Citation
Pollination and GA3 effects on fruit growth, sucrose metabolism, cell number, and cell size of blueberry fruits

Material Information

Title:
Pollination and GA3 effects on fruit growth, sucrose metabolism, cell number, and cell size of blueberry fruits
Creator:
Cano Medrano, Raquel
Publication Date:
Language:
English
Physical Description:
xi, 149 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Blueberries ( jstor )
Cell growth ( jstor )
Cells ( jstor )
Fruit set ( jstor )
Fruiting ( jstor )
Fruits ( jstor )
Mesocarp ( jstor )
Plants ( jstor )
Pollen ( jstor )
Pollination ( jstor )
Dissertations, Academic -- Horticultural Science -- UF
Horticultural Science thesis, Ph. D
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1994.
Bibliography:
Includes bibliographical references (leaves 131-148).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
Raquel Cano Medrano.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
021587384 ( ALEPH )
33379741 ( OCLC )

Downloads

This item has the following downloads:


Full Text










POLLINATION AND GA3 EFFECTS ON FRUIT GROWTH, SUCROSE METABOLISM, CELL NUMBER, AND CELL SIZE OF BLUEBERRY FRUITS














By

RAQUEL CANO MEDRANO


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


1994













ACKNOWLEDGEMENTS


Special thanks go to Dr. Rebecca Darnell, my advisor, for her guidance and support. I was lucky to work under her supervision. I also thank the rest of the members of my committee, Drs. W. B. Sherman, K. E. Koch, D. J. Huber, and K. J. Boote.

I extend my appreciation to Steve Hiss for his friendship and technical support, specially with the photographs.

I enjoyed sharing this time with Alejandra Guti~rrez, Don Merhaut, and Kurt Nolte. I also thank Ma. Elena Su~rez, Carlos Garcfa, and Carmen Gispert for their friendship and moral support.

Finally, sincere thanks go to my parents, Elisa and Bony; my husband, Jorge; my brother and sisters, J. Domingo, Olivia, Nelly, Irma, and Norma for for their love and unconditional support.













TABLE OF CONTENTS


ACKNOWLEDGEMENTS ...........


LIST OF TABLES ........................


LIST OF FIGURES ..............


ABSTRACT ...................................


CHAPTERS


1 INTRODUCTION ......


2 REVIEW OF LITERATURE .............


Flower Development .............
Pollination and Fertilization ......... Fruit Growth and Development ......
Endogenous Growth Regulators
Exogenous Growth Regulators
Effect of Exogenous Growth Regulators


Sink Tissues ..........
Sink Size .........
Sink Activity ......


3 EFFECT OF GA3 AND POLLINATION ON FRUIT SET .


..... 31


Introduction ....... Materials and Methods
Experiment I
Experiment II .
Experiment III
Results and Discussion
Experiment I .
Experiment II .
Experiment III
Conclusions .......


V


. . . . . . . . . . . . . . . . . . . . v i


. . . . . . X


. . . . . . . . . . . . . . . . . 1


...........
...........
...........













4 CARBOHYDRATE METABOLISM AND FRUIT GROWTH IN
GIBBERELLIC ACID-TREATED VS POLLINATED FRUITS


OF BLUEBERRY


. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 8


Introduction ..................
Materials and Methods ...........
Plant material ............
Treatments ..............
Fruit carbon dioxide exchange Calculations of carbon budget
Sugar determination ........
Enzyme extraction .........
Enzyme assays ...........
Results .....................
Fruit growth .............
Fruit carbon dioxide exchange .
Fruit carbon budget ........
Carbohydrate accumulation ...
Enzyme activity ...........
Discussion ....................
Fruit growth and carbon budget


Carbohydrate accumulation and sucrose enzym e activities ........................


5 CELL NUMBER AND CELL SIZE IN GIBBERELLIC ACIDTREATED VS POLLINATED FRUITS OF BLUEBERRY . .


Introduction ....... Materials and Methods Results ..........
Fruit growth . .
Cell number and Discussion ........


.....�......o...


�.....o...oo.....
cell size ...........
..............


6 SUM M ARY ..............................

LITERATURE CITED ............................

BIOGRAPHICAL SKETCH .........................


.... 91

.... 91
.... 93
.... 96
.... 96
.... 97
... 101

... 127

... 131

... 149














LIST OF TABLES


Table Paae

3-1 Effect of GA3 on fruit set and development in 'Beckyblue'
rabbitye blueberry in 1992 ........................ 55

3-2 Effect of GA3 on fruit set and development in 'Beckyblue'
rabbiteye blueberry in 1993 ........................ 56

3-3 Effect of GA3, pollen source, and pollen amount on fruit
fresh weight (FW) and seed number/seed weight in
'Beckyblue' rabbiteye blueberry ..................... 57

4-1 Estimated C Budget for developing GA3-treated, pollinated,
and nonpotlinated 'Beckyblue' rabbiteye blueberry fruits . . . 90

5-1 Pericarp cell size increases in whole cross-sectional area of 'Beckyblue' rabbiteye blueberry fruits .......... 126













LIST OF FIGURES


Figure Page

3-1 Time-course and extent of fruit abscission in 'Beckyblue'
rabbiteye blueberry as affected by gibberellic acid (GA3), pollination (POLL), or nonpollination (NP). 3x=3 applications of GA3; application times indicated by arrows.
Values are means � SE, n =3; SE bars present only when
larger than symbol ............................ 50

3-2 Time-course and extent of fruit abscission in 'Beckyblue'
rabbiteye blueberry as affected by gibberellic acid (GA3), pollination (POLL), or nonpollination (NP). 2x and 4x= 2 and 4 applications of GA3; application times indicated by arrows. Values are means � SE, n =3; SE bars present only
when larger than symbol ......................... 51

3-3 Time-course and extent of fruit abscission in 'Beckyblue'
rabbiteye blueberry as affected by gibberellic acid (GA3), pollination (POLL), or nonpollination (NP). 2x and 4x = 2 and 4 applications of GA3; application times indicated by arrows. Values are means � SE, n =9; SE bars present only
when larger than symbol ......................... 52

3-4 Time-course and extent of fruit abscission in 'Beckyblue'
rabbiteye blueberry as affected by gibberellic acid (GA3), pollen source, and pollen amount. X = cross-pollination (200 pollen tetrads); 1/4X = 50 pollen tetrads; S = self-pollination (200 pollen tetrads). GA3=0.7 mM GA3; GA3 application times indicated by arrows. Values are means � SE, n =9; SE
bars present only when larger than symbol ............ 53

3-5 Effect of GA3, pollen source, and pollen amount on final
fruit set of 'Beckyblue' rabbiteye blueberry fruits.
X=cross-pollination (200 pollen tetrads); 14X= cross pollination (50 pollen tetrads); S=self-pollination (200 pollen tetrads); GA3 =0.7 mM GA3 applied at FB and FB+7








DAFB. Values are means �SE, n =9; SE bars present only
when larger than symbol ......................... 54

4-1 Changes in dry weight (A) and fresh weight (B) in GA3treated (GA3), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means � SE, n =3; SE bars present only when larger than symbol. Arrow indicates when abscission of nonpollinated
fruits occurred .................................. 82

4-2 Relative growth rate (RGR) of GA3-treated (GA3),
pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Mean � SE, n = 3; SE bars present
only when larger than symbol ...................... 83

4-3 Net carbon dioxide exchange in GA3-treated (GA3),
pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means � SE, n = 3; SE
bars present only when larger than symbol ............ .84

4-4 Daily estimated C cost in GA3-treated (GA3) and pollinated
(POLL) 'Beckyblue' rabbiteye blueberry fruits. Values were calculated from data in Figures 4-1 and 4-3 and are based on a 12 h photoperiod and 25/210C light/dark
tem peratures ................................ 85

4-5 Daily estimated C supply in GA3-treated (GA3) and pollinated
(POLL) 'Beckyblue' rabbiteye blueberry fruits. Values were calculated from data in Figures 4-1 and 4-3 and are based
on a 12 h photoperiod and 25/21�C light/dark
tem peratures ................................ 86


4-6 Hexose accumulation in GA3-treated (GA3), pollinated
(POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means�SE, n=3; SE bars
present only when larger than symbol ............... 87

4-7 Activities of (A) sucrose phosphate synthase (SPS) and (B)
sucrose synthase (SS) in GA3-treated (GA3), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means�SE, n=3; SE bars
present only when larger than symbol ............... 88









4-8 Activities of (A) soluble acid and (B) insoluble acid
invertases in GA3-treated (GA3), pollinated (POLL), and nonpollinated (NP) 'Beckyblue'rabbiteye blueberry fruits.
Values are means �SE, n =3; SE bars present only when
larger than symbol ............................... 89

5-1 Developmental changes in fresh weight of GA3-treated
(GA3), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means � SE, n = 3; SE bars present only when larger than symbol. Arrows indicate
treatment application times ....................... 107

5-2 Median cross section of a 'Beckyblue' rabbiteye
blueberry fruit at 0 DAB. Bar= 150 jm. ep=epidermis; hp=hypodermis; om=outer mesocarp; mm=middle mesocarp; im = inner mesocarp; en = endocarp; vb = vascular
bundles; Ic = locule; pl = placental tissue; ov = ovule ..... .109

5-3 Cell number changes per median cross sectional area
(x.s.) in the pericarp of GA3-treated (GA3), pollinated (POLL), and nonpollinated (NP) blueberry fruits. Mean separation
across treatment and time by LSMean's, p =0.05 ...... 110

5-4 Increases in cell size in developing (A, B) epicarp and
(C) endocarp tissues of GA3-treated (GA3), pollinated (POLL), and nonpollinated (NP) blueberry fruits
Significance level at 0.05, n = 9 .................... 111

5-5 Median cross section of (A) GA3-treated, (B) nonpollinated,
and (C) pollinated 'Beckyblue' rabbiteye blueberry fruit at 3 DAB. Bar=150 pm. ep=epidermis; hp=hypodermis;
om=outer mesocarp; mm=middle mesocarp; im=inner
mesocarp; en = endocarp; vb = vascular bundles;
Ic = locule . . . . . . . .. .. . . . . . . . .. . . . . . . . . . . . . . . 113


5-6 Median cross section of (A) GA3-treated, (B) nonpollinated,
and (C) pollinated 'Beckyblue' rabbiteye blueberry fruit at 10
DAB. Bar = 150 pm. ep = epidermis; hp = hypodermis;
om=outer mesocarp; mm=middle mesocarp; im=inner
mesocarp; en = endocarp; vb = vascular bundles ........ 115







5-7 Median cross section of (A) GA3-treated, (B) nonpollinated,
and (C) pollinated 'Beckyblue' rabbiteye blueberry fruit at 24 DAB. Bar=150 pm. ep=epidermis; hp=hypodermis; om = outer mesocarp; mm = middle mesocarp; im = inner
mesocarp; en = endocarp; vb = vascular bundles ........ 117

5-8 Median cross section of (A) GA3-treated and (B)
pollinated 'Beckyblue' rabbiteye blueberry fruit at 45 DAB.
Bar = 150 pm. ep = epidermis; hp = hypodermis; or =outer mesocarp; mm = middle mesocarp; im = inner mesocarp;
en = endocarp; vb = vascular bundles; sc = sclereids . . . 119

5-9 Median cross section of (A) GA3-treated and (B)
pollinated 'Beckyblue' rabbiteye blueberry fruit at ripening (87 and 72 DAB, respectively). Bar= 150 pm.
ep = epidermis; hp=hypodermis; om=outer mesocarp; mm middle mesocarp; im= inner mesocarp; en = endocarp;
vb =vascular bundles; sc = sclereids ................. 121

5-10 Increases in cell size in developing mesocarp tissues of GA3treated (GA3), pollinated (POLL), and nonpollinated (NP)
blueberry fruits. Significance level at 0.05, n = 9 ....... 122

5-11 Average cell enlargement rate in developing (A, B)
epidermal, hypodermal, and (C) endocarp tissues of GA3treated (GA3), pollinated (POLL), and nonpollinated (NP)
blueberry fruits. Calculated from data in Figure 5-4 ..... 123

5-12 Average cell enlargement rate in developing mesocarp
tissues of GA3-treated (GA3), pollinated (POLL), and nonpollinated (NP) blueberry fruits. Calculated from data in
Figure 5-5 ................................. 124

5-13 Developmental changes in mesocarp cell size, fruit
fresh weight, and pericarp cell number in (A) pollinated, (B)
GA3-treated, and (C) nonpollinated rabbiteye blueberry
fruits ........................................ 125













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
POLLINATION AND GA3 EFFECTS ON FRUIT GROWTH, SUCROSE
METABOLISM, CELL NUMBER, AND CELL SIZE OF BLUEBERRY FRUITS

By

Raquel Cano Medrano

December, 1994

Chairperson: Dr. Rebecca L. Darnell Major Department: Horticultural Science

Research reported here examines the relationship between growth of pollinated and GA3-induced parthenocarpic blueberry fruits and differences in sucrose metabolizing ability and cell number and cell size of treated fruits.

Fruit set was induced and development initiated with all combinations of GA3 concentrations and application times tested, as well as in conjunction with cross- and self- pollination. However, compared to pollinated fruits, final fruit weight was decreased and fruit development was prolonged when GA3 was applied alone or in combination with self-pollination.

Pollinated fruits began importing and accumulating carbon at higher rates than GA3-treated fruits as early as 2 to 4 days after bloom (DAB). This higher rate of carbon import was sustained throughout development of pollinated fruits. The prolonged period II of growth in GA3-treated fruits resulted in








relatively higher respiratory carbon losses compared to pollinated fruits, and this may have negatively affected final fruit size. Carbon accumulation occurred mostly during the final stage of fruit growth in both treatments and was 3-fold greater in pollinated fruits compared to GA3-treated fruits.

Sucrose phosphate synthase, sucrose synthase, and insoluble and soluble acid invertase activities were similar throughout development for fruits of both treatments. SPS and SS activities were low throughout most of the fruit development period. Insoluble acid invertase activity was relatively high (ca. 15 pmol-gFW1 h') at anthesis, declining gradually throughout development. From 45 DAB to ripening, soluble acid invertase activity increased sharply, from 0 to 60pmol'gFW h'. Glucose and fructose accumulated in similar amounts and were strongly correlated with invertase activity. Thus, invertase was the main enzyme involved in soluble sugar accumulation during the final stage of development in blueberry fruits.

Pericarp cell number increased throughout fruit development, averaging 10300 cells/median cross sectional area (x.s) at ripening. Mesocarp composed about 70% of the total pericarp area and contributed an average of 8900 cells/x.s. Differences in size between GA3-treated and pollinated fruits were due to differences in cell size in middle and inner mesocarp cells. Cells from pollinated fruits increased an average of 33-fold in size compared to an average of 25-fold in GA3-treated fruits throughout fruit development.













CHAPTER 1
INTRODUCTION



Rabbiteye blueberries are native to north Florida and well adapted to the climate and acid mineral soils, as well as highly tolerant to Florida diseases. The main disadvantages of rabbiteye blueberries are low fruit set in some years and late fruit ripening relative to southern highbush cultivars. Poor set in rabbiteye blueberry has been attributed to several factors, including floral morphology, pollen viability, pollen incompatibility, and presence of pollinating insects.

In general, the mechanism of fruit set is unknown. In some cases, an increase in hormone levels following pollination and fertilization has been detected, and this stimulus can be replaced by exogenous applications of growth regulators. However, a direct relationship between hormone levels and fruit set has not yet been found.

Increased fruit set and parthenocarpic fruit development in lowbush and highbush blueberries have been obtained under experimental conditions using gibberellic acid (GA3). In rabbiteye blueberries, responses to GA3 applications have been variable depending on concentration, timing, and cultivar. Studies have shown that one of the main effects induced by GA3 applications to flowers







2

is stimulation of assimilate transport and accumulation of carbohydrates into the tissue at the application site. GA3 effect on assimilate import into sinks may be caused by effects on sink strength, i.e., sink activity and/or sink size. Although the effect of GA3 on sink strength has been studied to some extent in other organs, it has not been investigated extensively with respect to fruit set and development. Thus, the objectives of this study are a) to determine the effectiveness of GA3 in increasing fruit set and yield in rabbiteye blueberry, b) to determine the effect of GA3 on fruit carbon requirements and sucrose metabolizing enzymes in relation to fruit set and development, and c) to determine the effect of GA3 on cell number and cell size in blueberry fruit and the relationship to fruit set and development.













CHAPTER 2
REVIEW OF LITERATURE



Flower Development


The upper limit of fruit that can be produced by an individual plant during a reproductive season is determined by the number of female flowers, where the number of seeds is fixed by the number of ovules within those flowers (Stephenson, 1981). Flower bud initiation, a major factor affecting reproductive development, occurs the summer before flowering for most fruit crops, including blueberry. Floral differentiation normally takes place either in late winter or early spring.

Bell and Burchill (1955) found that floral initiation in lowbush blueberry started in June, under Nova Scotia conditions. Early in August, sepals, petals, stamens, and carpel primordia were fully formed and elevated to their epigynous position. Differentiation started in March. In the anthers, meiosis was completed during the first week of May and the vegetative and generative cells were formed 15 days later. In ovules, meiosis was completed during the third week of May.

In highbush blueberry, under Rhode Island conditions, floral differentiation began in August and by October all flower parts were visible (Gough et al.,







4

1978). Microspore and megaspore mother cell activity was apparent early in November and cell division continued through the autumn. By mid-March, the ovules began their final phase of development. Pollen grains appeared to be fully formed by mid-April and embryo sacs were formed just prior to bloom.

There is no information available about floral initiation and floral development in rabbiteye blueberry, but events are probably similar to those described above.

Pollination and Fertilization


Except in apomictic and parthenocarpic fruits, pollination and fertilization must occur in order for fruit set to occur. Factors such as floral morphology, insect visitation, pollen viability, pollen tube growth, embryo sac development, and pollen source affect pollination and fertilization, and therefore, fruit set.

In blueberry plants, each flower bud consists of about five individual flowers with each flower at a slightly different stage of development (Goldy, 1985). However, anthers within each individual flower are generally at the same meiotic stage. Flower structure is such that insect pollination must occur in order to achieve fruit set (Eck and Mainland, 1971). Individual flowers hang in a pendant position and the pistil protrudes past stamens and corolla; thus, the receptive surface of the stigma is not exposed to the pollen from the anther.

Lee (1 958) found that the honeybee is an effective pollinator of lowbush blueberry. However, Wood (1966) found no significant increase in fruit set of lowbush blueberry by increasing the number of honeybee colonies from 2.5 to







5

7.4 colonies per hectare. In highbush blueberry, Howell et al. (1972) found an increase in yield with the use of 10 hives per hectare and suggested that hives be introduced into plantations no later than 25% full bloom. Although honeybees can pollinate rabbiteye blueberry flowers, Payne et al. (1989) and Cane and Payne (1990) found that pollination was more effective by bees that vibrate the flowers, such as bumblebees (Bombus spp.) and the southeastern blueberry bee (Habropoda laboriosa).

The total potential for production of viable pollen may affect fruit set in some blueberry cultivars. Stushnoff and Hough (1968) found differences in microspore development in two highbush blueberry cultivars that differed in fruit production. Under field conditions, 'Coville ' showed inconsistent fruit production while 'Bluecrop' did not. 'Coville'pollen did not dehisce readily from the anther sacs and did not germinate in vitro. Embryo sac development was normal in the two cultivars tested, and when fertilization occurred, the development of the endosperm and embryo was also normal. Eaton (1966) and Brewer and Dobson (1 969) found cultivar differences in both number of pollen grains released and pollen germination. Brewer and Dobson (1969) suggested that low pollen production coupled with poor pollen germination may reduce fruit set under field conditions. Later, Cockerham and Galletta (1976) found that pollen in tetraploid species, including V. corymbosum, was potentially more fertile (as judged by stainability) than in diploid species, but pollen stainability in hexaploid species, including V. ashei, was not different from the







6

diploid or tetraploids. There were no differences in pollen analysis when they compared pollen from plants located at different sites. They concluded that genetic, rather than seasonal or environmental, differences appeared to account for the major portion of the interclonal and interspecific variation observed in pollen stainability. Young and Sherman (1978) found that the rate of pollen tube growth in the styles of both rabbiteye and southern highbush cultivars was adequate with pollination up to 6 days after emasculation, under greenhouse conditions (33/180 C). Varying intervals between emasculation and pollination up to 6 days did not consistently influence percentage fruit set or number of seeds produced per berry. They concluded that flowers in the field should be able to set adequately even with 2 to 4 days of unfavorable pollination weather during bloom.

Embryo sac development and stigmatic receptivity also affect fruit set. Eaton and Jamont (1966), studying ovule and embryo sac development between anthesis and petal fall in highbush blueberry, found that at anthesis and 1 day afterward, 81 % of the normal embryo sacs had differentiated egg cells, and from 2 to 9 days after anthesis, 95% of these sacs had differentiated egg cells. Between anthesis and two days later, 15% of the ovules had degenerated embryo sacs. This percentage increased to 33% three days after anthesis. In lowbush blueberry, blossoms remained receptive until 7 days after anthesis (Wood, 1962), and pollen tube growth required about 4 days to grow from stigma to ovule (Bell, 1957). In highbush, flowers remained receptive to








7

pollination up to 8 days after anthesis (Moore, 1964). In rabbiteye blueberries, the receptive period was 4 days (Young and Sherman, 1978).

Pollen source and the extent of self-compatibility also affect fruit set in blueberry. Highbush blueberry is generally believed to be highly self-fruitful, while rabbiteye blueberry is self-unfruitful. However, both species have shown different levels of self-compatibility, and fruit set has varied depending upon this factor.

In highbush, the effect of pollen source on fruit set and development is contradictory. While Coville (1921), Bailey (1938), and Morrow (1943) reported than self-pollinated northern highbush flowers set fewer berries than cross-pollinated flowers, Merrill (1936) found that self-pollination gave a satisfactory commercial fruit set and White and Clark (1939) indicated that set from self-pollination was not significantly different from the set in openpollinated flowers. More recently, Gupton (1984) found that self-pollination of southern highbush cultivars generally resulted in fruit set equal to or better than cross-pollination. However, Lyrene (1989) found that fruit set was significantly lower for self-pollinated than for mixed and cross-pollinated flowers of the southern highbush cultivar, 'Sharpblue.' All genotypes of the recently introduced half-high blueberry (V. corymbosum x V. angustifolium) set fruit on at least 40% of the flowers following cross-pollination. However, fruit set percentage varied from 0 to 84% following self-pollination (Rabaey and Luby, 1988).







8

In rabbiteye blueberry, Meader and Darrow (1944) found that under greenhouse conditions, most varieties tested were either partially or completely self-fruitful. Only one variety was completely self-unfruitful. In the three varieties of rabbiteye blueberry tested by Tamada et al. (1977), cross-pollination usually increased fruit set more than self-pollination. EI-Agamy et al. (1982) in their study on intra- and inter-ploidy pollen incompatibility, reported that selfing reduced fruit set in highbush and rabbiteye blueberries by 28.5% and 36.3%, respectively, compared to crossing. Although self-incompatibility was apparent in both types, highbush clones were more self-fruitful than rabbiteye clones. Lyrene and Goldy (1983) found that fruit set of several rabbiteye cultivars ranged from 36% to 75% on open-pollinated branches and from 3% to 21 % on self-pollinated branches. In a study of native rabbiteye blueberry clones, Garvey and Lyrene (1987) found that fruit set averaged 15% after selfpollination and 58% after cross-pollination.

There is also cross-incompatibility between some rabbiteye blueberry cultivars. Darnell and Lyrene (1989) found that when two related rabbiteye cultivars, 'Beckyblue' and 'Aliceblue,' were reciprocally crossed, fruit set was 25%, while cross-pollination with 'Climax' pollen resulted in 59% and 72% fruit set, respectively.

Fruit Growth and Development

Pericarp development of blueberry is similar to that of peach, cherry, grape, and fig. The double sigmoid type of growth curve is divided into three







9

stages: Stage I, consisting primarily of cell division following fertilization; Stage II, a period of slow pericarp development and development of embryo and endosperm tissues; and Stage Ill, a second period of rapid pericarp growth (mainly cell enlargement) that continues to fruit maturity (Young, 1952; Hindle et al., 1957; Edwards et al., 1970; Spiers, 1981). Duration of individual stages vary with the cultivar, clone, and temperature. The length of each growth stage is important in determining the time required from anthesis to fruit maturity.

The duration of both cell division and cell enlargement affects final fruit size. Cell division might occur well before anthesis, as has been shown in cherry (Tukey and Young, 1939) and peach (Scorza et al., 1991). The difference in cell number between small- and large-fruited peach cultivars was established 175 days before bloom. In addition, the small-fruited types had a shorter Stage I and longer Stage II.

Cells from different tissues in the pericarp may divide and enlarge at different rates and planes with a consequent effect on fruit shape. In grape, pericarp cells around the vascular bundles elongate tangentially, while those toward the placenta elongate radially (Harris et al. 1968). Similarly, in cherry, hypodermal cells enlarged in a tangential direction, while middle and inner fleshy pericarp cells enlarged radially (Tukey and Young, 1939). In cucumber, although cell elongation occurred throughout the fruit development period, cells at the peduncle end were larger than those at the blossom end of the fruit (Marcelis and Hofman-Eijer, 1993). The growth rate at both ends slowed







10

sooner than the growth rate of the rest of the fruit. The contribution of cell division period and cell enlargement to fruit development in blueberry has not yet been documented. The potential success of exogenous applications of GA3 may depend on the stage of development and, since both cell division and cell enlargement can be altered by growth regulators, knowing when they are occurring becomes important.

The formation of seeds within the pericarp has a marked effect on fruit size and development (Ryugo, 1988). In apple, peach, kiwifruit, and grape, the size and shape of the fruit are a function of the number of seeds formed. In general, fruit with a higher number of seeds are longer and more symmetrical than those with fewer seeds. Seed formation also affects blueberry fruit development. Aalders and Hall (1961) found that seed number in lowbush blueberry was significantly correlated with berry weight and fruit development period. Heavier berries and earlier ripening were associated with more seeds.

In highbush blueberry, Coville (1921) stated that self-pollination produced fewer berries, which were smaller and later maturing than berries that were cross-pollinated, suggesting a direct effect of seeds on these characteristics. White and Clark (1939) found that seed number in large berries ranged from 55-68, while smaller berries had 21-41 seeds/berry. Morrow (1943) reported that cross-pollinated highbush blueberry fruits averaged 61-62 seeds/berry, while self-pollinated fruits averaged 45-58 seeds/berry. In rabbiteye blueberry, the average total weight of the seeds per berry was greater for cross-pollinated







11

(20-34 mg/berry) than for self-pollinated berries (4-18 mg/berry) (Meader and Darrow, 1944). Darrow (1958) reported that seed number in highbush blueberry ranged from 15.7 to 73.8 per berry, and although the varieties with the largest berries generally had the most seeds, this relationship was not always true. Eaton (1967) reported that variation in seed number accounts for less than 40% of the total variability in berry weight in cross-pollinated highbush blueberry. Fruit of the southern highbush 'Sharpblue' also had an increase in seed number per berry, increased berry size, and a decrease in the fruit development period when cross-pollinated (Lyrene, 1989). However, the effect of the increased seed number was greater at low seed level (5 seeds per berry) than at high seed level (10 seeds per berry).

The effect of genetic relationships on berry weight and seed number among cross-pollinated blueberry cultivars was investigated by Hellman and Moore (1983). They found a significant reduction in mean berry weight associated with an increase in the degree of relationship between parents for only one highbush ('Coville') and one rabbiteye ('Climax') cultivar. They also found a variable response to cross-pollination among related individuals with respect to number of seeds per berry, although in general, self-pollination reduced the number of seeds per berry. Gupton (1984) reported that in highbush blueberry, berry weight and number of viable seeds per berry were not significantly higher for fruit produced from half-sibling crosses than selfing and that outcrosses generally produced the greater berry weight and seed number.







12

In rabbiteye blueberry, seed weight per berry and number of large seed per berry were reduced by self-pollination and by pollination with related cultivars (Darnell and Lyrene, 1989). However, mean berry weights of reciprocal crosses between the related cultivars were not significantly different from weights of berries produced by pollination with an unrelated cultivar. Meader and Darrow (1944) found that cross-pollinated rabbiteye blueberries had larger fruit and a shorter fruit development period than berries from self-pollinations. Tamada et al. (1977) found an increase in the number of large seeds per berry as a result of cross-pollination, which was reflected in larger berries and earlier ripening. Endogenous Growth Regulators

There are two classical points of view related to the role of growth regulators in fruit set and development (Browning, 1989). The first states that there is a deficiency of auxin and gibberellin or an excess of inhibitor that inhibits ovary growth at anthesis. The promotory hormonal stimulus needed is supplied by pollination and fertilization or by other means in the set of parthenocarpic fruits. Fruit growth is controlled by hormones produced by the endosperm and embryo in developing seed or by maternal tissue in the ovules of fruits that set parthenocarpically. The second view states that the primary effect of pollination or hormone treatment of the flower is to increase the nutrient import needed for growth to become self-sustaining through the subsequent mobilizing action of hormones produced by the ovules or seeds.








13

In lowbush blueberry, Collins et al. (1966) reported a period of increased auxin-like activity in developing fruit, beginning about 3 weeks after anthesis. Auxin-like activity reached a maximum 6 to 7 weeks after anthesis, during Stage II of fruit growth, before declining. No conclusions can be drawn from this experiment because auxins are unlikely to be the one promoting substance that affects fruit growth. However, it can be speculated that auxins may either inhibit pericarp growth or promote development of the embryo, since activity peaked in Stage II of slow pericarp growth.

In highbush blueberry, Mainland and Eck (1971) found cultivar differences in the endogenous levels of auxins. Auxin activity increased in pollinated 'Coville' berries two weeks after anthesis, while 'Jersey' fruits exhibited no such increase. A second peak in auxin activity occurred 6 weeks after anthesis for pollinated berries of both varieties. GA3 at 500 ppm, applied at blossom time, increased endogenous auxin in fruits of both cultivars compared to the nontreated berries. Nonpollinated 'Jersey' flowers contained higher concentrations of GA-like substances than did 'Coville' flowers. GA-like activity in fruit decreased in both cultivars two to three weeks after pollination. A second peak in GA-like activity occurred three to four weeks after anthesis, corresponding with Stage II of fruit development in both pollinated and GA3 treatments. GA3 application at bloom period increased GA-like activity in both cultivars, but at maturity, GA-like activity was higher in pollinated than in GAtreated berries.







14

Natural parthenocarpy has been observed in 'Jersey' cultivars and may be related to the higher levels of auxin activity compared to 'Coville' (Mainland and Eck, 1971). Gustafson (1939) found that the auxin content in the ovaries of flower buds from oranges, lemons, and grapes that produced parthenocarpic fruits was higher than in the ovaries that do not produce this type of fruit. Gil et al. (1972) proposed that pear fruit set and development may be associated with a sequential role of different endogenous growth substances. After anthesis, gibberellins were the main growth-promoting substances detectable. When gibberellin levels decreased, auxins appeared and remained conspicuous for most of the fruit development period.

The stimulus for fruit set delivered by pollination has not been identified. In some cases, endogenous hormone levels are increased by pollination and fertilization, and these stimuli can be replaced by exogenous hormones in many cases. However, a direct relationship between hormonal content and fruit set has not yet been found.

Exogenous Growth Regulators

Increased fruit set and parthenocarpic fruits have been obtained in several species by using exogenous applications of gibberellins, auxins, or a combination of both (Leopold, 1962). Parthenocarpic fruit development has been induced in tomato with gibberellins (Wittwer et al. 1956; Gustafson, 1959). Webster and Goldwin (1984) enhanced fruit set in sweet cherry by 125-150% the first year and 75-90% the subsequent years using a mixture








15

containing 200 ppm GA3, 300 ppm NN'-diphenylurea (DPU) and either 10 ppm naphthalene acetic acid (NAA), 10 ppm 2,3,5-trichlorophenoxypropionic acid (2,4,5-TP), or 50 ppm 2-naphthoxyacetic acid (NOXA). In self-sterile plum cultivars, GA3 induced parthenocarpic fruit and prevented the drop of fruitlet with aborted embryos (Hartman and Anvari, 1986). In 'Agua de Aranjuez' pear fruits, GA3 at 10 ppm applied at balloon stage, anthesis, or petal fall induced parthenocarpic fruit set and improved final fruit set (Herrero, 1984). This implies that a fully developed embryo sac is not necessary for the hormone to act. On the other hand, 10 or 20 ppm GA3 at full bloom increased initial fruit set of 'Doyenne du Cornice' pear fruits, but there was no difference in final fruit set compared to the control (Marcelle, 1984).

Several exogenous growth regulators have been used experimentally to increase fruit set in blueberry. In lowbush blueberry, Barker and Collins (1965) induced parthenocarpic fruit set with GA3 in concentrations as low as 20 ppm applied directly to the flower. They observed that the flower was physiologically receptive to the GA from 2 to 3 days before anthesis to 7 to 8 days after anthesis. The fruit from GA treatments under field conditions appeared normal in all visible aspects, but under growth chamber conditions, the waxy bloom did not appear and the sugar concentration was decreased. They attributed these negative effects to the low light intensity and/or temperature in the environment provided by the growth cabinet; however, it is not possible to determine if the effects on fruit quality were due to treatment







16

effects or growth environment, since pollinated controls were not included in the growth chamber treatments.

In highbush blueberry cultivars, Mainland and Eck, (1 968a) found that 5, 50, and 500 ppm of potassium gibberellate (KGA3) applied to the base of the style of emasculated flowers 2 to 5 days after they opened promoted higher percent set (80%) and larger berry diameter than the hand-pollinated treatment. They attributed this higher set to the high temperature (32.2C) that occurred on the day of treatment and the two following days. High temperatures and dry weather conditions may limit the maximum period of pistil receptivity and promote premature flower abscission in pollinated treatments. However, the high temperature and low relative humidity might have been expected to decrease uptake of exogenous gibberellins (Greenberg and Goldschmit, 1989), resulting in lower fruit set in the GA treatments as well. Fruits resulting from KGA3 treatments were indistinguishable from pollinated berries in terms of shape; however, they were completely seedless and were delayed in maturation. Of the auxin materials evaluated, 2,4, 5-trichlorophenoxyacetic acid (2,4,5-T) produced the highest percent fruit set (40-50%), but berry diameter was much smaller than for the hand-pollinated treatments. Fruit maturation was also delayed in these treatments.

Increases in both fruit set and yield were found in highbush blueberry treated with 500 ppm KGA3 combined with either 3 ppm 2,4,5-T or 20 ppm 2,4,5 TP applied at full bloom (Doughty and Scheer, 1975). Mixtures of 5, 50,








17

and 500 ppm of KGA3 and NAA applied at anthesis increased fruit set of highbush blueberry (Mainland and Eck, 1968b). Application of 6-(benzylamino)9-(2-tetrahydroxyranil)-purine (BTP) at 100, 500, or 1000 ppm did not induce fruit set, nor did it have any synergistic effect when used in combination with KGA3 or NAA. KGA3 treatments produced consistent increases in diameter with increasing concentration, and the berries resulting from treatment with 500 ppm of KGA3 were only slightly smaller than pollinated fruits. NAA at 500 ppm increased fruit set compared to hand-pollination, but it increased diameter only slightly at the highest concentration tested. The number of days from treatment to maturity decreased with increasing levels of both NAA and KGA3, compared to the pollinated control. NAA and KGA3 increased fruit soluble solids compared to the pollinated berries. Growth curves of the parthenocarpic fruit were very similar to those of seeded fruits, particularly as gibberellin concentration increased.

In a later study, Mainland and Eck (1969b) applied 5, 50, 200, or 500 ppm GA3 to flowers of uncaged plants and GA3 at 500 ppm to flowers of caged plants under field conditions. Fruit set was increased by all GA3 treatments the first year; however, the weight of fruit harvested per bush was not increased. In the second year, there was an increase in yield in all treatments due to an increase in fruit weight, but there was not an increase in fruit set. There was also a trend towards delayed ripening in all GA3 treatments in both years, but only the caged plants receiving 500 ppm GA3 experienced a significant delay








18

in fruit maturation. Additionally, seed number and berry size were significantly reduced in both the first and second pickings with GA3 treatments. No differences in percentage mold, percentage weight loss, juice pH, or titratable acidity were found between control and GA-treated fruits. All GA-treated berries had lower soluble solids than did the control berries. There were no differences in the number of flower buds formed among treatments in either year.

To determine whether or not GA3 applied to blueberry foliage would induce parthenocarpic fruit development, Mainland and Eck (1 969a) applied 5, 50, 200, and 500 ppm to highbush blueberry foliage or foliage plus flowers at full bloom. In the foliage treatments, clusters were wrapped to shield them from any GA3 spill. Since low fruit set was found in treatments where the flowers were unsprayed, they concluded that GA3 was not translocated from foliage and stems into the flower. In two consecutive years, all treatments but the foliage one resulted in higher fruit set than in cross-pollinated plants.

In rabbiteye blueberry, Davies and Buchanan (1979) reported an increase of 50% final fruit set in 'Tifblue' with one application of 200 ppm GA3 at 3040% full bloom and with two applications of 50 and 200 ppm GA3 at 30-40% full bloom and 15 days later. In 'Delite' blueberries, two applications of 200 ppm GA3 at 60% full bloom and 15 days later were necessary to increase fruit set by 60%. In a later study (Davies, 1986), single applications of a mixture of 100 ppm GA3 and 10 ppm 2,4-dichlorophenoxyacetic acid (2,4-D) at the







19

time of leaf emergence, or double applications of the same chemicals at 1 % full bloom and at leaf emergence, had no effect on either fruit set or yield. This response could be related to the application time and/or an increase in ethylene production due to auxin applications. Vanerwegen and Krewer (1990) increased yield in rabbiteye blueberry by applying 243 ppm GA3 at 90% full bloom or 121 ppm GA3 twice, at 90% full bloom and 18 days later. Two applications of GA3 increased yield by 373% in 'Climax' and 562% in 'Tifblue.' Fruit size and soluble solids were reduced, and fruit development period was increased.

In summary, increased fruit set has been obtained in lowbush and highbush blueberry by using single applications of GA3, while in rabbiteye blueberry, the results have not been consistent. Optimum concentration and timing of application of GA3 for rabbiteye blueberry have not been established.

Effect of Exoqenous Growth Regulators on Sink Tissues

Carbohydrate allocation patterns depend on the competitive ability of various sink regions within the whole plant (Daie, 1985, 1987). A high sink import rate is a function of its mobilizing activity, which, in turn, is a function of sink size and activity.

One of the effects induced by treatment with plant growth regulators is a change of the pattern of assimilate distribution within the plant (Daie, 1985; Kinet et al., 1985; Aloni et al., 1986). This effect on assimilate import into sinks may be caused by affecting sink size and/or sink activity. Cell number








20

and cell size are the components of sink size, while phloem unloading, uptake of assimilates by sink tissues, and metabolism and carbohydrate storage are involved in sink activity (Yelle et al., 1988). Both of these characteristics are subject to hormonal regulation (Bunger-Kibler and Bangerth, 1983; Craighton et al., 1986; Kinet et al., 1985; Iwasaki et al., 1986; Lethman 1963; Vercher et al., 1984, 1987).

