturonase Colorimetric analyses using crude or desalted extracts from fresh sepal or petal tissues, regardless of extraction pH, did not contain components to hydrolyze polygalacturonic acid. Desalted extracts from fresh or frozen D. crumenatum column tissues however, demonstrated PG activity. It was concluded that sepals and petals of D. crumenatum had no or very insignificant levels of PG activity and no further studies were conducted. 4.4.4 Pectin methylesterase PME demonstrated the second highest overall activity when compared to the other tested enzymes, with specific activities ranging from 6 units (mg protein)-1 to 34 units (mg protein)1 (Fig. 12A). PME activity [(mg protein)-1 basis, Fig. 12A] decreased throughout floral bud development (Phase I) and flower opening (Phase II), reaching a minimum level of approximately 9 units (mg protein)-1. There was an approximately three fold increase in activity as the flowers senesced (Phase III). 61 Phase I II III Fig. 12. Changes in pectin methylesterase activity in sepals and petals during development of D. crumenatum floral buds. (A) Activity expressed on (mg protein)-1 basis, (B) activity expressed on (floral part)-1 basis, (C) activity expressed on gFW-1 basis. Phases of development – I: floral bud development, II: flower opening, III: flower senescence. Each value is mean ± SE (n = 5) and values with different alphabet(s) (sepals: a – d; petals: a’ – d’) are significantly different according to Fisher’s LSD test (P < 0.05). 62 On a per floral part basis (Fig. 12B), an increase in PME activity was observed during the early stages of floral bud development (Phase I); it decreased throughout flower opening (Phase II) and flower senescence (Phase III). On a per gram fresh weight basis (Fig. 12C), PME activity showed a decreasing trend during floral bud development (Phase I), and it remained at low and constant levels through flower opening (Phase II) and flower senescence (Phase III). 4.4.5 β-galactosidase β-gal exhibited the third highest overall activity when compared to the other tested enzymes, with specific activities ranging from 2 units (mg protein)-1 to 12 units (mg protein)-1 (Fig. 13A). β-gal activity [(mg protein)-1 basis, Fig. 13A)] remained unchanged and was approximately 4 units (mg protein)-1 during floral bud development (Phase I) and flower opening (Phase II). Upon flower senescence (Phase III), a significant increase (P < 0.05) in β-gal activity, from approximately 4 units (mg protein)-1 to 10 units (mg protein)-1 was observed. On a per floral part basis (Fig. 13B), an increase in β-gal activity was observed during early stages of floral bud development (Phase I); it decreased as development progressed, and the level of activity increased again during flower senescence (Phase III). On a per gram fresh weight basis, a drastic decrease in β-gal activity was observed during floral bud development (Phase I) and flower opening (Phase II), and it subsequently increased during flower senescence (Phase III) (Fig. 13C). 63 Phase I II III Fig. 13. Changes in β-galactosidase activity in sepals and petals during development of D. crumenatum floral buds. (A) Activity expressed on (mg protein)-1 basis, (B) activity expressed on (floral part)-1 basis, (C) activity expressed on gFW-1 basis. Phases of development – I: floral bud development, II: flower opening, III: flower senescence. Each value is mean ± SE (n = 5) and values with different alphabet(s) (sepals: a – c; petals: a’ – c’) are significantly different according to Fisher’s LSD test (P < 0.05). 64 4.4.6 β-glucosidase β-glu exhibited the highest overall activity when compared to the other tested enzymes, with specific activities ranging from 9 units (mg protein)-1 to 43 units (mg protein)-1 (Fig. 14A). During late stages of floral bud development (Phase I), β-glu activity [(mg protein)-1 basis, Fig. 14A] increased sharply by approximately two folds, and decreased during flower opening (Phase II) and flower senescence (Phase III). On a per floral part basis (Fig. 14B), an increase in β-glu activity was observed during floral bud development (Phase I); it subsequently decreased upon flower opening (Phase II) and flower senescence (Phase III). On a per gram fresh weight basis (Fig. 14C), a constant decrease in β-glu activity was observed throughout floral bud development (Phase I), flower opening (Phase II) and flower senescence (Phase III). 4.4.7 β-mannosidase Activity of β-man was very much lower compared to that of PME, β-gal and β-glu, with specific activities ranging from 0.4 units (mg protein)-1 to 1.8 units (mg protein)-1 (Fig. 15A). β-man activity [(mg protein)-1 basis, Fig. 15A] demonstrated a gradual increase throughout floral bud development (Phase I), flower opening (Phase II) and flower senescence (Phase III), with a maximum activity of approximately 1.8 units (mg protein)-1 in the senesced flower. On a per floral part basis (Fig. 15B), an increase in β-man activity was observed during early stages of floral bud development (Phase I), which then decreased over flower opening (Phase II) and flower senescence (Phase III). . 65 Phase I II III Fig. 14. Changes in β-glucosidase activity in sepals and petals during development of D. crumenatum floral buds. (A) Activity expressed on (mg protein)-1 basis, (B) activity expressed on (floral part)-1 basis, (C) activity expressed on gFW-1 basis. Phases of development – I: floral bud development, II: flower opening, III: flower senescence. Each value is mean ± SE (n = 5) and values with different alphabet(s) (sepals: a – c; petals: a’ – c’) are significantly different according to Fisher’s LSD test (P < 0.05). 66 Phase I II III Fig. 15. Changes in β-mannosidase activity in sepals and petals during development of D. crumenatum floral buds. (A) Activity expressed in mg protein-1 basis, (B) activity expressed in (floral part)-1 basis, (C) activity expressed in gFW-1 basis. Phases of development – I: floral bud development, II: flower opening, III: flower senescence. Each value is mean ± SE (n = 5) and values with different alphabet(s) (sepals: a – c; petals: a’ – c’) are significantly different according to Fisher’s LSD test (P < 0.05). 67 On a per gram fresh weight basis (Fig. 15C), β-man decreased sharply during the late stages of floral bud development (Phase I), and it increased slightly upon flower senescence (Phase III). 4.4.8 β-xylosidase Similar to β-man, activity of β-xyl was very much lower compared to that of PME, β-gal and β-glu, with specific activities ranging from 0.3 units (mg protein)-1 to 1.9 units (mg protein)-1 (Fig. 16A). β-xyl activity [(mg protein)-1 basis, Fig. 16A] was at low and relatively constant levels throughout floral bud development (Phase I) and flower opening (Phase II). Upon flower senescence (Phase III), the activity of β-xyl increased by approximately three to four folds. On a per floral part basis (Fig 16B), an increase in β-xyl activity was observed during early stages of floral bud development (Phase I); it remained constant till flower opening (Phase II), then decreased upon flower senescence (Phase III). On a per gram fresh weight basis (Fig. 16C), a constant decrease in β-xyl activity was observed during floral bud development (Phase I), it remained at low and constant levels during flower opening (Phase II) and flower senescence (Phase III). 4.5 Expression of cell wall-based enzyme gene transcripts Based on the results on the activities of the various cell wall-based enzymes investigated, βgal and PME were selected for further studies on the expression of their gene transcripts at various stages of floral bud/ flower development. Both β-gal and PME have been suggested to be key enzymes involved in various cell wall modification processes during fruit ripening and flower opening (Brummell and Harpster 2001; O’Donoghue 2006). Since the results 68 Phase I II III Fig. 16. Changes in β-xylosidase activity in sepals and petals during development of D. crumenatum floral buds. (A) Activity expressed in mg protein-1 basis, (B) activity expressed in (floral part)-1 basis, (C) activity expressed in gFW-1 basis. Phases of development – I: floral bud development, II: flower opening, III: flower senescence. Each value is mean ± SE (n = 5) and values with different alphabet(s) (sepals: a – c; petals: a’ – c’) are significantly different according to Fisher’s LSD test (P < 0.05). 69 from the enzyme activities demonstrated that both sepals and petals exhibited the same patterns of change (Fig. 12, 13), expression of the gene transcripts were analysed in samples consisting of sepals and petals together. 4.5.1 Total RNA integrity and quality Total RNA isolated from D. crumenatum samples from various floral bud/ flower developmental stages were assessed to be of high integrity and high quality. The RNA samples from each stage of development exhibited two distinct bands (28S ribosomal RNA and 18S ribosomal RNA) when examined on agarose gels, indicating that the RNA had high integrity with minimal degradation (Fig. 17). The A260/280 ratios of the RNA samples were between 1.63 – 2.02, and A260/230 ratios were above 2.0, indicating the absence of protein and carbohydrate contamination respectively (Table 5). 4.5.2 Optimization of PCR Gradient PCR for the control gene β-TUB (Fig. 18A) demonstrated that the optimum annealing temperature for the pair of primers used was in the range of 50°C – 55°C, and the primers exhbited high specificity, with only 1 distinct band of size 830 bp. A wide range of temperatures were suitable as optimum annealing temperature for the pair of primers for βGAL (Fig. 18B), ranging from 50°C – 59°C. The pair of primers was also specific, with only 1 distinct band size of 441 bp. Gradient PCR for PME (Fig. 18C) demonstrated that the optimum annealing temperature for the pair of primers used was in the range of 48°C – 54°C. Between 48°C – 50°C, however, a high degree of smearing was obseerved. The PME primers demonstrated high specificity, with only 1 distinct band of size 292 bp. Since the optimum annealing temperature for all three pairs of primers coincided within the same temperature range, 52°C was selected as the final temperature to be used for all further PCR 70 Fig. 17. Electrophoresis of total RNA from sepals and petals of D. crumenatum, with 28S and 18S ribosomal RNAs indicated. 71 Table 5. Quality of RNA obtained from D. crumenatum samples at various floral bud/ flower developmental stages (n = 4). Sample (days after induction) Ratio of absorbance 260nm/ 280 nm Ratio of absorbance 260nm/ 230nm 4 1.76 ± 0.16 2.19 ± 0.15 7 1.82 ± 0.15 2.27 ± 0.13 9 1.63 ± 0.09 2.06 ± 0.19 10 1.96 ± 0.19 2.21 ± 0.11 12 2.02 ± 0.12 2.40 ± 0.17 72 °C °C °C Fig. 18. Gradient PCR amplification of (A) β-TUB, (B) β-GAL and (C) PME transcripts, conducted at various annealing temperatures. 73 amplifications. Amplification of all three genes at various numbers of amplification cycles showed that their expression levels continued to increase up till at least 32 cycles (Fig. 19). Further PCR amplifications were thus conducted for 30 cycles for β-TUB, β-GAL and PME. 4.5.3 Expression of β-GAL and PME during floral bud and flower development β-GAL transcripts were expressed at low levels during floral bud development (Phase I), which then became highly expressed upon flower opening (Phase II) and flower senescence (Phase III) (Fig. 20A). No expression of PME transcripts was observed during the early stages of floral bud development (Phase I), but PME transcripts were expressed at very low levels during the later stages of floral bud development. Expression of PME transcripts continued to increase upon flower opening (Phase II) and flower senescence (Phase III) (Fig. 20B). 4.6 Membrane stability Significant increases in percentage of electrolyte leakage (P < 0.05) were observed during early stages of bud development (day 4 – day 9) (Fig. 21A) in sepals and petals, and the percentage of leakage continued to increase significantly up till senescence stage in the petals. Membrane stability (determined as membrane stability index, MSI) of sepals and petals declined with development of D. crumenatum floral buds (Fig. 21B). The highest decline in membrane stability was observed during early stages of development; it decreased from approximately MSI 80 on day 4 to MSI 69 on day 7 for sepals, and from MSI 72 on day 4 to MSI 62 on day 7 for petals. 74 Fig. 19. PCR amplifications of β-TUB, β-GAL and PME transcripts conducted at an annealing temperature of 52°C, and at 24, 26, 28, 30 and 32 cycles each. 75 Phase Fig. 20. Expression of (A) β-GAL and (B) PME transcripts at various developmental stages of D. crumenatum floral buds/ flowers. β-TUB was used as the control gene. Phases of development – I: floral bud development, II: flower opening, III: flower senescence. 76 Phase I II III Fig. 21. Membrane stability of sepals and petals of D. crumenatum at various floral bud/ flower developmental stages. (A) Electrolyte leakage of sepals and petals, (B) membrane stability index of sepals and petals. Phases of development – I: floral bud development, II: flower opening, III: flower senescence. Each value is mean ± SE (n = 5) and values with different alphabet(s) (sepals: a – c; petals: a’ – e’) are significantly different according to Fisher’s LSD test (P < 0.05). 77 4.7 Control of flower opening One day after treatment, treated day 9 floral buds exhibited either full flower opening (Fig. 22A), or remained as dormant floral buds with day 9 features showing no signs of senescence (Fig. 22B), or displayed incomplete flower opening (Fig. 22C, D). Treatments with water, GA, AOA and AOA + glucose resulted in almost 100 % full flower opening in D. crumenatum floral buds, while the presence of BA in treatment solutions greatly reduced the percentage of full flower opening (Table 6). The greatest reduction, of approximately 50 %, in full flower opening occurred when day 9 floral buds were treated with 10-8 M BA. The inhibitory effects of BA on full flower opening were also observed in the combined treatments consisting of both BA and GA. While GA alone resulted in almost 100 % full flower opening at all the tested GA concentrations, the inclusion of BA resulted in about 15 – 40 % reduction in full flower opening, depending on the concentration of BA used. The percentages of treated floral buds that remained dormant with day 9 features, and without any signs of senescence (Fig. 21B), were as shown in Table 7. Dormancy of floral buds occurred only in the presence of BA, with the highest percentage of approximately 45 % when treated with 10-8 M BA. All of these dormant floral buds aborted without showing any signs of flower opening. All of the dormant floral buds began to show signs of senescence, turning transparent and exhibiting signs of discolouration in the pedicel by day 11 (Fig. 22I). The presence of BA also resulted in incomplete flower opening in some floral buds (Fig. 22C, D). These floral buds did not attain full flower opening and senesced at the same time and rate as those that displayed full flower opening. All the fully opened and incompletely opened flowers began to show signs of senescence on day 11. Except for those treated with AOA + glucose, all flowers senesced 78 similar to the control, with discolouration of the pedicel and the closing or wilting of all sepals and petals (Fig. 22G). Approximately 25 % of the flowers from the AOA + glucose treatment exhibited abnormal senescing phenotype on day 11. They remained fully open with the sepals and petals turning translucent, and were thus still considered as senesced flowers (Fig. 22E). By day 12, the sepals and petals from these flowers became even more translucent and turned ‘outwards’, instead of the usual browning and closing of the flowers (Fig. 22F, H). 