FRUIT DETACHMENT IN BLUEBERRY

by

TRIPTI VASHISTH

(Under the Direction of Anish Malladi)

ABSTRACT

A better understanding of fruit detachment and the processes mediating it is essential to improve the efficiency of mechanical harvesting in blueberry (Vaccinium sp.). In blueberry, fruit detachment may occur either at the point of attachment of the pedicel to the peduncle (peduncle- pedicel junction, PPJ) or at the point of attachment of the pedicel to the fruit (fruit-pedicel junction, FPJ). Whether fruit detachment at these junctions is mediated by the physiological process of abscission or through physical separation of the organ from the parent plant is not well understood. Abscission is a physiological process that involves the programmed separation of entire organs at an anatomically distinct layer called the abscission zone (AZ). Additionally, the spatial and temporal changes in the cell wall composition and metabolism and the regulation of fruit detachment process is not completely understood. In this study, a series of experiments were performed to understand physiological, biochemical and molecular aspects of fruit detachment in blueberry. Anatomical, physiological and microscopic analysis revealed that PPJ is the true abscission zone of blueberry and detachment at FPJ is as result of physical breakage. Glycome profiling and immuno-localization of PPJ indicated that alteration in pectins and hemicellulose plays a key role in cell separation during abscission. A number of cell-wall carbohydrate metabolism related genes were altered upon induction of abscission. RNA-Seq analysis showed that abscission agent induced abscission was associated with extensive changes in the expression of genes associated with the biosynthesis and signaling of phytohormones such as ethylene, jasmonic acid and auxin. Also, potentially a plant hormone cross-talk and interaction plays an important role in abscission. Scanning electron microscopy analysis and fruit detachment in response to mechanical shaking of southern highbush cultivar 'Suziblue' and accession line

TH729 revealed that variability in ease of fruit detachment is due to stronger PPJ. A decrease in expression of cell wall hydrolysis related gene was observed with stronger PPJ. Potentially these phytohormone and cell wall hydrolysis genes play an important role in the genetic ease of fruit detachment/abscission.

INDEX WORDS: blueberry, abscission, fruit detachment, MeJa, ethephon, glycome profiling, immuno-localization, RNA-Seq FRUIT DETACHMENT IN BLUEBERRY

by

TRIPTI VASHISTH

B.Tech., Bundelkhand University, India, 2006

M.S., University of Georgia, 2009

A Dissertation Submitted to the Graduate Faculty of The University of Georgia in Partial

Fulfillment of the Requirements for the Degree

DOCTOR OF PHILOSOPHY

ATHENS, GEORGIA

2013 © 2013

Tripti Vashisth

All Rights Reserved FRUIT DETACHMENT IN BLUEBERRY

by

TRIPTI VASHISTH

Major Professor: Anish Malladi

Committee: Dayton Wilde Rakesh Singh Robert Shewfelt Scott NeSmith

Electronic Version Approved:

Maureen Grasso Dean of the Graduate School The University of Georgia December 2013 DEDICATION

I would like to dedicate all my work to my parents Mr. OP Sharma and Mrs. Urmila Sharma and my husband Mr. Vijendra Sharma. Without their love and support, I would never have been able to accomplish this.

iv ACKNOWLEDGEMENTS

I would like to thank Dr. Anish Malladi for giving me the opportunity to work and learn under his guidance. Thank you Dr. Malladi for your support, guidance, listening to me and always showing me the right direction. You have been always there to correct me, and helped in learning, to do the things in a remarkable way.

I would also like to thank Dr. NeSmith, Dr. Shewfelt, Dr. Singh, and Dr. Wilde, for serving as my committee members. I am also thankful to whole faculty and staff of Horticulture

Department for their support throughout my research.

Special thanks to Lisa Johnson and Madhumita Dash for being supportive and helping me throughout. I would also like to thank Justin Porter, Kristin Abney, Jim Gegogeine and all the student in Horticulture Department for helping me out specially in cutting abscission zone.

I want to especially thank my father and mother, and my husband for always supporting me, believing in me and being a light during dark phases of my life.

v TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS...... v

CHAPTER

1 INTRODUCTION AND REVIEW OF LITERATURE ...... 1

2 FRUIT DETACHMENT IN RABBITEYE BLUEBERRY: ABSCISSION OR

PHYSICAL SEPARATION ...... 16

3 CHANGES IN THE CELL WALL COMPOSITION AND WALL-RELATED

GENE EXPRESSION DURING BLUEBERRY FRUIT ABSCISSION ...... 42

4 TRANSCRIPTOME WIDE CHANGES DURING THE INDUCTION OF

ABSCISSION IN BLUEBERRY FRUIT...... 85

5 VARIABILITY IN THE EASE-OF-FRUIT DETACHMENT IN BLUEBERRY...125

6 CONCLUSION...... 148

vi CHAPTER 1

INTRODUCTION AND REVIEW OF LITERATURE

Blueberry (Vaccinium sps.) is an indigenous crop of the United States. Blueberry is a flowering plant with berries ranging from purple to black in color. North America is the largest producer contributing approximately 90% of the world’s blueberries. The most widely grown species of blueberry are northern highbush, southern highbush, and rabbiteye. In 2009, the total production of blueberries was around 207 million kilograms, which was worth over 500 million U.S. dollars

(USDA). According to the USDA, consumption of fresh blueberries in 2007 was 0.26 kg per person, while consumption of frozen blueberries was an additional 0.14 kg per person

(www.agmrc.org). The demand and market of fresh blueberries has increased tremendously in last few decades. Therefore, continuous efforts are being made to improve production and other aspects such as fruit quality in blueberry.

Hand harvesting of rabbiteye and highbush blueberries is labor intensive and requires as many as 520 h of labor per acre (Brown et al., 1996). Hand harvesting costs range from $1.10 to

1.54 per kg for southern highbush blueberries and $0.86 to 1.10 per kg for rabbiteye blueberries

(Safley et al., 2005). Harvesting of blueberries is the largest expense in the production of the blueberry crop (Yarborough, 1992). Hence there is a strong interest in reducing harvesting costs using mechanical harvesting. Mechanical harvesting can potentially reduce costs to about $0.26 per kg for rabbiteye blueberry.

Mechanical harvesting of blueberries has been done commercially since 1966 (Austin and Williamson 1977, Mainland et al. 1975). Mechanical harvesters are typically over the row

1 harvesters and detach the berries from the bush through physical shaking of the bush. It was reported by Austin and Williamson (1977) that a greater amount of ripe rabbiteye blueberries were lost on the ground compared to harvested berries as a result of mechanical harvesting.

Some green and unripe fruits were also removed by mechanical harvesting, resulting in additional losses. Sorting and grading of machine-harvested fruit on a commercial cleaning line further soften the berries resulting in greater decay in storage (Mainland et al. 1971, Mainland et al. 1975). Excessive vibration in mechanical harvesting operations often results in breakage of the pedicel, or detachment of the fruit away from the pedicel-berry junction. Therefore, such mechanical harvesting methods can ultimately lead to fruit injury and loss of overall fruit quality

(Howell et al., 1976). Hence, methods to increase the mechanical harvesting efficiency are required.

Abscission

Abscission is a physiological process that involves the programmed separation of entire organs, such as leaves, petals, fruit, and flowers. Abscission is a highly coordinated and regulated process in response to plant developmental cues, and various biotic and abiotic stresses. Organ separation occurs at abscission zones (AZ). Abscission involves the breakdown of cell walls. The spatial and temporal regulation of the dissolution of primary cell wall polysaccharides and middle lamella is not completely understood, but many studies have indicated that the loss of pectins and other polysaccharides from the middle lamella and the primary cell wall is associated with abscission (Clements and Atkins, 2001; Roberts et al., 2000; Uheda and Nakamura, 2000).

Fruit abscission in blueberry can occur either at the peduncle-pedicel junction (branch AZ) or the pedicel-berry junction (fruit AZ) (Gough and Litke, 1980). Enhancing the abscission characteristics of blueberry fruit can lead to increased efficiency of mechanical harvesting.

2 Abscission can be divided into four major steps (Patterson, 2007). During the first step formation of the AZ occurs. In step 2, AZ cells become responsive to phytohormones such as ethylene (by far the most extensively studied), jasmonates, auxin and abscissic acid (ABA). In response to the phytohormone signaling the middle lamella is dissolved. During step 3, cells in the AZ expand and cell wall loosening occurs. Finally in step 4, cell separation occurs followed by the suberization of the proximal layer of cells to form the protective layer.

1. Formation of abscission zone

The abscission zone is comprised of several layers of small, densely cytoplasmic cells at the point of attachment of the organ to the main body of the plant (Patterson, 2001). Abscission zone cells contain large deposits of starch, and have highly branched plasmodesmata as the process of abscission starts at the corners of cell and in regions dense in plasmodesmata. Most of the enzymatic activity is also directed towards the plasmodesmata (Sexton et al. 1976). Abscission zone cells often compose a complete layer in between the plant and the detaching organ (Sexton and Roberts, 1982). The number of cells forming the abscission zone appears to be fixed for each species but varies among different species (Taylor and Whitelaw, 2001). Mutants defective in

AZ formation have been identified. Loss of JOINTLESS, a MADS box gene, results in suppression of the pedicel AZ formation in tomato (Mao et al., 2000).

2. Changes in middle lamella and cell wall in response to phytohormones

During abscission, the AZ cells respond to the abscission signals, such as changing levels of phytohormones e.g. ethylene, jasmonate and auxin, and activate cell wall loosening proteins

(such as cell wall hydrolases). A major component of the cell walls is a network of cellulose microfibrils joined by cross-linking glycans. Mechanical properties of the cell walls are largely dependent upon this network (Whitney et al., 1999). In addition to the cellulose-glycan network,

3 a structurally complex matrix of pectin is also present within the primary cell walls and is abundant in the intercellular matrices of middle lamella. The middle lamellae of AZ cells are presumably loosened by the active mobilization of cell wall hydrolases including β-1, 4- glucanases (Lashbrook et al., 1994) and polygalacturonases that result in the weakening of the cell wall in the abscission zone (Brown, 1997), initiating organ detachment.

Phytohormones in abscission signaling

Ethylene

Ethylene has been widely recognized as a plant hormone that promotes or induces fruit ripening and abscission (Suttle et al., 1991; Abeles et al., 1992). Jackson and Osborne (1970) in their study on Prunus serrulata and Parthenocissus quinquefolia concluded that ethylene was not only responsible for accelerating and inducing abscission, but was also an essential regulator of abscission Gomez-cadenes et al. (2000) and Tudela and Primo-Millo (1992) reported a significant increase in the endogenous levels of ethylene, during abscission in reproductive and vegetative organs, under natural and stressed (carbohydrate and water) conditions. Burns et al.

(2002) reported the use of ethylene to promote fruit loosening to facilitate and coordinate mechanical harvesting of citrus fruit. In citrus leaf explants ethylene can induce as much as

100% leaf explant abscission within 72 h whereas air treated leaf explants takes more than 240 h

(Agustί et al., 2008). In rabbiteye blueberry, pre-harvest application of ethephon, accelerated fruit drop (Dekazos, 1976). A similar study by Ban et al. (2006) in rabbiteye blueberry also reported enhanced fruit drop.

Genes related to ethylene biosynthesis, signaling and perception are differentially expressed during abscission. Ethylene biosynthesis related genes such as S-ADENOSYL-L-

METHIONINE (SAM) SYNTHASE and 1-AMINOCYCLOPROPANE-1-CARBOXYLATE (ACC)

4 SYNTHASE are up-regulated in citrus during ethylene induced leaf abscission (Agustί et al.,

2008). They reported the differential expression of various ethylene related genes in the petiole and the leaf AZ. In a recent study, the expression profiles of genes related to ethylene biosynthesis, perception and cell wall degradation in fruit abscission and ripening of apple were studied by Li et al. (2010). They reported an increase in transcript levels of MdACS5A,

MdACO1, and ethylene receptors, MdETR2, and MdERS2, in the fruit abscission zone over the period of 3 months of fruit ripening.

Defects in ethylene signaling and perception pathways have been reported to affect abscission in various ways. In ethylene insensitive mutants such as etr1 and ein2, both floral organ abscission and senescence is delayed (Patterson and Bleecker, 2004). Overexpression of the ethylene biosynthesis gene, ACC SYNTHASE, in tomato plants resulted in premature flower abscission, while a delay in abscission was noted in never ripe tomato (mutant of the ethylene receptor

LeETR3) and in transgenic plants expressing an antisense version of the ethylene receptor

LeETR1 (Lanahan et al., 1994; Whitelaw et al., 2002).

Auxin

Auxin has been widely known to delay abscission (Taylor and Whitelaw, 2001).

Exogenous auxin was effective in delaying fruit abscission in apple (Yuan and Carbaugh, 2007;

Dal Cin et al., 2008). Auxin seems to be effective in delaying abscission only prior to the initiation of the process (Eo and Lee, 2009). Agustί et al. (2008) reported higher expression of auxin related genes like AUXIN-REPRESSED/DORMANCY-ASSOCIATED PROTEIN during ethylene-induced abscission in citrus leaf explants. Although the exact role of auxin in process of abscission is still unclear, cross talk between auxin and ethylene is considered to be one of the

5 most important signaling steps during the initiation of abscission. The AZ becomes receptive to ethylene only upon lowered auxin levels.

Jasmonates

Jasmonates are a class of phytohormones which are widely associated with regulation of pathogen responses. Many studies have shown the involvement of jasmonic acid (JA) and methyl jasmonate (Meja), the primary jasmonates, in regulating plant development and senescence.

Some evidence links jasmonates to the regulation of abscission. Curtis (1984) first reported the induction of petiole abscission in de-bladed explants of bean (Vigna radiata L) after treatment with methyl jasmonate. Ueda et al. (1992) showed that JA and Meja promoted petiole abscission in beans at concentrations higher than 10 µM. Abscission induced by Meja was not mediated by an increase in ethylene production (Ueda et al., 1996). This study also reported that the Meja treatment did not enhance ethylene production in bean petiole explants. Hartmond et al. (2000) reported that Meja at concentration 10, 20 and 100 mM resulted in significant loosening of citrus, but concentrations higher than 10 mM also resulted in excessive leaf abscission. Meja at concentration of 8-10 mM has been reported to result in abscission of grape berries within 48 h

(Fidelibus and Cathline, 2009). It is still unclear if the Meja and ethylene regulate abscission through a common signaling pathway.

Interaction among Phytohormones

The interaction among multiple phytohormones such as ethylene and auxin plays an important regulatory role in abscission. The AZ becomes sensitive to ethylene only when the auxin concentration goes below a critical level. Rasori et al. (2003) reported that the down regulation of PpAux/IAA2 is paralleled by an up-regulation of Pp-ACO1 in peach during the process of ripening and initiation of ripening. In apple, down regulation of Aux/IAA7 was

6 accompanied with decreased levels of auxin and enhanced fruitlet shedding (Costa et al., 2006).

Meir et al. (2010) reported that the process of abscission is initiated upon changes in auxin gradient across the abscission zone and is elicited by ethylene in the tomato flower abscission zone, indicating a cross-talk of auxin-ethylene signaling. In another study by Meir et al. (2006) in

Mirablis jalapa, a relation between auxin depletion and increase in ethylene sensitivity was observed.

Sanieweski et al. (2000) reported the formation of secondary abscission zone in the stem of Bryophyllum calycinum upon treatment with Meja, and the formation of secondary abscission zone induced by Meja was prevented upon treatment with Indole acetic acid (IAA), an auxin.

Although, 1-methylcyclopropene (1-MCP), an ethylene perception inhibitor, blocks the effect and expression of genes stimulated by Meja-induced abscission (Beno-Moualem et al., 2004), interaction between ethylene and jasmonates remains poorly understood.

3. Cell wall loosening

Many studies reported that ethylene and other phytohormones are involved in inducing cell wall- degrading enzymes such as endo-β-1, 4-glucanase (EG), polygalacturonases (PG), cellulases, and

β–galactosidases, as well as lipid-modifying phospholipases in abscission zones (Goren, 1993;

Malladi and Burns, 2008). An increase in abscission was related to an increase in EG activity in fruit crops such as apple (Pandita and Jindal, 1991), avocado (Tonutti et al., 1997), peach (Rascio et al., 1985) and raspberry (Sexton et al., 1997). A very diverse gene family encodes for EGs in tomato (Brummell et al., 1997, del Campillo and Bennet, 1996). Two members of this family,

CEL1 and CEL2, exhibit an increase in transcript abundance during fruit ripening and flower abscission (Lashbrook et al., 1994). Increase in the transcript levels of MdPG2, and MdEG1 in the fruit AZ has been found to be concomitant with the levels of transcript levels of MdACS5A,

7 MdACO1, MdETR2, and MdERS2 in apple over the period of ripening to abscission (Li et al,.

2010). In cherry tomato, expression of the EG and cellulase (ENDO-1, 4-β-GLUCANASES,

CEL1, CEL2, CEL3, CEL5, CEL7 and CEL8) in the abscission zone was enhanced during abscission of the fruit upon treatment with ethylene and Meja. 1-MCP blocked the expression of these genes in both cases of treatment with ethylene and Meja (Beno-Moualem et al., 2004).

Cellulase activities in the pulvinus and petiole of 10-day-old bean seedlings were enhanced by

JA treatment (Ueda et al., 1996). The total amounts of cellulosic polysaccharides in this region were reduced significantly by the addition of Meja in the light. Deng et al. (2006) reported that abscission in table grapes was correlated with an increase in the activity of cellulase and polygalacturonase, in abscission zones. In another study, conducted by Mishra et al. (2007) in cotton leaf, a several-fold increase in the activity of cellulase and polygalacturonase was observed during ethylene induced abscission in the leaf abscission zone. In the same study, treatment with 1-MCP prior to ethylene strongly suppressed the activity of cellulase and PGs.

It has been reported that the activity of PGs increase significantly in AZs prior to, as well as during the abscission process (Coupe et al., 1995). Roberts et al., (2002) reported delayed floral organ abscission in Arabidopsis thaliana mutants defective in abscission-related polygalacturonase. Silencing of polygalacturonase can result in increase in break strength for petiole abscission, whereas silencing of LeCEL1 and LeCEL2 (cellulase) and LeEXP11 and

LeEXP12 (expansins) had no discernible effect on the break strength in tomato, even when two of these genes were silenced concurrently, indicating the important role of PG in abscission

(Jiang et al. 2008).

Expansins are a class of proteins which posses the unique ability to induce cell-wall extension without hydrolytic breakdown of the major structural components of the cell wall (Cosgrove,

8 1999). Cho and Cosgrove (2000) reported the role of expansins in cell-wall loosening during a variety of developmental processes including abscission in Arabidopsis thaliana. In ethylene promoted leaf abscission up to a seven fold increase in expansin expression can occur in cells undergoing separation, indicating an important role of expansins in ethylene promoted abscission

(Belfield et al., 2004).

DNA binding with one finger (DOF) is a family of transcription factors. In Arabidopsis, a member of this family DOF4.7 is expressed in floral abscission zones. Arabidopsis AtDOF4.7 lines which display constitutive expression of AtDOF4.7, exhibit an ethylene-independent deficiency in floral organ abscission (Wei et al., 2010). Overexpression of AtDOF4.7 results in abscission deficiency in floral organ in Arabidopsis by inhibiting dissolution of middle lamella in

AZ cells indicating that it controls abscission by regulating enzymes related to cell wall hydrolysis (Wei et al., 2010). In Arabidopsis, INFLORESCENCE DEFICIENT IN ABSCISSION

(IDA) is another gene family which is also known to regulate abscission through an ethylene insensitive pathway (Butenko et al., 2003).

The fundamental understanding of genes involved in abscission, genetic control of abscission, and its relation to plant growth regulators can help us in enhancing and developing tools for mechanical harvesting of blueberries. Therefore, the aim of this research is to understand the fruit detachment in blueberries. The specific objectives of my research are:

1. To identify the natural abscission zone of mature blueberry fruit 2. To identify, characterize and understand the alterations occurring in cell wall carbohydrates during the induction of abscission 3. To identify and characterize genes altered in expression upon induction of fruit abscission 4. To study the variability in ease-of-fruit detachment in blueberry

9 References Abeles FB, Morgan PW, Saltveit ME. 1992. Ethylene in Plant Biology (second ed.), Academic Press Inc, San Diego, CA .182–221.

Agustí J, Merelo P, Cercós M, Tadeo FR, Talón M. 2008. Ethylene-induced differential gene expression during abscission of citrus leaves. Journal of Experimental Botany 59, 2717-2733.

Austin ME, and Williamson RE. 1977. Comparison of harvest methods of rabbiteye blueberries. Journal of American Society of Horticultural Sciences 102,454-456.

Ban T, Kugishima M, Ogata T, Shiozaki S, Horiuchi S, Ueda H. 2007. Effect of ethephon (2- chloroethylphosphonic acid) on the fruit ripening characters of rabbiteye blueberry. Scientia Horticulturae 112, 278–281.

Belfield EJ, Ruperti B, Roberts JA, McQueen-Mason SJ. 2005.Changes in expansin activity and gene expression during ethylene-promoted leaflet abscission in Sambucus nigra. Journal of Experimental Botany 56, 817-823.

Beno-Moualem D, Gusev L, Dvir O, Pesis E, Meir S, Lichter A. 2004. The effects of ethylene, methyl jasmonate and 1-MCP on abscission of cherry tomatoes from the bunch and expression of endo-1, 4-β-glucanases. Plant Scencei 167 , 499-507.

Brown GK, Shulte NL, Timm EJ, Beaudrv RM, Peterson DL, Han-cock JF, Takeda F. 1996. Estimates of mechanization effects on fresh blueberry quality. Applied Engineering in Agriculture 12, 21-26.

Brown KM. 1997. Ethylene and abscission. Physiologia Plantarum 100, 567-576

Brummell DA, Bird CR, Schuch W, Bennett AA. 1997. An endo-1, 4-βglucanase expressed at high levels in rapidly expanding tissues. Plant Molecular Biology 33, 87-95. Burns JK. 2002. Using molecular biology tools to identify abscission materials for citrus. Hortscience 37, 459-464.

Butenko MA, Patterson SE, Grini PE, Stenvik GE, Amundsen SS, Mandal A, Aalen, RB. 2003. INFLORESCENCE DEFICIENT IN ABSCISSION controls floral organ abscission in Arabidopsis and identifies a novel family of putative ligands in plants. Plant Cell 15, 2296–2307.

Cho HT, Cosgrove DJ. 2000. Altered expression of expansin modulates leaf growth and pedicel abscission in Arabidopsis thaliana. Proceedings of National Academy of Science 97, 9783-9788.

Clements J, Atkins C. 2001. Characterization of a non-abscission mutant in Lupinus angustifolius. I. genetic and structural aspects. American Journal of Botany. 88, 31–42.

Cosgrove DJ.1999. Annual Review of Plant Physiology. Plant Molecular Biology 50, 391– 417.

10 Costa G, Dal Cin V, Ramina A. 2006.Physiological, molecular and practical aspects of fruit abscission. In: Webset, A.D. and H.Ramirez (eds.). Proc. Xth IS on Plant Bioregulators in Fruit. Acta Hort. 727. ISHS.

Coupe SA, Taylor JE, Roberts JA. 1995. Characterization of a messenger-RNA encoding a metallothionein-like protein that accumulates during ethylene-promoted abscission of Sambucus nigra L. leaflets. Planta 197, 442–447.

Curtis R.W. 1984. Abscission-inducing properties of methyl jasmonate, ABA, and ABA- methylester and their interactions with ethephon, AgNO3, and malformin. Journal of Plant Growth Regulators 3, 157–168.

Dal Cin V, Danesin M, Botton A, Boschetti A, Dorigoni A and Ramina A. 2008. Ethylene and preharvest drop: the effect of AVG and NAA on fruit abscission in apple (Malus domestica L. Borkh).Plant Growth Regulator. 56, 317–325.

Dekazos ED. 1976. Effects of preharvest applications of Ethephon and SADH on ripening, firmness and storage quality of rabbiteye blueberries ('T-19'). Proceedings of Florida State Horticulture. Set. 89:266-270. del Campillo E, Bennett AB.1996. Pedicel breakstrength and cellulase gene expression during tomato flower abscission. Plant Physiology 111, 813–820.

Deng Y, Wuc Y, Li Y, Yang M, Zheng CSC. 2007. Studies of postharvest berry abscission of ‘Kyoho’ table grapes during cold storage and high oxygen atmospheres. Postharvest Biology and Technology 43, 95-101.

Eo J, Lee BY .2009. Effects of ethylene, abscisic acid and auxin on fruit abscission in water dropwort (Oenanthe stolonifera DC.). Scientia Horticulturae 123, 224–227.

Fidelibus M, Cathline K. 2010. Dose and time dependent effects of Methyl Jasmonate on abscission of grapes. Acta Horticulturae (ISHS) 884,725-728.

Gomez-Cadenas A, Mehouachi J, Tadeo FR, Primo-Millo E, Talon M. 2000. Hormonal regulation of fruitlet abscission induced by carbohydrate shortage in citrus. Planta 210, 636-643.

Goren R. 1993. Anatomical, physiological and hormonal aspects of abscission in citrus. Horticultural Reviews 145-182.

Gough RE and Litke W. 1980. An anatomical and morphological study of abscission in highbush blueberry fruit. Journal of American Society of Horticultural Science 105, 335-341.

Hartmond U, Yuan R, Burns JK, Grant A, Kender WJ. 2000. Citrus fruit abscission induced by methyl-jasmonate. Journal of American Society of Horticultural Science 125, 547-552.

11 Howell GS, Stergios BG, Stackhouse SS, Bittenbender HC, Burton CL. 1976. Ethephon as a mechanical harvesting aid for highbush blueberries (Vaccinium austral Small). Journal of American Society of Horticultural Science 101, 111-115.

Jackson MB, Osborne DJ. 1970. Ethylene, the natural regulator of leaf abscission. Nature 225, 1019-1022.

Jiang CZ, Lu F, Imsabai, W, Meir S, Reid MS. 2008. Silencing polygalacturonase expression inhibits tomato petiole abscission. Journal of Experimental Botany 59, 973–979.

Lanahan MB, Yen HC, Giovannoni JJ, Klee HJ. 1994. The Never Ripe mutation blocks ethylene perception in tomato. Plant Cell 6, 521-530.

Lashbrook CC, Gonzalez-Bosch C, Bennett AB. 1994. Two divergent endo-β-1,4-glucanase genes exhibit overlapping expression in ripening fruit and abscising flowers. Plant Cell 6, 1485– 1493

Lashbrook CC, Giovannoni JJ, Hall BD, Fisher RL, Bennett AB. 1998. Transgenic analysis of tomato endo-β-1,4-glucanase gene function: role of Cel1 in floral abscission. Plant Journal 13, 303–310.

Li J, Zhu H, Rongcai Y. 2010. Profiling the expression of genes related to ethylene biosynthesis, ethylene perception, and cell wall degradation during fruit abscission and fruit ripening in Apple. Journal of American Society of Horticultural Science 135(5), 391-401.

Mainland CM, Kushman LJ, Ballinger WE. 1975. The effect of mechanical harvesting on yield, quality of fruit and bush damage of highbush blueberry. Journal of American Society of Horticultural Science 100, 129-134.

Mainland CM, Ballinger WE, Kushman LJ. 1971. Recovery firmness, and shelf-life of hand and machine harvested highbush blueberries. Hortscience 6,49.

Malladi A, Burns JK. 2008. CsPLDα1 and CsPLDγ1 are differentially induced during leaf and fruit abscission and diurnally regulated in Citrus sinensis. Journal of Experimental Botany 59, 3729-3739.

Mao L, Begum D, Chuang H, Budiman M, Szymkowiak E, Irish E, Wing R. (2000). JOINTLESS is a MADS-box gene controlling flower abscission zone development. Nature 406, 910–913.

Meir S, Hunter DA, Chen JC, Halaly V, Reid MS. 2006. Molecular changes occurring during acquisition of abscission competence following auxin depletion in Mirabilis jalapa. Plant Physiology 141, 1604–1616.

12 Meir S, Philosph-Hadas S, Sundaresan S, Selvara VKS, Burd S, Ophor R, et al. 2010.Microarray Analysis of the Abscission-Related Transcriptome in the Tomato Flower Abscission Zone in Response to Auxin Depletion. Plant Physiology 154 (4), 1929-1956.

Mishra A, Khare S, Trivedi PK, Nath P. 2007. Effect of ethylene, 1-MCP, ABA and IAA on break strength, cellulase and polygalacturonase activities during cotton leaf abscission. South African Journal of Botany 74, 282-287.

Pandita VK, Jindal KK. 1991. Enzymatic and anatomical changes in abscission zone cells of apple fruits induced by ethephon. Plant Biology 33, 20–25.

Pattathil S, Avci U, Baldwin D, Swennes AG, McGill JA, Popper Z, Bootten T, Albert A, Davis RH, Chennareddy C, et al.2010. A comprehensive toolkit of plant cell wall glycan-directed monoclonal antibodies. Plant Physiology 153, 514–525.

Patterson SE. 2001. Cutting loose. Abscission and dehiscence in Arabidopsis. Plant Physiology 126, 494-500.

Patterson SE, Bleecker A B. 2004. Ethylene-dependent and -independent processes associated with floral organ abscission in Arabidopsis. Plant Physiology 134,194 -203.

Patterson S, Butenko M, Kim J. 2007. Ethylene responses in abscission and other processes of cell separation in Arabidopsis. In Advances in Plant Ethylene Research (ed. A. Ramina, C. Chang, J. Giovannoni, H. Klee, P. Perata and E. Woltering), 271-278. Netherlands: Springer.

Rascio N, Casadoro G, Ramina A, Masia A. 1985. Structural and biochemical aspect of peach fruit abscission (Prunuspersica L. Batsch). Planta 164, 1–11.

Rasori A, Bertolasi B, Furini A, Bonghi C, Tonutti P, Ramina A. 2003.Functional analysis of peach ACC-oxidase promoters in transgenic tomatoes and ripening peach fruit. Plant Science 165, 523-530.