Sink Size

Sink size is the physical constraint of sink strength and in general, can be measured by cell number and cell size. Plant growth regulators may directly affect cell division and/or cell elongation. In tomato fruit, exogenous applications of 10% indole acetic acid (IAA) or 0.1 % 4-chlorophenoxyacetic acid (4-CPA) to emasculated flowers transiently increased the rate of cell division, while 0.6% GA3 decreased cell division rate through development (Bunger-Kibler and Bangert, 1983). However, cell size was increased by using gibberellins compared to using auxins. In pea ovaries, GA3 applied at anthesis induced mesocarp cell enlargement and division and differentiation of endocarp cells (Vercher et al., 1984). GA3 application to pea endocarp enhanced synthesis of primary and secondary cell walls (Vercher et al., 1987). Nishitani and Masuda (1982) found that synthesis of polyuronides, xyloglucans, and cellulose was regulated by auxins in Vigna angularis, whereas synthesis of xylans and cellulose during cell maturation was regulated by GA. In isolated Zinnia mesophyll cells, GA3s inhibited cell division and delayed the initiation of








21

DNA synthesis only when cells were in a resting state and not when cells were dividing continuously (Iwasaki et al., 1986). Cell elongation depends on the extensibility of the cell wall, which in turn is affected by the orientation of cellulose microfibrils (Mita and Katsumi, 1986). GA3 induced transversely oriented microtubules in mesocotyl epidermal cells of Zea mays L. mutants, stimulating elongation in this region (Mita and Katsumi, 1986). In flank meristems of Hedera, GA3 stimulated the frequency of tangential divisions as well as the expansion in the radial-longitudinal direction (Marc and Hackett, 1992). Keyes et al., (1990) found that exogenous GA3 increased cell wall extensibility and elongation in nonmutant (GA3-sensitive) control wheat compared to GA-insensitive genotypes. They concluded that, although chemirheological processes may be involved in cell wall loosening, the mechanical properties of the cell wall are strongly associated with leaf cell expansion potential, and GA is involved in the determination of the mechanical wall properties. Similar results were obtained in Phaseolus vulgaris leaves by Brock and Cleland (1990) and in Pisum sativum L. seedlings by Cosgrove and Sovonick-Dunford (1989).

In summary, the effect of growth regulators on cell division and enlargement depends upon species, stage of development, and tissue type within a fruit. Cell number at anthesis, the length of the cell division period after anthesis, and the extent of cell enlargement determine final fruit size in a number of fruits, including blueberry. The contribution of cell number and cell







22

size to the establishment of sink strength during fruit set and fruit development period has not been investigated in blueberry. Sink Activity

Increased cell number and cell size results in more sites for dry matter accumulation (Brenner et al., 1989). This dry matter accumulation will depend on the concentration gradient established between the source and the sink. Specifically, dry matter accumulation in sink tissue depends on phloem loading, phloem transport, phloem unloading, and uptake and compartmentation by the sink tissue. Hormones may affect assimilate (primarily carbon [CI) partitioning at different points along this pathway. Growth stimulation by gibberellic acid has been observed in pea subhook (Pisum sativum L.) within 6 h after application. Growth was positively correlated with soluble sugar accumulation, suggesting that GA3 stimulated the translocation from cotyledons to the subhook, thereby maintaining a low osmotic potential and enhanced growth (Miyamoto and Kamisaka, 1988). Similar results were observed in dwarf watermelon seedlings, although the GA-mediated increase in dry matter in hypocotyls was not the result of increased translocation out of the cotyledons but rather was the result of decreased accumulation of dry matter in the roots (Zack and Loy, 1984). In pea ovary explants, GA3 in the presence of sucrose induced fruit set and development over GA3 alone (Garcfa-Martfnez and Carbonell, 1985). The effect of GA3 on ovary development was a function of the concentration and duration of the GA3 treatment. The lower the GA3








23

concentration, the longer the time of incubation in GA3 necessary to obtain the maximum effect. GA3 also stimulated ovary development when sucrose was substituted by glucose or fructose.

Achhireddy et al. (1984) investigated the translocation of 14C-sucrose from the stipule to the 10 day-old fruit of excised Pisum sativum fruit/shoot systems. They found that 25 to 100 ppm GA3 enhanced "4C translocation into the fruit by about 25% but did not alter 14C levels in the stipule. However, GA3 increased the levels of 4C obtained from untreated leaflets and from water washings. Hayes and Patrick (1985) found that GA3 promoted '4Cphotosynthate transport to the site of hormone application in Phaseolus vulgaris decapitated stems. This was associated with significant increases in the pool size of free space sugars at the hormone-treated region, suggesting that GA3 stimulated unloading of the sieve element-companion cell complexes into the stem free space.

The mechanism by which GA3 enhances transport and accumulation of sucrose to the site of application is unknown. However, anything that steepens the sucrose gradient between the loading and unloading sites (i.e., decreases the concentration of sucrose in the unloading area) would, in theory, enhance sucrose transport to that site (Aloni et al., 1986; Daie et al., 1986; Miyamoto and Kamisaka, 1988, 1990; Morris and Arthur 1985).

The level of sucrose in the plant cell depends on four enzymes that are involved in its synthesis and degradation (Avigad, 1982): sucrose phosphate








24

synthase (UDP-glucose: d-fructose-6-phosphate 2 glucosyltransferase, EC 2,4,1,14), sucrose phosphate phosphatase (Sucrose-6-phosphate phosphohydrolase, EC 3.1.3.00); sucrose synthase (UDP-glucose: D-fructose 2-glucosyltransferase EC 2.4.2.13), and invertase (B-D-fructofuranoside fructohydrolase EC 3.2.1.26).

Sucrose synthase is distributed in all plant tissues and is found at high levels particularly in nonphotosynthetic and storage tissues. It catalyzes the reversible reaction UDP-glucose + fructose T sucrose + UDP. UDP-glucose released by this cleavage and its derivatives are substrates for synthesis of cell walls and storage compounds (Avigad, 1982). In some tissues, sucrose synthase is considered a key enzyme of metabolism in sink organs (Claussen et al., 1985, 1986), but in others, no differences in sucrose synthase activity have been observed between growing and fully expanded leaves (Huber, 1989; Schmalstig and Hitz, 1987). In acid lime, Echeverria and Burns (1989, 1990) and Echeverria (1990) found that during the early stages of fruit growth, sucrose was catabolized enzymatically by sucrose synthase, acid, and/or alkaline invertase, while in mature fruits, sucrose breakdown occurred by nonenzymatic acid hydrolysis. In 'Valencia' orange juice sacs, sucrose synthase activity was absent in vacuole and cytoplasm (Echeverria and Valich, 1988) and sucrose and hexoses were accumulated mainly in the vacuole. In muskmelon fruits, there is a lack of agreement whether sucrose synthase or sucrose phosphate synthase is involved in sucrose accumulation. McCollum et








25
al. (1988) reported a decrease in invertase activity and an increase in sucrose synthase activity as muskmelon fruit developed and accumulated sucrose. Hubbard et al. (1989) found low activity of sucrose synthase during muskmelon fruit development. Invertase activity decreased with time, while sucrose phosphate synthase activity increased throughout fruit development. They concluded that acid invertase and sucrose phosphate synthase determined the sucrose accumulation in these fruits. Similar results were found by Miron and Schaffer (1991) in sucrose accumulating tomato fruits.

When comparing sucrose-accumulating vs hexose-accumulating tomato fruits, Dali et al. (1992) found that both fruit types had high sucrose synthase activity when immature, but activity was nondetectable 55 and 70 days after anthesis. Invertase activity increased continuously from 2000 to 15000 nmol.gFW-l.min1 in hexose-accumulators, while in sucrose-accumulators it declined constantly from about 100 to almost nondetectable levels. Sucrose phosphate synthase was low in both types, but sucrose-accumulating types showed an increase in activity at ripening. They suggest that sucrose phosphate synthase is a key enzyme in sucrose accumulation in these fruits and that sucrose is being hydrolyzed in the apoplast and resynthesized in the symplast. On the other hand, Stommel (1992) did not find a direct relationship between sucrose phosphate synthase activity and sugar accumulation during early stages of development in wild and cultivated tomato, which showed marked differences in sugar accumulation. Wang et al. (1993) found that








26

sucrose synthase and not invertase was the enzyme responsible for the metabolism of imported sucrose in growing tomato. Sucrose synthase activity, relative growth rate, and starch accumulation were positively correlated at 0 to 13 days after anthesis as well as 20 to 39 days after anthesis. This may be a typical pattern of enzyme activity for storage tissues like tomato, where imported sucrose is stored as starch during the early stages of development. If sucrose is broken down by sucrose synthase, a step that requires half the energy of that catalyzed by invertase, the UDP-glucose released might be used for, among other things, synthesis of starch. This will decrease the sugar content in the fruit and increase the sucrose gradient between leaves and fruit, leading to an increase in sucrose transport. High sucrose synthase activity has also been found in potato tubers, which are starch accumulating sinks throughout fruit development, and hexose reservoirs when excised and stored at low temperatures (Sung et al., 1989; Ross and Davies, 1992).

Moriguchi et al. (1992) studied the pattern of sugar accumulation and sucrose synthase, sucrose phosphate synthase, and invertase activities in highand low-sucrose-accumulating types of Asian pear fruits. There was no correlation between sucrose content and invertase activity in high-sucroseaccumulating types, but they found a positive correlation between sucrose synthase and sucrose (r=.633) and between sucrose and sucrose phosphate synthase (r =.445). Acid invertase accounted for most of the invertase activity








27
and showed high activity at early stages of development, decreasing in activity at later phases of growth in both fruit types.

Invertase competes with sucrose synthase for the same substrate, and in several sink organs, carbohydrate import is correlated with high invertase activity (Morris and Arthur, 1984b). In pea leaves, cell growth rates were positively correlated with specific activity of soluble acid invertase (Morris and Arthur, 1984a). However, mature, nonimporting leaves of some species, such as soybean and Swiss chard, have shown high activities of acid invertase (Huber, 1989).

Several growth regulators have induced invertase activity. Auxins stimulated invertase activity (Morris and Arthur, 1986; Poovaiah and Veluthambi, 1985; Schaffer et al., 1987; Tanaka and Uritani, 1979), as well as its synthesis de novo (Gordon and Flood, 1980; Pressey and Avants, 1980) in a number of tissues. Increased invertase activity has also been observed in soybean fruits following abscissic acid application (Ackerson, 1985). Broughton and McComb (1971) found that GA3 stimulated amylase and invertase activities in pea internodes. Their activities were closely correlated with internode growth. Since the injection of glucose mimicked the effect of gibberellic acid on fresh and dry weight accumulation, cell elongation, cell division, and cell wall synthesis, the authors concluded that the overall effect of GA3 was to provide more substrate for general cell metabolism and wall synthesis. In sweet potato roots, GA3 stimulated invertase activity (Tanaka and







28

Uritani, 1979). Kaufman et al. (1973) found that in response to the GA3 treatment, invertase activity increased, then decreased, in Avena segments. The increase in activity was correlated with the active growth phase, whereas the decrease in activity was initiated when growth of the segments slowed. A continuous supply of GA3 retarded the decline of enzyme activity, but growth rate remained constant. In pea internode, both the amount of acid invertase per internode and the specific activity increased after GA3 application (Morris and Arthur, 1985). The maximum specific activity of the enzyme coincided with the time of peak internode elongation rate. Hexose sugars were correlated with total acid invertase per node. Sucrose concentration decreased as hexose and invertase activity increased. Miyamoto et al. (1993) reported that GA3 and not auxins are responsible for the accumulation of sugar at the subhook region in Pisum sativum. Both substances promoted elongation, but the increases in soluble sugar and soluble invertase activity were observed only upon application of GA3. Interestingly, the excision of cotyledons inhibited soluble sugar accumulation and decreased invertase activity. This effect was not reversed by application of GA3 or IAA. In previous studies Miyamoto et al. (1988) found that auxins stimulated growth by increasing cell wall loosening, while GA did not have this effect. This suggests that the mechanisms by which these two enzymes promote elongation are different. Following the application of O.5L of 15 pM GA3 to Pisum sativum shoots, Wu et al. (1993) found that invertase mRNA level increased before any increase in invertase activity. Subsequent








29
elongation of the pea shoots occurred 12 h after the increase in invertase activity was detected. They concluded that GA3 might regulate the cell wall invertase gene at the transcriptional and/or translational levels. On the other hand, in detached eggplant leaves, no effects of GA3 could be detected on the activity of invertase. In other plants, such as Zinnia elegans, invertase activity was reduced upon application of an inhibitor of GA3 biosynthesis, and its effects were counteracted by a subsequent treatment with GA3 (Kim and Suzuki, 1989).

The effect of GA3 on invertase activity and/or levels is inconsistent, and appears to depend on species, organs, and/or stage of development. Increased SS activity has been reported in orchids treated with GA3 (Chen et al., 1994), but there is no information available on gibberellin effects on sucrose phosphate synthase activity or levels.

In summary, fruits can be seen both as utilization sinks (like meristems) during the early stages of development and as storage sinks in later phases of growth. In both cases, fruit development depends upon the continuous supply of photoassimilates. The eventual carbohydrate partitioning in a plant is related to the competitive ability of the various sink regions. Sink strength has been investigated from different points of view, including environmental factors, enzyme activities, and source-sink relationships. However, the role of the number of cells and cell size, i.e., actual sink size, has not been studied in relation to fruit set and fruit development. Furthermore, the relationship








30
between fruit set/development and GA3 effects on sink size or sink activity (i.e., soluble sugar levels and sucrose enzymes activity/levels) has not been investigated.

The objectives of the present study are a) to determine the effectiveness of GA3 in increasing fruit set and yield in rabbiteye blueberry, b) to determine the effect of GA3 on soluble sugar accumulation and sucrose metabolizing enzymes in relation to fruit set and development, and c) to determine the effect of GA3 on cell number and cell size in blueberry fruit and their relationship to fruit set and development.













CHAPTER 3
EFFECT OF GA3 AND POLLINATION ON FRUIT SET



Introduction


The stimulus for fruit set elicited by pollination and fertilization has not been identified. In some cases, endogenous hormone levels increase after pollination and/or fertilization. Additionally, in many cases, pollination and fertilization can be replaced by exogenous applications of growth regulators. However, a direct relationship between fruit set and hormone levels has not been found. Response varies with several factors, including type of hormone, concentration, application timing, and species and/or cultivar.

Increased fruit set and parthenocarpic fruit development have been obtained in several species by using exogenous applications of gibberellins, auxins or a combination of both (Leopold, 1962; Wittwer et al., 1956; Webster and Goldwin, 1984; Hartman and Anvari, 1986). Gibberellic acid (GA3) or potassium gibberellate (KGA3) have been used successfully to induce fruit set and parthenocarpic fruit development in lowbush and highbush blueberry (Barker and Collins, 1965; Mainland and Eck 1968a,b; 1969 a,b; Doughty and Scheer, 1975). In rabbiteye blueberry, usually self-unfruitful, the response to gibberellin applications has been inconsistent. Increased fruit set has been







32

observed some years with single or multiple GA3 applications (Davies and Buchanan, 1979; Davies, 1986; Vanerwegen and Krewer, 1990), but the response appears to be cultivar dependent and optimum concentration and timing of GA3 application have not been established. Additionally, the sometimes erratic response may be due to excessive concentration of gibberellins, since the above experiments were carried out under field conditions where different levels of cross- and self- pollination occur. It is known that endogenous gibberellin levels increase in pollinated/fertilized ovaries of several fruits (Iwahori et al., 1968; Jackson and Coombe, 1966; Mainland and Eck, 1971; Mapelli et al., 1978; Martin et al., 1982). Thus, application of GA3 to fertilized fruit may lead to supraoptimal GA3 levels as suggested by Pharis and King (1985). Weaver and Pool (1971) found that increasing concentrations of GA3 at blossom time reduced fruit set in grape. Similar results were found by Edgerton (1981) in cross-pollinated 'Delicious' apples sprayed with GA + BA at blossom time. Since this effect was counteracted by aminoethoxyvinyl glycine (AVG), they concluded that ethylene was involved in such response. On the other hand, Lane (1984) found that foreign as well as pioneer pollen stimulated fruit set and suggested that gibberellin enrichment of the stigma may be the mechanism by which foreign or mentor pollen improved fruit set of apple, sweet cherry and apricot. The combined effect of GA3, pollen source, and pollen has not been studied in blueberry. Thus, the objectives of the present experiments are a) to evaluate the effect of GA3 concentration and application time on fruit







33
set, fruit development, and fruit quality in 'Beckyblue' rabbiteye blueberry and b) to evaluate the interaction between GA3 and pollination on fruit set and fruit development.

Materials and Methods

Experiment I

Five uniform, 4 year-old rabbiteye blueberry plants, 'Beckyblue' were grown in 12L containers in 1:1 peat:pine bark. Plants were watered every other day and fertilized with 20N-5.6P- 11 K water soluble fertilizer once a week. On 1 November, plants were placed in a dark cooler at 7 � 1�C for 30 days. After the chilling period, the plants were transferred to a greenhouse with average day/night temperature of 17/16'C to force budbreak. GA3 solutions were prepared from Pro Gibb (Pro Gibb 4% a.i., Abbot Laboratories, Chicago Ill) in Mclllvaine buffer, pH 3.5 (0.02 M Na2HPO4:0.01 M citric acid) and 0.01 % Tween-80 as a surfactant. Upon opening, individual florets were treated as follows:

Treatments Application times

1. 0.6 mM GA3 (1x) Full bloom (FB)

2. 1.4 mM GA3 (lx) FB

3. 0.3 mM GA3 (2x) FB + 7 days after full bloom (DAFB)

4. 0.7 mM GA3 (2x) FB + 7 DAFB

5. 0.2 mM GA3 (3x) FB + 7 DAFB + 21 DAFB 6. 0.5 mM GA3 (3x) FB + 7 DAFB + 21 DAFB









7. 0.1 mM GA3 (4x) FB + 7 DAFB + 21 DAFB + 42 DAFB 8. 0.4 mM GA3 (4x) FB + 7 DAFB + 21 DAFB + 42 DAFB

9. Hand-pollinated (Poll) FB

10. Nonpollinated (NP) FB

Pollinated treatments were hand pollinated once with 'Climax' pollen. Pollen was collected on the thumb nail by twirling the flowers. Nonpollinated treatments were sprayed at bloom with Mclllvaine buffer (pH 3.5). All treatments were applied on a single plant, 48 to 99 florets/treatment/plant, in a randomized complete block design, with five replications. Fruit abscission was measured on a weekly basis throughout fruit development and final fruit set was determined 10 weeks after anthesis. Fruit development period was calculated when 80% of the fruits had been harvested. For all harvests, fruit from each treatment were pooled for fresh weight, soluble solids, and pH measurements. Soluble solids were determined with an Abbe refractomer Mod. 10460 (Cambridge Instrument Inc., Bufalo, N.Y., U.S.A.). Total yield was calculated as a product of fruit fresh weight and final fruit number. Experiment II

Forty five 1-year-old 'Beckyblue' rabbiteye blueberry plants, grown in 2L containers in a 1:1 peat:pine bark mixture, were chilled and forced to budbreak in the same way as described above, except the average day/night temperatures in the greenhouse were 25/22�C. Nine plants were assigned to each of the following treatments selected from experiment I:










Treatments Application times

1. 0.3 mM GA3 (2x) 90% full bloom (FB) + 7 DAFB

2. 0.7 mM GA3 (2x) 90% FB + 7 DAFB

3. 0.4 mM GA3 (4x) 90% FB + 7 DAFB + 21 DAFB + 42 DAFB

4. Hand-pollinated (Poll) 90% FB 5. Nonpollinated (NP) 90% FB

Pollinated, nonpollinated, and GA3 treatments were applied as described previously except that each plant received only one treatment in order to eliminate competion effects among treatments. Prior to treatment application, each cluster was tagged and the number of florets and date was recorded. The number of florets per cluster varied from four to seven and the total number of florets per treatment ranged from 235 to 313. All treatments were applied in the afternoon. The experiment was a randomized block design, with plant size as the block, 2 blocks per treatment and four or five plants per block.

Fruit abscission was monitored on a weekly basis and final fruit set was determined 48 days after bloom. Fruit development period was determined when 80% of the fruits were harvested. All harvests were pooled to determine titratable acidity, soluble solids, and pH.

Titratable acidity was determined by titrating 5 mL juice with 0.1 N NaOH to pH 8.2 using an automatic titrator Mod. 380 (Fisher Scientific Co. U.S.A.). Soluble solids were measured as described.









Experiment III

Nine uniform, 2 year-old rabbiteye blueberry plants 'Beckyblue', grown in a 1:1 peat:pine bark mixture in 12L containers were chilled and forced to budbreak as described above. GA3 was prepared as described and applied at full bloom (FB) and at FB+ 7 DAFB in the following array of treatments:

1. Cross-pollination (saturation, =200 pollen tetrads) (X)

2. Cross-pollination (saturation) + 0.7 mM GA3 at FB+7DAFB (X+GA)

3. AX cross-pollination (50 pollen tetrads) (%X)

4. 1X cross-pollination (50 pollen tetrads) + 0.7 mM GA3 at FB+7DAFB

(Y4X+GA).

5. Self-pollination (saturation) (S)

6. Self-pollination (saturation) + 0.7 mM GA3 at FB+7DAFB (S+GA)

There were 96 florets per treatment and all treatments were applied on the same plant. Results were analyzed as a randomized complete block design. GA3 treatments were applied in the evening, at least four hours after pollination.

The number of pollen tetrads selected for the treatments was based on work by Parrie and Lang (1992, unpublished) indicating that blueberry stigmatic pollen saturation, i.e., the maximum amount of pollen grains that the stigmatic fluid can hydrate, ranges from 200 to 300 pollen tetrads. We considered that 50 pollen tetrads represent a quarter of the saturation level. Pollen tetrads were counted on a 100 mm2 black lid under a stereoscope, 1 5x magnification. The pollen was transferred by touching the black lid to the stigma. After each







37

transfer, pollen tetrads left on the black lid were transferred onto the stigma with a spatula. This process was repeated until all the pollen tetrads were on the stigma.

Fruit abscission was recorded as above, and final fruit set was determined at 56 DAFB. Fruit was harvested every other day and weekly harvestings were pooled to determine fruit development period and fresh weight. Subsequently, fruit was macerated with a mortar and pestle and seeds were separated and air dried for 24 hours. Total seed number and total seed weight per fruit were recorded. Seeds were separated into small and large seeds using a sieve of 1 mm2, counted and weighed.

Results and Discussion

Experiment I

The time-course and extent of fruit abscission for all treatments is shown in Fig. 3-1 and 3-2. Pollinated and nonpollinated controls are included in both figures for comparison. The majority of fruit abscission in the GA3 and pollinated treatments occurred in 2 stages: from 0 to 35 DAB and from 56 to 70 DAB. There was no difference in fruit abscission among treatments from 0 to 42 DAFB, except at 14 DAFB when abscission in pollinated treatments was significantly lower than in treatments with one application of 1.4 mM GA3 and three applications of 0.5 mM GA3 (Fig. 3-1) and 2 applications of 0.3 mM GA3 (Fig. 3-2). From 49 to 70 DAFB, there were no differences in abscission between GA3 and pollinated treatments; however, the extent of abscission in








38

the nonpollinated treatments was significantly greater than in the other treatments during this time. Our results are different from those reported by Davies (1986) who found only one peak of fruit abscission at 28-42 days after anthesis in 'Tifblue', 'Woodard' and 'Bluegem'.

Final fruit set was significantly lower in the nonpollinated treatment compared to the pollinated and GA3 treatments (Table 3-1). There was no difference in final fruit set among any of the GA3 or pollinated treatments. These results are difficult to compare with previous experiments since application timing was different. However, Davies and Buchanan (1979) reported that GA3 increased fruit set in field grown rabbiteye blueberry compared to the naturally pollinated control when GA3 was applied at 30-40% bloom and 15 days later or when a single application of GA3 was applied at 60% bloom. Similarly, NeSmith and Krewer (1991) observed an increase in fruit set in 'Tif blue' rabbiteye blueberry with a single application of 0.9 mM GA3 at balloon stage (just before opening) or 2 applications at full bloom and 10 or 18 days later. In both cases, set of the naturally pollinated control was significantly less than that obtained by hand pollination in the present experiment.

Fruit fresh weight was greatly reduced by all GA3 treatments (Table 3-1). Fruits from the pollinated control were four times the size of those from GA3 treatments. Unlike the results reported by Davies and Buchanan (1979), where increased fruit set and yield tended to reduce fruit fresh weight, in the present







39

experiment neither fruit set nor total yield were correlated with smaller fruit size. Fruit set in the pollinated treatment was as high as fruit set in the GA3 treatments, while total yield was double that of GA3-treated fruits. The initial number of flowers was smaller for pollinated treatments (avg. 48 flowers/plant) than for GA3 treatments (avg. 73 flowers/plant). The decrease in fruit size of the GA3-treated fruits may be due to the increased fruit load in these treatments. However, calculations of fruit number within each treatment reveal no significant differences in fruit number between the pollinated and several of the GA3 treatments (1 application of 0.6 mM or 1.4 mM GA3, 4 applications of 0.4 mM GA3), yet fruit weight was still significantly less in all of the GA3 treatments. Seed number strongly affects blueberry fruit size (Moore et al., 1972; Lyrene, 1989; Payne et al., 1989) and parthenocarpic blueberry fruits are smaller than seeded fruits (Mainland and Eck, 1969b). Although seed number was not measured in this experiment, absence of seeds was observed in all fruits from GA3 treatments.

Total yield, measured as fruit times fruit fresh weight, was less for GA3 treatments compared to pollinated treatments. Yield is a function of flower number, fruit set and fruit weight. In this case, fruit weight appears to be the determining factor in total yield. This is supported by a high correlation between yield and fresh weight (r=.71, p_5.01), while correlations between yield and fruit set or yield and flower number were much lower (r =0.49, p < .01 and r=-0.02, p_5.91, respectively).







40
GA3 treatments increased the fruit development period by 34 to 49 days compared to the pollinated control. Lower GA3 rates appeared to have a slightly more negative effect than higher rates, although there were no statistical differences between them. A delay in harvest date has been reported as a result of gibberellin applications to blueberry (Mainland and Eck, 1969, Vanerwegen and Krewer, 1990), but not to such an extent as observed in the present experiment. In blueberry, fruit development period and fruit seed number are negatively correlated (Lang and Danka, 1991 ; Lyrene, 1989). In the present experiment, GA3-treated fruits were mostly seedless and this probably accounts for the increased fruit development period.

There were no differences in soluble solids and pH among GA3 treatments (Table 3-1). However, pollinated fruits had significantly higher soluble solids than GA3 treatments, with no difference in pH. Similar results were obtained by Davies and Buchanan (1979) and Vanerwegen and Krewer (1990). No explanation for this observation has been offered. Crane and Grossi (1 960) reported that fig sweetness decreased as GA3 concentration increased, and they attributed this to the promotory effect of GA3 on elongation of new shoots that competed with fruits for assimilates.

The decreased fruit weight and, as a consequence, the decreased yield observed in the GA3-treatments compared to the pollinated treatment, may be related to the competitive ability of the different sinks to attract assimilates. In the present experiment, all treatments were applied on the same plant, and as







41

Goldwin (1984) observed, pollinated fruit are stronger sinks than parthenocarpic fruits and thus would accumulate more assimilates. This is supported, in part by the increased soluble solids content in pollinated fruit compared to GA3treated fruits. A decrease in soluble solids in fully colored seedless, GA3-treated 'Delaware' grapes was also reported by Clore (1965). Kishi et al. (1958), cited by Clore (1965), suggested that GA3-treated fruits must be harvested when they show a "deeper" color than the controls in order to have the same amount of soluble solids.

Experiment II

Extent of fruit abscission was lower for GA3 treatments than for pollinated treatments throughout the fruit development period (Fig. 3-3). In GA3 treatments, about 20% of the fruit abscised during the first 20 DAFB, after which very little abscission occurred. Maximum fruit abscission in pollinated and nonpollinated fruits occurred within 10 DAFB, reaching 66% and 100% for pollinated and nonpollinated treatments, respectively.

Final fruit set was increased significantly by multiple GA3 applications compared to pollinated controls (Table 3-2). While percent fruit set induced by GA3 was basically the same as in the previous experiment, fruit set of the pollinated control was comparatively lower.

Average fruit size in the pollinated treatment was twice as great as fruit size of the GA3 treatments. This may be related to fruit load as has been observed in other fruit crops (Coggins et al., 1960; Forshey and Elfving, 1977;







42
Garcfa-Martfnez and Garcfa-Papf, 1979). By 20 DAFB, 50% fruit abscission occurred in the pollinated treatments, while only 20% abscission occurred in the GA3 treatments. In fact, fresh weight and fruit set were negatively correlated (r=-0.65, p_.01). Davies and Buchanan (1979) found that GA3 treatments that increased fruit set and yield tended to decrease fruit weight in rabbiteye blueberry. The smaller fruit size in GA3 treatments may also be related to absence of seeds, as discussed previously. Total yield was similar among all treatments. Low fruit set in pollinated controls was compensated for by a higher fruit fresh weight, a determinant factor of yield in this case.

A delay in harvest of about 10 days was observed with 2 applications of 0.3 mM GA3 and four applications of 0.4 mM GA3 (applied at 24 and 42 DAFB, respectively) compared to the pollinated control. However, there was no difference in the fruit development period between 2 applications of 0.7 mM GA3 and the pollinated control. The different effects of GA3 treatments may be related to the concentration of GA3 used. Mainland and Eck (1968) found that in highbush blueberry, fruit development period was 17 days shorter in fruits treated with 1.4 mM KGA3 compared to the pollinated treatments, but fruits treated with lower concentrations of KGA3 ripened much later than pollinated fruits. Vanerwegen and Krewer (1990) also found a delay in maturation with 2 applications of 0.9 mM GA3 compared to one application at the same rate.

There was no difference in soluble solids and titratable acidity among treatments (Table 3-2). Juice pH tended to be lower as the GA3 concentration







43

and/or application number increased. Lower pH may be related to maturation, but the differences in pH do not correspond to the values for titratable acidity or soluble solid contents, i.e., low pH usually corresponds with higher values of titratable acidity and lower soluble solids. In general, soluble solids and titratable acidity were within the ranges reported by Spiers (1981) for rabbiteye blueberry.

Results from Experiments I and II were different in several respects, including fruit set, fruit weight, total yield, and fruit development period. Higher fruit set was observed in pollinated fruits in Experiment I than in Experiment II, but fruit set was about the same in GA3-treated fruits in both experiments. Fruit fresh weight was higher in all treatments in Experiment II than in Experiment I. Furthermore, fruit weight in the pollinated treatment was 4-fold greater than fruit weight in the GA3 treatments in Experiment I, but only 2-fold greater in Experiment II. Consequently, total yield was significantly increased in the pollinated treatment compared to the GA3 treatment in Experiment I, but not in Experiment II. Finally, fruit development period was longer for all treatments in Experiment I compared to Experiment II, with GA3treated fruits having a more delayed maturation period than pollinated fruits.

Several factors may account for some of these differences. All treatments were applied to the same plant in Experiment I and this may explain the relatively greater decrease in weight of GA3-treated fruits compared to pollinated fruits. Since pollinated fruits are stronger sinks than parthenocarpic







44

(i.e., GA3-treated) fruits (Goldwin, 1984), pollinated fruits would outcompete GA3-treated fruits for assimilates. Because fruit size was the dominant factor in yield in the present experiments, this would also explain the differences in yield between the pollinated and GA3 treatments in the 2 experiments. The decrease in fruit set in the pollinated treatment and the decreased FDP in all treatments observed in Experiment II relative to Experiment I may be partially due to differences in temperature conditions between the 2 experiments. Temperature may affect both fruit set and fruit development period in several fruit crops (Batjer and Martin, 1965; Jackson and Coombe, 1965). Day/night temperatures of 26/210C compared to 26/10�C decreased fruit set 23% in rabbiteye blueberry, but had no effect on fruit development period (Williamson et al., 1994). In peach, high night temperatures (21�C) decreased the fruit development period compared to 13'C night temperatures (Batjer and Martin, 1965). Thus the higher day/night temperature in Experiment II (25/220C) compared to Experiment 1 (17/160C) may account for the decreased fruit set in the pollinated treatment and the decreased fruit development period in all treatments. Differences in root system due to differences in plant age and container size may also account for some of the observed differences between Experiment I and Experiment I1. Cytokinins are synthesized in the roots and exert a strong influence on cell division in all organs, including shoots and fruits (Torrey, 1976). Thus, differences in cytokinin biosynthesis and translocation to the shoots may have occurred in the two experiments.









Experiment III

Fruit abscission was minimal for X, X + GA, 1/ X + GA, and S + GA from 0 to 28 DAFB, averaging 5% for all treatments (Fig. 3-4). This was followed by a slow increase in fruit abscission from 28 to 42 DAFB, after which fruit abscission ceased. In contrast, fruit abscission in the X and S treatments increased slowly from 0 to 28 DAFB, followed by a dramatic increase in fruit abscission from 28 to 42 DAFB. This pattern of fruit abscission resembles the one-wave abscission pattern obtained by Davies and Buchanan (1979) for 'Woodard', 'Bluegem', and 'Tifblue' under field conditions.

There were no differences in final fruit set among X, X + GA, Y4 X + GA and S + GA treatments, with fruit set averaging about 80% (Fig. 3-5). Thus, GA3 applications did not interact negatively with pollen source and decrease fruit set as has been observed in other crops like grape (Weaver and Pool, 1991) and apple (Edgerton, 1981). Final fruit set of %X and S treatments was significantly less (p_ 0.05) than the other treatments, as well as between each other. Fruit set in 4X treatment was fairly good, averaging 60%, while set in S treatments averaged less than 20%. Similar fruit set for self-pollinated rabbiteye blueberry was reported by EI-Agamy et al., (1982), Lyrene and Goldy (1983) and Garvey and Lyrene (1987). The relatively high fruit set under conditions of 'AX pollination suggests that adequate yields may be obtained in the field with sub-optimal pollination.








46
There were no differences in berry weight among X, X + GA, S, and S + GA treatments (Table 3-3). Fruits from X and X + GA treatments had the highest weight, averaging 1.7 g. Self-pollination alone did not decrease fresh weight, in contrast to reports by Coville (1921) and Meader and Darrow (1944) in highbush blueberry, and Lyrene (1989) in rabbiteye blueberry.

Fruits from 'AX and '/ X + GA treatments were the smallest, averaging 1.3 g. Thus, although a sub-optimal (i.e., saturation) pollination does not appear to limit fruit set, it does negatively affect fruit size, which would be expected to decrease total yield.

The number and weight of large seed per fruit was greatest in the crosspollinated (X) treatment compared to the other treatments (Table 3-3). There were no differences in large seed number or weight between X + GA and 'A X treatments. Seed number and weight were significantly less in the 4 X + GA and the S + GA treatments compared to X, X + GA and 1/4X treatments. The number and weight of small seeds per fruit were not statistically different among treatments. There were no differences in total seed number among treatments, but total seed weight per fruit in the cross-pollinated (X) treatment was significantly higher than the other treatments. There was a positive correlation between fresh weight and total seed number (r=0.60, p<.01), as previously reported by Meader and Darrow (1944), Alders and Hall (1961), and Darnell and Lyrene (1989). There was also correlation between fresh weight and number of small seeds per fruit (r=0.55, p<.01). A weaker correlation







47

between fresh weight and number (r=0.45, p<.01) and weight of large seeds (r=0.42, p<.01) was observed.

A comparison of the X and X + GA treatments, and the X and %X + GA3 treatments indicates that GA3 inhibits the development of large seeds in previously pollinated flowers. GA3 may be acting as a pollenicide, as reported by Weaver and McCune (1960) in grape flowers. Alternatively, since GA3 was applied 4-6 h after pollination, there is a possibility of a pollen wash-off effect. In either event, however, the number of small seeds as well as the total number of seeds were not affected by any of the GA3 treatments.

The fruit development period was similar among all treatments except for the self-pollinated (S) (Table 3-3). Eighty percent of the fruit in X, X + GA, and X+GA treatments was harvested in the period 63-72 DAFB (data not shown). Harvesting was delayed 15 days in the X and S + GA treatments and only 43% of the S fruits were harvested in the period of 72-92 DAFB (data not shown). Seed number and fruit development period are closely related. Alders and Hall (1961) reported that in lowbush blueberry, increased seed number was significantly associated with increased fresh weight and shorter fruit development period. Coville (1921) found similar results in highbush blueberry, as well as Lyrene (1989) in southern highbush and Meader and Darrow (1944) and Tamada et al. (1977) in rabbiteye blueberry. In this experiment, as seed number decreased, fruit development period increased (r=-.64, p_5.01).







48
Our results indicate that under the conditions of these experiments, GA3 does not interact negatively with any type of pollination tested. Moreover, under situations of partial pollination and/or self-pollination, GA3 can successfully replace the stimuli for fruit set delivered by normal pollination and fertilization in rabbiteye blueberry.

Conclusions

Maximum fruit abscission occurred during the early stage of fruit development, presumably the stage when cell division is occurring. The application of GA3 alone or in combination with pollination prevented fruit abscission compared to the nonpollinated and self-pollinated treatments, suggesting that GA3 can replace the stimulus for fruit set delivered by pollination and fertilization. Multiple applications of GA3 did not change the pattern of fruit abscission compared to single applications or pollination. The lack of response to increasing GA3 concentrations and/or number of applications indicate that factors other than gibberellins may be involved in the fruit set of rabbiteye blueberry.

Fruit size was decreased by GA3 treatments alone compared to pollinated treatments, but GA3 application in combination with full cross- or self-pollination did not affect negatively fruit size in 'Beckyblue'. The decreased fruit size obtained by %X pollination was partially reversed by GA3. Thus, the combination of GA3 with partial pollination can apparently provide a similar







49

stimulus as full pollination for growth after flower opening, while GA3 alone cannot.

An increase in the fruit development period was observed when GA3 was applied alone compared to pollinated treatments. However, no such negative effect was observed when GA3 was applied in combination with any pollinated treatments. This suggests that under optimal pollination conditions, no detrimental effects of GA3 treatments on fruit set, fruit size, or fruit development period are to be expected. Furthermore, GA3 applications could be beneficial under suboptimal pollination conditions, since the presence of some seeds may increase fruit size and decrease fruit development period.

Soluble solids, titratable acidity and pH were not consistently affected by GA3 treatments. These aspects need more careful consideration. A decrease in soluble solids has been observed in, other studies as one of the negative effects associated with GA3 applications. No explanations have been advanced as to the basis for this observation. It is possible that harvest criteria, usually color, may need modification if GA3 is used.







50
100

POLL 0.6 mM GA 3 1.4 mM GA 3 NP
80

/
60 /


40


g 20


"Z 0 to
"0
CU,
- I POLL 0.2 mM GA3~ (3x) 0.5 mM GA 'N
UL 80 -4.

/
60 /
/

40

00
207


0
0 20 40 60 80 Days after anthesis Figure 3-1. Time-course and extent of fruit abscission in 'Beckyblue' rabbiteye
blueberry as affected by gibberellic acid (GA3), pollination (POLL), or nonpollination (NP). 3x=3 applications of GA3; application times indicated by arrows. Values are means � SE, n =3; SE bars present only
when larger than symbol.









100


80


60 /
/

40


~20

"50

n I
3POLL 0.1mMGA(4x) 0.4mM GA3(4x) NP "_ 80


/
60 /
/

40




20 X-AiriIX

0 20 40 60 80 Days after anthesis Figure 3-2. Time-course and extent of fruit abscission in 'Beckyblue' rabbiteye
blueberry as affected by gibberellic acid (GA3), pollination (POLL), or nonpollination (NP). 2x and 4x = 2 and 4 applications of GA3; application times indicated by arrows. Values are means � SE, n = 3; SE bars present
only when larger than symbol.







