Citation
Structure and sucrose metabolizing enzymes of the transport path

Material Information

Title:
Structure and sucrose metabolizing enzymes of the transport path
Creator:
Lowell, Cadance Ann, 1957-
Copyright Date:
1986
Language:
English

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Cadance Ann Lowell. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
15624686 ( OCLC )
AEK6517 ( LTUF )
0029841483 ( ALEPH )

Downloads

This item has the following downloads:


Full Text
STRUCTURE AND SUCROSE METABOLIZING ENZYMES OF THE TRANSPORT PATH:
IMPLICATIONS FOR ASSIMILATE TRANSLOCATION IN GRAPEFRUIT
By
CADANCE ANN LOWELL

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA

1986

I




ACKNOWLEDGMENTS
I wish to express my deepest appreciation to Dr. Karen E. Koch for her guidance and helpful criticisms during this study. I also wish to thank the other members of my committee, Dr. James Soule, Dr. Thomas Humphreys, Dr. Hilton Biggs and Dr. James Kimbrough, for their assistance and helpful suggestions concerning both the interpretation of the data and this manuscript. I am grateful to Patricia Tolson-Tomlinson whose guidance allowed me to finally complete the enzymatic portion of this research and to Wayne Avigne and Andrew K. McCullers for their technical aid. I also wish to thank Dr. Kenneth Curry and Debra Akin for their help with the morphometric analysis in this manuscript.
Finally, my deepest thanks go to my husband, James William Williams for his time and expertise in drawing the figures in the text, and his support for this endevor, even when he did not understand it.




TABLE OF CONTENTS
ACKNOWLEDGMENTS ..............................................
LIST OF ABBREVIATIONS ..........................................
ABSTRACT .......................................................
CHAPTERS

page
ii
v
vi

I INTRODUCTION .........................................
II LITERATURE REVIEW ....................................
Anatomy of Citrus Fruit ..............................
Assimilate Partitioning in Citrus ..................
Photosynthate Translocation and Partitioning
in Sink Tissues .................................
Photosynthate Translocation and Partitioning
in Citrus Fruit Partitioning ...................
Sucrose Synthetase, Invertase and Assimilate Partitioning ................................
III THE VASCULAR SUPPLY AND ASSIMILATE TRANSPORT
IN GRAPEFRUIT ....................................
Introduction ...........................................
Materials and Methods .................................
Results and Discussion ................................
Conclusions ...........................................
IV NON-VASCULAR TISSUES IN GRAPEFRUIT DURING
DEVELOPMENT ......................................
Introduction ..........................................
Materials and Methods .................................
Results and Discussion ................................
Conclusions ...........................................
V SUCROSE METABOLIZING ENZYMES AND ASSIMILATE
PARTITIONING IN SINK TISSUES OF
DEVELOPING GRAPEFRUIT ............................
Introduction ...........................................




VI
APPENDICES

Materials and Methods..................................
Results and Discussion.................................
Conclusions............................................
OVERALL CONCLUSIONS....................................

A TECHNICAL DATA FOR SUCROSE SYNTHETASE IN
TISSUES OF 'MARSH' GRAPEFRUIT ....................
B TECHNICAL DATA FOR INVERTASE IN TISSUES
OF 'MARSH' GRAPEFRUIT.............................
LITERATURE CITED.................................................
BIOGRAPHICAL SKETCH..............................................

57 60 71 73
76
84 92
104




LIST OF ABBREVIATIONS ALB albedo (outer and inner mesocarp)
DTT dithiothreitol
EDTA ethylenediaminetetraacetic acid
Hepes N-2-Hydroxyethylpiperaxine-N'-2-ethanesulfonic acid
JV juice vesicles
Mes 2-(N-morpholino) ethane sulfonic acid
mM millimolar
ml milliliter
ul microliter
Pipes 1,4-piperazinediethanesulfonic acid
PVPP polyvinyl-poly pyrrolidine (insoluble)
Seg E segment epidermis
Tris tris(hydroxymethyl)aminomethane
UDP uridine 5-diphosphate
UDPG uridine 5-diphosphoglucose




Abstract of Dissertation Presented to the Graduate School of the
University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
STRUCTURE AND SUCROSE METABOLIZING ENZYMES OF THE TRANSPORT PATH:
IMPLICATIONS FOR ASSIMILATE TRANSLOCATION IN GRAPEFRUIT By
Cadance Ann Lowell
Chairman: Dr. Karen E. Koch
Major Department: Horticultural Sciences (Fruit Crops)
Structure of the transport path, activities of sucrose metabolizing enzymes, accumulation of dry weight and levels of soluble sugars were compared during development of 'Marsh' grapefruit (Citrus paradisi Macf.) to elucidate features related to mechanisms of photosynthate import.
Aspects of vascular anatomy which appear to be associated with assimilate translocation include 1) direct alignment of individual vascular strands within stems and specific fruit juice segments, 2) an increase in the phloem:xylem ratio of vascular bundles with distance from their point of entry into fruit and 3) developmental changes in vascular bundles supplying photosynthates to juice tissues. In the latter, an asymmetric proliferation of primary and secondary phloem and formation of lignified bundle caps coincide with a shift in rate of dry weight accumulation by adjacent tissues.




Assimilates enroute to juice vesicles must move through extensive non-vascular portions of the translocation path after phloem unloading. No structural specializations for either apoplastic or symplastic transport were evident in parenchyma surrounding the vascular bundles or in the phloem-free, elongated vesicle stalks. Frequiencies of plasmodesmata mature juice vesicle stalks ranged from 0.24 to 0.38 um (1.96 to 3.78 um ). Photosynthate transport along other paths would probably be restricted by the combination of cellular discontinuity and by cutinization of the epidermis surrounding both segment and individual juice vesicles. Cells of this continuous epidermal layer appear to be metabolically active during early fruit growth, but later become highly vacuolated. Cells of mature juice vesicles, the final site of assimilate storage, have low cytoplasm to vacuole ratios and show no anatomical modification for secretion.
Of the 5 sucrose metabolizing enzymes examined during grapefruit development, only alkaline invertase and sucrose synthetase remained active during the period of most rapid gains in dry weight. The extent to which sucrose is hydrolyzed and resynthesized during import into citrus fruit is unclear, but the in vitro capacity of these two enzymes is such that either or both could account for this process. A rapid decrease to minimal levels in the activity of soluble and insoluble acid invertase during the first 3 months of growth suggest acid invertases may be associated more closely with cell division than assimilate transfer. In addition, non-enzymatic sucrose hydrolysis can occur at low pH, and total hexose pools measured in the present study increased as juice pH decreased.




CHAPTER I
INTRODUCTION
Total soluble solids and the total soluble solids: total
(titratable) acid ratio are major factors in the harvest and market value of Citrus. Despite importance of the timing and quantity of assimilates entering citrus fruit, this process and its control are little understood. Crop productivity is influenced by rate of translocation (Chatterton, 1973; Liu et al., 1973), carbohydrate content of source leaves (Habeshaw, 1973; Geiger, 1976) and/or sink demand (Geiger and Fondy, 1980; Ho and Baker, 1982). Assimilate transfer from phloem to storage tissues is associated with the spatial arrangement of vascular tissue relative to sink cells and the effects of sugar gradients, pH and enzymes in this zone (Jenner, 1974; Eschrich, 1980; Felker and Shannon, 1980).
The spatial arrangement of vascular and non-vascular tissues in Citrus make this fruit a unique system with which to observe sugar movement into a sink. Primary vascular tissue in the citrus fruit consists of dorsal, septal and central carpellary bundles (traces) (Roth, 1977). A fine network of minor traces extends from dorsal and septal bundles and terminates near peripheral oil glands in the flavedo. This vascular tissue persists through fruit maturation (Ford, 1942). Citrus fruit lacks vascular connections, however, between the albedo and juice vesicles (Ford,1942). The presence of this non-vascular zone between




vascular bundles and juice vesicles suggests that assimilates do not enter juice vesicles directly through phloem translocation. Subsequent to phloem unloading, photosynthates must pass through the inner albedo (mesocarp) and juice vesicle stalks before entering the juice vesicle heads. Comparable movement of assimilates into sink structures of other species has been hypothesized to be symplastic, apoplastic or both (Jenner, 1980).
Sucrose presumably can be unloaded from phloem via apoplastic and symplastic pathways. In both instances, the process is believed to occur along a decreasing sucrose gradient. This gradient in symplastic systems could be developed either by metabolic conversions of sucrose in sink tissues (Geiger and Fondy, 1980) or by compartmentalization of sucrose into vacuoles (Fisher and Outlaw, 1979). Where photosynthates move into the apoplast, sucrose unloading is believed to involve one-way movement at a rate dependent on the hydrolytic action of free-space acid invertase (cell wall bound) and soluble neutral invertase (cytoplasmic) (Hawker and Hatch, 1965; Glasziou and Gayler, 1972; Eschrich, 1980). Direct unloading of sucrose into the apoplast and subsequent storage in the symplast of sink cells can occur without invertase activity, as in sugar beet (Giaquinta, 1979; Wyse, 1979). The cytoplasmic enzymes sucrose phosphate synthetase and sucrose synthetase also are associated with sucrose storage in sink tissues (Glasziou and Gayler, 1972; Giaquinta, 1979). Sucrose phosphate synthetase catalyzes sucrose synthesis, but sucrose synthetase readily catalyzes both the cleavage and synthesis of sucrose (Hawker, 1985). The activity ratio of sucrose synthetase to sucrose phosphate synthetase typically is high in carbon importers




(Giaquinta, 1979). Reversibility of sucrose synthetase activity possibly is a key to regulation of assimilate storage in sink tissues (Giaquinta, 1979).
Little data are available to support or refute these hypotheses in Citrus. The major storage carbohydrate in citrus juice vesicles is sucrose (Kefford and Chandler, 1970), but the presence of increased levels of glucose and fructose in grapefruit, particularly in the latter part of development, suggests sucrose metabolism could be involved in photosynthate storage. Sucrose synthetase and sucrose phosphate synthetase activity was determined for immature and mature 'Eureka' lemon and mature 'Valencia' orange fruit tissues (Bean, 1960). Activity of sucrose synthetase varied with development and tissue, while no sucrose phosphate synthetase activity was detected. Invertase activity has been measured in grapefruit flavedo in conjunction with freezing (Purvis and Rice, 1983) and during development in juice vesicles (Kato and Kubota, 1978). The presence of sucrose synthetase and invertase activity in Citrus may be associated with phloem unloading and storage of sugars in juice vesicles.
The objective of this research was to elucidate features of the
anatomy and sucrose metabolizing system involved in the transport pathway and storage of sugars during development of 'Marsh' grapefruit (Citrus paradisi Macf.). It is envisioned that this research will enhance current knowledge of sugar transport and storage in sink tissues and may be valuable to the search for regulation of timing and quantity of sugar entry into grapefruit and possibly other citrus fruits.




CHAPTER II
LITERATURE REVIEW
Anatomy of Citrus Fruit
Current knowledge of the anatomy of citrus fruit has been detailed in several overviews (Schneider, 1968; Roth, 1977) and in developmental studies of specific cultivars (Ford, 1942; Bain, 1958; Holtzhausen, 1969). By far, the majority of citrus anatomical and morphological research at the light and ultrastructural level have dealt with the outer fruit peel (flavedo). These have included studies of its development (Scott and Baker, 1947), stomata (Klotz and Haas, 1933; Turrell and Klotz, 1940; Scott and Baker, 1947), wax production (Scott and Baker, 1947; Schulman and Monselise, 1970; Albrigo, 1972; Freeman et al., 1979), chloroplast and chromoplast interconversions (Thomson, 1965) and oil glands (Thomson, 1966a; 1966b; 1969; Brown and Barmore, 1981; Bosabolidis and Tsekos, 1982a; 1982b). Anatomical research on juice tissues primarily has addressed secondary fruit development (Lima, 1983), wax production (Ford, 1942; Dodd, 1944; King, 1947; Fahn et al., 1974; Espelie et al., 1980; Shomer et al., 1980), the presence of vascular tissue (Kordan, 1964) and the origin and ontogeny of juice vesicles or glandular hairs (Davis, 1932; Ford, 1942; Banerji, 1954; Amelunxen and Arbeiter, 1967; Roth and Lindorf, 1972; Roth, 1977). Discrepancies arise in the literature with respect to vascular tissue distribution, cellular composition and origin of tissues within the citrus fruit. Few published studies have related the physiology of citrus fruit growth patterns to anatomy (Bain, 1958; Bouma, 1959; Holtzhausen, 1969; Lima,




1983). Physiological aspects of sucrose metabolizing enzymes and dry matter accumulation, and the distribution of vascular and non-vascular tissues in this fruit have not been addressed in the literature.
Flowers of the Rutaceae, subfamily Aurantoideae typically are
perfect (unisexual by abortion), actinomorphic and 3 to 5 merous (Tillson and Bamford, 1938). The pistil of the flower arises as 9 to 10 connate carpels in 'Eureka' lemon (Ford, 1942) and 8 to 12 carpels in 'Valencia' orange (Bain, 1958). Ovules generally are formed in two rows in each locule on the marginal central placenta (Roth, 1977).
The vascular supply within citrus flowers persists in the fruit (Ford, 1942; Roth, 1977) and has been studied extensively in lemon flowers and immature fruit (Ford, 1942). Dorsal, septal and central vascular traces supplying the pistil in 'Eureka' lemon kCitrus limon (L.) Burm.] diverge almost simultaneously from the stele of the receptacle (Ford, 1942). Each carpel is associated with one dorsal, two septal and two central traces from which smaller traces diverge. The dorsal vascular trace occurs medianly along the tangential surface of a carpel, extends into the style and terminates below the stigma. Septal traces curve inward along septa separating the carpels and merge with central traces. At this point the primary phloem and xylem of these septal traces are inverted. Central traces supply the ovules and extend through the central axis of the ovary (Ford, 1942). At the stylar end of the ovary, these traces curve outward and also form inverted bundles. Fusion of bundles forms amphicribral bundles (phloem completely surrounding xylem) (Ford, 1942). Central bundles may extend into the style and terminate below the stigma (Ford, 1942; Tillson and Bamford, 1938).




The anastomose network of vascular tissue in the citrus fruit is confined to the mesocarp or albedo (Ford, 1942; Scott and Baker, 1947; Bartholomew and Sinclair, 1951; Schneider, 1968). Primary xylem of traces associated with carpels includes vessel members and tracheids with spiral thickenings, tracheids with reticulate thickenings and xylem parenchyma cells (Ford, 1942; Scott and Baker, 1947). As the immature 'Eureka' lemon fruit undergoes rapid expansion, dorsal, septal and central bundles may develop cambia and produce secondary phloem and xylem elements (Ford, 1942). Cross sectional areas of the dorsal bundles also may increase in the 'Washington' navel orange during the growth season (Holtzhausen, 1969). Xylem elements have been noted in juice tissues in lemon (Kordan, 1964), although these elements occur individually or in small clusters in relatively few juice vesicle stalks and do not connect with vascular traces in the rest of the fruit (Kordan, 1964).
The citrus fruit is a hesperidium that consists of exocarp
(flavedo), mesocarp (albedo, intersegmental membranes, central axis) and endocarp (segments, juice vesicles) (Bain, 1958). The pigmented flavedo occurs internal to a uniseriate epidermis and an uni- to triseriate hypodermis (Scott and Baker, 1947). Flavedo is composed of polygonal parenchyma cells that contain plastids. Oil glands and vascular trace endings also occur within this tissue (Ford, 1942; Scott and Baker, 1947). Intercellular spaces increase in number near the outer albedo. The abundant plastids usually convert from chloroplasts to chromoplasts as the fruit matures depending on environmental conditions. Leucoplasts and starch grains also occur in the flavedo (Scott and Baker, 1947). These starch grains are confined to chloroplasts and chromoplasts (Thomson, 1965). Fewer plastids are present in cells of the inner

M




flavedo (Roth, 1977). Although oil glands include derivatives of the epidermal and subepidermal tissue, the bulk of these glands are located in the flavedo (Bosabolidis and Tsekos, 1982a; Scott and Baker, 1947).
Albedo is the white, aerenchymatous tissue interior to the flavedo that contains the majority of the vascular supply for the citrus fruit (Ford, 1942; Scott and Baker, 1947). This tissue consists of thin-walled, irregularly-shaped parenchyma cells which typically have eight lobes. Intercellular spaces are large and numerous (Ford, 1942; Scott and Baker, 1947). Interfaces between adjacent parenchyma cells possess primary pit fields and plasmodesmata (Scott and Baker, 1947). The albedo is segregated by number of intercellular spaces and parenchyma cell lobes into a regular, compact outer layer, an intermediate zone and a loosely organized inner portion (Ford, 1942; Scott and Baker, 1947). Pectin increases in albedo cells of 'Valencia' orange during fruit development and results in the appearance of a dense albedo in edible fruit (Bain, 1958). Starch grains are prominent in parenchyma cells of 'Valencia' orange albedo (Bain, 1958). In pummelo, chloroplasts are present in this tissue early in fruit development, while later in the season these plastids differentiate into chromoplasts (Gross et al., 1983). Mitochondria and nuclei are common in albedo cells, while endoplasmic reticulum, dictyosomes and extracellular material are rare (Brown and Barmore, 1981).
The edible portion of the citrus fruit is the endocarp. Each carpel or juice segment is a locule surrounded by the carpellary wall or segment epidermis and filled with juice vesicles and seeds (Schneider, 1968). The segment epidermis consists of a uniseriate parenchymatous layer lacking stomata and covered with a thin cuticle (Roth, 1977). These




8
epidermal cells are thick-walled and form a "net-like pattern" (Roth, 1977). The segment epidermis of two adjoining segments is separated by compact parenchyma cells, and these three layers comprise the interlocule septae. Intercellular spaces form in the parenchyma cells at the center of these septae during later fruit development (Ford, 1942).
Juice tissues are multicellular vesicles derived from the carpellary wall (segment epidermis and subepidermal cell layers) (Roth and Lindorf, 1972). These juice vesicles are attached to the segment epidermis and are associated with vascular bundles, although the vascular network has not been reported to enter juice tissues (Dodd, 1944). Juice vesicles are initiated in locules of young orange ovaries 3 mm in diameter or less (Roth and Lindorf, 1972). Each juice vesicle has a "club-shaped" sac or head with a multicellular stalk at the enlarged end (Turrell and Bartholomew, 1939; Roth, 1977). The juice vesicle head has an external layer of elongated epidermal cells that enclose large, thin-walled "juice cells" (Fahn, 1979). The epidermal cells are elongated parallel to the longitudinal axis of the sac, thick-walled and covered with a cuticle of cutin, epicuticular wax and suberin (Ford, 1942; Dodd, 1944; King, 1947; Fahn et al., 1974; Roth, 1977). Parenchyma cells within the juice vesicle are composed of compact peripheral cells and loosely-organized central cells. Overall, these cells have been separated into four zones based on shape, size and orientation in immature grapefruit (Dodd, 1944; Fahn, 1979). These parenchymatous layers, beginning with the vesicle exterior, are (1) hypodermis, (2) concentrically-arranged flattened cells, (3) large, isodiametric cells and (4) centrally-located, thin-walled cells with lipophilic droplets. The central, highly vacuolated cells are similar in many respects to those of "spouting




glands" in Dictamnus albus (Rutaceae). Cells in spouting glands disrupt at maturity when cavities form within the mature juice sac (Davis, 1932; Banerji, 1954; Ameluxen and Arbeiter, 1967). These cavities in 'Eureka' lemon juice vesicles result from high turgor pressure and are not comparable to gland cells (Ford, 1942). Nuclei and variously-shaped plastids are present in the innermost cells of most citrus juice vesicles (Turrell and Bartholomew, 1939; Dodd, 1944; Scott and Baker, 1947) and these plastids may contain starch in immature citrus fruit (King, 1947).
Assimilate Partitioning in Citrus
Major reviews of citrus carbohydrate literature have been authored by Sinclair (1972; 1984) and Kefford and Chandler (1970). Carbohydrates in citrus have been studied in relation to tree and fruit cold hardening (Cameron, 1933; Sharples and Burkhart, 1953; Guy et al., 1981; Purvis and Rice, 1983), alternate bearing (Schaffer et al., 1985), flower and fruit set (Hilgeman et al., 1966; Powell and Krezdorn, 1977; Guardiola et al., 1984; Sinclair, 1984; Goldschmidt et al., 1985), juice quality (Widdowson and McCance, 1935; Bartholomew and Sinclair, 1941; Roy, 1945; Birdsall et al., 1961; Ting, 1969; Stepak and Lifshitz, 1971; Syvertsen and Albrigo, 1980), and most recently, assimilate translocation and partitioning within the fruit (Koch, 1984a; 1984b; 1985; Koch and Avigne, 1984). Sucrose is the major storage sugar in oranges and grapefruit, while reducing sugars predominate in lemon fruit (Sinclair, 1984). In addition to sucrose, fructose and glucose, sugars in citrus fruit include rhamnose and xylose (Stepak and Lifshitz, 1971). The ratio of sucrose to hexoses in grapefruit generally decreases in juice tissues and albedo during development (Harvey and Rygg, 1936; Hilgeman and Smith, 1940), and hexoses will sometimes predominate in mature peel (Harvey and Rygg,




10
1936). At a given stage of development, sugar composition is reported to be relatively constant throughout in the peel or in immature juice vesicles (Harvey and Rygg, 1936; Bartholomew and Sinclair, 1941). Edible oranges and grapefruit, however, show an increasing gradient of total soluble solids (primarily sugars and organic acids) from their stem to stylar ends, and from central core to periphery of juice tissues (Bartholomew and Sinclair, 1941; Ting, 1969). Ratios of sucrose to hexoses decrease from the ends and periphery towards the central core (Ting, 1969). Existence of sugar gradients within citrus juice tissues may suggest variability in ripening and the capability to accumulate assimilates (Sinclair, 1984).
Starch has been reported in citrus plants and immature fruit
(Cameron, 1933; Kordan, 1971a; 1971b; Yelenosky and Guy, 1977), but only in very low percentages. Conversion of photosynthates to starch in Citrus is likely to be only a minor component of early storage.
Photosynthate Translocation and Partitioning in Sink Tissues
Non-vascular tissues and apoplastic barriers in sink tissues can
separate sites of phloem unloading from assimilate storage. The presence of high plasmodesmatal densities typically suggest a symplastic transport path for photosynthates subsequent to phloem unloading (Helder and Boerma, 1969; Eleftheriou and Hall, 1983; Hayes et al., 1985; Gunning and Hughes, 1976; Warmbrodt, 1985). In contrast, apoplastic transfer is characteristic in sink tissues that include discontinuous cell layers (Felker and Shannon, 1980; Thorne, 1979), transfer cells (Felker and Shannon, 1980) and suberized or cutinized barriers (Eleftheriou and Hall, 1983; Gunning and Hughes, 1979). Developmental changes in vascularization (Hardham, 1976), plasmatubules (Harris et al., 1982) and




11
symplastic connections (Gunning, 1978; Seagull, 1983; Juniper, 1977; Schnepf and Sych, 1983) also may affect the translocation and storage of assimilates subsequent to phloem unloading.
Autoradiographic and 14C-photosynthate studies in soybean (Thorne, 1980) and corn (Orr et al., 1981a; 1981b) have demonstrated functional significance for cell layers along the transport path subsequent to phloem unloading. In soybean fruit, assimilates initially must pass into and throughout the seed coat before entering the extracellular space surrounding the embryo (Thorne, 1980). Photosynthates likewise must enter the intermediary pedicel region of corn kernels before final deposition (Orr et al., 1981a; 1981b).
Photosynthate Translocation and Partitioning in Citrus Fruit
Recent research of the translocation of 14C-photosynthates into grapefruit show that photosynthate transport from a source leaf to a fruit is restricted to juice segments aligned directly with the leaf (Koch, 1984a; Koch and Avigne, 1984). The predominant site of phloem unloading of assimilates to juice tissues is the dorsal vascular bundle with minor contributions from septal bundles. Central bundles primarily supply developing seeds (Koch, 1984b). Time course studies of labeled assimilate transfer into juice tissues have determined that transport slows dramatically subsequent to phloem unloading. Movement through the segment epidermis and juice vesicle stalks is extremely slow (Koch, 1984b; unpublished data). The ratio of labelled sucrose to hexoses decreases with distance from the site of phloem unloading (Koch, 1984b).




Sucrose Synthetase, Invertase and Assimilate Partitioning
Sucrose synthetase (UDPG:D-fructose 2-a-D-glucosyltransferase, EC
2.4.1.13) catalyzes the readily reversible synthesis and cleavage of sucrose:
Fructose + UDPG <------> Sucrose + UDP
The molecular weight of this enzyme varies from 350 to 540 KD (Graham and Johnson, 1978; Hisajima, 1979; Gross and Pharr, 1982; Morell and Copeland, 1985). A tetrameric structure with subunits of 90 KD has been proposed (Graham and Johnson, 1978; Morell and Copeland, 1985). This enzyme typically is studied in a soluble form, suggesting a cytoplasmic location. Both membrane-bound and insoluble forms are known to occur, however (Graham and Johnson, 1978; Bean, 1960).
Invertase (B D-fructofuranosidase, EC 3.21.26) is specific for fructofuranose moieties and acts by cleaving the glycosidic linkage between the bridge oxygen and the fructose residue (Sum et al., 1980):
Glucose + Fructose < ------ Sucrose + H20
This enzyme typically is considered a glycoprotein (Sum et al., 1980) and in yeast and Neurospora either lacks carbohydrate or contains at least 50% carbohydrate (Metzenberg, 1963; Gascon et al., 1968; Holbein et al., 1976). Up to five different forms of invertase have been found in higher plants (Sasaki et al., 1971), but the majority of plant tissues studied possess one or usually two forms. Acid and either neutral or alkaline forms of invertase are most common. Both can be soluble (cytoplasmic and/or vacuolar) or insoluble (cell wall or membrane bound). The molecular weight of invertase in different tissues varies from 15 to 380




KD (del Rosario and Santisopasri, 1977; Sum et al., 1980; Howard and Witham, 1983). The acid form generally is either in the range of 57 to 65 KD as in radish cotyledons (Howard and Witham, 1983) or between 200 to 380 KD as in banana (Sum et al., 1980) and sugar cane (del Rosario and Santisopasri, 1977). The molecular weight of neutral, soluble invertase forms in sugar cane were 125 and 160 KD with monomers of 66, 35 and 15 KD (del Rosario and Santisopasri, 1977).
Sucrose synthetase is considered to catalyze sucrose breakdown in vivo to provide UDPG for synthesis of non-starch reserve polysaccharides, starch, glycoproteins and glycolipids (Hawker, 1985). Invertase, on the other hand, has been studied in relation to cell expansion growth (Altman et al., 1982; Morris and Arthur, 1985), cold stress (Purvis and Rice, 1983), cold storage (Sasaki et al., 1971) and wounding (Matsushita and Uritani, 1974). Both of these enzymes also have been associated with photoassimilate accumulation in sink tissues (Glaziou and Gaylor, 1972; Dick and ap Rees, 1976; Giaquinta, 1979; Silvius and Snyder, 1979; Eschrich, 1980; Gross and Pharr, 1982; Claussen, 1983b). Two hypotheses have been advanced for their function in sugar-storing sinks, based on sugarcane stalks (Glaziou and Gaylor, 1972) and sugarbeet taproot (Giaquinta, 1979). Both suggest apoplastic phloem unloading and subsequent storage. In sugarcane, a free-space acid invertase is proposed to cleave sucrose unloaded from phloem. Glucose and fructose then are transported into the symplast of storage parenchyma cells and phosphorylated. The resulting phosphorylated fructose presumably is converted to sucrose phosphate by sucrose phosphate synthetase. Sucrose phosphate synthetase typically is considered a cytoplasmic enzyme (Glaziou and Gaylor, 1972); however, recent reports on sugar transport




14
into vacuoles isolated from red beet suggest this enzyme may be part of a "group translocator" located on the tonoplast (Thom et al. 1986). Sucrose subsequently is stored in the vacuole where a soluble acid invertase again hydrolyzes it. A soluble alkaline invertase believed to be located in the cytoplasm also has the capacity to cleave sucrose (Glasziou and Gayler, 1972). Activity of an intracellular (vacuolar) invertase in immature sugarcane stalks disappears during a concomitant increase in alkaline invertase activity. Soluble alkaline invertase may regulate sucrose storage and utilization (Glasziou and Gayler, 1972).
A somewhat different scenario has been proposed for import and storage of sugars in sugar beet taproot. In this plant, sucrose is believed to be translocated and accumulated intact, so substantial sucrose phosphate synthetase activity is unnecessary. The onset of sucrose storage in sugar beet taproot coincides with loss of soluble acid invertase activity and the appearance and increase of sucrose synthetase activity. Sucrose synthetase then, is believed to play a role in regulating the partitioning of assimilate storage and utilization in this system. The presence of sucrose synthetase also has been positively correlated with growth rate and dry matter accumulation in eggplant fruit (Claussen, 1983b). Activities of insoluble acid invertase, soluble alkaline invertase and sucrose phosphate synthetase were all low throughout development of the sugar beet taproot (Giaquinta, 1979).
Activity of sucrose synthetase has been demonstrated in young and edible 'Eureka' lemon and edible 'Valencia' orange [Citrus sinensis (L.) Osbeck] fruits (Bean, 1960). The soluble portion of orange and the particulate portion of lemon juice vesicles have shown this activity, but it is lacking in the soluble fraction of lemon and the particulate




15
portion of orange juice vesicles. No sucrose phosphate synthetase activity was detected in juice vesicles, although patterns of radiolabeling of sucrose from 14C-fructose in tissue slices suggested sucrose formation was from a phosphorylated form of fructose (Bean, 1960). Incorporation of radiolabeled glucose and fructose into sucrose in fresh tissue slices from various aged flavedo, albedo and juice vesicles and intact young fruit suggest that the albedo is the most active site of sucrose synthesis in lemons, although each tissue makes some contribution to sucrose synthesis. Lemon albedo tissue maintained a relatively constant rate of glucose conversion into sucrose throughout fruit growth, while the incorporation rate in flavedo and juice vesicles declined rapidly with age (Bean, 1960).
Acid invertase activity has been quantified in actively growing
juice vesicle explants of lemon (Altman et al., 1982) and both immature and mature juice vesicles of Satsuma (Kato and Kubota, 1978). Altman et al. (1982) suggested acid soluble invertase increased in lemon juice vesicles in the presence of active callus growth. Kato and Kubota (1978) found that the presence of both acid and alkaline invertases corresponded to low sugar content in immature juice vesicles, but that increasing sugar content in edible fruit took place when alkaline invertase was active. Highest acid invertase activity was found in the soluble fractions, while alkaline invertase activity occurred in the insoluble fraction. The distribution of alkaline invertase between the soluble and insoluble fractions changed with fruit development and shifted toward soluble fraction.




