1. 5:4.
.3 .Euar

in?

fin...

‘1
.4

a
".an‘wuxn s
.55.". a .v:
3 H 5 ,
K.

.V:

.3 ....u.m3.. 9
I. d..!i91fl.VV.

.I,

.‘T...

.3. _
if. 1.

3‘» 751?
«v u 1 it]

www.mfiur.
ct:l:...fl;u .15. 3...
...w . 2 ‘
€50? ,

. .
Ps3 .2 3....

r. 4.,

.*

a

. a or $1,361:

rvr ..

a.

 

' “" LIBRARY
Michigan State
University

This is to certify that the
dissertation entitled

MODIFICATION OF SEX EXPRESSION IN TRANSGENIC MELON
VIA ALTERED ETHYLENE PRODUCTION AND PERCEPTION

presented by

HOLLY A. LITTLE

has been accepted towards fulfillment
of the requirements for the

Doctoral ' degree in The Plant Breeding and'
Genetics Program and
Department of Horticulture

 

 

 

264““ Maw”—
Major Professor’s Signature
.3 - 2 2 —0 5’
Date

MSU is an Afiirmative Action/Equal Opportunity Institution

'--n----:---a-o-o-o--o-.—~

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 cElRmJt-rdd-pls

 

MODIFICATION OF SEX EXPRESSION IN TRANSGENIC MELON

VIA ALTERED ETHYLENE PRODUCTION AND PERCEPTION

By

Holly A. Little

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Plant Breeding and Genetics Program
and Department of Horticulture

2005

ex.
firs
don
Aral
C [01-
regio
femal

fonna

ABSTRACT
MODIFICATION OF SEX EXPRESSION IN TRANSGENIC MELON
VIA ALTERED ETHYLENE PRODUCTION AND PERCEPTION
By

Holly A. Little

Cucurbit species are noted for their diversity of sexual phenotypes which are
subject to control by developmental, genetic, and hormonal factors. Understanding the
basis of sex expression is of basic interest to elucidate underlying mechanisms of sex
determination and is of applied interest as pistillate flowers are necessary for fruit
production. Studies with exogenous hormones have indicated that ethylene is the
primary hormone influencing sex determination in cucurbits, with increased ethylene
causing increased femaleness. In this dissertation, I sought to: examine the role of
ethylene perception by the developing flower in sex determination, test the ability to
increase femaleness in the field through increased endogenous ethylene production, and
examine possible ecological impacts of altered ethylene sensitivity or production. In the
first study, melon (Cucumis melo L.) plants were transformed with the Arabidopsis
dominant mutant ethylene receptor etrI -1 , under the control of the petal/stamen targeted
Arabidopsis Apetela3 (AP3) promoter, or carpel/nectary targeted Arabidopsis Crab '3
Claw (CRC) promoter, to evaluate the effects of blocked ethylene perception in different
regions of the developing flower. CRC::etrI-1 melon plants showed increased
femaleness as measured by earlier and increased pistillate bud production, and the

formation of female rather than hermaphrodite buds. AP3::etrI-1 melon plants showed

Iii

pc

im

increased maleness as measured by almost exclusive staminate flower production and the
presence of poorly developed carpels in the few pistillate buds that were produced. The
results of these experiments suggest a model for sex determination in melon whereby
ethylene perception by the stamen is required for carpel development while ethylene
perception by the carpel is required for stamen development. In the second study, field
grown transgenic melon constitutively overexpressing an ACS gene for ethylene
biosynthesis showed earlier and increased production of mature pistillate flowers as well
as earlier fruit set and increased sequential fi'uit set on the main stem. These results
indicate that ethylene is important for pistillate bud development and also may suggest an
additional role in signaling associated with patterns of fi'uit set. Lastly, published
literature was examined to conduct a case study on modified ethylene production and
perception in order to provide insight into possible secondary effects and ecological

impacts of altered signaling pathways.

 

C(

thc
Pa;
Am

for 1

flier

unde
l0}' tc

zest fi

ACKNOWLEDGEMENTS

I would like to thank my committee members; Drs. Rebecca Grumet, Randy
Beaudry, Fred Gifford, Amy Iezzoni, and Michael Thomashow, for their guidance
throughout my graduate studies. I would like to give a special thanks to my advisor,
Rebecca Grumet, for providing me with the support I needed to piece together a rather
complex puzzle.

I would like to express my gratitude to my fellow labmates, past and present, for
their technical and moral support, particularly Zakir Ullah, Xiaofeng Wang, and Katerina
Papadopoulou for teaching me many of the techniques I used for my research and Kaori
Ando for being such a good friend. I owe a special acknowledgement to Sue Harnmar,
for providing an enormous amount of assistance with my research as well as a great
friendship and support when I needed it.

Lastly but certainly not least, I wish to thank Joel for his friendship,
understanding, and endless love and support; Jackson, for bringing a new dimension of
joy to my life and helping me to put things in perspective; and Allie, for her energy and

zest for life.

iv

 

LIE
LIS

CH.

C HA.

ethylc

TABLE OF CONTENTS

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

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

CHAPTER 1. Literature Review: Sex determination in Cucumis .............................. 1
Introduction ................................................................................... 1
Unisexual flower development in Cucumis ................................................ 3
Genetics of sex expression ................................................................... 5
Hormonal control of sex expression ........................................................ 7
Role of ethylene in sex expression ........................................................ 11
Objectives of the following research ..................................................... 18
References .................................................................................... 22

CHAPTER 2. Modified sex expression in transgenic melon expressing the mutant

ethylene receptor, At-etrl -1 , under the control of floral targeted promoters ................. 28
Abstract ....................................................................................... 28
Introduction .................................................................................. 29
Materials and Methods ...................................................................... 38

Preparation of Plasmid Constructs ............................................... 38
CRC::etr1-I ............................................................... 38
AP3::etr1-I ................................................................ 38

Plant Transformation .............................................................. 40

PCR analysis ....................................................................... 41

Northern Hybridization Analysis ................................................ 42

f7
é

 

Ethylene Measurements ........................................................... 42

Sex Expression ..................................................................... 43
Floral Morphology Evaluation ................................................... 44
Results ........................................................................................ 44
Gene integration and expression analysis ...................................... 44
Ethylene Measurement ............................................................ 49

Sex expression and floral morphology .......................................... 51
CRC::etrI-1 ................................................................. 51

AP3.-:etr1 -1 ............................................................... 58

Discussion .................................................................................... 63
References ................................................................................... 70

CHAPTER 3. Evaluation of the effect on modified endogenous ethylene production

on sex expression and fruit production in field grown transgenic

Cucumis melo ............................................................................... 75
Abstract ........................................................................................ 75
Introduction ................................................................................. 76
Materials and Methods ..................................................................... 78
Northern analysis .................................................................. 78
Field experiments .................................................................. 79
Results ....................................................................................... 80
Northern analysis ................................................................. 80
Sex expression and hermaphrodite bud development ........................ 80
Fruit development ................................................................. 82

vi

 

 

CO)

APP

Discussion .................................................................................... 91
References ................................................................................... 95

CHAPTER 4. Risk assessment of transgenic plants with altered signaling pathways:

A case study of modified ethylene ........................................................ 99
Abstract ...................................................................................... 99
Introduction ................................................................................ 100

How can we begin to assess these secondary effects that may influence

gene flow, invasiveness, or non-target impacts? ................................................. 104
Why modify ethylene? ......................................................................................... 105
What are the genetic components of ethylene biosynthesis and signaling? ........ 106
How are these genes used to modify ethylene? ................................................... 108
What phenotypic effects are associated with modified ethylene signaling? ........ 110
Plant growth and development .................................................. 110
Plant-microbe interactions ...................................................... 1 14
Disease response ................................................................. l 15
Abiotic stress ..................................................................... 117

Cross talk with other hormones ................................................ 121
What pleitropic and secondary effects have been observed? ............................... 126
Conclusions: implications for risk assessment ....................................... 130
References ................................................................................. 132
CONCLUSIONS AND FUTURE RESEARCH .............................................. 143

APPENDIX A. Floral targeted expression of a heterologous ACS in andromonoecious

melon ....................................................................................... 147

vii

AF

Introduction ................................................................................ 147

Materials and Methods .................................................................... 147
Plasmid Construction ............................................................ 147

Plant Transformation ............................................................ 148

PCR analysis ..................................................................... 148
Northern Analysis ............................................................... 149
Ethylene Experiments ........................................................... 149
Results ...................................................................................... 150
References ................................................................................. 155
APPENDD( B. Cloning of a partial genomic sequence of CS-A CSI/3 ..................... 156
Introduction ................................................................................ 1 56
Materials and Methods ................................................................... 157
Results and Discussion .................................................................. 159
References ................................................................................. 1 60

viii

 

Tabl

Table

Table 2

Table 3.1

In

Table A, 1

LIST OF TABLES

Table 1.1. Summary of etrI -1 constructs and predicted effects on sex expression in
transgenic melon ............................................................................ 20
Table 2.1. Segregation analysis of AP3 ::etr1 -1 and CRC::etrI-1 self-pollinated
transgenic melon ............................................................................ 46
Table 2.2. Combined data for all leaf and male bud ethylene measurement experiments
ofAP3::etr1-I and CRC::etr1 -I transgenic melon lines ............................... 50
Table 3.1. Flowering, fruit set and yield characteristics of field grown 35S::A CS
transgenic and control melon plants ..................................................... 86

Table A.1. Segregation analysis of CRC::A CS T1 transgenic melon ...................... 151

ix

 

tranSg

Figurf

and
CRC
Figure 2.7, (

“are 2.8. ca

(ARISE/1

LIST OF FIGURES

Figure 1.1. Yin and Quinn model for the effects of ethylene on sex expression in
cucumber based on the predicted roles of the F and M loci ........................... 10
Figure 2.1. Expected sexual phenotypes of transgenic melon lines based on effects
of ethylene treatment of wild type andromonoecious melon ......................... 37
Figure 2.2. Schematic of binary vector construct CRC::etrI-1/AP3::etr1-1 ................. 39
Figure 2.3. . PCR amplification of etrI-I gene construct from genomic DNA of T1
transgenic lines with CRC::etr1-I and AP3::etr1-1 Specific primers ........................ 45
Figure 2.4. Northern hybridization of heterologous etrI-I gene expression from
greenhouse grown AP3::etr1 -1 and CRC::etr1-1 melon with a DIG-labeled
etrI -1 probe ................................................................................... 47
Figure 2.5. Main stem sex expression patterns of melon plants transformed with
CRC::etr1-1. .............................................................................. 52
Figure 2.6. Size reached by carpel-bearing buds before senescing for Hale’s Best Jumbo
and azygous non-transgenic plants (WT and AZY), and CRC: :etrI -1 plants (lines
CRCS, CRCl 1, and CRCIS). ............................................................ 55
Figure 2.7. Carpel-bearing flower morphology of non-transgenic (WT and 4nNEG) and
transgenic (35S::etr1-1 and CRC::etr1-1) melon plants. ............................. 56
Figure 2.8. Carpel-bearing bud formation on AP3 ::etr1 -1 trangenic melon plants ....... 59
Figure 2.9. Hermaphrodite flower morphology of non-transgenic (WT) and transgenic

(AP3::etrI-I) melon plants. ............................................................. 61

‘F'a

 

Figure

Figure}

Figure 2.10. Revised model for hypothesized role of ethylene perception in sex
determination which suggests a reciprocal relationship between petals and
stamens ....................................................................................... 66

Figure 3.1. Northern hybridization of heterologous ACS gene expression in leaf tissue
fi'om field grown T2 transgenic 35S::ACS lines ACSl, ACS3, and ACS4 .......... 81

Figure 3.2. Sex expression pattern along the main stem of wild type Hale’s Best
Jumbo (WT), azygous segregants (AZY), and transgenic T2 355::A CS (ACS3
and ACS4) field grown melon plants .................................................... 83

Figure 3.3. Pattern of mature hermaphrodite flower and fi'uit set on main stem of field
grown wild type (WT), azygous segregants (AZY) and 35S::A CS (ACS3 and
ACS4) melon plants ........................................................................ 85

Figure 3.4. Fruit set and ripening pattern of field grown wild type Hale’s Best Jumbo
(WT), azygous segregant (AZ), and T2 35S::A CS melon plants (ACS3 and
ACS4) ........................................................................................ 87

Figure 4.5. Consecutive node fruit set on main stem of field grown Hale’s Best Jumbo
wild type (WT) melon and 355::A CS transgenic melon .............................. 90

Figure A.1. Northern hybridization of heterologous ACS gene expression in tissue
from greenhouse grown T1 transgenic CRC::A CS melon ........................... 152

Figure A.2. Ethylene production from leaves of wild type (WT) Hale’s Best Jumbo,
transgenic 35S::ACS (ACS4), and CRC::A CS (M168 and M169) melon... .. ....154

Figure B]. Layout of primers designed for inverse PCR based on CS-A CS1 and
CS-A CS3 sequences ....................................................................... 158

Figure B2. 2.0 kb 5’ noncoding sequence of CS-ACSI(CS—A CS3) ........................ 161

xi

 

 

In trod

female

' Althoug
result in
evolutior
al., 1994:
al., 1998;
strategies
female rep
gametes, 31
Ainsworth.
“Wilber of I
We. The n
Species, indi
MC"fibers of .-

(plsilllate) flO

Chapter 1

Literature Review: Sex Determination in Cucumis

Introduction

Most flowering plants produce hermaphroditic flowers in which both male and
female gametes are produced by providing them with the capacity to self-fertilize.
Although self-fertilization can be an advantage for a species, it also has the potential to
result in loss of fitness arising from inbreeding depression. This may have been an
evolutionary force for plant species to develop strategies to promote outcrossing (Grant et
al., 1994; Dellaporta and Calderon-Unea, 1993; Juarez and Banks, 1998; Ainsworth et
al., 1998; Barrett, 2002; Tanurdzic and Banks, 2004). For hermaphroditic species,
strategies to minimize inbreeding depression include differences in position of male and
female reproductive organs, differences in the timing of maturity of male and female
gametes, and genetic self-incompatibility (Dellaporta and Calderon-Unea, 1993;
Ainsworth et al., 1998; Barrett, 2002). Another mechanism, which has appeared in a
number of plant genera, is unisexual Species, such that flowers form only one gamete
type. The most common unisexual types are monoecy and dioecy. Within a dioecious
species, individual plants will produce exclusively either male or female flowers.
Members of a monoecious species produce separate male (staminate) and female
(pistillate) flowers on the same plant (Grant, 1994). Agronomically important dioecious
species include papaya, spinach, yarn, and asparagus; and monoecious species of
agronomic importance include maize, cucumber, melon, castor bean, and hazelnut

(Frankel and Galun, 1977). Elucidation of the genetics of sex determination is

 

 

 

fund;

breec'

31365
detem
Cuczm.
determ.
(Tanurc
gynoecii
hennaph
sex types
andromon
genetic. en
Em
much of the
Particular] y c
generally tho;
gained regardi
Phenolmes‘ WI
technolongS en.-

Control of 56X dc,-

fundamental to filrther understanding of plant evolution as well as advancement of plant
breeding efforts.

Among monoecious species, cucurbit species show tremendous diversity in sexual
types and have become a valuable tool in research to further understand sex
determination. Driven by the efforts of plant breeding programs to produce hybrid seed,
Cucumis sativus (cucumber) became a model system for studies to understand sex
determination in the 1950s and 19603 and remains the best understood and most studied
(Tanurdzic and Banks, 2004). The most common sexual types in cucumber are
gynoecious and monoecious, however androecious, andromonoecious, and
hermaphroditic types also exist (Rudich, 1990). While not as well understood, a range of
sex types can be produced in Cucumis melo (muskmelon), which is predominately
andromonoecious. Research on sex expression in cucurbits has evaluated the roles of
genetic, environmental, hormonal, and developmental factors.

Extensive research of the 1950’s and 1960’s established a solid foundation for
much of the role of plant growth regulators, both endogenous and exogenous on cucurbit,
particularly cucumber, sex expression, as well as effects of the environment which are
generally thought to be mediated via hormones (Perl-Treves, 1999). Insight also was
gained regarding the genetic basis of components controlling the various sexual
phenotypes, which exist among cultivars (Perl-Treves, 1999). The advent of molecular
technologies enables the possibility for further elucidation of the underlying genetic

control of sex determination.

 

 

Unise:

the flor
and are
inhibitio
results in
of carpel
to develop
the precise
differences
reproducti v:
Base
three classes
four Whorls 01
thomeotic [1.
genes control I]
stamens; and C
might into the 3

W011) d Predict PC!

 

Unisexual flower development in Cucumis

Despite having separate male and female flowers, at a young deveIOpmental age
the floral buds of monoecious cucumber plants have both stamen and carpel primordia
and are therefore considered bisexual (Atsmon and Galun, 1960). The subsequent
inhibition of carpel primordia leads to a male flower while arrest of stamen primordia
results in pistillate flower development. Similarly in andromonoecious melon, inhibition
of carpel primordia results in a male flower while stamen and carpel primordia continue
to develop in pistillate buds to form perfect flowers. Further investigation to determine
the precise timing of sexual differentiation of cucumber buds showed morphological
differences beginning at the 1mm stage while changes in the growth rate of respective
reproductive organs begins prior to the visible difference (Goffinet, 1990).

Based on the ABC model of flower development, combination of the activities of
three classes of homeotic genes (A, B, and C) are responsible for organ identity of the
four whorls of the flower (N g and Yanofsky, 2000; Jack, 2001 ). Through the evaluation
of homeotic flower mutants of Arabidopsis and snapdragon, studies have Shown that A
genes control the development of the sepals; A and B genes, petals; B and C genes,
stamens; and C genes, carpels. Study of homeotic mutants in cucumber has provided
insight into the spatial control of unisexual flower development (Kater et al., 2001). One
would predict petals and stamens to be replaced by sepals and carpels, respectively, in a
class B mutant. However, female flowers of the class B mutant had sepals in the first two
whorls, carpels in the fourth, and an undeveloped third whorl, while in node positions
normally destined to males, flowers had sepals in the first two whorls, carpels in the third,

and an undeveloped fourth whorl. These results suggest that the developmental fate of

5

 

 

 

 

stamens or carpels of the unisexual cucumber flower is dependent on their position and
not their sexual identity.

Additional research has identified and evaluated the expression of MADS-box
homeotic genes in developing male and female cucumber buds to investigate their
possible roles in selective differentiation of sexual organs (Perl-Treves et al., 1998 and
Ando et al., 2001). Three C-function Arabidopsis AGAMOUS (AG) homologs were
isolated fiom cDNA libraries of male and female buds at the bisexual stage and
designated CAGI, CA 02, and CAG3 (Perl-Treves et al., 1998). CAGI and CAG3
expression followed a pattern typical of C-function genes with transcript detection in the
stamens, carpels, and nectary while CA 62 exhibited carpel-specific expression. Since
CA 62 expression was not detected until visible sexual differentiation, it did not appear to
play a role in the “designation” of female flowers. In another study of cucumber,
differential display was used to identify a female specific MADS-box gene (ERAFI 7)
(Ando et al., 2001). ERAFI 7 was found to be expressed at the apex upon hormonal
induction of female flowers, and its expression was maintained in tissue of female buds
throughout development. It was suggested that ERAFI 7 may act in a feminizing cascade.

A recent study focused on morphological, cellular, and molecular changes in the
arrested inappropriate sexual organ in floral buds of cucumber (Hao et al., 2003; Bai et
al., 2004). Detailed microscopic analysis revealed that divergence from a bisexual to
unisexual flower begins following the initiation of the carpel in both male and female
buds (Bai et al., 2004). Using a TUNEL assay, DNA damage was detected in cells of the
primordial anther, coinciding with arrest in stamen development in female buds (Hao et

al., 2003). This finding suggests a pathway of unisexual flower development in which

 

_r"=

 

fiirtht
activi
Stigm;
feature
spore-I
cascadi
carpel c

1

follows I
Galun, 19
during the
pistillate fl
an initial pl
Staminate a1
'Ph}'siol0gic
bem'een you}

a gradient. 5

Of each Phase (

Genetics Of Sex

Vaoing 1.

rise 10 312 maYOf:

h .
allng SmCtly pjsn'

further growth of the inappropriate organ is hindered by DNA damage resulted from the
activity of a specific DNase. Evaluation of the arrest of carpels in male buds, showed
stigma-like primordia but no differentiation of the ovary (Bai et al., 2004). The common
features of the inappropriate reproductive organ arrest, such as the timing and targeting of
Spore-bearing parts, were suggested to indicate that a common signal may initiate the
cascade of events leading to inhibited stamen development in pistillate buds and inhibited
carpel development in staminate buds (Bai et al., 2004).

In monoecious C. sativus cultivars, flower development along the main shoot
follows three phases of sex expression creating a ‘physiological gradient’ (Atsmon and
Galun, 1962; Shifriss, 1961). During the first phase, staminate flowers are formed;
during the second, a mixture of staminate and pistillate flowers; and during the third,
pistillate flowers. A similar gradient is observed in andromonoecious C. melo in which
an initial phase of vegetative nodes is followed by staminate nodes then a mixture of
staminate and perfect flowers. Atsmon and Galun (1962) attempted to describe the
‘physiological gradient,’ suggesting that a growth substance varying in distribution
between young and mature leaves along the shoot affect the sex of adjacent buds creating
a gradient. Sex genotype, environment, and plant growth regulators influence the length

of each phase (Perl-Treves, 1999).

Genetics of sex expression
Varying levels of femaleness or maleness exists among cucurbit genotypes giving
rise to an array of sexual phenotypes. Among these phenotypes are gynoecious lines

having strictly pistillate flowers, androecious lines having only staminate flowers,

 

andror

interm

control 1‘
1961; K
locus is .
results in
with an 0
al., 1976;
influences
formation 1
I969; revie
homozygou
hybrid seed
the seed ban
Paternal line
thehormonaj j
ememdy use.
p Wide an earl
The gen

andromonoecious lines having both perfect and staminate flowers, and a variety of
intermediate intersexual phenotypes.

Studies have shown that the majority of the sexual phenotypes in cucumber are
controlled by three major genes, although minor sexual modifier genes do exist (Galun,
1961; Kubucki, 1969; Shifi'iss, 1961; Rudich et al., 1976; Perl-Treves, 1999). The F
locus is a partially dominant gene that influences the degree of femaleness such that F -
results in gynoecy. The A locus is associated with increased maleness such that plants
with an aafl genotype are androecious and do not produce pistillate flowers (Rudich et
al., 1976; reviewed in Perl-Treves, 1999). The M locus is a dominant gene that
influences the development of stamens, whereby 3 MM or Mm genotype results in
formation of a female flower while mm results in perfect flower formation (Kubicki,
1969; reviewed in Perl-Treves, 1999). Various sexual genotypes, particularly those
homozygous dominant for the F locus, have been utilized commercially for inbred and
hybrid seed production. The use of a female line for hybrid seed production ensures that
the seed harvested from the maternal parent is the product of cross-pollination with a
paternal line. Maintenance of gynoecious lines by self-pollination is possible through
thehorrnonal induction of staminate flowers. Furthermore, the F locus has been
extremely useful to breeders in order to generate and maintain gynoecious cultivars that
provide an earlier and more uniform fruit-set (Rudich, 1990).

The genetic basis of sex expression in C. melo is thought to be controlled by
interaction of at least three genes, A, G, and M (Kenigsbuch and Cohen, 1989; Roy and
Saran, 1990). Interactions between A and G are proposed to be responsible for the

inheritance of many of the sexual phenotypes, such that A inhibits the development of the

 

 

fiilly

reces
genot
locus

lines 0

Burma

1

1990). 5
Plant hon
ethylene a
gibberemn
BukOVac, 1
6t al., 1969;
Possible to $1
7776 al
cucumber Ie d .
’egUIation Ofui

was found to pro

' def/Opingl

androecium in female flowers while G inhibits the development of the gynoeciurn in
male flowers (Roy and Saran, 1990). Therefore, genotype A-G- designates monoecy;
aaG—, andromonecy; A-gg, gynomonoecy, and aagg, herrnaphrodism. Inheritance studies
of gynoecy in muskrnelon have suggested that the M (maleness) locus in a homozygous
recessive form is required for a complete and stable gynoecious phenotype; therefore, the
genotype of gynoecious melon is A-ggmm (Kenigsbuch and Cohen, 1989). The lack of a
locus analogous to the F locus in cucumber has made efforts to maintain stable breeding

lines of gynoecious melon more difficult.

Hormonal control of sex expression

Another level of control of cucurbit sex expression is by plant hormones (Rudich,
1990). Exogenous application as well as changes in the endogenous level of a variety of
plant hormones is capable of influencing sex determination. Auxin, brassinosteroids, and
ethylene are growth regulators which promote pistillate flower formation, while
gibberellins and ethylene inhibitors promote staminate flower production (Wittwer and
Bukovac, 1959; Galun, 1959; Atsmon et al., 1968; McMurray and Miller, 1968; Rudich
et al., 1969; Papadopoulou, 2002). Therefore, with appropriate timing and quantity, it is
possible to Shift sex in either direction by application of plant growth regulators.

The ability of exogenous treatment with gibberellins (GA) to modify sex in
cucumber led researchers to suggest the possibility of a role of endogenous GA in the
regulation of unisexual flower development (Atsmon et al., 1968). Treatment with GA
was found to promote male flower initiation as well as prevent existing female buds from

fully developing (Atsmon et al., 1968; Perl-Treves, 1999) while treatment with inhibitors

 

ofGA 1
addition
lines slit

is involvl
TI

and melor
addition. e
have been i
of auxin as
However~ st

induction of

Therefore: then
Inflated “'3 etl

has been “alum,

of GA biosynthesis has been observed to promote femaleness (Yin and Quinn, 1995). In
addition, evaluation of GA content in monoecious and isogenic gynoecious cucumber
lines showed higher levels in monoecious plants, supporting the idea that endogenous GA
is involved in sex determination (Atsmon et al., 1968).

The feminizing effect of exogenous auxin has been demonstrated in cucumber
and melon (Galun et al., 1963; Galun et al., 196; Trebitsh et al., 1987; Rudich, 1990). In
addition, endogenous levels of auxin from varying sex genotypes of cucumber plants
have been evaluated and compared, and findings provided support for a hypothesized role
of auxin as a regulator of sex in cucumber (Galun et al., 1965; Rudich et al., 1972).
However, studies demonstrating the ability of auxin to stimulate ethylene production via
induction of ethylene biosynthetic enzyme, ACC synthase (ACS), in plant tissues led to
fiirther characterization of the feminizing effect of auxin (Trebitsh et al., 1987).
Treatment with auxin was found to stimulate ethylene in gynoecious and monoecious
cucumber, and, in addition, the feminizing effect of exogenous auxin treatment could be
blocked by treatment with inhibitors of ethylene action or biosynthesis while treatment of
gynoecious cucumber with antiauxins failed to inhibit female flower formation.
Therefore, there is strong evidence suggesting that the feminizing effect of auxin is
mediated via ethylene.

Most recently, the feminizing effect of brassinosteroids (BR) on sex expression
has been evaluated in cucumber, melon, and zucchini (Papadopoulou, 2002). Upon
treatment of seedling apices with a synthetic BR, epibrassinolide, monoecious cucumber
showed earlier and increased female bud formation, while sex expression of zucchini and

melon plants was not affected. Evaluation of ethylene production demonstrated increased

 

 

ethylene 1
expressiOJ
Or
hormone n
there are bc
other, when
range ofhor
ofont? mm“
other, Yin am
in the next 56C
different genot
that the com bin
led them to con.
therefore. be the
femaleness is inc
(St), while any le
Conversely. male:
Sensitivity (Sm). u
development. The

Ha-UICI'V F f and V ~71

ethylene production from BR treated seedlings, suggesting that the effect on sex
expression may be a result of increased ethylene production.

One possible mechanistic model for the regulation of sex expression is the one
hormone model proposed by Yin and Quinn (1992, 1995). According to this model,
there are both male and female receptors for a hormone to inhibit one sex and induce the
other, whereby differing sensitivity or levels of expression of the receptors respond to a
range of hormone concentrations to regulate sex expression (Figure 1.1). By this model
of one hormone producing a dual response of inhibition of one sex and induction of the
other, Yin and Quinn (1995) identified ethylene (which will be described in greater detail
in the next section) and not GA as the ‘sex hormone’ for cucumber by evaluating
different genotypes and their various hormone treatments. Furthermore, results showing
that the combined GA/ethylene treatment had results similar to ethylene treatment alone,
led them to conclude that ethylene had a more direct effect on sex expression and must,
therefore, be the ‘sex hormone’ (Yin and Quinn, 1995). According to their explanation,
femaleness is induced by an ethylene level higher than a threshold of female sensitivity
(Sf), while any level below Sf results in the lack of female organ development.
Conversely, maleness is inhibited by an ethylene level higher than a threshold of male
sensitivity (Sm), while levels lower than Sm result in induction of male organ
development. The mechanism of two major loci controlling sex expression in cucumber,
namely F -f and M-m, which have long been regarded as critical loci in sex determination
can be interpreted based on this model (Galun, 1961; Kubucki, 1969; Shifiiss, 1961).
The M-m gene is suggested to control the male sensitivity level such that the M allele

represents Sm at a lower level of hormone concentration (higher sensitivity)

 

 

MonO

AndfOr

GynOCCl

Hennaphj

Figure 1.1. W
cucumber based
represents the ph.
box represents the
depicts the threshc

development. Sm ‘

the genotype of the.