52


0.7 mM GA3 (2x) 0.3 mM GA3(2x) 0.4 mM GA3(4x)
120 - . ...... -- -,POLL NP



100



0 80
0
Co


60
LL1 .00
=- I LII


40




20





0 10 20 30 40 50 Days after anthesis Figure 3-3. Time-course and extent of fruit abscission in 'Beckyblue' rabbiteye
blueberry as affected by gibberellic acid (GA3), pollination (POLL), or nonpollination (NP). 2x and 4x = 2 and 4 applications of GA3; application times indicated by arrows. Values are means � SE, n =9; SE
bars present only when larger than symbol.













100
1001... A ----* - ---*.80





0
Ci



40 /




0
.o6 - a U
40i














0 14 28 42 56 Days after anthesis
Figure 3-4. Time-course and extent of fruit abscission in 'Beckyblue' rabbiteye
blueberry as affected by gibberellic acid (GA3), pollen source, and pollen amount. X =cross-pollination (200 pollen tetrads); 1/4X=50 pollen tetrads; S =self-pollination (200 pollen tetrads). GA3=0.7 mM GA3; GA3 application times indicated by arrows. Values are means �SE, n =9; SE
bars present only when larger than symbol.








54




100





80




4-0
(D60

4
C



40





20





0_
X X+GA3 1/4X+GA3 1/4X S S+GA3 Treatments Figure 3-5. Effect of GA3, pollen source, and pollen amount on final fruit set of
'Beckyblue' rabbiteye blueberry fruits. X =cross-pollination (200 pollen tetrads); 4X = cross-pollination (50 pollen tetrads); S =self-pollination (200 pollen tetrads); GA3 =0.7 mM GA3 applied at FB and FB+7DAFB.
Values are means�SE, n=9; SE bars present only when larger than
symbol.








Table 3-1. Effect of GA3 on fruit set and development in 'Beckyblue' rabbiteye blueberry in 1992.


Treatment Flower Fruit set FW Total yield FDPz SSY pH no./plant (%) (g/fruit) (g/plant) (days) ('Brix)
1 applicationx
0.6 mM GA3 71.0�20.7 62.8wvb .492 b 28.2 b 134.6 a 10.4 b 3.7 a 1.4 mM GA3 57.0� 18.8 74.6ab .532 b 25.7 b 126.4ab 11.4 b 3.9 a
2 applications
0.3 mM GA3 77.0 �20.2 77.6 ab .514 b 31.9 b 131.4ab 10.6 b 3.6 a 0.7 mM GA3 72.2� 17.3 77.8 ab .464 b 26.9 b 120.8ab 11.7 ab 3.7 a
3 applications
0.2 mM GA3 75.8� 15.4 79.8 ab .566 b 36.0 b 129.8ab 10.6 b 3.6 a 0.5 mM GA3 98.8 �22.9 84.6 ab .420 b 36.2 b 124.4ab 10.5 b 3.6 a
4 applications
0.1 mM GA3 81.2�20.3 77.0 ab .516 b 31.4 b 126.8ab 11.7 b 3.9 a 0.4 mM GA3 53.8�5.2 85.8 ab .464 b 22.1 b 119.8 b 12.3 ab 3.6 a Hand pollinated 48.2 � 6.0 78.4 ab 1.670 a 79.0 a 85.0 c 13.4 a 3.7 a Non-pollinated 52.0 � 11.1 29.0 c - - - zFDP=fruit development period. Y SS= soluble solids. XGA3 application times: 1 application - full bloom (FB); 2 applications - FB and FB+7 days after full bloom (DAFB); 3 applications - FB, FB+7 DAFB, FB+7+21 DAFB; 4 applications - FB, FB+7 DAFB, FB+7+21 DAFB, FB+7+21 +42 DAFB. WData were arcsin transformed prior to analysis.
VMean separation within a column by LSMeans's test, p =0.05











Table 3-2. Effect of GA3 on fruit set and development in 'Beckyblue' rabbiteye blueberry in 1993.

Treatments Flower Fruit FW Total yield FDPz SS" TA(%)x pH no./plant set (%) (g/fruit) (g/plant) (days)

2 applicationsw
0.3 mM GA3 32.9�4.7 76.2Vua 1.18b 30.3a 64.4a 11.6a 0.69 a 3.lab 0.7 mM GA3 28.2�5.5 74.8a 1.19b 25.0a 57.8b 12.0a 0.98 a 3.Ob
4 applications
0.4 mM GA3 34.8 � 6.6 68.8a 1.11 b 26.4a 66.1 a 1 2.6a 0.89 a 2.9b Hand pollinated 34.8�3.7 43.Ob 2.04a 23.7a 55.6b 12.1a 0.69 a 3.2a Nonpollinated 26.1 �2.5 3.3c - - ZFDP=fruit development period.
YSS= soluble solids.
XTA = titratable acidity.


WGA3 application times: 2 applications - Full bloom (FB) FB+7 DAFB, FB+7+21 DAFB, FB+7+21 +42 DAFB. VData were arcsin transformed prior to analysis.


and FB+7 days after full bloom (DAFB); 4 applications FB,


UMean separation within columns by LSMean's test, p =0.05.














Table 3-3. Effect of GA3, pollen source, and pollen in 'Beckyblue' rabbiteye blueberry.


amount on fruit fresh weight (FW) and seed number/seed weight


Treatmentz FW FDP Seed No. Seed weight (mg) Total Total (g/fruit) (days) seed no. seed Large Small Large Small weight
(mg)
X 1.69Yxab 72 a 10.2 a 29.1 a 7.4 a 9.7 a 39.4 a 15.2 a X+GA3 1.82a 72 a 3.5 b 28.2 a 2.0 b 8.3 a 31.7 a 9.4 b 1/4X 1.15c 87 a 3.3 b 23.9 a 2.8 c 5.2 a 27.2 a 7.2 b 1/4X+GA3 1.43bc 72 a 0.9 c 26.6 a 0.5 c 4.3 a 27.6 a 3.9 b S 1.38abc 92 b 2.0 bc 38.2 a 1.2 bc 2.8 a 40.2 a 3.9 b S+GA3 1.57ab 87 a 0.3 c 27.6 a 0.2 c 6.3 a 27.9 a 6.8 b zMeans of five harvest pooled for all parameters.
YTreatments: X=cross pollination with 'Climax' pollen (200 pollen tetratads), 1/4X= 50 pollen tetrads, S=self pollination with 'Beckyblue' pollen (200 pollen tetrads), GA3 = 0.7 mM GA3 applied at full bloom (FB) and FB + 7DAFB. xMean separation within columns by LSMean's test, p =0.05.













CHAPTER 4
CARBOHYDRATE METABOLISM AND FRUIT GROWTH IN GIBBERELLIC
ACID-TREATED VS POLLINATED FRUITS OF BLUEBERRY



Introduction


Parthenocarpic fruit set and development can be induced in blueberry by exogenous applications of gibberellic acid (GA3); however, final fruit size of parthenocarpic fruit is often smaller than that of pollinated fruit. The mechanism by which GA3 induces fruit set and development is unknown. It has been observed, however, that gibberellins, as well as other plant hormones, alter the pattern of assimilate distribution within the plant (Daie, 1985; Kinet et al., 1985; Aloni et al., 1986). Changes in assimilate distribution, specifically, enhancement of carbohydrate accumulation at the site of hormone application, have been correlated with fruit growth and development in a number of species (Marr6 and Murneck, 1952; Stembridge and Gambrell, 1972; Adedipe et al., 1976; Craighton et al., 1986; Peret6 and Beltr~n, 1987; Stutte and Gage, 1990).

Sucrose is the major translocated carbohydrate in most plants, including blueberry (Avigad, 1982). The extent of sucrose translocation to fruit depends, in part, on the concentration gradient between the loading and the unloading







59
sites (i.e., developing fruit). Factors that decrease the sucrose concentration in the unloading area may, in theory, enhance sucrose transport to that site (Aloni et al., 1986; Daie et al., 1986; Miyamoto and Kamisaka, 1988, 1990; Morris and Arthur, 1985).

The metabolism of sucrose in fruit tissues can be initiated by three enzymes that are temporally and spatially compartmentalized, depending on species. These enzymes are sucrose phosphate synthase (SPS, EC 2.4.1.14), which is involved in sucrose synthesis; invertase (EC 3.2.1.26), which hydrolyses sucrose to glucose and fructose; and sucrose synthase (SS, EC 2.4.2.13), which is thought to function primarily in a sucrose degrading direction. Changes in activities of these enzymes have been associated with developmental processes in sink organs (McCollum et al., 1988; Hubbard et al., 1989, 1991; Nielsen et al., 1991; Yelle et al., 1991; Stommel, 1992; Sin et al., 1992; Wang et al., 1993). In fact, increases in invertase activity have been reported to precede growth of pea ovaries (Estruch and Beltr~n 1991), stem segments (Morris and Arthur, 1984), lima bean embryos (Xu et al., 1989), and developing fruits of Cucumis (Schaffer et al., 1987a). Plant hormones may exert their effects on carbohydrate accumulation and subsequent fruit development by affecting activities and/or levels of one or more of these enzymes. For example, auxins have been reported to stimulate both invertase activity (Morris and Arthur, 1986; Pooviaiah, 1985; Schaffer et al., 1987b; Tanaka and Uritani, 1979) and its de novo synthesis (Gordon and Flood, 1980;







60

Pressey and Avants, 1980) in a number of tissues. Increased invertase activity has also been observed in soybean fruit following abscissic acid application (Akerson, 1985). Gibberellic acid stimulated invertase activity in pea internodes (Broughton and McComb, 1971; Morris and Arthur, 1985), sweet potato roots (Tanaka and Uritani, 1979), and Avena segments (Kaufman et al., 1973). In the orchid Phalaenopsis, GA3 stimulated sucrose synthase activity (Chen et al., 1994). There are no reports on plant hormones affecting activities or levels of SPS.

In the present study, developmental changes in activities of sucrosemetabolizing enzymes and carbohydrate accumulation in fruit were assessed and correlated with growth obtained by exogenous application of GA3 or pollination. The objective was to determine if differences in fruit growth could be attributed to differences in enzyme activities and carbohydrate accumulation in fruit.

Materials and Methods

Plant Material

Softwood cuttings of 'Beckyblue' rabbiteye blueberries were rooted in Spring 1990 and grown outdoors in 22L pots in a 1:1 peat:pine bark mix. Plants were watered every other day and fertilized with 20N-5.6P-1 1 K water soluble fertilizer once a week. In December 1991, 9 uniform plants were transferred to a dark cooler and chilled at 7 � 1�C for 30 days. After the







61

chilling period, plants were placed in a greenhouse with average day/night temperatures of 27/16'C to force budbreak. Treatments

Flower clusters were thinned to 4 to 5 florets/cluster at bloom, removing the most developed and least developed florets. The total number of florets ranged from 740 to 920 florets/plant. Treatments consisted of hand-pollination at bloom with 'Climax' pollen (a cross-compatible cultivar), application of 0.7 Mm GA3 (Pro Gibb 4%, Abbott Laboratories, Chicago, ILL, prepared in Mclllvaine buffer [Dawson et al., 19861 pH 3.5, and .01 % Tween-80) sprayed at bloom and again 7 DAFB. The nonpollinated treatment consisted of a spray application of buffer and surfactant at bloom. Treatments were arranged in a randomized design with 3 single plant replications per treatment. From 0-45 DAFB, data from the 3 treatments were analyzed by ANOVA. Following the abscission of nonpollinated fruits at 45 DAFB, data from the pollinated and GA3 treatments were analyzed by 1-way ANOVA.

Three sub-samples/plant were taken at 0, 2, 4, 10, 24, and 45 DAFB for all treatments. Subsequent to the abscission of the nonpollinated fruits (45 DAFB), data were taken on the remaining two treatments based on phenological development. Data were taken at the end of period II of fruit growth (estimated to be 45 DAFB for pollinated fruits and 72 DAFB for GA3), at fruit color break (65 DAFB for pollinated fruits and 80 DAFB for GA3-treated







62
fruits), and at ripening (72 DAFB for pollinated fruits and 87 DAFB for GA3treated fruits.

Fruit Carbon Dioxide Exchange

Fruit growth was determined for each treatment by estimating carbon (C) cost (fruit DW gain plus respiratory loss) throughout development. The C supply was broken down into C supplied by fruit photosynthesis and C imported into the fruit. This was done to determine if differences in C source existed among treatments. Fruit respiration and photosynthesis were determined by measuring net CO2 exchange of attached flowers and fruits in the laboratory under light (1000 /umol m2s1 photosynthetic photon flux [PPF] emitted by a 400-W high-pressure sodium vapor lamp [Lucalox]) and dark conditions.

At each sampling, the cluster (two to 4 florets/fruits) was enclosed in a 104 ml plexiglass chamber and flushed for 20 min with ambient air at 1 L m1. The chamber was then closed and fruit photosynthesis was quantified by monitoring the decrease in CO2 concentration after 30 min in the closed system. Dark conditions were established by wrapping the chamber with aluminum foil and black plastic. After air flushing for 20 min, the chamber was closed and respiration was determined after 30 min (unripe fruits) and 5 min (ripe fruits) by measuring the increase in CO2 concentration. Temperatures were maintained at 25�20C and 21 � 1�C light/dark conditions, respectively. Gas samples (500 pl) were taken with a 1 ml syringe and CO2 was analyzed by







63

gas chromatography (Fisher Scientific, Pittsburgh; Model 1200), using an 80 to 100 mesh Colmpak PQ column, injector and column temperatures of 600C, and a thermal conductivity detector.

After taking fruit CO2 exchange measurements, fruits were harvested and FW recorded. Samples were then lyophilized to constant weight for dry weight determination.

Calculations of Carbon Budget

CO2 exchange data in the light and dark were extrapolated to a per-day basis assuming a 12-h photoperiod, saturating PPF (1000 pmol.m2.s'), and mean temperatures of 25/210C (day/night). The following calculations were derived from the equations established by Birkhold et al., (1992).

The average daily C cost for a given stage of fruit development was calculated from fruit C accumulation plus respiratory C loss. Daily fruit C accumulation was determined from fruit DW measurements. Previous work indicates that C content of blueberry fruits averages 43% of fruit DW throughout development (Birkhold et al., 1992). Thus the average daily fruit C accumulation was calculated by:



Mean daily C accumulation =

(0.43 DW(t2) - 0.43 DW(tl)/ t2-t1 [11








64

Daily fruit C respiratory loss was calculated by multiplying the hourly dark CO2 exchange rate by 24. The average daily fruit C respiratory loss between times tj and t2 was done by interpolation between the two measurement points and was calculated by:



Mean daily respiratory C loss =

[daily resp C(tl) + daily resp Ct2)]/2 [21



Thus, the average daily C cost for a given stage of fruit development was the sum of the mean daily C accumulation and mean daily respiratory C loss.

The average daily C supply is composed of fruit photosynthetic C gain plus imported C. The daily fruit photosynthetic contribution was calculated as the difference between net CO2 exchange under light and dark conditions and multiplied by the 12 h photoperiod. The average daily fruit photosynthesis between times tj and t2 was done by interpolation between the two measurement points and was calculated by:



Mean daily fruit PS C =

[daily fruit PS C (t) + daily fruit PS C (t2)/2 [3]



The average daily imported C was determined by subtracting the average daily fruit photosynthetic C from the average daily C cost.











Sugar Determination

Fruit samples were collected on the sampling dates previously indicated and stored at -800C until analysis. Frozen samples were ground in liquid N2 and extracted in boiling 80% ethanol (1:10 w/v) for 1 min. Mannitol (100 mg) was added as an internal standard. Extracts were centrifuged at 600g, the supernatant decanted, and the pellet was re-extracted twice. The combined supernatant was partitioned against chloroform, and the aqueous fraction was vacuum dried, resuspended in water, and passed through Dowex-1 and Dowex50 ion resins. The final supernatant was dried under vacuum, resuspended in HPLC water and filtered through a 0.45 um filter. Samples were injected into a Bio-Rad HPLC (Richmond, Calif) and sucrose, glucose, and fructose were determined using a Bio-Rad HPX 87C cation-exchange column at 85�C and flow rate of 0.6 ml water/min.

Enzyme Extraction

Frozen blueberry tissue (500 mg) was ground in liquid N2, transferred to a chilled mortar containing 5 ml of extraction buffer (50 Mm MOPS, pH 7.5; 5 Mm MgCI2, 2.5 DTT, 1 Mm EDTA, and 10% w/w PVPP), mixed thoroughly, and vacuum filtered. For SPS and SS assays, crude extract was desalted through a chilled 5-ml Sephadex column equilibrated with extraction buffer without EDTA, and centrifuged for 1 min at 200g. For soluble invertase assays, aliquots of crude extract were dialyzed (25,000 mol wt cutoff) for 24 h at 40C







66
against 5 Mm K2HPO4 (pH 7.5). The insoluble pellet was rinsed several times with 200 Mm K2HPO4, vacuum filtered, and used to measure insoluble acid invertase activity.

Enzyme Assays

SPS and SS activities were determined according to the methods of Hubbard et al., (1989) with slight modifications. SPS activity was determined in a 70pl reaction volume containing 50 Mm MOPS (pH 7.5), 14 Mm Mg Cl2, 10 Mm Fru6P, 50 Mm GIc6P, 7 Mm UDPGIc, and 45 pl extract. Mixtures were incubated for 30 min at 30C, terminated with 70 pl 30% KOH, followed by boiling for 15 min. Sucrose production was quantified by the anthrone method (Van Handel, 1968). SS was determined in the degradative direction. Two hundred microliters of desalted extract were assayed in a 500 pl volume containing 200 Mm MES (pH5.5), 3 Mm NaF, 5 Mm UDP, and 50 Mm sucrose. Reactions were terminated after 30 min with 500 pl 100 Mm TRIS (pH 8.7) + 0.2% EDTA followed by boiling. UDPG production was quantified by measuring the UDPG-dehydrogenase-specific synthesis of NADH (Lowell et al., 1989). Enzyme controls in both assays were treated as described, but terminated at 0 min. Boiled controls were avoided due to floculation.

Soluble and insoluble acid invertase activities were assayed in a 500 pl volume consisting of 2 Mm acetic acid (pH 4.5), 100 Mm sucrose, and 200 pl extract (or 20-50 mg insoluble pellet). Reactions proceeded for 15 min at 30C and were terminated by adding 250 pl of 400 mM K2PO2, pH 7.5 and boiling for







67

2 min. The same reaction medium, but adjusted to pH 7.5 with 5 mM K2PO4, was used to assay neutral invertase activity. Enzyme controls contained all assay reagents, but were terminated at 0 min. Glucose production was determined by the glucose oxidase method (Sigma).

Results

Fruit Growth

There was no difference in fruit dry weight accumulation between GA3treated and pollinated berries from 0 to 10 DAFB (Fig 4-1 A). From 25 DAFB to ripening, GA3-treated fruit accumulated significantly less dry matter than pollinated fruits. Pollinated fruits accumulated an average of 390 mg DW while GA3-treated fruits accumulated an average of 180 mg DW throughout development. Dry weight of nonpollinated fruits did not change throughout fruit development. Fruit fresh weight accumulation followed the same pattern as that of dry weight (Fig. 4-1B).

Although blueberry fruit growth is reported to follow a double sigmoid curve (Birkhold et al., 1992), the sampling times used in the present experiment were too few to clearly show this. Nonetheless, there seemed to be 3 periods of growth: period I lasted about 24 days in both GA3 and pollinated treatments; period II in GA3-treated fruits extended from 24-72 DAFB, while in pollinated fruit, period II was significantly shorter, appearing to extend from 25 to 45 DAFB; period III was 12 days shorter in GA3-treated fruits (72-87 DAFB) compared to pollinated fruits (45-72 DAFB).







68
Relative growth rate (RGR) of pollinated fruits was greater than that of GA3-treated fruits throughout much of development (Fig. 4-2). The maximum RGR for both treatments occurred at the end of period I. This was followed by a phase of sharply declining growth rate during period II in both treatments. A second peak just before ripening was observed for both GA3-treated and pollinated fruits.

Fruit Carbon Dioxide Exchange

Developmental patterns of fruit respiration were similar between GA3treated and pollinated fruits (Fig. 4-3). Respiration rates increased from 12 to 20 pmol.gFW-'.h-' between 0 and 10 DAB for GA3-treated fruits, then decreased to 5 pmol.gFW-l.h-1 by 24 DAFB and remained constant thereafter. A similar increase in respiration rates was observed in pollinated fruits between 0 and 4 DAFB, declining to 12/mol.gFW-1.h-1 at 10 DAFB and then to 5 /mol. gFWl.h1 at 45 DAFB, with little change from 45 to 72 DAFB. Respiration rates in nonpollinated fruits averaged between 12 and 14/mol.gFW1.h1 from 0-45 DAFB and were significantly greater than rates of GA3-treated and pollinated fruits at 24 and 45 DAB.

Net photosynthesis occurred in fruit from all treatments throughout much of development (Fig 4-3). Fruit photosynthesis declined from 6pmol.gFW1.h-1 at 0 DAB to less than 1 /mol.gFW'.h1 by 24 DAFB, for both GA3-treated and pollinated fruits. Net photosynthesis ceased during period III of fruit growth in







69
both treatments. In nonpollinated fruits, photosynthetic rates were significantly higher than rates in GA3-treated and pollinated fruits from 4 to 45 DAB. Fruit Carbon Budget

The developmental pattern of the average daily fruit C cost (DW C + respiratory C) (Fig. 4-4) was similar to the developmental pattern of fruit DW gain for both GA3-treated and pollinated fruits. The C cost for nonpollinated fruit was negligible and is not depicted in Fig. 4-4. The average daily C accumulation (DW C) in pollinated fruits was greater than in GA3-treated fruits, and this difference was evident as early as the 2-4 DAB interval. Maximum rates of C accumulation reached 3.6 mg C fruit-l.day-1 in GA3-treated fruits at 72-87 DAFB, while in pollinated fruits, maximum rates reached 5.3 mg C fruit '.day1 at 45-72 DAFB. Overall, total DW C accumulation was 2.5-fold greater in pollinated fruits (180 mg C) than in GA3-treated fruits (73 mg C) (Table 4-1).

Daily respiratory C loss (Resp C) was maximum during the last interval of fruit growth for both GA3-treated and pollinated fruits (Fig.4-4). Absolute values were similar for both treatments throughout fruit development, increasing from 0.07 mg C fruit'.day at 0-4 DAB to 1 to 2 mg C fruit-.day-' during the final stage of development. Total respiratory C loss was 35 mg C for GA3-treated fruits and 64 mg C for pollinated fruits (Table 4-1). This represented 33% of the total C cost for GA3-treated fruits compared to 26% for pollinated fruits. Analysis of the data for individual fruits indicated that these percentages were not significantly different from each other.








70
The average daily fruit C cost (DW C + Resp C) for GA3-treated fruits increased from 0.06 mg C fruitl'.day1 to 4.5 and 7.1 mg C fruit'.day1 for GA3-treated and pollinated fruits, respectively, as development proceeded (Fig. 4-4). The total C cost (DW C + Resp C) for GA3-treated fruit was 109 mg C fruit1 while the C cost for pollinated fruit was 244 mg C fruit' (Table 4-1).

Daily fruit photosynthetic C supply reached its maximum between 24 and 45 DAFB in both GA3 and pollinated fruits, averaging 0.21 mg C fruit-'.day' in GA3-treated fruits and 0.32 mg C fruit'.day- in pollinated fruits (Fig. 4-5). Imported C was maximum during the last interval of fruit growth, reaching 4.5 mg C fruitl.day1 in GA3-treated fruits and 6.9 mg C fruitl'.day' in pollinated fruits. From 0 to 10 DAB, fruit photosynthesis supplied 51% of the fruit C requirement for GA3-treated fruits, and 36% of the C requirement for pollinated fruits (Table 4-1). Overall, fruit photosynthesis supplied 12% and 7% of the fruit C requirement for GA3-treated and pollinated fruits, respectively. Carbohydrate Accumulation

Hexoses. Glucose and fructose are the main soluble carbohydrates accumulated in blueberry fruits, and are present in equimolar amounts throughout fruit development. Thus, data are presented as total hexoses (Fig. 4-6). There were no differences in hexose concentration among treatments from 0 to 45 DAFB, nor between GA3-treated and pollinated treatments from 45 DAFB to ripening, when comparing phenological stages. Hexose concentration declined from 30 to about 4 mg.gFW1 in all treatments from 0







71
to 24 DAFB. Subsequently, there was a continuous increase in hexose concentration, up to about 100 mg.gFW1 in pollinated fruits by 72 DAFB (ripening). In GA3-treated fruits, hexose concentration remained constant from 24 to 72 DAFB, followed by a sharp increase to about 100 mg.gFW1 at 87 DAFB (ripening).

Sucrose. Sucrose accumulation was negligible throughout most of the fruit development and accounted for only a small fraction of the total soluble carbohydrate accumulation (data not shown). GA3-treated fruits accumulated

8.6 mg sucrose.gFW1 vs 4.4 mg.gFW1 in pollinated fruits.

Total carbohydrate content was correlated with both fruit FW and RGR. From 0-24 DAFB, carbohydrate accumulation was negatively correlated with dry weight gain (r=-.56, p<.01) and RGR (r=-.66, p_5.01), while from 45 DAFB to ripening, carbohydrate accumulation was positively correlated only with DW gain (r=.69, p< =.01).

Enzyme Activity

SPS and SS activities. In general, SPS activity was low and not significantly different among treatments, when phenological stages are compared (Fig 4-7A). The highest enzyme activity was reached during the final stage of fruit development in GA3-treated (3.7 pmol.gFW-1h1) and pollinated fruits (2.5 pmol.gFW1h-1). In nonpollinated fruits, maximum activity of 3.2 /pmol.gFW-1h-1 occurred at 10 DAFB. SPS activity was not correlated with fruit FW (r=.58, p< =.12), but was possitively correlated with non-structural







72
carbohydrate accumulation (r=-.78, p_<.05) during the final stage of fruit growth.

There were no consistent differences in SS activity among treatments (Fig 4-7B). SS activity remained low throughout fruit development, increasing to 4.3 and 2.1 pmol.gFWl'h1 in GA3-treated and pollinated fruits, respectively, during final fruit growth. In nonpollinated fruits, SS remained almost undetectable.

Invertases. There were no significant differences in activities of soluble acid or insoluble acid invertase between GA3-treated and pollinated fruits, when phenological stages are compared (Fig. 4-8). Soluble acid invertase activity was low throughout much of development, increasing to 65 pmol.gFW-1hlin GA3-treated fruit and 54jumol.gFW-1.h-1in pollinated fruits during the final stage of development (Fig. 4-8A). Nonpollinated fruits exhibited significantly reduced soluble acid invertase activity (0.12 pmol.gFW'h-') by 24 DAFB compared to GA3 (1.76 pmol.gFW'h-1) and pollinated treatments (2.39 pmol.gFW-1-h'). By 45 DAFB, no soluble acid invertase activity was detected in nonpollinated fruits.

There were no significant differences in insoluble acid invertase activity among treatments from 0 to 45 DAFB, except for a 2-fold increase in activity in nonpollinated fruits at 10 DAFB (Fig. 4-8B). This increase was not seen in pollinated or GA3-treated fruits. Enzyme activity in GA3-treated and pollinated fruits declined from about 16pmol.gFW-1.h-' at 4 DAB to about 3 /mol.gFW1h-1 at the beginning of the final stage of fruit growth. Acid insoluble invertase







73
activity increased at ripening, averaging 9.5,pmol.gFWl'h' in GA3-treated fruits and 5.1 pmol.gFW'h-' in pollinated fruits. In nonpollinated fruits, the declining pattern of enzyme activity was less dramatic compared to the other treatments, and at 45 DAB, when most of the fruits had abscised, activity of insoluble acid invertase was 50% of the activity observed at 0 DAB and 2-fold higher than the rest of the treatments.

There were no significant differences in neutral invertase activity among treatments (data not shown). Activity was low throughout most of fruit development (< 1.0 pmol.gFW-h'h-). During the final stage of fruit growth, activity increased to about 5 to 6 pmol.gFW-.h1 in both GA3-treated and pollinated fruits.

Discussion

Fruit Growth and Carbon Budget

Fruit growth in blueberry has been described as a double sigmoid pattern (Edwards et al., 1970; Hindle et al., 1947; Spiers 1952; Young 1952). However, the frequency of sampling in the present experiment does not allow clear determination of the three typical stages of growth, although 3 periods of growth can be described. The decreased final weight of GA3-treated fruits compared to pollinated fruits has been observed in most studies in which blueberry fruit set has been induced by GA3 applications (Mainland and Eck, 1968; 1969a; Davies and Buchanan, 1979; Vanerwegen and Krewer, 1990.







74

Decreased fruit weight in GA3-treated berries was due in part to a relatively slower growth rate during period II in GA3 ( from 24 to 72 DAFB) vs pollinated fruits (from 24 to 62 DAFB). The decrease in fruit growth during this period has been associated with an increase in the growth rate of the embryo and endosperm (Edwards et al., 1970). However, factors other than that must account for the lack of fruit growth in GA3-treated fruit, since pollination and fertilization were prevented and thus, seeds were absent in GA3-treated berries. It is unknown if seed development causes a decrease in pericarp growth, or if the pericarp ceases growth in order to allow seed development. It seems, in this particular case, that the signal for cessation of pericarp growth arose from the pericarp itself, since seeds were not present. It appears also that seeds or the products synthesized in them are necessary to sustain a continuous pericarp growth since GA3-treated fruits had a prolonged phase of slow growth compared to pollinated fruits. It is possible that once the seeds and embryos start developing, they synthesize gibberellins that are then translocated to the pericarp (Sastry and Muir, 1963). Alternatively, seed development may stimulate gibberellin synthesis in the pericarp, as has been found in pea fruit (Ozga et al., 1992). Another possibility is that gibberellins stimulate the synthesis of auxins and the concerted action of both hormones may influence cell wall properties and/or increase carbohydrate accumulation and water flow into the fruit and, in this way, stimulate growth (Coombe, 1960).







75

In general, blueberry fruit size has been positively correlated with seed number (Alders and Hall, 1961; Tamada et al., 1977; Darnell and Lyrene, 1989; Lyrene, 1989). Seed number also dramatically affects the fruit development period in blueberry (Lang, 1992; Williamson et al., 1994). Fruit with more seeds have a shorter fruit development period than unseeded fruits, and this was corroborated in this study.

Fruit CO2 exchange values were in general agreement with those obtained by Birkhold et al., (1991) in small- and large fruited-genotypes of rabbiteye blueberry. The transient increase in fruit respiration at 4 DAB and 10 DAB for pollinated and GA3-treated berries, respectively, may have been due to increased rates of cell division in these two treatments, since such an increase was not observed in nonpollinated fruits. During the early stages of development, fruits act as utilization sinks and respiration is their main metabolic activity (Ho, 1988). Frequently, the rate of import to these sinks parallels the respiration rate (Kallarackal and Milburn, 1985). In our case, the highest respiration rates occurred during the early stage of fruit growth I, when RGR is increasing; however, the peak of maximum respiration preceded that of maximum RGR. In peach, RGR has been correlated with fruit growth patterns, respiration, and source-sink relationships (DeJong and Goudriaan, 1989; Pavel and DeJong, 1993). In our case, RGR was not linearly correlated with respiration and the plotting of log RGR against time revealed 3 stages of








76
growth (data not shown) instead of two as reported in peach (DeJong and Goudriaan, 1989).

Respiration was relatively constant during the second period of fruit growth (24 to 45 DAFB and 24 to 72 DAFB for pollinated and GA3-treated fruits, respectively) when RGR was declining to its lowest point. By this stage, most of the cell structures have been laid down, and presumably, mostly maintenance respiration is occurring. This prolonged growth period in GA3treated fruits resulted in a greater C loss than in pollinated fruits (56% and 30%, respectively) (Table 4-1). Although statistical analysis indicated these percentages were not significantly different, biologically the greater respiratory C loss in GA3-treated fruits may have contributed to the smaller fruit size compared to pollinated fruits. The last increase in respiration at later stages of development may be associated with ripening and represented about the same percentage of C loss for both treatments.

As in other rabbiteye blueberry cultivars (Birkhold et al., 1992), net photosynthesis was observed in 'Beckyblue' fruit throughout most of the developmental period and contributed 12% and 7% of the total carbon supply in GA3-treated and pollinated fruits, respectively. These values are in the range observed by Birkhold et al. (1992) in rabbiteye blueberry as well as by Pavel and DeJong (1993) in peach fruits. The higher photosynthetic contribution by GA3-treated fruits may be due to the longer period that these fruits remained







77
green compared to pollinated fruits, as well as the higher surface area/volume ratio due to its smaller size.

Fruits from pollinated treatments began importing carbon at higher rates than GA3-treated fruits very early in development (2-4 DAB). Pollinated fruit sustained this higher rate of C import throughout development, importing ca. 3-fold more assimilates than GA3-treated fruits during the final stages of growth (45 to 72 DAB). These differences cannot be attributed to differences in source supply, since foliation and leaf canopy development occurred at the same time and to the same extent in both treatments. Additionally, fruit loads were similar for both treatments. This suggests that C import into GA3-treated fruits is sink limited.

Carbohydrate Accumulation and Sucrose Enzyme Activities

The ability of the sink to receive or attract assimilates is genetically determined, but can be regulated by the metabolic activity of the sink organ during development. The activities of sucrose-metabolizing enzymes in the unloading area are one component of metabolic activity that can influence the extent of sucrose translocation to fruits. In the present study, however, the increased rate and amount of imported C into pollinated vs GA3-treated fruits did not appear to be related to differences in activities of sucrose-metabolizing enzymes and/or subsequent carbohydrate accumulation.

In both pollinated and GA3-treated fruits, SPS and SS activities were similar. Activities of both enzymes were low throughout development and







78

increased slightly during the final stage of growth. Similar results were reported previously by Darnell et al. (1994) for blueberry fruit. The increase in SPS activity was coincident with the increase in sucrose concentration in maturing fruits. SPS activity is associated with tissues that accumulate sucrose and has been reported in sugar beet roots (Fiew and Willenbrike, 1987), sweet melon (McCollum et al., 1988; Hubbard et al., 1989) and some tomato species (Miron and Schaffer, 1991). Hubbard et al. (1989) reported that SPS was the dominant enzyme associated with sucrose accumulation in muskmelon.

Although the increase in SS was also coincident with the increase in fruit sucrose concentration, most research suggests that SS acts primarily in the degradative direction (cleaving sucrose to fructose and UDP-glucose) in vivo (Xu et al., 1989; Sun et al., 1992; Wang et al., 1993). SS activity has been strongly correlated with sink strength in potatoes (Sung et al., 1989; Xu et al., 1989), wild tomato (Sun et al., 1992), and cultivated tomato (Stommel, 1992; Wang et al., 1993). In these fruits, sucrose is broken down for starch synthesis during the early stages of development. SS has an advantage over invertase activity for this process, since SS results in UDP-glucose production, a precursor for starch biosynthesis. In other fruits, however, particularly hexose accumulators, SS activity does not appear to be related to growth (Nielsen et al., 1991). The increased SS activity observed in maturing blueberry fruits might be involved in the hydrolysis of sucrose since at this stage extensive hexose accumulation is taking place. However, this would result in futile







79

cycling of sucrose between SS and SPS. Nevertheless, the roles of SS and SPS in the regulation of carbohydrate metabolism in blueberry fruits appear to be minor.

The high activities of both soluble and insoluble acid invertases during blueberry fruit development and the strong correlation between invertase activity, hexose accumulation, and fruit growth, indicate that these are the major enzymes involved in sucrose metabolism in GA3-treated and pollinated fruits. This agrees with previous work in which invertase has been reported as the dominant enzyme throughout development of hexose accumulating fruits (Manning and Maw, 1975; Leigh et al., 1979; Nielsen et al., 1991).

Both insoluble (presumably cell wall associated) and soluble acid invertase may be important in decreasing the sucrose concentration at the unloading site. This, in turn, would enhance sucrose translocation to that site by maintaining a sucrose concentration gradient from source to sink. The presence of insoluble acid invertase, which hydrolyzes sucrose before it is taken up by the cells, ensures continuous sucrose unloading by maintaining low turgor in sink phloem and increased osmotic potential in the appoplast during the early stages of blueberry fruit development when readily available energy is needed to sustain rapid growth and respiration. Soluble acid invertase, which hydrolyzes sucrose once it is stored in vacuoles, is the enzyme most closely associated with hexose accumulation in blueberry fruits.

Insoluble invertase activity and fruit hexose concentrations were







80

relatively high at anthesis. Thus, it seems unlikely that differences in fruit set/early fruit development among treatments were limited by carbohydrate availability. Similar results were reported in strawberry by Darnell and Martin (1988) and in orange by Rufz and Guardiola (1994). This high initial carbohydrate content may explain the high initial fruit set in blueberry, as observed by the "swelling" of the receptacle that occurred in all treatments. It seems that this event is "programmed" at least under the conditions of this experiment, since nonpollinated fruits did not receive any stimulus compared to pollinated and GA3-treated fruits. Similarly, Martin et al. (1982) suggested that all the chemical factors required for fruit set of pear are present at anthesis, but that subsequent development of the fruit requires an extra stimulus.

Subsequent fruit development in the present study does not appear to be limited by invertase activity and/or carbohydrate availability in either GA3treated or pollinated fruits. GA3-induced fruit development was accompanied by similar increases in soluble acid invertase activity and fruit hexose concentration as was pollinated fruit development. Thus, the smaller size of GA3-treated fruit does not appear to be the result of a reduced capacity for sucrose metabolism. On the other hand, calculation of invertase activity and hexose accumulation on a per cell basis (refer to Chapter 5) indicates that cells in pollinated fruits exhibited a 1.8-fold increase in soluble invertase activity and a 2.2-fold increase in hexose accumulation compared to cells in GA3-treated fruits. Since cell number was similar between pollinated and GA3-fruits, this







81

increase reflects the increase in cell size of pollinated fruits. Thus, the larger cells in pollinated fruits apparently had increased protein (i.e., invertase) capacity compared to the smaller cells in GA3-treated fruits.

The role of invertases in nonpollinated fruits is less clear. The decrease in soluble invertase at 24 DAFB, as well as the decrease in insoluble invertase at 2 and 24 DAFB, may indicate that those fruits will be senescent, as has been reported by Estruch and Beltr~n (1991) in nonpollinated pea ovaries.

In summary, fruit size differences between GA3-treated and pollinated fruits were due to limited dry matter accumulation during the mid to late stages of development in GA3-treated fruits compared to pollinated fruits. This was reflected in a different carbon requirement for large (pollinated) and small (GAtreated) fruits. GA3-treated fruits appeared to be sink-limited and imported less carbon that pollinated fruits. However, the apparent sink-limitation and resultant decrease in fruit weight of GA3-treated compared to pollinated fruits were not associated with any differences in the capacity of sucrose-metabolizing enzyme activities in fruit extracts and/or subsequent soluble sugar accumulation as determined on a fruit weight basis.









0.5

GA 3 POLL NP
0.4 A

0)
0.3
._6
0.2 .



0.1

........... . .................. ....
0

3.0

2.5 B

. 2.0

FD 1.5 C,)
U

0.5
~~...... ............... ...............
0 20 40 60 80 Days after bloom Figure 4-1. Changes in dry weight (A) and fresh weight (B) in GA3-treated
(GA3), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means � SE, n = 3; SE bars present only when larger than symbol. Arrow indicates when abscission of nonpollinated
fruits occurred.







83



GA 3 POLL NP
0.5 .





0.4





~0.3



0)
CC 0.2




0.1
02





0.1 .


�"




0 10 20 30 40 50 60 70 80 90 Days after bloom Figure 4-2. Relative growth rate (RGR) of GA3-treated (GA3), pollinated (POLL),
and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits.
Means �SE, n =3; SE bars present only when larger than symbol.