79 Fig. 22. General physical features of day 9 D. crumenatum floral buds after treatments. (A) Fully opened flower, (B) dormant floral bud with day 9 features and no signs of senescence, (C) incompletely opened flower with slight opening on the sides and tip of the floral bud, (D) incompletely opened flower, (E) fully opened flower with translucent sepals and petals, (F) senescent flower turned ‘outwards’ with translucent sepals and petals, (G) wilted flower, (H) magnified view of flower in F, (I) senescent floral bud. Scale bar = 0.5 cm. 80 Table 6. Percentages of D. crumenatum floral buds that displayed full flower opening one day after treatment. Treatments were applied on day 9 floral buds that normally proceeded to flower opening the next day. Treatment Total number of floral buds 20 Number of fully opened flowers 20 Percentage flowered (%) ± SE 100 ± 0 BA (10-6 M) 20 11 55 ± 22 BA (10-8 M) 20 10 50 ± 6 BA (10-10 M) 20 12 60 ± 14 GA (10-6 M) 20 19 95 ± 5 GA (10-8 M) 20 20 100 ± 0 GA (10-10 M) 20 19 95 ± 5 BA/GA (10-6 M) 20 13 65 ± 10 BA/GA (10-8 M) 20 12 60 ± 8 BA/GA (10-10 M) 20 17 85 ± 9 AOA (10-6 M) 20 20 100 ± 0 AOA (10-8 M) 20 20 100 ± 0 AOA (10-6 M) + 0.1 M glucose 20 20 100 ± 0 AOA (10-8 M) + 0.1 M glucose 20 20 100 ± 0 Distilled water (control) 81 Table 7. Percentages of D. crumenatum floral buds that displayed dormancy after one day of treatment, and percentages of dormant floral buds that subsequently aborted two days after treatment. Floral buds were considered to be dormant if they maintained the typical features of a normal day 9 floral bud as described in Table 1, without any signs of senescence. Treatment Distilled water (control) Total number of floral buds 20 Number of white floral buds 0 Percentage of dormant floral buds (%) ± SE 0 Percentage of aborted floral buds (%) 0 BA (10-6 M) 20 3 15 ± 5 100 BA (10-8 M) 20 9 45 ± 6 100 BA (10-10 M) 20 2 10 ± 6 100 GA (10-6 M) 20 0 0 0 GA (10-8 M) 20 0 0 0 GA (10-10 M) 20 0 0 0 BA/GA (10-6 M) 20 4 20 ± 11 100 BA/GA (10-8 M) 20 2 10 ± 6 100 BA/GA (10-10 M) 20 0 0 0 AOA (10-6 M) 20 0 0 0 AOA (10-8 M) 20 0 0 0 AOA (10-6 M) + 0.1 M glucose 20 0 0 0 AOA (10-8 M) + 0.1 M glucose 20 0 0 0 82 Chapter 5. Discussion The cell wall is a dynamic structure that can undergo diverse changes during different developmental phases of the plant. Many important physiological functions are associated with cell wall modifications or cell wall remodelling. These include structural and mechanical support, maintenance of cell shape, control rate of growth, protection against dehydration, defence against pathogens and cell-cell interactions (Carpita and Gibeaut 1993; Cosgrove 1999; Carpita and McCann 2000). While the roles of cell wall modifications have been studied in great detail in various plant processes such as seed germination (Edwards et al. 1985; Crombie et al. 1998; Buckeridge et al. 2000; Tine et al. 2000), abscission (Lashbrook et al. 1994; Patterson 2001; Fulton and Cobbett 2003) and fruit ripening (Brummell and Harpster 2001; Brummell 2006), the association of cell wall remodelling and flowering remains rather unclear, with limited studies conducted to date. The main interest was in determining if there were distinctive changes, occurring in sepals and petals of D. crumenatum floral buds and flowers, which might contribute to the opening and senescing of the flowers. 5.1 Cell wall changes related to D. crumenatum floral bud/ flower development The development of D. crumenatum floral buds and opening of the flowers were found to be accompanied by increasing disorganization of parenchyma cells and appearance of intercellular spaces (Fig. 5). It was hypothesized that such anatomical changes were due to alterations in cell wall constituents that commenced in young floral buds, and that such cell wall modifications might regulate the development, opening and senescing of floral buds/flowers. 83 The ‘looseness’ observed in the structures of sepals and petals during development of D. crumenatum floral buds (Phase I) (Fig. 5) appeared to be due to the breakdown of cell wall materials. On the contrary, the results obtained demonstrated that the net amount of cell wall materials (in the form of ethanol-insoluble residue, EIR) increased during floral bud development, which was proportional to the increase in weight and size of the sepals and petals. Loss of cell wall materials occurred only during flower opening (Phase II) and flower senescence (Phase III) (Fig. 4, 7A, 8A). The synthesis of cellulose, the principal scaffold in cell walls, was observed to increase in conjunction with the increase in cell wall materials during floral bud development (Fig. 4, 7B, 8B). The net amount of cellulose subsequently decreased slightly during flower senescence, indicating cellulose hydrolysis. However, enzymatic analyses demonstrated the absence of cellulase activity at all developmental stages (Fig. 11). It was possible that cellulase was present in sepals and petals of D. crumenatum, but its activity was at very low and undetectable levels, or carboxymethylcellulose and filter paper could be unsuitable substrates for cellulase in D. crumenatum. Substrate specificities of cellulases have been studied with proteins purified from avocado, strawberry and tomato, and results demonstrated that different substrates resulted in varying levels of cellulase activity within each species (Hayashi et al. 2005; Urbanowicz et al. 2007). The net quantity of hemicelluloses increased during D. crumenatum floral bud development and decreased during flower opening and flower senescence (Fig. 4, 7C, 8C). The results indicated that although hemicelluloses were synthesized during floral bud development, the re-distribution of hemicelluloses due to increase in size of sepals and petals was not in synchrony with the rate of synthesis. As a result, there existed a dilution of hemicelluloses and a decrease in concentration when expressed on a per floral part basis (Fig. 84 7C). Hemicelluloses play important structural roles in the cell walls by contributing to the cross-linking of cellulose microfibrils, creating a network between the various microfibrils (Carpita and McCann 2000). One of the most significant networks, the cellulose-xyloglucan network, has been suggested to be a major load-bearing structure in the cell walls, and is an important constraint to cell wall loosening (Rose and Bennett 1999). The redistribution and breakdown of hemicelluloses may result in disassembling or relaxation of the cellulosexyloglucan network, causing cell wall swelling and altering the movement of cell wall enzymes within the cell wall matrix, varying the access to their respective substrates (Rose and Bennett 1999; Brummell 2006). The breakdown of hemicelluloses can be attributed to the actions of various glycosidases (de Vetten et al. 1991; Minic and Jouanin 2006). The specific activities of β-gal (Fig. 14A), β-man (Fig. 15A) and β-xyl (Fig. 16A), together with the changes in amount of soluble proteins with development of D. crumenatum floral buds and flowers (Fig. 9), resulted in the upregulation of the enzyme activities [units (floral part)-1] during the later stages of floral bud development (Fig. 14B, 15B, 16B), and might account for the subsequent breakdown of hemicelluloses during flower opening and senescence. The family of glycosidases is known to play a crucial role in the degradation of various cell wall polysaccharides, allowing the remodelling of the cell wall structure (Minic 2008). The breakdown of cell walls, catalysed by glycosidases, for the mobilization of cell wall storage polysaccharides in germinating or post-germinating seeds have been widely studied (Buckeridge et al. 2000). In guar (Cyamopsis tetragonolobus), fenugreek (Trigonella foenum-greacum) and carob (Ceretonia siliqua), the degradative effects of β-man has been shown to be important and associated with galactomannan mobilization, following 85 germination of the seeds (Reid 1971, 1985; McClendon et al. 1976). The mobilisation of xyloglucan in seeds of Tropaeolum majus and Copaifera langsdorffi has also been shown to be accompanied by the rise in β-gal activity (Edwards et al. 1985; Buckeridge et al. 1992). Softening of fruits during ripening has been associated with disassembly of the cell wall due to changes in hemicellulose structure or content. In the ripening of avocado, Japanese pear (Pyrus pyrifolia) and strawberry (Ronen et al. 1991; Itai et al. 1999; Martinez et al. 2004), increases in β-xyl activities have been shown to be related to the degradation of arabinan and xylan during the softening processes. Pectin hydrolysis in the development of D. crumenatum floral buds and flowers were not observed (Fig. 4, 7D, 8D). In support of this observation, activities of the enzyme that catalyses pectin hydrolysis, PG, was also undetected throughout development of the orchid floral buds/flowers. Pectins are heteropolysaccharides in the cell wall, and are important in regulating wall porosity, cell-cell adhesion at the middle lamella and movement of enzymes within the cell wall matrix (Baron-Epel et al. 1988; Pena and Carpita 2004; Brummell 2006). Degradation of cell wall pectins in the context of fruit ripening and softening has been widely studied. In the ripening of carambola and grapes, decreases in pectin content was accompanied by increases in PG activity (Chin et al. 1999; Deng et al. 2005). Pectin hydrolysis is, however, not the only pectin modification that can result in the remodelling of the cell wall structure. Pectin solubilisation, resulting from changes in the types of bonds and/or bond strengths existing between pectin molecules and/ or other matrix molecules, has been correlated with cell wall swelling, increasing the accessibility of cell wall enzymes to their respective substrates (Redgewell et al. 1997; Brummell 2006). Water-soluble and CDTA-soluble pectins are relatively weakly bound to cell wall polysaccharides by molecular 86 entanglements, hydrophobic forces, weak ionic bonds or ionic calcium bridges, while Na2CO3-soluble pectins are more strongly attached to the cell wall via covalent bonds (Brummell 2006). Increases in pectin solubilisation have been demonstrated in the ripening of tomato and avocado that were accompanied by changes in cell wall structure (Carrington et al. 1993; Wakabayashi et al. 2000). In D. crumenatum, pectin solubilisation changed as the floral buds and flowers developed. There were increasing proportions of pectins that were susceptible to the solubilising agents during floral bud development (Phase I), and decreased pectin solubility during flower opening (Phase II) and senescence (Phase III) (Table 3, 4). PME, which de-esterifies pectins, has been suggested to increase pectin solubilisation by creating electronic repulsion between negatively-charged molecules that could result in the loosening of weakly attached pectins from the cell wall (Grignon and Sentenac 1991). In D. crumenatum, the specific activity of PME (Fig. 12A) together with changes in the quantity of soluble proteins (Fig. 9), resulted in a significant increase in PME activity [units (floral part)-1] during the early stages of floral bud development (Phase I) (Fig. 12B). This maximum level of PME activity coincided with the largest increase in pectin solubility (Table 3, 4). However, PME transcript expression could not be related to the activity profile (Fig. 20B). PME might be coded by more than one gene in D. crumenatum such that the enzyme activity profile obtained was a quantification of a composite of PME proteins from two or more highly homologous genes. Studies have demonstrated that PME isoforms in cell walls are encoded by a multigene family, and the analysis of the genome sequence in arabidopsis has identified 67 PME-related genes PME (The Arabidopsis Genome Initiative 2000; Micheli 2001). In tomato, PME has also been shown to consist of at least four genes, some of which are highly homologous (Harriman et al. 1991; Hall et al. 1994; Turner et al. 87 1996; Gaffe et al. 1997). Cell wall modifications, occurring during the ripening of carambola and grapes, were the result of an increase in PME activity, which was in synchrony with pectin solubilisation (Chin et al. 1999; Deng et al. 2005). The activity of PME resulting in increased pectin solubilisation has also been studied during cell separation of root border cells in pea (Peasum sativum) (Stephenson and Hawes 1994). It was suggested that pectin solubilisation in the middle lamella caused changes in cellular adhesion and resulted in separation of the cells. Furthermore, increase in PME activity has been shown to accompany dormancy breakage and germination of yellow cedar seeds (Chamaecyparis nootkatensis), and although the precise role of the enzyme in regulating seed germination remains unknown, the actions of PME must had altered and weakened the cell walls of the megagametophyte, aiding radicle protrusion and seed germination (Ren and Kermode 2000). There are many other factors that may also regulate pectin solubilisation and other polymer modifications. One such candidate could be β-gal, the enzyme responsible for the degradation of cell wall galactan, regulating cell wall flexibility, intercellular connections and cell wall porosity, and affecting the mobility of enzymes within the cell wall matrix (Brummell and Harpster 2001; Brummell 2006). In vitro treatments of cell wall preparations from papaya with β-gal resulted in increased pectin solubilisation (Ali et al. 1998). In D. crumenatum, specific activity of β-gal (Fig. 13A) together with changes in the quantity of soluble proteins (Fig. 9), resulted in significant increases in β-gal activity [units (floral part)1 ] during the early stages of floral bud development (Phase I) and flower senescence (Phase III) (Fig. 13B). β-gal transcript expression, however, peaked only during flower opening (Phase II) and flower senescence (Phase III) (Fig. 20A). Similar to PME, β-gal might also be encoded by more than one homologous gene in D. crumenatum. In tomato, at least seven β- 88 gal genes were expressed during development and fruit ripening (Smith and Gross 2000). Post-transcriptional and post-translational modifications may also result in the differences between the enzyme activity profile and the expression of the gene transcript. It is possible that non-enzymic mechanisms might also be involved in cell wall modifications. In tomato, ascorbate-generated hydroxyl radicals demonstrated non-enzymic scission of cell wall polysaccharides, and caused an increase in pectin solubilisation (Dumville and Fry 2003). In the study, physiological concentrations of ascorbate gradually solubilised pectins present in the cell wall materials of the fruit. In the presence of DMSO, which is a scavenger for hydroxyl radicals, pectin solubilisation was inhibited, demonstrating that the mechanism of action of ascorbate was via hydroxyl radicals (Dumville and Fry 2003). 5.2 Model for cell changes accompanying D. crumenatum floral bud/ flower development The data obtained suggested that cell wall modifications determining floral bud development, flower opening and flower senescence in D. crumenatum were accompanied by modifications in both the cellulose-hemicellulose networks and pectin networks. The results suggested that temporal changes in cell wall modifications occurred in two stages (Fig. 23). An early stage corresponding to floral bud development (Phase I), that is characterised by cellulose synthesis, slight hemicellulose synthesis and major pectin solubilisation. Although net hemicellulose content increased during this period, the re-distribution of the polysaccharides due to increases in size of sepals/ petals resulted in an overall dilution of hemicellulose content per floral part (Fig. 7C), and could have resulted in the loosening of 89 Fig. 23. Proposed model for cell wall modifications accompanying D. crumenatum floral bud and flower development. Phases of development – I: floral bud development, II: flower opening, III: flower senescence. Black bars indicate the occurrences of the particular events. 90 the cellulose/ hemicellulose networks. It was also possible that pectin solubilisation, which occurred at significant levels during this developmental phase could have loosen the pectin networks, enhancing enzyme mobilities within the cell wall matrix, resulting in subsequent cell wall changes. Pectins are the main components of the middle lamella (Brummell 2006; Liepman et al. 2007). Thus, the occurrence of high pectin solubilisation could also indicate the dissolution of the middle lamella, resulting in decreased cell adhesion. The second stage of cell wall modifications coincided with flower opening (Phase II) and flower senescence (Phase III) (Fig. 23), with extensive breakdown of cellulose and hemicellulose. These degradative effects probably resulted in the almost total dissolution of the cell walls and collapse of the parenchyma layers upon senescence of the D. crumenatum flowers (Fig. 5E). 