<%BANNER%>

Effects of Sugar Metabolism Mutations on Ethylene Production and Related Transcript Levels in Developing Maize Seeds

Permanent Link: http://ufdc.ufl.edu/UFE0024674/00001

Material Information

Title: Effects of Sugar Metabolism Mutations on Ethylene Production and Related Transcript Levels in Developing Maize Seeds
Physical Description: 1 online resource (107 p.)
Language: english
Creator: Funk, Andrew
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: development, ethylene, hexose, hormone, kernel, maize, metabolism, miniature1, signaling, sugary1
Plant Pathology -- Dissertations, Academic -- UF
Genre: Plant Pathology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The seeds of cereal crops are critically important to the global food supply. Maize, one of the most important U.S. field crops, is an ideal model system to study cereal seed development due to its large size and well-defined tissue and cell types. Seed development is a complex process with distinct developmental phases, including cell division, cell expansion, endoreduplication, starch and protein storage, and kernel maturation. The timing and progression of kernel development is tightly regulated by factors such as sugar status and phytohormone activity. Interactions between sugars and hormones, forms of ?cross-talk,? have emerged as an important subject of study in recent years. This report focused primarily on the Mn1 gene, which encodes the INCW2 cell-wall invertase protein. The miniature1 (mn1) kernel mutation eliminates INCW2 activity, which causes decreased endosperm growth and a final seed weight 30% of the wild-type value. Samples were also collected from sugary1 (su1) mutant kernels deficient in a starch-debranching enzyme and a related wild-type accession as a control. Both ethylene production and related transcript levels were analyzed in these genotypes in order to compare the possible effects of genetic backgrounds and the effect of early- and late-acting metabolism mutations. Transcript levels of three metabolic genes, Mn1, Sus2 and HXK2, were analyzed via quantitative polymerase-chain-reaction (qPCR) techniques. Additionally, genes critical to ethylene biosynthesis and perception were included in the qPCR analysis in order to correlate transcript levels with ethylene activity. Genes under consideration were the ACC synthase genes ACS2, ACS6 and ACS7; two ACC oxidase genes ACO20 and ACO35; three ethylene receptors ERS1-14, ETR2-9 and ETR2-40; and the DNA binding protein EIL1-1 that is involved in downstream ethylene signaling. Results indicated distinct bursts of ethylene in all sample series, although the quantity and timing of ethylene production varied between genetic backgrounds. The Mn1 samples produced an increase in ethylene 13 days after pollination (DAP) that correlated with maximum published INCW2 activity. The smaller mn1 kernels did not display a 13 DAP ethylene burst but produced 2-fold higher levels of ethylene than the wild-type between 16 and 25 DAP, coinciding with increased sucrose content in the defective endosperms. The wild-type Su1 kernels generated two distinct peaks of ethylene production, one at 16 DAP and the second at 24 DAP. The su1 samples did not produce the second 25 DAP burst of ethylene, despite increased sucrose levels. Results of transcript analysis indicated HXK2 was reduced up to 2-fold in mn1 samples during early development, consistent with the deficiency in hexoses present in that genotype. Expression of both the ACC synthase and ACC oxidase family members appeared to be related to developmental stage more than kernel genotype. The ethylene receptors displayed relatively constitutive expression, while the transcription factor EIL1-1 was more highly expressed late in development.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Andrew Funk.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Chourey, Prem S.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024674:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024674/00001

Material Information

Title: Effects of Sugar Metabolism Mutations on Ethylene Production and Related Transcript Levels in Developing Maize Seeds
Physical Description: 1 online resource (107 p.)
Language: english
Creator: Funk, Andrew
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: development, ethylene, hexose, hormone, kernel, maize, metabolism, miniature1, signaling, sugary1
Plant Pathology -- Dissertations, Academic -- UF
Genre: Plant Pathology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The seeds of cereal crops are critically important to the global food supply. Maize, one of the most important U.S. field crops, is an ideal model system to study cereal seed development due to its large size and well-defined tissue and cell types. Seed development is a complex process with distinct developmental phases, including cell division, cell expansion, endoreduplication, starch and protein storage, and kernel maturation. The timing and progression of kernel development is tightly regulated by factors such as sugar status and phytohormone activity. Interactions between sugars and hormones, forms of ?cross-talk,? have emerged as an important subject of study in recent years. This report focused primarily on the Mn1 gene, which encodes the INCW2 cell-wall invertase protein. The miniature1 (mn1) kernel mutation eliminates INCW2 activity, which causes decreased endosperm growth and a final seed weight 30% of the wild-type value. Samples were also collected from sugary1 (su1) mutant kernels deficient in a starch-debranching enzyme and a related wild-type accession as a control. Both ethylene production and related transcript levels were analyzed in these genotypes in order to compare the possible effects of genetic backgrounds and the effect of early- and late-acting metabolism mutations. Transcript levels of three metabolic genes, Mn1, Sus2 and HXK2, were analyzed via quantitative polymerase-chain-reaction (qPCR) techniques. Additionally, genes critical to ethylene biosynthesis and perception were included in the qPCR analysis in order to correlate transcript levels with ethylene activity. Genes under consideration were the ACC synthase genes ACS2, ACS6 and ACS7; two ACC oxidase genes ACO20 and ACO35; three ethylene receptors ERS1-14, ETR2-9 and ETR2-40; and the DNA binding protein EIL1-1 that is involved in downstream ethylene signaling. Results indicated distinct bursts of ethylene in all sample series, although the quantity and timing of ethylene production varied between genetic backgrounds. The Mn1 samples produced an increase in ethylene 13 days after pollination (DAP) that correlated with maximum published INCW2 activity. The smaller mn1 kernels did not display a 13 DAP ethylene burst but produced 2-fold higher levels of ethylene than the wild-type between 16 and 25 DAP, coinciding with increased sucrose content in the defective endosperms. The wild-type Su1 kernels generated two distinct peaks of ethylene production, one at 16 DAP and the second at 24 DAP. The su1 samples did not produce the second 25 DAP burst of ethylene, despite increased sucrose levels. Results of transcript analysis indicated HXK2 was reduced up to 2-fold in mn1 samples during early development, consistent with the deficiency in hexoses present in that genotype. Expression of both the ACC synthase and ACC oxidase family members appeared to be related to developmental stage more than kernel genotype. The ethylene receptors displayed relatively constitutive expression, while the transcription factor EIL1-1 was more highly expressed late in development.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Andrew Funk.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Chourey, Prem S.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024674:00001


This item has the following downloads:


Full Text





EFFECTS OF SUGAR METABOLISM MUTATIONS ON ETHYLENE PRODUCTION AND
RELATED TRANSCRIPT LEVELS IN DEVELOPING MAIZE SEEDS





















By

ANDREW JOSEPH FUNK


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2009


































2009 Andrew Joseph Funk



































To my family









ACKNOWLEDGMENTS

There are a host of people that contributed to my development, both personally and

professionally, during the pursuit of this degree. I would like to acknowledge the work of my

advisor, Dr. Prem Chourey, and his valuable advice throughout the process of designing,

conducting, and communicating this research. Thanks to the other members of my advisory

committee, Dr. Harry Klee and Dr. Wen-Yuan Song, for keeping their doors open to me even

when I was too naive to ask for their help, and graciously providing assistance and critique

during the final stages of my studies. Thanks to Dr. Peter Teal for believing that somewhere

inside me is a worthwhile employee that just needs some direction. Thanks to Dr. Don Huber for

the use of his gas chromatography equipment, and much appreciation for the expert assistance

and instruction of James Lee in utilizing that equipment. Thanks to Qin Bao Li for making his

wealth of experience available to me while teaching me what is important and what "doesn't

matter." A special thanks to my friend and original scientific mentor, Mukesh Jain, and his wife

Rani, who took me into their home and expanded my understanding of kindness and hospitality.

I would like to acknowledge the good example provided by Dr. Fonsie Guilaran and his wife

Lesley, whose friendship has been a source of encouragement and admonishment as I have

grown into adulthood. I gladly acknowledge the unfailing love and support provided by my

father Joesph, my mother Donna, and my sister Marjorie, without whom my life would be a

fractured shard instead of part of a whole. Over all of these things I acknowledge God, who

offers life and heals what is broken; even me.










TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ..............................................................................................................4

ABSTRACT ........... ................... ................................. ......... 10

CHAPTER

1 IN TRODU CTION ................... .......................................... .. ........ ..............12

Kernel Development................. .. ................12
Ethylene Biosynthesis, Perception and Signaling ...................................... ........... ....15
H exokinase and Sugar Signaling .................................................................................. ... .....18
D description of M utant Lines ................................................................... .......19
The mn] M station ................. ..............................19
The sul Mutation............................... .. .... .......... 21

2 M ATERIALS AND M ETHODS ................................................. ...............22

Field W ork and Fresh M material .................................................................................... ..... .....22
Planting......................................................... ................ ........ 22
Harvest......................... ............................... 22
Gas Chrom atography Analysis.................................... ............... 23
Nucleic Acid Preparation.................. ............. .........23
RNA Isolation................................ ........23
DNase treatment...................................... ................................ ........24
R reverse transcription ................... ............... .. ......... ...................25
G ene-specific A analysis ................. ......................................................... 25
Prim er D design ................................................................... ................... 25
Cloning ....................... .... .. ........... .......... 26
DN A Sequencing Reaction................................................... 27
A absolute Quantitative PCR .................................................. ............... 28

3 RE SU LTS ........................................................................... 32

The M n and m n G enotypes ........................................................................................ ......32
Ethylene Production ............ .......................................... .. ....32
Transcript Accumulation in Mn] and mn] Genotypes..................................................34
M etabolic genes..................... .............. ......... 34
The ACC synthase gene family............................................................... ...... 36
The ACC oxidase gene family ................................................ 38
The ethylene receptor gene family and EIL7-1..................................................39
The Sul and sul G enotypes...............................................41
Ethylene Production ......................................... ........... ...... 41
Transcript Accumulation in Sul and sul Genotypes .................................................43
M etabolic genes.................................... .......... 43









The ACC synthase gene family.......................................................... ... ...... 45
The ACC oxidase gene family ....................................... ......... 47
The ethylene receptor gene family and EIL7-1............ ...... ...............48

4 D ISCU SSION .................. .......................................................... ........ 79

Ethylene Accumulation in Developing Seeds .................................80
Mn] Seeds Produced a Distinct Peak of Ethylene between 12 and 14 DAP ...............81
Trends in mn] Kernel Ethylene Production Were Varied Over Two Consecutive
Y ears ............................... ....... .. ... .. .......... .. ... ................8 1
Hexose Deficiency and Increased Sucrose Lead to Pleiotropic Effects in Maize
Seeds .................. .... ... .................. .... ...... .. ...............82
Sul and sul Kernels Displayed Inconsistent Ethylene Production over Two
Consecutive Years ............................. ............ ..... ... .... .......83
Sul and sul kernels showed two clear peaks of ethylene production prior to 30
D A P ....................... .......... .... .. ........ ...............84
A second peak of ethylene hormone in Sul samples from year 2008 showed
unique tim ing ............... ......... ......... .... .. ..... ....... ......... 86
Transcript Accumulation and Correlation with Ethylene Hormone Evolution.....................86
Transcript Accumulation in Mnl and mn] Genotypes..................................................87
M etabolic genes............... .............. .......... .. .. .. .. .......... 87
Ethylene biosynthesis genes of the ACS and ACO families........................89
Ethylene receptor genes and EIL -1 ................. ................. ............... 91
Transcript Accumulation in Sul and sul Genotypes ...................................... 92
M etabolic genes............... .............. .......... .. .. .. .. .......... 92
Ethylene biosynthesis genes of the ACS and ACO families........................94
Ethylene receptor genes and EIL -1 ................. ................. ............... 96

5 CONCLUSIONS ................................................ ........ 98

LIST OF REFERENCES ................................................................. ........ 101

BIO GR A PH ICA L SK ETCH .............................................................................................. ........ 107









LIST OF TABLES
Table page

2-1 Components of 50mL RNA isolation buffer. .................................... ...............30

2-2 List of qPCR prim ers 5' to 3' ..................... .................................................................31









LIST OF FIGURES
Figure paMe

3-1 Ethylene produced by Mn] (blue) and mn] (pink) kernels, nmols of ................ .......51

3-2 Ethylene produced by Mn] (blue) and mn] (pink) kernels, nmols of ..............................52

3-3 Mn] transcript levels in Mn] (blue) and mn] (pink) kernels...........................................53

3-4 Sus2 transcript levels in Mn] (blue) and mn] (pink) kernels............................... 54

3-5 HXK2 transcript levels in Mn] (blue) and mn] (pink) kernels.......................................55

3-6 ACS2 transcript levels in Mn] (blue) and mn] (pink) kernels........................................56

3-7 ACS6 transcript levels in Mn] (blue) and mn] (pink) kernels........................................57

3-8 ACS7 transcript levels in Mn] (blue) and mn] (pink) kernels........................................58

3-9 AC020 transcript levels in Mn] (blue) and mn] (pink) kernels.....................................59

3-10 AC035 transcript levels in Mn] (blue) and mn] (pink) kernels.....................................60

3-11 ERS1-14 transcript levels in Mn] (blue) and mn] (pink) kernels...................................61

3-12 ETR2-9 transcript levels in Mn] (blue) and mn] (pink) kernels.....................................62

3-13 ETR2-40 transcript levels in Mn] (blue) and mn] (pink) kernels...................................63

3-14 EIL1-1 transcript levels in Mn] (blue) and mn] (pink) kernels......................................64

3-15 Ethylene produced by Sul (blue) and sul (pink) kernels, nmols of...............................65

3-16 Ethylene produced by Sul (blue) and sul (pink) kernels, nmols of...............................66

3-17 Sul transcript levels in Sul (blue) and sul (pink) kernels...............................................67

3-18 Sus2 transcript levels in Sul (blue) and sul (pink) kernels .........................................68

3-19 HXK2 transcript levels in Sul (blue) and sul (pink) kernels............... .. ............69

3-20 ACS2 transcript levels in Sul (blue) and sul (pink) kernels ..............................................70

3-21 ACS6 transcript levels in Sul (blue) and sul (pink) kernels...........................................71

3-22 ACS7 transcript levels in Sul (blue) and sul (pink) kernels...........................................72

3-23 AC020 transcript levels in Sul (blue) and sul (pink) kernels................ .............73









3-24 AC035 transcript levels in Sul (blue) and sul (pink) kernels................ ............. 74

3-25 ERS1-14 transcript levels in Sul (blue) and sul (pink) kernels......................................75

3-26 ETR2-9 transcript levels in Sul (blue) and sul (pink) kernels .......................................76

3-27 ETR2-40 transcript levels in Sul (blue) and sul (pink) kernels .....................................77

3-28 EIL1-1 transcript levels in Sul (blue) and sul (pink) kernels .......................................78










Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

EFFECTS OF SUGAR METABOLISM MUTATIONS ON ETHYLENE PRODUCTION AND
RELATED TRANSCRIPT LEVELS IN DEVELOPING MAIZE SEEDS

By

Andrew Joseph Funk

August 2009

Chair: Prem S. Chourey
Major: Plant Pathology

The seeds of cereal crops are critically important to the global food supply. Maize, one of

the most important U.S. field crops, is an ideal model system to study cereal seed development

due to its large size and well-defined tissue and cell types. Seed development is a complex

process with distinct developmental phases, including cell division, cell expansion,

endoreduplication, starch and protein storage, and kernel maturation. The timing and progression

of kernel development is tightly regulated by factors such as sugar status and phytohormone

activity. Interactions between sugars and hormones, forms of "cross-talk," have emerged as an

important subject of study in recent years.

This report focused primarily on the Mn] gene, which encodes the INCW2 cell-wall

invertase protein. The miniature] (mn]) kernel mutation eliminates INCW2 activity, which

causes decreased endosperm growth and a final seed weight 30% of the wild-type value.

Samples were also collected from sugary] (sul) mutant kernels deficient in a starch-debranching

enzyme and a related wild-type accession as a control. Both ethylene production and related

transcript levels were analyzed in these genotypes in order to compare the possible effects of

genetic backgrounds and the effect of early- and late-acting metabolism mutations. Transcript









levels of three metabolic genes, Mn], Sus2 and HXK2, were analyzed via quantitative

polymerase-chain-reaction (qPCR) techniques. Additionally, genes critical to ethylene

biosynthesis and perception were included in the qPCR analysis in order to correlate transcript

levels with ethylene activity. Genes under consideration were the ACC synthase genes ACS2,

ACS6 and ACS7; two ACC oxidase genes AC020 and AC035; three ethylene receptors ERS1-14,

ETR2-9 and ETR2-40; and the DNA binding protein EIL1-1 that is involved in downstream

ethylene signaling.

Results indicated distinct bursts of ethylene in all sample series, although the quantity and

timing of ethylene production varied between genetic backgrounds. The Mn] samples produced

an increase in ethylene 13 days after pollination (DAP) that correlated with maximum published

INCW2 activity. The smaller mn] kernels did not display a 13 DAP ethylene burst but produced

2-fold higher levels of ethylene than the wild-type between 16 and 25 DAP, coinciding with

increased sucrose content in the defective endosperms. The wild-type Sul kernels generated two

distinct peaks of ethylene production, one at 16 DAP and the second at 24 DAP. The sul

samples did not produce the second 25 DAP burst of ethylene, despite increased sucrose levels.

Results of transcript analysis indicated HXK2 was reduced up to 2-fold in mn] samples during

early development, consistent with the deficiency in hexoses present in that genotype.

Expression of both the ACC synthase and ACC oxidase family members appeared to be related

to developmental stage more than kernel genotype. The ethylene receptors displayed relatively

constitutive expression, while the transcription factor EIL1-1 was more highly expressed late in

development.









CHAPTER 1
INTRODUCTION

Kernel Development

Maize seed formation is the result of coordinated development of various specialized

tissues. Progress of kernel development can be monitored in terms of overlapping phases related

to cell proliferation, differentiation, and maturation (Bosnes et al. 1992). Pollination results in a

double-fertilization event common in angiosperms: one of the two sperm nuclei contained in a

single male pollen grain fuses with an egg nuclei contained in the female megagametophyte,

creating a 2N zygote. The remaining haploid sperm nucleus fuses with the two female polar

nuclei, creating a 3N endosperm cell (Kiesselbach 1949; Diboll 1968). These two new cells

undergo distinct but related processes, ultimately leading to a quiescent embryo ready to

germinate, utilizing storage reserves contained in the terminal endosperm.

Immediately after fertilization the zygote undergoes rapid cell division to form the mass of

cells that will become the embryo. The 3N endosperm cell nucleus also divides rapidly, but the

absence of cytokinesis leads to a large, syncytial, highly multinucleate endosperm cell. By

approximately three days after pollination (DAP) the endosperm begins cellularization, starting

with the nuclei at the periphery of the cell and proceeding centripetally (Kiesselbach 1949;

Kowles and Phillips 1988). Once the endosperm cells are all cellularized, they begin both cell

division and expansion. From 8-12 DAP the endosperm reaches peak cell division, and after 12

DAP the cells begin the transition from differentiation to maturation, with the exception of the

subaleurone region continuing division until -20 DAP (Kiesselbach 1949; Kyle and Styles

1977). The maturation phase involves storage protein synthesis and starch loading, beginning in

the central endosperm and proceeding toward the periphery ca. 15 DAP. As a part of maturation,

endosperm cells undergo endoreduplication, a repetitive amplification of nuclear DNA without









normal cell division. The main onset of endoreduplication picks up as mitosis ends

approximately 12 DAP, with a peak in endopolyploidy levels at 16-18 DAP (Kowles and Phillips

1985). This process is widespread in cells with high metabolic activity and is thought to facilitate

the increased levels of transcription required for rapid cell enlargement and reserve storage

(Grime and Mowforth 1982), although some reports have noted disparity between

endoreduplication levels and cell division and enlargement (Vilhar et al. 2002). Ultimately, the

purpose of endoreduplication is unclear.

Between 12 and 22 DAP, maize kernels begin accumulation of dry matter, with a

coordinate increase in the presence of enzymes related to starch and storage protein biosynthesis.

Expression of starch synthesis and protein synthesis genes occurs in concert with a peak at 20-25

DAP, resulting in coordinated starch synthesis in the central endosperm cells as well as oil and

protein production increasing toward the aleurone and subaleurone layers (Tsai et al. 1970). The

main storage carbohydrates in maize kernels are homopolymers of D-glucose: the moderately

branched amylopectin, and the predominantly straight-chain amylose. The two polymers are

thought to be synthesized concurrently, with the coordinated action of multiple enzymes leading

to a final ratio of 3:1 amylopectin to amylose (Shannon et al. 1970; Myers et al. 2000). A third,

highly-branched, water-soluble polysaccharide phytoglycogen, is detected in very small

quantities but is greatly increased when various combinations of starch synthesis genes are

defective (Black et al. 1966).

The first committed step in starch biosynthesis is the largely cytosolic conversion of

glucose-1-phosphate (GIP) to ADP-glucose (ADP-Glc) by the enzyme ADP-glucose

pyrophosphorylase (AGP) (Dickinson and Preiss 1969). AGP is a heterotetrameric enzyme

encoded by the large subunit \,/ inIik/.'I (Sh2) and small subunit Brittle2 (Bt2) genes in maize









(Bhave et al. 1990; Preiss et al. 1990). Distinct cytosolic and plastidial forms of AGP are thought

to exist, with >85% of the activity in maize endosperm coming from the cytosolic enzyme

(Denyer et al. 1996; Beckles et al. 2001). ADP-glucose is transported into the amyloplast by

membrane-bound BRITTLE1 (BT1) encoded by the Btl gene (Shannon et al. 1998).

Inside the amyloplast, ADP-Glc monomers are linked through the formation of a-1,4

glycosidic bonds via the action of a family of at least five starch synthase (SS) enzymes. These

include a granule-bound starch synthase (GBSS) encoded by the Waxy (Wx) gene and four

soluble SSs designated SS1, SSIIa, SSIIb and DULL1/SSIII. Amylose is produced by GBSS1,

and amylopectin is primarily the result of zSSI and DU1/SSIII activity (Shure et al. 1983; Cao et

al. 1999). Two additional classes of enzymes are required to achieve the specific structure

allowing amylopectin to crystallize into insoluble starch grains: starch branching enzyme (BE)

and starch debranching enzyme (DBE) (Creech 1965; Nelson and Pan 1995).

Maize contains three known BEs with the ability to form branching a-1,6 glycosidic bonds

via cleavage and transfer of a-1,4 bonds of linear glucose polymers (Boyer and Preiss 1978).

These BEs comprise two classes in maize; BEI shows 10-fold preference for amylose (long-

chain polymer) as a substrate, while BEIIa and BEIIb transfer shorter chains. It is possible that

BEI provides substrate for BEIIa and BEIIb (Nelson and Pan 1995). Evidence suggests that SSs

and BEs associate in multi-subunit complexes, most notably SSI/SSIIa and SSI/BEIIb (Hennen-

Beirwagen et al. 2008).

The action of SSs and BEs alone results in substantial production of phytoglycogen at the

expense of amylose and amylopectin (Black 1966). DBEs are required to achieve proper

distribution of chain length. Two families are present in plants; isoamylases and pullulanases.

The maize isoamylase is represented by Sugary] (Sul), with specificity for amylopectin but not









pullulan. The pullulanase is represented by Zpul, having affinity for pullulan yet still able to

attack amylopectin in a lesser fashion (Dinges et al. 2001; Wu et al. 2002). The maize sul

mutation eliminates both isoamylase and pullulanase activity (Beatty et al. 1999).

The final stage of endosperm development is the initiation of programmed cell death

(PCD). Between 16 and 20 DAP cells in the central endosperm begin to lose viability.

Endonucleases dismantle DNA into fragments with sizes in multiples of 180-200 bp, first

detectable at 28 DAP (Young et al. 1997). The progression of PCD follows the spatial pattern of

endoreduplication and starch synthesis, initiating in the crown-proximal region and central

endosperm, then proceeding down toward the base of the kernel and out toward the periphery.

The only viable cells in the mature kernel are in the aleurone layer and quiescent embryo (Young

and Gallie 2000a).

Ethylene Biosynthesis, Perception and Signaling

The two-carbon molecule ethylene is the simplest phytohormone identified to date, yet is

involved in diverse processes related to seed germination, root and shoot elongation, senescence,

fruit ripening, and both biotic and abiotic stress response (Johnson and Ecker 1998). It is a

gaseous molecule able to diffuse freely through membranes, and thus the response must be

tightly regulated via control of the biosynthetic and signaling machinery. Ethylene is produced in

a two-step sequence beginning with ACC synthase enzymatically cleaving a structure off S-

adenosyl-L-methionine to form 1 -aminocyclopropane-l-carboxylic acid (ACC). The resulting

ACC is rapidly converted to ethylene via the enzyme ACC oxidase. The ACC synthase step is

considered rate-limiting for ethylene production (Yang and Hoffman 1984).

Ethylene is perceived via interaction with endoplasmic-reticulum (ER)-localized receptors

resembling bacterial histidine and serine/threonine kinases. Five types of ethylene receptors have

been identified in Arabidopsis to date and can be grouped into two subfamilies based on









homology (Hua and Meyerowitz 1998). Screens for Zea mays have yielded orthologs of two of

the five receptor types established in Arabidopsis (Gallie and Young 2004). The two groups are

ETHYLENE-RESPONSE2 (ETR2) and ETHYLENE RESPONSE SENSOR (ERS 1). Both

maize types have two members; ZmERS-14/ ZmERSl-25 and ZmETR2-9 ZmETR2-40. While the

gene products within the subgroups are 97% and 92% identical, respectively, the similarity

between ERS1 and ETR2 is only 51% (41% identical) at the amino acid level (Gallie and Young

2004).

Ethylene receptors are negative regulators of ethylene signaling (Hua and Meyerowitz

1998). In the absence of the hormone, the receptors activate the Raf-like kinase CTR1, which is

thought to control a phosphorylation cascade that ultimately represses the transcriptional

regulator ETHYLENE-INSENSITIVE-3 (EIN3) (Kieber et al. 1993; Clark et al. 1998) EIN3

and a homologous EIN3-LIKE1 (EIL1) subfamily control a transcriptional cascade related to

production of diverse ethylene-responsive genes (Chao et al. 1997). EIN3 and EIL are regulated

via phosphorylation state (Yoo et al. 2008) and degradation via the 26S proteasome pathway,

with known F-box proteins ETHYLENE-F-BOX-1 (EBF1) and ETHYLENE-F-BOX-2 (EBF2)

providing targeting specificity (Gou and Ecker 2003). ETHYLENE-INSENSIVITE-2 (EIN2)

functions downstream of CTR1 to repress EBF1/2 action, while EIN2 itself is repressed by

CTR1 (Alonso et al. 1999). Thus ethylene binding to the receptors inactivates CTR1, releases

EIN2 from repression, hinders EBF 1/2 and allows EIN3/EIL1 to initiate downstream ethylene

responses.

While the majority of ethylene research has been done in Arabidopsis and other model

systems such as tomato, some progress has been made in understanding ethylene activity in

developing maize seed. Early reports quantified the pattern of ethylene production in relation to









PCD, showing two peaks of hormone accumulation in many lines tested (Young et al. 1997). It is

of special importance that the patterns and absolute values of ethylene production varied between

genetic backgrounds, both for wild-types and identical single-gene mutant lines. A common

trend is that ethylene peaks between 12 and 16 DAP, coinciding with the first observation of cell

death in the central endosperm (Young et al. 1997). A second peak of ethylene is generated

roughly 28-36 DAP, accompanying the appearance of internucleosomal DNA fragmentation. In

sh2 mutant lines containing increased sucrose and glucose levels, ethylene generation is more

abundant than the control: this occurs in conjunction with earlier induction of cell death as well

as earlier and increased nuclease activity and a reduction in germination rate (Young et al. 1997).

In addition, it has been demonstrated that PCD can be modulated by a variety of ethylene-related

effects. Application of ethylene increased the severity of cell death and induction of nucleases in

all lines tested, and was sufficient to cause some cell death in tissues that normally remained

viable. Ethylene biosynthesis or perception inhibitors such as 2-aminoethoxyvinyl glycine

(AVG) and 1-methylcyclopropene (MCP), respectively, delayed the onset and severity of PCD

(Young et al. 1997; Young and Gallie 2000a). Finally, abscisic acid (ABA) has been shown to

antagonize ethylene, and kernels with impaired ABA synthesis or perception generate more

ethylene than controls and also show increased internucleosomal DNA fragmentation (Young

and Gallie 2000b).

Gallie and Young (2004) isolated and characterized the expression of many critical

components of the ethylene pathway from a collection of maize genomic and cDNA libraries.

They identified three ACS genes, four ACO genes, two ERS1-family receptors, two ETR2-family

receptors, a single EIN2 and two EIL1 members. It is of note that maize appears to lack ETR1,

ERS2, and NR family receptors. Their attempts at isolating a CTR gene were unsuccessful.









Hexokinase and Sugar Signaling

The hexokinase family of enzymes catalyzes the conversion of hexose to hexose-6-

phosphate in an ATP-dependent reaction. This enzyme is conserved across diverse animal, plant,

and yeast species (Slein et al. 1950; Saltman 1953; Dai et al. 1995). Hexokinase is positioned at

the gateway of hexose utilization as the first step in glycolysis, and has been linked to the

phenomenon of hexoses as a signaling molecule in the control of diverse processes, including

germination, seedling development, photosynthetic gene regulation, source/sink partitioning,

flowering and senescence (Foyer 1988; Sheen 1990; Jang and Sheen 1994; Pego et al. 1999;

Ohto et al. 2001). Furthermore, it was established that glucose and other hexoses, but not sugar

phosphates, are the direct sugar signals causing repression of photosynthetic genes (Jang and

Sheen 1994). With the discovery of distinct and separable catalytic and regulatory domains (Jang

et al. 1997), hexokinases emerged as a critical component of sugar signaling. Characterization of

the GLUCOSE-INSENSITIVE1 (gin]) mutant revealed overlap of ethylene and abscisic acid

hormone pathways, with ABA2 (GIN]) acting downstream of HXK1 to antagonize a branch of

ethylene response related to germination, cotyledon and leaf development, and flowering (Zhou

et al. 1998).

Subsequent research garnered support for the hypothesis that three sugar signaling

pathways exist in higher plants: one dependent on HXK catalytic function, another related to

HXK signaling, and a third functioning independent of HXKs (Xiao et al. 2000). Work by Moore

et al. (2003) clearly demonstrated that HXK 1-deficient Arabidopsis plants have altered responses

to light, glucose, nitrates, and the hormones auxin and cytokinin. In addition, mutations that

abolished kinase function but not glucose binding were still able to convey sugar signals,

conclusively establishing the multi-functional role of hexokinases as enzymes and sugar signal

transducers. Hexokinase2 from yeast can compliment HXK]-deficient plants in regard to









enzymatic functions but does not restore sugar signaling in transgenic Arabidopsis (Moore et al.

2003). Also, AtHXK1, but not the yeast analog, associates with nuclear fractions of Arabidopsis

cells. While only a small fraction of total HXK is found in the nucleus, there the protein interacts

with two unconventional partners VHA-B1 vacuolarr H -ATPase Bl) and RPT5B (19S

regulatory particle of proteasome subunit) to directly bind DNA and regulate transcription of

glucose-responsive genes (Cho et al. 2006). The key ethylene transcription factor EIN3 is

specifically degraded in the presence of glucose, but this EIN3 repression is abolished in gin2

(hxkl) mutants (Yanagisawa et al. 2003).

Description of Mutant Lines

The mnl Mutation

The primary subject of this study is the miniature] seed mutation, first described by Lowe

and Nelson (1946). They described two groups of maize kernel mutations: the first group

produced relatively normal vegetative tissues in regard to plant height, color, or vigor, yet had

negative traits with respect to kernel development and nutrient quality. The second group was

lethal, semi-lethal or otherwise defective during sporophytic growth. The miniature] locus is a

member of the first group, with plants showing generally normal (albeit delayed) seedling

development but producing kernels with >70% reduction in seed weight. Given the importance

of maize kernels in the global food supply, seed-specific mutations involving kernel

development, sugar metabolism and/or starch production are valuable tools for understanding

and subsequently modifying biochemical processes for human benefit.

The Mn] locus encodes the INCW2 protein, an invertase enzyme anchored to the cell wall

through ionic bonds (Cheng et al. 1996). Other forms of invertases are found in the cytoplasm

and vacuoles, and all three types are represented in diverse plant species, catalyzing the

unidirectional cleavage of sucrose into constituent glucose and fructose molecules. The MnI









gene is maize endosperm-specific, with the enzyme localizing to the basal endosperm transfer

cell (BETC) region that acts as a gateway for photosynthate entering the developing seed. This

cell-wall invertase is thought to play a role in source-sink partitioning, acting in the apoplast to

maintain a gradient for phloem unloading via sucrose cleavage, leading to hexose uptake across

the BETC layer. The mnn gene is a naturally occurring non-lethal mutant that limits enzyme

activity to less than 1% of wild-type (Miller and Chourey 1992; Chourey et al. 2006).

The earliest detectable mnn seed phenotype occurs at -9-10 DAP, assessed via histological

techniques showing gap formation between the pedicel and BETC layer. The phenotype is

readily visualized with the naked eye as miniature seed on the cob, relative to wild-type, at the

10-12 DAP stage (Lowe and Nelson 1946; Chourey et al. 1992). In addition, in situ localization

of both MnJ RNA and INCW2 protein demonstrate the absence of these molecules (Cheng et al.

1996; Li et al. 2008). These observations coincide with a significant increase in sucrose content

of mnn seeds between 8 and 20 DAP, along with a marked decrease in glucose and fructose

quantity over the same period (Li et al. 2008).

Vilhar et al. (2002) further associate the mnn phenotype with reduced mitotic activity and

cell expansion, with no significant alteration in endoreduplication. They speculate that a high

sucrose to hexose ratio during early mnn seed growth is a possible cause of cell differentiation

and maturation at the expense of normal cell division and expansion. Recently LeClere et al.

(2008) demonstrated that the auxins, indole-3-acetic acid (IAA) and IAA conjugates, are

significantly reduced in mnn seed. Their results suggest that INCW2 directly or indirectly

influences auxin biosynthesis in developing maize kernels, and this auxin deficiency could be a

factor influencing reduced cell size and division in mnn kernels.









The sul Mutation

The sugary] locus was first identified over a century ago as a maize kernel mutation

leading to a glassy, translucent mature seed (Correns 1901). Later analysis revealed an

accumulation of simple sugars as well as phytoglycogen in the endosperm of homozygous sul

kernels, corresponding to a large decrease of the predominant starch amylopectin (Creech 1965;

Evensen and Boyer 1986). Beginning at 16 DAP, Creech (1965) showed that sul kernels

maintained double the percentage of sucrose compared to wild-type. Reducing sugars were

similar at 16 DAP but were 4-fold more abundant at 24 and 28 DAP in sul seeds. Total sugar

content in sul seeds was double that of Sul kernels throughout development as a percent of dry

matter, and water-soluble polysaccharides (WSPs) were increased 10-fold by 24 and 28 DAP.

However, total carbohydrate was only 5-8% reduced in the sul genotype.

Cloning and characterization of the transcript and protein of sugary] led to its

identification as a starch debranching enzyme (DBE) targeting the a-1,6 linkages of amylopectin

and glycogen, part of the a-amylase superfamily of starch hydrolytic enzymes (James et al.

1995). The original allele sul-Ref contains two point mutations that lead to normal transcript

levels but no protein accumulation (Dinges et al. 2001). Wild-type and sul transcripts are

detectable as early as 8 DAP, with a slightly higher accumulation in the mutant. Enzyme levels

peak toward 20 DAP in Sul lines but are undetectable for the duration of sul kernel development

(Rahman et al. 1998).









CHAPTER 2
MATERIALS AND METHODS

Field Work and Fresh Material

Planting

Maize kernels from six homozygous inbred lines were planted at the University of

Florida/Institute of Food and Agricultural Sciences (UF/IFAS) Plant Science Research and

Education Unit (PSREU) in Citra, FL for two consecutive years. During April 2007 three

different single-mutation lines, mn], sh2 and sul were planted. In addition to the mutant

genotypes, wild-type controls were planted in equal proportion for each of the three related

inbred backgrounds from which the mutants were derived. Developing ears were shoot-bagged

prior to silk emergence to prevent non-specific pollination. As tassels matured and anthers began

to shed pollen, tassel bags were attached to the male donor plants one day before pollen

collection. Each of the six lines was self-pollinated or sibling-pollinated as material allowed.

Each bag was marked with the male and female genotype used in the cross as well as the date of

pollination. Once a plant was successfully pollinated, additional ears were stripped to promote

growth of the target ear. After assessing the total number of ears pollinated, a harvest schedule

was drawn up that would attempt to maximize sample coverage from 8 DAP through 32 DAP.

During 2008 lines were self- and sib-pollinated to maintain homozygous inbred genotypes

for analysis. Pollinations were conducted between 8:30am and 10:30 am. Harvesting was done as

close to 10:30 am as possible for every individual sample.