CHAPTER III
THE VASCULAR SUPPLY AND ASSIMILATE TRANSPORT IN GRAPEFRUIT Introduction
Structure and function appear to be closely related in many
transport tissues. Anatomical features in 'Marsh' grapefruit appear to influence the kinetics of photosynthate translocation (Koch, 1984a; Koch, unpublished data) and patterns of deposition (Koch and Avigne, 1984). Overall distribution of 14C-photosynthates within grapefruit depends on segment alignment to specific source leaves and the positioning of intervening vascular strands (Koch and Avigne, 1984). Assimilate movement to juice tissues is limited to three major vascular traces positioned outside a given segment (Koch, 1984a) (Fig. 3-1). Approximately half of the photosynthates entering juice tissues are unloaded from phloem in dorsal vascular strands and pass through the innermost cell layers of albedo, segment epidermis, juice vesicle stalks and finally to vesicle heads in stage III fruit (Koch, 1984a; Koch and Avigne, 1986) (Fig. 3-1). The remaining assimilates arrive via the two septal bundles and follow a similar route, although septal parenchyma cells are present instead of cells from the inner albedo. The central trace in the innermost portion of the fruit (central axis) supplies assimilates to ovules and seeds of a given carpel, but is completely separate from nearby juice tissues (Koch, 1984a) (Fig. 3-1).
Further structure/function associations are evident along the
post-unloading path of photosynthate transfer. Assimilates move slowly from phloem to juice vesicles and do so only by traversing phloem-free 16




Figure 3-1. Diagram of a transverse section through grapefruit showing the relationship between vascular and non-vascular tissues. Shaded area represents juice tissues. a=albedo; c=central bundle; d=dorsal bundle; jv=juice vesicle; s=septal bundle; se=segment epidermis.

M




18
parenchyma in a juice vesicle stalk (Koch, 1984a; Koch and Lowell, unpublished). This vascular distribution, a cutinized segment epidermis and complete lack of vascular tissue inside juice segments contributes to massive phloem unloading at highly localized sites.
Photosynthate import in numerous other sink tissues circumstantially has been linked to structural features in antomical studies. This has been particularly true for the zone of phloem unloading and tissues separating this zone from the site of final assimilate deposition. Vascularization associated with assimilate deposition can be discontinuous (Felker and Shannon, 1980; Oparka and Gates, 1981; 1982; Thorne, 1981; Kuo and Pate, 1985). Vascular tissue supplying photosynthates to the final site of deposition may contain only phloem (Hardham, 1976; Thorne, 1981; Kuo and Pate, 1985) or modifications in cells of the phloem (Sauter and Braun, 1972; Hardham, 1976; Oparka and Gates, 1981a; Hayes et al., 1985). Cross sectional areas of vascular strands associated with juice tissues in citrus fruit increase during development (Ford, 1942; Holtzhausen, 1969) by additional primary or secondary growth within bundles (Roth, 1977). These associations and those noted for citrus have motivated a closer examination of transport structures within this fruit.
Materials and Methods
Bain (1958) recognized three general developmental stages in
'Valencia' orange; stage I--cell division, stage II--cell enlargement and stage III-non-climacteric maturation. Similar growth stages were determined for 'Marsh' grapefruit (Citrus paradisi Macf.) in the present study, based on anatomical, morphological, volume and fresh and dry weight changes in the entire fruit and individual tissues (albedo,




segment epidermis and juice vesicles). 'Marsh' grapefruit (5 fruit minimum) were collected between 3 and 9 months of fruit growth from the outer canopy on the southern exposure of mature trees at approximately 3 m height on a monthly basis at commercial groves in Lake Wales, Florida (1985). Polar and equatorial diameters as well as fresh weights (excluding calyx) were measured for each fruit. Whole fruit volumes were calculated assuming the fruit to be spheres. Relative water content was determined for albedo, segment epidermis and juice tissues of each fruit.
Fruit of 'Marsh' grapefruit were collected at monthly intervals at the University of Florida (1983) and Lake Alfred, Florida (1984). The zone of phloem unloading and tissues involved in subsequent assimilate transfer were killed and fixed in half-strength Karnovsky's fluid (Karnovsky, 1965) with 0.075 M Pipes, pH 7.5, or 1.5% acrolein-3% glutaraldehyde-.5% paraformaldehyde (Hayat, 1981) followed by 1% Os04, dehydrated with an ethanol and acetone series and infiltrated with Spurr's resin (Spurr, 1969). One um sections were stained with 0.5% toluidine blue 0 in 0.1% NaCO3. Ultrathin sections (100 nm) were cut with glass knives, and then stained with uranyl acetate and lead citrate and mounted on formvar coated slot grids. Sections were analysed on a Philips 301 electron microscope. Fruit tissue also was killed and fixed in a 70% formalin-acetic acid-ethanol solution (FAA), and then dehydrated with a tertiary butanol series (Johansen, 1940) and infiltrated with Paraplast (56. 'C) (Sherwood Medical Industries). Sections (10-15 pm) were stained with a safranin-fast green series and examined with a Nikon light microscope.
Fresh tissue also was sectioned (40-80 pm) on a sliding microtome
and stained with safranin-fast green or 2% phloroglucinol in 95% ethanol




and concentrated HCl. Distances between the dorsal bundle and juice vesicle stalk base and stalk lengths were measured with a calibrated ocular micrometer from 15 pm sections and fresh or frozen material, respectively. Diagrams of whole juice segments and the relation of vascular tissue to juice vesicles were drawn to scale from fresh mature juice segments.
Results and Discussion
Fresh and dry weight changes are shown for whole fruit and
individual tissues of 'Marsh' grapefruit (Fig. 3-2) in relation to the three stages of growth defined by Bain (1958). Rates of increase in total fruit volume and fresh weight were similar throughout fruit growth. Only small gains in fresh and dry weight were observed for fruit tissues during the first 8- to 9-week cell division stage (stage I), which was early March to mid-May in central Florida. Increases were approximately linear for total fruit volume and fresh weight during cell expansion (stage II) which lasted 20-21 weeks (mid-May to late September). Fresh and dry weight accumulation by individual tissues also increased during this time. Based on dry matter accumulation, the albedo appeared to be the main sink for photosynthates within the fruit during stages I and II, while juice tissues became the predominant sink late in stage II. These findings are similar to those of Holtzhausen (1969). A slower rate of increase was observed in both fresh and dry weights of component fruit tissues during fruit maturation (stage III) lasting 9 to 12 weeks in this study.

M




I30(
0
20( 10( .D

I
140
w
IJ
100 C,,
Il _ALBEDO
C'60
0 SEG. EPIDERMIS
20*
18 C JUICE VESICLES
14
I
Co 14ALBEDO$
10
S6 SEG. EPIDERMIS
2
APRIL JUNE AUGI I OCT I
Figure 3-2. Fresh weights of whole fruit (A) and fresh and dry weights of albedo, segment epidermis and juice vesicles at biweekly to monthly intervals. Data represent the average of 5 fruit per measurement. Standard error (SE): 0.01< SE< (15.7 for fresh weight measurements,
0.003 (SE (1.1 for dry weight measurements.




Vascular Distribution in the Fruit
Distribution of major vascular bundles entering the fruit from stem to stylar end is diagramed in figure 3-3a. This vasculature is confined primarily to the albedo and central axis of the fruit. Vascular tissue entering the fruit radiates directly from that of the pedicel. This radial distribution coincides with a narrow distribution of 14C-photosynthates translocated from one source leaf to a specific portion of the grapefruit sink (Koch, 1984b). Vascular strands which supply assimilates to the peel diverge from the pedicel stele before those associated with juice tissues. Vascularization of the albedo is the most extensive in the fruit. More than 4 times the number of traces associated with juice segments enter the albedo, although the number of traces does not necessarily reflect the quantity of phloem tissue entering the albedo as compared to the rest of the fruit. The number of vascular traces branching from the main stele typically is in multiples of the number of juice segments present, usually 12 (range: 10 to 14) in the fruit (Fig. 3-3c).
Vascular traces comprising the stele in the grapefruit pedicel and at the point of entry into the fruit are collateral (phloem on the outer tangential surface of the xylem). A narrow transition zone is present interior to the fruit-pedicel juncture in the stem end of the fruit where collateral bundles become amphicribral (completely surrounded by phloem) (Fig. 3-3b, 3-3c, 3-6). The presence of additional primary xylem and phloem in bundles further into the fruit suggest that adjacent parenchyma cells may redifferentiate to form additional primary vascular tissue. The ratio of phloem to xylem in vascular bundles that enter and anastomose in the albedo increases as veinlets decrease in size with




Figure 3-3. The vascular organization within 'Marsh' grapefruit after approximately 12 weeks of development (early stage II). c=central bundle; d=dorsal bundle; l=locule; s=septal bundle.
A. Longitudinal diagram of a grapefruit depicting positions of major vascular bundles within the fruit. Arrows refer to the approximate location from which each section was taken.
B. Distribution of vascular bundles in the pedicel. Bar=20 pm.
C. Whorls of vascular traces diverge from the main stele and enter the albedo. Bar=lO pm.
D. Approximately 12 vascular traces branch from the stele to form dorsal bundles. One septal trace diverges from each dorsal trace. Central axis traces are derived from remaining stelar vascular tissues. Bar=f0 pm.
E. Each juice segment is associated with 1 dorsal, 2 septal and 1 central vascular trace. Bar=10 Fm.
F. Septal, dorsal and central vascular traces fuse to form a ring of approximately 12 vascular traces. Numerous traces from the albedo subsequently fuse with these bundles. Bar=1O pm.




pedicel (31 sections)

pedicelfruit juncture 19 sections)

..t:4'77

albedo above juice segments (7 sections) I
I

16 sections)

I
below cement A.---




distance from their site of entry as with other fleshy fruits (Roth, 1977). Near the flavedo, for example, these bundles may consist of a single primary xylem vessel member surrounded by small amounts of xylem parenchyma, possibly a vascular cambium and mostly primary and secondary phloem parenchyma. This increase in phloem may represent a morphological adaptation to minimize fruit desiccation and to increase the capacity of phloem unloading in the albedo.
Types of vascular bundles may vary within the same fruit. Apple may have collateral, bicollateral, amphicribral bundles and inverted central bundles (MacDaniels, 1940) similar to those in grapefruit and 'Eureka' lemon (Ford, 1942), Amphicribral bundles are common in fruit of such plant families as Rosaceae, Leguminosae, Anacardiaceae, Solanaceae, Cucurbitaceae (Roth, 1977) and Rutaceae (Ford, 1942). Inverted bundles also frequently occur in fruit, especially in those fruit where each locule of a compound ovary morphologically represents a modified leaf or carpe'l (MacDaniels, 1940). Primary xylem and phloem appear inverted at the stylar end of 'Eureka' lemon at the point of septal and central trace fusion, and prior to fusion of these traces with dorsal bundles (Ford, 1942). Bundle inversion does not appear to occur in grapefruit. However, traces at this stage are amphicribral, so inversion would be difficult to detect. The pattern of vascular fusion and bundle inversion does not necessarily indicate the direction of photosynthate movement at the stylar end of the fruit.
Vascular Supply to Juice Tissues
Four amphicribral traces are positioned around the exterior of each juice segment; 1 dorsal, 2 septal and 1 central bundle (Fig. 3-3d, 3-3e). Dorsal bundles branch from central vascular strands and subsequently give




rise to septal bundles (Fig. 3-3a, 3-3d). Septal bundles are shared between adjacent juice segments. Anastomoses between dorsal and septal vascular bundles occur, but primarily are confined to the stem end of the fruit (Fig. 3-3a). The locules or juice segments end and septal, dorsal and central bundles fuse to form a ring of discrete amphicribral bundles near the stylar end of the grapefruit (Fig. 3-3f).
Distance between the dorsal bundle (a site of phloem unloading) and the bases of juice vesicle stalks varies with the stage of fruit development (Fig. 3-4). The distance from dorsal bundle to stalk base decreases from stage I to stage III, averaging 260.0 pm + 16.33 (range 34.28-605.70 um) in mature stage III fruit. This results in the dorsal bundle being considerably closer to the segment epidermis (Fig. 3-4b). Septal bundles also are repositioned during development from positions in the albedo to midway between the albedo and central axis and between the segment epidermis of adjoining juice segments (Fig. 3-4). This change in position of the dorsal and septal bundles occurs during stage II growth as juice vesicles and segment epidermis expand. The actual distance to juice vesicle stalks may not change with development, but vascular bundles are adjacent to the segment epidermis on both radial surfaces. The heavily cutinized segment epidermis may limit the flow of assimilates into the albedo.
Each juice vesicle joins tissues exterior to the juice segment at a point near a dorsal or septal vascular strand, and areas of juncture thus occur primarily in three rows (Fig. 3-5). A small portion of the vesicle stalks also are associated with much smaller vascular traces that branch between dorsal and septal strands. The bases of these vesicles are at most 6 mm away from the nearest vascular strand. At least twice as many




A
S
d
S

p

I B

'a

Figure 3-4. Change in position of dorsal and septal vascular bundles with respect to albedo and segment epidermis during development. Shaded area represents juice tissues. a=albedo; c=central bundle; d=dorsal bundle; s=septal bundle; se=segment epidermis.
A. Transverse section of juice segment and associated vascular tissue during stage I growth (approx. 12 weeks).
B. Transverse section of juice segment and associated vascular tissue during stage III growth (approx. 24 wks).

0




Figure 3-5. Diagrams of segment epidermis and vascular bundles from the exterior of three grapefruit locules. Each point designates the site of the attachment of a single juice vesicle. Longitudinal lines represent dorsal, septal and central vascular traces. Transverse connections between these represent branching between dorsal and septal traces.







juice vesicles are present in the stem versus stylar half of the fruit and two-fold more juice vesicles also are associated with the dorsal versus septal surfaces of the juice segment (Fig. 3-5). Development of the Dorsal Bundle
Developmental changes in the dorsal vascular bundle are shown in
figure 3-6a and 3-6b. Primary phloem of young dorsal (and septal) traces includes parenchyma cells interspersed with narrow sieve tube members and companion cells. The majority of the primary phloem in dorsal bundles is adjacent to the largest portion of the albedo (Fig. 3-6a). Primary xylem in dorsal bundles occurs in radiating rows of narrow-lumened vessel elements and tracheids interspersed with xylem parenchyma (Fig. 3-6a). This also is true in septal bundles (data not shown). Dorsal, septal and central bundles increase in size by additional primary growth during the latter part of stage I and early part of stage II. A vascular cambium differentiates during stage II in these bundles and produces new cell layers of secondary phloem and xylem parenchyma (Fig. 3-6b). This growth in dorsal bundles is most extensive in the phloem adjacent to the segment epidermis and juice vesicle stalks. Vascular cambia occur less frequently in septal traces and rarely in central traces. The vascular supply in 'Washington' navel orange (Holtzhausen, 1969), lemon (Ford, 1942) and many other fruits (Roth, 1977) also increases by additional primary or secondary growth in individual vascular bundles. A similar increase in xylem and phloem tissues in developing seeds of Pisum sativum L. coincides with a rise in the supply of assimilates in this fruit (Hardham, 1976).
Unlignified phloem fibers typically develop in the portion of the dorsal bundle furthest from the juice vesicles during stage I growth.




Figure 3-6. Transverse sections of dorsal vascular bundles at different developmental stages. (Inner albedo at top of page).
A. Dorsal vascular bundle after approximately 12 weeks of fruit development (early stage II). Note amphicribral structure. p'=primary phloem; x'=primary xylem. Bar=5 pm.
B. Dorsal vascular bundle after approximately 24 weeks of fruit development (stage III). bc=bundle cap; c=vascular cambium; p=primary and secondary phloem; x=primary and secondary xylem. Bar=10 pm.




32
These fibers subsequently lignify to form a band or bundle cap that separates primary phloem into two zones during stage II growth. Primary phloem peripheral to the bundle cap is crushed by late stage II to early stage III growth (Fig. 3-6b). Fibers also lignify sporadically around the perimeter of the dorsal bundle during this time. A similar developmental sequence is not evident in other vascular bundles supplying juice segments. Phloem fibers occur occasionally in septal bundles, but none are found in central bundles.
Formation of a bundle cap in dorsal bundles coincides with the
switch in dominant sink tissue from albedo to pulp on a fresh and a dry weight basis during stage II development (Fig. 3-2). Unlike pigment strand suberization in rice caryopsis (Oparka and Gates, 1982), fibers of the bundle cap in grapefruit lack symplastic continuity and form a barrier to intercellular movement of assimilates. Assimilates entering the fruit from the dorsal bundle prior to this time, therefore, may unload from phloem tangentially into both albedo and juice tissues during early fruit development. Transfer subsequent to stage II is hindered in the direction of the albedo.
Conclusions
A functional significance in relation to photosynthate transfer is suggested by several structural features of vascular and non-vascular portions of the transport path in grapefruit. Three highly localized zones of phloem unloading are indicated for juice segments by 1) lack of vascular strands in the interior of juice vesicles; 2) limitation of cellular continuity between phloem outside segments to the 3 major vascular bundles (1 dorsal, 2 septal); and 3) isolation of juice tissues by a continuous, cutinized epidermis surrounding segments and juice




33
vesicles. Other aspects of vascular structure and distribution that appear most closely associated with photosynthate deposition include direct alignment of vascular tissue entering the fruit to juice segments, increase in the phloem to xylem ratio in individual vascular traces and changes in the dorsal vascular structure. In the latter, changes in position and vascular tissue and formation of a bundle cap in dorsal bundles coincide with the stage II switch in sink strength (based or dry matter partitioning) between albedo and juice tissues.




CHAPTER IV
NON-VASCULAR TISSUES IN GRAPEFRUIT DURING DEVELOPMENT Introduction
The spatial arrangement of vascular and non-vascular tissues in citrus fruit influences the kinetics of photosynthate translocation (Koch, 1984a; Koch, 1985; Koch, unpublished) and patterns of deposition (Koch and Avigne, 1984). Assimilate movement to juice tissues also is limited to three major vascular traces positioned outside a given juice segment (Koch, 1984a). This vascular distribution, a cutinized segment epidermis and the complete lack of vascular tissue inside juice tissues contributes to the apparently extensive phloem unloading at highly localized sites (see previous chapter). Assimilates move slowly from phloem to juice vesicles only by traversing the length of phloem-free parenchyma in a juice vesicle stalk (Koch, 1984a; Koch, 1985).
Anatomical studies of other sink tissues have suggested there is a close relationship between structural features and processes of photosynthate import. This has been particularily true in the zone of phloem unloading and along the path of subsequent assimilate movement to the site of final deposition. Non-vascular portions of this path typically include modifications such as numerous symplastic connections (Helder and Boerma, 1969; Gunning and Hughes, 1976; Eleftheriou and Hall, 1983; Hayes et al., 1985; Warmbrodt, 1985), transfer cells (Felker and Shannon, 1980; Harris et al., 1982), discontinuous cell layers (Felker and Shannon, 1980) or suberized/cutinized layers (Gunning and Hughes, 1979; Oparka and Gates,




35
1981a; 1982; Eleftheriou and Hall, 1983). Developmental changes also may occur in the extent of symplastic connections (Juniper, 1977; Gunning, 1978; Schnepf and Sych, 1983; Seagull, 1983) or cellular modifications such as plasmatubules (Harris et al., 1982).
The citrus fruit, a hesperidium, is composed of pigmented exocarp (flavedo), white mesocarp (predominantly albedo) and endocarp (segment epidermis and juice vesicles) (Bain, 1958). Albedo is the highly vascularized, spongy aerenchyma of the inner peel (Ford, 1942; Scott and Baker, 1947). Endocarp is that portion of the citrus fruit which typically is eaten and, unlike albedo, it lacks vascular tissue (Dodd, 1944). This zone in oranges is divided into 8-12 locules or segments (Bain, 1958; Holtzhausen, 1969). Each of these contains numerous multicellular juice vesicles (collectively the pulp) and a variable number of seeds. Individual locules and their vesicles are surrounded by a single, continuous epidermal layer with a thin cuticle (Schneider, 1968; Roth, 1977). Juice vesicles are derived from cell divisions of the outer and tangential segment epidermis and subepidermal cell layers (Bain, 1958; Schneider, 1968; Roth and Lindorf, 1972). Expanding juice vesicles become variously "club-shaped" (Roth, 1977) with a thin, parenchymatous stalk joining the vesicle head to tissues outside the juice segment at a site near a major vascular strand. Three vascular strands are associated closely with the exterior of each juice segment (Ford, 1942; Koch, 1984a). All juice tissues inside a single juice segment, therefore, are supplied with photosynthates by one dorsal bundle on the outer, tangential surface or one of the two septal bundles located between septae on either of the sides shared by adjacent segments (Koch,




36
1984a). Seeds of a given segment receive assimilates from one vascular strand in the central axis.
Citrus juice vesicles have been compared to "spouting glands" or glandular hairs on leaves and inflorescences of Dictamnus albus L. (Rutaceae) (Anelunxen and Arbeiter, 1967). These glands also consist of a multicellular stalk and globular body, but include a secretory cavity. It is lined with several layers of secretory cells and contains essential oils (Anelunxen and Arbeiter, 1967).
The following study was undertaken to examine what appeared to be a close association between structure and function for assimilate transport into citrus fruit, and because of the highly localized nature of those processes in this system. The objective of this research was to elucidate anatomical features of the non-vascular transport pathway and sites of sugar deposition at different developmental stages in 'Marsh' grapefruit (Citrus paradisi Macf.).
Materials and Methods
Materials and methods were detailed in chapter III. Presence or
absence of primary xylem tracheary elements was determined in 900 juice vesicle stalks from the middle 1 cm of 30 mature grapefruit by staining fresh cut stalks with 2% phloroglucinol in 95% ethanol and concentrated HCI. Stalks in which xylem elements were found were washed with deionized water, stained with toluidine blue 0 and fresh mounted for photographs. Absence of phloem cells was verified by observation of preserved fruit tissue.
Observations of plasmodesmata in cells of juice vesicle stalks were taken from ultrathin sections (approximately 60-70 nm) of stage II fruit (4-5 months) cut with a diamond knife and mounted on 300 mesh copper

I




37
grids. These sections were stained with uranyl acetate and lead citrate followed by carbon coating for stabilization.
Plasmodesmatal densities were determined morphometrically in juice vesicle stalks of mature, stage III fruit (8 months). Ten juice vesicle stalks were randomly selected from juice tissues in the median centimeter of four fruit and fixed, dehydrated and embedded in Spurr's resin as described in chapter III. To obtain a random sample of cell wall interfaces, 5 ultrathin (approximately 100 nm) serial sections were cut with a diamond knife followed by at least ten I um sections and repeated according to Weibel (1969). Ten non-consecutive sets of five serial sections from each of 10 juice vesicle stalks (100 grids, 500 sections total) were mounted on formvar coated slot grids and stained with uranyl acetate and lead citrate. A maximum of 16 random photographs were taken of one section per grid with a Philips 301 transmission electron microscope according to Weibel (1969). Transverse sections, 1 um thick, were taken from the same 10 juice vesicle stalks, stained with 0.5% toluidine blue in 0.1% NaCO 3and photographed with a Nikon light microscope. Partial and total cross sectional areas and plasmodesmatal densities per length cell wall interface (longitudinal and transverse) were calculated for peripheral and central stalk cells. Values were determined from 8X10 photographs using a digitizer and morphometric analysis software designed by Dr. Kenneth Curry, University of Florida. Apoplast to symplast ratios were estimated using a double square (6mm2and 24 mm2) grid (Weibel, 1969). Total number of measurements was determined in each set of data by a cumulative standard error of less than 10% expressed as percent of the mean (Bolender, 1978). Plasmodesmatal frequencies were determined per cell wall interface and used to




38
estimate frequencies per cell wall pm-2according to Robins and Juniper (1980).
Plasmodesmatal distribution in juice vesicle stalks was
non-parametric. The Wilcoxon-Mann-Whitney two-sample test for unequal sample sizes was used for analysis of significant differences (5% level) among frequencies of longitudinal and transverse cell walls in both peripheral and central juice vesicle stalk cells (Mann and Whitney, 1947).
Results and Discussion
Parenchyma Adjacent to Vascular Bundles
The first non-vascular tissue traversed by assimilates enroute to juice vesicles is a narrow parenchymatous layer surrounding dorsal and septal bundles (Fig 4-1). This tissue is the innermost albedo around dorsal vascular strands. Most albedo is composed of irregularly-shaped parenchyma cells separated by numerous, large intercellular spaces as previously described by Bain (1958). In contrast, the innermost albedo cells that separate dorsal bundles and juice vesicles are densely packed and are thinner-walled with a more regular elongated shape (Fig. 4-1). Intercellular spaces remain small in this area and plasmodesmata persist throughout fruit development. Thin-walled parenchyma cells also surround the septal bundles and are structurally similar to inner albedo cells. Parenchyma cells immediately adjacent to any of the three major vascular strands develop a large vacuole to cytoplasm ratio during stage II growth (5-6 months). The large vacuoles in these cells indicate a small, but potentially significant, capacity for photosynthate storage subsequent to phloem unloading. This capacity may or may not be analagous to the




Figure 4-i. Zone of phloem unloading and subsequent non-vascular transport tissues within the grapefruit. a=albedo; ia=inner albedo; st=juice vesicle stalk. Bar=10 pm.







41
transient, reservoir-type function ascribed to parenchyma of the maize pedicel (Felker and Shannon, 1980).
Segment Epidermis
The segment epidermis or carpel wall shown in figures 4-1 and 4-2 is not in the physical path of photosynthate transfer into juice vesicles, but 14C-assimilate transport studies in grapefruit suggest the segment epidermis plays a important role in photosynthate movement (Koch, 1984a). The uni- to biseriate epidermal layer contains no stomata and is continuous with that of the entire juice vesicle. Cells of the segment epidermis, however, differentiate earlier and are more highly vacuolated than those surrounding juice vesicles. An irregular uni- to biseriate hypodermal cell layer occurs adjacent to the segment epidermis (Fig. 4-2a). Epidermal cells initially are columnar and become tabular with maturity, while hypodermal cells may be either columnar or tabular throughout development (Fig. 4-2a, 4-2b). Plasmodesmata are present along the radial faces of cells within the segment epidermis, and also extend from their outer tangential faces to join cells of the hypodermis. Cavities may develop within a single plasmodesma or inside radial walls where several plasmodesmata meet. Epidermal cells become highly vacuolated during development and their appearance shifts from that of metabolically active cells to one more similar to storage cells.
Inner tangential walls (adjacent to juice vesicles) of epidermal cells are extremely thick and possess a striate cuticle. The striate cuticle often is 2-parted; an electron dense outer layer and an inner, less electron dense layer. The inner portion of the cuticle becomes invaginated upon fruit maturity. This cuticular layer consists of cutin and wax, but no suberin in grapefruit (Espelie et al., 1980). The




Figure 4-2. Transverse section of the segment epidermis, adjacent albedo cells at different developmental stages. h=hypodermis; se=segment epidermis.

hypodermis and a= albedo;

A. Stage I segment epidermis (approx. 12 wks). Bar=1 pm. B. Stage III segment epidermis (approx. 28 wks). Bar= 0.1 pm.

q a
4k 4Lw A---!U
6=6




continuous, cuticular epidermal layers of the segment epidermis and juice vesicle are likely to limit photosynthate passage into juice segments to points of juice vesicle stalk attachment to the carpel wall.
Cell walls between segment epidermis and hypodermis thicken during the late phase (12 mo after anthesis) of grapefruit maturation (stage III) and separate along the middle lamella. Thin cuticles then develop along cell walls of these adjoining surfaces and form what would appear to be at least a partial a barrier to the apoplastic and/or symplastic transfer of photosynthates between these cell layers. At this point in development, the segment epidermis therefore may have a decreased role in photosynthate storage and/or translocation. Juice Vesicles and Vesicle Stalks
No vascular connections exist within juice vesicles or between juice vesicles and the albedo, but small clusters or isolated primary xylem tracheary elements occasionally may form in stalks (Fig. 4-3). When present, these cells are narrow-lumened with oblique endwalls, simple perforation plates and bordered pits. Only up to 13.3% of the mature juice vesicle stalks possess these isolated tracheary elements.
Mature vesicle stalks vary greatly in length (Table 4-1), but the transport path of photosynthates from sites of phloem unloading (dorsal and septal vascular bundles) to juice vesicle storage cells in mature tissues may be 21 mm or greater (data not shown). Labeling studies show 14C-photosynthates move through the innermost albedo and the entire length of these parenchyma strands before final deposition in vesicles (Koch, 1984a; 1985). Analogous non-vascular areas, with varying degrees of specialization and function also occur in other sink tissues such as Pisum sativum L. (Hardham, 1976), maize root (Warmbrodt, 1985), rice




44
Figure 4-3. Longitudinal fresh section of juice vesicle stalk with clusters of isolated primary xylem tracheary elements in juice vesicle stalks. Bar=0.2 Fm.




45
Table 4-1. Dimensions of Mature Juice Vesicle Stalks from 'Marsh' grapefruit.

Length (mm): 6.62 Cross sectional area
total
peripheral
cells
central
cells

(+ 0.5z) (pm2 ): 29.0 (+ 1.7)
19.4 (+ 0.6) 9.6 (+ 1.3)

Standard error of the mean




46
caryopsis (Oparka and Gates, 1982) and maize kernels (Felker and Shannon, 1980). However, the length of the translocation path between the last phloem cells and site of final assimilate storage in grapefruit is one of the longest described (Koch, 1985).
Juice vesicle stalks are part of the multicellular vesicles which are the final site of deposition for assimilates in grapefruit segments (Fig. 4-4) Elongation of vesicle stalks begins early in stage I and is complete by stage II growth (Fig. 4-4c, 4-4d). As a result, the length of non-vascular tissue that photosynthates must traverse subsequent to phloem unloading increase during fruit growth. Cell length can contribute to factors determining whether transport through a given tissue is apoplastic or symplastic. In extremely long parenchyma cells, the rate of cytoplasmic streaming could limit intercellular transport (Gunning and Overall, 1983).
Structure of parenchyma cells varies within vesicle stalks.
Peripheral cells are narrow, elongate, and thin-walled, while central cells are larger and thinner-walled (Fig. 4-4a, 4-4b). The total cross sectional area occupied by peripheral cells within the stalk is 2-fold greater than that of central cells (Table 4-1). Peripheral cells remain smaller and more densely cytoplasmic than interior cells. Morphometric measurements show that the total apoplast, including cell walls and intercellular spaces, comprises approximately 18% of the total cross sectional area of central cells but 24% near the perimeter of the vesicle stalk. Parenchyma cells in the vesicle stalks are initially isodiametric to slightly oblong, but elongate extensively during stage II development, proceding from peripheral to central cells (Fig. 4-4c, 4-4d).