5.3 Species-specific variations in cell wall modifications associated with flowering The influence of cell wall modifications in the regulation of flowering has only been studied in a few species of flowers, including carnation, sandersonia and daylily. Cell wall modifications in petals of carnation and sandersonia were studied in great detail, with known cell wall composition and cell wall enzyme activities, whereas there were no published data on cell wall composition of daylily (de Vetten and Huber 1990; de Vetten et al. 1991; Panavas et al. 1998; O’Donoghue et al. 2002, 2005). The different species of flowers differ in their time-frames of flower development, i.e. rate of floral bud development, time of flower opening, duration of flower opening processes, longevity of flowers (Fig. 24). Of the tested species, carnation has the longest period of floral bud development, flower opening process and longevity, while daylily has the shortest 91 Fig. 24. Comparison of flower development events in carnation (de Vetten and Huber 1990), sandersonia (Eason and Webster 1995; O’Donoghue et al. 2002), daylily (Biesleki and Reid 1992; Panavas et al. 1998) and D. crumenatum. Time of fully opened flowers was denoted as day 0 for all species. 92 developmental period. Due to the variations in the time-frames of flower development of these flowers, comparisons of flower regulatory factors at each pre-defined developmental phase (floral bud development – Phase I; flower opening – Phase II; flower senescence – Phase III) are possible, but not for comparisons at each time point of development. Comparisons of the cell wall changes between the flowers of different species show that they share some features, but differ in others (Table 8). In general, cell wall modifications in the forms of alterations in cell wall composition and cell wall enzyme activities occurred during development of all the flowers. Increases in cellulose deposition during flower opening (Phase II) and increases in β-gal activities (presumably resulting in loss of galactose and changes in pectin networks) during flower senescence (Phase III) have been observed in all four species. One of the most significant differences between the flowers was the presence of cellulase and PG in daylily, while both enzymes were absent (or at very low and undetectable levels) in all the other three species. There appears to be a species-specific variation in the pattern of cell wall modifications during flower development, but a complete comparison is difficult owing to the lack of cell wall composition data from daylily and also the lack of data on various other cell wall modifications (e.g. activities of β-glu, β-man and β-xyl) from other species. Speciesspecific variation in cell wall modifications has also been reported in ripening fruits. Pectin solubilisation in ripening avocado was observed to be three times more than in kiwifruit (Actinidia deliciosa), five times more than in blackberry (Rubus fruticosus), nine times more than in persimmon (Diospyros kaki) and plum (Prunus domestica), and 12 times more than in tomato and strawberry; whereas pectin solubilisation was absent or occurred at very low and insignificant levels in apple (Malus domestica cv. Cox’s Orange Pippin and Malus domestica 93 Table 8. Summary of modifications of cell wall polysaccharides (per floral part basis) and cell wall enzyme activities (per floral part basis or per mg protein basis) during floral bud development (Phase I), flower opening (Phase II) and flower senescence (Phase III) of D. crumenatum, carnation (de Vetten and Huber 1990; de Vetten et al. 1991), sandersonia (O’Donoghue et al. 2002) and daylily (Panavas et al. 1998). Analysis of cell wall composition and enzymatic activities during Phase I were not conducted in carnation and sandersonia. Analysis of cell wall composition was not conducted in daylily throughout development. Symbols*, abbreviations† and legend# are outlined in footnote. Phase I (Young floral bud – Mature floral bud) Phase II (Mature floral bud – Mature flower) Phase III (Mature flower – Senesced flower) Composition Cellulose Hemicellulose Total pectin WSP CSP SSP + + + + + + Ø + + + X - + Ø - + - X - + - X - X Ø Ø + Ø + Ø Ø Ø + - - X X Enzyme activity Cellulase PG PME ̀β-gal β-glu β-man β-xyl Ο Ο Ο Ο Ο Ο X Ο + Ο Ο Ο Ο Ø Ο Ο X Ο Ο Ο Ο Ο +- +- Ø + + +- + +Ø Ø Ø - X - Ø Ø X Ø Ø + - - - + Ø + X X X X X X X X X + + + + + X + + + + + X X + X X X X X *Symbols – +: factor increased in quantity or activity during the developmental phase; -: factor decreased in quantity or activity during the developmental phase; Ø: factor assayed and remained unchanged; Ο: factor assayed and absent; X: factor not assayed. † Abbreviations – WSP: water-soluble pectin; CSP: CDTA-soluble pectin; SSP: Na2CO3-soluble pectin. # Legend – various font colours indicate different flower species and their respective expression units: D. crumenatum (per floral part basis); D. crumenatum (per mg protein basis); carnation (per floral part basis); sandersonia (per floral part basis); daylily (per mg protein basis). 94 cv. Braeburn), watermelon (Citrullus lanatus cv. Charisma) and Nashi pear (Pyrus serotina cv. Nijisseiki) (Redgwell et al. 1997). The species-specific variations in pectin solubilisation in these fruits were correlated to cell wall swelling and were suggested to result in the differences in the texture of the fruits (Redgwell et al. 1997). 5.4 Relationship between flowering and senescence Flower senescence, which is the end point of flower life, is often a rapid and synchronous process, and has frequently been linked to the various catabolic and ‘death’ processes that occur after flower opening (O’Donoghue 2006). Visible markers of flower senescence often include wilting, sepal/ petal abscission, browning and drying (O’Donoghue 2006). Physiological changes associated with flower senescence include membrane degradation, alterations in cell wall structures, protein remobilization and DNA laddering (Rubinstein 2000; Wagstaff et al. 2003; Zhou et al. 2005). Previously, studies on carnation and sandersonia suggested that just prior to flower opening (late floral bud stage), a certain degree of cell wall dismantling had already started (de Vetten and Huber 1990; de Vetten et al. 1991; O’Donoghue et al. 2002). In D. crumenatum, senescence-related events such as cell wall modifications occurred as early as during young floral bud development (Fig. 7, 8, Table 8). Also, membrane stability began to decrease during the period of floral bud development (Fig. 21), indicating the breakdown of the cell membrane. The data suggested that the onset of flower senescence occurred as early as during floral bud development. Early commencement of flower senescence was also observed in morning glory (Ipomoea tricolor), where alterations to cell shape and cell wall thickness occurred even before flower opening (Phillip and Kende 1980). Petal senescence 95 was also shown to begin extremely early in Alstromeria peruviensis var. Samora (Wagstaff et al. 2003). In the study, indicators of senescence such as DNA laddering and nuclear degradation appeared as early as two days before the onset of flower opening, and these senescence processes proceeded throughout flower opening and the eventual visible senescence of the flowers (Wagstaff et al. 2003). Pollination and fertilization of flowers promote sepal and petal senescence, while keeping the fertilized ovaries viable; in nonpollinated and unfertilized flowers, whole flowers senesce and die (van Doorn 1997). The possible onset of senescence, prior to flower opening or in non-pollinated/ unfertilized flowers, would thus infer a modification of the senescence programme, due to a cascade of signals generated upon pollination and/ or fertilization that results in the senescence of the sepals and petals but not the ovaries. 5.5 Effects of growth regulators on the control of flower opening The roles of PGRs in controlling various plant growth processes has been intensely studied and many have been identified as putative floral signals (van Doorn and van Meeteren 2003; Corbesier and Coupland 2006; Sim et al. 2008). These include cytokinins and gibberellin (Corbesier and Coupland 2006). Cytokinins and gibberellins have also been shown to be capable, through their interactions, of regulating floral evocation and floral bud/ flower development (Bernier 1988; Day et al. 1995; Setyadjit et al. 2004; Kim and Miller 2008). It is interesting that out of all the treatments used in the study, reductions in percentages of flower opening and signs of floral bud dormancy occurred only in the presence of BA (Fig. 22, Table 6, 7). All of these BA-induced dormant floral buds eventually aborted, with none proceeding on to flower opening (Table 7). The data suggested that BA (at least at the tested 96 concentrations) had the capacity to induce certain signals cascade(s) that altered the physiological processes and/or biochemical processes and/or genetic regulatory pathways that usually result in the normal flowering phenotype of D. crumenatum. On the other hand, BA (at least at the tested concentrations) was also too potent, causing irreversible changes or damages to the various regulatory processes and/or pathways, finally resulting in the abortion of flowering. Similar to the results obtained from D. crumenatum, exogenous application of BA to harvested Grevillea Sylvia inflorescences resulted in a decrease in percentages of flower opening (Setyadjit et al. 2004). Furthermore, in Boronia heterophylla, treatment of the plants with BA resulted in delayed flowering (Richards 1985). It appears that flower opening is suppressed by exogenous BA. In Cosmos sulphureus, endogenous cytokinin concentrations, in particularly zeatin and zeatin riboside, were shown to be at low levels during initial floral bud development, increased prior to flower opening, and decreased upon full bloom (Saha et al. 1985). In Boronia megastima, the rapid floral bud development stages were accompanied by increases in cytokinin concentrations, in particularly zeatin, and decreases in carbohydrate concentration; upon anthesis, zeatin concentrations decreased while zeatin riboside concentrations increased (Day et al. 1995). Cytokinins have been suggested to regulate floral bud/ flower development by controlling the metabolism and distribution of carbohydrates to the floral buds/ flowers (Munoz et al. 1990; Zieslin and Khayat 1990). The influence of carbohydrates distribution on cytokinin synthesis may also be possible (Day et al. 1995). Previous studies have demonstrated that exogenous application of BA to raceme tissues of soybean (Glycine max L. Merr.) prevented flower abortion (Reese et al. 1995; Nagel et al. 2001). It was suggested that the action of BA involved redirecting the 97 movements of assimilates and resources to the treated floral buds, thus increasing the sink strength and growth rates of the floral buds, and consequently reducing abortion of flowering (Reese et al. 1995; Nagel et al. 2001). In D. crumenatum, however, exogenous BA was found to have flower abortion-inductive effects (Table 7). To our knowledge, this is the first report on the ability of exogenous BA to cause abortion of flowering during the final stages of floral bud development. In the opening of flowers, carbohydrate traffic and osmolarity status are considered important driving forces behind petal movements (van Doorn and van Meeteren 2003). It could be possible that the exogenous application of BA to excised D. crumenatum floral buds, just one day prior to flower opening, caused severe changes to the partitioning and metabolism of carbohydrates in the floral buds. Such changes would result in incomplete sepal/ petal expansion that is required to evoke floral bud crack and flower opening (van Doorn and van Meeteren 2003). Further investigations are required to elucidate this possibility. The data also suggested that flower opening and anthesis are active processes that involve the fine concerted balance of various regulators. In D. crumenatum floral buds that were already committed to anthesis, alterations to this balance as late as during the final stages of floral bud development, just prior to the onset of flower opening, might have caused total disruption to the flower opening processes. While exogenous application of GA to harvested D. crumenatum floral buds demonstrated no effects on flower opening, treatment of excised iris (Iris x hollandica cv. Blue Magic) with GA promoted flower opening (Celikel and van Doorn 1995). Similarly, pulsing with GA in cut Polianthes tuberosa L. cv. Double also enhanced flower opening (Su et al. 2001). It was suggested that GA-stimulated ethylene production resulted in the promotion of flower bud opening (Su et al. 2001). In D. crumenatum, the applications of BA 98 or GA, alone or in combinations, produced different results in the reduction of flower opening and induction of bud dormancy (Table 6, 7). The stimulatory or inhibitory effects that BA and GA could have on each another were also demonstrated in the flowering of Miltoniopsis orchid hybrids (Matsumoto Brower 2006). In the study, inhibitory effects of BA on flowering were rescued by the addition of GA (Matsumoto Brower 2006). Furthermore, in Paphiopedilum (Macabre x glanduliferum), the stimulatory effects of GA on flowering were shown to be reduced by BA (Miguel et al. 2008). AOA + glucose treatments have been shown to promote flower opening and suppress senescence in D. ‘Jew Yuan Tew’ inflorescences (Rattanawisalanon et al. 2003). Longevity of pollinated D. ‘Heang Beauty’ flowers also increased with AOA + glucose treatments (Chandran et al. 2006). In D. crumenatum, however, effects of AOA + glucose on flower opening and senescence of flowers were not observed (Table 6). The differences in AOA + glucose effects observed in the previous two studies could be due to the presence of more sinks (floral buds/ flowers), resulting in partitioning and dilution of AOA and/ or glucose, and pollination, resulting in changes in overall physiology, respectively. It is interesting that AOA + glucose treatments resulted in an abnormal senescing phenotype in treated D. crumenatum floral buds, where the sepals and petals turned translucent but remained fully open (Fig. 22E, F, H). AOA has been proposed to act as an anti-microbial agent, enabling the continuous uptake of water in flowers (Rattanawisalanon et al. 2003; Chandran et al. 2006). The resultant increases in water and glucose contents in D. crumenatum could have resulted in the maintenance of the fully open stance, even in senescing flowers. 99 5.6 Further works Cell wall modifications are very diverse (Brummell 2006), and it is highly possible that other cell wall changes, such as depolymerization of hemicelluloses and pectins, occur and are important in regulating floral bud development, flower opening and senescence in D. crumenatum. The presence and activities of various other cell wall enzymes, such as glycanases and expansin, also remain to be elucidated. The identification of these cell wall modifications would provide a clearer understanding of the importance and roles that cell walls have in regulating flower development. Also, investigating the possible roles of nonenzymic mechanisms in controlling cell wall alterations could provide further insight to the involvement of the cell wall in controlling flowering. Whether variations in patterns of cell wall modifications are the cause for the diversity in the rates of floral bud development, time of flower opening and longevity of the flowers in different species remain to be investigated. Furthermore, determining the presence of changes in patterns of cell wall modifications between BA-induced dormant/aborted floral buds and untreated wild-type floral buds may provide useful information on the role of BA in controlling flower development in D. crumenatum. 100 Chapter 6. Conclusion Co-ordinated cell wall modifications, such as alterations in specific cell wall components and activities of cell wall-based enzymes, have important roles in the regulation of floral bud development, flower opening and flower senescence in D. crumenatum. In particular, floral bud development in D. crumenatum appeared to be accompanied by the loosening of cellulose/ hemicellulose and pectin networks, and solubilisation of the middle lamella. Such events probably resulted in decreases in cell-cell adhesion, loosening of the cell wall matrix, and increases in mobility of cell wall-based enzymes and their access to the respective substrates. In D. crumenatum sepals and petals, breakdown or total hydrolysis of the cell wall only occurred upon flower opening and flower senescence. Expression of the cell wall enzyme transcripts, however, did not correspond with the activity profiles of the enzymes, and this might be explained by the presence of more than one gene coding for each enzyme. Senescence-indicative events such as cell wall modifications and decreases in membrane stability were found to commence during early floral bud development in D. crumenatum, suggesting the early onset of a senescence programme. This study demonstrated that flower opening in mature D. crumenatum floral buds could be reduced, halted or aborted in the presence of exogenous BA. The importance of the cytokinin in regulating floral bud development, flower opening, anthesis and the various processes which precede senescence, and the mechanism through which it alters flower opening remains to be determined. Specifically, whether BA induced delayed and/ or aborted flowering occurred due to changes in patterns of cell wall modifications is of interest. 