(figure adapted from

 

Monoecious W-

 

 

 

 

 

 

 

 

 

 

 

 

3 >
Sf Sm
luau") ‘____
Andromonoecrous firm I 1 4
Sm Sf
‘— ........ )
G oecious FFM- I -
y“ E- - _, -3 ,
Sf Sm
........ ) ‘—
Hermaphrodite FF mm [L ....... f3 ’

Figure 1.1. Yin and Quinn model for the effects of ethylene on sex expression in
cucumber based on the predicted roles of the F and M loci. The long horizontal arrow
represents the physiologically possible range of ethylene concentration. The rectangular
box represents the range of ethylene actually produced, as controlled by the F locus. Sf
depicts the threshold of ethylene production which must be exceeded to promote carpel
development. Sm represents the threshold of ethylene sensitivity, which is dependent of
the genotype of the M locus, and must be exceeded to inhibit stamen development.

(figure adapted fiom Yin and Quinn, 1995)

10

 

 

while the n
senSIfifl'ty).
ParticulaI ’9‘
allele l0 3 IO
Suggestaj IO

Wehnefi 199‘

Role of ethylf
The fer

has been of top
ethylene-r61 easi
useful method fl
andromonoeciou
application of inl
flower formation
melon and is, ther
and Visserl980; (
aIHIDOCIhOX}-'Vln)'l ,
biosynthesis, has be
IRUdICll, I990; Atsn

With the val L

has focused on eralu.’

I0 their subsequent eff

while the m allele represents Sm at a higher level of hormone concentration (lower
sensitivity). The F -f gene is suggested to control the normal range of ethylene for a
particular genotype such that the F allele leads to a higher level of ethylene and the f
allele to a lower level. Further interaction of the F -locus with the m and a loci has been

suggested to contribute to the range of sexual phenotypes seen in cucumber (Pierce and

Wehner, 1990).

Role of ethylene in sex expression

The feminizing effect of ethylene on sex determination in cucumber and melon
has been of topic of extensive study that is still under investigation. Application of the
ethylene-releasing compound, 2-chloroethanephosphonic acid (ethrel), to plants is a
useful method for increasing the number of female flowers in monoecious and
andromonoecious lines (Robinson et al., 1969; Rudich et al., 1969). Conversely,
application of inhibitors of ethylene perception, such as silver ions, induces staminate
flower formation in gynoecious cucumber and hermaphrodite flowers in gynoecious
melon and is, therefore, used to self-pollinate and maintain female genotypes (Den Nijs
and Visser,l980; Owens, 1980). In addition, application of exogenous
aminoethoxyvinyl glycine (AVG) or or-aminoxyacetic acid (AOA) which inhibit ethylene
biosynthesis, has been shown to increase maleness in cucumber and melon plants
(Rudich, 1990; Atsmon and Tabbak, 1979).

With the value of such applications recognized by breeders, considerable effort
has focused on evaluating the effectiveness of treatment timing and quantity with regard

to their subsequent effect on sex (Rudich, et al., 1969; Byers et al., 1972; Den Nijs and

11

V1555” 1

1979)- V
coLVIE‘dOr
leaf stage
et al., 19?
stage with
foUITh nod‘

treatments :

appearanCe

 

whey nodef
LhCSC studies
positions affe
hormone treat
subsequent dil
upon treatment

demonstrated tl

 

timing of stame
Early ext

endogenous eth v

melon. Studies or

g}noecious cucuml

‘ asstitrated with the.

l
l
) ethylene producti
l
l

Visser, 1980; Iwahori et al., 1970; McMurray and Miller, 1968; Tolla and Peterson et al.,
1979). While no effect of sex was observed following treatment with ethrel at the
cotyledon stage, treatment of monoecious cucumber seedlings with 50 ppm at the first
leaf stage reduced the appearance of the first pistillate flower from node 13 to 4 (Iwahori
et al., 1970). In contrast, treatment of gynoecious cucumber seedlings at the fourth leaf
stage with 50 ppm silver nitrate was sufficient for induction of staminate flOwers at the
fourth node and the four subsequent nodes (Tolla and Peterson, 1979). Additional
treatments and/or increased concentration did not result in earlier staminate flower
appearance but did prolong the phase of staminate flower production, suggesting that the
earlier nodes had developed beyond the critical stage for sex determination. Furthermore,
these studies have demonstrated that depending on the timing on application, the node
positions affected varies, indicating that there are critical times in development when
hormone treatment can have an effect. More recently, microscopic evaluation of the
subsequent differentiation of developing monoecious and gynoecious cucumber buds
upon treatment with ethylene-releasing ethephon or ethylene production inhibitor, AVG,
demonstrated that the influence of ethylene on sex determination is concurrent with the
timing of stamen primordia differentiation (Y arnasaki et al., 2003 a).

Early experiments to establish whether or not a relationship existed between
endogenous ethylene and sex determination in cucurbits consisted of quantification of
ethylene production by gas chromatography in various sex genotypes of cucumber and
melon. Studies comparing ethylene production from apices of monoecious and
gynoecious cucumber seedlings have shown two— to three-fold higher ethylene levels

associated with the increased femaleness of gynoecious lines (Rudich et al., 1972, 1976;

12

 

 

direct rel

Presence of

Makus ‘3‘ a
genotwes j
reduce end
that endoge]
The I
feminizing e
two Stet) pr 0C
mediated by 4
oxidase (ACO
as a key factor
and A C 0 are 3‘
induced by a V3?
ethylene. wound
I994; Bleeker an
The overa
cucumber. and ad'
Ethylene prod uctio

molecular level. II

ationship b

identification of a pa

her analysis ofne.‘

Makus et al., 1975). Despite the failure to observe a similar difference between sex
genotypes in melon, when melon seedlings were grown under hypobaric conditions to
reduce endogenous gas levels, gynoecious melon showed increased maleness, suggesting
that endogenous ethylene is important in promoting femaleness (Byers et al., 1972).

The understanding of ethylene biosynthesis has led to further evaluation of the
feminizing effect. Ethylene is synthesized from S-adenosyl methionine (AdoMet) in a
two step process which includes cyclization to l-aminocyclopane carboxylate (ACC),
mediated by ACC synthase (ACS), followed by an oxidation step mediated by ACC
oxidase (ACO) (Adams and Yang, 1979). The level of ACC synthase has been proposed
as a key factor in regulating ethylene biosynthesis (Yang and Hoffman, 1984). Both ACS
and ACO are generally encoded by multigene families in which different genes are
induced by a variety of environmental and developmental stimuli, such as: auxin,
ethylene, wounding, fi'uit ripening, pathogens, and stress (Zarembinski and Theologis,
1994; Bleeker and Kende, 2000).

The overall consensus of the key role of ethylene in the sex determination of
cucumber, and advancements in the understanding of biochemistry and regulation of
ethylene production, have led to fitrther investigation to link the two together at a
molecular level. Work in cucumber by Trebitsh et a1. (1997) provided evidence for a
direct relationship between the F -locus and ethylene levels. They reported the
identification of a partial ACC synthase sequence in cucumber (CS-A CS1 ), which upon
further analysis of near isogenic lines by Southern blot hybridization revealed the
presence of an additional copy of ACC synthase (CS-A CS1 G) present only in gynoecious

cucumber lines. Linkage analysis indicated co-segregation of CS-A CSI G with the F -

l3

 

locus with
duplicated
locus using
continued
ACSIG (M
suggest tha‘
exclusive e:
Resr
mammcDN
Since previo
from apices 2
expression of
cucumber. A;
apices in Whig
I0 Trebitsh (‘5
I:UI'ther
and CS~A C52 ,
apices oflSOger

locus with 0% recombination leading the authors to speculate the possibility of the
duplicated ACC synthase being the F -locus. Recent molecular characterization of the F
locus using RF LP (restriction fragment length polymorphisms) analysis firrther
confirmed the presence of a duplication of CS-A CS] in gynoecious genotypes, CS-
ACSI G (Mibus and Tatlioglu, 2004). Subsequent cloning and sequence evaluation
suggest that variation in the promoter region of CS-A CS1 G is responsible for its
exclusive expression in genotypes with a dominant F allele (Mibus and Tatlioglu, 2004).

Results published by Kamachi et a1. (1997) reported the cloning of the three
distinct cDNAs encoding ACC synthase (CS-A CSZ, CS-A CS3, CS-A CS4) in cucumber.
Since previous research indicated a strong correlation between the evolution of ethylene
fiom apices and femaleness (Rudich et al., 1972, 1976), Kamachi et a1. (1997) analyzed
expression of the three ACC synthase genes at the apices of monoecious and gynoecious
cucumber. Among the three ACS genes, only CS-A CS2 mRNA was detected at the
apices in which female flowers were developing while CS-A CS3 (with sequence identical
to Trebitsh CS-A CSI ) and CS-A CS4 were not detected.

Further research focused on unraveling the correlation of CS-A CS1 (CS-A CS 1 G)
and CS-A CSZ with sex determination in cucumber by examining their expression in the
apices of isogenic monoecious and gynoecious lines (Kamachi et al., 2000). While CS-
ACSI and CS-A C52 transcripts were detected at the apices of the gynoecious line, only
CS-A CSZ transcript could be detected in that of the monoecious line. In addition, CS-
ACSZ was limited to floral buds that would develop into female flowers. These results

suggested that ethylene produced as a result of CS-A CSZ expression was responsible for

14

 

inducing the
may hasten 3
AC C
also been Slur
et al., 1999; P
distinct cDNA
leaves and sho
transcript level
one may predic
relative to the l(
However. rm.
Correlation to tht
higher/1C0.” let
iIldicate that the 5
other than ACO (S

genot3pes (Kai) an;

inducing the development of female flowers while CS-A CS1 (CS-A CS1 G) expression
may hasten and increase CS-A C52 expression in gynoecious apices.

ACC oxidase, the enzyme that catalyzes the final step of ethylene formation, has
also been studied at a molecular in an effort to interpret its role in sex expression (Kahana
et al., 1999; Perl-Treves et al., 2000). Kahana et al (1999) reported the cloning of three
distinct cDNAs of ACC oxidase (A CO-I, ACO-2, ACO-3), which are all expressed in the
leaves and shoot apices of cucumber plants. Further evaluation of A C02 showed higher
transcript levels in the first unfolded leaf of gynoecious and hermaphrodite plants, which
one may predict were associated with the higher ethylene evolution of female genotypes
relative to the lower transcipt levels measured in monoecious and androecious plants.
However, AC02 transcript levels measured in the shoot apices showed an inverse
correlation to the sexual genotype; therefore, the monoecious and androecious plants had
higher AC02 levels in their apices. The authors concluded that these findings might
indicate that the shoot apex is not the site critical for ethylene synthesis or gene products
other than ACO (such as ACS) are responsible for increased ethylene levels in female
genotypes (Kahana et al., 1999; Perl-Treves et al., 2000).

Studies also have focused on gaining insight into the signal transduction
mechanism of ethylene on the promotion of femaleness in cucumber (Y amasaki et al.,
2000). The signal transduction pathway of ethylene has been well studied in Arabidopsis
thaliana by evaluating mutants with altered perception or response to ethylene (Bleeker
and Kende, 2000; Steptanova and Ecker, 2000; Wang et al., 2002). A multigene family
of ethylene receptors (in Arabidopsis E T R1 , E T R2, ERSI, ERSZ, and EIN4) was

identified by this approach, which upon cloning and characterization was found to share a

15

 

 

 

 

 

 

  
  
  
  
  
   
    
 
  
  

high degree C
contain hIStid

polilxv’l’tide (IE

 

in order t0 trig
ethylene reci’l‘I
binding are use

by ethylene (B1 I

were elevated in
CS—ERS were mor

analogous to that o

in the signal transdu

In a related st
Provided insight into
determination (Yamat
locus confers bisexual

Immfl) treatment with

 

del'elopment. In a con

 

 

high degree of homology to bacterial two-component signal transduction proteins that
contain histidine kinase (sensor) and receiver (response regulator) domains on the same
polypeptide (Bleeker et al., 1988; Chang et al., 1993). Since ethylene must be perceived
in order to trigger a downstream ethylene response, studies evaluating the expression of
ethylene receptors as well as implementing receptor mutants to confer reduced ethylene
binding are useful for furthering the understanding of processes in plants that are affected
by ethylene (Bleeker and Kende, 2000).

To this end, ethylene receptor homologues, CS—E T R1 , CS—E T R2, and CS-ERS
were cloned from cucumber and their expression pattern analyzed in monoecious and
gynoecious cucumber (Y amasaki et al., 2000). While transcript levels of all three genes
were elevated in gynoecious relative to monoecious lines, differences in CS-E T R2 and
CS-ERS were more pronounced and followed a time course of expression in shoot apices
analogous to that of CS-A CS2, ethylene production, and female flower development. In
addition, upon treatment with ethrel or AVG, CS—E T R2 and CS-ERS exhibited subsequent
changes in expression reflecting their regulation by ethylene. Furthermore, since receptor
binding of ethylene initiates a signaling cascade, CS—E T R2 and CS-ERS may be important
in the signal transduction involved in cucumber sex determination.

In a related study, the expression patterns of CS-A C52, CS—E T R2, and CS-ERS
provided insight into the underlying function of the M locus in cucumber sex
determination (Y arnasaki et al., 2001). As already described, loss-of-function at the M
locus confers bisexual flower development, such that in andromonoecious cucumber
(mmfl) treatment with ethylene induces pistil development but does not inhibit stamen

development. In a comparison of monoecious (M-fl), gynoecious (M-F-) and

16

 

andromonomc
cs-ErRJ. C54
mmff. These fit
Supports implic
To furth
ethylene, a serie
CS-ETRJ, and C
by in situ hybrid:
CS-ACS.’ mRNA
was suggested thz
perception are ide
detected in petals.
suggested that for .
Perception need no
Another stu
gene ertpression anc
Obsm'ed to promote
pruduction and exp”
POW 0f 8 or 16 hour

demonstrating a photr

ethylene evolution an

bu
t such a phenomeno

andromonoecious (mmfi) cucumber genotypes, ethylene treatment resulted in increased
CS—ET R2, CS-ERS, and CS-A CSZ transcript levels in M—fl” and M-F— plants but not in
mmfli These findings suggest that response to ethylene is reduced in mmfl' plants and
supports implication of the M locus in the mediation of ethylene signals.

To further elucidate the relationship between tissues producing and perceiving
ethylene, a series of experiments characterized the localization of CS—A CS2, CS-E T R1 ,
CS-E T R2, and CS-ERS transcripts within cucumber flower buds throughout development
by in situ hybridization analysis (Yamasaki et al., 2003a). Since overlap of CS-ETR2 and
CS-A C52 mRNA was observed in pistil regions of monoecious and gynoecious buds, it
was suggested that for the induction of pistils, tissues involved in ethylene production and
perception are identical. In addition, in monoecious buds, since CS-A CS2 message was
detected in petals, while CS-E T R] and CS—ERS were detected in the stamens, it was
suggested that for development of stamens, cells responsible for ethylene production and
perception need not be identical.

Another study has evaluated the environmental regulation of ethylene-related
gene expression and its correlation to sex expression as short day conditions have been
observed to promote femaleness (Y amasaki et al., 2003b). The peak of ethylene
production and expression of CS—A CS2, CS-A CS4, and CS—ERS were found at the mid
point of 8 or 16 hour light period in monoecious and andromonoecious cucumber
demonstrating a photoperiodic rhythm. However, in monoecious cucumber, the peak of
ethylene evolution and CS-A C52 was greater under short-day conditions than long-days,

but such a phenomenon was not seen in andromonoecious plants. These results provide

17

 

support for Cm)l

short day 00mm

ObjeCtives of th(

AS is SUQ‘S
demonstrated to P
studies to elucidat
cucumber and ban
analyses of ethleF
laboratory IO Pmd“
endogenous ethyl er
determination of se‘

The signal tr
thali‘ana by evaluati:
characteristic of the 1
mutant causing ethyl
Use as a tool to evalu.
our laboratory. stud yr
acoustitutive (355-0

ethv ' '
,lene senSitn'itv. i r

 

 

a g . .
bseismon. and fail ur
L.

addition, Papadopoult

support for ethylene playing a role in the modulation sex expression in cucumber under

short day conditions.

Objectives of the thesis research

As is suggested in the preceding literature review, even though ethylene has been
demonstrated to play an important role as a “sex hormone” in muskmelon, much of the
studies to elucidate the role of ethylene in sex determination of cucurbits have focused on
cucumber and have relied on correlative studies of exogenous ethylene treatment or
analyses of ethylene-related gene expression. With reliable methodology available in our
laboratory to produce transgenic melon, the opportunity exists to directly modify
endogenous ethylene production and perception in order to further evaluate its role in the
determination of sex.

The signal transduction pathway of ethylene has been studied in Arabidopsis
thaliana by evaluating mutants with altered perception or response to ethylene. The
characteristic of the ethylene receptor mutant, etr] -1, to act as a dominant negative
mutant causing ethylene insensivity (Bleeker et al., 1988; Chang et al., 1993), enables its
use as a tool to evaluate processes in plants that are affected by ethylene. Recent work in
our laboratory, studying transgenic melons haboring the etrI-I receptor mutant driven by
a constitutive (3SS-CaMV) promoter, showed phenotypes characteristic of decreased
ethylene sensitivity, including: poor root formation, delayed flower senescence and
abscission, and failure to respond to exogenous ethylene (Papadopoulou, 2002). In

addition, Papadopoulou (2002) showed that 35S::etrI-I melon fail to produce pistillate

18

 

 

buds. demonsti
floral org‘mS'
Thereft‘
of ethylenc 13¢r
subsequent sex
alter ethylene p ‘
isolated from Ar
petals and stamt"
also isolated fro”
nectaries ofgmot
Predicted r
tipe melon and cut
changes from a pha
1999), suggesting tl
unaffected. ln cucu:
tlower production ( F
While stamen det'elo;

5l’eCulate that percept

 

buds, demonstrating that the perception of ethylene is critical for development of female
floral organs.

Therefore, one objective of my research was to firrther evaluate the requirement
of ethylene perception by specific regions of the developing floral primordia for
subsequent sex determination in melon. Constructs for transformation were prepared to
alter ethylene perception in stamens or carpels. The Apetela3 promoter, which was
isolated from Arabidopsis thaliana, is targeted to primordial cells destined to become the
petals and stamens (Irish and Yamamoto, 1995). The Crab ’s Claw promoter, which was
also isolated from Arabidopsis thaliana, is specifically expressed in the carpels and
nectaries of gynoecial primordia from their inception (Bowman and Smyth, 1999).

Predicted results were based on the resulting changes in sex expression in wild-
type melon and cucumber upon the addition of ethylene. In melon, sex expression
changes from a phase of staminate to a phase of perfect flower production (Perl-Treves,
1999), suggesting that carpel development is induced while stamen development is
unaffected. In cucumber, sex expression changes from a phase of staminate to pistillate
flower production (Perl-Treves, 1999), suggesting that carpel development is induced
while stamen development is suppressed. Therefore, in the flower bud of melon, one can
speculate that perception of ethylene is critical in the carpel while in cucumber,
perception is critical in both the carpel and stamen. Table 1.1 summarizes the
hypothesized results of experiments using plants transformed with constructs of the air]-

1 mutant and the subsequent effects on sex expression.

19

 

Table 1.1- 5‘

transgenic mel

 

M y second
ethylene productior
(producing male an
transformed to cons
smthase) gene from
production in melon
ethylene evolution f
overexpressed .4 C S

of the sex expressio

 

hermaphroditic bud .~

 
 
 
 
 

altey role in the con
it was observed that '

Interestingly. howe\

Table 1.1. Summary of etrI -1 constructs and predicted effects on sex expression in

 

 

 

 

 

 

 

transgenic melon.
wild-type Staminate to perfect
1. carpel development
2. stamens unaffected
35S-etr1-1 Staminate only
(constitutive) 1. no carpel development
2. stamens unaffected
AP3-etr1-1 Staminate and Perfect
(petal/stamen) 1. carpel development
2. stamens unaffected
CRC—etrI-I Staminate only
(carpel/nectary) 1. no carpel development
2. stamens unaffected

 

My second objective was to further evaluate the effects of increased endogenous
ethylene production on sex expression in melon. Recently, an andromonoecious
(producing male and hermaphroditic flowers) melon genotype, Hale’s Best Jumbo, was
transformed to constitutively express an ACS (1 -aminocyclopropane-1-carboxylic acid
synthase) gene from petunia in order to study the effect of modified endogenous ethylene
production in melon (Papadoupoulou et al., 2005). Greenhouse studies showed increased
ethylene evolution from leaves and floral buds of transgenic plants confirming that the
overexpressed ACS conferred increased ethylene biosynthesis. Evaluation of the pattern
of the sex expression along the main stem showed both earlier and increased numbers of
hermaphroditic buds relative to non-transgenic controls, consistent with ethylene playing
a key role in the commitment to form pistillate flowers. In performing these experiments,
it was observed that most hermaphrodite buds abort before reaching anthesis.

Interestingly, however, an increased percentage of hermaphroditic buds on the transgenic

20

 

plants reachcc
developmtfnt 5
observations 0
characterizatio:
levels on fruit p
The thin
effects on plant 1
an emphasis on e
ethylene producti
characteristics inc
tum result in secor
changes in aroma
in mm. alter the lik
and 'or alter reprodr

be important to co

 

 

plants reached firll anthesis, suggesting a role for ethylene in hermaphrodite bud
development as well as hermaphrodite bud initiation. To further substantiate greenhouse
observations on sex determination and flower development and to continue
characterization of the transgenic melon to include the influence of modified ethylene
levels on fi'uit production, I conducted a field evaluation of the 35S—A CS melon lines.
The third objective of this thesis was to provide a case study of the secondary
effects on plant phenotypes associated with genetically modified ethylene signaling, with
an emphasis on effects that may be relevant in terms of risk assessment. For example,
ethylene production and perception have been modified to alter plant physiological
characteristics including ripening, senescence, and flowering. These alterations can in
turn result in secondary effects such as, altered stress and disease response as well as
changes in aroma, floral characteristics, and growth habit. Such secondary effects may,
in turn, alter the likelihood of gene flow into neighboring non-transgenic populations
and/or alter reproductive fitness of arising hybrids in the event of gene flow and so may

be important to consider from an environmental risk assessment standpoint.

21

 

 

 

References
Adams. 00. an

aminocy
methioni

Ainsworth C.. Pa

Curr. To;ll

Ando, 5.. Sato. Y
(HUI .7) a
female flo
943-952.

Atsmon. D. 1965

Stem elongl
882.

Atsmon. D. and Q
hmaphfllk

Atsmon, D. and Ca
Patterns antI

Am“ D. and T.
- ‘l
ammOCllle
(CUCumis 5U

Bai, S,L_, Pen 2. “l

2004' DCVcl

reprOdUClch I

   
  

 

 

References

Adams, DD. and Yang, SF. 1979. Ethylene biosynthesis: identification of 1-
arninocycloprpane-l-carboxylic acid as an intermediate in the conversion of
methionine to ethylene. Proc. Natl. Acad. Sci. 76: 170-174.

Ainsworth C., Parker J ., Buchanan-Wollaston V. 1998. Sex determination in plants. In
Curr. Topics Devel. Biol., Vol 38. Academic Press, San Diego, ppl67-223.

Ando, S., Sato, Y., Kamachi, S., and Sakai, S. 2001. Isolation of a MADS-box gene
(ERAI 7) and correlation of its expression with the induction of formation of

female flowers by ethylene in cucumber plants (Cucumis sativus L.). Planta 213:
943-952.

Atsmon, D. 1968. The interaction of genetic, environmental, and hormonal factors in
stem elongation and floral development of cucumber plants. Ann. Bot. 32: 877-
882.

Atsmon, D. and Galun E. 1960. A morphological study of staminate, pistillate, and
hermaphrodite flowers in Cucumis sativus L. Phytomorphology 10: 110-115.

Atsmon, D. and Galun E. 1962. Physiology of sex in Cucumis sativus L. Leaf age
patterns and sexual differentiation of floral buds. Ann. Bot. 26: 137-146.

Atsmon, D. and Tabbak C. 1979. Comparative effects of gibberellins, silver nitrate and
aminoethoxyvinyl glycine on sexual tendency and ethylene in the cucumber plant
(Cucumis sativus L.). Plant Cell Physiology 20: 1547-1555.

Bai, S.L., Peng, Y.B., Cui, J .X., Gu, H.T., Xu, L.Y., Li, Y.Q., Xu, Z.H., and Bai, SN.
2004. Developmental analyses reveal early arrest of the spore-bearing parts of
reproductive organs in unisexual flowers of cucumber (Cucumis sativus L.).
Planta 220: 230-240.

Barrett, S. 2002. The evolution of plant sexual diversity. Nature Reviews 3: 274-284.
Bleeker, A.B., Estelle, M.A., Somerville, C., and Kende, H. 1988. Insensitivity to
ethylene conferred by a dominant mutation in Arabidopsis thaliana. Science 241:

1086-1089.

Bleeker, AB. and Kende, H. 2000. Ethylene: a gaseous signal molecule in plants.
Annu. Rev. Cell Dev. Biol. 16: 1-18.

Bowman, J .L. and Smyth DR. 1999. CRABS CLA W, a gene that regulates carpel and

nectary development in Arabidopsis, encodes a novel protein with zinc finger and
helix-loop-helix domains. Development 126: 2387-2396.

22

 

 

Byers9 R.E.
natur
717-".

Chang. C.. K
ethyl:
regul;

Dellaporta, S
The P

Den Nijs, A. z
(Cucw

Frankel, R. an.
Breedir

Galun. E. 1959
12: 48-6

Galun, E. 1961
interactit
Genetica

Galun, 5.. lshar,

hermaphr
326.

Cajuns E-g Yung, ‘
CUllUred in

Gofilnet’ «\4 C 19'

satitw. In:
RoblnSOn. a

 

Byers, R.E., Baker, L.R., Sell, H.M., Hemer, RC, and Dilley, DR. 1972. Ethylene: a
natural regulator of sex expression of Cucumis melo L. Proc. Natl Acad. Sci. 69:
717-720.

Chang, C., Kwok, S.F., Bleeker, A.B., and Meyerowitz, EM. 1993. Arabidopsis
ethylene-response gene E T R] : similarity of products to two-component
regulators. Science 262: 539-544.

Dellaporta, S.L. and Calderon-Unea, A. 1993. Sex determination in flowering plants.
The Plant Cell 5: 1241-1251.

Den Nijs, A. and Visser, D. 1980. Induction of male flowering in gynoecious cucumbers
(Cucumis sativus L.) by silver ions. Euphytica 29: 237-280.

Frankel, R. and Galun, E. 1977. Pollination Mechanisms, Reproduction and Plant
Breeding. Springer-Verlag, Berlin.

Galun, E. 1959. The role of auxin in the sex expression of the cucumber. Physiol. Plant.
12: 48-61.

Galun, E. 1961. Study of the inheritance of sex expression in the cucumber, the
interactions of major genes with modifying genetic and non-genetic factors.
Genetica 32: 134-163.

Galun, E., Ishar, S., and Atsmon, D. (1965) Determination of relative auxin content in
hermaphrodite and andromonoecious Cucumis sativus L. Plant Physiol. 40: 321-
326.

Galun, E., Yung, Y., and Lang, A. 1963. Morphogenesis of floral buds of cucumber
cultured in vitro. Developmental Biol. 6: 370-387.

Goffinet, MC. 1990. Comparative ontogeny of male and female flowers in Cucumis
sativus. In: Biology and Utilization of the Cucurbitaceae (eds DM Bates, RW
Robinson, and C Jeffrey). Cornell University Press, New York, pp 288-304.

Grant, 8., Houben, A., Vyskot, B., Siroky, J ., Pan, W.H., Macas, J ., and Saedler, H.
1994. Genetic of sex determination in flowering plants. Developmental Genetics
15: 214-230.

Hao, Y., Wang, D., Peng, Y., Bai, S., Xu, L., Li, Y., Xu, 2., and Bai, S. 2003. DNA
damage in the early primordial anther is closely correlated with stamen arrest in
the female flower of cucumber (Cucumis sativus L.). Planta 217: 888-895.

Iwahori, S., Lyons J .M., and Smith QB. 1970. Sex expression in cucumber plants as

affected by 2-chloroethylphosphonic acid, ethylene, and growth regulators. Plant
Physiol. 46: 412-415.

23

311111;. and Yamam‘
ln betwefin Arabld

lackl. 2000. Relearr
Science 6: 310-

luau C. and Banks .1.
Biology 1: 6S-

Kanachi. 5.. Seltimoto
aminocycloprur
of female 110M
1197-1206.

Kamachi. S.. Mizusaw
arninocyclopro]
correlated with
Biotechnology

Kahana A.. Silbersteir
ofACC oxidas
Plant Molecula

Rater. MM" Pranken.
determination i
whorls. Plant C

Kerigsbuch. D. and C t

Genome 33: 31

mes com;
M9115 [11».th D
between mODO
Production [)1
h‘lCMufiay‘ AL an ‘
Cthl'Oeth
1.1
thus. mam Tatliog'
e fOT fe \
No ‘1
“k 'and Yan .
Plant 0&1“

Irish, V.F. and Yamamoto, Y.T. 1995. Conservation of floral homeotic gene function
between Arabidopsis and Antirrhinum. Plant Cell Rep. 9:160-164.

Jack, T. 2000. Relearning our ABCs: new twists on an old model. Trends in Plant
Science 6: 310-316.

Juarez, C. and Banks J .A. 1998. Sex determination in plants. Current Opinion in Plant
Biology 1: 68-72.