Figure 4-4. Non-vascular juice tissues at different developmental stages. c=central cells; e=epidermis; pl=peripheral cells.
A. Transverse section of stage I juice vesicle stalk (approx 12 wks). Bar=1 pm.
B. Transverse section of stage III juice vesicle stalk (approx 20 wks). Bar=1 pm.
C. Longitudinal section of stage I juice vesicle stalk (approx 12 wks). Bar=0.2 pm.
41
D. Longitudinal section of stage III juice vesicle stalk (approx. 27 wks). Bar=0.2 pm.
E. Transverse section of stage I juice vesicle head (approx. 11 wks). Bar=0.2 pm.
F. Transverse section of stage III juice vesicle head (approx. 25 wks). Bar=0.2 Vm.







49
Each stalk cell possesses a large nucleus with one to several
nucleoli. Multivesicular bodies and small vesicles are common, as are plastoglobuli and long strands of endoplasmic reticulum. Sub-cellular vesicles increase slightly in number upon maturity, especially in peripheral parenchyma cells. Golgi apparati are extremely rare, but appear active based on numbers of associated sub-cellular vesicles. The majority of parenchyma cells in the vesicle stalk posses one large or several small vacuoles by 3 months of growth (early stage II). Central cells in particular become highly vacuolated by stage II and may act as temporary reservoirs for assimilates enroute to juice vesicle heads. Pulse-chase labeling studies have shown an extremely slow turn-over period for 14C-photosynthates moving through juice vesicle stalks (Koch, 1984a; Koch, 1985; Koch, unpublished data).
Juice vesicles also contain chloroplasts with starch granules
during stage I and early stage II growth. Unlike starch grains in the albedo of 'Valencia' orange (Brown and Barmore, 1981), starch grains in grapefruit tissues disappear by early stage II growth during chloroplast differentiation into chromoplasts. Chromoplasts typically include achlorophyllous membranes and numerous plastoglobuli. Starch has been reported in citrus stems, roots and fruit (Cameron, 1933; Kordan, 1971a; 1971b; Yelenosky and Guy, 1977; Brown and Barmore, 1981), but only in very low amounts. Major storage products in grapefruit are soluble sugars (Sinclair, 1984), thus conversion of photosynthates to starch probably is a minor component of early storage in the fruit.
Both young and mature heads of juice vesicles share most features of juice vesicle stalks, yet differ in some important respects (Fig. 4-4d, 4-4f). A major difference is a proliferation of central cells.




50
Parenchyma cells in vesicle heads also enlarge considerably more and elongate much less than in vesicle stalks. Cell walls become increasingly thin during late stage II growth (6-7 months) and central cells possibly may rupture in mature fruit (Fig. 4-4f). Such cavity formation in juice vesicles could result from lysigeneous (Banerji, 1954) or schizolysigeneous cell breakdown (Davis, 1932). Other authors contend that central cell loss in juice vesicles of 'Eureka' lemon was caused by high turgor pressure and that juice vesicles should not be compared to glands (Ford, 1942).
Citrus juice vesicles have been grouped into a broad category of glandular hairs (Davis, 1932; Banerji, 1954; Anelunxen and Arbeiter, 1967; Fahn, 1979). Secretion from symplast to apoplast within other glandular structures is envisioned to occur in one of two ways. Granulocrine secretion takes place when sugars are packaged into sub-cellular vesicles which subsequently fuse with the plasmalemma and dispell their contents. The eccrine mechanism involves only the relatively direct transport of sugars across the plasmalemma (Fahn, 1979). Circumstantial evidence for glandular secretion in cells other than those of citrus juice vesicles includes the presence of abundant dictyosomes, ribosomes and mitochondria, a well-developed endoplasmic reticulum and in the case of granulocrine secretion, numerous membrane vesicle fusions at the plasmalemma (Kuo and Pate, 1985). These characteristics were not observed in maturing and mature citrus juice vesicles, which instead, contained cells with large vacuoles and few organelles. These data suggest that the physiology and sub-cellular structure of maturing and mature juice vesicles probably are not comparable to glandular hairs.




Plasmodesmata
Plasmodesmata are present in lateral and end walls of cells
throughout juice vesicle stalks. These subcellular structures are not necessarily confined to primary pit fields and are clustered primarily in groups of varying size (Fig. 4-5a, 4-5c). Both simple and bifurcated plasmodesmata occur(Fig. 4-5b), with outer diameters ranging from approximately from 36 to 46 nm. These values are somewhat smaller than the 60 nm diameter cited as standard for a plasmodesma by Gunning and Overall (1983), but diameters are known to vary widely with position in the cell wall (Gunning and Overall, 1983).
Plasmodesmatal frequencies are significantly higher among central
versus peripheral cells in mature (stage III) juice vesicle stalks (Table 4-2). No significant differences were evident in this respect between longitudinal and end walls of cells within a given area of tissue. If photosynthate transport through these stalks was symplastic, end wall specialization could be expected. In reviews by Robards (1976) and Gamalei (1985), plasmodesmatal densities are described as typically increasing with the degree of cellular specialization for assimilate translocation. Plasmodesmatal densities reportedly range from 7 to 140
-2
Fl cell wall where the symplast is believed to be the primary transport path. Examples studied have been meristematic cells (Arisz, 1969; Warmbrodt, 1985), tangential walls of endodermal cells in roots (Helder et al., 1969; Seagull, 1983; Warmbrodt, 1985), secretory cells (Gunning and Hughes, 1976; Meyberg and Kristen, 1981) and bundle sheath cells in C-4 leaves (Evert et al., 1977; Fisher et al., 1982). In contrast, parenchymatous cells with less obvious roles in direct symplastic transport of assimilates have plasmodesmatal densities ranging from 0.1




52
i +~ d. N;+ + +
- ;t.Figure 4-5. Plasmodesmata in the juice vesicle stalk. A. Peripheral cells of a juice vesicle stalk. Arrows denote plasmodesmata. Bar=1 Fm. B. Plasmodesmata in mature juice vesicle stalk (approx. 27 wks) can be simple, form a central node or bifurcate as pictured here. Bar=0.2 pm. C. Primary pit field in cell wall of juice vesicle stalk cell. Bar=O.1 pm.




Table 4-2. Plasmodesmatal frequencies within parenchyma tissues of juice
vesicle stalks from 'Marsh' grapefruit.Z

Position of parenchyma tissue within juice vesicle stalk

plasmodesmatal frequency
at interface between 2 cells

-l
(pron wall)

-2
(pm~i wall)

Peripheral cells:
longitudinal walls
transverse walls
Central cells:
longitudinal walls
transverse walls

0.27 (+ 0.08)** 0.24 (+ 0.07)*
0.39 (+ 0.12)** 0.38 (+ 0.09)*

2.20 1.96
3.78 2.96

(+ 0.72) (+ 0.58)
(+ 1.05) (+ 0.76)

ZPlasmodesmatal frequency along a linear portion of the interface between two cells was determined from 202 to 505 um cell wall length as described by Weibel. Subsequent conversion to frequency per unit area was done with equations of Robins et al. (1980) which yielded values 3% less than those of Robards (1976). Values otherwise represent maximum frequencies which could be obtained from these tissues because a plasmodesma was counted even if not wholely visible in a given section.
** Different at p 0.05.
* Different at p 0.10.




54
to 10 pnm2cell wall. Densities vary widely because of differences in species and methodology. Offler and Patrick (1984) relate low plasmodesmatal frequencies to a primarily apoplastic radial transfer of photosynthates in bean stems. Plasmodesmatal frequencies in grapefruit juice vesicle stalks then, may support a relatively low rate of photosynthate inflow to juice vesicle heads, but these rates probably are not significant of symplastic specialization for assimilate transport.
Conclusions
The longest component of the phloem-free transport path in
grapefruit is the parenchymatous juice vesicle stalk. This elongates typically up to 16 mm in mature fruit. Cells are long and narrow-lumened at the stalk periphery, but short and wide-lumened in the center. Plasmodesmatal frequencies in mature stalk cells suggest a lack of specialization for symplastic transfer. Other parenchyma nearer the vascular bundles also remain non-specialized relative to assimilate transport. Segment epidermis does not appear to be directly in the path of photosynthate transfer, but several features suggest it may be important in this process. This cutinized, uni- to biseriate cell layer is continuous with that of juice vesicles and is likely to limit photosynthate passage into these structures to points of juice vesicle stalk attachment to the carpel wall. Segment epidermal cells become highly vacuolated during development and their appearance shifts from that of metabolically active cells to one more similar to storage cells. Cells of mature juice vesicles, the final site of assimilate storage, have low cytoplasm to vacuole ratios and show no anatomical modifications for secretion.




CHAPTER V
SUCROSE METABOLIZING ENZYMES AND ASSIMILATE PARTITIONING IN SINK TISSUES OF DEVELOPING GRAPEFRUIT
Introduction
Data from several plant systems have suggested widely varied roles for sucrose metabolizing enzymes in phloem unloading and subsequent storage of photoassimilates in sink tissues (Glaziou and Gaylor, 1972; Dick and ap Rees, 1976; Giaquinta, 1979; Silvius and Snyder, 1979; Eschrich, 1980; Claussen, 1983b; Wolswinkel, 1985). The three major enzymes that catalyze either synthesis or hydrolysis of sucrose are sucrose phosphate synthetase, invertase and sucrose synthetase. Sucrose phosphate synthetase (sucrose synthesis) and invertase (sucrose cleavage) generally are considered irreversible enzymes in vivo (Hawker, 1985). Sucrose synthetase, however, is readily reversible (Hawker, 1985).
The onset of assimilate accumulation in various sink tissues often is accompanied by the appearance of either acid invertase (Shannon and Dougherty, 1972; Echeverria and Humphreys, 1984) or alkaline invertase (Kato and Kubota, 1978; Giaquinta, 1979). Photosynthate-importing tissues frequently also have a higher ratio of sucrose synthetase activity to that of sucrose phosphate synthetase (Giaquinta, 1979). The photosynthetic product most commonly transported in plants is sucrose, but sink tissues store large quantities of assimilates in additional forms such as other soluble sugars and starch (Jenner, 1980; Hawker, 1985). Sucrose synthetase is believed to invert sucrose in vivo thus providing substrate for starch synthesis in starch-storing 55




56
tissues (Hawker, 1985). Still, starch content and sucrose synthetase activity in cotton are not necessarily related (Hendrix and Huber, 1986). In addition, sucrose synthetase may function in the synthetic direction in sugar-storing tissues, such as sugar beet taproot, mature sugar cane stalks, cucumber fruit and eggplant fruit (Hatch, 1963; Giaquinta, 1979; Gross and Pharr, 1982; Claussen, 1983b, 1985). Further study is required in both starch and sugar-storing sink tissues to determine the function of sucrose synthetase in photosynthate accumulation.
Sucrose synthetase activity has been demonstrated in the soluble portion of 'Valencia' orange and the particulate portion of 'Eureka' lemon juice tissues (Bean, 1960). No sucrose phosphate synthetase activity has been detected in these tissues. Incorporation of radiolabeled glucose and fructose into sucrose in fresh tissue slices from various aged flavedo, albedo and juice tissue, and intact young fruit suggest that the albedo is the most active site of sucrose synthesis in lemons, although each tissue makes some contribution to sucrose synthesis (Bean, 1960). Lemon albedo tissue maintained a relatively constant rate of sucrose synthesis throughout fruit growth, while this rate declined rapidly during development of flavedo and juice tissues (Bean, 1960).
Acid invertase activity has been measured in extracts from growing juice vesicle explants of lemon (Altman et al., 1982) and both immature and mature juice vesicles of Satsumma (Kato and Kubota, 1978). Kato and Kubota (1978) found acid invertase from immature juice vesicles was active only when their sugar content was low. Alkaline invertase, however, was active during periods of increasing cellular sugar content in mature fruit. High acid and alkaline invertase activities in citrus




57
both could contribute to production of hexoses for activily growing cells (Kato and Kubota, 1978).
Most citrus fruit accumulate sugars over a relatively long growth period (8-9 months) (Sinclair, 1972; 1984). Assimilates move into both the heavily-vascularized and non-vascular portions of the fruit during development, but a mid-season change occurs in dry matter accumulation between these tissues. In the present work, activities of sucrose synthetase and invertase (acid, alkaline, soluble and insoluble forms) were assayed during development of individual 'Marsh' grapefruit tissues in conjunction with a study of sugar accumulation and dry matter partitioning. The presence and developmental fluctuations of sucrose synthetase and invertase activity in these tissues are discussed in relation to variations in relative sink strength. Possible roles are suggested for sucrose synthetase and the four forms of invertase in phloem unloading and/or subsequent storage of sugars in these tissues.
Materials and Methods
'Marsh' grapefruit (Citrus paradisi Macf.) were collected from the outer, southern canopy of 6 mature trees in Lake Wales, Florida biweekly in May and June and monthly from July to November, 1985. Each replication consisted of 4 fruit from 2 trees. Fruit also were collected in a similar manner from 9 mature trees in Lake Alfred, Florida in 1984. In this case, 6 fruit from 3 trees constituted one replication. Each measurement was replicated 2 or 3 times. Equatorial and longitudinal diameters were measured, then the individual fruit were weighed and separated into albedo (inner peel), segment epidermis (including inner albedo and central axis) and juice vesicles. Relative water content of each tissue also was determined. Sucrose and glucose levels were




58
measured by thiobarbituric acid (Percheron, 1962) and glucose oxidase methods (Sigma, Bull. #510), respectively. Tissues were frozen and kept at -806C until use. Enzyme activity varied little between extracts from fresh versus frozen tissues, and remained constant under stated conditions at least one year (data not shown). Based on results from 1984, fruit with an average equatorial diameter between 30-40 mm, 79-85 mm and 89-95 mm were designated stage I, stage II and stage III fruit, respectively.
Extraction and partial purification of enzymes were conducted at 0 to 5 C. Sucrose synthetase was extracted by homogenizing for 5 min with a Brinkman polytron tissue homogenizer using a ratio of 1 g:5 ml 200 mM Hepes, pH 7.5, 20 mM sodium ascorbate, 10 mM cysteine-HCl, 5mM MgC2 5mM DTT and 10% PVPP. Homogenate was filtered through 4 layers of cheesecloth, rinsed with 5 ml extraction buffer minus PVPP and the filtrate centrifuged at 20,000g for 10 min. Proteins in the supernatant were precipitated with 80% (NF1 )2S0 and centrifuged as above for 20 min. The pH of the solution was adjusted to pH 7.5 when necessary with 50 mM (NH4)OH. The precipitate was resuspended in 80% (NH 4 ) SO4 in the above extraction buffer, diluted 1:5 with additional buffer and centrifuged for an additional 20 min. The precipitate was resuspended in 10 mM Hepes, pH
7.5, 0.25 mM MgCI2, 0.25 mM DTT and 5 mM EDTA and centrifuged as above for 10 min. Proteins were desalted on a sephadex G-25 column (1.0 x 6.0 cm) with 10 mM Hepes, pH 7.5, 0.25 mM MgCl 2 and DTT. The eluent between
1 and 1.5 Ve:V 0 (elution volume:void volume) was collected and protein content determined by the method of Bradford (1976) using BSA as a standard.




59
Soluble invertase was extracted similarily except potassium
phosphate (monobasic) buffer was substituted for Hepes, sodium ascorbate and cysteine-HCl were eliminated from the extraction buffer and 5% PVPP was used. Also, the precipitate was not resuspended in 80% (NH2 SO4 Soluble acid invertase also was extracted according to Purvis and Rice (1983), and utilized in an additional study of biweekly to monthly activity. The buffer used was 100 mM Hepes, pH 8.0, 5 mM EDTA and 5% PVPP. The crude supernatant was dialysed for 20-24 hrs in 3 volumes of 10 mM phosphate buffer, pH 7.5. Cell wall material for insoluble invertase determinations was washed in 150-200 mls extraction buffer, diluted 1:40, and excess buffer was removed in a Buchner funnel.
Optima of pH were determined for sucrose synthetase in grapefruit to be 8.5 and 5.5 when assayed in the synthetic and clevage directions, respectively. Sucrose synthetase was assayed in the synthetic direction in a reaction medium containing extract, 80 mM Hepes, pH 8.5, 5 mM KCN, 5 mM NaF, 100 mM fructose and 15 mM UDPG in a total volume of 0.5 ml. Sucrose synthetase also was assayed in the cleavage direction using an incubation medium of extract, 80 mM Mes, pH 5.5, 5 mM KCN, 5 mM NaF, 10 mM sucrose and 5 mM UDP in a total volume of 0.5 ml. Reactions took place at 30 C and were terminated after 15 min by boiling for 1 min. Product degradation occurred after 2 or more min (data not shown). UDP and UDPG production was quantified according to Bergmeyer (1965).
Optima of pH were determined for acid and alkaline invertases in grapefruit to be 4.5 and 7.5, respectively. Soluble invertases were assayed in a reaction medium containing extract, 80 mM acetate-K 2PO4 (pH 4.5 or 7.5 for acid and alkaline forms, respectively) and 100 mM sucrose in a total volume of 0.5 ml. Insoluble invertases were assayed in a




60
similar manner except that 0.2-0.5 g cell wall material was used and the final reaction volume was 1.0 ml. Reactions were incubated for 15 to 30 min at 450C. Biweekly to monthly measurements of soluble acid invertase were made at 371C using 0.67 M acetate buffer (pH 4.7). Reactions were terminated by the addition of Nelson's reagent A (Nelson, 1944). Glucose was quantified by the method of Nelson (1944).
Results and Discussion
Both the albedo and juice vesicles are major sinks for photosynthates in the fruit based on dry matter accumulation during development (Fig. 3-2). Albedo is the tissue which imports the largest amount of photosynthate in early fruit development, while juice vesicles predominate during mid-stage II (5-6 months). The segment epidermis accumulated little dry matter when compared to the other tissues and therefore, did not constitute a major site of final assimilate deposition in grapefruit. However, levels of sucrose and glucose in segment epidermis were similar to those of the other tissues throughout development (Fig. 5-1, 5-2). In contrast, a disproportionate amount of 14C-photosynthates accumulated in the segment epidermis during time course studies (Koch, 1984a; unpublished data). Although the segment epidermis is not a major sink for photosynthates, this tissue may play an important role in the transport of photoassimilates into juice segments.
Both sucrose and hexoses increased in albedo, segment epidermis and juice vesicles during development (Fig. 5-1, 5-2). The ratio of sucrose to hexoses, however, generally decreased in juice tissues and albedo during grapefruit development (Hilgeman and Smith, 1940). The presence of hexoses in these tissues does not necessarily suggest enzymatic sucrose cleavage. In juice tissues, for example, the pH often is such




70
:50 IL C,1
O
a S30

* ALBEDO (ALB) SEGMENT EPIDERMIS (SEG E) o JUICE VESICLES (JV)

MAY I JUNE I JULY AUGUST SEPT I OCT NOV I
Figure 5-1. Sucrose levels in albedo, segment epidermis and juice vesicles during development of grapefruit.




120
0
0 80J S.

* ALBEDO (ALB) h SEGMENT EPIDERMIS (SEG E) o JUICE VESICLES (JV)

40+

MAY I JUNEI JULY AUGUST SEPTI OCT NOV
Figure 5-2. Glucose levels in albedo, segment epidermis and juice vesicles during development of grapefruit.

160'




that sucrose hydrolysis can occur non-enzymatically, and its rate increases with decreasing pH (Sinclair,1984). The ratio of sucrose to hexoses also decreases during early grapefruit development as the juice pH decreases from approximately 5.5 to 3.0 (Caldwell, 1934). Sucrose cleavage in these juice tissues, therefore, may result at least partially from decreasing pH during development.
Both soluble and insoluble (cell-wall bound) acid invertases
(sucrose cleavage) were active in all three tissues of the immature grapefruit during the first three months of growth. Kato and Kubota (1978) found similar results for Satsuma juice vesicles. Activities of both forms of acid invertase in grapefruit were highest in the albedo extracts, followed by those from segment epidermis and juice vesicles. These activities decreased at least 80% by the fourth month of growth (stage II) and remained low through the remainder of fruit development. However, both dry matter and glucose per unit fresh weight continued to accumulate throughout the subsequent growing period (Fig. 5-1, 5-2). Only a slight decrease in glucose level (Fig. 5-2) coincided with the earlier decrease in soluble and insoluble acid invertase activity (Fig. 5-3a, 5-4, 5-5a). If the mechanism for sucrose import into citrus fruit involves acid invertase, then a mid-season drop in activity to minimal levels of this enzyme would be expected to decrease photosynthate accumulation. The concommittant slowing of assimilate import did not occur, so either the mechanism of photosynthate accumulation changed between stage I and II, or insoluble acid invertase was not essential to assimilate import. Insoluble acid invertase activity has been associated with phloem unloading in other species (Glaziou and Gaylor, 1972; Eschrich, 1980), but its action in grapefruit tissues also could be




64
9-
>t ALBEDO (ALB)
- SEGMENT EPIDERMIS (BEG E) o0 o-- JUICE VESICLES (JV)
160
g
ALS
100
BEG E su
60
JV
IVI
MY JNE I UL ALT SEPTEMBER OCTOBER NOVEMBER
Figure 5-3. Activity of soluble acid invertase in albedo, segment epidermis and juice vesicles during development of grapefruit.




90.0
ALB
SEG E
80.0
70.0 t
10.0 JV
Z
m 5.0
ALB
0 SEG E
... ALB JV
C ... ..i iiSEG E JV. p>B
5 ALB
o 5.0
S-JJV
0
-J
O 3.0 -SEG E 2ALB
2 JV SEG E J
JV
1.0 SEG E
STAGE I STAGE II STAGE III
Figure 5-4. Activity of partially purified enzyme in individual tissues during 3 stages in development of grapefruit.
A. Soluble acid invertase activity. B. Soluble alkaline invertase activity.




SEG E

75.0
ALB
. ..~S ESEE ALBSG JV SEG E
O 3.0
w JV
w
Vn SEG E
0
_j
-J
,,J
0 2.0
1.0 ALB
ii: v
.. ALB
ALB SEG E SEGE JV
STAGE I STAGE II STAGE II
Figure 5-5. Activity of cell-wall bound enzyme in individual tissues during 3 stages in development of grapefruit. A. Insoluble acid invertase activity. B. Insoluble alkaline invertase activity.




related to processes of cell division rather than assimilate partitioning. The high activity and cell-wall localization of insoluble acid invertase (Fig. 5-3a) during the cell division phase of fruit development are consistent with the suggestion that this enzyme may be involved in providing metabolites for cell wall precursors (Altman, et al., 1982; Morris and Arthur, 1985).
A comparable situation appears to exist for soluble acid invertase. Data shown here indicate any involvement of this enzyme in the overall accumulation of photosynthates by sink tissues is either minor or occurs only during early fruit development (Fig. 5-4, 5-5a). Soluble acid invertase is believed to be compartmentalized primarily in vacuoles of immature storage parenchyma cells (Glasziou, 1962; Sacher et al.., 1963; Glasziou and Gaylor, 1972).
Soluble alkaline invertase and sucrose synthetase were the only
sucrose-metabolizing enzymes that remained active during the period when the rate of dry matter accumulation was greatest (Fig. 3-2, 5-5b, 5-6, 5-7). Soluble alkaline invertase trends appeared to be variable in the three tissues studied (5-5b). Activity of this enzyme did not change significantly during development if extracted from albedo, but a slight decrease had occurred in activity from extracts of segment epidermis by stage III. However, a significant rise in the activity of this enzyme took place when it was assayed from juice vesicles during stage II growth (Fig. 5-5b). A similar pattern was observed during development of juice vesicles in Satsuma (Kato and Kubota, 1978).
The functional significance of soluble alkaline or neutral
invertases in plant tissue are not well understood (Kato and Kubota, 1978). It has been suggested that multiple forms of alkaline invertase




I
c 5.0 ALB JV SEG E
w ALB
S2.0
CJV oT
U r ::ii :::::.
ALB
0.30
O
:-i:: ... : :!is~i~ : i~ i i :i:!:.......i ........
0.10
STAGE I STAGE II STAGE III
Figure 5-6. Activity of sucrose synthetase assayed in the degradative direction in individual tissues during 3 stages of development of grapefruit.




69
SEG E 10.00 SEG E
I
5.00 JV SEG E
0 0
...... .....
JV
0.0ALB
ALS
0.10.
STAGE I STAGE II STAGE III
Figure 5-7. Activity of sucrose synthetase assayed in the synthetic direction in individual tissues during 3 stages of development of grapefruit.




are responsible for sucrose cleavage in sweet potato (Matsushita and Uritani, 1974). A further possibility has been considered in mature sugar cane stalks where neutral soluble invertase, presumably a cytoplasmic enzyme, could couple sucrose hydrolysis to subsequent hexose phosphorylation followed by resynthesis of sucrose by sucrose phosphate synthetase (Glaziou and Gayler, 1972). Another consideration is that sucrose hydrolysis in the cytoplasm also may be involved in osmoregulation of storage cells. Cell turgor is important in the regulation of sucrose uptake and may be an additional determinant of sink strength in the sugar beet taproot (Wyse et al., 1986). Activities of soluble alkaline invertase from grapefruit are greatest when rates of accumulation are most rapid and thus, this enzyme may function in the regulation of sugar storage subsequent to phloem unloading.
Insoluble (cell-wall bound) alkaline invertase followed the same trends observed for the soluble form, although developmental change of this activity in albedo was comparable to that of juice vesicles (Fig. 5-3b). Data of Kato and Kubota (1978) suggest much of the soluble and insoluble activity may be that of the same enzyme.
Overall, activity of sucrose synthetase from grapefruit tissues
decreased during development (Fig. 5-6, 5-7). This decrease in activity did not occur concurrently with the increase in glucose or sucrose accumulation during the period of study. The direction of the sucrose synthetase reaction, depending on compartmentalization, could either resynthesize sucrose hydrolyzed by alkaline invertase or provide substrates (UDPG) for a sucrose phosphate synthetase-group translocator complex of the tonoplast (Thom et al., 1986). Activity of this enzyme was roughly similar when assayed in both cleavage and synthetic




directions, but significantly lower values were obtained for the latter with extracts from albedo and juice vesicles of young fruit (stage I or II and stage I, respectively). This disparity may suggest a problem in the product determination procedure for UDP or the presence of an inhibitor in the protein extract from young fruit that most strongly affects activity in the synthetic direction.
Relative activity of sucrose synthetase was greatest when extracted from segment epidermis and considerably less if from juice vesicles or albedo (Fig. 5-6, 5-7). Interestingly, activity of the enzyme from segment epidermis was many-fold greater that the other tissues and remained so through fruit development. Segment epidermis retained a disproportionately high percentage of 14C-photosynthates in transport studies when compared to albedo and juice vesicles (Koch, 1984a).
Activity of sucrose phosphate synthetase extracted from all three
tissues also tends to decrease during development, but increases slightly in the segment epidermis (data not shown). Overall, however, sucrose phosphate synthetase activity was minimal relative to that of sucrose synthetase in maturing and mature tissues. A near lack of sucrose phosphate synthetase and acid invertase, and the presence of alkaline invertase and sucrose synthetase activities during dry matter accumulation suggest the latter two enzymes may function in the regulation of photosynthate storage in grapefruit.
Conclusions
Five sucrose metabolizing enzymes were examined in conjunction with soluble sugar and dry weight accumulation during development of individual tissues in grapefruit. Only alkaline invertase and sucrose synthetase remained active during the period of most rapid increases in




dry matter. The extent to which sucrose is hydrolyzed and resynthesized during import into citrus fruit is unclear, but the in vitro capacity of these two enzymes is such that either could account for this process. The direction of the sucrose synthetase reaction, depending on compartmentalization, could either resynthesize sucrose hydrolyzed by alkaline invertase or provide substrate (UDPG) for a sucrose phosphate synthetase-group translocator complex of the tonoplast. Sucrose synthetase activity was highest in segment epidermis and remained relatively constant throughout development. These data, together with structural features (chapter IV) and radiolabeling studies (Koch, 1984a) suggest segment epidermis may play a major role in the transport of photosynthates into juice vesicles following phloem unloading.
Sucrose in developing grapefruit also appears to be cleaved in
processes not directly related to photosynthate import. A rapid decrease to minimal levels in the activity of soluble and insoluble acid invertase during the first 3 months of growth suggested acid invertases may be associated more closely with cell division than assimilate transfer. In addition, non-enzymatic sucrose hydrolysis can occur at low pH, and total hexose pools measured in the present study increased as juice pH decreased.