Harvest

For both years, harvesting consisted of breaking off whole ears with husk intact, sealing

them in their identification bags, and depositing material in a lab refrigerator maintaining 40C.

Kernels were harvested whole onto moist paper towels for gas chromatography (GC) analysis









and then the remainder replaced in the refrigerator until they could be flash-frozen whole directly

into liquid nitrogen. Any damaged kernels were discarded. Once frozen, kernels were deposited

into labeled 50mL or 15mL Fisherbrand disposable conical-bottom centrifuge tubes and stored at

-800C for further analysis. Each ear was stored in a separate tube.

Gas Chromatography Analysis

The GC analysis performed in this study followed the protocol outlined by Young et al.

(1997). Kernels were harvested from fresh ears and allowed to rest between moist paper towels

for one hour to alleviate possible wound responses. Approximately 15 kernels were sealed in a

20mL I-CHEM borosilicate vials with 0.125 inch thick septa caps (Fischer #05-719-111), with

up to three replicates per ear. Samples were allowed 3-4 hours to evolve ethylene gas. During

this time a Tracor 540 GC unit feeding into a Hewlett-Packard 3396 Series III integrator was

calibrated using an ethylene standard of known composition. Using a 1 mL gas-tight syringe, 0.5

mL of gas was removed from the vial head space and injected into the GC to produce a reading

of parts-per-million ethylene. Data was entered into Microsoft Excel for subsequent calculations

including standard error.

Nucleic Acid Preparation

RNA Isolation

Total RNA was isolated from frozen tissue using an acid-phenol lithium chloride technique

adapted from Maniatis et al. (1982). A mortar and pestle were treated with chloroform, allowed

to dry, and then cooled with liquid nitrogen. Two or three kernels were weighed then ground

under liquid N2 by hand, homogenized with 3 mL of isolation buffer (Table 2) and allowed to

thaw. The mixture was pipetted into 14 mL round-bottom disposable polypropylene Falcon tubes

(Fischer #14-959-11B), 1 mL of phenol: chloroform:isoamyl alcohol (IA) added, then mixed on a

rotary shaker for 5 minutes. Samples were centrifuged at 5000 rpm for 5 minutes after which the









supernatant was mixed with 2 mL of the phenol: chloroform:IA mixture. The samples were

centrifuged for 5 minutes at 5000 rpm, then the supernatant was transferred by pipette into a

clean 14 mL Falcon tube along with 3 mL of chloroform:IA. The solution was mixed thoroughly

and then centrifuged for 10 minutes at 5000 rpm. The aqueous phase (-2.4 mL) was then

transferred to a fresh Falcon tube, mixed with an equal amount of 6M lithium chloride, and

incubated at -200C for at least one hour to precipitate RNA. The samples were then centrifuged

for 30 minutes at 5000 rpm and the supernatant was discarded. The pellet was resuspended with

2 mL of 2% potassium acetate and the samples were incubated for 5 minutes at 500C, after which

2 mL of phenol: chloroform:IA was added. The samples were mixed thoroughly and centrifuged

5 minutes at 5000 rpm. Another 2 mL of chloroform:IA was added, mixed, and centrifuged 5

additional minutes at 5000 rpm. The supernatant was carefully transferred to fresh 30 mL Corex

glass tubes (DuPont Instruments #00156) containing 3 mL absolute ethanol and incubated at -

200C overnight (at least 8 hours). The samples were centrifuged at 10,000 rpm for 25 minutes,

the supernatant discarded, and samples air dried. Finally the pellets were resuspended in a known

volume (100-175 uL) of diethylpyrocarbonate (DEPC)-treated water, incubated at 600C for 5

minutes, transferred to clean Eppendorf 1.5mL tubes, and quantified using a NanoDrop ND-1000

spectrophotometer (Thermo Fischer Scientific, Wilmington, DE) to calculate RNA recovered per

kernel. The isolated RNA was resolved on an 0.8% agarose gel to check for quality.

DNase treatment

After initial concentration was recorded, 50 uL aliquots were taken from each sample for

routine DNase treatment using the Ambion DNAfree kit (Ambion # AM 1906). Five microliters

of 10X DNase I buffer and one microliter rDNase I enzyme were added to each sample, mixed,

and incubated at 370 for 20-30 minutes. 5.5 uL DNase Inactivation Reagent was then added and

mixed by shaking and inversion several times for two minutes at room temperature. Samples









were spun down in a tabletop centrifuge for 1.5 minutes at 10,000 x g and the supernatant

removed to new tubes for downstream applications. RNA quality was again verified in the same

way as described in the RNA isolation protocol.

Reverse transcription

Total mRNA was converted into cDNA using the SuperScriptIII First-Strand Synthesis

System (Invitrogen #18080-051). 3.5 ug DNAfree RNA was brought to a total volume of 8 uL in

200 uL PCR tubes. A master mix of dNTPs (10 mM) and Oligo(dT)2o (50 uM) was prepared

with a 1:1 ratio, and then 2 uL of the mixture was added to each 8 uL (RNA + H20) sample.

Samples were incubated at 650C for 5 minutes then placed on ice for at least one minute.

Meanwhile a cDNA synthesis master mix was prepared consisting of, per sample: 2 uL 10X RT

buffer (200 mM Tris-HCI at pH 8.4, 500 mM KC1), 4 uL 25 mM MgCl2, 2 uL 0.1 M DTT, 1 uL

RNase OUTTM (40 U/uL), and 1 uL SuperScript TM III reverse-transcriptase (200 U/uL). A total

of 10 uL cDNA synthesis mix was added to each sample, mixed, briefly spun in a table-top

microcentrifuge, then incubated at 500C for 50 minutes. To stop the reaction, samples were

incubated at 850C for 5 minutes then chilled on ice. Original RNA remaining in the reaction was

digested by adding 1 uL (2 U/uL) Escherichia coli RNase H and incubating at 370C for 20

minutes. Two RT reactions were performed for each sample and used for downstream qPCR

analysis.

Gene-specific Analysis

Primer Design

Primers for genes of interest were acquired by aligning gene families using Vector NTI

software (Invitrogen #12605099) and searching for suitable regions of conservation or

divergence, depending on desired use. For five genes related to ethylene biosynthesis, reverse

primers were selected from those designed by Gallie and Young (2004). Forward primers









targeting conserved regions of homologous genes were designed using PrimerQuest software

(Integrated DNA Technologies online). For ACS2, ACS6, and ACS7, new forward and reverse

primers were designed targeting sub-300 bp amplification fragments. Mn], Hexokinase (HXK)

and Sucrose-\yin1h/i\e2 (Sus2) primers were designed by Qin Bao Li in the lab. Primers for

Sugary] (Sul) were adapted from James et al. (1995). See Table 2 for a complete list of primers.

Cloning

Following RT-PCR amplification of target gene fragments, PCR product was resolved on a

0.8% agarose gel stained with ethidium bromide. The Promega Benchtop 1kb DNA ladder

(Fischer #PR-G7541) was used as a marker for fragment size. Once a primer pair was confirmed

to produce a single band of expected size, the fragment was either purified from the remaining

PCR reaction using the QIAGEN Minielute PCR Purification Kit (QIAGEN #28006) or the band

was excised directly from the agarose gel and recovered using the QIAquick agarose cleanup kit

(QIAGEN #28706). The purified fragment (generally 1 uL) was used in TOPO TA cloning

(Invitrogen #45-0030) The fragment was incubated at room temperature for 5 minutes in the

presence of 1 uL salt solution, 1 uL TOPO2.1 PCR cloning vector, and 3 uL purified water in

order to ligate the amplicon into the TOPO plasmid, forming a circular DNA structure. A vial of

TOP 10 OneShot chemically competent E. coli cells (Invitrogen #44-0301) was thawed on ice,

and 2 uL of the ligation reaction was added. After 5 minutes on ice, the cells were heat-shocked

at 42C for exactly 30 seconds to induce uptake of the recombinant DNA then placed back on ice

for 1 minute to recover. In a laminar-flow hood, 250 uL of SOC medium was added to the

TOP10 cells and incubated on a rotary shaker at 370C for 1 hour. A petri plate with solidified

Luria-Bertani medium containing kanamycin was pre-warmed at 370C, and then the transformed

TOP10 cells were spread evenly on the plate and incubated at 37 C overnight.









Single transformant E. coli colonies were selected and transferred to liquid LB medium

containing kanamycin and incubated at 370C in a rotary shaker for at least 8 hours. The QIAprep

Spin Miniprep Kit (QIAGEN #27106) was used to isolate plasmid DNA. The samples were

transferred to 2 mL microcentrifuge tubes and spun at 10,000 rpm for 1.5 minutes to pellet cells.

The supernatant was discarded, and 750 mL lysis buffer was added to each sample, vortexed at

maximum speed for 30 seconds, and then allowed to rest at room temperature for 3 minutes. The

solution was applied to spin columns with collection tubes and centrifuged at 14,000 rpm for 1

minute. The flow-through was discarded, and 750 mL wash buffer was applied. The samples

were spun for 1 minute, flow-through discarded, and then spun one final time to remove residual

wash buffer still in the spin column. The columns were transferred to fresh 1.5 mL collection

tubes, 50 uL of elution buffer added to the filters, then collected by centrifugation for 1 minute at

14,000 rpm.

To verify the presence of inserts, 10 uL of plasmid eluate from each sample was digested

using 1 uL EcoRl, 2 uL ReAct3 buffer and 7 uL water. The reaction was incubated at 370C for 1

hour then visualized on a 0.8% agarose gel as previously described.

DNA Sequencing Reaction

Plasmids containing correctly-sized inserts were sequenced using M13 primers included in

the TOPO TA cloning kit (see Table 2 for primer sequence). Inserts were sequenced using the

Applied Biosystems BigDye Terminator vl. 1 Cycle Sequencing Kit (Fischer #NC9008533) as

follows: 2uL TOPO forward primer, 2uL plasmid, 2uL BigDye and 4uL water were mixed and

briefly centrifuged. The sequencing reaction consisted of 3 minutes at 950C, followed by 25

cycles of 95C for 25 seconds, 50'C for 15 seconds, and 60'C for 4 minutes. The samples were

held at 10oC until ready for processing using the QIAGEN DyeEx 2.0 Spin Kit (QIAGEN

#63206). Once the sequencing reactions were dried, they were sent to the University of Florida









Interdisciplinary Center for Biotechnology Research (ICBR) for processing by Sanger

sequencing, after which FASTA and fluorescent waveform results were returned. Sequences

were verified using BLAST searches against the National Center for Biotechnology Information

(NCBI) nucleotide database.

Absolute Quantitative PCR

Fresh cDNA was diluted 10-fold and stored at -200C for analysis. Both an MJ Research

PTC-200 and an Applied Biosystems 7300 real-time PCR machine were used in this study.

Reactions were run according to the instructions supplied with the Finnzymes DYNAMO HS

410 SYBR-Green detection system (Fischer #50-995-143). A master-mix was prepared

consisting of lOuL DYNAMO reagent, 5.6 uL H2O, 1 uL 5uM forward + reverse primer and 0.4

uL 50X ROX passive reference dye, each multiplied by n+1 reactions. Each well contained 17

uL of master mix, to which was added 3 uL of template. A standard curve was generated for each

gene using amplicon-containing plasmids diluted on a 10-fold gradient from 107 through 102.

The template was either 10-fold-diluted RT reaction (50 ng RT reaction per well), a diluted

plasmid standard, or water blank. The reaction components were identical for both machines. See

Table 2 for primer annealing temperatures.

The program for thermal cycling was kept constant with the exception of annealing

temperature. First samples were warmed to 500C for 2 minutes, then denatured by 15 minutes at

950C. Amplification consisted of 40 cycles of denaturation (20 seconds at 950C), annealing (20

seconds at the primer-specific temperature), and extension (30 seconds at 720C). One last

extension step (10 minutes at 720C) was run at the end of each reaction. Finally a DNA

dissociation curve was generated by ramping the thermocycler temperature from 550C to 950C

with a plate read every one degree. Each sample, plasmid standard and water blank was run in

triplicate to minimize experimental error.









Absolute qPCR data was entered into Microsoft Excel and triplicate results were averaged

into single values for each sample. Biological replicates were averaged and graphed including

standard error calculations.









Table 2-1. Components of 50mL RNA isolation buffer.
IM Tris-HCI buffer pH 7 5 mL 100 mM
4M NaCL 2.5 mL 200 mM
DL-Dithiothreitol (DTT) 40 mg 5 mM
N-lauroyl sarcosine 500 mg 34 mM
0.5M Ethylenediamine tetraacetic acid 2 mL 20 mM
H20 treated with 1% Diethylpyrocarbonate (DEPC) 40.5 mL









Table 2-2. List of qPCR primers 5' to 3'


Gene
ZmInCW2
ZmSul
ZmSus2
ZmHXK2
ZmACS2
ZmACS6
ZmACS7
ZmERS]-14
ZmETR2-9
ZmETR2-40
ZmAC020
ZmAC035
ZmEIL-1-


Forward
GGTGACCGGGATACAAACGGCACA
ACCAGAGGATGCAGTCTATG
ACTTTCCACATACCGAGAAGGCCA
ATAGCAAGCAGAGGGAACTGGGTT
CCACAGCTCAAACAACTTCACCCT
TGCACTGCACGAGCGGCAA
CTCGAACAACTTCACCCTCACCA
ACTCGAGGATGGAAGCCTTGAACT
GCTATGTATGTGTGAAATTTGAGATTAGGA
GCTATGTATGTGTGAAATTTGAGATTAGGA
CGCCGACGCCGTCATCTT
CGCCGACGCCGTCATCTT
GCAGCAGCAGCAGTTCTTCATCC


Reverse
GACAAATCCTGCAAATGTCGGGCG
CCATTCCACTCTGACCAAACG
AAGGTTTACCAGCTCCCTCAGCTT
ATCAGCATATCTCCCACCAGCCAA
GTGCTCCGTGGCGAACCT
CGCTCCGTGGCGAACGC
CACCAGGTGGATGCCCTTGG
TCTCCCGTCGGGCAGCAC
CTCGTACAAATCTGAGGACGCTCCAG
TCAAGTCTGAAGACGCCGCGGAGGAG
TCCACGATACACGCATAACCACCGT
ACACACATAACTGTGCCACTATAAGCA
GTTTATGGCTGGCCGGACATACAAGT


Ta(F)
60
56
56
60
60
60
60
60
58
57
60
56
57









CHAPTER 3
RESULTS

The Mnl and mnl Genotypes

Ethylene Production

A key component of this study was the determination of ethylene generation in developing

seeds. GC analysis was performed for two consecutive years, 2007 (Fig. 3-1) and 2008 (Fig. 3-

2). Due to differences in kernel physiology, data was plotted on both a nanomols per gram fresh

weight (nmol/g/hr) and nanomols per whole kernel basis (nmol/kemel/hr).

Figure 3-1A shows that for the 2007 crop, on a per-gram fresh weight basis, mnn produced

more ethylene at all stages, with an early peak of -70 nmol/g/hr at 8 DAP and a maximum rate

of 135 nmol/g/hr at 29 DAP. The early peak fell to -37 nmol/g/hr at 14 DAP before a linear rise

to the 29 DAP maximum. The mnn ethylene peak at 29 DAP was followed by a steady 85

nmol/g/hr rate through 36 DAP. MnI ethylene production also started with a peak at 8 DAP of

over 45 nmol/g/hr, before dropping to half the initial value, remaining constant up to 30 DAP

before a sharp increase at 33 DAP that reached over 55 nmol/g/hr. At both 20 and 29 DAP the

mnn ethylene production was at least 3-fold higher, with an overall hormone peak in mnl four

days before that shown in MnI.

Figure 3-1B shows the same ethylene data calculated on a per kernel basis. This

calculation method highlights the smaller kernel size of the mnn genotype due to the miniature

mutation. The Mn] data still showed an initial drop between 8 DAP and 10 DAP, but the starting

value of -3.6 nmol/kernel/hr remained similar between 12 and 20 DAP, reaching -4

nmol/kernel/hr at this time. The ethylene levels at 33 DAP in MnI rose to over 17.5

nmol/kemel/hr, nearly double the amount in the mnn kernels. The mnn genotype was similar to

Mnl in that initial values of -4 nmol/kernel/hr were more stable before 20 DAP on a per kernel









basis. The 29 DAP peak in mn] was still present, but the absolute differences at 20 and 29 DAP

were only 50% higher in mn] at both time points. It is important to note that where error bars are

not shown, the data point was the result of a single biological sample and might not represent the

genotype as a whole. Factors that led to the loss of replicate data included technical difficulties

with the GC apparatus, poor seed set on pollinated ears, and human error.

During 2008 (Fig. 3-2) Mn] and mn] samples were more abundant, allowing improved

replicate coverage of each time point, especially between 8 and 20 DAP. Figure 3-2A shows that

on a per-gram basis the Mn] and mn] hormone values were similar between 8 and 13 DAP. The

Mn] ethylene production started at a rate of -90 nmol/g/hr at 8 DAP, followed by a linear

decline to -28 nmol/g/hr at 20 DAP. This -28 nmol/g/hr value was maintained for the remainder

of observed time points. The mn] line had an 8 DAP ethylene production rate of -70 nmol/g/hr,

and after a 20% decrease at 11 DAP, ethylene remained between 45 and 60 nmol/g/hr until 25

DAP, a level nearly twice that of Mnl. The mn] samples exhibited a 4-fold decrease in ethylene

production at 30 DAP versus 25 DAP. When plotted on a per kernel basis (Fig. 3-2B), the Mn]

genotype produced more ethylene than the mn] line at all points tested. This was highlighted by

a distinct peak of ethylene in Mn] at 13 DAP (-11 nmol/kemel/hr). Hormone levels dropped to

-6.5 nmol/kernel/hr by 16 DAP and remained steady for the remaining time points, with a final

measurement of -4.5 nmol/kernel/hr at 34 DAP. Overall, ethylene production per mn] kernel

was reduced by 2-fold between 8 and 13 DAP when compared to the wild-type, which remained

-4 nmol/kemel/hr. The mn] line showed a -6 nmol/kernel/hr peak at 16 DAP followed by a

gradual downward trend ending at -4.7 nmol/kernel/hr at 25 DAP. Ethylene levels in mn]

kernels were below 1 nmol/kemel/hr at 30 DAP, less than half of the wild-type (Fig. 3-2B).









Transcript Accumulation in Mnl and mnl Genotypes

In order to better understand the relationship between hormone production and kernel

physiology, gene expression analysis was performed in order to quantify RNA levels of the

genes related to sugar metabolism, ethylene biosynthesis and ethylene perception. Transcript

levels were investigated for the following genes: three metabolic genes Mn], Sus2 and HXK2,

three ACC synthase genes ACS2, ACS6 and ACS7, two ACC oxidase genes AC020 and AC035,

three ethylene receptors ERS1-14, ETR2-9 and ETR2-40, and finally one transcription factor

EIL1-1. Transcript levels were analyzed using two biological replicates from the 2007 field crop

and three replicates from the 2008 crop. All qPCR data from the 2007 harvest were reported as

the number of transcripts per nanogram total RNA (# transcripts/ng total RNA). The results from

the 2008 crop were reported as both # transcripts/ng total RNA and the number of transcripts per

kernel (# transcripts/kernel). In addition, AC020, AC035, ETR2-40 and EIL1-1 were cloned and

included in the real-time PCR analysis for year 2008, so year 2007 data for these genes is absent.

Metabolic genes

The causal basis of the mn] seed phenotype is the loss of INCW2 enzyme activity, which

is encoded by the Mn] gene (Cheng et al. 1996).Quantifying this gene provides an internal

control for RNA quality and reaction efficiency during sample preparation. Figure 3-3 shows the

difference in MnJ transcript levels in the Mn] and mn] lines. The Mn] transcript was abundant

between 8 and 13 DAP in Mn] kernels and reduced by -95% in the mn] background for both

2007 and 2008 samples, correlating with timing of maximum enzyme activity (Chourey et al.

2006). Comparing absolute quantification between year 2007 and year 2008 (Fig. 3-3A and 3-

3B, respectively), the 2008 data showed a 50% higher transcript abundance at 8 DAP, reaching

-27,000 transcripts/ng total RNA. At 12 DAP the 2008 samples produced -9,200 transcripts/ng

total RNA, 16% more than year 2007. The transcript levels at later stages were similar between









the years, with -3,000 transcripts/ng total RNA at 16 DAP and -1,600 transcripts/ng total RNA

at 20 DAP. In Figure 3-3C the trend was similar when calculated per kernel, with peak levels of

Mn] transcript occurring at 8 DAP at nearly 1.13 x10^9 transcripts/kernel. The only divergence

from a steadily downward Mn] expression level was at 13 DAP, when # transcripts/kernel

showed a slight rise before continuing to recede.

A sucrose synthase gene Sus2 was chosen as a second internal control (Fig. 3-4). The 2007

results showed very little difference between Mn] and mnn lines (Fig. 3-4 A), with levels

between 600 and 800 transcripts/ng total RNA and sizable standard error at all time points.

Figure 3-4B shows the 2008 Sus2 transcripts/ng total RNA consistently increased over year 2007

amounts, starting with -1,500 transcripts/ng total RNA at 8 DAP, afterwards increasing in both

MnI and mnn. In the Mn] line, Sus2 gene peaked at -3,100 transcripts/ng total RNA by 11 DAP

then decreased 20% through 16 DAP before a rise to -2,700 transcripts/ng total RNA at 20 DAP.

The mnn line showed a similar trend, with a later, reduced peak of -2,500 transcripts/ng total

RNA at 13 DAP, and a decline to -1,500 transcripts/ng total RNA for 16 and 20 DAP. On a per

kernel basis (Fig. 3-4C) the Sus2 transcript levels in Mn] samples peaked at 13 DAP and

remained unchanged thereafter, at -3x10"8 transcripts/kernel. The mnn samples produced a peak

Sus2 transcript level at 11 DAP, two days earlier than Mnl, with a noticeable linear decline that

ended at lxl108 transcripts/kernel at 20 DAP, a value 3-fold lower than Mnl.

Hexokinase2 transcript levels (Fig. 3-5) were investigated because members of this gene

family are reported to be essential components of sugar sensing and signaling pathways to this

study (Saltman 1953; Jang et al. 1997; Yanigisawa et al. 2003). The HXK2 transcript levels from

years 2007 and 2008 showed similarity in the number of transcripts/ng total RNA (Fig. 3-5A and

B).Additionally, HXK2 transcript levels were consistently higher in Mn] kernels than the mnl.









Figure 3-5A depicts a trend of decreasing expression from 8 to 20 DAP, with an initial transcript

level of -1,500 transcripts/ng total RNA shifting to a final value of -550 transcripts/ng total

RNA in Mn] and mn] samples. There was no statistically significant difference between the two

genotypes tested for that year, although average levels of HXK2 in mn] samples only reached

-70% of the levels in Mn]. In Figure 3-5B the Mn] value was increased 30% at 8 DAP

compared to the same Mn] time point for the previous year, but the remaining results for Mn]

were similar. However, the year 2008 mn] data at 8 DAP (-850 transcripts/ng total RNA)

showed a 3-fold decrease in transcripts/ng total RNA versus Mn] kernels. Figure 5-5B also

shows that HXK2 transcript levels in mn] from 13 to 20 DAP were between 50-60% of the

related amounts in Mnl. Figure 3-5C shows two peaks of HXK2 transcript accumulation in Mn]

kernels, the largest at 8 DAP (-1.5x10"8 transcripts/ng total RNA) and another at 13 DAP.

There is a 60% decrease overall from 8 to 20 DAP. The mn] data showed a constant value for

HXK2 of roughly 3x10^7 transcripts/kernel throughout development. Excluding 13 DAP, HXK2

transcripts/kernel were at least 3-fold lower in the mn] line relative to the wild-type.

The ACC synthase gene family

The ACC synthases catalyze the rate-limiting step in ethylene biosynthesis. Real-time PCR

data for the maize ACC synthase family, ACS2, ACS6 and ACS7 are shown in Figures 3-6

through 3-8, respectively. The expression level of the family as a whole was low, with ACS2 the

most abundant (up to 400 transcripts/ng total RNA), followed by ACS7, and ACS6 appearing as

low as 2 and 3 transcripts/ng total RNA (Fig. 3-8). Figure 3-6 reveals subtle variation in ACS2

transcript between Mn] and mn] lines, with consistently higher transcript levels in the Mn]

kernels at 8 DAP. The # of transcripts/ng total RNA in the year 2007 samples (Fig. 3-6A) was

approximately half the of those from year 2008 (Fig.3-6B) for both genotypes, but the trends

were similar. The Mn] kernels showed highest accumulation of ACS2 at 8 DAP (-400









transcripts/ng total RNA) then dropped over 50% by 12 DAP. The decline continued, with

lowest transcript levels at 20 DAP (-75 transcripts/ng total RNA) (Fig. 3-6B). The mn] line had

transcript levels of -110 transcripts/ng total RNA at 8 DAP, then showed a relative peak 12-13

DAP. After maximum ACS2 transcript accumulation of 100 and 200 transcripts/ng in mn] at 13

DAP (Fig. 3-6A and B respectively), ACS2 levels were reduced at 20 DAP. On a per-kernel basis

(Fig. 3-6C) the trend for the ACS2 RNA profile in the Mn] samples was more variable, but still

showed an overall high-to-low progression between 8 and 20 DAP. Upon calculation of

transcript levels per kernel, the 13 DAP peak in mn] samples became more prominent (Fig. 3-6).

ACS6 transcripts/ng total RNA in the Mn] line (Fig. 3-7A and B) had maxima at 8 DAP

followed by progressively lower abundance throughout development, to a 20 DAP value of ~4

transcripts/ng total RNA. The mn] line, for the year 2007 crop (Fig. 3-7A), showed peak levels

of ACS6 transcript at 13 DAP, similar to that of ACS2. This trend was absent in the 2008 samples

(Fig. 3-7B), which instead produced two plateaus, one 8-11 DAP (8 transcripts/ng total RNA)

and the second 13-20 DAP (~3 transcripts/ng total RNA). Figure 3-7C shows ACS6 transcript on

a per-kernel basis, which reveals a distinct peak in both Mn] and mn] lines that was not seen in

terms of transcripts/ng total RNA. The Mn] line showed no change in transcript amounts at 8

and 11 DAP, then rose to 1.07x10^6 transcripts/kernel at 13 DAP. Following this increase,

transcript levels fell to ~5x105 transcripts/kernel and remained constant to 20 DAP. In the mn]

line, ACS6 increased nearly 2-fold between 8 and 11 DAP, reaching 6.64x10A5

transcripts/kernel. After this 11 DAP peak, transcript levels dropped over 60% by 13 DAP and

maintained a consistent level of accumulation through 20 DAP.

The third ACC synthase ACS7 (Fig. 3-8), showed intermediate levels of transcription

compared with the other two family members. Both transcript/ng and transcript/kernel showed









little difference between Mn] and mn] genotypes with respect to ACS7. Figures 3-8A and B

demonstrate maximum ACS7 transcript at 8 DAP for Mn] (-60 transcripts/ng total RNA) and

mn] (-40 transcripts/ng total RNA). Following 8 DAP, transcript levels declined in a linear

fashion to final values of -8-10 transcripts/ng total RNA for both genotypes. Results from year

2007 and year 2008 samples are virtually identical, with similar absolute levels as well as trends.

On a per kernel basis (Fig. 3-8C), Mn] samples generated a slight peak of ACS7 transcript at 13

DAP. The mn] samples produced maximum ACS7 transcript levels at 11 DAP, but the difference

was not statistically significant. This transcript showed a trend of general decline from 8 to 20

DAP, and both genotypes ended at a final value of -8x10^5 transcripts/kernel.

The ACC oxidase gene family

Because the ACC synthases catalyze the first step in ethylene biosynthesis, it is important

to asses the state of ACC oxidase transcripts in order to represent the final step in ethylene

production. Figures 3-9 and 3-10 show results for AC020 and AC035, the two ACC oxidase

genes investigated in this study. All ACO data are from the year 2008 crop. In Figure 3-9A,

AC020 transcript levels in the Mn] genotype remained flat throughout all stages, with a slight

decrease between 11 and 13 DAP from -4,300 to -3,400 transcripts/ng total RNA. In the same

figure mn] developed a peak of AC020 transcript at 13 DAP of -5,100 transcripts/ng total RNA,

an increase of over 60% versus Mnl. The highest expression in mn] was -6,000 transcripts/ng

total RNA at 20 DAP, nearly 2-fold higher than Mnl. Figure 3-9B demonstrates that on a per

kernel basis there was a steady increase in AC020 transcript from 8 to 13 DAP in both lines.

From 13 to 20 DAP, AC020 levels remained unchanged in the wild-type samples at 4x10A8

transcripts/kernel. The mn] sample produced 60% of the AC020 transcript relative to Mn] at 16

DAP, but transcript levels were identical in the two lines at 20 DAP.









The AC035 transcript accumulation data (Fig. 3-10) showed similar expression in both

Mn] and mn]. Although Mn] produced maximum transcripts/ng total RNA at 8 DAP, and mn]

produced an AC035 peak at 13 DAP, both lines generate approximately 1,000 transcripts/ng

total RNA for the first three time points, with similar decreases at 16 and 20 DAP. When

calculated on a per kernel basis (Fig. 3-10B), AC035 transcript levels were effectively identical

between genotypes, with a clear parabolic peak centered on 13 DAP.

The ethylene receptor gene family and EIL1-1

Ethylene is perceived by membrane-bound receptors, three of which have been included in

this study: ERS1-14, ETR2-9 and ETR2-40. The fourth ethylene receptor in maize, ERS1-25, is

expressed at a lower level than the others (Young and Gallie 2004) and attempts at cloning the

gene were unsuccessful during this study. For year 2007 and 2008 samples, the three receptor

genes ERS1-14, ETR2-9 and ETR2-40 were expressed to similar levels, with ETR2-9 slightly

more abundant on a per ng total RNA basis (Figs. 3-11 through 3-13).

For ERS1-14 in year 2007 samples (Fig. 3-1 1A) both Mn] and mn] genotypes produced

similar transcript results, with peak accumulation at 8 DAP (-560 transcripts/ng total RNA).

After a drop to -300 transcripts/ng total RNA at 12 DAP, subsequent time points remained

consistent ending with -250 transcripts/ng total RNA at 20 DAP. Results from the year 2008

samples (Fig. 3-1 1B) showed an increase of -20% over year 2007 data. However, the trend for

Mn] kernels was similar, with an 8 DAP peak of 630 transcripts/ng total RNA followed by a

30% decline at 11 DAP, remaining constant thereafter. The mn] kernels from year 2008 diverged

from the previous year's results, with 8 DAP ERS1-14 levels that started at -530 transcript/ng,

then increased 10% to peak at 11 DAP, generating 50% more transcript than Mn] at that time

point (Fig. 3-11B). After the 11 DAP peak, ERS1-14 levels in mn] returned to wild-type levels at

16 and 20 DAP. The peak in ERS1-14 transcripts at 11 DAP was visible on a per kernel basis as









well (Fig. 3-11C), marking the only point that mn] demonstrated higher values than Mn]. Both

genotypes showed identical plateaus from 13 to 20 DAP, though ERS1-14 transcripts/kernel

were insignificantly lower in mn]-1 than Mnl.

Figure 3-12 shows the transcript results for the most abundant receptor, ETR2-9. Year

2007 samples produced an identical profile of ETR2-9 generation between Mn] and mnl,

decreasing from -860 transcripts/ng total RNA at 8 DAP to just over 500 transcripts/ng total

RNA at 20 DAP (Fig. 3-12A). No statistically significant differences were observed. Figure 3-

12B shows data from the 2008 harvest on a per ng total RNA basis. A similar trend emerged as

in the previous year; ETR2-9 transcript declined as development progressed. In addition, the

accelerated decline of ETR2-9 transcript in mn] samples was significant at 13 DAP, when mn]

kernels produced only 65% of the ETR2-9 transcript versus the wild-type. This observation is

supported on a per kernel basis by the results shown in Figure 3-12C. Both genotypes produced a

similar 2-fold transcript increase from 8 to 11 DAP, after which the two lines diverged: The Mn]

samples generated -8x10^7 transcripts/kernel at 13 DAP, roughly 3-fold higher than mn]. The

wild-type kernels subsequently maintained at least 50% higher ETR2-9 transcripts/kernel at 16

and 20 DAP.

Figure 3-13A depicts year 2008 results for ETR2-40, showing a -646 transcript/ng peak in

the Mn] kernels at 8 DAP that was not reflected in mn] samples. Other than this difference, both

genotypes produced no significant variation in amounts of transcript, remaining between 300 and

350 transcripts/ng total RNA from 11 to 20 DAP. Per kernel results for ETR2-40 (Fig. 3-13B)

showed an increase in both genotypes from 8 to 11 DAP. At 13 DAP, the Mn] samples showed

increased transcript accumulation of -4x10^7 transcripts/kernel and maintained that level for the









remaining time points. At 13 DAP the mnn kernels decreased accumulation to ~2.23x10A7

transcripts/kernel through 20 DAP, 2-fold lower than MnI results.

The transcription factors EIN3 and EIL1 are downstream transducers of ethylene

perception. EIL1-1 was included in this study as a representative of this group of genes (Fig. 3-

14). For both MnI and mnn samples, EIL1-1 exhibited the trend of showing lowest transcript

levels at 8 DAP and steadily rising to a peak at the latest stage measured, 20 DAP. Figure 3-14A

shows Mnl remained constant from 8 to 16 DAP, maintaining 740-780 transcripts/ng total RNA.

Over the same period mnn samples produced a peak of EIL1-1 transcript accumulation at 13

DAP -30% higher than the MnI kernels. Both lines eventually rose to their maximum values of

1,200 transcripts/ng total RNA at 20 DAP. On a per kernel basis (Fig. 3-14B) the two lines

produced similar increases in EIL1-1 transcript during development. From 13 to 16 DAP, MnI

transcript production was unchanged (-9.2x10^7 transcripts/kernel) but the mnn line decreased

temporarily at 16 DAP to 4.89x10A7 transcripts/kernel. The subsequent increase at 20 DAP was

proportionally similar between genotypes, but the MnI value of 1.35x10A8 transcripts/kernel at

20 DAP was -30% higher than mnn at the same time point.

The Sul and sul Genotypes

Ethylene Production

The Sul and sul genotypes were included in this study in order to provide data for a late-

acting starch synthesis mutation, as compared to the mnn seed mutation that affects carbohydrate

metabolism during the early stages of seed development. The ethylene production results for the

Sul and sul genotypes were reported using the same convention as the MnI and mnn data.

Figure 3-15 shows ethylene production for the year 2007 samples. On a per gram fresh weight

basis (Fig. 3-15A) both genotypes produced -35 nmol/g/hr ethylene at 12 DAP, the highest

amount in either line for all time points tested. There was a -40% decline in ethylene at 16 DAP









in both lines. At 20 DAP sul kernels produced 27 nmol/g/hr ethylene, 50% more than Sul seeds.

Both lines reached their lowest production rate, ~11 nmol/g/hr, at 28 DAP. Figure 3-15B shows

results in terms of # nmol/kernel/hr, revealing a decrease in ethylene production for both

genotypes from 8-16 DAP. The Sul kernels produced a minor peak of -5.5 nmol/kernel/hr at 12

DAP and a major peak of 8 nmol/kernel/hr at 24 DAP. The ethylene level in the sul mutant at 12

DAP was similar to that in the wild-type, as was the subsequent decrease at 16 DAP and rise to

~7 nmol/kernel at 20 DAP. Due to technical difficulties the data for sul kernels at 24 DAP were

not recovered. At 28 DAP the results of 4.5 nmol/kernel/hr for Sul and sul kernels were similar,

showing a downward trend at this stage in both lines.

During 2008, the Sul line grew abundantly, providing excellent coverage of all time

points. The sul mutant line, however, showed poor germination, allowing only one or two

replicates per stage. Bars represent standard error of between two and four biological replicates

for Sul and two replicates for sul. Figure 3-15A shows that the highest level of ethylene

production was at the earliest time point sampled for both Sul (-45 nmol/g/hr, 8 DAP) and sul

(-40 nmol/g/hr, 12 DAP). Hormone levels dropped sharply in Sul from 20 nmol/g/hr at 12 DAP

to 10 nmol/hr at 20 DAP. There was a slight increase of ethylene 20-25 DAP before production

reached the observed minimum of -2 nmol/g/hr from 28-30 DAP. This lowest value in Sul

kernels was followed by a rise to -5 nmol/g/hr at 34 and 37 DAP. The sul line produced a

similar trend as Sul, decreasing 3-fold from 12 to 20 DAP. The sul minimum value was slightly

earlier than the wild-type but similar in rate (-2 nmol/g/hr). As in the Sul samples, ethylene

production showed an upward trend at 35 DAP. When considered on a per kernel basis (Fig. 3-

16B), the Sul kernels did not have a maximum at 8 DAP, but instead produced distinct peaks

above 4 nmol/kernel/hr at 16 and 24-25 DAP. The timing of the later peak corresponded well









with 2007 observations (Fig. 3-15B and 3-16B). Later development showed similar trends as the

per gram results, with a minimum at 28-30 DAP followed by a -2-fold increase at 34-37 DAP.