Kamachi, S., Sekimoto, H., Kondo, N., and Sakai, S. 1997. Cloning of a cDNA for a 1-
aminocyclopropane-l -carboxylate synthase that is expressed during development

of female flowers at the apices of Cucumis sativus L. Plant Cell Physiology 38:
1197-1206.

Kamachi, S., Mizusawa, H., Matsuura, S., and Sakai, S. 2000. Expressionof two 1-
aminocyclopropane-carboxylate synthase genes, CS-A CS1 and CS-A C52,
correlated with sex phenotypes in cucumber plants (Cucumis sativus L.). Plant
Biotechnology 17: 69-74.

Kahana, A., Silberstein, L., Kessler, N., Goldstein, RS, and Perl-Treves, R. Expression
of ACC oxidase genes differs among sex genotypes and sex phase in cucumber.
Plant Molecular Biology 41: 517-528.

Kater, M.M., Franken, J ., Carney, K.J., Clombo, L., Angenent, G.C. 2001. Sex
determination in the monoecious species cucumber is confined to specific floral
whorls. Plant Cell 13: 481-493.

Kenigsbuch, D. and Cohen, Y. 1989. The inheritance of gynoecy in muskrnelon.
Genome 33: 317-320.

Kubicki, B. 1969. Investigations on sex determination in cucumber (Cucumz's sativus L.)
V. Genes controlling intensity of femaleness. Genet. P01. 10: 69-85.

Makus, D.J., Pharr, D.M., and Lower, R.L. 1975. Some morphogenic differences
between monoecious and gynoecious cucumber seedlings as related to ethylene
production. Plant Physiol. 55: 352-355

McMurray, AL. and Miller, CH. 1968. Cucumber sex expression modified by 2-
chloroethane phosphonic acid. Science 162: 1397-1398.

Mibus, H. and Tatlioglu, T. 2004. Molecular characterization and isolation of the F/f
gene for femaleness in cucumber (Cucumis sativus L.). Theor. Appl. Genet. 109:
1669-1676.

Ng, M. and Yanofsky, M.F. 2000. Three ways to learn the ABCs. Current Opinion in
Plant Biology 3: 47-52.

24

Owens. K.W.. Peterso r
on gsnoecious :
HortScience 15

 

Papadopoulou E. 200:
perception and >
University.

Papadopoulou E.. Littl'c
endogenous eth}!
production in mtl

Perl-lrei'es. R. 1999. _\
to study sex geneI
Determination ir.

Perl-Trees. R.. Kahana.
three cDNAs enct
cucumber floral b

Perl-lrei-es. R“ Kahana
EXPression of mu
cucumber (01(th

Perl-lret’es. R.. Sanf-Le‘
Gal-On. A. 2000.
cucumber. ln: Prt
Acta Hort. pp 463-

Pl .‘ r
We. Us. and \l'ehner. '
HonScience 25: 61 .

 

Owens, K.W., Peterson, CE. and Tolla, GE. 1980. Prodution of hermaphrodite flowers
on gynoecious muskrnelon by silver nitrate and aminoethyoxyvinyl glycine.
HortScience 15: 654-655.

Papadopoulou, E. 2002. Sex expression in cucurbits: the role of ethylene synthesis and
perception and sex determination genes. PhD Dissertation. Michigan State
University.

Papadopoulou, E., Little, H.A., Harnmar, S.A., and Grumet, R. (2005) Effect of modified
endogenous ethylene production on hermaphrodite flower development and fruit
production in melon (Cucumis melo L.). in submission.

Perl-Treves, R. 1999. Male to female conversion along the cucumber shoot: approaches
to study sex genes and floral development in Cucumis sativus. In: Sex
Determination in Plants (ed CC Ainsworth). BIOS Scientific, Oxford pp 189-215.

Perl-Treves, R., Kahana, A., Rosemann, N., Yu, X., and Silberstein, L. 1998. Cloning of
three cDNAs encoding l-aminocyclopropane-1-carboxylate oxidases from
cucumber floral buds. Plant Physiology 116: 1192.

Perl-Treves, R., Kahana, A., Rosemann, N.,Yu, X., Xiang, Y., and Silberstein, L. 1998.
Expression of multiple AGAMOUS-like genes in male and female flowers of
cucumber (Cucumis sativus L.). Plant Cell Physiol. 39: 701-710.

Perl-Treves, R., Sarif-Levy, T., Kahana, A., Kessler, N., Silberstein, L., Wang, Y., and
Gal-On, A. 2000. Genes involved in ethylene sythesis and perception in
cucumber. In: Proceedings of Cucurbitaceae 2000 (eds N Katzir and HS Paris).
Acta Hort. pp 463-469.

Pierce, L.K. and Wehner, TC. 1990. Review of genes and linkage groups in cucumber.
HortScience 25: 605-615.

Robinson, R.W., Shannon, S., and DeLaGuardia, MD. 1969. Regulation of sex
expression in the cucumber. BioScience 19: 141-142.

Roy, RP. and Saran, S. 1990 Sex expression in the Cucurbitaceae. In Biology and
utilization of the Cucurbitaceae (eds Bates, D.M., Robinson, R.W., Jeffrey, C.,
eds), pp. 251-268. Ithaca NY: Cornell Univ. Press.

Rudich, J. 1990. Biochemical aspects of hormonal regulation of sex expression in
cucurbits. In Biology and utilization of the Cucurbitaceae (eds Bates, D.M.,
Robinson, R.W., Jeffrey, C., eds), pp. 288-304. Ithaca NY: Cornell Univ. Press.

Rudich, J ., Halevy, A.H., and Kedar, N. 1969. Increase in femaleness of three cucurbits
by treatment with Etherel, an ethylene-releasing compound. Planta 86: 69-76.

25

 

Rudich. J., Hales): Al
as related to se.‘

Rudich. 1.. Baker. LR.
ethylene evolut

Sinfn'ss. O. 1961. Sex

Steptanova AN. and Er
Current Opiniur

Tanurdzic M. and Bank
Cell. 161561-57

Tolla 0E. and Petersor
for induction of:
14:542-544.

Trebitsh T.. Rudich. J .. .
expression in cuc

Trebitsh T.. Staub. J .E. z
aminocyclopropa

locus that enhanct
987-995.

Wang K.I.C.. Li. H.. and

networks. Plam (
Wilmer. SH an

E d BUkOV.‘

con. Bot. 12; a].

l'irnasakj g

Rudich, J ., Halevy, A.H., and Kedar, N. 1972. Ethylene evolution from cucumber plants
as related to sex expression. Plant Physiology 49: 998-999.

Rudich, J ., Baker, L.R., Scott, J .W., and Sell, HM. 1976. Phenotypic stability and
ethylene evolution in androecious cucumber. J. Am. Soc. Hortic. Sci. 181: 48-51.

Shifriss, O. 1961. Sex control in cucumbers. J. Hered. 52: 5-12.

Steptanova AN. and Ecker J .R. 2000. Ethylene signaling: from mutants to molecules.
Current Opinion in Plant Biology 3: 353-360.

Tanurdzic M. and Banks J .A. 2004. Sex-deterrnining mechanisms in land plants. Plant
Cell. 16:S61-S71.

Tolla, GE. and Peterson, CE. 1979. Comparison of gibberellin A4/A7 and silver nitrate
for induction of staminate flowers in a gynoecious cucumber line. HortScience
14: 542-544.

Trebitsh, T., Rudich, J ., and Rivoc, J. 1987. Auxin, biosynthesis of ethylene and sex
expression in cucumber (Cucumis sativus). Plant Growth Regul. 5: 105-113.

Trebitsh, T., Staub, J .E. and O’Neill, SD. 1997. Identification of a 1-
arninocyclopropane-l-carboxylic acid synthase gene linked to the Female (F)
locus that enhances female sex expression in cucumber. Plant Physiology 113:
987-995.

Wang K.L.C., Li, H., and Ecker, J .R. 2002. Ethylene biosynthesis and signaling
networks. Plant Cell Sl31-S151.

Wittwer, SH. and Bukovac, M.J. 1958. The effect of gibberellins on economic crops.
Econ. Bot. 12: 213-255.

Yamasaki, S., Fujii, N., and Takahashi, H. 2000. The ethylene-regulated expression of
CS-E T R2 and CS-ERS genes in cucumber plants and their possible involvement
with sex expression in flowers. Plant Cell Physiol. 41: 608-616.

Yamasaki, S., Fujii, N., Matsuura, S., Mizusawa, H., and Takahashi, H. 2001. The M
locus and ethylene-controlled sex determination in andromonoecious cucumber
plants. Plant Cell Physiol. 42: 608-619.

Yamasaki, S., Fujii, N. and Takahashi, H. 2003a. Characterization of ethylene effects on
sex determination in cucumber plants. Sex Plant Reproduction 16: 103-111.

Yamasaki, S., Fujii, N. and Takahashi, H. 2003b. Photoperiodic regulation of C5-A C52,
CS-A C54 and CS-ERS gene expression contributes to the femaleness of cucumber

26

 

flowers throng?
Cell. and Envir

Yang. SF. and Hoffnra
plants. Annu. 1i

 

Yin T. and Quinn. J.A.
sexes in floss-cm

Yin T.. and Quinn. 1 .A.
both sexes in ('1.

Zarembinslti. T.l. and Ti
conservation. Pl;

flowers through diurnal ethylene production under short-day conditions. Plant,
Cell, and Environment 26: 537-546.

Yang, SF. and Hoffman, NE. 1984. Ethylene biosynthesis and its regulation in higher
plants. Annu. Rev. Plant Physiol. 35: 155-189.

Yin, T. and Quinn, J .A. 1992. A mechanistic model of a single hormone regulating both
sexes in flowering plants. Bulletin of the Torey Botanical Club 119: 431-441.

Yin, T., and Quinn, J .A. 1995. Tests of a mechanistic model of one hormone regulating
both sexes in Cucumis sativus (Cucurbitaceae). Am. J. Bot. 82: 1537-1546.

Zarembinski, TI. and Theologis, A. 1994. Ethylene biosynthesis and action: a case of
conservation. Plant Mol. Biol. 26: 1579-1597.

27

 

Modifi
un

Abstract

The diversif3 0“
'0)- emjronmental. horm
aieyhonnonal regulatc
melo L.) plants such thLi‘
hermaphrodite flower pr
that Ethylene is critical ft
lnthis study. I sought to
developing flower bud.

expressing the Arabidops

of '
.umohpes consrstent xvi

Patten showed almost ex
0f99.99116 Perception in

3999011 cts under the

“3191 get developing st
f8Mirelv Gr

OUSCI

lurid
eds srupnsmg results

thinn- mwb
3 earlier and me

Chapter 2
Modified sex expression in transgenic melon expressing
the mutant ethylene receptor, At-etrI-I,

under the control of floral targeted promoters

Abstract

The diversity of sexual phenotypes observed within cucurbit species is influenced
by environmental, hormonal, developmental, and genetic factors. Ethylene is regarded as
a key hormonal regulator in the sex determination of andromonoecious melon (Cucumis
melo L.) plants such that application of ethylene causes a shift from staminate to
hermaphrodite flower production. These effects of exogenous ethylene treatment suggest
that ethylene is critical for carpel development, but does not affect stamen development.
In this study, I sought to determine the critical sites of ethylene perception in the
developing flower bud. A previous study of transgenic melon plants constitutively
expressing the Arabidopsis dominant mutant ethylene receptor, etrI-I, displayed
phenotypes consistent with ethylene insensitivity, and evaluation of their sex expression
pattern showed almost exclusively male flower production. To further decipher the role
of ethylene perception in melon sex determination, melon was transformed with etrI-I
gene constructs under the direction of Apetala3 (AP3) and Crab ’3 Claw (CRC) promoters
which target developing stamens and petals, and carpels and nectaries in Arabidopsis,
respectively. Greenhouse characterization of sex expression and floral morphology
provided surprising results. CRC::etr1-1 melon plants showed increased femaleness as

measured by earlier and increased pistillate bud production, and the formation of female

28

rathfir than hmaphrod

 

assured by nearly exc
dereloped carpels in the
experiments suggest a r.
by the stamen is require.

muted for stamen d“

Introduction

lhe C ucubitaceac'
phenonpes. has served a
determination. Cucurbit
and female flowers which

patterns. Cucumber (C no

rule flowers precedes a p

l’ .,
finale). androecious (all r

ltrinaphroditic t3pes also
reach is typi call 3' andomo
Itrons sexual phenot\pes

tar '
tronmental factors. Tht

almond

 

rather than hermaphrodite buds. AP3::etr1-1 melon plants showed increased maleness as

measured by nearly exclusive staminate flower production and the presence of poorly
developed carpels in the few pistillate buds that were produced. The results of these
experiments suggest a model for sex determination in melon whereby ethylene perception
by the stamen is required for carpel development, while ethylene perception by the carpel

is required for stamen development.

Introduction

The Cucubitaceae family, which is noted for diversity in sex expression
phenotypes, has served as a model for understanding of processes involved in sex
determination. Cucurbit species exhibit various combinations of hermaphrodite, male,
and female flowers which can influence outcrossing behavior and fi'uit set timing and
patterns. Cucumber (Cucumis sativus L.) is typically monoecious, in which a phase of
male flowers precedes a phase of both male and female flowers; however, gynoecious (all
female), androecious (all male), andomonoecious (male followed by perfect flowers), and
hermaphroditic types also exist (Rudich, 1990; Perl-Treves, 1999). Melon (C. melo),
which is typically andomonoecious, also exhibits a variety of sexual phenotypes. These
various sexual phenotypes are under the control of genetic, developmental, hormonal, and
environmental factors. The underlying molecular processes are now beginning to be
understood.

Sex expression in both cucumber and melon is influenced by at least three major
genes. In cucumber, sex is largely controlled by the F, M, and A loci (reviewed in Perl-

Treves, 1999). The partially dominant F (female) locus has been used commercially to

29

 

develop g}noecious CL
and Nienhuis, 1990). '
buds such that .11- cool

hermaphrodite buds (re

ncreased maleness suc

   
  
   
 

exclusively. Although 1
sexual thOUPC'S 1“ mi-
renot completely analr
member confem'n g g}'?

Developmentall}
physiological gradient' _
ofsuminate and pistillat
bisexual flowers in andro

Despite this obsen'ed ch
undated floral buds con
3515mm flower (Go f t
29‘33; Bai et al., 2004 ).
declopmem continues.
ill-ll? carpel developmc:
brrmocmutants of coca

first '
or rather than sema/ "

     

Itilefi .
011 m/fiy/E/I‘Om S‘t

develop gynoecious cultivars, which provide earlier and more uniform fi'uit-set (Lower
and Nienhuis, 1990). The M locus affects the development of stamens in pistillate floral
buds such that M— confers the development of unisexual, female buds while mm confers
hermaphrodite buds (reviewed in Perl-Treves, 1999). The A locus is associated with
increased maleness such that a plant with an aafl genotype produces staminate flowers
exclusively. Although three major loci are also primarily responsible for inheritance of
sexual phenotypes in melon (Kenigsbuch and Cohen, 1989; Roy and Saran, 1990), they
are not completely analogous, as an equivalent locus to the semi-dominant F locus in
cucumber conferring gynoecy has not been identified.

Developmentally, sex expression in cucumber and melon typically follows a
‘physiological gradient’ such that a phase of staminate flowers is followed by a mixture
of staminate and pistillate flowers in monoecious cucumber or a mixture of staminate and
bisexual flowers in andromonoecious melon (Atsmon and Galun, 1962; Shifriss, 1961 ).
Despite this observed chronology in flower sex, in early stages of development all
initiated floral buds contain the four whorls (sepals, petals, stamen, and carpels) typical of
a bisexual flower (Goffinet, 1990; Kater et al., 2001; Yarnasaki et a1, 2003; Hao etal.,
2003; Bai et al., 2004). The subsequent arrest of carpel primordia, while stamen
development continues, leads to a male flower; similarly, the arrest of stamen primordia,
while carpel development continues, results in a bisexual flower. An evaluation of floral
homeotic mutants of cucumber revealed that sex determination is dependent on the whorl
position rather than sexual identity, such that the prevention of carpel development in

male flowers results from suppression of differentiation in whorl 4, where carpels

30

normally form. Simil
stamens normally dex t

Detailed morp?
have divided the procc
into 12 stages to detenr

Bai et a1. 2004,). Dr" C

 

apparent morphological
of carpel primordia (Bai
Degas immediately follt
liferentiation of the inn
uoditierentiation of the
filament and anther is of
Primordial anther (Hao c
maturing arrest of inap;

dig . -
erentiatron follows If

The Cfi‘ects 0f ho

late A.
r“med Consideral

nonnally form. Similarly, female flowers result from suppression of whorl 3, where
stamens normally develop (Kater et al., 2001).

Detailed morphological studies of male and female cucumber bud development
have divided the process of floral development from floral meristem initiation to anthesis
into 12 stages to determine the precise timing of sexual differentiation (Hao et al., 2003;
Bai et al., 2004). Divergence from a bisexual to unisexual flower begins to become
apparent morphologically at stage 6 in both male and female buds following the initiation
of carpel primordia (Bai et al., 2004). In male flowers, the arrest of carpel primordia
begins immediately following its inception (Bai et al., 2004). Despite some subsequent
differentiation of the inappropriate carpel to form the nectary and stigma-like primordia,
no differentiation of the ovary was observed. In female flowers, differentiation of a
filament and anther is observed, followed by the detection of DNA damage in the
primordial anther (Hao et al., 2003). It was suggested that the response to the signal
conferring arrest of inappropriate male or female floral organs may differ because carpel
differentiation follows that of stamens (Bai et al., 2004).

The effects of hormones on sex determination of cucurbits, particularly cucumber
have received considerable attention (Rudich, 1990; Perl-Treves, 1999). Auxin, ethylene,
and brassinosteroids have been shown to promote femaleness while gibberellins promote
maleness (Wittwer and Bukovac, 1958; Galun, 1959; Atsmon, 1968; McMurray and
Miller, 1968; Rudich et al., 1969; Papadopoulou, 2002). However, ethylene appears to
be the primary hormonal component affecting sex expression with other hormones acting
via ethylene. Application of ethylene or ethylene-releasing compounds has been

demonstrated to increase pistillate flower production in cucumber, muskrnelon, and

31

squash while treatmcr
increase in maleness 1
stage does not confer t
critical stages in devel
et a1. 1970; Tolla and .
sulies to characterize .
releasing compounds or
proceeding or immedia
sex determination (Yam:

further SUppon :
OfEIh.l'lene evolution. w ‘
Elmeo'ous cucumber St" I
sawmill£15113y'ers et al., 1 r

molecular support for eth

cu: '
umber, increased mR \

frame. ACC synthase ( :1

fiT- . .

rioting in cucumber ( K
in. ' '

. tenors lines were fou

.lfl’

15) Co-segregatin g wi
1'1” It
( ~
studies in melt.
in» ’

«$1335

ed femaleness as m

C

librn'
‘rddopoulou er al 700 -
'9 ~ . 3 ll

 

squash while treatment with inhibitors of ethylene biosynthesis or action confers an
increase in maleness (Rudich, 1990; Perl-Treves, 1999). Since treatment at the 4-leaf
stage does not confer changes in sex until node 8-10, it has been proposed that there are
critical stages in development at which ethylene can modulate sex determination (Iwahori
et al., 1970; Tolla and Peterson, 1979; Den Nijs and Visser, 1980). Morphological
studies to characterize the stage of floral development affected by treatment with ethylene
releasing compounds or inhibitors of ethylene biosynthesis, indicated that the stage just
preceeding, or immediately following, differentiation of stamen primordia, is critical for
sex determination (Y amasaki et al., 2003).

Further support for the feminizing effect of ethylene is provided by measurements
of ethylene evolution, which showed 2-3 fold higher ethylene production fiom apices of
gynoecious cucumber seedlings compared to monoecious and andromonoecious
seedlings (Byers et al., 1972; Rudich et al., 1976). Additional studies have provided
molecular support for ethylene’s role in promoting pistillate flower development. In
cucumber, increased mRN A levels of a gene encoding a key ethylene biosynthetic
enzyme, ACC synthase (ACS), (CS-A C52) has been shown to coincide with female
flowering in cucumber (Kamachi et al., 1997, 2000; Yamasaki etal., 2001), and
gynoecious lines were found to carry an additional copy of the C5-A C51 gene (CS-

ACSI G) co-segregating with the F locus (Trebitsh etal., 1997; Mibus and Tatlioglu et al.,
2004). Our studies in melon demonstrated that overexpression of an ACS gene conferred
increased femaleness as measured by earliness and number of hermaphrodite buds

(Papadopoulou et al., 2005)

32

   
  
  
  
  

Yin ant

female organs U
Therefore. perm.
induce the other if.

to different ranges ‘
locus controls sensiti
Quinn 1995). The .4
seams in pistillate f
development while or
development (Roy ant
tothe ll locus in cucu
Genetic and mi
receptors in the transdr
Elmore and Ecker.
Potptors homologous t
Perception of ethylene.

ERSZ. and ElN-l) act to

violent and become in:

his ~
.rncreased expressit
:l‘riv, ‘
:‘ssron can lead to co;
Evaluation of the r

*‘Wrrberincluded the i

it
and CS-ERS) and an:

Yin and Quinn (1992, 1995) proposed that the developmental fate of male and
female organs in cucumber is affected by the level of, and perceptiveness to, ethylene.
Therefore, perception of ethylene by male and female primordia could inhibit one sex and

induce the other independently, and differing sensitivity or receptor levels could respond
to different ranges of hormone concentration. It has been proposed that the cucumber M
locus controls sensitivity to ethylene and the F locus controls ethylene levels (Yin and
Quinn, 1995). The A locus in melon is also proposed to influence the development of the
stamens in pistillate flowers such that an A- genotype confers the inhibition of stamen
development while an allows stamen development resulting in hermaphrodite bud
development (Roy and Saran, 1990). Therefore, the A locus in melon may be analogous
to the M locus in cucumber.

Genetic and molecular studies in Arabidopsis have elucidated the role of ethylene
receptors in the transduction of the ethylene signal (Bleecker and Kende, 2000;
Steptanova and Ecker, 2000; Wang et al., 2002). A family of membrane-localized
receptors homologous to bacterial two-component histidine kinases is responsible for
perception of ethylene. The five ethylene receptors in Arabidopsis (ETRl , ETR2, ERS 1,
ERSZ, and EIN4) act to repress downstream ethylene responses in the absence of
ethylene and become inactivated by binding ethylene to de-repress ethylene responses.
Thus, increased expression can lead to decreased sensitivity to ethylene and reduced
expression can lead to constitutive ethylene response (Hua and Meyerowitz, 1998).

Evaluation of the relationship between ethylene perception and sex determination
in cucumber included the isolation of three ethylene receptor homologs (CS-E TR 1 , CS-

E 7R2, and CS-ERS) and analysis of expression patterns in monoecious (fill/L),

33

gynoecious (F
3001). CS-Elr
gynoecious tha
gnoecious sht
regulation of C
releasing comp
mRNA. demon
receptor expres
Ethrel treatmen
todromonoecio
hemhen the .1! l
”993. ll was l
may be ll'lSlJlllCl
de‘ilopment of
111 sltu h:
W E 7 R3 l or
he lime Critical

mallOnghip bets

gynoecious (F-M-), and andromonoecious (fiinm) genotypes (Y amasaki et al., 2000;
2001). CS-ETRZ and CS-ERS were expressed at higher levels in shoot apices of
gynoecious than monoecious seedling (Y amasaki et al., 2000). In addition, treatment of

gynoecious shoot apices with the ethylene biosynthesis inhibitor AVG, resulted in down-
regulation of CS-E T R2 and CS-ERS while treatment of monoecious apices with ethylene-
releasing compound, Ethrel, led to increased accumulation of CS—E T R2 and CS—ERS
mRNA, demonstrating their regulation by ethylene. In contrast, induction of ethylene
receptor expression was not observed in an andromonoecious genotype in response to
Ethrel treatment (Y amasaki et al., 2001). The absence of receptor induction in the
andromonoecious genotype in response to ethylene provided support for the relationship
between the M locus and ethylene response as proposed by the Yin and Quinn model
(1995). It was proposed that reduced ethylene perception by the stamen of mm genotypes
may be insufficient for inhibition of stamen development in pistillate buds, leading to
development of hermaphrodite buds (Y amasaki et al., 2001).

In situ hybridization studies evaluating the expression of ethylene receptors (ERS,
E T R1 , ETRZ) and biosynthetic gene (A CS2) following initiation of all four floral whorls,
the time critical for divergence to a unisexual flower, provided further insight into the
relationship between ethylene-producing and ethylene-perceiving tissues (Yamasaki et
al., 2003). Buds fiom monecious plants at the stage following initiation of primordia for
all four floral whorls showed expression of E T R] in both stamen and pistil primordia and
ERS in stamen primordia; E T R2 was expressed just below the pistil primordia and in the
adaxial side of petals. In contrast, at the same developmental stage, buds from

andromonoecious plants showed expression of E TRI only in the region just below pistil

34

prime
expre

respo:

REY-gar
ethyle
and th
etltylet

melon,

primordia and no detectable expression of ERS and E T R2. The lack of receptor
expression detected in the stamens of andromonoecious buds was suggested to prevent
response to the ethylene signal which would typically inhibit stamens in pistillate buds.

Further evidence for the importance of ethylene perception for pistillate flower
formation was obtained from recent studies of andromonoecious and gynoecious melon
transformed with the mutant Arabidopsis etrI -1 gene (Papadopoulou et al., 2002). The
receptor mutant etrI-I, which is unable to bind ethylene, confers a dominant ethylene-
insensitive phenotype as a result of its inability to turn off the repression of ethylene
responses. Constitutive expression of etrI -1 caused phenotypes associated with reduced
ethylene sensitivity, including reduced adventitious rooting, delayed flower senescence,
and the inability to respond to exogenous ethylene, indicating that etrI -1 is able to confer
ethylene insensitivity in the heterologous melon system. In addition, the 35S::etrI-1
melons showed an altered sex expression pattern such that the formation of carpel-
bearing buds was largely abolished. These studies demonstrated that ethylene perception
is necessary to promote bisexual flower development in melon.

In this study, I sought to investigate the effect of targeted inhibition of ethylene
perception. Andromonoecious melon was transformed with etrI -1 under the direction of
Arabidopsis floral-organ specific Crab ’3 Claw (CRC) and Apetela3 (AP3) promoters. In
Arabidopsis, Crab ’s Claw (CRC) encodes a regulator of carpel and nectary development
which is expressed specifically in primordia of these organs from their inception
(Bowman and Smyth, 1999; Lee et al., 2005). The class B floral identity homeotic gene
Apetala3 (AP3) is expressed in cells destined to become petals and stamens (Irish and

Yamamoto, 1995).

35

in.

plants t
and me
stamina
that car
anal 0 iii
produc

dex'elo;

321ml!

The predicted sex expression patterns of AP3 ::etr1 -1 and CRC::etr1-1 melon
plants were based on the observed effects of exogenous ethylene treatment in cucumber
and melon. In cucumber, ethylene treatment promotes the transition from a phase of
staminate to pistllate flower production (Rudich, 1990; Perl-Treves, 1999), suggesting
that carpel development is induced while stamen development is suppressed. In contrast,
analogous treatment of melon results in a shift from staminate to bisexual flower
production (Rudich, 1990), suggesting that carpel development is induced while stamen
development is unaffected. The production of virtually all staminate flowers by
35S::etr1 -1 melon plants (Papadopoulou et al., 2002) was consistent with this proposed
effect of ethylene, such that the inability to perceive ethylene in carpels prevented their
development while blocked perception in the stamens had no effect. Therefore, as shown
in Figure 2.1, ethylene insensitivity in the stamens and petals conferred by AP3::etrI-1
was expected to have no effect on sex expression in melon while the carpel insensitivity
conferred by CRC::etr1—I was expected to prevent carpel development, resulting in
staminate flower production. Phenotypic characterization of sex expression of the
transgenic CRC::etr1 -1 and AP3::etrI-1 provided the unexpected, opposite observation
of female rather than bisexual flowers on CRC::etr1-1 plants, and almost exclusive
staminate flower production on AP3::etrI-I plants. These results and a revised model for

the role of ethylene perception in sex determination of melon will be discussed.

36

Wilt

355

AP

Cl

 

Genotype

Wild-type

3SS-etr1

AP3-etr1

CRC—etr]

Expected Result

Staminate and bisexual
1. Carpel development
2. Stamens unaffected

 

Staminate only
1. No carpel development
2. Stamens unaffected

 

Staminate and bisexual
1. Carpel development
2. Stamens unaffected

 

Staminate only
1. No carpel development
2. Stamens unaffected

Figure 2.1. Expected sexual phenotype of transgenic melon lines based on effects of

ethylene treatment of wild type andromonoecious melon.

37

Materials
Preparati

Cr
Research
form pClI
John Boo
ligated int
et al. 198
(RC. C o
to the errl
Etolil to 1
Vector pC

Ar
Primer Ct
site at the
(R6177 l
PGEM w
PTOmoter
PUB? to
here Scre
Selected p

into pcm

Materials and Methods
Preparation of Plasmid Constructs

CRC: :etrI -1 : The 355 promoter was removed from pCIB7lO (CIBA-GEIGY,
Research Triangle Park, NC) (Rothstein etal., 1987) as an XbaI-BamHI fiagment to
form pCIB710A35S. The CRC promoter was cut from 8-12pbj-6O (kindly provided by
John Bowman, University of California, Riverside) as an XbaI-BamHI fragment and
ligated into pCIB710A35$ to form pCIB7lO-CRC. Arabidopsis etrI -1 cDNA (Bleecker
et al., 1988) was cut from pBS-etrI-I as a BamHI fragment and cloned into pCIB7lO-
CRC. Colonies were screened for proper etrI -1 orientation based on an XbaI site internal
to the etrI -1 gene. The selected pCIB7lO-CRC-etr1-I clone was digested with PstI-
EcoRI to move the CRC-etrI -1 fragment into the binary Agrobacterium transformation
vector pCambia2300 (CAMBIA, Australia.) (Figure 2.2).