CHAPTER VI
OVERALL CONCLUSIONS
Several aspects of vascular structure and distribution appear to be associated with assimilate translocation. First, individual vascular strands in stems are directly aligned with specific fruit juice segments. Second, the phloem to xylem ratio of vascular bundles increases with distance from their point of entry into the fruit. Third, the vast majority of assimilates entering the non-vascular juice segments must be unloaded from phloem into one of 3 vascular bundles outside the segment epidermis. The position and structure of these strands changes during development. An asymmetric proliferation of primary and secondary phloem occurs in dorsal vascular bundles, and a bundle cap of lignified fibers develops in the primary phloem. Differential rates of dry weight accumulation in individual tissues are consistant with these changes in positioning and relative amount of phloem.
The longest component of the phloem-free transport path in
grapefruit is the parenchymatous juice vesicle stalk. This elongates typically up to 16 mm in mature fruit. Cells are long and narrow-lumened at the stalk periphery, but short and wide-lumened in the center. Plasmodesmatal frequencies in mature stalk cells suggest a lack of specialization for symplastic transfer. Other parenchyma nearer the vascular bundles also remain non-specialized relative to assimilate transport. The segment epidermis does not appear to be directly in the path of photosynthate transfer; however, this cutinized, uni- to 73




biseriate cell layer is continuous with that of juice vesicles and is likely to limit photosynthate passage into these structures to points of juice vesicle stalk attachment to the carpel wall. Segment epidermal cells become highly vacuolated during development and their appearance shifts from that of metabolically active cells to one more similar to storage cells.
Cells of mature juice vesicles, the final site of assimilate
storage, have low cytoplasm to vacuole ratios and show no anatomical modification for secretion.
Five sucrose metabolizing enzymes were examined in conjunction with soluble sugar and dry weight accumulation during development of individual tissues in grapefruit. Only alkaline invertase and sucrose synthetase remained active during the period of most rapid increases in dry matter. The extent to which sucrose is hydrolyzed and resynthesized during import into citrus fruit is unclear, but the in vitro capacity of these two enzymes is such that either or both could account for this process. The direction of the sucrose synthetase reaction, depending on compartmentalization, could either resynthesize sucrose hydrolyzed by alkaline invertase or provide substrates (UDPG) for a sucrose phosphate synthetase-group translocator complex of the tonoplast. Sucrose synthetase activity was highest in segment epidermis and remained relatively constant throughout development. This data, together with structural features and radiolabeling studies (Koch, 1984a) suggest segment epidermis may play a major role in the transport of photosynthates into juice vesicles following phloem unloading.
Sucrose in developing grapefruit also appears to be cleaved in processes not directly related to photosynthate import. The rapid




75
decrease to minimal levels in the activity of soluble and insoluble acid invertase during the first 3 months of growth suggests acid invertases may be associated more closely with cell division than assimilate transfer. In addition, non-enzymatic sucrose hydrolysis can occur at low pH, and total hexose pools measured in the present study increased as juice pH decreased.




APENDIX A
TECHNICAL DATA FOR SUCROSE SYNTHASE IN TISSUES OF 'MARSH' GRAPEFRUIT Introduction
Optimal conditions for assay of sucrose synthetase differ depending on the direction of the reaction (synthesis or cleavage). The optimum pH of the sucrose cleavage assay is 6.0-6.5 in a number of tissues (Hisajima, 1979; Gross and Pharr, 1982; Claussen, 1983a; Morell and Copeland, 1985). Soybean nodules have a broader range of pH 5.0-7.0 for assay of sucrose cleavage (Morell and Copeland, 1985), while in corn scutellum, greatest activity was measured from pH 6.0 to 8.5 (Echeverria, 1983).
A pH range from 7.2-9.5 generally favors measurement of sucrose synthetase in the synthetic direction (Cardini et al., 1955; Rorem et al., 1960; Avigad, 1964; Hawker et al., 1976; Hisajima, 1979; Huber, 1981; Gross and Pharr, 1982; Claussen, 1983a; Morell and Copeland, 1985). Activity was optimum from pH 8.5 to 10.0 in soybean nodules (Morell and Copeland, 1985) and pH 7.5 to 8.5 in Jerusalem artichoke tubers (Avigad, 1964). Two isozymes of sucrose synthetase were distinguished in tissues of cucumber fruit based on two optima, one at pH 7.5 and one at 9.3 (Gross and Pharr, 1982). This pH difference may have been an effect of different buffers, however, because synthetic activity in glycine-NaOH buffer was 25% higher than in tris-HCl buffer in eggplant vegetative and reproductive tissues (Claussen, 1983a)
Divalent cations, heavy metals, end products, metabolites and other proteins affect sucrose synthetase activity. Divalent cations generally




stimulate both synthesis and cleavage (Gross and Pharr, 1982; Morell and Copeland, 1985), although 8 mM Mg+2 and Mn+2 inhibited sucrose synthesis in Jerusalem artichoke tubers (Avigad, 1964). All activity was quite sensitive to inhibition by heavy metals (Morell and Copeland, 1985). Arsenate, arsenite, fluoride, iodoacetate, citrate and pyrrophosphate have no affect on sucrose synthetase activity assayed in the synthetic direction (Cardini et al., 1955). Substrate inhibition of the synthetic reaction occurrs at fructose concentrations greater than 15 mM (Morell and Copeland, 1985). Sucrose synthesis and cleavage are both inhibited by various amounts of the end products UDP and sucrose, and UDPG and fructose, respectively (Gross and Pharr, 1982; Echeverria, 1983; Morell and Copeland, 1985). Metabolites such as glucose also inhibit sucrose synthetase activity when measured in either direction. In addition, isolated protein factors from wheat seeds are reported to stimulate sucrose synthesis, but inhibit cleavage activity by changing the affinity of sucrose synthetase for UDP (Pontis and Salerno, 1982). The inhibition of sucrose cleavage suggests the existence of a regulatory mechanism controlling the activity of sucrose synthetase in vivo.
The following parameters, pH, linearity with extract amount and
time, product recovery and inhibition by phenyl-B glucoside, were tested on partially purfied extracts of sucrose synthetase to optimize methods for reaction incubation and product determination.
Materials and Methods
Materials and methods are described in chapter V. Buffers used for determinations of pH optima include 100 mM acetate and 200 mM Mes for sucrose cleavage at pH 4.0-7.0, and 200 mM Hepes and 200 mM tris for sucrose synthesis at pH 7.0-9.0. Linearity with quantity of enzyme




extract was determined for aliquots between 100 and 200 pls (approximately 14.7-29.4 pg protein). Final extract volumes were adjusted to 200 ul with column buffer (see chapter V). Linearity of partially purified enzyme also was measured over time, from 15 to 60 min. For analysis of product recovery, 100 nmoles of the respective nucleotide end product was substituted for the nucleotide substrate. Assay procedure in both directions also was conducted with the addition of 3 nmoles phenyl-B-glucoside (a sucrose synthetase inhibitor) to the reaction medium in a final volume of 0.5 ml to ascertain that sucrose synthetase activity solely was responsible for measured changes.
Results and Discussion
Low levels of phenyl-B-glucoside completely inhibited sucrose synthetase when cleavage was assayed and inhibited 61% when sucrose synthesis was measured. Sucrose synthetase, therefore, is the primary reactant. Sucrose synthetase from 'Marsh' grapefruit tissue has an optimum of pH 5.0-6.0 when assayed in the degradative direction (Fig. A-i). In this assay, Mes buffer appears to stimulate sucrose synthetase activity at pH 5.5 by 27%. The pH optimum for the assay of sucrose synthetase in the synthetic direction is less clear, but also appears to be broad, from pH 8.0 to 9.0 (Fig. A-2). Tris buffer enhanced sucrose synthesis 36 and 39% at pH 8.0 and 8.5, respectively.
Sucrose synthetase activity in both directions is linear with
quantity of extract from 100 to 200 pl (Fig. A-3). Assays are linear with time for 45 or 60 min when measured in the synthetic and degradative directions, respectively (Fig. A-4). Product recovery for sucrose synthetase cleavage and synthesis activities was 93.12% + 0.019 and 55.84% + 0.075 after 15 min incubations, respectively. A low product




ACETATE BUFFER ---e MES BUFFER
e
I
cc
z
g 0.20
0
0
9 0.10
0

6.0 7.0

Figure A-1. Effect of pH on the cleavage activity of sucrose synthetase.




TRIS BUFFER HEPES BUFFER

pp
0-c

7.0 8.0 9.0
pH

Figure A-2. Effect of pH on the synthesis activity of sucrose synthetase.




150
UL EXTRACT

Figure A-3. synthetase.

Effect of extract amount on the activity of sucrose

A. Sucrose cleavage r2= 0.990. B. Sucrose synthesis, r2= 0.943.

50.0+

30.01

10.0+

40.0
30.0 20.0
10.01

100

200




82
8.01
z 6.0
0
.0
0
2.0
"r,
- 3.0
n
0
cc
2.0
C.
a
1.0.
,.I
0
15 30 45 60
TIME (MIN)
Figure A-4. Effect of time on the activity of sucrose synthetase. A. Sucrose cleavage, r2= 0.994. B. Sucrose synthesis, r2= 0.947 up to 45 min.




83
recovery for the synthetic direction suggests some component of the enzyme preparation is using UDP as a substrate. Sucrose was included in the reaction medium, thus sucrose synthetase in the degradative direction is the probable cause. Preliminary data on sucrose synthetase cleavage at pH 8.5 in Hepes buffer suggest a low level of activity (data not shown).
Parameters of pH, linearity and product recovery suggest the methods of in vitro reaction and product determination are sufficient for a developmental study of sucrose synthetase activity in tissues of 'Marsh'

grapefruit.




APENDIX B
TECHNICAL DATA FOR INVERTASE IN TISSUES OF 'MARSH' GRAPEFRUIT Introduction
Invertase has been widely studied in a variety of lower and higher plant tissues and parameters for optimal in vivo activity have been established in these tissues. Optima of pH for soluble and insoluble invertase varies with the plant tissue in two general ranges, 2.5 to 6.8 (Gascon et al., 1965; Pressey, 1966; Sasaki et al., 1971; Klis et al., 1974; Chan et al., 1976; del Rosario et al., 1977; Kato and Kubota, 1978; Humphreys et al., 1980; Sum et al., 1980; Jacob et al., 1982; Prado et al., 1982; Howard et al., 1983; Echeverria et al., 1984) and 7.0 to 7.7 (Matsushita et al., 1974; del Rosario et al., 1977; Kato and Kubota, 1978; Prado et al., 1982; Dey, 1986). Temperature optima are physiologically high at 400C (Sum et al., 1980) and 550C (Chan et al., 1975). Invertase inhibitors include a wide range of heavy metals, metallic and non-metallic ions, sugars (glucose-6-phosphate, glucose), tris buffer, lauryl sulfate, metasilicate and specific detergents (Arnold, 1965; Metzenberg, 1963; Kato et al., 1978; del Rosario et al., 1977; Matsushita et al., 1974). Stimulators of invertase include potassium and sodium nitrates and most phosphates and thiols (Jacob et al., 1982). Inhibitors and stimulators of invertase appear to vary widely with origin and form of the enzyme.




85
Product recovery, pH optima and linearity with enzyme concentration and time were determined to optimize the invertase assay for use in a developmental study of invertase activity in 'Marsh' grapefruit tissues.
Materials and Methods
Materials, extraction and assay procedures are described in chapter V. The pH optima were determined with 300 mM acetate-potassium phosphate (dibasic) buffer in the pH range of 4.0 to 9.0. Soluble acid invertase from stage I albedo and soluble alkaline invertase from stage II segment epidermis were used to determine optimal pH between 4.0 and 9.0, and linearity with time up to 60 min in 'Marsh' grapefruit. Two replications were used for the soluble acid invertase pH curve. Duplicate analyses from stage I and III albedo were used to determine linearity with concentration of enzyme for 15 to 30 min assays of all forms of invertase. Recovery of added product to the reaction incubation was measured by adding 277.6 nmoles of glucose to assay medium with soluble acid invertase from stage I segment epidermis or soluble alkaline invertase from stage II segment epidermis. Water was substituted for sucrose in these reactions. Standard errors were determined from means of 2-3 replications.
Results and Discussion
Soluble invertases from grapefruit have specific pH optima at 4.5 and 7.5 for acid and alkaline forms, respectively (Figs. B-1, B-2). Activity of soluble acid invertase at pH 4.0 is 75% of the optimal and overlapping standard errors suggest a broad range of activity between pH 4.0 and 5.0. Similarily, soluble alkaline invertase activity at pH 7.0 and 8.0 is at least 70% of the maximum and, again, overlapping standard




140.0
I
X 120.0
I
z M.U
0
2 80.0
60.0
w
0
.)
2 20.0
4.0 5.0 6.0 7.0
pH
Figure B-1. Activity of soluble acid invertase from stage I albedo at pH
4.0 to 7.0.




1 C"
x
S5.0
10
0
1.0
w
CO 0 0
6.5 7.0 7.5 8.0 8.5 9.0
pH
Figure B-2. Activity of soluble alkaline invertase from stage II segment epidermis at pH 6.5 to 9.0.




88
errors suggest a broad range of optimal activity between pH 7.0 and 8.0. Kato and Kubota (1978) also demonstrated similar broad pH ranges for optimal activity of soluble and insoluble acid invertases (pH 4.8-5.3) and alkaline invertases (pH 7.2-7.7) in orange juice vesicles. Assays were linear with concentration of enzyme between 25 and 250 ul enzyme and between 0.01 and 0.5 g of pellet (Figs. B-3, B-4). Soluble acid invertase activity also is linear with time to 60 min (Fig. B-5a). Soluble alkaline invertase activity is linear with time between 15 and 60 min, while linearity is not immediately apparent at times less than 15 min (Fig. B-5b). Linearity from zero time up to 60 min, however, is within the limits of the large standard errors obtained for this procedure. Product recovery for soluble acid and alkaline forms was 94% and 95%, respectively.
Activities measured for soluble and insoluble, acid and alkaline forms of invertase in grapefruit tissues demonstrate linearity with enzyme amount and time at pH 4.5 and 7.5 and the conditions given for the reactions. Product recovery also is approximately 95% with the given conditions.




a ALBEDO, STAGE I (pH 4.5)
* ALBEDO, STAGE I (pH 7.5)
* ALBEDO, STAGE III(pH 7.5)

/

150

250

UL ENZYME EXTRACT
Figure B-3. Glucose equivalents produced during the soluble acid invertase assay at different concentrations of partially purified enzyme extract.

1300+

1000
500,
j
O
O
w
t,
0
_ 100.0
0
C

50.0 1

20.01




1.50+

1.00
O ILl V)
0
0
-J
0 0.50
0.10

* ALBEDO, STAGE I (pH 7.5) o ALBEDO, STAGE I, (pH 5.0)

0.01

0.05 G PELLET

Figure B-4. Glucose equivalents produced during the insoluble invertase assay at different amounts of pellet.

i i 'I




100.0
A
50.0
0
0
2 10.0
CO
0
o 5.0
-1
o 4.0
3.0
2.0
1.0
15 30 45 60
TIME (MIN)
Figure B-5. Glucose equivalents produced by partially purified enzyme extract at different time intervals. A. Soluble acid invertase from stage I segment epidermis at pH 4.5. r 2=.999 B. Soluble alkaline invertase from stage II segment epidermis at pH 7.5. r = .983.




LITERATURE CITED

ALBRIGO LG 1972 Distribution of stomata and epicuticular wax on oranges
as related to stem end rind breakdown and water loss. J Amer Soc
Hort Sci 97:220-223
ALTMAN A, Y GULSEN, R GOREN 1982 Growth and metabolic activity of lemon
juice vesicle explants in vitro. Plant Physiol 69:1-6
ANELUNXEN F, H ARBEITER 1967 Untersuchungen an den Spritzdrusen von
Dictamnus albus L. Z Pflanzenphysiol 58:49-69
ARNOLD WN 1965 B-fructo furanosidase from grape berries. Biochim Biophys
Acta 110:134-147
ARISZ WH 1969 Intercellular polar transport and the role of the
plasmodesmata in coleoptiles and Vallisneria leaves. Acta Bot Neerl
18:14-38
AVIGAD G 1964 Sucrose-uridine diphosphate glucosyltransferase from
Jerusalem artichoke tubers. J Biol Chem 239:3613-3618
BAIN JM 1958 Morphology, anatomical and physiological changes in the
developing fruit of the Valencia orange, Citrus sinensis (L.)
Osbeck. Aust J Bot 6:1-28
BANERJI I 1954 Morphological and cytological studies on Citrus grandis
Osbeck. Phytomorphology 4:390-396
BARTHOLOMEW ET, WB SINCLAIR 1941 Unequal distribution of soluble solids
in the pulp of citrus fruits. Plant Physiol 16:293-312
BARTHOLOMEW ET, WB SINCLAIR 1951 The lemon fruit. University of
California Press, Berkeley
BEAN RC 1960 Carbohydrate metabolism of citrus fruits. I. Mechanisms of
sucrose synthesis in oranges and lemons. Plant Physiol 35:429-434
BERGMEYER H 1965 In H BERGMEYER ed Methods of enzymatic analysis Academic
Press, New York
BIRDSALL JJ, PH DERSE, LJ TEPLY 1961 Nutrients on California lemons and
oranges. II. Vitamin, mineral, and proximate composition. J Amer
Dietet Assoc 38:555-559
BOLENDER RP 1978 Correlation of morphometry and stereology with
biochemical analysis of cell fractions. Int Rev Cytol 55:247-289




BOSABALIDIS A, I TSEKOS 1982a Ultrastructural studies on the secretory
cavities of Citrus-Deliciosa Ten. 1. Early stages of the gland cells
differentiation. Protoplasma 112:55-62
BOSABALIDIS A, I TSEKOS 1982b Ultrastructure studies on the secretory
cavities of Citrus-Deliciosa Ten. 2. Development of the essential oil-accumulating central space of the gland and process of active
secretion 112:63-70
BOUMA D 1959 The development of the fruit of the "Washington Navel"
orange. Aust J Agr Res 10:804-817
BRADFORD MM 1976 A rapid and sensitive method for quantification of
microgram quantities of protein utilizing the principle of protein
dye-binding. Anal Biochem 72:248-254
BROWN GE, CR Barmore 1981 Ultrastructure of the response of citrus
epicarp to mechanical injury. Bot Gaz 142:477-481
CAMERON SH 1933 Starch in the young orange tree. Proc Amer Soc Hort Sci
29:110-114
CARDINI CE, LF LELOIR, J CHIRIBOGA 1955 Biosynthesis of sucrose. J Biol
Chem 214:149-156
CHAN HT, SCM KWOL, CWQ LEE 1975 Sugar composition and invertase activity
in Lychee. J Food Sci 40::772-774
CHATTERTON NJ 1973 Product inhibition of photosynthesis in alfalfa leaves
as related to specific leaf weight. Crop Sci 1:284-285
CLAUSSEN W 1983a Untersuchungen uber den Zusammenhang zwischen der
Verteilung der Assimilate und der Saccharose-Synthetase-Aktivitat in
Solanum melongena L. 1. Charakterisierung und Verteilung der
Saccharose-Synthetase. Z Pflanzenphysiol 110:165-173
CLAUSSEN W 1983b Untersuchengen uber den Zusammenhang zwuschen der
Verteilung der Assimilate und der Saccharose-Synthetase-Aktivitat in
Solanum melongena L. 2. Assimilatverteilung und
Saccharose-Synthetase-Aktivitat. Z Pflanzenphysiol 110:175-182
CLAUSSEN W, BR LOVEYS, JS HAWKER 1985 Comparative investigations on the
distribution of sucrose synthase activity and invertase activity within growing, mature and old leaves of some C-3 and C-4 plant
species. Physiol Plant 65:275-280
DEL ROSARIO EJ, V SANTISOPASRI 1977 Characterization and inhibition of
invertases in sugar cane juice. Phytochem 16:443-445
DEY PM 1986 Changes in the forms of invertase during germination of mung
bean seeds. Phytochem 25:51-53
DICK PS, T AP REES 1976 Sucrose metabolism by roots of Pisum sativum.
Phytochem 15:255-259




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EIUXPQFBG_GQHW1N INGEST_TIME 2017-07-11T21:09:45Z PACKAGE UF00101055_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

STRUCTURE AND SUCROSE METABOLIZING ENZYMES OF THE TRANSPORT PATH: IMPLICATIONS FOR ASSIMILATE TRANSLOCATION IN GRAPEFRUIT By CADANCE ANN LOWELL A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1986

PAGE 2

ACKNOWLEDGMENTS I wish to express my deepest appreciation to Dr. Karen E. Koch for her guidance and helpful criticisms during this study. I also wish to thank the other members of my committee, Dr. James Soule, Dr. Thomas Humphreys, Dr. Hilton Biggs and Dr. James Kimbrough, for their assistance and helpful suggestions concerning both the interpretation of the data and this manuscript. I am grateful to Patricia Tolson-Tomlinson whose guidance allowed me to finally complete the enzymatic portion of this research and to Wayne Avigne and Andrew K. McCullers for their technical aid. I also wish to thank Dr. Kenneth Curry and Debra Akin for their help with the morphometric analysis in this manuscript. Finally, my deepest thanks go to my husband, James William Williams for his time and expertise in drawing the figures in the text, and his support for this endevor, even when he did not understand it.

PAGE 3

TABLE OF CONTENTS page ACKNOWLEDGMENTS ii LIST OF ABBREVIATIONS v ABSTRACT vi CHAPTERS I INTRODUCTION 1 II LITERATURE REVIEW 4 Anatomy of Citrus Fruit 4 Assimilate Partitioning in Citrus 9 Photosynthate Translocation and Partitioning in Sink Tissues 10 Photosynthate Translocation and Partitioning in Citrus Fruit Partitioning 11 Sucrose Synthetase, Invertase and Assimilate Partitioning 12 III THE VASCULAR SUPPLY AND ASSIMILATE TRANSPORT IN GRAPEFRUIT 16 Introduction 16 Materials and Methods 18 Results and Discussion 20 Conclusions 32 IV NON-VASCULAR TISSUES IN GRAPEFRUIT DURING DEVELOPMENT 34 Introduction 34 Materials and Methods 36 Results and Discussion 38 Conclusions 54 V SUCROSE METABOLIZING ENZYMES AND ASSIMILATE PARTITIONING IN SINK TISSUES OF DEVELOPING GRAPEFRUIT 55 Introduction 55

PAGE 4

Materials and Methods ->7 Results and Discussion 60 Conclusions '* VI OVERALL CONCLUSIONS 73 APPENDICES A TECHNICAL DATA FOR SUCROSE SYNTHETASE IN TISSUES OF 'MARSH' GRAPEFRUIT 76 B TECHNICAL DATA FOR INVERTASE IN TISSUES OF 'MARSH' GRAPEFRUIT 84 LITERATURE CITED 92 BIOGRAPHICAL SKETCH 104

PAGE 5

ALB DTT EDTA Hepes JV Mes mM ml ul Pipes PVPP Seg E Tris UDP UDPG LIST OF ABBREVIATIONS albedo (outer and inner mesocarp) dithiothreitol ethylenediaminetetraacetic acid N-2-Hydroxyethylpiperaxine-N' -2-ethanesulf onic acid juice vesicles 2(N-morpholino) ethane sulfonic acid millimolar milliliter microliter 1,4-piperazinediethanesulfonic acid polyvinyl-poly pyrrolidine (insoluble) segment epidermis tris (hydroxy methyl ) am inome thane uridine 5-diphosphate uridine 5-diphosphoglucose

PAGE 6

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy STRUCTURE AND SUCROSE METABOLIZING ENZYMES OF THE TRANSPORT PATH: IMPLICATIONS FOR ASSIMILATE TRANSLOCATION IN GRAPEFRUIT By Cadance Ann Lowell Chairman: Dr. Karen E. Koch Major Department: Horticultural Sciences (Fruit Crops) Structure of the transport path, activities of sucrose metabolizing enzymes, accumulation of dry weight and levels of soluble sugars were compared during development of 'Marsh' grapefruit (C itru s para disi Macf.) to elucidate features related to mechanisms of photosyn thate import. Aspects of vascular anatomy which appear to be associated with assimilate translocation include 1) direct alignment of individual vascular strands within stems and specific fruit juice segments, 2) an increase in the phloem :xylem ratio of vascular bundles with distance from their point of entry into fruit and 3) developmental changes in vascular bundles supplying photosynthates to juice tissues. In the latter, an asymmetric proliferation of primary and secondary phloen and formation of lignified bundle caps coincide with a shift in rate of dry weight accumulation by adjacent tissues.

PAGE 7

Assimilates enroute to juice vesicles must move through extensive non-vascular portions of the translocation path after phloem unloading. No structural specializations for either apoplastic or symplastic transport were evident in parenchyma surrounding the vascular bundles or in the phloem-free, elongated vesicle stalks. Frequiencies of plasmodesmata mature juice vesicle stalks ranged from 0.24 to 0.38 urn (1.96 to 3.78 urn ). Photosynthate transport along other paths would probably be restricted by the combination of cellular discontinuity and by cutinization of the epidermis surrounding both segment and individual juice vesicles. Cells of this continuous epidermal layer appear to be metabolically active during early fruit growth, but later become highly vacuolated. Cells of mature juice vesicles, the final site of assimilate storage, have low cytoplasm to vacuole ratios and show no anatomical modification for secretion. Of the 5 sucrose metabolizing enzymes examined during grapefruit development, only alkaline invertase and sucrose synthetase remained active during the period of most rapid gains in dry weight. The extent to which sucrose is hydrolyzed and resynthesized during import into citrus fruit is unclear, but the in vitro capacity of these two enzymes is such that either or both could account for this process. A rapid decrease to minimal levels in the activity of soluble and insoluble acid invertase during the first 3 months of growth suggest acid invertases may be associated more closely with cell division than assimilate transfer. In addition, non-enzymatic sucrose hydrolysis can occur at low pH, and total hexose pools measured in the present study increased as juice pH decreased

PAGE 8

CHAPTER I INTRODUCTION Total soluble solids and the total soluble solids: total (titratable) acid ratio are major factors in the harvest and market value of Citrus Despite importance of the timing and quantity of assimilates entering citrus fruit, this process and its control are little understood. Crop productivity is influenced by rate of translocation (Chatterton, 1973; Liu et al., 1973), carbohydrate content of source leaves (Habeshaw, 1973; Geiger, 1976) and/or sink demand (Geiger and Fondy, 1980; Ho and Baker, 1982). Assimilate transfer from phloem to storage tissues is associated with the spatial arrangement of vascular tissue relative to sink cells and the effects of sugar gradients, pH and enzymes in this zone (Jenner, 1974; Eschrich, 1980; Felker and Shannon, 1980). The spatial arrangement of vascular and non-vascular tissues in Citrus make this fruit a unique system with which to observe sugar movement into a sink. Primary vascular tissue in the citrus fruit consists of dorsal, septal and central carpellary bundles (traces) (Roth, 1977). A fine network of minor traces extends from dorsal and septal bundles and terminates near peripheral oil glands in the flavedo. This vascular tissue persists through fruit maturation (Ford, 1942). Citrus fruit lacks vascular connections, however, between the albedo and juice vesicles (Ford, 1942). The presence of this non-vascular zone between

PAGE 9

vascular bundles and juice vesicles suggests that assimilates do not enter juice vesicles directly through phloem translocation. Subsequent to phloem unloading, photosynthates must pass through the inner albedo (mesocarp) and juice vesicle stalks before entering the juice vesicle heads. Comparable movement of assimilates into sink structures of other species has been hypothesized to be symplastic, apoplastic or both (Jenner, 1980). Sucrose presumably can be unloaded from phloem via apoplastic and symplastic pathways. In both instances, the process is believed to occur along a decreasing sucrose gradient. This gradient in symplastic systems could be developed either by metabolic conversions of sucrose in sink tissues (Geiger and Fondy, 1980) or by compartmentalization of sucrose into vacuoles (Fisher and Outlaw, 1979). Where photosynthates move into the apoplast, sucrose unloading is believed to involve one-way movement at a rate dependent on the hydrolytic action of free-space acid invertase (cell wall bound) and soluble neutral invertase (cytoplasmic) (Hawker and Hatch, 1965; Glasziou and Gayler, 1972; Eschrich, 1980). Direct unloading of sucrose into the apoplast and subsequent storage in the symplast of sink cells can occur without invertase activity, as in sugar beet (Giaquinta, 1979; Wyse, 1979). The cytoplasmic enzymes sucrose phosphate synthetase and sucrose synthetase also are associated with sucrose storage in sink tissues (Glasziou and Gayler, 1972; Giaquinta, 1979). Sucrose phosphate synthetase catalyzes sucrose synthesis, but sucrose synthetase readily catalyzes both the cleavage and synthesis of sucrose (Hawker, 1985). The activity ratio of sucrose synthetase to sucrose phosphate synthetase typically is high in carbon importers

PAGE 10

(Giaquinta, 1979). Reversibility of sucrose synthetase activity possibly is a key to regulation of assimilate storage in sink tissues (Giaquinta, 1979). Little data are available to support or refute these hypotheses in Citrus. The major storage carbohydrate in citrus juice vesicles is sucrose (Kefford and Chandler, 1970), but the presence of increased levels of glucose and fructose in grapefruit, particularly in the latter part of development, suggests sucrose metabolism could be involved in photosynthate storage. Sucrose synthetase and sucrose phosphate synthetase activity was determined for immature and mature 'Eureka' lemon and mature 'Valencia' orange fruit tissues (Bean, 1960). Activity of sucrose synthetase varied with development and tissue, while no sucrose phosphate synthetase activity was detected. Invertase activity has been measured in grapefruit flavedo in conjunction with freezing (Purvis and Rice, 1983) and during development in juice vesicles (Kato and Kubota, 1978). The presence of sucrose synthetase and invertase activity in Citrus may be associated with phloem unloading and storage of sugars in juice vesicles. The objective of this research was to elucidate features of the anatomy and sucrose metabolizing system involved in the transport pathway and storage of sugars during development of 'Marsh' grapefruit (Citrus paradisi Macf.). It is envisioned that this research will enhance current knowledge of sugar transport and storage in sink tissues and may be valuable to the search for regulation of timing and quantity of sugar entry into grapefruit and possibly other citrus fruits.