The sul mutant kernels generated more ethylene than wild-type at 12 and 14 DAP but fell more

sharply at 20 and 25 DAP, preceding both the initial peak and initial decline of the wild-type by

~5 days. The sul samples showed an increasing trend late in development that resembled Sul

kernels, and reached a 2 nmol/kernel/hr rate at 35 DAP.

Transcript Accumulation in Sul and sul Genotypes

Due to difficulties in generating complete time courses for Sul and sul genotypes, each of

the data points representing transcript levels were derived from single biological samples. Less

emphasis will be placed on slight differences in transcript levels because of the questionable

reproducibility of the results. Error bars represent experimental error for the real-time PCR

reactions.

Metabolic genes

The Sul transcript was used as an internal control for RNA quality and reaction efficiency

during qPCR sample preparation. This was due to the published similarity in Sul transcript

levels in Sul and sul genotypes (Dinges et al. 2001). Samples from each year demonstrated little

difference in transcript level between mutant and wild-type lines on a per ng total RNA basis

(Fig. 3-17A and B). The highest observed transcript levels were at the 12-14 DAP stage for both

years followed by a 3-fold decline as development progressed. In the year 2007 samples, 12

DAP results showed -8x10^3 transcripts/ng total RNA, and there was a second minor peak at 20

DAP (-5.5x10^3 transcripts/ng total RNA) before the downward trend resumed, which caused

the Sul and sul transcript levels to mirror trends in ethylene production (Fig. 3-16). For year

2008 samples, # transcripts on both a per ng total RNA and per kernel basis dropped steadily

from early to late development. A delay in transcript reduction in the Sul genotype caused 2-fold









higher accumulation at 20 DAP over levels in the sul mutant. Absolute amounts of Sul

transcript were shown to be three-fold lower in the 2008 crop, which produced from 2,500 to 500

transcripts/ng total RNA between 14 and 35 DAP.

The gene Sus2 encodes a member of the sucrose synthase family, which catalyzes the

unidirectional cleavage of sucrose into fructose and UDP-glucose. In Figure 3-18, Sus2 transcript

accumulation is shown to vary between year 2007 and 2008 samples. In the Sul kernels, for year

2007 (Fig. 3-18A), Sus2 transcript was lowest at 8 DAP then had a peak of -1,700 transcripts/ng

total RNA at 12 DAP. After this time point, transcript levels dropped 40% at 16 DAP then

reached their maximum level of 2,500 transcripts/ng total RNA at 20 DAP. Finally, Sus2 levels

declined to -1,400 transcripts/ng total RNA at 28 DAP. The sul kernels showed an upward trend

at 8 and 12 DAP, similar to wild-type, but generated a peak 4 days earlier at 16 DAP (-2,250

transcripts/ng total RNA). After this peak, Sus2 levels remained between 1,500 and 1,800

transcripts/ng total RNA for the rest of development. The Sul and sul samples from year 2008

showed similar trends when compared by # transcripts/ng total RNA as well as #

transcripts/kernel (Fig. 3-17 B and C). In Sul seeds the Sus2 transcript level was elevated 2-fold

over that of sul samples. The trend for Sus2 showed consistent levels of accumulation

throughout development. Sul kernels maintained between 2,000 and 2,500 transcripts/ng total

RNA, and sul produced over 1,000 transcripts/ng total RNA at all time points. Figure 3-17C

shows the same 2-fold increase in Sus2 transcript in the wild-type samples (-2x10^8

transcripts/kernel) over those in the sul mutant (-1x108 transcripts/kernel) at all stages. At 35

DAP, the Sul sample reached a peak of 2.67x108 Sus2 transcripts/kernel.

Another sugar-related transcript, HXK2, showed consistent patterns of accumulation

between years and genotypes. The highest transcript accumulation, 1,500-1,800 transcripts/ng









total RNA, was from 12-15 DAP, and decreased as development progressed (Fig. 3-19A and B).

For year 2007 samples, there was a 2.5-fold higher peak at 8 DAP in the sul sample that was not

present during the following year. Figure 3-19C shows that, on a per kernel basis, both genotypes

underwent steady decline in HXK2 transcript levels from an initial peak of -1.5x10A8

transcripts/kernel at 14 DAP. The Sul kernels showed a 2-fold higher HXK2 transcript level than

sul kernels at 20 DAP. By 25 DAP the two genotypes contained similar amounts of transcript

(-5x10^A7 transcripts/kernel) that was unchanged for the remainder of development.

The ACC synthase gene family

The ACC synthase enzymes catalyze the first committed step in ethylene biosynthesis

(Yang and Hoffman 1984). The ACC synthase gene family retained a similar distribution as that

seen in Mn] and mn] kernels, in that ACS transcripts were comprised mainly of ACS7 and ACS2,

followed by a 2-fold less ACS6 (Figs. 3-20 to 3-22). Figure 3-20A contains year 2007 data for

ACS2, showing peaks of 100 transcripts/ng total RNA at 12 and 20 DAP in the Sul line.

Between these peaks were intervening lows of -25 transcripts/ng total RNA at 16 and 28 DAP.

The sul samples were expressed at -35 transcripts/ng total RNA between 12 and 20 DAP. The

maximum expression levels were at the first and last stages tested; first over 450 transcripts/ng

total RNA at 8 DAP and last 3-fold lower at 28 DAP. The kernels from year 2008 (Fig. 3-20B

and C) showed reduced ACS2 transcript levels compared to year 2007, with expression under 50

transcripts/ng total RNA for 14 and 20 DAP. Both genotypes produced rising levels of transcript

as development progressed. Sul kernels produced 2-fold more ACS2 at 25 and 35 DAP. The

maximum transcript levels were at 35 DAP, of 150 transcripts/ng total RNA for Sul kernels and

72 transcripts/ng total RNA for sul.

ACS6 was the least-expressed member of this gene family (Fig. 3-21). All three plots of

transcript accumulation showed highest levels before 15 DAP for both Sul and sul lines. During









year 2007 (Fig. 3-21A) Sul seeds produced less than 4 transcripts/ng total RNA at 8 DAP, 6-fold

lower than sul. All Sul results from this year showed less than 10 transcripts/ng total RNA, with

a steadily decreasing trend after 12 DAP. The sul line produced a clear maximum ACS6

transcript accumulation at 8 DAP, then showed consistently lower amounts from 12-28 DAP (~6

transcripts/ng total RNA). The year 2008 samples showed highly similar distributions of

transcript in both Sul and sul kernels (Fig. 3-21B). The highest levels for both lines were again

at the earliest point assessed; 20-25 transcripts/ng total RNA at 14 DAP. ACS6 accumulation

dropped to ~8 transcripts/ng total RNA at 25 DAP, and continued at this level through 35 DAP.

When calculated on a per kernel basis (Fig. 3-21C), the Sul genotype maintained a high level of

ACS6 transcript at 14 and 20 DAP, ~ 2.5x10A6 transcripts/kernel. In contrast, sul kernels

showed a 2-fold decrease of ACS6 transcript between 14 and 20 DAP. For each time point from

25 to 35 DAP, both lines generated -7x10^5 transcripts/kernel.

The third ACC synthase gene, ACS7, was the highest-expressed member of the family in

Sul and sul genotypes (Fig. 3-22). For year 2007, results indicated constant expression, between

100 and 200 transcripts/ng total RNA, from 12 through 28 DAP (Fig. 3-22A).In sul samples

there was a 2-fold higher accumulation at 8 DAP, and in Sul samples the transcript level

dropped sharply at 28 DAP (24 transcripts/ng total RNA). Figure 3-22B shows that the 2008

samples had lower absolute levels of ACS7 transcript than year 2007, and maintained 50-125

transcripts/ng total RNA for all time points. These samples produced no peaks, which led to

constant ACS7 levels in both genotypes from 14 through 35 DAP. When calculated on a per

kernel basis (Fig. 3-22C), both Sul and sul genotypes had highest ACS7 accumulation at 14

DAP, greater than Ix1OA7 transcripts/kernel. Both lines' transcript levels fell to ~6x10A6

transcripts/kernel at 20 DAP. Sul samples maintained this amount until 30 DAP, then produced a









peak of IxlO7 transcripts/kernel at 35 DAP. The sul line had a slight downward trend from 20

to 35 DAP, ending at 5x10^6 transcripts/kernel.

The ACC oxidase gene family

ACC oxidase enzymes catalyze the final step in ethylene biosynthesis. Figure 3-23 shows

transcript levels ofAC020 in the year 2008 field samples. Both Sul and sul genotypes produced

-1,200 transcripts/ng total RNA at 14 DAP (Fig. 3-23A). There was a peak of nearly 2,000

transcripts/ng total RNA in sul samples at 20 DAP, a 50% higher transcript accumulation than

the Sul line. At 25 DAP both genotypes returned to parallel levels of 1,600 transcripts/ng total

RNA. After this time point there was a decrease in AC020 transcript levels in all samples that

ended with 1,000 and 1,200 transcripts/ng total RNA at 35 DAP for Sul and sul, respectively.

When data was calculated on a per kernel basis, both lines produced ~1.2x10A8 transcripts/kernel

at all stages, with the exception of 20 DAP peaks of 1.83x10A8 transcripts/kernel in Sul and

1.5x10A8 transcripts/kernel in sul (Fig. 3-23B).

For the AC035 gene (Fig. 3-24), transcript levels in Sul kernels increased from 500

transcripts/ng total RNA between 14 and 20 DAP to a peak of -1,750 transcripts/ng total RNA

by 35 DAP. The sul line also produced -500 transcripts/ng total RNA from 14 to 20 DAP, but

had a smaller increase over time, reaching -900 transcripts/ng total RNA at 30 and 35 DAP.

When calculated on a per kernel basis, the trend remained consistent, with increasing transcript

accumulation from 5x10A7 transcripts/kernel at 15 DAP to 7.5x10^7 transcripts/kernel at 30

DAP (Fig. 3-24B). From 30 to 35 DAP the Sul line produced a 2-fold higher accumulation of

AC035 transcript. However, in the sul line transcript levels remained constant from 30 to 35

DAP, 65% lower than wild-type.









The ethylene receptor gene family and EIL1-1

Ethylene receptors are the initial point of ethylene perception in plants. Three ethylene

receptor genes, ERS1-14, ETR2-9 and ETR2-40, were quantified in the Sul and sul

developmental series. Figure 3-25 shows ERS1-14 levels for kernels from both 2007 and 2008

field harvests. For year 2007 samples (Fig. 3-25A) accumulation was between 200 and 360

transcripts/ng total RNA for both genotypes throughout development, with the exception of sul

kernels, which contained over 800 transcripts/ng total RNA at 8 DAP. Data from year 2008

samples (Fig. 3-25B) showed similar amounts of ERS1-14 accumulation as the previous year.

Both Sul and sul series experienced a 50% decrease in transcript amount from 14 to 35 DAP.

The Sul line produced -450 transcripts/ng total RNA at 8 DAP; 40% more than sul kernels at

the same stage. This 40% higher ERS1-14 transcript accumulation occurred at all time points

except 30 DAP, at which point the results for the two genotypes were identical. Figure 3-25C

shows that the Sul series produced a peak of 4.5x10A7 transcripts/kernel at 20 DAP and a low of

1.5x10^7 transcripts/kernel at 30 DAP, before a final increase to 3x10^7 transcripts/kernel. In

sul kernels the ERS1-14 transcript levels were highest at 8 DAP, accumulating to 3x10A7

transcripts/kernel. By 20 DAP abundance had dropped to 1.6x10^7 transcripts/kernel, 3-fold

lower than wild-type, and remained constant through 35 DAP.

As in the Mn] and mn] samples, ETR2-9 was the most abundant receptor transcript, with

up to 10-fold higher levels than the other two family members (Fig. 3-26). The results from the

year 2007 harvest resembled the figure for ethylene production (Figs.3-15 and 3-26A), with two

peaks clearly defined in both genotypes; -800 transcripts/ng total RNA at 12 DAP and -650

transcripts/ng total RNA at 20 DAP. Both lines showed decreased accumulation of ETR2-9

transcript at 28 DAP (-325 transcripts/ng total RNA). For year 2008 samples (Fig. 3-26B),

transcript levels were 5-fold higher than those from the previous year. ETR2-9 decreased in









nearly linear fashion throughout development of both genotypes, from over 3,500 transcripts/ng

total RNA at 14 DAP to -2,000 transcripts/ng total RNA at 35 DAP. Figure 3-26C demonstrates

peaks of ETR2-9 in the Sul samples at 20 and 35 DAP, similar to ERS1-14 transcript (Fig. 3-

25C). Levels in Sul kernels were 2-fold higher than sul kernels at 20 DAP. Results for ETR2-9

transcript levels in sul kernels showed highest levels at 14 DAP (3.5x10O8 transcripts/kernel),

which decreased to 2x108 transcripts/kernel at 20 DAP and -1.3x108 transcripts/kernel at 35

DAP.

For ETR2-40 (Fig. 3-27), the data did not correlate to the other two receptor trends, with

transcript levels in Sul samples higher at 14 DAP but lower at 30 DAP as a portion of total RNA

(Fig. 3-27A). ETR2-40 transcripts in Sul kernels decreased in a linear fashion from -175

transcripts/ng total RNA at 14 DAP to -80 transcripts/ng total RNA at 30 DAP. There was an

upward trend at 35 DAP ( -140 transcripts/ng total RNA). The sul line displayed an increasing

level of ETR2-40 transcript from 100 transcripts/ng total RNA at 14 DAP to -145 transcripts/ng

total RNA at 30 and 35 DAP. Per kernel (Fig. 3-27C), ETR2-40 transcript levels displayed the

same peak in 20 DAP Sul kernels that was seen for ERS1-14 and ETR2-9 transcripts. This peak

reached -2x10^7 transcripts/kernel and was followed by a decrease in transcript accumulation at

30 DAP, as well as a secondary peak at 35 DAP that paralleled trends in the other receptor

transcript levels. The sul results for ETR2-40 showed an overall increase during development,

from below 1x10^7 transcripts/kernel during early stages to -1.2x10^7 transcripts/kernel by 35

DAP. Both ERS1-14 and ETR2-9 decreased during the same period (Figs. 3-25C and 3-26C).

Levels of EIL]-1 transcript from year 2008 samples are shown in Figure 3-28. For Sul

kernels, transcript levels were between 700 and 800 transcripts/ng total RNA at 14, 20, and 30

DAP (Fig. 3-28A). These points were separated by peaks of -1,250 transcripts/ng total RNA at









25 and 35 DAP. For sul samples, there was a decrease from -1,140 transcripts/ng total RNA at

14 DAP to -860 transcripts/ng total RNA at 25 DAP. By 30 DAP EIL1-1 transcript levels

peaked in sul at -1,200 transcripts/ng total RNA, and produced a similar amount at 35 DAP as

well. Upon calculation of transcript amounts per kernel, observations remained unchanged.

Figure 3-28B shows that EIL1-1 accumulation in Sul kernels increased after 14 DAP to a plateau

of -1x10^8 transcripts/kernel from 20-25 DAP. This amount was decreased 50% at 30 DAP,

then rose to 1.5x10A8 transcripts/kernel at 35 DAP. In sul samples the maximum transcript

accumulation was over 1x10A8 transcripts/kernel at 14 and 35 DAP, with a minimum value, -6.3

transcripts/kernel, at 25 DAP.











A 160
140
120
100

60--

S40
| 20 ---


5 10 15 20 25 30 35 40

B 20

16-

12

8


0I I \

5 10 15 20 25 30 35 40

Days-after-pollination

Figure 3-1. Ethylene produced by Mn] (blue) and mn] (pink) kernels, nmols of A) nmol/g/hour
B) nmol/kernel/hour (Summer 2007 field harvest)











A 120

100
S-80

60
u 40
0
1 20
0


5 10 15 20 25 30 35


5 10 15 20 25 30 35

Days-after-pollination

Figure 3-2. Ethylene produced by Mn] (blue) and mn] (pink) kernels, nmols of A) nmol/g/hour
B) nmol/kernel/hour (Summer 2008 field harvest)











2. O0E+04


1. 5E+04


1. OE+04


5. OE+03


0.OE+00


4. 0E+04


3. OE+04


2. OE+04


1. OE+04


0.OE+00


1. 6E+09


1.2E+09


8.0E+08


4. O0E+08


0.OE+00


20 2


6 8


20 2


N.


K.

>1F,~
I Y-----


6 8 10 12 14 16 18 20 22
Days-after-pollination


Figure 3-3. Mnl transcript levels in Mnl (blue) and mnl (pink) kernels A) transcripts per
nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008
samples) C) transcripts per kernel (Summer 2008 samples)


2













2


6 8 10 12


p *.
I I I I Y-~v- !











1.2E-+03


9.OE-+02

6.0E+02


3.OEH-02


0.OE-+00 I I I I
6 8 10 12 14 16 18 20 22
B 4.0E+03 ,


3. 0E+03

2. 0E+03


1. OE+03

0.OE+00


5. 0E+08

4. 0E+08

3. 0E+08

2.0E+08

1. 0E+08

0.OE+00


6 8 10 12


14 16 18 20 22


6 8 10 12 14 16 18 20 2


Days-after-pollination


Figure 3-4. Sus2 transcript levels inMnl (blue) and mnl (pink) kernels A) transcripts per
nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008
samples) C) transcripts per kernel (Summer 2008 samples)










3. 0E+03


2. 0E+03


1. OE+03


0.OE+00


B 3.OE+03


2. OE+03


1. OE+03


O.OE+00


C 3.OE+08


2. OE+08


1. OE+08


0.OE+00


6 8 10 12 14 16 18 20 2



.6 8 10 12 14 16 18 20 2









6 8 10 12 14 16 18 20 2


6 8 10 12 14 16 18 20


Days-after-pollination


Figure 3-5. HXK2 transcript levels in Mn] (blue) and mn] (pink) kernels A) transcripts per
nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008
samples) C) transcripts per kernel (Summer 2008 samples)










3. 0E+02


2.0E+02


1. OE+02


0.OE+00


B 8.OE+02


6.OE+02

4. OE+02

2.OE+02

0.OE+00


6 8 10 12 14 16 18 20 22













6 8 10 12 14 16 18 20 22


L 2.5E+07

2.0E+07 -

1.5E+07 '-

c 1.* O.07

E 5.OE+06

0.OE+00 -
6 8 10 12 14 16 18 20 22
Days-after-pollination


Figure 3-6. ACS2 transcript levels in Mn] (blue) and mn] (pink) kernels A) transcripts per
nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008
samples) C) transcripts per kernel (Summer 2008 samples)











1.5E+01


1.OE+01


5.OE+00


0.OE+00


B 4.0E+01


S3.0E+01

2.0E+01


1. OE+01
E-*


0. OE+00


C 1.6E+06


1.2E+06


8 .OE+05


4.OE+05


0.OE+00


6 8 10 12 14 16 18 20 22













6 8 10 12 14 16 18 20 22


6 8 10 12 14 16 18 20 22


Days-after-pollination


Figure 3-7. ACS6 transcript levels in Mn] (blue) and mn] (pink) kernels A) transcripts per
nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008
samples) C) transcripts per kernel (Summer 2008 samples)










1.0E+02

8. 0E+01

6. 0E+01

4. 0E+01

2. 0E+01

0.OE+00


B 1.OE+02


8. OE+01

6. OE+01

4. OE+01

2. OE+01

0.OE+00


4.OE+06

3. OE+06

2. OE+06

1. OE+06

0. OE+00


6 8 10 12 14 16 18 20 22


6 8 10 12 14 16 18 20 22












6 8 10 12 14 16 18 20 22


Days-after-pollination


Figure 3-8. ACS7 transcript levels in Mn] (blue) and mnl (pink) kernels A) transcripts per
nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008
samples) C) transcripts per kernel (Summer 2008 samples)










8. 0E+03

6. 0E+03

4. 0E+03

2. O0E+03

0.0E+00


6 8 10 12 14 16 18 20 22


B 6.0E+08


4. O0E+08


2.0E+08


0.OE+00


6 8 10 12 14 16 18 20 22


Days-after-pollination


Figure 3-9. AC020 transcript levels in Mn] (blue) and mn] (pink) kernels (Summer 2008
samples) A) transcripts per nanogram RNA B) transcripts per kernel










2. 0E+03

1. 5E+03

1. OE+03

5. 0E+02

0.0E+00


6 8 10 12 14 16 18 20 22


B 1.2E+08


8. OE+07


4. O0E+07


0.OE+00


6 8 10 12 14 16 18 20 22


Days-after-pollination


Figure 3-10. AC035 transcript levels in Mn] (blue) and mn] (pink) kernels (Summer 2008
samples) A) transcripts per nanogram RNA B) transcripts per kernel










8.0E+02

6.0E+02

4. 0E+02

2.0E+02

0.OE+00


8.0E+02

6.0E+02

4. 0E+02

2.0E+02

0.OE+00


8. OE+07

6. OE+07

4. OE+07

2. OE+07

0. OE+00


6 8 10 12 14 16 18 20 2


6 8 10 12 14 16 18 20 22


6 8 10 12 14 16 18 20 22
Days-after-pollination


Figure 3-11. ERS1-14 transcript levels in Mn] (blue) and mnl (pink) kernels A) transcripts per
nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008
samples) C) transcripts per kernel (Summer 2008 samples)










1.3E+03

1. 0E+03

7.5E+02

5. 0E+02

2.5E+02


0.OE+00


B 1.OE+03


7.5E+02

5. O0E+02

2.5E+02

0.OE+00


C 1.2E+08


8. OE+07


4. O0E+07


0.OE+00


6 8 10 12 14 16 18 20 22












6 8 10 12 14 16 18 20 22


6 8 10 12 14


16 18 20 22


Days-after-pollination


Figure 3-12. ETR2-9 transcript levels in Mn] (blue) and mnl (pink) kernels A) transcripts per
nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008
samples) C) transcripts per kernel (Summer 2008 samples)










1. OE+03

7.5E+02

5. 0E+02

2.5E+02

0. OE+00


6 8


B 6.0E+07


5. OE+07

4. O0E+07
3. OE+07
2. O0E+07

1. OE+07
0.OE+00


6 8 10 12 14 16 18 20 22


Days-after-pollination


Figure 3-13. ETR2-40 transcript levels in Mn] (blue) and mnl (pink) kernels (Summer 2008
samples) A) transcripts per nanogram RNA B) transcripts per kernel


SA~. ~~-j


20 22












1. 6E+03

1.2E+03

8. 0E+02

4. O0E+02

0. OE+00


6 8 10 12 14 16 18 20 22


B 2.0E+08


1. 6E+08

1.2E+08

8. 0E+07

4. O0E+07

0.OE+00


6 8 10 12 14 16 18 20 22


Days-after-pollination


Figure 3-14. EIL1-1 transcript levels in Mnl (blue) and mnl (pink) kernels (Summer 2008
samples) A) transcripts per nanogram RNA B) transcripts per kernel











A 100

80

V 60

40


0

5 10 15 20 25 30

B 10





o-

^ 2
0
*S 40 ___________________________________










5 10 15 20 25 30

Days-after-pollination

Figure 3-15. Ethylene produced by Sul (blue) and sul (pink) kernels, nmols of A) nmol/g/hour
B) nmol/kernel/hour (Summer 2007 field harvest)











A 100


5 10 15 20 25 30 35 4'













5 10 15 20 25 30 35 41


Days-after-pollination


Figure 3-16. Ethylene produced by Sul (blue) and sul (pink) kernels, nmols of A) nmol/g/hour
B) nmol/kernel/hour (Summer 2008 field harvest)


(











1.3E+04

1. OE+04

7.5E+03

5. O0E+03

2.5E+03

0.OE+00


3. 0E+03

2.5E+03

2. O0E+03

1. 5E+03

1. OE+03

5. O0E+02

0.OE+00


3. O0E+08

2.5E+08

2.0E+08

1. 5E+08

1. OE+08

5. OE+07

0.OE+00


5 10 15 20 25 30 35 4C












5 10 15 20 25 30 35 4C


5 15 25 35 45

Days-after-pollination

Figure 3-17. Sul transcript levels in Sul (blue) and sul (pink) kernels A) transcripts per
nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008
samples) C) transcripts per kernel (Summer 2008 samples)











3. O0E+03


2. O0E+03


1. OE+03


0.OE+00


3. O0E+03


2. O0E+03


1. OE+03


0.OE+00


4. O0E+08

3. OE+08

2.0E+08

1. OE+08

0.OE+00


/







5 10 15 20 25 30 35 4C


.-S-









5 10 15 20 25 30 35 4C





1.






5 10 15 20 25 30 35 4C


Days-after-pollination

Figure 3-18. Sus2 transcript levels in Sul (blue) and sul (pink) kernels A) transcripts per
nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008
samples) C) transcripts per kernel (Summer 2008 samples)










4. 0E+03

3. 0E+03

2. 0E+03

1. OE+03

0.OE+00

2.5E+03

2. 0E+03

1. 5E+03

1. OE+03

5. 0E+02

0. OE+00


2.0E+08

1. 5E+08

1. OE+08

5. 0E+07

0.OE+00


15 20 25 30 35


Days-after-pollination

Figure 3-19. HXK2 transcript levels in Sul (blue) and sul (pink) kernels A) transcripts per
nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008
samples) C) transcripts per kernel (Summer 2008 samples)


9

\






5 10 15 20 25 30 35 4C







r


5 10 15 20 25 30 35 40



:\-t


:\ N^


5 10

























D" 250

200

Jx 150

100

50

0


C 3.E+07

t 2.E+07

S2.E+07

1.E+07

5.E+06

O.E+00


5 10 15 20 25 30 35 4C












5 10 15 20 25 30 35 4C











5 10 15 20 25 30 35 40
5 10 15 20 25 30 35 4C


Days-after-pollination

Figure 3-20. ACS2 transcript levels in Sul (blue) and sul (pink) kernels A) transcripts per
nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008
samples) C) transcripts per kernel (Summer 2008 samples)



































0


C 3.E+06


3.E+06

2.E+06
2.E+06
1.E+06
5.E+05
n -c-i-nn


U



------------

I -


5 10 15 20 25 30 35 4C












5 10 15 20 25 30 35 4C





^41


5 10 15 20 25 30 35 40
Figure 3-21. ACS6 transcript levels in Sul (blue) and sul (pink) kernels A) transcripts per
nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008
samples) C) transcripts per kernel (Summer 2008 samples)


.1 1

























oD 400
350
300
bo 250
200
S 150
100
H 50
0


C 2.E+07

| 2.E+07


f 1.E+07
C)

E 5.E+06

O.E+00


5 10 15 20 25 30 35 40


5 10 15 20 25 30 35 40


5 10 15 20 25 30 35 40
Figure 3-22. ACS7 transcript levels in Sul (blue) and sul (pink) kernels A) transcripts per
nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008
samples) C) transcripts per kernel (Summer 2008 samples)











2.5E+03

2. OE+03

1. 5E+03

1. OE+03

5. O0E+02

0.OE+00


2.5E+08

2.0E+08

1. 5E+08

1. OE+08

5. 0E+07

0. OE+00


5 10


Days-after-pollination


Figure 3-23. ACO20 transcript levels in Sul (blue) and sul (pink) kernels (Summer 2008
samples) A) transcripts per nanogram RNA B) transcripts per kernel


5 10 15 20 25 30 35 4C




x

I!-=











2.5E+03

2. 0E+03

1. 5E+03

1. 0E+03

5. 0E+02

0.OE+00


2.5E+08

2.0E+08

1. 5E+08

1. OE+08

5. O0E+07

0. OE+00


5 10 15 20 25 30 35 40


5 10 15 20 25 30 35 40

Days-after-pollination

Figure 3-24. AC035 transcript levels in Sul (blue) and sul (pink) kernels (Summer 2008
samples) A) transcripts per nanogram RNA B) transcripts per kernel.











1000

800

600

400

200


oD 800
700
600
gtx 500
400
S300
200
100
0


C 5.E+07

- 4.E+07

, 3.E+07

2.E+07

1.E+07

O.E+00


5 10 15 20 25 30 35 4C








r -U


5 10 15 20 25 30 35 4C
5 10 15 20 25 30 35 4C













5 10 15 20 25 30 35 4C


Days-after-pollination


Figure 3-25. ERS1-14 transcript levels in Sul (blue) and sul (pink) kernels A) transcripts per
nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008
samples) C) transcripts per kernel (Summer 2008 samples)











A 1.OE+03

S8. OE+02 -

S 6.0E+02

0 4.0E+02--

E- 2.OE+02

0. OE+00
5 10 15 20 25 30 35 40
B 5.0E+03

S4.0E+03 -

? 3.OE+03

2.0E+03

E 1.OE+03

0. OE+00 -
5 10 15 20 25 30 35 40
C 5.0E+08

4.0E+08 ,

3.OE+08 7

2.0E+08

1. OE+08

0.OE+00
5 10 15 20 25 30 35 40

Days-after-pollination

Figure 3-26. ETR2-9 transcript levels in Sul (blue) and sul (pink) kernels A) transcripts per
nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008
samples) C) transcripts per kernel (Summer 2008 samples)






















2.5E+07

2. O0E+07

1. 5E+07

1. OE+07

5. OE+06

0. OE+00


5 10 15 20 25 30 35 40





:~ a.,

I I I I


5 10


15 20 25


Days-after-pollination

Figure 3-27. ETR2-40 transcript levels in Sul (blue) and sul (pink) kernels (Summer 2008
samples) A) transcripts per nanogram RNA B) transcripts per kernel











A 2.0E+03

S1.5E+03

S1.0OE+03
C) L
S5.OE+02

0.OE+00
5 10 15 20 25 30 35 40
B 2.0E+08

1.5E+08

S1.OE+08 .

5.OE+07

0.OE+00 -
5 10 15 20 25 30 35 40

Days-after-pollination

Figure 3-28. EIL1-1 transcript levels in Sul (blue) and sul (pink) kernels (Summer 2008
samples) A) transcripts per nanogram RNA B) transcripts per kernel









CHAPTER 4
DISCUSSION

The seeds produced by Zea mays are important sources of energy; for human consumption,

animal feed, sugar production, and recently as a source of ethanol. Given the importance of

cereal crops as a whole, research dealing with seed development is valuable not only for basic

science but also for many social and economic purposes. The large size of the Zea mays kernel

makes it a good model system for cereals.

Much work has already been done to unravel the processes involved in growth and starch

biosynthesis in maize seeds. Many metabolic pathways are established, from sucrose and hexose

import to starch and protein synthesis (Nelson and Pan 1995; Shewry and Halford 2002; James et

al. 2003). Also, the basic developmental program has been resolved, providing an outline of cell

division, expansion, endoreduplication, embryo development, starch and storage protein

synthesis, and finally programmed cell death of the endosperm (reviewed in Lopes and Larkins

1993; Young and Gallie 2000a). Now the most pressing questions are related to regulation and

timing of these processes. Specifically, sugar and hormone interactions have emerged as critical

control elements of cereal seed development. The purpose of this study is to observe the effects

of two established maize kernel metabolism mutations, mn] and sul, on ethylene hormone

production and related transcript levels. This should provide a basis for comparison relating

sugar metabolism and hormone effects during kernel development.

The investigation of interactions between sugars and phytohormones is complicated by the

fact that multiple pathways can mutually influence the processes under consideration. This

phenomenon, commonly referred to as 'crosstalk,' allows slight modifications in physiology to

have far-reaching consequences and effects. This also causes difficulty in determining the

difference between direct and indirect interactions. Furthermore, hormone effects can be









inconsistent depending on concentrations, tissue localization, and developmental stages. For

example, what holds true for germination or seedling development might be irrelevant or

reversed in root meristems (Moore et al. 2003; Arteca and Arteca 2008; Gallie et al. 2009).

This report demonstrates the variable nature of ethylene production between different

genetic backgrounds and developmentally altered kernel genotypes. A consistent theme is the

observance of periodic bursts of ethylene production that could be related to transition through

stages of kernel development.

Ethylene Accumulation in Developing Seeds

The observation of two peaks of ethylene production in developing maize kernels has been

well-documented over the last 12 years (Young et al 1997; Young and Gallie 2000a; Young and

Gallie 2000b). However, the specific timing, relative amplitude, and absolute levels of ethylene

biosynthesis have proven variable depending on the genetic background analyzed. Young et al.

(1997) suggest a partial explanation for this occurrence lies with differences in kernel mass

between genotypes. In this study ethylene values were reported as a function of kernel weight

and also kernel number, allowing for comparison of kernels both as biological units and as

masses of tissue. In keeping with published data, discussion will focus on nmoles/kemel/hr

measurements

Levels of ethylene accumulation were quantified for four genotypes during two

consecutive years of field planting and harvest, generating eight individual sample series. Each

of the eight series produced some form of peak of ethylene production before 20 DAP on a per

kernel basis. The sizes of these peaks were between 4 and 10 nmols/kernel/hr, which is in good

agreement with published values and timing (Young et al 1997; Young and Gallie 2000a; Young

and Gallie 2000b).









Mnl Seeds Produced a Distinct Peak of Ethylene between 12 and 14 DAP

Analysis of the results from Mn] kernels indicate that this genotype produced a peak of

ethylene between 12 and 14 DAP. This is in line with the earliest recorded bursts of ethylene

evolution in published reports (Young et al. 1997; Gallie and Young 2000a). Data for the Mn]

genotype for year 2007 ethylene production paralleled the rates from the 1145 lb genetic

background depicted in Young et al. (1997). Hormone results of the Mn] kernel analysis from

year 2008 more closely resembled the pattern of the Oh43 results published in that same report,

which generated nearly 3-fold more ethylene than the year 2007 samples before 20 DAP. This

current investigation shows that, while early ethylene biosynthesis was more subtle in year 2007

samples versus the following year, the later peak at 33 DAP was more well-defined (Fig. 3-1).

This is in contrast to the steady 5-7 nmol/kernel/hr of ethylene produced after 16 DAP in the year

2008 samples (Fig. 3-2B). It is possible that the 33 DAP sample from year 2007 was an anomaly

since the value was derived from a single biological sample. It is also possible that sampling for

year 2008 ended before a clear second developmental peak of ethylene biosynthesis could be

established, given that in most previous reports, ethylene levels remained low at 32 DAP before

a second peak was resolved (Young et al. 1997; Young and Gallie 2000b). Finally, differences in

the technical aspects of ethylene measurement could have varied between years.

Trends in mnl Kernel Ethylene Production Were Varied Over Two Consecutive Years

As in the Mn] results, the mn] kernels produced a major late peak of ethylene for year

2007 samples and a dominant early peak for year 2008 samples (Figs. 3-1 and 3-2). The mn]

genotype generated levels in the 4 nmol/kernel/hr range prior to 16 DAP during both years.

However, year 2007 samples produced an apparent peak at 29 DAP more than 2-fold higher than

levels at 12 DAP. As mentioned previously, the last three data points for mn] kernels from year

2007 were all single replicates, raising the possibility of aberrant results. Still, this pattern is in









agreement with previously published reports (Young et al. 1997), although the strongest

examples of late-stage ethylene production are in high-sugar mutant lines such as sh2, sul and

sulsel, not sugar-deficient genotypes such as mnn. For year 2008 samples, mnn seeds exhibited

a maximum rate of ethylene generation at 16 DAP, roughly four days after maximal levels of

INCW2 activity and ethylene peak in the wild-type kernels (Cheng et al. 1996; Fig. 3-2B).

Ethylene production rates remained constant until a sharp decline at 30 DAP, as reported

previously. The period from 16 to 25 DAP showed that, although ethylene production in mnl

kernels was lower than wild-type, the smaller kernels produced nearly 2-fold higher ethylene

levels on a per gram fresh weight basis. This rate was maintained throughout the period

associated with initiation and progression of starch loading in wild-type kernels. Increased

endogenous ethylene production as well as treatment of wild-type kernels with ethylene has been

shown to accelerate cell death in developing maize seeds (Young et al. 1997; Young and Gallie

2000b).

Neither Mn] nor mnl lines showed any periods of sharp decline in ethylene levels that

would clearly separate distinct peaks (Figs. 3-1 and 3-2). This relatively constant level of

production is not unique when considered against previously published data, but even maize

genotypes with high ethylene production rates exhibit substantial transient reductions in ethylene

levels between 20 and 32 DAP (Young et al. 1997; Young and Gallie 2000b). The lack of mnn

material in the field prevented sampling past 32 DAP; a stage that frequently involves increased

hormone levels (Young et al. 1997). It is possible that 30 DAP represents the effective endpoint

of mnn kernel development.

Hexose Deficiency and Increased Sucrose Lead to Pleiotropic Effects in Maize Seeds

While the significance of rising and falling ethylene levels as developmental cues is

unknown, and sensitivity of different tissues to ethylene is another important factor, a









consistently high level of ethylene has traditionally been thought of as a stress signal promoting

cell death in this tissue and stage of growth (reviewed in Young and Gallie 2000a). Auxin is

shown to promote ethylene biosynthesis in various systems (Arteca and Arteca 2008). The auxin

IAA, and IAA-conjugates, have been shown to increase sharply in maize kernels between 9 and

11 DAP, coinciding with reduced cytokinin (Lur and Setter 1993; LeClere et al. 2008). It is

possible that the substantial rise in auxin levels 9-11 DAP contributed to the timing of the early

ethylene peak in wild-type kernels. From this it follows that the auxin-deficient mn] line could

have delayed ethylene biosynthesis at this critical stage of kernel development (LeClere et al.