AP3: :etrI-I : The AP3 promoter (Jack et al., 1994) was PCR amplified with 5’
primer CCTCTAGAAAGC'I'I‘MGAA'ITATAGTAGC (R6176) to introduce a XbaI
site at the S’end and 3’ primer CCCGGATCCATTCTTCTCTCTI‘TGTTI‘AATC
(RG177) to introduce a BamHI site at the 3’end. The PCR fragment was cloned into
pGEM With pGEM®-T-Easy Vector System (Promega Corporation, Madison, WI). The
promoter was then cut fiom pGEM as an XbaI-BamHI fragment and ligated into
pCIB710A35S. etrI -1 was cloned into pCIB710—AP3 as a BamHI fragment. Colonies
were screened for proper etrI -1 orientation based on the internal etrI -1 XbaI site. The
seleCted PCIB71 O-AP3-etr1-I clone was digested with PstI-EcoRI to move AP3-etr1-1

““0 PCambia23oo (Figure 2.2).

38

 

Xbal

Flgllre 2.3

llemid ire,

Xbal BamHl BamHl EcoRl

CRC (3.6 kb)
or AP3 (1.8 kb) promoter etr1-1 cDNA (2.2 kb) 358 t

‘3‘}

 

 

 

 

 

 

 

 

 

 

 

{abs -' ass p Nerutxam’r .53 . eh—
pCambiaZ300
KanR

 

 

 

 

 

 

 

Figure 2.2. Schematic of binary vector construct CRC: :etrI -1/AP3::etrI -I in binary

plasmid vector pCambia23OO as an XbaI-EcoRI fragment.

39

Plant Transformation

The pCambia-CRC-etrI -1 and —AP3-etr1-1 constructs were introduced into
Agrobacterium tumefaciens strain C58 (Deblaere et al.,l985) for subsequent
transformation into melon. Andromonoecious muskrnelon, Hale’s Best Jumbo (I-lollar
Seed, Rocky Ford, Colorado), was transformed via Agrobacterium-mediated
transformation of cotyledon tissue via the method of Fang and Grurnet (1990) with
revisions based on the cucumber transformation method of Tabei et a1. (1998), as follows.
Seeds were decoated, sterilized in a 15% (v/v) bleach solution (5.25% sodium
hypochlorite) with a drop of Tween-20 for 10 minutes, rinsed three times with sterile
distilled H20, and then germinated in darkness at room temperature in a Petri plate with 5
mL sterile dH20 for 17-19 hours.

Cotyl edons from the partially germinated seeds were separated and cut into four
Pieces. The explants were soaked for 10 minutes in a diluted suspension (1:10 in LB) of
an Agrobacterz’um culture grown overnight, blotted dry on sterile filter paper, and
transferred to M1 media (MS media (B10101) with 5 uM BAP, 1 uM ABA, and 2.5 g /L
P11340861 (Si gma)). Plates were cultured for 3 days at 26°C in darkness. Explants were
rinsed three times with deO, blotted dry on sterile filter paper, transferred to shoot
induction media, M2 (M1 with 400 mg/L timentin). M2 plates were cultured under
growth I‘OOIn conditions (26°C) for 7 days before transferring to M3 media (M1 with 400
mg/L timentin and 200 mg/L kanarnycin) then returned to the growth room. After 4-5
Weeks, regenerated shoots were separated fiom explants and cultured on MS media with
ZHM BAP, 400 mg/L timentin, 200 mg/L kanarnycin, and 8 g/L agar in Magenta boxes

(Magenta Corp, Chicago, IL) for selection and shoot elongation. Unbleached shoots

40

 

lit?

for

go.

Wt

51'

ll.)

were transferred to MS media with 400 mg/L timentin and 8 g/L agar in Magenta boxes
for rooting under grth room conditions. Plantlets were transplanted to Baccto soil as
soon as roots appeared. Potted plantlets were sealed in bags for one week before slowly
acclirnating to lower humidity conditions and transferring to the greenhouse.

Regenerated plants (T0) were evaluated for the presence of the neomycin
phosphotransferase protein by NPTII ELISA (Agia®, Elkhart, Indiana) and the etr1-1
gene by polymerase chain reaction (PCR) (described below). PCR-positive To plants
were transferred to the greenhouse and self-pollinated to produce T1 progeny for further
evaluation.
PCR analysis

The Wizard Genomic Plant DNA Purification Kit (Promega Corporation,
Madison, WI) was used to extract genomic DNA from leaf tissue of melon. PCR was
used to amplify the inserted etrI-I transgene using etrI -1 specific 5’ and 3’ primers (5’-
GGGGAGGTGGTCGCTGTGA (RG305) and 5’-GCTCATGGGACACAACTCGG
(R6306), respectively). CRC::etr1-l was amplified with CRC specific 5’ primer 5’-
CTTGCAATCCCTAGCCAG (RG 299) and 3’ etrI-I primer RG306. AP3:.°etr1-1 was
amplified with AP3 specific 5’ primer 5’-GCTTTGGTCCCCCTCT'I"I'TACC (R6300)
and 3’ etrI-I primer RG306. The amplification reactions of etrl -1 were carried out with
Taq DNA Polymerase (Invitrogen, Carlsbad, California) in 3.0 mM MgClz for 1 cycle of
2 minutes at 95°C followed by 40 cycles of 95°C for 1 minute, 62°C for 1 minute, and
72°C for 1 minute, then 1 cycle of 5 minutes at 72°C (Sambrook and Russell, 2001).
Amplification reactions of CRC::etr1-I and AP3 ::etr1 -1 were carried out as with etrI -1

but with the following program: 1 cycle of 2 minutes at 95°C followed by 40 cycles of

41

Cl

in

hi

to

Oh

95°C for 50 seconds, 58°C for 50 seconds, and 72°C for 2:20 minutes, then 1 cycle of 5
minutes at 72°C.
Northern Hybridization Analysis

Fresh young leaf tissue was collected from T1 greenhouse grown melon plants to
evaluate etrl -I transgene expression. RNA was isolated from 100 mg fresh leaf tissue
frozen in liquid nitrogen using ConcertTM Plant RNA Reagent (Invitrogen, Carlsbad,
California). RNA quantification and purity was determined by absorbance at 260 nm and
by gel electrophoresis. Total RNA (15 ug) was electrophoresed on a 1.5% agarose
formaldehyde gel and blotted to nylon membrane (Hybond-N+; Amersham-Pharmacia
Biotech Ltd, England) (Sambrook and Russell, 2001). The membrane was hybridized
overnight at 61°C with a digoxigenin-UTP labeled RNA probe of etrI -1 (Roche
Diagnositics, Germany). Stringency washes and detection via chemiluminescent
visualization with CDP-Star were carried out following protocols developed for the DIG
System (Roche Diagnostics, Germany). RNA also was isolated fi'om young male and
female buds (3-5 mm) and older male buds (10-12 mm) from greenhouse grown plants.
Petals were excised from the older male buds with a razor blade and designated as
“petals” While the remaining portion of the male bud containing stamens, nectaries, and
floral cup Was designated “lower-bud.”
Ethylene Measurements

Young leaves (approximately 4-7 cm wide) were removed from lateral branches
0f greenhouse grown plants with a razor blade, weighed, and placed in a 50 ml syringe
Containing ‘A piece of Whatman #2 filter paper (90 mm) moistened with 200 1.1L distilled

Water (dHZQ). The syringe plunger was adjusted to 30 mL, and the end of the syringe

42

 

021C:
beer

can

was fitted with a rubber septum and sealed with parafilm. Gas samples were removed
using an airtight syringe (lmL) inserted through the rubber septa into the airspace of the
50mL syringe. Samples were analyzed by gas chromatography (Hach Carle series
IOOGC, Linear 1200 recorder) with an activated alumina column and flame ionization
detector. Likewise, male buds (1-2 days pre-anthesis) were removed from lateral
branches with a razor blade, weighed, placed in 10 mL syringes containing a 1 cm square
of Whatman #2 filter paper moistened with 40 uL H20, plunger adjusted to 5mL, and the
opposite end fitted with a rubber septa and parafilm. Leaf and male bud samples were
collected three times from four randomly sampled plants per genotype.

Sex Expression

Plants for all experiments were grown in 35 cm diameter plaster pots containing
commercial Baccto soil mix (Michigan Peat C., Houston, TX). Plants were fertilized
once per week with 300 ppm of 20-20-20 fertilizer mix. Experiments were conducted
between March and October, during seasons of adequate light to observe flowering
patterns.

Each node of the mainstem of greenhouse grown melon plants was evaluated for
the presence of staminate and/or carpel-bearing buds. The development of each bud was
followed to note if anthesis was reached, and, for aborted carpel-bearing buds, the size
reached Prior to senescing. Plants were observed for 30 nodes.

A11 exPeriments were conducted in a randomized complete block design with 5-8
replicates Per gentoype. Each genotype was tested in at least 4 experiment with the
exception 0f 35S::etr1 -I , AP3::etr1-1 line 3, and non-transgenic tetraploid controls. The

SpeCifiC genotypes and number of replicates are indicated in the figure legends.

43

 

{'71

(LC?

E“:

Pit

Wt:

Floral Morphology Evaluation
Flower buds were removed from greenhouse grown plants with a razor blade.
Buds were cut longitudinally and viewed under a dissecting microscope fitted with a

digital camera (Olympus Optical Co., Melville, NY).

Results
Gene integration and expression analysis

Transgenic To melon plants haboring the AP3::etr1-I or CRC::etr1-1 constructs
were produced by Agrobacterium-mediated transformation and verified by PCR analysis.
Examination of pollen morphology indicated that AP3::etrI-1 lines had four-sided pollen
grains indicative of tetraploidy (F assuliotus and Nelson, 1992), while CRC::etr1-1 were
diploid. Tetraploidy is a common phenomenon in melons regenerated from tissue culture
(Ezura et al., 1992; Yadav et al., 1996) and does not interfere with sex phenotypes as
non-transgenic tetraploid segregants show normal sex phenotypes (Papadopoulou et al.,
2005). Transgenic To plants were self-pollinated in the greenhouse to produce seed.
Presence of the etrI-I gene and appropriate AP3 or CRC promoter was verified in T1
progeny by PCR analysis (Figure 2.3). Segregation ratios of T. progeny are consistent
with gene insertion at one site, with the exception of AP3-3, which could have one or two
transgene insertions depending on when the chromosome doubling event occurred (Table
2.1).

Northern hybridization analysis was conducted to evaluate etrl -1 transgene

expression in leaves and floral tissue (Figure 2.4). In AP3 :.-etr1 -1 lines, etr-I -1

K \‘0 $5} N43 55.73?)
'0’ 'o’

rifle QsiieLOX 3” V26 flash its”

see 000 q' eve ooosq’q' s

    

l—APS-etrl —l l—CRC-etr1 —l

Figure 2.3. PCR amplification of etrI -1 gene construct from genomic DNA of T1
transgenic lines with CRC::etr1-1 and AP3::etr1-1 specific primers. The picture shows
amplification of the same set of plant DNA samples with both CRC::etr1-I and
AP3::etr1-1 specific primers. WT=Hale’s Best Jumbo wild type plant control; p-CRC—

etrI-I and p-AP3-etr1-I = plasmid controls.

45

Table 2.1. Segregation analysis of AP3 ::etr1 -1 and CRC:.'etr1-1 self-pollinated

transgenic melon progeny.

 

 

 

 

 

 

Line Segregation Ploidy Expected x7
Observed Ratio

CRCS 25:14 2n 3:1 1.92ns

CRC11 28:16 2n 3:1 1.98ns

CRC15 37:11 2n 3:1 0.03ns

AP3-3 12:0 4n 15:1 0.61ns

AP3-8 T1 6:6 4n 3:1 2.77ns
T2 19:6 3:1 0.097ns

 

All progeny were T1 generation except AP3-8, which included T2. Data are based on the

presence of the AP3::etr1-1 or CRC::etr1-1 by PCR amplification.

ns- )8 value not significantly different at p=0.05

46

Figure 2.4. Northern hybridization of heterologous etrl-I gene expression from
greenhouse grown AP3::etr1-1 and CRCssetrI-I melon with a DIG-labeled etrI-I
probe. (A) Hybridization of total RNA isolated from leaf tissue of non-transgenic
(WT), 35S::etr1-I (35S), CRC::etr1-I (CRC11 and CRC15), and AP3::etr1-I
(AP3-3 and AP3-8) plants. (B) Hybridization of total RNA isolated from young
male buds of non-transgenic (WT), AP3::etr1-1 (AP3-3 and AP3-8), and
CRC::etr1-1 (CRC5 and CRC15) plants and young female buds of non-transgenic
(WT) and CRC::etr1-1 (CRC5 and CRC15) plants. (C) Hybridization of total
RNA isolated from mature male buds of non-transgenic (WT), CRC::etrI-I
(CRC5 and CRC11), and AP3::etrI -1 (AP3-8) plants. Mature male buds were
separated into petal tissue and lower bud, which included sepals, stamens,

nectary, and floral cup. The bottom panel in each pair shows rRNA stained with

ethidium bromide.

47

 

A Lear

 

B. You

In-

A. Leaves
WT 35$ AP3-3 AP3-8 CRC5 CRC11 CRC15

 

B. Young male buds Young female buds

WT AP3-3 AP3-8 5 WT CRC5 CRC
. ~ H Attemga‘hfl’ 3W _ . W E. '3“ 1 .

 

C. Mature male buds

Petals: Lower-bud:

 

48

expression on
and stamen spt
tested for lerei
till he discus
lefty express
We detected

Were OllSt’rl e

EXp'
inability to
Mutton
l997; Papa
ethylene pr
Eltl‘ated e
CRC11, c
fail eupr.
prOduction
Ethylene Pr

bun.

expression was absent in leaves and present in male buds as expected based on the petal
and stamen specificity of AP3 in Arabidopsis (Jack etal., 1994). AP3 plants could not be
tested for levels of transcript in female buds due to an absence of pistillate buds, which
will be discussed later. CRC::etr1-1 lines showed hybridization in leaves, indicating
leaky expression from the promoter in the heterologous melon system. etrI-I mRN A
was detected in both young male and female buds of CRC::etr1-1 plants; higher levels
were observed in the females buds, consistent with expected carpel expression. Detection
of etrI -1 message in male CRC buds could be the result of expression in the nectaries
(Bowman and Symth, 1999; Lee et al., 2005) or leakiness, as is suggested by expression
in the petals.
Ethylene Measurement

Expression of etrI -1 is expected to confer increased ethylene production since the
inability to perceive ethylene interferes with the negative feedback inhibition of ethylene
production, as has been observed in 35S::etr1-1 petunia and melon (Wilkinson et al.,
1997; Papadopoulou 2002). As expected, 35S::etr1-I plants showed elevated levels of
ethylene production from leaves and male buds (1-2 days pro-anthesis) (Table 2.2).
Elevated ethylene evolution from leaves and male buds of CRC::etr1-1 lines (CRC5,
CRC11, CRC15), relative to non-transgenic controls (WT, Azygous) was consistent with
leaky expression observed by northern hybridization (Table 2.2). Similarly, ethylene
production in AP3::etr1-1 plants was generally consistent with northern analysis.
Ethylene production was not elevated in leaves and was slightly elevated in mature male

buds.

49

lrblell. Cor

of.lP3.'.'etrl-l

 

fl

\EG

in

E

,ous

Hm!

\t \n WKL

R
C

D313 Shown

Table 2.2. Combined data for all leaf and male bud ethylene measurement experiments

of AP3::etr1 -1 and CRC::etr1-1 transgenic melon lines.

 

 

 

 

 

 

 

 

 

 

 

 

Experiments Leaf Male Blld
(Total number
of plants) Samples C2H4 Samples C2114
evaluated (pmol/g) :1: SE evaluated (pmol/g) :t SE
WT 4 (23) 51 53.7 i: 3.1 47 39.9 i 2.8
4nNEG 2 (11) 24 58.1 i 5.0 24 63.0 :I: 6.2
Azygous 2 (10) 26 66.6 i 24.9 22 60.0 :t 9.0
358 1(5) 12 511.1 :1: 109.1 12 188.3 i 50.5
AP3-3 l (7) 12 50.1 :1: 11.7 12 65.6 :1: 9.2
AP3-8 3 (17) 42 82.2 i 7.6 36 108.0 :1: 31.0
CRC5 2 (11) 24 197.8 1 28.5 23 117.7 :1: 16.8
CRC11 2 (11) 24 130.6 i 15.5 20 169.4 i 29.6
CRC15 4 (23) 52 190.8 a: 20.7 48 99.6 :b 6.6

 

 

 

 

 

 

Data shown are i standard error.

50

 

So expression a
(RC‘rtr
The ttpi
ndromonoecio
henamirture
l melon plants
ethylene (Pap:
tourist to the
llll‘tlllll Carpel
buds. Initial
burl bearing t
Wired to
ctufted out 2
“emotions
lllese exalt
on [ht mair

"tOIllIOlg (\l

Sex expression and floral morphology

CRC:.'etrI-I

The typical progression of sex expression along the main stem of
andromonoecious melon is 3-4 vegetative nodes, followed by phase of staminate buds,
then a mixture of staminate and carpel-bearing buds. As previously observed, 35S::etr1-
I melon plants, which display several phenotypes consistent with the inability to perceive
ethylene (Papadopoulou, 2002), failed to produce carpel-bearing buds (Figure 2.5A). In
contrast to the prediction that inhibition of ethylene perception in the carpel would also
inhibit carpel-bearing buds, CRC.°:etr1-1 lines did not fail to produce carpel-bearing
buds. Initial evaluation showed no significant difference in the appearance of the first
bud bearing a carpel on plants of CRC::etr1-1 lines CRC5, CRC11, and CRC15
compared to wild type (WT) plants (Figure 2.5A). While the initial experiment was
carried out at the end of the growing season, late August to early October, subsequent
evaluations were carried out during the spring and summer months, May through August.
These evaluations showed significantly earlier appearance of the first bud with a carpel
on the main stem of CRC plants (CRC5, CRC11, CRC15) than on non-transgenic
controls (WT and AZY) by 6-10 nodes (Figure 2.5B).

Once the main stem begins producing pistillate buds, both male and female buds
are generally produced at the same node for the remainder of the plant’s life. Evaluation
of the total number of carpel-bearing buds produced in the first 30 nodes on the main
stem of CRC plants demonstrated this typical flowering pattern. The increase in number

of total buds bearing carpels on CRC plants (CRC5, CRC] 1, CRC15) compared to

51

Figure 2.5. Main stem sex expression patterns of melon plants transformed with
CRC::etr1-1. (A) Node position of first carpel-bearing bud on the main stem of
Hale’s Best Jumbo 2n and 4n non-transgenic plants (WT and 4n-neg), 35S::etrI-I
plants (358), and CRC::etr1-1 plants (lines CRC5, CRC11, and CRC15). The
experiment was conducted in a randomized complete block design. Data are the
mean i: standard error of seven replicate plants per genotype (except 35S, n=5).
(B) Node of first carpel-bearing bud on main stem and (C) Total number of node
with carpel-bearing buds on the main stem for Hale’s Best Jumbo and azygous
non-transgenic plants (WT and AZY), and CRC::etr1-1 lines CRC5, CRC11, and
CRC15. The experiment was repeated three times in a replicated complete block
design, with each giving equivalent results. The data were pooled from the three
experiments. Data are means i standard error of 16 plants of each genotype
(except n=15 AZY). Bars marked with different letters are significantly different,

ANOVA, LSD, P=0.05.

52

‘l-J— n__n-__

Numbar no '1- "a-

Node Position

Node Posltlon

Number of buds

 

First Carpel-bearing Bud on Main Stem

 

 

 

WT 4-neg 35$ CRC5 CRC11CRC15

 

 

 

 

Total Main Stem Car el-bearin Buds

 

18-

 

 

WT AZY CRC5 CRC11 CRC15

53

 

mew.
pregen.
senesc:
the m
panem

ajom
M or
control
Prom-n

control:

beefing
i5 IEpic
the CR.
17). T
Since 1}
CRC Ca

$0qu 2

 

 

controls (WT, AZY) reflects the earliness in production of the first pistillate bud (Figure
2.5C). These results indicate that CRC::etr1-I plants showed enhanced rather than
inhibited femaleness.

Typically, the majority of the buds bearing carpels initiated on the main stem of
melon plants abort prior to reaching anthesis (Papadopoulou et al., 2005). Despite the
presence of carpel-bearing buds, preliminary observation suggested that premature
senescence may be even more pronounced in CRC plants. Measurement of the size of
the immature carpel-bearing bud at the time of abortion showed a distinctly different
pattern for CRC buds than non-transgenic controls (WT and AZY) (Figure 2.6). The
majority (80%) of CRC buds with carpels aborted early in development. Carpel-bearing
buds on CRC plants were four times more likely than buds on wild type and azygous
control plants to senescence at 4 mm or less, and a similar decrease was observed in
proportion of buds on CRC plants that mature beyond 4 mm and reach anthesis relative to
controls (Figure 2.6).

An even more surprising finding came in the examination of the few carpel-
bearing buds that did reach anthesis. Instead of developing into a bisexual flower, which
is typical of the carpel-bearing bud formed on andromonoecious melon, examination of
the CRC flowers containing a carpel frequently showed an absence of stamens (Figure
2.7). Thus, the CRC plants exhibited a sex conversion from hermaphrodite to female.
Since the pistil of bisexual buds is largely hidden by stamens, the absence of stamens in
CRC carpel-bearing flowers was quite obvious (Figure 2.7A). Subsequent evaluation of
the rare flowers bearing a carpel produced on lateral branches of 35S::etr1-1 plants also

showed an absence of stamens indicating that in the cases where flowers containing a

54

Carpel-bearing Bud Maturity

 

I <4mm a
80 ' a I >4mm a’
70 l B open

501

40‘

% Carpel-beefing Buds

20‘

10‘

 

 

     

CRC5 CRC] 1

Figure 2.6. Size reached by carpel-bearing buds before senescing for Hale’s Best Jumbo
and azygous non-transgenic plants (WT and AZY), and CRCssetrI-I plants (lines CRC5,
CRC11, and CRC15). The experiment was repeated three times in a replicated complete
block design with equivalent results. The data were pooled from the three experiments.
Data are means i standard error of 16 plants of each genotype (except n=15 AZY). Bars
marked with different letters are significantly different, ANOVA, LSD, P=0.05; with a

and b corresponding to <4mm floral buds, a’ and b’ corresponding to >4mm floral buds,

and a” and b” corresponding to open flowers.

55

 

Figure 2.7. Carpel-bearing flower morphology of non-transgenic (WT and
4nNEG) and transgenic (35S::etr1-1 and CRC::etrI-1) melon plants. (A) Flowers
of WT, 4n—NEG, CRC::etr1-1, and 35S::etr1-I at anthesis and at 6, 10 and 16
mm. (B) Developing buds bearing carpels of WT plants. (C) Developing carpel-

bearing buds of CRC::etr1-1 plants at 6, 10, and 19mm. S=stamen; P=pistil.

56

 

 

57

carpel were formed, inhibition of ethylene perception prevented stamen development
(Figure 2.7A). Microscopic evaluation of carpel-bearing buds at various stages of
development was consistent with the observed inhibition of stamen development in CRC
plants (figure 2.7BC); 55% of the CRC flowers/buds evaluated lackied stamens versus
0% of the non-transgenic buds. Stamens can be observed in the region between the
enlarging pistil and petals throughout the development of the WT hermaphrodite buds,
while in pistillate CRC buds this entire region is occupied by the pistil (Figure 2.7B and
2.7C).

AP3::etr1 -1

Evaluation of the sex expression pattern on the main stem ofAP3::etr1-I plants
also provided surprising results. While the inhibition of ethylene perception in the
stamens conferred by AP3::etr1-1 was not predicted to affect sex determination in melon,
all AP3-3 plants and the majority of AP3-8 plants (61%) failed to produce any carpel-
bearing buds within the first 30 nodes (Figure 2.8A). Those AP3-8 plants, which did
produce buds bearing carpels on the main stem, reverted back to staminate buds, after
producing only 1-2 buds containing a carpel. The first node of lateral branches from the
main stem also was examined for flower sex. Hermaphrodite buds typically form at these
positions in wild type andromonoecious melon; however, AP3::etr1-I plants also showed
staminate buds in these positions (Figure 2.9A). Overall, AP3 plants showed
hermaphrodite bud formation on less than 5% of the nodes evaluated while more than
50% of nodes evaluated in non-transgenic plants (WT and AZY) produced buds

containing a carpel (Figure 2.8B). As was observed in previous experiments, CRC::etr1-

58

Figure 2.8. Carpel-bearing bud formation on AP3::etr1-1 trangenic melon plants.
(A) Node position of first carpel-bearing bud on main stem and (B) Percentage of
nodes bearing buds containing a carpel for Hale’s Best Jumbo (WT), azygous
(AZY), AP3:.-etr1-1 (AP3-8) and CRC::etr1 -1 (CRC15) plants. The experiment
was repeated three times (with the exception of AP3-3 which was only only
included in one of the experiments), in a randomized complete block design, with
a total of n=21 AP3-8, n=18 WT and CRC15 and n=14 AZY plants, and n=7
AP3-3.

All data are combined means :t standard error, and bars marked with different

letters are significantly different, ANOVA, LSD, P=0.05.

59

Carpel-bearing Bud Formation

 

C...
%%%

0....
0..

C.......
IFPQQQQQVQDEVoIe
coco

0..
§¢

 

(«XV

.\ V 3v»?!(«vvvvvvx Vvvxvvvbvvvvxqv

0.... 00...... O

C
O OOOOOOOOOOHO“OO

fififivoe

O.
§%%%%§€€€.
00.......
90.0.0.0?
0.0.00.0...
.0.......
‘0’...
90.90.?
000.0900
Q.......OO.
.90......
0.0.90?
0 COQQOOO......QOOQ
o»? 555%...
t.‘¢

, H . H N x . .
y»»»»»«

.5; 5.«.3,(«weeieie.«.«.«.«.«.3¢.««.eeiwie.
MvvxJ“one“«we“«“uvvvvvvvvvv“«v3 VVVVV“ VVVVVVVR

«5. . .5. .5

eve???“ « a \
.«883338?
weusuwxukwwwww

I

Lia...
pl

25.25 .15? .«eee? .«.«.«.5.«.«.«.
(«x .«“«M«M«M«M«~«VVM«“¢“ vvvvvvVVVVVV“6.“
«Kyyyyyppbvyfi%¥MiVVnggkyv

V» by
> 2 .

«eggwage»?we?«wwwuwkxw

 

 

_ - 4
AU :0 nu
’—

1 1

:oEuom ouoz

 

 

 

 

3 AP3-8 CRC15

WT AZY AP3

60

Figure 2.9. Hermaphrodite flower morphology of non-transgenic (WT) and
transgenic (AP3::etr1-1) melon plants. (A) First node on lateral of WT and
AP3::etr1-I at anthesis. (B) Developing hermaphrodite buds of WT plants. (C)
Developing hermaphrodite buds of AP3::etr1-1 plants. S=stamen; P=pistil,

=ovary, Ov=ovule.

61

AP3::etr1-1

 

62

1 plants that were included for comparison had earlier and increased formation of buds
containing carpels (Figure 2.8).

Unlike the CRC::etr1-1 and 35S::etr1-1 carpel-bearing buds, the occasional buds
containing a carpel of AP3 : :etrl -1 plants showed well-developed stamens; however,
development of carpels was altered relative to wild type (Figure 2.9BC). In the upper
region of the wild type hermaphrodite bud, the stigma occupies a large portion of this
space while the stamens are tightly nested between the stigma and petals (Figure 2.9B).
In hermaphrodite buds of AP3 plants, the stamens occupied most of the upper region of
the bud and the stigrnas appeared smaller and less developed than those in WT plants at
similar stages of development (Figure 2.9C). Evaluation of longitudinal sections of
developing wild type hermaphrodite buds also showed distinctive structural development
of the ovary and progressive maturation of the ovules. In contrast, the AP3 bisexual buds
had a shorter, poorly differentiated ovary and a lack of ovule development, indicating a
difference in ethylene perception requirements for the development of ovaries versus
stamens. It should also be noted, that consistent with the rare occurrence of
hermaphrodite buds and the poorly developed carpels, it was difficult to produce progeny

on AP3::etr1 -1 plants.

Discussion

The results presented demonstrate that expression of the dominant mutant
ethylene receptor gene, At—etrI -1 , under the direction of Arabidopsis floral-targeted
promoters of APE TALA3 and CRAB ’S CLA W in andromonoecious melon conferred

altered flowering phenotypes strikingly different from each other, fiom wild type, and

63

from 35S::etr1-1 melon, with respect to sex expression patterns, and flower maturation
and morphology.