Roberts JA, Whitelaw CA, Gonzalez-Carranza ZH, McManus MT. 2000. Cell separation processes in plants – models, mechanisms and manipulation. Annals of botany 86, 223–235.

Roberts JA, Elliott KA, Gonzalez-Carranza ZH. 2002. Abscission, dehiscence, and other cell separation processes. Annual review of Plant Biology 53, 131–158.

Safley CD, WO Cline, Mainland GM. 2005. Estimated costs ofproducing, harvesting, and marketing blueberries in the southeastern United States. Proceedings. 12th Biennial Southeast Blueberry Conference, Savannah, GA, 33-49.

Saniewski M, Ueda J, Miyamoto K. 2000. Methyl jasmonate induces the formation of secondary abscission zone in stem of Bryophyllum calycinum Salisb. Acta Physiology of Plant 22, 17–23.

13 Sexton R. 1976. Some ultrastructural observations on the nature of foliar abscission in Impatiens sultani. Planta 128, 49-58.

Sexton R, Roberts JA. 1982. Cell biology of abscission. Annual Review of Plant Physiology 33, 133- 162. Sexton R, Palmer JM, Whyte NA, Littlejhons S.1997. Cellulase, fruit softening and abscission in red raspberry Rubus idaeus L. cv Glen Clova. Annals of Botany 80, 371–376.

Suttle JC, Hultstrand. 1991. Ethylene-induced leaf abscission in cotton seedlings. The physiological bases for age-dependent differences in sensitivity, Plant Physiology 95, 29–33.

Taylor JE, Whitelaw CA. 2001. Signals in abscission. New Phytologist 151, 323-339.

Tonutti P, Bonghi C, Ruperti B, Ramina A. 1997. The modulation of ethylene biosynthesis and ACC oxidase gene expression during peach fruit development and fruitlet abscission. In: Kanellis AK, Chang C, Kende H, Grierson D (eds) Biology and Biotechnology of the Plant Hormone Ethylene. NATO ASI Series, Vol 3/34. Dordrecht/Boston/London: Kluwer Academic Publisher, 149–153.

Tudela D, Primo-Millo E. 1992. 1-Aminocyclopropane-1-carboxylic acid transported from roots to shoots promotes leaf abscission in Cleopatra mandarin (Citrus unshiu Hort. ex Tan.) seedlings rehydrated after water stress. Plant Physiology 100,131-137.

Ueda J, Miyamoto K, Momotani Y, Kato J, Kamisaka S. 1992. Promotive effect of jasmonic acid on the abscission of bean petiole explants in relation to the metabolism of cell wall polysaccharide. In: Masuda Y (ed) Plant Cell Wall as Biopolymers with Physiological Functions. Yamada Science Foundation, Osaka, 307–312.

Ueda J, Miyamoto K, Hashimoto M. 1996. Jasmonates promote abscission in bean petiole explants: Its relationship to the metabolism of cell wall polysaccharides and cellulase activity. Journal of Plant Growth Regulators 15, 189–195.

USDA-NASS Noncitrus Fruits and Nuts Summary 2009. http://usda.mannlib.cornell.edu/usda/nass/NoncFruiNu//2010s/2010/NoncFruiNu-07-07- 2010.pdf

Uheda E, Nakamura S. 2000. Abscission of Azolla branches induced by ethylene and sodium azide. Plant Cell Physiology 41, 1365–1372.

Vashisth T, Johnson LK, Malladi A, 2011. An efficient RNA extraction protocol and identification of reference genes for normalization of gene expression blueberry. Submittted to Plant Cell Reports.

14 Wei PC, Tan F, Gao XQ, Zhang XQ, Wang GQ, Xu H, et al.2010.Overexpression of AtDOF4.7, an Arabidopsis DOF Family Transcription Factor, Induces Floral Organ Abscission Deficiency in Arabidopsis. Plant Physiology 153, 1031-1045.

Whitelaw CA, Lyssenko NN, Chen L, D. Zhou, A.K. Mattoo, Tucker ML. 2002. Delayed abscission and shorter internodes correlated with a reduction in the ethylene receptor LeETR1 transcript in transgenic tomato. Plant Physiology 128,978-987

Whitney S E C , Gothard M G E, Mitchell JT, Gidley M J. 1999. Roles of cellulose and xyloglucan in determining the mechanical properties of primary plant cell walls. American Society of Plant Physiologists 121,657-664. www.agmrc.org

Yarborough DE. 1992. Blueberry Enterprise Budget. Wild Blueberry Fact Sheet No. 260. University of Maine Cooperative Extension, Orono, Maine, USA. 3. Yarborough DE. 1997. Production Trends in the Wild Blueberry Industry in North America. Acta Horticulturae. 446:33-35.

Yuan RC, Carbaugh DH,.2007 Effects of NAA, AVG, and 1-MCP on ethylene biosynthesis, preharvest fruit drop, fruit maturity, and quality of ‘Golden supreme’ and ‘Golden delicious’ apples. Hortscience 42, 101–105.

15 CHAPTER 2

FRUIT DETACHMENT IN RABBITEYE BLUEBERRY: ABSCISSION AND PHYSICAL

SEPARATION

Vashisth, T. and A. Malladi. 2013. Journal of ASHS. 138: 95-101

Reprinted here with permission of publisher

16 Abstract

A better understanding of fruit detachment and the processes mediating it is essential to improve the efficiency of mechanical harvesting in blueberry (Vaccinium sp.). In blueberry, fruit detachment may occur either at the point of attachment of the pedicel to the peduncle (peduncle- pedicel junction, PPJ) or at the point of attachment of the pedicel to the fruit (fruit-pedicel junction, FPJ). The fruit detachment responses of the PPJ and the FPJ to different conditions are not entirely clear. Additionally, whether fruit detachment at these junctions is mediated by the physiological process of abscission or through physical separation of the organ from the parent plant is not well understood. In this study, a series of experiments were performed to determine the abscission zone (AZ) corresponding to the point of mature fruit detachment and to determine whether fruit detachment occurs due to abscission or physical separation in rabbiteye blueberry

(V. ashei). Anatomical studies indicated the presence of an AZ at the PPJ. Greater than 92% of the natural detachment of mature fruit occurred at the PPJ. The morphology of the fracture plane at the PPJ in naturally detached fruit was even and uniform, consistent with fruit detachment through abscission at this location. Abscission agents such as methyl jasmonate (20 mM) and ethephon (2-chloroethylphosphonic acid, 1000 mg·L-1) enhanced the extent of fruit detachment at the PPJ, further indicating that mature fruit detachment through abscission occurred primarily at this location. Additionally, the fracture plane at the PPJ during fruit detachment in response to abscission agent applications was flattened and even, further supporting the conclusion that fruit detachment at this location occurred through abscission. In contrast, the majority of the fruit detachment in response to mechanical shaking occurred at the FPJ. Analysis of the morphology of the fracture plane at the FPJ during detachment in response to mechanical shaking indicated that fruit detachment at this location was associated with extensive tearing and mechanical

17 disruption of cells, consistent with physical separation. Together, data from this study indicate that mature fruit detachment due to abscission occurs primarily at the PPJ while fruit detachment during mechanical shaking occurs primarily at the FPJ due to physical breakage at this weak junction.

Index Words. abscission agent, abscission zone, mechanical harvesting, plant growth regulators,

Vaccinium sp .

Introduction

Blueberry, a fruit crop indigenous to the United States, was valued at over $780 million in 2011

(USDA, 2012). In spite of its emerging importance, various aspects of blueberry physiology including the process of fruit detachment are not well understood. Enhancing our knowledge of the process of fruit detachment can greatly aid in developing methods to increase the efficiency of mechanical harvesting, which can in turn reduce the costs associated with blueberry production (Austin and Williamson 1977; Howell et al., 1976; Mainland et al. 1975; Takeda et al., 2008).

Fruit detachment may occur either through the physiological process of abscission or through tearing and physical separation of the fruit from the parent plant. Abscission involves the programmed separation of entire organs in response to plant developmental cues, and various types of biotic and abiotic stress. Abscission occurs at specific regions within the plant termed as abscission zones (AZs; Roberts et al., 2000; 2002; Sexton and Roberts, 1982). The cells in the

AZ are morphologically and biochemically distinct prior to the initiation of the process of separation. The AZs generally consist of one to many layers of small, rounded, densely cytoplasmic cells at the union of the organ destined for detachment and the main body of the

18 plant (Goren, 1993; Roberts et al., 2002; Sexton and Roberts, 1982). The AZ cells generally contain large deposits of starch and have highly branched plasmodesmata (Sexton and Roberts,

1982). These cells also display several ultra-structural changes during differentiation (Goren

1993; Iwahori and Van Steveninck, 1976; Patterson, 2001; Roberts et al., 2002; Stösser et al.,

1969; Webster, 1968). The progression of abscission at the pre-formed AZs involves multiple phases such as a gain in competence to respond to abscission signals, activation of the AZ resulting in the initiation of cell separation in response to the abscission signals, and cell separation followed by formation of a protective layer (Patterson, 2001).

Fruit crops may possess multiple AZs where fruit detachment can occur, often depending on the maturity of the separating organ. For example, in citrus fruit two AZs are associated with the fruit: 1. the shoot-peduncle AZ (AZ-A), and 2. the peduncle-fruit AZ (AZ-C) (Goren, 1993).

While the AZ-A is active during early fruit development, the AZ-C is the active zone associated with mature fruit detachment (Goren, 1993; Kazokas and Burns, 1998). Similarly, in the sour cherry (Prunus cerasus) two AZs have been described, one at the junction of the pedicel and the peduncle, and another at the union of the pedicel and the fruit (Stösser et al., 1969; Wittenbach and Bukovac, 1974). Mature sour cherry fruit abscission occurs primarily at the AZ located at the junction of the pedicel and the fruit (Stösser et al., 1969; Wittenbach and Bukovac, 1974).

Also in sweet cherry (P. avium), immature fruit detachment occurs at the pedicel-peduncle AZ while mature fruit detachment occurs at the receptacle-fruit AZ (Wittenbach and Bukovac,

1972).

Plant organs can also separate from the plant as a result of mechanical breakage and cell disruption within weak points of attachment in response to physical force instead of abscission.

During hand or mechanical harvesting, the fruit may be separated from the plant due to cell

19 rupture and breakage at the AZ or at other physically weak points of attachment of the fruit to the plant. For example, mechanical harvesting in sweet cherry can result in mature fruit detachment between the pedicel and the peduncle (Norton et al., 1962) whereas mature fruit normally abscise at the receptacle-fruit AZ (Wittenbach and Bukovac, 1972).

Fruit detachment in blueberry can occur at the point of attachment of the pedicel to the peduncle or at the point of attachment of the pedicel to the berry. Gough and Litke (1980) reported the presence of an AZ at the point of attachment of the pedicel to the berry and indicated that mature fruit detachment in northern highbush blueberry (V. corymbosum) occurs primarily at this location. This junction is referred to hereafter as the fruit-pedicel junction (FPJ).

Hand harvesting and the majority of mechanical harvesting of blueberry cultivars typically result in fruit separation at the FPJ (Howell et al., 1976; Takeda et al., 2008). Recent studies indicate that fruit drop in response to the application of abscission agents such as ethephon (2- chloroethylphosphonic acid) and methyl jasmonate (MeJa) occurs primarily at the point of attachment of the pedicel to the peduncle in different types of blueberries (Malladi et al., 2012).

This junction is referred to hereafter as the peduncle-pedicel junction (PPJ). The point of natural detachment of mature fruit is not well understood. Additionally, it is unclear whether fruit separation at these junctions occurs as a result of abscission or due to physical tissue disruption.

A better understanding of the point of fruit detachment and the processes mediating it in blueberry is essential to improve the efficiency of mechanical harvesting. Such information can aid in determining the applicability of abscission agents as harvest aids, and in the breeding and selection of genotypes better suited for mechanical harvesting. Hence, the main objectives of this study were to determine the point of mature fruit detachment in blueberry, and to determine if fruit detachment occurs through the physiological process of abscission or through physical

20 separation. To achieve these objectives, the anatomy of the points of fruit detachment following natural fruit detachment, fruit detachment in response to abscission agents, and fruit detachment in response to mechanical shaking were investigated in rabbiteye blueberry.

Materials and Methods

Anatomy of the PPJ and the FPJ. To study the anatomy of the potential points of fruit detachment, the PPJ and the FPJ tissues from mature (blue/black) fruit ready for harvest were collected from 6-year-old ‘Powderblue’ rabbiteye blueberry plants (n = 4) grown at the

University of Georgia, Horticulture Farm, Watkinsville, GA in 2011. The fruit were collected along with the fruiting branch and the PPJ and the FPJ tissues were dissected. The tissues were fixed in FAA (50% ethyl alcohol, 10% formalin, 5% acetic acid) and stored at room temperature until further analysis. Two to three PPJ and FPJ samples from each replicate plant were used in this study. The samples were removed from the FAA solution and rinsed thoroughly in de- ionized water. The samples were dehydrated through a graded ethanol series. Paraffin infiltration and embedding was performed as described in Ruzin (1999) with the modification of using histoclear (Electron Microscopy Science, Hatsfield, PA) during the infiltration process.

Longitudinal sections (10 μm) were collected using a rotary microtome. After removal of paraffin from the sections using histoclear, the sections were rehydrated and stained with 1% aqueous toluidine blue. Images of the sections were captured using a light microscope (BX51;

Olympus, Center Valley, PA) fitted with a digital camera (DP70; Olympus).

Natural fruit detachment. Mature (4- to 6-year-old) rabbiteye blueberry plants of ‘Premier’ and

‘Briteblue’ grown at the University of Georgia, Horticulture Farm, Watkinsville, GA were used in this study. This study was performed in 2011 and in 2012 with both of the genotypes (n = 8 plants for ‘Premier’ in 2011 and n = 4 plants in all other experiments). A 1 m2 region (square)

21 was marked underneath each plant and all fruit debris were removed from this region. The number of mature fruit that naturally detached from the plant and remained within the marked area was counted at regular intervals. From among these fruit, the proportion that detached along with the pedicel (i.e., at the PPJ) was determined. After each observation, the detached fruit were removed from within the marked area. Data collection was performed for 2-3 weeks in each growing season. Cumulative fruit detachment at the PPJ over this period is presented here.

In 2012, the morphology of the fracture plane in naturally detached fruit was studied in

‘Briteblue.’ The area under the bushes was cleared and the bushes were manually monitored over a period of several hours until at least two mature fruit per replicate plant (n = 4) naturally detached from the plant. These fruit were collected immediately after separation from the plant.

All of the detached fruit separated at the PPJ. The pedicel from the detached fruit was immediately separated from the berry, fixed in 5% glutaraldehyde:0.1 M potassium phosphate buffer (1:1), and stored at 4 °C until further analysis using scanning electron microscopy (SEM).

The pedicel end of the PPJ (pedicel part detached from the peduncle) was observed in this analysis. Samples were processed for SEM according to Mims (1981). Briefly, the samples were cut into approximately 1-2 mm upright sections and were rinsed three times with the fixative buffer, 5% glutaraldehyde:0.1 M potassium phosphate buffer (1:1) for 15 min. The samples were immediately treated with osmium tetroxide for 2 h and rinsed. The rinsed samples were dehydrated through a graded ethanol series. For critical point drying, the ethanol was replaced by liquid CO2 which was brought to the critical point in the Autosamdri-814 Critical Point Dryer

(Tousimis Research Corporation, Rockville, MD). Subsequently, samples mounted on aluminum stubs were coated with gold in the SPI-Module Sputter Coater and Carbon-Coater (SPI Supplies

22 /Structure Probe, Inc. West Chester, PA). Samples were observed using the Zeiss 1450EP scanning electron microscope (Carl Zeiss MicroImaging Inc., Thornwood, NY).

Fruit detachment in response to abscission agent applications and mechanical shaking. Mature

(5-year-old) rabbiteye blueberry plants of ‘Briteblue’ grown at the University of Georgia,

Horticulture Farm, Watkinsville, GA were used in this study in 2011. The experiment was designed as a completely randomized design with three treatments and three replicates

(individual plants). The treatments were: 1) control; 2) methyl jasmonate (MeJa, 20 mM; Sigma-

Aldrich, St. Louis, MO); and 3) ethephon (1000 mg·L-1; Bayer CropScience, Kansas City, MO).

These abscission agent treatments have been previously shown to induce rapid fruit detachment in multiple rabbiteye blueberry genotypes (Malladi et al., 2012). The treatments were performed when the majority (~75%) of the fruit on the plant was mature. All treatments, including the control, were applied along with 0.15% of the adjuvant (Latron B-1956, Rohm and Haas,

Philadelphia, PA) and were performed using a hand pump sprayer until run-off around 0830 HR.

The average daily temperature on the day of application was 27 °C. At 24 h after treatment, two branches per plant with around 50 fruit each were shaken using a hand-held mechanical shaker

(described in Malladi et al., In press) until all of the berries were detached. While MeJa and

Ethephon treated plants required 3 and 3.5 s of mechanical shaking, respectively, 14 s of shaking was required for fruit detachment from the control plants (Malladi et al., In press). The immature fruit were removed prior to mechanical shaking. The detached berries were captured in a catch frame and the proportion of fruit that detached along with the pedicel at the PPJ was determined.

An additional branch on each plant was tagged and the fruit number was counted to determine the extent of drop as a result of the treatments. After 72 h of treatment 13% fruit drop was

23 observed in the control while MeJa and Ethephon treatments resulted in 65% and 66% fruit drop, respectively.

In 2012, mature 7-year-old rabbiteye blueberry plants of ‘Powderblue’ grown at the

University of Georgia, Horticulture Farm, Watkinsville, GA were used to study the morphology of the fracture plane during fruit detachment in response to abscission agent applications (n = 4).

This experiment was designed as a completely randomized design with three treatments: 1)

-1 Control; 2) MeJa (20 mM); and 3) ethephon (1000 mg·L ). All treatments included 0.15% of the adjuvant (Latron B-1956) and were performed using a hand pump sprayer until run-off around

0900 HR. The average daily temperature on the day of application was 28 °C. At 24 h after treatment, the pedicels were separated manually from the peduncle or from the mature fruit to obtain the PPJ and the FPJ fracture planes, respectively. Three samples of the PPJ and the FPJ were collected from each replicate plant. These samples were immediately fixed in 5% glutaraldehyde:0.1 M potassium phosphate buffer (1:1), stored at 4 °C and used for SEM analysis, as described above. The pedicel ends of the PPJ and the FPJ (pedicel part detached from the fruit) were observed in this study.

Fruit detachment in response to mechanical shaking. Mature (6-year-old) rabbiteye blueberry plants of ‘Briteblue’ grown at the University of Georgia, Horticulture Farm, Watkinsville, GA were used to understand fruit detachment in response to mechanical shaking in 2012. One branch per plant with approximately 50 fruit was shaken using a hand-held mechanical shaker (Malladi et al., In press) for 5 s and the detached fruit were captured in a catch frame (n = 3). The proportion of the detached fruit that separated at the PPJ was determined. Additionally, the PPJ tissues were collected from the mature berries that separated with the pedicel attached (collected in a catch frame), while the FPJ tissues were collected from the pedicels that were left on the

24 branch after the detachment of the fruit. Two PPJ and FPJ tissues were dissected and collected from each replicate. These samples were immediately fixed in 5% glutaraldehyde:0.1 M potassium phosphate buffer (1:1) and were stored at 4 °C until further analysis. The pedicel ends of these junctions were used for SEM analysis, as described above, to better understand the mechanism of fruit detachment due to mechanical shaking.

Statistical analysis. The data from the study on fruit detachment in response to abscission agent applications and mechanical shaking were analyzed using analysis of variance (ANOVA) followed by mean separation using Fisher's least significant difference (LSD; α = 0.05). All analyses were performed using SigmaPlot 11 (Systat Software Inc., San Jose, CA). Data collected from two branches per plant were treated as subsamples and averaged within the replicate.

Results and Discussion

Anatomy of the PPJ and the FPJ. At the point of attachment of the pedicel to the peduncle (PPJ), multiple layers of typically small and iso-diametric cells were observed on either side of the vascular tissue (Fig. 1A). These cells were distinct from the surrounding cells within the peduncle and pedicel which were generally larger in size. Indentations were observed on either side of the point of attachment and the cells within this region were also characteristically smaller, especially within the peduncle. AZs are typically characterized by the presence of such multiple layers of small rounded cells which also display intense staining (Roberts et al., 2002;

Sexton and Roberts, 1982). Hence, the above observations are consistent with the presence of an

AZ at the PPJ and suggest that progression of fruit detachment at this location can occur through the physiological process of abscission. However, a line of fracture or cell rupture was not

25 observed within the PPJ samples analyzed here, indicating that the progression of cell separation had not yet been initiated at this region in these samples. Additionally, a zone of cell separation was not apparent across the vascular bundles connecting the pedicel and the peduncle, suggesting that detachment along this region may occur through mechanical fracture.

A clear AZ was not apparent at the FPJ (Fig. 1B). The majority of the cells at this junction appeared to be large and rounded and similar to the neighboring cells. Although an indentation between the pedicel and the fruit was observed towards the periphery, no clear plane of separation was evident in this region. These observations are not consistent with the presence of an AZ at the union of the pedicel and the fruit. However, Gough and Litke (1980) reported the presence of an AZ at the FPJ in northern highbush blueberry. It is possible that these differences are due to differences between the Vaccinium species used in these studies. Anatomical analysis of the FPJ across different species may be required to address the possibility of the presence of an AZ at this location in other Vaccinium species. Such a study may be complemented by the analysis of the localized expression of genes and immuno-histochemical analysis using markers specifically associated with the AZs as has been recently described in tomato (Iwai et al., 2012).

Natural fruit detachment. The majority of natural fruit detachment occurred at the point of attachment of the pedicel to the peduncle, i.e., at the PPJ, in both of the rabbiteye blueberry cultivars (Fig. 2). In ‘Premier,’ 92% and 98% of the berries detached at the PPJ in 2011 and

2012, respectively. In ‘Briteblue,’ 98% of the fruit detached at the PPJ in both years of this study. Similar trend of natural fruit detachment was also observed in ‘Powderblue’ (data not shown). These data indicate that natural detachment of mature fruit occurs at the PPJ across the different rabbiteye blueberry genotypes analyzed.

26 The PPJ was used for SEM analysis of the morphology of the fracture plane as natural fruit detachment occurred primarily at this location. The pedicel end of the PPJ in naturally detached fruit displayed an even fracture plane (Fig. 3). Small and rounded cells were visible at the surface and at the periphery of the fracture plane. These observations are consistent with the involvement of abscission during natural fruit detachment. Cell rupture and breakage, and some tearing of the tissues were observed at the center of fracture plane, primarily within the vascular tissue, suggesting that abscission progressed through the disruption of cells within this region.

Similarly, breakage of cells within the vascular tissue was associated with the final stages of abscission progression in cherry (Stösser et al., 1969; Wittenbach and Bukovac, 1972). Also,

Gough and Litke (1980) reported that final fruit separation in blueberry occurs by rupture and mechanical tearing of the vascular bundles, although these authors indicated that fruit detachment occurred at the FPJ in northern highbush blueberry. Together, the natural fruit detachment data and the SEM analysis support the involvement of the physiological process of abscission during natural fruit detachment in the rabbiteye blueberry cultivars studied.

Fruit detachment in response to abscission agent applications and mechanical shaking. The majority of the fruit in the control treatment detached at the FPJ after mechanical shaking (Fig.

4), in contrast to the point where the majority of natural fruit detachment occurred. While some of this difference may be due to the different modes of fruit detachment, it may also be likely that fruit that naturally detach at the PPJ may be primed for detachment owing to a slightly advanced degree of fruit maturity. Mechanical shaking at 24 h after the application of the abscission agents, MeJa and ethephon, clearly enhanced the extent of fruit detachment at the PPJ

(Fig. 4). The applications of MeJa and ethephon resulted in 81% and 38% fruit detachment at the

PPJ, respectively, in response to mechanical shaking. Similarly, Malladi et al. (2012) reported

27 that the majority of the fruit detachment in response to MeJa and ethephon treatments occurred at the PPJ in multiple rabbiteye blueberry genotypes as well as in southern highbush blueberry

(hybrids largely derived from V. corymbosum and V. darrowi). Together, these data strongly indicate that fruit detachment in response to abscission agent applications were consistent across different cultivars and species of blueberry, and occurred primarily at the PPJ. The difference in the extent of fruit detachment at the PPJ between MeJa and ethephon may be a reflection of the time taken for peak fruit loosening effects of these agents. The applications of MeJa generally resulted in high fruit detachment within 24 h while fruit drop in response to ethephon applications was generally high at around 48 h after treatment (Malladi et al., 2012). In earlier studies, pre-harvest treatment of ethephon followed by mechanical harvesting of highbush blueberry was reported to result in less than 5% stemmy fruit (Howell et al., 1976), in contrast to the observations described here and the study of Malladi et al. (2012). In the study by Howell et al. (1976), observations on fruit detachment were performed at approximately one week after treatment, while the observations in the current study were performed at 24 h after treatment. It may be likely that fruit detachment at later stages (1 week after treatment) does not reflect the direct effects of ethephon on abscission. Alternatively, the differences among these studies may be attributed to the different blueberry genotypes used.

The morphology of the fracture planes on the pedicel ends of the PPJ and the FPJ from the control, MeJa and ethephon treated samples were observed after manual separation of the pedicel from the peduncle and the fruit, respectively (Fig. 5). The fracture plane of the pedicel end of the PPJ in the control treatment was uneven and consisted of a large portion of the peduncle tissue (Fig. 5A), indicating that fruit detachment at the PPJ involved extensive mechanical breakage at the junction. The fracture plane of the pedicel end of the PPJ in the

28 ethephon treatment also displayed the presence of parts of the peduncle, although this was lesser than that observed in the control (Fig. 5C). These data suggest that some of the separation in response to ethephon applications was associated with mechanical breakage. In MeJa treated samples, the pedicel end of the PPJ displayed a smoother fracture plane with a few broken cells towards the center of pedicel (Fig. 5E). The peduncle tissue was not attached to the fracture plane indicating uniform fruit detachment involving lesser mechanical disruption in response to

MeJa applications. The morphological features of the fracture plane observed here are consistent with that reported during ethylene-induced citrus leaf abscission (Agustí et al., 2009). In this study, prior to the ethylene treatment, the fracture plane displayed a ragged surface consisting of broken cell walls, indicating forcible separation at the laminar AZ, similar to that observed in the control treatment in the current study. At 24 h after ethylene treatment, the AZ planes were mostly flattened with some breakage within the pith and vascular bundles, similar to that observed in the MeJa treatment in the current study. These observations suggest that fruit detachment at the PPJ in response to the abscission agents, especially MeJa, was associated with the physiological process of abscission. In response to the application of the abscission agents, the progression of abscission may be accelerated, resulting in a smoother fracture plane at the

PPJ, especially in response to MeJa applications. Differences in the morphology of the fracture plane between the MeJa and ethephon treatments may be attributed to differences in the timing of peak efficacy of the treatments (Malladi et al., 2012).

In the control treatment, the fracture plane of the pedicel end of the FPJ contained parts of the fruit tissue in addition to extensively disrupted cells (Fig. 5B). Neither of the abscission agent treatments altered the appearance of the fracture plane at the pedicel end of the FPJ (Fig.

5D, F). These data clearly indicate that the abscission agent applications did not affect

29 detachment characteristics at the FPJ. Hence, fruit detachment at the FPJ in response to mechanical shaking after abscission agent applications was likely associated with mechanical tearing and physical separation of the berry.

Fruit detachment in response to mechanical shaking. Mechanical shaking for 5 s resulted in the detachment of around 65% of the fruit in ‘Briteblue’ (Fig. 6). Interestingly, the majority of the fruit detachment in response to mechanical shaking occurred at the point of attachment of the pedicel to the fruit, i.e., at the FPJ, while only around 14% of the detached fruit separated at the

PPJ. These data indicate that fruit detachment during mechanical shaking occurs at a junction distinct from the point of fruit detachment associated with natural or abscission agent-induced fruit drop.

Among the few fruit that detached at the PPJ during mechanical shaking, the fracture plane of the pedicel end of the PPJ appeared to be flattened and even, similar to that observed during natural fruit detachment (Fig. 7A). The few fruit that detached at the PPJ during mechanical shaking may likely represent mature fruit in which the physiological process of natural abscission had already been initiated. Weakening of this junction due to the progression of abscission may aid in the detachment of these fruit at this location upon the application of physical force during mechanical shaking. The fracture plane at the pedicel end of the FPJ was also observed using fruit that detached at the FPJ during mechanical shaking. This fracture plane consisted of extensively disrupted cells (Fig. 7B). Importantly, part of the fruit tissue was often found to be attached to the pedicel at this fracture plane. The above data indicate that fruit detachment at the FPJ during mechanical shaking occurred as a result of physical breakage and tearing, processes not consistent with abscission.

30 Conclusions

Anatomical analyses clearly indicated the presence of an AZ at the PPJ, but did not support the presence of an AZ at the FPJ in rabbiteye blueberry. The PPJ appears to be the main point of fruit detachment when the mature fruit naturally drop as a result of the physiological and developmental progression of abscission. This conclusion is supported by the presence of a largely smooth fracture plane along with mild tissue disruption, primarily within the vascular tissue in the naturally detached fruit. The application of abscission agents which may be expected to accelerate the progression of abscission resulted in fruit detachment primarily at the

PPJ, further supporting the conclusion that fruit detachment at this junction is associated with abscission. However, mechanical shaking alone resulted in fruit detachment at the FPJ. Such detachment of the fruit was not consistent with abscission but was generally associated with extensive physical disruption of cells. Hence it is likely that fruit detachment at the FPJ is a result of mechanical breakage and physical separation. The FPJ may present a weak junction and the oscillations of the berry around this point during mechanical shaking may result in physical separation at this junction. Similarly, manual fruit removal may also result in physical separation at this weak junction.