2008). In addition, Li et al. (2008) report that, beginning at 12 DAP and extending through 20

DAP, mn] kernels exhibit increased sucrose levels versus the wild-type. Combined with data

from Young et al. (1997) that demonstrated increased ethylene levels in conjunction with

increased sucrose levels in sh2 and sulsel mutant lines, it is possible that a relationship exists

between the hexose/sucrose ratio, auxin, and ethylene levels in developing maize kernels.

Previous studies of Arabidopsis suggest complex interactions between sugars and

phytohormones, supporting the possibility of this relationship in maize kernels (Zhou et al. 1998;

Cheng et al. 2002; Leon and Sheen 2003; Moore et al. 2003; Yanigisawa et al. 2003; Gibson

2004)

Sul and sul Kernels Displayed Inconsistent Ethylene Production over Two Consecutive
Years

According to Creech (1965) the sul mutation leads to 2-fold higher sucrose than wild-type

by 16 DAP. At this same stage, water-soluble polysaccharides (WSP) are increased nearly 4-

fold. Despite differences in sugar and starch content during early development, Sul and sul

ethylene production in this report was parallel until 20 DAP (Figs. 3-15 and 3-16) and has been

reported to be similar until 32 DAP (Young et al. 1997). The data presented in this report showed









a parallel relationship between Sul and sul at most stages of development (Figs. 3-15 and 3-16).

However, the timing and relative amplitude between first and second ethylene peaks were

dissimilar when comparing results from year 2007 and year 2008 GC analysis. This could be the

result of different environmental conditions between years, leading to variations in stress factors

such as drought and heat. It is possible that improved handling of samples during the year 2008

harvest increased the accuracy of the analysis. However, the range of ethylene production from

year 2007 is higher than that of year 2008 for the Sul and sul kernels (Figs. 3-15 and 3-16),

while the opposite is true of the Mn] and mn] samples (Figs. 3-1 and 3-2), making it difficult to

attribute changes between years to systemic effects.

Sul and sul kernels showed two clear peaks of ethylene production prior to 30 DAP

The differences between Sul and sul ethylene levels for year 2007 field samples were

statistically insignificant (Fig. 3-15). Technical difficulties with the GC apparatus prevented

collection of data for sul kernels at 28 DAP. The highest levels of hormone in Sul kernels were

at 12 and 24 DAP, with a small decline during the intervening 12 days (Fig. 3-15B). Cumulative

ethylene exposure is measured in part by ethylene receptor degradation in climacteric fruit

(Kevany et al. 2007). Whether or not this paradigm holds true in maize kernels is unknown. It is

important to note that the 12-DAP peak was in the 5 nmol/kernel range, and the later Sul peak

reached 8 nmol/nL. This is in good agreement with Mn] and mn] data concerning

physiologically relevant concentrations. However, the amount of ethylene required to trigger a

developmental responses is unknown. In addition, ethylene perception and signaling are

mediated by many factors downstream of ethylene production, and therefore inferring

physiological effects based solely on hormone data is impossible (Tatsuki and Mori 2001;

Yanigisawa et al. 2003; Yoo et al. 2008).









While the year 2008 data concerning Sul and sul were similar between genotypes prior to

20 DAP (Fig. 3-16), analysis revealed one major difference during development. Most

importantly, the samples of Sul kernels from the year 2008 revealed two clear peaks, one each at

16 and 22-24 DAP, before falling 4-fold at 28-30 DAP (Fig. 3-16B). The sul line had a first

peak of similar level two days earlier at 14 DAP (though this was a single replicate and was not

enough on which to base any conclusions). However, the second peak was completely absent in

the sul line from 20 to 25 DAP, with a 4-fold hormone reduction in the mutant kernels. This

increase after a deep trough at the 25-30 DAP stage was similar to previous reports (Young et al.

1997; Young and Gallie 2000b) although the most closely-related data on sul from Young et al.

(1997) does not show such a decline in this genotype. In that report sul ethylene levels rise to

-3-fold at 40 DAP over the levels in the related control genotype. Insufficient amounts of field

samples prevented analysis of this later stage.

It is important to note that the most disparate levels of ethylene between the Sul and sul

samples were at 24-25 DAP. These sul measurements were taken on July 18th and July 20th,

2008. Upon further investigation, each of the ten ears harvested on these dates generated

abnormally low ethylene values. These included the sul samples at 24-25 DAP (Fig. 3-16), the

Sul samples at 28 and 30 DAP (Fig. 3-16) and the mn] samples at 30 and 31 DAP (Fig. 3-2). It

is possible that all these ages of kernels coincidently produced low amounts of ethylene at their

respective stages of development, considering that published data support a transient cessation of

ethylene biosynthesis (Young et al. 1997; Young and Gallie 2000b). After consulting the hard-

copies of GC data from these two harvest dates, no abnormalities were apparent in either the

calibration or function of the GC apparatus. The average morning temperatures for these two

dates were 83.6F and 82.4oF, respectively. This was similar to the temperatures on the dates that









the Sul ethylene peak between 22 and 25 DAP were recorded. Despite this similarity in

temperature, it seems plausible that some environmental variation caused the low ethylene levels

on July 18th and 20th. Because data was recorded over two different days for ten different ears,

with multiple replicates per ear, experimental error is less likely.

A second peak of ethylene hormone in Sul samples from year 2008 showed unique timing

The occurrence of a second ethylene peak in year 2008 in Sul samples, just eight days

after the initial peak at 16 DAP (Fig. 3-16B), was unexpected because none of the published

literature reports such a quickly repeated burst of ethylene (Young et al. 1997; Gallie and Young

2000b). Each of the three data points from 22 to 25 DAP was the average of two biological

replicates (Fig. 3-16B), recorded across three different dates, lending credibility to the

observance of this peak. Data from Young et al. (1997) show a consistent dip in ethylene

production at 24 DAP in the wild-type and three mutant lines; sul, sulsel, and sh2. Despite this

similarity, none of those lines show such a rapid shift in hormone production. This could be

partly due to the practice in all previously published work of sampling kernels at 4 DAP

intervals. This methodology limits resolution of rising and falling hormone production to 8 DAP

periods, possibly masking rapid bursts of ethylene generation. Still, most published ethylene data

in maize kernels show sweeping curves without indication of rapid shifts. Data from various

genetic backgrounds and genotypes highlight the complex nature of ethylene hormone

production and measurement, and this report adds to the body of evidence supporting the

sensitive nature of ethylene activity in maize kernels.

Transcript Accumulation and Correlation with Ethylene Hormone Evolution

Hormone synthesis, perception and action are regulated at many levels, with complex

signals being integrated and modulated continuously. Transcript accumulation is one way to

connect the involvement of a gene with certain tissues and processes. Despite the frequent









occurrence of post-transcriptional and post-translational regulation, the quantity of mRNA for a

particular gene serves as a foundation for understanding the processes that affect phenotypes of

biological systems.

In this study, transcript levels were analyzed for four metabolic genes and four families of

genes related to ethylene biosynthesis, perception and signal transduction. Samples were

collected from two consecutive years of field harvest, and transcript levels were recorded as

absolute transcript levels per nanogram total RNA. The second year, additional data was

recorded to allow for calculation of transcripts on a per kernel basis in order to assess the state of

the kernels as biological units.

Transcript Accumulation in Mnl and mnl Genotypes

Metabolic genes

Given the centrality of the Mn] gene for explanation of the mn] phenotype, it is important

to understand how Mn] transcript levels relate to published data on INCW2 enzyme activity. In

addition, the Mn] transcript was used as a control in order to compare qPCR data between

biological replicates. At 8 DAP, the Mn] transcript in the Mn] genotype exhibited maximum

expression as a proportion of total RNA, which is consistent with maximum enzyme activity and

the role of INCW2 in providing hexoses to the rapidly dividing cells between 8 and 12 DAP

(Fig. 3-3A and B; Cheng et al. 1996). As cells proliferate and grow, the amount of Mn]

transcript is down-regulated as other developmentally important genes are transcribed. On a per

kernel basis, there was a temporary plateau at 13 DAP, which coincides with a period of

transition from cell division to cell expansion and elongation (Kowles and Phillips 1985;

Sreenivasulu et al. 2004). In the mn] genotype, the presence of Mn] transcript was less than 10%

of wild-type at all stages (Fig. 3-3), though this level of transcript accumulation was well above

published enzyme activity and protein levels (Cheng et al. 1996). This suggests that, while the









mn] transcript levels were less than 10% of wild-type (Fig. 3-3), there exists one or more

regulatory processes that reduced enzyme activity beyond the reduction in transcript level.

Another set of important enzymes in kernel metabolism are the sucrose synthases, which

are associated with production of substrates for cell expansion as well as cell wall and starch

biosynthesis (Chourey et al. 1998). Originally, Sus2 was included in this study in order to

provide an internal control for RNA quality and sample preparation efficiency due to supposed

consistent expression profiles in both Mn] and mn] genotypes. The margin of error in year 2007

samples prevented any meaningful comparison of Mn] and mn] lines. Year 2008 samples

showed clear differences in the timing and abundance of transcript accumulation. Sus2

transcripts in the Mn] genotype increased as a portion of total RNA at 11 DAP, and on a per

kernel basis at 13 DAP (Fig. 3-4). This pattern is consistent with the concept of increased Sus2

activity promoting transition from cell division to cell enlargement around 12 DAP. Conversely,

in the mn] line, Sus2 had a delayed and lower abundance of transcript at 13 DAP, possibly as a

result of low hexose signaling fueling developmental progress (Fig. 3-4B). On a per kernel basis,

mn] samples displayed a brief increase of Sus2 transcript at 11 DAP, with a subsequent decline

that probably reflected the cessation of kernel expansion after -12 DAP (Fig. 3-4C).

The HXK2 gene showed an insignificantly higher expression in the wild-type versus the

mn] genotype for year 2007 samples. This pattern was more clearly resolved in year 2008, when

the 8 DAP stage showed 2-fold higher levels in Mn] kernels (Fig 3-5). This difference in

expression was similar to that of the Mn] gene, in that levels were highest at early stages and

declined steadily. The same plateau in the wild-type was seen between 11 and 13 DAP on a per

kernel basis, further supporting this period as significant in developmental progression.









Overall, the hexose-related genes, Mn] and HXK2, were expressed early and declined

during development in the wild-type, while Sus2 expression rose at the 11-13 DAP transition

period. Hexokinase inhibitors have been shown to impair sucrose- and hexose-dependent

induction of a major sucrose synthase gene, Sus], in Arabidopsis leaves, and specific forms of

the sucrose synthase enzyme family are reported to be individually regulated by sugar status

(Koch et al. 1992; Ciereszko and Kleczkowski 2002). The published data provide insight into

possible causes of delayed and reduced Sus2 expression in mnl seeds as a result of low

hexose/sucrose ratio. These changes in INCW2-deficient kernels highlight the pleiotropic effects

of the mnn seed mutation and resulting hexose deficiency during a critical stage of early kernel

development.

Ethylene biosynthesis genes of the ACS and ACO families

As actuators of the rate-limiting step in ethylene biosynthesis, ACC synthase enzymes are

important targets for regulation of hormone effects. Levels for all three of the ACS genes

currently identified in maize were an order of magnitude higher than published results, expressed

both as a function of total RNA and as whole kernel data (Figs. 3-6 to 3-8; Gallie and Young

2004). This could be partially due to differences in genetic background used in the two studies.

Additionally, the soil composition, temperature, water supply and light characteristics of the

environment were possible factors in physiological variation. However, the relative contributions

of each ACS gene to total ACS transcript levels were in agreement with Gallie and Young (2004):

ACS2 was most abundant, followed by ACS7, and ACS6 was the least-expressed.

Few significant differences were seen between Mn] and mnn samples for the ACS family

(Figs. 3-6 to 3-8). The ACS2 and ACS6 transcripts were more abundant in wild-type than mnn at

8 DAP, but large standard error prevented solid conclusions. All three family members displayed

a constant or slightly downward trend in both genotypes, on a per ng total RNA and per kernel









basis. This trend was not dissimilar to the wild-type results of Gallie and Young (2004), given

that data in the current study covered only early development for Mn] and mn]. The most

compelling observation was from ACS6 data, which showed a clear increase in transcript levels

at 13 DAP in the wild-type from year 2008 (Fig 3-7B and C). The transcript levels at this stage

coincided with the peak in ethylene evolution in the same samples (Fig 3-2B) and also matched

the timing of ACS6 transcript in Gallie and Young (2004).

A disproportionate contribution of the three ACS genes to ethylene biosynthesis is

demonstrated in maize leaf tissue (Young and Gallie 2004).The loss of acs2 function reduces

ethylene levels by 50%, compared with a mutation in the acs6 gene that causes a 90% reduction

in hormone levels. Even though the ACS proteins are subject to extensive post-transcriptional

regulation (Tatsuki and Mori 2001; Liu and Zhang 2004; Sebastia et al. 2004), the data presented

in Figures 3-6 through 3-8 are in agreement with the disproportionate contribution of ACS6

transcript levels to increased ethylene biosynthesis described previously (Young and Gallie

2004). This observation is based primarily on the correlation of ACS6 transcript with ethylene

evolution in 13 DAP Mn] kernels (Fig. 3-7).

The ACC oxidase family is comprised of four members in maize, three of which are

expressed in kernels (Gallie and Young 2004). These enzymes convert ACC produced by ACC

synthase into ethylene, which is then free to diffuse through tissues and interact with receptors.

Two genes of the ACC oxidase family were analyzed in year 2008 samples only. The highly-

similar AC020 and AC035 genes showed relative expression rates similar to those published by

Gallie and Young (2004), although absolute abundance shown here was again an order of

magnitude higher than their published work. The peak levels in AC035 transcripts, shown here

at 13 DAP, were similar in both Mn] and mn] genotypes (Fig 3-10B) and were in agreement









with data published in Gallie and Young (2004). Compared to the wild-type, a higher proportion

ofAC020 transcript was found in mn] kernels at 13 and 20 DAP based on # transcripts/ng total

RNA (Fig. 3-9). There is no reason to conclude that this is related to higher ethylene synthesis in

the mn] line from 16 DAP onward, although the difference at 13 DAP could be indicative of

increased induction of ethylene-related transcripts at the expense of normal storage gene

synthesis. AC020 transcript levels in wild-type kernels maintained constant expression per ng

total RNA during development, which supports the observation of Gallie and Young (2004) that

this gene is not induced early in endosperm growth.

Ethylene receptor genes and EIL1-1

Ethylene receptors are critical components of ethylene signaling. They function as negative

regulators of ethylene response that are inactivated upon binding of the ethylene molecule (Hua

and Meyerowitz 1998). All four of the known maize ethylene receptors are expressed in maize

kernels (Gallie and Young 2004). The three receptor transcripts examined here, ERS1-14, ETR2-

9 and ETR2-40, followed a general pattern of decline from 8 to 20 DAP in terms of relative

mRNA abundance (Figs. 3-11 to 3-13), with the ETR transcripts slightly more numerous than

ERS1-14. Both of these observations are in agreement with results published by Gallie and

Young (2004).

For the ERS114 transcript, little difference between the Mn] and mn] genotypes was

apparent in year 2007 samples (Fig. 3-11A). Data from year 2008 showed a distinct and

consistent difference between Mn] and mn] kernels at 11 DAP (Figure 3-11B). This is another

example of a variation in the mn] samples that coincides with a critical period in kernel

development and transcriptional reprogramming. While increased ERS1-14 transcript in itself

does not reveal the status of the ERS1-14 protein, and the direct cause of transcriptional









modulation for this gene is unknown, correlative evidence connects a transient increase in ERS1-

14 levels to decreased glucose and/or auxin levels as a result of the mnn mutation.

The two ETR2 genes showed similar expression between Mnl and mnn genotypes at 8 and

11 DAP on a per kernel basis. While the Mnl samples showed an increase in ETR2 transcript at

13 DAP, the mnn samples transcript levels decreased, and both lines remained level during

subsequent development (Figs. 3-12C and 3-13C). This could be explained by differences in

kernel size. Mnl and mnn kernels have similar characteristics until 11 DAP, at which point the

miniature phenotype leads to impaired growth in the hexose-deficient mutant seeds. This trend of

transcript reduction in mnn kernels was visible in the ERS1-14 results as well, but differences

were not statistically significant. Overall it seems that the mnn mutation did not affect

transcription of ETR2 ethylene receptor genes in developing maize seeds, and only impacted

ERS1-14 transcript levels during the 11-13 DAP transition period.

Gallie and Young (2004) note similarities between the pattern of receptor expression and

the downstream signaling components EIN2, EIL1-1 and EIL1-3. While only EIL1-1 transcript

data is reported here, the trend was similar to that of the receptor genes. At all stages the trends

of EIL1-1 transcript levels were similar between Mnl and mnn lines, with a notable reduction

visible in mnn kernels at 13 DAP that correlated with reduction in kernel size relative to wild-

type (Fig. 3-14). EIL1-1 transcripts increased from 8 to 20 DAP in both lines, possibly

facilitating increased ethylene signaling as kernels mature. These data do not support a

possibility for transcriptional modulation of EIL1-1 due to the mnn mutation.

Transcript Accumulation in Sul and sul Genotypes

Metabolic genes

Due to difficulty producing an adequate supply of field-grown samples, the transcript

levels for Sul and sul genotypes were the results of single biological sample analysis at each









time point. Dinges et al. (2001) report that the reference sul mutation reduces SU1 protein

accumulation while leaving transcript levels unchanged. The use of the Sul transcript as an

internal control is of critical importance because of the expected similarity in Sul transcript

levels between the mutant and wild-type kernels. However, the published data focuses on a

single time point (20 DAP) and the use of RNA gel blot analysis to quantify transcript levels

(Dinges et al. 2001). With that in mind, both year 2007 and year 2008 results showed similar

levels of Sul and sul transcript as a proportion of total RNA for all stages (Fig. 3-17 and 3-18 A

and B), lending credibility to the accuracy of the assay. The observation that sul kernels

exhibited fewer Sul transcripts per kernel at 20 DAP could be a result of chance selection of

smaller kernels for that time point. Additionally, sul kernels could have had higher water content

but reduced dry weight during this stage, leading to an overall lower production of RNA per

kernel. When the singular nature of the 20 DAP sul sample is considered, a lower efficiency of

RNA isolation would also have led to an inaccurate determination of total RNA per kernel, thus

skewing transcript results. Also, transcriptional control of Sul is poorly understood, and could be

related to the drastically altered sugar and starch profile present in the sul genotype (Creech

1965; Dinges et al. 2001).

The hexokinase HXK2 transcript levels paralleled the trend of Sul transcript results (Fig.

3-17 and 3-19). HXK2 showed similar trends of transcript expression between Sul and sul

genotypes (Fig. 3-19). Highest levels of transcript accumulation were during times of peak cell

division during early development. There was no apparent transcriptional modulation of HXK2

as a result of the sul mutation.

While Sul and HXK2 transcript levels were similar between wild-type and sul kernels, the

levels of Sus2 showed different responses based on genotype in the single replicates shown here









(Fig. 3-18). The most consistent response was in year 2008 samples, where Sul samples

produced more Sus2 transcript than the sul mutant at all stages except 20 DAP (Fig 3-18B and

C). Higher sucrose levels in the sul genotype could be one explanation for lower Sus2

expression. It is important to note that in this study, both the mn] and sul genotypes caused

higher levels of sucrose relative to the wild-type kernels, and there were decreased levels of Sus2

transcript that correlated with that modification of sugar status. Because the correlation between

Sus2 transcript levels and SUS2 enzyme activity is unknown, further analysis is needed to

understand the function of Sus2 in this system.

Ethylene biosynthesis genes of the ACS and ACO families

The ACC synthase family showed largely unrelated patterns of expression over two

consecutive years. For year 2007 samples, all three ACS genes were strongly expressed in sul

kernels at 8 and 28 DAP versus wild-type (Figs. 3-20A to 3-22A). The period from 12 to 20

DAP was more similar between lines, to the point that no significant difference was apparent.

Why the earliest and latest stages would show such variation is perplexing, considering that

ethylene production was similar at all stages (even though the 8 DAP sul sample was

unavailable for ethylene analysis). The lack of replication leads to the conclusion that

experimental error and random variation are as likely as any biological explanation for the

differences at 8 and 28 DAP. One consistency is that ACS7 and ACS2 mRNAs were more

abundant than ACS6, in agreement with other data presented in this report (Figs. 3-20A to 3-

22A).

Concerning year 2008 samples, the relative contributions of each family member to total

transcript amounts were in agreement with year 2007 data, in that the ACS2 and ACS7 transcripts

showed up to 10-fold higher expression than ACS6 (Figs 3-20 to 3-22). For ACS2, expression

was identical between Sul and sul samples at 15 and 20 DAP (Fig. 3-20 B and C), which is a









period marked by higher sucrose in the sul line. It is logical to hypothesize that an effect due to

sugar status would manifest itself during this stage, but this was not reflected in the data. At 25

DAP both ACS2 transcript and ethylene levels were elevated in the Sul genotype. At the same

time point ACS2 transcript and ethylene levels were low in the sul mutant (Figs. 3-15 and 3-20).

This suggests a possible correlation between this ACC synthase isoform and ethylene production

at this stage in development. Overall levels of ACS2 transcript rose throughout development in

both genotypes, further supporting ACS2 as a developmentally-modulated gene.

The ACS7 transcript, when translated, shares 95% amino acid identity with translated

ACS2 (Gallie and Young 2004). However, ACS7 showed a different pattern of expression than

the highly-similar family member. There was a slight decline in ACS7 transcript levels during

development. Overall there was no significant difference in ACS7 transcript abundance between

Sul and sul kernels (Fig. 3-22B and C). While the contribution of the ACS7 gene product to

ethylene levels in maize kernels cannot be deduced from transcript data alone, the observation of

relatively constitutive expression does not support the possibility of transcriptional regulation

due to developmental or metabolic cues.

The levels of ACS6 transcript were similar between wild-type and sul samples at all stages

except 20 DAP (Fig. 3-21B and C). In contrast to the other family members, ACS6 was

maximally expressed at the earliest stage tested. This is in agreement with transcript levels

published in Gallie and Young (2004) and supports the possibility that ACS6 is specifically

involved in early ethylene biosynthesis, and is in part specifically up-regulated at this stage.

Ultimately, few conclusions can be drawn from such preliminary data concerning

transcriptional regulation of the ACC synthase family in developing maize kernels. ACS6

appeared to be more highly expressed during early stages, while ACS2 was more prominent later









in development. ACS7 transcript showed little variation during the period of development

investigated here. These observations are in general agreement with results from Gallie and

Young (2004a), but further analysis is required for validation.

Transcripts for the ACC oxidase genes AC020 and AC035 were present at all stages of

development tested. AC020 was again the more highly expressed of the two family members

(Figs 3-23 and 3-24). The Sul and sul genotypes were similar with respect to AC020

accumulation; with relatively constant expression from 15 to 35 DAP. The AC035 transcript

levels in both genotypes showed a modest increase as kernels developed (Fig. 3-24), possibly

providing ACO enzymes in order to increase ethylene evolution during maturation. Results for

the two ACO members investigated here showed similarity between Sul and sul genotypes.

However, AC035 transcript levels were highest at early stages in the report by Gallie and Young

(2004), contrary to data presented in this report. One possibility is that isoforms of the ACO

family could be differently regulated as a result of changes in genetic background. With only

single biological samples to draw data from, it is beyond the scope of this study to make such a

claim. Similarities between Sul and the mutant genotype lead to the conclusion that

transcriptional modulation of the ACO genes as a result of the sul mutation was unlikely during

the stages tested.

Ethylene receptor genes and EIL1-1

The ETR2-9 and ERS1-14 transcripts quantified in Sul and sul samples from year 2007

showed dissimilar patterns of expression (Figs 3-25A and 3-26A). The ETR2-9 transcript was

most abundant and coincided with ethylene levels (Fig. 3-15). ERS1-14 had a constant level of

expression except for a substantial increase in 8 DAP sul kernels (Fig. 3-25A). This singular

variation was not repeated in any other receptor data, casting doubt on the accuracy of this result









and emphasizing the need for biological replication. There was no significant difference between

Sul and sul genotypes for either receptor other than the anomalous 8 DAP ERS1-14 time point.

For year 2008 samples, both ERS1-14 and ETR2-9 had similar trends and quantities of

expression in both wild-type and sul kernels (Figs 3-25 and 3-26). The ETR2-40 results showed

more variation between genotypes, with a higher amount of transcript in the Sul line at the

earliest stage (Fig. 3-27). The most significant difference was the higher receptor transcript

levels in Sul versus sul at 35 DAP. This is the time when sul kernels begin to produce

substantially more ethylene than wild-type (Young et al. 1997), which would be exacerbated by

lower receptor levels leading to increased sensitivity. The observation that receptor levels began

to rise as ethylene increased in the Sul genotype raised the possibility that ethylene activity

increased transcription of the receptors. However, no changes in ethylene levels were observed

in these samples, contrary to published data (Young et al. 2007) Also, receptor transcript levels

do not always correlate with protein accumulation (Kevany et al. 2007). It is possible that

ethylene action simultaneously induces receptor transcript and prevents receptor protein

accumulation depending on the activity of tissue-specific control mechanisms, as seen in

transcript and protein levels of ripening tomato fruits (Kevany et al. 2007).

The EIL1-1 transcript level increased modestly during development, with similar

expression in Sul and sul samples (Fig. 3-28). An increase in the EIL7-1 transcript, if it led to

increased protein accumulation and activity, could facilitate increased ethylene signaling during

the later developmental stage associated with kernel maturation. However, since quantitative

protein data is unavailable, further work is needed to connect transcript levels with ethylene

responses. No transcriptional differences between wild-type and sul genotypes were clear from

the data shown.









CHAPTER 5
CONCLUSIONS

This report has demonstrated that patterns of biosynthesis of the phytohormone ethylene

were affected by the mn] mutation in developing maize seeds. In addition, data presented here

documented variance in hormone levels based on the genetic background of wild-type and

mutant genotypes. The mn] mutation led to an increase in ethylene production rates in the

smaller kernels during the 16-25 DAP stage, adding to the list of pleiotropic effects attributed to

loss of INCW2 enzyme activity. Variance based on genetic background was highlighted by the

novel results of Sul kernel analysis, which showed a second burst of ethylene production at 24

DAP, shortly after the initial peak at 16 DAP. This pattern is unique when compared to the

results of Mn] and mn] kernels analyzed during the same period, as well as previously published

reports.

Transcript levels of several genes were shown to be affected by the mn] mutation. For the

first time, to the author's knowledge, the Mn] transcript was quantified in developing seeds and

shown to be present in the mn] mutant kernels at levels between 5 and 10% of the related wild-

type samples. Both Sus2 and HXK2 transcripts were reduced in the hexose-deficient mn]

genotype. Of the three ACC synthase genes, transcript levels of ACS6 appeared to be the most

closely-correlated to rates of ethylene production. This is in agreement with previous reports on

the importance of this gene in ethylene biosynthesis in other tissues. The levels of ACS7

transcripts were largely constant throughout development in all genotypes tested, leading to the

possibility that this ACC synthase transcript was not regulated via developmental cues.

Transcript levels of the ACC oxidase genes were similar in mn] kernels compared to the wild-

type control. However, a slightly higher level of transcript was evident in the mutant genotype at

13 DAP for both AC020 and AC035. The difference in ERS1-14 transcript level, which was









increased in the mn] samples over the levels in Mn] kernels, was only observed at the 11-13

DAP period associated with transition from cell division to cell expansion. Transcript data for

the ETR2 receptors were similar in both Mn] and mn] genotypes. The EIL1-1 transcript was

more abundant as development progressed in both genotypes in what appeared to be a

developmentally-regulated process.

Transcript levels for isoforms of the ACC synthase, ACC oxidase and ethylene receptor

gene families showed some developmental regulation between family members, but little

transcriptional variation was clearly attributable to the mn] mutation.

Due to the lack of biological replication, conclusions based on transcript data from Sul and

sul genotypes are tenuous. The HXK2 transcript levels appeared unchanged, while the Sus2

levels were clearly modified in the high-sucrose sul kernels. Transcripts of the ACC synthase

genes ACS2 and ACS7 were much more abundant than those of ACS6, in agreement with data

from Mn] and mn] kernels. A higher proportion of ACS6 transcript was present during early

stages, while ACS2 transcripts were more abundant during later development. ACS7 transcript

levels appeared to be constitutive in both genotypes throughout kernel growth. The ACC oxidase

transcripts were similarly expressed in both sul and the wild-type kernels, with AC035 increased

during later stages of growth. The ethylene receptor expression in both Sul and sul genotypes

decreased during development, with little difference in transcription between wild-type and

mutant kernels. Conversely, EIL1-1 transcript levels rose in both genotypes over time, consistent

with a role in ethylene signaling during PCD and kernel maturation during late development.

In summary, the biosynthesis of ethylene in developing maize kernels is a complex result

of many interacting factors, one of which is sugar status in developing seeds. Correlations

between hormone levels and transcript levels of sugar- and ethylene-related genes provide a









basis for further investigation of the process involved in maize kernel development. Both the

similarities and differences between wild-type and mutant maize genotypes offer additional

insight into the genetic interactions influencing kernel development.









LIST OF REFERENCES


Alonso JM, Hirayama T, Roman G, Nourizadeh S, Ecker JR (1999) EIN2, a bifunctional
transducer of ethylene and stress responses in Arabidopsis. Science 284:2148-2152

Beatty MK, Rahman A, Cao H, Woodman W, Lee M, Myers AM, James MG (1999) Purification
and molecular genetic characterization of ZPU1, a pullulanase-type starch Debranching
enzyme from maize. Plant Physiol 119:225-266

Beckles DM, Smith AM, Rees T (2001) A cytosolic ADP-glucose pyrophosphorylase is a feature
of graminaceous endosperms, but not of other starch-storing organs. Plant Physiol
125:818-827

Bhave MR, Lawrence S, Barton C, Hannah LC (1990) Identification and molecular
characterization of,.\///nkevnl-2 cDNA clones of maize. Plant Cell 2:581-588

Black RC, Loerch JD, McArdle FJ, Creech RG (1966) Genetic interactions affecting maize
phytoglycogen and the pyytoglycogen-forming branching enzyme. Genetics 53:661-668

Bosnes M, Weideman F, Olsen OA (1992) Endosperm differentiation in barley wild-type and
sex-mutants. Plant J 2:661-674

Boyer CD, Preiss J (1978) Multiple forms of starch branching enzyme of maize: evidence for
independent genetic control. Biochem Biophys Res Comm 80:169-175

Cao H, Imparl-Radosevich J, Guan H, Keeling PL, James MG, Myers AM (1999) Identification
of the soluble starch synthase activities of maize endosperm. Plant Physiol 120:205-215

Chao Q, Rothenberg M, Solano R, Roman G, Terzaghi W, Ecker JR (1997) Activation of the
ethylene gas response pathway in Arabidopsis by the nuclear protein ETHYLENE-
INSENSITIVE3 and related proteins. Cell 89:1133-44

Cheng WH, Tallercio EW, Chourey PS (1996) The Miniature] seed locus of maize encodes a
cell wall invertase required for normal development of endosperm and maternal cells in the
pedicel. Plant Cell 8:971-983

Cho YH, Yoo SD, Sheen J (2006) Regulatory functions of nuclear Hexokinase 1 complex in
glucose signaling. Cell 127:579-589

Chourey PS, Jain M, Li QB, Carlson SJ (2006) Genetic control of cell wall invertases in
developing endosperm of maize. Planta 223:159-167

Chourey PS, Nelson OE (1996) The enzymatic deficiency conditioned by the shrunken-]
mutations in maize. Biochem Genet 14:1041-1055

Chourey PS, Taliercio EW, Carlson SJ, Ruan YL (1998) Genetic evidence that two isozymes of
sucrose synthase present un developing maize endosperm are critical, one for cell wall
integrity and the other for starch biosynthesis. Mol Gen Genet 259:88-96









Ciereszko I, Kleczkowski LA (2002) Glucose and mannose regulate the expression of a major
sucrose synthase gene in Arabidopsis via hexokinase-dependent mechanisms. Plant Phy
Biochem 40:907-911

Clark KL, Larsen PB, Wang X, Chang C (1998) Association of the Arabidopsis CTR1 Raf-like
kinase with the ETR1 and ERS ethylene receptors. Proc Natl Acad Sci USA 95:5401-5406

Creech RG (1965) Genetic control of carbohydrate synthesis in maize endosperm. Genetics
52:1175-1186

Correns, C (1901) Bastarde zwischen maisrassen, mit besonder Berucksichtung der Xenien. Bibl
Bot 53, 1-161

Dai N, Schaffer AA, Petreikov M, Granot D (1995) Arabidopsis thaliana hexokinase cDNA
isolated by complementation of yeast cells. Plant Physiol 108:879-880

Denyer K, Dunlap F, Thorbjornsen T, Keeling P, Smith AM (1996) The major form of ADP-
glucose pyrophosphorylase in maize endosperm is extra-plastidial. Plant Physiol 112:7779-
785

Diboll AG (1968) Fine structure development of the megagametophyte of Zea mays following
fertilization. Amer J Bot 55:787-806

Dickinson DB, Preiss J (1969) ADP glucose pyrophosphorylase from maize endosperm. Arch
Biochem Biophys 130:119-128

Dinges JR, Colleoni C, Myers AM, James MG (2001) Molecular structure of three mutations at
the maize sugary] locus and their allele-specific phenotypic effects. Plant Physiol
152:1406-1418

Evensen KB, Boyer CD (1986) Carbohydrate composition and sensory quality of fresh and
stored sweet corn. J Amer Soc Hort Sci 111:734-738

Foyer CH (1988) Feedback inhibition of photosynthesis through source-sink regulation in leaves.
Plant Physiol Biochem 26:483-492

Gallie DR, Young TE (2004) The ethylene biosynthetic and perception machinery is
differentially expressed during endosperm and embryo development in maize. Mol Gen
Genet 271:267-281

Gallie DR, Geisler-Lee J, Chen J, Jolley Blair (2009) Tissue-specific expression of the ethylene
biosynthetic machinery regulates root growth in maize. Plant Mol Biol 69:195-211

Gou H, Ecker JR (2003) Plant responses to ethylene gas are mediated by SCF EBF1EBF2-dependent
proteolysis of EIN3 transcription factor. Cell 115:667-677

Grime JP, Mowforth MA (1982) Variation in genome size- an ecological interpretation. Nature
299:151-153









Hennen-Beirwagen TA, Liu F, Marsh RS, Kim S, Gan Q, Tetlow IJ, Emes MJ, James MG,
Myers AM (2008) Starch biosynthetic enzymes from developing maize endosperm
associate in multisubunit complexes. Plant Physiol 146:1892-1908

Hua J, Meyerowitz EM (1998) Ethylene responses are negatively regulated by a receptor gene
family in Arabidopsis thaliana. Cell 94:261-271

James MG, Robertson DS, Myers AM (1995) Characterization of the maize gene sugary], a
determinant of starch composition in kernels. Plant Cell 7:417-429

Jang JC, Leon P, Zhou L, Sheen J (1997) Hexokinase as a sugar sensor in higher plants. Plant
Cell 9:5-19

Jang JC, Sheen J (1994) Sugar sensing in higher plants. Plant Cell 6:1665-1679

Johnson PR, Ecker JR (1998) The ethylene gas signal transduction pathway: a molecular
perspective. Annu Rev Genet 32:227-254

Kevany BM, Tieman DM, Taylor MG, Dal Sin V, Klee HJ (2007) Ethylene receptor degradation
controls the timing of ripening in tomato fruit. Plant J 51:458-467

Kieber JJ, Rothenberg M, Roman G, Feldmann KA, Ecker JR (1993) CTR1, a negative regulator
of the ethylene response pathway in Arabidopsis, encodes a member of the Raf family of
protein kinases. Cell 72:427-41

Kiesselbach TA (1949) The structure and reproduction of corn. Univ. of Nebraska Press:
Lincoln, NE

Koch KE, Nolte KD, Duke ER, McCarty DR, Avigne WT (1992) Sugar levels modulate
differential expression of maize sucrose synthase genes. Plant Cell 4:59-69

Kowles RV, Phillips RL (1985) DNA amplification patterns in maize endosperm nuclei during
kernel development. Proc Natl Acad Sci USA 82:7010-7014

Kowles RV, Phillips RL (1988) Endosperm development in maize. Int Rev Cytol 112:97-136

Kyle DJ, Styles ED (1977) Development of aleurone and sub-aleurone layers in maize. Planta
137:185-193

LeClere S, Schmelz EA, Chourey PS (2008) Cell wall invertase-deficient miniature] kernels
have altered phytohormone levels. Phytochem 69:692-699

Li QB, LcClere S, Chourey P (2008) Plieotropic effects of a single-gene mutation in sucrose
utilization pathway extend beyond its tissue- and metabolic-specificity in developing seed
of maize. Fl Genet Conf Abstract/Poster 55









Liu Y, Zhang S (2004) Phosphorylation of 1-aminocyclopropane-l-carboxylic acid synthase by
MPK6, a stress-responsive mitogen-activated protein kinase, induces ethylene biosynthesis
is Arabidopsis. Plant Cell 16:3386-3399

Lowe J, Nelson OE Jr (1946) Miniature seed- a study in the development of a defective
caryopsis in maize. Genetics 31:525-533

Lur HS, Setter TL (1993) Role of auxin in maize endosperm development. Plant Physiol
130:273-280

Maniaitis T, Fritsch EF, Sambrook J (1982) Molecular cloning: a laboratory manual. Cold Spring
Harbor Laboratory, Cold Spring Harbor, New York.