101 Chapter 7. References Abu-Bakr A, Abu-Goukh, Hind AB (2003) Changes in pectic enzymes and cellulase activity during guava fruit ripening. Food Chemistry 83, 213 – 218. Ahmed AER, Labavitch JM (1977) A simple method for accurate determination of cell wall uronide content. Journal of Food Chemistry 1, 361 – 365. Ahmed AER, Labavitch JM (1980) Cell wall metabolism in ripening pear. I. Cell wall changes in ripening Bartlett pears. Plant Physiology 65, 1009 -1013. Ali ZM, Armugam S, Lazan H (1995) β-galactosidase and its significance in ripening mango fruit. Phytochemistry 38, 1109 – 1114. Ali ZM, Ng SY, Othman R, Goh LY, Lazan H (1998) Isolation, characterization and significance of papaya β-galactanases to cell wall modification and fruit softening during ripening. Physiologia Plantarum 104, 105 – 115. Almeida DPF, Huber DJ (1999) Apoplastic pH and inorganic ion levels in tomato fruit: a potential means for regulating cell wall metabolism during ripening. Physiologia Plantarum 105, 506 – 512. Baron-Epel O, Gharyal PK, Schindler M (1988) Pectins as mediators of wall porosity in soybean cells. Planta 175, 389 – 395. Bartlett AA (1997) Reflections on sustainability, population growth and the environment – revisited. Renewable Resources Journal 15, 6 – 23. Bernier G (1988) The control of floral evocation and morphogenesis. Annual Review of Plant Physiology and Plant Molecular Biology 39, 175 – 219. Bernier G, Havelange A, Houssa C, Petikean A, Lejeune P (1993) Physiological signals that induce flowering. Plant Cell 5, 1147 – 1155. Bieleski RL (1993) Fructan hydrolysis drives petal expansion in the ephemeral daylily flower. Plant Physiology 103, 213 – 219. Bieleski RL, Elgar J, Heyes J (2000) Mechanical aspects of rapid flower opening in Asiatic lily. Annals of Botany 86, 1175 – 1183. Bieleski RL, Reid MS (1992) Physiological changes accompanying senescence in the ephemeral daylily flower. Plant Physiology 98, 1042 – 1049. Blumenkrantz N, Asboe-Hansen G (1973) New method for quantitative determination of uronic acids. Analytical Biochemistry 54, 484 – 489. 102 Boo CM, Kartini OH, Oh-Yang CL (2006) 1001 Garden Plants in Singapore. Singapore: National Parks Board. Bordenave M, Goldberg R (1993) Purification and characterization of pectin methylesterases from mung bean hypocotyls cell walls. Phtyochemistry 33, 999 -1003. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principles of protein-dye binding. Analytical Biochemistry 72, 248 – 259. Brummell DA (2006) Cell wall disassembly in ripening fruit. Functional Plant Biology 33, 103 – 119. Brummell DA, Dal Cin V, Crisoto CH, Labavitch JM (2004) Cell wall metabolism during maturation, ripening and senescence of peach fruit. Journal of Experimental Botany 55, 2029 – 2039. Brummell DA, Harpster MH (2001) Cell wall metabolism in fruit softening and quality and its manipulation in transgenic plants. Plant Molecular Biology 47, 311 – 339. Brummell DA, Lashbrook CC, Bennett AB (1994) Plant eno-1,4-β-D-glucanases: structure, properties and physiological function. In ‘Enzymatic conversion of biomass for fuels production (American Chemical Society Symposium Series 566)’. (Eds ME Himmel, JO Baker, RP Overend) pp. 100 – 129. Oxford: Oxford University Press. Buckeridge MS, Pessoa dos Santos H, Tine MAS (2000) Mobilization of storage cell wall polysaccharides in seeds. Plant Physiology and Biochemistry 38, 141 – 156. Buckeridge MS, Rocha DC, Reid JSG, Dietrich SMC (1992) Xyloglucan structure and postgerminative metabolism in seeds of Copaifera langsdorffi from savanna and forest populations. Physiologia Plantarum 86, 145 – 151. Burger WC (2006) Flowers: how they changed the world. New York: Promesthus Books. Carpita NC, Gibeaut DM (1993) Structural models of primary cell walls in flowering plants: consistency of molecular structure with the physical properties of the walls during growth. The Plant Journal 3, 1 – 30. Carpita NC, McCann MC (2000) The cell wall. In ‘Biochemistry and molecular biology of plants’. (Eds BB Buchanan, W Gruissem, RL Jones) pp. 52 – 108. Rockville: American Society of Plant Physiologist. Carrington CMS, Greve LC, Labavitch JM (1993) Cell wall metabolism in ripening fruit. VI. Effect of the antisense polygalacturonase gene on cell wall changes accompanying ripening in transgenic tomatoes. Plant Physiology 103, 429 – 434. 103 Catoire L, Pierron M, Morvan C, du Penhoat CH, Goldberg R (1998) Investigation of the action patterns of pectin methylesterase isoforms through kinetic analyses and NMR spectroscopy. Implications in cell wall expansion. Journal of Biological Chemistry 50, 33150 – 33156. Celikel FG, van Doorn WG (1995) Effects of water stress and gibberellin on flower opening in Iris × hollandica. Acta Horticulturae 405, 246 – 252. Chandran S, Toh CL, Zuliana R, Yip YK, Nair H, Boyce AN (2006) Effects of sugars and aminooxyacetic acid on the longevity of pollinated Dendrobium (Heang Beauty) flowers. Journal of Applied Horticulture 8, 117 – 120. Chin LH, Mohd Ali Z, Lazan H (1999) Cell wall modifications, degrading enzymes and softening of carambola fruits during ripening. Journal of Experimental Botany 50, 767 – 777. Chun JP, Huber DJ (1998) Polygalacturonase-mediated solubilisation and depolymerization of pectic polymers in tomato fruit cell walls. Regulation by pH and ionic conditions. Plant Physiology 117, 1293 – 1299. Cleland CF (1982) The chemical control of flowering – a status report. In ‘Plant growth substances’. (Ed PF Wareing) pp. 635 – 644. London: Academic Press. Collier DE (1997) Changes in respiration, protein and carbohydrates of tulip tepals and Alstroemeria petals during development. Journal of Plant Physiology 150, 446 – 451. Corbesier L, Coupland G (2006) The quest for florigen: a review of recent progress. Journal of Experimental Botany 57, 3395 – 3403. Corner EJH (1988) Wayside Trees of Malaya Vol. 1, 3rd edn. Malaysia: United Selangor Press. Cosgrove DJ (1999) Enzymes and other agents that enhance cell wall extensibility. Annual Review of Plant Physiology and Plant Molecular Biology 50, 391 – 417. Crombie HJ, Chengappa S, Hellyer A, Reid JSG (1998) A xyloglucan oligosaccharideactive, transglycosylating β-D-glucosidase from the cotyledons of nasturtium (Tropaeolum majus L.) seedlings – purification, properties and characterization of a cDNA clone. Plant Journal 15, 27 – 38. Davies G, Henrissat B (1995) Structure and mechanism of glycosyl hydrolases. Structure 3, 853 – 859. Day JS, Loveys BR, Aspinall D (1995) Cytokinin and carbohydrate changes during flowering of Boronia megastigma. Australian Journal of Plant Physiology 22, 57 – 65. 104 de Vetten NC, Huber DJ (1990) Cell wall changes during the expansion and senescence of carnation (Dianthus caryophyllus) petals. Physiologia Plantarum 78, 447 – 454. de Vetten NC, Huber DJ, Gross KC (1991) Endoglycanase-catalysed degradation of hemicelluloses during development of carnation (Dianthus caryophyllus L.) petals. Plant Physiology 95, 853 – 860. Della Penna D, Kates DS, Bennett AB (1987) Polygalacturonase gene expression in Rutgers, rin, nor, and Nr tomato fruits. Plant Physiology 85, 502 – 507. Della Penna D, Lashbrook CC, Toenjes K, Giocannoni JJ, Fischer RL, Bennett AB (1990) Polygalacturonase isozymes and pectin depolymerization in transgenic rin tomato fruit. Plant Physiology 94, 1882 – 1886. Della Penna D, Lincoln JE, Fischer RL, Bennett AB (1989) Transcriptional analysis of polygalacturonase and other ripening associated genes in Rutgers, rin, nor, and Nr tomato fruit. Plant Physiology 90, 1372 – 1377. Denes JM, Baron A, Renard CM, Pean C, Drilleau JF (2000) Different action pattern for apple pectin methylesterase at pH 7.0 and 4.5. Carbohydrate Research 327, 385 – 393. Deng Y, Wu Y, Li Y (2005) Changes in firmness, cell wall composition and cell wall hydrolases of grapes stored in high oxygen atmospheres. Food Research International 38, 769 – 776. Dumville JC, Fry SC (2003) Solubilisation of tomato fruits pectins by ascorbate: a possible non-enzymic mechanism of fruit softening. Planta 217, 951 – 961. Eason JR, Webster D (1995) Development and senescence of Sandersonia aurantiaca (Hook.) flowers. Scientia Horticulturae 63, 113 – 121. Edwards M, Dea ICM, Bulpin PV, Reid JSG (1985) Xyloglucan (amyloid) mobilization in the cotyledons of Tropaeolum majus L. seeds following germination. Planta 163, 133 – 140. Espinosa A, Harden R, Walker J (2008) A complexity approach to sustainability – Stafford Beer revisited. European Journal of Operational Research 187, 636 – 651. Evans RY, Reid MS (1988) Changes in carbohydrates and osmotic potential during rhythmic expansion of rose petals. Journal of the American Society of Horticultural Science 113, 884 – 888. Ferrari TE, Arnison PG (1974) Extraction and partial characterization of cellulases from expanding pea epicotyls. Plant Physiology 54, 487 – 493. 105 Fischer RL, Bennett AB (1991) Role of cell wall hydrolases in fruit ripening. Annual Review of Plant Physiology and Plan Molecular Biology 42, 675 – 703. Francis KE, Lam SY, Copenhaver GP (2006) Separation of Arabidopsis pollen tetrads is regulated by QUARTET 1, pectin methylesterase gene. Plant Physiology 142, 1004 – 1013. Fry SC (1986) Cross-linking of matrix polymers in the growing cell walls of angiosperms. Annual Review of Plant Physiology 37, 165 – 186. Fulton LM, Cobbett CS (2003) Two alpha-L-arabinofuranosidase genes in Arabidopsis thaliana are differentially expressed during vegetative growth and flower development. Journal of Experimental Botany 54, 2467 – 2475. Futamura N, Mori H, Kouchi H, Shinohara K (2000) Male flower-specific expression of genes for polygalacturonase, pectin methylesterase and β-1,3-glucanase in a dioecious willow (Salix gilgliana Seemen). Plant Cell and Physiology 41, 151 – 160. Gaffe J, Tiznado ME, Handa AK (1997) Characterization and functional expression of a ubiquitously expressed tomato pectin methylesterase. Plant Physiology 114, 1547 – 1556. Geertsen V (1990) The keeping quality of Ortacanthus coeruleus – a potential new cut flower. Scientia Horticulturae 43, 145 – 153. Ghose TK (1987) Measurement of cellulase activities. Pure and Applied Chemistry 59, 257 – 268. Goh CJ (1979) Hormonal regulation of flowering in a sympodial orchid hybrid Dendrobium Lousiae. New Phytologist 82, 375 – 380. Goldberg R, Morvan C, Jauneau A, Jarvis MC (1996) Methyl-esterification, de-esterification and gelation of pectins in the primary cell wall. In ‘Pectins and Pectinases: Progress in Biotechnology’. (Eds J Visser, Voagen AGJ) pp. 151 – 172. Amsterdam: Elsevier Science. Goulao LF, Santos J, de Sousa I, Oliveria CM (2007) Patterns of enzymatic activity of cell wall-modifying enzymes during growth and ripening of apples. Postharvest Biology and Technology 43, 307 – 318. Grignon C, Sentenac H (1991) pH and ionic conditions in the apoplast. Annual Review of Plant Physiology and Plant Molecular Biology 42, 103 – 128. Gross KC, Wallner SJ (1979) Degradation of cell wall polysaccharides during tomato fruit ripening. Plant Physiology 63, 117 – 120. 106 Hadfield KA, Bennett AB (1998) Polygalacturonases: many genes in search of a function. Plant Physiology 117, 363 – 373. Hagerman AE, Austin PJ (1986) Continuous spectrophotometric assay for plant pectin methyl esterase. Journal of Agricultural and Food Chemistry 34, 440 – 444. Hall LN, Bird CR, Picton S, Tucker GA, Seymour GB, Grierson D (1994) Molecular characterization of cDNA clones representing pectin esterase isozymes from tomato. Plant Molecular Biology 25, 313 – 318. Harriman RW, Tieman DM, Handa AK (1991) Molecular cloning of tomato pectin methylesterasegene and its expression in Rutgers. Ripening inhibitor, nonripening, and Never Ripe tomato fruits. Plant Physiology 97, 80 – 87. Hayashi T, Yoshida K, Park YW, Konishi T, Baba K (2005) Cellulose metabolism in plants. International Review of Cytology 247, 1 – 34. Hew CS, Yong WH (2004) The Physiology of Tropical Orchids in Relation to the Industry, 2nd edn. Singapore: World Scientific. Holttum RE (1953) Flora of Malaya – Orchids of Malaya. Singapore: Govt. Printing Office. Hrmova M, Fincher GB (1998) Barley β-D-glucan exohydrolases. Substrate specificity and kinetic properties. Carbohydrate Research 305, 209 – 221. Hrmova M, Fincher GB (2001) Structure-function relationships of β-D-glucan endo- and exohydrolases from higher plants. Plant Molecular Biology 47, 73 – 91. Huber DJ (1983a) Polyuronide degradation abd hemicellulose modifications in ripening tomato fruit. Journal of American Society for Horticultural Science 108, 405 – 409. Huber DJ (1983b) The role of cell wall hydrolases in fruit softening. Horticultural Review 5, 169 – 219. Huber DJ (1984) Strawberry fruit softening: the potential roles of polyuronides and hemicelluloses. Journal of Food Science 49, 1310 – 1315. Huber DJ (1992) The inactivation of pectin depolymerase associated with isolated tomato fruit cell wall: implications for the analysis of pectin solubility and molecular weight. Physiologia Plantarum 86, 25 – 32. Huberman M, Goren R, Birk Y (1975) The effects of pH, ionic strength, and ethylene on the extraction of cellulase from abscission zones of citrus leaf extracts. Plant Physiology 55, 941 – 945. 107 Hughes A (2000) Retailers, knowledges and changing commodity networks: the case of the cut flower trade. Geoforum 31: 175 – 190. Ichimura K, Suto K (1998) Environmental factors controlling flower opening and closing in a Portulaca hybrid. Annals of Botany 82, 67 – 70. Iiyama K, Lam TM-T, Stone BA (1994) Covalent cross-links in the cell wall. Plant Physiology 104, 315 – 320. Imoto K, Yokoyama R, Nishitani K (2005) Comprehensive approach to genes involved in cell wall modifications in Arabidopsis thaliana. Plant Molecular Biology 58, 177 – 192. Itai A, Yoshida K, Tanabe K, Tamura F (1999) A β-D-xylosidase-like gene is expressed during fruit ripening in Japanese pear (Pyrus pyrifolia Nakai). Journal of Experimental Botany 50, 877 – 878. Jarvis MC (1984) Structure and properties of pectin gels in plant cell walls. Plant, Cell and Environment 26, 977 – 989. Jordi W, Stoopen M, Kelepouris K, van der Krieken WM (1995) Gibberellin-induced delay of leaf senescence of Alstromeria cut flowering stems is not caused by an increase in the endogenous cytokinin content. Journal of Plant Growth Regulation 14, 121 – 127. Kaihara S, Takimoto A (1981) Effects of light and temperature on flower-opening in Pharbitis nil. Plant and Cell Physiology 32, 21 – 26. Kim DS, Infante Ferreira CA (2008) Solar refrigeration option – a state-of-the-art review. International Journal of Refrigeration 31, 3 – 15. Kim HJ, Miller WB (2008) Effects of GA4+7 and benzyladenine application on postproduction quality of ‘Seadov’ pot tulip flowers. Postharvest Biology and Technology 47, 416 – 421. Knapp J, Moureau P, Schuch W, Grierson D (1989) Organization and expression of polygalacturonase and other ripening related genes in Ailsa Craig ‘Neverripe’ and ‘Ripening inhibitor’ tomato mutants. Plant Molecular Biology 12, 105 – 116. Koslanund R, Archbold DD, Pomper KW (2005) Pawpaw (Asimina triloba L. Dunal) fruit ripening. II. Activity of selected cell-wall degrading enzymes. Journal of American Society for Horticultural Science 130, 643 – 648. Lashbrook CC, Gonzalez-Bosch C, Bennett AB (1994) Two divergent endo-β-1,4-glucanase genes exhibit overlapping expression in ripening fruit and abscising flowers. Plant Cell 6, 1485 – 1493. 