Knowledge of fruit detachment gained from this study has potential implications for mechanical harvesting of blueberry fruit. Abscission agents have been shown to considerably enhance the extent of fruit detachment (Malladi et al., 2012). However, as the physiological process of abscission is involved in such detachment, it is likely that the applications of these agents will result in ‘stemmy’ fruit which will subsequently require de-stemming as the presence of the pedicel on the fresh fruit is considered to be a defect that reduces its quality (USDA,

1995). Additionally, breeding programs aiming to improve mechanical harvesting traits may

31 need to select genotypes with stronger pedicel-peduncle junctions so that fruit detachment during mechanical harvesting in such genotypes can occur at the weak union along the FPJ. In conjunction with the above, selection for a dry or a small stem scar at this junction may help maintain the post-harvest quality of the fruit.

Acknowledgements

The authors thank Lisa K. Johnson for help with experiments in this study. This research was supported by a grant from the U.S Department of Agriculture, Specialty Crops Research

Initiative (SCRI): 2008-51180-19579.

32 Figure 1. Anatomy of the peduncle-pedicel and the fruit-pedicel junctions in rabbiteye blueberry

(‘Powderblue’). (A) the peduncle-pedicel junction is shown. The black arrow indicates the indentation at the point of attachment of the peduncle (PE) and the pedicel (PL). The white arrows on either side of the vascular bundle (VB) indicate multiple layers of small rounded cells within the abscission zone. (B) the fruit-pedicel junction is presented. The black arrow indicates the indentation at the point of attachment of the pedicel (PL) and the fruit (FT). The white arrow indicates normal size cells next to the vascular bundle (VB) at the junction of the pedicel and the fruit. Bar = 200 μm.

33 Figure 2. Cumulative natural detachment of mature fruit at the peduncle-pedicel junction in the rabbiteye blueberry cultivars, ‘Premier’ and ‘Briteblue’ in 2011 and 2012. Natural fruit drop within a 1 m2 area below the plant was determined at regular intervals for 2-3 weeks. The proportion of naturally dropped fruit which detached at the peduncle-pedicel junction is presented. Error bars indicate the SE of the mean.

34 Figure 3. Morphology of the fracture plane of the pedicel end of the peduncle-pedicel junction in naturally detached rabbiteye blueberry fruit (‘Briteblue’). Fruit naturally detached from the plant at the peduncle-pedicel junction were collected and the fracture plane at the point of detachment was observed using scanning electron microscopy (bar = 100 μm). The arrow indicates the region of extensive cell rupture and breakage within the vascular tissue.

35 Figure 4. Effect of abscission agents on the point of fruit detachment in rabbiteye blueberry.

-1 Control (adjuvant only), methyl jasmonate (MeJa; 20 mM), and ethephon (1000 mg·L ) applications were performed on ‘Briteblue’ blueberry plants (n = 3). At 24 h after application, individual branches with around 50 fruit each were shaken using a mechanical shaker. The proportion of detached fruit that retained the pedicel was determined. ANOVA was performed followed by mean separation using Fisher’s least significant difference (α = 0.05). Error bars represent the SE of the mean. Similar letters above the bars indicate no significant difference between treatments.

36 Figure 5. Effect of abscission agents on the morphology of the fracture plane of the peduncle- pedicel and the fruit-pedicel junctions. Control (A and B), ethephon (C and D; 1000 mg·L-1), and methyl jasmonate (E and F; 20 mM) applications were performed on ‘Powderblue’ blueberry plants. At 24 h after treatment, the pedicel was detached manually from the peduncle to obtain the fracture plane at the peduncle-pedicel junction (A, C and E). Similarly, the pedicel was detached manually from the fruit to obtain the fracture plane of the fruit-pedicel junction (B, D and F). The fracture plane of the pedicel end of the junctions was analyzed using scanning electron microscopy (bar = 100 μm).

37 Figure 6. Fruit detachment in response to mechanical shaking in rabbiteye blueberry

(‘Briteblue’). Individual branches were shaken using a mechanical shaker for 5 s. The fruit number prior to and after shaking was used to determine the extent of total fruit detachment (n =

3). Detached fruit were collected in a catch frame and the proportion of fruit which detached at the peduncle-pedicel junction (with the pedicel) was determined. Error bars represent the SE of the mean.

38 Figure 7. The morphology of the fracture plane of the peduncle-pedicel and the fruit-pedicel junctions in response to mechanical shaking in rabbiteye blueberry (‘Briteblue’). Fruit detached during mechanical shaking were collected and the fracture plane at the point of detachment was observed using scanning electron microscopy (bar = 100 μm). (A) Fracture plane of the pedicel end from fruit that detached at the peduncle-pedicel junction. (B) Fracture plane of the pedicel end of fruit that detached at the fruit-pedicel junction.

39 References

Agustí, J., P. Merelo, M. Cercós, F.R. Tadeo, and M. Talón. 2009. Comparative transcriptional survey between laser-microdissected cells from laminar abscission zone and petiolar cortical tissue during ethylene-promoted abscission in citrus leaves. BMC Plant Biol 9: 127. Austin, M.E. and R.E. Williamson. 1977. Comparison of harvest methods of rabbiteye blueberries. J. Amer. Soc. Hort. Sci. 102:454-456.

Goren, R. 1993. Anatomical, physiological, and hormonal aspects of abscission in citrus. Hort. Rev. 15:145-182. Gough, R.E. and W. Litke. 1980. An anatomical and morphological study of abscission in highbush blueberry fruit. J. Amer. Soc. Hort. Sci. 105:335-341. Howell, G.S., B.G. Stergios, S.S. Stackhouse, H.C. Bittenbender, and C.L. Burton. 1976. Ethephon as a mechanical harvesting aid for highbush blueberries (Vaccinium austral Small). J. Amer. Soc. Hort. Sci. 101:111-115. Iwahori, S. and R. F. Van Steveninck. 1976. Ultrastructural observation of lemon fruit abscission. Scientia Hort. 4:235-246. Iwai, H., A. Terao, and S. Satoh. 2012. Changes in distribution of cell wall polysaccharides in floral and fruit abscission zones during fruit development in tomato (Solanum lycopersicum). J. Plant Res. DOI: 10.1007/s10265-012-0536-0. Kazokas, W.C. and J. K. Burns. 1998. Cellulase activity and gene expression in citrus fruit abscission zones during and after ethylene treatment. J. Amer. Soc. Hort. Sci. 123:781-786. Mainland, C.M., L.J. Kushman, and W.E. Ballinger. 1975. The effect of mechanical harvesting on yield, quality of fruit and bush damage of highbush blueberry. J. Amer. Soc. Hort. Sci. 100:129-134.

Malladi. A., T. Vashisth, and L.K. Johnson. 2012. Ethephon and methyl jasmonate affect fruit detachment in rabbiteye and southern highbush blueberry. HortScience. 47:1745-1749.

Malladi.A.,T. Vashisth, and S. NeSmith. 2013. Development and evaluation of a portable, hand- held mechanical shaker to study fruit detachment in blueberry. HortScience. In press.

Mims, C.W. 1981. SEM of aeciospore formation in Puccinia bolleyana. Scanning Electron Microsc. III: 299-303.

Norton, R. A., L. L. Claypool, S. J. Leonard, P. A. Adrain, R. B Fridley, and F. M. Charles. 1962. Mechanical harvest of sweet cherries. California Agr. 16:8–10.

Patterson, S.E. 2001. Cutting loose. Abscission and dehiscence in Arabidopsis. Plant Physiol. 126:494-500.

40 Roberts, J.A., K.A. Elliott, and Z.H. Gonzalez-Carranza. 2002. Abscission, dehiscence and other cell separation processes. Annu. Rev. Plant Biol. 53:131–158. Roberts, J.A., C.A. Whitelaw, Z.H. Gonzalez-Carranza, and M.T. McManus. 2000. Cell separation processes in plants – models, mechanisms and manipulation. Ann. Bot. 86:223–235.

Ruzin, S.E.1999. Plant microtechnique and microscopy. Oxford University Press, New York, NY. Sexton, R. and J.A. Roberts. 1982. Cell biology of abscission. Annu. Rev. Plant Physiol. 33:133- 162. Stösser, R., H.P. Rasmussen, and M.J. Bukovac. 1969. A histological study of abscission layer formation in cherry fruit during maturation. J. Amer. Soc. Hort. Sci. 94:239–243. Takeda, F., G. Krewer, E.L. Andrews, B. Mullinix, and D.L. Peterson. 2008. Assessment of the V45 blueberry harvester on rabbiteye blueberry and southern highbush blueberry pruned to V- shaped canopy. HortTechnology. 18:130-138. U.S. Department of Agriculture. 1995. United States standards for grades of blueberries. Agricultural Marketing Service, Washington, DC. U.S. Department of Agriculture. 2012. Non-citrus fruits and nuts 2011 summary. National Agricultural Statistics Service, Washington, DC. Webster, B.D. 1968. Anatomical aspects of abscission. Plant Physiol. 43:1512-1544. Wittenbach, V.A. and M.J. Bukovac. 1972. An anatomical and histochemical study of abscission in maturing sweet cherry fruit. J. Amer. Soc. Hort. Sci. 97:214-219. Wittenbach, V.A. and M.J. Bukovac. 1974. Cherry fruit abscission: Evidence for time of initiation and involvement of ethylene. Plant Physiol. 54:494-498.

41 CHAPTER 3

CHANGES IN THE CELL WALL COMPOSITION AND WALL-RELATED GENE

EXPRESSION DURING BLUEBERRY FRUIT ABSCISSION1

1Vashisth, T., S.Pattathil, U. Avci, M.G.Hahn and A. Malladi. 2013. To be submitted to Journal of ASHS

42 Abstract

The spatial and temporal changes in the cell wall composition and metabolism within the abscission zone (AZ) during organ separation are not completely understood. In blueberry, fruit abscission occurs at the pedicel-peduncle junction (PPJ). The objective of this study was to obtain a comprehensive understanding of changes in the cell wall during blueberry fruit abscission. Ethephon (1000 mg L-1) and Methyl Jasmonate (MJ; 20 mM) were used to induce abscission. Around 60% of the fruits abscised in response to Ethephon and MeJa treatments. The

PPJ tissue was collected at 24 h after treatment and cell wall glycome profiling using over 200 glycan-directed monoclonal antibodies was performed. More Xyloglucan (XG) epitopes were released in MeJa treated PPJ tissues, suggesting that loosening of XG is a PPJ-specific phenomenon induced by MeJa mediated mechanisms. Also, the relative abundance of hemicellulosic epitopes (Xylan and XG) released in the chlorite extract of treated walls was reduced during abscission induction by Ethephon and MeJa, potentially as a result of reduced lignin-hemicellulose association within the cell walls of the active AZ. An increase was observed in CCRC-M 104 and LM 15 labeling of XG in ethephon and MJ treated PPJ tissue after 48 h of application, though labeling was throughout tissue and not confined to AZ. No labeling was observed with JIM 7, medium labeling intensity was observed with JIM 5 in PPJ tissue. This suggests that PPJ tissue is mostly de-esterifed and potentially XG play a key role in the abscission process. Next generation sequencing of the AZ transcriptome was performed to identify genes expressed within this region. Nineteen AZ specific cell wall carbohydrate metabolism-related genes were identified and were studied for changes in their expression in response to treatment with Ethephon and MJ. Expression of POLYGALCTURONASE ISOZYME increased by 3-fold in response to MJ and ethephon treatments, while PECTATE LYASE, BETA-

43 GLUCANASE, and BETA-GALACTOSIDASE increased by more than 8 to 15-fold at 48 h after treatment. Expression of PECTIN METHYLESTERASE and INVERTASE was down-regulated by in response to these treatments. Together, these data suggests that blueberry fruit abscission is mediated by specific changes in hemicelluloses and pectins in cell wall at the PPJ.

Keywords: Abscission, blueberry, plant cell wall, cell – wall carbohydrate metabolism, glycome profiling, immunolocalization, plant cell wall glycan- directed antibody

Introduction

Abscission is a physiological process that involves the programmed separation of entire organs, such as leaves, petals, fruit, and flowers. Abscission is a highly coordinated and regulated process in response to plant developmental cues, and various biotic and abiotic stresses. Organ separation occurs at abscission zones (AZ). Abscission can be divided into four major steps

(Patterson, 2007). During the first step, formation of the AZ occurs. Earlier studies have identified several MADS-box genes are required for the AZ formation, such as JOINTLESS. A mutation in this gene suppresses the development of pedicel AZ in tomato (Mao et al., 2000). In step 2, AZ cells become responsive to phytohormones such as ethylene, jasmonates, auxin and abscissic acid (ABA). In response to phytohormone signaling the middle lamella dissolves.

During step 3, cells in the AZ expand and cell wall loosening and degradation occur. Finally in step 4, cell separation occurs followed by the suberization of the proximal layer of cells to form a protective layer. In the process of abscission, breakdown/degradation of cell wall is one of the key steps to allow for the completion of cell separation. The spatial and temporal regulation of the dissolution of primary cell wall polysaccharides and middle lamella is not completely understood. However, many studies have indicated that loss of pectins and other polysaccharides

44 from the middle lamella, and degradation of primary cell wall are commonly associated with abscission (Clements and Atkins, 2001; Roberts et al., 2000; Uheda and Nakamura, 2000).

A major component of the cell walls is a network of cellulose microfibrils joined by hemicelluloses. Mechanical properties of the cell walls are largely dependent upon this network

(Whitney et al., 1999). Hemicelluloses are defined chemically as cell wall polysaccharides that are structurally homologous to cellulose because of their backbone composed of 1, 4-linked β-D- pyranosyl residues (O'Neill and York, 2003). Hemicelluloses are not solubilized by water or chelating agents but are solubilized by aqueous alkaline solvents (Selvendran and O'Neill, 1985).

Hemicelluloses include xyloglucans, glucomannans, mannans, xylans, arabinoxylans, and arabinogalactans. Xyloglucan is the most abundant hemicellulose in the primary walls of higher plants, often comprising 20% of the dry mass of the wall (O'Neill and York, 2003). Xyloglucan has a backbone composed of 1, 4-linked β-D-Glucose residues. Xyloglucans are classified as

XXXG-type or XXGG-type depending on the number of backbone residues that are branched.

Xylans, including arabinoxylan, glucuronoxylans, and glucuronoarabinoxylans, are quantitatively minor components of the primary walls of dicots (Darvill et al., 1980; Ebringerova and Heinze, 2000). Many studies have reported increase in expression of hemicellulose and cellulose hydrolysis related genes during the process of abscission. An increase in abscission was related to an increase in EG activity in fruit crops such as apple (Pandita and Jindal, 1991), avocado (Tonutti et al., 1997), peach (Rascio et al., 1985) and raspberry (Sexton et al., 1997). A very diverse gene family encodes for EGs in tomato (Brummell et al., 1997; del Campillo and

Bennet, 1996). Two members of this family, CEL1 and CEL2, exhibit an increase in transcript abundance during fruit ripening and flower abscission (Lashbrook et al., 1994).

45 In addition to the cellulose-hemicellulose network, a structurally complex matrix of pectin is also present within the primary cell walls and is abundant in the intercellular matrices of middle lamella. Pectins are polysaccharides that contain a high proportion of 1, 4-linked α-D- galactosyluronic acid residues (Ridley et al., 2001). These polysaccharides are typically solubilized by treating walls with aqueous buffers, chelating agents, and dilute mineral acids. All primary walls contain three pectic polysaccharides - homogalacturonan (HG), rhamnogalacturonan I (RG-I), and rhamnogalacturonan II (RG-II) (O'Neill et al 1990).

Homogalacturonan (HG) is composed of 1, 4 linked α-D-galactosyluronic acid residues. Some of the carboxyl groups of HG can be methyl esterified. Lee et al. (2008) reported that AZ- associated de-esterification of homogalcturonan components of pectin was detected upon induction of abscission in poinsettia. They also reported abundance of lignin and xylan in poinsettia AZ and lower levels of cellulose, arabinose and pectin. During tomato flower abscission, change in the distribution of xyloglucan, pectic galactan and arabinose have been found, suggesting a key role of these polysachharides in bringing about the completion of abscission (Iwai et al., 2012).

Numerous studies have suggested that the alterations in cell wall polysaccharides are co- related with the activity of cell wall hydrolysis enzymes and change in cell wall metabolism- related genes. The middle lamellae of AZ cells are presumably loosened by the active mobilization of cell wall hydrolases, including Endo β-1, 4-glucanases (EGs; Lashbrook et al.,

1994) and polygalacturonases (PGs) that result in the weakening of the cell wall in the AZ

(Brown, 1997), initiating organ detachment. Many studies reported that ethylene and other phytohormones are involved in inducing cell wall-degrading enzymes such as EG, PGs,

Cellulases, and β–Galactosidases, as well as lipid-modifying phospholipases in AZs (Goren,

46 1993; Malladi and Burns, 2008). A very diverse gene family encodes for EGs in tomato

(Brummell et al., 1997; del Campillo and Bennet, 1996). Two members of this family, CEL1 and

CEL2, exhibit an increase in transcript abundance during fruit ripening and flower abscission

(Lashbrook et al., 1994). Increase in the transcript levels of MdPG2, and MdEG1 in the fruit AZ was concomitant with an increase in the expression of MdACS5A, MdACO1, MdETR2, and

MdERS2 in apple during ripening and abscission (Li et al., 2010). In cherry tomato, expression of the EG and cellulose genes (ENDO-1, 4-β-GLUCANASES, CEL1, CEL2, CEL3, CEL5, CEL7 and CEL8) in the AZ was enhanced during abscission of the fruit upon treatment with ethylene and Meja. 1-MCP blocked the expression of these genes in both cases of treatment with the growth regulators (Beno-Moualem et al., 2004). Roberts et al. (2002) reported delayed floral organ abscission in Arabidopsis thaliana mutants defective in abscission-related PG. Silencing of

PG resulted in an increase in the break strength required for petiole abscission, whereas silencing of LeCEL1 and LeCEL2 (cellulase) and LeEXP11 and LeEXP12 (expansins) had no discernible effect on the break strength in tomato, even when two of these genes were silenced concurrently, indicating the important role of PG in abscission (Jiang et al. 2008). Expansins are a class of proteins which posses the unique ability to induce cell-wall extension without hydrolytic breakdown of the major structural components of the cell wall (Cosgrove, 1999). Cho and

Cosgrove (2000) reported the role of expansins in cell-wall loosening during a variety of developmental processes including abscission in Arabidopsis thaliana. In ethylene promoted leaf abscission, up to a 7-fold increase in expansin expression was observed in cells undergoing separation, indicating an important role of expansins in ethylene mediated abscission (Belfield et al., 2004).

47 Blueberry is a fruit crop indigenous to the United States and was valued at over $780 million in 2011 (USDA, 2012). Fruit abscission in rabbiteye blueberry occurs at the pedicel- peduncle junction (PPJ; Vashisth and Malladi, 2013). Ethephon and Methyl Jasmonate (MeJa) are two abscission agents that are effective in inducing fruit abscission in blueberry (Malladi et al., 2012). Enhancing our knowledge of the process of fruit detachment can greatly aid in developing methods to increase the efficiency of mechanical harvesting, which can in turn reduce the costs associated with blueberry production (Austin and Williamson 1977; Howell et al., 1976; Mainland et al., 1975; Takeda et al., 2008). In spite of its emerging importance, various aspects of blueberry physiology including the alteration in the cell wall chemical composition and expression of cell wall metabolism related genes during induction of abscission are not well understood. This study aimed to obtain a comprehensive understanding of changes in the cell wall during blueberry fruit abscission. Three different approaches were used to achieve this objective: 1) Glycome profiling- A technique in which cell wall glycan-directed monoclonal antibodies (mAbs) were used to monitor structural/extractable changes in cell wall polysaccharides upon induction of abscission; 2) Immuno-localization – Spatial changes in the distribution of cell wall polysaccharides in response to abscission were studied; and 3) Gene expression profiling - The expression pattern of cell wall metabolism related genes upon induction of abscission was analyzed.

Materials and Methods

Plant Material for Glycome Profiling and Immuno-localization:

Mature (5-year-old) rabbiteye blueberry plants of ‘Briteblue’ grown at the University of Georgia,

Horticulture Farm, Watkinsville, GA were used in this study in 2011. The experiment was designed as a completely randomized design with three treatments and three replicates

48 (individual plants). The treatments were: 1) control; 2) methyl jasmonate (MeJa, 20 mM; Sigma-

Aldrich, St. Louis, MO); and 3) ethephon (1000 mg·L-1; Bayer CropScience, Kansas City, MO).

The treatments were performed when majority (~75%) of the fruit on the plant was mature. All treatments, including the control, were applied along with 0.15% of an adjuvant (Latron B-1956,

Rohm and Haas, Philadelphia, PA), and were performed using a hand pump sprayer until run-off around 0830 HR. The average daily temperature on the day of application was 27.5 °C. At 24 h after treatment, the PPJ and pedicel tissue from about 50 fruit were manually dissected, although

1 mm or less on either side of AZ remained attached to the PPJ. Dissected PPJ and pedicel tissue were pooled within each replicate and frozen in liquid N2 for glycome profiling. For immuno- localization, intact control, ethephon and MeJa treated PPJ tissue were fixed in fixed in 1.6%

(v/v) paraformaldehyde and 0.2% (v/v) glutaraldehyde in 25 mM sodium phosphate buffer (pH:

7.1) and were stored at 4 °C.

In 2013, the above experiment was repeated with similar parameters to perform a time series immuno-localization study. Mature (7-year-old) rabbiteye blueberry plants of ‘Briteblue’ grown at the University of Georgia, Horticulture Farm, Watkinsville, GA were used in 2013.

Intact PPJ tissue from control, ethephon and MeJa were collected at 0, 24 and 48 h after treatment. PPJ tissues were fixed in 1.6% (v/v) paraformaldehyde, 0.2% (v/v) glutaraldehyde and

25 mM sodium phosphate buffer (pH: 7.1) and were stored at 4°C.

ELISA sample preparation and analysis: Approximately 200 mg of control, ethephon and MeJa treated PPJ and pedicel tissue were ground in pestle and mortar and were sequentially washed with absolute ethanol and 100% acetone. The washed residues were then vacuum-dried overnight. The dried samples were subjected to sequential extraction with increasingly harsher reagents to isolate fractions enriched in various cell wall components. All extractions were

49 performed in 10 mg mL-1 suspensions based on the starting dry biomass weight used. First, the biomass was suspended in 50 mM ammonium oxalate (pH 5.0) and incubated overnight with constant mixing at room temperature. After incubation, the mixture was centrifuged at 3400 g for

15 min at room temperature. The resulting supernatant fraction was decanted and saved as the ammonium oxalate fraction, and the pellet was subsequently washed by re-suspension in the same volume of deionized water and centrifuged again as previously described except that the subsequent supernatant fraction was decanted and discarded. Following the same protocol, the pellet was then subjected to additional sequential extractions using in turn 50 mM sodium carbonate (pH 10) containing 0.5% (w/v) sodium borohydride, 1 M KOH, and 4 M KOH, each containing 1% (w/v) sodium borohydride. The pellet remaining after the final KOH extraction was then treated with sodium chlorite (100 mM) to breakdown lignin polymers into smaller components. Lastly, the pellet left following the sodium chlorite treatment was subjected to a final extraction with 4 M KOH containing 1% (w/v) sodium borohydride to isolate material that had previously been secured within the walls by lignin (4 M KOH PC). The resulting residual pellet was discarded. The 1 M KOH, 4 M KOH, and 4 M KOH PC extracts were neutralized with glacial acetic acid. All extracts were dialyzed against four changes of DI water (with an approximate sample to water ratio of 1:60) for 48 h at room temperature and subsequently lyophilized. After estimating the total sugar contents of the cell wall extracts using the phenol- sulfuric acid method, the extracts were dissolved in de-ionized water to a concentration of 0.2 mg mL-1. Subsequently, all extracts were diluted to the same sugar concentration of 60 mg mL-1 for loading onto ELISA plates (Costar 3598). Diluted extract (50 mL) was added to each well and allowed to evaporate overnight at 37 °C until dry. The ELISA assays were performed using an array of 150 monoclonal antibodies specific to epitopes from most major groups of plant cell

50 wall polysaccharides as described in Pattathil et al., (2010). Negative controls consisting of water blanks without the antigen were included in all assays and their absorbance subtracted from all samples. The OD of each well was read as the difference in A450 and A655 using a model 680 microplate reader (Bio-Rad). ELISA data are presented as heat maps in which antibodies are grouped based on a hierarchical clustering analysis of their binding specificities against a diverse set of plant glycans.

Statistical Analysis: Statistical analysis for differential binding specificities of antibodies against plant cell wall glycan was performed using Microsoft Excel [Redmond, Washington: Microsoft,

2007 (Computer Software)]. Average value for replicates was used for analyses. The OD values for control, ethephon and MeJa treated PPJ tissues were normalized against their respective pedicel tissues. Further, the OD values normalized to the pedicel were normalized using the PPJ control treatment. A 30% cut-off (0.66-1.5; no change) was set to determine the differential binding specificities of antibodies for the resulting normalized OD values for ethephon and MeJa

PPJ tissue. Subsequently, only the antibodies for which at least three out of the six fractions (AO,

SC, 1 M KOH, 4M KOH, CH, and PC) displayed the similar trend of alteration of OD values in both ethephon and MeJa were identified.

Immuno-localization: Samples for immuno-localization were prepared as described in Pattathil et al., 2010. Briefly, samples were washed with the same buffer (3 times, 15 min each) and water (2 times, 10 min each). Then, samples were dehydrated through a 35%, 50%, 70%, 95% (v/v), and

100% ethanol series for 25 min each and gradually infiltrated with LR White resin (1:3 resin:100% ethanol; 1:1 resin:100% ethanol; 3:1 resin:100% ethanol; 3 times resin; each step 24 h). Samples were placed into gelatin capsules with fresh LR White and polymerized under ultraviolet light at 4 °C for 48 h. Sections of 250 nm thickness were obtained using a Leica EM

51 UC6 ultra-microtome (Leica Microsystems) and mounted on glass slides (Colorfrost/plus; Fisher

Scientific). Sections were blocked with 3% (w/v) nonfat dry milk in KPBS (0.01 M potassium phosphate, pH 7.1, containing 0.5 M NaCl) for 45 min and then were washed with KPBS for 5 min. Undiluted hybridoma supernatant fraction of the mAb was applied and incubated for 60 min. Sections were then washed with KPBS three times for 5 min each, and goat anti-mouse IgG or goat anti-rat IgG conjugated to Alexa-fluor 488 (Invitrogen) diluted 1:100 in KPBS was applied and incubated for 90 to 120 min. Sections were then washed with KPBS for 5 min and distilled water for 5 min. Prior to applying a coverslip, Citifluor antifade mounting medium AF1

(Electron Microscopy Sciences) was applied.

Light Microscopy: Sections were observed using a light microscope (BX51; Olympus, Center

Valley, PA) equipped with epifluorescence optics and a digital camera (DP70; Olympus). Images were assembled without further processing.

Plant Material for Gene Expression Analysis of Cell Wall Metabolism Related Genes

In 2012, mature (6-year-old) rabbiteye blueberry plants of ‘Briteblue’ grown at the University of

Georgia, Horticulture Farm, Watkinsville, GA were used. The experiment was designed as a completely randomized design with three treatments and three replicates (individual plants). The treatments were: 1) control; 2) MeJa, 20 mM (Sigma-Aldrich, St. Louis, MO); and 3) ethephon

(1000 mg·L-1; Bayer CropScience, Kansas City, MO). The treatments were performed when majority (~75%) of the fruit on the plant was mature. All treatments, including the control, were applied along with 0.15% of the adjuvant (Latron B-1956, Rohm and Haas, Philadelphia, PA) and were performed using a hand pump sprayer until run-off around 0830 HR. The average daily temperature on the day of application was 27 °C. PPJ samples from approximately 50 fruit were collected at 0, 24 and 48 h. PPJ samples were manually dissected to collect the AZ.

52 Approximately 1 mm of tissue on either side of the AZ was also included owing to limitations of manual dissection. PPJ tissues from all the 50 fruit within each replicate were pooled and immediately frozen in liquid N2 and then stored at -80 °C until further analysis.

RNA Extraction and Reverse Transcription: RNA extraction was performed as described by

Vashisth et al., 2011 with some modifications. Briefly, 0.1 g of the sample was ground to fine powder and was added to 0.75 ml of the extraction buffer. The RNA extraction buffer consisted of 2% cetyltrimethylammonium bromide (CTAB), 2% polyvinylpyrrollidone (PVP), 100 mM

Tris-HCl, 25 mM ethylenediaminetetraacetic acid (EDTA), 2 M sodium chloride (NaCl), 3.44 mM spermidine, and 2% β-mercaptoethanol. All solutions were prepared using 0.1% diethylpyrocarbonate (DEPC) treated water. The buffer was warmed to 65 °C prior to the addition of the ground tissue. The sample-buffer slurry was vortexed briefly and incubated at 65

°C for 10 min, vortexed again and incubated at room temperature for 5 min. The mixtures were extracted twice with an equal volume of chloroform: isoamyl alcohol (24:1) with centrifugation at 4 °C for 15 min (5000 x g). The supernatant fraction was transferred to sterile tubes followed by the addition of 0.25 volumes of lithium chloride (LiCl; 10 M). The samples were mixed gently and precipitated overnight at 4 °C. Samples were subsequently centrifuged at 4 °C for 20 min (12000 x g). After discarding the supernatant fraction, the RNA pellets were washed twice with 500 µL of ice cold 70% ethanol, and dissolved in DEPC and stored at -80 °C. For cDNA synthesis genomic DNA contamination was removed by treating 1 µg of total RNA was with

DNase (Promega; 37 °C for 34 min). The DNase-treated RNA was reverse transcribed using Im-

Prom II Reverse Transcriptase (Promega) and oligo dT (Promega) in a 20 µL reaction volume.