Miller ME, Chourey PS (1992) The maize invertase-deficient miniature-] seed mutation is
associated with aberrant pedicel and endosperm development. Plant Cell 4:297-305

Moore B, Zhou L, Rolland F, Hall Q, Cheng WH, Liu YX, Hwang I, Jones T, Sheen J (2003)
Role of the Arabidopsis glucose sensor HXK1 in nutrient, light, and hormonal signaling.
Science 300:332-336

Myers AM, Morell MK, James MG, Ball SG (2000) Recent progress toward understanding
biosynthesis of the amylopectin crystal. Plant Physiol 122:989-997

Nelson 0, Pan D (1995) Starch synthesis in maize endosperm. Annu Rev Plant Physiol Plant
Mol Bio 46:475-496

Ohto M, Onai K, Furukawa Y, Aoki E, Araki T, Nakamura K (2001) Effects of sugar on
vegetative development and floral transition in Arabidopsis. Plant Physiol 127:252-261

Pego JV, Weisbeek PJ, Smeekens SCM (1999) Mannose inhibits Arabidopsis germination via a
hexokinase-mediated step. Plant Physiol 119:1017-1023

Preiss J, Danner S, Summers PS, Morell M, Barton CR, Yang L, Nieder M (1990) Molecular
characterization of the Brittle-2 gene effect on maize endosperm ADPglucose
pyrophosphorylase subunits. Plant Physiol 92:881-885

Rahman A, Wong Ks, Jane Jl, Myers AM, Jane MG (1998) Characterization of SU1 isoamylase,
a determinant of storage starch structure in maize. Plant Physiol 117:425-435

Saltman P (1953) Hexokinase in higher plants. J Biol Chem 200(1):145-154

Sebastia CH, Hardin SC, Clouse SD, Kieber JJ, Huber SC (2004) Identification of a new motif
for CDPK phosphorylation in vitro that suggests ACC synthase may be a CDPK substrate.
Arch Biochem Biophys 428:81-91

Shannon JC, Creech RG, Loerch JD (1970) Starch synthesis studies in Zea mays. Plant Physiol
45:163-168









Shannon JC, Pien FM, Cao H, Liu KC (1998) Brittle-1, an adenylate translocator, facilitates
transfer of extraplastidial synthesized ADP-glucose into amyloplasts of maize endosperms.
Plant Physiol 117:1235-1252

Sheen J. 1990 Metabolic repression of transcription in higher plants. Plant Cell 2:1027-1038

Shewry PR, Halford NG. (2002) Cereal seed storage proteins: structures, properties and role in
grain utilization. Journal Exp Bot 53:947-958

Shure M, Wessler S, Fedoroff N (1983) Molecular identification of the Waxy locus in maize.
Cell 35:225-233

Slein MW, Cori GT, Cori CF (1950) A comparative study of hexokinase from yeast and animal
tissues. JBiol Chem 186:763-780

Tatsuki M, Mori H (2001) Phosphorylation of tomato 1-aminocyclopropane-l-carboxylic acid
synthase, LeACS2, at the C-terminal region. J Biol Chem 276:28051-28057

Tsai CY, Salamini F, Nelson OE (1970) Enzymes of carbohydrate metabolism in the developing
endosperm of maize. Plant Physiol 46:299-306

Vilhar B, Kladnik A, Blejec A, Chourey PS, Dermastia M (2002) Cytometrical evidence that the
loss of seed weight in the miniature] seed mutant of maize is associated with reduced
mitotic activity in the developing endosperm. Plant Physiol 129:23-30

Wu C, Colleoni C, Myers AM, James MG (2002) Enzymatic properties and regulation of ZPU1,
the maize pullulanase-type starch Debranching enzyme. Arch Biochem Biophys 406:21-32

Xiao W, Sheen J, Jang JC (2000) The role of hexokinase in plant sugar signal transduction and
growth and development. Plant Mol Bio 44:451-461

Yang S, Hoffman N (1984) Ethylene biosynthesis and its regulation in higher plants. Annu Rev
Plant Physiol 35:155-89

Yanigisawa S, Yoo SD, Sheen J (2003) Differential regulation of EIN3 stability by glucose and
ethylene signaling in plants. Nature 425:521-525

Yoo SD, Cho YH, Tena G, Xiong Y, Sheen J (2008) Dual control of nuclear EIN3 by bifurcate
MAPK cascades in C2H4 signalling. Nature 451:789-796

Young TE, Gallie DR (2000a) Programmed cell death during endosperm development. Plant
Mol Bio 44:283-301

Young TE, Gallie DR (2000b) Regulation of programmed cell death in maize endosperm by
abscisic acid. Plant Mol Bio 42:397-414

Young TE, Gallie DR (2004) ACC synthase expression regulates leaf performance and drought
tolerance in maize. Plant J 40:813-825









Young TE, Gallie DR, DeMason DA (1997) Ethylene-mediated programmed cell death during
maize endosperm development of wild-type and shrunken2 genotypes. Plant Physiol
115:737-751

Zhou L, Jang JC, Jones TL, Sheen J (1998) Glucose and ethylene signal transduction crosstalk
revealed by an Arabidopsis glucose-insensitive mutant. Proc Natl Acad Sci 95:10294-
10299









BIOGRAPHICAL SKETCH

Andrew Joseph Funk was born to Joseph William and Donna Carol Funk during the year

1983 in the town of Decatur, IL. In 1990 he moved with his parents and sister to Tallahassee, FL.

During high school he developed a love for biology under the teaching of Janice Ouimet. This

led to his receipt of a bachelor's degree in microbiology and cell science from the University of

Florida in 2006, where he worked as a research assistant in the laboratory of Dr. Prem Chourey.

At Dr. Chourey's invitation, Andrew applied to the plant pathology program at the University of

Florida, and worked for the United States Department of Agriculture while fulfilling

requirements for the degree of Master of Science.





PAGE 1

EFFECTS OF SUGAR METABOLISM MUTATIONS ON ETHYLEN E PRODUCTION AND RELATED TRANSCRIPT LEVELS IN DEVELOPING MAIZE SEEDS By ANDREW JOSEPH FUNK A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2009 1

PAGE 2

2009 Andrew Joseph Funk 2

PAGE 3

To my family 3

PAGE 4

ACKNOWLEDGMENTS There are a host of people that contributed to my development, both personally and professionally, during the pursuit of this degree. I would like to acknowledge the work of my advisor, Dr. Prem Chourey, and his valuable advice throughout the process of designing, conducting, and communicating this research. Thanks to the other members of my advisory committee, Dr. Harry Klee and Dr. Wen-Yuan S ong, for keeping their doors open to me even when I was too naive to ask for their help, a nd graciously providing assistance and critique during the final stages of my studies. Thanks to Dr. Peter Teal for believing that somewhere inside me is a worthwhile employee that just n eeds some direction. Thanks to Dr. Don Huber for the use of his gas chromatography equipment, a nd much appreciation for the expert assistance and instruction of James Lee in utilizing that eq uipment. Thanks to Qin Bao Li for making his wealth of experience available to me while teach ing me what is important and what doesnt matter. A special thanks to my friend and origin al scientific mentor, M ukesh Jain, and his wife Rani, who took me into their home and expande d my understanding of kindness and hospitality. I would like to acknowledge the good example provided by Dr. F onsie Guilaran and his wife Lesley, whose friendship has been a source of encouragement and admonishment as I have grown into adulthood. I gladly acknowledge th e unfailing love and support provided by my father Joesph, my mother Donna, and my sister Marjorie, without whom my life would be a fractured shard instead of part of a whole. Over all of thes e things I acknowledge God, who offers life and heals what is broken; even me. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 ABSTRACT...................................................................................................................................10 CHAPTER 1 INTRODUCTION................................................................................................................. .12 Kernel Development...............................................................................................................12 Ethylene Biosynthesis, Perception and Signaling..................................................................15 Hexokinase and Sugar Signaling............................................................................................18 Description of Mutant Lines...................................................................................................1 9 The mn1 Mutation...........................................................................................................19 The su1 Mutation.............................................................................................................21 2 MATERIALS AND METHODS...........................................................................................22 Field Work and Fresh Material...............................................................................................22 Planting....................................................................................................................... .....22 Harvest.............................................................................................................................22 Gas Chromatography Analysis........................................................................................23 Nucleic Acid Preparation........................................................................................................23 RNA Isolation..................................................................................................................23 DNase treatment..............................................................................................................24 Reverse transcription.......................................................................................................25 Gene-specific Analysis......................................................................................................... ..25 Primer Design.................................................................................................................. 25 Cloning............................................................................................................................26 DNA Sequencing Reaction..............................................................................................27 Absolute Quantitative PCR.............................................................................................28 3 RESULTS...................................................................................................................... .........32 The Mn1 and mn1 Genotypes.................................................................................................32 Ethylene Production........................................................................................................32 Transcript Accumulation in Mn1 and mn1 Genotypes....................................................34 Metabolic genes........................................................................................................34 The ACC synthase gene family................................................................................36 The ACC oxidase gene family.................................................................................38 The ethylene receptor gene family and EIL1-1 ........................................................39 The Su1 and su1 Genotypes....................................................................................................41 Ethylene Production........................................................................................................41 Transcript Accumulation in Su1 and su1 Genotypes......................................................43 Metabolic genes........................................................................................................43 5

PAGE 6

The ACC synthase gene family................................................................................45 The ACC oxidase gene family.................................................................................47 The ethylene receptor gene family and EIL1-1 ........................................................48 4 DISCUSSION................................................................................................................... ......79 Ethylene Accumulation in Developing Seeds........................................................................80 Mn1 Seeds Produced a Distinct Peak of Ethylene between 12 and 14 DAP..................81 Trends in mn1 Kernel Ethylene Production Were Varied Over Two Consecutive Years............................................................................................................................81 Hexose Deficiency and Increased Sucrose Lead to Pleiotropic Effects in Maize Seeds............................................................................................................................82 Su1 and su1 Kernels Displayed Inconsistent Ethylene Production over Two Consecutive Years.......................................................................................................83 Su1 and su1 kernels showed two clear peaks of ethylene production prior to 30 DAP......................................................................................................................84 A second peak of ethylene hormone in Su1 samples from year 2008 showed unique timing........................................................................................................86 Transcript Accumulation and Correlation with Ethylene Hormone Evolution......................86 Transcript Accumulation in Mn1 and mn1 Genotypes....................................................87 Metabolic genes........................................................................................................87 Ethylene biosynthesis genes of the ACS and ACO families.....................................89 Ethylene receptor genes and EIL1-1 ........................................................................91 Transcript Accumulation in Su1 and su1 Genotypes......................................................92 Metabolic genes........................................................................................................92 Ethylene biosynthesis genes of the ACS and ACO families.....................................94 Ethylene receptor genes and EIL1-1 ........................................................................96 5 CONCLUSIONS.................................................................................................................. ..98 LIST OF REFERENCES.............................................................................................................101 BIOGRAPHICAL SKETCH.......................................................................................................107 6

PAGE 7

LIST OF TABLES Table page 2-1 Components of 50mL RNA isolation buffer.....................................................................30 2-2 List of qPCR primers 5 to 3............................................................................................31 7

PAGE 8

LIST OF FIGURES Figure page 3-1 Ethylene produced by Mn1 (blue) and mn1 (pink) kernels, nmols of...............................51 3-2 Ethylene produced by Mn1 (blue) and mn1 (pink) kernels, nmols of...............................52 3-3 Mn1 transcript levels in Mn1 (blue) and mn1 (pink) kernels.............................................53 3-4 Sus2 transcript levels in Mn1 (blue) and mn1 (pink) kernels.............................................54 3-5 HXK2 transcript levels in Mn1 (blue) and mn1 (pink) kernels..........................................55 3-6 ACS2 transcript levels in Mn1 (blue) and mn1 (pink) kernels...........................................56 3-7 ACS6 transcript levels in Mn1 (blue) and mn1 (pink) kernels...........................................57 3-8 ACS7 transcript levels in Mn1 (blue) and mn1 (pink) kernels...........................................58 3-9 ACO20 transcript levels in Mn1 (blue) and mn1 (pink) kernels........................................59 3-10 ACO35 transcript levels in Mn1 (blue) and mn1 (pink) kernels........................................60 3-11 ERS1-14 transcript levels in Mn1 (blue) and mn1 (pink) kernels......................................61 3-12 ETR2-9 transcript levels in Mn1 (blue) and mn1 (pink) kernels........................................62 3-13 ETR2-40 transcript levels in Mn1 (blue) and mn1 (pink) kernels......................................63 3-14 EIL1-1 transcript levels in Mn1 (blue) and mn1 (pink) kernels.........................................64 3-15 Ethylene produced by Su1 (blue) and su1 (pink) kernels, nmols of..................................65 3-16 Ethylene produced by Su1 (blue) and su1 (pink) kernels, nmols of..................................66 3-17 Su1 transcript levels in Su1 (blue) and su1 (pink) kernels.................................................67 3-18 Sus2 transcript levels in Su1 (blue) and su1 (pink) kernels...............................................68 3-19 HXK2 transcript levels in Su1 (blue) and su1 (pink) kernels.............................................69 3-20 ACS2 transcript levels in Su1 (blue) and su1 (pink) kernels..............................................70 3-21 ACS6 transcript levels in Su1 (blue) and su1 (pink) kernels..............................................71 3-22 ACS7 transcript levels in Su1 (blue) and su1 (pink) kernels..............................................72 3-23 ACO20 transcript levels in Su1 (blue) and su1 (pink) kernels...........................................73 8

PAGE 9

3-24 ACO35 transcript levels in Su1 (blue) and su1 (pink) kernels...........................................74 3-25 ERS1-14 transcript levels in Su1 (blue) and su1 (pink) kernels.........................................75 3-26 ETR2-9 transcript levels in Su1 (blue) and su1 (pink) kernels..........................................76 3-27 ETR2-40 transcript levels in Su1 (blue) and su1 (pink) kernels........................................77 3-28 EIL1-1 transcript levels in Su1 (blue) and su1 (pink) kernels...........................................78 9

PAGE 10

Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science EFFECTS OF SUGAR METABOLISM MUTATIONS ON ETHYLEN E PRODUCTION AND RELATED TRANSCRIPT LEVELS IN DEVELOPING MAIZE SEEDS By Andrew Joseph Funk August 2009 Chair: Prem S. Chourey Major: Plant Pathology The seeds of cereal crops are critically impor tant to the global food supply. Maize, one of the most important U.S. field crops, is an idea l model system to study ce real seed development due to its large size and well-de fined tissue and cell types. Se ed development is a complex process with distinct developmental pha ses, including cell division, cell expansion, endoreduplication, starch and protein storage, and kernel matu ration. The timing and progression of kernel development is tightly regulated by factors such as sugar status and phytohormone activity. Interactions betw een sugars and hormones, forms of cross-talk, have emerged as an important subject of study in recent years. This report focused primarily on the Mn1 gene, which encodes the INCW2 cell-wall invertase protein. The miniature1 ( mn1 ) kernel mutation eliminates INCW2 activity, which causes decreased endosperm growth and a final seed weight 30% of the wild-type value. Samples were also collected from sugary1 ( su1) mutant kernels deficien t in a starch-debranching enzyme and a related wild-type accession as a control. Both ethylene production and related transcript levels were analyzed in these genot ypes in order to compare the possible effects of genetic backgrounds and the effect of earlyand late-acting metabolism mutations. Transcript 10

PAGE 11

11 levels of three metabolic genes, Mn1 Sus2 and HXK2 were analyzed via quantitative polymerase-chain-reaction (qPCR) techniques. Additionally, genes cr itical to ethylene biosynthesis and perception were in cluded in the qPCR analysis in order to correlate transcript levels with ethylene activity. Genes under c onsideration were the ACC synthase genes ACS2, ACS6 and ACS7; two ACC oxidase genes ACO20 and ACO35 ; three ethylene receptors ERS1-14 ETR2-9 and ETR2-40; and the DNA binding protein EIL1-1 that is involved in downstream ethylene signaling. Results indicated distinct burst s of ethylene in all sample series, although the quantity and timing of ethylene production varied between genetic backgrounds. The Mn1 samples produced an increase in ethylene 13 days after pollinati on (DAP) that correlated with maximum published INCW2 activity. The smaller mn1 kernels did not display a 13 DAP ethylene burst but produced 2-fold higher levels of ethylen e than the wild-type between 16 and 25 DAP, coinciding with increased sucrose content in the de fective endosperms. The wild-type Su1 kernels generated two distinct peaks of ethylen e production, one at 16 DAP a nd the second at 24 DAP. The su1 samples did not produce the second 25 DAP burst of ethylene, despite increased sucrose levels. Results of transcript analysis indicated HXK2 was reduced up to 2-fold in mn1 samples during early development, consistent with the defi ciency in hexoses present in that genotype. Expression of both the ACC syntha se and ACC oxidase family members appeared to be related to developmental stage more than kernel genotyp e. The ethylene receptors displayed relatively constitutive expression, while the transcription factor EIL1-1 was more highly expressed late in development.

PAGE 12

CHAPTER 1 INTRODUCTION Kernel Development Maize seed formation is the result of coor dinated development of various specialized tissues. Progress of kernel development can be monitored in terms of overlapping phases related to cell proliferation, differentiation, and maturati on (Bosnes et al. 1992). Pollination results in a double-fertilization event common in angiosperms: one of the two sperm nuclei contained in a single male pollen grain fuses with an egg nuc lei contained in the female megagametophyte, creating a 2N zygote. The remaining haploid sp erm nucleus fuses with the two female polar nuclei, creating a 3N endosperm cell (Kiess elbach 1949; Diboll 1968). These two new cells undergo distinct but related processes, ultim ately leading to a quiescent embryo ready to germinate, utilizing storage reserves contained in the terminal endosperm. Immediately after fertilization the zygote undergoes rapid cell di vision to form the mass of cells that will become the embryo. The 3N endosperm cell nucleus also divides rapidly, but the absence of cytokinesis leads to a large, sync ytial, highly multinucleate endosperm cell. By approximately three days after pollination (DAP ) the endosperm begins cellularization, starting with the nuclei at the periphery of the cell and proceeding centripetally (Kiesselbach 1949; Kowles and Phillips 1988). Once the endosperm cel ls are all cellularized, they begin both cell division and expansion. From 812 DAP the endosperm reaches peak cell division, and after 12 DAP the cells begin the transition from differen tiation to maturation, with the exception of the subaleurone region continuing division until ~2 0 DAP (Kiesselbach 1949; Kyle and Styles 1977). The maturation phase involves storage protein synthesis and starch loading, beginning in the central endosperm and proceeding toward the periphery ca. 15 DAP. As a part of maturation, endosperm cells undergo endoreduplication, a re petitive amplification of nuclear DNA without 12

PAGE 13

normal cell division. The main onset of e ndoreduplication picks up as mitosis ends approximately 12 DAP, with a peak in endopolyplo idy levels at 16-18 DAP (Kowles and Phillips 1985). This process is widespread in cells with high metabolic activ ity and is thought to facilitate the increased levels of transc ription required for rapid cell en largement and reserve storage (Grime and Mowforth 1982), although some reports have noted disparity between endoreduplication levels and cell division and en largement (Vilhar et al. 2002). Ultimately, the purpose of endoreduplication is unclear. Between 12 and 22 DAP, maize kernels begi n accumulation of dry matter, with a coordinate increase in the presence of enzymes rela ted to starch and storag e protein biosynthesis. Expression of starch synthesis and protein synthesis genes occurs in concert with a peak at 20-25 DAP, resulting in coordinated starch synthesis in the central endosperm cells as well as oil and protein production increasing towa rd the aleurone and subaleurone layers (Tsai et al. 1970). The main storage carbohydrates in maize kernels are homopolymers of D-glucose: the moderately branched amylopectin, and the predominantly straight-chain amylose. The two polymers are thought to be synthesized concurrently, with th e coordinated action of multiple enzymes leading to a final ratio of 3:1 amylopectin to amylos e (Shannon et al. 1970; Myers et al. 2000). A third, highly-branched, water-soluble polysaccharide phytoglycogen, is detected in very small quantities but is greatly increased when various combinations of starch synthesis genes are defective (Black et al. 1966). The first committed step in starch biosynthesis is the largely cytosolic conversion of glucose-1-phosphate (G1P) to ADP-glucose (ADP-Glc) by the enzyme ADP-glucose pyrophosphorylase (AGP) (Dickinson and Preiss 1969). AGP is a heterotetrameric enzyme encoded by the large subunit Shrunken2 ( Sh2) and small subunit Brittle2 ( Bt2 ) genes in maize 13

PAGE 14

(Bhave et al. 1990; Preiss et al 1990). Distinct cytosolic and pl astidial forms of AGP are thought to exist, with >85% of the activity in mai ze endosperm coming from the cytosolic enzyme (Denyer et al. 1996; Beckles et al. 2001). ADP-g lucose is transported into the amyloplast by membrane-bound BRITTLE1 (BT1) encoded by the Bt1 gene (Shannon et al. 1998). Inside the amyloplast, ADP-Glc monomers are linked through the formation of -1,4 glycosidic bonds via the ac tion of a family of at least five st arch synthase (SS) enzymes. These include a granule-bound starch s ynthase (GBSS) encoded by the Waxy ( Wx ) gene and four soluble SSs designated SS1, SSIIa, SSIIb and DULL1/SSIII. Amylose is produced by GBSS1, and amylopectin is primarily the result of zSSI and DU1/SSIII activity (Shure et al. 1983; Cao et al. 1999). Two additional classes of enzymes ar e required to achieve the specific structure allowing amylopectin to crystallize into insolubl e starch grains: starch branching enzyme (BE) and starch debranching enzyme (DBE ) (Creech 1965; Nelson and Pan 1995). Maize contains three known BEs with the ability to form branching -1,6 glycosidic bonds via cleavage and transfer of -1,4 bonds of linear glucose polymers (Boyer and Preiss 1978). These BEs comprise two classes in maize; BEI shows 10-fold preference for amylose (longchain polymer) as a substrate, while BEIIa and BEIIb transfer shorter chains It is possible that BEI provides substrate for BEIIa and BEIIb (Nel son and Pan 1995). Evidence suggests that SSs and BEs associate in multi-subunit complexes, mo st notably SSI/SSIIa and SSI/BEIIb (HennenBeirwagen et al. 2008). The action of SSs and BEs alone results in substantial production of phytoglycogen at the expense of amylose and amylopectin (Black 1966). DBEs are required to achieve proper distribution of chain length. Two families are pr esent in plants; isoamylases and pullulanases. The maize isoamylase is represented by Sugary1 ( Su1), with specificity for amylopectin but not 14

PAGE 15

pullulan. The pullulanase is represented by Zpu1, having affinity for pullulan yet still able to attack amylopectin in a lesse r fashion (Dinges et al. 2001; Wu et al. 2002). The maize su1 mutation eliminates both isoamylase and pullulanase activity (B eatty et al. 1999). The final stage of endosperm development is the initiation of programmed cell death (PCD). Between 16 and 20 DAP cells in the cen tral endosperm begin to lose viability. Endonucleases dismantle DNA into fragments with sizes in multiples of 180-200 bp, first detectable at 28 DAP (Young et al. 1997). The progr ession of PCD follows the spatial pattern of endoreduplication and starch synthesis, initiating in the cr own-proximal region and central endosperm, then proceeding down toward the base of the kernel and out toward the periphery. The only viable cells in the mature kernel are in the aleurone layer and quiescent embryo (Young and Gallie 2000a). Ethylene Biosynthesis, Perception and Signaling The two-carbon molecule ethylene is the simplest phytohormone identified to date, yet is involved in diverse processes re lated to seed germination, root and shoot elongation, senescence, fruit ripening, and both biotic and abiotic st ress response (Johnson and Ecker 1998). It is a gaseous molecule able to diffuse freely thr ough membranes, and thus the response must be tightly regulated via control of the biosynthetic and signaling m achinery. Ethylene is produced in a two-step sequence beginning w ith ACC synthase enzymatically cleaving a structure off Sadenosyl-L-methionine to form 1-aminocyclopr opane-1-carboxylic acid (ACC). The resulting ACC is rapidly converted to ethylene via the en zyme ACC oxidase. The ACC synthase step is considered rate-limiting for ethylene production (Yang and Hoffman 1984). Ethylene is perceived via inte raction with endoplasmic-reticu lum (ER)-localized receptors resembling bacterial histidine and serine/threonine kinases. Five t ypes of ethylene receptors have been identified in Arabidopsis to date and can be grouped in to two subfamilies based on 15

PAGE 16

homology (Hua and Meyerowitz 1998). Screens for Zea mays have yielded orthologs of two of the five receptor types established in Arabidopsis (Gallie and Young 2004) The two groups are ETHYLENE-RESPONSE2 (ETR2) and ETHYLENE RESPONSE SENSOR1 (ERS1). Both maize types have two members; ZmERS1-14/ZmERS1-25 and ZmETR2-9/ZmETR2-40 While the gene products within the subgroups are 97% a nd 92% identical, respect ively, the similarity between ERS1 and ETR2 is only 51% (41% identic al) at the amino acid level (Gallie and Young 2004). Ethylene receptors are negative regulators of ethylene signa ling (Hua and Meyerowitz 1998). In the absence of the hormone, the receptors activate the Raf-like kinase CTR1, which is thought to control a phosphorylation cascade that ultimately represses the transcriptional regulator ETHYLENE-INSENSITIVE-3 (EIN3) (Kie ber et al. 1993; Clark et al. 1998) EIN3 and a homologous EIN3-LIKE1 (EIL1) subfamily c ontrol a transcriptional cascade related to production of diverse ethylene-responsive genes (Chao et al. 1997). EIN3 and EIL are regulated via phosphorylation state (Yoo et al. 2008) and degradation via the 26S proteasome pathway, with known F-box proteins ETHYLENE-F-BOX1 (EBF1) and ETHYLENE-F-BOX-2 (EBF2) providing targeting specificity (Gou and Ecker 2003). ETHY LENE-INSENSIVITE-2 (EIN2) functions downstream of CTR1 to repress EBF1 /2 action, while EIN2 itself is repressed by CTR1 (Alonso et al. 1999). Thus ethylene bindin g to the receptors inactivates CTR1, releases EIN2 from repression, hinders EBF1/2 and allows EIN3/EIL1 to initiate downstream ethylene responses. While the majority of ethylene research has been done in Arabidopsis and other model systems such as tomato, some progress has b een made in understandi ng ethylene activity in developing maize seed. Early reports quantified th e pattern of ethylene production in relation to 16

PAGE 17

PCD, showing two peaks of hormone accumulation in many lines tested (Young et al. 1997). It is of special importance that the patterns and abso lute values of ethylene production varied between genetic backgrounds, both for wild-types and id entical single-gene mutant lines. A common trend is that ethylene peaks be tween 12 and 16 DAP, coinciding with the first observation of cell death in the central endosperm (Young et al. 1997). A second p eak of ethylene is generated roughly 28-36 DAP, accompanying the appearance of internucleosomal DNA fragmentation. In sh2 mutant lines containing increased sucrose and glucose levels, ethylene generation is more abundant than the control: this occurs in conjun ction with earlier induction of cell death as well as earlier and increased nuclease activity and a reduction in ge rmination rate (Young et al. 1997). In addition, it has been demonstrat ed that PCD can be modulated by a variety of ethylene-related effects. Application of ethylene increased the seve rity of cell death and in duction of nucleases in all lines tested, and was sufficient to cause some cell death in tissues that normally remained viable. Ethylene biosynthesis or perception in hibitors such as 2aminoethoxyvinyl glycine (AVG) and 1-methylcyclopropene (MCP), respectively, delayed the onset and severity of PCD (Young et al. 1997; Young and Gallie 2000a). Finally, abscisic acid (ABA) has been shown to antagonize ethylene, and kernels with impaired ABA synthesis or perception generate more ethylene than controls and also show incr eased internucleosomal DNA fragmentation (Young and Gallie 2000b). Gallie and Young (2004) isolated and charac terized the expression of many critical components of the ethylene pathway from a co llection of maize genomic and cDNA libraries. They identified three ACS genes, four ACO genes, two ERS1-family receptors, two ETR2-family receptors, a single EIN2 and two EIL1 members. It is of note that maize appears to lack ETR1 ERS2, and NR family receptors. Their attempts at isolating a CTR gene were unsuccessful. 17

PAGE 18

Hexokinase and Sugar Signaling The hexokinase family of enzymes cataly zes the conversion of hexose to hexose-6phosphate in an ATP-dependent reaction. This enzyme is conserved across diverse animal, plant, and yeast species (Slein et al. 1950; Saltman 1953; Dai et al. 1995). Hexokinase is positioned at the gateway of hexose utilization as the first st ep in glycolysis, and has been linked to the phenomenon of hexoses as a signa ling molecule in the control of diverse processes, including germination, seedling development, photosynthe tic gene regulation, source/sink partitioning, flowering and senescence (Foyer 1988; Sheen 1990; Jang and Sheen 1994; Pego et al. 1999; Ohto et al. 2001). Furthermore, it was establishe d that glucose and othe r hexoses, but not sugar phosphates, are the direct sugar signals causi ng repression of photosynt hetic genes (Jang and Sheen 1994). With the discovery of distinct and separable cataly tic and regulatory domains (Jang et al. 1997), hexokinases emerged as a critical co mponent of sugar signa ling. Characterization of the GLUCOSE-INSENSITIVE1 ( gin1 ) mutant revealed overlap of ethylene and abscisic acid hormone pathways, with ABA2 ( GIN1) acting downstream of HXK1 to antagonize a branch of ethylene response related to germ ination, cotyledon and leaf de velopment, and flowering (Zhou et al. 1998). Subsequent research garnered support fo r the hypothesis that three sugar signaling pathways exist in higher plants: one dependent on HXK catalytic functi on, another related to HXK signaling, and a third functioning independen t of HXKs (Xiao et al. 2000). Work by Moore et al. (2003) clearly demons trated that HXK1-deficient Arabidopsis plants have altered responses to light, glucose, nitrates, and the hormones auxin and cytokinin. In addition, mutations that abolished kinase function but not glucose bindin g were still able to convey sugar signals, conclusively establishing the multi-functional ro le of hexokinases as enzymes and sugar signal transducers. Hexokinase2 from yeast can compliment HXK1 -deficient plants in regard to 18

PAGE 19

enzymatic functions but does not rest ore sugar signaling in transgenic Arabidopsis (Moore et al. 2003). Also, AtHXK1, but not the yeast analog, associates with nuclear fractions of Arabidopsis cells. While only a small fraction of total HXK is found in the nucleus, ther e the protein interacts with two unconventional partners VHA-B1 (vacuolar H+-ATPase B1) and RPT5B (19S regulatory particle of proteasome subunit) to directly bind DNA and regul ate transcription of glucose-responsive genes (Cho et al. 2006). The key ethylene transcrip tion factor EIN3 is specifically degraded in the pres ence of glucose, but this EI N3 repression is abolished in gin2 ( hxk1 ) mutants (Yanagisawa et al. 2003). Description of Mutant Lines The mn1 Mutation The primary subject of this study is the miniature1 seed mutation, first described by Lowe and Nelson (1946). They described two groups of maize kernel muta tions: the first group produced relatively normal vegetative tissues in rega rd to plant height, color, or vigor, yet had negative traits with respect to kernel development and nutrien t quality. The second group was lethal, semi-lethal or otherwise de fective during sporophytic growth. The miniature1 locus is a member of the first group, with plants show ing generally normal (albeit delayed) seedling development but producing kernels with >70% re duction in seed weight. Given the importance of maize kernels in the global food supply, seed-specific mutations involving kernel development, sugar metabolism and/or starch production are valuable tools for understanding and subsequently modifying biochemical processes for human benefit. The Mn1 locus encodes the INCW2 protein, an invertase enzyme anchored to the cell wall through ionic bonds (Cheng et al. 1996). Other form s of invertases are found in the cytoplasm and vacuoles, and all three types are represen ted in diverse plant sp ecies, catalyzing the unidirectional cleavage of sucr ose into constituent glucose and fructose molecules. The Mn1 19

PAGE 20

gene is maize endosperm-specific, with the enzy me localizing to the basal endosperm transfer cell (BETC) region that acts as a gateway for photosynthate entering the developing seed. This cell-wall invertase is thought to play a role in source-sink partitioning, ac ting in the apoplast to maintain a gradient for phloem unloading via sucrose cleavage, leading to hexose uptake across the BETC layer. The mn1 gene is a naturally occurring non-l ethal mutant that limits enzyme activity to less than 1% of wild-type (Miller and Chourey 1992; Chourey et al. 2006). The earliest detectable mn1 seed phenotype occurs at ~910 DAP, assessed via histological techniques showing gap formation between th e pedicel and BETC layer. The phenotype is readily visualized with the naked eye as miniatur e seed on the cob, relative to wild-type, at the 10-12 DAP stage (Lowe and Nelson 1946; Chourey et al. 1992). In addition, in situ localization of both Mn1 RNA and INCW2 protein demonstrate the ab sence of these molecules (Cheng et al. 1996; Li et al. 2008). These observations coincide with a signif icant increase in sucrose content of mn1 seeds between 8 and 20 DAP, along with a ma rked decrease in glucose and fructose quantity over the same period (Li et al. 2008). Vilhar et al. (2002) further associate the mn1 phenotype with reduced mitotic activity and cell expansion, with no significan t alteration in endoreduplicati on. They speculate that a high sucrose to hexose ratio during early mn1 seed growth is a possible cause of cell differentiation and maturation at the expense of normal cell di vision and expansion. Recently LeClere et al. (2008) demonstrated that the auxins, indole-3-acetic acid (IA A) and IAA conjugates, are significantly reduced in mn1 seed. Their results suggest that INCW2 directly or indirectly influences auxin biosynthesis in developing maize kernels, and this auxin deficiency could be a factor influencing reduced cell size and division in mn1 kernels. 20

PAGE 21

21 The su1 Mutation The sugary1 locus was first identified over a century ago as a maize kernel mutation leading to a glassy, translucent mature seed (Correns 1901). Later analysis revealed an accumulation of simple sugars as well as phytoglycogen in the endosperm of homozygous su1 kernels, corresponding to a large decrease of th e predominant starch amylopectin (Creech 1965; Evensen and Boyer 1986). Beginning at 16 DAP, Creech (1965) showed that su1 kernels maintained double the percentage of sucrose co mpared to wild-type. Reducing sugars were similar at 16 DAP but were 4-fold more abundant at 24 and 28 DAP in su1 seeds. Total sugar content in su1 seeds was double that of Su1 kernels throughout development as a percent of dry matter, and water-soluble polysaccharides (WSP s) were increased 10-fold by 24 and 28 DAP. However, total carbohydrate wa s only 5-8% reduced in the su1 genotype Cloning and characterization of th e transcript and protein of sugary1 led to its identification as a starch debranching enzyme (DBE) targeting the -1,6 linkages of amylopectin and glycogen, part of the -amylase superfamily of starch hy drolytic enzymes (James et al. 1995). The original allele su1-Ref contains two point mutations th at lead to normal transcript levels but no protein accumulation (Dinges et al. 2001). Wild-type and su1 transcripts are detectable as early as 8 DAP, w ith a slightly higher accumulation in the mutant. Enzyme levels peak toward 20 DAP in Su1 lines but are undetecta ble for the duration of su1 kernel development (Rahman et al. 1998).