108 Lazan H, Mohd Ali Z, Liang KS, Yee KL (1989) Polygalacturonase activity and variation in ripening of papaya fruit with tissue depth and heat treatment. Physiologia Plantarum 77, 93 – 98. Liepman AH, Cavalier DM, Lerouxel O, Keegrsta K (2007) Cell wall structure, biosynthesis and assembly. In ‘Plant cell separation and adhesion’. (Eds JA Roberts, Z GonzalezCarranza) pp. 8 – 39. Oxford: Blackwell Publishing. Lohani S, Trivedi PK, Nath P (2004) Changes in activities of cell wall hydrolases during ethylene-induced ripening in banana: effect of 1-MCP, ABA and IAA. Postharvest Biology and Technology 31, 119 – 126. Lukaszewska AJ, Bianco J, Barthe P, Page-Degivry MTI (1994) Endogenous cytokinins in rose petals and the effect of exogenously applied cytokinins on flower senescence. Plant Growth Regulation 14, 119 – 126. Maclachlan G, Brady C (1994) Endo-1,4-β-glucanase, xyloglucanase and xyloglucan endotransglycosylase activities versus potential substrates in ripening tomatoes. Plant Physiology 105, 965 – 974. Maricle BR, Cobos DR, Campbell CS (2007) Biophysical and morphological leaf adaptations to drought and salinity in salt marsh grasses. Environmental and Experimental Botany 60, 458 – 467. Markovic O, Kohn R (1984) Mode of pectin deesterification by Trichoderma reesei pectinesterase. Experientia 40, 842 – 843. Martinez GA, Chaves AR, Civello PM (2004) β-xylosidase activity and expression of βxylosidase gene during strawberry fruit ripening. Plant Physiology and Biochemistry 42, 89 – 96. Matsumoto Brower TK (2006) Gibberellic acid and benzyladenine promote early flowering and vegetative growth of Miltoniopsis orchid hybrids. HortScience 41, 131 – 135. McClendon JH, Nolan WG, Wenzler HF (1976). The role of the endosperm in the germination of legumes: galactomannan, nitrogen and phosphorus changes in the germination of guar (Cyamopsis tetragonolobus, Leguminosae). American Journal of Botany 63, 790 – 797. McCollum TG, Huber DJ, Cantlife DJ11 (1989) Modification of polyuronides and hemicelluloses during muskmelon fruit softening. Physiologia Plantarum 76, 303 – 308. Micheli F (2001) Pectin methylesterases: cell wall enzymes with important roles in plant physiology. TRENDS in Plant Science 6, 414 – 419. 109 Micheli F, Holliger C, Goldberg G, Richard L (1998) Characterization of the pectin methylesterase-like gene AtPME3: a new member of a gene family comprising at least 12 genes in Arabidopsis thaliana. Gene 220, 13 – 20. Miguel TP, Sakai WS, Fang J (2008) Gibberellic acid induced flowering of Paphiopedilum (Macabre x glanduliferum). Acta Horticulturae 766, 279 – 282. Minic Z (2008) Physiological roles of plant glycoside hydrolases. Planta 227, 723 – 740. Minic Z, Jouanin L (2006) Plant glycoside hydrolases involved in cell wall polysaccharide degradation. Plant Physiology and Biochemistry 44, 435 – 449. Minic Z, Rihouey C, Do CT, Lerouge P, Jouanin L (2004) Purification and characterization of enzymes exhibiting beta-D-xylosidase activities in stem tissues of Arabidopsis. Plant Physiology 135, 867 – 878. Molhoj M, Ulvskov P, Dal Degan F (2001) Characterization of a functional soluble form of a Brassica napus membrane-anchored endo-1,4-beta-glucanase heterologously expressed in Pichia pastoris. Plant Physiology 127, 674 – 684. Mor Y, Spiegelstein H, Halevy AH (1983) Inhibition of ethylene biosynthesis in carnation petals by cytokinin. Plant Physiology 71, 541 – 546. Munoz JL, Martin L, Nicholas G, Villalobos N (1990) Influence of endogenous cytokinins on reverse mobilization in cotyledons of Cicer arietinum L. Plant Physiology 93, 1011 – 1016. Nagel L, Brewster R, Riedell WE, Reese RN (2001) Cytokinin regulation of flower and pod set in soybeans (Glycine max (L.) Merr.). Annals of Botany 88, 27 – 31. Nell TA (2007) Floriculture research in the 21st century: It’s more than just science. Acta Horticulturae 755, 27 – 32. O’Donoghue EM (2006) Flower petal cell walls: changes associated with opening and senescence. New Zealand Journal of Forestry Science 36, 134 – 144. O’Donoghue EM, Eason JR, Somerfield SD, Ryan DA (2005) Galactosidases in opening, senescing and water-stressed Sandersonia aurantiaca flowers. Functional Plant Biology 32, 911 – 922. O’Donoghue EM, Somerfield SD, de Vre LA, Heyes JA (1997) Developmental and ripeningrelated effects on the cell wall of pepino (Solanum muricatum) fruits. Journal of the Science of Food and Agriculture 73, 455 – 463. O’Donoghue EM, Somerfield SD, Heyes JA (2002) Organization of cell walls in Sandersonia aurantiaca floral tissue. Journal of Experimental Botany 53, 513 – 523. 110 O’Donoghue EM, Somerfield SD, Sinclair BK, King GA (1998) Characterization of the harvest-induced expression of β-galactosidase in Asparagus officinalis. Plant Physiology and Biochemistry 36, 721 – 729. Ohmiya Y, Takeda T, Nakamura S, Sakai F, Hayashi T (1995) Purification and properties of a wall-bound endo-1,4-β-glucanase from suspension-cultured poplar cells. Plant and Cell Physiology 36, 607 – 614. Panavas T, Reid PD, Rubinstein B (1998) Programmed cell death of daylily petals: activities of wall-based enzymes and effects of heat shock. Plant Physiology and Biochemistry 26, 379 – 388. Panavas T, Rubinstein B (1998) Oxidative events during programmed cell death of daylily (Hemerocallis hybrid) petals. Plant Science 133, 125 – 138. Panavas T, Walker EL, Rubinstein E (1998) Possible involvement of abscisic acid in senescence of daylily petals. Journal of Experimental Botany 49, 1987 – 1997. Patterson SE (2001) Cutting loose. Abscission and dehiscence in Arabidopsis. Plant Physiology 126, 494 – 500. Pauly M, Albersheim P, Darvill A, York WS (1999) Molecular domains of the cellulose/ xyloglucan network in the cell walls of higher plants. The Plant Journal 20, 629 – 639. Pena MJ, Carpita NC (2004) Loss of highly branched arabinans and debranching of rhamnogalacturonan I accompanying loss of firm tecture and cell separation during prolonged storage of apple. Plant Physiology 135, 1305 – 1313. Phillips HL, Kende H (1980) Structural changes in flowers of Ipomoea tricolor during flower opening and closing. Protoplasma 102, 199 – 215. Popper ZA, Fry SC (2005) Widespread occurrence of a covalent linkage between xyloglucan and acidic polysaccharides in suspension-cultured angiosperms cells. Annals of Botany 96, 91 – 99. Pressey R (1983) β-galactosidase in ripening tomatoes. Plant Physiology 71, 132 – 135. Qi XY, Behrens BX, West PR, Mort AJ (1995) Solubilization and partial characterization of extensin fragments from cell walls of cotton suspension cultures: evidence for a covalent cross-link between extensin and pectin. Plant Physiology 108, 1691 – 1701. Raab MM, Koning RE (1987) Interacting roles of gibberellin and ethylene in corolla expansion of Ipomoea nil (Convolvulaceae). American Journal of Botany 74, 921 – 927. 111 Rattanawisalanon C, Ketsa S, van Doorn WG (2003) Effect of aminooxyacetic acid and sugars on the vase life of Dendrobium flowers. Postharvest Biology and Technology 29, 93 – 100. Redgwell RJ, MacRae EA, Hallett I, Fischer M, Perry J, Harker R (1997) In vivo and in vitro swelling of cell walls during fruit ripening. Planta 203, 162 – 173. Redman PB, Dole JM, Maness NO, Anderson JA (2002) Postharvest handling of nine specialty cut flower species. Scientia Horticulturae 92, 293 – 303. Reese RN, Dybing CD, White CA, Page SM, Larson JE (1995) Expression of vegetative protein (VSP-β) in soybean raceme tissues in response to flower set. Journal of Experimental Botany 46, 957 – 964. Reid JSG (1971) Reserve carbohydrate metabolism in germinating seeds of Trigonella foenum-graecum L. (Legum). Planta 100, 131 – 142. Reid JSG (1985) Structure and function in legume-seed polysaccharides. In ‘Biochemistry of the plant cell walls’. (Eds C Brett, R Hillman) pp. 259 – 268. Cambridge: Cambridge University Press. Ren C, Kermode AR (2000) An increase in pectin methylesterase activity accompanies dormancy breakage and germination of yellow cedar seeds. Plant Physiology 124, 231 – 242. Rhizopoulou S, Ioannidi E, Alexandrades N, Argiropoulos A (2006) A study on functional and structural traits of the nocturnal flowers of Capparis spinosa L. Journal of Arid Environments 66, 635 – 647. Richards D (1985) Effects of cytokinin application and environment on growth and flowering of Boronia heterophylla F. Meull. Scientia Horticulturae 27, 325 – 334. Ronen R, Zauberman G, Akerman M, Weksler A, Rot I, Fuchs Y (1991) Xylanase and xylosidase activities in avocado fruit. Plant Physiology 95, 961 – 964. Rose JKC, Bennett AB (1999) Cooperative disassembly of the cellulose-xyloglucan network of plant cell walls: parallels between cell expansion and fruit ripening. Trends in Plant Science 4, 176 – 183. Rose JKC, Hadfield KA, Labavitch JM, Bennett AB (1998) Temporal sequence of cell wall disassembly in rapidly ripening melon fruit. Plant Physiology 117, 345 – 361. Roy S, Vian B, Roland JC (1992) Immunocytochemical study of the deesterification patterns during cell wall autolysis in the ripening of cherry tomato. Plant Physiology and Biochemistry 30, 139 – 146. 112 Rubinistein B (2000) Regulation of cell death in flower petals. Plant Molecular Biology 44, 303 – 318. Saha S, Naga PK, Sircar PK (1985) Changes in cytokinin activity during flower development in Cosmos sulphureus Cav. Plant Growth Regulation 3, 27 – 35. Saito M, Yamaki T (1967) Retardation of flower opening in Oenothera lamarckiana caused by blue and green light. Nature 214, 1027. Saks Y, van Staden J (1993) Evidence for the involvement of gibberellins in developmental phenomena associated with carnation flower senescence. Plant Growth Regulation 12, 105 – 110. Sakurai N, Nevins DJ (1997) Relationship between fruit softening and wall polysaccharides in avocado (Persea americana Mill) mesocarp tissues. Plant and Cell Physiology 38, 603 – 610. Scott TA, Melvin EH (1953) Determination of dextran with anthrone. Analytical Chemistry 25, 1656 – 1661. Seidenfaden G, Wood JJ (1992) The Orchids of Peninsular Malaysia and Singapore. Fredensborg: Olsen and Olsen. Sethu KMP, Prabha TN, Tharanathan RN (1996) Post-harvest biochemical changes assicaited with the softening phenomenon in Capsicum annum fruits. Phytochemistry 42, 961 – 966. Setyadjit, Joyce DC, Irving DE, Simons DH (2004) Effects of 6-benzylaminopurine treatments on the longevity of harvested Grevillea ‘Sylvia’ inflorescences. Plant Growth Regulation 43, 9 -14. Setyadjit, Joyce DC, Irving DE, Simons DH (2006) Vase treatments containing gibberellic acid fo not increase longevity of cut Grevillea ‘Sylvia’ inflorescences. Australian Journal of Experimental Agriculture 46, 1535 – 1539. Seymour GB, Harding SE, Taylor AJ, Hobson GE, Tucker GA (1987) Polyuronide solubilisation during ripening of normal and mutant tomato fruit. Phytochemistry 26, 1871 – 1875. Sim GE, Goh CJ, Loh CS (2008) Induction of in vitro flowering in Dendrobium Madame Thong-In (orchidaceae) seedlings in associated with increase in endogenous N6-(Δ2isopentenyl)-adenine (iP) and N6-(Δ2-isopentenyl)-adenosine (iPA) levels. Plant Cell Reports 27, 1281 – 1289. Smith DL, Gross KC (2000) A family of at least seven β-galactosidase genes is expressed during tomato fruit development. Plant Physiology 123, 1173 – 1183. 113 Spala A, Bagiorgas HS, Assimakopoulos MN, Kalavrouziotis J, Matthopoulos D, Mihalakakou G (2008) On the green roof system. Selection, state of the art and energy potential investigation of a system installed in an office building in Athens, Greece. Renewable Energy 33, 173 – 177. Stephenson MB, Hawes MC (1994) Correlation of pectin methylesterase activity in root caps of pea with root border cell separation. Plant Physiology 106, 739 – 745. Su WR, Huang KR, Chang PS, Chen WS (2001) Improvement of postharvest vase life and flower bud opening in Polianthes tuberosa using gibberellic acid and sucrose. Australian Journal of Experimental Agriculture 41, 1227 – 1230. Sun DW, Brosnan T (1999) Extension of the vase life of cut daffodil flowers by rapid vacuum cooling. International Journal of Refrigeration 22, 472 – 478. Sun DW, Brosnan T (2001) Precooling techniques and applications for horticultural products. International Journal of Refrigeration 24, 154 – 170. Tan HTW, Hew CS (1993) A Guide to the Orchids of Singapore. Singapore: Science Centre. Tan H, Liu X, Ma N, Xue J, Lu W, Bai J, Gao J (2006) Ethylene-influenced flower opening and expression of genes encoding Etrs, Ctrs, and Ein3s in two cut rose cultivars. Postharvest Biology and Technology 40, 97 – 105. Taverner E, Letham DS, Wang J, Cornish E, Willcocks DA (1999) Influence of ethylene on cytokinin metabolism in relation to Petunia corolla senescence. Phytochemistry 51, 341 – 347. Teoh ES (2005) Orchids of Asia. Singapore: Times Editions. The Arabidopsis Genome Initiative (2000) Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408, 796 – 815. Thompson JE, Fry SC (2000) Evidence for covalent linkages between xyloglucan and acidic pectins in suspension-cultured rose cells. Planta 211, 275 – 286. Tieman DM, Harriman RW, Ramamohan G, Handa AK (1992) An antisense pectin methylesterase gene alters pectin chemistry and soluble solids in tomato fruit. Plant Cell 4, 667 – 679. Tine MAS, Cortelazzo AL, Buckeridge MS (2000) Xyloglucan mobilization in cotyledons of developing plantlets of Hymenaea courbaril L. (Leguminosae-Caesalpinoideae). Plant Science 154, 117 – 126. 114 Tuna AV, Kaya C, Ashraf M, Altunlu H, Yokas I, Yagmur B (2007) The effects of calcium sulphate on growth, membrane stability and nutrient uptake of tomato plants grown under salt stress. Environmental and Experimental Botany 59, 173 – 178. Turner LA, Harriman RW, Handa AK (1996) Isolation and nucleotide sequence of three tandemly arranged pectin methylesterase genes (Accession Nos. U70675, U70676 and U70677) from tomato. Plant Physiology 112, 1398. Updegraff DM (1969) Semimicro determination of cellulose in biological materials. Analytical Biochemistry 32, 420 – 424. Urbanowicz BR, Catala C, Irwin D, Wilson DB, Ripoli DR, Rose JKC (2007) A tomato endo-β-1,4-glucanase, SlCel9C1, represents a distinct subclass with a new family of carbohydrate binding modules (CBM49). Journal of Biological Chemistry 282, 12066 – 12074. van Doorn WG (1997) Effects of pollination on floral attraction and longevity. Journal of Experimental Botany 48, 1615 – 1622. van Doorn WG, van Meeteren U (2003) Flower opening and closure: a review. Journal of Experimental Botany 54, 1801 – 1812. van Meeteren U (2007) Why do we treat flowers the way we do? A system analysis approach of the cut flower postharvest chain. Acta Horticulturae 755, 61 – 73. Vergauwen R, van den Ende W, van Laere A (2000) The role of fructan in flowering of Campanula rapunculoides. Journal of Experimental Botany 51, 1261 – 1266. Vicente AR, Lowell A, Greve LC, Labavitch JM (2007) Cell wall disassembly events in boysenberry (Rubus idaeus L. × Rubus ursinus Cham. & Schldl.) fruit development. Functional Plant Biology 34, 614 – 623. Vincken JP, Schols HA, Oomen RJFJ, McCann MC, Ulvskov P, Voragen AGJ, Visser RGF (2003) If homogalacturonan were a side chain of rhamnogalacturonan I. Implications for cell wall architecture. Plant Physiology 132, 1781 – 1789. Wagstaff C, Malcolm P, Rafiq A, Leveremtz M, Griffths G, Thomas B, Stead A, Rogers H (2003) Programmed cell death (PCD) processes begin extremely early in Alstromeria petal senescence. New Phytologist 160, 49 – 59. Wakabayashi K, Chun JP, Huber DJ (2000) Extensive solubilization and depolymerization of cell wall polysaccharides during avocado (Persea americana) ripening involves concerted action of polygalacturonase and pectinmethylesterase. Physiologia Plantarum 108, 345 – 352. 115 Wakeley PR, Rogers HJ, Rozycka M, Greenland AJ, Hussey PJ (1998) A maize pectin methylesterase-like gene, ZmC5, specifically expressed in pollen. Plant Molecular Biology 37, 187 – 192. Wood PJ, Erfle JD, Teather RM (1988) Use of complex formation between congo red and polysaccharides in detection and assay of polysaccharide hydrolases. Methods in Enzymology 160, 59 – 74. Wood TM, Bhat KM (1988) Methods for measuring cellulase activities. Methods in Enzymology 160, 87 – 112. Wood WML (1953) Thermonasty in tulip and crocus flowers. Journal of Experimental Botany 4, 65 – 77. Woodson W, Brandt A (1991) Role of the gynoecium in cytokinin-induced carnation petal senescence. Journal of American Society for Horticultural Science 116, 676 – 679. Yap YM (2006) The flowering physiology of Dendrobium crumenatum. B.Sc. (Hons) dissertation. Department of Biological Sciences. The National University of Singapore. Zhou Y, Wang CY, Ge H, Hoeberichts FA, Visser PB (2005) Programmed cell death in relation to petal senescence in ornamental plants. Journal of Integrative Plant Biology 47, 641 – 650. Zieslin N, Khayat E (1990) Effects of alternating night temperature and cytokinin application on the activity of enzymes of sucrose metabolism in leaves, young shoots and petals of roses. Plant and Cell Physiology 31, 845 – 849. Zykwinska AW, Ralet MCJ, Garnier CD, Thibault JFJ (2005) Evidence for in vitro binding of pectin side chains to cellulose. Plant Physiology 139, 397 – 407. 