84


15 Light Dark
10- GA 3
5
0
-5 -.... .... ............................. .....
-10 . .. ...................

.c: -15 4-20


E POLL


-U � . . . . . . . . . . . . . . . . . . ..... . . . . . . .

"-0 10

.-15
c: -20
,.
C



15
(D 10- NP
0





0
-5"
10Z




-i ...4 ........... ............ ...
-20

0 10 20 30 40 50 60 70 80 90 Days after bloom Figure 4-3. Net carbon dioxide exchange in GA3-treated (GA3), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits.
Values are means � SE, n =3; SE bars present only when larger than
symbol.











6RespC DWC
6 F


GA 3




E
0-' Co
0
U

C.

>4
V


2-4


4-10
Days


10-24 24-45 after bloom


H


45-72 72-87


Figure 4-4. Daily estimated C cost in GA3-treated (GA3) and pollinated (POLL)
'Beckyblue' rabbiteye blueberry fruits. Values were calculated from data in Figures. 4-1 and 4-3 and are based on a 12 h photoperiod and 25/21�C
light/dark temperatures.


POLL










___-_


41-


0-2


L L A







86
8

I * Photosynthetic C Imported C

6

GA3

4



2
L)
E

0. 0
o.
8
0


6 POLL

Ca
a)
> 4-.



2



0
0-2 2-4 4-10 10-24 24-45 45-72 72-87 Days after bloom Figure 4-5. Daily estimated C supply in GA3-treated (GA3) and pollinated (POLL) 'Beckyblue' rabbiteye blueberry fruits. Values were calculated from data in Figures 4-1 and 4-3 and are based on a 12 h photoperiod and 25/21�C
light/dark temperatures.











100





80 60





40


0 10 20 30 40 50 Days after bloom


60 70 80 90


Figure 4-6. Hexose accumulation in GA3-treated (GA3), pollinated (POLL), and
nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are
means � SE, n =3; SE bars present only when larger than symbol.










GA3 POLL NP
.. .. �-


SPS


SS


0 10 20 30 40 50 60 70 80 90 Days after bloom
Figure 4-7. Activities of (A) sucrose phosphate synthase (SPS) and (B) sucrose
synthase (SS) in GA3-treated (GA3), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means � SE, n = 3;
SE bars present only when larger than symbol.























LL 0)

E



Ca) a) Co

o,
*0


0 10 20 30 40 50 60 Days after bloom


70 80


Figure 4-8. Activities of (A) soluble acid and (B) insoluble acid invertases in
GA3-treated (GA3), pollinated (POLL), and nonpollinated (NP) 'Beckyblue'rabbiteye blueberry fruits. Values are means �SE, n = 3; SE
bars present only when larger than symbol.




Full Text

PAGE 1

POLLINATION AND GA 3 EFFECTS ON FRUIT GROWTH, SUCROSE METABOLISM, CELL NUMBER, AND CELL SIZE OF BLUEBERRY FRUITS By RAQUEL CANO MEDRANO 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 1994

PAGE 2

ACKNOWLEDGEMENTS Special thanks go to Dr. Rebecca Darnell, my advisor, for her guidance and support. I was lucky to work under her supervision. I also thank the rest of the members of my committee, Drs. W. B. Sherman, K. E. Koch, D. J. Huber, and K. J. Boote. I extend my appreciation to Steve Hiss for his friendship and technical support, specially with the photographs. I enjoyed sharing this time with Alejandra Gutierrez, Don Merhaut, and Kurt Nolte. I also thank Ma. Elena Suarez, Carlos Garcia, and Carmen Gispert for their friendship and moral support. Finally, sincere thanks go to my parents, Elisa and Bony; my husband, Jorge; my brother and sisters, J. Domingo, Olivia, Nelly, Irma, and Norma for for their love and unconditional support. n

PAGE 3

TABLE OF CONTENTS ACKNOWLEDGEMENTS " LIST OF TABLES v LIST OF FIGURES vi ABSTRACT * CHAPTERS 1 INTRODUCTION 1 2 REVIEW OF LITERATURE 3 Flower Development 3 Pollination and Fertilization 4 Fruit Growth and Development 8 Endogenous Growth Regulators 12 Exogenous Growth Regulators 14 Effect of Exogenous Growth Regulators on Sink Tissues 19 Sink Size 20 Sink Activity 22 3 EFFECT OF GA 3 AND POLLINATION ON FRUIT SET 31 Introduction 31 Materials and Methods 33 Experiment I 33 Experiment II 34 Experiment III 36 Results and Discussion 37 Experiment I 37 Experiment II 41 Experiment III 45 Conclusions 48 iii

PAGE 4

4 CARBOHYDRATE METABOLISM AND FRUIT GROWTH IN GIBBERELLtC ACID-TREATED VS POLLINATED FRUITS OF BLUEBERRY 58 Introduction 58 Materials and Methods 60 Plant material 60 Treatments 61 Fruit carbon dioxide exchange 62 Calculations of carbon budget 63 Sugar determination 65 Enzyme extraction 65 Enzyme assays 66 Results 67 Fruit growth 67 Fruit carbon dioxide exchange 68 Fruit carbon budget 69 Carbohydrate accumulation 70 Enzyme activity 71 Discussion 73 Fruit growth and carbon budget 73 Carbohydrate accumulation and sucrose enzyme activities 77 5 CELL NUMBER AND CELL SIZE IN GIBBERELLIC ACIDTREATED VS POLLINATED FRUITS OF BLUEBERRY 91 Introduction 91 Materials and Methods 93 Results 96 Fruit growth 96 Cell number and cell size 97 Discussion 101 6 SUMMARY 127 LITERATURE CITED 131 BIOGRAPHICAL SKETCH 149 IV

PAGE 5

LIST OF TABLES Table Page 3-1 Effect of GA 3 on fruit set and development in 'Beckyblue' rabbitye blueberry in 1992 55 3-2 Effect of GA 3 on fruit set and development in 'Beckyblue' rabbiteye blueberry in 1993 56 33 Effect of GA 3 , pollen source, and pollen amount on fruit fresh weight (FW) and seed number/seed weight in 'Beckyblue' rabbiteye blueberry 57 41 Estimated C Budget for developing GA 3 -treated, pollinated, and nonpollinated 'Beckyblue' rabbiteye blueberry fruits ... 90 51 Pericarp cell size increases in whole cross-sectional area of 'Beckyblue' rabbiteye blueberry fruits 126 v

PAGE 6

LIST OF FIGURES Figure Page 3-1 Time-course and extent of fruit abscission in 'Beckyblue' rabbiteye blueberry as affected by gibberellic acid (GA 3 ), pollination (POLL), or nonpollination (NP). 3x = 3 applications of GA 3 ; application times indicated by arrows. Values are means ±SE, n = 3; SE bars present only when larger than symbol 50 3-2 Time-course and extent of fruit abscission in 'Beckyblue' rabbiteye blueberry as affected by gibberellic acid (GA 3 ), pollination (POLL), or nonpollination (NP). 2x and 4x = 2 and 4 applications of GA 3 ; application times indicated by arrows. Values are means ± SE, n = 3; SE bars present only when larger than symbol 51 3-3 Time-course and extent of fruit abscission in 'Beckyblue' rabbiteye blueberry as affected by gibberellic acid (GA 3 ), pollination (POLL), or nonpollination (NP). 2x and 4x = 2 and 4 applications of GA 3 ; application times indicated by arrows. Values are means ± SE, n = 9; SE bars present only when larger than symbol 52 3-4 Time-course and extent of fruit abscission in 'Beckyblue' rabbiteye blueberry as affected by gibberellic acid (GA 3 ), pollen source, and pollen amount. X = cross-pollination (200 pollen tetrads); 1 /4X = 50 pollen tetrads; S = self-pollination (200 pollen tetrads). GA 3 = 0.7 mM GA 3 ; GA 3 application times indicated by arrows. Values are means ± SE, n = 9; SE bars present only when larger than symbol 5v 3-5 Effect of GA 3 , pollen source, and pollen amount on final fruit set of 'Beckyblue' rabbiteye blueberry fruits. X = cross-pollination (200 pollen tetrads); %X= cross pollination (50 pollen tetrads); S = self-pollination (200 pollen tetrads); GA 3 = 0.7 mM GA 3 applied at FB and FB + 7 VI

PAGE 7

DAFB. Values are means ±SE, n = 9; SE bars present only when larger than symbol 54 4-1 Changes in dry weight (A) and fresh weight (B) in GA 3 treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means ±SE, n = 3; SE bars present only when larger than symbol. Arrow indicates when abscission of nonpollinated fruits occurred 4-2 Relative growth rate (RGR) of GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Mean ± SE, n = 3; SE bars present only when larger than symbol 83 4-3 Net carbon dioxide exchange in GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means ± SE, n = 3; SE bars present only when larger than symbol 84 4-4 Daily estimated C cost in GA 3 -treated (GA 3 ) and pollinated (POLL) 'Beckyblue' rabbiteye blueberry fruits. Values were calculated from data in Figures 4-1 and 4-3 and are based on a 12 h photoperiod and 25/21°C light/dark temperatures 85 4-5 Daily estimated C supply in GA 3 -treated (GA 3 ) and pollinated (POLL) 'Beckyblue' rabbiteye blueberry fruits. Values were calculated from data in Figures 4-1 and 4-3 and are based on a 12 h photoperiod and 25/21°C light/dark temperatures 86 4-6 Hexose accumulation in GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means ±SE, n = 3; SE bars present only when larger than symbol 87 4-7 Activities of (A) sucrose phosphate synthase (SPS) and (B) sucrose synthase (SS) in GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means ±SE, n = 3; SE bars present only when larger than symbol 88 vii

PAGE 8

48 Activities of (A) soluble acid and (B) insoluble acid invertases in GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) 'Beckyblue'rabbiteye blueberry fruits. Values are means ±SE, n = 3; SE bars present only when larger than symbol 51 Developmental changes in fresh weight of GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means ± SE, n = 3; SE bars present only when larger than symbol. Arrows indicate treatment application times 5-2 Median cross section of a 'Beckyblue' rabbiteye blueberry fruit at 0 DAB. Bar =150 /v m. ep = epidermis; hp = hypodermis; om = outer mesocarp; mm = middle mesocarp; im = inner mesocarp; en = endocarp; vb=vascular bundles; lc = locule; pi = placental tissue; ov = ovule . . . . 5-3 Cell number changes per median cross sectional area (x.s.) in the pericarp of GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) blueberry fruits. Mean separation across treatment and time by LSMean's, p = 0.05 5-4 Increases in cell size in developing (A, B) epicarp and (C) endocarp tissues of GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) blueberry fruits Significance level at 0.05, n = 9 5-5 Median cross section of (A) GA 3 -treated, (B) nonpollinated, and (C) pollinated 'Beckyblue' rabbiteye blueberry fruit at 3 DAB. Bar = 150 //m. ep = epidermis; hp = hypodermis; om = outer mesocarp; mm = middle mesocarp; im = inner mesocarp; en=endocarp; vb=vascular bundles; Ic = locule 5-6 Median cross section of (A) GA 3 -treated, (B) nonpollinated, and (C) pollinated 'Beckyblue' rabbiteye blueberry fruit at 1 DAB. Bar=150/ym. ep = epidermis; hp = hypodermis; om = outer mesocarp; mm = middle mesocarp; im = inner mesocarp; en = endocarp; vb=vascular bundles viii 89 107 109 110 1 1 1 113 0 115

PAGE 9

5-7 Median cross section of (A) GA 3 -treated, (B) nonpollinated, and (C) pollinated 'Beckyblue' rabbiteye blueberry fruit at 24 DAB. Bar =150 /vm. ep = epidermis; hp = hypodermis; om = outer mesocarp; mm = middle mesocarp; im = inner mesocarp; en = endocarp; vb=vascular bundles 5-8 Median cross section of (A) GA 3 -treated and (B) pollinated 'Beckyblue' rabbiteye blueberry fruit at 45 DAB. Bar=150/vm. ep = epidermis; hp = hypodermis; om = outer mesocarp; mm = middle mesocarp; im = inner mesocarp; en = endocarp; vb = vascular bundles; sc = sclereids . . . 5-9 Median cross section of (A) GA 3 -treated and (B) pollinated 'Beckyblue' rabbiteye blueberry fruit at ripening (87 and 72 DAB, respectively). Bar = 150 //m. ep = epidermis; hp = hypodermis; om = outer mesocarp; mm = middle mesocarp; im = inner mesocarp; en = endocarp; vb = vascular bundles; sc = sclereids 5-10 Increases in cell size in developing mesocarp tissues of GA 3 treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) blueberry fruits. Significance level at 0.05, n = 9 5-1 1 Average cell enlargement rate in developing (A, B) epidermal, hypodermal, and (C) endocarp tissues of GA 3 treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) blueberry fruits. Calculated from data in Figure 5-4 . . . . 5-12 Average cell enlargement rate in developing mesocarp tissues of GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) blueberry fruits. Calculated from data in Figure 5-5 5-13 Developmental changes in mesocarp cell size, fruit fresh weight, and pericarp cell number in (A) pollinated, (B) GA 3 -treated, and (C) nonpollinated rabbiteye blueberry fruits 117 119 121 122 123 124 . 125 IX

PAGE 10

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 POLLINATION AND GA 3 EFFECTS ON FRUIT GROWTH, SUCROSE METABOLISM, CELL NUMBER, AND CELL SIZE OF BLUEBERRY FRUITS By Raquel Cano Medrano December, 1994 Chairperson: Dr. Rebecca L. Darnell Major Department: Horticultural Science Research reported here examines the relationship between growth of pollinated and GA 3 -induced parthenocarpic blueberry fruits and differences in sucrose metabolizing ability and cell number and cell size of treated fruits. Fruit set was induced and development initiated with all combinations of GA 3 concentrations and application times tested, as well as in conjunction with crossand selfpollination. However, compared to pollinated fruits, final fruit weight was decreased and fruit development was prolonged when GA 3 was applied alone or in combination with self-pollination. Pollinated fruits began importing and accumulating carbon at higher rates than GA 3 -treated fruits as early as 2 to 4 days after bloom (DAB). This higher rate of carbon import was sustained throughout development of pollinated fruits. The prolonged period II of growth in GA 3 -treated fruits resulted in x

PAGE 11

relatively higher respiratory carbon losses compared to pollinated fruits, and this may have negatively affected final fruit size. Carbon accumulation occurred mostly during the final stage of fruit growth in both treatments and was 3-fold greater in pollinated fruits compared to GA 3 -treated fruits. Sucrose phosphate synthase, sucrose synthase, and insoluble and soluble acid invertase activities were similar throughout development for fruits of both treatments. SPS and SS activities were low throughout most of the fruit development period. Insoluble acid invertase activity was relatively high (ca. 1 5 /vmol gFW' 1 h' 1 ) at anthesis, declining gradually throughout development. From 45 DAB to ripening, soluble acid invertase activity increased sharply, from 0 to 60 /vmol gFW Â’-h' 1 . Glucose and fructose accumulated in similar amounts and were strongly correlated with invertase activity. Thus, invertase was the main enzyme involved in soluble sugar accumulation during the final stage of development in blueberry fruits. Pericarp cell number increased throughout fruit development, averaging 1 0300 cells/median cross sectional area (x.s) at ripening. Mesocarp composed about 70% of the total pericarp area and contributed an average of 8900 cells/x.s. Differences in size between GA 3 -treated and pollinated fruits were due to differences in cell size in middle and inner mesocarp cells. Cells from pollinated fruits increased an average of 33-fold in size compared to an average of 25-fold in GA 3 -treated fruits throughout fruit development. XI

PAGE 12

CHAPTER 1 INTRODUCTION Rabbiteye blueberries are native to north Florida and well adapted to the climate and acid mineral soils, as well as highly tolerant to Florida diseases. The main disadvantages of rabbiteye blueberries are low fruit set in some years and late fruit ripening relative to southern highbush cultivars. Poor set in rabbiteye blueberry has been attributed to several factors, including floral morphology, pollen viability, pollen incompatibility, and presence of pollinating insects. In general, the mechanism of fruit set is unknown. In some cases, an increase in hormone levels following pollination and fertilization has been detected, and this stimulus can be replaced by exogenous applications of growth regulators. However, a direct relationship between hormone levels and fruit set has not yet been found. Increased fruit set and parthenocarpic fruit development in lowbush and highbush blueberries have been obtained under experimental conditions using gibberellic acid (GA 3 ). In rabbiteye blueberries, responses to GA 3 applications have been variable depending on concentration, timing, and cultivar. Studies have shown that one of the main effects induced by GA 3 applications to flowers 1

PAGE 13

2 is stimulation of assimilate transport and accumulation of carbohydrates into the tissue at the application site. GA 3 effect on assimilate import into sinks may be caused by effects on sink strength, i.e., sink activity and/or sink size. Although the effect of GA 3 on sink strength has been studied to some extent in other organs, it has not been investigated extensively with respect to fruit set and development. Thus, the objectives of this study are a) to determine the effectiveness of GA 3 in increasing fruit set and yield in rabbiteye blueberry, b) to determine the effect of GA 3 on fruit carbon requirements and sucrose metabolizing enzymes in relation to fruit set and development, and c) to determine the effect of GA 3 on cell number and cell size in blueberry fruit and the relationship to fruit set and development.

PAGE 14

CHAPTER 2 REVIEW OF LITERATURE Flower Development The upper limit of fruit that can be produced by an individual plant during a reproductive season is determined by the number of female flowers, where the number of seeds is fixed by the number of ovules within those flowers (Stephenson, 1981). Flower bud initiation, a majorfactor affecting reproductive development, occurs the summer before flowering for most fruit crops, including blueberry. Floral differentiation normally takes place either in late winter or early spring. Bell and Burchill (1955) found that floral initiation in lowbush blueberry started in June, under Nova Scotia conditions. Early in August, sepals, petals, stamens, and carpel primordia were fully formed and elevated to their epigynous position. Differentiation started in March. In the anthers, meiosis was completed during the first week of May and the vegetative and generative cells were formed 1 5 days later. In ovules, meiosis was completed during the third week of May. In high bush blueberry, under Rhode Island conditions, floral differentiation began in August and by October all flower parts were visible (Gough et al.. 3

PAGE 15

4 1978). Microspore and megaspore mother cell activity was apparent early in November and cell division continued through the autumn. By mid-March, the ovules began their final phase of development. Pollen grains appeared to be fully formed by mid-April and embryo sacs were formed just prior to bloom. There is no information available about floral initiation and floral development in rabbiteye blueberry, but events are probably similar to those described above. Pollination and Fertilization Except in apomictic and parthenocarpic fruits, pollination and fertilization must occur in order for fruit set to occur. Factors such as floral morphology, insect visitation, pollen viability, pollen tube growth, embryo sac development, and pollen source affect pollination and fertilization, and therefore, fruit set. In blueberry plants, each flower bud consists of about five individual flowers with each flower at a slightly different stage of development (Goldy, 1985). However, anthers within each individual flower are generally at the same meiotic stage. Flower structure is such that insect pollination must occur in order to achieve fruit set (Eck and Mainland, 1971). Individual flowers hang in a pendant position and the pistil protrudes past stamens and corolla; thus, the receptive surface of the stigma is not exposed to the pollen from the anther. Lee (1958) found that the honeybee is an effective pollinator of lowbush blueberry. However, Wood (1966) found no significant increase in fruit set of lowbush blueberry by increasing the number of honeybee colonies from 2.5 to

PAGE 16

5 7.4 colonies per hectare. In highbush blueberry, Howell et al. (1 972) found an increase in yield with the use of 10 hives per hectare and suggested that hives be introduced into plantations no later than 25% full bloom. Although honeybees can pollinate rabbiteye blueberry flowers, Payne et al. (1989) and Cane and Payne (1990) found that pollination was more effective by bees that vibrate the flowers, such as bumblebees (Bombus spp.) and the southeastern blueberry bee ( Habropoda laboriosa). The total potential for production of viable pollen may affect fruit set in some blueberry cultivars. Stushnoff and Hough (1968) found differences in microspore development in two highbush blueberry cultivars that differed in fruit production. Under field conditions, 'Coville ' showed inconsistent fruit production while 'Bluecrop' did not. 'Coville'pollen did not dehisce readily from the anther sacs and did not germinate in vitro. Embryo sac development was normal in the two cultivars tested, and when fertilization occurred, the development of the endosperm and embryo was also normal. Eaton (1 966) and Brewer and Dobson ( 1 969) found cultivar differences in both number of pollen grains released and pollen germination. Brewer and Dobson (1969) suggested that low pollen production coupled with poor pollen germination may reduce fruit set under field conditions. Later, Cockerham and Galletta (1976) found that pollen in tetraploid species, including V. corymbosum, was potentially more fertile (as judged by stainability) than in diploid species, but pollen stainability in hexaploid species, including V. ashei, was not different from the

PAGE 17

6 diploid or tetraploids. There were no differences in pollen analysis when they compared pollen from plants located at different sites. They concluded that genetic, rather than seasonal or environmental, differences appeared to account for the major portion of the interclonal and interspecific variation observed in pollen stainability. Young and Sherman (1978) found that the rate of pollen tube growth in the styles of both rabbiteye and southern highbush cultivars was adequate with pollination up to 6 days after emasculation, under greenhouse conditions (33/18° C). Varying intervals between emasculation and pollination up to 6 days did not consistently influence percentage fruit set or number of seeds produced per berry. They concluded that flowers in the field should be able to set adequately even with 2 to 4 days of unfavorable pollination weather during bloom. Embryo sac development and stigmatic receptivity also affect fruit set. Eaton and Jamont (1966), studying ovule and embryo sac development between anthesis and petal fall in highbush blueberry, found that at anthesis and 1 day afterward, 81% of the normal embryo sacs had differentiated egg cells, and from 2 to 9 days after anthesis, 95% of these sacs had differentiated egg cells. Between anthesis and two days later, 15% of the ovules had degenerated embryo sacs. This percentage increased to 33% three days after anthesis. In lowbush blueberry, blossoms remained receptive until 7 days after anthesis (Wood, 1 962), and pollen tube growth required about 4 days to grow from stigma to ovule (Bell, 1957). In highbush, flowers remained receptive to

PAGE 18

7 pollination up to 8 days after anthesis (Moore, 1 964). In rabbiteye blueberries, the receptive period was 4 days (Young and Sherman, 1978). Pollen source and the extent of self-compatibility also affect fruit set in blueberry. Highbush blueberry is generally believed to be highly self-fruitful, while rabbiteye blueberry is self-unfruitful. However, both species have shown different levels of self-compatibility, and fruit set has varied depending upon this factor. In highbush, the effect of pollen source on fruit set and development is contradictory. While Coville (1921), Bailey (1938), and Morrow (1943) reported than self-pollinated northern highbush flowers set fewer berries than cross-pollinated flowers, Merrill (1936) found that self-pollination gave a satisfactory commercial fruit set and White and Clark (1 939) indicated that set from self-pollination was not significantly different from the set in openpollinated flowers. More recently, Gupton (1 984) found that self-pollination of southern highbush cultivars generally resulted in fruit set equal to or better than cross-pollination. However, Lyrene (1 989) found that fruit set was significantly lower for self-pollinated than for mixed and cross-pollinated flowers of the southern highbush cultivar, 'Sharpblue.' All genotypes of the recently introduced half-high blueberry (V. corymbosum x \/. angustifolium ) set fruit on at least 40% of the flowers following cross-pollination. However, fruit set percentage varied from 0 to 84% following self-pollination (Rabaey and Luby, 1988).

PAGE 19

8 In rabbiteye blueberry, Meader and Darrow (1944) found that under greenhouse conditions, most varieties tested were either partially or completely self-fruitful. Only one variety was completely self-unfruitful. In the three varieties of rabbiteye blueberry tested by Tamada et al. (1 977), cross-pollination usually increased fruit set more than self-pollination. El-Agamy et al. (1 982) in their study on intraand inter-ploidy pollen incompatibility, reported that selfing reduced fruit set in highbush and rabbiteye blueberries by 28.5% and 36.3%, respectively, compared to crossing. Although self-incompatibility was apparent in both types, highbush clones were more self-fruitful than rabbiteye clones. Lyrene and Goldy (1983) found that fruit set of several rabbiteye cultivars ranged from 36% to 75% on open-pollinated branches and from 3% to 21% on self-pollinated branches. In a study of native rabbiteye blueberry clones, Garvey and Lyrene (1987) found that fruit set averaged 15% after selfpollination and 58% after cross-pollination. There is also cross-incompatibility between some rabbiteye blueberry cultivars. Darnell and Lyrene (1989) found that when two related rabbiteye cultivars, 'Beckyblue' and 'Aliceblue,' were reciprocally crossed, fruit set was 25%, while cross-pollination with 'Climax' pollen resulted in 59% and 72% fruit set, respectively. Fruit Growth and Development Pericarp development of blueberry is similar to that of peach, cherry, grape, and fig. The double sigmoid type of growth curve is divided into three

PAGE 20

9 stages: Stage I, consisting primarily of cell division following fertilization; Stage II, a period of slow pericarp development and development of embryo and endosperm tissues; and Stage III, a second period of rapid pericarp growth (mainly cell enlargement) that continues to fruit maturity (Young, 1 952; Hindle et al., 1 957; Edwards et al., 1 970; Spiers, 1981). Duration of individual stages vary with the cultivar, clone, and temperature. The length of each growth stage is important in determining the time required from anthesis to fruit maturity. The duration of both cell division and cell enlargement affects final fruit size. Cell division might occur well before anthesis, as has been shown in cherry (Tukey and Young, 1939) and peach (Scorza et al., 1991). The difference in cell number between smalland large-fruited peach cultivars was established 175 days before bloom. In addition, the small-fruited types had a shorter Stage I and longer Stage II. Cells from different tissues in the pericarp may divide and enlarge at different rates and planes with a consequent effect on fruit shape. In grape, pericarp cells around the vascular bundles elongate tangentially, while those toward the placenta elongate radially (Harris et al. 1968). Similarly, in cherry, hypodermal cells enlarged in a tangential direction, while middle and inner fleshy pericarp cells enlarged radially (Tukey and Young, 1939). In cucumber, although cell elongation occurred throughout the fruit development period, cells at the peduncle end were larger than those at the blossom end of the fruit (Marcelis and Hofman-Eijer, 1993). The growth rate at both ends slowed

PAGE 21

10 sooner than the growth rate of the rest of the fruit. The contribution of cell division period and cell enlargement to fruit development in blueberry has not yet been documented. The potential success of exogenous applications of GA 3 may depend on the stage of development and, since both cell division and cell enlargement can be altered by growth regulators, knowing when they are occurring becomes important. The formation of seeds within the pericarp has a marked effect on fruit size and development (Ryugo, 1 988). In apple, peach, kiwifruit, and grape, the size and shape of the fruit are a function of the number of seeds formed. In general, fruit with a higher number of seeds are longer and more symmetrical than those with fewer seeds. Seed formation also affects blueberry fruit development. Aalders and Hall (1961) found that seed number in lowbush blueberry was significantly correlated with berry weight and fruit development period. Heavier berries and earlier ripening were associated with more seeds. In highbush blueberry, Coville (1 921 ) stated that self-pollination produced fewer berries, which were smaller and later maturing than berries that were cross-pollinated, suggesting a direct effect of seeds on these characteristics. White and Clark (1939) found that seed number in large berries ranged from 55-68, while smaller berries had 21-41 seeds/berry. Morrow (1943) reported that cross-pollinated highbush blueberry fruits averaged 61-62 seeds/berry, while self-pollinated fruits averaged 45-58 seeds/berry. In rabbiteye blueberry, the average total weight of the seeds per berry was greater for cross-pollinated

PAGE 22

11 (20-34 mg/berry) than for self-pollinated berries (4-18 mg/berry) (Meader and Darrow, 1944). Darrow (1958) reported that seed number in highbush blueberry ranged from 1 5.7 to 73.8 per berry, and although the varieties with the largest berries generally had the most seeds, this relationship was not always true. Eaton (1967) reported that variation in seed number accounts for less than 40% of the total variability in berry weight in cross-pollinated highbush blueberry. Fruit of the southern highbush 'Sharpblue' also had an increase in seed number per berry, increased berry size, and a decrease in the fruit development period when cross-pollinated (Lyrene, 1 989). However, the effect of the increased seed number was greater at low seed level (5 seeds per berry) than at high seed level (10 seeds per berry). The effect of genetic relationships on berry weight and seed number among cross-pollinated blueberry cultivars was investigated by Heilman and Moore (1983). They found a significant reduction in mean berry weight associated with an increase in the degree of relationship between parents for only one highbush ('Coville') and one rabbiteye ('Climax') cultivar. They also found a variable response to cross-pollination among related individuals with respect to number of seeds per berry, although in general, self-pollination reduced the number of seeds per berry. Gupton (1984) reported that in highbush blueberry, berry weight and number of viable seeds per berry were not significantly higher for fruit produced from half-sibling crosses than setting and that outcrosses generally produced the greater berry weight and seed number.

PAGE 23

12 In rabbiteye blueberry, seed weight per berry and number of large seed per berry were reduced by self-pollination and by pollination with related cultivars (Darnell and Lyrene, 1 989). However, mean berry weights of reciprocal crosses between the related cultivars were not significantly different from weights of berries produced by pollination with an unrelated cultivar. Meader and Darrow (1944) found that cross-pollinated rabbiteye blueberries had larger fruit and a shorter fruit development period than berries from self-pollinations. Tamada et al. (1977) found an increase in the number of large seeds per berry as a result of cross-pollination, which was reflected in larger berries and earlier ripening. Endogenous Growth Regulators There are two classical points of view related to the role of growth regulators in fruit set and development (Browning, 1 989). The first states that there is a deficiency of auxin and gibberellin or an excess of inhibitor that inhibits ovary growth at anthesis. The promotory hormonal stimulus needed is supplied by pollination and fertilization or by other means in the set of parthenocarpic fruits. Fruit growth is controlled by hormones produced by the endosperm and embryo in developing seed or by maternal tissue in the ovules of fruits that set parthenocarpically. The second view states that the primary effect of pollination or hormone treatment of the flower is to increase the nutrient import needed for growth to become self-sustaining through the subsequent mobilizing action of hormones produced by the ovules or seeds.

PAGE 24

13 In lowbush blueberry, Collins et al. (1966) reported a period of increased auxin-like activity in developing fruit, beginning about 3 weeks after anthesis. Auxin-like activity reached a maximum 6 to 7 weeks after anthesis, during Stage II of fruit growth, before declining. No conclusions can be drawn from this experiment because auxins are unlikely to be the one promoting substance that affects fruit growth. However, it can be speculated that auxins may either inhibit pericarp growth or promote development of the embryo, since activity peaked in Stage II of slow pericarp growth. In highbush blueberry, Mainland and Eck (1971) found cultivar differences in the endogenous levels of auxins. Auxin activity increased in pollinated 'Coville' berries two weeks after anthesis, while 'Jersey' fruits exhibited no such increase. A second peak in auxin activity occurred 6 weeks after anthesis for pollinated berries of both varieties. GA 3 at 500 ppm, applied at blossom time, increased endogenous auxin in fruits of both cultivars compared to the nontreated berries. Nonpollinated 'Jersey' flowers contained higher concentrations of GA-like substances than did 'Coville' flowers. GA-like activity in fruit decreased in both cultivars two to three weeks after pollination. A second peak in GA-like activity occurred three to four weeks after anthesis, corresponding with Stage II of fruit development in both pollinated and GA 3 treatments. GA 3 application at bloom period increased GA-like activity in both cultivars, but at maturity, GA-like activity was higher in pollinated than in GAtreated berries.

PAGE 25

14 Natural parthenocarpy has been observed in 'Jersey' cultivars and may be related to the higher levels of auxin activity compared to 'Coville' (Mainland and Eck, 1971). Gustafson (1939) found that the auxin content in the ovaries of flower buds from oranges, lemons, and grapes that produced parthenocarpic fruits was higher than in the ovaries that do not produce this type of fruit. Gil et al. (1972) proposed that pear fruit set and development may be associated with a sequential role of different endogenous growth substances. After anthesis, gibberellins were the main growth-promoting substances detectable. When gibberellin levels decreased, auxins appeared and remained conspicuous for most of the fruit development period. The stimulus for fruit set delivered by pollination has not been identified. In some cases, endogenous hormone levels are increased by pollination and fertilization, and these stimuli can be replaced by exogenous hormones in many cases. However, a direct relationship between hormonal content and fruit set has not yet been found. Exogenous Growth Regulators Increased fruit set and parthenocarpic fruits have been obtained in several species by using exogenous applications of gibberellins, auxins, or a combination of both (Leopold, 1962). Parthenocarpic fruit development has been induced in tomato with gibberellins (Wittwer et al. 1956; Gustafson, 1959). Webster and Goldwin (1984) enhanced fruit set in sweet cherry by 125-150% the first year and 75-90% the subsequent years using a mixture

PAGE 26

15 containing 200 ppm GA 3 , 300 ppm NN'-diphenylurea (DPU) and either 10 ppm naphthalene acetic acid (NAA), 10 ppm 2,3,5-trichlorophenoxypropionic acid (2,4,5-TP), or 50 ppm 2-naphthoxyacetic acid (NOXA). In self-sterile plum cultivars, GA 3 induced parthenocarpic fruit and prevented the drop of fruitlet with aborted embryos (Hartman and Anvari, 1 986). In 'Agua de Aranjuez' pear fruits, GA 3 at 10 ppm applied at balloon stage, anthesis, or petal fall induced parthenocarpic fruit set and improved final fruit set (Herrero, 1984). This implies that a fully developed embryo sac is not necessary for the hormone to act. On the other hand, 10 or 20 ppm GA 3 at full bloom increased initial fruit set of 'Doyenne du Comice' pear fruits, but there was no difference in final fruit set compared to the control (Marcelle, 1984). Several exogenous growth regulators have been used experimentally to increase fruit set in blueberry. In lowbush blueberry, Barker and Collins (1 965) induced parthenocarpic fruit set with GA 3 in concentrations as low as 20 ppm applied directly to the flower. They observed that the flower was physiologically receptive to the GA from 2 to 3 days before anthesis to 7 to 8 days after anthesis. The fruit from GA treatments under field conditions appeared normal in all visible aspects, but under growth chamber conditions, the waxy bloom did not appear and the sugar concentration was decreased. They attributed these negative effects to the low light intensity and/or temperature in the environment provided by the growth cabinet; however, it is not possible to determine if the effects on fruit quality were due to treatment

PAGE 27

16 effects or growth environment, since pollinated controls were not included in the growth chamber treatments. In highbush blueberry cultivars, Mainland and Eck, (1 968a) found that 5, 50, and 500 ppm of potassium gibberellate (KGA 3 ) applied to the base of the style of emasculated flowers 2 to 5 days after they opened promoted higher percent set (80%) and larger berry diameter than the hand-pollinated treatment. They attributed this higher set to the high temperature (32. 2C) that occurred on the day of treatment and the two following days. High temperatures and dry weather conditions may limit the maximum period of pistil receptivity and promote premature flower abscission in pollinated treatments. However, the high temperature and low relative humidity might have been expected to decrease uptake of exogenous gibberellins (Greenberg and Goldschmit, 1 989), resulting in lower fruit set in the GA treatments as well. Fruits resulting from KGA 3 treatments were indistinguishable from pollinated berries in terms of shape; however, they were completely seedless and were delayed in maturation. Of the auxin materials evaluated, 2,4, 5-trichlorophenoxyacetic acid (2,4, 5-T) produced the highest percent fruit set (40-50%), but berry diameter was much smaller than for the hand-pollinated treatments. Fruit maturation was also delayed in these treatments. Increases in both fruit set and yield were found in highbush blueberry treated with 500 ppm KGA 3 combined with either 3 ppm 2,4, 5-T or 20 ppm 2,4,5 TP applied at full bloom (Doughty and Scheer, 1 975). Mixtures of 5, 50,

PAGE 28

17 and 500 ppm of KGA 3 and NAA applied at anthesis increased fruit set of highbush blueberry (Mainland and Eck, 1 968b). Application of 6-(benzylamino)9-(2-tetrahydroxyranil)-purine (BTP) at 100, 500, or 1000 ppm did not induce fruit set, nor did it have any synergistic effect when used in combination with KGA 3 or NAA. KGA 3 treatments produced consistent increases in diameter with increasing concentration, and the berries resulting from treatment with 500 ppm of KGA 3 were only slightly smaller than pollinated fruits. NAA at 500 ppm increased fruit set compared to hand-pollination, but it increased diameter only slightly at the highest concentration tested. The number of days from treatments maturity decreased with increasing levels of both NAA and KGA 3 , compared to the pollinated control. NAA and KGA 3 increased fruit soluble solids compared to the pollinated berries. Growth curves of the parthenocarpic fruit were very similar to those of seeded fruits, particularly as gibberellin concentration increased. In a later study, Mainland and Eck (1969b) applied 5, 50, 200, or 500 ppm GA 3 to flowers of uncaged plants and GA 3 at 500 ppm to flowers of caged plants under field conditions. Fruit set was increased by all GA 3 treatments the first year; however, the weight of fruit harvested per bush was not increased. In the second year, there was an increase in yield in all treatments due to an increase in fruit weight, but there was not an increase in fruit set. There was also a trend towards delayed ripening in all GA 3 treatments in both years, but only the caged plants receiving 500 ppm GA 3 experienced a significant delay

PAGE 29

18 in fruit maturation. Additionally, seed number and berry size were significantly reduced in both the first and second pickings with GA 3 treatments. No differences in percentage mold, percentage weight loss, juice pH, or titratable acidity were found between control and GA-treated fruits. All GA-treated berries had lower soluble solids than did the control berries. There were no differences in the number of flower buds formed among treatments in either year. To determine whether or not GA 3 applied to blueberry foliage would induce parthenocarpic fruit development, Mainland and Eck (1969a) applied 5, 50, 200, and 500 ppm to highbush blueberry foliage or foliage plus flowers at full bloom. In the foliage treatments, clusters were wrapped to shield them from any GA 3 spill. Since low fruit set was found in treatments where the flowers were unsprayed, they concluded that GA 3 was not translocated from foliage and stems into the flower. In two consecutive years, all treatments but the foliage one resulted in higher fruit set than in cross-pollinated plants. In rabbiteye blueberry, Davies and Buchanan (1 979) reported an increase of 50% final fruit set in 'Tifblue' with one application of 200 ppm GA 3 at 3040% full bloom and with two applications of 50 and 200 ppm GA 3 at 30-40% full bloom and 15 days later. In 'Delite' blueberries, two applications of 200 ppm GA 3 at 60% full bloom and 1 5 days later were necessary to increase fruit set by 60%. In a later study (Davies, 1986), single applications of a mixture of 100 ppm GA 3 and 10 ppm 2,4-dichlorophenoxyacetic acid (2,4-D) at the

PAGE 30

19 time of leaf emergence, or double applications of the same chemicals at 1 % full bloom and at leaf emergence, had no effect on either fruit set or yield. This response could be related to the application time and/or an increase in ethylene production due to auxin applications. Vanerwegen and Krewer (1990) increased yield in rabbiteye blueberry by applying 243 ppm GA 3 at 90% full bloom or 121 ppm GA 3 twice, at 90% full bloom and 18 days later. Two applications of GA 3 increased yield by 373% in 'Climax' and 562% in 'Tifblue.' Fruit size and soluble solids were reduced, and fruit development period was increased. In summary, increased fruit set has been obtained in lowbush and highbush blueberry by using single applications of GA 3 , while in rabbiteye blueberry, the results have not been consistent. Optimum concentration and timing of application of GA 3 for rabbiteye blueberry have not been established. Effect of Exogenous Growth Regulators on Sink Tissues Carbohydrate allocation patterns depend on the competitive ability of various sink regions within the whole plant (Daie, 1985, 1987). A high sink import rate is a function of its mobilizing activity, which, in turn, is a function of sink size and activity. One of the effects induced by treatment with plant growth regulators is a change of the pattern of assimilate distribution within the plant (Daie, 1 985; Kinet et al., 1985; Aloni et al., 1986). This effect on assimilate import into sinks may be caused by affecting sink size and/or sink activity. Cell number