PAGE 22

CHAPTER 2 MATERIALS AND METHODS Field Work and Fresh Material Planting Maize kernels from six homozygous inbred lines were planted at the University of Florida/Institute of Food and Agricultural Sciences (UF/IFAS) Plant Science Research and Education Unit (PSREU) in Citra, FL for tw o consecutive years. During April 2007 three different single-mutation lines, mn1 sh2 and su1 were planted. In addition to the mutant genotypes, wild-type controls we re planted in equal proportion fo r each of the three related inbred backgrounds from which the mutants we re derived. Developing ears were shoot-bagged prior to silk emergence to prev ent non-specific pollination. As tassels matured and anthers began to shed pollen, tassel bags were attached to the male donor plants one day before pollen collection. Each of the six lines was self-pollinated or sibling-pollinated as material allowed. Each bag was marked with the male and female ge notype used in the cross as well as the date of pollination. Once a plant was successfully pollinated, additional ears were stripped to promote growth of the target ea r. After assessing the total number of ears pollinated, a harvest schedule was drawn up that would attempt to maximize sample coverage from 8 DAP through 32 DAP. During 2008 lines were selfand sib-pollinated to maintain homozygous inbred genotypes for analysis. Pollinations were conducted between 8:30am and 10:30 am. Harvesting was done as close to 10:30 am as possible for every individual sample. Harvest For both years, harvesting consisted of breaki ng off whole ears with husk intact, sealing them in their identification bags, and depositing material in a lab refrigerator maintaining 4C. Kernels were harvested whole onto moist paper towels for gas chromatography (GC) analysis 22

PAGE 23

and then the remainder replaced in the refrigerator until they could be flash-frozen whole directly into liquid nitrogen. Any damaged kernels were discarded. Once frozen, kernels were deposited into labeled 50mL or 15mL Fisherbrand disposable conical-bottom centrifuge tubes and stored at -80C for further analysis. Each ear was stored in a separate tube. Gas Chromatography Analysis The GC analysis performed in this study followed the protocol outlined by Young et al. (1997). Kernels were harvested from fresh ears a nd allowed to rest between moist paper towels for one hour to alleviate possible wound responses Approximately 15 kernels were sealed in a 20mL I-CHEM borosilicate vials with 0.125 inch thick septa caps (Fischer #05-719-111), with up to three replicates per ear. Samples were al lowed 3-4 hours to evolve ethylene gas. During this time a Tracor 540 GC unit feeding into a Hewlett-Packard 3396 Series III integrator was calibrated using an ethylene standard of know n composition. Using a 1 mL gas-tight syringe, 0.5 mL of gas was removed from the vial head spac e and injected into the GC to produce a reading of parts-per-million ethylene. Data was entered into Microsoft Excel for subsequent calculations including standard error. Nucleic Acid Preparation RNA Isolation Total RNA was isolated from frozen tissue us ing an acid-phenol lithium chloride technique adapted from Maniatis et al. ( 1982). A mortar and pestle were tr eated with chloroform, allowed to dry, and then cooled with liquid nitrogen. Two or three ke rnels were weighed then ground under liquid N2 by hand, homogenized with 3 mL of isol ation buffer (Table 2) and allowed to thaw. The mixture was pipetted into 14 mL round-bottom disposable polypropylene Falcon tubes (Fischer #14-959-11B), 1 mL of phenol:chlorofor m:isoamyl alcohol (IA) added, then mixed on a rotary shaker for 5 minutes. Samples were centr ifuged at 5000 rpm for 5 minutes after which the 23

PAGE 24

supernatant was mixed with 2 mL of the phenol:chloroform:IA mixture. The samples were centrifuged for 5 minutes at 5000 rpm, then the supernatant was transfer red by pipette into a clean 14 mL Falcon tube along with 3 mL of ch loroform:IA. The solution was mixed thoroughly and then centrifuged for 10 minutes at 5000 rp m. The aqueous phase (~2.4 mL) was then transferred to a fresh Falcon tube, mixed with an equal amount of 6M lithium chloride, and incubated at -20C for at least one hour to precipitate RNA. The samples were then centrifuged for 30 minutes at 5000 rpm and the supernatant wa s discarded. The pellet was resuspended with 2 mL of 2% potassium acetate and the samples were incubated for 5 minutes at 50C, after which 2 mL of phenol:chloroform:IA was added. The samples were mixed thoroughly and centrifuged 5 minutes at 5000 rpm. Another 2 mL of chlo roform:IA was added, mixed, and centrifuged 5 additional minutes at 5000 rpm. The supernatant was carefully transferred to fresh 30 mL Corex glass tubes (DuPont Instruments #00156) containing 3 mL absolute ethanol and incubated at 20C overnight (at least 8 hours). The samples were centrifuged at 10,000 rpm for 25 minutes, the supernatant discarded, and samples air dried. Finally the pellets were resuspended in a known volume (100-175 uL) of diethylpyrocarbonate (DEPC )-treated water, incubated at 60C for 5 minutes, transferred to clean Eppendorf 1.5mL tubes, and qua ntified using a NanoDrop ND-1000 spectrophotometer (Thermo Fischer Scientific, Wilmington, DE) to calculate RNA recovered per kernel. The isolated RNA was resolved on an 0.8% agarose gel to check for quality. DNase treatment After initial concentration was recorded, 50 uL aliquots were taken from each sample for routine DNase treatment using the Ambion DNA free kit (Ambion # AM1906). Five microliters of 10X DNase I buffer and one microliter rDNase I enzyme were added to each sample, mixed, and incubated at 37 for 20-30 minutes. 5.5 uL DNa se Inactivation Reagent was then added and mixed by shaking and inversion several times fo r two minutes at room temperature. Samples 24

PAGE 25

were spun down in a tabletop centrifuge fo r 1.5 minutes at 10,000 x g and the supernatant removed to new tubes for downstream applications. RNA quality was again verified in the same way as described in the RNA isolation protocol. Reverse transcription Total mRNA was converted into cDNA using the SuperScriptIII First-Strand Synthesis System (Invitrogen #18080-051). 3.5 ug DNAfree RNA was brought to a total volume of 8 uL in 200 uL PCR tubes. A master mix of dNTPs (10 mM) and Oligo(dT)20 (50 uM) was prepared with a 1:1 ratio, and then 2 uL of the mixture was added to each 8 uL (RNA + H2O) sample. Samples were incubated at 65C for 5 minutes then placed on ice for at least one minute. Meanwhile a cDNA synthesis master mix was prepar ed consisting of, per sample: 2 uL 10X RT buffer (200 mM Tris-HCl at pH 8.4, 500 mM KCl), 4 uL 25 mM MgCl2, 2 uL 0.1 M DTT, 1 uL RNase OUTTM (40 U/uL), and 1 uL SuperScript TM III reverse-transcriptase (200 U/uL). A total of 10 uL cDNA synthesis mix was added to each sample, mixed, briefly spun in a table-top microcentrifuge, then incubated at 50C for 50 minutes. To stop the reaction, samples were incubated at 85C for 5 minutes then chilled on i ce. Original RNA remaining in the reaction was digested by adding 1 uL (2 U/uL) Escherichia coli RNase H and incubating at 37C for 20 minutes. Two RT reactions were performed fo r each sample and used for downstream qPCR analysis. Gene-specific Analysis Primer Design Primers for genes of interest were acquired by aligning gene families using Vector NTI software (Invitrogen #12605099) and searching for suitable regions of conservation or divergence, depending on desired us e. For five genes related to ethylene biosynthesis, reverse primers were selected from those designed by Gallie and Young (2004). Forward primers 25

PAGE 26

targeting conserved regions of homologous gene s were designed using PrimerQuest software (Integrated DNA Technologies online). For ACS2, ACS6, and ACS7, new forward and reverse primers were designed targeting su b-300 bp amplification fragments. Mn1 Hexokinase ( HXK ) and Sucrose-synthase2 ( Sus2) primers were designed by Qin Bao Li in the lab. Primers for Sugary1 ( Su1 ) were adapted from James et al. (1995). See Table 2 for a complete list of primers. Cloning Following RT-PCR amplification of target ge ne fragments, PCR product was resolved on a 0.8% agarose gel stained with ethidium bromide. The Promega Benchtop 1kb DNA ladder (Fischer #PR-G7541) was used as a marker for fragment size. Once a primer pair was confirmed to produce a single band of expected size, the fragment was either purified from the remaining PCR reaction using the QIAGEN Mi nielute PCR Purification Kit (QIAGEN #28006) or the band was excised directly from the agarose gel and recovered using the QIA quick agarose cleanup kit (QIAGEN #28706). The purified fr agment (generally 1 uL) wa s used in TOPO TA cloning (Invitrogen #45-0030) The fragment was incubated at room temperature for 5 minutes in the presence of 1 uL salt solution, 1 uL TOPO2.1 P CR cloning vector, and 3 uL purified water in order to ligate the amplicon into the TOPO plasmid, forming a circ ular DNA structur e. A vial of TOP10 OneShot chemically competent E. coli cells (Invitrogen #44-0301) was thawed on ice, and 2 uL of the ligation reaction was added. After 5 minutes on ice, the cells were heat-shocked at 42C for exactly 30 seconds to induce uptake of the recombinant DNA then placed back on ice for 1 minute to recover. In a laminar-flow hood, 250 uL of SOC medium was added to the TOP10 cells and incubated on a rotary shaker at 37C for 1 hour. A petri plate with solidified Luria-Bertani medium containing kanamycin was pr e-warmed at 37C, and then the transformed TOP10 cells were spread evenly on the plate and incubated at 37C overnight. 26

PAGE 27

Single transformant E. coli colonies were selected and tran sferred to liquid LB medium containing kanamycin and incubated at 37C in a rotary shaker for at least 8 hours. The QIAprep Spin Miniprep Kit (QIAGEN #27106) was used to isolate plasmid DNA. The samples were transferred to 2 mL microcentr ifuge tubes and spun at 10,000 rpm for 1.5 minutes to pellet cells. The supernatant was discarded, and 750 mL lysis buffer was added to each sample, vortexed at maximum speed for 30 seconds, and then allowed to rest at room temperature for 3 minutes. The solution was applied to spin columns with co llection tubes and centrifuged at 14,000 rpm for 1 minute. The flow-through was discarded, and 750 mL wash buffer was applied. The samples were spun for 1 minute, flow-throug h discarded, and then spun one final time to remove residual wash buffer still in the spin column. The colu mns were transferred to fresh 1.5 mL collection tubes, 50 uL of elution buffer added to the filter s, then collected by centr ifugation for 1 minute at 14,000 rpm. To verify the presence of inserts, 10 uL of plasmid eluate from each sample was digested using 1 uL EcoR1, 2 uL ReAct3 buffer and 7 uL water. The reaction was incubated at 37C for 1 hour then visualized on a 0.8% agaros e gel as previously described. DNA Sequencing Reaction Plasmids containing correctly-sized inserts we re sequenced using M13 primers included in the TOPO TA cloning kit (see Table 2 for primer sequence). Inserts were sequenced using the Applied Biosystems BigDye Terminator v1.1 Cycle Sequencing Kit (Fischer #NC9008533) as follows: 2uL TOPO forward primer, 2uL plasmi d, 2uL BigDye and 4uL water were mixed and briefly centrifuged. The sequencing reaction cons isted of 3 minutes at 95C, followed by 25 cycles of 95C for 25 seconds, 50C for 15 sec onds, and 60C for 4 minutes. The samples were held at 10C until ready for processing us ing the QIAGEN DyeEx 2.0 Spin Kit (QIAGEN #63206). Once the sequencing reactions were dried, they were sent to the University of Florida 27

PAGE 28

Interdisciplinary Center fo r Biotechnology Research (ICBR) for processing by Sanger sequencing, after which FASTA and fluorescent waveform results were returned. Sequences were verified using BLAST searches against the National Center for Biotechnology Information (NCBI) nucleotide database. Absolute Quantitative PCR Fresh cDNA was diluted 10-fold and stored at -20C for analysis. Both an MJ Research PTC-200 and an Applied Biosystems 7300 real-t ime PCR machine were used in this study. Reactions were run according to the instruct ions supplied with the Finnzymes DYNAMO HS 410 SYBR-Green detection system (Fischer #50-995-143). A master-mix was prepared consisting of 10uL DYNAMO reagent, 5.6 uL H2O, 1 uL 5uM forward + reverse primer and 0.4 uL 50X ROX passive reference dye, each multiplied by n+1 reactions. Each well contained 17 uL of master mix, to which was added 3 uL of template. A standard curve was generated for each gene using amplicon-containing plasmids diluted on a 10-fold gradient from 107 through 102. The template was either 10-fold-diluted RT r eaction (50 ng RT reaction per well), a diluted plasmid standard, or water blank. The reaction components were identical for both machines. See Table 2 for primer annealing temperatures. The program for thermal cycling was kept constant with the exception of annealing temperature. First samples were warmed to 50C for 2 minutes, then denatured by 15 minutes at 95C. Amplification consisted of 40 cycles of denaturation (20 seconds at 95C), annealing (20 seconds at the primer-specific temperature), and extension (30 seconds at 72C). One last extension step (10 minutes at 72C) was r un at the end of each reaction. Finally a DNA dissociation curve was generated by ramping the thermocycler temperature from 55C to 95C with a plate read every one degree. Each sample plasmid standard and water blank was run in triplicate to minimize experimental error. 28

PAGE 29

Absolute qPCR data was entered into Microsoft Excel and triplicate results were averaged into single values for each sample. Biological replicates were averag ed and graphed including standard error calculations. 29

PAGE 30

30 Table 2-1. Components of 50 mL RNA isolation buffer. 1M Tris-HCl buffer pH 7 5 mL 100 mM 4M NaCL 2.5 mL 200 mM DL-Dithiothreitol (DTT) 40 mg 5 mM N-lauroyl sarcosine 500 mg 34 mM 0.5M Ethylenediamine tetraacetic acid 2 mL 20 mM H2O treated with 1% Diethyl pyrocarbonate (DEPC) 40.5 mL

PAGE 31

Table 2-2. List of qPCR primers 5 to 3 Gene Forward Reverse Ta (F) ZmInCW2 GGTGACCGGGATACAAACGGCACA GACAAATCCTGCAAATGTCGGGCG 60 ZmSu1 ACCAGAGGATGCAGTCTATG CCATTCCACTCTGACCAAACG 56 ZmSus2 ACTTTCCACATACCGAGAAGGCCA AAGGTTTACCAGCTCCCTCAGCTT 56 ZmHXK2 ATAGCAAGCAGAGGGAACTGGGTT AT CAGCATATCTCCCACCAGCCAA 60 ZmACS2 CCACAGCTCAAACAACTTCACCCT GTGCTCCGTGGCGAACCT 60 ZmACS6 TGCACTGCACGAGCGGCAA CGCTCCGTGGCGAACGC 60 ZmACS7 CTCGAACAACTTCACCCTCACCA CACCAGGTGGATGCCCTTGG 60 ZmERS1-14 ACTCGAGGATGGAAGCCTTGAACT TCTCCCGTCGGGCAGCAC 60 ZmETR2-9 GCTATGTATGTGTGAAATTTGAGATTAGGA CT CGTACAAATCTGAGGACGCTCCAG 58 ZmETR2-40 GCTATGTATGTGTGAAATTTGAGATTAGGA TCAAGTCTGAAGACGCCGCGGAGGAG 57 ZmACO20 CGCCGACGCCGTCATCTT TCCAC GATACACGCATAACCACCGT 60 ZmACO35 CGCCGACGCCGTCATCTT ACACACATAACTGTGCCACTATAAGCA 56 ZmEIL1-1 GCAGCAGCAGCAGTTCTTCATCC GTTTA TGGCTGGCCGGACATACAAGT 57 31

PAGE 32

CHAPTER 3 RESULTS The Mn1 and mn1 Genotypes Ethylene Production A key component of this study was the determin ation of ethylene generation in developing seeds. GC analysis was performed for two cons ecutive years, 2007 (Fig. 3-1) and 2008 (Fig. 32). Due to differences in kernel physiology, data was plotted on both a na nomols per gram fresh weight (nmol/g/hr) and nanomols per whol e kernel basis (nmol/kernel/hr). Figure 3-1A shows that for the 2007 cr op, on a per-gram fresh weight basis, mn1 produced more ethylene at all stages, with an early peak of ~70 nmol/g/hr at 8 DAP and a maximum rate of 135 nmol/g/hr at 29 DAP. The early peak fell to ~37 nmol/g/hr at 14 DAP before a linear rise to the 29 DAP maximum. The mn1 ethylene peak at 29 DAP was followed by a steady 85 nmol/g/hr rate through 36 DAP. Mn1 ethylene production also started with a peak at 8 DAP of over 45 nmol/g/hr, before dropping to half the in itial value, remaining constant up to 30 DAP before a sharp increase at 33 DAP that reached over 55 nmol/g/hr. At both 20 and 29 DAP the mn1 ethylene production was at least 3-fold hi gher, with an overall hormone peak in mn1 four days before that shown in Mn1 Figure 3-1B shows the same ethylene data calculated on a per kernel basis. This calculation method highlights the smaller kernel size of the mn1 genotype due to the miniature mutation. The Mn1 data still showed an initial drop between 8 DAP and 10 DAP, but the starting value of ~3.6 nmol/kernel/hr remained si milar between 12 and 20 DAP, reaching ~4 nmol/kernel/hr at this time. Th e ethylene levels at 33 DAP in Mn1 rose to over 17.5 nmol/kernel/hr, nearly double the amount in the mn1 kernels. The mn1 genotype was similar to Mn1 in that initial values of ~4 nmol/kernel/hr we re more stable before 20 DAP on a per kernel 32

PAGE 33

basis. The 29 DAP peak in mn1 was still present, but the abso lute differences at 20 and 29 DAP were only 50% higher in mn1 at both time points. It is important to note that where error bars are not shown, the data point was the result of a single biological sa mple and might not represent the genotype as a whole. Factors that led to the loss of replicate da ta included technical difficulties with the GC apparatus, poor seed set on pollinated ears, and human error. During 2008 (Fig. 3-2) Mn1 and mn1 samples were more abundant, allowing improved replicate coverage of each time point, especial ly between 8 and 20 DAP. Figure 3-2A shows that on a per-gram basis the Mn1 and mn1 hormone values were similar between 8 and 13 DAP. The Mn1 ethylene production started at a rate of ~90 nmol/g/hr at 8 DAP, followed by a linear decline to ~28 nmol/g/hr at 20 DAP. This ~28 nm ol/g/hr value was maintained for the remainder of observed time points. The mn1 line had an 8 DAP ethylene produc tion rate of ~70 nmol/g/hr, and after a 20% decrease at 11 DAP, ethylene re mained between 45 and 60 nmol/g/hr until 25 DAP, a level nearly twice that of Mn1 The mn1 samples exhibited a 4-fold decrease in ethylene production at 30 DAP versus 25 DAP When plotted on a per kern el basis (Fig. 3-2B), the Mn1 genotype produced more ethylene than the mn1 line at all points test ed. This was highlighted by a distinct peak of ethylene in Mn1 at 13 DAP (~11 nmol/kernel/hr). Hormone levels dropped to ~6.5 nmol/kernel/hr by 16 DAP and remained steady for the remaining time points, with a final measurement of ~4.5 nmol/kernel/hr at 34 DAP. Overall, ethylene production per mn1 kernel was reduced by 2-fold between 8 and 13 DAP when compared to the wild-type, which remained ~4 nmol/kernel/hr. The mn1 line showed a ~6 nmol/kernel/h r peak at 16 DAP followed by a gradual downward trend ending at ~4.7 nmol/k ernel/hr at 25 DAP. Ethylene levels in mn1 kernels were below 1 nmol/kernel/hr at 30 DAP, le ss than half of the wild-type (Fig. 3-2B). 33

PAGE 34

Transcript Accumulation in Mn1 and mn1 Genotypes In order to better understand the relationshi p between hormone production and kernel physiology, gene expression analys is was performed in order to quantify RNA levels of the genes related to sugar metabolism, ethylene bi osynthesis and ethylene perception. Transcript levels were investigated for the following genes: three metabolic genes Mn1 Sus2 and HXK2 three ACC synthase genes ACS2, ACS6 and ACS7, two ACC oxidase genes ACO20 and ACO35 three ethylene receptors ERS1-14 ETR2-9 and ETR2-40, and finally one tr anscription factor EIL1-1 Transcript levels were an alyzed using two biological re plicates from the 2007 field crop and three replicates from the 2008 crop. All qPCR data from the 2007 harvest were reported as the number of transcripts per nanogram total RNA (# transcripts/ng total RNA). The results from the 2008 crop were reported as both # transcripts/ ng total RNA and the number of transcripts per kernel (# transcripts/ kernel). In addition, ACO20 ACO35, ETR2-40 and EIL1-1 were cloned and included in the real-time PCR analysis for year 20 08, so year 2007 data for these genes is absent. Metabolic genes The causal basis of the mn1 seed phenotype is the loss of INCW2 enzyme activity, which is encoded by the Mn1 gene (Cheng et al. 1996).Quantifying this gene provides an internal control for RNA quality and reaction efficiency during sample preparation. Figure 3-3 shows the difference in Mn1 transcript levels in the Mn1 and mn1 lines. The Mn1 transcript was abundant between 8 and 13 DAP in Mn1 kernels and reduced by ~95% in the mn1 background for both 2007 and 2008 samples, correlating with timing of maximum enzyme activity (Chourey et al. 2006). Comparing absolute quantif ication between year 2007 and year 2008 (Fig. 3-3A and 33B, respectively), the 2008 data showed a 50% higher transcript abundance at 8 DAP, reaching ~27,000 transcripts/ng total RNA. At 12 DAP the 2008 samples produced ~9,200 transcripts/ng total RNA, 16% more than year 2007. The transcript levels at la ter stages were similar between 34

PAGE 35

the years, with ~3,000 transcripts/ng total R NA at 16 DAP and ~1,600 tr anscripts/ng total RNA at 20 DAP. In Figure 3-3C the tre nd was similar when calculated per kernel, with peak levels of Mn1 transcript occurring at 8 DAP at nearly 1.13 x10^9 transcripts/kernel. The only divergence from a steadily downward Mn1 expression level was at 13 DAP when # transcripts/kernel showed a slight rise before continuing to recede. A sucrose synthase gene Sus2 was chosen as a second intern al control (Fig. 3-4). The 2007 results showed very little difference between Mn1 and mn1 lines (Fig. 3-4 A), with levels between 600 and 800 transcripts/ng total RNA and sizable standard error at all time points. Figure 3-4B shows the 2008 Sus2 transcripts/ng total RNA consis tently increased over year 2007 amounts, starting with ~1,500 transcripts/ng total R NA at 8 DAP, afterwards increasing in both Mn1 and mn1 In the Mn1 line, Sus2 gene peaked at ~3,100 transc ripts/ng total RNA by 11 DAP then decreased 20% through 16 DAP before a rise to ~2,700 transcripts/ng total RNA at 20 DAP. The mn1 line showed a similar trend, with a later, reduced peak of ~2,500 transcripts/ng total RNA at 13 DAP, and a decline to ~1,500 transcripts/ng total RNA for 16 and 20 DAP. On a per kernel basis (Fig. 3-4C) the Sus2 transcript levels in Mn1 samples peaked at 13 DAP and remained unchanged thereafter, at ~3x10^8 transcripts/kernel. The mn1 samples produced a peak Sus2 transcript level at 11 DAP, two days earlier than Mn1 with a noticeable linear decline that ended at 1x10^8 transcripts/kernel at 20 DAP, a value 3-fold lower than Mn1 Hexokinase2 transcript levels (Fig. 35) were investigated because members of this gene family are reported to be essent ial components of suga r sensing and signaling pathways to this study (Saltman 1953; Jang et al. 1997 ; Yanigisawa et al. 2003). The HXK2 transcript levels from years 2007 and 2008 showed similarity in the number of transcript s/ng total RNA (Fig. 3-5A and B).Additionally, HXK2 transcript levels were consistently higher in Mn1 kernels than the mn1 35

PAGE 36

Figure 3-5A depicts a trend of decreasing expressi on from 8 to 20 DAP, with an initial transcript level of ~1,500 transcripts/ng total RNA shifting to a final value of ~550 transcripts/ng total RNA in Mn1 and mn1 samples. There was no statistically significant difference between the two genotypes tested for that year although average levels of HXK2 in mn1 samples only reached ~70% of the levels in Mn1 In Figure 3-5B the Mn1 value was increased 30% at 8 DAP compared to the same Mn1 time point for the previous year but the remaining results for Mn1 were similar. However, the year 2008 mn1 data at 8 DAP (~850 transcripts/ng total RNA) showed a 3-fold decrease in transcripts/ng tota l RNA versus Mn1 kernels. Figure 5-5B also shows that HXK2 transcript levels in mn1 from 13 to 20 DAP were between 50-60% of the related amounts in Mn1 Figure 3-5C shows two peaks of HXK2 transcript accumulation in Mn1 kernels, the largest at 8 DAP (~1.5x10^8 transcripts/ng total RNA) and another at 13 DAP. There is a 60% decrease overall from 8 to 20 DAP. The mn1 data showed a constant value for HXK2 of roughly 3x10^7 transcripts/kernel throughout development. Excluding 13 DAP, HXK2 transcripts/kernel were at least 3-fold lower in the mn1 line relative to the wild-type. The ACC synthase gene family The ACC synthases catalyze the rate-limiting step in ethylene biosynt hesis. Real-time PCR data for the maize ACC synthase family, ACS2, ACS6 and ACS7 are shown in Figures 3-6 through 3-8, respectively. The expression level of the family as a whole was low, with ACS2 the most abundant (up to 400 transc ripts/ng total RNA), followed by ACS7, and ACS6 appearing as low as 2 and 3 transcripts/ng total RNA (Fig. 3-8). Figure 3-6 reveals subtle variation in ACS2 transcript between Mn1 and mn1 lines, with consistently higher transcript levels in the Mn1 kernels at 8 DAP. The # of tran scripts/ng total RNA in the year 2007 samples (Fig. 3-6A) was approximately half the of those from year 2008 (Fig.3-6B) for both genotypes, but the trends were similar. The Mn1 kernels showed highest accumulation of ACS2 at 8 DAP (~400 36

PAGE 37

transcripts/ng total RNA) then dropped ove r 50% by 12 DAP. The decline continued, with lowest transcript levels at 20 DAP (~75 tr anscripts/ng total RNA) (Fig. 3-6B). The mn1 line had transcript levels of ~110 tran scripts/ng total RNA at 8 DAP, th en showed a relative peak 12-13 DAP. After maximum ACS2 transcript accumulation of 100 and 200 transcripts/ng in mn1 at 13 DAP (Fig. 3-6A and B respectively), ACS2 levels were reduced at 20 DAP. On a per-kernel basis (Fig. 3-6C) the trend for the ACS2 RNA profile in the Mn1 samples was more variable, but still showed an overall high-to-low progression between 8 and 20 DAP Upon calculation of transcript levels per kernel, the 13 DAP peak in mn1 samples became more prominent (Fig. 3-6). ACS6 transcripts/ng total RNA in the Mn1 line (Fig. 3-7A and B) had maxima at 8 DAP followed by progressively lower abundance thro ughout development, to a 20 DAP value of ~4 transcripts/ng total RNA. The mn1 line, for the year 2007 crop (Fig. 3-7A), showed peak levels of ACS6 transcript at 13 DAP, similar to that of ACS2. This trend was absent in the 2008 samples (Fig. 3-7B), which instead produced two plateau s, one 8-11 DAP (8 transcripts/ng total RNA) and the second 13-20 DAP (~3 transcript s/ng total RNA). Figure 3-7C shows ACS6 transcript on a per-kernel basis, which reve als a distinct peak in both Mn1 and mn1 lines that was not seen in terms of transcripts/ng total RNA. The Mn1 line showed no change in transcript amounts at 8 and 11 DAP, then rose to 1.07x10^6 transcripts/ kernel at 13 DAP. Following this increase, transcript levels fell to ~5x10^ 5 transcripts/kernel and remained constant to 20 DAP. In the mn1 line, ACS6 increased nearly 2-fold between 8 and 11 DAP, reaching 6.64x10^5 transcripts/kernel. After this 11 DAP peak, tr anscript levels dropped over 60% by 13 DAP and maintained a consistent level of accumulation through 20 DAP. The third ACC synthase ACS7 (Fig. 3-8), showed intermediate levels of transcription compared with the other two family members. Both transcript/ng and tr anscript/kernel showed 37

PAGE 38

little difference between Mn1 and mn1 genotypes with respect to ACS7. Figures 3-8A and B demonstrate maximum ACS7 transcript at 8 DAP for Mn1 (~60 transcripts/ ng total RNA) and mn1 (~40 transcripts/ng total RNA). Following 8 DAP transcript levels declined in a linear fashion to final values of ~8-10 transcripts/ng total RNA for both genotypes. Results from year 2007 and year 2008 samples are virtually identical, with similar absolute levels as well as trends. On a per kernel basis (Fig. 3-8C), Mn1 samples generated a slight peak of ACS7 transcript at 13 DAP. The mn1 samples produced maximum ACS7 transcript levels at 11 DAP, but the difference was not statistically significant. This transcript showed a trend of general decline from 8 to 20 DAP, and both genotypes ended at a fina l value of ~8x10^5 transcripts/kernel. The ACC oxidase gene family Because the ACC synthases catalyz e the first step in ethylene biosynthesis, it is important to asses the state of ACC oxidase transcripts in order to repres ent the final step in ethylene production. Figures 3-9 and 3-10 show results for ACO20 and ACO35 the two ACC oxidase genes investigated in this study. All ACO data are from the year 2008 crop. In Figure 3-9A, ACO20 transcript levels in the Mn1 genotype remained flat throughout all stages, with a slight decrease between 11 and 13 DAP from ~4,300 to ~3,400 transcripts/ng total RNA. In the same figure mn1 developed a peak of ACO20 transcript at 13 DAP of ~ 5,100 transcripts/ng total RNA, an increase of over 60% versus Mn1 The highest expression in mn1 was ~6,000 transcripts/ng total RNA at 20 DAP, nearly 2-fold higher than Mn1 Figure 3-9B demonstrates that on a per kernel basis there was a steady increase in ACO20 transcript from 8 to 13 DAP in both lines. From 13 to 20 DAP, ACO20 levels remained unchanged in the wild-type samples at 4x10^8 transcripts/kernel. The mn1 sample produced 60% of the ACO20 transcript relative to Mn1 at 16 DAP, but transcript levels were iden tical in the two lines at 20 DAP. 38

PAGE 39

The ACO35 transcript accumulation data (Fig. 310) showed similar expression in both Mn1 and mn1 Although Mn1 produced maximum transcripts/ng total RNA at 8 DAP, and mn1 produced an ACO35 peak at 13 DAP, both lines generate approximately 1,000 transcripts/ng total RNA for the first three time points, w ith similar decreases at 16 and 20 DAP. When calculated on a per kern el basis (Fig. 3-10B), ACO35 transcript levels were effectively identical between genotypes, with a clear pa rabolic peak centered on 13 DAP. The ethylene receptor gene family and EIL1-1 Ethylene is perceived by membrane-bound recepto rs, three of which have been included in this study: ERS1-14, ETR2-9 and ETR2-40. The fourth ethylene receptor in maize, ERS1-25 is expressed at a lower level than the others (Young and Gallie 2004) and attempts at cloning the gene were unsuccessful during this study. For year 2007 and 2008 samples, the three receptor genes ERS1-14, ETR2-9 and ETR2-40 were expressed to similar levels, with ETR2-9 slightly more abundant on a per ng total RNA basis (Figs. 3-11 through 3-13). For ERS1-14 in year 2007 samples (Fig. 3-11A) both Mn1 and mn1 genotypes produced similar transcript results, with peak accumu lation at 8 DAP (~560 transcripts/ng total RNA). After a drop to ~300 transcripts/ng total RNA at 12 DAP, subsequent time points remained consistent ending with ~250 transcripts/ng total RNA at 20 DAP. Results from the year 2008 samples (Fig. 3-11B) showed an increase of ~ 20% over year 2007 data. However, the trend for Mn1 kernels was similar, with an 8 DAP peak of 630 transcripts/ng total RNA followed by a 30% decline at 11 DAP, remaining constant thereafter. The mn1 kernels from year 2008 diverged from the previous years results, with 8 DAP ERS1-14 levels that started at ~530 transcript/ng, then increased 10% to peak at 11 DAP, generating 50% more transcript than Mn1 at that time point (Fig. 3-11B). After the 11 DAP peak, ERS1-14 levels in mn1 returned to wild-type levels at 16 and 20 DAP. The peak in ERS1-14 transcripts at 11 DAP was visi ble on a per kernel basis as 39

PAGE 40

well (Fig. 3-11C), marki ng the only point that mn1 demonstrated higher values than Mn1 Both genotypes showed identical plateaus from 13 to 20 DAP, though ERS1-14 transcripts/kernel were insignificantly lower in m n1-1 than Mn1 Figure 3-12 shows the transcript results for the most abundant receptor, ETR29. Year 2007 samples produced an identical profile of ETR2-9 generation between Mn1 and mn1 decreasing from ~860 transcripts/ng total RNA at 8 DAP to just over 500 transcripts/ng total RNA at 20 DAP (Fig. 3-12A). No statistically significant diffe rences were observed. Figure 312B shows data from the 2008 harvest on a per ng total RNA basis. A similar trend emerged as in the previous year; ETR2-9 transcript declined as development progressed. In addition, the accelerated decline of ETR2-9 transcript in mn1 samples was significant at 13 DAP, when mn1 kernels produced only 65% of the ETR2-9 transcript versus the wild -type. This observation is supported on a per kernel basis by the results sh own in Figure 3-12C. Both genotypes produced a similar 2-fold transcript increase from 8 to 11 DAP, after which the two lines diverged: The Mn1 samples generated ~8x10^7 transcripts/kern el at 13 DAP, roughly 3-fold higher than mn1 The wild-type kernels subsequently ma intained at least 50% higher ETR2-9 transcripts/kernel at 16 and 20 DAP. Figure 3-13A depicts year 2008 results for ETR2-40, showing a ~646 transcript/ng peak in the Mn1 kernels at 8 DAP that was not reflected in mn1 samples. Other than this difference, both genotypes produced no significant variation in am ounts of transcript, remaining between 300 and 350 transcripts/ng total R NA from 11 to 20 DAP. Per kernel results for ETR2-40 (Fig. 3-13B) showed an increase in both genotyp es from 8 to 11 DAP. At 13 DAP, the Mn1 samples showed increased transcript accumulation of ~4x10^7 transcri pts/kernel and maintained that level for the 40

PAGE 41

remaining time points. At 13 DAP the mn1 kernels decreased accumulation to ~2.23x10^7 transcripts/kernel through 20 DAP, 2-fold lower than Mn1 results. The transcription factors EIN3 and EIL1 are downstream transducers of ethylene perception. EIL1-1 was included in this study as a represen tative of this group of genes (Fig. 314). For both Mn1 and mn1 samples, EIL1-1 exhibited the trend of s howing lowest transcript levels at 8 DAP and steadily rising to a peak at the latest stage measur ed, 20 DAP. Figure 3-14A shows Mn1 remained constant from 8 to 16 DAP, ma intaining 740-780 transc ripts/ng total RNA. Over the same period mn1 samples produced a peak of EIL1-1 transcript accumulation at 13 DAP ~30% higher than the Mn1 kernels. Both lines eventually rose to their maximum values of 1,200 transcripts/ng total RNA at 20 DAP. On a per kernel basi s (Fig. 3-14B) the two lines produced similar increases in EIL1-1 transcript during development. From 13 to 16 DAP, Mn1 transcript production was unchanged (~ 9.2x10^7 transcripts/kernel) but the mn1 line decreased temporarily at 16 DAP to 4.89x10^7 transcripts/kern el. The subsequent in crease at 20 DAP was proportionally similar between genotypes, but the Mn1 value of 1.35x10^8 transcripts/kernel at 20 DAP was ~30% higher than mn1 at the same time point. The Su1 and su1 Genotypes Ethylene Production The Su1 and su1 genotypes were included in this study in order to provide data for a lateacting starch synthesis muta tion, as compared to the mn1 seed mutation that affects carbohydrate metabolism during the early stages of seed deve lopment. The ethylene pr oduction results for the Su1 and su1 genotypes were reported using the same convention as the Mn1 and mn1 data. Figure 3-15 shows ethylene producti on for the year 2007 samples. On a per gram fresh weight basis (Fig. 3-15A) both genotypes produced ~35 nmol/g/hr ethylene at 12 DAP, the highest amount in either line for all time points tested. There was a ~40% declin e in ethylene at 16 DAP 41

PAGE 42

in both lines. At 20 DAP su1 kernels produced 27 nmol/g/h r ethylene, 50% more than Su1 seeds. Both lines reached their lowest production rate, ~11 nmol/g/hr, at 28 DAP. Figure 3-15B shows results in terms of # nmol/kernel/hr, reveali ng a decrease in ethylene production for both genotypes from 8-16 DAP. The Su1 kernels produced a minor peak of ~5.5 nmol/kernel/hr at 12 DAP and a major peak of 8 nmol/kernel/hr at 24 DAP. The ethylene level in the su1 mutant at 12 DAP was similar to that in the wild-type, as wa s the subsequent decrease at 16 DAP and rise to ~7 nmol/kernel at 20 DAP. Due to technical difficulties the data for su1 kernels at 24 DAP were not recovered. At 28 DAP the re sults of 4.5 nmol /kernel/hr for Su1 and su1 kernels were similar, showing a downward trend at this stage in both lines. During 2008, the Su1 line grew abundantly, providing excellent coverage of all time points. The su1 mutant line, however, showed poor germination, allowing only one or two replicates per stage. Bars represent standard e rror of between two and f our biological replicates for Su1 and two replicates for su1. Figure 3-15A shows that th e highest level of ethylene production was at the earliest time point sampled for both Su1 (~45 nmol/g/hr, 8 DAP) and su1 (~40 nmol/g/hr, 12 DAP). Hormone levels dropped sharply in Su1 from 20 nmol/g/hr at 12 DAP to 10 nmol/hr at 20 DAP. There was a slight in crease of ethylene 20-25 DAP before production reached the observed minimum of ~2 nmol/g /hr from 28-30 DAP. This lowest value in Su1 kernels was followed by a rise to ~5 nmol/g/hr at 34 and 37 DAP. The su1 line produced a similar trend as Su1, decreasing 3-fold from 12 to 20 DAP. The su1 minimum value was slightly earlier than the wild-type but similar in rate (~2 nmol/g/hr). As in the Su1 samples, ethylene production showed an upward trend at 35 DAP. When considered on a per kernel basis (Fig. 316B), the Su1 kernels did not have a maximum at 8 DAP but instead produced distinct peaks above 4 nmol/kernel/hr at 16 and 24-25 DAP. Th e timing of the later peak corresponded well 42