PAGE 11

CHAPTER II LITERATURE REVIEW Anatomy of Citrus Fruit Current knowledge of the anatomy of citrus fruit has been detailed in several overviews (Schneider, 1968; Roth, 1977) and in developmental studies of specific cultivars (Ford, 1942; Bain, 1958; Holtzhausen, 1969). By far, the majority of citrus anatomical and morphological research at the light and ultrastructural level have dealt with the outer fruit peel (flavedo). These have included studies of its development (Scott and Baker, 1947), stomata (Klotz and Haas, 1933; Turrell and Klotz, 1940; Scott and Baker, 1947), wax production (Scott and Baker, 1947; Schulman and Monselise, 1970; Albrigo, 1972; Freeman et al. 1979), chloroplast and chromoplast interconversions (Thomson, 1965) and oil glands (Thomson, 1966a; 1966b; 1969; Brown and Barmore, 1981; Bosabolidis and Tsekos, 1982a; 1982b). Anatomical research on juice tissues primarily has addressed secondary fruit development (Lima, 1983), wax production (Ford, 1942; Dodd, 1944; King, 1947; Fahn et al., 1974; Espelie et al., 1980; Shomer et al., 1980), the presence of vascular tissue (Kordan, 1964) and the origin and ontogeny of juice vesicles or glandular hairs (Davis, 1932; Ford, 1942; Banerji, 1954; Amelunxen and Arbeiter, 1967; Roth and Lindorf, 1972; Roth, 1977). Discrepancies arise in the literature with respect to vascular tissue distribution, cellular composition and origin of tissues within the citrus fruit. Few published studies have related the physiology of citrus fruit growth patterns to anatomy (Bain, 1958; Bouma, 1959; Holtzhausen, 1969; Lima, 4

PAGE 12

5 1983). Physiological aspects of sucrose metabolizing enzymes and dry matter accumulation, and the distribution of vascular and non-vascular tissues in this fruit have not been addressed in the literature. Flowers of the Rutaceae, subfamily Aurantoideae typically are perfect (unisexual by abortion), actinomorphic and 3 to 5 merous (Tillson and Bamford, 1938). The pistil of the flower arises as 9 to 10 connate carpels in 'Eureka' lemon (Ford, 1942) and 8 to 12 carpels in 'Valencia' orange (Bain, 1958). Ovules generally are formed in two rows in each locule on the marginal central placenta (Roth, 1977). The vascular supply within citrus flowers persists in the fruit (Ford, 1942; Roth, 1977) and has been studied extensively in lemon flowers and immature fruit (Ford, 1942). Dorsal, septal and central vascular traces supplying the pistil in 'Eureka' lemon i;Citrus limon (L.) Burm.] diverge almost simultaneously from the stele of the receptacle (Ford, 1942). Each carpel is associated with one dorsal, two septal and two central traces from which smaller traces diverge. The dorsal vascular trace occurs medianly along the tangential surface of a carpel, extends into the style and terminates below the stigma. Septal traces curve inward along septa separating the carpels and merge with central traces. At this point the primary phloem and xylem of these septal traces are inverted. Central traces supply the ovules and extend through the central axis of the ovary (Ford, 1942). At the stylar end of the ovary, these traces curve outward and also form inverted bundles. Fusion of bundles forms amphicribral bundles (phloem completely surrounding xylem) (Ford, 1942). Central bundles may extend into the style and terminate below the stigma (Ford, 1942; Tillson and Bamford, 1938).

PAGE 13

6 The anastomose network of vascular tissue in the citrus fruit is confined to the niesocarp or albedo (Ford, 1942; Scott and Baker, 1947; Bartholomew and Sinclair, 1951; Schneider, 1968). Primary xylem of traces associated with carpels includes vessel members and tracheids with spiral thickenings, tracheids with reticulate thickenings and xylem parenchyma cells (Ford, 1942; Scott and Baker, 1947). As the immature 'Eureka' lemon fruit undergoes rapid expansion, dorsal, septal and central bundles may develop cambia and produce secondary phloem and xylem elements (Ford, 1942). Cross sectional areas of the dorsal bundles also may increase in the 'Washington' navel orange during the growth season (Holtzhausen, 1969). Xylem elements have been noted in juice tissues in lemon (Kordan, 1964), although these elements occur individually or in small clusters in relatively few juice vesicle stalks and do not connect with vascular traces in the rest of the fruit (Kordan, 1964). The citrus fruit is a hesperidium that consists of exocarp (flavedo), mesocarp (albedo, intersegmental membranes, central axis) and endocarp (segments, juice vesicles) (Bain, 1958). The pigmented flavedo occurs internal to a uniseriate epidermis and an unito triseriate hypodermis (Scott and Baker, 1947). Flavedo is composed of polygonal parenchyma cells that contain plastids. Oil glands and vascular trace endings also occur within this tissue (Ford, 1942; Scott and Baker, 1947). Intercellular spaces increase in number near the outer albedo. The abundant plastids usually convert from chloroplasts to chromoplasts as the fruit matures depending on environmental conditions. Leucoplasts and starch grains also occur in the flavedo (Scott and Baker, 1947). These starch grains are confined to chloroplasts and chromoplasts (Thomson, 1965). Fewer plastids are present in cells of the inner

PAGE 14

flavedo (Roth, 1977). Although oil glands include derivatives of the epidermal and subepidermal tissue, the bulk of these glands are located in the flavedo (Bosabolidis and Tsekos, 1982a; Scott and Baker, 1947). Albedo is the white, aerenchymatous tissue interior to the flavedo that contains the majority of the vascular supply for the citrus fruit (Ford, 1942; Scott and Baker, 1947). This tissue consists of thin-walled, irregularly-shaped parenchyma cells which typically have eight lobes. Intercellular spaces are large and numerous (Ford, 1942; Scott and Baker, 1947). Interfaces between adjacent parenchyma cells possess primary pit fields and plasmodesmata (Scott and Baker, 1947). The albedo is segregated by number of intercellular spaces and parenchyma cell lobes into a regular, compact outer layer, an intermediate zone and a loosely organized inner portion (Ford, 1942; Scott and Baker, 1947). Pectin increases in albedo cells of 'Valencia' orange during fruit development and results in the appearance of a dense albedo in edible fruit (Bain, 1958). Starch grains are prominent in parenchyma cells of 'Valencia' orange albedo (Bain, 1958). In pummelo, chloroplasts are present in this tissue early in fruit development, while later in the season these plastids differentiate into chromoplasts (Gross et al., 1983). Mitochondria and nuclei are common in albedo cells, while endoplasmic reticulum, dictyosomes and extracellular material are rare (Brown and Barmore, 1981). The edible portion of the citrus fruit is the endocarp. Each carpel or juice segment is a locule surrounded by the carpellary wall or segment epidermis and filled with juice vesicles and seeds (Schneider, 1968). The segment epidermis consists of a uniseriate parenchymatous layer lacking stomata and covered with a thin cuticle (Roth, 1977). These

PAGE 15

8 epidermal cells are thick-walled and form a "net-like pattern" (Roth, 1977). The segment epidermis of two adjoining segments is separated by compact parenchyma cells, and these three layers comprise the interlocule septae. Intercellular spaces form in the parenchyma cells at the center of these septae during later fruit development (Ford, 1942). Juice tissues are multicellular vesicles derived from the carpellary wall (segment epidermis and subepidermal cell layers) (Roth and Lindorf, 1972). These juice vesicles are attached to the segment epidermis and are associated with vascular bundles, although the vascular network has not been reported to enter juice tissues (Dodd, 1944). Juice vesicles are initiated in locules of young orange ovaries 3 mm in diameter or less (Roth and Lindorf, 1972). Each juice vesicle has a "club-shaped" sac or head with a multicellular stalk at the enlarged end (Turrell and Bartholomew, 1939; Roth, 1977). The juice vesicle head has an external layer of elongated epidermal cells that enclose large, thin-walled "juice cells" (Fahn, 1979). The epidermal cells are elongated parallel to the longitudinal axis of the sac, thick-walled and covered with a cuticle of cutin, epicuticular wax and suberin (Ford, 1942; Dodd, 1944; King, 1947; Fahn et al. 1974; Roth, 1977). Parenchyma cells within the juice vesicle are composed of compact peripheral cells and loosely-organized central cells. Overall, these cells have been separated into four zones based on shape, size and orientation in immature grapefruit (Dodd, 1944; Fahn, 1979). These parenchymatous layers, beginning with the vesicle exterior, are (1) hypodermis, (2) concentrically-arranged flattened cells, (3) large, isodiametric cells and (4) centrally-located, thin-walled cells with lipophilic droplets. The central, highly vacuolated cells are similar in many respects to those of "spouting

PAGE 16

9 glands" in Dictamnus albus (Rutaceae) Cells in spouting glands disrupt at maturity when cavities form within the mature juice sac (Davis, 1932; Banerji, 1954; Ameluxen and Arbeiter, 1967). These cavities in 'Eureka' lemon juice vesicles result from high turgor pressure and are not comparable to gland cells (Ford, 1942). Nuclei and variously-shaped plastids are present in the innermost cells of most citrus juice vesicles (Turrell and Bartholomew, 1939; Dodd, 1944; Scott and Baker, 1947) and these plastids may contain starch in immature citrus fruit (King, 1947). Assimilate Partitioning in Citrus Major reviews of citrus carbohydrate literature have been authored by Sinclair (1972; 1984) and Kefford and Chandler (1970). Carbohydrates in citrus have been studied in relation to tree and fruit cold hardening (Cameron, 1933; Sharpies and Burkhart, 1953; Guy et al. 1981; Purvis and Rice, 1983), alternate bearing (Schaffer et al., 1985), flower and fruit set (Hilgeman et al., 1966; Powell and Krezdorn, 1977; Guardiola et al., 1984; Sinclair, 1984; Goldschmidt et al., 1985), juice quality (Widdowson and McCance, 1935; Bartholomew and Sinclair, 1941; Roy, 1945; Birdsall et al., 1961; Ting, 1969; Stepak and Lifshitz, 1971; Syvertsen and Albrigo, 1980), and most recently, assimilate translocation and partitioning within the fruit (Koch, 1984a; 1984b; 1985; Koch and Avigne, 1984). Sucrose is the major storage sugar in oranges and grapefruit, while reducing sugars predominate in lemon fruit (Sinclair, 1984). In addition to sucrose, fructose and glucose, sugars in citrus fruit include rhamnose and xylose (Stepak and Lifshitz, 1971). The ratio of sucrose to hexoses in grapefruit generally decreases in juice tissues and albedo during development (Harvey and Rygg, 1936; Hilgeman and Smith, 1940), and hexoses will sometimes predominate in mature peel (Harvey and Rygg,

PAGE 17

10 1936). At a given stage of development, sugar composition is reported to be relatively constant throughout in the peel or in immature juice vesicles (Harvey and Rygg, 1936; Bartholomew and Sinclair, 1941). Edible oranges and grapefruit, however, show an increasing gradient of total soluble solids (primarily sugars and organic acids) from their stem to stylar ends, and from central core to periphery of juice tissues (Bartholomew and Sinclair, 1941; Ting, 1969). Ratios of sucrose to hexoses decrease from the ends and periphery towards the central core (Ting, 1969). Existence of sugar gradients within citrus juice tissues may suggest variability in ripening and the capability to accumulate assimilates (Sinclair, 1984). Starch has been reported in citrus plants and immature fruit (Cameron, 1933; Kordan, 1971a; 1971b; Yelenosky and Guy, 1977), but only in very low percentages. Conversion of photosynthates to starch in Citrus is likely to be only a minor component of early storage. Photosynthate Translocation and Partitioning in Sink Tissues Non-vascular tissues and apoplastic barriers in sink tissues can separate sites of phloem unloading from assimilate storage. The presence of high plasmodesmatal densities typically suggest a symplastic transport path for photosynthates subsequent to phloem unloading (Helder and Boerma, 1969; Eleftheriou and Hall, 1983; Hayes et al. 1985; Gunning and Hughes, 1976; Warmbrodt, 1985). In contrast, apoplastic transfer is characteristic in sink tissues that include discontinuous cell layers (Felker and Shannon, 1980; Thorne, 1979), transfer cells (Felker and Shannon, 1980) and suberized or cutinized barriers (Eleftheriou and Hall, 1983; Gunning and Hughes, 1979). Developmental changes in vascularization (Hardham, 1976), plasmatubules (Harris et al., 1982) and

PAGE 18

11 symplastic connections (Gunning, 1978; Seagull, 1983; Juniper, 1977; Schnepf and Sych, 1983) also may affect the translocation and storage of assimilates subsequent to phloem unloading. Autoradiographic and 14C-photosynthate studies in soybean (Thorne, 1980) and corn (Orr et al., 1981a; 1981b) have demonstrated functional significance for cell layers along the transport path subsequent to phloem unloading. In soybean fruit, assimilates initially must pass into and throughout the seed coat before entering the extracellular space surrounding the embryo (Thorne, 1980). Photosynthates likewise must enter the intermediary pedicel region of corn kernels before final deposition (Orr et al., 1981a; 1981b). Photosynthate Translocation and Partitioning in Citrus Fruit Recent research of the translocation of 14C-photosynthates into grapefruit show that photosynthate transport from a source leaf to a fruit is restricted to juice segments aligned directly with the leaf (Koch, 1984a; Koch and Avigne, 1984). The predominant site of phloem unloading of assimilates to juice tissues is the dorsal vascular bundle with minor contributions from septal bundles. Central bundles primarily supply developing seeds (Koch, 1984b). Time course studies of labeled assimilate transfer into juice tissues have determined that transport slows dramatically subsequent to phloem unloading. Movement through the segment epidermis and juice vesicle stalks is extremely slow (Koch, 1984b; unpublished data). The ratio of labelled sucrose to hexoses decreases with distance from the site of phloem unloading (Koch, 1984b).

PAGE 19

12 Sucrose Synthetase, Invertase and Assimilate Partitioning Sucrose synthetase (UDPG:D-f ructose 2-a-D-glucosyltransf erase, EC 2.4.1.13) catalyzes the readily reversible synthesis and cleavage of sucrose: Fructose + UDPG < > Sucrose + UDP The molecular weight of this enzyme varies from 350 to 540 KD (Graham and Johnson, 1978; Hisajima, 1979; Gross and Pharr, 1982; Morell and Copeland, 1985). A tetrameric structure with subunits of 90 KD has been proposed (Graham and Johnson, 1978; Morell and Copeland, 1985). This enzyme typically is studied in a soluble form, suggesting a cytoplasmic location. Both membrane-bound and insoluble forms are known to occur, however (Graham and Johnson, 1978; Bean, 1960). Invertase (B D-f ructof uranosidase, EC 3.21.26) is specific for fructofuranose moieties and acts by cleaving the glycosidic linkage between the bridge oxygen and the fructose residue (Sum et al. 19S0): Glucose + Fructose < Sucrose + HO This enzyme typically is considered a glycoprotein (Sum et al., 19SC) ana in yeast and Neurospora either lacks carbohydrate or contains at least 50% carbohydrate (Metzenberg, 1963; Gascon et al., 1968; Holbein et al., 1976). Up to five different forms of invertase have been found in higher plants (Sasaki et al., 1971), but the majority of plant tissues studied possess one or usually two forms. Acid and either neutral or alkaline forms of invertase are most common. Both can be soluble (cytoplasmic and/or vacuolar) or insoluble (cell wall or membrane bound). The *~ ^iffprent tissues varies from 15 to 380 molecular weight of invertase in different

PAGE 20

13 KD (del Rosario and Santisopasri 1977; Sum et al. 1980; Howard and Witham, 1983). The acid form generally is either in the range of 57 to 65 KD as in radish cotyledons (Howard and Witham, 1983) or between 200 to 380 KD as in banana (Sum et al., 1980) and sugar cane (del Rosario and Santisopasri, 1977). The molecular weight of neutral, soluble invertase forms in sugar cane were 125 and 160 KD with monomers of 66, 35 and 15 KD (del Rosario and Santisopasri, 1977). Sucrose synthetase is considered to catalyze sucrose breakdown in vivo to provide UDPG for synthesis of non-starch reserve polysaccharides, starch, glycoproteins and glycolipids (Hawker, 1985). Invertase, on the other hand, has been studied in relation to cell expansion growth (Altman et al., 1982; Morris and Arthur, 1985), cold stress (Purvis and Rice, 1983), cold storage (Sasaki et al., 1971) and wounding (Matsushita and Uritani, 1974). Both of these enzymes also have been associated with photoassimilate accumulation in sink tissues (Glaziou and Gaylor, 1972; Dick and ap Rees, 1976; Giaquinta, 1979; Silvius and Snyder, 1979; Eschrich, 1980; Gross and Pharr, 1982; Claussen, 1983b). Two hypotheses have been advanced for their function in sugar-storing sinks, based on sugarcane stalks (Glaziou and Gaylor, 1972) and sugarbeet taproot (Giaquinta, 1979). Both suggest apoplastic phloem unloading and subsequent storage. In sugarcane, a free-space acid invertase is proposed to cleave sucrose unloaded from phloem. Glucose and fructose then are transported into the symplast of storage parenchyma cells and phosphorylated. The resulting phosphorylated fructose presumably is converted to sucrose phosphate by sucrose phosphate synthetase. Sucrose phosphate synthetase typically is considered a cytoplasmic enzyme (Glaziou and Gaylor, 1972); however, recent reports on sugar transport

PAGE 21

14 into vacuoles isolated from red beet suggest this enzyme may be part of a "group translocator" located on the tonoplast (Thorn et al. 1986). Sucrose subsequently is stored in the vacuole where a soluble acid invertase again hydrolyzes it. A soluble alkaline invertase believed to be located in the cytoplasm also has the capacity to cleave sucrose (Glasziou and Gayler, 1972). Activity of an intracellular (vacuolar) invertase in immature sugarcane stalks disappears during a concomitant increase in alkaline invertase activity. Soluble alkaline invertase may regulate sucrose storage and utilization (Glasziou and Gayler, 1972). A somewhat different scenario has been proposed for import and storage of sugars in sugar beet taproot. In this plant, sucrose is believed to be translocated and accumulated intact, so substantial sucrose phosphate synthetase activity is unnecessary. The onset of sucrose storage in sugar beet taproot coincides with loss of soluble acid invertase activity and the appearance and increase of sucrose synthetase activity. Sucrose synthetase then, is believed to play a role in regulating the partitioning of assimilate storage and utilization in this system. The presence of sucrose synthetase also has been positively correlated with growth rate and dry matter accumulation in eggplant fruit (Claussen, 1983b). Activities of insoluble acid invertase, soluble alkaline invertase and sucrose phosphate synthetase were all low throughout development of the sugar beet taproot (Giaquinta, 1979). Activity of sucrose synthetase has been demonstrated in young and edible 'Eureka' lemon and edible 'Valencia' orange [ Citrus sinensis (L.) Osbeck] fruits (Bean, 1960). The soluble portion of orange and the particulate portion of lemon juice vesicles have shown this activity, but it is lacking in the soluble fraction of lemon and the particulate

PAGE 22

15 portion of orange juice vesicles. No sucrose phosphate synthetase activity was detected in juice vesicles, although patterns of radiolabeling of sucrose from 14C-fructose in tissue slices suggested sucrose formation was from a phosphory lated form of fructose (Bean, 1960). Incorporation of radiolabeled glucose and fructose into sucrose in fresh tissue slices from various aged flavedo, albedo and juice vesicles and intact young fruit suggest that the albedo is the most active site of sucrose synthesis in lemons, although each tissue makes some contribution to sucrose synthesis. Lemon albedo tissue maintained a relatively constant rate of glucose conversion into sucrose throughout fruit growth, while the incorporation rate in flavedo and juice vesicles declined rapidly with age (Bean, 1960). Acid invertase activity has been quantified in actively growing juice vesicle explants of lemon (Altman et al., 1982) and both immature and mature juice vesicles of Satsuma (Kato and Kubota, 1978). Altman et al. (1982) suggested acid soluble invertase increased in lemon juice vesicles in the presence of active callus growth. Kato and Kubota (1976) found that the presence of both acid and alkaline invertases corresponded to low sugar content in immature juice vesicles, but that increasing sugar content in edible fruit took place when alkaline invertase was active. Highest acid invertase activity was found in the soluble fractions, while alkaline invertase activity occurred in the insoluble fraction. The distribution of alkaline invertase between the soluble and insoluble fractions changed with fruit development and shifted toward soluble fraction.

PAGE 23

CHAPTER III THE VASCULAR SUPPLY AND ASSIMILATE TRANSPORT IN GRAPEFRUIT Introduction Structure and function appear to be closely related in many transport tissues. Anatomical features in 'Marsh' grapefruit appear to influence the kinetics of photosynthate translocation (Koch, 1984a; Koch, unpublished data) and patterns of deposition (Koch and Avigne, 1984). Overall distribution of 14C-photosynthates within grapefruit depends on segment alignment to specific source leaves and the positioning of intervening vascular strands (Koch and Avigne, 1984). Assimilate movement to juice tissues is limited to three major vascular traces positioned outside a given segment (Koch, 1984a) (Fig. 3-1). Approximately half of the photosynthates entering juice tissues are unloaded from phloem in dorsal vascular strands and pass through the innermost cell layers of albedo, segment epidermis, juice vesicle stalks and finally to vesicle heads in stage III fruit (Koch, 1984a; Koch and Avigne, 1986) (Fig. 3-1). The remaining assimilates arrive via the two septal bundles and follow a similar route, although septal parenchyma cells are present instead of cells from the inner albedo. The central trace in the innermost portion of the fruit (central axis) supplies assimilates to ovules and seeds of a given carpel, but is completely separate from nearby juice tissues (Koch, 1984a) (Fig. 3-1). Further structure/function associations are evident along the post-unloading path of photosynthate transfer. Assimilates move slowly from phloem to juice vesicles and do so only by traversing phloem-free 16

PAGE 24

17 Figure 3-1. Diagram of a transverse section through grapefruit showing the relationship between vascular and non-vascular tissues. Shaded area represents juice tissues. a=albedo; c=central bundle; d=dorsal bundle; jv=juice vesicle; s=septal bundle; se=segment epidermis.

PAGE 25

18 parenchyma in a juice vesicle stalk (Koch, 1984a; Koch and Lowell, unpublished). This vascular distribution, a cutinized segment epidermis and complete lack of vascular tissue inside juice segments contributes to massive phloem unloading at highly localized sites. Photosynthate import in numerous other sink tissues circumstantially has been linked to structural features in antomical studies. This has been particularly true for the zone of phloem unloading and tissues separating this zone from the site of final assimilate deposition. Vascularization associated with assimilate deposition can be discontinuous (Felker and Shannon, 1980; Oparka and Gates, 1981; 1982; Thorne, 1981; Kuo and Pate, 1985). Vascular tissue supplying photosynthates to the final site of deposition may contain only phloem (Hardham, 1976; Thorne, 1981; Kuo and Pate, 1985) or modifications in cells of the phloem (Sauter and Braun, 1972; Hardham, 1976; Oparka and Gates, 1981a; Hayes et al., 1985). Cross sectional areas of vascular strands associated with juice tissues in citrus fruit increase during development (Ford, 1942; Holtzhausen, 1969) by additional primary or secondary growth within bundles (Roth, 1977). These associations and those noted for citrus have motivated a closer examination of transport structures within this fruit. Materials and Methods Bain (1958) recognized three general developmental stages in 'Valencia' orange; stage I--cell division, stage II--cell enlargement and stage Ill-non-climacteric maturation. Similar growth stages were determined for 'Marsh' grapefruit ( Citrus paradisi Macf.) in the present study, based on anatomical, morphological, volume and fresh and dry weight changes in the entire fruit and individual tissues (albedo,

PAGE 26

19 segment epidermis and juice vesicles). 'Marsh' grapefruit (5 fruit minimum) were collected between 3 and 9 months of fruit growth from the outer canopy on the southern exposure of mature trees at approximately 3 m height on a monthly basis at commercial groves in Lake Wales, Florida (1985). Polar and equatorial diameters as well as fresh weights (excluding calyx) were measured for each fruit. Whole fruit volumes were calculated assuming the fruit to be spheres. Relative water content was determined for albedo, segment epidermis and juice tissues of each fruit. Fruit of 'Marsh' grapefruit were collected at monthly intervals at the University of Florida (1983) and Lake Alfred, Florida (1984). The zone of phloem unloading and tissues involved in subsequent assimilate transfer were killed and fixed in half-strength Karnovsky's fluid (Karnovsky, 1965) with 0.075 M Pipes, pH 7.5, or 1.57. acrolein-37. glutaraldehyde-1.57. paraformaldehyde (Hayat, 1981) followed by 17. OsO^ dehydrated with an ethanol and acetone series and infiltrated with Spurr's resin (Spurr, 1969). One urn sections were stained with 0.57. toluidine blue in 0.17. NaCO,. Ultrathin sections (100 nm) were cut with glass knives, and then stained with uranyl acetate and lead citrate and mounted on formvar coated slot grids. Sections were analysed on a Philips 301 electron microscope. Fruit tissue also was killed and fixed in a 707. formalin-acetic acid-ethanol solution (FAA), and then dehydrated with a tertiary butanol series (Johansen, 1940) and infiltrated with Paraplast (56.5C) (Sherwood Medical Industries). Sections (10-15 pm) were stained with a saf ranin-f ast green series and examined with a Nikon light microscope. Fresh tissue also was sectioned (40-80 pm) on a sliding microtome and stained with saf ranin-f ast green or 27. phloroglucinol in 957. ethanol

PAGE 27

20 and concentrated HC1. Distances between the dorsal bundle and juice vesicle stalk base and stalk lengths were measured with a calibrated ocular micrometer from 15 urn sections and fresh or frozen material, respectively. Diagrams of whole juice segments and the relation of vascular tissue to juice vesicles were drawn to scale from fresh mature juice segments. Results and Discussion Fresh and dry weight changes are shown for whole fruit and individual tissues of 'Marsh' grapefruit (Fig. 3-2) in relation to the three stages of growth defined by Bain (1958). Rates of increase in total fruit volume and fresh weight were similar throughout fruit growth. Only small gains in fresh and dry weight were observed for fruit tissues during the first 8to 9-week cell division stage (stage I), which was early March to mid-May in central Florida. Increases were approximately linear for total fruit volume and fresh weight during cell expansion (stage II) which lasted 20-21 weeks (mid-May to late September). Fresh and dry weight accumulation by individual tissues also increased during this time. Based on dry matter accumulation, the albedo appeared to be the main sink for photosynthates within the fruit during stages I and II, while juice tissues became the predominant sink late in stage II. These findings are similar to those of Holtzhausen (1969). A slower rate of increase was observed in both fresh and dry weights of component fruit tissues during fruit maturation (stage III) lasting 9 to 12 weeks in this study.

PAGE 28

21 APR 111 LuNEl I AUG I I OCT I Figure 3-2. Fresh weights of whole fruit (A) and fresh and dry weights of albedo, segment epidermis and juice vesicles at biweekly to monthly intervals. Data represent the average of 5 fruit per measurement. Standard error (SE) : 0.01
PAGE 29

22 Vascular Distribution in the Fruit Distribution of major vascular bundles entering the fruit from stem to stylar end is diagramed in figure 3-3a. This vasculature is confined primarily to the albedo and central axis of the fruit. Vascular tissue entering the fruit radiates directly from that of the pedicel. This radial distribution coincides with a narrow distribution of lAC-photosynthates translocated from one source leaf to a specific portion of the grapefruit sink (Koch, 1984b). Vascular strands which supply assimilates to the peel diverge from the pedicel stele before those associated with juice tissues. Vascularization of the albedo is the most extensive in the fruit. More than 4 times the number of traces associated with juice segments enter the albedo, although the number of traces does not necessarily reflect the quantity of phloem tissue entering the albedo as compared to the rest of the fruit. The number of vascular traces branching from the main stele typically is in multiples of the number of juice segments present, usually 12 (range: 10 to 14) in the fruit (Fig. 3-3c). Vascular traces comprising the stele in the grapefruit pedicel and at the point of entry into the fruit are collateral (phloem on the outer tangential surface of the xylem). A narrow transition zone is present interior to the fruit-pedicel juncture in the stem end of the fruit where collateral bundles become amphicribral (completely surrounded by phloem) (Fig. 3-3b, 3-3c, 3-6). The presence of additional primary xylem and phloem in bundles further into the fruit suggest that adjacent parenchyma cells may redif f erentiate to form additional primary vascular tissue. The ratio of phloem to xylem in vascular bundles that enter and anastomose in the albedo increases as veinlets decrease in size with

PAGE 30

Figure 3-3. The vascular organization within 'Marsh' grapefruit after approximately 12 weeks of development (early stage II). c=central bundle; d=dorsal bundle; l=locule; s=septal bundle. A. Longitudinal diagram of a grapefruit depicting positions of major vascular bundles within the fruit. Arrows refer to the approximate location from which each section was taken. B. Distribution of vascular bundles in the pedicel. Bar=20 urn. C. Whorls of vascular traces diverge from the main stele and enter the albedo. Bar=10 urn. D. Approximately 12 vascular traces branch from the stele to form dorsal bundles. One septal trace diverges from each dorsal trace. Central axis traces are derived from remaining stelar vascular tissues. Bar=10 urn. E. Each juice segment is associated with 1 dorsal, 2 septal and 1 central vascular trace. Bar=10 urn. F. Septal, dorsal and central vascular traces fuse to form a ring of approximately 12 vascular traces. Numerous traces from the albedo subsequently fuse with these bundles. Bar=10 um.

PAGE 31

24 pedicel (31 sections) pedicelfruit juncture (19 sections) albedo above juice segment 1 (7 sections) juice segment (16 sections) albedo below juice segment (17 sections)

PAGE 32

25 distance from their site of entry as with other fleshy fruits (Roth, 1977). Near the flavedo, for example, these bundles may consist of a single primary xylem vessel member surrounded by small amounts of xylem parenchyma, possibly a vascular cambium and mostly primary and secondary phloem parenchyma. This increase in phloem may represent a morphological adaptation to minimize fruit desiccation and to increase the capacity of phloem unloading in the albedo. Types of vascular bundles may vary within the same fruit. Apple may have collateral, bicollateral, amphicribral bundles and inverted central bundles (MacDaniels, 1940) similar to those in grapefruit and 'Eureka' lemon (Ford, 1942), Amphicribral bundles are common in fruit of such plant families as Rosaceae, Leguminosae, Anacardiaceae, Solanaceae, Cucurbitaceae (Roth, 1977) and Rutaceae (Ford, 1942). Inverted bundles also frequently occur in fruit, especially in those fruit where each locule of a compound ovary morphologically represents a modified leaf or carpel (MacDaniels, 1940). Primary xylem and phloem appear inverted at the stylar end of 'Eureka' lemon at the point of septal and central trace fusion, and prior to fusion of these traces with dorsal bundles (Ford, 1942). Bundle inversion does not appear to occur in grapefruit. However, traces at this stage are amphicribral, so inversion would be difficult to detect. The pattern of vascular fusion and bundle inversion does not necessarily indicate the direction of photosynthate movement at the stylar end of the fruit. Vascular Supply to Juice Tissues Four amphicribral traces are positioned around the exterior of each juice segment; 1 dorsal, 2 septal and 1 central bundle (Fig. 3-3d, 3-3e). Dorsal bundles branch from central vascular strands and subsequently give

PAGE 33

26 rise to septal bundles (Fig. 3-3a, 3-3d). Septal bundles are shared between adjacent juice segments. Anastomoses between dorsal and septal vascular bundles occur, but primarily are confined to the stem end of the fruit (Fig. 3-3a). The locules or juice segments end and septal, dorsal and central bundles fuse to form a ring of discrete amphicribral bundles near the stylar end of the grapefruit (Fig. 3-3f). Distance between the dorsal bundle (a site of phloem unloading) and the bases of juice vesicle stalks varies with the stage of fruit development (Fig. 3-4). The distance from dorsal bundle to stalk base decreases from stage I to stage III, averaging 260.0 pm + 16.33 (range 34.28-605.70 urn) in mature stage III fruit. This results in the dorsal bundle being considerably closer to the segment epidermis (Fig. 3-4b). Septal bundles also are repositioned during development from positions in the albedo to midway between the albedo and central axis and between the segment epidermis of adjoining juice segments (Fig. 3-4). This change in position of the dorsal and septal bundles occurs during stage II growth as juice vesicles and segment epidermis expand. The actual distance to juice vesicle stalks may not change with development, but vascular bundles are adjacent to the segment epidermis on both radial surfaces. The heavily cutinized segment epidermis may limit the flow of assimilates into the albedo. Each juice vesicle joins tissues exterior to the juice segment at a point near a dorsal or septal vascular strand, and areas of juncture thus occur primarily in three rows (Fig. 3-5). A small portion of the vesicle stalks also are associated with much smaller vascular traces that branch between dorsal and septal strands. The bases of these vesicles are at most 6 mm away from the nearest vascular strand. At least twice as many

PAGE 34

27 Figure 3-4. Change in position of dorsal and septal vascular bundles with respect to albedo and segment epidermis during development. Shaded area represents juice tissues. a=albedo; c=central bundle; d=dorsal bundle; s=septal bundle; se=segment epidermis. A. Transverse section of juice segment and associated vascular tissue during stage I growth (approx. 12 weeks). B. Transverse section of juice segment and associated vascular tissue during stage III growth (approx. 24 wks).