As described in the Introduction, the predicted sex expression patterns of
CRC::etr1 -1 and AP3::etr1-1 transgenic melon were based on the changes in sex
subsequent to treatment with ethylene. The transition from staminate to female flower
production upon ethylene treatment in cucumber (Perl-Treves, 1999) suggested that
ethylene promotes carpel development while suppressing stamen development. In melon,
an observed shifi from staminate to hermaphrodite flower production (Rudich, 1990)
suggested that ethylene facilitates carpel development but has no effect on stamens.
Therefore, blocking ethylene perception in the stamens of melon plants was not expected
to result in a modified flowering phenotype since ethylene was not predicted to play a
role in stamen development. In contrast, blocking ethylene perception in the carpels was
expected to result in the inhibition of carpel-bearing flower production since ethylene was
predicted to be critical for carpel development. Consistent with the suggested
requirement of ethylene perception by the carpel, hermaphrodite bud formation in
35S::etr1 -1 melon plants was largely abolished (Papadopoulou et al., 2002).

Despite leaky expression, CRC::etrI -1 plants were phenotypically different fi'om
35S::etr1-1 plants and did not show the expected inhibition of carpel-bearing flowers.
Instead, CRC::etr1-1 plants showed enhanced femaleness as indicated by earlier and
increased formation of buds containing carpels and a conversion from hermaphrodite to
female bud production. The difference in the sex expression of 35S::etr1-I and
CRC::etrI -1 plants may result fiom differences in timing and/or location of etrI-I

expression at key stages in sex determination

64

 

 

pa

214
(P

Conversely, and in contradiction to the predicted result that blocked ethylene
perception in the stamen would not affect the production of hermaphrodite buds,
AP3::etr1-1 plants almost exclusively produced staminate flowers. In addition,
longitudinal sections of limited pistillate buds produced by AP3::etr1-1 plants, exhibited
inhibited carpel development compared to wild type buds of similar size, while stamen
development within the pistillate buds appeared normal.

Collectively, the findings of this study were not consistent with expected sexual
phenotypes and suggest a revised model for the role of ethylene perception in sex
determination. The lack of pistillate buds on AP3.°:etrI-I plants suggests that ethylene
perception by stamen primordia is important for development of carpels. On the other
hand, the absence of stamens in carpel-bearing buds, as observed on CRC::etr1-1 plants,
may suggest that perception of ethylene by carpel primordia is important for sustained
stamen development. Therefore, a revised model would suggest that a reciprocal
relationship exists between the stamen and carpel in the development of hermaphrodite
buds, such that ethylene perception by the stamen is required for subsequent development
of the carpel and vice versa (Figure 2.10). According to this model, andromoneocious
melon plants produce a phase of staminate flowers until sufficient levels of ethylene are
present to initiate hermaphrodite buds. Within the developing floral bud, stamen
primordia must perceive ethylene at a critical time to promote the development of
carpels. Similarly, continued stamen development within the hermaphrodite bud requires
perception of ethylene by the developing carpel. Based on this model, carpel arrest

precedes stamen arrest, which would allow for the occasional female flower observed on

65

 

All genotypes bisexual at
early stages of development

Do stamens perceive C2H4?

No Yes

   

Carpels suppressed. Carpels develop.

0’ I

Do carpels perceive C2H4?

N Yes
.4. °

Male
Flowers

   

Stamens suppressed, Stamens not suppressed.

$3 33”

Female Hermaphrodite
Flowers Flowers

Figure 2.10. Revised model for hypothesized role of ethylene perception in sex

determination which suggests a reciprocal relationship between petals and stamens.

llli

Elli

Ff

35S::etrI-1 plants if the carpel managed to begin developing, but did not perceive enough
ethylene at the critical stage to maintain stamen development.

Several lines of evidence from recent studies are consistent with the proposed
relationship between stamens and carpel of this revised model. Morphological studies of
male and female cucumber bud development have shown that the divergence fi'om a
bisexual to unisexual flower begins after the initiation of carpel primordia (Bai et al.,
2004) and that the stage of floral development at which sex determination can be
modulated by ethylene treatment just preceeds, or immediately follows, differentiation of
stamen primordia (Y arnasaki et al., 2003). Since both stamens and carpels initiate prior
to sexual divergence, it is possible that signaling, stimulated via ethylene perception,
between these two adjacent whorls is responsible for the sexual fate of the flower bud. In
cucumber, the arrest of carpel development in staminate buds begins to become
morphologically apparent at the stage just prior to carpel differentiation into the stigma
and ovary (Bai et al., 2004). The arrest of stamen development in female buds becomes
apparent just afier differentiation between the anther and filament of stamen (Hao et al.,
2003; Bai etal., 2004).

In Arabidopsis, ethylene receptors act redundantly to negatively regulate ethylene
response such that knock-out of receptors leads to constitutive ethylene response (Hua
and Meyerowitz, 1998). The comparison of gynoecious (F -M-), monoecious (fiM—), and
andromonoecious ([finm) cucumber cultivars demonstrated a lack of E TRI, E T R2, and
ERS ethylene receptor expression in stamen and carpel primordia of mm plants at the time
critical for sex determination (Y amasaki et al., 2000, 2001, 2003). Therefore, it is

possible that the absence of ethylene receptors in mm stamen or carpel primordia

67

(Y arnasaki et al., 2001, 2003) could confer constitutive ethylene response, resulting in
the promotion of both carpel and stamen development. This possibility would also
explain how carpels could develop in mm plants in the absence of ethylene perception by
carpel primodia. Similarly, the arrest of the stamen development conferred by the M
allele could be the result of receptor mediated inhibition of response to ethylene at the
time and/or location critical for divergence from bisexual to unisexual flower
development. The finding that CRC::etrI-1 caused a conversion from perfect to female
flowers suggests that the inability of the carpel to perceive ethylene at the time critical for
sex produced a phenotype similar to observed stamen arrest in M- cucumber and A-
melon. Clearly, a critical factor will be the relative location and timing of expression of
both ethylene production and perception components within floral primordia.

Analysis of CRC::etr1-1 melon plants showed an additional phenotype of interest.
Although carpel-bearing buds were able to initiate, virtually all buds containing a carpel
on CRC plants aborted prior to anthesis, most before reaching 4 mm in length, suggesting
that ethylene plays a role in sustained maturation of the carpel. These results provide
support for observations made in transgenic melon constitutively overexpressing ACC
synthase, ACS, a key enzyme in ethylene biosynthesis. The 35S::A CS melon plants
showed earliness and increased production of carpel-bearing buds, and an increase in the
proportion reaching anthesis. Thus, it appears that ethylene is important for both sex
determination and pistillate bud maturation (Papadopoulou et al., 2005).

In summary, the study of CRC::etrI-I and AP3::etr1-I transgenic melon
provided unexpected results. CRC::etr1-1 melon plants showed increased femaleness as

demonstrated by earlier and increased carpel-bearing bud production and the conversion

68

of hermaphrodite to female buds. In contrast, increased maleness was observed in
AP3::etrI -1 melon plants as demonstrated by rare production of hermaphrodite buds and
poorly developed carpels in many of those which did form. These results suggest that the
critical site for ethylene perception in the determination of stamen or pistil development
does not reside within the affected organ. Instead, there appears to be a previously
unknown reciprocial relationship between stamens and carpels in their requirement for
ethylene perception for sexual organ development in hermaphrodite buds, which has led
to a revised model of ethylene’s role in sex expression of melon. Further study is
necessary to the determine the precise timing and localization of etrI -1 expression at key
stages of sex determination and to establish if the effects of altered ethylene perception in

cucumber are in line with observations made in melon.

69

leierenc

.ltcmori
s
8

Armor

BiS.‘

Bled

Box!

Bye

Debi

Den.)

5’
(I‘D

References

Atsmon, D. 1968. The interaction of genetic, environmental, and hormonal factors in
stem elongation and floral development of cucumber plants. Ann. Bot. 32: 877-
882.

Atsmon, D. and Galun E. 1962. Physiology of sex in Cucumz's sativus L. Leaf age
patterns and sexual differentiation of floral buds. Ann. Bot. 26: 137-146.

Bai, S.L., Peng, Y.B., Cui, J .X., Gu, H.T., Xu, L.Y., Li, Y.Q., Xu, Z.H., and Bai, SN.
2004. Developmental analyses reveal early arrest of the spore-bearing parts of

reproductive organs in unisexual flowers of cucumber (Cucumis sativus L.).
Planta 220: 230-240.

Bleeker, A.B., Estelle, M.A, Somerville, C., and Kende, H. 1988. Insensitivity to
ethylene conferred by a dominant mutation in Arabidopsis thaliana. Science.
241: 1086-1089.

Bleeker, AB. and Kende, H. 2000. Ethylene: a gaseous signal molecule in plants.
Annu. Rev. Cell Dev. Biol. 16: 1-18.

Bowman, IL. and Smyth, DR. 1999. CRABS CLA W, a gene that regulates carpel and
nectary development in Arabidopsis, encodes a novel protein with zinc finger and
helix-loop-helix domains. Development 126: 2387-2396.

Byers, R.E., Baker, L.R., Sell, H.M., Hemer, RC, and Dilley, DR. 1972. Ethylene: a
natural regulator of sex expression of Cucumis melo L. Proc. Natl Acad. Sci. 69:
717-720.

Deblaere, R., Bytebier, B., De Greve, H., Deboeck, F ., Schell, J ., Van Montagu, M., and
Leemans, J. (1985) Efficient octopine Ti plasmid-derived vectors for

Agrobacterium-mediated gene transfer to plant. Nucleic Acids Res. 13: 4777-
4788.

Den Nijs, A., Visser, D. 1980. Induction of male flowering in gynoecious cucumbers
(Cucumis sativus L.) by silver ions. Euphytica 29: 237-280.

Ezura, H.H., Amagai, D., Yoshioka, D., and Oosawa, K. 1992. Highly frequent
appearance of tetraploidy in regenerated melon plants, a universal phenomenon in

tissue cultures of melon (Cucumis melo). Plant Science 85: 209-213.

Fang, G. and Grumet, R. 1990. Agrobacterium tumefasciens mediated transformation
and regeneration of muskmelon plants. Plant Cell Rep. 9: 160-164.

70

Fassuliotus, G. and Nelson, B.V. 1992. Regeneration of tetraploid muskmelons from
cotyledons and their morphological differences from two diploid muskmelon
genotypes. J. Amer. Soc. Hort. Sci. 117: 863-866.

Galun, E. 1959. The role of auxin in the sex expression of the cucumber. Physiol. Plant.
12: 48-61.

Goffinet, M.C. 1990. Comparative ontogeny of male and female flowers in Cucumz's
sativus. In: Biology and Utilization of the Cucurbitaceae (eds DM Bates, RW
Robinson, and C Jeffrey). Cornell University Press, New York, pp 288-304.

Hao, Y., Wang, D., Peng, Y., Bai, S., Xu, L., Li, Y., Xu, 2., and Bai, S. 2003. DNA
damage in the early primordial anther is closely correlated with stamen arrest in
the female flower of cucumber (Cucumis sativus L.). Planta 217: 888-895.

Hua, J. and Meyerowitz, EM. 1998. Ethylene responses are negatively regulated by a
receptor gene family in Arabidopsis thaliana. Cell 94: 261-271.

Iwahori, S., Lyons J .M., and Smith OF. 1970. Sex expression in cucumber plants as
affected by 2-chloroethylphosphonic acid, ethylene, and growth regulators. Plant
Physiol. 46: 412-415.

Irish, V.F. and Yamamoto, Y.T. 1995. Conservation of floral homeotic gene function
between Arabidopsis and Antirrhinum. Plant Cell Rep. 9: 160-164.

Jack, T., Fox, G.L., and Meyerwitz, EM. 1994. Arabidopsis homeotic gene APETALA3
ectopic expression: transcriptional and posttranscriptional regulation determine
floral organ identity. Cell 76: 703-716.

Kamachi, S., Sekimoto, H., Kondo, N., and Sakai, S. 1997. Cloning of a cDNA for a 1-
aminocyclopropane-l-carboxylate synthase that is expressed during development
of female flowers at the apices of Cucumis sativus L. Plant Cell Physiology 38:
1 197-1206.

Kamachi, S., Mizusawa, H., Matsuura, S., and Sakai, S. 2000. Expressionof two 1-
aminocyclopropane—carboxylate synthase genes, CS-A CS1 and CS-A CS2,
correlated with sex phenotypes in cucumber plants (Cucumis sativus L.). Plant
Biotechnology 17: 69-74.

Kater, M.M., Franken, J ., Carney, K.J., Clombo, L., and Angenent, G.C. 2001. Sex
determination in the monoecious species cucumber is confined to specific floral

whorls. Plant Cell 13: 481-493.

Kerrigsbuch, D. and Cohen, Y. 1989. The inheritance of gynoecy in muskmelon.
Genome 33: 317-320.

71

Lee, J-. Y., Baum, S.F., Alvarez, J ., Patel, A., Chitwood, D.H., and Bowman, J .L. 2005.
Activation of CLAWS CLAW in nectaries and carpels of Arabidopsis. Plant Cell
1 7: 25-36.

Lower, R.L., and Nienhuis, J. 1990. Prospects for increasing yilds of cucumber via
Cucumis sativus var. hardwickii germplasm. In: Biology and Utilization of the
Curcurbitaceae. Eds. Bates, D.M., Robinson, R.W., J efferey, C. Ithaca NY:
Cornell Univ. Press. Pp. 397-405.

McMurray, AL. and Miller, CH. 1968. Cucumber sex expression modified by 2-
chloroethane phosphonic acid. Science 162: 1397-1398.

Mibus, H. and Tatlioglu, T. 2004. Molecular characterization and isolation of the F/f
gene for femaleness in cucumber (Cucumis sativus L.). Theor. Appl. Genet. 109:
1669-1676.

Papadopoulou, E. 2002. Sex expression in cucurbits: the role of ethylene synthesis and
perception and sex determination genes. PhD Dissertation. Michigan State
University.

Papadopoulou, E., Little, H.A., Hammar, S.A., and Grumet, R. (2005) Effect of modified
endogenous ethylene production on hermaphrodite flower development and fruit
production in melon (Cucumis melo L.). in submission.

Perl-Treves, R. 1999. Male to female conversion along the cucumber shoot: approaches
to study sex genes and floral development in Cucumis sativus. In: Sex
Determination in Plants (ed CC Ainsworth). BIOS Scientific, Oxford pp 189-215.

Rothstein, S.J., Lanhners, K.N., Lotstein, R.J., Carozzi, N.B., Jayne, S.M., and Rice DA.
1987. Promoter cassettes, antibiotice-resistance genes, and vectors for
transformation. Gene 53: 153-161.

Roy, R.P. and Saran, S. 1990 Sex expression in the Cucurbitaceae. In Biology and
utilization of the Cucurbitaceae (eds Bates, D.M., Robinson, R.W., Jeffrey, C.,
eds), pp. 251-268. Ithaca NY: Cornell Univ. Press.

Rudich, J. 1990. Biochemical aspects of hormonal regulation of sex expression in
cucurbits. In Biology and utilization of the Cucurbitaceae (eds Bates, D.M.,
Robinson, R.W., Jeffrey, C., eds), pp. 288-304. Ithaca NY: Cornell Univ. Press.

Rudich, J ., Halevy, A.H., and Kedar, N. 1969. Increase in femaleness of three cucurbits
by treatment with Etherel, an ethylene-releasing compound. Planta 86: 69-76.

Rudich, J ., Baker, L.R., Scott, J .W., and Sell, HM. 1976. Phenotypic stability and
ethylene evolution in androecious cucumber. J. Am. Soc. Hortic. Sci. 181: 48-51.

72

Sambrook, J. and Russell, D.W. 2001. Molecular Cloning A Laboratory Manual, 3rd ed.
Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.

Shifiiss, O. 1961. Sex control in cucumbers. J. Hered. 52: 5-12.

Tabei, Y., Kitade, S., Nishizawa, Y., Kikuchi, N., Kayano, T., Hibi, T., and Akustu, K.
1998. Transgenic cucumber plants harboring a rice chitinase gene exhibit
enhanced resistance to gray mold (Botrytis cinerea). Plant Cell Reports 17: 159-
164.

Steptanova AN. and Ecker J .R. 2000. Ethylene signaling: from mutants to molecules.
Current Opinion in Plant Biology 3: 353-360.

Tolla, GE. and Peterson, CE. 1979. Comparison of gibberellin A4/A7 and silver nitrate
for induction of staminate flowers in a gynoecious cucumber line. HortScience
14: 542-544.

Trebitsh, T., Staub, J .E., and O’Neill, SD. 1997. Identification of a 1-
aminocyclopropane-l-carboxylic acid synthase gene linked to the Female (F)
locus that enhances female sex expression in cucumber. Plant Physiology 113:
987-995.

Wang K.L.C., Li, H., and Ecker, J .R. 2002. Ethylene biosynthesis and signaling
networks. Plant Cell S13l-S151.

Wilkinson, J .Q., Lanahan, M.B., Clark, D.G., Bleeker, A.B., Chang, C., Meyerowitz,
E.M., and Klee, H]. 1997. A dominant mutant receptor from Arabidopsis confers
ethylene insensitivity in heterologous plants. Nature Biotechnology 15; 444-447.

Wittwer, SH. and Bukovac, M.J. 1958. The effect of gibberellins on economic crops.
Econ. Bot. 12:213-255.

Yadav, R.C., Salah, M.T., and Grumet, R. 1996. High frequency shoot regeneration from
leaf explants of muskmelon. Plant Cell Tiss. Org. Cult. 45: 207-214.

Yamasaki, S., Fujii, N., and Takahashi, H. 2000. The ethylene-regulated expression of
CS-E T R2 and CS-ERS genes in cucumber plants and their possible involvement
with sex expression in flowers. Plant Cell Physiol. 41: 608-616.

Yamasaki, S., Fujii, N., Matsuura, S., Mizusawa, H., and Takahashi, H. 2001. The M
locus and ethylene-controlled sex determination in andromonoecious cucumber
plants. Plant Cell Physiol. 42: 608-619.

Yamasaki, S., Fujii, N., and Takahashi, H. 2003. Characterization of ethylene effects on
sex determination in cucumber plants. Sex Plant Reproduction 16: 103-111.

73

Yin, T. and Quinn, J .A. 1992. A mechanistic model of a single hormone regulating both
sexes in flowering plants. Bulletin of the Torey Botanical Club 119: 431-441.

Yin, T., and Quinn J .A. 1995. Tests of a mechanistic model of one hormone regulating
both sexes in Cucumis sativus (Cucurbitaceae). Am. J. Bot. 82: 1537-1546.

74

Chapter 3
Evaluation of the effect of modified endogenous ethylene production
on sex expression and fruit production in

field grown transgenic Cucumis melo

Abstract

Sex expression in cucurbit species is affected by environmental, hormonal,
developmental, and genetic factors. Ethylene is considered the primary hormone
promoting femaleness in cucumber, melon, and squash; exogenous ethylene increases
femaleness while inhibition of ethylene perception or production decreases femaleness.
Greenhouse experiments with transgenic andromonoecious melon constitutively
overexpressing an ACS ( l -aminocyclopropane-l -carboxylic acid synthase) gene for
ethylene biosynthesis showed increased ethylene evolution from leaves and floral buds,
increased numbers of hermaphroditic buds, and an increased percentage of hermaphrodite
buds that reached full anthesis. In this study I sought to determine whether increased
femaleness would be observed in plants grown in field conditions and to determine the
effect of increased ethylene and modified sex expression on fruit production. Despite no
enhanced earliness in hermaphrodite bud initiation, field grown transgenic 35S:A CS
melon plants exhibited significantly earlier and increased mature hermaphroditic flower
production than non-transgenic controls. In addition, ACS plants showed earlier fruit set
and increased sequential fruit set on the main stem. These experiments confirmed the
importance of endogenous ethylene production for hermaphrodite flower production and
maturation, and also suggest a relationship between of ethylene levels and fruit set

patterns on the main stem.

75

Introduction

The Cucurbitaceae family is comprised of species that exhibit a range of heritable
patterns of sex expression (reviewed in Perl-Treves, 1999 and Roy and Saran, 1990).
For example, cucumber (Cucumis sativus) is typically monoecious, in which a phase of
male flowers precedes a phase of both male and female flowers, and melon is typically
andromonoecious with male flowers followed by male and perfect flowers. Despite
differentiation into unisexual flowers, in early stages of development, all initiated floral
buds contain the four whorls typical of a hermaphrodite flower, sepals, petals, stamens,
and carpels (Goffinet, 1990; Kater et al., 2001; Bai et al., 2004). The subsequent
inhibition of carpel primordia results in staminate flower development, while arrest of
stamen primordia results in pistillate flower development.

Sexual phenotypes observed in cucumber and melon are subject to the control of
various developmental, genetic, hormonal, and environmental factors. Application of
auxin, brassinosteroids, ethylene, or gibberellin can influence the sexual fate of
developing flower buds (reviewed by Rudich, 1990 and Perl-Treves, 1999; Papadopoulou
and Grumet, 2005). Ethylene appears to be the primary hormone affecting sex
expression with other hormones acting via ethylene. Application of ethylene or ethylene-
releasing compounds has been demonstrated to increase pistillate flower production in
cucumber, muskmelon, and squash (Robinson et al., 1969; Rudich et al., 1969; Yin and
Quinn, 1995) while treatment with inhibitors of ethylene biosynthesis or action confers
an increase in maleness (Rudich, 1990; Atsmon and Tabbak, 1979). In addition, support
for the feminizing effect of ethylene is provided by measurements of ethylene evolution,

which shown 2-3 fold higher ethylene production from gynoecious compared to

76

monoecious and andromonoecious cucumber seedlings (Byers et al., 1972; Rudich et al.,
1976). While an analogous difference in ethylene production among melon sexual
phenotypes has not been observed, the application of hypobaric conditions to melon
plants to reduce internal gas concentrations has been shown to reduce femaleness (Byers
et al., 1972).

Ethylene is synthesized from S-adenosyl methionine (AdoMet) in a two step
process which includes cyclization to l-aminocyclopropane carboxylate (ACC) mediated
by ACC synthase (ACS) followed by an oxidation step mediated by ACC oxidase (ACO)
(Adams and Yang, 1979). Although ACS and ACO are both encoded by multigene
families and, as such, are subject to regulation by various inducers, including auxin,
ethylene, wounding, fruit ripening, pathogens, and stress, the level of ACS is generally
regarded as rate-limiting in ethylene production (Zarembinski and Theologis, 1994;

, Bleeker and Kende, 2000).

Comparative studies of near-isogenic gynoecious and monoecious cucumber lines
has led to the identification and molecular characterization of an additional ACS gene
(CS-A CS1 G) present only in gynoecious genotypes and completely co-segregating with
the semi-dominant F (Female) locus (Trebitsh etal., 1997; Mibus and Tatlioglu, 2004).
Gynoecious (F -) cucumber cultivars provide earlier, more uniform fruit set (Lower and
Nienhuis, 1990). An equivalent to the F locus allowing stable inheritance of gynoecy has
not been identified in melon (Kenigsbuch and Cohen, 1989).

To determine whether a similar feminizing effect could be achieved in melon via
increased endogenous ethylene production, transgenic melons were produced haboring an

ACC synthase fi'om petunia driven by a constitutive (3SS-CaMV) promoter

77

(Papadopoulou et al., 2005). Greenhouse studies of the 35S::A CS melon plants showed
phenotypes characteristic of increased endogenous ACS, which included increased ACC
content and ethylene evolution fiom leaves, staminate buds, and hermaphroditic buds.
These studies also demonstrated that increased endogenous ethylene production
conferred earlier and increased perfect flowering to andromonoecious melon. In
addition, greenhouse study showed earlier and increased numbers of hermaphrodite buds
reaching anthesis. This observation was a novel finding and suggested a previous
unknown role of ethylene in the maturation of carpel-bearing buds.

In this study, I sought to examine whether the sex-related phenotypes also would
be observed under field conditions and to test for effects on fi'uit set and development.
The results of this field evaluation further demonstrate the role of ethylene the
development of mature hermaphrodite buds and offer insight into resource allocation

associated with fruit set.

Materials and Methods

Northern analysis

Fresh young leaf tissue of 35S::ACS lines 1, 3, and 4, non-transgenic segregants, and wild
type andromonoecious Hale’s Best Jumbo was collected from T2 field-grown melon
plants to evaluate ACS transgene expression. RNA was isolated fiom 100 mg fresh leaf
tissue frozen in liquid nitrogen using ConcertTM Plant RNA Reagent (Invitrogen,
Carlsbad, California). RNA quantification and purity was determined by measuring the
absorbance of the samples at 260 nm and by gel electrophoresis. Total RNA (20 pg) was

electrophoresed on a 1.5% agarose formaldehyde gel and blotted (Sambrook and Russell,

78

2001). The membrane was hybridized overnight at 55°C with a digoxigenin-UTP labeled
RNA probe of ACS (Roche Diagnositics, Germany). Stringency washes and detection
via chemiluminescent visualization with CDP-Star was carried out following protocols
developed for the DIG system (Roche Diagnostics, Germany).

Field experiments

Field trials were carried out in East Lansing, Michigan during the summer of 2002
using a randomized complete block design with three replications to evaluate flowering
and fruit-set of 35S::A CS lines 3 and 4, non-transgenic segregants, and Hales Best
Jumbo. Plants were started in the greenhouse 19 days before hand transplanting, and
transgenic plants were selected based on NPTII ELISA (Agia®, Elkhart, Indiana). Each
replication contained five plants of each genotype (exception: NT segregants, two
plants/rep). Rows of black plastic mulch were spaced three meters apart; plants were
spaced at 0.9 meter intervals within rows. Within each row, genotypes were separated by
two Hale’s Best Jumbo plants. The field was surrounded by two border rows of non-
transgenic Hale’s Best Jumbo. Plot management included standard fertilizer application,
weekly fungicide treatment, and irrigation of 2.5 cm per week.

Each node along the main stem of individual plants was examined to determine
the node position of the first visible hermaphrodite bud and first mature hermaphrodite
flower, the total number of hermaphrodite buds and mature hermaphrodite flowers, and
the position and number of fi'uit set. For evaluation of ripening, all fi'uit within a plot
were scored for maturity. Fruit were placed in five categories: overripe, ripe (full-slip),
not ripe (tan and fully netted but still attached to vine), immature (green netted), and

small (green not netted). All fi'uit was hand harvested ten weeks after field planting.

79

Fruit within a single plot were pooled by genotype and evaluated for total number, total
weight, and size at maturity. Analysis of variance and least square difference

comparisons were carried out using SAS software (SAS Institute, Cary, North Carolina).

Results
Northern analysis

Northern hybridization analysis of leaf tissue confirmed the expression of
introduced ACS gene in field grown transgenic ACS plants (Figure 3.1). The lack of
detection of transcript in the Hales wild type and non-transgenic segregant (NT) controls
demonstrates the specificity of the probe for the transgene. The levels of expression
varied between transgenic families with ACS4 and ACS3 showing higher expression.
This northern data is consistent with previous segregation data suggesting two copies of
the ACS transgene in ACS3 and ACS4 (Papadopoulou et al., 2005).
Sex expression and hermaphrodite bud development

Field grown transgenic ACS3 and ACS4 were compared to non-transformed
Hale’s Best Jumbo (WT) and azygous, non-transgenic segregants (AZ) with regard to sex
expression patterns and hermaphrodite bud maturation. Hale’s Best Jumbo plants are
diploid, and 35S::A CS transgenic lines and their azygous siblings are tetraploid
(Papadopoulou, 2002). Tetraploidy is a common occurrence in melons regenerated from
tissue culture (Ezura et al., 1992; Yadav et al., 1996) but does not affect sex expression

patterns of melon in greenhouse studies (Papadopoulou et al., 2005).

80

 

WT AZY ACS1 ACS4 . j

 

Figure 3.1. Northern hybridization of heterologous ACS gene expression in leaf tissue
from field grown T2 transgenic 35S::A CS lines ACSl, ACS3, and ACS4. The top panel
shows hybridization of total RNA from non-transgenic wild-type Hale’s Best Jumbo
plants (WT), azygous non-transgenic segregants (AZY), and transgenic lines (ACSl,
ACS3, and ACS4) with DIG-labeled ACS probe. The bottom panel shows rRNA stained

with ethidium bromide.

81

All genotypes produced hermaphrodite buds earlier in the field than in the
greenhouse, node 5-6 in the field versus node 6-9 for 35S::ACS plants and node 10-11 for
non-transgenic control plants in the greenhouse (Papadopoulou et al., 2005). While there
were no differences in the time of appearance or number of hermaphroditic buds in the
field, there was a significant difference between transgenic 35S plants and non-transgenic
wild type Hale’s Best Jumbo (WT) and non-transgenic tetraploid segregant (azygous,
AZY) plants in the appearance and number of mature hermaphroditic flowers on the first
25 nodes of the main stem (Figure 3.2, 3.3). ACS plants produced mature hermaphroditic
flowers 2-3 nodes earlier than WT and AZ plants and had higher percentage of the
hermaphrodite buds reach maturity. On the non-transgenic plants, only about 30% of the
hermaphroditic buds reached anthesis, compared to 55% on transgenic plants (Table 3.1).
In addition, ACS3 and ACS4 plants were more likely to produce mature flowers on
closely spaced nodes of the main stem than non-transgenic lines (Figure 3.3, Table 3.1).
Fruit development

The field grown melon plants also were observed for fruit set and ripening.
Consistent with earlier appearance of mature hermaphrodite flowers on the main stem,
ACS plants set fi'uit significantly earlier along the main stem with fruit appearing on
ACS3 and ACS4 at approximately node 8 and 9, respectively, compared to nodes 16 and
21 in azygous and Hale’s Best Jumbo controls (Figure 3.4A). In addition, total number
of fruit set along the main stem was 2-4 fold higher in ACS plants than on non-transgenic
controls (Figure 3.48). Evaluation of the pattern of fi'uit on the main stem of ACS3 and

ACS4 plants also demonstrated more closely spaced fruit, with 60-70% of the plants

82

 

Figure 3.2. Sex expression pattern along the main stem of wild type Hale’s Best
Best Jumbo (WT), azygous segregants (AZY), and transgenic T2 35S::ACS
(ACS3 and ACS4) field grown melon plants. (A) Node position of first
hermaphrodite bud. (B) Total number of hermaphrodite buds on first 25 nodes.
(C) Node position of the first hermaphrodite flower to reach anthesis. (D) Total
number of hermaphrodite flowers that reached anthesis. Experiments were
performed using a randomized complete block design with three replicates. (All
data are means i standard error; bars marked with different letters are

significantly different, Least square difference, P=0.05.)