PAGE 31

20 and cell size are the components of sink size, while phloem unloading, uptake of assimilates by sink tissues, and metabolism and carbohydrate storage are involved in sink activity (Yelle et al. , 1988). Both of these characteristics are subject to hormonal regulation (Bunger-Kibler and Bangerth, 1983; Craighton et al., 1 986; Kinet et al., 1 985; iwasaki et al., 1 986; Lethman 1 963; Vercher et al., 1984,1987). Sink Size Sink size is the physical constraint of sink strength and in general, can be measured by cell number and cell size. Plant growth regulators may directly affect cell division and/or cell elongation. In tomato fruit, exogenous applications of 10% indole acetic acid (IAA) or 0.1% 4-chlorophenoxyacetic acid (4-CPA) to emasculated flowers transiently increased the rate of cell division, while 0.6% GA 3 decreased cell division rate through development (Bunger-Kibler and Bangert, 1 983). However, cell size was increased by using gibberellins compared to using auxins. In pea ovaries, GA 3 applied at anthesis induced mesocarp cell enlargement and division and differentiation of endocarp cells (Vercher et al., 1984). GA 3 application to pea endocarp enhanced synthesis of primary and secondary cell walls (Vercher et al., 1987). Nishitani and Masuda (1982) found that synthesis of polyuronides, xyloglucans, and cellulose was regulated by auxins in Vigna angularis, whereas synthesis of xylans and cellulose during cell maturation was regulated by GA. In isolated Zinnia mesophyll cells, GA 3 s inhibited cell division and delayed the initiation of

PAGE 32

21 DNA synthesis only when cells were in a resting state and not when cells were dividing continuously (Iwasaki et al. , 1986). Cell elongation depends on the extensibility of the cell wall, which in turn is affected by the orientation of cellulose microfibrils (Mita and Katsumi, 1986). GA 3 induced transversely oriented microtubules in mesocotyl epidermal cells of Zea mays L. mutants, stimulating elongation in this region (Mita and Katsumi, 1986). In flank meristems of Hedera, GA 3 stimulated the frequency of tangential divisions as well as the expansion in the radial-longitudinal direction (Marc and Hackett, 1992). Keyes et al., (1990) found that exogenous GA 3 increased cell wall extensibility and elongation in nonmutant (GA 3 -sensitive) control wheat compared to GA-insensitive genotypes. They concluded that, although chemirheological processes may be involved in cell wall loosening, the mechanical properties of the cell wall are strongly associated with leaf cell expansion potential, and GA is involved in the determination of the mechanical wall properties. Similar results were obtained in Phaseo/us vulgaris leaves by Brock and Cleland (1990) and in Pisum sativum L. seedlings by Cosgrove and Sovonick-Dunford (1989). In summary, the effect of growth regulators on cell division and enlargement depends upon species, stage of development, and tissue type within a fruit. Cell number at anthesis, the length of the cell division period after anthesis, and the extent of cell enlargement determine final fruit size in a number of fruits, including blueberry. The contribution of cell number and cell

PAGE 33

22 size to the establishment of sink strength during fruit set and fruit development period has not been investigated in blueberry. Sink Activity Increased cell number and cell size results in more sites for dry matter accumulation (Brenner et al. ( 1 989). This dry matter accumulation will depend on the concentration gradient established between the source and the sink. Specifically, dry matter accumulation in sink tissue depends on phloem loading, phloem transport, phloem unloading, and uptake and compartmentation by the sink tissue. Hormones may affect assimilate (primarily carbon [C]) partitioning at different points along this pathway. Growth stimulation by gibberellic acid has been observed in pea subhook (Pisum sativum L.) within 6 h after application. Growth was positively correlated with soluble sugar accumulation, suggesting that GA 3 stimulated the translocation from cotyledons to the subhook, thereby maintaining a low osmotic potential and enhanced growth (Miyamoto and Kamisaka, 1988). Similar results were observed in dwarf watermelon seedlings, although the GA-mediated increase in dry matter in hypocotyls was not the result of increased translocation out of the cotyledons but rather was the result of decreased accumulation of dry matter in the roots (Zack and Loy, 1984). In pea ovary explants, GA 3 in the presence of sucrose induced fruit set and development over GA 3 alone (Garcia-Martmez and Carbonell, 1 985). The effect of GA 3 on ovary development was a function of the concentration and duration of the GA 3 treatment. The lower the GA 3

PAGE 34

23 concentration, the longer the time of incubation in GA 3 necessary to obtain the maximum effect. GA 3 also stimulated ovary development when sucrose was substituted by glucose or fructose. Achhireddy et al. (1984) investigated the translocation of 14 C-sucrose from the stipule to the 1 0 day-old fruit of excised Pisum sativum fruit/shoot systems. They found that 25 to 100 ppm GA 3 enhanced 14 C translocation into the fruit by about 25% but did not alter 14 C levels in the stipule. However, GA 3 increased the levels of 14 C obtained from untreated leaflets and from water washings. Hayes and Patrick (1985) found that GA 3 promoted 14 Cphotosynthate transport to the site of hormone application in Phaseolus vulgaris decapitated stems. This was associated with significant increases in the pool size of free space sugars at the hormone-treated region, suggesting that GA 3 stimulated unloading of the sieve element-companion cell complexes into the stem free space. The mechanism by which GA 3 enhances transport and accumulation of sucrose to the site of application is unknown. However, anything that steepens the sucrose gradient between the loading and unloading sites (i.e. , decreases the concentration of sucrose in the unloading area) would, in theory, enhance sucrose transport to that site (Aloni et al., 1 986; Daie et al., 1 986; Miyamoto and Kamisaka, 1988, 1990; Morris and Arthur 1985). The level of sucrose in the plant cell depends on four enzymes that are involved in its synthesis and degradation (Avigad, 1982): sucrose phosphate

PAGE 35

24 synthase (UDP-glucose: d-fructose-6-phosphate 2 glucosyltransferase, EC 2,4,1,14), sucrose phosphate phosphatase (Sucrose-6-phosphate phosphohydrolase, EC 3.1.3.00); sucrose synthase (UDP-glucose: D-fructose 2-glucosyltransferase EC 2.4.2.13), and invertase (B-D-fructofuranoside fructohydrolase EC 3.2.1.26). Sucrose synthase is distributed in all plant tissues and is found at high levels particularly in nonphotosynthetic and storage tissues. It catalyzes the reversible reaction UDP-glucose + fructose +* sucrose + UDP. UDP-glucose released by this cleavage and its derivatives are substrates for synthesis of cell walls and storage compounds (Avigad, 1982). In some tissues, sucrose synthase is considered a key enzyme of metabolism in sink organs (Claussen et al., 1985, 1986), but in others, no differences in sucrose synthase activity have been observed between growing and fully expanded leaves (Huber, 1 989; Schmalstig and Hitz, 1987). In acid lime, Echeverria and Burns (1989, 1990) and Echeverria (1990) found that during the early stages of fruit growth, sucrose was catabolized enzymatically by sucrose synthase, acid, and/or alkaline invertase, while in mature fruits, sucrose breakdown occurred by nonenzymatic acid hydrolysis. In 'Valencia' orange juice sacs, sucrose synthase activity was absent in vacuole and cytoplasm (Echeverria and Valich, 1988) and sucrose and hexoses were accumulated mainly in the vacuole. In muskmelon fruits, there is a lack of agreement whether sucrose synthase or sucrose phosphate synthase is involved in sucrose accumulation. McCollum et

PAGE 36

25 al. (1988) reported a decrease in invertase activity and an increase in sucrose synthase activity as muskmelon fruit developed and accumulated sucrose. Hubbard et al. (1 989) found low activity of sucrose synthase during muskmelon fruit development. Invertase activity decreased with time, while sucrose phosphate synthase activity increased throughout fruit development. They concluded that acid invertase and sucrose phosphate synthase determined the sucrose accumulation in these fruits. Similar results were found by Miron and Schaffer (1991) in sucrose accumulating tomato fruits. When comparing sucrose-accumulating vs hexose-accumulating tomato fruits, Dali et al. (1992) found that both fruit types had high sucrose synthase activity when immature, but activity was nondetectable 55 and 70 days after anthesis. Invertase activity increased continuously from 2000 to 15000 nmol. gFW' 1 .min 1 in hexose-accumulators, while in sucrose-accumulators it declined constantly from about 100 to almost nondetectable levels. Sucrose phosphate synthase was low in both types, but sucrose-accumulating types showed an increase in activity at ripening. They suggest that sucrose phosphate synthase is a key enzyme in sucrose accumulation in these fruits and that sucrose is being hydrolyzed in the apoplast and resynthesized in the symplast. On the other hand, Stommel (1992) did not find a direct relationship between sucrose phosphate synthase activity and sugar accumulation during early stages of development in wild and cultivated tomato, which showed marked differences in sugar accumulation. Wang et al. (1993) found that

PAGE 37

26 sucrose synthase and not invertase was the enzyme responsible for the metabolism of imported sucrose in growing tomato. Sucrose synthase activity, relative growth rate, and starch accumulation were positively correlated at Oto 1 3 days after anthesis as well as 20 to 39 days after anthesis. This may be a typical pattern of enzyme activity for storage tissues like tomato, where imported sucrose is stored as starch during the early stages of development. If sucrose is broken down by sucrose synthase, a step that requires half the energy of that catalyzed by invertase, the UDP-giucose released might be used for, among other things, synthesis of starch. This will decrease the sugar content in the fruit and increase the sucrose gradient between leaves and fruit, leading to an increase in sucrose transport. High sucrose synthase activity has also been found in potato tubers, which are starch accumulating sinks throughout fruit development, and hexose reservoirs when excised and stored at low temperatures (Sung et al. , 1989; Ross and Davies, 1992). Moriguchi et al. (1992) studied the pattern of sugar accumulation and sucrose synthase, sucrose phosphate synthase, and invertase activities in highland low-sucrose-accumulating types of Asian pear fruits. There was no correlation between sucrose content and invertase activity in high-sucroseaccumulating types, but they found a positive correlation between sucrose synthase and sucrose (r = .633) and between sucrose and sucrose phosphate synthase (r = .445). Acid invertase accounted for most of the invertase activity

PAGE 38

27 and showed high activity at early stages of development, decreasing in activity at later phases of growth in both fruit types. Invertase competes with sucrose synthase for the same substrate, and in several sink organs, carbohydrate import is correlated with high invertase activity (Morris and Arthur, 1984b). In pea leaves, cell growth rates were positively correlated with specific activity of soluble acid invertase (Morris and Arthur, 1 984a). However, mature, nonimporting leaves of some species, such as soybean and Swiss chard, have shown high activities of acid invertase (Huber, 1989). Several growth regulators have induced invertase activity. Auxins stimulated invertase activity (Morris and Arthur, 1986; Poovaiah and Veluthambi, 1 985; Schaffer et al . , 1 987; Tanaka and Uritani, 1 979), as well as its synthesis de novo (Gordon and Flood, 1980; Pressey and Avants, 1980) in a number of tissues. Increased invertase activity has also been observed in soybean fruits following abscissic acid application (Ackerson, 1985). Broughton and McComb (1971) found that GA 3 stimulated amylase and invertase activities in pea internodes. Their activities were closely correlated with internode growth. Since the injection of glucose mimicked the effect of gibberellic acid on fresh and dry weight accumulation, cell elongation, cell division, and cell wall synthesis, the authors concluded that the overall effect of GA 3 was to provide more substrate for general cell metabolism and wall synthesis. In sweet potato roots, GA 3 stimulated invertase activity (Tanaka and

PAGE 39

28 Uritani, 1979). Kaufman et at. (1973) found that in response to the GA 3 treatment, invertase activity increased, then decreased, in Avena segments. The increase in activity was correlated with the active growth phase, whereas the decrease in activity was initiated when growth of the segments slowed. A continuous supply of GA 3 retarded the decline of enzyme activity, but growth rate remained constant. In pea internode, both the amount of acid invertase per internode and the specific activity increased after GA 3 application (Morris and Arthur, 1985). The maximum specific activity of the enzyme coincided with the time of peak internode elongation rate. Hexose sugars were correlated with total acid invertase per node. Sucrose concentration decreased as hexose and invertase activity increased. Miyamoto et al. (1993) reported that GA 3 and not auxins are responsible for the accumulation of sugar at the subhook region in Pisum sativum. Both substances promoted elongation, but the increases in soluble sugar and soluble invertase activity were observed only upon application of GA 3 . Interestingly, the excision of cotyledons inhibited soluble sugar accumulation and decreased invertase activity. This effect was not reversed by application of GA 3 or IAA. In previous studies Miyamoto et al. (1 988) found that auxins stimulated growth by increasing cell wall loosening, while GA did not have this effect. This suggests that the mechanisms by which these two enzymes promote elongation are different. Following the application of 0.5L of 15 /vM GA 3 to Pisum sativum shoots, Wu et al. (1993) found that invertase mRNA level increased before any increase in invertase activity. Subsequent

PAGE 40

29 elongation of the pea shoots occurred 12 h after the increase in invertase activity was detected. They concluded that GA 3 might regulate the cell wall invertase gene at the transcriptional and/or translational levels. On the other hand, in detached eggplant leaves, no effects of GA 3 could be detected on the activity of invertase. In other plants, such as Zinnia elegans, invertase activity was reduced upon application of an inhibitor of GA 3 biosynthesis, and its effects were counteracted by a subsequent treatment with GA 3 (Kim and Suzuki, 1 989). The effect of GA 3 on invertase activity and/or levels is inconsistent, and appears to depend on species, organs, and/or stage of development. Increased SS activity has been reported in orchids treated with GA 3 (Chen et al . , 1994), but there is no information available on gibberellin effects on sucrose phosphate synthase activity or levels. In summary, fruits can be seen both as utilization sinks (like meristems) during the early stages of development and as storage sinks in later phases of growth. In both cases, fruit development depends upon the continuous supply of photoassimilates. The eventual carbohydrate partitioning in a plant is related to the competitive ability of the various sink regions. Sink strength has been investigated from different points of view, including environmental factors, enzyme activities, and source-sink relationships. However, the role of the number of cells and cell size, i.e., actual sink size, has not been studied in relation to fruit set and fruit development. Furthermore, the relationship

PAGE 41

30 between fruit set/development and GA 3 effects on sink size or sink activity (i.e., soluble sugar levels and sucrose enzymes activity/levels) has not been investigated. The objectives of the present study are a) to determine the effectiveness of GA 3 in increasing fruit set and yield in rabbiteye blueberry, b) to determine the effect of GA 3 on soluble sugar accumulation and sucrose metabolizing enzymes in relation to fruit set and development, and c) to determine the effect of GA 3 on cell number and cell size in blueberry fruit and their relationship to fruit set and development.

PAGE 42

CHAPTER 3 EFFECT OF GA 3 AND POLLINATION ON FRUIT SET Introduction The stimulus for fruit set elicited by pollination and fertilization has not been identified. In some cases, endogenous hormone levels increase after pollination and/or fertilization. Additionally, in many cases, pollination and fertilization can be replaced by exogenous applications of growth regulators. However, a direct relationship between fruit set and hormone levels has not been found. Response varies with several factors, including type of hormone, concentration, application timing, and species and/or cultivar. Increased fruit set and parthenocarpic fruit development have been obtained in several species by using exogenous applications of gibberellins, auxins or a combination of both (Leopold, 1 962; Wittwer et al., 1 956; Webster and Goldwin, 1984; Hartman and Anvari, 1986). Gibberellic acid (GA 3 ) or potassium gibberellate (KGA 3 ) have been used successfully to induce fruit set and parthenocarpic fruit development in lowbush and highbush blueberry (Barker and Collins, 1965; Mainland and Eck 1968a,b; 1969 a,b; Doughty and Scheer, 1975). In rabbiteye blueberry, usually self-unfruitful, the response to gibberellin applications has been inconsistent. Increased fruit set has been 31

PAGE 43

32 observed some years with single or multiple GA 3 applications (Davies and Buchanan, 1979; Davies, 1986; Vanerwegen and Krewer, 1990), but the response appears to be cultivar dependent and optimum concentration and timing of GA 3 application have not been established. Additionally, the sometimes erratic response may be due to excessive concentration of gibberellins, since the above experiments were carried out under field conditions where different levels of crossand selfpollination occur. It is known that endogenous gibberellin levels increase in pollinated/fertilized ovaries of several fruits (Iwahori et al., 1968; Jackson and Coombe, 1966; Mainland and Eck, 1971; Mapelli et al., 1978; Martin et al., 1982). Thus, application of GA 3 to fertilized fruit may lead to supraoptimal GA 3 levels as suggested by Pharis and King ( 1 985). Weaver and Pool (1971) found that increasing concentrations of GA 3 at blossom time reduced fruit set in grape. Similar results were found by Edgerton (1981) in cross-pollinated 'Delicious' apples sprayed with GA + BA at blossom time. Since this effect was counteracted by aminoethoxyvinyl glycine (AVG), they concluded that ethylene was involved in such response. On the other hand, Lane (1 984) found that foreign as well as pioneer pollen stimulated fruit set and suggested that gibberellin enrichment of the stigma may be the mechanism by which foreign or mentor pollen improved fruit set of apple, sweet cherry and apricot. The combined effect of GA 3 , pollen source, and pollen has not been studied in blueberry. Thus, the objectives of the present experiments are a) to evaluate the effect of GA 3 concentration and application time on fruit

PAGE 44

33 set, fruit development, and fruit quality in 'Beckyblue' rabbiteye blueberry and b) to evaluate the interaction between GA 3 and pollination on fruit set and fruit development. Materials and Methods Experiment I Five uniform, 4 year-old rabbiteye blueberry plants, 'Beckyblue' were grown in 12L containers in 1:1 peat:pine bark. Plants were watered every other day and fertilized with 20N-5.6P-1 1 K water soluble fertilizer once a week. On 1 November, plants were placed in a dark cooler at 7±1°C for 30 days. After the chilling period, the plants were transferred to a greenhouse with average day/night temperature of 17/16°C to force budbreak. GA 3 solutions were prepared from Pro Gibb (Pro Gibb 4% a.i., Abbot Laboratories, Chicago III) in Mclllvaine buffer, pH 3.5 (0.02 M Na 2 HP0 4 :0.01 M citric acid) and 0.01 % Tween-80 as a surfactant. Upon opening, individual florets were treated as follows: Treatments 1 . 0.6 mM GA 3 (1 x) 2. 1.4 mM GA 3 (1 x) 3. 0.3 mM GA 3 (2x) 4. 0.7 mM GA 3 (2x) 5. 0.2 mM GA 3 (3x) 6. 0.5 mM GA 3 (3x) Application times Full bloom (FB) FB FB + 7 days after full bloom (DAFB) FB + 7 DAFB FB + 7 DAFB + 21 DAFB FB + 7 DAFB + 21 DAFB

PAGE 45

34 7. 0.1 mM GA 3 (4x) FB + 7 DAFB + 21 DAFB + 42 DAFB 8. 0.4 mM GA 3 (4x) FB + 7 DAFB + 21 DAFB + 42 DAFB 9. Hand-pollinated (Poll) FB 10. Nonpollinated (NP) FB Pollinated treatments were hand pollinated once with 'Climax' pollen. Pollen was collected on the thumb nail by twirling the flowers. Nonpollinated treatments were sprayed at bloom with Mclllvaine buffer (pH 3.5). All treatments were applied on a single plant, 48 to 99 florets/treatment/plant, in a randomized complete block design, with five replications. Fruit abscission was measured on a weekly basis throughout fruit development and final fruit set was determined 10 weeks after anthesis. Fruit development period was calculated when 80% of the fruits had been harvested. For all harvests, fruit from each treatment were pooled for fresh weight, soluble solids, and pH measurements. Soluble solids were determined with an Abbe refractomer Mod. 1 0460 (Cambridge Instrument Inc., Bufalo, N.Y., U.S.A.). Total yield was calculated as a product of fruit fresh weight and final fruit number. Experiment II Forty five 1 -year-old 'Beckyblue' rabbiteye blueberry plants, grown in 2L containers in a 1 : 1 peat:pine bark mixture, were chilled and forced to budbreak in the same way as described above, except the average day/night temperatures in the greenhouse were 25/22°C. Nine plants were assigned to each of the following treatments selected from experiment I:

PAGE 46

35 Treatments Application times 1. 0.3 mM GA 3 (2x) 90% full bloom (FB) + 7 DAFB 2. 0.7 mM GA 3 (2x) 90% FB + 7 DAFB 3. 0.4 mM GA 3 (4x) 90% FB + 7 DAFB + 21 DAFB + 42 DAFB 4. Hand-pollinated (Poll) 90% FB 5. Nonpollinated (NP) 90% FB Pollinated, nonpollinated, and GA 3 treatments were applied as described previously except that each plant received only one treatment in order to eliminate competion effects among treatments. Prior to treatment application, each cluster was tagged and the number of florets and date was recorded. The number of florets per cluster varied from four to seven and the total number of florets per treatment ranged from 235 to 313. All treatments were applied in the afternoon. The experiment was a randomized block design, with plant size as the block, 2 blocks per treatment and four or five plants per block. Fruit abscission was monitored on a weekly basis and final fruit set was determined 48 days after bloom. Fruit development period was determined when 80% of the fruits were harvested. All harvests were pooled to determine titratable acidity, soluble solids, and pH. Titratable acidity was determined by titrating 5 mL juice with 0.1 N NaOH to pH 8.2 using an automatic titrator Mod. 380 (Fisher Scientific Co. U.S.A.). Soluble solids were measured as described.

PAGE 47

36 Experiment III Nine uniform, 2 year-old rabbiteye blueberry plants 'Beckyblue', grown in a 1:1 peat:pine bark mixture in 12L containers were chilled and forced to budbreak as described above. GA 3 was prepared as described and applied at full bloom (FB) and at FB+ 7 DAFB in the following array of treatments: 1. Cross-pollination (saturation, =200 pollen tetrads) (X) 2. Cross-pollination (saturation) + 0.7 mM GA 3 at FB + 7DAFB (X + GA) 3. %X cross-pollination (50 pollen tetrads) (V+X) 4. Va X cross-pollination (50 pollen tetrads) + 0.7 mM GA 3 at FB + 7DAFB (%X + GA). 5. Self-pollination (saturation) (S) 6. Self-pollination (saturation) + 0.7 mM GA 3 at FB + 7DAFB (S + GA) There were 96 florets per treatment and all treatments were applied on the same plant. Results were analyzed as a randomized complete block design. G A 3 treatments were applied in the evening, at least four hours after pollination. The number of pollen tetrads selected for the treatments was based on work by Parrie and Lang ( 1 992, unpublished) indicating that blueberry stigmatic pollen saturation, i.e., the maximum amount of pollen grains that the stigmatic fluid can hydrate, ranges from 200 to 300 pollen tetrads. We considered that 50 pollen tetrads represent a quarter of the saturation level. Pollen tetrads were counted on a 100 mm 2 black lid under a stereoscope, 1 5x magnification. The pollen was transferred by touching the black lid to the stigma. After each

PAGE 48

37 transfer, pollen tetrads left on the black lid were transferred onto the stigma with a spatula. This process was repeated until all the pollen tetrads were on the stigma. Fruit abscission was recorded as above, and final fruit set was determined at 56 DAFB. Fruit was harvested every other day and weekly harvestings were pooled to determine fruit development period and fresh weight. Subsequently, fruit was macerated with a mortar and pestle and seeds were separated and air dried for 24 hours. Total seed number and total seed weight per fruit were recorded. Seeds were separated into small and large seeds using a sieve of 1 mm 2 , counted and weighed. Results and Discussion Experiment I The time-course and extent of fruit abscission for all treatments is shown in Fig. 3-1 and 3-2. Pollinated and nonpollinated controls are included in both figures for comparison. The majority of fruit abscission in the GA 3 and pollinated treatments occurred in 2 stages: from 0 to 35 DAB and from 56 to 70 DAB. There was no difference in fruit abscission among treatments from 0 to 42 DAFB, except at 14 DAFB when abscission in pollinated treatments was significantly lower than in treatments with one application of 1 .4 mM GA 3 and three applications of 0.5 mM GA 3 (Fig. 3-1 ) and 2 applications of 0.3 mM GA 3 (Fig. 3-2). From 49 to 70 DAFB, there were no differences in abscission between GA 3 and pollinated treatments; however, the extent of abscission in

PAGE 49

38 the nonpollinated treatments was significantly greater than in the other treatments during this time. Our results are different from those reported by Davies (1986) who found only one peak of fruit abscission at 28-42 days after anthesis in 'Tifblue', 'Woodard' and 'Bluegem'. Final fruit set was significantly lower in the nonpollinated treatment compared to the pollinated and GA 3 treatments (Table 3-1). There was no difference in final fruit set among any of the GA 3 or pollinated treatments. These results are difficult to compare with previous experiments since application timing was different. However, Davies and Buchanan (1979) reported that GA 3 increased fruit set in field grown rabbiteye blueberry compared to the naturally pollinated control when GA 3 was applied at 30-40% bloom and 1 5 days later or when a single application of GA 3 was applied at 60% bloom. Similarly, NeSmith and Krewer (1991) observed an increase in fruit set in 'Tifblue' rabbiteye blueberry with a single application of 0.9 mM GA 3 at balloon stage (just before opening) or 2 applications at full bloom and 10 or 18 days later. In both cases, set of the naturally pollinated control was significantly less than that obtained by hand pollination in the present experiment. Fruit fresh weight was greatly reduced by all GA 3 treatments (Table 3-1 ). Fruits from the pollinated control were four times the size of those from GA 3 treatments. Unlike the results reported by Davies and Buchanan (1 979), where increased fruit set and yield tended to reduce fruit fresh weight, in the present

PAGE 50

39 experiment neither fruit set nor total yield were correlated with smaller fruit size. Fruit set in the pollinated treatment was as high as fruit set in the GA 3 treatments, while total yield was double that of GA 3 -treated fruits. The initial number of flowers was smaller for pollinated treatments (avg. 48 flowers/plant) than for GA 3 treatments (avg. 73 flowers/plant). The decrease in fruit size of the GA 3 -treated fruits may be due to the increased fruit load in these treatments. However, calculations of fruit number within each treatment reveal no significant differences in fruit number between the pollinated and several of the GA 3 treatments (1 application of 0.6 mM or 1 .4 mM GA 3 , 4 applications of 0.4 mM GA 3 ), yet fruit weight was still significantly less in all of the GA 3 treatments. Seed number strongly affects blueberry fruit size (Moore et al., 1 972; Lyrene, 1 989; Payne et al., 1 989) and parthenocarpic blueberry fruits are smaller than seeded fruits (Mainland and Eck, 1969b). Although seed number was not measured in this experiment, absence of seeds was observed in all fruits from GA 3 treatments. Total yield, measured as fruit times fruit fresh weight, was less for GA 3 treatments compared to pollinated treatments. Yield is a function of flower number, fruit set and fruit weight. In this case, fruit weight appears to be the determining factor in total yield. This is supported by a high correlation between yield and fresh weight (r = .71, p<.01), while correlations between yield and fruit set or yield and flower number were much lower (r = 0.49, p < .01 and r = -0.02, p<.91, respectively).

PAGE 51

40 GA 3 treatments increased the fruit development period by 34 to 49 days compared to the pollinated control. Lower GA 3 rates appeared to have a slightly more negative effect than higher rates, although there were no statistical differences between them. A delay in harvest date has been reported as a result of gibberellin applications to blueberry (Mainland and Eck, 1 969, Vanerwegen and Krewer, 1 990), but not to such an extent as observed in the present experiment. In blueberry, fruit development period and fruit seed number are negatively correlated (Lang and Danka, 1991; Lyrene, 1 989). In the present experiment, GA 3 -treated fruits were mostly seedless and this probably accounts for the increased fruit development period. There were no differences in soluble solids and pH among GA 3 treatments (Table 3-1). However, pollinated fruits had significantly higher soluble solids than GA 3 treatments, with no difference in pH. Similar results were obtained by Davies and Buchanan (1979) and Vanerwegen and Krewer (1990). No explanation for this observation has been offered. Crane and Grossi (1960) reported that fig sweetness decreased as GA 3 concentration increased, and they attributed this to the promotory effect of GA 3 on elongation of new shoots that competed with fruits for assimilates. The decreased fruit weight and, as a consequence, the decreased yield observed in the GA 3 -treatments compared to the pollinated treatment, may be related to the competitive ability of the different sinks to attract assimilates. In the present experiment, all treatments were applied on the same plant, and as

PAGE 52

41 Goldwin ( 1 984) observed, pollinated fruit are stronger sinks than parthenocarpic fruits and thus would accumulate more assimilates. This is supported, in part by the increased soluble solids content in pollinated fruit compared to GA 3 treated fruits. A decrease in soluble solids in fully colored seedless, GA 3 -treated 'Delaware' grapes was also reported by Clore (1965). Kishi et al. (1958), cited by Clore (1965), suggested that GA 3 -treated fruits must be harvested when they show a deeper" color than the controls in order to have the same amount of soluble solids. Experiment II Extent of fruit abscission was lower for GA 3 treatments than for pollinated treatments throughout the fruit development period (Fig. 3-3). In GA 3 treatments, about 20% of the fruit abscised during the first 20 DAFB, after which very little abscission occurred. Maximum fruit abscission in pollinated and nonpollinated fruits occurred within 1 0 DAFB, reaching 66% and 1 00% for pollinated and nonpollinated treatments, respectively. Final fruit set was increased significantly by multiple GA 3 applications compared to pollinated controls (Table 3-2). While percent fruit set induced by GA 3 was basically the same as in the previous experiment, fruit set of the pollinated control was comparatively lower. Average fruit size in the pollinated treatment was twice as great as fruit size of the GA 3 treatments. This may be related to fruit load as has been observed in other fruit crops (Coggins et al., 1 960; Forshey and Elfving, 1 977;

PAGE 53

42 Garci'a-Martinez and Garcia-Papf, 1979). By 20 DAFB, 50% fruit abscission occurred in the pollinated treatments, while only 20% abscission occurred in the GA 3 treatments. In fact, fresh weight and fruit set were negatively correlated (r = -0.65, p<.01). Davies and Buchanan (1979) found that GA 3 treatments that increased fruit set and yield tended to decrease fruit weight in rabbiteye blueberry. The smaller fruit size in GA 3 treatments may also be related to absence of seeds, as discussed previously. Total yield was similar among all treatments. Low fruit set in pollinated controls was compensated for by a higher fruit fresh weight, a determinant factor of yield in this case. A delay in harvest of about 1 0 days was observed with 2 applications of 0.3 mM GA 3 and four applications of 0.4 mM GA 3 (applied at 24 and 42 DAFB, respectively) compared to the pollinated control. However, there was no difference in the fruit development period between 2 applications of 0.7 mM GA 3 and the pollinated control. The different effects of GA 3 treatments may be related to the concentration of GA 3 used. Mainland and Eck (1968) found that in highbush blueberry, fruit development period was 17 days shorter in fruits treated with 1.4 mM KGA 3 compared to the pollinated treatments, but fruits treated with lower concentrations of KGA 3 ripened much later than pollinated fruits. Vanerwegen and Krewer (1990) also found a delay in maturation with 2 applications of 0.9 mM GA 3 compared to one application at the same rate. There was no difference in soluble solids and titratable acidity among treatments (Table 3-2). Juice pH tended to be lower as the GA 3 concentration

PAGE 54

43 and/or application number increased. Lower pH may be related to maturation, but the differences in pH do not correspond to the values for titratable acidity or soluble solid contents, i.e., low pH usually corresponds with higher values of titratable acidity and lower soluble solids. In general, soluble solids and titratable acidity were within the ranges reported by Spiers (1981) for rabbiteye blueberry. Results from Experiments I and II were different in several respects, including fruit set, fruit weight, total yield, and fruit development period. Higher fruit set was observed in pollinated fruits in Experiment I than in Experiment II, but fruit set was about the same in GA 3 -treated fruits in both experiments. Fruit fresh weight was higher in all treatments in Experiment II than in Experiment I. Furthermore, fruit weight in the pollinated treatment was 4-fold greater than fruit weight in the GA 3 treatments in Experiment I, but only 2-fold greater in Experiment II. Consequently, total yield was significantly increased in the pollinated treatment compared to the GA 3 treatment in Experiment I, but not in Experiment II. Finally, fruit development period was longer for all treatments in Experiment I compared to Experiment II, with GA 3 treated fruits having a more delayed maturation period than pollinated fruits. Several factors may account for some of these differences. All treatments were applied to the same plant in Experiment I and this may explain the relatively greater decrease in weight of GA 3 -treated fruits compared to pollinated fruits. Since pollinated fruits are stronger sinks than parthenocarpic

PAGE 55

44 (i.e., GA 3 -treated) fruits (Goldwin, 1984), pollinated fruits would outcompete GA 3 -treated fruits for assimilates. Because fruit size was the dominant factor in yield in the present experiments, this would also explain the differences in yield between the pollinated and GA 3 treatments in the 2 experiments. The decrease in fruit set in the pollinated treatment and the decreased FDP in all treatments observed in Experiment II relative to Experiment I may be partially due to differences in temperature conditions between the 2 experiments. Temperature may affect both fruit set and fruit development period in several fruit crops (Batjer and Martin, 1 965; Jackson and Coombe, 1 965). Day/night temperatures of 26/21 °C compared to 26/1 0°C decreased fruit set 23% in rabbiteye blueberry, but had no effect on fruit development period (Williamson et al., 1994). In peach, high night temperatures (21°C) decreased the fruit development period compared to 13°C night temperatures (Batjer and Martin, 1965). Thus the higher day/night temperature in Experiment II (25/22°C) compared to Experiment I (17/1 6°C) may account for the decreased fruit set in the pollinated treatment and the decreased fruit development period in all treatments. Differences in root system due to differences in plant age and container size may also account for some of the observed differences between Experiment I and Experiment II. Cytokinins are synthesized in the roots and exert a strong influence on cell division in all organs, including shoots and fruits (Torrey, 1976). Thus, differences in cytokinin biosynthesis and translocation to the shoots may have occurred in the two experiments.

PAGE 56

45 Experiment III Fruit abscission was minimal for X, X + G A, % X + G A, and S + G A from 0 to 28 DAFB, averaging 5% for all treatments (Fig. 3-4). This was followed by a slow increase in fruit abscission from 28 to 42 DAFB, after which fruit abscission ceased. In contrast, fruit abscission in the 14 X and S treatments increased slowly from 0 to 28 DAFB, followed by a dramatic increase in fruit abscission from 28 to 42 DAFB. This pattern of fruit abscission resembles the one-wave abscission pattern obtained by Davies and Buchanan (1979) for 'Woodard', 'Bluegem', and 'Tifblue' under field conditions. There were no differences in final fruit set among X, X + GA, Â’AX + GA and S + GA treatments, with fruit set averaging about 80% (Fig. 3-5). Thus, GA 3 applications did not interact negatively with pollen source and decrease fruit set as has been observed in other crops like grape (Weaver and Pool, 1991) and apple (Edgerton, 1981). Final fruit set of % X and S treatments was significantly less (p<0.05) than the other treatments, as well as between each other. Fruit set in 14 X treatment was fairly good, averaging 60%, while set in S treatments averaged less than 20%. Similar fruit set for self-pollinated rabbiteye blueberry was reported by El-Agamy et al., (1982), Lyrene and Goldy (1983) and Garvey and Lyrene (1987). The relatively high fruit set under conditions of 14 X pollination suggests that adequate yields may be obtained in the field with sub-optimal pollination.

PAGE 57

46 There were no differences in berry weight among X, X + GA, S, and S + GA treatments (Table 3-3). Fruits from X and X + GA treatments had the highest weight, averaging 1.7 g. Self-pollination alone did not decrease fresh weight, in contrast to reports by Coville (1921) and Meader and Darrow ( 1 944) in highbush blueberry, and Lyrene (1989) in rabbiteye blueberry. Fruits from 'AX and 'A X + GA treatments were the smallest, averaging 1.3 g. Thus, although a sub-optimal (i.e., 'A saturation) pollination does not appear to limit fruit set, it does negatively affect fruit size, which would be expected to decrease total yield. The number and weight of large seed per fruit was greatest in the crosspollinated (X) treatment compared to the other treatments (Table 3-3). There were no differences in large seed number or weight between X + GA and 'AX treatments. Seed number and weight were significantly less in the % X + GA and the S + GA treatments compared to X, X + GA and 'AX treatments. The number and weight of small seeds per fruit were not statistically different among treatments. There were no differences in total seed number among treatments, but total seed weight per fruit in the cross-pollinated (X) treatment was significantly higher than the other treatments. There was a positive correlation between fresh weight and total seed number (r = 0.60, p<.01), as previously reported by Meader and Darrow (1944), Alders and Hall (1961), and Darnell and Lyrene (1989). There was also correlation between fresh weight and number of small seeds per fruit (r = 0.55, p<.01). A weaker correlation

PAGE 58

47 between fresh weight and number (r = 0.45, p < .01 ) and weight of large seeds (r = 0.42, p<.01) was observed. A comparison of the X and X + GA treatments, and the !4X and %X + GA 3 treatments indicates that GA 3 inhibits the development of large seeds in previously pollinated flowers. GA 3 may be acting as a pollenicide, as reported by Weaver and McCune (1960) in grape flowers. Alternatively, since GA 3 was applied 4-6 h after pollination, there is a possibility of a pollen wash-off effect. In either event, however, the number of small seeds as well as the total number of seeds were not affected by any of the GA 3 treatments. The fruit development period was similar among all treatments except for the self-pollinated (S) (Table 3-3). Eighty percent of the fruit in X, X + GA, and 14 X + GA treatments was harvested in the period 63-72 DAFB (data not shown). Harvesting was delayed 1 5 days in the 14 X and S + GA treatments and only 43% of the S fruits were harvested in the period of 72-92 DAFB (data not shown). Seed number and fruit development period are closely related. Alders and Hall (1 961 ) reported that in lowbush blueberry, increased seed number was significantly associated with increased fresh weight and shorter fruit development period. Coville (1 921 ) found similar results in highbush blueberry, as well as Lyrene ( 1 989) in southern highbush and Meader and Darrow ( 1 944) and Tamada et al. (1977) in rabbiteye blueberry. In this experiment, as seed number decreased, fruit development period increased (r = -.64, p<.01).

PAGE 59

48 Our results indicate that under the conditions of these experiments, GA 3 does not interact negatively with any type of pollination tested. Moreover, under situations of partial pollination and/or self-pollination, GA 3 can successfully replace the stimuli for fruit set delivered by normal pollination and fertilization in rabbiteye blueberry. Conclusions Maximum fruit abscission occurred during the early stage of fruit development, presumably the stage when cell division is occurring. The application of GA 3 alone or in combination with pollination prevented fruit abscission compared to the nonpollinated and self-pollinated treatments, suggesting that GA 3 can replace the stimulus for fruit set delivered by pollination and fertilization. Multiple applications of GA 3 did not change the pattern of fruit abscission compared to single applications or pollination. The lack of response to increasing GA 3 concentrations and/or number of applications indicate that factors other than gibberellins may be involved in the fruit set of rabbiteye blueberry. Fruit size was decreased by GA 3 treatments alone compared to pollinated treatments, but GA 3 application in combination with full crossor self-pollination did not affect negatively fruit size in 'Beckyblue'. The decreased fruit size obtained by 'AX pollination was partially reversed by GA 3 . Thus, the combination of GA 3 with partial pollination can apparently provide a similar

PAGE 60

49 stimulus as full pollination for growth after flower opening, while GA 3 alone cannot. An increase in the fruit development period was observed when GA 3 was applied alone compared to pollinated treatments. However, no such negative effect was observed when GA 3 was applied in combination with any pollinated treatments. This suggests that under optimal pollination conditions, no detrimental effects of GA 3 treatments on fruit set, fruit size, or fruit development period are to be expected. Furthermore, GA 3 applications could be beneficial under suboptimal pollination conditions, since the presence of some seeds may increase fruit size and decrease fruit development period. Soluble solids, titratable acidity and pH were not consistently affected by GA 3 treatments. These aspects need more careful consideration. A decrease in soluble solids has been observed in other studies as one of the negative effects associated with GA 3 applications. No explanations have been advanced as to the basis for this observation. It is possible that harvest criteria, usually color, may need modification if GA 3 is used.