116 Appendix A: Publication in Scientia Horticulturae ----Original Message----From: Scientia Horticulturae [mailto:sci_horti@elsevier.com] Sent: Monday, June 23, 2008 2:16 PM To: Ong Bee Lian Subject: Your Submission Ref.: Ms. No. HORTI2729R1 Regulation of flower development in Dendrobium crumenatum by changes in carbohydrate contents, water status and cell wall metabolism Scientia Horticulturae Dear Ong, I am pleased to tell you that your work has now been accepted for publication in Scientia Horticulturae. The manuscript will be transferred to our production site in Ireland for preparation for press. Proofs will be sent to you in due course. If there were any comments from the Editor and/or Reviewers, they can be found below. Thank you for submitting your work to this journal. With kind regards J.P. Bower Editor Scientia Horticulturae Comments from the Editors and Reviewers: Thank you for the revised manuscript. The paper is now ready for publication. 117 This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright Author's personal copy Scientia Horticulturae 119 (2008) 59–66 Contents lists available at ScienceDirect Scientia Horticulturae journal homepage: www.elsevier.com/locate/scihorti Regulation of flower development in Dendrobium crumenatum by changes in carbohydrate contents, water status and cell wall metabolism You-Min Yap, Chiang-Shiong Loh, Bee-Lian Ong * Department of Biological Sciences, National University of Singapore, 14 Science Drive 4, Singapore S117543, Singapore A R T I C L E I N F O A B S T R A C T Article history: Received 4 March 2008 Received in revised form 5 June 2008 Accepted 23 June 2008 The involvement of carbohydrates, water potential, cell wall components and cell wall-based enzymes in regulating flower development in Dendrobium crumenatum was investigated. Plants were subjected to cold treatment to release floral buds from dormancy, and the various parameters were investigated from young floral bud stage till flower senescence. Development of floral buds was accompanied by progressive decrease in concentrations of fructans and starch. Upon full flower opening, concentration of soluble sugars was maximum, accompanied by a more negative water potential. High pectin methylesterase activity was observed during early bud development and decreased thereafter. Significant increase in activities of b-galactosidase, b-mannosidase and b-xylosidase was also observed during floral bud development. The cell walls of sepals and petals were modified extensively during floral bud and flower development, as observed by changes in the amounts of celluloses, hemicelluloses and total pectin. Pectin solubilisation was also observed to commence during early floral bud development. These results indicated that carbohydrate hydrolysis, osmotic changes and cell wall dissolution that began early in young floral buds, all regulated flower development in this sympodial orchid. Possible applications of the findings in the horticultural industry are discussed. ß 2008 Published by Elsevier B.V. Keywords: Orchid Flower development Carbohydrates Osmolality Cell wall composition Cell wall hydrolases 1. Introduction Flowering is a critical event in the life-cycle of angiosperms, allowing for the reproduction of these plants. Flowers of various plants are also highly prized objects of beauty and are commercially valuable. Hence, understanding the various processes that regulate flower opening and senescence could help to enhance the visual quality and vase-life of flowers, and thus increasing their commercial value. Besides, with the world’s increasing interest in ‘green buildings’ to aid energy efficiency and the accompanying issue of using flowers for aesthetic benefits (Spala et al., 2008), understanding flower physiology is very important. However, publications on the physiology of tropical flowers are limited and the few detailed studies focus mainly on flowers of temperate species such as carnation (Dianthus caryophyllus L.), daylily (Hemerocallis spp.), Asiatic lily (Lilium hybrid), rose (Rosa) and sandersonia (Sandersonia aurantiaca (Hook.)). In the opening of flowers, changes in carbohydrate metabolism and cell osmolarity are considered important driving forces behind petal movements (van Doorn and van Meeteren, 2003). Rapid * Corresponding author. Tel.: +65 65162852; fax: +65 67792486. E-mail address: dbsongbl@nus.edu.sg (B.-L. Ong). 0304-4238/$ – see front matter ß 2008 Published by Elsevier B.V. doi:10.1016/j.scienta.2008.06.029 flower opening in many species, including roses (Ho and Nichols, 1977), daylily (Bieleski, 1993), Asiatic lilies (Bieleski et al., 2000) and creeping bellflowers (Campanula rapunculoides) (Vergauwen et al., 2000), was related to the hydrolysis of reserve carbohydrates. Rapid petal movements are also highly correlated with cell sap osmolarity changes, which regulate the direction of water movements, resulting in turgor changes and cell expansion (Ho and Nichols, 1977; Hew et al., 1989; Bieleski, 1993). Changes in carbohydrate metabolism and cell sap osmolarity are, therefore, intimately linked in the process of petal expansion and flower opening. In recent years, besides the roles of carbohydrates and cell osmolarity, there has been some focus on the possible involvement of cell wall metabolism in the regulation of flowering (de Vetten and Huber, 1990; O’Donoghue et al., 2002). These studies also addressed the question on whether flowering could be a process induced by the senescence programme of the plant and/or specific plant organ. Some supporting evidences included the upregulation of a phosphoenolpyruvate mutase mRNA, involved in regulating hydrolytic enzymes that resulted in membrane degradation in senescing carnation petals (Wang et al., 1996), the upregulation in levels of a mRNA that codes for proteins controlling the oxidation of membrane lipids, prior to or during flower senescence in daylily (Panavas et al., 1999), and the increasing trend of DNA laddering Author's personal copy 60 Y.-M. Yap et al. / Scientia Horticulturae 119 (2008) 59–66 throughout petal development in Alstroemeria (Wagstaff et al., 2003). While molecular evidence is increasing, structural and physiological evidences are quite limited. Studies on carnation and sandersonia flowers demonstrated that the transition of floral stages from opening to fully mature flower till senescence was accompanied by changes in the levels of various cell wall polymers such as cellulose and pectins, and activities of cell wall-based enzymes (de Vetten and Huber, 1990; O’Donoghue et al., 2002). These observations were similar to the loss of cell wall integrity in ripening fruits of carambola (Averrhoa carambola) and grapes (Vitis vinifera) (Chin et al., 1999; Deng et al., 2005). In daylily flowers, analyses of cell wall composition were not published, but reported changes in activities of cell wall-based enzymes during flower development suggested the involvement of cell wall metabolism in flowering (Panavas et al., 1998). While cellulase activity was detected in daylily flowers, it was reported to be absent in sandersonia flowers, indicating the possibility of a species-specific variation in cell wall metabolism that regulates flowering (O’Donoghue et al., 2002; Panavas et al., 1998). In the abovementioned studies on carnation and sandersonia flowers, cell wall changes were compared only between stages of late bud (just prior to opening), opening flower, mature flower, wilting flower and senesced flower (de Vetten and Huber, 1990; O’Donoghue et al., 2002). To fully understand if flowering is a consequence of a senescence programme that has already started, investigating the physiological changes occurring throughout the development of a newly induced young floral bud till flower senescence would be advantageous. Pollination and fertilisation of flowers promote sepal/petal senescence, while keeping the fertilised ovary viable; in non-pollinated and unfertilised flowers, whole flowers senesce and die (van Doorn, 1997). The possible onset of senescence prior to flower opening would thus infer a modification of the senescence programme, due to a cascade of signals generated upon pollination and fertilisation that results in the senescence of sepals/petals, but not the ovaries. Few studies on the physiology of flowering in tropical orchids have been conducted to date. As orchid cultivation continues to be a highly profitable commercial market (Hew and Yong, 2004), characterisation of tropical orchid flowering is of paramount importance. Dendrobium crumenatum (Swartz), also known as the pigeon orchid, is a common native epiphytic orchid species of South-east Asia. It exhibits an interesting and unique diversion of the normal flowering process: upon transition of the meristem from a vegetative to a reproductive phase, floral buds develop to a certain stage and then become ‘dormant’. These floral buds resume growth and development after cold-induction, such as after a heavy rainfall, and culminating into the opening of the flowers exactly nine days after (Holttum, 1953; Corner, 1988). Full flower opening is achieved before dawn and the flowers last for only 24 h under natural conditions. Senescence of the flowers is indicated by the flaccid sepals and petals. Knowledge on the physiological processes controlling dormancy release, floral bud development, flower opening and senescence in D. crumenatum can be applied to commercially important flowers where it would be advantageous to be able to control the timing of floral bud development. For example, it would be beneficial to be able to force floral buds into dormancy during shipment, releasing them from dormancy and to resume normal floral bud development when required. In this study, carbohydrates, water potential, cell wall components and activities of cell wall-based enzymes of D. crumenatum were analysed throughout the development of newly induced floral buds till flower senescence. These data would allow us a better understanding of the physiological processes, especially the possible involvement of cell wall metabolism, in the regulation of flowering. 2. Materials and methods 2.1. Plant material Dendrobium crumenatum (Swartz) plants were maintained under partially shaded conditions (PAR ranged from 100 to 250 mmol mÀ2 sÀ1; average air temperature ranged from 25 to 33 8C) in a planthouse of the Department of Biological Sciences, National University of Singapore. All plants were watered daily, and fertilised weekly with a foliar fertiliser (N18:P36:K18). Pots of D. crumenatum with inducible inflorescences carrying dormant floral buds were acclimatised at 30 8C for 24 h in temperaturecontrolled growth chambers. They were then subjected to a cold induction at 20 8C for 24 h. Growth chambers were maintained on a 12 h day/12 h night cycle and photosynthetic active radiation (PAR) ranged from 10 to 20 mmol mÀ2 sÀ1. Plants of D. crumenatum exhibit crassulacean acid metabolism, demonstrating different carbon dioxide exchange patterns during different times of the day. Thus, all plants were moved into the growth chambers at 1600 h, to minimise the effects of any possible temporal variations in the plant physiology. Plants were also watered daily to minimise dehydration stress. Floral buds or flowers for analyses of carbohydrates, water potential, cell wall composition and cell wall enzyme activities were selected according to their age and features (Table 1, Fig. 3). Sepals and petals of the harvested floral buds and flowers were separated and stored at À80 8C until use. 2.2. Carbohydrate analyses Sepals or petals (0.1 g) were boiled in 10 ml of distilled water for 90 min. The supernatant was collected as the total soluble sugar fraction. The residue was re-suspended in 10 ml of 10 mM sodium Table 1 Stages of floral bud development in D. crumenatum Features Exposure of dormant floral buds to cold induction at 20 8C for 24 h Green bud (ca. 1 cm long) with reddish brown tinges along ventral side, elongation of mentum, mentum reddish brown Light green bud (ca. 2.5 cm long), reddish brown tinges only at beginning and tip of mentum, further elongation of mentum, length of mentum almost half of length of whole bud White bud (ca. 3 cm long), no splitting of sepals, elongated mentum pointing downwards away from tip of bud Full flower opening, sepals and petals fully expanded, lip fully protruded with visible yellow ridges running down from midlobe to foot of column Sepals and petals shrivelled and brownish Timing of events is reported in relation to the time during which floral buds were released from dormancy by cold induction (denoted as day 0). Time (day) 0 4 7 9 10 12 Author's personal copy Y.-M. Yap et al. / Scientia Horticulturae 119 (2008) 59–66 acetate buffer (pH 4.5) at 30 8C for 24 h. The supernatant was collected as the fructan fraction. The remaining residue was ground in 10 ml of 10 mM sodium acetate buffer (pH 4.5). Amyloglucosidase (12.4 units, Rhizopus, Sigma) was added to the ground plant materials and incubated at 45 8C for 24 h. The suspension was centrifuged at 4000  g for 10 min at 4 8C. The supernatant was collected as the starch fraction. All fractions were kept frozen until use. Concentrations of total soluble sugars, fructans and starch were determined by the phenol–sulphuric acid method (Dubois et al., 1956). Glucose was used as the standard in the assays. 2.3. Water potential analyses Frozen sepals or petals were allowed to thaw at room temperature (22 8C) for 20 min and were centrifuged at 10,000  g for 10 min at 18 8C for the collection of cell sap. Sap osmolality was determined using a vapour pressure osmometer (Wescor, model 5520 VAPRO). The corresponding water potential was calculated using the Van’t-Hoff relation: water potential (MPa) = ÀRT (mol kgÀ1), where R is the ideal gas constant (0.00831 MPa kg molÀ1 KÀ1) and T is temperature (K) (Nobel, 1983; Maricle et al., 2007). 2.4. Light microscopy Segments of approximately 1 cm  0.5 cm, including the central vein, were cut about halfway from the tips of freshly harvested sepals or petals. The fresh tissues were fixed in a ethanol:acetic acid:formalin:water (10:1:2:7 by volume) mixture, dehydrated in a series of increasing concentrations of ethanol, infiltrated with melted paraffin wax, and then embedded in paraffin wax. Embedded tissues were sectioned using a microtome to a thickness of 10 mm. The sections were stained with toluidine blue (0.1%) and examined using light microscopy. 2.5. Cell wall compositional analyses Cell wall materials were prepared by using a modified method described by Huber (1992). Frozen sepal or petal tissues were homogenised in 95% ethanol at 4 8C and then chilled at À20 8C for 24 h. The homogenates were centrifuged at 8000  g for 10 min at 4 8C. Tris-buffered phenol (5 ml per gram fresh weight of sepal/ petal tissues) was added to the residue and the whole mixture was incubated at room temperature for 45 min. The suspensions were re-centrifuged as described above, and the residues were resuspended in 80% ethanol at À20 8C for 2 h, followed by centrifugation again. The remaining residues were washed once with 80% ethanol, followed by 80% acetone, and then chloroform:methanol (1:1) mixture. All organic washings were conducted at room temperature. The final residues were recovered by filtration, washed thrice with acetone until total whitening, yielding the crude cell wall material (ethanol-insoluble residue, EIR). The EIR was air-dried, weighed, and then stored at À80 8C. Cellulose content of the EIR (5 mg) was extracted following the procedure described by Updegraff (1969) involving an initial hydrolysis in 5 ml of acetic–nitric acid reagent (10 volumes of 80% acetic acid:1 volume of nitric acid) at 100 8C for 30 min, followed by a further hydrolysis in 67% sulphuric acid at room temperature for 1 h. Total pectin in EIR (5 mg) was extracted as described by Ahmed and Labavitch (1977), involving a complete hydrolysis in 2 ml of concentrated sulphuric acid at room temperature for 20 min. Soluble pectins and hemicelluloses were sequentially extracted from EIR by water (water-soluble pectin), 50 mM sodium acetate 61 buffer (pH 6.0) containing 50 mM CDTA (CDTA-soluble pectin), 50 mM sodium carbonate containing 20 mM sodium borohydride (Na2CO3-soluble pectin), and finally 6 N sodium hydroxide containing 0.13 mM sodium borohydride (hemicellulose). The proportion of EIR to extractant was 1 mg:0.5 ml. All extractions were conducted at room temperature for 2 h with constant shaking, and then centrifuged at 10,000  g for 15 min at 4 8C (O’Donoghue et al., 2002; Deng et al., 2005). Cellulose and hemicellulose contents were assayed by the anthrone method (Scott and Melvin, 1953). Cellulose was used as the standard in these assays. Uronic acid contents were determined by the m-hydroxydiphenyl method (Blumenkrantz and Asboe-Hansen, 1973), using galacturonic acid as the standard. 