PAGE 35

Figure 3-5. Diagrams of segment epidermis and vascular bundles from the exterior of three grapefruit locules. Each point designates the site of the attachment of a single juice vesicle. Longitudinal lines represent dorsal, septal and central vascular traces. Transverse connections between these represent branching between dorsal and septal traces.

PAGE 36

29

PAGE 37

30 juice vesicles are present in the stem versus stylar half of the fruit and two-fold more juice vesicles also are associated with the dorsal versus septal surfaces of the juice segment (Fig. 3-5). Development of the Dorsal Bundle Developmental changes in the dorsal vascular bundle are shown in figure 3-6a and 3-6b. Primary phloem of young dorsal (and septal) traces includes parenchyma cells interspersed with narrow sieve tube members and companion cells. The majority of the primary phloem in dorsal bundles is adjacent to the largest portion of the albedo (Fig. 3-6a). Primary xylem in dorsal bundles occurs in radiating rows of narrowlumened vessel elements and tracheids interspersed with xylem parenchyma (Fig. 3-6a). This also is true in septal bundles (data not shown). Dorsal, septal and central bundles increase in size by additional primary growth during the latter part of stage I and early part of stage II. A vascular cambium differentiates during stage II in these bundles and produces new cell layers of secondary phloem and xylem parenchyma (Fig. 3-6b). This growth in dorsal bundles is most extensive in the phloem adjacent to the segment epidermis and juice vesicle stalks. Vascular cambia occur less frequently in septal traces and rarely in central traces. The vascular supply in 'Washington' navel orange (Hoi tzhausen, 1969), lemon (Ford, 1942) and many other fruits (Roth, 1977) also increases by additional primary or secondary growth in individual vascular bundles. A similar increase in xylem and phloem tissues in developing seeds of Pisum sativum L. coincides with a rise in the supply of assimilates in this fruit (Hardham, 1976). Unlignified phloem fibers typically develop in the portion of the dorsal bundle furthest from the juice vesicles during stage I growth.

PAGE 38

31 Figure 3-6. Transverse sections of dorsal vascular bundles at different developmental stages. (Inner albedo at top of page). A. Dorsal vascular bundle after approximately 12 weeks of fruit development (early stage II). Note amphicribral structure. p'=primary phloem; x'=primary xylem. Bar=5 urn. B. Dorsal vascular bundle after approximately 24 weeks of fruit development (stage III). bc=bundle cap; c=vascular cambium; p=primary and secondary phloem; x=primary and secondary xylem. Bar=10 Jim.

PAGE 39

32 These fibers subsequently lignify to form a band or bundle cap that separates primary phloem into two zones during stage II growth. Primary phloem peripheral to the bundle cap is crushed by late stage II to early stage III growth (Fig. 3-6b). Fibers also lignify sporadically around the perimeter of the dorsal bundle during this time. A similar developmental sequence is not evident in other vascular bundles supplying juice segments. Phloem fibers occur occasionally in septal bundles, but none are found in central bundles. Formation of a bundle cap in dorsal bundles coincides with the switch in dominant sink tissue from albedo to pulp on a fresh and a dry weight basis during stage II development (Fig. 3-2). Unlike pigment strand suberization in rice caryopsis (Oparka and Gates, 1982), fibers of the bundle cap in grapefruit lack symplastic continuity and form a barrier to intercellular movement of assimilates. Assimilates entering the fruit from the dorsal bundle prior to this time, therefore, may unload from phloem tangentially into both albedo and juice tissues during early fruit development. Transfer subsequent to stage II is hindered in the direction of the albedo. Conclusions A functional significance in relation to photosynthate transfer is suggested by several structural features of vascular and non-vascular portions of the transport path in grapefruit. Three highly localized zones of phloem unloading are indicated for juice segments by 1) lack of vascular strands in the interior of juice vesicles; 2) limitation of cellular continuity between phloem outside segments to the 3 major vascular bundles (1 dorsal, 2 septal); and 3) isolation of juice tissues by a continuous, cutinized epidermis surrounding segments and juice

PAGE 40

33 vesicles. Other aspects of vascular structure and distribution that appear most closely associated with photosynthate deposition include direct alignment of vascular tissue entering the fruit to juice segments, increase in the phloem to xylem ratio in individual vascular traces and changes in the dorsal vascular structure. In the latter, changes in position and vascular tissue and formation of a bundle cap in dorsal bundles coincide with the stage II switch in sink strength (based or dry matter partitioning) between albedo and juice tissues.

PAGE 41

CHAPTER IV NON-VASCULAR TISSUES IN GRAPEFRUIT DURING DEVELOPMENT Introduction The spatial arrangement of vascular and non-vascular tissues in citrus fruit influences the kinetics of photosynthate translocation (Koch, 1984a; Koch, 1985; Koch, unpublished) and patterns of deposition (Koch and Avigne, 1984). Assimilate movement to juice tissues also is limited to three major vascular traces positioned outside a given juice segment (Koch, 1984a). This vascular distribution, a cutinized segment epidermis and the complete lack of vascular tissue inside juice tissues contributes to the apparently extensive phloem unloading at highly localized sites (see previous chapter). Assimilates move slowly from phloem to juice vesicles only by traversing the length of phloem-free parenchyma in a juice vesicle stalk (Koch, 1984a; Koch, 1985). Anatomical studies of other sink tissues have suggested there is a close relationship between structural features and processes of photosynthate import. This has been particularly true in the zone of phloem unloading and along the path of subsequent assimilate movement to the site of final deposition. Non-vascular portions of this path typically include modifications such as numerous symplastic connections (Helder and Boerma, 1969; Gunning and Hughes, 1976; Eleftheriou and Hall, 1983; Hayes et al., 1985; Warmbrodt, 1985), transfer cells (Felker and Shannon, 1980; Harris et al., 1982), discontinuous cell layers (Felker and Shannon, 1980) or suberized/cutinized layers (Gunning and Hughes, 1979; Oparka and Gates, 34

PAGE 42

35 1981a; 1982; Eleftheriou and Hall, 1983). Developmental changes also may occur in the extent of symplastic connections (Juniper, 1977; Gunning, 1978; Schnepf and Sych, 1983; Seagull, 1983) or cellular modifications such as plasmatubules (Harris et al., 1982). The citrus fruit, a hesperidium, is composed of pigmented exocarp (flavedo), white mesocarp (predominantly albedo) and endocarp (segment epidermis and juice vesicles) (Bain, 1958). Albedo is the highly vascularized, spongy aerenchyma of the inner peel (Ford, 1942; Scott and Baker, 1947). Endocarp is that portion of the citrus fruit which typically is eaten and, unlike albedo, it lacks vascular tissue (Dodd, 1944). This zone in oranges is divided into 8-12 locules or segments (Bain, 1958; Holtzhausen, 1969). Each of these contains numerous multicellular juice vesicles (collectively the pulp) and a variable number of seeds. Individual locules and their vesicles are surrounded by a single, continuous epidermal layer with a thin cuticle (Schneider, 1968; Roth, 1977). Juice vesicles are derived from cell divisions of the outer and tangential segment epidermis and subepidermal cell layers (Bain, 1958; Schneider, 1968; Roth and Lindorf, 1972). Expanding juice vesicles become variously "club-shaped" (Roth, 1977) with a thin, parenchymatous stalk joining the vesicle head to tissues outside the juice segment at a site near a major vascular strand. Three vascular strands are associated closely with the exterior of each juice segment (Ford, 1942; Koch, 1984a). All juice tissues inside a single juice segment, therefore, are supplied with photosynthates by one dorsal bundle on the outer, tangential surface or one of the two septal bundles located between septae on either of the sides shared by adjacent segments (Koch,

PAGE 43

36 1984a). Seeds of a given segment receive assimilates from one vascular strand in the central axis. Citrus juice vesicles have been compared to "spouting glands" or glandular hairs on leaves and inflorescences of Dictamnus albus L. (Rutaceae) (Anelunxen and Arbeiter, 1967). These glands also consist of a multicellular stalk and globular body, but include a secretory cavity. It is lined with several layers of secretory cells and contains essential oils (Anelunxen and Arbeiter, 1967). The following study was undertaken to examine what appeared to be a close association between structure and function for assimilate transport into citrus fruit, and because of the highly localized nature of those processes in this system. The objective of this research was to elucidate anatomical features of the non-vascular transport pathway and sites of sugar deposition at different developmental stages in 'Marsh' grapefruit ( Citrus paradisi Macf.). Materials and Methods Materials and methods were detailed in chapter III. Presence or absence of primary xylem tracheary elements was determined in 900 juice vesicle stalks from the middle 1 cm of 30 mature grapefruit by staining fresh cut stalks with 27. phloroglucinol in 957. ethanol and concentrated HC1. Stalks in which xylem elements were found were washed with deionized water, stained with toluidine blue and fresh mounted for photographs. Absence of phloem cells was verified by observation of preserved fruit tissue. Observations of plasmodesmata in cells of juice vesicle stalks were taken from ultrathin sections (approximately 60-70 nm) of stage II fruit (4-5 months) cut with a diamond knife and mounted on 300 mesh copper

PAGE 44

37 grids. These sections were stained with uranyl acetate and lead citrate followed by carbon coating for stabilization. Plasmodesmatal densities were determined morphometrically in juice vesicle stalks of mature, stage III fruit (8 months). Ten juice vesicle stalks were randomly selected from juice tissues in the median centimeter of four fruit and fixed, dehydrated and embedded in Spurr's resin as described in chapter III. To obtain a random sample of cell wall interfaces, 5 ultrathin (approximately 100 nm) serial sections were cut with a diamond knife followed by at least ten 1 urn sections and repeated according to Weibel (1969). Ten non-consecutive sets of five serial sections from each of 10 juice vesicle stalks (100 grids, 500 sections total) were mounted on formvar coated slot grids and stained with uranyl acetate and lead citrate. A maximum of 16 random photographs were taken of one section per grid with a Philips 301 transmission electron microscope according to Weibel (1969). Transverse sections, 1 um thick, were taken from the same 10 juice vesicle stalks, stained with 0.57. toluidine blue in 0.17. NaCO q and photographed with a Nikon light microscope. Partial and total cross sectional areas and plasmodesmatal densities per length cell wall interface (longitudinal and transverse) were calculated for peripheral and central stalk cells. Values were determined from 8X10 photographs using a digitizer and morphometric analysis software designed by Dr. Kenneth Curry, University of Florida. Apoplast to symplast ratios were estimated using a double square (6mm and o 24 mm ) grid (Weibel, 1969). Total number of measurements was determined in each set of data by a cumulative standard error of less than 107. expressed as percent of the mean (Bolender, 1978). Plasmodesmatal frequencies were determined per utfT-'cell wall interface and used to

PAGE 45

38 estimate frequencies per cell wall \im according to Robins and Juniper (1980). Plasmodesmatal distribution in juice vesicle stalks was non-parametric. The Wilcoxon-Mann-Whitney two-sample test for unequal sample sizes was used for analysis of significant differences (57„ level) among frequencies of longitudinal and transverse cell walls in both peripheral and central juice vesicle stalk cells (Mann and Whitney, 1947). Results and Discussion Parenchyma Adjacent to Vascular Bundles The first non-vascular tissue traversed by assimilates enroute to juice vesicles is a narrow parenchymatous layer surrounding dorsal and septal bundles (Fig 4-1). This tissue is the innermost albedo around dorsal vascular strands. Most albedo is composed of irregularly-shaped parenchyma cells separated by numerous, large intercellular spaces as previously described by Bain (1958). In contrast, the innermost albedo cells that separate dorsal bundles and juice vesicles are densely packed and are thinner-walled with a more regular elongated shape (Fig. 4-1). Intercellular spaces remain small in this area and plasmodesmata persist throughout fruit development. Thin-walled parenchyma cells also surround the septal bundles and are structurally similar to inner albedo cells. Parenchyma cells immediately adjacent to any of the three major vascular strands develop a large vacuole to cytoplasm ratio during stage II growth (5-6 months). The large vacuoles in these cells indicate a small, but potentially significant, capacity for photosynthate storage subsequent to phloem unloading. This capacity may or may not be analagous to the

PAGE 46

Figure 4-1. Zone of phloem unloading and subsequent non-vascular transport tissues within the grapefruit. a=albedo; ia=inner albedo; st=juice vesicle stalk. Bar=10 um.

PAGE 47

40

PAGE 48

41 transient, reservoir-type function ascribed to parenchyma of the maize pedicel (Felker and Shannon, I960). Segment Epidermis The segment epidermis or carpel wall shown in figures 4-1 and 4-2 is not in the physical path of photosynthate transfer into juice vesicles, but 14C-assimilate transport studies in grapefruit suggest the segment epidermis plays a important role in photosynthate movement (Koch, 1984a). The unito biseriate epidermal layer contains no stomata and is continuous with that of the entire juice vesicle. Cells of the segment epidermis, however, differentiate earlier and are more highly vacuolated than those surrounding juice vesicles. An irregular unito biseriate hypodermal cell layer occurs adjacent to the segment epidermis (Fig. 4-2a). Epidermal cells initially are columnar and become tabular with maturity, while hypodermal cells may be either columnar or tabular throughout development (Fig. 4-2a, 4-2b). Plasmodesmata are present along the radial faces of cells within the segment epidermis, and also extend from their outer tangential faces to join cells of the hypodermis. Cavities may develop within a single plasmodesma or inside radial walls where several plasmodesmata meet. Epidermal cells become highly vacuolated during development and their appearance shifts from that of metabolically active cells to one more similar to storage cells. Inner tangential walls (adjacent to juice vesicles) of epidermal cells are extremely thick and possess a striate cuticle. The striate cuticle often is 2-parted; an electron dense outer layer and an inner, less electron dense layer. The inner portion of the cuticle becomes invaginated upon fruit maturity. This cuticular layer consists of cutin and wax, but no suberin in grapefruit (Espelie et al 1980). The

PAGE 49

42 A .w^VUlY

PAGE 50

A3 continuous, cuticular epidermal layers of the segment epidermis and juice vesicle are likely to limit photosynthate passage into juice segments to points of juice vesicle stalk attachment to the carpel wall. Cell walls between segment epidermis and hypodermis thicken during the late phase (12 mo after anthesis) of grapefruit maturation (stage III) and separate along the middle lamella. Thin cuticles then develop along cell walls of these adjoining surfaces and form what would appear to be at least a partial a barrier to the apoplastic and/or symplastic transfer of photosynthates between these cell layers. At this point in development, the segment epidermis therefore may have a decreased role in photosynthate storage and/or translocation. Juice Vesicles and Vesicle Stalks No vascular connections exist within juice vesicles or between juice vesicles and the albedo, but small clusters or isolated primary xylem tracheary elements occasionally may form in stalks (Fig. 4-3). When present, these cells are narrow-lumened with oblique endwalls, simple perforation plates and bordered pits. Only up to 13.37 of the mature juice vesicle stalks possess these isolated tracheary elements. Mature vesicle stalks vary greatly in length (Table 4-1), but the transport path of photosynthates from sites of phloem unloading (dorsal and septal vascular bundles) to juice vesicle storage cells in mature tissues may be 21 mm or greater (data not shown). Labeling studies show 14C-photosynthates move through the innermost albedo and the entire length of these parenchyma strands before final deposition in vesicles (Koch, 1984a; 1985). Analogous non-vascular areas, with varying degrees of specialization and function also occur in other sink tissues such as Pisum sativum L. (Hardham, 1976), maize root (Warmbrodt, 1985), rice

PAGE 51

44 Figure 4-3. Longitudinal fresh section of juice vesicle stalk with clusters of isolated primary xylem tracheary elements in juice vesicle stalks. Bar=0.2 urn.

PAGE 52

45 Table 4-1. Dimensions of Mature Juice Vesicle Stalks from 'Marsh' grapefruit. Length (mm): 6.62 (+ 0.5 Z ) Cross sectional area (um^): total 29.0 (+ 1.7) peripheral cells 19.4 (+ 0.6) central cells 9.6 (+ 1.3) 'Standard error of the mean

PAGE 53

46 caryopsis (Oparka and Gates, 1982) and maize kernels (Felker and Shannon, 1980). However, the length of the translocation path between the last phloem cells and site of final assimilate storage in grapefruit is one of the longest described (Koch, 1985). Juice vesicle stalks are part of the multicellular vesicles which are the final site of deposition for assimilates in grapefruit segments (Fig. 4-4) Elongation of vesicle stalks begins early in stage I and is complete by stage II growth (Fig. 4-4c, 4-4d). As a result, the length of non-vascular tissue that photosynthates must traverse subsequent to phloem unloading increase during fruit growth. Cell length can contribute to factors determining whether transport through a given tissue is apoplastic or symplastic. In extremely long parenchyma cells, the rate of cytoplasmic streaming could limit intercellular transport (Gunning and Overall, 1953). Structure of parenchyma cells varies within vesicle stalks. Peripheral cells are narrow, elongate, and thin-walled, while central cells are larger and thinner-walled (Fig. 4-4a, 4-4b). The total cross sectional area occupied by peripheral cells within the stalk is 2-fold greater than that of central cells (Table 4-1). Peripheral cells remain smaller and more densely cytoplasmic than interior cells. Morphometric measurements show that the total apoplast, including cell walls and intercellular spaces, comprises approximately 187of the total cross sectional area of central cells but 247. near the perimeter of the vesicle stalk. Parenchyma cells in the vesicle stalks are initially isodiametric to slightly oblong, but elongate extensively during stage II development, proceding from peripheral to central cells (Fig. 4-4c, 4-4d).

PAGE 54

Figure 4-4. Non-vascular juice tissues at different developmental stages. c=central cells; e=epiderrnis ; pl=peripheral cells. A. Transverse section of stage I juice vesicle stalk (approx 12 wks). Bar=l urn. B. Transverse section of stage III juice vesicle stalk (approx 20 wks). Bar=l urn. C. Longitudinal section of stage I juice vesicle stalk (approx 12 wks). Bar=0.2 urn. D. Longitudinal section of stage III juice vesicle stalk (approx. 27 wks). Bar=0.2 pm. E. Transverse section of stage I juice vesicle head (approx. 11 wks). Bar=0.2 um. F. Transverse section of stage III juice vesicle head (approx. 25 wks). Bar=0.2 um.

PAGE 55

48

PAGE 56

49 Each stalk cell possesses a large nucleus with one to several nucleoli. Multivesicular bodies and small vesicles are common, as are plastoglobuli and long strands of endoplasmic reticulum. Sub-cellular vesicles increase slightly in number upon maturity, especially in peripheral parenchyma cells. Golgi apparati are extremely rare, but appear active based on numbers of associated sub-cellular vesicles. The majority of parenchyma cells in the vesicle stalk posses one large or several small vacuoles by 3 months of growth (early stage II). Central cells in particular become highly vacuolated by stage II and may act as temporary reservoirs for assimilates enroute to juice vesicle heads. Pulse-chase labeling studies have shown an extremely slow turn-over period for 14C-photosynthates moving through juice vesicle stalks (Koch, 1984a; Koch, 1965; Koch, unpublished data). Juice vesicles also contain chloroplasts with starch granules during stage I and early stage II growth. Unlike starch grains in the albedo of 'Valencia' orange (Brown and Barmore, 1981), starch grains in grapefruit tissues disappear by early stage II growth during chloroplast differentiation into chromoplasts Chromoplasts typically include achlorophy llous membranes and numerous plastoglobuli. Starch has been reported in citrus stems, roots and fruit (Cameron, 1933; Kordan, 1971a; 1971b; Yelenosky and Guy, 1977; Brown and Barmore, 1981), but only in very low amounts. Major storage products in grapefruit are soluble sugars (Sinclair, 1984), thus conversion of photosynthates to starch probably is a minor component of early storage in the fruit. Both young and mature heads of juice vesicles share most features of juice vesicle stalks, yet differ in some important respects (Fig. 4-4d, 4-4f). A major difference is a proliferation of central cells.

PAGE 57

50 Parenchyma cells in vesicle heads also enlarge considerably more and elongate much less than in vesicle stalks. Cell walls become increasingly thin during late stage II growth (6-7 months) and central cells possibly may rupture in mature fruit (Fig. 4-4f). Such cavity formation in juice vesicles could result from lysigeneous (Banerji, 1954) or schizolysigeneous cell breakdown (Davis, 1932). Other authors contend that central cell loss in juice vesicles of 'Eureka' lemon was caused by high turgor pressure and that juice vesicles should not be compared to glands (Ford, 1942). Citrus juice vesicles have been grouped into a broad category of glandular hairs (Davis, 1932; Banerji, 1954; Anelunxen and Arbeiter, 1967; Fahn, 1979). Secretion from symplast to apoplast within other glandular structures is envisioned to occur in one of two ways. Granulocrine secretion takes place when sugars are packaged into sub-cellular vesicles which subsequently fuse with the plasmalemma and dispell their contents. The eccrine mechanism involves only the relatively direct transport of sugars across the plasmalemma (Fahn, 1979). Circumstantial evidence for glandular secretion in cells other than those of citrus juice vesicles includes the presence of abundant dictyosomes, ribosomes and mitochondria, a well-developed endoplasmic reticulum and in the case of granulocrine secretion, numerous membrane vesicle fusions at the plasmalemma (Kuo and Pate, 1985). These characteristics were not observed in maturing and mature citrus juice vesicles, which instead, contained cells with large vacuoles and few organelles. These data suggest that the physiology and sub-cellular structure of maturing and mature juice vesicles probably are not comparable to glandular hairs.

PAGE 58

51 Plasmodesmata Plasmodesmata are present in lateral and end walls of cells throughout juice vesicle stalks. These subcellular structures are not necessarily confined to primary pit fields and are clustered primarily in groups of varying size (Fig. 4-5a, 4-5c). Both simple and bifurcated plasmodesmata occur(Fig. 4-5b), with outer diameters ranging from approximately from 36 to 46 nm. These values are somewhat smaller than the 60 nm diameter cited as standard for a plasmodesma by Gunning and Overall (1983), but diameters are known to vary widely with position in the cell wall (Gunning and Overall, 1983). Plasmodesmatal frequencies are significantly higher among central versus peripheral cells in mature (stage III) juice vesicle stalks (Table 4-2). No significant differences were evident in this respect between longitudinal and end walls of cells within a given area of tissue. If photosynthate transport through these stalks was symplastic, end wall specialization could be expected. In reviews by Robards (1976) and Gamalei (1985), plasmodesmatal densities are described as typically increasing with the degree of cellular specialization for assimilate translocation. Plasmodesmatal densities reportedly range from 7 to 140 urn "cell wall where the symplast is believed to be the primary transport path. Examples studied have been meristematic cells (Arisz, 1969; Warmbrodt, 1985), tangential walls of endodermal cells in roots (Helder et al., 1969; Seagull, 1983; Warmbrodt, 1985), secretory cells (Gunning and Hughes, 1976; Meyberg and Kristen, 1981) and bundle sheath cells in C-4 leaves (Evert et al., 1977; Fisher et al., 1982). In contrast, parenchymatous cells with less obvious roles in direct symplastic transport of assimilates have plasmodesmatal densities ranging from 0.1

PAGE 59

52 Figure 4-5. Plasmodesmata in the juice vesicle stalk. A. Peripheral cells of a juice vesicle stalk. Arrows denote plasmodesmata. Bar=l um. B. Plasmodesmata in mature juice vesicle stalk (approx. 27 wks) can be simple, form a central node or bifurcate as pictured here. Bar=0.2 pn. C. Primary pit field in cell wall of juice vesicle stalk cell. Bar=0.1 pm.

PAGE 60

53 Table 4-2. Plasmodesmatal frequencies within parenchyma tissues of juice z vesicle stalks from 'Marsh' grapefruit. Position of parenchyma tissue within juice vesicle stalk plasmodesmatal frequency at interface between 2 cells Peripheral cells: longitudinal walls transverse walls (urn wall) -2 (urn wall) 0.27 (+0.08)** 2.20 (+0.72) 0.24 (+ 0.07)* 1.96 (+ 0.58) Central cells: longitudinal walls transverse walls 0.39 (+0.12)** 3.78 (+1.05) 0.38 (+ 0.09)* 2.96 (+ 0.76) Z Plasmodesmatal frequency along a linear portion of the interface between two cells was determined from 202 to 505 ura cell wall length as described by Weibel. Subsequent conversion to frequency per unit area was done with equations of Robins et al. (1980) which yielded values 3% less than those of Robards (1976). Values otherwise represent maximum frequencies which could be obtained from these tissues because a plasmodesma was counted even if not wholely visible in a given section. ** Different at p 0.05. Different at p' 0.10.

PAGE 61

54 -2 to 10 urn" cell wall. Densities vary widely because of differences in species and methodology. Offler and Patrick (1984) relate low plasmodesmatal frequencies to a primarily apoplastic radial transfer of photosynthates in bean stems. Plasmodesmatal frequencies in grapefruit juice vesicle stalks then, may support a relatively low rate of photosynthate inflow to juice vesicle heads, but these rates probably are not significant of symplastic specialization for assimilate transport. Conclusions The longest component of the phloem-free transport path in grapefruit is the parenchymatous juice vesicle stalk. This elongates typically up to 16 mm in mature fruit. Cells are long and narrow-lumened at the stalk periphery, but short and wide-lumened in the center. Plasmodesmatal frequencies in mature stalk cells suggest a lack of specialization for symplastic transfer. Other parenchyma nearer the vascular bundles also remain non-specialized relative to assimilate transport. Segment epidermis does not appear to be directly in the path of photosynthate transfer, but several features suggest it may be important in this process. This cutinized, unito biseriate cell layer is continuous with that of juice vesicles and is likely to limit photosynthate passage into these structures to points of juice vesicle stalk attachment to the carpel wall. Segment epidermal cells become highly vacuolated during development and their appearance shifts from that of metabolically active cells to one more similar to storage cells. Cells of mature juice vesicles, the final site of assimilate storage, have low cytoplasm to vacuole ratios and show no anatomical modifications for secretion.