Reverse transcription was performed at 42 °C for 75 min, followed by incubation at 75 °C for 15

53 min. The cDNA samples were subsequently diluted 10-fold and stored at -20 °C until further analysis.

Gene Expression Analysis: Gene expression analysis was performed by quantitative Real Time-

PCR (qRT-PCR). Twenty eight AZ specific cell wall carbohydrate metabolism-related genes were selected from a next generation sequencing (454) AZ transcriptome dataset. Massive Go database and NCBI BLASTx mining was performed to determine their homology with respective genes from other plant species. The primers for qRT-PCR were designed using PrimerQuest

(Integrated DNA Technologies, Inc.). The amplicon size ranged from 120-200 bases. All qRT-

PCR reactions were performed using the Stratagene Mx3005P real-time PCR system (Agilent

Technologies, Inc., Santa Clara, CA). A reaction volume of 12 µL with 1 µL of diluted cDNA and 6 µL of 2X SYBR Green Master Mix (Affymetrix) was used. The thermal cycling conditions were: 95 °C for 10 min; 40 cycles of 94 °C for 30 s and 60 °C for 1 min. Melt-curve analysis was performed at the end of the qRT-PCR reactions to check for primer specificity. Controls without a template were run for each gene. The primer efficiency for each gene was determined using

LinRegPCR (Ramakers et al. 2003; Ruijter et al. 2009). Relative expression was calculated using a modified Pfaffl method (Pfaffl, 2001) and as described in Rieu and Powers (2009). Relative quantity (RQ) for each sample was calculated using the formula, 1/ECq, where Cq is the quantification cycle (threshold cycle). The RQ was normalized using RNA HELICASE-LIKE 8

(RH8) and CLATHRIN ADAPTOR COMPLEXES MEDIUM SUBUNIT FAMILY PROTEIN a

(CACSa) as described by Vashisth et al., (2011). The geometric mean of expression of the two reference genes (normalization factor) was used for normalization. The normalized RQ (NRQ) values were log2 transformed and used for statistical analyses. The standard error of the means was calculated as described in Rieu and Powers (2009).

54 Statistical Analysis: Statistical analyses were performed in SigmaPlot 11 (Systat Software Inc.,

San Jose, CA). Two-way ANOVA with the general linear model: time, treatment, and time×treatment (three levels of time, 0, 24, and 48 h after treatment; and three levels of treatment, Control, Ethephon and MeJa) was used for analyzing gene expression data (n = 3), and was followed by Fisher’s LSD multiple comparison test for mean separation (α = 0.05).

NRQ values (log2 transformed) were used for the above analyses.

Results

Induction of abscission

Two abscission agents, ethephon and MeJa, were used to induce fruit abscission in blueberry. In

2011, MeJa and ethephon resulted in significant (p<0.05) abscission of more than 50% of the fruit within 48 h of application (Figure 1). In 2012 and 2013, MeJa and Ethephon resulted in more than 60% of fruit drop within 48 h of treatment. In all the experiments, MeJa and ethephon resulted in fruit abscission at the PPJ.

Glycome Profiling

Glycome profiling was performed on the PPJ and pedicel tissue (Figure 2). PPJ tissue were manually dissected and potentially contained several layers of cells from adjacent tissues.

Therefore, the pedicel tissue was also included for glycome profiling. In glycome profiling, cell wall components isolated from PPJ and pedicel tissues were isolated and sequentially fractionated using increasingly harsher reagents to solubilize different carbohydrate components from the cell walls.

Overall, the glycome profiles of control, ethephon and MeJa treated samples from the

PPJ and pedicel tissue were largely similar in the glycan composition (Figure 3). All samples contained xyloglucans (XG), xylans, glucans and pectic polysaccharides like homogalacturonan

55 (HG), rhamnogalcturonan RG-1 and Arabinogalactans. Gylcome profiles of PPJ and pedicel also indicated the absence of galactomannans these tissues. However, significantly more homogalacturonan epitopes (a pectic epitope), HG backbone-1 were extracted in the oxalate fraction from the pedicel. Also, more hemicellulosic polysaccharides were extracted by the carbonate reagent in control pedicel compared to PPJ. Two specific differences were observed in the abscission agent treated PPJ tissues: 1) Hemicellulosic polysaccharides in carbonate fraction:

Greater extent of XG epitopes were released in MeJa treated PPJ tissue (Figure 3).

Hemicelluloses are soluble only in aqueous alkali therefore their presence in the carbonate fraction suggests that loosening of XG could be a PPJ specific phenomenon mediated by induction of abscission in response to MeJa.; 2) the relative abundance of hemicellulosic epitopes (xylan and XG) released in the chlorite fraction in case of treated PPJ was reduced

(Figure 3). In plant cell walls, lignins are linked with hemicelluloses and confer mechanical strength. During chlorite extraction lignin are removed from the cell wall residue. Such an observation could be potentially a result of reduced lignin-hemicellulose association in the cell walls of AZ due to the induction of abscission by ethephon and Meja.

The statistical analysis was performed on the average binding OD values of the each antibody for ethephon and MeJa treated PPJ. The OD values were normalized against respective pedicel values and then further normalized against the control. This approach was performed to ensure that the resulting differential binding of the antibodies was primarily associated with the induction of abscission. At least, 38 out of 150 antibodies displayed differential binding upon induction of abscission in response to ethephon and MeJa (Table 1). About 50% of the differentially expressed antibodies were specific for XG (fucosylated and non-fucosylated) and xylan families, suggesting that these polysaccharides are primarily altered or remodeled as a

56 result of induction of abscission. Low OD values but some differential expression was observed in the case of antibodies targeting galactomannans, suggesting its negligible presence in the PPJ.

Only about 30% of the antibodies with differential expression were specific to the pectin (HG and RG) family. This analysis suggests that during the induction of abscission remodeling and alteration of hemicelluloses is potentially a major contributor to the cell separation process. A relatively minor degree of alteration of pectins was also observed upon the induction of abscission.

Immuno-localization

Immuno-localization analysis using monoclonal antibodies was performed to investigate the distribution of cell wall pectic and hemicellulosic polysaccharides. Results from the 2011 immuno-localization study was used to further select antibodies of importance for a time- progression immuno-labeling study in 2013. In 2011, a total of 38 antibodies were selected for immuno-localization based on the results of glycome profiling and previous studies on abscission. Eleven out of the 38 antibodies showed labeling in the PPJ sections. PPJ sections labeled with JIM 5 displayed high binding intensity however the labeling was not specific to the

AZ alone (Figure 4). No labeling was observed in case of JIM 7. JIM 5 recognizes low or non- methylesterified HG whereas JIM 7 binds to only heavily methy-esterified HG. Such an observation suggests that the HG in AZ and entire pedicel are not heavily methyl-esterified.

CCRC- M137, CCRC-M139, CCRC-M153 and LM 11 showed high intensity of labeling in control, ethephon and MeJa treated PPJ sections (Figure 5). However the labeling was localized to the vascular bundles and no or little labeling was observed in the epidermis and the cortex. All of these four antibodies bind to the xylan epitopes indicating that epidermis and cortex lack the corresponding xylan epitopes for these antibodies (Figure 5). CCRC-M108 showed low intensity

57 labeling in the entire sections including vascular bundles, cortex and epidermis suggesting that xylan epitopes for CCRC-M108 are present throughout the PPJ section, though not in abundance. The intensity of CCRC-M88, CCRC-M51 and CCRC-M104 labeling appeared to be high but was evident throughout the tissue and was not affected by treatment with abscission agents. These three antibodies identify XG (non-fucosylated, NFXG) epitopes. Hence, such labeling pattern suggests that the PPJ sections have an abundance of various type of NFXG. LM

15 showed overall low intensity all over the tissue suggesting that the LM 15 XG (fucosylated,

FXG) epitopes were present within the AZ, but also in the pedicel including. CCRC-M84 displayed low binding intensity throughout the entire section although no binding was observed in the AZ indicating the absence of CCRC-M 84 FXG/pectic polysachharide epitope in AZ.

Immuno-localization pattern from CCRC-84 suggests that FXG epitopes are less abundant in PPJ and pedicel as compared to non-fucosylated xyloglucan epitopes.

Further these eleven antibodies were selected to investigate changes of labeling distribution during the progression of abscission. The PPJ samples used in this study were collected 0, 24 and

48 h after the treatment. CCRC- M137, CCRC-M139, CCRC-M153, LM 11, and CCRC-M108,

CCRC-M88 displayed no differences in their labeling patterns in response to the abscission agents or over the period of abscission progression. In case CCRC-M 84 low labeling intensity was observed throughout the section including the AZ, this was contrary to the preliminary results where CCRC-M 84 epitopes were missing in the AZ. Enhanced labeling was observed in case of CCRC-M104 and LM 15 in ethephon and MeJa treated PPJ at 48 h after the treatment

(Figure 6, 7). This observation suggests that the NFXG and FXG epitopes for these antibodies became readily available for binding upon induction of abscission. In CCRC-M51 binding intensity was reduced from 24 h to 48 h in case of all three: control, MeJa and ethephon treated

58 sections suggesting that non-fucosylated xyloglucan epitopes for CCRC-M51 were undergoing remodeling with progression of time.

Gene expression profiling

Gene expression profiling of cell wall metabolism genes was performed to determine the genes that are potentially instrumental in remodeling or alteration of cell wall polysaccharides during the induction of abscission by ethephon and MeJa. Next generation sequence data of the blueberry AZ transcriptome was mined for cell wall metabolism related genes. A total of 19 cell wall metabolism related genes were identified and further analyzed to determine their expression profiles upon induction of abscission at 0, 24 and 48 h after the treatment (Table 2). Out of the

19 genes analyzed, eight genes displayed significant (p< 0.05) interaction of time*treatment effect upon induction of abscission (Figure 8). Expression of PECTATE LYASE increased by 8- fold in case of MeJa and 7-fold in case of ethephon within 48 h of treatment suggesting that potentially PECTATE LYASE- a plays a key role in cell wall loosening by hydrolysis of pectin during induction of abscission. A 10 to 15-fold increase in expression of β-GLUCANASE was observed within 24-48 h of MeJa application, while more than 5-fold increase was observed within 48 h of ethephon application (Figure 8). This suggests that β-GLUCANASE potentially plays a critical role in enhanced abscission in response to MeJa and Ethephon. MeJa application also resulted in the up-regulation of a β-GALACTOSIDASE by 8-fold within 48 h whereas ethephon did not alter its expression. An increase of 2.5 to 3-fold in expression of

POLYGALCTURONASE ISOZYME was observed within 24-48 h of MeJa application whereas a sharp decrease in its expression was observed in response to ethephon (Figure 8). A decrease in the expression of PECTIN METHYLESTERASE (PME) was observed in response to MeJa and ethephon. The primary role of PME is to catalyze the de-methylesterification of

59 homogalacturonan components of pectins. Similar to PME, a decrease in the expression was observed in case of PLANT INVERSTASE/PECTIN METHYESTERASE INHIBITOR in response to MeJa and ethephon. PLANT INVERSTASE/PECTIN METHYESTERASE INHIBITOR super family is involved in cell wall biogenesis/degradation. More than 3-fold increase in the expression of POLYGALCTURONASE INHIBITING (PGI) was observed within 24 h of MeJa application while in response to ethephon, a similar 3-fold increase was observed at 48 h after treatment (Figure 9). This increase in the expression of PGI suggests that it potentially plays an important role in blueberry fruit abscission and hastening cell separation.

Few of the cell wall metabolism related genes were up-regulated in expression over time, irrespective of the treatment. XYLOGLUCAN ENDOTRANSGLUCOSYLASE expression increased by 5 fold at 48 h. PECTATE LYASE-b was also up-regulated by 5 to 8-fold between 24 to 48 h after treatment. A two-fold increase in expression of INVERTASE/PECTIN

METHYLESTERASE INHIBITOR was also observed at 48 h. Increase in the expression of these three genes suggest that this may be intrinsic to the PPJ of mature fruit.

Discussion

Three in-depth approaches have been used to investigate the changes in the cell wall composition and metabolism-related genes upon induction of abscission. Ethephon, an ethylene releasing compound and MeJa, a jasmonate hormone family member were used to induce fruit abscission in blueberry. Both the abscission agents resulted in significant amount of fruit abscission in blueberry. MeJa resulted in significant fruit drop within 24 h of application whereas ethephon resulted in abscission only within 48 h of application. Similar effect of ethephon and MeJa on fruit abscission has been reported by Malladi et al. (2012). Difference in the time taken by the

60 two abscission agents suggests that the even though both ethephon and MeJa result in abscission, the absorption and/or signaling pathways leading to this process are potentially different.

Glycome profiling

Hemicellulosic composition of PPJ and Pedicel tissue altered the most upon induction of abscission. Specifically, XG displayed the foremost changes in the extraction pattern upon induction of abscission. More XG was released through carbonate extraction upon treatment with

MeJa, potentially due to loosening of XG upon induction of abscission; no such observation was seen in response to ethephon. Such an observation is supported by fruit abscission data where

MeJa resulted in abscission within 24 h whereas ethephon took 48 h to result in abscission. A reduction in release of hemicelluloses was also observed in the chlorite fraction (majority of lignin are removed during chlorite treatment) in ethephon and MeJa treated PPJs. Reduction in hemicellulose release during the removal of lignin suggests that hemicelluloses were already loosened upon induction of abscission and were extracted in prior fractions. A large number of hemicelluloses (XGs and xylans) directed antibodies also showed differential binding in ethephon and MeJa treated PPJ. Xyloglucans are one of the most abundant hemicelluloses of the primary cell walls of non-graminaceous species. XGs and xylans were also abundant in the active AZs of poinsettia and tomato (Lee et al., 2008; Iwai et al., 2012). XGs are proposed to have a functional role in tethering the cellulose microfibrils together, potentially through hydrogen bonds (Oneil and York, 2003). This load-bearing hemicellulosic network maintains the strength of primary cell walls which is a crucial factor supporting expansive plant growth (Oneil and York, 2003; Cosgrove, 2005; Peňa et al., 2004; Obel et al., 2007). Therefore it is likely that during the process of abscission the hemicellulose-cellulose linkage is broken down resulting in

61 cell separation. Some of the pectin (HG and RG) directed antibodies also displayed differential binding in ethephon and MeJa treated PPJs. Pectins account for ~30% of the primary walls of dicotyledenous and non-graminaceous monocotyledenous plants Oneil et al. (1990). Loss of cell adhesion due to dissolution of pectin in the middle lamella between AZ cell walls is well documented (Addicott, 1982; Sexton and Roberts, 1982).

Together glycome profiling suggests that the alteration in hemicelluloses and pectins related to the process of abscission, conceivably these alteration results in cell separation during the final step of abscission. Overall no dramatic differences were observed in glycome profile probably due to fact that the AZ region is quite minute tissue and surrounding pedicel tissue can obscure the results.

Immuno-localization

Thirty-eight monoclonal antibodies were used to study spatial distribution of cell wall polysaccharides in the PPJ tissue. Out of 38, only 11 antibodies displayed labeling in the PPJ tissue. Most of these eleven antibodies recognized the hemicellulose antigens. JIM 5 and JIM 7 are antibodies specific for pectic HG epitopes (Willats et al., 2000). HG is generally highly methyl-esterified in most primary cell walls and cleavage of HG chain is only possible upon de- esterification of HG. Polygalacturonases are considered responsible for cleavage of the HG chain and activity of polygalacturonases is well associated with abscission (Kalaitzis et al., 1997;

Tucker et al., 2007). The medium intensity of JIM 5 labeling throughout the pedicel including

AZ indicates the presence of HG in these cells. However, no JIM 7 labeling was observed indicating that the HG present in pedicel were not methyl-esterified or were methyl-esterified to a low degree. No differences were observed among control, ethephon and MeJa treated PPJ tissue. Lee et al. (2008) also reported an increase in labeling intensity of JIM 5 accompanied by a

62 reduction in the labeling intensity of JIM 7 during poinsettia abscission. These data suggest that a reduction of HG methylesterification occurs in the cell wall and adjacent tissue during the process of abscission. There can be two possible reasons for such an observation 1) PPJ and pedicel tissue in blueberry is inherently low in the extent of HG methyl esterfication, or 2) As the

PPJ tissue is from mature fruit, the process of abscission may have already been initiated and may have involved de-esterification of the HG. Therefore no difference was observed among control, ethephon and MeJa treated PPJ. No labeling intensity was observed for other pectic polysaccharide specific antibodies including in LM 5 and LM 6. LM 5 and LM 6 are specific for pectic galactan and arabinan epitopes, respectively (Jones et al., 1997; Willats et al., 1998). Iwai et al. (2012) reported an increase in the labeling intensity of LM 5 and LM 6 during flower abscission in tomato whereas Lee et al. (2008) reported a decrease in labeling intensity of LM 5 and LM 6 upon induction of abscission in poinsettia. No labeling intensity of LM 5 and LM 6 in case of blueberry PPJ tissue suggests that either the epitopes for these are absent or that these epitopes are remodeled during within the AZ. An increase in CCRC-M104 and LM 15 labeling of xyloglucan epitopes was observed upon induction of abscission in response to ethephon and

MeJa. This suggests that XG (NON-FUCOSYLATED XYLOGLUCAN, NFXG) epitope were readily available for labeling during the induction of abscission. A dramatic increase in the labeling of LM15 in the flowers AZs of tomato and poinsettia has been observed during abscission (Iwai et al., 2012; Lee et al., 2008). In these studies, the labeling of LM 15 occurred throughout the pedicel and the AZ, similar to the observations of the current study. There are two possibilities for such an increase in XG epitopes during induction of abscission. First, cell degradation and synthesis are well known part of abscission process. Potentially, XG are involved in the cell wall remodeling process during abscission. Therefore during remodeling

63 process XG epitopes become readily available for binding due to the activity of cell wall protein that are associated with the AZ, such as endo-β- 1, 4- glucanases and expansins (Belfield et al.,

2005; Leslie et al., 2007). Second possibility is that, due to the decrease in methylesterified HG,

XG is exposed and available for antibody labeling. Marcus et al. (2008) showed that the decrease in labeling of methylesterified HG was accompanied by increase in the labeling of LM15.

Marcus et al. (2008) also reported that de-esterification as result of pectate lyase activity resulted in an increase in XG labeling by LM 15. A reduction in labeling intensity of CCRC-M51 was also observed. Similarly, the first possibility can also be used to explain this reduction, suggesting that XG chain gets remodeled due to activity of cell wall hydrolysis proteins and in this process epitopes for CCRC-M 51 were impacted.

Gene Expression Profiling

Loss of cell adhesion and degradation of primary cell wall are well documented physiological and metabolic process that takes place in AZ during abscission (Roberts et al., 2000, 2002).

The activities of cell wall-degrading enzymes, including cellulase (Cel), polygalacturonase (PG), expansin (EXP), and xyloglucan endohydrolase endotransglycosylase

(XET), have been shown to increase dramatically with the onset of abscission (Lashbrook et al.,

1994; Kalaitzis et al., 1997; Agustı´et al., 2008, 2009; Cai and Lashbrook, 2008; Roberts and

Gonzalez-Carranza, 2009). Cellulase /endo-glucanase and PG are two of the most widely reported active enzymes involved in cell separation processes associated with abscission

(Gonzalez-Carranza et al., 2002). Abscission induced expression of GLUCANASE and PG is reported in a number of plant species such as tomato (Brummell et al., 1999, del Campillo and

Bennet, 1996), apple (Pandita and Jindal, 1991, Li et al,. 2010), avocado (Tonutti et al., 1997), peach (Rascio et al., 1985), citrus (Burns et al., 1998) and raspberry (Sexton et al., 1997). Jiang

64 et al. (2008) reported that silencing of PG can result in increase in break strength for petiole abscission and delayed abscission indicating the important role of PG in abscission.

In the current study an up-regulation in the expression of PG ISOZYME and PL was observed within 24-48 h of application of MeJa and ethephon. A dramatic increase in expression of β-GLUCANASE and β-GALACTOSIDASE was observed only in case of MeJa treated PPJ within 24-48 h of application. In response to ethephon, a significant 5-fold increase was observed in β-GLUCANASE expression. This observation suggests that β-GLUCANASE plays an important role in abscission induced by MeJa and ethephon, while β-GALACTOSIDASE plays a role in response to MeJa. Interestingly, Agusti et al. (2008) reported an increase in the expression of a β-GALACTOSIDASE in citrus during ethylene induced abscission. Involvement of ENDO-

GLUCANASE, PG and PL in the process of abscission is reported in many studies. Tucker et al.

(2007) reported an increase in the expression of PG and PL in soybean petiole abscission.

Kalaitzis et al. (1997) reported an increase in the expression of PG in leaf and flower AZs of tomato. Meir et al. (2010) also reported an increase in the expression of PG and GLUCANASES during ethylene induced flower abscission in tomato. Similar results were reported by Beno-

Moualem et al. (2004) in cherry tomato bunches, where the expression of ENDO- 1, 4-β-

GLUCANASE was increased by MeJa and ethylene during abscission. The results from the current study are therefore consistent with these earlier reports and indicate that the progression of abscission in blueberry involves the altered expression of these genes.

A decrease in the expression of PECTIN METHYLESTERSE (PME) and

INVERTASE/PECTIN METHYLESTERASE INHIBITOR (PMEI) was observed during ethephon and MeJa induced abscission. PME catalyses the de-esterification of pectin into pectate and methanol and participates in the degradation of the middle lamella (Addicott, 1982). Since both

65 PGs and PLs favor de-esterified substrates, control of methylesterification level is undoubtedly a key factor in controlling AZ cell wall integrity. In some ripening fruits, PME-mediated de- esterification is necessary for subsequent PG action (Lashbrook, 2005). There are multiple conflicting reports about the involvement of PME in the process of abscission. In citrus, increase the expression of PME has been associated with fruit abscission (Agusti et al., 2008). On the contrary in the present study a decrease in PME expression was observed, suggesting that expression of PME may be suppressed by MeJa and Ethephon during abscission. Suppressed expression of PMEI has also been observed during Arabidopsis stamen abscission (Cai and

Lashbrook, 2008), although the significance of this alteration remains unclear and requires further investigation.

All together the results from glycome profiling, immunolocalization and gene expression profiling, suggest extensive cell wall remodeling upon the induction of abscission by MeJa and ethephon. Xyloglucans and pectins were altered greatly during this process. Xyloglucans and pectins are both quantitatively important polymers of plant cell walls comprising approximately a third each of the polysaccharides of primary cell walls of dicotyledons. Xyloglucans are known to attach to cellulose microfibrils by means of hydrogen bonds and may tether adjacent microfibrils providing the mechanical basis of cell wall strength. The pectic network of several polymers (the major one being HG) embeds the cellulose-xyloglucan network and impacts cell wall properties including cell wall porosity. Multiple models of cell wall structure have proposed a linkage between xyloglucan and pectin. It is proposed that during blueberry fruit abscission the linkage between XGs and cellulose, and XGs and pectins are altered in the AZ in addition to the alteration of the pectic polymers, thereby leading to cell separation.

66 Table 1. Plant cell wall glycan-directed monoclonal antibodies with differential binding OD values upon induction of abscission by ethephon and MeJa.

Monoclonal Antibody Epitope CCRC-M 52 NON-FUCOSYLATED XYLOGLUCAN CCRC-M 54 NON-FUCOSYLATED XYLOGLUCAN CCRC-M 49 NON-FUCOSYLATED XYLOGLUCAN CCRC-M 102 FUCOSYLATED XYLOGLUCAN CCRC-M 39 FUCOSYLATED XYLOGLUCAN CCRC-M 84 FUCOSYLATED XYLOGLUCAN CCRC-M 1 FUCOSYLATED XYLOGLUCAN CCRC-M 111 FUCOSYLATED XYLOGLUCAN CCRC-M 115 XYLAN CCRC-M 117 XYLAN CCRC-M 118 XYLAN CCRC-M 114 XYLAN CCRC-M 154 XYLAN CCRC-M 150 XYLAN CCRC-M 151 XYLAN CCRC-M 140 XYLAN CCRC-M 139 XYLAN CCRC-M 138 XYLAN CCRC-M 137 XYLAN CCRC-M 75 GALACTOMANNAN CCRC-M 174 GALACTOMANNAN CCRC-M 169 ACETYLATED MANNAN CCRC-M 170 ACETYLATED MANNAN JIM 136 HOMOGALACTURONAN JIM 7 HOMOGALACTURONAN CCRC-M 35 RHAMNOGALACTURONAN JIM 3 LM RHAMNOGALACTURONAN CCRC-M 164 LM RHAMNOGALACTURONAN CCRC-M 98 PSCYHOMITRELLA PECTIN JIM 137 RHAMNOGALACTURONAN JIM 101 RHAMNOGALACTURONAN CCRC-M 19 RHAMNOGALACTURONAN CCRC-M 56 RHAMNOGALACTURONAN CCRC-M 16 RHAMNOGALACTURONAN CCRC-M 77 RHAMNOGALACTURONAN JIM 8 ARABINOGALACTURONAN MAC 266 ARABINOGALACTAN

67 Table 2. List of genes and primer sequences used for gene expression profiling cell of wall metabolism related genes in abscission induced blueberry PPJ.

GENE Primer Sequence POLYGALACTURONASE INHIBITING F: GCATGTTTCGGGAGAAATCCAATGTCG R: TTGGACCGGAACAAGCTGACC EXPANSIN LIKE PROTEIN F: TGCTGTAGAAGTGTTTCAGGCGGA R: ATCAGGTTCCAAGTGAGTGGCAGT ALPHA EXPANSIN 8 F: TGACGTCCATTCGGTGTCAGTCAA R: ATTAGAGGGTGCCGCGTTGTAAGA β GLUCANASE -1 F: TGGCGTCGCTTAGGAACCAGATTA R: GGAATGTCCACAATGCCCTTGCTT β GLUCANASE F : GGCTGCTTGTTGGGTAAGA R : CTGGGAAGGCTATTGAGACATAC PHOSPHOLIPASE C F: AGAAGGAAGAGAAAGCCACCGTGA R: AATAGCGAGACAACGGAGCTGTCA PHOSPHOLIPASE D α F: TCAGCCATCCAAATAGCGTCGAGT R: TTCCATTCCTGGAAGCTCCGTGAT PECTATE LYASE -1 F : CTCGTGCTTTGTCACCTCTT R: GGTGAACAACGACTACACTCAT PECTATE LYASE -2 F:CACCGGTGGTGATCGTACCG R: GGTGACTAAACATGAGGATGCACCAG GLUCAN ENDO GLUCOSIDASE -1 F: CTGGAGGACCGTGCTTCAACC R: CTCAAGCCGTGTATTTACACGCACC GLUCAN ENDO GLUCOSIDASE -2 F : CCTCACCCAGAACTGAAGAATAA R : GGCCACTCTCTCAACCAATAA PECTIN METHYLESTERASE -1 F: TCGAGGAGCAGCAGTACACAAACA R: AGGCCTGTGAATCAGTCCATCCAA PECTIN METHYLESTERASE -2 F: GCCATGTCTCCTGCATTTCGGTTT R: CCATGCAAGATTTCAGGTTGGCGT INVERTASE/PECTIN METHYLESTERASE F : ACAACACCCAGTCCACTTC INHIBITOR-1 R : ATTCACCCATCTCCTCAACAC INVERTASE/PECTIN METHYLESTERASE F: GGTTTGTGTGTCATCTGGCCTTCA INHIBITOR-2 R: AGGGAATCAATCTGGGCCATCCAA β GALACTOSIDASE 1 PRECURSOR F: AGTACTGCATTGGGCAGGAATCGT R: AGCGTGTTCTGGATCCATCAGCTA XYLOGLUCAN F: GTTGCCCAATCGTCAGCATTCCAT ENDOTRANSGLUCOSYLASE R: CGTCCTTTGGAACCCACAAAGCAT CELLULOSE SYNTHASE LIKE PROTEIN F: TCTCATCGCTCATCACCTTTCCGT R: AGCCTTACTCGTGCCCATTTGATG POLYGALACTURONASE ISOZYME 1 β F: AAACGAAGGCCAAGATCAACCACG R: ACAATTGGCTCAATCAGCGGTAGC

68 90 Control 80 MeJa Ethephon 70

60

50

40 % Fruit Fruit % Drop 30

20

10

0 24 48 72 Time After Application

Figure1. Fruit drop in response to application of MeJa and Ethephon in ‘Briteblue’ (2011).

69 Peduncle

Pedicel

Figure 2. Anatomy of the PPJ. Red box indicates the approximate region that was sampled in this study.

70 Figure 3. Glycome profiles of Control, MeJa and Ethephon treated PPJ (branch abscission zone) and Pedicel tissues.

71 Figure 4. Immuno-localization of homogalacturonan at 24 h after application of MeJa and Ethephon using JIM 5 and JIM 7 antibodies.

72 Figure 5. Immuno-localization of Xyloglucan at 48 h after application of MeJa and Ethephon using CCRC-M137, CCRC M139, CCRC M153 and LM 11 antibodies.

73 Figure 6. Immuno-localization of Xyloglucan at 24 and 48 h after application of MeJa and Ethephon using CCRC M104 antibodies. The intensity of labeling in ethephon and MeJa treated PPJ increased at 48 h after the treatment application.

74 Figure 7. Immuno-localization of Xyloglucan at 24 and 48 h after application of MeJa and Ethephon using LM 15 antibodies. The intensity of labeling in ethephon and MeJa treated PPJ increased at 48 h after the treatment application.

75 76 Figure 8. Gene expression profiles of cell wall metabolism related genes upon induction of abscission in blueberry.

77 References:

Adams, D. and S. Yang. 1979. Ethylene biosynthesis: identification of 1-aminocyclopropane-1- carboxylic acid as an intermediate in the conversion of methionine to ethylene. Proc.Natl. Acad. Sci. 76(1):170-174.

Agusti, J., P. Merelo, M. Cercos, F.R. Tadeo, and M. Talon. 2008. Ethylene-induced differential gene expression during abscission of citrus leaves. J. Exp. Bot. 59:2717-2733.