2.6. Cell wall enzyme analyses Frozen sepals or petals were ground to fine powder in liquid nitrogen, then homogenised for 30 min in 20 mM sodium phosphate buffer (pH 7.5) containing 1.5 M sodium chloride, using 10 ml of extraction buffer per g fresh weight of tissues. All enzyme extractions were performed at 4 8C. The homogenates were centrifuged at 12,000  g for 20 min at 4 8C, and the supernatants were used for the various enzymatic assays. Pectin methylesterase (EC 3.1.1.11) activity was determined by a continuous spectrophotometric method as described by Hagerman and Austin (1986) with some modifications. The reaction mixture consisted of 2 ml citrus pectin solution (0.5%), 0.15 ml bromothymol blue solution (0.01%) and 100 ml crude enzyme extract. The pH of the solutions was adjusted to pH 7.5 each time before assay. After adding the enzyme extract, the reaction mixture was mixed well and allowed to stabilise for 1 min before measuring the absorbance at 620 nm. The absorbance was again measured after 21 min of incubation. PME activity was determined by the difference in absorbances and was calibrated against a galacturonic acid standard curve. One unit of PME activity represents 1 mmol of methylester liberated per hour. The activities of b-galactosidase (EC 3.2.1.23), b-glucosidase (EC 3.2.1.21), b-mannosidase (EC 3.2.1.25) and b-xylosidase (EC 3.2.1.37) were assayed by an adapted method described by Chin et al. (1999). Each of the reaction mixture consisted of 0.5 ml 5 mM p-nitrophenyl derivatives of b-D-galactopyranoside, b-D-glucopyranoside, b-D-mannopyronoside and b-D-xylopyranoside (Sigma) as substrates respectively, 50 ml crude enzyme extract and 50 mM sodium acetate buffer (pH 4.5) in a total volume of 2 ml. After incubation at 37 8C for 30 min, each of the reaction was stopped by the addition of 1 ml 0.2 M sodium carbonate, and the amount of pnitrophenol formed was determined spectrophotometrically at 415 nm. One unit of the respective glycosidase activity represents 1 mmol of p-nitrophenol released per hour. In all enzymatic assays, boiled enzyme extracts were included as controls. 3. Results 3.1. Growth changes Sepals and petals demonstrated similar patterns of weight changes. Total fresh weight increased during early stages of floral bud development and decreased upon full flower opening on day 10, due to changes in both water and dry matter contents (Fig. 1). Increase in fresh weight: dry weight ratio observed during late stage bud development (day 9) was due to a drastic increase in water content, approximately 32 mg sepalÀ1 and 37 mg petalÀ1 on day 7 to 100 mg sepalÀ1 and 97 mg petalÀ1 on day 9. A decrease in the ratio was also observed during the senescence stage (day 12) where water loss was much greater than dry matter loss. Author's personal copy 62 Y.-M. Yap et al. / Scientia Horticulturae 119 (2008) 59–66 Fig. 1. Characteristics of weight changes of (A) sepals and (B) petals of D. crumenatum at each developmental stage. Each value is the mean Æ S.E. (n = 5). 3.2. Carbohydrate and water status changes during development During floral bud development (day 4–day 9), total soluble sugar concentrations in both sepals and petals increased gradually, reaching a maximum level (approximately three times that of newly induced floral buds at day 4) upon full flower opening on day 10. Total soluble sugar concentrations subsequently decreased by approximately 50% during senescence on day 12 (Fig. 2A). The reserve carbohydrates, fructans and starch, decreased in concentrations throughout floral bud and flower development, up till senescence (Fig. 2B–C). The accumulation of total soluble sugars during flower opening was accompanied by decreases in water potential of sepal and petal cell sap, which subsequently increased during senescence (Fig. 2C). 3.3. Changes in anatomy during development Cross-sections of sepals and petals from young, newly induced floral buds (day 4) showed round parenchyma cells packed tightly and neatly between the upper and lower epidermes (Fig. 3A). By day 7, the parenchyma cells had enlarged and were less tightly packed together, exhibiting some intercellular spaces (Fig. 3B). This layer became progressively disorganised as floral buds continued to develop and open into flowers, with larger intercellular spaces and membrane disintegration becoming apparent (Fig. 3C and D). By senescence stage on day 12, both sepals and petals showed almost complete absence of parenchyma cells, although the epidermal layers remained intact (Fig. 3E). 3.4. Cell wall composition Sepals and petals demonstrated similar patterns of changes in cell wall composition. The amount of EIR increased from approximately 0.2 mg (floral part)À1 to 0.8 mg (floral part)À1 throughout early floral bud development and flower opening (day 4–day 9), but decreased to approximately 0.4 mg (floral part)À1 as flowers senesced on day 12 (Fig. 4A). Cellulose levels increased Fig. 2. Changes in amounts of (A) total soluble sugars, (B) fructans, and (C) starch and water potential during development of D. crumenatum flowers. Each value is mean Æ S.E. (n = 10). from about 5 mg (floral part)À1 to 30 mg (floral part)À1 as the floral buds developed to maturity and decreased to about 25 mg (floral part)À1 during senescence (Fig. 4B). Hemicellulose levels decreased steadily from approximately 30 mg (floral part)À1 to 2 mg (floral part)À1 throughout bud and flower development (Fig. 4C). The amount of total pectin generally increased during floral bud development (day 4–day 9), and remained constant (approximately 90 mg sepalÀ1 and 60 mg petalÀ1) throughout flower opening and senescence on day 10 and day 12 respectively (Fig. 4D). The amount of water-soluble pectin increased by about three times during early floral bud development on day 7, remained at these levels (approximately 6 mg sepalÀ1 and 3 mg petalÀ1) in mature buds, then decreased upon flower opening and senescence (Fig. 4E). On the other hand, the quantity of CDTAsoluble pectin increased during bud development and remained at high levels (approximately 20 mg sepalÀ1 and 8 mg petalÀ1) even during senescence (Fig. 4F). The levels of Na2CO3-soluble pectin also increased progressively during bud development, remaining at high levels (approximately 60 mg sepalÀ1 and 40 mg petalÀ1) during flower opening and then decreased during senescence (Fig. 4G). Maximum amounts of water-soluble pectin, CDTAsoluble pectin and Na2CO3-soluble pectin were all observed during late floral bud stage at day 9, indicating a maximum level of pectin solubility during this developmental period. The amount of Author's personal copy Y.-M. Yap et al. / Scientia Horticulturae 119 (2008) 59–66 63 4. Discussion Fig. 3. Changes in anatomical features of D. crumenatum during flower development. (I) Developmental stages of floral buds/flowers: A, day 4; B, day 7; C, day 9; D, day 10; E, day 12 floral buds/flowers. Transverse sections of (II) sepals and (III) petals corresponding to those in I were stained with toluidine blue. Arrows represent areas with parenchyma cell disruption and/or intercellular spaces. insoluble pectin decreased during floral bud development (day 4– day 9), and subsequently increased upon flower opening up till senescence (Fig. 4H). 3.5. Activities of wall-based enzymes Colorimetric analyses using fresh tissue extracts demonstrated the extracts did not contain components to hydrolyse carboxymethylcellulose, filter paper and polygalacturonic acid. As such, we concluded that sepals and petals of D. crumenatum had no or very insignificant levels of cellulase and polygalacturonase activities. Pectin methylesterase activity increased during early floral bud development and decreased drastically during late floral bud development, and remained at low levels throughout flower opening and senescence stages (Fig. 5A). Maximum activities of bgalactosidase, b-glucosidase, b-mannosidase and b-xylosidase were observed during floral bud development (day 7 or day 9, Fig. 5B–E). Of all the enzymes tested, the major ones were bglucosidase [0.3–5 units (floral part)À1], pectin methylesterase [0.5–4.3units (floral part)À1] and b-galactosidase [0.4–0.7 units (floral part)À1]. The activities of b-mannosidase and b-xylosidase were lower [0.04–0.2 units (floral part)À1]. With the exception of b-galactosidase, activities of all other tested glycosidases exhibited significant decreases during flower senescence (Fig. 5B–E). bgalactosidase activity decreased slightly during flower opening, and then increased again upon senescence. There is great variation among flowering plants in the physiology regulating floral bud development, flower opening and senescence (O’Donoghue, 2006). The results obtained from the present study provided novel insights into the floral bud development and flower opening processes in tropical orchids. Flower opening and senescence in D. crumenatum were found to be dependent on carbohydrate metabolism and water relations (Fig. 2). The high accumulation of reserve carbohydrates in sepals and petals of D. crumenatum during early stages of bud development was not unexpected as inflorescences and flowers are considered to be sinks for assimilates (Hew and Yong, 2004). During the flowering process in orchids, the priority of assimilate partitioning usually follows this decreasing order: inflorescences ) young leaves ) shoots (Hew and Yong, 2004). Prior to the onset of flower opening in D. crumenatum, the significant breakdown of reserve carbohydrates and increase in levels of soluble sugars resulted in a more negative water potential. This resulted in a larger water potential gradient, possibly causing a greater water influx into the expanding cells and subsequent turgor changes, which had been suggested to be a major force in driving flower opening (van Doorn and van Meeteren, 2003). Flower opening regulated by changes in carbohydrate levels and water relations has also been reported in species such as carnation (Acock and Nichols, 1979), daylily (Bieleski, 1993) and creeping bellflower (Vergauwen et al., 2000). Contradictory to our results, however, Sonia roses did not show a relationship between carbohydrate and water status during flower opening (Evans and Reid, 1988). During rapid petal expansion in roses, starch hydrolysis occurred, but osmotic potential of the cell sap became less negative, reducing the water potential gradient (Evans and Reid, 1988). Hence, it was proposed that mobilisation of storage carbohydrates and water uptake were not the main controlling factors in rose petal growth, but rather other physiological factors such as cell wall extensibility might be involved (Evans and Reid, 1988). Cell wall metabolism in flowers has been studied in great detail in only two other species—sandersonia and carnation (de Vetten and Huber, 1990; de Vetten et al., 1991; O’Donoghue et al., 2002), but cell wall alterations were not investigated in young developing floral buds in these studies. Studies conducted on daylily focused on activities of cell wall hydrolases from young floral buds till senesced flowers, but no accompanying analyses on cell wall compositional changes were reported (Panavas and Rubinstein, 1998; Panavas et al., 1998). With these in mind, and the proposals that cell wall metabolism might be involved during flower opening and senescence (O’Donoghue, 2006), we examined the changes in anatomy of sepals and petals of D. crumenatum during floral bud and flower development. The increasing disorganisation of parenchyma cells and appearance of intercellular spaces observed (Fig. 3) during the development of D. crumenatum floral buds resembled those observed during the opening of sandersonia and daylily flowers (Panavas et al., 1998; O’Donoghue et al., 2002), indicating alterations in cell wall constituents. In our study, cell wall composition and activities of cell wallbased enzymes were expressed as ‘per floral part’ basis and similar trends were obtained when expressed as ‘per flower basis’ (data not shown). The accumulation of cell wall cellulose during floral bud development and subsequent breakdown during senescence in D. crumenatum (Fig. 4A) were similar to those of carnation and sandersonia (de Vetten and Huber, 1990; O’Donoghue et al., 2002). However, while both D. crumenatum and sandersonia flowers showed no cellulase activity, the enzyme was detected in daylily flowers, indicating possible hydrolysis of cellulose by cellulase in daylily (Panavas et al., 1998). Hemicellulose content decreased Author's personal copy 64 Y.-M. Yap et al. / Scientia Horticulturae 119 (2008) 59–66 Fig. 4. D. crumenatum sepal and petal cell wall components, expressed on a per floral part (sepal or petal) basis, at each developmental stage: (A) EIR, (B) cellulose, (C) hemicellulose, (D) pectin, (E) water-soluble pectin, (F) CDTA-soluble pectin, (G) Na2CO3-soluble pectin, and (H) insoluble pectin. Each value is mean Æ S.E. (n = 5). steadily in D. crumenatum, while that in sandersonia showed no apparent changes (O’Donoghue et al., 2002). The breakdown of hemicelluloses in D. crumenatum resembled the changes observed during the ripening of fruits such as grapes (Deng et al., 2005) and carambola (Chin et al., 1999). Hemicelluloses (or glycans) play important structural roles in the cross-linking of cellulose in the cell walls and the breakdown of hemicelluloses may contribute to the alterations in primary cell wall structure (Brummell, 2006). Activities of various glycosidases were detected in D. crumenatum and were observed to alter with the development of the floral buds (Fig. 5C–E). Various glycosidases had also been reported in carnation and were suggested to be involved in the hydrolysis of hemicelluloses (de Vetten et al., 1991). Glycosidases had also been reported in fruits such as capsicum (Sethu et al., 1996) and carambola (Chin et al., 1999). The family of glycosidases is known to play a crucial role in the degradation of various cell wall polysaccharides, allowing the remodelling of the cell wall structure (Minic, 2008). Pectin hydrolysis and polygalacturonase activity were not detected in D. crumenatum flowers (Fig. 5). These were similar to those observed in sandersonia flowers (O’Donoghue et al., 2002). However, daylily flowers were reported to exhibit increasing polygalaturonase activity in senescing flowers, indicating the possibility of pectin hydrolysis in these flowers (Panavas et al., 1998). While pectin synthesis or accumulation occurred in D. crumenatum, there was also considerable pectin solubilisation. Water-soluble and CDTA-soluble pectins are relatively weakly bound to cell wall polysaccharides by molecular entanglements, Author's personal copy Y.