83

 

Number of Node

Number of Buds

00m ‘1 0

N0) 5

Number of Node

Number of Flowers

 

I

 

Total

 

 

 

 

 

 

WT

1 “at"?
.mztza
‘ 1'.‘

k. c
:43? l
,5} inr‘ll‘L-l‘lu a

AZY ACS3 A

84

p:

i‘l Ti

.2 in;
3:: {El angst:
l‘ I‘ 3;: O. 01;,

:e'mr‘r
: damage 5
:a: y; uv 3
a ‘.‘.‘,{:fi "
(an. H"
0" I’d):
‘n! :-

'i‘f-Zil
.2‘

It
.

0

.g:}..::.:

 

S4

 

 

ACS3 AZY WT

ACS4

 

10 5 10 5
Node Position Node Position Node Position

10 5

Figure 3.3. Pattern of mature hermaphrodite flower and fruit set on main stem of field
grown wild type (WT), azygous segregants (AZY) and 35S::A CS (ACS3 and ACS4)

melon plants. Nodes bearing mature hermaphrodite flowers that did not develop into

fruit are depicted by O and nodes bearing fruit by 0. Horizontal lines represent the
main stem of individual plants. Each of the three sets of plants represents a replicated

plot.

85

Table 3.1. Flowering, fi'uit set, and yield characteristics of field grown 35S::A CS and

control melon plants.

 

 

 

 

 

 

 

Characteristic WT AZY ACS3 ACS4
(ploidy) (2n) (4n) (4n) (4n)

% mature perfect flowers

24.0:t2.0 a 39.1i1.1 b 57.4il.l c 52.0:t3.4 c
(main stem)
Instances of adjacent mature

0.9i 0.3 a 3.2i 0.7 b 6.2i 0.6 c 5.73: 0.6 c
perfect flowers (main stem)
% mature perfect flowers

15.0t4.0 a 20.8i 8.1 a 27.8i1.8 a 22.22il.6 a
setting fruit (main stem)
% plants with 2 fruit within 4

6713.8 3 16.7:t9.6 a 60.0:t6.7 b 71.2i3.5 b
nodes (main stem)
Average fruit size (kg) 0.911 0.17 a 0.47zt0.01 b 0.45i0.02 b 0.43i0.01 b
Total number of fruit set per

9.6110 a 6.83:2] a l 1.63:1 .0 a 9.9il.5 a
plant

 

Data are mean of three replicate plots istandard error.

Values within a row followed by a different letter are significantly different by LSD,

P=0.05.

86

Figure 3.4. Fruit set and ripening pattern of field grown wild type Hale’s Best Jumbo
(WT), azygous segregants (AZY), and T2 35S::A CS melon plants (AC83 and ACS4). (A)
Node position of first fruit on the main stem. (B) Total fruit set on first 25 nodes of main
stem. (C) Profile of fruit development. All data are means of three replicate plots i
standard error. Bars marked with different letters are significantly different, Least square
difference, P=0.05. Ripening data were collected on the fruit within the plot (adjusted to
per plant basis) at 8 ‘/2 weeks afier planting. Individual fruit were scored as: OR, over-
ripe; Ripe, full-slip; Not Ripe, tan and fully netted but still attached to plant; Immature,

green netted; or Small, green not netted.

87

A

Appearance of First Fruit

 

I HALES

 

 

 

0
3

5

050
4|

ouoz we non—.52

Total Fruit on Main Stem

 

 

. .
may...
.nswu.

um. 1

.mwm-.~.-un.a.

2:}. i

u»
n...“

 

 

z . .

4 3 2 1

=2". so 85::

Profile of Frui

6 5 4 3 2 4| 0
=3". .6 .2252

 

Not Ripe Immature Small

Ripe

OR

88

setting fi'uit within four nodes versus 7-17% in non-transgenic controls (Figure 3.3, Table
3.1). In several cases, fi'uit set on three consecutive nodes and in one extreme case on
five consecutive nodes (Figure 3.3, Figure 3.5). All lines showed similar proportions of
mature hermaphrodite buds producing fruit on the main stem (Table 3.1).

The fi'uits within each plot were evaluated for stage of ripening at 8 1/2 weeks post-
transplanting. Fruit on ACS3 and ACS4 plants ripened sooner than on non-transgenic
plants as depicted by no non-transgenic fruit in the over-ripe category and ahnost
exclusively ACS fruit in the ripe category at the time of observation (Figure 3.4C). The
profile also shows a more uniform distribution among stages of ripening of ACS3 and
ACS4, while the majority of fruit from the non-transgenic plants cluster in the not-ripe
category, suggesting different temporal patterns of hit set (Figure 3.4C).

At harvest, all fi'uit from the plants were collected. Despite higher numbers of
fruit on the main stem of ACS plants, the genotypes did not differ in the total number of
fruit set per plant (Table 3.1), indicating a difference in fruit distribution on the plant,
such that more fruit was set on lateral branches of non-transgenic than transgenic plants.
Fruit from wild type Hale’s Best Jumbo plants were larger than fruit fi'om the azygous
and transgenic plants (Table 3.1). The smaller fruit size from azygous and transgenic
plants can be attributed to the effects of tetraploidy, as has previously been observed

(Nugent, 1994a,b).

89

 

Figure 3.5. Consecutive node fruit set on main stem of field grown Hale’s Best Jumbo
wild type (WT) melon and 35S::ACS transgenic melon. (A) Typical main stem fruit set of
Hale’s Best Jumbo WT. (B) Extreme fruit set on consecutive nodes of main stem in

ACS3.

90

Discussion

Treatment with ethylene or ethylene-releasing compounds, such as ethephon, has
a feminizing effect on a range of cucurbits, including Cucumis melo (Robinson et al.,
1969; Rudich, 1969; Owens et al., 1980; Sadhu and Das, 1978), and so it was predicted
that increased endogenous ethylene would similarly enhance femaleness, potentially
allowing for earlier hermaphrodite flower appearance and fruit set, and a shorter growing
season. Our earlier greenhouse studies with 35S::A CS melon plants verified that the ACS
transgene conferred increased endogenous ethylene production and earlier appearance of
the first hermaphroditic bud on the main stem (Papadopoulou et al., 2005). In addition to
an effect on sex determination, the transgenic 35S::A CS plants showed an increase in the
earliness and proportion of hermaphrodite buds that matured to anthesis suggesting an
additional role of ethylene in carpel-bearing bud maturation.

In the field, 35S::A CS melon plants produced mature hermaphrodite flowers
sooner and a higher percentage of hermaphrodite buds reached anthesis on the main stem.
The increase and earliness of mature hermaphrodite buds supports greenhouse results
indicating that ethylene is needed for hermaphrodite bud to reach anthesis deve10pment,
and rules out the possibly that failure of buds to mature in the greenhouse was an artifact
of the greenhouse environment. A possible role of ethylene in carpel development is not
unprecedented. In orchid, ethylene biosynthesis stimulated by pollination is critical for
subsequent ovary maturation (Bui and O’Neill, 1998; Zhang and O’Neill, 1993), and in
tobacco, down-regulation of a pistil-specific ACC oxidase gene resulted in the arrest of

ovule development that could be reversed with exogenous ethylene (DeMartinis and

91

Mariani, 1999). Whether ethylene is needed for ovule development or other aspects of
hermaphrodite bud maturation in melon remains to be determined.

The 35S::A CS melon plants also showed earlier and greater fruit set on the main
stem and exhibited fruit set on closely spaced nodes. While earlier appearance of mature
hermaphroditic flowers is a prerequisite for earlier fi'uit set, the plants must also be
developmentally ready to bear fi'uit. They must, therefore, have the photosynthetic
capacity to establish and maintain the sink a developing fruit creates. The earlier fi'uit set
on the main stem of ACS plants seen in the field trial demonstrated that the transgenic
plants had the resource capacity to set fruit earlier.

The ACS melon plants showed a strikingly different pattern of fruit set on the
main stem. Multiple ACS plants set four or more fruit on the main stem. In one extreme
case, five fruit were set on consecutive nodes, and in another case, eight fruit were on the
main stem with no more than two nodes separating any of the fi'uit. This is a sharp
contrast to what was observed on the main stems of non-transgenic plants where typically
only one fruit was set and in only two instances were two fruit set within four nodes.
Previous work in the area of plant reproductive ecology has suggested that in some
monoecious plant species, sex expression is a reflection of resource availability, such that
a developing fruit results in a reduction in pistillate flower production while the lack of
fruit results in increased pistillate flower production (Stephenson, 1981; Schapendonk
and Brouwer, 1984; Krupnick and Weis, 1998; Krupnick et al., 1999). A developing fi'uit
in squash and gourd was shown to lead to reduced subsequent pistillate flower production
and increased fruit abortions (Stephenson et al., 1988). The relationship of developing

fruits to endogenous ethylene levels has been studied in monoecious Cucurbita texana

92

(Krupnick et al., 1999). These studies show that internal levels of ethylene were highest
adjacent to tips of branches bearing fruit nearing maturity and lowest when branches
canied two or more fruit, suggesting that the number of developing fruit and their stages
of development regulate endogenous ethylene levels (Krupnick et al., 1999). If an
analogous relationship between ethylene production and fi'uit exists in Cucumis melo,
then the increased production of endogenous ethylene in the transgenic lines may
interfere with the signaling which would normally result in lower ethylene and decreased
pistillate flowering on branches carrying fi'uit.

The fruit on ACS plants also showed a different trend of maturation as fruit
ripened sooner on ACS3 and ACS4 plants than on non-transgenic controls. Confounding
factors may play a role in the earlier ripening observed in the fi'uit on ACS plants. The
earlier fruit set on transgenic plants relative to non-transgenic would confer earlier
ripening regardless of any effect on the time required for maturation. Secondly, in
climacteric fi'uit, a concomitant burst in respiration and ethylene initiates the ripening of
mature fruit (Giovannoni, 2004); thus, the increased levels of endogenous ethylene in
ACS plants may hasten the climacteric and speed up the ripening process.

In summary, evaluation in a field setting made possible further phenotypic
analysis of transgenic melon over-expressing an ACC synthase gene. Field grown
35S::A CS melon plants produced earlier and greater numbers of mature hermaphrodite
flowers and showed increased fruit set on the main stem. These experiments demonstrate
the importance of endogenous ethylene production for hermaphrodite flower
development in andromonoecious melon and suggest a role of ethylene in signaling

associated with fruit set patterns. 35S::A CS melon plants also exhibited earlier fruit set, a

93

characteristic which may be horticulturally advantageous in areas, such as Michigan, with
a short growing season. Engineering of melon with ACS gene constructs conferring
localization of A CS expression to regions of the developing floral organ at the times
critical for sex determination and maturation may help to achieve the benefits of
increased earlier fi'uit set without potentially agronomically underdesirable effects of

constitutively elevated ethylene production.

94

References

Adams, DO. and Yang, SF. 1979. Ethylene biosynthesis: identification of 1-
aminocycloprpane-l-carboxylic acid as an intermediate in the conversion of
methionine to ethylene. Proc. Natl. Acad. Sci. 76: 170-174.

Atsmon, D. and Tabbak, C. 1979. Comparative effects of gibberellins, silver nitrate and
aminoethoxyvinyl glycine on sexual tendency and ethylene in the cucumber plant
(Cucumis sativus L.). Plant Cell Physiology 20: 1547-1555.

Bai, S.L., Peng, Y.B., Cui, J .X., Gu, H.T., Xu, L.Y., Li, Y.Q., Xu, Z.H., and Bai, SN.
2004. Developmental analyses reveal early arrest of the spore-bearing parts of

reproductive organs in unisexual flowers of cucumber (Cucumis sativus L.).
Planta 220: 230-240.

Bleeker, AB. and Kende, H. 2000. Ethylene: a gaseous signal molecule in plants.
Annu. Rev. Cell Dev. Biol. 16: 1-18.

Bui, A.Q. and O’Neill, SD. 1998. Three 1-aminocyclopropane—l-carboxylate synthase
genes regulated by primary and secondary pollination signals in orchid flowers.
Plant Physiol. 116: 419-428.

Byers, R.E., Baker, L.R., Sell, H.M., Herner, R.C., and Dilley, DR. 1972. Ethylene: a
natural regulator of sex expression of Cucumis melo L. Proc. Natl Acad. Sci. 69:
717-720.

De Martinis, D. and Mariani, C. 1999. Silencing gene expression of the ethylene
forming enzyme results in a reversible inhibition of ovule development in
transgenic tobacco plants. Plant Cell 11: 1061-1071.

Ezura, H.H., Amagai, D., Yoshioka, D., and Oosawa, K. 1992. Highly fi'equent
appearance of tetraploidy in regenerated melon plants, a universal phenomenon in
tissue cultures of melon (Cucumis melo). Plant Science 85: 209-213.

Giovannoni, J .J . (2004) Genetic regulation of fruit development and ripening. Plant Cell
16: $170-$180.

Goffrnet, M.C. 1990. Comparative ontogeny of male and female flowers in Cucumis
sativus. In: Biology and Utilization of the Cucurbitaceae (eds DM Bates, RW
Robinson, and C Jeffrey). Cornell University Press, New York, pp 288-304.

Kater, M.M., Franken, J ., Carney, K.J., Clombo, L., and Angenent, G.C. 2001. Sex

determination in the monoecious species cucumber is confined to specific floral
whorls. Plant Cell 13: 481-493.

95

Kerrigsbuch, D. and Cohen, Y. 1989. The inheritance of gynoecy in muskmelon.
Genome 33: 317-320.

Krupnick, GA. and Weis A.E. 1998. Floral herbivore effect on the sex expression of an
andromonecious plant, Isomeris arborea (Capparaceae). Plant Ecology 134:
1 51-162.

Krupnick, G.A., Brown, K.M., and Stephenson, A.G. 1999. The influence of fruit on the
regulation of internal ethylene concentrations and sex expression in Cucurbita
texana. International Journal of Plant Sciences 160: 321-330.

Lower, R.L. and Nienhuis J. 1990. Prospects for increasing yields of cucumber via
Cucumis sativus var. hardwickii germplasm. In: Biology and Utilization of
Cucurbitaceae. (Bates, D.M., Robinson, R.W., Jeffrey, C., eds) pp. 397-405.
Ithaca NY: Cornell Univ. Press.

Mibus, H. and Tatlioglu, T. 2004. Molecular characterization and isolation of the F/f
gene for femaleness in cucumber (Cucumis sativus L.). Theor. Appl. Genet. 109:
1669-1676.

Owens, K.W., Peterson, CE. and Tolla, GE. 1980. Prodution of hermaphrodite flowers
on gynoecious muskmelon by silver nitrate and aminoethyoxyvinyl glycine.
HortScience 15: 654-655.

Nugent, P.E. 1994a. Monoecious flowering, tetraploid, virescent melon C879-J2-4X.
HortScience 29: 47-48.

Nugent, P.E. 1994b. Tetraploid ‘Planters Jumbo” melon lines C883-m6-4x and 67-m6-
100-4x. HortScience 29: 48-49.

Papadopoulou, E. 2002. Sex expression in cucurbits: the role of ethylene synthesis and
perception and sex determination genes. PhD Dissertation. Michigan State
University.

Papadopoulou, E., Little, H.A., Hammar, S.A., and Grumet, R. (2005) Effect of modified
endogenous ethylene production on hermaphrodite flower development and film
production in melon (Cucumis melo L.). in submission.

Perl-Treves, R. 1999. Male to female conversion along the cucumber shoot: approaches
to study sex genes and floral development in Cucumis sativus. In: Sex
Determination in Plants (ed CC Ainsworth). BIOS Scientific, Oxford pp 189-215.

Robinson, R.W., Shannon, S., and DeLaGuardia, MD. 1969. Regulation of sex
expression in the cucumber. BioScience 19: 141-142.

96

Roy, R.P. and Saran, S. 1990. Sex expression in Cucurcitaceae. In Biology and
utilization of the Cucurbitaceae (eds Bates, D.M., Robinson, R.W., Jeffrey, C.,
eds), pp. 252-268. Ithaca NY: Cornell Univ. Press.

Rudich, J. 1990. Biochemical aspects of hormonal regulation of sex expression in
cucurbits. In Biology and utilization of the Cucurbitaceae (eds Bates, D.M.,
Robinson, R.W., Jeffrey, C., eds), pp. 288-304. Ithaca NY: Cornell Univ. Press.

Rudich, J ., Halevy, AH. and Kedar, N. 1969. Increase in femaleness of three cucurbits
by treatment with Etherel, an ethylene-releasing compound. Planta 86: 69-76.

Rudich, J ., Baker, L.R., Scott, J .W., and Sell, HM. 1976. Phenotypic stability and
ethylene evolution in androecious cucumber. J. Am. Soc. Hortic. Sci. 181: 48-51.

Sadhu, M.K. and Das, PC. 1978. Effect of ethrel (ethephon) on the growth, flowering,
and fruiting of three cucurbits. Journal of Horticultural Science 53: 1-4.

Sambrook, J. and Russell, D.W. 2001. Molecular Cloning A Laboratory Manual, 3rd ed.
Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.

Schapendonk, A.H.C.M. and Brouwer P. 1984 Fruit growth of cucumber in relation to
assimilate supply and sink activity. Sci. Hortic. 23:21-33.

Stephenson, A.G. 1981. Flower and fi'uit abortion: proximate causes and ultimate
functions. Ann. Rev. Ecol.Syst. 12: 253-79.

Stephenson, A.G., Devlin B., and Horton, J .B. 1988. The effects of seed number and
prior fi'uit dominance on the pattern of fruit production in Curcurbita pepo
(zucchini squash). Ann. Bot. 62: 653-661.

Trebitsh, T., Staub, J .E. and O’Neill, SD. 1997. Identification of a 1-
aminocyclopropane-l -carboxylic acid synthase gene linked to the Female (F)
locus that enhances female sex expression in cucumber. Plant Physiology 113:
987-995.

Yadav, R.C., Salah, M.T., and Grumet, R. 1996. High frequency shoot regeneration from
leaf explants of muskmelon. Plant Cell Tiss. Org. Cult. 45: 207-214.

Yin, T. and Quinn, J .A. 1992. A mechanistic model of a single hormone regulating both
sexes in flowering plants. Bulletin of the Torey Botanical Club 119: 431-441.

Yin, T., and Quinn, J .A. 1995. Tests of a mechanistic model of one hormone regulating
both sexes in Cucumis sativus (Cucurbitaceae). Am. J. Bot. 82: 153 7-1546.

Zarembinski, T.I. and Theologis, A. 1994. Ethylene biosynthesis and action: a case of
conservation. Plant Mol. Biol. 26: 1579-1597.

97

Zhang, X.S. and O’Neill, SD. 1993. Ovary and gametophyte development are

coordinately regulated by auxin and ethylene following pollination. Plant Cell 5:
403-418.

98

Chapter 4
Risk assessment of transgenic plants with altered signaling pathways:

A case study of modified ethylene

Abstract

The past decade has seen a tremendous increase in the types of genes and variety
of species used for crop improvement via genetic engineering. Many current efforts
include transformation with genes that encode transcription factors, members of signal
transduction networks, or key enzymes in metabolic pathways. Since these genes can
affect the expression of additional (often many) genes and pathways, their introduction
may have secondary effects on plant phenotypes, which may be important to consider
fiom an environmental risk assessment standpoint. The goal of this case study is to focus
on the effects associated with genetic engineering to modify traits regulated by ethylene.
Ethylene production and perception have been modified to alter plant physiological
characteristics including ripening, senescence, and flowering. These alterations may in
turn result in secondary effects such as altered stress and disease response, as well as
changes in floral characteristics and growth habit. In addition, pleiotropic effects may
alter the likelihood of gene flow into neighboring non-transgenic populations and/or alter
reproductive fitness of emerging hybrids in the event of gene flow. These possibilities

will be discussed within the context of modification of the ethylene signaling pathway.

99

Introduction

With 67.7 million hectare of biotech crops grown worldwide in 2003, the
commercialization and widespread cultivation of genetically engineered crops is a reality
(ISAAA; http://www.isaaa.org). The majority of transgenics in cultivation are
engineered for two traits, herbicide resistance and insect resistance, with a small amount
of acreage for virus resistant plants. In this initial wave of transgenic crops, the genes
transformed into the plant encode a protein product which itself confers the trait of
interest. For example, the widely cultivated Roundup ReadyTM soybean harbors a single
transgene isolated from a soil bacterium, which renders the plant resistant to the
biodegradable glyphosate herbicide Roundup® (Padgette et al., 1995). Another example
is Bt-corn, which is engineered to express a cry gene that encodes a protein toxic to
certain species of insects, therefore, conferring insect resistance to the plant (Peferoen,
1997). In these cases, even though the protein is present throughout the plant, it does not
play a role in the plant’s own growth and development. Changes in the plant phenotype
other than the direct effects of the introduced trait are likely to be a result of a gene
position effect and would be selected against in the early stages of evaluations of the
transgenic plant material.

A much broader range of traits and variety of genes are under current research and
development in plant biotechnology laboratories, including genes for increased stress
tolerance, enhanced disease resistance, modified metabolism, and altered growth and
development (Wolfenberger and Grumet, 2003). Such modifications could result in the
ability to cultivate crops in regions previously not possible, reduce chemical input, or

tailor the content of the plants for nutritional or industrial needs. Unlike the genes used

100

to engineer traits of the initial wave of transgenics, these newer traits often utilize genes
which affect regulatory, metabolic, or signaling pathways of the plant. For example,
engineering increased tolerance to dehydration stress by overexpression of the C-
repeat/dehydration-responsive element binding factor (At-CBF , -DREB), a transcription
factor, stimulates the expression of a regulon of genes involved in the acclimation to
environmental stresses related to dehydration (ie fieezing, drought, salinity) (J aglo-
Ottosen et al., 1998; Thomashow, 1999; Fowler and Thomashow, 2002). Another
example is the modification of the signaling pathway of the plant hormone ethylene,
through the expression of a mutant ethylene receptor or antisense biosynthetic gene in an
effort to delay ripening or flower senescence (Shaw et al., 2002).

In these cases, the introduced trans gene, as a component of a transcription or
signaling pathway, will affect the activities of many additional gene products and may,
therefore, have a greater probably for unanticipated secondary effects on the engineered
plant (Wolfenbarger and Grumet, 2003). For example, a detailed microarray analysis of
gene expression in response to constitutive expression of At-CBF in Arabidopsis showed
that approximately 300 genes of the 8000 evaluated were affected (Fowler and
Thomashow, 2002). Such changes in gene expression could in turn cause ecologically
relevant changes in phenotype. As conscientious scientists, it is important to continually
assess the risk associated with the introduction of new plant varieties, in an effort to
determine if they pose new or additional risks, which might affect their impact on the
environment. Relevant changes include those which might affect gene flow as well as
those affecting the fitness of the resulting hybrid plant in the event of gene escape (Snow

2002b; Hancock, 2003).

101

In terms of gene flow, it can be generally assumed that if crops are grown within
close proximity to wild relatives, hybridization events will occur (Ellstrand, 2001;
Hancock, 2003). This concept has been a reality in conventional agiculture and is,
therefore, not unique to genetically modified crop plants. The ability of most cultivated
crops to mate with wild relatives is well documented (Snow, 2002a; Ellstrand et al.,
1999). Many factors influence gene flow, such as mating system, mode of pollination,
mode of dispersal of seed or vegetative propagules, and environmental characteristics;
therefore, phenotypic changes to multiple plant structures need to be carefully evaluated
when characterizing gene flow (Snow, 2002b; Messeguar, 2003). Since secondary
effects resulting from the introduced transgene may cause one or more of these factors to
deviate from what is observed in their non-transgenic source, careful evaluation of
attributes affecting gene flow should be considered in the establishment of confinement
criteria for field gown transgenic plants. For example, dispersal of pollen alone can be
affected by many components, including number of flowers per plant, amount of pollen
per flower, number of anthers per flower, and longevity of pollen, as well as changes in
floral structure that alter the attractiveness and/or accessibility to flowers by pollinators
(some which may not otherwise visit). Moreover, since pollen is not the only mechanism
of gene flow, phenotypic alterations affecting the movement of seed and vegetative
propagules are also relevant from a risk assessment perspective and may include changes
in seed dormancy, timing of ripening, seed shedding or shattering, and dispersal by
animals and environmental elements (Snow, 2002b).

In addition to understanding how secondary effects impact gene flow, knowledge

regarding how the transgene may impact fitness of the recipient organism and the

102

ecosystem is crucial to risk assessment. Two of the most important questions to address
when considering risk assessment of transgenic crops are: 1. Will the transgene
contribute to the evolution of increased invasiveness in natural populations?; and 2. How
will the transgene impact non-target species? (Snow and Palma, 1997; Hails, 2000;
Ellstrand, 2001). If the transgene has the potential to increase the invasiveness of a
natural population, high levels of successful introgession could result in reduced natural
genetic diversity and even contribute to extinction (W olfenbarger and Phifer, 2000).
Whether or not the introgession of the transgene poses an ecological risk depends on if it
confers a selective fitness advantage (Hails, 2000). The fitness impact may be
detrimental, neutral, or positive and may depend on the environment in which the crop is
introduced (Hancock, 2003). For example, the introgession of an herbicide resistance
gene into a wild population is likely to confer a positive fitness advantage only in areas
where that herbicide is used for weed control, but in areas where the herbicide is not
routinely applied, the gene is likely to be neutral and, therefore, have no impact on
fitness. When the transgene alters the environmental tolerance or patterns of plant
gowth and development (which is often the goal of modifying a sigraling pathway),
sigrificant shifts in adaptability could result, having major impacts on plant fitness
(Hancock, 2003). Assessments of the risks associated with transgenes may become even
more complex if the trait of interest confers a selective advantage while the secondary
effects have a detrimental fitness effect or vise versa.

For this discussion of genetically engineered plants, the concept of non-target
effects encompasses a range of potential ecological impacts that can be a direct or

indirect result of the transgenic plant’s introduction. Possible effects include, but are not

103

limited to: toxicity to non-target organisms, shift in the food supplies available to
organisms, and impact on natural habitats (Poppy, 2000; Dale of al., 2002). It is
important to note that these risks can be quite variable, difficult to predict, and must take
into account the locale of plant introduction on a case by case basis (Snow and Palma,
1997). For example, the direct effects of ingestion of pollen from Bt plants by the non-
target monarch butterfly in the United States has been intensely studied with field and
laboratory evaluations giving contrasting results (Barton and Dracup, 2000; Poppy,
2000). These studies have demonstrated the importance of conducting realistic risk
assessment analysis to determine the likelihood of non-target species exposure and

susceptibility (Poppy, 2000).

How can we begin to assess these secondary effects that may influence gene flow,
invasiveness, or non-target impacts?

As research in plant molecular biology progesses at a rapid rate, information
regarding the complexity of sigraling pathways is becoming available that may help to
anticipate how genetically engineered modification will affect the plant. One example is
the research that has expanded the understanding of the molecular biology of the plant
hormone ethylene. Ethylene has a range of diverse roles throughout plant gowth and
development. These roles include but are not limited to seed germination, root and shoot
gowth, xylem differentiation, flower development, senescence, stress response, and fruit
ripening (as reviewed by: Arshad and Frankenberger, 2002; Abeles et al., 1992; Matteo
and Suttle, 1991). Because of its diverse array of functions there has been interest in the

manipulation of ethylene production or action in order to modify plant phenotypes.

104

However, altering ethylene sigraling for a characteristic such as increased flower
longevity may in turn result in secondary changes in seed dormancy, stress response, or
disease resistance. Therefore, the goal of this paper is to conduct a case study focusing
on ethylene as a sigraling molecule to examine the complexity of its sigral transduction
and interconnectedness with other pathways within the context of pleiotropic effects that
may have ecological sigrificance. Ethylene is a good candidate for this case study as
there is an extensive body of literature examining the effects of ethylene through mutant
or transgenic approaches. Pleiotropic effects on plant gowth and development, which
have been observed as secondary to the initial desired goal of genetic modification, also
will be reviewed. In addition, how these secondary effects may be relevant to risk

assessment of plants engineered to modify signaling pathways will be discussed.

Why modify ethylene?