PAGE 62

CHAPTER V SUCROSE METABOLIZING ENZYMES AND ASSIMILATE PARTITIONING IN SINK TISSUES OF DEVELOPING GRAPEFRUIT Introduction Data from several plant systems have suggested widely varied roles for sucrose metabolizing enzymes in phloem unloading and subsequent storage of photoassimilates in sink tissues (Glaziou and Gaylor, 1972; Dick and ap Rees, 1976; Giaquinta, 1979; Silvius and Snyder, 1979; Eschrich, 1980; Claussen, 1983b; Wolswinkel, 1985). The three major enzymes that catalyze either synthesis or hydrolysis of sucrose are sucrose phosphate synthetase, invertase and sucrose synthetase. Sucrose phosphate synthetase (sucrose synthesis) and invertase (sucrose cleavage) generally are considered irreversible enzymes in vivo (Hawker, 1985). Sucrose synthetase, however, is readily reversible (Hawker, 1985). The onset of assimilate accumulation in various sink tissues often is accompanied by the appearance of either acid invertase (Shannon and Dougherty, 1972; Echeverria and Humphreys, 1984) or alkaline invertase (Kato and Kubota, 1978; Giaquinta, 1979). Photosynthate-importing tissues frequently also have a higher ratio of sucrose synthetase activity to that of sucrose phosphate synthetase (Giaquinta, 1979). The photosynthetic product most commonly transported in plants is sucrose, but sink tissues store large quantities of assimilates in additional forms such as other soluble sugars and starch (Jenner, 1980; Hawker, 1985). Sucrose synthetase is believed to invert sucrose in vivo thus providing substrate for starch synthesis in starch-storing 55

PAGE 63

56 tissues (Hawker, 1985). Still, starch content and sucrose synthetase activity in cotton are not necessarily related (Hendrix and Huber, 1986). In addition, sucrose synthetase may function in the synthetic direction in sugar-storing tissues, such as sugar beet taproot, mature sugar cane stalks, cucumber fruit and eggplant fruit (Hatch, 1963; Giaquinta, 1979; Gross and Pharr, 1982; Claussen, 1983b, 1985). Further study is required in both starch and sugar-storing sink tissues to determine the function of sucrose synthetase in photosynthate accumulation. Sucrose synthetase activity has been demonstrated in the soluble portion of 'Valencia' orange and the particulate portion of 'Eureka' lemon juice tissues (Bean, 1960). No sucrose phosphate synthetase activity has been detected in these tissues. Incorporation of radiolabeled glucose and fructose into sucrose in fresh tissue slices from various aged flavedo, albedo and juice tissue, and intact young fruit suggest that the albedo is the most active site of sucrose synthesis in lemons, although each tissue makes some contribution to sucrose synthesis (Bean, 1960). Lemon albedo tissue maintained a relatively constant rate of sucrose synthesis throughout fruit growth, while this rate declined rapidly during development of flavedo and juice tissues (Bean, 1960). Acid invertase activity has been measured in extracts from growing juice vesicle explants of lemon (Altman et al., 1982) and both immature and mature juice vesicles of Satsumma (Kato and Kubota, 1978). Kato and Kubota (1978) found acid invertase from immature juice vesicles was active only when their sugar content was low. Alkaline invertase, however, was active during periods of increasing cellular sugar content in mature fruit. High acid and alkaline invertase activities in citrus

PAGE 64

57 both could contribute to production of hexoses for activily growing cells (Kato and Kubota, 1978). Most citrus fruit accumulate sugars over a relatively long growth period (8-9 months) (Sinclair, 1972; 1984). Assimilates move into both the heavily-vascularized and non-vascular portions of the fruit during development, but a mid-season change occurs in dry matter accumulation between these tissues. In the present work, activities of sucrose synthetase and invertase (acid, alkaline, soluble and insoluble forms) were assayed during development of individual 'Marsh' grapefruit tissues in conjunction with a study of sugar accumulation and dry matter partitioning. The presence and developmental fluctuations of sucrose synthetase and invertase activity in these tissues are discussed in relation to variations in relative sink strength. Possible roles are suggested for sucrose synthetase and the four forms of invertase in phloem unloading and/or subsequent storage of sugars in these tissues. Materials and Methods 'Marsh' grapefruit ( Citrus paradisi Macf.) were collected from the outer, southern canopy of 6 mature trees in Lake Wales, Florida biweekly in May and June and monthly from July to November, 1985. Each replication consisted of 4 fruit from 2 trees. Fruit also were collected in a similar manner from 9 mature trees in Lake Alfred, Florida in 1984. In this case, 6 fruit from 3 trees constituted one replication. Each measurement was replicated 2 or 3 times. Equatorial and longitudinal diameters were measured, then the individual fruit were weighed and separated into albedo (inner peel), segment epidermis (including inner albedo and central axis) and juice vesicles. Relative water content of each tissue also was determined. Sucrose and glucose levels were

PAGE 65

58 measured by thiobarbituric acid (Percheron, 1962) and glucose oxidase methods (Sigma, Bull. #510), respectively. Tissues were frozen and kept at -8Gc until use. Enzyme activity varied little between extracts from fresh versus frozen tissues, and remained constant under stated conditions at least one year (data not shown). Based on results from 1984, fruit with an average equatorial diameter between 30-40 mm, 79-85 mm and 89-95 mm were designated stage I, stage II and stage III fruit, respectively. Extraction and partial purification of enzymes were conducted at to 5 C. Sucrose synthetase was extracted by homogenizing for 5 min with a Brinkman polytron tissue homogenizer using a ratio of 1 g:5 ml 200 mM Hepes, pH 7.5, 20 mM sodium ascorbate, 10 mM cysteine-HCl, 5mM MgCl 5mM DTT and 107. PVPP. Homogenate was filtered through 4 layers of cheesecloth, rinsed with 5 ml extraction buffer minus PVPP and the filtrate centrifuged at 20,000g for 10 min. Proteins in the supernatant were precipitated with 80% (NH )jS0 and centrifuged as above for 20 min. The pH of the solution was adjusted to pH 7.5 when necessary with 50 mM (NH^)OH. The precipitate was resuspended in 807. (NH ) SO in the above extraction buffer, diluted 1:5 with additional buffer and centrifuged for an additional 20 min. The precipitate was resuspended in 10 mM Hepes, pH 7.5, 0.25 mM MgCl 2 0.25 mM DTT and 5 mM EDTA and centrifuged as above for 10 min. Proteins were desalted on a sephadex G-25 column (1.0 x 6.0 cm) with 10 mM Hepes, pH 7.5, 0.25 mM MgCl and DTT. The eluent between 1 and 1.5 V :V (elution volume:void volume) was collected and protein e o content determined by the method of Bradford (1976) using BSA as a standard

PAGE 66

59 Soluble invertase was extracted similarily except potassium phosphate (monobasic) buffer was substituted for Hepes, sodium ascorbate and cysteine-HCl were eliminated from the extraction buffer and 57. PVPP was used. Also, the precipitate was not resuspended in 807. (NH .) SO 2 4 Soluble acid invertase also was extracted according to Purvis and Rice (1983), and utilized in an additional study of biweekly to monthly activity. The buffer used was 100 mM Hepes, pH 8.0, 5 mM EDTA and 57. PVPP. The crude supernatant was dialysed for 20-24 hrs in 3 volumes of 10 mM phosphate buffer, pH 7.5. Cell wall material for insoluble invertase determinations was washed in 150-200 mis extraction buffer, diluted 1:40, and excess buffer was removed in a Buchner funnel. Optima of pH were determined for sucrose synthetase in grapefruit to be 8.5 and 5.5 when assayed in the synthetic and clevage directions, respectively. Sucrose synthetase was assayed in the synthetic direction in a reaction medium containing extract, 80 mM Hepes, pH 8.5, 5 mM KCN, 5 mM NaF, 100 mM fructose and 15 mM UDPG in a total volume of 0.5 ml. Sucrose synthetase also was assayed in the cleavage direction using an incubation medium of extract, 80 mM Mes, pH 5.5, 5 mM KCN, 5 mM NaF, 10 mM sucrose and 5 mM UDP in a total volume of 0.5 ml. Reactions took place at 30 C and were terminated after 15 min by boiling for 1 min. Product degradation occurred after 2 or more min (data not shown). UDP and UDPG production was quantified according to Bergmeyer (1965). Optima of pH were determined for acid and alkaline invertases in grapefruit to be 4.5 and 7.5, respectively. Soluble invertases were assayed in a reaction medium containing extract, 80 mM acetate-K P0 (pH 4.5 or 7.5 for acid and alkaline forms, respectively) and 100 mM sucrose in a total volume of 0.5 ml. Insoluble invertases were assayed in a

PAGE 67

60 similar manner except that 0.2-0.5 g cell wall material was used and the final reaction volume was 1.0 ml. Reactions were incubated for 15 to 30 min at 45C. Biweekly to monthly measurements of soluble acid invertase were made at 37C using 0.67 M acetate buffer (pH 4.7). Reactions were terminated by the addition of Nelson's reagent A (Nelson, 1944). Glucose was quantified by the method of Nelson (1944). Results and Discussion Both the albedo and juice vesicles are major sinks for photosynthates in the fruit based on dry matter accumulation during development (Fig. 3-2). Albedo is the tissue which imports the largest amount of photosynthate in early fruit development, while juice vesicles predominate during mid-stage II (5-6 months). The segment epidermis accumulated little dry matter when compared to the other tissues and therefore, did not constitute a major site of final assimilate deposition in grapefruit. However, levels of sucrose and glucose in segment epidermis were similar to those of the other tissues throughout development (Fig. 5-1, 5-2). In contrast, a disproportionate amount of 14C-photosynthates accumulated in the segment epidermis during time course studies (Koch, 1984a; unpublished data). Although the segment epidermis is not a major sink for photosynthates, this tissue may play an important role in the transport of photoassimilates into juice segments. Both sucrose and hexoses increased in albedo, segment epidermis and juice vesicles during development (Fig. 5-1, 5-2). The ratio of sucrose to hexoses, however, generally decreased in juice tissues and albedo during grapefruit development (Hilgeman and Smith, 1940). The presence of hexoses in these tissues does not necessarily suggest enzymatic sucrose cleavage. In juice tissues, for example, the pH often is such

PAGE 68

61 70 £ 50 30 10• ALBEDO (ALB) SEGMENT EPIDERMIS (SEG E) 1 JUNE j JULY | AUGUST j SEPT | OCT f MAY NOV Figure 5-1. Sucrose levels in albedo, segment epidermis and juice vesicles during development of grapefruit.

PAGE 69

62 160 120 80 40 10 • ALBEDO (ALB) SEGMENT EPIDERMIS (SEG E) o JUICE VESICLES (JV) MAY ] JUNE j JULY j AUGUST | SEPT j OCT f NOV Figure 5-2. Glucose levels in albedo, segment epidermis and juice vesicles during development of grapefruit.

PAGE 70

63 that sucrose hydrolysis can occur non-enzymatically and its rate increases with decreasing pH (Sinclair ,1984) The ratio of sucrose to hexoses also decreases during early grapefruit development as the juice pH decreases from approximately 5.5 to 3.0 (Caldwell, 1934). Sucrose cleavage in these juice tissues, therefore, may result at least partially from decreasing pH during development. Both soluble and insoluble (cell-wall bound) acid invertases (sucrose cleavage) were active in all three tissues of the immature grapefruit during the first three months of growth. Kato and Kubota (1978) found similar results for Satsuma juice vesicles. Activities of both forms of acid invertase in grapefruit were highest in the albedo extracts, followed by those from segment epidermis and juice vesicles. These activities decreased at least 807. by the fourth month of growth (stage II) and remained low through the remainder of fruit development. However, both dry matter and glucose per unit fresh weight continued to accumulate throughout the subsequent growing period (Fig. 5-1, 5-2). Only a slight decrease in glucose level (Fig. 5-2) coincided with the earlier decrease in soluble and insoluble acid invertase activity (Fig. 5-3a, 5-4, 5-5a). If the mechanism for sucrose import into citrus fruit involves acid invertase, then a mid-season drop in activity to minimal levels of this enzyme would be expected to decrease photosynthate accumulation. The concommittant slowing of assimilate import did not occur, so either the mechanism of photosynthate accumulation changed between stage I and II, or insoluble acid invertase was not essential to assimilate import. Insoluble acid invertase activity has been associated with phloem unloading in other species (Glaziou and Gaylor, 1972; Eschrich, 1980), but its action in grapefruit tissues also could be

PAGE 71

64 — i 200 gi tti ui • |l!0 > z > 5 100 ALBEDO (ALB) SEGMENT EPIDERMIS (8EQ E) JUICE VESICLES (JV) 8EQ E AUGUST SEPTEMBER I OCTOBER Figure 5-3. Activity of soluble acid invertase in albedo, segment epidermis and juice vesicles during development of grapefruit.

PAGE 72

65 90.0A ALB 80.0 SEG E 70.0 X10.0 + i z UJ 5.0 O cr a (D > O 5.0 LU UJ CO o o o O 3.0 + 2 Lr 1 1.0 in JV B ALB SEGE JV STAGE I ALB ALB SEG E JV r4^SEG_E^ ALB JV SEGE STAGE II SEGE JV ftjiiifi'iil ALB JV SEG E STAGE III Figure 5-4. Activity of partially purified enzyme in individual tissues during 3 stages in development of grapefruit. A. Soluble acid invertase activity. B. Soluble alkaline invertase activity.

PAGE 73

66 20.0 15.0 10.0 i r 5.0 + o > o LU UJ CO o u ID _i o _l o a. 3.0 2.0 1.0 SEGE ALB JV ALB SEG E jv ALB SEGE JV JV SEGE ALB ALB JV SEGE ALB SEG E JV STAGE I STAGE II STAGE III Figure 5-5. Activity of cell-wall bound enzyme in individual tissues during 3 stages in development of grapefruit. A. Insoluble acid invertase activity. B. Insoluble alkaline invertase activity.

PAGE 74

67 related to processes of cell division rather than assimilate partitioning. The high activity and cell-wall localization of insoluble acid invertase (Fig. 5-3a) during the cell division phase of fruit development are consistent with the suggestion that this enzyme may be involved in providing metabolites for cell wall precursors (Altman, et al., 1982; Morris and Arthur, 1985). A comparable situation appears to exist for soluble acid invertase. Data shown here indicate any involvement of this enzyme in the overall accumulation of photosynthates by sink tissues is either minor or occurs only during early fruit development (Fig. 5-4, 5-5a). Soluble acid invertase is believed to be compartmentalized primarily in vacuoles of immature storage parenchyma cells (Glasziou, 1962; Sacher et al.., 1963; Glasziou and Gaylor, 1972). Soluble alkaline invertase and sucrose synthetase were the only sucrose-metabolizing enzymes that remained active during the period when the rate of dry matter accumulation was greatest (Fig. 3-2, 5-5b, 5-6, 5-7). Soluble alkaline invertase trends appeared to be variable in the three tissues studied (5-5b). Activity of this enzyme did not change significantly during development if extracted from albedo, but a slight decrease had occurred in activity from extracts of segment epidermis by stage III. However, a significant rise in the activity of this enzyme took place when it was assayed from juice vesicles during stage II growth (Fig. 5-5b). A similar pattern was observed during development of juice vesicles in Satsuma (Kato and Kubota, 1978). The functional significance of soluble alkaline or neutral invertases in plant tissue are not well understood (Kato and Kubota, 1978). It has been suggested that multiple forms of alkaline invertase

PAGE 75

68 10.0 SEG E tr 5.0 x UJ 5 2.0 + d a. o O Q. D D _J< O 5 ALB JV ,0.30 0.10 /I In ^ SEG E r+ ALB (5 P JV k/1 M^ It 1 SEG E r s JV ALB £_ STAGE I STAGE II STAGE III Figure 5-6. Activity of sucrose synthetase assayed in the degradative direction in individual tissues during 3 stages of development of grapefruit.

PAGE 76

69 10.00SEGE SJ00-Z iu to g 2JDO030' 0.10-SEGE s ALB ALB MH JV SEGE ALB JV STAGE I STAGE STAGE III Figure 5-7. Activity of sucrose synthetase assayed in the synthetic direction in individual tissues during 3 stages of development of grapefruit.

PAGE 77

70 are responsible for sucrose cleavage in sweet potato (Matsushita and Uritani, 1974). A further possibility has been considered in mature sugar cane stalks where neutral soluble invertase, presumably a cytoplasmic enzyme, could couple sucrose hydrolysis to subsequent hexose phosphorylation followed by resynthesis of sucrose by sucrose phosphate synthetase (Glaziou and Gayler, 1972). Another consideration is that sucrose hydrolysis in the cytoplasm also may be involved in osmoregulation of storage cells. Cell turgor is important in the regulation of sucrose uptake and may be an additional determinant of sink strength in the sugar beet taproot (Wyse et al., 1986). Activities of soluble alkaline invertase from grapefruit are greatest when rates of accumulation are most rapid and thus, this enzyme may function in the regulation of sugar storage subsequent to phloem unloading. Insoluble (cell-wall bound) alkaline invertase followed the same trends observed for the soluble form, although developmental change of this activity in albedo was comparable to that of juice vesicles (Fig. 5-3b). Data of Kato and Kubota (1978) suggest much of the soluble and insoluble activity may be that of the same enzyme. Overall, activity of sucrose synthetase from grapefruit tissues decreased during development (Fig. 5-6, 5-7). This decrease in activity did not occur concurrently with the increase in glucose or sucrose accumulation during the period of study. The direction of the sucrose synthetase reaction, depending on compartmentalization, could either resynthesize sucrose hydrolyzed by alkaline invertase or provide substrates (UDPG) for a sucrose phosphate synthetase-group translocator complex of the tonoplast (Thorn et al., 1986). Activity of this enzyme was roughly similar when assayed in both cleavage and synthetic

PAGE 78

71 directions, but significantly lower values were obtained for the latter with extracts from albedo and juice vesicles of young fruit (stage I or II and stage I, respectively). This disparity may suggest a problem in the product determination procedure for UDP or the presence of an inhibitor in the protein extract from young fruit that most strongly affects activity in the synthetic direction. Relative activity of sucrose synthetase was greatest when extracted from segment epidermis and considerably less if from juice vesicles or albedo (Fig. 5-6, 5-7). Interestingly, activity of the enzyme from segment epidermis was many-fold greater that the other tissues and remained so through fruit development. Segment epidermis retained a disproportionately high percentage of 14C-photosynthates in transport studies when compared to albedo and juice vesicles (Koch, 1984a). Activity of sucrose phosphate synthetase extracted from all three tissues also tends to decrease during development, but increases slightly in the segment epidermis (data not shown). Overall, however, sucrose phosphate synthetase activity was minimal relative to that of sucrose synthetase in maturing and mature tissues. A near lack of sucrose phosphate synthetase and acid invertase, and the presence of alkaline invertase and sucrose synthetase activities during dry matter accumulation suggest the latter two enzymes may function in the regulation of photosynthate storage in grapefruit. Conclusions Five sucrose metabolizing enzymes were examined in conjunction with soluble sugar and dry weight accumulation during development of individual tissues in grapefruit. Only alkaline invertase and sucrose synthetase remained active during the period of most rapid increases in

PAGE 79

72 dry matter. The extent to which sucrose is hydrolyzed and resynthesized during import into citrus fruit is unclear, but the _in vitro capacity of these two enzymes is such that either could account for this process. The direction of the sucrose synthetase reaction, depending on compartmentalization, could either resynthesize sucrose hydrolyzed by alkaline invertase or provide substrate (UDPG) for a sucrose phosphate synthetase-group translocator complex of the tonoplast. Sucrose synthetase activity was highest in segment epidermis and remained relatively constant throughout development. These data, together with structural features (chapter IV) and radiolabeling studies (Koch, 1984a) suggest segment epidermis may play a major role in the transport of photosynthates into juice vesicles following phloem unloading. Sucrose in developing grapefruit also appears to be cleaved in processes not directly related to photosynthate import. A rapid decrease to minimal levels in the activity of soluble and insoluble acid invertase during the first 3 months of growth suggested acid invertases may be associated more closely with cell division than assimilate transfer. In addition, non-enzymatic sucrose hydrolysis can occur at low pH, and total hexose pools measured in the present study increased as juice pH decreased.

PAGE 80

CHAPTER VI OVERALL CONCLUSIONS Several aspects of vascular structure and distribution appear to be associated with assimilate translocation. First, individual vascular strands in stems are directly aligned with specific fruit juice segments. Second, the phloem to xylem ratio of vascular bundles increases with distance from their point of entry into the fruit. Third, the vast majority of assimilates entering the non-vascular juice segments must be unloaded from phloem into one of 3 vascular bundles outside the segment epidermis. The position and structure of these strands changes during development. An asymmetric proliferation of primary and secondary phloem occurs in dorsal vascular bundles, and a bundle cap of lignified fibers develops in the primary phloem. Differential rates of dry weight accumulation in individual tissues are consistant with these changes in positioning and relative amount of phloem. The longest component of the phloem-free transport path in grapefruit is the parenchymatous juice vesicle stalk. This elongates typically up to 16 mm in mature fruit. Cells are long and narrow-lumened at the stalk periphery, but short and wide-lumened in the center. Plasmodesmatal frequencies in mature stalk cells suggest a lack of specialization for symplastic transfer. Other parenchyma nearer the vascular bundles also remain non-specialized relative to assimilate transport. The segment epidermis does not appear to be directly in the path of photosynthate transfer; however, this cutinized, unito 73

PAGE 81

Ik biseriate cell layer is continuous with that of juice vesicles and is likely to limit photosynthate passage into these structures to points of juice vesicle stalk attachment to the carpel wall. Segment epidermal cells become highly vacuolated during development and their appearance shifts from that of metabolically active cells to one more similar to storage cells. Cells of mature juice vesicles, the final site of assimilate storage, have low cytoplasm to vacuole ratios and show no anatomical modification for secretion. Five sucrose metabolizing enzymes were examined in conjunction with soluble sugar and dry weight accumulation during development of individual tissues in grapefruit. Only alkaline invertase and sucrose synthetase remained active during the period of most rapid increases in dry matter. The extent to which sucrose is hydrolyzed and resynthesized during import into citrus fruit is unclear, but the _in vitro capacity of these two enzymes is such that either or both could account for this process. The direction of the sucrose synthetase reaction, depending on compartmentalization, could either resynthesize sucrose hydrolyzed by alkaline invertase or provide substrates (UDPG) for a sucrose phosphate synthetase-group translocator complex of the tonoplast. Sucrose synthetase activity was highest in segment epidermis and remained relatively constant throughout development. This data, together with structural features and radiolabeling studies (Koch, 1984a) suggest segment epidermis may play a major role in the transport of photosynthates into juice vesicles following phloem unloading. Sucrose in developing grapefruit also appears to be cleaved in processes not directly related to photosynthate import. The rapid

PAGE 82

75 decrease to minimal levels in the activity of soluble and insoluble acid invertase during the first 3 months of growth suggests acid invertases may be associated more closely with cell division than assimilate transfer. In addition, non-enzymatic sucrose hydrolysis can occur at low pH, and total hexose pools measured in the present study increased as juice pH decreased.

PAGE 83

APENDIX A TECHNICAL DATA FOR SUCROSE SYNTHASE IN TISSUES OF 'MARSH' GRAPEFRUIT Introduction Optimal conditions for assay of sucrose synthetase differ depending on the direction of the reaction (synthesis or cleavage). The optimum pH of the sucrose cleavage assay is 6.0-6.5 in a number of tissues (Hisajima, 1979; Gross and Pharr, 1982; Claussen, 1983a; Morell and Copeland, 1985). Soybean nodules have a broader range of pH 5.0-7.0 for assay of sucrose cleavage (Morell and Copeland, 1985), while in corn scutellum, greatest activity was measured from pH 6.0 to 8.5 (Echeverria, 1983). A pH range from 7.2-9.5 generally favors measurement of sucrose synthetase in the synthetic direction (Cardini et al., 1955; Rorem et al., I960; Avigad, 1964; Hawker et al., 1976; Hisajima, 1979; Huber, 1981; Gross and Pharr, 1982; Claussen, 1983a; Morell and Copeland, 1985). Activity was optimum from pH 8.5 to 10.0 in soybean nodules (Morell and Copeland, 1985) and pH 7.5 to 8.5 in Jerusalem artichoke tubers (Avigad, 1964). Two isozymes of sucrose synthetase were distinguished in tissues of cucumber fruit based on two optima, one at pH 7.5 and one at 9.3 (Gross and Pharr, 1982). This pH difference may have been an effect of different buffers, however, because synthetic activity in glycine-NaOH buffer was 25% higher than in tris-HCl buffer in eggplant vegetative and reproductive tissues (Claussen, 1983a) Divalent cations, heavy metals, end products, metabolites and other proteins affect sucrose synthetase activity. Divalent cations generally 76

PAGE 84

77 stimulate both synthesis and cleavage (Gross and Pharr, 1982; Morell and Copeland, 1985), although 8 mM Mg+2 and Mn+2 inhibited sucrose synthesis in Jerusalem artichoke tubers (Avigad, 1964). All activity was quite sensitive to inhibition by heavy metals (Morell and Copeland, 1985). Arsenate, arsenite, fluoride, iodoacetate, citrate and pyrophosphate have no affect on sucrose synthetase activity assayed in the synthetic direction (Cardini et al. 1955). Substrate inhibition of the synthetic reaction occurrs at fructose concentrations greater than 15 mM (Morell and Copeland, 1985). Sucrose synthesis and cleavage are both inhibited by various amounts of the end products UDP and sucrose, and UDPG and fructose, respectively (Gross and Pharr, 1982; Echeverria, 1983; Morell and Copeland, 1985). Metabolites such as glucose also inhibit sucrose synthetase activity when measured in either direction. In addition, isolated protein factors from wheat seeds are reported to stimulate sucrose synthesis, but inhibit cleavage activity by changing the affinity of sucrose synthetase for UDP (Pontis and Salerno, 1982). The inhibition of sucrose cleavage suggests the existence of a regulatory mechanism controlling the activity of sucrose synthetase in vivo The following parameters, pH, linearity with extract amount and time, product recovery and inhibition by phenyl-B glucoside, were tested on partially purfied extracts of sucrose synthetase to optimize methods for reaction incubation and product determination. Materials and Methods Materials and methods are described in chapter V. Buffers used for determinations of pH optima include 100 mM acetate and 200 mM Mes for sucrose cleavage at pH 4.0-7.0, and 200 mM Hepes and 200 mM tris for sucrose synthesis at pH 7.0-9.0. Linearity with quantity of enzyme

PAGE 85

78 extract was determined for aliquots between 100 and 200 uls (approximately 14.7-29.4 jig protein). Final extract volumes were adjusted to 200 ul with column buffer (see chapter V). Linearity of partially purified enzyme also was measured over time, from 15 to 60 min. For analysis of product recovery, 100 nmoles of the respective nucleotide end product was substituted for the nucleotide substrate. Assay procedure in both directions also was conducted with the addition of 3 nmoles phenyl-B-glucoside (a sucrose synthetase inhibitor) to the reaction medium in a final volume of 0.5 ml to ascertain that sucrose synthetase activity solely was responsible for measured changes. Results and Discussion Low levels of phenyl-B-glucoside completely inhibited sucrose synthetase when cleavage was assayed and inhibited 617. when sucrose synthesis was measured. Sucrose synthetase, therefore, is the primary reactant. Sucrose synthetase from 'Marsh' grapefruit tissue has an optimum of pH 5.0-6.0 when assayed in the degradative direction (Fig. A-l). In this assay, Mes buffer appears to stimulate sucrose synthetase activity at pH 5.5 by 2771. The pH optimum for the assay of sucrose synthetase in the synthetic direction is less clear, but also appears to be broad, from pH 8.0 to 9.0 (Fig. A-2). Tris buffer enhanced sucrose synthesis 36 and 397. at pH 8.0 and 8.5, respectively. Sucrose synthetase activity in both directions is linear with quantity of extract from 100 to 200 ul (Fig. A-3). Assays are linear with time for 45 or 60 min when measured in the synthetic and degradative directions, respectively (Fig. A-4). Product recovery for sucrose synthetase cleavage and synthesis activities was 93.127c _+ 0.019 and 55.847. + 0.075 after 15 min incubations, respectively. A low product

PAGE 86

79 ACETATE BUFFER MES BUFFER cc X o o O 0. 0.200.10 Figure A-l. Effect of pH on the cleavage activity of sucrose synthetase.

PAGE 87

80 TRIS BUFFER HEPES BUFFER 10.0 i GC X l z III O 6.0 GC a CL o O 2.0 7.0 8.0 PH 9.0 Figure A-2 Effect of pH on the synthesis activity of sucrose synthetase

PAGE 88

81 O a o -I o 5 c 50.030.0 10.0• 40.0 10.0100 150 UL EXTRACT 200 Figure A-3. Effect of extract amount on the activity of sucrose synthetase. A. Sucrose cleavage, r = 0.990. B. Sucrose synthesis, r^= 0.943.

PAGE 89

82 8.0| z ai O c Q. C 5 6 Gl o 3 -J O 5 D 6.0 4.0 2.0O 5 3.02.0 1.0 30 TIME (MIN) Figure A-4. Effect of time on the activity of sucrose synthetase, A. Sucrose cleavage, r = 0.994. B. Sucrose synthesis, r = 0.947 up to 45 min.

PAGE 90

83 recovery for the synthetic direction suggests some component of the enzyme preparation is using UDP as a substrate. Sucrose was included in the reaction medium, thus sucrose synthetase in the degradative direction is the probable cause. Preliminary data on sucrose synthetase cleavage at pH 8.5 in Hepes buffer suggest a low level of activity (data not shown) Parameters of pH, linearity and product recovery suggest the methods of _in vitro reaction and product determination are sufficient for a developmental study of sucrose synthetase activity in tissues of 'Marsh' grapefruit.

PAGE 91

APENDIX B TECHNICAL DATA FOR INVERTASE IN TISSUES OF 'MARSH' GRAPEFRUIT Introduction Invertase has been widely studied in a variety of lower and higher plant tissues and parameters for optimal in vivo activity have been established in these tissues. Optima of pH for soluble and insoluble invertase varies with the plant tissue in two general ranges, 2.5 to 6.8 (Gascon et al., 1965; Pressey, 1966; Sasaki et al., 1971; Klis et al., 1974; Chan et al. 1976; del Rosario et al., 1977; Kato and Kubota, 1978; Humphreys et al., 1980; Sum et al., 1980; Jacob et al., 1982; Prado et al. 19S2; Howard et al., 1983; Echeverria et al., 1984) and 7.0 to 7.7 (Matsushita et al., 1974; del Rosario et al., 1977; Kato and Kubota, 1978; Prado et al., 1982; Dey, 1986). Temperature optima are physiologically high at 40C (Sum et al., 1980) and 55C (Chan et al., 1975). Invertase inhibitors include a wide range of heavy metals, metallic and non-metallic ions, sugars (glucose-6-phosphate, glucose), tris buffer, lauryl sulfate, metasilicate and specific detergents (Arnold, 1965; Metzenberg, 1963; Kato et al., 1978; del Rosario et al., 1977; Matsushita et al., 1974). Stimulators of invertase include potassium and sodium nitrates and most phosphates and thiols (Jacob et al. 1982). Inhibitors and stimulators of invertase appear to vary widely with origin and form of the enzyme. 84

PAGE 92

85 Product recovery, pH optima and linearity with enzyme concentration and time were determined to optimize the invertase assay for use in a developmental study of invertase activity in 'Marsh' grapefruit tissues. Materials and Methods Materials, extraction and assay procedures are described in chapter V. The pH optima were determined with 300 mM acetate-potassium phosphate (dibasic) buffer in the pH range of 4.0 to 9.0. Soluble acid invertase from stage I albedo and soluble alkaline invertase from stage II segment epidermis were used to determine optimal pH between 4.0 and 9.0, and linearity with time up to 60 min in 'Marsh' grapefruit. Two replications were used for the soluble acid invertase pH curve. Duplicate analyses from stage I and III albedo were used to determine linearity with concentration of enzyme for 15 to 30 min assays of all forms of invertase. Recovery of added product to the reaction incubation was measured by adding 277.6 nmoles of glucose to assay medium with soluble acid invertase from stage I segment epidermis or soluble alkaline invertase from stage II segment epidermis. Water was substituted for sucrose in these reactions. Standard errors were determined from means of 2-3 replications. Results and Discussion Soluble invertases from grapefruit have specific pH optima at 4.5 and 7.5 for acid and alkaline forms, respectively (Figs. B-l, B-2). Activity of soluble acid invertase at pH 4.0 is 757. of the optimal and overlapping standard errors suggest a broad range of activity between pH 4.0 and 5.0. Similarily, soluble alkaline invertase activity at pH 7.0 and 8.0 is at least 707= of the maximum and, again, overlapping standard

PAGE 93

86 140.0 120.0• z UJ IO cc a. o > O HI LU CO o O Z) _l a _j o 3. 80.0 60.0 20.0 4.0 5.0 6.0 7.0 PH Figure B-l 4.0 to 7.0, Activity of soluble acid invertase from stage I albedo at pH

PAGE 94

8? CC X T 5.0 + LU 5 DC Q. O > 8 LU LU CO O o D -1 O _J o 3.0-1.0-6.5 7.0 7.5 8.0 8.5 9.0 PH Figure B-2. Activity of soluble alkaline invertase from stage II segment epidermis at pH 6.5 to 9.0.