PAGE 43

with 2007 observations (Fig. 3-15B and 3-16B). Later development s howed similar trends as the per gram results, with a minimum at 28-30 DAP followed by a ~2-fold increase at 34-37 DAP. The su1 mutant kernels generated more ethylene th an wild-type at 12 and 14 DAP but fell more sharply at 20 and 25 DAP, preceding both the initial peak and initial declin e of the wild-type by ~5 days. The su1 samples showed an increasing trend late in development that resembled Su1 kernels, and reached a 2 nmo l/kernel/hr rate at 35 DAP. Transcript Accumulation in Su1 and su1 Genotypes Due to difficulties in generating complete time courses for Su1 and su1 genotypes, each of the data points representing transcript levels we re derived from single biological samples. Less emphasis will be placed on slight differences in transcript levels because of the questionable reproducibility of the results. Error bars repres ent experimental error for the real-time PCR reactions. Metabolic genes The Su1 transcript was used as an internal cont rol for RNA quality and reaction efficiency during qPCR sample preparation. This was due to the published similarity in S u1 transcript levels in Su1 and su1 genotypes (Dinges et al. 2001). Samples from each year demonstrated little difference in transcript level between mutant and wild-type lines on a per ng total RNA basis (Fig. 3-17A and B). The highest observed transcript levels were at the 12-14 DAP stage for both years followed by a 3-fold decline as development progressed. In the year 2007 samples, 12 DAP results showed ~8x10^3 transcripts/ng total RNA, and there was a second minor peak at 20 DAP (~5.5x10^3 transcripts/ng to tal RNA) before the downward trend resumed, which caused the Su1 and su1 transcript levels to mirror trends in ethylene production (Fig. 3-16). For year 2008 samples, # transcripts on both a per ng tota l RNA and per kernel basis dropped steadily from early to late development. A delay in transcript reduction in the Su1 genotype caused 2-fold 43

PAGE 44

higher accumulation at 20 DAP over levels in the su1 mutant. Absolute amounts of Su1 transcript were shown to be three-fold lower in the 2008 crop, which produced from 2,500 to 500 transcripts/ng total RNA between 14 and 35 DAP. The gene Sus2 encodes a member of the sucrose s ynthase family, which catalyzes the unidirectional cleavage of su crose into fructose and UDP-glucose. In Figure 3-18, Sus2 transcript accumulation is shown to vary between year 2007 and 2008 samples. In the Su1 kernels, for year 2007 (Fig. 3-18A), Sus2 transcript was lowest at 8 DAP then had a peak of ~1,700 transcripts/ng total RNA at 12 DAP. After this time point, tr anscript levels dropped 40% at 16 DAP then reached their maximum level of 2,500 tran scripts/ng total RNA at 20 DAP. Finally, Sus2 levels declined to ~1,400 transcript s/ng total RNA at 28 DAP. The su1 kernels showed an upward trend at 8 and 12 DAP, similar to wild-type, but ge nerated a peak 4 days earlier at 16 DAP (~2,250 transcripts/ng total RNA). After this peak, Sus2 levels remained between 1,500 and 1,800 transcripts/ng total RNA for the rest of development. The Su1 and su1 samples from year 2008 showed similar trends when compared by # transcripts/ng total RNA as well as # transcripts/kernel (F ig. 3-17 B and C). In Su1 seeds the Sus2 transcript level was elevated 2-fold over that of su1 samples. The trend for Sus2 showed consistent levels of accumulation throughout development. Su1 kernels maintained between 2,000 and 2,500 transcripts/ng total RNA, and su1 produced over 1,000 transcripts/ng total RN A at all time points. Figure 3-17C shows the same 2-fold increase in Sus2 transcript in the wild-type samples (~2x10^8 transcripts/kernel) over those in the su1 mutant (~1x10^8 transcripts/ke rnel) at all stages. At 35 DAP, the Su1 sample reached a peak of 2.67x10^8 Sus2 transcripts/kernel. Another sugar-related transcript, HXK2 showed consistent patterns of accumulation between years and genotypes. The highest transcript accumulation, 1,500-1,800 transcripts/ng 44

PAGE 45

total RNA, was from 12-15 DAP, and decreased as development progressed (Fig. 3-19A and B). For year 2007 samples, there was a 2.5fold higher peak at 8 DAP in the su1 sample that was not present during the following year. Figure 3-19C shows that, on a per kernel basis, both genotypes underwent steady decline in HXK2 transcript levels from an initial peak of ~1.5x10^8 transcripts/kernel at 14 DAP. The Su1 kernels showed a 2-fold higher HXK2 transcript level than su1 kernels at 20 DAP. By 25 DAP the two genot ypes contained similar amounts of transcript (~5x10^7 transcripts/kernel) that was unchanged for the remainder of development. The ACC synthase gene family The ACC synthase enzymes catalyze the first committed step in ethylene biosynthesis (Yang and Hoffman 1984). The ACC synthase gene fa mily retained a similar distribution as that seen in Mn1 and mn1 kernels, in that ACS transcripts were comprised mainly of ACS7 and ACS2, followed by a 2-fold less ACS6 (Figs. 3-20 to 3-22). Figure 3-20A contains year 2007 data for ACS2, showing peaks of 100 transcripts/ng total RNA at 12 and 20 DAP in the Su1 line. Between these peaks were intervening lows of ~25 transcripts/ng total RNA at 16 and 28 DAP. The su1 samples were expressed at ~35 transcri pts/ng total RNA between 12 and 20 DAP. The maximum expression levels were at the first and last stages test ed; first over 450 transcripts/ng total RNA at 8 DAP and last 3-fold lower at 28 DAP. The kernels from year 2008 (Fig. 3-20B and C) showed reduced ACS2 transcript levels compared to year 2007, with expression under 50 transcripts/ng total RNA for 14 and 20 DAP. Both genotypes produced rising le vels of transcript as development progressed. Su1 kernels produced 2-fold more ACS2 at 25 and 35 DAP. The maximum transcript levels were at 35 DAP, of 150 transcripts/ng total RNA for Su1 kernels and 72 transcripts/ng total RNA for su1. ACS6 was the least-expressed member of this ge ne family (Fig. 3-21). All three plots of transcript accumulation showed highe st levels before 15 DAP for both Su1 and su1 lines. During 45

PAGE 46

year 2007 (Fig. 3-21A) Su1 seeds produced less than 4 transcri pts/ng total RNA at 8 DAP, 6-fold lower than su1 All Su1 results from this year showed less than 10 transcripts/ng total RNA, with a steadily decreasing tr end after 12 DAP. The su1 line produced a clear maximum ACS6 transcript accumulation at 8 DAP, then showed consistently lower amounts from 12-28 DAP (~6 transcripts/ng total RNA). The year 2008 samp les showed highly sim ilar distributions of transcript in both Su1 and su1 kernels (Fig. 3-21B). The highest levels for both lines were again at the earliest point assessed; 20-25 transcripts/ng total RNA at 14 DAP. ACS6 accumulation dropped to ~8 transcripts/ng total RNA at 25 DAP, and continued at this level through 35 DAP. When calculated on a per kernel basis (Fig. 3-21C), the Su1 genotype maintained a high level of ACS6 transcript at 14 and 20 DAP, ~ 2.5x10^6 transcripts/kernel. In contrast, su1 kernels showed a 2-fold decrease of ACS6 transcript between 14 and 20 DAP. For each time point from 25 to 35 DAP, both lines generated ~7x10^5 transcripts/kernel. The third ACC synthase gene, ACS7, was the highest-expressed member of the family in Su1 and su1 genotypes (Fig. 3-22). For year 2007, result s indicated constant expression, between 100 and 200 transcripts/ng total RNA, from 12 through 28 DAP (Fig. 3-22A).In su1 samples there was a 2-fold higher accumulation at 8 DAP, and in Su1 samples the transcript level dropped sharply at 28 DAP (24 transcripts/ng total RNA). Figure 3-22B shows that the 2008 samples had lower absolute levels of ACS7 transcript than year 2007, and maintained 50-125 transcripts/ng total RNA for all time points. These samples produced no peaks, which led to constant ACS7 levels in both genotypes from 14 through 35 DAP. When calculated on a per kernel basis (Fig. 3-22C), both Su1 and su1 genotypes had highest ACS7 accumulation at 14 DAP, greater than 1x10^7 transcripts/kernel. Both lines transcript levels fell to ~6x10^6 transcripts/kernel at 20 DAP. Su1 samples maintained this amount until 30 DAP, then produced a 46

PAGE 47

peak of 1x10^7 transcripts/kernel at 35 DAP. The su1 line had a slight downward trend from 20 to 35 DAP, ending at 5x10^ 6 transcripts/kernel. The ACC oxidase gene family ACC oxidase enzymes catalyze the final step in ethylene biosynthesi s. Figure 3-23 shows transcript levels of ACO20 in the year 2008 field samples. Both Su1 and su1 genotypes produced ~1,200 transcripts/ng total RNA at 14 DAP (Fi g. 3-23A). There was a peak of nearly 2,000 transcripts/ng total RNA in su1 samples at 20 DAP, a 50% higher transcript accumulation than the Su1 line. At 25 DAP both genotypes returned to parallel levels of 1,600 transcripts/ng total RNA. After this time point there was a decrease in ACO20 transcript levels in all samples that ended with 1,000 and 1,200 transcript s/ng total RNA at 35 DAP for Su1 and su1, respectively. When data was calculated on a per kernel basis, both lines produced ~1.2x10^8 transcripts/kernel at all stages, with the exception of 20 DAP peaks of 1.83x10^8 transcripts/kernel in Su1 and 1.5x10^8 transcripts/kernel in su1 (Fig. 3-23B). For the ACO35 gene (Fig. 3-24), transcript levels in Su1 kernels increased from 500 transcripts/ng total RNA between 14 and 20 DAP to a peak of ~1,750 transcripts/ng total RNA by 35 DAP. The su1 line also produced ~500 transcripts/ ng total RNA from 14 to 20 DAP, but had a smaller increase over time, reaching ~ 900 transcripts/ng total RNA at 30 and 35 DAP. When calculated on a per kernel basis, the trend remained consistent, with increasing transcript accumulation from 5x10^7 transcripts/kernel at 15 DAP to 7.5x10^7 transcripts/kernel at 30 DAP (Fig. 3-24B). From 30 to 35 DAP the Su1 line produced a 2-fold higher accumulation of ACO35 transcript. However, in the su1 line transcript levels remained constant from 30 to 35 DAP, 65% lower than wild-type. 47

PAGE 48

The ethylene receptor gene family and EIL1-1 Ethylene receptors are the ini tial point of ethylene percepti on in plants. Three ethylene receptor genes, ERS1-14 ETR2-9 and ETR2-40, were quantified in the Su1 and su1 developmental series. Figure 3-25 shows ERS1-14 levels for kernels from both 2007 and 2008 field harvests. For year 2007 samples (Fig. 3-25A) accumulation was between 200 and 360 transcripts/ng total RNA for both genotypes th roughout development, with the exception of su1 kernels, which contained over 800 transcripts/ ng total RNA at 8 DAP. Data from year 2008 samples (Fig. 3-25B) showed similar amounts of ERS1-14 accumulation as the previous year. Both Su1 and su1 series experienced a 50% decrease in transcript amount from 14 to 35 DAP. The Su1 line produced ~450 transcripts/ng tota l RNA at 8 DAP; 40% more than su1 kernels at the same stage. This 40% higher ERS1-14 transcript accumulation occu rred at all time points except 30 DAP, at which point the results for the two genotypes were id entical. Figure 3-25C shows that the Su1 series produced a peak of 4.5x10^7 transc ripts/kernel at 20 DAP and a low of 1.5x10^7 transcripts/kernel at 30 DAP, before a fi nal increase to 3x10^7 transcripts/kernel. In su1 kernels the ERS1-14 transcript levels were highest at 8 DAP, accumulating to 3x10^7 transcripts/kernel. By 20 DAP abundance had dr opped to 1.6x10^7 transc ripts/kernel, 3-fold lower than wild-type, and rema ined constant through 35 DAP. As in the Mn1 and mn1 samples, ETR2-9 was the most abundant re ceptor transcript, with up to 10-fold higher levels than the other two fa mily members (Fig. 3-26). The results from the year 2007 harvest resembled the figure for ethyle ne production (Figs.3-15 and 3-26A), with two peaks clearly defined in both genotypes; ~800 transcripts/ng total RNA at 12 DAP and ~650 transcripts/ng total RNA at 20 DAP. Both lines showed decreased accumulation of ETR29 transcript at 28 DAP (~325 tr anscripts/ng total RNA). For year 2008 samples (Fig. 3-26B), transcript levels were 5-fold highe r than those from the previous year. ETR2-9 decreased in 48

PAGE 49

nearly linear fashion throughout development of both genotypes, from over 3,500 transcripts/ng total RNA at 14 DAP to ~2,000 transcripts/ng tota l RNA at 35 DAP. Figure 3-26C demonstrates peaks of ETR2-9 in the Su1 samples at 20 and 35 DAP, similar to ERS1-14 transcript (Fig. 325C). Levels in Su1 kernels were 2-fold higher than su1 kernels at 20 DAP. Results for ETR2-9 transcript levels in su1 kernels showed highest levels at 14 DAP (3.5x10^8 transcripts/kernel), which decreased to 2x10^8 transcripts/kernel at 20 DAP and ~1.3x10^8 tran scripts/kernel at 35 DAP. For ETR2-40 (Fig. 3-27), the data did not correlate to the other two receptor trends, with transcript levels in Su1 samples higher at 14 DAP but lower at 30 DAP as a portion of total RNA (Fig. 3-27A). ETR2-40 transcripts in Su1 kernels decreased in a linear fashion from ~175 transcripts/ng total RNA at 14 DAP to ~80 tr anscripts/ng total RNA at 30 DAP. There was an upward trend at 35 DAP ( ~140 tr anscripts/ng total RNA). The su1 line displayed an increasing level of ETR2-40 transcript from 100 transcripts/ng to tal RNA at 14 DAP to ~145 transcripts/ng total RNA at 30 and 35 DAP. Per kernel (Fig. 3-27C), ETR2-40 transcript levels displayed the same peak in 20 DAP Su1 kernels that was seen for ERS1-14 and ETR2-9 transcripts. This peak reached ~2x10^7 transcripts/kernel and was follo wed by a decrease in transcript accumulation at 30 DAP, as well as a secondary peak at 35 DAP th at paralleled trends in the other receptor transcript levels. The su1 results for ETR2-40 showed an overall increase during development, from below 1x10^7 transcripts/ke rnel during early stages to ~1.2x10^7 transcripts/kernel by 35 DAP. Both ERS1-14 and ETR2-9 decreased during the same period (Figs. 3-25C and 3-26C). Levels of EIL1-1 transcript from year 2008 sample s are shown in Figure 3-28. For Su1 kernels, transcript levels were between 700 and 800 transcripts/ng total RNA at 14, 20, and 30 DAP (Fig. 3-28A). These points were separated by peaks of ~1,250 transcripts/ng total RNA at 49

PAGE 50

25 and 35 DAP. For su1 samples, there was a decrease from ~1,140 transcripts/ng total RNA at 14 DAP to ~860 transcripts/ng to tal RNA at 25 DAP. By 30 DAP EIL1-1 transcript levels peaked in su1 at ~1,200 transcripts/ng total RNA, a nd produced a similar amount at 35 DAP as well. Upon calculation of transcript amounts per kernel, observations remained unchanged. Figure 3-28B shows that EIL1-1 accumulation in Su1 kernels increased after 14 DAP to a plateau of ~1x10^8 transcripts/kernel from 20-25 DAP. This amount was decreased 50% at 30 DAP, then rose to 1.5x10^8 transcri pts/kernel at 35 DAP. In su1 samples the maximum transcript accumulation was over 1x10^8 transcripts/kernel at 14 and 35 DAP, with a minimum value, ~6.3 transcripts/kernel, at 25 DAP. 50

PAGE 51

Figure 3-1. Ethylene produced by Mn1 (blue) and mn1 (pink) kernels, nmols of A) nmol/g/hour B) nmol/kernel/hour (Summer 2007 field harvest) 51

PAGE 52

Figure 3-2. Ethylene produced by Mn1 (blue) and mn1 (pink) kernels, nmols of A) nmol/g/hour B) nmol/kernel/hour (Summer 2008 field harvest) 52

PAGE 53

Figure 3-3. Mn1 transcript levels in Mn1 (blue) and mn1 (pink) kernels A) transcripts per nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008 samples) C) transcripts per kernel (Summer 2008 samples) 53

PAGE 54

Figure 3-4. Sus2 transcript levels in Mn1 (blue) and mn1 (pink) kernels A) transcripts per nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008 samples) C) transcripts per kernel (Summer 2008 samples) 54

PAGE 55

Figure 3-5. HXK2 transcript levels in Mn1 (blue) and mn1 (pink) kernels A) transcripts per nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008 samples) C) transcripts per kernel (Summer 2008 samples) 55

PAGE 56

Figure 3-6. ACS2 transcript levels in Mn1 (blue) and mn1 (pink) kernels A) transcripts per nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008 samples) C) transcripts per kernel (Summer 2008 samples) 56

PAGE 57

Figure 3-7. ACS6 transcript levels in Mn1 (blue) and mn1 (pink) kernels A) transcripts per nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008 samples) C) transcripts per kernel (Summer 2008 samples) 57

PAGE 58

Figure 3-8. ACS7 transcript levels in Mn1 (blue) and mn1 (pink) kernels A) transcripts per nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008 samples) C) transcripts per kernel (Summer 2008 samples) 58

PAGE 59

Figure 3-9. ACO20 transcript levels in Mn1 (blue) and mn1 (pink) kernels (Summer 2008 samples) A) transcripts per nanogram RNA B) transcripts per kernel 59

PAGE 60

Figure 3-10. ACO35 transcript levels in Mn1 (blue) and mn1 (pink) kernels (Summer 2008 samples) A) transcripts per nanogram RNA B) transcripts per kernel 60

PAGE 61

Figure 3-11. ERS1-14 transcript levels in Mn1 (blue) and mn1 (pink) kernels A) transcripts per nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008 samples) C) transcripts per kernel (Summer 2008 samples) 61

PAGE 62

Figure 3-12. ETR2-9 transcript levels in Mn1 (blue) and mn1 (pink) kernels A) transcripts per nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008 samples) C) transcripts per kernel (Summer 2008 samples) 62

PAGE 63

Figure 3-13. ETR2-40 transcript levels in Mn1 (blue) and mn1 (pink) kernels (Summer 2008 samples) A) transcripts per nanogram RNA B) transcripts per kernel 63

PAGE 64

Figure 3-14. EIL1-1 transcript levels in Mn1 (blue) and mn1 (pink) kernels (Summer 2008 samples) A) transcripts per nanogram RNA B) transcripts per kernel 64

PAGE 65

Figure 3-15. Ethylene produced by Su1 (blue) and su1 (pink) kernels, nmols of A) nmol/g/hour B) nmol/kernel/hour (Summer 2007 field harvest) 65

PAGE 66

Figure 3-16. Ethylene produced by Su1 (blue) and su1 (pink) kernels, nmols of A) nmol/g/hour B) nmol/kernel/hour (Summer 2008 field harvest) 66

PAGE 67

Figure 3-17. Su1 transcript levels in Su1 (blue) and su1 (pink) kernels A) transcripts per nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008 samples) C) transcripts per kernel (Summer 2008 samples) 67

PAGE 68

Figure 3-18. Sus2 transcript levels in Su1 (blue) and su1 (pink) kernels A) transcripts per nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008 samples) C) transcripts per kernel (Summer 2008 samples) 68

PAGE 69

Figure 3-19. HXK2 transcript levels in Su1 (blue) and su1 (pink) kernels A) transcripts per nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008 samples) C) transcripts per kernel (Summer 2008 samples) 69

PAGE 70

Figure 3-20. ACS2 transcript levels in Su1 (blue) and su1 (pink) kernels A) transcripts per nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008 samples) C) transcripts per kernel (Summer 2008 samples) 70

PAGE 71

Figure 3-21. ACS6 transcript levels in Su1 (blue) and su1 (pink) kernels A) transcripts per nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008 samples) C) transcripts per kernel (Summer 2008 samples) 71

PAGE 72

Figure 3-22. ACS7 transcript levels in Su1 (blue) and su1 (pink) kernels A) transcripts per nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008 samples) C) transcripts per kernel (Summer 2008 samples) 72

PAGE 73

Figure 3-23. ACO20 transcript levels in Su1 (blue) and su1 (pink) kernels (Summer 2008 samples) A) transcripts per nanogram RNA B) transcripts per kernel 73

PAGE 74

Figure 3-24. ACO35 transcript levels in Su1 (blue) and su1 (pink) kernels (Summer 2008 samples) A) transcripts per nanogram RNA B) transcripts per kernel. 74

PAGE 75

Figure 3-25. ERS1-14 transcript levels in Su1 (blue) and su1 (pink) kernels A) transcripts per nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008 samples) C) transcripts per kernel (Summer 2008 samples) 75

PAGE 76

Figure 3-26. ETR2-9 transcript levels in Su1 (blue) and su1 (pink) kernels A) transcripts per nanogram RNA (Summer 2007 samples) B) transcripts per ng RNA (Summer 2008 samples) C) transcripts per kernel (Summer 2008 samples) 76

PAGE 77

Figure 3-27. ETR2-40 transcript levels in Su1 (blue) and su1 (pink) kernels (Summer 2008 samples) A) transcripts per nanogram RNA B) transcripts per kernel 77

PAGE 78

78 Figure 3-28. EIL1-1 transcript levels in Su1 (blue) and su1 (pink) kernels (Summer 2008 samples) A) transcripts per nanogram RNA B) transcripts per kernel

PAGE 79

CHAPTER 4 DISCUSSION The seeds produced by Zea mays are important sources of energy; for human consumption, animal feed, sugar production, and recently as a source of et hanol. Given the importance of cereal crops as a whole, research dealing with seed development is valuable not only for basic science but also for many social and economic purposes. The large size of the Zea mays kernel makes it a good model system for cereals. Much work has already been done to unravel the processes involved in growth and starch biosynthesis in maize seeds. Many metabolic path ways are established, fr om sucrose and hexose import to starch and protein synthesis (Nelson and Pan 1995; Shewry and Halford 2002; James et al. 2003). Also, the basic developmental program has been resolved, providing an outline of cell division, expansion, endoredupli cation, embryo development, st arch and storage protein synthesis, and finally programmed cell death of the endosperm (reviewed in Lopes and Larkins 1993; Young and Gallie 2000a). Now the most pressi ng questions are related to regulation and timing of these processes. Specifi cally, sugar and hormone interactio ns have emerged as critical control elements of cereal seed development. Th e purpose of this study is to observe the effects of two established maize kernel metabolism mutations, mn1 and su1, on ethylene hormone production and related transcript levels. This should provide a basis for comparison relating sugar metabolism and hormone effects during kernel development. The investigation of interact ions between sugars and phytoh ormones is complicated by the fact that multiple pathways can mutually influence the processes under consideration. This phenomenon, commonly referred to as crosstalk, allows slight m odifications in physiology to have far-reaching consequences and effects. Th is also causes difficulty in determining the difference between direct and indirect intera ctions. Furthermore, hormone effects can be 79

PAGE 80

inconsistent depending on concentrations, tissue localization, and developmental stages. For example, what holds true for germination or seedling development might be irrelevant or reversed in root meristems (Moore et al. 2003; Arteca and Arteca 2008; Gallie et al. 2009). This report demonstrates the variable natu re of ethylene production between different genetic backgrounds and developmentally altered kernel genotypes. A consistent theme is the observance of periodic bursts of ethylene production that could be related to transition through stages of kernel development. Ethylene Accumulation in Developing Seeds The observation of two peaks of ethylene produ ction in developing maize kernels has been well-documented over the last 12 years (Young et al 1997; Young and Gallie 2000a; Young and Gallie 2000b). However, the specific timing, relativ e amplitude, and absolute levels of ethylene biosynthesis have proven variab le depending on the genetic b ackground analyzed. Young et al. (1997) suggest a partial explana tion for this occurrence lies w ith differences in kernel mass between genotypes. In this study ethylene values were reported as a function of kernel weight and also kernel number, allowing for comparison of kernels both as biological units and as masses of tissue. In keeping with published data, discussion will focus on nmoles/kernel/hr measurements Levels of ethylene accumulation were quantified for four genotypes during two consecutive years of field planting and harvest, generating eight individual sample series. Each of the eight series produced some form of p eak of ethylene production before 20 DAP on a per kernel basis. The sizes of these peaks were be tween 4 and 10 nmols/kernel/hr, which is in good agreement with published values and timing (Young et al 1997; Young and Gallie 2000a; Young and Gallie 2000b). 80

PAGE 81

Mn1 Seeds Produced a Distinct Peak of Ethylene between 12 and 14 DAP Analysis of the results from Mn1 kernels indicate that this genotype produced a peak of ethylene between 12 and 14 DAP. This is in line with the earlie st recorded bursts of ethylene evolution in published reports (Young et al. 1997; Gallie and Young 2000a). Data for the Mn1 genotype for year 2007 ethylene production pa ralleled the rates from the Il451b genetic background depicted in Young et al (1997). Hormone results of the Mn1 kernel analysis from year 2008 more closely resembled the pattern of the Oh43 results published in that same report, which generated nearly 3-fold more ethylene th an the year 2007 samples before 20 DAP. This current investigation shows that, while early ethylene biosynthesis was more subtle in year 2007 samples versus the following year, the later peak at 33 DAP was more well-defined (Fig. 3-1). This is in contrast to the st eady 5-7 nmol/kernel/hr of ethylene produced after 16 DAP in the year 2008 samples (Fig. 3-2B). It is possible that th e 33 DAP sample from year 2007 was an anomaly since the value was derived from a single biological sample. It is also possible that sampling for year 2008 ended before a clear second developmen tal peak of ethylene biosynthesis could be established, given that in most previous reports, ethylene levels remained low at 32 DAP before a second peak was resolved (Young et al. 1997; Young and Gallie 2000b). Finally, differences in the technical aspects of ethylene measurem ent could have varied between years. Trends in mn1 Kernel Ethylene Production Were Varied Over Two Consecutive Years As in the Mn1 results, the mn1 kernels produced a major late peak of ethylene for year 2007 samples and a dominant early peak for y ear 2008 samples (Figs. 3-1 and 3-2). The mn1 genotype generated levels in the 4 nmol/kernel/hr range prior to 16 DAP during both years. However, year 2007 samples produced an apparent peak at 29 DAP more than 2-fold higher than levels at 12 DAP. As mentioned previ ously, the last three data points for mn1 kernels from year 2007 were all single replicates, rais ing the possibility of aberrant resu lts. Still, this pattern is in 81

PAGE 82

agreement with previously published reports (Young et al. 1997), although the strongest examples of late-stage ethyl ene production are in high-s ugar mutant lines such as sh2, su1 and su1se1 not sugar-deficient genotypes such as mn1 For year 2008 samples, mn1 seeds exhibited a maximum rate of ethylene generation at 16 DAP roughly four days after maximal levels of INCW2 activity and ethylene peak in the wild -type kernels (Cheng et al. 1996; Fig. 3-2B). Ethylene production rates remained constant until a sharp decline at 30 DAP, as reported previously. The period from 16 to 25 DAP showed that, although ethylene production in mn1 kernels was lower than wild-typ e, the smaller kernels produced nearly 2-fold higher ethylene levels on a per gram fresh weight basis. This rate was maintained throughout the period associated with initiation and progression of starch loading in wild-type kernels. Increased endogenous ethylene production as we ll as treatment of wild-type ke rnels with ethylene has been shown to accelerate cell death in developing maize seeds (Young et al. 1997; Young and Gallie 2000b). Neither Mn1 nor mn1 lines showed any periods of sharp decline in ethylene levels that would clearly separate distinct peaks (Figs. 3-1 and 3-2). This relatively constant level of production is not unique when considered agains t previously published data, but even maize genotypes with high ethylene produ ction rates exhibit substantial transient reductions in ethylene levels between 20 and 32 DAP (Young et al. 1997; Young and Gallie 2000b). The lack of mn1 material in the field prevented sampling past 32 DAP; a stage that freque ntly involves increased hormone levels (Young et al. 1997). It is possibl e that 30 DAP represents the effective endpoint of mn1 kernel development. Hexose Deficiency and Increased Sucrose Lead to Pleiotropic Effects in Maize Seeds While the significance of rising and falling et hylene levels as developmental cues is unknown, and sensitivity of different tissues to ethylene is another important factor, a 82

PAGE 83

consistently high level of ethylen e has traditionally been thought of as a stress signal promoting cell death in this tissu e and stage of growth (reviewed in Young and Gallie 2000a). Auxin is shown to promote ethylene biosynthesis in va rious systems (Arteca and Arteca 2008). The auxin IAA, and IAA-conjugates, have been shown to increase sharply in maize kernels between 9 and 11 DAP, coinciding with reduced cytokinin (Lur and Setter 1993; LeClere et al. 2008). It is possible that the substantial rise in auxin levels 9-11 DAP contri buted to the timing of the early ethylene peak in wild-type kernels. From this it follows that the auxin-deficient mn1 line could have delayed ethylene biosynthesis at this critical stage of kern el development (LeClere et al. 2008). In addition, Li et al. (2008) report that beginning at 12 DAP and extending through 20 DAP, mn1 kernels exhibit increa sed sucrose levels versus the w ild-type. Combined with data from Young et al. (1997) that demonstrated in creased ethylene levels in conjunction with increased sucrose levels in sh2 and su1se1 mutant lines, it is possible that a relationship exists between the hexose/sucrose ratio, auxin, and ethy lene levels in developing maize kernels. Previous studies of Arabidopsis suggest complex interactions between sugars and phytohormones, supporting the possibility of this relationship in maize kernels (Zhou et al. 1998; Cheng et al. 2002; Leon and Sheen 2003; Moore et al. 2003; Yanigisawa et al. 2003; Gibson 2004) Su1 and su1 Kernels Displayed Inconsistent Ethy lene Production over Two Consecutive Years According to Creech (1965) the su1 mutation leads to 2-fold hi gher sucrose than wild-type by 16 DAP. At this same stage, water-soluble polysaccharides (WSP) are increased nearly 4fold. Despite differences in sugar and st arch content during early development, Su1 and su1 ethylene production in this report was parallel until 20 DAP (Figs. 3-15 and 3-16) and has been reported to be similar until 32 DAP (Young et al. 1997) The data presented in this report showed 83

PAGE 84

a parallel relationship between Su1 and su1 at most stages of development (Figs. 3-15 and 3-16). However, the timing and relative amplitude be tween first and second ethylene peaks were dissimilar when comparing results from year 2007 a nd year 2008 GC analysis. This could be the result of different environmental conditions between years, leading to vari ations in stress factors such as drought and heat. It is possible that improved handling of sa mples during the year 2008 harvest increased the accuracy of the analysis. However, the range of ethylene production from year 2007 is higher than th at of year 2008 for the Su1 and su1 kernels (Figs. 3-15 and 3-16), while the opposite is true of the Mn1 and mn1 samples (Figs. 3-1 and 3-2), making it difficult to attribute changes between y ears to systemic effects. Su1 and su1 kernels showed two clear peaks of ethylene production prior to 30 DAP The differences between Su1 and su1 ethylene levels for year 2007 field samples were statistically insignificant (Fig. 3-15). Technica l difficulties with the GC apparatus prevented collection of data for su1 kernels at 28 DAP. The highe st levels of hormone in Su1 kernels were at 12 and 24 DAP, with a small decline during th e intervening 12 days (Fig. 3-15B). Cumulative ethylene exposure is measured in part by ethyl ene receptor degradation in climacteric fruit (Kevany et al. 2007). Whether or not this paradigm holds true in maize kernels is unknown. It is important to note that the 12-DAP peak was in the 5 nmol/kernel range, and the later Su1 peak reached 8 nmol/nL. This is in good agreement with Mn1 and mn1 data concerning physiologically relevant concentrations. However, the amount of ethylene required to trigger a developmental responses is unknown. In addi tion, ethylene percepti on and signaling are mediated by many factors downstream of ethylene production, and therefore inferring physiological effects based solely on hormone da ta is impossible (Tatsuki and Mori 2001; Yanigisawa et al. 2003 ; Yoo et al. 2008). 84

PAGE 85

While the year 2008 data concerning Su1 and su1 were similar between genotypes prior to 20 DAP (Fig. 3-16), analysis revealed one major difference during development. Most importantly, the samples of Su1 kernels from the year 2008 revealed two clear peaks, one each at 16 and 22-24 DAP, before falling 4-fold at 28-30 DAP (Fig. 3-16B). The su1 line had a first peak of similar level two days earlier at 14 DAP (though this was a single replicate and was not enough on which to base any conclusions). However, the second peak was completely absent in the su1 line from 20 to 25 DAP, with a 4-fold horm one reduction in the mutant kernels. This increase after a deep trough at the 25-30 DAP stage was similar to previous reports (Young et al. 1997; Young and Gallie 2000b) although the most closely-related data on su1 from Young et al. (1997) does not show such a decline in this genotype. In that report su1 ethylene levels rise to ~3-fold at 40 DAP over the levels in the rela ted control genotype. Insufficient amounts of field samples prevented analysis of this later stage. It is important to note that the most disparate levels of ethylene between the Su1 and su1 samples were at 24-25 DAP. These su1 measurements were taken on July 18th and July 20th, 2008. Upon further investigation, each of the ten ears harvested on th ese dates generated abnormally low ethylene values. These included the su1 samples at 24-25 DAP (Fig. 3-16), the Su1 samples at 28 and 30 DAP (Fig. 3-16) and the mn1 samples at 30 and 31 DAP (Fig. 3-2). It is possible that all these ages of kernels coinci dently produced low amount s of ethylene at their respective stages of development, considering that published data support a transient cessation of ethylene biosynthesis (Young et al. 1997; Young and Gallie 2000b). After consulting the hardcopies of GC data from these two harvest date s, no abnormalities were apparent in either the calibration or function of the GC apparatus. Th e average morning temperatures for these two dates were 83.6F and 82.4F, respectively. This was similar to the temperatures on the dates that 85

PAGE 86

the Su1 ethylene peak between 22 and 25 DAP were recorded. Despite this similarity in temperature, it seems plausible that some enviro nmental variation caused the low ethylene levels on July 18th and 20th. Because data was recorded over two di fferent days for ten different ears, with multiple replicates per ear, experimental error is less likely. A second peak of ethylene hormone in Su1 samples from year 2008 showed unique timing The occurrence of a second et hylene peak in year 2008 in Su1 samples, just eight days after the initial peak at 16 DAP (Fig. 3-16B), was unexpected because none of the published literature reports such a quickly repeated burst of ethylene (Y oung et al. 1997; Gallie and Young 2000b). Each of the three data points from 22 to 25 DAP was the average of two biological replicates (Fig. 3-16B), recorded across thre e different dates, lending credibility to the observance of this peak. Data from Young et al (1997) show a consiste nt dip in ethylene production at 24 DAP in the wild -type and three mutant lines; su1, su1se1 and sh2. Despite this similarity, none of those lines show such a rapid shift in hormone production. This could be partly due to the practice in all previously published work of sampling kernels at 4 DAP intervals. This methodology limits resolution of rising and falling hormone production to 8 DAP periods, possibly masking rapid bursts of ethylene generation. Still, most published ethylene data in maize kernels show sweeping curves without indication of rapid shifts. Data from various genetic backgrounds and genotypes highlight the complex nature of ethylene hormone production and measurement, and this report ad ds to the body of evidence supporting the sensitive nature of ethylene activity in maize kernels. Transcript Accumulation and Correlat ion with Ethylene Hormone Evolution Hormone synthesis, perception and action ar e regulated at many levels, with complex signals being integrated and modulated continuously. Transcript accumulation is one way to connect the involvement of a gene with certain tissues and processes. Despite the frequent 86