Modification of ethylene to manipulate plant gowth and development is not a
new concept. Ethylene-releasing compounds as well as inhibitors of ethylene synthesis
and perception are tools that have been used broadly in agiculture to control multiple
facets of crop production. Ethephon (2-chloroethylphosphonic acid) is an ethylene-
releasing compound that has a range of applications including the treatment of wheat and
barley to retard gowth and prevent lodging, initiate uniform flowering in pineapple,
induce female sex expression in cucurbits, increase bud hardiness and bloom delay in
sweet cherry and plum, reduce curing time in tobacco, promote fruit abscission in cotton,
and promotedefoliation in tree species (reviewed in: Arshad and Frankenberger, 2002;

Abeles et al., 1992). Alternatively, ethylene production can be inhibited through

105

treatment with aminoethyloxyvinyl glycine (AVG) or arninooxyacetic acid (AOA)
(Abeles et al., 1992), and inhibition of ethylene response at the level of perception is also
possible through chemical treatment with compounds such as silver thiosulfate, 2,5-
norborrridiene, and l-methylcyclopropene (l-MCP) (Abeles et al., 1992; Sisler and Serek,
1997). Applications for these compounds include the delay of fruit ripening, the
prevention of flower senescence or fruit abscission, and increasing the shelf life of leafy
vegetables (Arshad and Frankenberger, 2002.).

Thus, there are many reasons for modifying ethylene, and it is not difficult to see
why the possibility of genetically engineering plants with altered ethylene would be

desirable.

What are the genetic components of ethylene biosynthesis and signaling?

Much effort has been devoted to elucidation of the genetics and molecular
mechanisms underlying ethylene biosynthesis and sigraling and has led to the cloning of
key gene components in multiple species (Kende, 1993; Zarembinski and Theologis,
1994). Methionine is the amino acid precursor to ethylene. Methionine is converted to
S-adenosyl-L-methionine (S-AdoMet), mediated by methionine adenosyltransferase

(Adams and Yang, 1979; Kende, 1993; Bleeker and Kende, 2000).

 

 

 

Met 4. e ACC s
Met SAM ACC ACC C’H“
adenosyl synthase oxidase
transferase

The synthesis of 1-aminocyclopropane-l-carboxylic acid (ACC) from S-Ado-Met is
catalyzed by ACC synthase (ACS) (Adams and Yang, 1979; Kende, 1993; Bleeker and

Kende, 2000). 5’-Methyl-thioadenosine is also produced, which is used to synthesize a

106

new methionine (Kende, 1993). The reaction of ACC with oxygen, catalyzed by ACC
oxidase (ACO), produces ethylene. ACS is generally regarded to be rate-limiting in
ethylene biosynthesis and is encoded by a multigene family in which genes are
differentially regulated by developmental, environmental, and hormonal inducers (Kende,
1993; Bleecker and Kende, 2000; Zarembinski and Theologis, 1994; Johnson and Ecker,
1998)

The path of ethylene perception and sigraling also has been well characterized,
primarily in Arabidopsis, through the evaluation of mutants displaying either an ethylene
insensitive or a constitutive response phenotype (Wang et al., 2002). A family of
receptors (in Arabidopsis: ETR] , ETR2, ERSI, ERS2, and EIN4) homologous to
bacterial two-component histidine kinases are responsible for binding ethylene (Johnson
and Ecker, 1998; Bleeker and Kende, 2000). Mutations that eliminate ethylene binding
of at least one receptor result in a dominant, ethylene-insensitive phenotype, indicating
that unbound receptors function to negatively regulate downstream ethylene responses
and are, therefore, inactivated by binding (Zarembinski and Theologis, 1994; Johnson
and Ecker, 1998; Bleeker and Kende, 2000). The signaling component directly
downstream of the receptors is CTR] , which was isolated as a loss-of—function mutant
displaying a constitutive ethylene response and is a member of the Raf family Ser/Thr
protein kinases. Since loss-of-function mutations in CTRl result in the plant responding
as if it were exposed to ethylene even in ethylene’s absence, CTRl functions as a
negative regulator of downstream signaling events (Bleeker and Kende, 2000; Wang et
al., 2002). Further downstream is EIN2, which upon knock out, results in a loss of

ethylene responsiveness, suggesting that it functions as a positive regulator of ethylene

107

sigraling (Johnson and Ecker, 1998; Wang et al., 2002). Transcription factors involved
in changes in the expression of ethylene response genes lie downstream of EIN2 and

include the EIN3 family and ERF 1 (Wang et al., 2002; Bleeker and Kende, 2000).

How are these genes used to modify ethylene?

The availability of cloned genes involved in endogenous ethylene production and
sigraling have created opportunities for evaluating the potential of genetic engineering to
modify ethylene-controlled processes in plants. Although the primary goals of modifying
endogenous ethylene have been for post-harvest characteristics such as delayed ripening
and flower senescence in an effort to extend shelf-life and reduce spoilage (reviewed in:
Yang and Oetiker, 1998; Bleeker and Kende, 2000), increasing ethylene biosynthesis in
order to prevent lodging in barley, increase latex flow in rubber tress, increase pistillate
flower production in melon, and promote leaf abscission prior to tree harvest have been
suggested as possible goals as well (Hedden and Phillips, 2000; Papadopoulou et al.,
2005).

Transgenic plants evaluated for characteristics involving reduced ethylene
production or activity have included apple, melon, tomato, broccoli, carnation, and
petunia (reviewed by Stearns and Glick, 2003). Approaches to attain this end have
included down-regulation of genes encoding ethylene biosynthetic enzymes ACS and
ACO (via antisense or cosuppression); introduction of genes encoding enzymes to reduce
ethylene production, ACC degading enzyme ACC dearninase, SAM inactivating SAM
hydrolase; and introduction of mutant ethylene receptors (Yang and Oetiker, 1998;

Dandekar et al., 2002; Steams and Glick, 2003). For example, ethylene induced ripening

108

and senescence was inhibited in fi'uits of tomato plants engineered with an antisense copy
of an endogenous ACS gene which reduced fruit ethylene levels by 95.5%; however,
treatment with exogenous ethylene could reverse the inhibition and lead to normal
ripening (Oeller et al., 1991). Similarly, but through the heterologous expression of the
Arabidopsis etrI-I mutant ethylene receptor, senescence of carnations flowers was
delayed three-fold relative to controls (Bovy et al., 1999). A review of many of the plants
and genes used to transgenically modify ethylene biosynthesis or perception has recently
been compiled (Steams and Glick, 2003).

Four plants engineered for modified ethylene have already been deregulated, and
therefore approved for environmental release. A carnation has been approved for
environmental release in Australia and the European Union, which has been transformed
for reduced ethylene content to provide a longer life to cut flowers through the
introduction of a truncated ACS gene (htpp://www.agbios.com). Similarly, a tomato
engineered for delayed ripening by introducing a truncated ACS gene has been approved
in the United States. Two other tomato lines genetically engineered for delayed ripening
have been approved for release in the United States. In these cases, ethylene biosynthesis
is thwarted through the introduction of a SAM hydrolase or ACC deaminase gene
(htpp://www.agbios.com).

While these plant materials have undergone extensive field testing and have been
shown to be safe for environmental release, one of the ways to maintain safe
environmental release of these and new plant introductions is to have as much knowledge
about the introduced transgene as possible. As current research continually demonstrates

the complexity of ethylene’s role as a signaling molecule, highlighting some of the recent

109

findings on ethylene’s role in plant gowth and development may help to stimulate
thoughts regarding what to look for in terms of secondary effects when assessing the risk

of genetically engineered plants.

What phenotypic effects are associated with modified ethylene signaling?

Plant growth and development

Transgenic and mutant lines have implicated ethylene to play a role in virtually all
stages of plant gowth and development, e. g. germination, vegetative gowth,
reproduction development, and ripening and senescence, (Smalle and Van Der Straeten,
1997). Evidence for a role of endogenous ethylene perception in germination has been
established in evaluations of ethylene insensitive etrI mutant Arabidopsis and transgenic
etr1 petunia showing decreased seed germination (Bleeker et al., 1988; Clevenger et al.,
2004). However, there appears to be considerable variation in the requirement of
ethylene for germination among species (Smalle and Van Der Straeten, 1997).

The role of ethylene on further plant gowth and establishment has also been
studied. In fact, the characteristic ‘triple response’ phenotype of dark gown seedlings
under continuous exposure to ethylene, which includes thick and short roots and
hypocotyls, and an exaggerated apical hook, provided a key bioassay for screening
mutant populations in the initial efforts to identify primary genes in ethylene sigraling
(Stepanova and Ecker, 2000). While results have been mixed, several studies also
suggest that ethylene affects leaf expansion by suppressing cell enlargement (Kieber et
al., 1993; Hua et al., 1995; Grbic and Bleeker, 1995; Tholen et al., 2004).

Characterization of gowth of ethylene insensitive Arabidopsis mutants based on rosette

110

leaf area has shown an increase in gowth relative to wild type plants (Grbic and
Bleecker, 1995) while the constitutive ethylene response mutant ctrI exhibits reduction in
size (Kieber et al., 1993). However, further study on gowth in Arabidopsis, tobacco, and
petunia showed no difference in leaf area between ethylene-sensitive and —insensitive
plants (Tholen et al., 2004).

The effects of ethylene on root formation have also been studied. The
development of adventitious roots was geatly inhibited in vegetative cuttings of
transgenic ethylene-insensitive tomato, petunia, and melon (Clark et al., 1999;
Papadopoulou et al., 2002). Secondary effects associated with adventitious rooting may
become particularly relevant in plant species which depend on vegetative propagation. In
addition, reduced root hair length in tomato and Arabidopsis and impairment in root
penetration in less than optimal soil medium, such as sand, in tomato was reported for
ethylene-insensitive genotypes (Clark et al., 1999). The presence of longer and more root
hairs in the Arabidopsis ctr] provide further support for ethylene as a regulator of root
hair development (Tanimoto et al., 1995). Such effects on rooting may affect the plant’s
ability to become established and thrive in certain soils or environmental conditions.

A function of ethylene in reproductive development has been demonstrated in
several species. Ethylene is considered the primary hormone influencing female floral
development in cucumber, melon, and squash, with cucumber being the most studied.
The strong correlation between ethylene and female flower deve10pment in the
moneocious species cucumber is now supported with molecular evidence including the
cosegetion of an ACS gene with a locus used for breeding gynoecious (female)

cucumber as well as correlation of ACS mRNA with female flowering (Trebitsh et al.,

111

1997; Kamachi et al., 1997; Kamachi etal., 2000; Yamasaki et al., 2003; Mibus and
Tatlioglu, 2004). In addition, increasing endogenous ethylene production via
heterologous expression of an ACS transgene in andromonoecious melon conferred
earlier and increased production of hermaphrodite flowers (Papadopoulou et al., 2005)
while ethylene insensitivity as a result of heterologous expression of ethylene receptor
mutant, etrI -1, modified sex expression, with effects dependent on the location of gene
expression (Chapter 2). In another study, transgenic tobacco expressing an antisense
ACO under the control of a pistil-specific promoter demonstrated the importance of
ethylene for ovule development in tobacco. Transgenic plants with reduced ethylene
production displayed female sterility as a result of arrested ovule development; however,
the phenotype could be overcome with treatment with ethylene (DeMartinis and Mariani,
1999). In addition, studies suggest that ethylene production that is triggered upon
pollination in orchid flowers initiates sigraling events important for ovary development,
and several ACS genes that respond to pollination sigrals in the orchid flower have been
identified (Bui and O’Neill, 1998).

Based on these studies, reduced ethylene production or sensitivity is likely to be
detrimental to a plant’s reproductive capacity, and from an ecological risk assessment
perspective would not result in increased fitness. However, the observation of earlier and
increased numbers of hermaphrodite buds reaching anthesis in melon overexpressing
ACS (Papadopoulou et al., 2005), may be relevant to ecological risk assessment.
Evaluation of these melon plants in a field setting also showed earlier fruit maturity and

modified patterns of fruit set on the main stem. Such effects may warrant close

112

evaluation as a reproductive characteristic is being altered in a way that might confer a
fitness advantage in the event of gene flow.

As rrright be assumed by the types of traits in the deregulated transgenic plants
described above, a primary function of horticultural importance is the role of ethylene in
senescence and ripening. Floral and leaf senescence is a complex biochemical process of
catabolism, mobilization of metabolites, and physical disintegation leading to
progammed cell death that is under the control of many hormonal factors and
developmental cues (Abeles et al., 1992, Bovy et al., 1999). An increase in sensitivity
and autocatalytic production of ethylene in an aging tissue typically precedes the onset of
senescence, and ethylene is regarded as the strongest promoter of senescence among the
plant hormones (Bovy et al., 1999). Ripening is characterized by changes in biochemical
and physical attributes that take place at fi'uit maturity including changes in color, texture,
flavor, aroma, and nutritional content in preparation to attract seed-dispersing organisms
(Gray et al., 1994; Adams-Philips et al., 2004; Giovannoni, 2004). In climacteric fruit
such as tomato, banana, apple, and stone fi'uit, ripening is initiated by a sharp increase in
respiration that is accompanied by a burst in ethylene production (Giovannoni, 2004).
The role of ethylene in fruit ripening is exemplified by isolation of the Never ripe (Nr)
locus of the tomato mutant, which was characterized by a non-ripening phenotype.
Cloning of theNR gene established its identity as an ETRl ethylene receptor homologue
impaired in the ability to bind ethylene (Wilkinson et al., 1995). Since positive feedback
regulation of ethylene biosynthesis is an important component of senescence and ripening
progams, controlling production and perception of ethylene has been an area of

economic and agicultural interest.

113

Plant-microbe interactions

Symbiotic associations between the plant species, such as legumes, clover and
alfalfa, and soil borne nitrogen-fixing microbes enable the plant to access nitrogen in a
useable form (Crawford et al., 2000). Root nodules are the sites of plant-rhizobia
symbiosis, and studies suggest that ethylene is involved in nodule formation (Wang et al.,
2002). Studies of hyper-nodulation mutant, sickle (skl), plants of Medicago truncatula
(alfalfa) show that skl plants also have ethylene insensitive phenotypes, including delayed
senescence, decreased abscission, and insensitivity of seedling to ACC treatment
(Penmetsa and Cook, 1997). Nodule formation has also been studied in legume, Lotus
japonicus, engineered to express a mutant ethylene receptor conferring ethylene
insensitivity (Nukui et al., 2004). Upon inoculation with Mesorhizobium loti, transgenic
plants demonstrated enhanced formation of infection threads and nodule promordia
(Nukui et al., 2004). These studies suggest that ethylene is involved in the regulation of
rhizobial infection to balance symbiosis and control the number of modulation events
(Penmetsa and Cook, 1997; Nukui et al., 2004). The implications of effects on these
symbiotic interactions may be of interest from a risk assessment perspective on multiple
levels. Enhancement of interactions, which allows the plant to develop more vigorously,
may contribute to increased fitness and invasiveness in events of gene flow. In addition,
nontarget effects on microbe populations may result from changes in plant-rhizobia

associations.

114

Disease response

While accelerated ethylene production is a typical response to a pathogen attack
by fungi, bacteria, viruses, and nematodes (Arshad and Frankenberger, 2002), its role in
disease sigraling and, subsequent, susceptibility or resistance has been more difficult to
elucidate (Hoffman et al., 1999; Thomma et al., 1999; Diaz etal., 2002). Research to
decipher the role of ethylene in disease response has incorporated the use of plants with
altered ethylene sensitivity and, depending on the type of pathogen and plant species,
reduced sensitivity has resulted in increased susceptibility or increased resistance (Wang
et al., 2002). For example, study of mutant soybean lines with reduced ethylene
sensitivity showed that the mutants had reduced or no difference in disease severity in
response to virulent bacterial Psuedomonas syringae and fungal Phytophthora sojae
compared to wild type lines (Hoffman et al., 1999). In contrast, observation of more
severe symptoms in response to fungal Septoria glycines and Rhizoctonia solani seemed
to suggest that ethylene-insensitivity increased disease susceptibility to other pathogens
in soybean (Hoffman et al., 1999).

Results showing contrasting effects of ethylene involvement in disease response
depending on the pathogen type have also been reported in studies of ethylene-insensitive
ein2-1 mutants of Arabidopsis. Mutants showed enhanced susceptibility to strains of the
gay mold fungus Botrytis cinerea; however, the mutation caused no change in the plant’s
response to other firngal pathogens, Peronospora parasitica or Alternaria brassicola
(Thomma et al., 1999). In yet another study, evaluation of ethylene-insensitive
transgenic etrI tobacco and mutant ein2-I Arabidopsis demonstrated increased disease

susceptibility to soilbome pathogens, including several Pythium strains, which had been

115

previously isolated fiom diseased ethylene-insensitive plants gown in nonautoclaved
potting soil (Geraats et al., 2002). Further investigation of the altered disease resistance
of ethylene-insensitive tobacco and Arabidopsis included a range of root and leaf
pathogens and concluded that the role of ethylene sigraling in disease development is
dependent on the type of host-pathogen interaction rather than site of infection or
pathogen species (Geraats et al., 2003).

Despite a lack of clarity in the role of ethylene in response to infection, studies
have demonstrated ethylene’s function as an effector molecule in the signaling process of
disease response and resistance. For example, study of mutant einZ-I Arabidopsis
(Thomma et al., 1999; Penninckx etal., 1996) and transgenic etrI tobacco (Knoester et
al., 1998) have demonstrated that these ethylene insensitive lines lack pathogen-inducible
expression of a subset of PR (pathogen-related) genes, indicating dependence on ethylene
for some host responses. In contrast to studies of ethylene insensitivity, Ciardi et a1.
(2001) evaluated antisense LeE T R4 tomato plants, which exhibited a phenotype of
increased sensitivity to ethylene. Upon infection with Xanthomonas campestris,
transgenic plants gave a more rapid disease response and exhibited accelerated necrosis;
however, pathogen levels decreased more quickly and to a lower level in transgenic
plants (Ciardi et al., 2001). In addition, transgenic plants showed stronger and more rapid
increases in ethylene synthesis and PR gene expression following inoculation, providing
support for ethylene’s role in regulating the defense response to X. campestris in tomato.

Clearly, the function of ethylene in plant defense against disease is complex, and
studies suggest that manipulating ethylene production or perception may increase

tolerance to some diseases while increasing susceptibility to others. From a risk

116

assessment point of view, an evaluation of how an engineered trait modifying ethylene
may affect plant fitness, with regard to disease resistance, will likely be variable, and the
effect might be detrimental, positive, or neutral. Secondary effects on disease resistance
will be dependent on the pathogen pressures on a specific species in specific locations
and should, therefore, be assessed on a case by case basis. For example, hypothetically,
if a transgenic plant having a secondary effect of increased tolerance to Phytophora is
introduced into a region where disease pressure by Phytophora controls populations of
related species of the introduced plant, hybridization between the two species may result
in a fitness advantage and lead to increased invasiveness in the natural populations.
Conversely, if an engineered plant that has the secondary effect of increased
susceptibility to Phytophora is introduced into the same region, hybrids may have
reduced fitness and be, subsequently, rogued out of natural populations, as the disease

pressure would inhibit their establishment.

Abiotic stress

There are many examples of increased ethylene production following an array of
abiotic disturbances including temperature extremes, drought, ozone, hypoxia, wounding,
and crowding (Abeles et al., 1992). As has been shown in response to biotic stress,
understanding of the function of such increases is not complete. In many cases, the
stimulation of ethylene biosynthesis in response to environmental stresses is reported to
involve reactive oxygen species (ROS), which act as preliminary sigraling molecules
(Wang et al., 2002). In addition, studies to evaluate the mechanisms underlying plant

responses to abiotic stressors suggest similarities to pathogen response pathways.

117

In a study of heat-induced oxidative stress, the Arabidopsis ethylene-insensitive
etr1 mutant showed increased damage and reduced survival following heat treatment,
while pretreatment of wild type plants with ACC reduced damage and enhanced survival
(Larkindale and Knight, 2002). This study provides support for a role of ethylene in the
protection against or repair of heat-induced oxidation stress, in Arabidopsis. However,
somewhat conflicting results were reported in studies of transgenic tobacco and potato
with suppressed expression of a ozone(O3)-inducible ACS (N akajima et a1. 2002; Sinn et
al., 2004). Transgenic lines showed increased ozone tolerance, and the extent of 0;;-
induced damage was positively correlated with the level of 03-induced ethylene
production, suggesting that ethylene increases sensitivity to ozone.

The role of ethylene in the acclimation of wetland and flood-tolerant species to
the oxygen-deprived status of submergence has been well studied. As observed in
species, such as maize and rice, ethylene production promotes the formation of
aerenchyma in root tissue as well as adventitious roots, which allows transport of oxygen
to the hypoxic roots to facilitate root gowth and minimize damage resulting fi'om
hypoxic conditions (Bray et al., 2000). Such a relationship between ethylene and the
response to flooding was exemplified in a study which reported better performance under
waterlogged conditions of wild type tobacco over transgenic ethylene-insensitive (etrI-I)
tobacco, as measured by adventitious root formation (MacDonald and Visser, 2003). In
addition, ethylene production has been correlated with the activities of hydrolytic
enzymes and enhanced formation of aerenchyma and adventitious roots in response to

root submersion of maize seedlings (Bragina et al., 2001, 2003).

118

Ethylene also has been shown to be involved in the rapid elongation response of
deepwater rice to submergence (Kende et al., 1998) as well as the hyponastic leaf gowth
in semiaquatic species such as Rumex palustis (Cox et al., 2004). Studies to characterize
the rapid increase in ethylene biosynthesis in deepwater rice triggered by flooding have
identified that the induction and sustained expression levels of a particular ACS gene
family member is correlated with the gowth response (Van der Straeten et al., 2001;
Zhou et al., 2001; Zhou et al., 2002). The role of ethylene in the induction of petiole
elongation in R. palustris is supported by molecular studies showing up-regulation of
ACS and AC 0 genes as well as increases in their enzymatic activities concomitant with
submergence (V oesenek et al., 2002).

In contrast, in flood-sensitive species, such as tomato, ethylene induced upon
flooding is associated with deleterious consequences, including epinasty, leaf chlorosis,
necrosis, and reduced fruit yield, often referred to as ‘ethylene syndrome’ (Grichko and
Glick, 20013). An evaluation of flood tolerance in transgenic tomato expressing a root-
specific ACC deaminase gene to confer reduced ethylene production demonstrated
increased tolerance as measured by shoot gowth, leaf chlorophyll content, and epinastic
curvature, further substantiating the adverse effects of stress ethylene on sensitive
species.

In addition, recent work has studied the ability of root-associated bacteria with
ACC deaminase (ACD) activity to confer resistance to otherwise sensitive crop plants to
the effects of flooding, salt, and drought stress by reducing levels of ACC available for
oxidation to ethylene within the plant, therefore alleviating effects associated with

ethylene syndrome (Grichkco and Glick, 2001b; Mayak et al., 2004a; Mayak eta1.,

119

2004b). The soil bacterium, Achromebacter piechaudii, was evaluated for its effect on
tomato and pepper under drought stress (Mayak et al., 2004a) and tomato under salt stress
(Mayak et al., 2004b), and results, based on measurements of fresh and dry weight as
well as recovery, indicated that treatment with the bacteria allowed continued gowth
throughout the duration of the stress and a faster recovery upon removal of the stress. A
related study of the changes in genes expressed in canola roots treated with rhizobacteria,
Enterbacter cloacae, which also has ACC deaminase (ACD) activity, in comparison to
plants treated with an ACD knockout mutant, suggests that the effect of the bacteria on
the plants is primarily the result of lowering ethylene levels (Hontzeas et al., 2004). A
down-regulation of stress genes and up-regulation of genes involved in cell division and
proliferation was observed in roots treated bacteria with ACD activity.

Research also has demonstrated an ethylene-mediated decrease in leaf expansion
and shoot gowth in response to soil compaction by evaluating transgenic tomato
harboring an antisense ACO (Hussain et al., 1999). Reduced ethylene producing tomato
plants were able to maintain uncompacted control gowth rates while wild type plant
exhibited geatly reduced shoot gowth (Hussain etal., 1999). Recent studies have also
evaluated the shade avoidance response, which includes enhanced shoot elongation and
elevated leaf angles enabling a plant to optimize light capture in crowded gowth
conditions, using transgenic ethylene-insensitive tobacco (Tetr) (Pierek et al., 2003).
When gown within a crowded canopy along with wild type plants, Tetr plant gowth was
severely suppressed as a result of delayed shade avoidance response, suggesting that
ethylene is required for plants to optimally respond to neighbors and successfully

compete for light (Pierek et al., 2003; Pierek et al., 2004). Collectively, these studies

120

demonstrate that the effects of ethylene on gowth can be variable and are dependent on
environmental conditions.

While an increase in ethylene production concurrent with abiotic stress cannot be
argued, a dichotomy exists in the nature of the downstream responses triggered by such
an increase and appears to be dependent on properties of individual species. As will be
addressed in the next section, much of the difficulty in assessing the firnction of ethylene
in stress response is related to its interactions with other hormones in synergistic and
antagonistic ways. Nonetheless, efforts characterizing the acclimation of tolerant species
to particular abiotic stresses, such as flooding, suggest that ethylene is involved in the
sigraling of particular mechanisms that have evolved to allow species to cope with
specific stresses (Peeters et al., 2002). With regard to risk assessment, as was discussed
with disease response, the impact of modified ethylene on a plant’s ability to respond to
abiotic stress is likely to be variable and evaluation should, therefore, be considered on a
case-by-case basis. Difficulty arises in the lack of resolution of the signal transduction
pathways accounting for the mechanisms of adaption, resulting in the inability to assess

with certainty how gene flow may impact the invasiveness of a related species.

Cross talk with other hormones

The study of hormone-response mutants has repeatedly demonstrated that the path
of hormone perception or response is not a distinct linear one. Instead hormone sigraling
pathways intersect at many levels to create a complex network of overlapping sigrals

~ (Gazzarrini and McCourt, 2003). Research has demonstrated interaction of ethylene

121

synthesis, signaling and response with that of auxin, abscissic acid (ABA), cytokinin
(CK), gibberellin (GA), jasmonic acid (JA), and salicylic acid (SA).

Cross talk between ethylene and auxin is apparent in the regulation of ethylene
biosynthesis and response. While low concentrations of auxin reduce ethylene formation
and promote gowth, high concentrations induce ethylene production and ethylene-
associated responses (Grossman and Hansen, 2001). Characterization of regulatory
regions of A CS genes has resulted in the identification of auxin response elements
(ARES) responsible for the upregulation of ethylene synthesis by auxin (Ishiki etal.,
2000). Induction of ACO gene expression and post-translation modification in response
to auxin treatment provides further support and additional mechanisms for modulation of
auxin-induced ethylene biosynthesis (Chae et al., 2000). Interaction between auxin and
ethylene signaling pathways has become apparent through an evaluation of ethylene-
insensitive mutant ein2 and an auxin influx mutant, which showed that responsiveness to
ethylene is important for auxin-induced root hair elongation (Rahman et al., 2002). In
addition, studies of R. palustris suggest an interaction between ethylene and auxin sigrals
in flood-induced adventitious root formation (V oesenek et al., 2003).

Ethylene and ABA interactions have been shown to affect germination, rooting,
and shoot gowth (Gazzarrini and McCourt, 2003; LeNoble et al., 2004; Hussain et al.,
2000). Characterization of the shoot gowth in Arabidopsis ABA-deficient mutant, aba2-
1, showed that the mutant had increased ethylene production and sensitivity, and
demonstrated that ABA plays a role in the maintenance of shoot gowth by suppression
of ethylene synthesis (and, perhaps, sensitivity) (LeNoble et al., 2004). An antagonistic

interaction between ethylene and ABA has also been observed in studies of germination

122

as ethylene-response mutants, such as Arabidopsis etrI -1 , increase the sensitivity of the
seed to ABA (Gazzarini and McCourt, 2003). In contrast, ethylene-response mutants
have shown reduced sensitivity to ABA in root tissue, demonstrating an additive
interaction between the two hormones with respect to root gowth (Beaudoin et al.,
2000).

Interactions between ethylene and cytokinin (CK) pathways have been identified
through study of factors affecting ethylene production, mutants, and transgenics. Studies
of mutants overproducing ethylene have provided support for the post-transcriptional
regulation of ACC synthase levels by cytokinin (Woeste et al., 1999). In addition, altered
development including smaller leaves with reduced color and modified shape in
transgenic Arabidopsis CK overproducing lines was partially attributed to increased
ethylene levels (van der Graaff et al., 2001). However, an antagonistic relationship
between ethylene and CK during senescence is suggested by the phenotype of petunia
transformed to produce increased levels of CK at senescence (Chang et al., 2003). In
these transgenics, higher levels of CK at flower maturity, resulted in lower levels of
ethylene production, reduced ethylene sensitivity, and delayed senescence.

Communication between sigraling pathways of ethylene and gibberellin have
been demonstrated in the release of seed dormancy as well as gowth processes including
shade avoidance and response to flooding (Kende et al., 1998; Brady and McCourt, 2003;
Cox et al., 2004; Rao et al., 2004). While the molecular mechanism of the interplay
between ethylene and GA in the breaking of seed dormancy is not completely
understood, the ability of ethylene to rescue the Arabidopsis GA-deficient mutant go] at

germination suggests that there is cross talk between their sigraling pathways at this

123

stage of development (Brady and McCourt, 2003). Similarly, the ability of gibberellin to
enhance the germination of etrI Arabidopsis suggests a reciprocal relationship between
GA and ethylene (Bleeker et al., 198 8). In addition, it was recently shown in tobacco that
the gowth-promoting effects of ethylene involved in phytochrome-mediated shade
avoidance response are dependent on the action of GA (Pierik et al., 2004). An
analogous dependence on GA in the mediation of an ethylene-induced gowth response
has been demonstrated in the studies of semiaquatic plants such as R. palustris and
deepwater rice. Studies suggest that upon flooding ethylene mediates the rapid gowth
response in semiaquatic plants by increasing the plant’s sensitivity to GA and, also, by
increasing the ratio of GA to ABA by reducing ABA levels (Kende et al., 1998).