PAGE 61

Fruit abscission (%) 50 100 80 60 40 20 POLL 0.6 mM GA 3 1.4 mM GA 3 NP * * * k v « i i-i f I T I' f '' z £. O' _ Figure 3-1 . Time-course and extent of fruit abscission in 'Beckyblue' rabbiteye blueberry as affected by gibberellic acid (GA 3 ), pollination (POLL), or nonpollination (NP). 3x = 3 applications of GA 3 ; application times indicated by arrows. Values are means ± SE, n = 3; SE bars present only when larger than symbol.

PAGE 62

Fruit abscission (%) 51 Figure 3-2. Time-course and extent of fruit abscission in 'Beckyblue' rabbiteye blueberry as affected by gibberellic acid (GA 3 ), pollination (POLL), or nonpollination (NP). 2x and 4x = 2 and 4 applications of GA 3 ; application times indicated by arrows. Values are means ± SE, n = 3; SE bars present only when larger than symbol.

PAGE 63

52 Figure 3-3. Time-course and extent of fruit abscission in 'Beckyblue' rabbiteye blueberry as affected by gibberellic acid (GA 3 ), pollination (POLL), or nonpollination (NP). 2x and 4x = 2 and 4 applications of GA 3 ; application times indicated by arrows. Values are means ± SE, n = 9; SE bars present only when larger than symbol.

PAGE 64

53 Figure 3-4. Time-course and extent of fruit abscission in 'Beckyblue' rabbiteye blueberry as affected by gibberellic acid (GA 3 ), pollen source, and pollen amount. X = cross-pollination (200 pollen tetrads); 1/4X = 50 pollen tetrads; S = self-pollination (200 pollen tetrads). GA 3 = 0.7 mM GA 3 ; GA 3 application times indicated by arrows. Values are means ±SE, n = 9; SE bars present only when larger than symbol.

PAGE 65

54 Treatments Figure 3-5. Effect of GA 3 , pollen source, and pollen amount on final fruit set of 'Beckyblue' rabbiteye blueberry fruits. X = cross-pollination (200 pollen tetrads); !4X= cross-pollination (50 pollen tetrads); S = self-pollination (200 pollen tetrads); GA 3 = 0.7 mM GA 3 applied at FB and FB + 7DAFB. Values are means ±SE, n = 9; SE bars present only when larger than symbol.

PAGE 66

Table 3-1. Effect of GA 3 on fruit set and development in 'Beckyblue' rabbiteye blueberry in 1992. 55 CO CO CO CO CO CO CO CO CO X Q_ 03 X r* X X CD X 1 CO CO 00 X CO 00 00 00 co >. X XI X X X CO X X X X CO co CO \_ ct 't X X X X 1 CO CO o o r— o o o CM X % — ^ — ^ GO T " " LL CO W CO X X X X X X X o C N 'w > CO CO CO CO CO CO ro X o Q_ CO CO CO 00 CO o V-> (D Q ro CO r-1 o 03 't X 03 X 1 O u_ X CO CM X CM CM X CM c X 1 + T T y— T“ LL m LL X ro 2 X c 03 X c X X X X X X X X co CO 0Q LL ro CM 03 CD o CM *ct O x < "ro CL 00 LO r-1 X X X T-! CM 03 LL Q o ^03 CM CM X CM X X X CM C/3 KC CM o + ’M ro £ ’3 kX CM X CM X X X X X o X X X CO O r\ o "5. Q. + m LL soc LL H— 03 03 CO X X X ''dX X sjX X X 0 CQ II • • , — CM LL . Cl . * < X C/3 X X LL o LL < 4-» if) CD 4-^ CD X X X X X X X X X o "o r^» O 4—* CO Cp O' > 5 00 co CO CO X CO CO CO CO co X co o CO X CO •3o w 0 E o + m CM Tj< A CO "co *D k_ CM 03 X 00 03 CM X o LL + c/) > c CD LL X 1 Cv CO X O' 3 X OQ CD CD o — LL CM C ^ C/3 3 H— i + ro X II ( 0 o X M 00 CM X CD X CM o X c C o + 4—' k_ > k_ c d 00 o X CM o X X , — X o w X o -LJ X 4-^ ro LL L_ c $ +1 +1 +1 +1 +1 +1 +1 +1 +1 +| . ro o "a. X o Q. E _o LL 6 c o o o CM CM CO X X 00 CM X X X CM 00 r+ o CM X o k_ 0 X "o. X ro X LL < "O CD E 3 O o X rX co \T X ro X a i— o ro +-< * r _ *4— c . t J— ^ CM c 0 w CQ c X E 0 LL + CD k_ X E < 4— » o '*-> Q + C 4-* r— X 0 c X c n c o L_ 0 o E LL o L_ X ro uro X 0 C/3 E +-» CD CD k_ ho CO o o E cd E o CO o cd E CD E o CO u CD E CD E o X CO o CD E CD E CO c "5 X •M 3 i— H— ro o X Q. O o X) x' LL < a CD CD k_ CD "a. a CO CO o "o. X CO X o r* o "5. Cl CO CM o X o "5. a CO i — o o X c co X 1 c o 11 Q. CO co < ZD k_

PAGE 67

Table 3-2. Effect of GA 3 on fruit set and development in 'Beckyblue' rabbiteye blueberry in 1993. JO I Q_ 0 -Q o -Q CD 0 CM 1 CO CO CM CO X 0 0 0 0 CD 00 cn CD < ICO cn 00 CO 1 o o o o 0 0 0 0 > CO CO o CD T CO 4— CM CM CM r " Dl £ 0 JO 0 JO 00 t — CO Q 0 CO id 1 Lc "o CO LO CO LO T3 .E £ 0 0 0 0 > 0 CO q "3"0 TjL o LO CD CO 1 -4-3 o C33 HCO CM CM CM JO JO JO 0 FW /fru CO cn 4 o 1 .E? r ~ i — T— CM 0 .t: ^ > ra 0 JO 3.3c D ' — 1 1 LL % CM CD 00 "Ct 00 00 o CO (/) r-* CO «-* LO CD i^LO 0 0 LO CO CO csi ^ Q. +1 -H +1 +1 +1 o 0 o "5. CL (D DO LL < Q E o JO JO 0 4-> 0 (/) > CO -a r-' + DO T3 C CO CO CQ E o _o JO < Q C\l '3+ LO o o w 0 -= « W « V T > CM + + DO ~o O 0 Q. 44 C 0 E Q. . O 0 o > w 8 £ 5| +w 0_ 11 Q C/3 U. C/3 0 ll o = CO SLJ CL < ro Q CM > W 4-4 0 ig E o ro « « ro ro 0 s ^ CO O _l 4-» L. > o JO ;r w CL C T3 E FE o 5 ° « .E c J= 0 .-t: c o 0 ™ 4—* CJ 0 L= >Q. -CL ~ 0 II Jp < < ‘ £ P C o CD 2 QQ) CD <: w _ c 2 ro 0 Q3 Q ^ 56

PAGE 68

57 -C 05 05 £ -o 05 05 C/5 05 JO E 3 c "O 05 05 C/5 T5 C (0 -C 05 *05 $ -C C/5 05 C o <-' c 3 o E co c _Q5 "5 Q. TD C co o' 05 3 ° ^ CO >“ II II , JO n < > O o s.t; O -O M o to o LLI O • JO 00 > 1 O 05 O S“ £ c "m TJ JZ — iS 05 C 55 05 CO JO J 2 X! JO JO £ Sgl CM LCD 't CM OD OD 00 5 OD 1 ^' 00 00 CD c c CO CO CO CO CO CO "co "C 't CM CD CM OD +-• 0 • O 0 OD t — r» 6 1 — CO CO CO CM CM 'Ct CM 05 co E CO CO CO CO CO CO _E CO CM CO 00 CO (-> C/D OD 00 LCD CM CD sz 05 05 s "O 05 05 CD JO O o O JO o 05 05 k_ CD o 00 in CM CM c n -J CM CM 6 d co CO CO co CO CO "CD C rCM OD CD CM CO 6 c to 05 00 CO CD 00 h-’ 2 CM CM CM CM CO CM TJ 05 05 CO CD O) k_ CO CM -Q JO CD CD JO ° CD | O LTD CO OD o P 00 00 CD CM o D> CO CO CO CO JO co = Q co CM CM CM CM LL TD r>« 00 Cv OD 00 >+ i j $1 JO CO X > 05 co CM o LD o J 0 CO O JO CO 00 CO JO CO LL. it; CO 00 + ID _D 5 i j N CO M c < L Q) CO a HC*) E < + < c 1 CD X X CD CD k_ + + c K X X T C/D C/D S CQ LL H_ < 0 Q co II + C/D 00 „ LL CO -O -o co c «o £ S c £. 1 E O -Q LCD = ,, 3 X co TD o LCD ,o co co cl co Cl "O ro CO (*> ffl < J= u c E =2 ^ 0.0 o g < O || c o. Q5 CO = -o +0 (0 <0 Q. 05 > 05 X ^ <0 CO c c E 05 CO L= = 05 <-> g .5 C /5 -c O tc o > CM > > — J 3 1 1 £ 1 a! ^ § s a S.E 8 0 o > o ® s II P c 1 o X £ -43 CO L_ CO a 05 CO £ CO ^ c o c 05 E CO c co C CO Q 5 L_ Q 5 H > LX x

PAGE 69

CHAPTER 4 CARBOHYDRATE METABOLISM AND FRUIT GROWTH IN GIBBERELLIC ACID-TREATED VS POLLINATED FRUITS OF BLUEBERRY Introduction Parthenocarpic fruit set and development can be induced in blueberry by exogenous applications of gibberellic acid (GA 3 ); however, final fruit size of parthenocarpic fruit is often smaller than that of pollinated fruit. The mechanism by which GA 3 induces fruit set and development is unknown. It has been observed, however, that gibberellins, as well as other plant hormones, alter the pattern of assimilate distribution within the plant (Daie, 1 985; Kinet et al. , 1985; Aloni et al. , 1986). Changes in assimilate distribution, specifically, enhancement of carbohydrate accumulation at the site of hormone application, have been correlated with fruit growth and development in a number of species (Marre and Murneck, 1952; Stembridge and Gambrell, 1972; Adedipe et al., 1976; Craighton et al., 1986; Pereto and Beltran, 1987; Stutte and Gage, 1990). Sucrose is the major translocated carbohydrate in most plants, including blueberry (Avigad, 1 982). The extent of sucrose translocation to fruit depends, in part, on the concentration gradient between the loading and the unloading 58

PAGE 70

59 sites (i.e., developing fruit). Factors that decrease the sucrose concentration in the unloading area may, in theory, enhance sucrose transport to that site (Aloni et al. , 1986; Daie et al. , 1986; Miyamoto and Kamisaka, 1988, 1990; Morris and Arthur, 1985). The metabolism of sucrose in fruit tissues can be initiated by three enzymes that are temporally and spatially compartmentalized, depending on species. These enzymes are sucrose phosphate synthase (SPS, EC 2.4.1 . 1 4), which is involved in sucrose synthesis; invertase (EC 3.2.1.26), which hydrolyses sucrose to glucose and fructose; and sucrose synthase (SS, EC 2.4.2.13), which is thought to function primarily in a sucrose degrading direction. Changes in activities of these enzymes have been associated with developmental processes in sink organs (McCollum et al., 1 988; Hubbard et al., 1989, 1991; Nielsen et al., 1991; Yelle et al., 1991; Stommel, 1992; Sin et al., 1 992; Wang et al., 1 993). In fact, increases in invertase activity have been reported to precede growth of pea ovaries (Estruch and Beltran 1991), stem segments (Morris and Arthur, 1 984), lima bean embryos (Xu et al., 1 989), and developing fruits of Cucumis (Schaffer et al., 1987a). Plant hormones may exert their effects on carbohydrate accumulation and subsequent fruit development by affecting activities and/or levels of one or more of these enzymes. For example, auxins have been reported to stimulate both invertase activity (Morris and Arthur, 1986; Pooviaiah, 1985; Schaffer et al., 1987b; Tanaka and Uritani, 1 979) and its de novo synthesis (Gordon and Flood, 1 980;

PAGE 71

60 Pressey and Avants, 1 980) in a number of tissues. Increased invertase activity has also been observed in soybean fruit following abscissic acid application (Akerson, 1 985). Gibberellic acid stimulated invertase activity in pea internodes (Broughton and McComb, 1971; Morris and Arthur, 1 985), sweet potato roots (Tanaka and Uritani, 1979), and Avena segments (Kaufman et al. , 1973). In the orchid Phalaenopsis, GA 3 stimulated sucrose synthase activity (Chen et al., 1 994). There are no reports on plant hormones affecting activities or levels of SPS. In the present study, developmental changes in activities of sucrosemetabolizing enzymes and carbohydrate accumulation in fruit were assessed and correlated with growth obtained by exogenous application of GA 3 or pollination. The objective was to determine if differences in fruit growth could be attributed to differences in enzyme activities and carbohydrate accumulation in fruit. Materials and Methods Plant Material Softwood cuttings of 'Beckyblue' rabbiteye blueberries were rooted in Spring 1990 and grown outdoors in 22L pots in a 1:1 peat:pine bark mix. Plants were watered every other day and fertilized with 20N-5.6P-1 IK water soluble fertilizer once a week. In December 1991, 9 uniform plants were transferred to a dark cooler and chilled at 7 ± 1°C for 30 days. After the

PAGE 72

61 chilling period, plants were placed in a greenhouse with average day/night temperatures of 27/1 6°C to force budbreak. Treatments Flower clusters were thinned to 4 to 5 florets/cluster at bloom, removing the most developed and least developed florets. The total number of florets ranged from 740 to 920 florets/plant. Treatments consisted of hand-pollination at bloom with 'Climax' pollen (a cross-compatible cultivar), application of 0.7 Mm GA 3 (Pro Gibb 4%, Abbott Laboratories, Chicago, ILL, prepared in Mclllvaine buffer [Dawson et al., 1 986] pH 3.5, and .01 % Tween-80) sprayed at bloom and again 7 DAFB. The nonpollinated treatment consisted of a spray application of buffer and surfactant at bloom. Treatments were arranged in a randomized design with 3 single plant replications per treatment. From 0-45 DAFB, data from the 3 treatments were analyzed by ANOVA. Following the abscission of nonpollinated fruits at 45 DAFB, data from the pollinated and GA 3 treatments were analyzed by 1-way ANOVA. Three sub-samples/plant were taken at 0, 2, 4, 1 0, 24, and 45 DAFB for all treatments. Subsequent to the abscission of the nonpollinated fruits (45 DAFB), data were taken on the remaining two treatments based on phenological development. Data were taken at the end of period II of fruit growth (estimated to be 45 DAFB for pollinated fruits and 72 DAFB for GA 3 ), at fruit color break (65 DAFB for pollinated fruits and 80 DAFB for GA 3 -treated

PAGE 73

62 fruits), and at ripening (72 DAFB for pollinated fruits and 87 DAFB for GA 3 treated fruits. Fruit Carbon Dioxide Exchange Fruit growth was determined for each treatment by estimating carbon (C) cost (fruit DW gain plus respiratory loss) throughout development. The C supply was broken down into C supplied by fruit photosynthesis and C imported into the fruit. This was done to determine if differences in C source existed among treatments. Fruit respiration and photosynthesis were determined by measuring net C0 2 exchange of attached flowers and fruits in the laboratory under light (1000yc/mol m V 1 photosynthetic photon flux [PPF] emitted by a 400-W high-pressure sodium vapor lamp [Lucalox]) and dark conditions. At each sampling, the cluster (two to 4 florets/fruits) was enclosed in a 104 ml plexiglass chamber and flushed for 20 min with ambient air at 1 L m \ The chamber was then closed and fruit photosynthesis was quantified by monitoring the decrease in C0 2 concentration after 30 min in the closed system. Dark conditions were established by wrapping the chamber with aluminum foil and black plastic. After air flushing for 20 min, the chamber was closed and respiration was determined after 30 min (unripe fruits) and 5 min (ripe fruits) by measuring the increase in C0 2 concentration. Temperatures were maintained at 25±2°C and 21 ±1°C light/dark conditions, respectively. Gas samples (500 /vl) were taken with a 1 ml syringe and C0 2 was analyzed by

PAGE 74

63 gas chromatography (Fisher Scientific, Pittsburgh; Model 1 200), using an 80 to 1 00 mesh Colmpak PQ column, injector and column temperatures of 60°C, and a thermal conductivity detector. Aftertaking fruit C0 2 exchange measurements, fruits were harvested and FW recorded. Samples were then lyophilized to constant weight for dry weight determination. Calculations of Carbon Budget C0 2 exchange data in the light and dark were extrapolated to a per-day basis assuming a 12-h photoperiod, saturating PPF (1000 /ymol m^s' 1 ), and mean temperatures of 25/21°C (day/night). The following calculations were derived from the equations established by Birkhold et al. , (1992). The average daily C cost for a given stage of fruit development was calculated from fruit C accumulation plus respiratory C loss. Daily fruit C accumulation was determined from fruit DW measurements. Previous work indicates that C content of blueberry fruits averages 43% of fruit DW throughout development (Birkhold et al., 1 992). Thus the average daily fruit C accumulation was calculated by: Mean daily C accumulation = (0.43 DW (t2) 0.43 DW (t1) )/ t 2 -t, [ 1 ]

PAGE 75

64 Daily fruit C respiratory loss was calculated by multiplying the hourly dark C0 2 exchange rate by 24. The average daily fruit C respiratory loss between times t, and t 2 was done by interpolation between the two measurement points and was calculated by: Mean daily respiratory C loss = [daily resp C, t1) + daily resp C( t2) l/2 [2] Thus, the average daily C cost for a given stage of fruit development was the sum of the mean daily C accumulation and mean daily respiratory C loss. The average daily C supply is composed of fruit photosynthetic C gain plus imported C. The daily fruit photosynthetic contribution was calculated as the difference between net C0 2 exchange under light and dark conditions and multiplied by the 12 h photoperiod. The average daily fruit photosynthesis between times t, and t 2 was done by interpolation between the two measurement points and was calculated by: Mean daily fruit PS C = [daily fruit PS C (t1) + daily fruit PS C (t2) ]/2 [3] The average daily imported C was determined by subtracting the average daily fruit photosynthetic C from the average daily C cost.

PAGE 76

65 Sugar Determination Fruit samples were collected on the sampling dates previously indicated and stored at -80°C until analysis. Frozen samples were ground in liquid N 2 and extracted in boiling 80% ethanol (1:10 w/v) for 1 min. Mannitol ( 1 00 mg) was added as an internal standard. Extracts were centrifuged at 600g, the supernatant decanted, and the pellet was re-extracted twice. The combined supernatant was partitioned against chloroform, and the aqueous fraction was vacuum dried, resuspended in water, and passed through Dowex-1 and Dowex50 ion resins. The final supernatant was dried under vacuum, resuspended in HPLC water and filtered through a 0.45 um filter. Samples were injected into a Bio-Rad HPLC (Richmond, Calif) and sucrose, glucose, and fructose were determined using a Bio-Rad HPX 87C cation-exchange column at 85°C and flow rate of 0.6 ml water/min. Enzyme Extraction Frozen blueberry tissue (500 mg) was ground in liquid N 2 , transferred to a chilled mortar containing 5 ml of extraction buffer (50 Mm MOPS, pH 7.5; 5 Mm MgCI 2 , 2.5 DTT, 1 Mm EDTA, and 1 0% w/w PVPP), mixed thoroughly, and vacuum filtered. For SPS and SS assays, crude extract was desalted through a chilled 5-ml Sephadex column equilibrated with extraction buffer without EDTA, and centrifuged for 1 min at 200g. For soluble invertase assays, aliquots of crude extract were dialyzed (25,000 mol wt cutoff) for 24 h at 4°C

PAGE 77

66 against 5 Mm K 2 HP0 4 (pH 7.5). The insoluble pellet was rinsed several times with 200 Mm K 2 HP0 4 , vacuum filtered, and used to measure insoluble acid invertase activity. Enzyme Assays SPS and SS activities were determined according to the methods of Hubbard et al., (1989) with slight modifications. SPS activity was determined in a 70 / j \ reaction volume containing 50 Mm MOPS (pH 7.5), 14 Mm Mg Cl 2 , 10 Mm Fru6P, 50 Mm Glc6P, 7 Mm UDPGIc, and 45 /vl extract. Mixtures were incubated for 30 min at 30C, terminated with 70 /vl 30% KOH, followed by boiling for 1 5 min. Sucrose production was quantified by the anthrone method (Van Handel, 1968). SS was determined in the degradative direction. Two hundred microliters of desalted extract were assayed in a 500 /vl volume containing 200 Mm MES (pH5.5), 3 Mm NaF, 5 Mm UDP, and 50 Mm sucrose. Reactions were terminated after 30 min with 500 /vl 100 Mm TRIS (pH 8.7) + 0.2% EDTA followed by boiling. UDPG production was quantified by measuring the UDPG-dehydrogenase-specific synthesis of NADH (Lowell et al., 1989). Enzyme controls in both assays were treated as described, but terminated at 0 min. Boiled controls were avoided due to floculation. Soluble and insoluble acid invertase activities were assayed in a 500 /vl volume consisting of 2 Mm acetic acid (pH 4.5), 100 Mm sucrose, and 200 /vl extract (or 20-50 mg insoluble pellet). Reactions proceeded for 1 5 min at 30C and were terminated by adding 250 /vl of 400 mM K 2 P0 2 , pH 7.5 and boiling for

PAGE 78

67 2 min. The same reaction medium, but adjusted to pH 7.5 with 5 mM K 2 P0 4 , was used to assay neutral invertase activity. Enzyme controls contained all assay reagents, but were terminated at 0 min. Glucose production was determined by the glucose oxidase method (Sigma). Results Fruit Growth There was no difference in fruit dry weight accumulation between GA 3 treated and pollinated berries from 0 to 1 0 DAFB (Fig 4-1 A). From 25 DAFB to ripening, GA 3 -treated fruit accumulated significantly less dry matter than pollinated fruits. Pollinated fruits accumulated an average of 390 mg DW while GA 3 -treated fruits accumulated an average of 180 mg DW throughout development. Dry weight of nonpollinated fruits did not change throughout fruit development. Fruit fresh weight accumulation followed the same pattern as that of dry weight (Fig. 4-1 B). Although blueberry fruit growth is reported to follow a double sigmoid curve (Birkhold et al. , 1 992), the sampling times used in the present experiment were too few to clearly show this. Nonetheless, there seemed to be 3 periods of growth: period I lasted about 24 days in both GA 3 and pollinated treatments; period II in GA 3 -treated fruits extended from 24-72 DAFB, while in pollinated fruit, period II was significantly shorter, appearing to extend from 25 to 45 DAFB; period III was 12 days shorter in GA 3 -treated fruits (72-87 DAFB) compared to pollinated fruits (45-72 DAFB).

PAGE 79

68 Relative growth rate (RGR) of pollinated fruits was greater than that of GA 3 -treated fruits throughout much of development (Fig. 4-2). The maximum RGR for both treatments occurred at the end of period I. This was followed by a phase of sharply declining growth rate during period II in both treatments. A second peak just before ripening was observed for both GA 3 -treated and pollinated fruits. Fruit Carbon Dioxide Exchange Developmental patterns of fruit respiration were similar between GA 3 treated and pollinated fruits (Fig. 4-3). Respiration rates increased from 12 to 20 /ymol.gFW' 1 .h 1 between 0 and 10 DAB for GA 3 -treated fruits, then decreased to 5 /ymol.gFW \h 1 by 24 DAFB and remained constant thereafter. A similar increase in respiration rates was observed in pollinated fruits between 0 and 4 DAFB, declining to 1 2 /ymol.gFW Vh' 1 at 10 DAFB and then to 5 /ymol. gFW \h 1 at 45 DAFB, with little change from 45 to 72 DAFB. Respiration rates in nonpollinated fruits averaged between 12 and 14//mol.gFW \h 1 from 0-45 DAFB and were significantly greater than rates of GA 3 -treated and pollinated fruits at 24 and 45 DAB. Net photosynthesis occurred in fruit from all treatments throughout much of development (Fig 4-3). Fruit photosynthesis declined from 6 /ymol.gFW ^.h' 1 at 0 DAB to less than 1 //mol. gFW' 1 . h' 1 by 24 DAFB, for both GA 3 -treated and pollinated fruits. Net photosynthesis ceased during period III of fruit growth in

PAGE 80

69 both treatments. In nonpollinated fruits, photosynthetic rates were significantly higher than rates in GA 3 -treated and pollinated fruits from 4 to 45 DAB. Fruit Carbon Budget The developmental pattern of the average daily fruit C cost (DW C + respiratory C) (Fig. 4-4) was similar to the developmental pattern of fruit DW gain for both GA 3 -treated and pollinated fruits. The C cost for nonpollinated fruit was negligible and is not depicted in Fig. 4-4. The average daily C accumulation (DW C) in pollinated fruits was greater than in GA 3 -treated fruits, and this difference was evident as early as the 2-4 DAB interval. Maximum rates of C accumulation reached 3.6 mg C fruit' 1 . day' 1 in GA 3 -treated fruits at 72-87 DAFB, while in pollinated fruits, maximum rates reached 5.3 mg C fruit \day 1 at 45-72 DAFB. Overall, total DW C accumulation was 2.5-fold greater in pollinated fruits (180 mg C) than in GA 3 -treated fruits (73 mg C) (Table 4-1). Daily respiratory C loss (Resp C) was maximum during the last interval of fruit growth for both GA 3 -treated and pollinated fruits (Fig. 4-4). Absolute values were similar for both treatments throughout fruit development, increasing from 0.07 mg C fruit' 1 . day' 1 at 0-4 DAB to 1 to 2 mg C fruit' 1 . day 1 during the final stage of development. Total respiratory C loss was 35 mg C for GA 3 -treated fruits and 64 mg C for pollinated fruits (Table 4-1). This represented 33% of the total C cost for GA 3 -treated fruits compared to 26% for pollinated fruits. Analysis of the data for individual fruits indicated that these percentages were not significantly different from each other.

PAGE 81

70 The average daily fruit C cost (DW C + Resp C) for GA 3 -treated fruits increased from 0.06 mg C fruit 1 . day' 1 to 4.5 and 7.1 mg C fruit' 1 . day 1 for GA 3 -treated and pollinated fruits, respectively, as development proceeded (Fig. 4-4). The total C cost (DW C + Resp C) for GA 3 -treated fruit was 109 mg C fruit 1 while the C cost for pollinated fruit was 244 mg C fruit' 1 (Table 4-1). Daily fruit photosynthetic C supply reached its maximum between 24 and 45 DAFB in both GA 3 and pollinated fruits, averaging 0.21 mg C fruit' 1 . day' 1 in GA 3 -treated fruits and 0.32 mg C fruit 1 . day' 1 in pollinated fruits (Fig. 4-5). Imported C was maximum during the last interval of fruit growth, reaching 4.5 mg C fruit' 1 . day' 1 in GA 3 -treated fruits and 6.9 mg C fruit' 1 . day 1 in pollinated fruits. From 0 to 10 DAB, fruit photosynthesis supplied 51% of the fruit C requirement for GA 3 -treated fruits, and 36% of the C requirement for pollinated fruits (Table 4-1). Overall, fruit photosynthesis supplied 12% and 7% of the fruit C requirement for GA 3 -treated and pollinated fruits, respectively. Carbohydrate Accumulation Hexoses. Glucose and fructose are the main soluble carbohydrates accumulated in blueberry fruits, and are present in equimolar amounts throughout fruit development. Thus, data are presented as total hexoses (Fig. 4-6). There were no differences in hexose concentration among treatments from 0 to 45 DAFB, nor between GA 3 -treated and pollinated treatments from 45 DAFB to ripening, when comparing phenological stages. Hexose concentration declined from 30 to about 4 mg.gFW' 1 in all treatments from 0

PAGE 82

71 to 24 DAFB. Subsequently, there was a continuous increase in hexose concentration, up to about 100 mg.gFW' 1 in pollinated fruits by 72 DAFB (ripening). In GA 3 -treated fruits, hexose concentration remained constant from 24 to 72 DAFB, followed by a sharp increase to about 100 mg.gFW' 1 at 87 DAFB (ripening). Sucrose . Sucrose accumulation was negligible throughout most of the fruit development and accounted for only a small fraction of the total soluble carbohydrate accumulation (data not shown). GA 3 -treated fruits accumulated 8.6 mg sucrose. gFW 1 vs 4.4 mg.gFW 1 in pollinated fruits. Total carbohydrate content was correlated with both fruit FW and RGR. From 0-24 DAFB, carbohydrate accumulation was negatively correlated with dry weight gain (r = -.56, p<.01) and RGR (r = -.66, p<.01), while from 45 DAFB to ripening, carbohydrate accumulation was positively correlated only with DW gain (r = .69, p < =.01). Enzyme Activity SPS and SS activities . In general, SPS activity was low and not significantly different among treatments, when phenological stages are compared (Fig 4-7A). The highest enzyme activity was reached during the final stage of fruit development in GA 3 -treated (3.7 //mol. gFW' 1 h 1 ) and pollinated fruits (2.5 //mol. gFW' 1 h' 1 ). In nonpollinated fruits, maximum activity of 3.2 //mol.gFW 1 h 1 occurred at 1 0 DAFB. SPS activity was not correlated with fruit FW (r = .58, p<=.12), but was possitively correlated with non-structural

PAGE 83

72 carbohydrate accumulation (r = -.78, p<.05) during the final stage of fruit growth. There were no consistent differences in SS activity among treatments (Fig 4-7B). SS activity remained low throughout fruit development, increasing to 4.3 and 2.1 //mol.gFW' 1 h 1 in GA 3 -treated and pollinated fruits, respectively, during final fruit growth. In nonpollinated fruits, SS remained almost undetectable. Invertases. There were no significant differences in activities of soluble acid or insoluble acid invertase between GA 3 -treated and pollinated fruits, when phenological stages are compared (Fig. 4-8). Soluble acid invertase activity was low throughout much of development, increasing to 65 //mol.gFW' 1 h _1 in GA 3 -treated fruit and 54//mol.gFW' 1 .h' 1 in pollinated fruits during the final stage of development (Fig. 4-8A). Nonpollinated fruits exhibited significantly reduced soluble acid invertase activity (0.12 //mol.gFW' 1 h' 1 ) by 24 DAFB compared to GA 3 (1 .76 //mol.gFW 1 h 1 ) and pollinated treatments (2.39 //mol.gFW 1 h' 1 ). By 45 DAFB, no soluble acid invertase activity was detected in nonpollinated fruits. There were no significant differences in insoluble acid invertase activity among treatments from 0 to 45 DAFB, except for a 2-fold increase in activity in nonpollinated fruits at 10 DAFB (Fig. 4-8B). This increase was not seen in pollinated or GA 3 -treated fruits. Enzyme activity in GA 3 -treated and pollinated fruits declined from about 1 6//mol.gFW" 1 .h' 1 at 4 DAB to about 3 //mol.gFW 1 h' 1 at the beginning of the final stage of fruit growth. Acid insoluble invertase

PAGE 84

73 activity increased at ripening, averaging 9.5//mol.gFW 1 h" 1 in GA 3 -treated fruits and 5.1 //mol.gFW 1 h 1 in pollinated fruits. In nonpollinated fruits, the declining pattern of enzyme activity was less dramatic compared to the other treatments, and at 45 DAB, when most of the fruits had abscised, activity of insoluble acid invertase was 50% of the activity observed at 0 DAB and 2-fold higher than the rest of the treatments. There were no significant differences in neutral invertase activity among treatments (data not shown). Activity was low throughout most of fruit development (<1.0 /ymol.gFW 1 h' 1 )During the final stage of fruit growth, activity increased to about 5 to 6 /ymol.gFW' 1 . h' 1 in both GA 3 -treated and pollinated fruits. Discussion Fruit Growth and Carbon Budget Fruit growth in blueberry has been described as a double sigmoid pattern (Edwards et al., 1970; Hindle et al., 1947; Spiers 1952; Young 1952). However, the frequency of sampling in the present experiment does not allow clear determination of the three typical stages of growth, although 3 periods of growth can be described. The decreased final weight of GA 3 -treated fruits compared to pollinated fruits has been observed in most studies in which blueberry fruit set has been induced by GA 3 applications (Mainland and Eck, 1968; 1969a; Davies and Buchanan, 1979; Vanerwegen and Krewer, 1990.

PAGE 85

74 Decreased fruit weight in GA 3 -treated berries was due in part to a relatively slower growth rate during period II in GA 3 ( from 24 to 72 DAFB) vs pollinated fruits (from 24 to 62 DAFB). The decrease in fruit growth during this period has been associated with an increase in the growth rate of the embryo and endosperm (Edwards et al. , 1 970). However, factors other than that must account for the lack of fruit growth in GA 3 -treated fruit, since pollination and fertilization were prevented and thus, seeds were absent in GA 3 -treated berries. It is unknown if seed development causes a decrease in pericarp growth, or if the pericarp ceases growth in order to allow seed development. It seems, in this particular case, that the signal for cessation of pericarp growth arose from the pericarp itself, since seeds were not present. It appears also that seeds or the products synthesized in them are necessary to sustain a continuous pericarp growth since GA 3 -treated fruits had a prolonged phase of slow growth compared to pollinated fruits. It is possible that once the seeds and embryos start developing, they synthesize gibberellins that are then translocated to the pericarp (Sastry and Muir, 1963). Alternatively, seed development may stimulate gibberellin synthesis in the pericarp, as has been found in pea fruit (Ozga et al., 1992). Another possibility is that gibberellins stimulate the synthesis of auxins and the concerted action of both hormones may influence cell wall properties and/or increase carbohydrate accumulation and water flow into the fruit and, in this way, stimulate growth (Coombe, 1960).

PAGE 86

75 In general, blueberry fruit size has been positively correlated with seed number (Alders and Hall, 1 961 ; Tamada et al., 1 977; Darnell and Lyrene, 1 989; Lyrene, 1989). Seed number also dramatically affects the fruit development period in blueberry (Lang, 1992; Williamson et al., 1994). Fruit with more seeds have a shorter fruit development period than unseeded fruits, and this was corroborated in this study. Fruit C0 2 exchange values were in general agreement with those obtained by Birkhold et al., (1991) in smalland large fruited-genotypes of rabbiteye blueberry. The transient increase in fruit respiration at 4 DAB and 10 DAB for pollinated and GA 3 -treated berries, respectively, may have been due to increased rates of cell division in these two treatments, since such an increase was not observed in nonpollinated fruits. During the early stages of development, fruits act as utilization sinks and respiration is their main metabolic activity (Ho, 1988). Frequently, the rate of import to these sinks parallels the respiration rate (Kallarackal and Milburn, 1985). In our case, the highest respiration rates occurred during the early stage of fruit growth I, when RGR is increasing; however, the peak of maximum respiration preceded that of maximum RGR. In peach, RGR has been correlated with fruit growth patterns, respiration, and source-sink relationships (DeJong and Goudriaan, 1 989; Pavel and DeJong, 1993). In our case, RGR was not linearly correlated with respiration and the plotting of log RGR against time revealed 3 stages of

PAGE 87

76 growth (data not shown) instead of two as reported in peach (DeJong and Goudriaan, 1 989). Respiration was relatively constant during the second period of fruit growth (24 to 45 DAFB and 24 to 72 DAFB for pollinated and GA 3 -treated fruits, respectively) when RGR was declining to its lowest point. By this stage, most of the cell structures have been laid down, and presumably, mostly maintenance respiration is occurring. This prolonged growth period in GA 3 treated fruits resulted in a greater C loss than in pollinated fruits (56% and 30%, respectively) (Table 4-1). Although statistical analysis indicated these percentages were not significantly different, biologically the greater respiratory C loss in GA 3 -treated fruits may have contributed to the smaller fruit size compared to pollinated fruits. The last increase in respiration at later stages of development may be associated with ripening and represented about the same percentage of C loss for both treatments. As in other rabbiteye blueberry cultivars (Birkhold et al., 1992), net photosynthesis was observed in 'Beckyblue' fruit throughout most of the developmental period and contributed 12% and 7% of the total carbon supply in GA 3 -treated and pollinated fruits, respectively. These values are in the range observed by Birkhold et al. (1992) in rabbiteye blueberry as well as by Pavel and DeJong (1993) in peach fruits. The higher photosynthetic contribution by GA 3 -treated fruits may be due to the longer period that these fruits remained

PAGE 88

77 green compared to pollinated fruits, as well as the higher surface area/volume ratio due to its smaller size. Fruits from pollinated treatments began importing carbon at higher rates than GA 3 -treated fruits very early in development (2-4 DAB). Pollinated fruit sustained this higher rate of C import throughout development, importing ca. 3-fold more assimilates than GA 3 -treated fruits during the final stages of growth (45 to 72 DAB). These differences cannot be attributed to differences in source supply, since foliation and leaf canopy development occurred at the same time and to the same extent in both treatments. Additionally, fruit loads were similar for both treatments. This suggests that C import into GA 3 -treated fruits is sink limited. Carbohydrate Accumulation and Sucrose Enzyme Activities The ability of the sink to receive or attract assimilates is genetically determined, but can be regulated by the metabolic activity of the sink organ during development. The activities of sucrose-metabolizing enzymes in the unloading area are one component of metabolic activity that can influence the extent of sucrose translocation to fruits. In the present study, however, the increased rate and amount of imported C into pollinated vs GA 3 -treated fruits did not appear to be related to differences in activities of sucrose-metabolizing enzymes and/or subsequent carbohydrate accumulation. In both pollinated and GA 3 -treated fruits, SPS and SS activities were similar. Activities of both enzymes were low throughout development and

PAGE 89

78 increased slightly during the final stage of growth. Similar results were reported previously by Darnell et al. (1994) for blueberry fruit. The increase in SPS activity was coincident with the increase in sucrose concentration in maturing fruits. SPS activity is associated with tissues that accumulate sucrose and has been reported in sugar beet roots (Fiew and Willenbrike, 1987), sweet melon (McCollum et al., 1 988; Hubbard et al., 1 989) and some tomato species (Miron and Schaffer, 1991). Hubbard et al. (1989) reported that SPS was the dominant enzyme associated with sucrose accumulation in muskmelon. Although the increase in SS was also coincident with the increase in fruit sucrose concentration, most research suggests that SS acts primarily in the degradative direction (cleaving sucrose to fructose and UDP-glucose) in vivo (Xu et al., 1989; Sun et al., 1992; Wang et al., 1993). SS activity has been strongly correlated with sink strength in potatoes (Sung et al., 1 989; Xu et al., 1 989), wild tomato (Sun et al., 1 992), and cultivated tomato (Stommel, 1 992; Wang et al., 1993). In these fruits, sucrose is broken down for starch synthesis during the early stages of development. SS has an advantage over invertase activity for this process, since SS results in UDP-glucose production, a precursor for starch biosynthesis. In other fruits, however, particularly hexose accumulators, SS activity does not appear to be related to growth (Nielsen et al., 1991). The increased SS activity observed in maturing blueberry fruits might be involved in the hydrolysis of sucrose since at this stage extensive hexose accumulation is taking place. However, this would result in futile

PAGE 90

79 cycling of sucrose between SS and SPS. Nevertheless, the roles of SS and SPS in the regulation of carbohydrate metabolism in blueberry fruits appear to be minor. The high activities of both soluble and insoluble acid invertases during blueberry fruit development and the strong correlation between invertase activity, hexose accumulation, and fruit growth, indicate that these are the major enzymes involved in sucrose metabolism in GA 3 -treated and pollinated fruits. This agrees with previous work in which invertase has been reported as the dominant enzyme throughout development of hexose accumulating fruits (Manning and Maw, 1975; Leigh et al., 1979; Nielsen et al., 1991). Both insoluble (presumably cell wall associated) and soluble acid invertase may be important in decreasing the sucrose concentration at the unloading site. This, in turn, would enhance sucrose translocation to that site by maintaining a sucrose concentration gradient from source to sink. The presence of insoluble acid invertase, which hydrolyzes sucrose before it is taken up by the cells, ensures continuous sucrose unloading by maintaining low turgor in sink phloem and increased osmotic potential in the appoplast during the early stages of blueberry fruit development when readily available energy is needed to sustain rapid growth and respiration. Soluble acid invertase, which hydrolyzes sucrose once it is stored in vacuoles, is the enzyme most closely associated with hexose accumulation in blueberry fruits. Insoluble invertase activity and fruit hexose concentrations were

PAGE 91

80 relatively high at anthesis. Thus, it seems unlikely that differences in fruit set/early fruit development among treatments were limited by carbohydrate availability. Similar results were reported in strawberry by Darnell and Martin (1988) and in orange by Ruiz and Guardiola (1994). This high initial carbohydrate content may explain the high initial fruit set in blueberry, as observed by the "swelling" of the receptacle that occurred in all treatments. It seems that this event is "programmed" at least under the conditions of this experiment, since nonpollinated fruits did not receive any stimulus compared to pollinated and GA 3 -treated fruits. Similarly, Martin et al. (1982) suggested that all the chemical factors required for fruit set of pear are present at anthesis, but that subsequent development of the fruit requires an extra stimulus. Subsequent fruit development in the present study does not appear to be limited by invertase activity and/or carbohydrate availability in either GA 3 treated or pollinated fruits. GA 3 -induced fruit development was accompanied by similar increases in soluble acid invertase activity and fruit hexose concentration as was pollinated fruit development. Thus, the smaller size of GA 3 -treated fruit does not appear to be the result of a reduced capacity for sucrose metabolism. On the other hand, calculation of invertase activity and hexose accumulation on a per cell basis (refer to Chapter 5) indicates that cells in pollinated fruits exhibited a 1 .8-fold increase in soluble invertase activity and a 2.2-fold increase in hexose accumulation compared to cells in GA 3 -treated fruits. Since cell number was similar between pollinated and GA 3 -fruits, this

PAGE 92

81 increase reflects the increase in cell size of pollinated fruits. Thus, the larger cells in pollinated fruits apparently had increased protein (i.e., invertase) capacity compared to the smaller cells in GA 3 -treated fruits. The role of invertases in nonpollinated fruits is less clear. The decrease in soluble invertase at 24 DAFB, as well as the decrease in insoluble invertase at 2 and 24 DAFB, may indicate that those fruits will be senescent, as has been reported by Estruch and Beltran (1991) in nonpollinated pea ovaries. In summary, fruit size differences between GA 3 -treated and pollinated fruits were due to limited dry matter accumulation during the mid to late stages of development in GA 3 -treated fruits compared to pollinated fruits. This was reflected in a different carbon requirement for large (pollinated) and small (GAtreated) fruits. GA 3 -treated fruits appeared to be sink-limited and imported less carbonthat pollinated fruits. However, the apparent sink-limitation and resultant decrease in fruit weight of GA 3 -treated compared to pollinated fruits were not associated with any differences in the capacity of sucrose-metabolizing enzyme activities in fruit extracts and/or subsequent soluble sugar accumulation as determined on a fruit weight basis.