Agusti, J., P. Merelo, M. Cercos, F. Tadeo, and M. Talon. 2009. Comparative transcriptional survey between laser-microdissected cells from laminar abscission zone and petiolar cortical tissue during ethylene-promoted abscission in citrus leaves. BMC Plant Biol. 9(1):127.

Atkinson, R.G., R. Schröder, I.C. Hallett, D. Cohen, and E.A. MacRae. 2002. Overexpression of polygalacturonase in transgenic apple trees leads to a range of novel phenotypes involving changes in cell adhesion. Plant Physiol. 129(1):122-133.

Austin, M.E., and Williamson, R.E. 1977. Comparison of harvest methods of rabbiteye blueberries. J. Amer. Soc. Hort. Sci. 102,454-456.

Belfield, E.J., Ruperti, B., Roberts, J.A. and McQueen-Mason, S.J. 2005.Changes in expansin activity and gene expression during ethylene-promoted leaflet abscission in Sambucus nigra. J. Exp. Bot.56, 817-823.

Beno-Moualem, D., Gusev, L., Dvir, O., Pesis, E., Meir, S. and Lichter, A. 2004. The effects of ethylene, methyl jasmonate and 1-MCP on abscission of cherry tomatoes from the bunch and expression of endo-1, 4-β-glucanases. Plant Sci.167 , 499-507. Brummell, D.A., B.D. Hall, and A.B. Bennett. 1999. Antisense suppression of tomato endo-1, 4- β-glucanase Cel2 mRNA accumulation increases the force required to break fruit abscission zones but does not affect fruit softening. Plant Mol. Biol. 40(4):615-622.

Burns, J.K. 2002. Using molecular biology tools to identify abscission materials for citrus. HortScience 37:459-463.

Burns, J.K., D.J. Lewandowski, C.J. Nairn, and G.E. Brown. 1998. Endo 1,4-beta-glucanase gene expression and cell wall hydrolase activities during abscission in Valencia orange. Physiol. Plantarum 102:217-225.

Cai, S. and C.C. Lashbrook. 2008. Stamen abscission zone transcriptome profiling reveals new candidates for abscission control: enhanced retention of floral organs in transgenic plants overexpressing Arabidopsis ZINC FINGER PROTEIN2. Plant Physiol. 146:1305-1321.

Chandler, J.W., M. Cole, A. Flier, and W. Werr. 2009. BIM1, a bHLH protein involved in brassinosteroid signalling, controls Arabidopsis embryonic patterning via interaction with DORNRÖSCHEN and DORNRÖSCHEN-LIKE. Plant Mol. Biol. 69(1-2):57-68.

78 Chini, A., Fonseca, S., Fernandez, G., Adie, B., Chico, J.M., Lorenzo, O., Garcia-Casado, G., Lopez-Vidriero, I., Lozano, F.M., Ponce, M.R., Micol, J.L., and Solano, R. 2007. The JAZ family of repressors is the missing link in jasmonate signalling. Nature 448: 666–671. Botton, A., G. Eccher, C. Forcato, A. Ferrarini, M. Begheldo, M. Zermiani, S. Moscatello, A. Battistelli, R. Velasco, and B. Ruperti. 2011. Signaling pathways mediating the induction of apple fruitlet abscission. Plant Physiol. 155(1):185-208.

Cosgrove, D.J., L.C. Li, H. Cho, S. Hoffmann-Benning, R.C. Moore, and D. Blecker. 2002. The growing world of expansins. Plant Cell Physiol. 43(12):1436-1444.

Costa, G., Dal Cin, V. and Ramina, A. 2006.Physiological, molecular and practical aspects of fruit abscission. In: Webset, A.D. and H.Ramirez (eds.). Proc. Xth IS on Plant Bioregulators in Fruit. Acta Hort. 727. ISHS. del Campillo, E. and Bennett, A.B.1996. Pedicel breakstrength and cellulase gene expression during tomato flower abscission. Plant Physiol. 111, 813–820. Dombrecht, B., G.P. Xue, S.J. Sprague, J.A. Kirkegaard, J.J. Ross, J.B. Reid, G.P. Fitt, N. Sewelam, P., M. Schenk, and J.M. Manners. 2007. MYC2 differentially modulates diverse jasmonate-dependent functions in Arabidopsis. The Plant Cell Online 19(7):2225-2245.

Eisen, M.B., P.T. Spellman, P.O. Brown, and D. Botstein. 1998. Cluster analysis and display of genome-wide expression patterns. Proc. Natl. Acad. Sci. 95(25):14863-14868.

Fry, S., R. Smith, K. Renwick, D. Martin, S. Hodge, and K. Matthews. 1992. Xyloglucan endotransglycosylase, a new wall-loosening enzyme activity from plants. Biochem. J. 282:821- 828.

Gahagan, H., R. Holm, and F. Abeles. 1968. Effect of ethylene on peroxidase activity. Physiol. Plant. 21(6):1270-1279. Gil-Amado, J.A. and M.C. Gomez-Jimenez. 2013. Transcriptome analysis of mature fruit abscission control in olive. Plant Cell Physiol. 54(2):244-269.

González-Carranza, Z.H., C.A. Whitelaw, R. Swarup, and J.A. Roberts. 2002. Temporal and spatial expression of a polygalacturonase during leaf and flower abscission in oilseed rape and Arabidopsis. Plant Physiol. 128(2):534-543.

Gonzalez-Carranza, Z., U. Rompa, J.L. Peters, A.M. Bhatt, C. Wagstaff, A.D. Stead, and J.A. Roberts. 2007. HAWAIIAN SKIRT: An F-Box gene that regulates organ fusion and growth in Arabidopsis. Plant Physiol. 144:1370-1382.

Gonzalez-Carranza, Z.H. and J.A. Roberts. 2012. Ethylene and cell separation processes. Annual Plant Reviews, The Plant Hormone Ethylene. 44:246.

79 Goren, R. 1993. Anatomical, physiological, and hormonal aspects of abscission in citrus. Hort. Rev. 15:145-182.

Hartmond U, Yuan R, Burns JK, Grant A, Kender WJ. 2000. Citrus fruit abscission induced by methyl-jasmonate. J. Amer. Soc. Hort. Sci. 125, 547-552. He, Y., H. Fukushige, D.F. Hildebrand, and S. Gan. 2002. Evidence supporting a role of jasmonic acid in Arabidopsis leaf senescence. Plant Physiol. 128(3):876-884.

Hirsch, S. and G.E. Oldroyd. 2009. GRAS-domain transcription factors that regulate plant development. Plant signaling & behavior 4(8):698-700.

Hinman, R. and J. Lang. 1965. Peroxidase-catalyzed Oxidation of Indole-3-acetic Acid*. Biochem. 4(1):144-158.

Howell GS, Stergios BG, Stackhouse SS, Bittenbender HC, Burton CL. 1976. Ethephon as a mechanical harvesting aid for highbush blueberries (Vaccinium austral Small). J. Amer. Soc. Hort. Sci. 101, 111-115. Iwai, H., A. Terao, and S. Satoh. 2012. Changes in distribution of cell wall polysaccharides in floral and fruit abscission zones during fruit development in tomato (Solanum lycopersicum). J. Plant Res. :1-11.

Jiang, C., F. Lu, W. Imsabai, S. Meir, and M.S. Reid. 2008. Silencing polygalacturonase expression inhibits tomato petiole abscission. J. Exp. Bot. 59(4):973-979.

Kalaitzis, P., S.M. Koehler, and M.L. Tucker. 1995. Cloning of a tomato polygalacturonase expressed in abscission. Plant Mol. Biol. 28(4):647-656.

Kalaitzis, P., T. Solomos, and M.L. Tucker. 1997. Three different polygalacturonases are expressed in tomato leaf and flower abscission, each with a different temporal expression pattern. Plant Physiol. 113(4):1303-1308.

Kende, H. and J. Zeevaart. 1997. The Five" Classical" Plant Hormones. Plant Cell 9(7):1197. Kranz, H.D., M. Denekamp, R. Greco, H. Jin, A. Leyva, R.C. Meissner, K. Petroni, A. Urzainqui, M. Bevan, and C. Martin. 1998. Towards functional characterisation of the members of theR2R3 MYBgene family from Arabidopsis thaliana. The Plant Journal 16(2):263-276.

Lee, Y., P. Derbyshire,‐ J.P. Knox, and A.K. Hvoslef Eide. 2008. Sequential cell wall transformations in response to the induction of a pedicel abscission event in Euphorbia pulcherrima (poinsettia). The Plant Journal 54(6):993‐-1003.

Lewis, M.W., M.E. Leslie, and S.J. Liljegren. 2006. Plant separation: 50 ways to leave your mother. Curr. Opin. Plant Biol. 9:59-65.

80 Li, J. and R. Yuan. 2008. NAA and ethylene regulate expression of genes related to ethylene biosynthesis, perception, and cell wall degradation during fruit abscission and ripening in ‘Delicious’ apples. J. Plant Growth Regul. 27(3):283-295.

Li J, Zhu H, Rongcai Y. 2010. Profiling the expression of genes related to ethylene biosynthesis, ethylene perception, and cell wall degradation during fruit abscission and fruit ripening in Apple. J. Amer. Soc. Hort. Sci. 135(5), 391-401. Li, Y., L. Jones, and S. McQueen-Mason. 2003. Expansins and cell growth. Curr. Opin. Plant Biol. 6(6):603-610.

Lorenzo, O., R. Piqueras, J.J. Sánchez-Serrano, and R. Solano. 2003. ETHYLENE RESPONSE FACTOR1 integrates signals from ethylene and jasmonate pathways in plant defense. The Plant Cell Online 15(1):165-178.

Mainland CM, Kushman LJ, Ballinger WE. 1975. The effect of mechanical harvesting on yield, quality of fruit and bush damage of highbush blueberry. J. Amer. Soc. Hort. Sci. 100, 129-134.

Malladi A, Burns JK. 2008. CsPLDα1 and CsPLDγ1 are differentially induced during leaf and fruit abscission and diurnally regulated in Citrus sinensis. J. Exper. Bot. 59, 3729-3739. Malladi, A., T. Vashisth, and L.K. Johnson. 2012. Ethephon and methyl jasmonate affect fruit detachment in rabbiteye and southern highbush blueberry. HortSci. 47(12):1745-1749.

Mao L, Begum D, Chuang H, Budiman M, Szymkowiak E, Irish E, Wing R. (2000). JOINTLESS is a MADS-box gene controlling flower abscission zone development. Nature 406, 910–913. Marynick, M.C. 1977. Patterns of ethylene and carbon dioxide evolution during cotton explant abscission. Plant Physiol. 59(3):484-489.

McQueen-Mason, S.J. and D.J. Cosgrove. 1995. Expansin mode of action on cell walls (analysis of wall hydrolysis, stress relaxation, and binding). Plant Physiol. 107(1):87-100.

McQueen-Mason, S. and D.J. Cosgrove. 1994. Disruption of hydrogen bonding between plant cell wall polymers by proteins that induce wall extension. Proc. Natl. Acad. Sci. 91(14):6574- 6578.

Meir, S., D.A. Hunter, J. Chen, V. Halaly, and M.S. Reid. 2006. Molecular changes occurring during acquisition of abscission competence following auxin depletion in Mirabilis jalapa. Plant Physiol. 141:1604-1616. Meir, S., S. Philosoph-Hadas, S. Sundaresan, K.V. Selvaraj, S. Burd, R. Ophir, B. Kochanek, M.S. Reid, C. Jiang, and A. Lers. 2010. Microarray analysis of the abscission-related transcriptome in the tomato flower abscission zone in response to auxin depletion. Plant Physiol. 154(4):1929-1956.

81 Miyamoto, K., J. Ueda, K. Yamada, S. Kosemura, S. Yamamura, and K. Hasegawa. 1997. Inhibition of abscission of bean petiole explants by lepidimoide. J. Plant Growth Regul. 16(1):7- 9.

Olsen, A.N., H.A. Ernst, L.L. Leggio, and K. Skriver. 2005. NAC transcription factors: structurally distinct, functionally diverse. Trends Plant Sci. 10(2):79-87.

Pfaffl, M.W. 2001. A new mathematical model for relative quantification in real-time RT–PCR. Nucleic Acids Res. 29(9):e45-e45.

Patterson SE, Bleecker A B. 2004. Ethylene-dependent and -independent processes associated with floral organ abscission in Arabidopsis. Plant Physiol. 134,194 -203.

Patterson, S., M. Butenko, and J. Kim. 2007. Ethylene responses in abscission and other processes of cell separation in Arabidopsis. p. 271-278. In: Advances in Plant Ethylene Research. Springer.

Pech, J., M. Bouzayen, and A. Latché. 2008. Climacteric fruit ripening: ethylene-dependent and independent regulation of ripening pathways in melon fruit. Plant Sci. 175(1):114-120.

Rajani, S. and V. Sundaresan. 2001. The Arabidopsis myc/bHLH gene ALCATRAZ enables cell separation in fruit dehiscence. Current Biol. 11(24):1914-1922.

Ramakers, C., J.M. Ruijter, R.H.L. Deprez, and A.F. Moorman. 2003. Assumption-free analysis of quantitative real-time polymerase chain reaction (PCR) data. Neurosci. Lett. 339(1):62-66.

Rasori, A., B. Bertolasi, A. Furini, C. Bonghi, P. Tonutti, and A. Ramina. 2003. Functional analysis of peach ACC oxidase promoters in transgenic tomato and in ripening peach fruit. Plant Sci. 165(3):523-530.

Rieu, I. and S.J. Powers. 2009. Real-time quantitative RT-PCR: design, calculations, and statistics. The Plant Cell Online 21(4):1031-1033.

Roberts, J.A., K.A. Elliott, and Z.H. Gonzalez-Carranza. 2002. Abscission, dehiscence, and other cell separation processes. Annu. Rev. Plant Biol. 53(1):131-158.

Ruijter, J., S. van der Velden, and A. Ilgun. 2009. LinReg PCR (11.0). Analysis of quantitative.

Sakamoto, M., I. Munemura, R. Tomita, and K. Kobayashi. 2008. Involvement of hydrogen peroxide in leaf abscission signaling, revealed by analysis with an in vitro abscission system in Capsicum plants. The Plant Journal 56(1):13-27.

Sane, A.P., S.K. Tripathi, and P. Nath. 2007. Petal abscission in rose (Rosa bourboniana var Gruss an Teplitz) is associated with the enhanced expression of an alpha expansin gene, RbEXPA1. Plant Sci. 172(3):481-487.

82 Saniewski, M., J. Ueda, and K. Miyamoto. 2000. Methyl jasmonate induces the formation of secondary abscission zone in stem of Bryophyllum calycinum Salisb. Acta Physiol. Plantarum 22(1):17-23.

Sasaki, Y., E. Asamizu, D. Shibata, Y. Nakamura, T. Kaneko, K. Awai, M. Amagai, C. Kuwata, T. Tsugane, and T. Masuda. 2001. Monitoring of methyl jasmonate-responsive genes in Arabidopsis by cDNA macroarray: self-activation of jasmonic acid biosynthesis and crosstalk with other phytohormone signaling pathways. DNA Research 8(4):153-161.

Schaller, F., C. Biesgen, C. Müssig, T. Altmann, and E.W. Weiler. 2000. 12-Oxophytodienoate reductase 3 (OPR3) is the isoenzyme involved in jasmonate biosynthesis. Planta 210(6):979-984.

Spellman, PT. 2003. Cluster analysis and display. In D Bowtell, J Sambrook, eds, DNA Microarrays: A Molecular Cloning Manual. Cold Spring Harbor Laboratory Press, Woodbury, NY, pp 569–581

Steiner Lange, S., U.S. Unte, L. Eckstein, C. Yang, Z.A. Wilson, E. Schmelzer, K. Dekker, and H. Saedler. 2003. Disruption of Arabidopsis thaliana MYB26 results in male sterility due to non dehiscent‐ anthers. The Plant Journal 34(4):519-528.

Takeda,‐ F., G. Krewer, E. Andrews, D. Peterson, and J. Mullinix. 2008. Assessment of v45 over- the-row mechanical harvester on pruned rabbiteye and southern highbush blueberry. Hort. Technol. 18:130-138.

Taylor, J.E. and C.A. Whitelaw. 2001. Signals in abscission. New Phytol. 151:323-339. Terol, J., A. Conesa, J.M. Colmenero, M. Cercos, F.R. Tadeo, J. Agusti, E. Alos, F. Andres, G. Soler, J. Brumos, and D.J. Iglesias. 2007. Analysis of 13000 unique Citrus clusters associated with fruit quality, production and salinity tolerance. BMC Genomics 8:31.

Tonutti, P., L. Cass, and R. Christoffersen. 1995. The expression of cellulase gene family members during induced avocado fruit abscission and ripening. Plant, Cell Environ. 18(6):709- 713.

Tucker, M.L., S.L. Baird, and R. Sexton. 1991. Bean leaf abscission: tissue-specific accumulation of a cellulase mRNA. Planta 186(1):52-57.

Ueda, J., K. Miyamoto, and M. Hashimoto. 1996. Jasmonates promote abscission in bean petiole expiants: Its relationship to the metabolism of cell wall polysaccharides and cellulase activity. J. Plant Growth Regul. 15(4):189-195.

U.S. Department of Agriculture. 2012. Non-citrus fruits and nuts 2011 summary. National Agricultural Statistics Service, Washington, DC.

Vanholme, B., W. Grunewald, A. Bateman, T. Kohchi, and G. Gheysen. 2007. The tify family previously known as ZIM. Trends Plant Sci. 12(6):239-244.

83 Vashisth, T., L.K. Johnson, and A. Malladi. 2011. An efficient RNA isolation procedure and identification of reference genes for normalization of gene expression in blueberry. Plant Cell Rep. 30(12):2167-2176.

Vashisth, T. and A. Malladi. 2013. Fruit Detachment in Rabbiteye Blueberry: Abscission and Physical Separation. J. Am. Soc. Hort. Sci. 138(2):95-101.

Ward, D.L., E.P. Beers, R.E. Byers, and R.P. Marini. 1999. Cutting apple fruits induces cellulase activity in the abscission zone. Hort. Sci. 34(4):601-603.

Wei PC, Tan F, Gao XQ, Zhang XQ, Wang GQ, Xu H, et al.2010.Overexpression of AtDOF4.7, an Arabidopsis DOF Family Transcription Factor, Induces Floral Organ Abscission Deficiency in Arabidopsis. Plant Physiology 153, 1031-1045. Whitney, S.E., M.G. Gothard, J.T. Mitchell, and M.J. Gidley. 1999. Roles of cellulose and xyloglucan in determining the mechanical properties of primary plant cell walls. Plant Physiol. 121(2):657-664.

Woodson, W.R., K.Y. Park, A. Drory, P.B. Larsen, and H. Wang. 1992. Expression of ethylene biosynthetic pathway transcripts in senescing carnation flowers. Plant Physiol. 99(2):526-532.

Yang, S., B. Vanderbeld, J. Wan, and Y. Huang. 2010. Narrowing down the targets: towards successful genetic engineering of drought-tolerant crops. Molecular plant 3(3):469-490.

84 CHAPTER 4

TRANSCRIPTOME WIDE CHANGES DURING THE INDUCTION OF ABSCISSION IN

BLUEBERRY FRUIT1

1Vashisth, T. and A. Malladi. 2013. To be submitted to Journal of ASHS

85 Abstract

Abscission is a physiological process that involves the programmed separation of entire organs at an anatomically distinct layer called the abscission zone (AZ). Induction of abscission depends on a complex interplay among hormones and other developmental signals. Our objective was to identify the transcriptome wide changes that occur during fruit abscission in blueberry. Ethephon

(1000 mg L-1) and Methyl Jasmonate (MeJa; 20 mM) were used to induce abscission. Around

60% of the fruit abscised in response to Ethephon and MeJa treatments. The AZ tissue was collected at 48 h after treatment. RNA-Seq analysis was performed to characterize changes in the transcriptome of blueberry AZ in response to Ethephon and MeJa. Abscission agent induced abscission was associated with extensive changes in the expression of genes associated with the biosynthesis and signaling of phytohormones such as ethylene, jasmonic acid and auxin. In case of MJ, genes from the transcription factors families such as AP2/ERF, GRAS, MYB, MYC, and

WRKY were up-regulated while zinc finger (ZF) were down- regulated. However, in case of

Ethephon, only the transcription factors belonging to the AP2/ERF and GRAS showed increased expression. Most of the cell wall metabolism-related genes such as POLYGALACTURONASE, β-

GLUCANASE, XYLANASE, and EXPANSIN were up-regulated upon induction of abscission by

Ethephon and MJ. Further, quantitative real time PCR analysis was performed on 22 selected genes to validate the RNA-Seq results. More than 50% of qRT-PCR data validates RNA-Seq data. Overall, the transcriptomics analyses suggest that fruit abscission in blueberry is associated with changes in phytohormone signaling and cell wall metabolism. Also, potentially a plant hormone cross-talk and interaction plays an important role in abscission.

Key words: abscission, blueberry, RNA-seq, ethephon, MeJa, transcriptome

86 Introduction

Abscission is a process of detachment of plant organs such as leaves, flowers, flower parts, fruits and seeds from the parent body. The process of separation is primarily under developmental and hormonal control, but is also strongly affected by the environment (Taylor and Whitelaw 2001;

Lewis et al., 2006). Organ separation occurs at abscission zones (AZ). Generally, abscission can be divided into four major steps (Patterson, 2007). During the first step formation of the AZ occurs. In step 2, AZ cells become responsive to phytohormones such as ethylene (by far the most extensively studied), jasmonates, auxin and abscisic acid (ABA). In response to the phytohormone signaling the middle lamella is dissolved. Also, many studies reported that ethylene and other phytohormones are involved in inducing cell wall-degrading enzymes such as endo-β-1, 4-glucanase (EG), polygalacturonases (PG), cellulases, and β–galactosidases, as well as lipid-modifying phospholipases in abscission zones (Goren, 1993; Malladi and Burns, 2008).

Increase in the transcript levels of MdPG2, and MdEG1 in the fruit AZ has been found to be concomitant with the levels of transcript levels of MdACS5A, MdACO1, MdETR2, and MdERS2 in apple over the period of ripening and abscission (Li et al., 2010). During step 3, cells in the

AZ expand and cell wall loosening occurs. Finally in step 4, cell separation occurs followed by the suberization of the proximal layer of cells to form the protective layer. Mutants defective in

AZ formation have been identified. Loss of JOINTLESS, a MADS box gene, results in suppression of the pedicel AZ formation in tomato (Mao et al., 2000). During abscission, the AZ cells respond to the abscission signals, such as changing levels of phytohormones e.g. ethylene, jasmonate and auxin, and activate cell wall loosening proteins (such as cell wall hydrolases). It has been reported that abscission is regulated by two pathways: ethylene dependent and ethylene independent (Patterson, 2004). It is well documented that ethylene promotes abscission in many

87 crops like citrus (Agusti et al. 2008, 2009; Burns, 2002), blueberry (Malladi et al. 2012) and tomato (Beno-Moualem et al., 2004). Genes related to ethylene biosynthesis, signaling and perception are differentially expressed during abscission (Agustί et al., 2008). They reported the differential expression of various ethylene related genes in the petiole and the leaf AZ. In a recent study, the expression profiles of genes related to ethylene biosynthesis, perception and cell wall degradation in fruit abscission and ripening of apple were studied by Li et al. (2010). They reported an increase in transcript levels of MdACS5A, MdACO1, and ethylene receptors,

MdETR2, and MdERS2, in the fruit abscission zone over the period of 3 months of fruit ripening.

Defects in ethylene signaling and perception pathways have been reported to affect abscission in various ways. In ethylene insensitive mutants such as etr1 and ein2, both floral organ abscission and senescence is delayed (Patterson and Bleecker, 2004). DNA binding with one finger (DOF) is a family of transcription factors. In Arabidopsis, a member of this family, DOF4.7, is expressed in floral abscission zones. Arabidopsis plants constitutively expressing AtDOF4.7, exhibit an ethylene-independent deficiency in floral organ abscission (Wei et al., 2010).

Overexpression of AtDOF4.7 results in abscission deficiency in floral organ in Arabidopsis by inhibiting dissolution of middle lamella in the AZ cells indicating that it controls abscission by regulating enzymes related to cell wall hydrolysis (Wei et al., 2010). In Arabidopsis,

INFLORESCENCE DEFICIENT IN ABSCISSION (IDA) is another gene family which is also known to regulate abscission through an ethylene insensitive pathway (Butenko et al., 2003).

Another group of plant hormones reported to be involved in abscission is the Jasmonates family consisting of jasmonic acid (JA) and methyl jasmonate (MeJa) among others. Many studies have shown the involvement of JA and MeJa, the primary jasmonates, in regulating plant development and senescence. MeJa has been reported to result in loosening and abscission in

88 blueberry (Malladi et al., 2012), citrus (Hartmond et al., (2000), and cherry tomato (Beno-

Moualem et al., 2004).

The interaction among multiple phytohormones such as ethylene, auxin and possibly jasmonic acid plays an important regulatory role in abscission. The AZ becomes sensitive to ethylene only when the auxin concentration goes below a critical level. Rasori et al. (2003) reported that the down regulation of PpAux/IAA2 is paralleled by an up-regulation of Pp-ACO1 in peach during the initiation and progression of ripening. In apple down regulation of Aux/IAA7, was accompanied with decreased levels of auxin and enhanced fruitlet shedding (Costa et al.,

2006). Meir et al. (2010) reported that the process of abscission is initiated upon changes in auxin gradient across the abscission zone and is elicited by ethylene in the tomato flower abscission zone, indicating a cross-talk of auxin-ethylene signaling. In another study by Meir et al. (2006) in Mirablis jalapa, a relation between auxin depletion and increase in ethylene sensitivity was observed. Sanieweski et al. (2000) reported the formation of secondary abscission zone in the stem of Bryophyllum calycinum upon treatment with MeJa, and the formation of secondary abscission zone induced by Meja was prevented upon treatment with Indole-3-acetic acid (IAA), an auxin. Ueda et al. (1996) reported abscission induced by MeJa is not mediated by an increase in ethylene production. This study also reported that the MeJa treatment did not enhance ethylene production in bean petiole explants. However, Beno-Moualem et al. (2004) reported that 1-methylcyclopropene (1-MCP), an ethylene perception inhibitor, blocks the effect and expression of genes stimulated by MeJa during abscission. In Arabidopsis, exogenous application of JA results in premature senescence, by potentially lowering auxin levels or inducing ethylene biosynthesis (He et al., 2002). Cai and Lashbrook (2008) concluded that MeJa or JA biosynthesis is not involved in stamen abscission. Interaction between ethylene and

89 jasmonates remains poorly understood and it is still unclear if the MeJa and ethylene regulate abscission through a common or through diverse signaling pathways.

Blueberry is a fruit crop indigenous to the United States valued at over $780 million in

2011 (USDA, 2012). Fruit abscission in rabbiteye blueberry occurs at the pedicel-peduncle junction (PPJ; Vashisth and Malladi, 2013). Ethephon and Methyl Jasmonate (MeJa) are two abscission agents that are effective in inducing abscission in blueberry (Malladi et al., 2012).

Due to the limited availability of blueberry genomic information, horticulturally important traits like abscission/fruit detachment are not well studied. Currently there is no information on transcriptome-wide changes that occur during the induction of abscission in blueberry or any other closely related species. Enhancing our knowledge of the process of fruit detachment mechanism can greatly aid in developing methods to increase the efficiency of mechanical harvesting, which can in turn reduce the costs associated with blueberry production (Austin and

Williamson 1977; Howell et al., 1976; Mainland et al., 1975; Takeda et al., 2008). RNA-Seq is a relatively new and powerful transcriptomics tool that can provide valuable insights into the transcriptome-wide changes in gene expression. Therefore the objective of this study was to identify and compare transcriptome wide changes that occur during ethylene- and Meja-induced fruit abscission in blueberry.

Materials and Methods

Plant Material for RNA-Seq and Quantitative Real Time-PCR (qRT-PCR) analyses

Mature rabbiteye blueberry plants of ‘Powderblue’ grown at the University of Georgia,

Horticulture Farm, Watkinsville, GA were used in this study. Experiments were performed in

2010 for RNA-Seq analysis and in 2013 for qRT-PCR analysis. Experiments were designed as completely randomized design with three treatments. The treatments were: 1) control; 2) methyl

90 -1 jasmonate (MeJa, 20 mM; Sigma-Aldrich, St. Louis, MO); and 3) ethephon (1000 mg·L ; Bayer

CropScience, Kansas City, MO). The treatments were performed when majority (~75%) of the fruit on the plant was mature. All treatments, including the control, were applied along with

0.15% of the adjuvant (Latron B-1956, Rohm and Haas, Philadelphia, PA) and were performed using a hand pump sprayer until run-off around 0900 HR. The average daily temperature on the day of application was 26 and 27.5 °C in 2010 and 2013 respectively. For RNA-Seq analysis, the

PPJ tissue were collected from about 50 fruit at 48 h after treatment and for qRT-PCR analysis

PPJ tissue from about 50 fruit were collected at 0, 24 and 48 h after the treatment. PPJ tissues were manually dissected. Three replicates were used for RNA-Seq analysis while four replicates were used for the qRT-PCR analysis. Dissected PPJ tissues were pooled within each replicate and frozen in liquid N2.

RNA Extraction:

RNA extraction was performed as described by Vashisth et al., (2011) with some modifications.

Briefly, 0.1 g of the sample was ground to a fine powder and was added to the 0.75ml of the extraction buffer. The RNA extraction buffer consisted of 2% cetyltrimethylammonium bromide

(CTAB), 2% polyvinylpyrrollidone (PVP), 100 mM Tris-HCl, 25 mM ethylenediaminetetraacetic acid (EDTA), 2 M sodium chloride (NaCl), 3.44 mM spermidine, and

2% β-mercaptoethanol. All solutions were prepared using 0.1% diethylpyrocarbonate (DEPC) treated water. The buffer was warmed to 65 °C prior to the addition of the ground tissue. The sample-buffer slurry was vortexed briefly and incubated at 65 °C for 10 min, vortexed again and incubated at room temperature for 5 min. The mixtures were extracted twice with an equal volume of chloroform: isoamyl alcohol (24:1) with centrifugation at 4 °C for 15 min (5000 x g).