-M. Yap et al. / Scientia Horticulturae 119 (2008) 59–66 Fig. 5. Changes in activities, expressed on a per floral part (sepal or petal) basis, of (A) pectin methylesterase, (B) b-galactosidase, (C) b-glucosidase, (D) bmannosidase and (E) b-xylosidase during development of D. crumenatum flowers. Each value is mean Æ S.E. (n = 5). hydrophobic forces, weak ionic bonds or ionic calcium bridges, while Na2CO3-soluble pectins are more strongly attached to the cell wall via covalent bonds (Brummell, 2006). In D. crumenatum, the types of bonds and bond strengths that interlink pectin molecules appeared to alter as the floral buds and flowers continued to develop. There were increasing proportions of pectins that were susceptible to the solubilising agents during floral bud development, and decreased pectin solubility during flower opening and senescence (Fig. 4E–H). Pectin solubilisation is one 65 ripening-associated cell wall modification that has been widely reported in an array of fruits, and different species of fruits also exhibited variations in pectin modifications (Chin et al., 1999; Brummell and Harpster, 2001; Deng et al., 2005; Vicente et al., 2007). Cell wall swelling, which would enhance access of hydrolytic enzymes to their substrates and promote polymer disassembling, had been correlated to increased pectin solubilisation (Redgwell et al., 1997). Pectin methylesterase, which de-esterifies pectins, had been suggested to increase pectin solubilisation by creating electrostatic repulsion between negatively charged molecules that could result in the loosening of weakly attached pectins from the cell wall (Grignon and Sentenac, 1991). In D. crumenatum, the largest increase in pectin solubilisation, which was observed during floral bud development, coincided with maximum pectin methylesterase activity (Fig. 5A). However, the significant decrease in pectin methylesterase activity, shortly after the peak, was not accompanied by a drastic decrease in pectin solubility (Fig. 5E–H). In sandersonia, pectin methylesterase activity (units flowerÀ1 basis) peaked upon mature flower stage, then decreased till senescence (O’Donoghue et al., 2002). In daylily, pectin methylesterase activity (units mg proteinÀ1 basis) decreased from young floral buds stage till flower opening, similar to D. crumenatum (units mg proteinÀ1 basis, data not shown). Such variations among the different species may be the result of, or cause(s) for, the differences in timeframes of floral bud development, flower opening, senescence and longevity of the flowers (O’Donoghue, 2006). Different species of fruits also demonstrated variations in cell wall pectic modifications during ripening (Brummell, 2006). There are many other factors that may also regulate pectin solubilisation and other polymer modifications. One such candidate could be b-galactosidase, an enzyme responsible for the degradation of cell wall b-galactan, regulating cell wall flexibility, intercellular connections and cell wall porosity, and affecting mobility of enzymes within the cell wall matrix (Brummell and Harpster, 2001; Brummell, 2006). In vitro treatments of cell wall preparations from papaya with b-galactosidase resulted in increased pectin solubilisation (Ali et al., 1998). In D. crumenatum, the activity of b-galactosidase remained relatively constant during late floral bud and flower opening stages, and increased slightly during senescence (Fig. 5B). This might account for the maintenance of CDTA-soluble pectins (Fig. 4F). In sandersonia, bgalactosidase activity (units flowerÀ1 basis) increased during late floral bud up till wilted flower stage, and decreased significantly during senescence (O’Donoghue et al., 2002). On the other hand, bgalactosidase activity (units mg proteinÀ1) in daylily was constant throughout bud development and flower opening, and only increased during wilting and senescence stages, similar to that observed in D. crumenatum (units mg proteinÀ1, data not shown). It is possible that non-enzymic mechanisms might also be involved in cell wall modifications. In tomato, ascorbate-generated hydroxyl radicals demonstrated non-enzymic scission of cell wall polysaccharides, and caused an increase in pectin solubilisation (Dumville and Fry, 2003). Cell wall modification is thus a complex process, and it would be worthwhile to identify other non-enzymic factors, and also investigate the possible interactions between the various enzymic and non-enzymic mechanisms involved in the process. There are questions on whether flowering is part of a senescence programme that has already started (O’Donoghue et al., 2002). Carbohydrate and cell wall polymer alterations are considered to be events associated with cell death and senescence (Rubinstein, 2000). Previously, studies on carnation and sandersonia suggested that just prior to flower opening (late floral bud stage), a certain degree of cell wall dismantling had already started Author's personal copy 66 Y.-M. Yap et al. / Scientia Horticulturae 119 (2008) 59–66 (de Vetten and Huber, 1990; de Vetten et al., 1991; O’Donoghue et al., 2002). Our results showed that senescence-related events such as cell wall disassembly occurred as early as during young floral bud stage (day 7), suggesting the early onset of a senescence programme during floral bud development. Flower opening, thus, appears to be a continuum of a senescence programme that had already commenced. It remains to be tested if inhibitors of cell wall hydrolases and/or senescence-inhibiting chemicals could delay floral bud development and flower opening. Such techniques would be beneficial to the horticultural industry that commonly depends on refrigeration to delay flower opening; the techniques could also help lower energy consumption. In conclusion, floral bud development, flower opening and flower senescence in D. crumenatum were tightly regulated by changes in carbohydrate and water status, and cell wall metabolism. Senescence-indicative events such as cell wall disassembling were found to occur during early floral bud development, suggesting the early onset of a senescence programme. Cell wall metabolism is a common feature among D. crumenatum, sandersonia, carnation and daylily but the timing and manner of modifications vary between the species. Whether these differences are the cause for the diversity in the rates of floral bud development, time of flower opening, and longevity of the flowers in the various plants remain to be investigated. Investigating the possible roles of non-enzymic mechanisms in controlling cell wall disassembly could provide further insight to the involvement of cell wall metabolism in controlling flowering, and the relation between flowering and the in-built senescence programme of the plant. Acknowledgements We thank H.H. Yeoh for providing the various enzymatic substrates, P.L. Chong and B.L. Tay for their technical help in the experiments, and T.K. Ong, W.P. Yap, H.Q. Lim and S.L. Ng for their assistance in many ways. We also thank the National University of Singapore for sponsoring the Research Scholarship for Y.M. Yap. References Acock, B., Nichols, R., 1979. Effects of sucrose on water relations of cut, senescing, carnation flowers. Ann. Bot. 44, 221–230. Ahmed, A.E.R., Labavitch, J.M., 1977. A simple method for accurate determination of cell wall uronide content. J. Food Chem. 1, 361–365. Ali, Z.M., Ng, S.Y., Othman, R., Goh, L.Y., Lazan, H., 1998. Isolation, characterization and significance of papaya b-galactanases to cell wall modification and fruit softening during ripening. Physiol. Plant. 104, 105–115. Bieleski, R.L., 1993. Fructan hydrolysis drives petal expansion in the ephemeral daylily flower. Plant Physiol. 103, 213–219. Bieleski, R.L., Elgar, J., Heyes, J., 2000. Mechanical aspects of rapid flower opening in Asiatic lily. Ann. Bot. 86, 1175–1183. Blumenkrantz, N., Asboe-Hansen, G., 1973. New method for quantitative determination of uronic acids. Anal. Biochem. 54, 484–489. Brummell, D.A., 2006. Cell wall disassembly in ripening fruit. Funct. Plant Biol. 33, 103–119. Brummell, D.A., Harpster, M.H., 2001. Cell wall metabolism in fruit softening and quality and its manipulation in transgenic plants. Plant Mol. Biol. 47, 311–339. Chin, L.H., Ali, Z.M., Lazan, H., 1999. Cell wall modifications, degrading enzymes and softening of carambola fruits during ripening. J. Exp. Bot. 50, 767–777. Corner, E.J.H., 1988. Wayside Trees of Malaya, vol. 1. United Selangor Press, Malaysia. de Vetten, N.C., Huber, D.J., 1990. Cell wall changes during the expansion and senescence of carnation (Dianthus caryophyllus) petals. Physiol. Plant. 78, 447–454. de Vetten, N.C., Huber, D.J., Gross, K.C., 1991. Endoglycanase-catalysed degradation of hemicelluloses during development of carnation (Dianthus caryophyllus L.) petals. Plant Physiol. 95, 853–860. Deng, Y., Wu, Y., Li, Y., 2005. Changes in firmness, cell wall composition and cell wall hydrolases of grapes stored in high oxygen atmospheres. Food Res. Int. 38, 769– 776. Dubois, M., Gilles, K.A., Kamilton, J.K., Rebers, P.A., Smith, F., 1956. Colorimetric method for determination of sugars and related substances. Anal. Chem. 28, 350–356. Dumville, J.C., Fry, S.C., 2003. Solubilisation of tomato fruit pectins by ascorbate: a possible non-enzymic mechanism of fruit softening. Planta 217, 951–961. Evans, R.Y., Reid, M.S., 1988. Changes in carbohydrates and osmotic potential during rhythmic expansion of rose petals. J. Am. Soc. Hort. Sci. 113, 884–888. Grignon, C., Sentenac, H., 1991. pH and ionic conditions in the apoplast. Ann. Rev. Plant Physiol. Plant Mol. Biol. 42, 103–128. Hagerman, A.E., Austin, P.J., 1986. Continuous spectrophotometric assay for plant pectin methyl esterase. J. Agric. Food Chem. 34, 440–444. Hew, C.S., Wee, K.H., Wong, S.M., Ong, T.K., Lee, F.Y., 1989. Water content and longevity of orchid cut flowers. Malay. Orch. Rev. 23, 36–43. Hew, C.S., Yong, W.H., 2004. The Physiology of Tropical Orchids in Relation to the Industry. World Scientific, Singapore. Ho, L.C., Nichols, R., 1977. Translocation of 14C-sucrose in relation to changes in carbohydrate content in rose corollas cut at different stages of development. Ann. Bot. 41, 227–242. Holttum, R.E., 1953. Flora of Malaya–Orchids of Malaya. Govt. Printing Office, Singapore. Huber, D.J., 1992. The inactivation of pectin depolymerase associated with isolated tomato fruit cell wall: implications for the analysis of pectin solubility and molecular weight. Physiol. Plant. 86, 25–32. Maricle, B.R., Cobos, D.R., Campbell, C.S., 2007. Biophysical and morphological leaf adaptations to drought and salinity in salt marsh grasses. Environ. Exp. Bot. 60, 458–467. Minic, Z., 2008. Physiological roles of plant glycoside hydrolases. Planta 227, 723– 740. Nobel, P.S., 1983. Biophysical Plant Physiology and Ecology. WH Freeman and Company, San Francisco. O’Donoghue, E.M., 2006. Flower petal cell walls: changes associated with opening and senescence. New Zeal. J. Forest Sci. 36, 134–144. O’Donoghue, E.M., Somerfield, S.D., Heyes, J.A., 2002. Organization of cell walls in Sandersonia aurantiaca floral tissue. J. Exp. Bot. 53, 513–523. Panavas, T., Pikula, A., Reid, P.D., Rubinstein, B., Walker, E.L., 1999. Identification of senescence-associated genes from daylily petals. Plant Mol. Biol. 40, 237–248. Panavas, T., Reid, P.D., Rubinstein, B., 1998. Programmed cell death of daylily petals: activities of wall-based enzymes and effects of heat shock. Plant Physiol. Biochem. 26, 379–388. Panavas, T., Rubinstein, B., 1998. Oxidative events during programmed cell death of daylily (Hemerocallis hybrid) petals. Plant Sci. 133, 125–138. Redgwell, R.J., MacRae, E.A., Hallett, I., Fischer, M., Perry, J., Harker, R., 1997. In vivo and in vitro swelling of cell walls during fruit ripening. Planta 203, 162–173. Rubinstein, B., 2000. Regulation of cell death in flower petals. Plant Mol. Biol. 44, 303–318. Scott, T.A., Melvin, E.H., 1953. Determination of dextran with anthrone. Anal. Chem. 25, 1656–1661. Sethu, K.M.P., Prabha, T.N., Tharanathan, R.N., 1996. Post-harvest biochemical changes associated with the softening phenomenon in Capsicum annum fruits. Phytochemistry 42, 961–966. Spala, A., Bagiorgas, H.S., Assimakopoulos, M.N., Kalavrouziotis, J., Matthopoulos, D., Mihalakakou, G., 2008. On the green roof system. Selection, state of the art and energy potential investigation of a system installed in an office building in Athens. Greece. Renew. Energ. 33, 173–177. Updegraff, D.M., 1969. Semimicro determination of cellulose in biological materials. Anal. Biochem. 32, 420–424. van Doorn, W.G., 1997. Effects of pollination on floral attraction and longevity. J. Exp. Bot. 48, 1615–1622. van Doorn, W.G., van Meeteren, U., 2003. Flower opening and closure: a review. J. Exp. Bot. 54, 1801–1812. Vergauwen, R., van den Ende, W., van Laere, A., 2000. The role of fructan in flowering of Campanula rapunculoides. J. Exp. Bot. 51, 1261–1266. Vicente, A.R., Lowell, A., Greve, L.C., Labavitch, J.M., 2007. Cell wall disassembly events in boysenberry (Rubus idaeus L.  Rubus ursinus Cham. & Schldl.) fruit development. Funct. Plant Biol. 34, 614–623. Wagstaff, C., Malcolm, P., Rafiq, A., Leverentz, M., Griffiths, G., Thomas, B., Stead, A., Rogers, H., 2003. Programmed cell death (PCD) processes begin extremely early in Alstroemeria petal senescence. New Phyto. 160, 49–59. Wang, H., Li, J., Bostock, R.M., Gilchrist, D.G., 1996. Apoptosis: a functional paradigm for programmed plant cell death induced by a host-selective phytotoxin and invoked during development. Plant Cell 8, 375–391. [...]... variation in cell wall metabolism that regulates flowering (Panavas et al 1998; O’Donoghue et al 2002) In the above-mentioned studies on the possible involvement of cell wall modifications and cell wall remodelling in regulating flowering, cell wall changes were compared only between stages of mature bud (just prior to opening), opening flower, mature flower, wilting flower and senesced flower (de... (Chin et al 1999; Deng et al 2005) In daylily flowers, analyses of cell wall composition were not published, but reported changes in activities of cell wall- based enzymes during flower development suggested the involvement of cell wall metabolism in flowering (Panavas et al 1998) While cellulase activity was detected in daylily flowers, it was reported to be absent in sandersonia flowers, indicating... between the flowers shows certain similarities and differences in cell wall changes associated with flowering In carnation flowers, full flower opening was accompanied by increases in contents of cell wall cellulose, total pectins, chelator-soluble pectins, carbonate-soluble pectins and neutral sugars Upon senescence of the flowers, all of these cell wall components, except chelator-soluble pectins, decreased... the cells cease growth (Cosgrove 1999) The dynamic structure of the cell wall is important in regulating cell expansion and cell growth For example, cell wall loosening is a pre-requisite for the incorporation of newly synthesized wall polymers during cell expansion and cell growth (Carpita and McCann 2000) Plant cell walls also have important roles in controlling fruit ripening The degree of cell wall. .. (Rubinstein 2000; Wagstaff et al 2003) In Ipomoea, dynamic structural changes such as cell enlargement, modification of cell shape and reduction in cell wall thickness occurred in the inner epidermal cells even before flower opening (Phillips and Kende 1980) In sandersonia, intercellular air spaces and increasingly disorganized packing of parenchyma cells also occurred prior to flower opening (O’Donoghue... that may involve analysing and comparing cell wall modifications between floral buds showing normal and abnormal opening patterns 3 Chapter 2 Literature Review 2.1 Plant cell wall Each plant cell comprises a specialised and complex cell wall that serves many functions (Cosgrove 1999) The cell walls provide structural support and maintain the shape of the cells, act as a protective barrier against water... published regarding cellulose degradation As the main component of the cell wall matrix, cellulose is expected to undergo distinct alterations during cell wall changes (Fischer and Bennett 1991; Rose and Bennett 1999) Cell wall cellulose content decreased during ripening of grapes (Deng et al 2005) However, during the ripening of pear (Pyrus communis), tomato and avocado, cellulose levels remained constant... hemicelluloses held tightly by hydrogen bonds (Brummell 2006) 7 2.2 Cell wall metabolism Various regulated cell wall architecture changes occur with the development of plants (Carpita and McCann 2000) Fruit ripening is one developmental event whereby many changes occur in the cell walls, resulting in the final texture of the fruit (Brummell 2006) Consequently, the majority of the information on cell wall. .. Pectin solubilisation One of the major changes in cell wall pectins with the development of plants is the increasing solubilisation of pectins, as commonly observed during the ripening of fruits (Brummell and Harpster 2001) Pectin solubilisation is usually measured as the increase in ease of extractability of pectins by various extractants, and can be extrapolated to bond changes within the cell wall. .. question on whether flowering could be a process regulated by the senescence programme of the plant Some supporting evidence include the increase in oxidation of membrane components prior to flower opening in daylily (Panavas and Rubinstein 1998), and the increasing trend of DNA laddering throughout petal development in alstroemeria (Wagstaff et al 2003) Studies on carnation and sandersonia 1 flowers demonstrated ... the flowers shows certain similarities and differences in cell wall changes associated with flowering In carnation flowers, full flower opening was accompanied by increases in contents of cell wall. .. changes in activities of cell wall- based enzymes during flower development suggested the involvement of cell wall metabolism in flowering (Panavas et al 1998) While cellulase activity was detected in. .. 1998; O’Donoghue et al 2002) In the above-mentioned studies on the possible involvement of cell wall modifications and cell wall remodelling in regulating flowering, cell wall changes were compared