PAGE 93

Fresh weight (g) Dry weight (g) 82 Figure 4-1. Changes in dry weight (A) and fresh weight (B) in GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means ± SE, n = 3; SE bars present only when larger than symbol. Arrow indicates when abscission of nonpollinated fruits occurred.

PAGE 94

RGR (gDW (gDWj ’day' 1 ) 83 Figure 4-2. Relative growth rate (RGR) of GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Means ±SE, n = 3; SE bars present only when larger than symbol.

PAGE 95

15 10 5 0 -5 10 15 20 15 10 5 0 -5 10 15 20 15 10 5 ! 0 -5 10 1 15 84 S 20 30 40 50 60 Days after bloom NP 70 80 90 4-3. Net carbon dioxide exchange in GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means ±SE, n = 3; SE bars present only when larger than symbol.

PAGE 96

85 8 Days after bloom Figure 4-4. Daily estimated C cost in GA 3 -treated (GA 3 ) and pollinated (POLL) 'Beckyblue' rabbiteye blueberry fruits. Values were calculated from data in Figures. 4-1 and 4-3 and are based on a 1 2 h photoperiod and 25/21 °C light/dark temperatures.

PAGE 97

Average daily fruit C supply (mg fruited ) 86 Figure 4-5. Daily estimated C supply in GA 3 -treated (GA 3 ) and pollinated (POLL) 'Beckyblue' rabbiteye blueberry fruits. Values were calculated from data in Figures 4-1 and 4-3 and are based on a 1 2 h photoperiod and 25/21 °C light/dark temperatures.

PAGE 98

87 Figure 4-6. Hexose accumulation in GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means ±SE, n = 3; SE bars present only when larger than symbol.

PAGE 99

SPS and SS activities (umol gFW V 1 ) 88 Figure 4-7. Activities of (A) sucrose phosphate synthase (SPS) and (B) sucrose synthase (SS) in GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means ± SE, n = 3; SE bars present only when larger than symbol.

PAGE 100

Acid invertase activity (umol gFwV 1 ) 89 80 ga 3 POLL NP •*Q* * (• ) • 0 10 20 30 40 50 60 70 80 90 Days after bloom Figure 4-8. Activities of (A) soluble acid and (B) insoluble acid invertases in GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) 'Beckyblue'rabbiteye blueberry fruits. Values are means ±SE, n = 3; SE bars present only when larger than symbol.

PAGE 101

90 Table 4-1. Estimated C budget for developing GA 3 -treated, pollinated, and nonpollinated 'Beckyblue' rabbiteye blueberry fruits. Treatment DAB Carbon Cost Total Carbon Supply DW C Resp C Fruit Ps C Importod C ga 3 0-2 o.ow 0.12(100) 0.12 0.09(75) 0.03(25) 2-4 0.0(0) 0.14(100) 0.14 0.09(64) 0.05(36) 4-10 0.31(30) 0.71(70) 1.02 0.47(46) 0.54(54) 10-24 5.43(66) 2.84(34) 8.27 1.73(46) 6.54(79) 24-45 5.92(43) 7.91(57) 13.83 4.36(32) 9.47(68) 45-72 7.46(44) 9.49(56) 16.95 5.39(32) 11.56(68) 72-87 54.28(79) 14.13(21) 68.41 1.09(2) 67.32(98) Total 73.40(67) 35.34(33) 108.7 13.23(12) 95.51(88) POLL 0-2 0.0(0) 0.12(100) 0.12 0.09(75) 0.03(25) 2-4 0.26(63) 0.15(37) 0.41 0.1 1(27) 0.30(73) 4-10 0.66(51) 0.66(49) 1.32 0.46(34) 0.86(66) 10-24 8.31(66) 4.80(37) 13.11 2.83(22) 10.28(78) 24-45 26.04(70) 11.33(30) 37.37 6.65(18) 30.72(82) 45-72 144.93(76) 46.60(24) 191.5 5.85(3) 185.6(97) Total 180.20(74) 63.66(26) 243.8 15.99(7) 227.8(93) NP 0-2 0.0(0) 0.14(100) 0.14 0.10(71) 0.04(29) 2-4 0.0(0) 0.19(100) 0.19 0.13(68) 0.06(32) 4-10 0.0(0) 0.58(100) 0.58 0.43(74) 0.15(26) 10-24 0.26(19) 1.11(81) 1.37 0.89(65) 0.48(35) 24-45 0.17(7) 2.17(93) 2.34 1.70(73) 0.64(27) Total 0.43(9) 4.19(91) 4.62 3.25(70) 1.37(30) 2 Values represent milligrams C per fruit required for each stage of development. Budget assumes 25/21 °C for day/night and a 12 h photoperiod. POLL = pollinated; NP = nonpollinated. v Values in parenthesis represent the percentage contribution to the C cost or C supply for each stage of development.

PAGE 102

CHAPTER 5 CELL NUMBER AND CELL SIZE IN GIBBERELLIC ACID-TREATED VS POLLINATED FRUITS OF BLUEBERRY Introduction In blueberry, parthenocarpic fruit set and development is induced by exogenous applications of gibberellic acid (GA 3 ); however, final fruit size of parthenocarpic fruits is often smaller than that of pollinated fruits (Davies and Buchanan, 1 972; Vanerwegen and Krewer, 1 990). The small fruit size resulting from GA 3 applications has also been observed in other fruit crops, including grape (Iwahori et al. , 1 968), cranberry (Devlin and Demoranville, 1 968), peach (Stembridge and Gambrell, 1 972), and citrus (Garcia Martinez and Garcia-Papi, 1979). Our previous results indicate that the decrease in C import into GA 3 treated fruits compared to pollinated fruits was apparently not related to differences in sucrose-metabolizing enzyme activities and/or subsequent carbohydrate mobilization to the fruit. Cell number at anthesis, the length of the cell division period after anthesis, and the extent of cell enlargement determine final fruit size in a number of fruits (Coombe, 1976). Plant growth regulators may induce parthenocarpic fruit set and development by directly 91

PAGE 103

92 affecting cell division and/or cell enlargement. In tomato fruit, exogenous applications of 10% IAA or 0.1% 4-chlorophenoxyacetic acid (4-CPA) to emasculated flowers transiently increased the rate of cell division, while 0.6% GA 3 decreased cell division rate throughout fruit development (Bunger-Kibler and Bangert, 1983). Although cell size was increased by GA 3 compared to 4CPA, final fruit size of GA 3 -treated fruits was reduced by 55% compared to both pollinated and 4-CPA-treated fruits. In isolated Zinnia mesophyll cells, GA 3 inhibited cell division and delayed the initiation of DNA synthesis only when cells were in resting state and not when cells were dividing continuously (Iwasaki et ai . , 1986). In pea ovaries, GA 3 applied at anthesis induced mesocarp cell enlargement and division (Vercher et al., 1984) and enhanced synthesis of primary and secondary cell walls of endocarp cells (Vercher et al., 1987). A decreased weight and decreased cell size were observed in GA 3 treated ovaries compared to pollinated ovaries at 4 DAB; however, conclusions about final fruit size and/or cell size cannot be drawn since the experiment was terminated at 4 DAB. Nishitani and Masuda (1982) found that synthesis of polyuronides, xyloglucans and cellulose were regulated by auxins in Vigna anguiaris, whereas synthesis of xylans and cellulose during cell maturation were regulated by GA 3 . In flank meristems of Hedera, GA 3 stimulated the frequency of tangential divisions as well as the expansion in the radial-longitudinal direction (Marc and Hackett, 1992). Thus, the effect of growth regulators on

PAGE 104

93 cell division and enlargement depends upon species, stage of development and tissue type within a fruit. The objectives of this study were to assess the contribution of cell number and cell size to the final fruit size in blueberry and to determine the basis for differences in fruit size between pollinated and GA 3 -induced parthenocarpic fruits. Materials and Methods Softwood cuttings of 'Beckyblue' rabbiteye blueberries were rooted in Spring 1990 and grown outdoors in 22 L pots in a 1:1 peat:pine bark mix. Plants were watered every other day and fertilized with 20N-5.6P-1 1 K water soluble fertilizer. In December 1991, nine uniform plants were transferred to a dark cooler and chilled at 7±1°C for 30 days. After chilling, plants were placed in a greenhouse with average day/night temperatures of 27/1 6°C to force budbreak. Flower clusters were thinned to four to five florets/cluster at bloom, removing the most developed and least developed florets. The total number of florets ranged from 1 98 to 21 6 florets/plant. Treatments consisted of hand pollination at bloom with 'Climax' pollen (a cross-compatible cultivar), application of 0.7 mM GA 3 (Pro Gibb 4%, Abott Laboratories, Chicago, ILL, prepared in Mclllvaine buffer pH 3.5, and 0.01 % Tween-80), sprayed at bloom and again 7 days after bloom (DAB). The nonpollinated treatment consisted of a spray application of buffer and surfactant sprayed at bloom. Sampling dates were 0, 3, 10, and 24 DAB for all treatments. Subsequent to the abscission

PAGE 105

94 of the nonpollinated fruits (24 DAB), data were taken on the remaining two treatments at 45 DAB and at ripening. Three fruits per each of three replications were harvested and fixed separately in formalin-acetic acid-alcohol mixture (FAA). Fresh weight, longitude, and diameter were taken before fixation. Fruit median cross sections were dehydrated as follows: Solution Time 50% ethyl alcohol (ETOH) 2 hours 50% tertiary butyl alcohol (TBA) 2 hours 70% TBA 1 0 hours 85% TBA 2 hours 95% TBA 2 hours 1 00% TBA 2 hours Absolute TBA 1 hour Absolute TBA overnight Absolute TBA 2 hours 50/50 absolute TBA/paraffin oil (light mineral oil) 2 hours 75/25 melted paraffin (Paraplast)/ paraffin oil 2 hours after Melted paraffin (three changes) 2 hours each change

PAGE 106

95 Sections were embedded in paraplast, sectioned at 8 to 20 /ym, attached to slides with albumen fixative and stained with safranin and fast green. The staining procedure was as follows: Solution Time Xylene (two changes) 3 min each Absolute ETOH 3 min 95% ETOH (two changes) 3 min 70% ETOH 3 min 1 % safranin 2 h, rinsing with running distilled water 70% ETOH 1 min 0.5% picric acid-95% ETOH 1 0 sec 95% ETOH-ammonia (4 drops) 1 min Absolute ETOH (two changes) 1 min each 0.5% fast green FCF Clear with a mixture of 50/25/25 clove oil/absolute 10-30 sec, rinse with used clove oil ETOH/xylene 3 min Xylene (three changes) 3 min each After staining the sections were mounted in Permount. Cell number and cell size were obtained from microphotographs. A Lasico planimeter (Model #L1 0, series 81 191, Los Angeles Scientific Instrument Co. Inc. Los Angeles, CA) was used to measure total area of the amplified

PAGE 107

96 photograph as well as cell area of 1 0 cells considered representative of each of the following regions: epidermis, hypodermis, outer mesocarp, middle mesocarp, inner mesocarp and endocarp. Vascular bundle areas as well as locular area were subtracted to obtain total mesocarp area. Values were converted to //m 2 using a reference area of 1 mm 2 from a hemocytometer chamber that was microphotographed each time new film was used. Cell number in a given cross sectional area was determined by following the equations used by Scorza et al. (1991). Results Fruit Growth Growth of GA 3 -treated fruits followed a similar growth pattern as pollinated fruits; however, fruit weight was significantly lower in GA 3 -treated fruits compared to pollinated fruits from 45 DAB to ripening (Fig. 5-1). Nonpollinated fruits abscised between 24 and 45 DAB and weighed significantly less than both GA 3 and pollinated fruits. A delay in ripening was also observed in GA 3 -treated fruits compared to pollinated fruits. The fruit development period averaged 72 days for pollinated fruits and 87 days for GA 3 -treated fruits. Cell Number and Cell Size The histological composition of a blueberry ovary at 0 DAB in cross section is depicted in Fig. 5-2. Measurements were taken from the interlocular

PAGE 108

97 tissue and the following regions were observed: epicarp comprised of epidermis (ep) and hypodermis (hp); mesocarp, divided arbitrarily into outer mesocarp (om; region between epicarp and the first large vascular bundle (vb), middle mesocarp (mm; region between the two large vascular bundles), and inner mesocarp (im; the region between the second large vascular bundle and the endocarp); and endocarp (e), a single layer of cells lining the locules (Ic). Epicarp . Cell number in the epicarp at anthesis averaged 1250 cells/cross sectional area (x.s.) (Fig. 5-3). Cell number increased gradually for all treatments, and at ripening, epicarp cell number was the same for both GA 3 treated and pollinated fruits, averaging 1804 cells/x.s. and 2446 cells/x.s., respectively. Epicarp cell number increased 1.6-fold in GA 3 -treated fruits and 2.1 -fold in pollinated fruits throughout development. There was a 1.7-fold increase in epicarp cell number in nonpollinated fruits prior to abscission. Cell size in epidermal tissue did not change significantly from 0 to 10 DAB in any treatment (Fig. 5-4). However, at 24 DAB, epidermal cells of pollinated and nonpollinated fruits were significantly smaller than GA 3 -treated fruits (an average of 470 /ym 2 vs 600 /vm 2 ). There were no significant changes in cell size from 24 to 45 DAB. Epidermal cell size in both pollinated and GA 3 treated fruits were similar at ripening, averaging 1175 //m 2 . Total cell size increased about 3-fold from 0 DAB to ripening in both treatments. Increases in hypodermal cell size followed a similar pattern (Fig. 5-4). Differences among treatments were not manifested until 24 DAB, when cell size

PAGE 109

98 in GA 3 -treated fruits was about 1.6-fold greater than in pollinated and nonpollinated fruits. However, there were no differences in cell size between GA 3 -treated (1860 /vm 2 ) and pollinated fruits (2138 //m 2 ) at ripening. The overall increase in hypodermal cell size was about 3.5x from 0 DAB to ripening. Both epidermal and hypodermal cells were rectangular-shaped, but hypodermal cells were larger than epidermal cells. At 0 DAB only one layer of each tissue was easily distinguished (Fig. 5-2), but as development progressed, two or three cell tiers of hypodermal cells were observed, based on shape and pigment accumulation (Figs. 5-5 to 5-9). Cells in the epidermis appeared to divide anticlinally while hypodermal cells appeared to divide periclinally. Cells of both tissues elongated tangentially during development. Pigment accumulation was observed in hypodermis cells at 24 DAB, and a massive accumulation of pigment was observed at ripening in both epidermis and hypodermis (Figs. 5-7 to 5-9). Endocarp. There were no consistent differences in endocarp cell number among treatments or sampling dates (Fig. 5-3). The measured decrease in endocarp cell number in all treatments was probably due to loss of some cells upon removal of seed and/or lignified placental tissue. Endocarp cell size increased about 6.5-fold from 0 to ripening in both GA 3 -treated and pollinated fruits, with no difference between treatments (Fig 5-4). At ripening, cell size averaged 1 1 28 /ym 2 for both treatments. Cell size

PAGE 110

99 in nonpollinated fruits was significantly less than that of GA 3 -treated and pollinated fruits at 24 DAB. Mesocarp . There were no consistent differences in total mesocarp cell number between pollinated and GA 3 -treated fruits within a given sampling date or phenological stage (Fig. 5-3). Cell number in mesocarp tissues of both increased throughout fruit development. At anthesis, mesocarp cell number averaged 6900 cells/x.s. increasing to 8950 and 8840 cells/x.s in GA 3 -treated and pollinated fruits, respectively. Cell number in nonpollinated fruits did not increase between anthesis and fruit abscission. There were no differences in cell size in the outer mesocarp of GA 3 treated and pollinated fruits throughout fruit development (Fig 5-10). Cell size increased from an average of 898 //m 2 at 0 DAB to an average of 1 6266 /ym 2 at ripening. At 24 DAB, cell size in nonpollinated fruits was significantly smaller than the other treatments (4035 /ym 2 vs 5400 /ym 2 ). Cell size in both middle and inner mesocarp of pollinated fruits was significantly larger than in GA 3 treated fruits at 24 DAB and ripening (Fig. 5-10). Nonpollinated fruits had significantly smaller cells than both GA 3 -treated and pollinated fruits at 24 DAB. Overall, from anthesis to ripening, cell size in middle and inner mesocarp tissues increased 25-fold in GA 3 -treated and 33-fold in pollinated fruits, respectively. Average cell size increases for the fruit development period are depicted in Table 5-1 for individual pericarp tissues.

PAGE 111

100 The mean rate of cell enlargement was calculated for individual tissues in each treatment (Fig. 5-1 1). Cell enlargement rate was higher in epidermal and hypodermal tissues of GA 3 -treated fruits than in both pollinated and nonpollinated fruits at 24 DAB. However, from 45 DAB to ripening, cell enlargement rate in pollinated fruits exceeded that of GA 3 -treated fruits. Cell enlargement rate in endocarp cells was similar for GA 3 -treated and pollinated fruits, while enlargement rates in nonpollinated fruits slowed compared to the other treatments by 24 DAB. Outer mesocarp cells of GA 3 -treated and pollinated fruits had similar rates of cell enlargement from 0 to 45 DAB, while in middle and inner mesocarp regions, cells of GA 3 -treated fruits exhibited slower growth rates at 24 DAB than cells of pollinated fruits (Fig. 5-12). At 45 DAB, cell enlargement rate was the same for both GA 3 -treated and pollinated fruits in middle mesocarp, while inner mesocarp cells of GA 3 -treated fruits had a faster rate of enlargement than pollinated fruits. From 45 DAB to ripening, cells from all mesocarp regions of pollinated fruits enlarged at a faster rate than GA 3 -treated fruits. Mesocarp cells comprised about 78% of the total pericarp tissue in blueberry, while epicarp accounted for 14% and endocarp for only 8%. Increases in mesocarp cell size followed the same pattern as increases in fruit fresh weight (Fig. 5-13).

PAGE 112

101 Discussion The decrease in weight of GA 3 -induced parthenocarpic blueberry fruits is similar to results observed in cranberry fruits upon spray applications of GA 3 (Devlin and Demoranville, 1 967) or in natural parthenocarpic fruits like seedless tomato (Mapelli et al . , 1970) and seedless 'Tokay' grapes (Weaver and Pool, 1968), but they differed from those observed in seedless 'Delaware' grapes (Clore, 1965), peach and apricot (Jackson, 1968). In these latter fruits, final fruit weight was not negatively affected by GA 3 treatments, and fruit fresh weight was comparable to pollinated fruits. The fruit cell division period in epicarp and mesocarp tissues of pollinated and GA 3 -treated fruits appears to occur during the first 24 DAB. Cell division also occurred in the epicarp of nonpollinated fruits, but no division was apparent in the mesocarp tissue. These results are in agreement with those reported for other pollinated fruits such as peach (Jackson, 1 968; Scorza et al., 1992), tomato (Mapelli et al., 1978), grapes (Harris et al., 1969), apricot (Jackson and Coombe, 1966), and sour cherry (Tukey and Young, 1939), where cell division occurs during Stage I of development. Edwards (1970) reported increases of 2 to 3-fold in mesocarp cell number at the end of Stage I in some rabbiteye blueberry clones. However, only a 1.2-fold increase in mesocarp cell number was observed in our case. These differences may be due to the different counting methods. Edwards indicated that mesocarp cell number was determined as the average distance from the endocarp to the

PAGE 113

102 epicarp. It was unclear; however, whether intralocular shape was taken into account in his cell counts as it was in ours. Cell counts may also be affected by fruit shape. In some cases, elongated fruits have resulted from GA 3 applications (as cited by Schwabe and Mills, 1981; Weaver and McCune, 1 960) and since cell number in the present study was measured in a median cross section area, changes in the fruit length:diameter (L/D) ratio would affect the estimation of cell number among treatments. The L/D ratio was similar in fruits from all treatments, averaging about =0.9 at ripening; thus we assume cell counts were not influenced by L/D ratios. Significant increases in cell size were not observed in any tissues of any treatment until 10 DAB. The increased cell enlargement rate and concomitant increase in cell size in epidermis and hypodermis of GA 3 -treated fruits compared to pollinated and nonpollinated fruits from 10 to 24 DAB appears to be due to a very localized effect of GA 3 in these tissues. No such effect was observed in mesocarp tissue. Epidermal and hypodermal cell enlargement rates of pollinated and nonpollinated fruits were slower during this period. This appears to be a programmed' event, unrelated to pollination and/or fertilization, since pollinated and nonpollinated treatments showed similar cell enlargement rates. This situation was not the case for mesocarp cells, where cell enlargement rates and cell size between 10 and 24 DAB were significantly reduced in nonpollinated fruits compared to pollinated and GA 3 -treated fruits. Thus mesocarp cell enlargement is dependent on the stimulus provided by

PAGE 114

103 pollination or GA 3 applications. This stimulus was only partially provided by exogenous GA 3 application, however; since pollinated fruits exhibited a marked increase in mesocarp cell enlargement rates and cell size compared to GA 3 treated fruits between 45 DAB and ripening. In contrast, Bunger-Kibler and Bangert (1983) observed increased cell size in GA 3 -treated tomato fruits compared to pollinated fruits, even though final fruit size of GA 3 -treated fruits was smaller. They attributed the increased cell size to a higher level of endoploidy in GA 3 -treated fruits. No such effect was observed in the present study. The decreased enlargement rate in GA 3 -treated fruits from 45 DAB to ripening observed in the present study may indicate that the exogenous GA 3 supply was exhausted. However, previous work (Chapter 3) indicated that additional applications of exogenous GA 3 at 21 and 42 DAB were unsuccessful in stimulating additional pericarp growth compared to the applications used in the present study. Cell enlargement has been described in physical terms as a function of cell wall extensibility, osmotic potential of the cell, and/or turgor pressure (Lockhart, 1965; Wareing and Phillips, 1981). Gibberellins are believed to be involved in both cell wall extensibility and osmotic potential. Gibberellins have induced enlargement by increasing cell wall extensibility in Avena stem segments (Adams et al., 1975), lettuce hypocotyls (Stuart and Jones, 1977), Phaseolus vulgaris leaves (Brock and Cleland, 1 990), and wheat leaves (Keyes et al., 1 990). In cucumber hypocotyls; however, GA 3 induced enlargement by

PAGE 115

104 decreasing the osmotic potential rather than increasing extensibility (Katsumi and Kazama, 1 978) In our study, GA 3 -treated and pollinated fruits began accumulating considerable amounts of sugars at the same time (24 DAB) and in similar concentrations (Chapter 4), but the decrease in osmotic potential was not enough to elicit cell enlargement in GA 3 -treated fruits to the same extent as in pollinated fruits. This suggests that decreased osmotic potential in pollinated compared to GA 3 -treated fruits was not the basis for differences in cell enlargement. Decreased wall extensibility may be responsible for the decreased mesocarp cell enlargement rate and resultant decrease in cell size in GA 3 treated fruits. The contribution of cell size and cell enlargement to final fruit size is not easy to elucidate. In apple, fruit size appears to be determined primarily by the degree of cell division after pollination (Pearson and Robertson, 1952; Smith, 1940). Recently, Scorzaetal. (1991) reported that differences between smalland large-fruited peach genotypes were set at anthesis, with large fruits having more cells than small ones. These differences were enhanced by a prolonged Stage I, a period of active cell division, in large-fruited types compared to smallfruited types. However, Bain and Robertson (1951) argued that cell enlargement is continuous throughout apple fruit development and that in small fruits, the differences in fruit size are due to both cell size and cell enlargement. Similarly, final fruit size in Cucumis is due to both cell number and cell size

PAGE 116

105 (Sinnot, 1939). Larger fruits had a longer cell division period, a larger cell size at which cell division ceased, and increased cell enlargement compared to small fruits. In apricot, differences in fruit size within trees were due to differences in cell number, while differences between trees of the same cultivar were determined primarily by differences in cell enlargement (Jackson and Coombe, 1 966). Harris et al. (1965) reported that final fruit size in grape was determined by cell size rather than cell number. In our study, differences in fruit size among pollinated, GA 3 -treated, and nonpollinated fruits were due primarily to differences in cell size with cell number playing a minor role. This is supported by a high correlation between cell size and fruit fresh weight (r = .93, p<.01), while the correlation between cell number and fresh weight was insignificant (r = 0.38, p<.16). Additionally, the duration of the cell division period did not appear to be affected by treatment. At ripening, GA 3 -treated and pollinated fruits had similar mesocarp cell number, but both mesocarp cell size and fresh weight of GA 3 -treated fruits were reduced by about 67% compared to pollinated fruits. From our results, it appears that cell division was stimulated to the same extent by both GA 3 and pollination, and that nonpollinated fruits exhibited only a slight decrease in cell division compared to the other two treatments. Differences in final fruit size between GA 3 -treated and pollinated fruits were due to differences in cell enlargement rather than cell number. Cell enlargement was continuous throughout fruit development and was lower for GA 3 -treated

PAGE 117

106 fruits compared to pollinated fruits. The lower cell enlargement in GA 3 -treated fruits may have been due to a decreased cell wall extensibility.

PAGE 118

Fresh weight (g) 107 Figure 5-1. Developmental changes in fresh weight of GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) 'Beckyblue' rabbiteye blueberry fruits. Values are means ±SE, n = 3; SE bars present only when larger than symbol. Arrows indicate treatment application times.

PAGE 119

Figure 5-2. Median cross section of a 'Beckyblue' rabbiteye blueberry fruit at 0 DAB. Bar = 150 /jm. ep = epidermis; hp = hypodermis; om = outer mesocarp; mm = middle mesocarp; im = inner mesocarp; en=endocarp; vb = vascular bundles; Ic = locule; pl = placental tissue; ov = ovule.

PAGE 120

109

PAGE 121

Cell number/x.s. (xIO 3 ) 110 15 10 Endocarp Mesocarp 0 Epicarp 0 15 10 0 15 B c BC ®8888 aaaAaa. cb cb ba a a b B B ^888888 cb c c a a b GA; POLL A ba a a A NP 10 * BC 5 cb B cb BC cb cb 10 24 Days after bloom Figure 5-3. Cell number changes per median cross sectional area (x.s.) in developing pericarp of GA 3 -treated (GA 3 ), pollinated (POLL), and non pollinated (NP) blueberry fruits. Mean separation across treatment and time by LSMean's, p = 0.05.

PAGE 122

Cell size (um 2 ) x 10 111 Figure 5-4. Increases in cell size in developing (A, B) epicarp and (C) endocarp tissues of GA 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) blueberry fruits. Significance level at 0.05, n = 9.

PAGE 123

0 0 > 0 •f-t 73 73 JO E _Q 0 11 E '0 E 3 _Q ii > LJ*C 0 O O 0 O DO C / ) 0 73 E 0 k_ 4-> 0 0 4-* c D ' O 0 11 £ Q. O E 0 0 0 — 73 C E ~ 0 73' 0 4-> 0 (0 73 O 73 CL C 0 > 3 c -C XI — II 0 Q. Q. — JZ 3 C O O I/) « c 0 E > DO 0 " 2 > “O a 0 4-» 11 £ CD 0 0 0 +-» 1 0 0 to . 73 o I. ® 2rO JL < m £ O ii u. c iS s o to CD E O CQ +-i o m <13 < W Q to (0 00 o . 03 c M C 03 | 5 E 73 ^ Q3 > a 5 fc fc 0 o . .Q o in o) (/) i 3 0 ^ 5 E 0 i_ 3 03

PAGE 124

113

PAGE 125

0 _0 > 0 4-> T3 T3 [q E _Q 0 11 E '0 £ 3 -Q a > i— CO O 0 0 0 CQ CO 0 "O E 0 L~ +-> 0 0 +-< C 3 O "O II Q. p O 0 to' T3 C £ 0 i— 0 T3 T3 O 0 a. M 0 > c j= to 03 T3 C 3 Q o a «! 5 o O to to c = (0 _ E > m © II T3 JD -a' cl > 0 ® o -t— > || Qro 11 ;= 03 Q. ® ia) O v -S CO r^3 < E F O u. 0 c ca o o. o 03 '< to Q CO o o DQ (O 03 E 8° ® 2,c 0 to C.t r to 3 C T3 05 > CL 2 t 15 03 O . -Q O CD 03 tO 1 3 " 03 E 03 3 03 Ll

PAGE 126

115

PAGE 127

Q) CD -Q X! CD CD _3 XI > o 0) CD a CD M CD C "5 Q. o E E CD O O w CD qj II "O -Q -o' o. > II Q. CD 11 != cd cl ro i_ CD O v • -S < E C m ^ CD _ O II < CT) C CD ro m c m S .2 . o y m w O CD CD <• C W Q C ro ro I2i = .”] root T3 ‘t "... CD >. Q. ^ t ro CD O . XI o s 0 in i 3 CD X E CD 3 CD

PAGE 128

117

PAGE 129

Figure 5-8. Median cross section of (A) GA 3 -treated and (B) pollinated 'Beckyblue' rabbiteye blueberry fruit at 45 DAB. Bar=150/;m. ep = epidermis; hp = hypodermis; om = outer mesocarp; mm = middle mesocarp; im = inner mesocarp; en = endocarp; vb = vascular bundles; sc = sclereids.

PAGE 130

119

PAGE 131

Figure 5-9. Median cross section of (A) GA 3 -treated and (B) pollinated 'Beckyblue' rabbiteye blueberry fruit at ripening (87 and 72 DAB, respectively). Bar= 150>t/m. ep = epidermis; hp = hypodermis; om = outer mesocarp; mm = middle mesocarp; im = inner mesocarp; en=endocarp; vb=vascular bundles; sc =sclereids.

PAGE 132

121

PAGE 133

20 18 16 14 12 10 8 6 4 2 I 0 32 28 24 20 16 12 8 4 0 32 28 24 20 16 12 8 4 0 | 122 GA 3 POLL NP Outer mesocarp Middle mesocarp Inner mesocarp 30 40 50 60 Days after bloom 80 90 5-10. Increases in cell size in developing mesocarp tissues of GA 3 treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) blueberry fruits. Significance level at 0.05, n = 9.

PAGE 134

30 25 20 15 10 5 70 60 50 40 30 20 10 30 25 20 15 10 5 °, ure 123 GA, POLL NP \ Hypodermis 20 40 60 80 Days after bloom 5-11. Average cell enlargement rate in developing (A, B) epidermal, ypodermal, and (C) endocarp tissues of GA 3 -treated (GA 3 ), pollinated 3 OLL), and nonpollinated (NP) blueberry fruits. Calculated from data in igure 5-4.

PAGE 135

400 300 200 100 600 400 200 0i 800 600 400 200 0i 124 GA 3 poll NP ~80 Days after bloom i-12. Average cell enlargement rate in developing mesocarp tissues of A 3 -treated (GA 3 ), pollinated (POLL), and nonpollinated (NP) blueberry uits. Calculated from data in Figure 5-4.

PAGE 136

Cell number and cell size x 10 3 (urn 2 ) 125 3 2 1 3 2 1 3 2 1 0 Figure 5-13. Developmental changes in mesocarp cell size, fruit fresh weight, and pericarp cell number in (A) pollinated, (B) GA 3 -treated, and (C) nonpollinated rabbiteye blueberry fruits. Fresh weight (g)

PAGE 137

126 Table 5-1. Pericarp cell size increases in whole cross-sectional areas of 'Beckyblue' rabbiteye blueberry fruits. Tissue Treatment Cell size (/jm 2 ) Increase in size 0 DAB Ripening Epidermis ga 3 376. 9 yz 1 121.5a 2.9x Pollinated 400.7 1229.1a 3. lx Non-pollinated 359.5 494.5b 1.4x Hypodermis ga 3 577.1 1860.6a 3.2x Pollinated 531.2 2138.1a 4. Ox Non-pollinated 543.2 880.0b 1.6x Outer ga 3 872.1 15242.7b 17. 5x mesocarp Pollinated 924.7 17290.9a 18. 7x Non-pollinated 895.5 4035.3b 4.5x Middle ga 3 757.1 20892.1b 27. 6x mesocarp Pollinated 863.6 29890.5a 34. 6x Non-pollinated 725.5 6333.7b 8.7x Inner ga 3 993.7 23000.6b 23. lx mesocarp Pollinated 1021.6 32612.9a 31. 9x Non-pollinated 913.1 8120.3a 8.9x Endocarp ga 3 179.4 1207.3a 6.7x Pollinated 164.8 1049.4a 6.4x V D i r> n r» i r\ n _ 0*7 Non-pollinated 157.6 356.6b "> HAD ill 2.3x y Ripening = 87 DAB for GA 3 -treated fruits and 72 DAB for pollinated fruits. Data for non-pollinated fruits are reported at 25 DAB. z Mean separation within tissue type by LSMean's, p = 0.05.

PAGE 138

CHAPTER 6 SUMMARY The stimulus for fruit set elicited by pollination and fertilization has not been identified. In many cases, pollination and/or fertilization can be replaced by exogenous applications of growth regulators. Gibberellins have been used successfully to induce fruit set and parthenocarpic fruit development in a number of fleshy fruits (Leopold, 1962; Wittwer et al., 1956; Webster and Goldwin, 1984; Hartman and Anvari, 1986) including blueberry. However, in these cases, response to GA 3 has been inconsistent and appears to vary with concentration, timing, and genotype. Additionally, even under conditions that induce successful development of GA 3 4nduced parthenocarpic fruit, final fruit weight is often smaller than that of pollinated fruit. The mechanism by which GA 3 induces fruit set and development is unknown. It has been observed that gibberellins, as well as other plant hormones, alter the pattern of assimilate distribution within the plant (Daie, 1985; Kinet et al., 1985; Aloni et al., 1986). Changes in assimilate distribution, specifically, enhancement of carbohydrate accumulation at the site of hormone application, have been correlated with fruit growth and development in several fleshy fruits (Marre and Murnek, 1 952; Stembridge and Gambrell, 1972; Adedipe et al., 1976; Craighton et al., 1986; Pereto and 127

PAGE 139

128 Beltran, 1987; Stutte and Gage, 1990). GA 3 has also been implicated in regulation of cell division and/or cell enlargement. In some cases, cell division and cell enlargement has been increased upon exogenous applications of GA 3 (Vercher et al . , 1 984), but in others, only cell enlargement has been stimulated (Bunger-Kibler and Bangert, 1983). Final fruit size may be determined by the number of cells at anthesis (Scorza et al., 1991), by the extent of cell division after pollination (Pearson and Robertson, 1 952; Smith, 1 940), by the degree of cell enlargement (Harris et al., 1968), and/or a combination. (Sinnot, 1939). The first part of the research reported here was undertaken to determine the optimum GA 3 concentration and timing for successful development of parthenocarpic blueberry fruits. Parthenocarpic fruit set and development was induced with all combinations of GA 3 concentrations and times (0, 7, 21 , and 42 DAFB) tested as well as in combination with self or cross-pollination. Fruit size was reduced by GA 3 treatments alone compared to pollinated treatments, but GA 3 applications in combination with full crossor self-pollination did not negatively affect fruit size. This implies that under situations of partial pollination and/or self-pollination, GA 3 can successfully enhance the stimulus for fruit set delivered by normal pollination and fertilization. In the second part of the research, the basis for the decrease in size of GA 3 -induced parthenocarpic fruit compared to pollinated fruit was investigated. The first hypothesis was that the reduction in size of GA 3 -induced parthenocarpic fruits was due to differences in the ability to import and/or

PAGE 140

129 metabolize assimilates. Pollinated fruits began importing and accumulating C at higher rates than GA 3 -treated fruits as early as 2 to 4 DAB, and this high rate of C import was sustained throughout fruit development. The prolonged second period of growth in GA 3 -treated fruits resulted in relatively higher respiratory C losses compared to pollinated fruits, and this may have negatively affected final fruit size. GA 3 -treated fruits appeared to be sink-limited and imported only 44% as much carbon as pollinated fruits. Glucose and fructose accumulated in similar concentrations in fruits from both treatments, and hexose accumulation was strongly correlated with soluble invertase activity in fruits. SPS and SS activities remained low at all points examined during fruit development in both treatments and appeared to play a negligible role in carbohydrate accumulation in blueberry fruits. There were no differences in fruit invertase, SPS, and/or SS activities between pollinated and GA 3 -treated fruits throughout development. Thus, differences in fruit weight between treatments were not correlated with differences in sucrosemetabolizing enzyme activities in fruit extracts and/or non-structural carbohydrate accumulation. The second hypothesis was that the reduction in size of GA 3 -induced parthenocarpic fruit was due to limitations in GA 3 -induced cell division and or cell enlargement. Cell division was stimulated to the same extent by both GA 3 and pollination. However, the rate and extent of cell enlargement was significantly reduced in GA 3 -treated compared to pollinated fruits. Thus,

PAGE 141

130 differences in fruit size between pollinated and GA 3 -treated fruits were due primarily to differences in cell size, with cell number playing only a minor role. The lower cell enlargement in GA 3 -treated fruits may have been due to a decreased cell wall extensibility.