The supernatant fraction was transferred to sterile tubes followed by the addition of 0.25

91 volumes of lithium chloride (LiCl; 10 M). The samples were mixed gently and precipitated overnight at 4 °C. Samples were subsequently centrifuged at 4 °C for 20 min (12000 x g). After discarding the supernatant fraction, the RNA pellets were washed twice with 500 µL of ice cold

70% ethanol, and dissolved in DEPC and stored at -80 °C.

RNA-Seq analysis:

Three replicates within each treatment (control, MeJa, ethephon) were used for RNA-Seq analysis. cDNA library synthesis and RNA-Seq analysis was performed at Georgia Genomics

Facility, Athens, GA. cDNA library preparation was done using Trueseq stranded mRNA kit.

RNA-seq analysis was performed on Illumina Hi seq paired end sequencing (100 base pairs) was performed in this study. One of the replicates of MeJa was subsequently excluded owing to poor

RNA quality.

RNA-Seq data analysis:

The quality of sequencing reads was assessed using FastQC. Trimmomatic was used for quality trimming. Trinity (Version r2013-02-25) was used for assembling reads and all reads were mapped back to the assembly. Each library was mapped and differential expression was calculated using the Trinity/RSEM pipeline and the EdgeR modules. Average fragments per kilobase per million reads (FPKM) for each sample were generated. Expression clusters were generated manually using Trinity R plugin with the hierarchical cluster profiles as input.

Annotation of transcripts with blastx versus the NCBI_nr database and also against the gene ontology (GO) database using Blast2GO was performed. Further a cutoff for differential expression was set up at a false discovery rate (FDR) of 0.001 and Log2-fold relative expression.

92 Reverse Transcription and Gene Expression Analysis:

A total of 4 replicates within each treatment were used. For cDNA synthesis genomic DNA contamination was removed by treating 1 µg of total RNA was with DNase (Promega; 37 °C for

34 min). The DNase-treated RNA was reverse transcribed using Im-Prom II Reverse

Transcriptase (Promega) and oligo dT (Promega) in a 20 µL reaction volume. Reverse transcription was performed at 42 °C for 75 min, followed by incubation at 75 °C for 15 min.

The cDNA samples were subsequently diluted 10-fold and stored at -20 °C until further analysis.

Gene expression analysis was performed by quantitative Real Time-PCR (qRT-PCR). Twenty genes displaying differential expression in the RNA-Seq analysis were selected for this study.

The primers for qRT-PCR were designed using PrimerQuest (Integrated DNA Technologies,

Inc.). The amplicon size targeted from 120-200 bases. All qRT-PCR reactions were performed using the Stratagene Mx3005P real-time PCR system (Agilent Technologies, Inc., Santa Clara,

CA). A reaction volume of 12 µL with 1 µL of diluted cDNA and 6 µL of 2X Veriquest SYBR

Green Master Mix (Affymetrix) was used. The thermal cycling conditions were: 95 °C for 10 min; 40 cycles of 94 °C for 30 s and 60 °C for 1 min. Melt-curve analysis was performed at the end of the qRT-PCR reactions to check for primer specificity. Controls without a template and without the reverse transcriptase were run for each gene. The primer efficiency for each gene was determined using LinRegPCR (Ramakers et al., 2003; Ruijter et al., 2009). Relative expression was calculated using a modified Pfaffl method (Pfaffl, 2001) and as described in Rieu and Powers (2009). Relative quantity (RQ) for each sample was calculated using the formula,

1/ECq, where Cq is the quantification cycle (threshold cycle). The RQ was normalized using RNA

HELICASE-LIKE 8 (RH8) and CLATHRIN ADAPTOR COMPLEXES MEDIUM SUBUNIT

FAMILY PROTEIN a (CACSa) as described by Vashisth et al. (2011). The geometric mean of

93 expression of the above two reference genes (normalization factor) was used for normalization.

The normalized RQ (NRQ) values were log 2 transformed and used for statistical analyses. The standard error of the means was calculated as described in Rieu and Powers (2009).

Statistical Analysis for gene expression: Statistical analyses were performed SigmaPlot 11

(Systat Software Inc., San Jose, CA). Two-way ANOVA with the general linear model: time, treatment, and time×treatment (three levels of time, 0, 24, and 48 h after treatment; and three levels of treatment, ‘Control, Ethephon and MeJa) was used for analyzing gene expression data for (n = 4), and was followed by Fisher’s LSD multiple comparison test for mean separation (α =

0.05). NRQ values (log2 transformed) were used for the above analyses.

Results and Discussion

MeJa and Ethephon resulted in abscission at the pedicle–peduncle junction (PPJ). Both the abscission agents resulted in significant fruit drop in blueberry within 48 h of treatment in both year ‘2010’ and ‘2013’ (Figure 1). In preliminary studies it was observed that MeJa resulted in fruit drop by 24 h after treatment whereas ethephon resulted in significant fruit drop around 48 h after treatment. Similar effect of MeJa and ethephon in blueberry has been reported by Malladi et al. (2012). Since 48 h after the application of treatment was the optimal time point when both the treatments resulted fruit drop, therefore PPJ samples for RNA-Seq analysis were collected at 48 h after the application. To identify differences in the PPJ transcriptome in relation to induction of abscission, differential expression was calculated by comparison with control PPJ transcriptome.

After the RNA-Seq analysis 136,059 good quality reads were obtained from control, ethephon and MeJa treated PPJ samples. These sequences were assembled and all reads were mapped back to the assembly and resulting relative expression profile were used to parse out transcripts that represent at least 1% of the isoform percentage for each transcript resulting in 98,217 transcripts.

94 A total of 14,335 unique isotigs were found to be present in all three: control, MeJa and ethephon treated PPJ transcriptome out of which only about 10, 770 were identified by performing BLAST analysis of these sequences. About 25% of the isotigs did not match to any NCBI data entry and therefore represent a potential source for novel gene discovery. A heat map (Figure 2) was generated for control, ethephon and MeJa-treated PPJ transcriptomes to show the differential expression pattern using hierarchical clustering. MeJa and ethephon-treated PPJ transcriptomes were considerably different in comparison to the control. However, the overall transcriptome of the ethephon-treated PPJ resembled the control PPJ more closely than that of the MeJa PPJ transcriptome.

Cluster analysis was performed to identify distinct biological processes that were altered by the application of the abscission agents. Cluster analysis is a statistical procedure that identifies and organizes the patterns contained within complex data sets (Spellman, 2003). In a large gene transcript population, clusters are formed by genes that demonstrate similar gene expression pattern (Eisen et al., 1998). Often functionally related transcripts are housed together within the same cluster and thus knowledge of biological function of even a subset of transcripts within a cluster can provide important clues about potential roles for non-characterized transcript within the same cluster (Cai and Lashbrook, 2005). Figure 3 shows the eight expression clusters that were manually generated using the hierarchical cluster profile. Each cluster represents the group of genes with similar gene expression pattern within control, ethephon and MeJa treated

PPJ. Clusters 3, 5, 6, 7 and 8 represent the up-regulated genes with MeJa-treated PPJ and remaining clusters 1, 2, and 4 represents the clusters of genes that were down-regulated in MeJa- treated PPJ. Clusters for ethephon and control-treated PPJ looked similar for 1, 6 and 7. Clusters

3, 4 and 5 represents gene that were up-regulated in ethephon-treated PPJ and clusters 2 and 8

95 comprises of genes that were down-regulated in ethephon-treated PPJ. Clusters 4 and 8 were interesting as the genes in these clusters were inversely regulated by MeJa and ethephon treatments. Most of the genes in the up-regulated clusters belonged to carbohydrate metabolic processes, and response to stress and oxidation reduction process. Down-regulated clusters included genes with biological processes such as biosynthetic processes, various metabolic processes, regulation of transcription, and genes with unknown functions.

Differential expression of genes was determined for the following comparisons: control versus ethephon (ethephon group) and control versus MeJa (MeJa group). More than 6500 and

7500 genes were differentially expressed in ethephon and MeJa treated PPJ, respectively, in comparison to the control. To ensure that genes with differential expression are abscission induced, a set of stringent cut-offs including a FDR of 0.001 and a log 2 fold difference in expression was used. Using these cut-offs, 190 and 354 genes were differentially expressed in the ethephon and the MeJa-treated PPJ transcriptomes in comparison to the control.

Categorization based on the GO (Gene Ontology) ID of these differentially expressed genes in ethephon and MeJa-treated PPJ was performed (Figure 4). Genes were categorized on the basis of biological process (Figure 4a) and molecular function (Figure 4b). Categorization based on the biological processes revealed that majority of the genes was enriched with gene ontology terms

(GO) such as metabolic processes, response to stress, carbohydrate metabolic processes, oxidation-reduction processes and biosynthetic processes. Around 21% of the genes were involved in unknown processes in both ethephon and MeJa-treated PPJ transcriptomes.

Interestingly, in both differential gene expression groups, a similar percentage of genes were involved in response to stress, oxidation-reduction process, carbohydrate metabolic processes, plant hormone mediated signaling and cytokinesis. Higher numbers of differentially expressed

96 genes were involved in metabolic processes in response to ethephon. A higher number genes involved in the regulation of transcription were differentially expressed in response to the MeJa treatment. Enriched GO terms in ethephon and MeJa induced PPJ abscission under molecular function were binding, ion binding, oxidoreductase activity, hydrolase activity, transferase activity and transcription factor activity. Thirty five and twenty two percent of the genes were of unknown molecular function in MeJa and ethephon groups. Diverse molecular functions were observed in case of the MeJa group as compared to the ethephon group. A very similar GO categorization was found during mature fruit abscission of olive. In olive, the transcriptome was enriched in GO terms such as metabolic process, oxidation-reduction process, regulation of transcription, carbohydrate metabolic process and transport under biological processes (Gil-

Amado and Gomez-Jimnez, 2013). They also reported the most common molecular function GO terms were catalytic activity, transferase activity, oxidoreductase activity, binding, ion binding and hydrolase activity. The GO classification of genes during stamen abscission in Arabidopsis showed a significant increase in the transcription factor activity under molecular function and in biological process a significant increase was observed in genes involved in response to stress/stimuli, transcription and signal transduction (Cai and Lashbrook, 2008). The GO categorization of ethephon and MeJa induced genes is in agreement with the above studies in olive and Arabidopsis.

Carbohydrate and cell wall metabolic processes during induction of abscission

A large number of genes such as cell wall hydrolases, transferases and lyases were altered in the

PPJ 48 h after ethephon and MeJa treatments. In MeJa induced abscission, twenty four carbohydrate metabolic processes and cell wall modifications related genes were up-regulated.

Two genes involved in the pectinesterase inhibitor activity were down-regulated during MeJa-

97 induced abscission. The up-regulated genes included POLYGALACTURONASE, ENDO-β-

XYLANASE, β-GALACTOSIDASE, GLUCAN- β-GLUCOSIDASE, EXPANSIN, β-GLUCANASE,

INVERSTASE, XYLOGLUCAN ENDOTRANSGLUCOSYLASE and CELLULOSE SYNTHASE. In ethephon induced abscission, fifteen cell wall and carbohydrate metabolism related genes were differentially expressed. Out of these fifteen genes, thirteen genes were up-regulated and two genes were down-regulated. The down-regulated two genes were β-GLUCOSIDASE and β-

GLUCOSIDASE LIKE. The genes that were up-regulated included POLYGALACTURONASE,

ENDO-β-XYLANASE LIKE, α-GALACTOSIDASE, β-GLUCOSIDASE, β–EXPANSIN

PRECURSOR and β-GLUCANASE. Overall, ethephon and MeJa induced abscission resulted in up-regulation of genes from same families except for β-GLUCOSIDASE which was down- regulated in case of ethephon and up-regulated in MeJa induced abscission suggesting that β-

GLUCOSIDASE plays different roles in abscission induced by these abscission agents. In case of both ethephon and MeJa induced abscission a LACCASE LIKE gene was up-regulated.

LACCASE is a multi-copper glycoprotein oxidase. Its molecular function involves copper ion binding and it is involved in the biological process of lignin metabolism. However, its role in abscission is unknown. An increase in LACCASE was observed during mature olive fruit abscission (Gil-Amado and Gomez-Jimnez, 2013) further suggesting that LACCASE is associated with the progression of abscission. These results suggest that enzymes that are coded by these cell wall and carbohydrate metabolism related genes may be required for ensuring cell separation in mature fruit abscission and possibly for wall restructuring after AZ cell separation.

The above results agree with previous studies on the expression of genes encoding abscission-associated cell wall-hydrolyzing enzymes (Gonzalez-Carranza and Roberts, 2012). In the current study, an increase in the expression of ENDO- POLYGALACTURONASE (PG) and β-

98 GLUCANASE (EG; also commonly known as CELLULASE) was observed in response to MeJa and ethephon. Potentially this PG is directly involved in the process of abscission due to its similarity with ABSCISSION POLYGALCTURONASE PRESUCRSOR from Solanum lycopersicum. The role of PG and EG during the process of abscission has been implicated in many previous studies like mature olive fruit abscission (Gil-Amado and Gomez-Jimnez, 2013),

Arabidopsis stamen abscission (Cai and Lashbrook, 2008), apple fruit abscission (Li and Yuan,

2008) and tomato flower abscission (Meir et al., 2010). Tonutti et al. (1995) reported an increase in the expression of EG during mature avocado abscission. Three different PGs were shown to be involved in tomato flower abscission by Kalaitzis et al. (1997). Silencing of PG has been reported to result in delayed petiole abscission in tomato (Jing et al., 2008) and floral abscission in Arabidopsis (Gonzalez-Carranza et al., 2007), suggesting the importance of PG in leaf abscission. Burns et al. (1998) studied cellulase and polygalacturonase activity in the fruit abscission zone of Valencia orange and concluded that these enzyme activities were associated with the process of abscission. Apple fruit abscission induced by wounding or application of ethephon was associated with increased cellulase activity in the fruit AZ (Ward et al., 1999). An increase is the expression of PG and EG has been reported to be associated with the ethylene induced abscission in citrus (Agusti et al., 2009). An increase in the expression of EG in cherry tomato along with increased abscission after exogenous application of MeJa and ethylene was reported by Ben-Moualem et al. (2004). In the same study application of 1-MCP (an ethylene perception inhibitor) was shown to decrease the rate of abscission as well as expression of EG.

Li and Yuan (2008) also demonstrated that pre-harvest application of 1-MCP (an ethylene perception inhibitor) resulted in the decrease in the expression of PG and EG in apple along with decrease in the fruit abscission, suggesting that potentially ethylene plays an important role in

99 abscission. Hence, it may be proposed that these genes play important roles in cell separation during fruit abscission in blueberry.

GALACTOSIDASE and GLUCOSIDASE were up-regulated in ethylene induced citrus leaf abscission (Agusti et al., 2009). In the current study, an increase in the expression of

GALACTOSIDASE and GLUCOSIDASE was observed in ethephon and MeJa treated PPJ tissues indicating that these genes potentially play an important role in plant hormone induced abscission.

Xyloglucan-endotransglucosylases (XET) and Xylanses are enzymes which are involved in breaking down hemicelluloses (one of the major components of plant cell walls). XET play an important role in the alteration of cell wall mechanical properties and its reorganization (Fry,

1992). In the present study, an increase in the expression XYLANASE was observed after the application of ethephon and MeJa whereas the expression of XET increased only in MeJa-treated

PPJ. Up-regulation of XET expression upon induction of abscission has been shown to occur during citrus leaf abscission (Agusti et al., 2008), Arabidopsis stamen abscission (Cai and

Lashbrook, 2008) and mature olive fruit abscission (Gil-Amado and Gomez-Jimnez, 2013).

XYLANASE has not been previously shown to be related with abscission but alteration in the hemicelluloses is well correlated with the induction of abscission (Lee et al., 2008; Iwai et al.

2012). Therefore it is suggested that XYLANASE plays an important role in the process of abscission by altering cell wall hemi-celluloses.

Expansins are proteins that are involved in the regulation of wall extension during plant cell growth (Cosgrove et al., 2002; Li and McQueen, 2003). Expansins possibly act by disrupting hydrogen bonds between cellulose microfibrils and hemicelluloses that tether them to one another in the plant cell wall (McQueen-Mason and Cosgrove 1994, 1995; Whitney et al., 2000).

100 In present study an increase in the expression of EXPANSINs was observed in both ethephon and

MeJa treated PPJ, suggesting that enhanced abscission is complemented by up-regulation of

EXPANSINs. Similarly, EXPANSINS may play an important role in ethylene promoted leaflet abscission in Sambucus nigra (Belfield et al., 2005). This study reported an increase in the activity of expansin in tissues undergoing cell separation. Agusti et al. (2009) also reported an increase in the expression of EXPANSIN during ethylene promoted citrus leaf abscission. An increase in expression of EXPANSIN has also been reported during apple fruitlet abscission

(Botton et al., 2011).

Regulation of Transcription

A number of gene transcripts belonging to different families of transcription factors were differentially expressed in the PPJ within 48 h of ethephon and MeJa treatment. The number of transcription factors expressed in the MeJa-treated PPJ was significantly higher than those in ethephon-treated PPJ tissues. MeJa treatment resulted in the increased expression of four

AP2/ethylene response factors (ERFs), another AP2 domain transcription factor, MYB transcription factor (MYB), basic helix-loop-helix (bHLH) transcription factors- MYC2 like

(MYC) and BIM1 LIKE (BIM), a helix loop helix (HLH)transcription factor BPE-LIKE (BPE),

WRKY transcription factor (WRKY) and two transcription factors from NAC-Domain superfamily

(NAC). Transcription factors belonging to zinc-finger family (ZF) were down-regulated in MeJa- treated PPJ. Transcription factor of the GRAS family as well as a probable WRKY family transcription factor was down-regulated in response to MeJa treatment. A few other transcription factors were also up-regulated but thier family information remains unknown.

In ethephon-treated PPJ up-regulated transcription factors included - AP2/ERF, GRAS and NAC-domain family genes. Surprisingly, the transcription factor, BIM, was down-regulated

101 in ethephon treated PPJ. Few other transcription factors were altered but there family information still remains unknown.

Overall these results are in agreement with several previous studies on abscission.

Expression of ERF is known to be induced by jasmonic acid and ethylene during plant defense responses (Lorenzo et al., 2003). ERF has been shown to be up-regulated in stamen abscission in

Arabidopsis (Cai and Lashbrook, 2008), apple fruitlet abscission (Botton et al., 2011), and ethylene induced citrus leaf abscission (Agusti et al., 2009). On the contrary Gil-Amado and

Gomez-Jimenez (2013) reported down-regulation in AP2/ERF transcription factors during olive fruit abscission. In the present study, more than one AP2 Domain/ERF were several fold up- regulated in response to both ethephon and MeJa treatment, suggesting that AP2/ERF transcription factors possibly play important roles in plant growth regulator induced abscission.

MYB transcription factors were up-regulated in MeJa treated PPJ. MYB factors play a central role in the control of gene transcription involved in many processes including cell separation. Steiner-

Lange et al. (2003) demonstrated the involvement of AtMYB26 in the regulation of the swelling and lignification of the anther cell layer, which is essential for dehiscence. MYB transcription factors are also associated with the stamen abscission in Arabidopsis (Cai and Lashbrook, 2008).

Agusti et al. (2008) reported that a MYB transcription factor was expressed in the AZ only during the first phase of abscission, suggesting potential role of MYB in initiation of abscission. MYC2, a basic helix-loop-helix (bHLH) domain–containing transcription factor, acts as both an activator and a repressor of the expression of distinct JA-responsive genes in Arabidopsis (Lorenzo et al.,

2004). Dombrecht et al. (2007) demonstrated that MYC2 may modulate JA responses via differential regulation of an intermediate spectrum of transcription factors with activating or repressing roles in JA signaling. Rajani and Sunderesan (2001) described a mutation in

102 Arabidopsis called alcatraz (alc), which prevents dehiscence of fruit by specifically blocking the separation of the valve cells from the replum. The ALC gene was shown to encode a protein related to the MYC/bHLH family of transcription factors and is expressed in the valve margins of the silique, which is the site of cell separation during dehiscence. In the present study, an up- regulation in the expression of MYC2 LIKE transcription factor was observed suggesting that this up-regualtion is due to JA mediated signaling andthat it may be involved in MeJa-mediated abscission. Other altered transcription factors such as NAC, WRKY, GRAS, BPE and BIM are known to regulate the expression plant hormone–responsive genes (Chandler et al., 2009; Hirsch and Oldroyd, 2009; Olsen et al., 2005; Yang et al., 2010). Gil-Amado and Gomez-Jimenez

(2013) also reported similar transcription factor families to be altered during olive abscission. It can be concluded that these altered transcription factors possibly regulate downstream process related to plant hormone such as ethylene, JA, auxin, gibberellic acid and ABA and thereby regulate abscission.

Plant hormone related processes

Several genes involved in plant hormone- metabolism and signaling were altered in the PPJ within 48 h of application of MeJa and ethephon. Plant hormone –related gene transcripts that were altered in response to ethephon or MeJa treatment generally belonged to ethylene biosynthesis, ethylene-signaling, jasmonate biosynthesis, jasmonate signaling, auxin signaling, and ABA signaling. Few of the major genes that were up-regulated included BON-1

ASSOCIATED PROTEIN LIKE, a number of ETHYLENE RESPONSIVE REGULATED

NUCLEAR PROTEIN (ERF), 1AMINOCYCLOPROPANE-1-CARBOXYLATE OXIDASE,

LIPOXYGENSE, 12-OXOPHYTODIENOATE REDUCTASE and PROTEIN TIFY 9. ACC synthase (ACS) and ACC oxidase (ACO) are involved in regulating ethylene biosynthesis

103 (Kende and Zeevaart, 1997, Pech et al., 2004). The activities of ACS and ACO are both rate- limiting steps for ethylene production (Kende and Zeevaart, 1997). Increased transcription of

ACS and ACO genes leads to an increase in ethylene biosynthesis (Woodson et al., 1992).

Ethylene biosynthesis is increased in response to a number of biotic and abiotic stresses and ethylene production can be autocatalytic. In the present study we observed an increase in a number of ethylene biosynthesis and signaling genes in both ethephon and MeJa treated PPJ. In citrus exogenous application of MeJa or any other functional analog of JA induces abscission by stimulating levels of ethylene (Burns et al., 2003; Hartmond et al., 2000). Similar to our results

Gil-Amado and Gomez-Jimenez (2013) reported a strong induction of ACO and ERF expression in the olive AZ suggesting that both ACO and ERF may be instrumental in modifying ethylene biosynthesis and signaling during abscission. Increase in the ethylene biosynthesis and signaling related genes during the induction of abscission has been demonstrated in many studies including

Arabidopsis (Cai and Lashbrook, 2008), citrus (Agusti et al., 2008) and tomato (Meir et al.,

2010). LIPOXYGENASE and 12-OXOPHYTODIENOATE REDUCTASE are known for their roles in JA synthesis (Delker et al., 2006). Increase in JA biosynthesis related genes may be a mechanism that further augments the progression of abscission in response to exogenous application of MeJa. TIFY belongs to large family of proteins that are involved in jasmonate signaling. The TIFY family includes ZIM and ZIM-like proteins that contain a zinc-finger DNA- binding domain (Vanholme et al., 2007). Jasmonate Zim domains are proposed to exert their effects on gene expression through interaction with MYC2 (Chini et al., 2007).

Many previous studies have reported that ethylene treatment directly or indirectly results in biosynthesis of jasmonates (Agusti et al., 2008). Expression of ethylene response factor is induced by JA and ethylene in plant defense responses (Lorenzo et al., 2003). It is also well

104 documented that there is interaction between ethylene and jasmonate signaling and a positive feedback of ethylene on jasmonic acid biosynthesis has been postulated (Sasaki et al., 2001). In the present study we observed an increase in a number of ethylene and JA biosynthesis and signaling related genes in both ethephon and MeJa treated PPJ. These results suggest that the signaling pathways of ethylene and JA may be interconnected (may be directly or indirectly) and such an interplay may regulate the progression of abscission. Such observation supports the balance model which proposes that the induction of abscission depends on a complex interplay of plant hormone concentration in addition to factors that alter the responsiveness of the tissue or sensitivity to them (Gonzalez-Carranza and Roberts, 2012).

Defense response, pathogenesis and oxidation stress related genes

A large number of genes involved in stress response, pathogenesis, disease resistance and oxidative stress were altered in the PPJ within 48 h of ethephon and MeJa application. Majority of the genes related to pathogenesis or defense response were up-regulated except for the cold regulated proteins which were down-regulated. Cell separation has been previously associated with the accumulation of pathogenesis-related or defense response proteins (Roberts et al.,

2002). This accumulation is apparently related to a mechanism of defense organized to protect a potential wound. In both ethephon and MeJa-treated PPJ there was an over representation of pathogenesis, defense response and disease resistance genes. Similar results have been reported in citrus leaf abscission (Agusti et al., 2009) and olive fruit abscission (Gil-Amado and Gomez-

Jimenez, 2013).

Most of the genes involved in oxidation-reduction process or redox reaction were also up-regulated with very few exceptions. Some of the up-regulated genes were PEROXIDASE and genes related to response to hydrogen peroxide. There is increasing evidence suggesting that

105 reactive oxygen species (ROS) might be associated with ethylene induced abscission as well as with other physiological processes that can indirectly promote organ abscission (Agusti et al.,

2009). Increase in the peroxidase activity in the abscission zone during the onset of abscission is well documented (Hinman and Lang, 1965; Gahagan et al., 1968) although its role in the process of abscission is still unclear. Marynick (1977) concluded that oxidative respiration is essential to trigger abscission. Agusti et al. (2008) suggested that preferential expression of PEROXIDASE was detected only in the latter phase of abscission indicating that ROS scavenging is part of a defensive response. Recently Sakamato et al. (2008) reported that hydrogen peroxide was directly involved in ethylene-mediated abscission signaling in vitro in Capsicum leaves, where it appears to act as an intermediate molecule in the expression of ethylene-induced cell wall hydrolases. To minimize the damaging effect of ROS, plants have evolved non-enzymatic and enzymatic antioxidant defenses. Non enzymatic defenses include some secondary metabolites - compounds with intrinsic antioxidant properties such vitamin C, vitamin E, carotene and polyphenolic compounds. In present study a large number of genes involved in secondary metabolism were altered in expression suggesting that potentially these genes were expressed to combat oxidative stress during abscission.

Transport, metabolic and other process-related genes

A large number of genes involved in transporter activity and metabolic processes were altered in response to ethephon and MeJa applications. Agusti et al. (2008) reported preferential expression of ion, sugar and lipid transport genes during ethylene mediated abscission. Genes related to metabolism and phosphorylation such as RING-finger domain (RING), U-box domain,

F-box, ubiquitin –conjugating, ubiquitin- ligase and other families were altered in response to ethephon and MeJa applications. Increased metabolism and protein phophorylation has been

106 reported in many studies. In recent work, the activity of the F-box protein HAWAIIAN SKIRT has been linked to the control of floral abscission in Arabidopsis (Gonzlaez-Carranza et al., 2007).

Terol et al., (2007) reported that an E2 ubiquitin conjugating enzyme and two E3 ubiquitin ligases were present exclusively in abscission related libraries. All together these results suggest that transport and metabolism related genes play a role in ethylene and MeJa induced abscission, although their specific functions are still not clear.

Gene expression analysis

Twenty genes with differential expression were selected from the RNA-Seq data for further qRT-PCR analysis. The selected genes belonged to cell wall hydrolysis/metabolism, plant hormone metabolism and signaling, and transcription factor families. The qRT-PCR analysis was performed on control, ethephon and MeJa-treated PPJ tissue collected at 0, 24 and 48 h of application. Out of 20 genes 12 genes were found to be significantly differentially expressed in ethephon and MeJa-treated PPJ tissues. The cell wall metabolism/hydrolysis related genes that found to be differentially up-regulated within 48 h of ethephon and MeJa-induced abscission were ENDO-POLYGALCTURONASE, PROBABLE XYLOGLUCAN

ENDOTRANSGLUCOSYLASE and EXPANSIN (Figure 5). ENDO-POLYGALCTURONASE was found to be up-regulated in expression in both the ethephon and MeJa-treated PPJ transcriptome at 24 and 48 h after the treatment. XYLOGLUCAN ENDOTRANSGLUCOSYLASE was found to up-regulated in at 48 h after the MeJa treatment. EXPANSIN were observed to be up-regulated in

MeJa treated PPJ transcriptome at 24 h of application suggesting that activity changes might have been initiated within 24 h of application. Expression of PECTINESTERASE INHIBITOR was down-regulated in both RNA-Seq analysis and qRT-PCR analysis in response to MeJa induced abscission. Pectinesterase activity is known to be involved in dissolution of middle

107 lamella. These results are in agreement with RNA-Seq data and previous studies on abscission in other species. Abscission involves cell separation and dissolution of the middle lamella, cell- wall-modifying proteins and wall hydrolases such as polygalacturonases, expansins and pectate lyases are well associated with abscission (Atkinson et al. 2002; Belfield et al. 2005; Brummell et al. 1999; Burns et al. 1998; del Campillo and Bennett 1996; Gonzalez-Carranza et al. 2002,

2007; Jiang et al. 2008; Kalaitzis et al., 1995, 1997; Sane et al. 2007; Tucker et al., 1991).