PAGE 87

occurrence of post-transc riptional and post-transl ational regulation, the qu antity of mRNA for a particular gene serves as a foundation for unders tanding the processes that affect phenotypes of biological systems. In this study, transcript levels were analyzed for four metabolic genes and four families of genes related to ethylene biosynthesis, perc eption and signal transduction. Samples were collected from two consecutive years of field harv est, and transcript levels were recorded as absolute transcript levels per nanogram tota l RNA. The second year, additional data was recorded to allow for calculation of transcripts on a per kernel basis in order to assess the state of the kernels as biological units. Transcript Accumulation in Mn1 and mn1 Genotypes Metabolic genes Given the centrality of the Mn1 gene for explanation of the mn1 phenotype, it is important to understand how Mn1 transcript levels relate to publis hed data on INCW2 enzyme activity. In addition, the Mn1 transcript was used as a control in order to compare qPCR data between biological replicates. At 8 DAP, the Mn1 transcript in the Mn1 genotype exhibited maximum expression as a proportion of total RNA, which is consistent with maximum enzyme activity and the role of INCW2 in providing hexoses to th e rapidly dividing cells between 8 and 12 DAP (Fig. 3-3A and B; Cheng et al. 1996). As cells proliferate and grow, the amount of Mn1 transcript is down-regulated as other developmentally important genes are transcribed. On a per kernel basis, there was a temporary plateau at 13 DAP, which coincides with a period of transition from cell division to cell expansi on and elongation (Kowle s and Phillips 1985; Sreenivasulu et al. 2004). In the mn1 genotype, the presence of Mn1 transcript was less than 10% of wild-type at all stag es (Fig. 3-3), though this level of transcript accumulation was well above published enzyme activity and protein levels (Che ng et al. 1996). This su ggests that, while the 87

PAGE 88

mn1 transcript levels were less than 10% of w ild-type (Fig. 3-3), there exists one or more regulatory processes that reduced enzyme activ ity beyond the reduction in transcript level. Another set of important enzymes in kernel metabolism are the sucrose synthases, which are associated with production of substrates for cell expansion as well as cell wall and starch biosynthesis (Chourey et al. 1998). Originally, Sus2 was included in this study in order to provide an internal control for RNA quality and sample preparation efficiency due to supposed consistent expression profiles in both Mn1 and mn1 genotypes. The margin of error in year 2007 samples prevented any meaningful comparison of Mn1 and mn1 lines. Year 2008 samples showed clear differences in the timing and abundance of transcript accumulation. Sus2 transcripts in the Mn1 genotype increased as a portion of total RNA at 11 DAP, and on a per kernel basis at 13 DAP (Fig. 3-4) This pattern is consistent with the concept of increased Sus2 activity promoting transition from cell divisi on to cell enlargement around 12 DAP. Conversely, in the mn1 line, Sus2 had a delayed and lower abundance of transcript at 13 DAP, possibly as a result of low hexose signaling fueling developmenta l progress (Fig. 3-4B). On a per kernel basis, mn1 samples displayed a brief increase of Sus2 transcript at 11 DAP, wi th a subsequent decline that probably reflected the cessation of kern el expansion after ~12 DAP (Fig. 3-4C). The HXK2 gene showed an insignificantly higher expression in the wild-type versus the mn1 genotype for year 2007 samples. This pattern was more clearly resolved in year 2008, when the 8 DAP stage showed 2-fold higher levels in Mn1 kernels (Fig 3-5). This difference in expression was similar to that of the Mn1 gene, in that levels were highest at early stages and declined steadily. The same plateau in the wild-type was seen between 11 and 13 DAP on a per kernel basis, further supporting this period as significant in devel opmental progression. 88

PAGE 89

Overall, the hexose-related genes, Mn1 and HXK2 were expressed early and declined during development in the wild-type, while Sus2 expression rose at the 11-13 DAP transition period. Hexokinase inhibitors have been shown to impair sucroseand hexose-dependent induction of a major sucrose synthase gene, Sus1, in Arabidopsis leaves, and specific forms of the sucrose synthase enzyme family are reported to be individually re gulated by sugar status (Koch et al. 1992; Ciereszko a nd Kleczkowski 2002). The publishe d data provide insight into possible causes of delayed and reduced Sus2 expression in mn1 seeds as a result of low hexose/sucrose ratio. These changes in INCW2-defi cient kernels highlight the pleiotropic effects of the mn1 seed mutation and resulting hexose deficiency during a critical st age of early kernel development. Ethylene biosynthesis genes of the ACS and ACO families As actuators of the rate-limiting step in ethy lene biosynthesis, ACC synthase enzymes are important targets for regulation of hormone effects. Levels for all three of the ACS genes currently identified in maize were an order of ma gnitude higher than published results, expressed both as a function of total RNA and as whole ke rnel data (Figs. 3-6 to 3-8; Gallie and Young 2004). This could be partially due to differences in genetic ba ckground used in the two studies. Additionally, the soil co mposition, temperature, water supply and light characteristics of the environment were possible factors in physiologica l variation. However, th e relative contributions of each ACS gene to total ACS transcript levels were in agre ement with Gallie and Young (2004): ACS2 was most abundant, followed by ACS7, and ACS6 was the least-expressed. Few significant differences were seen between Mn1 and mn1 samples for the ACS family (Figs. 3-6 to 3-8). The ACS2 and ACS6 transcripts were more a bundant in wild-type than mn1 at 8 DAP, but large standard error prevented solid conclusions. All three fa mily members displayed a constant or slightly downward trend in bot h genotypes, on a per ng to tal RNA and per kernel 89

PAGE 90

basis. This trend was not dissimilar to the w ild-type results of Gallie and Young (2004), given that data in the current study covered only early development for Mn1 and mn1 The most compelling observation was from ACS6 data, which showed a clear increase in transcript levels at 13 DAP in the wild-type from year 2008 (Fig 3-7B and C). The transcript levels at this stage coincided with the peak in ethy lene evolution in the same samples (Fig 3-2B) and also matched the timing of ACS6 transcript in Gallie and Young (2004). A disproportionate cont ribution of the three ACS genes to ethylene biosynthesis is demonstrated in maize leaf tissue (Young and Gallie 2004).The loss of acs2 function reduces ethylene levels by 50%, compared with a mutation in the acs6 gene that causes a 90% reduction in hormone levels. Even though the ACS proteins are subject to extensive post-transcriptional regulation (Tatsuki and Mori 2001; Liu and Zhang 2004; Sebastia et al. 2004), the data presented in Figures 3-6 through 3-8 are in agreement with the disproporti onate contribution of ACS6 transcript levels to increased ethylene biosynthesis described previously (Young and Gallie 2004). This observation is based pr imarily on the correlation of ACS6 transcript with ethylene evolution in 13 DAP Mn1 kernels (Fig. 3-7). The ACC oxidase family is comprised of f our members in maize, three of which are expressed in kernels (Galli e and Young 2004). These enzymes convert ACC produced by ACC synthase into ethylene, which is then free to di ffuse through tissues and interact with receptors. Two genes of the ACC oxidase family were analyzed in year 2008 samples only. The highlysimilar ACO20 and ACO35 genes showed relative expression rates simila r to those published by Gallie and Young (2004), although absolute abundance shown here was again an order of magnitude higher than their publis hed work. The peak levels in ACO35 transcripts, shown here at 13 DAP, were similar in both Mn1 and mn1 genotypes (Fig 3-10B) and were in agreement 90

PAGE 91

with data published in Gallie and Young (2004). Compared to th e wild-type, a higher proportion of ACO20 transcript was found in mn1 kernels at 13 and 20 DAP ba sed on # transcripts/ng total RNA (Fig. 3-9). There is no reason to conclude that this is related to higher ethylene synthesis in the mn1 line from 16 DAP onward, although the differe nce at 13 DAP could be indicative of increased induction of ethylene-related transcri pts at the expense of normal storage gene synthesis. ACO20 transcript levels in wild-type kernel s maintained constant expression per ng total RNA during development, which supports the observation of Gallie and Young (2004) that this gene is not induced early in endosperm growth. Ethylene receptor genes and EIL1-1 Ethylene receptors are critical components of ethylene signaling. They function as negative regulators of ethylene response that are inactivated upon binding of the ethylene molecule (Hua and Meyerowitz 1998). All four of the known mai ze ethylene receptors are expressed in maize kernels (Gallie and Young 2004). The thre e receptor transcripts examined here, ERS1-14, ETR29 and ETR2-40, followed a general pattern of decline from 8 to 20 DAP in terms of relative mRNA abundance (Figs. 3-11 to 3-13), with the ETR transcripts slightly more numerous than ERS1-14 Both of these observations are in agre ement with results published by Gallie and Young (2004). For the ERS114 transcript, little difference between the Mn1 and mn1 genotypes was apparent in year 2007 samples (Fig. 3-11A). Data from year 2008 showed a distinct and consistent difference between Mn1 and mn1 kernels at 11 DAP (Figure 3-11B). This is another example of a variation in the mn1 samples that coincides with a critical period in kernel development and transcriptional reprogramming. While increased ERS1-14 transcript in itself does not reveal the status of the ERS1-14 protein, and the direct cau se of transcriptional 91

PAGE 92

modulation for this gene is unknown, correlative evidence connects a tr ansient increase in ERS114 levels to decreased glucose and/or auxin levels as a result of the mn1 mutation. The two ETR2 genes showed simila r expression between Mn1 and mn1 genotypes at 8 and 11 DAP on a per kernel basis. While the Mn1 samples showed an increase in ETR2 transcript at 13 DAP, the mn1 samples transcript levels decrease d, and both lines remained level during subsequent development (Figs. 3-12C and 3-13C). This could be explained by differences in kernel size. Mn1 and mn1 kernels have similar characteristics until 11 DAP, at which point the miniature phenotype leads to impaired growth in the hexose-deficient mutant seeds. This trend of transcript reduction in mn1 kernels was visible in the ERS1-14 results as well, but differences were not statistically signifi cant. Overall it seems that the mn1 mutation did not affect transcription of ETR2 ethylene receptor genes in developi ng maize seeds, and only impacted ERS1-14 transcript levels during the 11-13 DAP transition period. Gallie and Young (2004) note similarities betw een the pattern of receptor expression and the downstream signaling components EIN2 EIL1-1 and EIL1-3 While only EIL1-1 transcript data is reported here, the trend was similar to that of the receptor genes. At all stages the trends of EIL1-1 transcript levels were similar between Mn1 and mn1 lines, with a notable reduction visible in mn1 kernels at 13 DAP that correlated with reduction in kernel size relative to wildtype (Fig. 3-14). EIL1-1 transcripts increased from 8 to 20 DAP in both lines, possibly facilitating increased ethylene signaling as kernels mature. These data do not support a possibility for transcriptional modulation of EIL1-1 due to the mn1 mutation. Transcript Accumulation in Su1 and su1 Genotypes Metabolic genes Due to difficulty producing an adequate supply of field-grown samples, the transcript levels for Su1 and su1 genotypes were the results of single biological sample analysis at each 92

PAGE 93

time point. Dinges et al. (2001) report that the reference su1 mutation reduces SU1 protein accumulation while leaving transcript levels unchanged. The use of the Su1 transcript as an internal control is of critical importan ce because of the expected similarity in Su1 transcript levels between the mutant and wild-type kern els. However, the published data focuses on a single time point (20 DAP) and the use of RNA gel blot analysis to quantify transcript levels (Dinges et al. 2001). With that in mind, both year 2007 and y ear 2008 results showed similar levels of Su1 and su1 transcript as a proportion of total RNA for all stages (Fig. 3-17 and 3-18 A and B), lending credibility to the accur acy of the assay. The observation that su1 kernels exhibited fewer Su1 transcripts per kernel at 20 DAP could be a resu lt of chance selection of smaller kernels for that time point. Additionally, su1 kernels could have had higher water content but reduced dry weight during this stage, lead ing to an overall lower production of RNA per kernel. When the singular nature of the 20 DAP su1 sample is considered, a lower efficiency of RNA isolation would also have led to an inaccura te determination of total RNA per kernel, thus skewing transcript results. Al so, transcripti onal control of Su1 is poorly understood, and could be related to the drastically altered sugar and starch profile present in the su1 genotype (Creech 1965; Dinges et al. 2001). The hexokinase HXK2 transcript levels paralleled the trend of Su1 transcript results (Fig. 3-17 and 3-19). HXK2 showed similar trends of transcript expression between Su1 and su1 genotypes (Fig. 3-19). Highest leve ls of transcript accumulation were during times of peak cell division during early development. There was no apparent transcrip tional modulation of HXK2 as a result of the su1 mutation. While Su1 and HXK2 transcript levels were si milar between wild-type and su1 kernels, the levels of Sus2 showed different responses based on genot ype in the single re plicates shown here 93

PAGE 94

(Fig. 3-18). The most consistent resp onse was in year 2008 samples, where Su1 samples produced more Sus2 transcript than the su1 mutant at all stages ex cept 20 DAP (Fig 3-18B and C). Higher sucrose levels in the su1 genotype could be one explanation for lower Sus2 expression. It is important to not e that in this study, both the mn1 and su1 genotypes caused higher levels of sucrose relative to the wild-type kernels, and th ere were decreased levels of Sus2 transcript that correlated with that modification of sugar status. Because the correlation between Sus2 transcript levels and SUS2 enzyme activity is unknown, furt her analysis is needed to understand the function of Sus2 in this system. Ethylene biosynthesis genes of the ACS and ACO families The ACC synthase family showed largely unrelated patterns of expression over two consecutive years. For y ear 2007 samples, all three ACS genes were strongly expressed in su1 kernels at 8 and 28 DAP versus wild-type (Figs. 3-20A to 3-22A). The period from 12 to 20 DAP was more similar between lines, to the po int that no significant difference was apparent. Why the earliest and latest stages would show such variation is perplexing, considering that ethylene production was similar at all stages (even though the 8 DAP su1 sample was unavailable for ethylene analysis). The lack of replication leads to the conclusion that experimental error and random va riation are as likely as any biological explanation for the differences at 8 and 28 DAP. One consistency is that ACS7 and ACS2 mRNAs were more abundant than ACS6, in agreement with other data presente d in this report (Figs. 3-20A to 322A). Concerning year 2008 samples, the relative cont ributions of each family member to total transcript amounts were in agreemen t with year 2007 data, in that the ACS2 and ACS7 transcripts showed up to 10-fold higher expression than ACS6 (Figs 3-20 to 3-22). For ACS2, expression was identical between Su1 and su1 samples at 15 and 20 DAP (Fi g. 3-20 B and C), which is a 94

PAGE 95

period marked by higher sucrose in the su1 line. It is logical to hypothe size that an effect due to sugar status would manifest itself during this stage, but this was not reflected in the data. At 25 DAP both ACS2 transcript and ethylene le vels were elevated in the Su1 genotype. At the same time point ACS2 transcript and ethylene levels were low in the su1 mutant (Figs. 3-15 and 3-20). This suggests a possible correla tion between this ACC synthase isoform and ethylene production at this stage in developm ent. Overall levels of ACS2 transcript rose thr oughout development in both genotypes, further supporting ACS2 as a developmentally-modulated gene. The ACS7 transcript, when translated, shares 95% amino acid identity with translated ACS2 (Gallie and Young 2004). However, ACS7 showed a different pattern of expression than the highly-similar family member. There was a slight decline in ACS7 transcript levels during development. Overall there was no significant difference in ACS7 transcript abundance between Su1 and su1 kernels (Fig. 3-22B and C). While the contribution of the ACS7 gene product to ethylene levels in maize kernels cannot be deduced from transcript data alone, the observation of relatively constitutive expression does not support the possibility of transcriptional regulation due to developmental or metabolic cues. The levels of ACS6 transcript were similar between wild-type and su1 samples at all stages except 20 DAP (Fig. 3-21B and C). In contrast to the other family members, ACS6 was maximally expressed at the earliest stage tested. This is in agreement with transcript levels published in Gallie and Young (2004) and supports the possibility that ACS6 is specifically involved in early ethylene biosynthe sis, and is in part specifica lly up-regulated at this stage. Ultimately, few conclusions can be drawn from such preliminary data concerning transcriptional regulation of the ACC synt hase family in developing maize kernels. ACS6 appeared to be more highly expr essed during early stages, while ACS2 was more prominent later 95

PAGE 96

in development. ACS7 transcript showed little variation during the period of development investigated here. These observa tions are in general agreement with results from Gallie and Young (2004a), but further analysis is required for validation. Transcripts for the ACC oxidase genes ACO20 and ACO35 were present at all stages of development tested. ACO20 was again the more highly expressed of the two family members (Figs 3-23 and 3-24). The Su1 and su1 genotypes were similar with respect to ACO20 accumulation; with relatively constant expression from 15 to 35 DAP. The ACO35 transcript levels in both genotypes showed a modest increa se as kernels develope d (Fig. 3-24), possibly providing ACO enzymes in order to increase ethylene evoluti on during maturation. Results for the two ACO members investigated here showed similarity between Su1 and su1 genotypes. However, ACO35 transcript levels were hi ghest at early stages in th e report by Gallie and Young (2004), contrary to data presented in this repor t. One possibility is that isoforms of the ACO family could be differently regulated as a re sult of changes in genetic background. With only single biological samples to draw data from, it is beyond the scope of this study to make such a claim. Similarities between Su1 and the mutant genotype lead to the conclusion that transcriptional modulation of the ACO genes as a result of the su1 mutation was unlikely during the stages tested. Ethylene receptor genes and EIL1-1 The ETR2-9 and ERS1-14 transcripts quantified in Su1 and su1 samples from year 2007 showed dissimilar patterns of expr ession (Figs 3-25A and 3-26A). The ETR2-9 transcript was most abundant and coincided with ethylene levels (Fig. 3-15). ERS1-14 had a constant level of expression except for a substantial increase in 8 DAP su1 kernels (Fig. 3-25A). This singular variation was not repeated in any other receptor data, casting doubt on the accuracy of this result 96

PAGE 97

97 and emphasizing the need for biological replica tion. There was no significant difference between Su1 and su1 genotypes for either receptor other than the anomalous 8 DAP ERS1-14 time point. For year 2008 samples, both ERS1-14 and ETR2-9 had similar trends and quantities of expression in both wild-type and su1 kernels (Figs 3-25 and 3-26). The ETR2-40 results showed more variation between genotypes, with a higher amount of transcript in the Su1 line at the earliest stage (Fig. 3-27). The most significan t difference was the higher receptor transcript levels in Su1 versus su1 at 35 DAP. This is the time when su1 kernels begin to produce substantially more ethylene than wild-type (Young et al. 1997), which would be exacerbated by lower receptor levels leading to increased sensitivity. The observation that receptor levels began to rise as ethylene increased in the Su1 genotype raised the possibility that ethylene activity increased transcription of the receptors. However, no changes in ethylene levels were observed in these samples, contrary to published data (Y oung et al. 2007) Also, recep tor transcript levels do not always correlate with pr otein accumulation (Kevany et al. 2007). It is possible that ethylene action simultaneously induces receptor transcript and prevents receptor protein accumulation depending on the activity of tissue-specific control mechanisms, as seen in transcript and protein leve ls of ripening tomato fruits (Kevany et al. 2007). The EIL1-1 transcript level increased modestly during development, with similar expression in Su1 and su1 samples (Fig. 3-28). An increase in the EIL1-1 transcript, if it led to increased protein accumulation a nd activity, could facilitate incr eased ethylene signaling during the later developmental stage associated with kernel maturation. However, since quantitative protein data is unavailable, further work is need ed to connect transcript levels with ethylene responses. No transcriptional di fferences betwee n wild-type and su1 genotypes were clear from the data shown.

PAGE 98

CHAPTER 5 CONCLUSIONS This report has demonstrated that patterns of biosynthesis of the phytohormone ethylene were affected by the mn1 mutation in developing maize seeds. In addition, data presented here documented variance in hormone levels based on the genetic background of wild-type and mutant genotypes. The mn1 mutation led to an increase in ethylene production rates in the smaller kernels during the 16-25 DAP stage, adding to the list of pleiotropic effects attributed to loss of INCW2 enzyme activity. Variance based on genetic background was highlighted by the novel results of Su1 kernel analysis, which showed a sec ond burst of ethylene production at 24 DAP, shortly after the initial p eak at 16 DAP. This pa ttern is unique when compared to the results of Mn1 and mn1 kernels analyzed during the same pe riod, as well as previously published reports. Transcript levels of several genes were shown to be affected by the mn1 mutation. For the first time, to the authors knowledge, the Mn1 transcript was quantifie d in developing seeds and shown to be present in the mn1 mutant kernels at levels between 5 and 10% of the related wildtype samples. Both Sus2 and HXK2 transcripts were reduced in the hexose-deficient mn1 genotype. Of the three ACC synthase genes, transcript levels of ACS6 appeared to be the most closely-correlated to rates of ethylene production. This is in agreement with previous reports on the importance of this gene in ethylene biosynthesis in other tissues. The levels of ACS7 transcripts were largely constant throughout development in all ge notypes tested, leading to the possibility that this ACC synthase transcri pt was not regulated via developmental cues. Transcript levels of the ACC oxidase genes were similar in mn1 kernels compared to the wildtype control. However, a slightly higher level of transcript was evident in the mutant genotype at 13 DAP for both ACO20 and ACO35 The difference in ERS1-14 transcript level, which was 98

PAGE 99

increased in the mn1 samples over the levels in Mn1 kernels, was only observed at the 11-13 DAP period associated with transition from cell division to cell expansion. Transcript data for the ETR2 receptors were similar in both Mn1 and mn1 genotypes. The EIL1-1 transcript was more abundant as development progressed in both genotypes in what appeared to be a developmentally-regulated process. Transcript levels for isoforms of the ACC synthase, ACC oxidase and ethylene receptor gene families showed some developmental re gulation between family members, but little transcriptional variation was clearly attributable to the mn1 mutation. Due to the lack of biological replication, conclusions based on transcript data from Su1 and su1 genotypes are tenuous. The HXK2 transcript levels appeared unchanged, while the Sus2 levels were clearly modi fied in the high-sucrose su1 kernels. Transcripts of the ACC synthase genes ACS2 and ACS7 were much more abundant than those of ACS6, in agreement with data from Mn1 and mn1 kernels. A higher proportion of ACS6 transcript was present during early stages, while ACS2 transcripts were more abunda nt during later development. ACS7 transcript levels appeared to be constitutive in both genot ypes throughout kernel growth. The ACC oxidase transcripts were simila rly expressed in both su1 and the wild-type kernels, with ACO35 increased during later stages of growth. The et hylene receptor expression in both Su1 and su1 genotypes decreased during development, with little diffe rence in transcription between wild-type and mutant kernels. Conversely, EIL1-1 transcript levels rose in both genotypes over time, consistent with a role in ethylene signaling during PCD and kernel matura tion during late development. In summary, the biosynthesis of ethylene in developing maize kernels is a complex result of many interacting factors, one of which is sugar status in devel oping seeds. Correlations between hormone levels and transcript levels of sugarand ethylene-related genes provide a 99

PAGE 100

100 basis for further investigation of the process involved in maize kernel development. Both the similarities and differences between wild-typ e and mutant maize geno types offer additional insight into the genetic interactions influencing kernel development.

PAGE 101

LIST OF REFERENCES Alonso JM, Hirayama T, Roman G, Nourizadeh S, Ecker JR (1999) EIN2, a bifunctional transducer of ethylene and stress responses in Arabidopsis Science 284:2148-2152 Beatty MK, Rahman A, Cao H, Woodman W, Lee M, Myers AM James MG (1999) Purification and molecular genetic characte rization of ZPU1, a pullulanase-type starch Debranching enzyme from maize. Plant Physiol 119:225-266 Beckles DM, Smith AM, Rees T (2001) A cytoso lic ADP-glucose pyrophosphorylase is a feature of graminaceous endosperms, but not of ot her starch-storing organs. Plant Physiol 125:818-827 Bhave MR, Lawrence S, Barton C, Hannah LC (1990) Identification and molecular characterization of Shrunken-2 cDNA clones of maize. Plant Cell 2:581-588 Black RC, Loerch JD, McArdle FJ, Creech RG (1966) Genetic interac tions affecting maize phytoglycogen and the pyytoglycogen-formi ng branching enzyme Genetics 53:661-668 Bosnes M, Weideman F, Olsen OA (1992) Endosp erm differentiation in barley wild-type and sex-mutants. Plant J 2:661 Boyer CD, Preiss J (1978) Multip le forms of starch branching enzyme of maize: evidence for independent genetic control. Biochem Biophys Re s Comm 80:169-175 Cao H, Imparl-Radosevich J, Guan H, Keeli ng PL, James MG, Myers AM (1999) Identification of the soluble starch synthase activities of maize endosperm. Pl ant Physiol 120:205-215 Chao Q, Rothenberg M, Solano R, Roman G, Te rzaghi W, Ecker JR ( 1997) Activation of the ethylene gas response pathway in Arabi dopsis by the nuclear protein ETHYLENEINSENSITIVE3 and related proteins. Cell 89:1133 Cheng WH, Tallercio EW, Chourey PS (1996) The Miniature1 seed locus of maize encodes a cell wall invertase required for normal development of endosperm and maternal cells in the pedicel. Plant Cell 8:971-983 Cho YH, Yoo SD, Sheen J (2006) Regulatory f unctions of nuclear Hexokinase1 complex in glucose signaling. Cell 127:579-589 Chourey PS, Jain M, Li QB, Carlson SJ (2006) Genetic control of ce ll wall invertases in developing endosperm of maize. Planta 223:159-167 Chourey PS, Nelson OE (1996) The enzymatic deficiency conditioned by the shrunken-1 mutations in maize. Biochem Genet 14:1041-1055 Chourey PS, Taliercio EW, Carlson SJ, Ruan YL (1998) Genetic evidence that two isozymes of sucrose synthase present un developing mai ze endosperm are critical, one for cell wall integrity and the other for starch biosynthesis. Mol Gen Genet 259:88-96 101

PAGE 102

Ciereszko I, Kleczkowski LA ( 2002) Glucose and mannose regulat e the expression of a major sucrose synthase gene in Arabidopsis via hexokinase-dependent mechanisms. Plant Phy Biochem 40:907-911 Clark KL, Larsen PB, Wang X, Ch ang C (1998) Association of the Arabidopsis CTR1 Raf-like kinase with the ETR1 and ERS ethylene re ceptors. Proc Natl Acad Sci USA 95:5401-5406 Creech RG (1965) Genetic control of carbohydrat e synthesis in maize endosperm. Genetics 52:1175-1186 Correns, C (1901) Bastarde zwischen maisrassen, mit besonder Berucksichtung der Xenien. Bibl Bot 53, 1-161 Dai N, Schaffer AA, Petreikov M, Granot D (1995) Arabidopsis thaliana hexokinase cDNA isolated by complementation of yeast cells. Plant Physiol 108:879-880 Denyer K, Dunlap F, Thorbjornsen T, Keeli ng P, Smith AM (1996) The major form of ADPglucose pyrophosphorylase in maize endosperm is extra-plastidial. Plant Physiol 112:7779785 Diboll AG (1968) Fine structure development of the megagametophyte of Zea mays following fertilization. Amer J Bot 55:787-806 Dickinson DB, Preiss J (1969) ADP glucose pyrophosphorylase from maize endosperm. Arch Biochem Biophys 130:119-128 Dinges JR, Colleoni C, Myers AM, James MG (2001) Molecular structure of three mutations at the maize sugary1 locus and their allele-specific phenotypic effects. Plant Physiol 152:1406-1418 Evensen KB, Boyer CD (1986) Carbohydrate composition and sensory quality of fresh and stored sweet corn. J Amer Soc Hort Sci 111:734-738 Foyer CH (1988) Feedback inhibition of photosynthesis through source-sink regulation in leaves. Plant Physiol Biochem 26:483-492 Gallie DR, Young TE (2004) The ethylene biosynthetic and perception machinery is differentially expressed during endosperm a nd embryo development in maize. Mol Gen Genet 271:267-281 Gallie DR, Geisler-Lee J, Chen J, Jolley Blair (2009) Tissue-specific expression of the ethylene biosynthetic machinery regulates root grow th in maize. Plant Mol Biol 69:195-211 Gou H, Ecker JR (2003) Plant response s to ethylene gas are mediated by SCFEBF1/EBF2-dependent proteolysis of EIN3 transcription factor. Cell 115:667-677 Grime JP, Mowforth MA (1982) Va riation in genome sizean eco logical interpretation. Nature 299:151-153 102

PAGE 103

Hennen-Beirwagen TA, Liu F, Marsh RS, Kim S, Gan Q, Tetlow IJ, Emes MJ, James MG, Myers AM (2008) Starch biosynthetic enzymes from developing maize endosperm associate in multisubunit complexes. Plant Physiol 146:1892-1908 Hua J, Meyerowitz EM (1998) Ethylene responses are negatively regulated by a receptor gene family in Arabidopsis thaliana Cell 94:261-271 James MG, Robertson DS, Myers AM (1995) Characterization of the maize gene sugary1 a determinant of starch composition in kernels. Plant Cell 7:417-429 Jang JC, Leon P, Zhou L, Sheen J (1997) Hexokinase as a sugar sensor in higher plants. Plant Cell 9:5-19 Jang JC, Sheen J (1994) Sugar sensing in higher plants. Plant Cell 6:1665-1679 Johnson PR, Ecker JR (1998) The ethylene gas signal transduction pathway: a molecular perspective. Annu Rev Genet 32:227-254 Kevany BM, Tieman DM, Taylor MG, Dal Sin V, Klee HJ (2007) Ethylene receptor degradation controls the timing of ripening in tomato fruit. Plant J 51:458-467 Kieber JJ, Rothenberg M, Roman G, Feldmann KA, Ecker JR (1993) CTR1, a negative regulator of the ethylene response pathway in Arabidopsis encodes a member of the Raf family of protein kinases. Cell 72:427 Kiesselbach TA (1949) The stru cture and reproduction of corn Univ. of Nebraska Press: Lincoln, NE Koch KE, Nolte KD, Duke ER, McCarty DR, Avigne WT (1992) Sugar levels modulate differential expression of maize sucros e synthase genes. Plant Cell 4:59-69 Kowles RV, Phillips RL (1985) DNA amplificatio n patterns in maize e ndosperm nuclei during kernel development. Proc Natl Acad Sci USA 82:7010 Kowles RV, Phillips RL (1988) Endosperm deve lopment in maize. Int Rev Cytol 112:97-136 Kyle DJ, Styles ED (1977) Development of aleurone and sub-aleurone layers in maize. Planta 137:185-193 LeClere S, Schmelz EA, Chourey PS (2008) Cell wall inve rtase-deficient miniature1 kernels have altered phytohormone levels. Phytochem 69:692-699 Li QB, LcClere S, Chourey P ( 2008) Plieotropic effect s of a single-gene mutation in sucrose utilization pathway extend beyond its tissuean d metabolic-specificity in developing seed of maize. Fl Genet C onf Abstract/Poster 55 103

PAGE 104

Liu Y, Zhang S (2004) Phosphorylation of 1-ami nocyclopropane-1-carboxylic acid synthase by MPK6, a stress-responsive mitogen-activated prot ein kinase, induces et hylene biosynthesis is Arabidopsis Plant Cell 16:3386-3399 Lowe J, Nelson OE Jr (1946) Miniature seed a study in the development of a defective caryopsis in maize. Genetics 31:525-533 Lur HS, Setter TL (1993) Role of auxin in ma ize endosperm development. Plant Physiol 130:273-280 Maniaitis T, Fritsch EF, Sambrook J (1982) Molecu lar cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Miller ME, Chourey PS (1992) Th e maize invertase-deficient miniature-1 seed mutation is associated with aberrant pedicel and endosperm development. Plant Cell 4:297-305 Moore B, Zhou L, Rolland F, Hall Q, Cheng WH, Liu YX, Hwang I, Jones T, Sheen J (2003) Role of the Arabidopsis glucose sensor HXK1 in nutrient, light, and hormonal signaling. Science 300:332-336 Myers AM, Morell MK, James MG, Ball SG (2000) Recent progress toward understanding biosynthesis of the amylopectin crystal. Plant Physiol 122:989-997 Nelson O, Pan D (1995) Starch synthesis in maize endosperm. Annu Rev Plant Physiol Plant Mol Bio 46:475-496 Ohto M, Onai K, Furukawa Y, Aoki E, Ar aki T, Nakamura K (2001) Effects of sugar on vegetative development and floral transition in Arabidopsis Plant Physiol 127:252-261 Pego JV, Weisbeek PJ, Smeeken s SCM (1999) Mannose inhibits Arabidopsis germination via a hexokinase-mediated step. Plant Physiol 119:1017-1023 Preiss J, Danner S, Summers PS, Morell M, Ba rton CR, Yang L, Nieder M (1990) Molecular characterization of the Brittle-2 gene effect on maize endosperm ADPglucose pyrophosphorylase subunits. Plant Physiol 92:881-885 Rahman A, Wong Ks, Jane Jl, Myers AM, Jane MG (1998) Characterizati on of SU1 isoamylase, a determinant of storage starch stru cture in maize. Plant Physiol 117:425-435 Saltman P (1953) Hexokinase in higher plants. J Biol Chem 200(1):145-154 Sebastia CH, Hardin SC, Clouse SD, Kieber JJ, Huber SC (2004) Identification of a new motif for CDPK phosphorylation in vi tro that suggests ACC synthase may be a CDPK substrate. Arch Biochem Biophys 428:81-91 Shannon JC, Creech RG, Loerch JD (1970) Starch synthesis studies in Zea mays Plant Physiol 45:163-168 104

PAGE 105

Shannon JC, Pien FM, Cao H, Liu KC (1998) Brittle-1, an adenylate tran slocator, facilitates transfer of extraplastidial s ynthesized ADP-glucose into amyloplasts of maize endosperms. Plant Physiol 117:1235-1252 Sheen J. 1990 Metabolic repression of transcription in higher plants. Plant Cell 2:1027-1038 Shewry PR, Halford NG. (2002) Cere al seed storage proteins: stru ctures, properties and role in grain utilization. J ournal Exp Bot 53:947-958 Shure M, Wessler S, Fedoroff N (198 3) Molecular identification of the Waxy locus in maize. Cell 35:225-233 Slein MW, Cori GT, Cori CF ( 1950) A comparative study of hexokinase from yeast and animal tissues. J Biol Chem 186:763-780 Tatsuki M, Mori H (2001) Phosphorylation of to mato 1-aminocyclopropane-1-carboxylic acid synthase, LeACS2, at the C-termin al region. J Biol Chem 276:28051-28057 Tsai CY, Salamini F, Nelson OE (1970) Enzy mes of carbohydrate metabolism in the developing endosperm of maize. Plant Physiol 46:299 Vilhar B, Kladnik A, Blejec A, Chourey PS, De rmastia M (2002) Cytometrical evidence that the loss of seed weight in the miniature1 seed mutant of maize is associated with reduced mitotic activity in the developing endosperm. Plant Physiol 129:23-30 Wu C, Colleoni C, Myers AM, James MG (2002) Enzymatic properties a nd regulation of ZPU1, the maize pullulanase-type starch Debran ching enzyme. Arch Biochem Biophys 406:21-32 Xiao W, Sheen J, Jang JC (2000) The role of he xokinase in plant sugar signal transduction and growth and development. Plant Mol Bio 44:451-461 Yang S, Hoffman N (1984) Ethylen e biosynthesis and its regulati on in higher plants. Annu Rev Plant Physiol 35:155 Yanigisawa S, Yoo SD, Sheen J (2003) Differentia l regulation of EIN3 stability by glucose and ethylene signaling in plants. Nature 425:521-525 Yoo SD, Cho YH, Tena G, Xiong Y, Sheen J (2008) Dual control of nuclear EIN3 by bifurcate MAPK cascades in C2H4 signalling. Nature 451:789-796 Young TE, Gallie DR (2000a) Programmed cell d eath during endosperm development. Plant Mol Bio 44:283-301 Young TE, Gallie DR (2000b) Regulation of pr ogrammed cell death in maize endosperm by abscisic acid. Pl ant Mol Bio 42:397-414 Young TE, Gallie DR (2004) ACC synthase expre ssion regulates leaf performance and drought tolerance in maize. Plant J 40:813-825 105

PAGE 106

106 Young TE, Gallie DR, DeMason DA (1997) Ethylen e-mediated programmed cell death during maize endosperm development of wild-type and shrunken2 genotypes. Plant Physiol 115:737-751 Zhou L, Jang JC, Jones TL, Sheen J (1998) Glucos e and ethylene signal transduction crosstalk revealed by an Arabidopsis glucose-insensitive mutant Proc Natl Acad Sci 95:1029410299

PAGE 107

BIOGRAPHICAL SKETCH Andrew Joseph Funk was born to Joseph Willia m and Donna Carol Funk during the year 1983 in the town of Decatur, IL. In 1990 he moved with his parents and sist er to Tallahassee, FL. During high school he developed a love for biol ogy under the teaching of Janice Ouimet. This led to his receipt of a bachelor's degree in mi crobiology and cell science from the University of Florida in 2006, where he worked as a research assistant in the laboratory of Dr. Prem Chourey. At Dr. Chourey's invitation, Andrew applied to th e plant pathology program at the University of Florida, and worked for the United States Department of Agriculture while fulfilling requirements for the degree of Master of Science. 107