PAGE 142

LITERATURE CITED Aalders, L. E. and I. V. Hall. 1961. Pollen incompatibility and fruit set in lowbush blueberries. Can. J. Genet. Cytol. 3:300-307. Achkireddy, N. R., R. C. Kirkwood, and G. van den Berg. 1984. The development of Pisum sativum explant systems for studies concerning source-sink activities. Physiol. Plant. 61:130-134. Ackerson, R. C. 1985. Invertase activity and abscissic acid in relation to carbohydrate status in developing soybean reproductive structures. Crop Sci. 25:61 5-61 8. Adams, P. A., M. J. Montague, M. Tepfer, D. L. Rayle, H. Ikuma, and P. B. Kaufman. 1 975. Effect of gibberellic acid on the plasticity and elasticity of Avena stem segments. Plant Physiol. 56:757-760. Adedipe, N. O. 1956. Distribution of 14 C in cowpea (Vigna unguiculata L.) in relation to fruit abscission and treatment with benzyladenine. Ann. Bot. 40:731-737. Aloni, B., J. Daie, and R. E. Wyse. 1986. Enhancement of [ 14 C]Sucrose export from source leaves of Vicia faba by gibberellic acid. Plant Physiol. 82:962-966. Avigad, G. 1982. Sucrose and other disaccharides, pp. 216-234. In: Plant carbohydrates I. Intracellular carbohydrates. F. A. Loewus and W. Tanner (eds.). Vol13A. Springer. Berlin. Bailey, J. S. 1938. The pollination of the cultivated blueberry. Proc. Amer. Soc.Hort. Sci. 35:71-72. Barker, W. G., and W. B. Collins. 1965. Parthenocarpic fruit set in the lowbush blueberry. Proc. Amer. Soc. Hort. Sci. 87:229-233. 131

PAGE 143

132 Batjer, L. P. and G. C. Martin. 1965. The influence of night temperature on growth and development of early Redhaven peaches. Proc. Amer. Soc. Hort. Sci. 87:139-144. Bell, H. P. 1957. The development of the blueberry seed. Can. J. Bot. 35:139149. Bell, H. P. and J. Burchill. 1955. Flower development in the lowbush blueberry. Can. J. Bot. 33:251-258. Birkhold, K. T., K. E. Koch, and R. L. Darnell. 1992. Carbon and nitrogen economy of developing rabbiteye blueberry fruit. J. Amer. Soc. Hort. Sci. 117:139-145. Brenner, M. L., B. M. N. Schreiber, and R. J. Jones. 1989. Hormonal control of assimilate partitioning: regulation in the sink. Acta Hort. 239:141147. Brewer, J. W. and R. C. Dobson. 1969. Pollen analysis of two highbush blueberry varieties Vaccinium corymbosum. J. Amer. Soc. Hort. Sci. 94:251-252. Brock, T. G. and R. E. Cleland. 1990. Biophysical basis of growth promotion in primary leaves of Phaseo/us vulgaris L. by hormones versus light. Planta 82:427-431. Broughton, W. J. and A. J. McComb. 1971. Changes in the pattern of enzyme development in gibberellin-treated pea internodes. Ann. Bot. 35:213-228. Browning, G. 1989. The physiology of fruit set. pp. 195-217. In: Manipulation of fruiting. J. Wright (ed.). Butterworths. London. Bunger-Kibler, S. and F. Bangerth. 1983. Relationship between cell number, cell size and fruit size of seeded fruits of tomato (Lycopersicon esculentum Mill.), and those induced parthenocarpically by the application of plant growth regulators. Plant Growth Regul. 1:143-154. Cane, J. H. and J. A. Payne. 1 990. Native bee pollinates rabbiteye blueberry. Highlights Agric. Res. 37:4. University of Alabama. Chen, W-S., H-Y. Liu, Z-H. Liu, L. Yang, and W-H. Chen. 1994. Gibberellin and temperature influence carbohydrate content in flowering in Pha/aenopsis. Physiol. Plant. 90:391-395.

PAGE 144

133 Claussen, W., B. R. Loveys and J. S. Hawker. 1985. Comparative investigations on the distribution of sucrose synthase activity and invertase activity within growing, mature and old leaves of some C3 and C4 plant species. Physiol. Plant. 65:275-280. Claussen, W., B. R. Loveys, and J.S. Hawker. 1986. Influence of sucrose and hormones on the activity of sucrose synthase and invertase in detached leaves and leaf sections of eggplants (Solanum melongena) . J. Plant Physiol. 124:345-357. Clore, W. J. 1965. Responses of Delaware grapes to gibberellin. Proc Amer. Soc. Hort Sci. 87:259-263. Cockerham, L. E. and G. J. Galletta. 1976. A survey of pollen characteristics in certain Vaccinium species. J. Amer. Soc. Hort. Sci. 101:671-676. Collins, W. B., K. H. Irving, and W. G. Barker. 1966. Growth substances in the flower bud and developing fruit of Vaccinium angustifolium Ait. Proc. Amer. Soc. Hort. Sci. 89:243-247. Coggins Jr., C. W., H. Z. Hield, and M J. Garber. 1960. The influence of potassium gibberellate on Valencia orange trees and fruit. Proc. Amer. Soc. Hort. Sci. 76:193-198. Coombe, B. G. 1 960. Relationship of growth and development to changes in sugars, auxins, and gibberellins in fruit of seeded and seedless varieties of Vitis vinifera. Plant Physiol. 35:241-250. Coombe, B. G. 1976. The development of fleshy fruits. Ann. Rev. Plant Physiol. 27:507-528. Cosgrove, D. 1986. Biophysical control of plant cell growth. Ann. Rev. Plant Physiol. 37:337-405. Cosgrove, D. J. and S. A. Sovonick-Dunford. 1 989. Mechanism of gibberellindependent stem elongation in peas. Plant Physiol. 89:184-191. Coville, F. V. 1921. Directions for blueberry culture. U.S.D.A. Bull. 974. Craighton, S. M., M. G. Bausher, and G. Yelenosky. 1986. Influence of growth regulator treatments on dry matter production, fruit abscission, and 14 C-assimilate partitioning in citrus. J. Plant Growth Regul. 5:1 1 1 120 .

PAGE 145

134 Crane, J. C. and R. Blondeau. 1949. Controlled growth of fig fruits by synthetic hormone application. Proc. Amer. Soc. Hort. Sci. 54:102108. Crane, J. C., M. V. Brandley, and L. C. Luckwill. 1959. Auxins in parthenocarpic and non-parthenocarpic figs. J. Hort. Sci. 34:142-153. Crane J. C. and N. Grossi. 1960. Fruit and vegetative responses of the mission fig to gibberellin. Proc. Amer. Soc. Hort. Sci. 75:139-145. Daie, J. 1985. Carbohydrate partitioning and metabolism in crops. Hort. Rev. 7:69-108. Daie, J. 1987. Bioregulator enhancement of sink activity in sugar beet. Plant Growth Reg. 5:219-228. Daie, J., M. Watts, B. Aloni, and R. E. Wyse. 1986. In vitro and in vivo modification of sugar transport and translocation in celery by phytohormones. Plant Sci. 46:35-41. Dali, N., D. Michaud, and S. Yelle. 1992. Evidence for the involvement of sucrose phosphate synthase in the pathway of sugar accumulation in sucrose-accumulating tomato fruits. Plant Physiol. 99:434-438. Darnell, R. L., R. Cano-Medrano, K. E. Koch, and M. L. Avery. 1994. Differences in sucrose metabolism relative to accumulation of bird deterrent sucrose levels in fruits of wild and domestic Vaccinium species. Physiol. Plant. In Press. Darnell, R. L. and P. M. Lyrene. 1989. Cross-incompatibility of two related rabbiteye blueberry cultivars. HortScience 24:1017-1018. Darnell, R. L. and G. C. Martin. 1988. Role of assimilate translocation and carbohydrate accumulation in fruit set of strawberry. J. Amer. Hort. Sci. 1 13:1 14-1 18. Darrow, G. M. 1958. Seed number in blueberry fruits. Proc. Amer. Soc. Hort. Sci. 72:212-215. Davies, F. S. 1986. Flower possition, growth regulators, and fruit set of rabbiteye blueberries. J. Amer. Soc. Hort. Sci. 111:338-341.

PAGE 146

135 Davies, F. S. and D. W. Buchanan. 1979. Influence of GA 3 on rabbiteye blueberry fruit set, yield and quality, pp. 229-236. In: J.N. Moore (ed.). Proc. IV North Amer. Blueberry Res. Workers Conf., Fayetteville, Ark. Dawson, R. M. C., D. C. Elliot, W. C. Elliot, and K. C. Jones. 1986. Data for biochemical reseach. Oxford University Press. New York. Dennis Jr., F. G. 1981. Limiting factors in fruit set of 'Delicious' apple. Acta Hort. 120:1 19-123. DeJong, T. M. and J. Goudriaan. 1989. Modeling peach fruit growth and carbohydrate requirements: reevaluation of the double-sigmoid growth pattern. J. Amer. Soc. Hort. Sci. 114:800-804. Devlin, R. M. and L. E. Demoranville. 1967. Influence of gibberellic acid and gibrel on fruit set and yield in Vaccinium macrocarpon cv Early Black. Physiol. Plant. 20:587-597. Doll, S., F. Rodier, and J. Willenbrink. 1979. Accumulation of sucrose in vacuole isolated from red beet tissue. Planta 144-409-41 1 . Doughty, C. C. and W. P.A. Scheer. 1975. Growth regulators increase yield and reduce length of harvest of highbush blueberries. HortScience 10:260-261. Durley, R. C., J. MacMillan, and R. J. Pryce. 1971. Investigation of gibberellins and other growth substances in the seed of Phaseolus multiflorus and of Phaseolus vulgaris by gas chromatography and by gas chromatography-mass spectrometry. Phytochem. 10:1891-1908. Eaton, G. W. 1966. Production of highbush blueberry pollen and its germination in vitro as affected by pH and sucrose concentration. Can. J. Plant Sci. 46:207-209. Eaton, G. W. 1967. The relationship between seed number and berry weight in open-pollinated highbush blueberries. HortScience 2:14-15. Eaton, G. W. and A. M. Jamont. 1966. Megagametogenesis in Vaccinium corymhosum L. Can. J. Bot. 44:712-714. Echeverria, E. 1990. Developmental transition from enzymatic to acid hydrolysis of sucrose in acid limes (Citrus aurantifo/ia) . Plant Physiol. 92:168-171.

PAGE 147

136 Echeverria, E. and J.K. Burns. 1989. Vacuolar acid hydrolysis as a physiological mechanism for sucrose breakdown. Plant Physiol. 90:530533. Echeverria. E. and J.K. Burns. 1990. Sucrose breakdown in relation to fruit growth of acid lime (Citrus aurantifolia). J. Exp. Bot. 41:705-708. Echeverria, E. and J. Valich. 1988. Carbohydrate and enzyme distribution in protoplasts from 'Valencia' orange juice sacs. Phytochem. 27:73-76. Eck, P. and C. M. Mainland. 1971. Highbush blueberry fruit set in relation to flower morphology. HortScience 6:494-495. Edgerton, L. K. 1 981 . Some effects of aminoethoxyvinylglicine, 6-benzilamino purine and gibberellins on fruit set and vegetative growth of apple. Acta Hort. 120-125-130. Edwards, T. W. Jr. 1970. The morphology of fruit development in blueberry clones. M. S. Thesis. Edwards, T. W. Jr., W. B. Sherman, and R. H. Sharpe. 1970. Fruit development in short and long cycle blueberries. HortScience 5:274275. El-Abd, S. A., A. S. El-Beltagy, and M. A. Hall. 1986. Physiological studies on flowering and fruit set in tomatoes. Acta Hort. 179:349-354. El-Agamy, S. Z. A., W. B. Sherman, and P. M. Lyrene. 1982. Pollen incompatibility in blueberries ( Vaccinium spp). J. Palynology 18: 1 0S112 . Estruch, J.J. and J. P. Beltran. 1991. Changes in invertase activities precede ovary growth induced by gibberellic acid in Pissum sativum. Physiol. Plant. 81:319-326. Fiew, S. and J. Willenbrink. 1987. Sucrose synthase and sucrose phosphate synthase in sugar beet plants (Beta vulgaris L. ssp. altissima). J. Plant Physiol. 131:153-162. Forshey, C. G. and D. C. Elfving. 1977. Fruit numbers, fruit size, and yield relationships in 'McIntosh' apples. J. Amer. Soc. Hort. Sci. 102:399402.

PAGE 148

137 Garcia-Martinez, J. L. and J. Carbonell. 1985. Induction of fruit set and development in pea ovary explants by gibberellic acid. J. Plant Growth Regul. 4: 1 9-27. Garvey, E. J. and P. M. Lyrene. 1987. Self-incompatibility in 19 native blueberry selections. J. Amer. Soc. Hort. Sci. 112:856-858. Giaquinta, R. T. 1979. Sucrose translocation and storage in the sugar beet. Plant Physiol. 63:828-832. Gil, G. F., G. C. Martin, and W. H. Griggs. 1972. Fruit set and development in the pear: extractable endogenous hormones in parthenocarpic and seeded fruits. J. Amer. Soc. Hort. Sci 97:731-735. Goldwin, G. K. 1984. Factors affecting hormone-assited setting of Cox's apple. Acta Horticultural 149:161-171. Goldy, R. G. 1985. Relationship of morphological development to meiotic occurrence in various small fruits. HortScience 20: 255-256. Gordon, A. J. and A. E. Flood. 1 980. 2,4-dichlorophenoxyacetic acid and the de novo synthesis of invertase in chicory root tissue. Phytochem. 19:505-508. Gough, R. E., V. G. Shutak, and R. L. Hauke. 1978. Growth and development of highbush blueberry. II. Reproductive growth, histological studies. J. Amer. Soc. Hort. Sci 103:476-479. Greenberg, J. and E. E. Goldschmidt. 1989. Acidifying agents, uptake, and physiological activity of gibberellin A 3 in Citrus. HortScience 24:791793. Gupton, C. L. 1984. Effect of pollen source on fruit characteristics of low chilling highbush type blueberries. HortScience 19:531-532. Gustafson, F. G. 1939. The cause of natural parthenocarpy. Amer. J. Bot. 26:135-138. Gustafson, F. G. 1959. Influence of gibberellic acid on setting and development of fruits in tomato. Plant Physiol. 43:521-523. Harris, J. M., P. E. Kriedemann, and J. V. Possingham. 1968. Anatomical aspects of grape berry development. Vitis 7:106-1 19.

PAGE 149

138 Hartman, W. and S. F. Anvari. 1986. Effects of GA 3 on fruit and seed development of self-sterile plum cultivars. Acta Hort. 179:349-354. Hayes, P. M. and J. W. Patrick. 1985. Photosynthate transport in stems of Phaseo/us vulgaris L. treated with gibberellic acid, indole-3 acetic acid or kinetin. Planta 1 66:371 -379. Heilman, E. W., and J. N. Moore. 1983. Effect of genetic relationship to pollenizer on fruit, seed, and seedling parameters in highbush and rabbiteye blueberries. J. Amer. Soc. Hort. Sci. 108:401-405. Herrero, M. 1 984. Effect of time of GA 3 treatment on 'Agua de Aranjuez' pear fruit set. Acta Hort. 149:211-215. Herrero, M. and M. Gascon. 1 987. Prolongation of embryo sac viability in pear (Pyrus communis ) following pollination or treatment with gibberellic acid. Ann. Bot. 60:287-293. Hindle, R. Jr., V. G. Shutak, and E. P. Christopher. 1957. Growth studies of the highbush blueberry fruit. Proc. Amer. Soc. Hort. Sci. 68:282-289. Ho, L. C. 1 988. Metabolism and compartmentation of imported sugars in sink organs in relation to sink strength. Ann. Rev. Plant Physiol. Plant Mol. Biol. 39:355-378. Howell, G. S., M. W. Kilby, and J. E. Nelson. 1 972. Influence of timing of hive introduction on production of highbush blueberries. HortScience 7:1 29131. Hubbard, N. L., S. C. Huber, and D. M. Pharr. 1989. Sucrose phosphate synthase and acid invertase as determinants of sucrose concentration in developing muskmelon (Cucumis melo L.) fruits. Plant Physiol. 91 :1 5271534. Hubbard, N. L., D. M. Pharr, and S. C. Huber. 1991. Sucrose phosphate synthase and other sucrose metabolizing enzymes in fruits of various species. Physiol. Plant. 82:191-196. Huber, S. C. 1989. Biochemical mechanism for regulation of sucrose accumulation in leaves during photosynthesis. Plant Physiol. 91:656662. Iki, K., K. Segiguchi, K. Kurate, T. Tada, H. Nakagawa, N. Ogura, and H. Takehana. 1978. Immunological properties of fructofuranosidase from

PAGE 150

139 ripening tomato fruit. Phytochemistry 17:311-312. Iwahori, S., R. J. Weaver, and R. M. Pool. 1968. Gibberellin-like activity in berries of seeded and seedless Tokay grapes. Plant Physiol. 43:333-337. Iwasaki, T. H. Fakuda, and H. Shibaoka. 1986. Inhibition of cell division and DNA synthesis by gibberellin in isolated Zinnia mesophile cells. Plant Cell Physiol. 27:717-724. Jackson, D. I. 1968. Gibberellin and the growth of peach and apricot fruits. Aust. J. Biol. Sci. 21:209-215. Jackson, D. I. and B. G. Coombe. 1965. The growth of apricot fruit. I. Morphological changes during development and the effects of various tree factors. Aust. J. Agric. Res. 17:465-477. Kaufman, P. B., N. S. Ghosheh, J. D. La Croix, S. L. Soni, and H. Ikuma. 1973. Regulation of invertase levels in Avena segments by gibberellic acid, sucrose, glucose, and fructose. Plant Physiol. 52:221-228. Keyes, G., M. E. Sorrels, and T. L. Setter. 1 990. Gibberellic acid regulates cell wall extensibility in wheat (Triticum aestivum L.). Plant Physiol. 92:242245. Kaliarackal, J. and J. A. Milburn. 1985. Respiration and phloem translocation in the roots of chickpea [Cicer arietinum). Ann. Bot. 56:21 1-218. Katsumi, M. and H. Kazama. 1978. Gibberellin control of cell elongation in cucumber hypocotyl sections. Bot. Mag. Tokyo. Special Issue 1:141-158. Kim, H. Y. and Y. Suzuki. 1989. Changes in assimilated 13 C distribution and soluble acid invertase activity of Zinnia elegans induced by uniconazol, an inhibitor of gibberellin biosynthesis. Plant Physiol. 90:316-321. Kimura, Y. 1970. Identity of Pisum gibberellin. Agric. Food Chem. 1 8: 1 82186. Kinet, J. M., V. Zune, C. Linotte, A. Jacqmard, and G. Bernier. 1985. Resumption of cellular activity induced by cytokinin and gibberellin treatments in tomato flowers targeted for abortion in unfavorable light conditions. Physiol. Plant. 64:67-73. Komor, E., M. Thom, and A. Maretzki. 1981. The mechanism of sugar uptake by sugar suspension cells. Planta 153:181-192.

PAGE 151

140 Krewer, G., S. NeSmith, D. Stanaland, J. Clark, M. Bruorton, and J. E. Smith. 1991. Results of 1991 field trials with gibberellic acid on rabbiteye blueberries in south Georgia, pp. 9-15. In: Proc. of the S.E. Prof. Fruit Workers Conf. Tifton, Ga. Lane, W. D. 1984. Fruit set pre-treatment with foreign pollen compared with killed compatible pollen. Can. J. Bot. 62:1678-1681. Lang, G. A., and R. G. Danka. 1991. Honey-bee-mediated cross-versus selfpollination of 'Sharpblue' blueberry increases fruit size and hastens ripening. J. Amer. Soc. Hort. Sci. 116:770-773. Lee, W. R. 1958. Pollination studies on lowbush blueberries. J. Econ. Entom. 51:544-545. Leigh, R. A., T. ApRees, W. A. Fuller, and J. Banfield. 1979. The localization of acid invertase and sucrose in the vacuoles of storage roots of beetroot (Beta vulgaris). Biochem. J. 178:539-547. Leopold, A. C. 1962. The roles of growth substances in flowers and fruits. Can. J. Bot. 40:745-754. Lethman, D. S. 1963. Regulators of cell division in plant tissues. N. Z. J. Bot. 1:336-350. Lockhart, J. A. 1965. An analysis of irreversible plant cell elongation. J. Theor. Biol. 8:264-275. Looney, N. E. and R. P. Pharis. 1986. Gibberellins and reproductive development of tree fruits and grapes. Acta Hort. 179:59-71 Lyrene, P. M. 1 989. Pollen source influences fruiting of 'Sharpblue' blueberry. J. Amer. Soc. Hort. Sci. 144:995-999. Lyrene, P. M. and R. G. Goldy. 1 983. Cultivar variation in fruit set and number of flowers per cluster in rabbiteye blueberry. HortScience 1 8:228-229. Mainland, C. M. and P. Eck. 1968a. Growth regulator survey for activity in inducing parthenocarpy in the highbush blueberry. HortScience 3:170172. Mainland, C. M. and P. Eck. 1 968b. Induced parthenocarpic fruit development in highbush blueberry. Proc. Amer. Soc. Hort. Sci. 92:284-289.

PAGE 152

141 Mainland, C. M. and P. Eck. 1969a. Fruiting response of the highbush blueberry to gibberellic acid under field conditions. J. Amer. Soc. Hort. Sci. 94:21-23. Mainland, C. M. and P. Eck. 1969b. Fruit and vegetative responses of the highbush blueberry to gibberellic acid under greenhouse conditions. J. Amer. Soc. Hort. Sci. 94:19-20. Mainland, C. M., and P. Eck. 1971. Endogenous auxin and gibberellin-like activity in highbush blueberry flowers and fruit. J. Amer. Soc. Hort. Sci. 96:141-145. Manning, K. and G. A. Maw. 1975. Distribution of acid invertase in the tomato plant. Phytochemistry 14:1965-1969. Mapelli, S., C. Frova, G. Torti, and G. P. Soressi. 1978. Relationship between set, development and activities of growth regulators in tomato fruits. Plant and Cell Physiol. 7:1281-1288. Marc, J. and W. P. Hackett. 1992. Changes in the pattern of cell arrangenment at the surface of the shoot apical meristem in Hedera helix L. following gibberellin treatment. Planta 186:503-510. Marcelis, L.F. M. and L. R. Baan Hofman-Eijer. 1993. Cell division and expansion in the cucumber fruit. J. Hort. Sci. 68:665-671. Marcelle, R.D. 1 984. Effects of GA 3 , BA and growth retardants on fruit set in the pear cultivar 'Doyenne du Comice'. Acta Hort. 149:225-229. Marre, E. and A. E. Murneek. 1952. Carbohydrate metabolism in the tomato fruit as affected by pollination, fertilization, and applications of growth regulators. Plant. Physiol. 27:255-266. Martin, G. C., R. Horgan, and C. Nishijima. 1982. Changes in hormone content of pear receptacles from anthesis to shortly after fertilization as affected by pollination or GA 3 treatment. J. Amer. Soc. Hort. Sci. 107:479-482. Meader, E. M. and G.M. Darrow. 1 944. Pollination of the rabbiteye blueberry and related species. Proc. Amer. Soc. Hort. Sci. 44:268-274. Merrill, T. A. 1 936. Pollination of the highbush blueberry. Mich. Agr. Exp. Sta. Tech. Bull. 151.

PAGE 153

142 McCollum, T. G., D. J. Huber, and D. J. Cantliffe. 1988. Soluble sugar accumulation and activity of related enzymes during muskmelon fruit development. J.Amer. Soc. Hort Sci. 113:339-403. Miron, D. and A. A. Schaffer. 1991. Sucrose phosphate synthase, sucrose synthase, and invertase activities in developing fruit of Lycopersicon esculentum Mill, and the sucrose accumulating Lycopersicon hirsutum Humb and Bonpl. Plant Physiol. 95:623-627. Mita, T. and M. Katsumi. 1 986. Gibberellin control of microtubule arrangement in the mesocotyl epidermal cells of the d5 mutant of Zea mays L. Plant Cell Physiol. 27:651-659. Miyamoto, K. and S. Kamisaka. 1988. Growth and osmoregulation in Pisum sativum subhooks as affected by gibberellic acid and cotyledon excision. Physiol. Plant. 74:669-674. Miyamoto, K. and S. Kamisaka. 1990. Effect of gibberellic acid on epicotyl growth and carbohydrate distribution in derooted Pisum sativum cutting with or without cotyledons. Physiol. Plant. 80:357-364. Miyamoto, K., J. Ueda, and S. Kamisaka. 1993. Gibberellin-enhanced sugar accumulation in growing subhooks of etiolated Pisum sativum seedlings. Effects of gibberellic acid, indoleacetic acid and cycloheximide on invertase activity, sugar accumulation and growth. Physiol. Plant. 88:301-306. Moore, J. N. 1964. Duration of receptivity to pollination of flowers of the highbush blueberry and the cultivated strawberry. Proc. Amer. Soc. Hort. Sci. 85:295-301. Moore, J. H., B. D. Reynolds, and G. R. Brown. 1972. Effects of seed number, size, and development on fruit size of cultivated blueberries. HortScience 7:268-269. Moriguchi, T., K. Abe, T. Sanada, and S. Yamaki. 1992. Levels and role of sucrose synthase, sucrose phosphate synthase, and acid invertase in sucrose accumulation in fruit of Asian pear. J. Amer. Soc. Hort. Sci. 1 17:274-278. Morris, D. A., and E. D. Arthur. 1984a. Invertase activity in sinks undergoing cell expansion. Plant Growth Reg. 2:327-337.

PAGE 154

143 Morris, D. A. and E.D. Arthur. 1984b. An association between acid invertase activity and cell growth during leaf expansion in Phaseolus vulgaris L. J. Exp. Bot. 35:1369-1379. Morris, D. A. and E. D. Arthur. 1985. Effects of gibberellic acid on patterns of carbohydrate distribution and acid invertase activity in Phaseolus vulgaris. Physiol. Plant. 65:257-262. Morris, D. A. and E. D. Arthur. 1986. Stimulation of acid invertase activity by indol-3yl-acetic acid in tissues undergoing cell expansion. Plant Growth Reg. 4:259-271. Morrow, E. B. 1943. Some effects of cross-pollination versus self-pollination in the cultivated blueberry. Proc. Amer. Soc. Hort. Sci. 42:470-472. Nakagawa, H., Y. Kawasaki, N. Ogura, and H. Takena. 1971. Purification and some properties of two types of fructofuronosidase from tomato fruit. Agric. Biol. Chem. 36:18-26. Nielsen, T. M., H. C. Skjaerbaek, and P. Karlsen. 1991. Carbohydrate metabolism during fruit development in sweet pepper (Capsicum annuum). Physiol. Plant. 82:311-319. NeSmith, D. S. and G. Krewer. 1991. The influence of flower bud stage and chill hours on the activity of GA 3 applied to rabbiteye blueberry. HortScience 27:316-318. Nishitani, K. and Y. Masuda. 1982. Roles of auxin and gibberellic acid in growth and maturation of epicotyls of Vigna angularis : cell wall changes. Physiol. Plant. 56:38-45. Ozga, J. A., M. L. Brenner, and D. M. Reinecke. 1992. Seed effects on gibberellin metabolism in pea pericarp. Plant. Physiol. 100:88-94. Pavel, E. W., and T. M. DeJong. 1993. Estimating the photosynthetic contribution of developing peach ( Prunus persica) fruits to their growth and maintenance carbohydrate requirements. Physiol. Plant. 88:331338. Payne, J. A., J. H. Cane, A. A. Amis, and P. M. Lyrene. 1 989. Fruit size, seed size, seed viability and pollination of rabbiteye blueberries ( Vaccinium ashei Reade). Acta Hort. 241:38-43.

PAGE 155

144 Pereto, J. G. and J. P. Beltran. 1987. Hormone directed sucrose transport during fruit set induced by gibberellins in Pisum sativum. Physiol. Plant. 69:356-360. Pharis, R. P. and R. W. King. 1 985. Gibberellins and reproductive development in seed plants. Ann. Rev. Plant. Physiol. 36:517-568. Poovaiah, B. W., and K. Veluthambi. 1985. Auxin-regulated invertase activity in strawberry fruits. J. Amer. Soc. Hort. Sci. 110:258-261. Powell, L E. and C. Pratt. 1966. Growth promoting substances in the developing fruit of peach. J. Hort. Sci. 41:331-348. Pressey, R. and J. R. Avants. 1980. Invertase in oat seedlings. I. Separation, properties and changes in activities in seedling segments. Plant Physiol. 65:136-140. Rabaey, A., and J. Luby. 1988. Fruit set in half-high blueberry genotypes following selfand cross-pollination. Fruit Var. J. 42:126-129. Ross, H. A. and H. V. Davies. 1992. Sucrose metabolism in tubers of potato (Solarium tuberosum L). Plant Physiol. 98:287-293. Ruiz, R. and J. L. Guardiola. 1994. Carbohydrate and mineral nutrition of orange fruitlets in relation to growth and abscission. Physiol. Plant. 90:27-36. Ryugo, K. 1988. Fruit culture: Its science and art. J. Wiley & Sons. New York. Sanchez-Beltran, M-J., J. Carbonel, J. L. Garcia-Martfnez, and I. Lopez-Diaz. 1992. Gene expression during two alternative pathways of ovary development in Pisum sativum: fruit development and ovary senescence. Physiol. Plant. 85:69-76. Sastry, K., and R. Muir. 1963. Gibberellin: effect on diffusible auxin in fruit development. Science 140:494-495. Satti, S. M. E. and N. F. Oebker. 1986. Effects of benzyladenine and gibberellin (GA 4/7 ) on flowering and fruit set of tomato under high temperature. Acta Hort. 190:347-354.

PAGE 156

145 Schaffer, A. A., B. Aloni, and E. Fogelman. 1987. Sucrose metabolism and accumulation in developing fruit of Cucumis. Phytochem. 26:18831887. Schaffer, A. A., 0. Sagee, E.E. Goldschmidt and R. Goren. 1987. Invertase and sucrose synthase activity, carbohydrate status and endogenous IAA levels during Citrus leaf development. Physiol. Plant. 69:151-155. Schmalstig, J. G. and W. D. Hitz. 1987. Contributions of sucrose synthase and invertase to the metabolism of sucrose in developing leaves. Plant Physiol. 85:407-412. Scorza, R., L. G. May, B. Purnell, and B. Upchurch. 1991. Differences in number and area of mesocarp cells between smalland large-fruited peach cultivars. J. Amer. Soc. Hort. Sci. 116:861-864. Sinnott, E. W. 1939. A developmental analysis of the relation between cell size and fruit size in Cucurbitis. Amer. J. Bot. 26:179-189. Southwick, S. M. and F. S. Davies. 1982. Growth regulator effects on fruit set and fruit size in Navel orange. J. Amer. Soc. Hort. Sci. 107:395397. Spiers, J. M. 1981. Fruit development in rabbiteye blueberry cultivars. HortScience 16:175-176. Stembridge, G. E. and C. E. Gambrell, Jr. 1972. Peach fruit abscission as influenced by applied gibberellin and seed development. J. Amer. Soc. Hort. Sci. 97:708-711. Stephenson, A. G. 1981. Flower and fruit abortion: proximate causes and ultimate functions. Ann. Rev. Ecol. Syst. 12:253-79. Stommel, J. R. 1992. Enzymic components of sucrose accumulation in the wild tomato species Lycopersicon peruvianum. Plant Physiol. 99:324328. Stushnoff, C. and L. F. Hough. 1968. Sporogenesis and gametophyte development in 'Bluecrop' and 'Coville' highbush blueberries. Proc. Amer. Soc. Hort. Sci. 93:242-249. Stutte, G. W. and J. Gage. 1 990. Gibberellin inhibits fruit abscission following seed abortion in peach. J. Amer. Soc. Hort. Sci. 115:107-110.

PAGE 157

146 Sun, J. D., T. Loboda, S. J. Sung, and C. C. Black. 1992. Sucrose synthase in wild tomato Lycopersicon chmielewskii, and tomato fruit sink strength. Plant Physiol. 98:1163-1169. Sung, S-J.S., D-P. Xu, and C. C. Black. 1989. Identification of actively filling sucrose sinks. Plant Physiol. 89: 1117-1121. Tamada, T., S. Ishikawa, and H. Iwagaki. 1977. The pollination of rabbiteye blueberries in Tokyo. Acta Hort. 61:335-341. Tanaka, Y. and I. Uritani. 1979. Effect of auxin and other hormones on the metabolic response to wounding in sweet potato roots. Plant Cell Physiol. 20:1557-1564. Torrey, J. G. 1976. Root hormones and plant growth. Ann. Rev. Plant Physiol. 27:435-459. Tukey, H. B. and J. 0. Young. 1 939. Histological study of the developing fruit of the sour cherry. Bot. Gaz. 100:723-749. Vanerwegen, J. and G. Krewer. 1990. Gibberellic acid has potential for use in rabbiteye blueberries, pp. 23-28. In: Proc. V Biennial Southwest Blueberry Conf. Fletcher, NC. Vercher, Y., A. Molowny, and J. Carbonell. 1987. Gibberellic acid effect on the ultrastructure of endocarp cells of unpollinated ovaries of Pisum sativum. Physiol. Plant. 71:302-308. Vercher, Y., A. Molowny, C. Lopez, J. L. Garcia-Martmez, and J. Carbonell. 1 984. Structural changes in the ovary of Pisum sativum L. induced by pollination and gibberellic acid. Plant Sci. Letters. 36:87-91. Walker, A. J. and L. C. Ho. 1 977. Carbon translocation in the tomato: carbon import and fruit growth. Ann. Bot. 41:813-823. Wang, F., A. Sanz, M. L. Brenner, and A. Smith. 1993. Sucrose synthase, starch accumulation, and tomato fruit sink strength. Plant Physiol. 101:321-327. Wareing, P. F. and I. D. J. Phillips. 1981. Growth and Differentiation In Plants, pp. 85-88. Pergamon Press. New York. Weaver, R. J. and S. B. McCune. 1960. Further studies with gibberellin on Vitis vinifera grapes. Bot. Gaz. 121:155-162.

PAGE 158

147 Weaver, R. J. and R. M. Pool. 1971. Thinning 'Tokay' and 'Zinfandel' grapes by bloom sprays of gibberellin. J. Amer. Soc. Hort. Sci. 96.820-822. Webster, A. D. and G. K. Goldwin. 1984. The effect of setting mixtures on the cropping, flowering, and vegetative growth of sweet cherry. Acta Hort. 149:217-223. White, E., and H. J. Clark. 1939. Some results of self-pollination of the highbush blueberry at Whitesbog, New Jersey. Proc. Amer. Soc. Hort. Sci. 36:305-309. Williamson, J. G., R. L. Darnell, G. Krewer, J. Vanerwegen, and S. NeSmith. 1994. Gibberellic acid: a management tool for increasing yield of rabbiteye blueberry. J. Small Fruit & Vitic. (Submitted). Wittwer, S. H., M. J. Bukovac, H. M. Sell, and L. E. Weller. 1956. Some effects of gibberellin on flowering and fruit setting. Plant Physiol. 3941. Wood, G. W. 1962. The period of receptivity in flowers of the lowbush blueberry. Can. J. Bot. 40:685-686. Wood, G. W. 1 966. The influence of honey bee pollination on fruit set of the lowbush blueberry. Can. J. Plant Sci. 41:332-335. Wu, L-L., J. P. Mitchell, N. S. Cohn, and P. B. Kaufman. 1993. Gibberellin (GA 3 ) enhances cell wall invertase activity and mRNA levels in elongating dwarf pea (Pisum sativum) shoots. Int. J. Plant Sci. 154:280-289. Xu, D-P., S-J. S. Sung, and C. C. Black. 1989a. Sucrose metabolism in lima bean seeds. Plant Physiol. 89:1 106-1 1 16. Xu, D-P., S-J. S. Sung, T. Loboda, P. P. Kormanik, and C. C. Black. 1989b. Characterization of sucrolysis via the uridine diphosphate and pyrophosphate-dependent sucrose synthase pathway. Plant Physiol. 90:635-642. Yelle, S., R. T. Chetelat, M. Dorais, J. W. DeVerma, and A. B. Bennett. 1991. Sink metabolism in tomato fruit. Plant Physiol. 95:1026-1035. Yelle, S., J. D. Hewitt, N. L. Robinson, S. Damon, and A.B. Bennett. 1988. Sink metabolism in tomato fruit. III. Analysis of carbohydrate assimilation in a wild species. Plant Physiol. 87:737-740.

PAGE 159

148 Young, R. S. 1 952. Growth and development of the blueberry fruit ( Vaccinium corymbosun L. and Vaccinium angustifolium Ait). Proc. Amer. Soc. Hort. Sci. 59:167-172. Young, M. J. and W. B. Sherman. 1978. Duration of pistil receptivity, fruit set, and seed production in rabbiteye and tetraploid blueberries. HortScience 13:278-279. Zack, C. D. and J. B. Loy. 1984. Comparative effects of gibberellic acid and N 6 -benzyladenine on dry matter partitioning and osmotic and water potentials in seedling organs of dwarf watermelon. J. Plant Growth Regul. 3:65-73.

PAGE 160

BIOGRAPHICAL SKETCH Raquel Cano Medrano spent the early part of her life in Durango, Mexico, where she was born and her high school years in Chihuahua, Mexico. She received her bachelor's degree in agronomy from the Universidad de Chapingo, Texcoco, Mexico, in 1984 and her Master of Science degree in horticulture from the Colegio de Postgraduados, Texcoco, Mexico, in 1 987. She worked in that institution from 1 987 to 1 990, when she was admitted to the horticultural science program at the University of Florida, where she was awarded the Ph.D. in 1994. 149

PAGE 161

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. Rebecca L. Darnell, Chair Associate Professor of Horticultural Science 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. Wayne B. Sherman, Cochair Professor of Horticultural Science 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 gf Philosophy. Karen E. Koch Professor of Horticultural Science 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 gfJDoctor ^ Philosophy. Donald J. f-U^per Professor of Horticultural Science 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. ~/J Kenneth J. Boote Professor of Agronomy

PAGE 162

This dissertation was submitted to the Graduate Faculty 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, 1994 Dean, College of Agriculture Dean, Graduate School