According to the qRT-PCR analysis, transcription factors that were significantly expressed during MeJa induced abscission included two AP2/ERF domain transcription factors and two

MYC2 LIKE transcription factors (Figure 5). Both the MYC2 LIKE transcription factors were found to be up-regulated and followed the pattern of RNA-seq data where they were found to be up-regulated only by MeJa treatment. Interestingly, both the AP2/ERF domain transcription factors were observed to differentially express only at 24 h and not at 48 h in the qRT-PCR analysis. Moreover, one of the AP2/ERF domain transcription factors decreased in expression at

24 h of application after MeJa treatment. These observations were contrary to RNA-Seq data where both of the AP2/ERF domain transcription factors were significantly up-regulated at 48 h time point. One of the AP2 domain transcription factor that was significantly up-regulated by

MeJa treatment in RNA-Seq analysis was found to be up-regulated in ethephon-treated PPJ at 24 h of application (Figure 5).

Only two plant hormone related genes were observed to be differentially expressed in qRT-PCR analysis (Figure 5). According qRT-PCR analysis 12-OXOPHYTODIENOTE REDUCTASE was found to differentially up-regulated only in ethephon-treated PPJ after 48 h of treatment. 12-

OXOPHYTODIENOTE REDUCTASE is reported to be involved in jasmonate biosynthesis

(Schaller et al., 2000), indicating that ethylene mediated signaling directly or indirectly impact

108 jasmonic acid biosynthesis. Similar pattern was observed in RNA-Seq data thus validating the potential involvement of 12-OXOPHYTODIENOTE REDUCTASE in ethephon mediated abscission. Another gene that was found to be up-regulated is a RESPIRATORY BURST

OXIDASE, which was up-regulated both at 24 h and 48 h after Meja and ethephon treatment.

This gene is potentially involved in ethylene signaling, jasmonate signaling, salicylic acid signaling, and in response to stress. According to the RNA-Seq analysis, RESPIRATORY BURST

OXIDASE was up-regulated only in ethephon treated PPJ transcriptome but in qRT-PCR analysis

MeJa and ethephon-treated PPJ displayed an increase in the expression.

Overall, more thatn 50% of the genes analyzed in the qRT-PCR analysis were found to be in agreement with the RNA-Seq data. Also the qRT-PCR analyses provided insights into how the expression of these genes changed over the period of abscission progression. Some of the differences between RNA-Seq and qRT-PCR data may be attributed to differences in the sensitivity of these two techniques in terms of identifying differential gene expression.

Conclusion

RNA-Seq approach provides a comprehensive view on blueberry fruit abscission induced by two different plant growth hormones, the ethylene releasing compound, ethephon, and MeJa. Genes associated with cell wall biosynthesis and metabolism were differentially expressed in ethephon and MeJa induced abscission treatments potentially implicating them in organ shedding. These data also suggest that in addition to modification of pectins, hemicelluloses are also remodeled and altered during cell separation process. The data also suggests that large number of transcription factors (basic helix loop helix, MYB, MYC, NAC, ZF- domains) were altered in response to ethephon and MeJa treatments and may potentially contribute to different processes involved in the progression of abscission. Number of plant hormone biosynthesis, metabolism

109 and signaling pathway related genes were altered in response to ethephon and MeJa treatment.

This observation supports previous studies where a complex interplay among plant hormones has been suggested to contribute to the regulation abscission. The data also suggest that a number of pathogenesis and defense response genes were expressed during induction of abscission. A large number of unknown and uncharacterized genes were also differentially expressed during the induction of abscission which provides a potential source of new gene discovery.

110 Figure 1. Percent fruit drop as a result of MeJa and ethephon treatment in ‘Powderblue’ blueberry in year 2010(top) and year 2013 (bottom).

111 Figure 2. Heat Map of differentially expressed genes in the Control, MeJa and ethephon treated

PPJ transcriptome.

112 Figure 3. Cluster analysis of transcripts into eight clusters with similar gene expression patterns.

113 Figure 4a. Biological process categorization of differentially expressed genes in control vs. ethephon (left) and control vs. MeJa (right) comparisons.

114 Figure 4b.Molecular function categorization of differentially expressed genes in control vs. ethephon (left) and control vs. MeJa (right) comparisons.

115 116 Figure 5. Gene expression of various cell wall metabolism genes, phytohormone related genes and transcription factors in control, MeJa and ethephon PPJs.

117 References

Adams, D. and S. Yang. 1979. Ethylene biosynthesis: identification of 1-aminocyclopropane-1- carboxylic acid as an intermediate in the conversion of methionine to ethylene. Proc. Natl. Acad. Sci. 76(1):170-174.

Agusti, J., P. Merelo, M. Cercos, F.R. Tadeo, and M. Talon. 2008. Ethylene-induced differential gene expression during abscission of citrus leaves. J. Exp. Bot. 59:2717-2733.

Agusti, J., P. Merelo, M. Cercos, F. Tadeo, and M. Talon. 2009. Comparative transcriptional survey between laser-microdissected cells from laminar abscission zone and petiolar cortical tissue during ethylene-promoted abscission in citrus leaves. BMC Plant Biol. 9(1):127.

Atkinson, R.G., R. Schröder, I.C. Hallett, D. Cohen, and E.A. MacRae. 2002. Overexpression of polygalacturonase in transgenic apple trees leads to a range of novel phenotypes involving changes in cell adhesion. Plant Physiol. 129(1):122-133.

Austin, M.E., and Williamson, R.E. 1977. Comparison of harvest methods of rabbiteye blueberries. J. Amer. Soc. Hort. Sci. 102,454-456.

Belfield, E.J., Ruperti, B., Roberts, J.A. and McQueen-Mason, S.J. 2005.Changes in expansin activity and gene expression during ethylene-promoted leaflet abscission in Sambucus nigra. J. Exp. Bot.56, 817-823.

Beno-Moualem, D., Gusev, L., Dvir, O., Pesis, E., Meir, S. and Lichter, A. 2004. The effects of ethylene, methyl jasmonate and 1-MCP on abscission of cherry tomatoes from the bunch and expression of endo-1, 4-β-glucanases. Plant Sci.167 , 499-507.

Brummell, D.A., B.D. Hall, and A.B. Bennett. 1999. Antisense suppression of tomato endo-1, 4- β-glucanase Cel2 mRNA accumulation increases the force required to break fruit abscission zones but does not affect fruit softening. Plant Mol. Biol. 40(4):615-622.

Burns, J.K. 2002. Using molecular biology tools to identify abscission materials for citrus. HortScience 37:459-463.

Burns, J.K., D.J. Lewandowski, C.J. Nairn, and G.E. Brown. 1998. Endo 1,4-beta-glucanase gene expression and cell wall hydrolase activities during abscission in Valencia orange. Physiol. Plantarum 102:217-225.

Cai, S. and C.C. Lashbrook. 2008. Stamen abscission zone transcriptome profiling reveals new candidates for abscission control: enhanced retention of floral organs in transgenic plants overexpressing Arabidopsis ZINC FINGER PROTEIN2. Plant Physiol. 146:1305-1321.

Chandler, J.W., M. Cole, A. Flier, and W. Werr. 2009. BIM1, a bHLH protein involved in brassinosteroid signalling, controls Arabidopsis embryonic patterning via interaction with DORNRÖSCHEN and DORNRÖSCHEN-LIKE. Plant Mol. Biol. 69(1-2):57-68.

118 Chini, A., Fonseca, S., Fernandez, G., Adie, B., Chico, J.M., Lorenzo, O., Garcia-Casado, G., Lopez-Vidriero, I., Lozano, F.M., Ponce, M.R., Micol, J.L., and Solano, R. (2007). The JAZ family of repressors is the missing link in jasmonate signalling. Nature 448: 666–671.

Botton, A., G. Eccher, C. Forcato, A. Ferrarini, M. Begheldo, M. Zermiani, S. Moscatello, A. Battistelli, R. Velasco, and B. Ruperti. 2011. Signaling pathways mediating the induction of apple fruitlet abscission. Plant Physiol. 155(1):185-208.

Cosgrove, D.J., L.C. Li, H. Cho, S. Hoffmann-Benning, R.C. Moore, and D. Blecker. 2002. The growing world of expansins. Plant Cell Physiol. 43(12):1436-1444.

Costa G, Dal Cin V, Ramina A. 2006.Physiological, molecular and practical aspects of fruit abscission. In: Webset, A.D. and H.Ramirez (eds.). Proc. Xth IS on Plant Bioregulators in Fruit. Acta Hort. 727. ISHS. del Campillo E. and A.B. Bennett.1996. Pedicel breakstrength and cellulase gene expression during tomato flower abscission. Plant Physiol. 111, 813–820.

Dombrecht, B., G.P. Xue, S.J. Sprague, J.A. Kirkegaard, J.J. Ross, J.B. Reid, G.P. Fitt, N. Sewelam, P.M. Schenk, and J.M. Manners. 2007. MYC2 differentially modulates diverse jasmonate-dependent functions in Arabidopsis. Plant Cell Onl. 19(7):2225-2245.

Eisen, M.B., P.T. Spellman, P.O. Brown, and D. Botstein. 1998. Cluster analysis and display of genome-wide expression patterns. Proc. Natl. Acad. Sci. 95(25):14863-14868.

Fry, S., R. Smith, K. Renwick, D. Martin, S. Hodge, and K. Matthews. 1992. Xyloglucan endotransglycosylase, a new wall-loosening enzyme activity from plants. Biochem. J. 282:821- 828.

Gahagan, H., R. Holm, and F. Abeles. 1968. Effect of ethylene on peroxidase activity. Physiol. Plantarum 21(6):1270-1279.

Gil-Amado, J.A. and M.C. Gomez-Jimenez. 2013. Transcriptome analysis of mature fruit abscission control in olive. Plant Cell Physiol. 54(2):244-269.

González-Carranza, Z.H., C.A. Whitelaw, R. Swarup, and J.A. Roberts. 2002. Temporal and spatial expression of a polygalacturonase during leaf and flower abscission in oilseed rape and Arabidopsis. Plant Physiol. 128(2):534-543.

Gonzalez-Carranza, Z., U. Rompa, J.L. Peters, A.M. Bhatt, C. Wagstaff, A.D. Stead, and J.A. Roberts. 2007. HAWAIIAN SKIRT: An F-Box gene that regulates organ fusion and growth in Arabidopsis. Plant Physiol. 144:1370-1382.

Gonzalez-Carranza, Z.H. and J.A. Roberts. 2012. Ethylene and cell separation processes. Annual Plant Reviews, Plant Horm. Ethyl. 44:246.

119 Goren, R. 1993. Anatomical, physiological, and hormonal aspects of abscission in citrus. Hortic. Rev. 15:145-182.

Hartmond, U., R.Yuan, J.K.Burns, A. Grant and W.J. Kender. 2000. Citrus fruit abscission induced by methyl-jasmonate. J. Amer. Soc. Hort. Sci. 125, 547-552.

He, Y., H. Fukushige, D.F. Hildebrand, and S. Gan. 2002. Evidence supporting a role of jasmonic acid in Arabidopsis leaf senescence. Plant Physiol. 128(3):876-884.

Hirsch, S. and G.E. Oldroyd. 2009. GRAS-domain transcription factors that regulate plant development. Plant Sign. Beh. 4(8):698-700.

Hinman, R. and J. Lang. 1965. Peroxidase-catalyzed Oxidation of Indole-3-acetic Acid*. Biochem. 4(1):144-158.

Howell, G.S., B.G. Stergios, S.S. Stackhouse, H.C. Bittenbender and C.L. Burton. 1976. Ethephon as a mechanical harvesting aid for highbush blueberries (Vaccinium austral Small). J.Amer. Soc. Hort. Sci. 101, 111-115.

Iwai, H., A. Terao, and S. Satoh. 2012. Changes in distribution of cell wall polysaccharides in floral and fruit abscission zones during fruit development in tomato (Solanum lycopersicum). J. Plant Res. :1-11.

Jiang, C.Z., F. Lu, W. Imsabai, S. Meir and M.S. Reid. 2008. Silencing polygalacturonase expression inhibits tomato petiole abscission. J. Exp. Bot. 59, 973–979.

Kalaitzis, P., S.M. Koehler, and M.L. Tucker. 1995. Cloning of a tomato polygalacturonase expressed in abscission. Plant Mol. Biol. 28(4):647-656.

Kalaitzis, P., T. Solomos, and M.L. Tucker. 1997. Three different polygalacturonases are expressed in tomato leaf and flower abscission, each with a different temporal expression pattern. Plant Physiol. 113(4):1303-1308.

Kende, H. and J. Zeevaart. 1997. The Five" Classical" Plant Hormones. Plant Cell 9(7):1197.

Kranz, H.D., M. Denekamp, R. Greco, H. Jin, A. Leyva, R.C. Meissner, K. Petroni, A. Urzainqui, M. Bevan, and C. Martin. 1998. Towards functional characterisation of the members of theR2R3 MYBgene family fromArabidopsis thaliana. Plant Journal 16(2):263-276.

Lee, Y., P. Derbyshire,‐ J.P. Knox, and A.K. Hvoslef Eide. 2008. Sequential cell wall transformations in response to the induction of a pedicel abscission event in Euphorbia pulcherrima (poinsettia). Plant Journal 54(6):993-1003.‐

Lewis, M.W., M.E. Leslie, and S.J. Liljegren. 2006. Plant separation: 50 ways to leave your mother. Curr. Opin. Plant Biol. 9:59-65.

120 Li, J. and R. Yuan. 2008. NAA and ethylene regulate expression of genes related to ethylene biosynthesis, perception, and cell wall degradation during fruit abscission and ripening in ‘Delicious’ apples. J. Plant Growth Regul. 27(3):283-295.

Li, J., H. Zhu and Y. Rongcai. 2010. Profiling the expression of genes related to ethylene biosynthesis, ethylene perception, and cell wall degradation during fruit abscission and fruit ripening in Apple. J. Amer. Soc. Hort. Sci. 135(5), 391-401.

Li, Y., L. Jones, and S. McQueen-Mason. 2003. Expansins and cell growth. Curr. Opin. Plant Biol. 6(6):603-610.

Lorenzo, O., R. Piqueras, J.J. Sánchez-Serrano, and R. Solano. 2003. ETHYLENE RESPONSE FACTOR1 integrates signals from ethylene and jasmonate pathways in plant defense. Plant Cell Onl. 15(1):165-178.

Mainland, C.M., L.J. Kushman and W.E. Ballinger. 1975. The effect of mechanical harvesting on yield, quality of fruit and bush damage of highbush blueberry J. Amer. Soc. Hort. Sci. 100, 129-134.

Malladi A. and J.K. Burns. 2008. CsPLDα1 and CsPLDγ1 are differentially induced during leaf and fruit abscission and diurnally regulated in Citrus sinensis. J. Exp. Bot. 59, 3729-3739.

Malladi, A., T. Vashisth, and L.K. Johnson. 2012. Ethephon and methyl jasmonate affect fruit detachment in rabbiteye and southern highbush blueberry. HortScience 47(12):1745-1749.

Mao, L., D. Begum, H. Chuang, M. Budiman, E. Szymkowiak, E. Irish and R. Wing. 2000. JOINTLESS is a MADS-box gene controlling flower abscission zone development. Nature 406, 910–913.

Marynick, M.C. 1977. Patterns of ethylene and carbon dioxide evolution during cotton explant abscission. Plant Physiol. 59(3):484-489.

McQueen-Mason, S.J. and D.J. Cosgrove. 1995. Expansin mode of action on cell walls (analysis of wall hydrolysis, stress relaxation, and binding). Plant Physiol. 107(1):87-100.

McQueen-Mason, S. and D.J. Cosgrove. 1994. Disruption of hydrogen bonding between plant cell wall polymers by proteins that induce wall extension. Proc. Natl. Acad. Sci.91(14):6574- 6578.

Meir, S., D.A. Hunter, J.C. Chen, V. Halaly and M.S. Reid. 2006. Molecular changes occurring during acquisition of abscission competence following auxin depletion in Mirabilis jalapa. Plant Physiol. 141, 1604–1616.

Meir, S., S. Philosph-Hadas, S. Sundaresan, V.K.S. Selvara, S. Burd, R. Ophor, et al. 2010.Microarray Analysis of the Abscission-Related Transcriptome in the Tomato Flower Abscission Zone in Response to Auxin Depletion. Plant Physiol. 154 (4), 1929-1956.

121 Miyamoto, K., J. Ueda, K. Yamada, S. Kosemura, S. Yamamura, and K. Hasegawa. 1997. Inhibition of abscission of bean petiole explants by lepidimoide. J. Plant Growth Regul. 16(1):7- 9.

Olsen, A.N., H.A. Ernst, L.L. Leggio, and K. Skriver. 2005. NAC transcription factors: structurally distinct, functionally diverse. Trends Plant Sci. 10(2):79-87.

Pfaffl, M.W. 2001. A new mathematical model for relative quantification in real-time RT–PCR. Nucleic Acids Res. 29(9):e45-e45.

Patterson, S.E. and A.B. Bleecker. 2004. Ethylene-dependent and -independent processes associated with floral organ abscission in Arabidopsis. Plant Physiol. 134,194 -203.

Patterson, S., M. Butenko and J. Kim. 2007. Ethylene responses in abscission and other processes of cell separation in Arabidopsis. In Advances in Plant Ethylene Research (ed. A. Ramina, C. Chang, J. Giovannoni, H. Klee, P. Perata and E. Woltering), 271-278. Netherlands: Springer.

Pech, J., M. Bouzayen, and A. Latché. 2008. Climacteric fruit ripening: ethylene-dependent and independent regulation of ripening pathways in melon fruit. Plant Sci. 175(1):114-120.

Rajani, S. and V. Sundaresan. 2001. The Arabidopsis myc/bHLH gene ALCATRAZ enables cell separation in fruit dehiscence. Curr.Biol. 11(24):1914-1922.

Ramakers, C., J.M. Ruijter, R.H.L. Deprez, and A.F. Moorman. 2003. Assumption-free analysis of quantitative real-time polymerase chain reaction (PCR) data. Neurosci. Lett. 339(1):62-66.

Rasori A, Bertolasi B, Furini A, Bonghi C, Tonutti P, Ramina A. 2003.Functional analysis of peach ACC-oxidase promoters in transgenic tomatoes and ripening peach fruit. Plant Sci. 165, 523-530.

Rieu, I. and S.J. Powers. 2009. Real-time quantitative RT-PCR: design, calculations, and statistics. Plant Cell Onl. 21(4):1031-1033.

Roberts, J.A., K.A. Elliott and Z.H. Gonzalez-Carranza. 2002. Abscission, dehiscence, and other cell separation processes. Annu. Rev. Plant Biol. 53, 131–158.

Ruijter, J., S. van der Velden, and A. Ilgun. 2009. LinReg PCR (11.0). Analysis of quantitative.

Sakamoto, M., I. Munemura, R. Tomita, and K. Kobayashi. 2008. Involvement of hydrogen peroxide in leaf abscission signaling, revealed by analysis with an in vitro abscission system in Capsicum plants. Plant J. 56(1):13-27.

Sane, A.P., S.K. Tripathi, and P. Nath. 2007. Petal abscission in rose (Rosa bourboniana var Gruss an Teplitz) is associated with the enhanced expression of an alpha expansin gene, RbEXPA1. Plant Sci. 172(3):481-487.

122 Saniewski M, Ueda J, Miyamoto K. 2000. Methyl jasmonate induces the formation of secondary abscission zone in stem of Bryophyllum calycinum Salisb. Acta Physiol. Plant 22, 17–23.

Sasaki, Y., E. Asamizu, D. Shibata, Y. Nakamura, T. Kaneko, K. Awai, M. Amagai, C. Kuwata, T. Tsugane, and T. Masuda. 2001. Monitoring of methyl jasmonate-responsive genes in Arabidopsis by cDNA macroarray: self-activation of jasmonic acid biosynthesis and crosstalk with other phytohormone signaling pathways. DNA Res.8(4):153-161.

Schaller, F., C. Biesgen, C. Müssig, T. Altmann, and E.W. Weiler. 2000. 12-Oxophytodienoate reductase 3 (OPR3) is the isoenzyme involved in jasmonate biosynthesis. Planta 210(6):979-984.

Spellman PT (2003) Cluster analysis and display. In D Bowtell, J Sambrook, eds, DNA Microarrays: A Molecular Cloning Manual. Cold Spring Harbor Laboratory Press, Woodbury, NY, pp 569–581

Steiner Lange, S., U.S. Unte, L. Eckstein, C. Yang, Z.A. Wilson, E. Schmelzer, K. Dekker, and H. Saedler. 2003. Disruption of Arabidopsis thaliana MYB26 results in male sterility due to non dehiscent‐ anthers. Plant J. 34(4):519-528.

Takeda,‐ F., G. Krewer, E. Andrews, D. Peterson, and J. Mullinix. 2008. Assessment of v45 over- the-row mechanical harvester on pruned rabbiteye and southern highbush blueberry. HortTechnology 18:130-138.

Taylor, J.E. and C.A. Whitelaw. 2001. Signals in abscission. New Phytol. 151:323-339.

Terol, J., A. Conesa, J.M. Colmenero, M. Cercos, F.R. Tadeo, J. Agusti, E. Alos, F. Andres, G. Soler, J. Brumos, and D.J. Iglesias. 2007. Analysis of 13000 unique Citrus clusters associated with fruit quality, production and salinity tolerance. BMC Genomics 8:31.

Tonutti, P., L. Cass, and R. Christoffersen. 1995. The expression of cellulase gene family members during induced avocado fruit abscission and ripening. Plant, Cell Environ. 18(6):709- 713.

Tucker, M.L., S.L. Baird, and R. Sexton. 1991. Bean leaf abscission: tissue-specific accumulation of a cellulase mRNA. Planta 186(1):52-57.

Ueda J, Miyamoto K, Hashimoto M. 1996. Jasmonates promote abscission in bean petiole explants: Its relationship to the metabolism of cell wall polysaccharides and cellulase activity. J. Plant Growth. Reg. 15, 189–195.

U.S. Department of Agriculture. 2012. Non-citrus fruits and nuts 2011 summary. National Agricultural Statistics Service, Washington, DC.

Vanholme, B., W. Grunewald, A. Bateman, T. Kohchi, and G. Gheysen. 2007. The tify family previously known as ZIM. Trends Plant Sci. 12(6):239-244.

123 Vashisth, T., L.K. Johnson, and A. Malladi. 2011. An efficient RNA isolation procedure and identification of reference genes for normalization of gene expression in blueberry. Plant Cell Rep. 30(12):2167-2176.

Vashisth, T. and A. Malladi. 2013. Fruit Detachment in Rabbiteye Blueberry: Abscission and Physical Separation. J. Am. Soc. Hort. Sci. 138(2):95-101.

Ward, D.L., E.P. Beers, R.E. Byers, and R.P. Marini. 1999. Cutting apple fruits induces cellulase activity in the abscission zone. HortScience 34(4):601-603.

Wei P.C., F. Tan, X.Q. Gao, X.Q. Zhang, G.Q. Wang, H. Xu, et al.2010.Overexpression of AtDOF4.7, an Arabidopsis DOF Family Transcription Factor, Induces Floral Organ Abscission Deficiency in Arabidopsis. Plant Physiol. 153, 1031-1045.

Whitney, S. E. C. , M. G. E. Gothard, J.T. Mitchell and M .J. Gidley. 1999. Roles of cellulose and xyloglucan in determining the mechanical properties of primary plant cell walls. Am. Soc. Plant Physiol. 121,657-664.

Woodson, W.R., K.Y. Park, A. Drory, P.B. Larsen, and H. Wang. 1992. Expression of ethylene biosynthetic pathway transcripts in senescing carnation flowers. Plant Physiol. 99(2):526-532.

Yang, S., B. Vanderbeld, J. Wan, and Y. Huang. 2010. Narrowing down the targets: towards successful genetic engineering of drought-tolerant crops. Mol. plant 3(3):469-490.

124 CHAPTER 5

VARIABILITY IN THE EASE - OF- FRUIT DETACHMENT IN BLUEBERRY1

1Vashisth T., S.NeSmith and A. Malladi. To be submitted to Journal of ASHS

125 Abstract

Currently there is a large interest in selection of blueberry cultivars that are more suitable for mechanical harvesting. One of the desirable traits in selection for mechanical harvesting suited fruit is ease of fruit detachment/abscission. Abscission is a physiological process that involves the programmed separation of entire organs, such as fruit. In blueberry fruit detachment is desired at fruit-pedicel junction (FPJ) although fruit abscission is known to occur at pedicel- peduncle junction (PPJ), which makes fruit ‘stemmy’. Therefore in case of blueberry fruit detachment at FPJ is an important trait when selecting for mechanical harvesting. Suziblue, a southern highbush blueberry is new cultivar that is potentially suited for mechanical harvesting.

Suziblue and TH 729 are siblings derived from the same cross (same parents). TH 729 displays drop of mature fruit in response to a mild touch or shaking while suziblue does not. In TH 729 fruit detachment in response to mechanical shaking was found to occur at PPJ while suziblue consistently detach at FPJ. Scanning electron microscopy study revealed a very disrupted and uneven fracture plane in Suziblue when forcefully removed at PPJ as compared to TH 729. A comprehensive fifty eight gene quantitative real-time PCR gene profiling revealed that selective phytohormone hormone biosynthesis, metabolism and signaling pathway related genes were differentially expressed in the PPJ of two genotypes. In addition to phytohormone related genes a cell wall hydrolysis related gene GLUCAN ENDO 1-3 β GLUCANSE was significantly down- regulated in Suziblue. Overall, this data suggests that preferential detachment of Suziblue at FPJ can be related to these differentially expressed genes. Also, potentially these phytohormone and cell wall hydrolysis genes play an important role in the genetic ease of fruit detachment/abscission.

Keywords: blueberry, abscission, cell wall hydrolysis, phytohormone

126 Introduction

Blueberry (Vaccinium spp.) is an indigenous crop of the United States. Blueberry is a flowering plant with berries ranging from purple to black in color. North America is the largest producer contributing approximately 90% of the world’s blueberries. The most widely grown species of blueberry are the northern highbush, southern highbush, lowbush and rabbiteye.

Blueberry was valued at over $780 million in 2011 (USDA, 2012). Harvesting is generally one of the most expensive and labor intensive aspects of blueberry production, especially when performed using manual methods. Hand harvesting of rabbiteye and highbush blueberries is labor intensive and requires as many as 520 h of labor per acre (Brown et al., 1996). Hand harvesting costs range from $1.10 to 1.54 per kg for southern highbush blueberries and $0.86 to

1.10 per kg for rabbiteye blueberries (Safley et al., 2005). Mechanical harvesting can greatly reduce the costs associated with harvesting. Owing to the growing interest in the use of mechanical harvesting in blueberry, especially for fruit intended for the fresh fruit market, there are considerable research efforts underway to enhance the efficiency of this process. Currently, there is significant interest in the selection of blueberry cultivars suitable for mechanical harvesting. One of the desirable traits in cultivar selection for mechanical harvest-suitable blueberry is detachment at the fruit-pedicel junction (FPJ). Fruit detachment by abscission in blueberry occurs at the pedicle-peduncle junction (PPJ) whereas detachment at FPJ is due to mechanical breakage (Vashisth and Malladi, 2013). Figure 1 shows a diagrammatic representation of the two junctions.

Hand harvesting and majority of mechanical harvesting of blueberry typically result in fruit separation at the FPJ (Howell et al., 1976; Takeda et al., 2008). Mechanical harvesters are typically over the row harvesters and detach the berries from the bush through physical shaking

127 of the bush. Excessive vibration in mechanical harvesting operations often results in breakage of the pedicel, or detachment of the fruit away from the FPJ. Therefore, such mechanical harvesting methods can ultimately lead to fruit injury and loss of overall fruit quality (Howell et al., 1976).

Hence, methods to increase the mechanical harvesting efficiency are required.

‘Suziblue’, a southern highbush blueberry is relatively new cultivar that is potentially suited for mechanical harvesting (NeSmith, 2010). ‘Suziblue’ and TH 729 are siblings derived from the same parents. TH 729 displays drop of mature fruit in response to a mild touch or shaking while ‘Suziblue’ does not. Preliminary studies indicated that in TH 729 fruit detachment in response to mechanical shaking occurs at the PPJ while ‘Suziblue’ preferentially detaches at

FPJ. Therefore these two genotypes are of interest as they differ in their fruit detachment response. Such variation may be due to differences in phytohormone levels or signaling, or due to inherent differences in the AZ anatomy. Comparison of these genotypes may increase our understanding of physiology of abscission and may help in developing tools to improve mechanical harvesting of blueberry.

Therefore, the objective of this study was to study the genetic control of ease of fruit detachment in these two genotypes. The objective was approached through three experiments 1)

Fruit detachment in response to mechanical shaking; 2) Scanning electron microscopy analysis to study the fracture plane at the PPJ; and 3) Gene expression analysis using the PPJ tissue for phytohormone and cell wall metabolism- related genes.

128 Materials and Methods

Fruit Detachment in Response to Mechanical Shaking

Fruit detachment in TH729 and ‘Suziblue’ in response to mechanical shaking was performed in

2012 and 2013 at different locations. In 2012, mature TH729 and ‘Suziblue’ plants (n=4) grown at the Blueberry Research Farm, University of Georgia, Alapaha, GA were used. In 2013, TH729 and ‘Suziblue’ plants (n=5) grown at the Griffin experiment station, The University of Georgia,

Griffin, GA were used. A hand-held mechanical shaker as described by Malladi et al. (2013) was used to perform mechanical shaking. A branch with around 50 fruit was selected on each plant.

The adapter of the mechanical shaker was attached to the branch, several inches above the fruit and the branch was shaken using the instrument for 5 s. The number of fruit on the branch was determined prior to, and after mechanical shaking, and used to calculate the percent fruit detachment. The detached fruit were collected in a catch frame placed below the branch. From the collected fruit, the total number of mature and immature fruit, and the number of fruit with the pedicel attached (stemmy fruit) were determined. The percent stemmy fruit was calculated from these data.