PAGE 95

88 errors suggest a broad range of optimal activity between pH 7.0 and 8.0. Kato and Kubota (1978) also demonstrated similar broad pH ranges for optimal activity of soluble and insoluble acid invertases (pH 4.8-5.3) and alkaline invertases (pH 7.2-7.7) in orange juice vesicles. Assays were linear with concentration of enzyme between 25 and 250 ul enzyme and between 0.01 and 0.5 g of pellet (Figs. B-3, B-4). Soluble acid invertase activity also is linear with time to 60 min (Fig. B-5a). Soluble alkaline invertase activity is linear with time between 15 and 60 min, while linearity is not immediately apparent at times less than 15 min (Fig. B-5b). Linearity from zero time up to 60 min, however, is within the limits of the large standard errors obtained for this procedure. Product recovery for soluble acid and alkaline forms was 94% and 957„, respectively. Activities measured for soluble and insoluble, acid and alkaline forms of invertase in grapefruit tissues demonstrate linearity with enzyme amount and time at pH 4.5 and 7.5 and the conditions given for the reactions. Product recovery also is approximately 957. with the given conditions.

PAGE 96

89 1300 1000 500 ALBEDO, STAGE I (pH 4.5) ALBEDO, STAGE I (pH 7.5) • ALBEDO, STAGE IIKpH 7.5) 50.0 20.0 50 150 UL ENZYME EXTRACT 250 Figure B-3. Glucose equivalents produced during the soluble acid invertase assay at different concentrations of partially purified enzyme extract.

PAGE 97

9Q 1.50 ^ i.oot => O Ul HI CO o o => _J O O 0.50 30.10 • ALBEDO, STAGE I (pH 7.5) o ALBEDO, STAGE L (pH 5.0) 0.01 0.05 G PELLET 0.1 Figure B-A Glucose equivalents produced during the insoluble invertase assay at different amounts of pellet.

PAGE 98

9] O QC Q. O 5 o UJ UJ CO o o _l o -I o a. 100.0

PAGE 99

LITERATURE CITED ALBRIGO LG 1972 Distribution of stomata and epicuticular wax on oranges as related to stem end rind breakdown and water loss. J Amer Soc Hort Sci 97:220-223 ALTMAN A, Y GULSEN, R GOREN 1982 Growth and metabolic activity of lemon juice vesicle explants in vitro Plant Physiol 69:1-6 ANELUNXEN F, H ARBEITER 1967 Untersuchungen an den Spritzdrusen von Dictamnus albus L. Z Pf lanzenphysiol 58:49-69 ARNOLD WN 1965 B-fructo furanosidase from grape berries. Biochim Biophys Acta 110:134-147 ARISZ WH 1969 Intercellular polar transport and the role of the plasmodesmata in coleoptiles and Vallisneria leaves. Acta Bot Neerl 18:14-38 AVIGAD G 1964 Sucrose-uridine diphosphate glucosyltransf erase from Jerusalem artichoke tubers. J Biol Chem 239:3613-3618 BAIN JM 1958 Morphology, anatomical and physiological changes in the developing fruit of the Valencia orange, Citrus sinensis (L. ) Osbeck. Aust J Bot 6:1-28 BANERJI I 1954 Morphological and cytological studies on Citrus grandis Osbeck. Phytomorphology 4:390-396 BARTHOLOMEW ET, WB SINCLAIR 1941 Unequal distribution of soluble solids in the pulp of citrus fruits. Plant Physiol 16:293-312 BARTHOLOMEW ET, WB SINCLAIR 1951 The lemon fruit. University of California Press, Berkeley BEAN RC 1960 Carbohydrate metabolism of citrus fruits. I. Mechanisms of sucrose synthesis in oranges and lemons. Plant Physiol 35:429-434 BERGMEYER H 1965 In H BERGMEYER ed Methods of enzymatic analysis Academic Press, New York BIRDSALL JJ, PH DERSE, LJ TEPLY 1961 Nutrients on California lemons and oranges. II. Vitamin, mineral, and proximate composition. J Amer Dietet Assoc 38:555-559 BOLENDER RP 1978 Correlation of morphometry and stereology with biochemical analysis of cell fractions. Int Rev Cytol 55:247-289 92

PAGE 100

93 BOSABALIDIS A, I TSEKOS 1982a Ul trastructural studies on the secretory cavities of Citrus-Deliciosa Ten. 1. Early stages of the gland cells differentiation. Protoplasma 112:55-62 BOSABALIDIS A, I TSEKOS 1982b Ultrastructure studies on the secretory cavities of Citrus-Deliciosa Ten. 2. Development of the essential oil-accumulating central space of the gland and process of active secretion 112:63-70 BOUMA D 1959 The development of the fruit of the "Washington Navel" orange. Aust J Agr Res 10:804-817 BRADFORD MM 1976 A rapid and sensitive method for quantification of microgram quantities of protein utilizing the principle of protein dye-binding. Anal Biochem 72:248-254 BROWN GE, CR Barmore 1981 Ultrastructure of the response of citrus epicarp to mechanical injury. Bot Gaz 142:477-481 CAMERON SH 1933 Starch in the young orange tree. Proc Amer Soc Hort Sci 29:110-114 CARDINI CE, LF LELOIR, J CHIRIBOGA 1955 Biosynthesis of sucrose. J Biol Chem 214:149-156 CHAN HT, SCM KWOL, CWQ LEE 1975 Sugar composition and invertase activity in Lychee. J Food Sci 40::772-774 CHATTERTON NJ 1973 Product inhibition of photosynthesis in alfalfa leaves as related to specific leaf weight. Crop Sci 1:284-285 CLAUSSEN W 1983a Untersuchungen uber den Zusammenhang zwischen der Verteilung der Assimilate und der Saccharose-Synthetase-Aktivitat in Solanum melongena L. 1. Charakterisierung und Verteilung der Saccharose-Synthetase. Z Pf lanzenphysiol 110:165-173 CLAUSSEN W 1983b Untersuchengen uber den Zusammenhang zwuschen der Verteilung der Assimilate und der Saccharose-Synthetase-Aktivitat in Solanum melongena L. 2. Assimilatverteilung und Saccharose-Synthetase-Aktivitat. Z Pf lanzenphysiol 110:175-182 CLAUSSEN W, BR LOVEYS, JS HAWKER 1985 Comparative investigations on the distribution of sucrose synthase activity and invertase activity within growing, mature and old leaves of some C-3 and C-4 plant species. Physiol Plant 65:275-280 DEL ROSARIO EJ, V SANTISOPASRI 1977 Characterization and inhibition of invertases in sugar cane juice. Phytochem 16:443-445 DEY PM 1986 Changes in the forms of invertase during germination of mung bean seeds. Phytochem 25:51-53 DICK PS, T AP REES 1976 Sucrose metabolism by roots of Pisum sativum Phytochem 15:255-259

PAGE 101

94 DODD JD 1944 Three-dimensional cell shape in the carpel vesicles of Citrus grandis Amer J Bot 31:120-127 ECHEVERRIA E 1983 Regulation of sucrose metabolism in the maize scutellum. Dissertation, University of Florida ECHEVERRIA E, T HUMPHREYS 1984 Involvement of sucrose synthase in sucrose catabolism. Phytochem 23:2173-2178 ELEFTHERIOU EP, JL HALL 1983 The extrafloral nectaries of cotton. 1. Fine structure of the secretory papillae. J Exp Bot 34:105-119 ESCHRICH W 1980 Free space invertase, its possible role in phloem unloading ( Monstera deliciosa and other plants including forest trees). Ber Dtsch Bot Ges 93:363-378 ESPELIE KE, RW DAVIS, PE KOLATTUKUDY 1980 Composition, ultrastructure and function of the cutinand suberin-containing layers in the leaf, fruit peel, juice-sac and inner seed coat of grapefruit ( Citrus paradisi Macfed.). Planta 149:498-511 EVERT RF, W ESCHRICH, W HEYSER 1977 Distribution and structure of the plasmodesmata in mesophyll and bundle-sheath cells of Zea mays L. Planta 136:77-89 FAHN A 1979 Secretory tissues in plants. Academic Press, New York. FAHN A, I SHOMER, I BEN-GERA 1974 Occurrence and structure of epicuticular wax on the juice vesicles of citrus fruits. Ann Bot 38:869-872 FELKER CF, JC SHANNON 1980 Movement of 14C-labeled assimilates into kernels of Zea mays L. III. An anatomical examination and microautoradiographic study of assimilate transfer. Plant Physiol 65:864-870 FISHER DG, RF EVERT 1962 Studies on the leaf of Amaranthus retrof lexus (Amaranthaceae) : ultrastructure, plasmodesmatal frequency, and solute concentration in rrelation to phloem loading. Planta 155:377-387 FISHER DB, WH OUTLAW 1979 Sucrose compartmentation in the palisade parenchyma of Vicia faba L. Plant Physiol 64:563-568 FORD E 1942 Anatomy and histology of the Eureka lemon. Bot Gaz 104:288-305 FREEMAN B, LG ALBRIGO, RH BIGGS 1979 Ultrastructure and chemistry of cuticular waxes of developing citrus leaves and fruits (oranges, tangerines, lemons). J Ann Soc Hort Sci 104:801-808 GAMALEI YV 1985 Plasmodesmata: intercellular communication in plants. Sov Plant Physiol 32:134-146

PAGE 102

95 GASCON S, NP NEUMANN, JO LAMPEN 1968 Comparative study of the properties of the purified internal and external invertases from yeast. J Biol Chem 7:1573-1577 GE1GER DR 1976 Effects of translocation and assimilate demand on photosynthesis. Can J Bot 54:2337-2345 GEIGER DR, BR FONDY 1980Phloem loading and unloading: pathways and mechanisms. What's New Plant Physiol 11:25-28 GIAQUINTA RT 1979 Sucrose translocation and storage in the sugar beet. Plant Physiol 63:828-832 GIAQUINTA RT, W LIN, NL SADLER, VR FRANCESCHI 1983 Pathway of phloem unloading of sucrosein corn roots. Plant Physiol 72:362-367 GLASZIOU KT 1962 Accumulation and transformation of sugars in sugar cane stalks: mechanism of inversion of sucrose in the inner space. Nature 193:1100 GLASZIOU KT, KR GAYLER 1972 Sugar accumulation in sugarcane. Role of cell walls in sucrose transport. Plant Physiol 49:912-913 GOLDSCHMIDT EE, N ASCHKENAZI, Y HERZANO, AA SCHAFFER, SP MONSELISE 1985 A role for carbohydrate levels in the control of flowering in citrus. Sci Hort 26:159-166 GRAHAM LL, MA JOHNSON 1978 Sucrose synthetase from triploid quaking aspen callus. Phytochem 17:12331-1233 GROSS J, R TIMBERG, M GRAFF 1983 Pigment and ultrastructural changes in the developing pummelo Citrus grandis 'Goliath'. Bot Gaz 144:401-406 GROSS KC, DM PHARR 1982 Cucumber fruit sucrose synthase isozymes. Phytochem 21:1241-1244 GUARDIOLA JL, F GARCIA-MARI, M AGUSTI 1984 Competition and fruit set in the Washngton navel orange. Physiol Plant 62:297-302 GUNNING BES 1978 Age-related and origin-related control of the numbers pf plasmodesmata in cell walls of developing Azolla roots. Planta 143:181-190 GUNNING BES, JE HUGHES 1976 Quantitative assessment of symplastic transport of pre-nectar into the trichomes of Abutilon nectaries. Aust J Plant Physiol 3:619-637 GUNNING BES, RL OVERALL 1983 Plasmodesmata and cel-to-cell transport in plants. Bioscience 33:260-265 GUY CL, G YELENOSKY, HC SWEET 1981 Distribution of 14C photosynthetic assimilates in 'Valencia' orange seedlings at 10 and 25 C. J Amer Soc Hort Sci 106:433-437

PAGE 103

96 HABESHAW D 1973 Translocation and the control of photosynthesis in sugar beet. Planta 110:213-226 HARDHAM AR 1976 Structural aspects of the pathways of nutrient flow to the developing embryo and cotyledons of Pisum sativum L. Aust J Bo. 24:711-721 HARRIS N, KJ OPARKA, DJ WALKER-SMITH 1982 Plasmatubules: an alternative to transfer cells? Planta 156:461-465 HARVEY EM, GL RYGG 1936 Field and storage studies on changes in the composition of the rind of Marsh grapefruit in California. J Agr Res 52:747-787 HATCH MD 1963 A uridine diphosphatase from sugar cane. Biochem J 88:423-427 HAWKER JS 1965 Sucrose. In PM DEY, RA DIXON, eds, Biochemistry of storage carbohydrates in green plants. Academic Press, London pp 1-51 HAWKER JS, MD HATCH 1965 Mechanism of sugar storage by mature stem tissue of sugarcane. Physiol Plant 18:444-453 HAWKER JS, RR WALKER, HP RUFFNEN 1976 Invertase and sucrose synthase in flowers. Phytochem 15:1441-1443 HAYAT MA 1981 Principles and techniques of electron microscopy. Biological applications. Vol 1 University Park Press, New Jersey HAYES PM, CE OFFLER, JW PATRICK 1985 Cellular structures, plasma membrane surface areas and plasmodesmtal frequencies of the stem of Phaseolus vulgaris L. in relation to radial photosynthate transfer. Ann Bot 56:125-138 HELDER RJ, J BOERMA 1969 An electron-microscopical study of the plasmodesmata in the roots of young barley seedlings. Acta Bot Neerl 18:99-107 HENDRIX DL, SC HUBER 1986 Diurnal fluctuations in cotton leaf carbon export, carbohydrate content, and sucrose synthesizing enzymes. Plant Physiol 81:584-586 HILGEMAN RH, JA DUNLAP, GC SHARPLES 1966 Effect of time of harvest of Valencia oranges on leaf carbohydrate content and subsequent set of fruit. Amer Soc Hort Sci 90:110-115 HILGEMAN RH, JG SMITH 1940 Changes in invert sugar and sucrose during ripening of Arizona grapefruit. Proc Amer Soc Hort Sci 37.535 538 HISAJIMA S 1979 Studies on the sucrose metabolism in callus cells of Japanese morning-glory ( Pharbitis Nil L. ) Memoirs Tokyo Univ Agric 21:1-54

PAGE 104

97 HO LC, DA BAKER 1982 Regulation of loading and unloading in long distance transport systems. Physiol Plant 56:225-230 HOLBEIN BE, CW FORSBERG, DK KIDBY 1976 A modified procedure for studying enzyme secretion in yeast sphaeroplasts : subcellular distribution of invertase. Can J Microbiol 22:989-995 HOLTZHAUSEN LC 1969 Observations on the developing fruit of Citrus sinensis cultivar Washington Navel from anthesis to ripeness. Dept Agr Tech Serv South Afr Tech Commun No 91, pp 1-15 HOWARD HF, FH WITHAM 1983 Invertase activity and the kinetin-stimulated enlargement of detached radish cotyledons. Plant Physiol 73:304-308 HUBER SC 1981 Interspecific variation in activity and regulation of leaf sucrose-P-synthetase. Z Pf lanzenphysiol 102:443-450 HUMPHREYS T, E ECHEVERRIA 1980 Invertase and maltase in the free space of the maize scutellum. Phytochem 19:189-193 JACOB JL, JC PREVOT, J D'AUZAC 1982 Physiological activators of invertase from Hevea brasiliensis latex. Phytochem 21:851-853 JENNER CF 1974 Factors in the grain regulating the accumulation of starch. Roy Soc N Z Bull 12:901-908 JENNER CF 1980 The conversin of sucrose to starch in developing fruits. Ber Dtsch Bot Ges 93:289-298 JOHANSEN DA 1940 Plant microtechnique. McGraw-Hill Book Co, New York JUNIPER BE 1977 Some speculations on the possible roles of the plasmodesmata in the control of differentiation. J Theor Biol 66:583-592 KARNOVSKY MJ 1965 A f ormaldehyde-glutaraldehyde fixative of high osmolarity for use in electron microscopy. J Cell Biol 27:137A KATO T, S KUBOTA 1978 Properties of invertases in sugar storage of citrus fruit and changes in their activities during maturation. Physiol Plant 43:67-72 KEFFORD JF, BV CHANDLER 1970 The chemical constituents of citrus fruits. Academic press, New York KING GS 1947 Peripheral deposits of citrus fruit vesicles stained by oil-soluble dyes. Amer J Bot 34:427-431 KLIS FM, R DALHUIZEN, L SOL 1974 Wall-bound enzymes in callus of Convolvulus arvensis. Phytochem 13:55-57 KLOTZ LJ, ARC HAAS 1933 Some differences between button blossom halves of citrus fruits. Calif Citrograph 18:318, 324

PAGE 105

98 KOCH KE 1984a The path of photosynthate translocation into citrus fruit. Plant, Cell and Environ 7:647-653 KOCH KE 1984b Translocation of photosynthetic products from source leaves to aligned juice segments in citrus fruit. Hort-Science 19:260-261 KOCH KE 1985 Nonvascular transfer of assimilates in citrus juice tissues. In RL HEATH, J PRE1SS, eds, Regulation of Carbon Partitioning in Photosynthetic Tissue. Waverly Press, Baltimore, Maryland, pp 362-366 KOCH KE, WT AVIGNE 1984 Localized photosynthate deposition in citrus fruit segments relative to source-leaf position. Plant and Cell Physiol 25:859-860 KORDAN HA 1964 Vascular elements in juice vesicles of the lemon fruit. Bull Torrey Bot Club 91:271-274. KORDAN HA 1971a Starch synthesis in lemon fruits injected with sugar solutions. Z Pflanzen Physiol 65:183-187 KORDAN HA 1971b Starch synthesis in quiescent lemon fruit explants. Z Pflanzen Physiol 65:118-123 KUO J, J S PATE 1985 Unusual network of internal phloem in the pod mesocarp of cowpea [ Vigna unguiculata (L.) Walp. (Fabaceae)]. Ann Bot 55:635-647 LIMA J 1983 Navel orange fruit drop: secondary-fruit ontogeny, physiological studies, and growth regulator effects. Dissertation, Univ Florida, Gainesville, Florida LIU P, DH WALLACE, JL OZBUN 1973 Influence of translocation on photosynthetic efficiency of Phaseolus vulgaris L. Plant Physiol 52:412-415 MANN HB, DR WHITNEY 1947 On a test of whether one of two random variable is stochastically larger than the other. Ann Math Statist 18:50-60. MATSUSHITA K, I URITANI 1974 Change in invertse activity of sweet potato in response to wounding and purification and properties of its invertases. Plant Physiol 54:60-66 MEYBERG M, U KRISTEN 1981 The nectaries of Aptenia cordifolia ultrastructure, translocation of 14C-labelled sugars, and a possible pathway of secretion. Z Pf lanzenphysiol 104:139-147 METZENBERG RL 1963 The purification and properties of invertase of Neurospora. Arch Biochem Biophys 100:503-511 M0RELL M, L COPELAND 1984 Enzymes of sucrose breakdown in soybean nodules. Plant Physiol 74:1030-1034

PAGE 106

99 MORELL M, L COPELAND 1985 Sucrose synthase of soybean nodules. Plant Physiol 78:149-154 MORRIS D, E ARTHUR 1985 Invertase activity, carbohydrate metabolism and cell expansion in the stem of Phaseolus vulgaris L. J Exp Bot 36:623-633 NELSON N 1944 A photometric adaptation of the Somogyi method for the determination of glucose. J Biol Chem 153:357-380 OFFLER CE, JW PATRICK 1984 Cellular structures, plasma membrane surface areas and plasmodesmatal frequencies of seed coats of Phaseolus vulgaris L. in relation tophotosynthate transfer. Aust J Plant Physiol 11:79-99 OPARKA K L, P GATES 1981 Transport of assimilates in the eveloping caryopsis of rice ( Oryza sativa L.). Ultrastructure of the pericarp vascular bundle and its connections with the aleurone layer. Planta 151:561-573 OPARKA KJ, PJ GATES 1982 Ultrastructure of the developing pigment strand of rice (Oryza sativa L. ) in relation to its role in solute transport. Protoplasma 113:33-43 ORR PM, DP KNIEVEL, JC SHANNON 1981a Compartmental analysis of sugars in the transport tissues of developing maize kernels. Plant physiol 67S:716 ORR PM, DP KNIEVEL, JC SHANNON 1981b Measurement of pedicel cellular and freespace compartmentation of sugars in maize hybrids differing in kernel growth rates. Plant Physiol 6 7S : 71 7 PATE JS, BES GUNNING 1972 Transfer cells. Ann Rev Plant Physiol 23:173-196 PATRICK JW 1983 Photosynthate unloading from seed coats of Phaseolus vulgaris L. General characteristics and facilitated transfer. Z Pf lanzenphysiol 111:9-18 PERCHERON F 1962 Dosage colorimetrique du fructose et des f ructof uranosides par l'acide thiobarbiturique. Compt Rend 255:2521-2522 PONTIS HG, GL SALERNO 1982 Inhibition of sucrose synthetase cleavage activity by protein factors. FEBS Letters 141:120-123 POWELL AA, AH KREZDORN 1977 Influence of fruit-setting treatment on translocation of 14C-metabolites in citrus during flowering and fruiting. J Amer Soc Hort Sci 102:709-714 PRADO FE, OL FLEISCHMACHER, MA VATTUONE, AR SAMPIETRO 1982 Cell wall invertases of sugar cane. Phytochem 21:2825-2828

PAGE 107

100 PRESSEY R 1966 Separation and properties of potato invertase and invertase inhibitor. Arch Biochem Biophys 113:667-674 PURVIS A, J RICE 1983 Low temperature induction of invertase activity in grapefruit flavedo tissue. Phytochem 22:831-834 ROBARDS AW 1976 Plasmodesmata in higher plants; studies on plasmodesmata. In BES Gunning, AW Robards, eds, Intercellular communications in plants; studies on plastnodesmata. Springer, Berlin, pp 15-57 ROBINS RJ, BE JUNIPER 1980 Considerations of the estimation of plasmodesmatal frequencies. J Theor Biol 83:405-409 ROREM ES, HG WALKER, RM MCCREADY 1960 Biosynthesisof sucrose and sucrose-phosphate by sugar beet leaf extracts. Plant Physiol 35:269-272 ROTH I 1977 Fruits of angiosperms. Gebruder Borntrager, Berlin, pp 494-516 ROTH, I, H LINDORF 1972 Desarrollo y anatomia del fruto y de la semilla de Citrus Acta Bot Venez 7:163-186 ROY WR 1945 Effect of potassium deficiency and of potassium derived from different sources on the composition of the juice of Valencia oranges. J Agr Res 70:143-169 SACHER JA, MD HATCH, KT GLASZIOU 1963 Regulation of invertase synthesis in sugar cane by an auxinand sugar-mediated control system. Physiol Plantarum 16:836-842 SASAKI T, K TAD0K0R0, S SUZUKI 1971 Multiple forms of invertase of potato tuber stored at low temperature. Phytochem 10:2047-2050 SAUTER JJ, HJ BRAUN 1972 Cytochemische Untersuchung der Atmungsativitat in den Strasburger-Zellin von Larix und ihre Bedeutung fur den Assimilattransport. Z Pf lanzenphysiol 66:440-458 SCHAFFER AA, EE GOLDSCHMIDT, R GOREN, D GALILI 1985 Fruit set and carbohydrate status in alternate and nonalternate bearing Citrus cultivars. J Amer Soc Hort Sci 110:574-578 SCHNEIDER H 1968 The anatomy of citrus. In: W Reuther, LD Batchelor, HJ Webber, eds, The citrus industry. Vol II Univ Cal Press, Berkeley, pp 1-85 SCHNEPF E, A SYCH 1983 Distribution of plasmodesmata in developing Sphagnum leaflets. Protoplasma 116:51-56 SCHULMAN Y, SP MONSELISE 1970 Some studies of the cuticular wax of citrus fruits. J Hort Sci 45:471-478 SCOTT EM, KC BAKER 1947 Anatomy of Washington navel orange rind in relation to water spot. Bot Gaz 108:495-475

PAGE 108

101 SEAGULL RW 1983 Differences in the frequency and disposition of plasmodesmata resulting from root cell elongation. Planta 159:497-504 SHARPLES GC, L BURKHART 1953 Seasonal changes in carbohydrates in the Marsh grapefruit tree in Arizona. Amer Soc Hort Sci 63:74-80 SHANNON JC, TC DOUGHERTY 1972 Movement of 14C-labelled assimilates into kernels of Zea mays L. II. Invertase activity of the pedicel and placento-chalazal tissues. Plant Physiol 49:203-206 SHOMER I, I BEN-GERA, A FAHN 1975 Epicuticular wax on the juice sacs of citrus: a possible adhesive in the fruit segments. J Food Sci 40:925-930 S1LVIUS JE, FW SNYDER 1979 Comparative enzymic studies of sucrose metabolism in the taprooots and fibrous roots of Beta vulgaris L. Plant Physiol 64:1070-1073 SINCLAIR WB 1972 The grapefruit: its composition, physiology, and products. University of California, California SINCLAIR WB 1984 The biochemistry and physiology of the lemon. USA Division of Agriculture and Natural Resources, Oakland, California SPURR AR 1969 A low-viscosity epoxy resin embedding medium for electron microscopy. J Ultrastruct Res 26:31-43 STEPAK Y, A LIFSHITZ 1971 Identification and determination of sugars in some fruit juices. JA0AC 54:1215-1217 SU JC, JL WU, CL YANG 1977 Purification and characterization of sucrose synthetase from the shoot of bamboo Leleba oldhami Plant Physiol 60:17-21 SUM WF, PJ ROGERS, ID JENKINS, RD GUTHRIE 1980 Isolation of invertase from banana fruit ( Musa Cavendishii ) Phytochem 19:399-401 SYVERTSEN JP, LG ALBRIG0 1980 Some effects of grapefruit tree canopy position on microclimate, water relatins, fruit yield, and juice quality. J Amer Soc Hort Sci 105:454-459 THOM M, RA LEIGH, A MARETZKI 1986 Evidence for the involvement of a UDP-glucose-dependent group translocator in sucrose uptake into vacuoles of storage roots of red beet. Planta 167:410-413 THOMSON WW 1965 Observations on the ultrastructure of the plasmalemma in oranges. J Ultrastructure Res 16:640-650 THOMSON WW 1966a Observations on the ultrstructure of the plasmalemma in oranges. J Ultrastructure Res 16:640-650

PAGE 109

102 THOMSON WW 1966b Ultrastructural development of chromoplasts in Valencia oranges. Bot Gaz 127:133-139 THOMSON WW 1969 Ultrastructural studies on the epicarp of ripening oranges. Proc First International Citrus Symposium 3:1163-1169 THORNE JH 1979 Redistribution of assimilates from soybean pod walls during seed development Agron J 71:812-816 THORNE JH 1980 Kinetics of 14C-photosynthate uptake by developing soybean fruit. Plant Physiol 65:975-979 THORNE JH 1981 Morphology and ultrastructure of maternal seed tissues of soybean in relation to the import of photosynthate. Plant Physiol 67:1016-1025 TILLSON AH, R BAMFORD 1938 The floral anatomy of the Aurantioideae. Amer J Bot 25:780-793 TING SV 1969 Distribution of soluble components and quality factors in the edible portion of citrus fruits. J Amer Soc Hort Sci 94: 515-519 TING SV, EJ DESZYCK 1961 The carbohydrates in the peel of oranges and grapefruit. J Food Sci 26:146-152 TURRELL FM, ET BARTHOLOMEW 1939 Structural and microchemical changes in granulated orange vesicles. Calif Citrog 24:88,110 TURRELL FM, LJ KLOTZ 1940 Density of stomata and oil glands and incidence of water spot in the rind of Washington navel orange. Bot Gaz 101:862-871 WARMBRODT RD 1965 Studies on the root of Zea mays L. Structure of the adventitious roots with respect to phloem unloading. Bot Gaz 146:169-160 WEIBEL ER 1969 Stereological principles for morphometry in electron microscopic cytology. Int Rev Cytol 26:235-302 WIDDOWSON EM, RA MCCANCE 1935 The available carbohydrates of fruits. Determination of glucose, fructose, sucrose, and starch Biochem J 29:151-156 WOLSWINKEL P 1965 Phloem unloading and turgor-sensitive transport: factors involved in sink control of assimilate partitioning. Physiol Plant 85:274-283 WYSE R 1979 Sucrose uptake by sugar beet taproot tissue. Plant Physiol 61:172-182

PAGE 110

103 WYSE RE, E ZAMSKI, AD TOMOS 1986 Turgor regulation of sucrose transport in sugar beet taproot tissue. Plant Physiol 81:478-481 YELENOSKY G, CL GUY 1977 Carbohydrate accumulation in leaves and stems of 'Valencia' orange at progressively colder temperatures. Bot Gaz 138:13-17

PAGE 111

BIOGRAPHICAL SKETCH Cadance Ann Lowell was born in Washington DC, on 15 May, 1957. She attended public schools in Chevy Chase, Maryland, and graduated from Bethesda-Chevy Chase High School in 1975. In 1978, she received a Bachelor of Science degree in botany from Duke University. She was admitted to the Graduate School at the University of Florida in September, 1979 and received a Master of Science in plant anatomy in August, 1982. While pursuing her work toward this degree, she held a teaching assistantship in the Department of Botany, and also was employed part-time as a science aide in the Environmental Protection Agency and a lab assistant in the Departments of Fruit Crops and Astronomy. In July, 1982, she received a research assistantship and began work toward a Doctor of Philosophy degree in the Department of Fruit Crops at the University of Florida. She was awarded the Ph.D. in December ,1986. She is currently a member of the American Society of Plant Physiologists, Botanical Society of America, American Society for Horticultural Science and Sigma Xi. She is married to James William Williams and has one son, Taylor Lowell Williams. 104

PAGE 112

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Kai/en E. Koch, Chairman Associate Professor of Horticultural Sciences I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. c *L c>-t£^-Jaflies Soule Professor Emeritus of Horticultural Sciences I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. ;V\ ,— Thomas Humphreys t Professor of Horticultural Sciences I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Hilton Biggs Professor of Horticultural Sciences

PAGE 113

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. istructor Microbiology and Cell Science This dissertation was submitted to the Graduate Faulty of the College of Agriculture and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December, 1986 JLrtA^iAW of Agriculture Dean, Graduate School

PAGE 114

UNIVERSITY OF FLORIDA 3 1262 08553 4203