Scanning Electron Microscopy

Mature plants of TH729 and ‘Suziblue’ (n=4) grown at Blueberry Research Farm,

University of Georgia, Alapaha, GA were used for this study. The pedicels were separated manually from the peduncle or from the mature fruit to obtain the PPJ and the FPJ fracture planes, respectively. These samples were immediately fixed in 5% glutaraldehyde:0.1 M potassium phosphate buffer (1:1), stored at 4 °C and used for SEM analysis. The pedicel end of the PPJ and FPJ were observed in this analysis. Samples were processed for SEM according to

129 Mims (1981). Briefly, the samples were cut into approximately 1-2 mm upright sections and were rinsed three times with the fixative buffer [5% glutaraldehyde:0.1 M potassium phosphate buffer (1:1)] for 15 min. The samples were immediately treated with osmium tetroxide for 2 h and rinsed. The rinsed samples were dehydrated through a graded ethanol series. For critical point drying, the ethanol was replaced by liquid CO2 which was brought to the critical point in the Autosamdri-814 Critical Point Dryer (Tousimis Research Corporation, Rockville, MD).

Subsequently, samples mounted on aluminum stubs were coated with gold in the SPI-Module

Sputter Coater and Carbon-Coater (SPI Supplies / Structure Probe, Inc. West Chester, PA).

Samples were observed using the Zeiss 1450EP scanning electron microscope (Carl Zeiss

MicroImaging Inc., Thornwood, NY).

Plant Material for Gene Expression Profiling

Mature TH729 and ‘Suziblue’ plants (n=4) grown at Blueberry Research Farm, University of

Georgia, Alapaha, GA were used. The PPJ tissue from about 50 fruit were collected, and dissected manually to capture the AZ. Approximately, 1mm or less of the adjacent tissue was attached to the AZs. The dissected PPJ tissues were pooled within each replicate and frozen in liquid N2.

RNA Extraction, Reverse Transcription and Gene Expression Analysis

RNA extraction, cDNA synthesis and gene expression profiling was performed as described by

Vashisth et al., (2011) with some modifications. Briefly, 0.1 g of the sample was ground to fine powder and were added to 0.75 ml of the extraction buffer. The RNA extraction buffer consisted of 2% cetyltrimethylammonium bromide (CTAB), 2% polyvinylpyrrollidone (PVP), 100 mM

Tris-HCl, 25 mM ethylenediaminetetraacetic acid (EDTA), 2 M sodium chloride (NaCl), 3.44

130 mM spermidine, and 2% β-mercaptoethanol. All solutions were prepared using 0.1% diethylpyrocarbonate (DEPC) treated water. The buffer was warmed to 65 °C prior to the addition of the ground tissue. The sample-buffer slurry was vortexed briefly and incubated at 65

°C for 10 min, vortexed again and incubated at room temperature for 5 min. The mixtures were extracted twice with an equal volume of chloroform: isoamyl alcohol (24:1) with centrifugation at 4 °C for 15 min (5000 x g). The supernatant fraction was transferred to sterile tubes followed by the addition of 0.25 volumes of lithium chloride (LiCl; 10 M). The samples were mixed gently and precipitated overnight at 4 °C. Samples were subsequently centrifuged at 4 °C for 20 min (12000 x g). After discarding the supernatant fraction, the RNA pellets were washed twice with 500 µL of ice cold 70% ethanol, and dissolved in DEPC and stored at -80 °C. For cDNA synthesis genomic DNA contamination was removed by treating 1 µg of total RNA was with

DNase (Promega; 37 °C for 34 min). The DNase-treated RNA was reverse transcribed using Im-

Prom II Reverse Transcriptase (Promega) and oligo dT (Promega) in a 20 µL reaction volume.

Reverse transcription was performed at 42 °C for 75 min, followed by incubation at 75 °C for 15 min. The cDNA samples were subsequently diluted 10-fold and stored at -20 °C until further analysis. Gene expression analysis was performed by quantitative Real Time-PCR (qRT-PCR).

A total of 58 genes related to cell wall metabolism, phytohormone metabolism and signaling, and transcription factors were selected from next generation sequencing data. The primers for qRT-

PCR were designed using PrimerQuest (Integrated DNA Technologies, Inc.). The amplicon size targeted from 120-200 bases. All qRT-PCR reactions were performed using the Stratagene

Mx3005P real-time PCR system (Agilent Technologies, Inc., Santa Clara, CA). A reaction volume of 12 µL with 1 µL of diluted cDNA and 6 µL of 2X Veriquest SYBR Green Master

Mix (Affymetrix) was used. The thermal cycling conditions were: 95 °C for 10 min; 40 cycles of

131 94 °C for 30 s and 60 °C for 1 min. Melt-curve analysis was performed at the end of the qRT-

PCR reactions to check for primer specificity. Controls without a template and without the reverse transcriptase were run for each gene. The primer efficiency for each gene was determined using LinRegPCR (Ramakers et al., 2003; Ruijter et al., 2009). Relative expression was calculated using a modified Pfaffl method (Pfaffl, 2001) and as described in Rieu and Powers

(2009). Relative quantity (RQ) for each sample was calculated using the formula, 1/ECq, where

Cq is the quantification cycle (threshold cycle). The RQ was normalized using RNA HELICASE-

LIKE 8 (RH8) and CLATHRIN ADAPTOR COMPLEXES MEDIUM SUBUNIT FAMILY

PROTEIN a (CACSa) as described by Vashisth et al., (2011). The geometric mean of expression of these two reference genes (normalization factor) was used for normalization. The normalized

RQ (NRQ) values were log2 transformed and used for statistical analyses. The standard error of the means was calculated as described in Rieu and Powers (2009).

Statistical Analysis for gene expression: Statistical analyses were performed SigmaPlot 11

(Systat Software Inc., San Jose, CA). One way ANOVA with the general linear model was used for analyzing gene expression data for (n = 4), and was followed by Fisher’s LSD multiple comparison test for mean separation (α = 0.05). NRQ values (log2 transformed) were used for the above analyses.

Results and Discussion

Mechanical shaking for 5 s resulted in significant amount of fruit detachment in TH729 and

‘Suziblue’ in both the years. In 2012, more than 55% of the detached TH729 fruit detached at the

PPJ resulting in fruits to be stemmy. In comparison, only about 3% of the detached fruits were stemmy in ‘Suziblue’ (Figure 2). In 2013, in TH 729, 85% of the detached fruit was stemmy

132 whereas in ‘Suziblue,’ only 9% of the fruits were stemmy (Figure 2). The mechanical shaking data are in agreement with the field observations where usually TH 729 was observed to detach along with the pedicel on mild touch whereas ‘Suziblue’ did not. These data suggest that TH 729 preferentially detached at the PPJ whereas ‘Suziblue’ displayed lesser detachment at this junction. It has been previously reported that the true abscission zone of blueberry is the PPJ and that detachment at FPJ is due to mechanical breakage (Vashisth and Malladi, 2013). It was also reported that in response to mechanical shaking, blueberry fruit normally detach at the FPJ.

However, during abscission agent induced abscission, blueberry fruit primarily detached at the

PPJ upon mechanical shaking (Malladi et al., 2013), suggesting that when PPJ is loosened potentially due to the progression of abscission then the detachment takes place at the PPJ as it is weakened.

Scanning electron microscopy analysis was performed to observe the PPJ and FPJ fracture planes. The morphology of the fracture planes on the pedicel ends of the PPJ and the FPJ from TH 729 and ‘Suziblue’ were observed after manual separation of the pedicel from the peduncle and the fruit, respectively (Figures 3, 4). The pedicel end of the PPJ in TH729 displayed an even fracture plane with some disruption (Figure 3a). Small and rounded cells were visible at the surface and at the periphery of the fracture plane. These observations are consistent with the involvement of abscission during fruit detachment. Cell rupture and breakage, and some tearing of the tissues were observed at the center of fracture plane, the possible reason for this breakage could be the fact that the pedicel was manually separated from peduncle therefore the some breakage was involved within the vascular tissue. In blueberry, during natural fruit detachment and abscission induced detachment, the PPJ fracture plane appears to be even with little bit of disruption in the center possibly due to breakage of vascular bundles and also

133 abscission progresses through the disruption of cells within this region (Vashisth and Malladi,

2013). Similarly, breakage of cells within the vascular tissue was associated with the final stages of abscission progression in cherry (Stösser et al., 1969; Wittenbach and Bukovac, 1972). Also,

Gough and Litke (1980) reported that final fruit separation in blueberry occurs by rupture and mechanical tearing of the vascular bundles, although these authors indicated that fruit detachment occurred at the FPJ in northern highbush blueberry. The PPJ fracture plane of

‘Suziblue’ displayed extensive disruption and uneven surface (Figure 3b). PPJ fracture plane of

‘Suziblue’ looked very different from that of TH729 indicating that more force was involved in separating the pedicel from the peduncle in ‘Suziblue’. In both the genotypes, the fracture plane of the pedicel end of the FPJ contained parts of the fruit tissue in addition to extensively disrupted cells (Figure 4 a, b). Together, the SEM analysis indicates that fruit detachment in TH

729 occurs at the PPJ and lesser force is required to remove pedicel from peduncle possibly due to progression of abscission. On the contrary it appears that in ‘Suziblue’, the pedicel is attached strongly to the peduncle and therefore more force is required to remove the pedicel from peduncle resulting in extensive breakage and disruption of fracture plane. It is also likely that the abscission process was not yet advanced in ‘Suziblue.’ SEM analysis also indicated that detachment at FPJ is primarily due to mechanical breakage. All together mechanical shaking and

SEM data indicate that the PPJ in ‘Suziblue’ adheres strongly and therefore detachment during shaking primarily occurs at the comparatively weaker junction, FPJ.

Gene expression profiling was performed on PPJ tissue from TH729 and ‘Suziblue’. Fifty eight genes related to cell wall hydrolysis, phytohormone (ethylene, jasmonates, auxin) biosynthesis/metabolism/signaling pathway, and transcription factors were analyzed. Out of these fifty eight genes only eight were differentially expressed between TH729 and ‘Suziblue’

134 (Figure 4). One of the eight genes that was found to be differentially down-regulated in Suziblue was GLUCAN ENDO 1-3 GLUCANASE. This is a cell wall hydrolysis-related gene that has been reported to be involved in cell separation process (Abeles and Forrence, 1970). Abeles and

Forrence (1970) reported an increase in the activity of endo 1-3 glucanase in Phaseolus vulgaris during abscission. They also reported that ethylene promoted the activity of endo 1-3 glucanase and auxin suppressed the activity.The middle lamellae of AZ cells are presumably loosened by the active mobilization of cell wall hydrolases including β-1, 4-glucanases (Lashbrook et al.,

1994) initiating organ detachment. Many studies reported that ethylene and other phytohormones are involved in inducing cell wall-degrading enzymes such as glucanase (EG) in abscission zones (Goren, 1993). A very diverse gene family encodes for EGs in tomato

(Brummell et al., 1997, del Campillo and Bennet, 1996). Two members of this family, CEL1 and

CEL2, exhibit an increase in transcript abundance during fruit ripening and flower abscission

(Lashbrook et al., 1994). In cherry tomato, expression of the EG and cellulase (ENDO-1, 4-β-

GLUCANASES, CEL1, CEL2, CEL3, CEL5, CEL7 and CEL8) in the abscission zone was enhanced during abscission of the fruit upon treatment with ethylene and MeJa. 1-MCP blocked the expression of these genes in both cases of treatment with ethylene and MeJa (Beno-Moualem et al., 2004).

Other genes that were differentially expressed at lower levels in 'Suziblue' PPJ were 1-

AMINOCYCLOPROPANE -CARBOXYLATE-1-LIKE (ACO), RESPIRATORY BURST

OXIDASE, MYC 2 LIKE, CULLIN -1-ISOFORM 2, BIM-1 transcription factor, and AUXIN

DOWN REGULATED PROTEIN. All these genes are related phytohormone metabolism/signaling pathway. The role of ACO is well documented in the ethylene biosynthesis pathway. Increased transcription of ACS and ACO genes leads to an increase in ethylene

135 biosynthesis (Woodson et al., 1992). The activities of ACS and ACO are both rate-limiting steps for ethylene production (Kende and Zeevaart, 1997). During the induction of abscission increase in the ACO expression has been reported in many fruit such as citrus, apple and tomato (Agusti et al, 2009; Li et al., 2010; Meir et al., 2010). The MYC 2 LIKE gene codes for a basic helix- loop-helix (bHLH) domain containing transcription factor. It activates and represses the expression of distinct jasmonate-responsive genes in Arabidopsis (Lorenzo et al., 2003).

Dombrecht et al. (2007) demonstrated that MYC2 modulates JA responses via differential regulation of an intermediate spectrum of transcription factors with activating or repressing roles in JA signaling. Rajani and Sunderesan (2001) described a novel mutation in Arabidopsis called alcatraz (alc, ALC gene is shown to encode a protein related to the MYC family of transcription factors), which prevents dehiscence of fruit by specifically blocking the separation of the cells from the replum. RESPIRATORY BURST OXIDASE an up-regulated gene in TH729, is potentially involved in the ethylene mediated signaling pathway, jasmonic acid mediated signaling pathway, salicylic mediated signaling pathway as well as response to stress.

RESPIRATORY BURST OXIDASE was also found to be up-regulated in both ethephon and MeJa induced abscission suggesting that it may have a potential function related to abscission. BIM-1

(BES interacting Myc-like protein 1) transcription factor is a member of basic helix loop helix family and is known to be involved with brassinosteroid (BR) signaling (Yin et al., 2005).

Chandler et al., (2008) reported the loss of BIM1 function results in cell division defects in embryo development. Tissue elongation or cell expansion is considered an important response for understanding interactions between auxins and BR, but the molecular mechanisms by which they interact and regulate plant tissue elongation are poorly understood. It is suggested that BIM-

1 transcription factor is involved in phytohormone cross talk. An up-regulation in the expression

136 of AUXIN DOWN REGULATED PROTEIN was observed in TH729, potentially this gene is down-regulated by Auxin therefore an increase in TH729 may indicate decrease in auxin levels at the PPJ in this genotype. Another up-regulated gene was CULLIN -1-ISOFORM 2, which is involved in processes such as response to jasmonic acid stimulus, response to auxin stimulus, and may function in ubiquitin ligase binding. Tero et al., (2007) reported an E2 ubiquitin conjugating enzyme and two E3 ubiquitin ligases were found to be present exclusively in the abscission related libraries. Therefore, it the CULLIN -1-ISOFORM 2 may be associated with jasmonic acid and/or auxin signaling and thereby regulate abscission.

S-ADENOSYL METHYLTRANSFERASE (SAMT) expression was down-regulated in TH

729. SAMT is known to be involved in many processes metabolic/transferase processes including ethylene and polyamine biosynthesis. One of the most interesting genes in gene expression profiling was IAA AMIDO SYNTHETASE. TH 729 PPJ did not display any detectable expression of IAA AMIDO SYNTHETASE whereas expression of this gene was observed in

‘Suziblue.’ IAA–amido synthetase is known to maintain auxin homeostasis by conjugating excess IAA to amino acids. Nakano et al. (2013) reported preferential expression of IAA amide synthase homologues in tomato pre-abscissing AZs as compared to other regions. IAA AMIDO

SYNTHETASE genes are induced in auxin-abundant tissues and the gene products inactivate IAA to control auxin homeostasis (Ludwig-Muller, 2011). In abscission, the interplay between indole-3-acetic acid (IAA) and ethylene is well established (Abeles and Rubinstein, 1964;

Roberts et al., 2002; Sexton, 1997; Taylor and Whitelaw, 2001). The generally accepted model is that a basipetal IAA flux through the abscission zone (AZ) prevents abscission by rendering the

AZ insensitive to ethylene. If the source of IAA is removed, the AZ is sensitized to the action of ethylene and abscission commences (Abeles and Rubinstein, 1964; Addicott, 1982; Meir et al.,

137 2003, 2006, 2010; Rubinstein and Leopold, 1963; Sexton and Roberts, 1982). Based on these previous findings and our observation it can be suggested that potentially IAA AMIDO

SYNHTHETASE plays a critical role in maintaining optimum auxin level to prevent fruit detachment/abscission at PPJ in ‘Suziblue’ and lack of its expression in TH 729 results in fruit detachment to readily take place at the PPJ.

Together, mechanical shaking and SEM data indicate that PPJ in ‘Suziblue’ is stronger and therefore the detachment during shaking primarily occurs at the comparatively weaker FPJ.

Gene expression data indicates towards the balance model which proposes that the induction of abscission depends on a complex interplay of plant hormone concentration in addition to factors that alter the responsiveness of the tissue to the phytohormones (Gonzalez-Carranza and Roberts,

2012). Potentially in TH 729, differential expression of phytohormone related genes results in increased expression of cell wall metabolism genes such as GLUCAN ENDO GLUCOSIDASE.

This may lead to loosening and fruit detachment at the PPJ in TH729. In ‘Suziblue,’ the PPJ is stronger and adheres tightly, and therefore detachment occurs at the comparatively weaker junction, FPJ.

Conclusion

Overall from this study it can be concluded that a cross-talk or interplay of phytohormones mainly ethylene, jasmonic acid and auxin is involved in the process of abscission. It can be hypothesized that in TH 729, the lack or decreased expression of auxin related genes render the PPJ sensitive to ethylene, jasmonic acid and abscission related signals.

Also potentially GLUCAN ENDO GLUCOSIDASE plays an important in loosening of AZ or initiating cell separation in PPJ in TH729. In ‘Suziblue’ decreased levels of GLUCAN ENDO

138 GLUCOSIDASE may allow the PPJ to remain stronger. Present study suggests a high potential for breeding programs where cultivar selection for a stronger PPJ may be done to enhance the ease of mechanical harvesting. Another alternative can be use of plant growth regulators (for example, auxin based) that can make PPJ insensitive to abscission signal or which can make PPJ stronger.

139 Figure 1. A diagram of blueberry showing its potential fruit detachment points, PPJ-Pedicel-

Peduncle Junction and FPJ-Fruit-Pedicel Junction.

140 80

60

40

20 % Stemmy fruit Stemmy % 0 TH 729 Suziblue

100 80 60 40 20

% Stemmy Stemmy % fruit 0 TH 729 Suziblue

Figure 2. Percent stemmy fruit that detached in response to 5 s of mechanical shaking in TH729 and ‘Suziblue’ in 2012(top) and 2013 (bottom).

141 a b

Figure 3. Scanning electron micrographs of the pedicel end of the PPJ fracture plane in TH729

(a) and ‘Suziblue’ (b)

a b

Figure 4. Scanning electron micrographs of the pedicel end of the FPJ fracture plane in TH729

(a) and ‘Suziblue’ (b)

142 143 Figure 5. Gene expression of eight genes that were differentially expressed between TH729 and

‘Suziblue’ in the PPJ.

144 References

Abeles, F. and B. Rubinstein. 1964. Regulation of ethylene evolution and leaf abscission by auxin. Plant Physiol. 39(6):963. Addicott, F.T. 1982. Abscission. Univ of California Press, Agusti, J., P. Merelo, M. Cercos, F. Tadeo, and M. Talon. 2009. Comparative transcriptional survey between laser-microdissected cells from laminar abscission zone and petiolar cortical tissue during ethylene-promoted abscission in citrus leaves. BMC Plant Biol. 9(1):127. Beno-Moualem D, Gusev L, Dvir O, Pesis E, Meir S, Lichter A. 2004. The effects of ethylene, methyl jasmonate and 1-MCP on abscission of cherry tomatoes from the bunch and expression of endo-1, 4-β-glucanases. Plant Sci. 167, 499-507. Brummell, D.A., C.R. Bird, W. Schuch, and A.B. Bennett. 1997. An endo-1, 4-β-glucanase expressed at high levels in rapidly expanding tissues. Plant Mol. Biol. 33(1):87-95. Chandler, J.W., M. Cole, A. Flier, and W. Werr. 2009. BIM1, a bHLH protein involved in brassinosteroid signalling, controls Arabidopsis embryonic patterning via interaction with DORNRÖSCHEN and DORNRÖSCHEN-LIKE. Plant Mol. Biol. 69(1-2):57-68. del Campillo E, Bennett AB.1996. Pedicel breakstrength and cellulase gene expression during tomato flower abscission. Plant Physiol 111, 813–820. Dombrecht, B., G.P. Xue, S.J. Sprague, J.A. Kirkegaard, J.J. Ross, J.B. Reid, G.P. Fitt, N. Sewelam, P.M. Schenk, and J.M. Manners. 2007. MYC2 differentially modulates diverse jasmonate-dependent functions in Arabidopsis. The Plant Cell Online 19(7):2225-2245.

Gonzalez-Carranza, Z.H. and J.A. Roberts. 2012. Ethylene and cell separation processes. Annual Plant Reviews, The Plant Hormone Ethylene 44:246.

Goren, R. 1993. Anatomical, physiological, and hormonal aspects of abscission in citrus. Hort.Rev 15:145-182. Gough, R.E. and W. Litke. 1980. An anatomical and morphological study of abscission in highbush blueberry fruit. J. Amer. Soc. Hort. Sci. 105:335-341. Kende, H. and J. Zeevaart. 1997. The Five" Classical" Plant Hormones. Plant Cell 9(7):1197. Lashbrook, C.C., C. Gonzalez-Bosch, and A.B. Bennett. 1994. Two divergent endo-beta-1, 4- glucanase genes exhibit overlapping expression in ripening fruit and abscising flowers. The Plant Cell Online 6(10):1485-1493. Li J., H.Zhu and Y. Rongcai. 2010. Profiling the expression of genes related to ethylene biosynthesis, ethylene perception, and cell wall degradation during fruit abscission and fruit ripening in Apple. J. Am. Soc. Hort Sci 135(5), 391-401.

145 Lorenzo, O., R. Piqueras, J.J. Sánchez-Serrano, and R. Solano. 2003. ETHYLENE RESPONSE FACTOR1 integrates signals from ethylene and jasmonate pathways in plant defense. The Plant Cell Online 15(1):165-178. Ludwig-Muller J. Auxin conjugates: their role for plant development and in the evolution of land plants. J Exp Bot. 2011;62(6):1757–1773. Malladi, A., T. Vashisth, and S. NeSmith. 2013. Development and Evaluation of a Portable, Handheld Mechanical Shaker to Study Fruit Detachment in Blueberry. HortScience 48(3):394- 397. Meir, S., D.A. Hunter, J. Chen, V. Halaly, and M.S. Reid. 2006. Molecular changes occurring during acquisition of abscission competence following auxin depletion in Mirabilis jalapa. Plant Physiol. 141:1604-1616. Meir S., S. Philosph-Hadas, Sundaresan S, V.K.S. Selvara, S. Burd, R. Ophor, et al. 2010. Microarray Analysis of the Abscission-Related Transcriptome in the Tomato Flower Abscission Zone in Response to Auxin Depletion. Plant Physiol. 154 (4), 1929-1956. Mims, C.W. 1981. SEM of aeciospore formation in Puccinia bolleyana. Scanning Electron Microsc. III: 299-303.

Nakano, T., M. Fujisawa, Y. Shima, and Y. Ito. 2013. Expression profiling of tomato pre- abscission pedicels provides insights into abscission zone properties including competence to respond to abscission signals. BMC Plant Biol. 13(1):40.

NeSmith, D.S. 2010. 'Suziblue'Southern Highbush Blueberry. HortSci. 45:142-143.

Pfaffl, M.W. 2001. A new mathematical model for relative quantification in real-time RT–PCR. Nucleic Acids Res. 29(9):e45-e45.

Rajani, S. and V. Sundaresan. 2001. The Arabidopsis myc/bHLH gene ALCATRAZ enables cell separation in fruit dehiscence. Current Biol.11(24):1914-1922.

Ramakers, C., J.M. Ruijter, R.H.L. Deprez, and A.F. Moorman. 2003. Assumption-free analysis of quantitative real-time polymerase chain reaction (PCR) data. Neurosci. Lett. 339(1):62-66.

Rieu, I. and S.J. Powers. 2009. Real-time quantitative RT-PCR: design, calculations, and statistics. The Plant Cell Online 21(4):1031-1033.

Rubinstein, B. and A. Leopold. 1963. Analysis of the auxin control of bean leaf abscission. Plant Physiol. 38(3):262.

Ruijter, J., S. van der Velden, and A. Ilgun. 2009. LinReg PCR (11.0). Analysis of quantitative.

Sexton, R. and J.A. Roberts. 1982. Cell biology of abscission. Annu. Rev. Plant Physiol. 33:133- 162.

Sexton R. (1997). The role of ethylene and auxin in abscission. Acta Hortic463: 435–444.

146 Stösser, R., H.P. Rasmussen, and M.J. Bukovac. 1969. A histological study of abscission layer formation in cherry fruit during maturation. J. Amer. Soc. Hort. Sci. 94:239–243. Takeda, F., G. Krewer, E. Andrews, D. Peterson, and J. Mullinix. 2008. Assessment of v45 over- the-row mechanical harvester on pruned rabbiteye and southern highbush blueberry. HortTechnology 18:130-138.

Taylor, J.E. and C.A. Whitelaw. 2001. Signals in abscission. New Phytol. 151:323-339.

Terol, J., A. Conesa, J.M. Colmenero, M. Cercos, F.R. Tadeo, J. Agusti, E. Alos, F. Andres, G. Soler, J. Brumos, and D.J. Iglesias. 2007. Analysis of 13000 unique Citrus clusters associated with fruit quality, production and salinity tolerance. BMC Genomics 8:31.

U.S. Department of Agriculture. 2012. Non-citrus fruits and nuts 2011 summary. National Agricultural Statistics Service, Washington, DC.

Vashisth, T., L.K. Johnson, and A. Malladi. 2011. An efficient RNA isolation procedure and identification of reference genes for normalization of gene expression in blueberry. Plant Cell Rep. 30(12):2167-2176.

Vashisth, T. and A. Malladi. 2013. Fruit Detachment in Rabbiteye Blueberry: Abscission and Physical Separation. J. Am. Soc. Hort. Sci. 138(2):95-101.

Wittenbach, V.A. and M.J. Bukovac. 1972. An anatomical and histochemical study of abscission in maturing sweet cherry fruit. J. Amer. Soc. Hort. Sci. 97:214-219.

Woodson, W.R., K.Y. Park, A. Drory, P.B. Larsen, and H. Wang. 1992. Expression of ethylene biosynthetic pathway transcripts in senescing carnation flowers. Plant Physiol. 99(2):526-532.

Yin, Y., D. Vafeados, Y. Tao, S. Yoshida, T. Asami, and J. Chory. 2005. A New Class of Transcription Factors Mediates Brassinosteroid-Regulated Gene Expression in Arabidopsis. Cell 120(2):249-259.

147 CHAPTER 6

CONCLUSION

The process of fruit detachment in blueberry was studied. This research presents a better understanding of fruit detachment in blueberry. Anatomical analyses clearly indicated the presence of an abscission zone at the Pedicel-Peduncle junction (PPJ), but did not support the presence of an AZ at the fruit-pedicel Junction FPJ in rabbiteye blueberry. The PPJ appears to be the main point of fruit detachment when the mature fruit naturally drop as a result of the physiological and developmental progression of abscission. This conclusion is supported by the presence of a largely smooth fracture plane along with mild tissue disruption, primarily within the vascular tissue in the naturally detached fruit. Mechanical shaking alone resulted in fruit detachment at the FPJ. It is likely that fruit detachment at the FPJ is a result of mechanical breakage and physical separation. The FPJ may present a weak junction and the oscillations of the berry around this point during mechanical shaking may result in physical separation at this junction.

Glycome profiling, immunolocalization and gene expression profiling, suggest extensive cell wall remodeling upon the induction of abscission by induced by two different plant growth hormones, the ethylene releasing compound, ethephon, and Methyl Jasmonate (MeJa).

Xyloglucans (XG) and pectins were altered greatly during this process. Xyloglucans and pectins are both quantitatively important polymers of plant cell walls comprising approximately a third each of the polysaccharides of primary cell walls of dicotyledons. It is proposed that during blueberry fruit abscission the linkage between XGs and cellulose, and XGs and pectins are

148 altered in the AZ in addition to the alteration of the pectic polymers, thereby leading to cell separation.

RNA-Seq approach provides a comprehensive view on blueberry fruit abscission induced by Methyl Jasmonate (MeJa) and ethephon. Genes associated with cell wall biosynthesis and metabolism were differentially expressed in ethephon and MeJa induced abscission treatments potentially implicating them in organ shedding. These data also suggest that in addition to modification of pectins, hemicelluloses are also remodeled and altered during cell separation process. RNA-Seq data supports previous studies where a complex interplay among plant hormones has been suggested to contribute to the regulation abscission. A large number of unknown and uncharacterized genes were also differentially expressed during the induction of abscission which provides a potential source of new gene discovery.

Present study also showed that variability in ease of fruit detachment can be due to stronger PPJ junction. A stronger PPJ junction can be due of differential phyto-hormone interaction and reduction in cell wall hydrolysis related gene expression.

Overall this studies shows that PPJ is the true abscission zone of blueberry. Alteration and remodeling of pectins and hemicellulose potentially results in cell separation. A complex interplay among plant hormones potentially regulates abscission. Present study suggests a high potential for breeding programs where cultivar selection for a stronger PPJ may be done to enhance the ease of mechanical harvesting.

Knowledge of fruit detachment gained from this study has potential implications for mechanical harvesting of blueberry fruit. Abscission agents have been shown to considerably enhance the extent of fruit detachment. However, as the physiological process of abscission is involved in such detachment, it is likely that the applications of these agents will result in

149 ‘stemmy’ fruit which will subsequently require de-stemming as the presence of the pedicel on the fresh fruit is considered to be a defect that reduces its quality. Use of plant growth regulators which make PPJ insensitive to abscission signals can also be an alternative. Additionally, breeding programs aiming to improve mechanical harvesting traits may need to select genotypes with stronger pedicel-peduncle junctions so that fruit detachment during mechanical harvesting in such genotypes can occur at the weak union along the FPJ. In conjunction with the above, selection for a dry or a small stem scar at this junction may help maintain the post-harvest quality of the fruit.

150