Cross talk between ethylene and jasmonic acid is a key component in the
regulation of various plant processes including response to pathogen attack, wounding,
stress and apical hook formation (Lorenzo et al., 2003). The negative interaction that
exists between ethylene and jasmonic acid in a plant’s response to the stress of ozone has
been characterized through the study of J A- and ethylene-insensitive mutants of
Arabidopsis. J A and ethylene appear to modulate each other’s activity in ozone-induced
cell death, whereby ethylene stimulates cell death while J A protects tissues (Tuominen et
al., 2004). An analogous antagonistic relationship is apparent in the formation of an
apical hook by emerging seedlings, as has been shown in studies in which the typically
exaggerated apical hook of the Arabidopsis constitutive ethylene response mutant (ctrI)
of Arabidopsis can prevented by jasmonate treatment (Ellis and Turner, 2001).

A plant’s defense response to pathogen attack involves a complex network of

sigraling that includes salicylic acid (SA), jasmonic acid (J A), and ethylene. The

124

sigraling pathways of ethylene and jasmonic acid interact synergistically to respond to
some pathogens, and recent study of Arabidopsis J A and ethylene response mutants
demonstrated that sensitivity to both is required for the expression of transcription factor
ERF 1 , which in turn is responsible for the activation of defense-related genes (Lorenzo et
al., 2003). In addition, characterization of Arabidopsis mutant cevI revealed its
constitutively active JA and ethylene sigraling pathways as well as enhanced resistance
to pathogens, providing further support for cooperative regulation of JA/ethylene signal
pathways (Ellis and Turner, 2001).

Although contradictory, research has provided evidence for cross-talk between
SA and ethylene (Kunkel and Brooks, 2002). In a study utilizing tomato lines that were
ethylene-insensitive, ethylene-deficient, or SA-deficient, ethylene was shown is be
required for the accumulation of SA in plants infected with Xanthomonas campestris
(O’Donnell et al., 2001). In addition, results of evaluation of the interaction of Tobacco
mosaic virus and tobacco in transgenic ethylene-insensitive lines demonstrates that
ethylene perception is necessary to initiate the sigraling cascade, which triggers SA
accumulation, pathogen-related gene expression, and systemic acquired resistance (SAR)
(V erbeme et al., 2003). However, in contrast, based on a series of experiments
investigating SAR in ethylene-insensitive Arabidopsis, SA-mediated responses leading to
the development of acquired resistance were found to not require ethylene sensitivity and,
therefore, suggested to be independent of ethylene (Lawton et al., 1994). Another level
of interaction between ethylene and SA signaling pathways during disease response has
been identified by studying the expression of ERFs (ethylene response factors) from

tomato in Arabidopsis. This work suggested that the depending on the type of pathogen

125

interaction, the ERF transcription factor, Ptil, may be positively regulated either by J A
and ethylene or by SA (Gu et al., 2002). In addition, Gu et al. proposed that SA can
repress the activity of J A/ethylene-induced Pti4, giving rise to an addition level of
pathway interaction.

As research continues to reveal cross-talk between hormones, it is becoming
increasingly difficult to deduce independent roles of any one hormone. Interactions
between pathways can be antagonistic, synergistic, or cooperative, and may depend of the
plant species or the developmental or physiological status of the plant (Rojo et al., 2003;
Chow and McCourt, 2004). When considering the secondary effects associated with
modifying ethylene pathways, it is very likely that the sigraling pathways of additional
hormones could be altered in some way as a result of disruption of their interactions with
ethylene and may result in pleitropic effects, reflected developmentally or in responses to
environmental stimuli. Consideration of cross talk between hormone sigraling pathways
is relevant from a risk assessment perspective as phenotypic changes in the engineered
plant may be unexpected and subtle enough to not be identified in routine

characterization.

What pleiotropic and secondary effects have been observed?

This wide-reaching range of developmental, disease, and stress responses and
cross talk with other hormones suggests that alteration of ethylene for one purpose may
have secondary effects on other phenotypes. Observations of effects secondary to the
phenotype of interest sometimes are included in published work and may be relevant

from a risk assessment perspective. In addition, expression profiling of ethylene related

126

genes by microarray analysis firrther demonstrates the high degee of transcriptional
regulation that is affected by ethylene sigraling.

As already indicated, flower longevity and delayed ripening have been primary
targets of modified ethylene. Although not surprising from the preceding description of
ethylene-related phenomena, evaluation of transgenic petunia, tomato, melon, and potato
has revealed a range of unintended consequences of modified endogenous ethylene
production or perception including effects on seed germination, rooting, fertility, and
disease resistance, some of which were unexpected. For example, upon trying to
vegetatively propagate ethylene-insensitive lines of tomato, petunia, and melon through
cuttings, the requirement of ethylene for adventitious root formation became apparent
(Clark et al., 1999; Klee, 2002; Papodopoulou, 2002). In addition, difficulty of the roots
of ethylene-insensitive lines penetrating heavy soils has been observed (Klee, 2002).
Other unexpected effects noted in characterization of ethylene-insensitive petunia and
melons include the formation of larger flowers (Shaw et al., 2002), production of a
geater number of flowers (Klee, 2002), and the conversion of hermaphrodite to female
flowers (Chapter 2). An unexpected phenotype observed in melon engineered to
overexpress an ACS gene to increase femaleness was the earlier and increased production
of mature hermaphrodite flowers providing support for a previously unknown role of
ethylene in pistillate bud development (Papadopoulou et a1, 2005). In addition, earlier
and increased sequential fruit set on the main stem was also observed, suggesting that
ethylene sigraling may be associated with fruit set.

Clevenger et al. (2004) recently investigated factors affecting seed production in

transgenic ethylene-insensitive petunia, including pollen viability, pollen tube gowth,

127

 

seed yield, seed germination, and whole plant vigor. Ethylene-insensitive pollen showed
reduced viability; however, pollen tube gowth was not affected. In addition, seeds
produced on an ethylene-insensitive maternal parent showed reduced seed weight,
germination, and seedling vigor.

On a slightly different note, potato that was engineered to express the Arabidopsis
E T R] ethylene receptor gene in an effort to better understand ethylene’s role in tuber
dormancy gave rise to a range of abnormalities in gowth and development (Haines et al.,
2003). Preliminary phenotypic characterization of transforrnants revealed changes in leaf
size and shape, plant gowth habit, and tuber size and shape that were attributed to the
altered sensitivity to ethylene.

Additional pleitropic effects associated with modified ethylene that have been
reported relate to the plant’s response to biotic and abiotic stress. Increased disease
susceptibility has been observed in reduced ethylene-producing tomato as well as
ethylene-insensitive petunia (Klee, 2002; Shaw et al., 2002; Shibuya et al., 2004).
However, as already mentioned modified ethylene can also result in enhanced tolerance
to certain pathogens. Melon engineered for delayed ripening by inhibiting ethylene
biosynthesis (Ayub et al., 1996) was subsequently found to have increased resistance to
the abiotic stress of chilling (Ben-Amer et al., 1999).

Microarray analysis has provided a valuable tool for investigating gene activity by
allowing examination of the expression levels of thousands of genes simultaneously (Van
Zhong and Burns, 2003). Comparison of expression profiles in Arabidopsis of ethylene-
insensitive mutant etr] -] , constitutive response mutant ctr], ethylene-treated wild type,

and untreated wild type plants has provided insight into the extent of gene transcriptional

128

regulation that is affected by ethylene (Van Zhong and Burns, 2003). These studies
revealed that 7% of the approximately 6000 unique genes included in the microarray
were regulated by ethylene, which if extrapolated to the whole Arabidopsis genome,
would suggest the presence of ca. 1800 ethylene-responsive genes. A wide variety of
genes were found to be differentially regulated by ethylene, including those encoding its
own biosynthetic and signaling components, transcription factors, components of other
hormone pathways, primary metabolic proteins, and defense-related proteins, as well as
many of unknown function.

Additional microarray studies have provided information about genes regulated
by ethylene in response to environmental stresses (Schenk et al., 2000; Chen et al., 2002).
Results from both of these studies strongly support the presence of a complex sigraling
network of positive and negative interactions between multiple sigraling pathways to
achieve an appropriate defense response. The inclusion of hormone response or
biosynthetic mutants in evaluations provides evidence for the effects of ethylene on the
expression of genes involved in the plant’s response to pathogen attack (Chen et al.,
2002)

Despite a tremendous volume of literature supporting ethylene’s involvement in
virtually all facets of a plant’s life cycle, there has been limited report of secondary
effects associated with engineered ethylene pathways in publications describing plants
transformed for potential commercial applications. This may be a result of research and
publication focus on the intended goals of the genetically engineered product. Another

factor may be the subtleness of the pleitropic effects such that characterization of these

129

secondary phenotypes is not a part of the research which is presented in the published
results.

Nonetheless, the pleitropic effects associated with modified ethylene pathways
pose interesting possible secondary plant phenotypes, which may be relevant in the
context of risk assessment. For example, a change in the size or number of flowers on
the transgenic plant may impact pollinator behavior resulting in increased gene flow.
However, the difficulty arises in predicting how the transgene will impact fitness of
resulting hybrids or surrounding ecosystem in the event of gene flow into adjacent related
populations as effects will largely be dependent on the stresses imposed in that
environment. This calls for scrutiny in the assessment of plant phenotypes and careful

consideration of environmental effects.

Conclusions: implications for risk assessment

This case study of ethylene has provided an analysis of the dynamic nature of
signaling pathways throughout a plant’s life. As efforts in plant biotechnology continue
to expand to include genes encoding transcription factors and sigraling or metabolic
components, the likelihood of impacting the expression of many genes with the
transgene’s introduction is quite high. As has been reviewed in the case of ethylene,
genetic engineering these pathways has the potential to impact plant development, stress
response, and interactions with other pathways. In addition, advances in plant molecular
biology continue to reveal the complexity of plant sigraling.

As a result of the interconnectedness of regulatory, metabolic and sigraling

pathways, modification of individual components may have pleiotropic effects, some of

130

which are unexpected and unpredicted. In addition, phenotypes secondary to the trait of
interest which may be relevant from an ecological risk perspective, such as changes in
attractiveness to pollinators or changes in pollen morphology, amount, or viability, may
go unnoticed in laboratory, geenhouse, and field trials (Hokanson, 2002). Therefore,
such effects could go unrecogrized until large scale commercial production. Due to the
difficulty in predicting and identifying some secondary effects, one recommendation for
the genetic engineering of sigraling pathways is the use of tissue-targeted or temporally
regulated promoters rather than constitutive promoters for transgene expression. This
strategy could confer localization or timing of transgene expression to meet the specific
requirements of the trait of interest in an effort to minimize pleiotropic effects of the
expressed transgene. Another recommendation is the establishment and enforcement of a
set of guidelines to collect baseline data for many of the plant life cycle characteristics of
both transgenic and non-transgenic plants at various stages of crop development from
laboratory to large scale field evaluation. The implementation of such data into a
publically accessible database may help to identify common secondary phenomenon
associated certain engineered traits or transgene types and assist in the development of
appropriate confinement and risk assessment protocols. In conclusion, the preceding
information suggests that careful and thorough characterization of the effects of
transgenes on crop biology is imperative for adequate assessment of the ecological

impact of genetically engineered plants destined for commercial production.

131

References

Abeles, F.B., Morgan, P.W., and Saltveit, M.E. Jr. 1992. Ethylene in Plant Biology, 2nd
ed. Academic Press, San Diego, CA.

Adam-Phillips, L., Barry, C., and Giovannoni, J. 2004. Sigral transduction systems
regulating fruit ripening. Trends in Plant Science 92331-338.

Andow, DA. and Hilbeck, A. 2004. Science-based risk assessment for nontarget effects
of transgenic crops. Bioscience 54: 63 7-649.

Arshad, M. and Frankenberger, W.T. 2002. Ethylene Agicultural Sources and
Applications. Kluwer Academic/Plenum Publishers, New York.

Ayub, R., Guis, M., Ben-Amor, M., Gillot, L., Roustan, J .P., Latche, A., Bouzayen, M.,
and Pech, J .C. 1996. Expression of ACC oxidase antisense gene inhibits ripening
of cantaloupe melon fi'uits. Nature Biotechnology 14: 826-866.

Barton, J .E. and Dracup, M. 2000. Genetically modified crops and the environment.
Agon. J. 92: 797-803.

Beaudoin, N., Serizet, C., Gosti, F ., and Graudat, J. 2000. Interaction between abscisic
acid and ethylene sigraling cascades. Plant Cell 12: 1103-1115.

Ben-Amor, M., Flores, B., Latche, A., Bouzayen, F ., Pech, J .C., and Romojaro, F. 1999.
Inhibition of ethylene biosynthesis by antisense ACC oxidase RNA prevents

chilling injury in Charentais cantaloupe melons. Plant, Cell and Environment 22:
1 579-1 586.

Bleeker, A.B., Estelle, M.A., Somerville, C., and Kende, H. 1988. Insensitivity to
ethylene conferred by a dominant mutation in Arabidopsis thaliana. Science 241:
1086-1 089.

Bleeker, AB. and Kende, H. 2000. Ethylene: a gaseous sigral molecule in plants. Annu.
Rev. Cell Dev. Biol. 16: 1-18.

Bovy, A.G., Angenent, G.C., Dons, H.J.M., and van Altrost, A. 1999. Heterologous
expression of the Arabidopsis etrI-I . Molecular Breeding 5: 301-308.

Brady, SM. and McCourt, P. 2003. Hormone cross-talk in seed dormancy. J. Plant
Growth Regul. 22: 25-31.

Bragina, T.V., Martinovich, L.I., Rodionova, N.A., Bezborodov, A.M., and Grineva,
GM. 2001. Ethylene-induced activation of xylanase in adventitious roots of

maize as a response to the stress effect of root submersion. Applied Biochemistry
and Microbiology 37: 618-621.

132

Bragina, T.V., Rodionova, N.A., and Grineva, GM. 2003. Ethylene production and
activation of hydrolytic enzymes during acclimation of maize seedlings to partial
flooding. Russian Journal of Plant Physiology 50: 794-798.

Bui, A.Q. and O’Neill, SD. 1998. Three 1-aminocyclopropane-1-carboxylate synthase
genes regulated by primary and secondary pollination sigrals in orchid flowers.
Plant Physiol. 116: 419-428.

Chae, H.S., Cho, Y.G., Park, M.Y., Lee, M.C., Eun, M.Y., Kang, B.G., and Kim, W.T.
2000. Hormonal cross-talk between auxin and ethylene differentially regulates
the expression of two members of the 1-aminocyclopropane-1-carboxylate
oxidase family in rice (Oryza sativa L.). Plant and Cell Physiology 41: 354-362.

Chang, H., Jones, M.L., Banowetz, G.M., and Clark, D.G. 2003. Overproduction of
cytokinins in petunia flowers with Psag12-IPT delays corolla senescence and
decreases sensitivity to ethylene. Plant Physiology 132: 2174-2183.

Chen, W., Provart, N.J., Glazebrook, J ., Katagiri, F ., Chang, H-.S., Eulgem, T., Mauch
F., Luan, S., Zou, G., Whitman, S.A., Budworth, P.R., Tao, Y., Xie, Z., Chen, X.,
Lam, S., Kreps, J .A., Harper, J .F., Si-Ammour, A., Mauch-Mani, B., Heilein, M.,
Kobayashi, K., Hohn, T., Dangl, J .L., Wang, X., and Zhu, T. 2002. Expression
profile matrix of Arabidopsis transcription factor genes suggests their putative
functions in response to environmental stresses. Plant Cell 14: 559-574.

Chow, B. and McCourt, P. 2004. Hormone sigraling from a developmental context.
Journal of Experimental Botany 55: 247-251.

Ciardi, J .A., Tieman, D.M., Jones, J .B., and Klee, H]. 2001. Reduced expression of the
tomato ethylene receptor gene LeETR4 enhances the hypersensitive response to
Xanthomonas campestris pv. vesicatoria. MPMI 14: 487-495.

Clark, D.G., Gubrium, E.K., Barrett, J .E., Nell, T.A., and Klee, H]. 1999. Root
formation in ethylene-insensitive plants. Plant Physiology 121: 53-59.

Clevenger, D.J., Barrett, J .E., Klee, H.J., and Clark, D.G. 2004. Factors affecting seed
production in transgenic ethylene-insensitive petunias. J. Amer. Soc. Hort. Sci.
129: 401-406.

Cox, M.C.H., Benschop, J .J ., Vreeburg, R.A.M., Wagemaker, C.A.M., Moritz, T.,
Peeters, A.J.M., and Voesenek, A.C.J. 2004. The roles of ethylene, auxin,
abscisic acid, and gibberellin in the hyponastic growth of submerged Rumex
palustris petioles. Plant Phys. 136: 2948-2960.

Conner, A.J., Glare, TR, and Nap, J. 2003. The release of genetically modified crops

into the environment, Part 11. Overview of ecological risk assessment. Plant J. 33:
19-46.

133

Crawford, N.M., Kahn, M.L., Leustek, T., and Long, SR. 2000. Nitrogen and sulfur. In
S. Buchanan, W. Gruissem, R Jones, eds. Biochemistry and Molecular Biology of
Plants, American Society of Plant Physiologists, Rockville, Maryland.

Dale, P.J., Clarke, B., and F ontes, E.M.G. 2002. Potential for the environmental impact
of transgenic cr0ps. Nature Biotechnology 20:56-574.

De Martinis, D. and Mariarri, C. 1999. Silencing gene expression of the ethylene-
forrning enzyme results in a reversible inhibition of ovule development in
transgenic plants. Plant Cell 11: 1061-1071.

Diaz, J ., den Have, A., and van Kan, J AL. 2002. The role of ethylene and wound
sigraling in resistance of tomato to Botrytis cinerea. Plant Physiology 129: 1341-
1351.

Ellis, C. and Turner, J .G. 2001. The Arabidopsis mutant cev] has constitutively active

jasmonate and ethylene sigral pathways and enhanced resistance to pathogens.
Plant Cell 13: 1025-1033

Ellstrand, NC. 2001. When transgenes wander, should we worry? Plant Physiology 125:
1543-1545.

Ellstrand, N.C., Prentice, HO, and Hancock, J .F. 1999. Gene flow and introgession
from domesticated plants into their wild relatives. Annu. Rev. Ecol. Syst. 30: 539-
563.

Fowler, S. and Thomashow, M.F. 2002. Aradidopsis transcriptome profiling indicated
that multiple regulatory pathways are activated during cold acclimation in
addition to the CBF cold response pathway. Plant Cell 14: 1675-1690.

Gazzarrini, S. and McCourt, P. 2003. Cross-talk in plant hormone sigraling: what
Arabidopsis mutants are telling us. Annals of Botany 91: 605-612.

Geraats, B.P.J., Bakker, P.A.H.M., and van Loon, LC. 2002. Ethylene insensitivity
impairs resistance to soilbome pathogens in tobacco and Arabidopsis thaliana.
MPMI 15: 1078-1085.

Giovannoni, J .J . 2004. Genetic regulation of fruit development and ripening. Plant Cell
16: $170-$180.

Gray, J .E., Picton, S., Giovannoni, J.J., and Grierson, D. 1994. The use of transgenic
and naturally occurring mutants to understand and manipulate tomato fi'uit

ripening. Plant, Cell and Environment 17: 557-571.

Grbic, V. and Bleeker, AB. 1995. Ethylene regulates the timing of leaf senescence in
Arabidopsis. Plant J. 8: 595-602.

134

Grichko, VP. and Glick, B.R. 2001a. Amelioration of flooding stress by ACC
deaminase-containing gowth-promoting bacteria. 39: 11-17.

Grichko, VP. and Glick, B.R. 2001b. Flooding tolerance of transgenic tomato plants
expressing the bacterial enzyme ACC deaminase controlled by the 355, rolD or
PRB-lb promoter. Plant Physiol. Biochem. 39: 19-25.

Grimmig, B., Gonzalez-Perez, M.N., Leubner-Metzger, G., Vogeli-Lange, R., Meins, F.,
Hain, R., Penuelas, J ., Heidenreich, B., Langebartels, C., Ernst, D., and
Sanderrnann, H. 2003. Ozone-induced gene expression occurs via ethylene-
dependent and —independent sigraling. Plant Molecular Biology 51: 599-607.

Grossmann, K. and Hansen, H. 2001. Ethylene-triggered abscisic acid: a principle in
plant gowth regulation. Physiologia Plantarurn 113: 9-14.

Gu, Y.Q., Wildermuth, M.C., Chakravarthy, S., Loh, Y.T., Yang, H., He, X., Han, Y.,
and Martin, GB. 2002. Tomato transcription factors Pti4, PtiS, and Pti6 activate
defense responses when expressed in Arabidopsis. Plant Cell 14: 817-831.

Guo, H. and Ecker, J .R. 2004. The ethylene sigraling pathway: new insights. Current
Opinion in Plant Biology 7: 40-49.

Hails, RS. 2000. Genetically modified plants—the debate continues. Trends in
Ecological Evolution 15: 14-18.

Hails, RS. 2002. Assessing the risks associated with new agicultural practices. Nature
418: 685-688.

Haines, M.M., Shiel, P.J., Fellman, J .K., and Berger, PH. 2003. Abnormalities in
gowth, development and physiological responses to biotic and abiotic stress in
potato (Solanum tuberosum) transformed with Arabidopsis E T R]. Journal of
Agicultural Science 141: 333-347.

Hancock, J .F . 2003. A framework for assessing the risk of transgenic crops. Bioscience
53: 512-519.

Hedden, P. and Phillips, AL. 2000. Manipulation of hormone biosynthetic genes in
transgenic plants. Current Opinion in Biotechnology 11: 130-137.

Henzi, M.X., Christey, M.C., and McNeil, D.L. 2000. Morphological characterization
and agonorrric evaluation of transgenic broccoli (Brassica oleracea L. var.
italica) containing an antisense ACC oxidase gene. Euphytica 113: 9-18.

Hoffinan, T., Schmidt, J .S., Zheng, X., and Bent, AF. 1999. Isolation of ethylene-
insensitive soybean mutants that are altered in pathogen susceptibility and gene-
for-gene disease resistance. Plant Physiology 119: 935-949.

135

Hokanson, K. 2002. Report of the altered flowering goup. In LL Wolfenbarger, ed,
Proceedings of a Workshop on: Criteria for Field Testing of Plants with
Engineered Regulatory, Metabolic, and Sigraling Pathways, June 3-4, 2002.
Information Systems for Biotechnology, Blacksberg, Virgina, pp 19-26.

Hontzeas, N., Saleh, SS, and Glick, BR. 2004. Changes in gene expression in canola
roots induced by ACC-deaminase-containing plant-gowth-promoting bacteria.
MPMI 17: 865-871.

Hua, J ., Chang, C., Sun, Q., and Meyerowitz, EM. 1995. Ethylene insensitivity
conferred by Arabidopsis ERS gene. Science 269: 1712-1714.

Hussain, A., Black, CR, Taylor, 1.8., and Roberts, J .A. 2000. Does an antagonistic
relationship between ABA and ethylene mediate shoot gowth when tomato

(Lycoperisicon esculentum Mill.) plants encounter compacted soil? Plant, Cell
and Environment 23: 1217-1226.

Ishiki, Y., Oda, A., Yaegashi, Y., Orihara, Y., Arai, T., Hirahayashi, T., Nakagawa, H.,
and Sato, T. 2000. Cloning of an auxin-responsive l-aminocyclopropane-l-
carboxylate synthase gene (CMe-A CS2) from melon and the expression of ACS
genes in etiolated melon seedlings and melon fruits. Plant Science 159: 173-181.

J aglo-Ottosen, K.R., Gilmour, S.J., Zarka, D.G., Schabenberger, O., and Thomashow,
M.F. 1998. Arabidopsis CBF] overexpression induces COR genes and enhances
freezing tolerance. Science 280: 104-106

J ank, B., and Gaugitsch, H. 2001. Assessing the environmental impacts of transgenic
plants. Trends in Biotechnology 19: 371-3 72.

Johnson, PR. and Ecker, J .R. 1998. The ethylene gas sigral transduction pathway: a
molecular respective. Annu. Rev. Genet. 32: 227-254.

Kamachi, S., H. Sekimoto, N. Kondo, and S. Sakai. 1997. Cloning of a cDNA for a 1-
aminocyclopropane-l -carboxylate synthase that is expressed during development
of female flowers at the apices of Cucumis sativus L. Plant Cell Physiology 38:
1197-1206

Kamachi, S., Mizusawa, H., Matsuura, S., and Sakai, S. 2000. Expression of two 1-
aminocyclopropane-carboxylate synthase genes, CS-A CS] and CS-A C52,
correlated with sex phenotypes in cucumber plants (Cucumis sativus L.). Plant
Biotechnology 17: 69-74.

Kende, H. 1993. Ethylene biosynthesis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 44:
283-307.

136

Kende, H., van der Knaap, E., and Cho, HT. 1998. Deepwater rice: 3 model plant to
study stem elongation. Plant Physiology 118: 1105-1110.

Klee, H.J. 1993. Ripening physiology of fruit from transgenic tomato (Lycopersicon
esculentum) plants with reduced ethylene synthesis. Plant Physiology 102: 911-
916.

Klee, H.J. 2002. Engineered changes in ethylene sigral transduction pathways. In LL
Wolfenbarger, ed, Proceedings of a Workshop on: Criteria for Field Testing of
Plants with Engineered Regulatory, Metabolic, and Sigraling Pathways, June 3-4,
2002. Information Systems for Biotechnology, Blacksberg, Virgina, pp 81-84.

Klee, H.J., Hayford, M.B., Kretzrner, K.A., Barry, GE, and Kishore, GM. 1991. F:
Control of ethylene synthesis by expression of a bacterial enzyme in transgenic '
tomato plants. Plant Cell 3: 1187-1193.

 

Kieber, J .J ., Rothenberg, M., Roman, G., Feldmann, K.A., and Ecker, J .R. 1993. CTRl, l
a negative regulator of the ethylene response pathway in Arabidopsis encodes a
member of the Raf family of protein kinases. Cell 72: 427-441.

Knoester, M., van Loon, L.C., van den Heuvel, J ., Hennig, J ., and Bo], J .F. 1998.
Ethylene-insensitive tobacco lacks nonhost resistance against soil-home fungi.
Proc. Natl. Acad. Sci. 95: 1933-1937.

Kunkel, B.K. and Brooks, D.M. 2002. Cross talk between sigraling pathways in
pathogen defense. Current Opinion in Plant Biology 5: 325-331.

Larkindale, J. and Knight, MR. 2002. Protection against heat stress-induced oxidative
damage in Arabidopsis involves calcium, abscissic acid, ethylene, and salicylic
acid. Plant Physiology 128: 682-695.

Lawton, K.A., Potter, S.L., Uknes, S., and Ryals, J. 1994. Acquired resistance sigral
Transduction in Arabidopsis is ethylene-independent. Plant Cell 6: 581-588.

LeNoble, M.B., Spollen, W.G., and Sharp, RE. 2004. Maintenance of shoot gowth by
endogenous ABA: genetic assessment of the involvement of ethylene suppression.
Journal of Experimental Botany 55: 237-245.

Lorenzo, O., Piqueras, R., Sanchez-Serrano, J .J ., and Solano, R. 2003. ETHYLENE
RESPONSE FACTORI integates signals from ethylene and jasmonate pathways
in plant defense. Plant Cell 15: 165-178.

Mayak, S., Tirosh, T., and Glick, B.R. 2004a. Plant gowth-promoting bacteria that

confer resistance to water stress in tomatoes and peppers. Plant Science 166: 525-
530.

137

Mayak, S., Tirosh, T., and Glick, B.R. 2004b. Plant gowth-promoting bacteria that
confer resistance in tomato plants to salt stress. Plant Physiology and
Biochemisn'y 42: 565-572.

McDonald, MP. and Visser, E.J.W. 2003. A study of the interaction between auxin and
ethylene in wild type and transgenic ethylene-insensitive tobacco during
adventitious root formation induced by stagnant root zone conditions. Plant
Biology 5: 550-556.

Messeguar, J. 2003. Gene flow assessment in transgenic plants. Plant, Cell and Organ
Culture 73: 201-212.

Mibus, H. and Tatlioglu, T. 2004. Molecular characterization and isolation of the F/f
gene for femaleness in cucumber (Cucumis sativus L.). Theor Appl Genet 109:
1669-1 676.

Nap, J ., Metz, P.L.J., Escaler, M., and Conner, AJ. 2003. The release of genetically
modified crops into the environment, Part I. Overview of current status and
regulations. Plant J. 33: 1-18.

Nishiuchi, T., Suzuki, K., Kitajima, S., Sato, F., and Shinshi, H. 2002. Wounding
activates immediate early transcription of genes for ERFs in tobacco plants. Plant
Molecular Biology 49: 473-482.

Nakajima, N., Itoh, T., Asai, N., Aona, M., Kubo, A., Azumi, Y., Karnada, H., and Saji,
H. 2002. Improvement in ozone tolerance of tobacco plants with an antisense
DNA for 1-aminocyclopropane-l-carboxylate synthase. Plant, Cell and
Environment 25: 727-735.

Nukui, N., Ezura, H., and Minarnisawa, K. 2004. Transgenic Lotus japonicus with an
ethylene receptor gene Cm-ERS/H 70A enhances formation of infection threads
and nodule primordia. Plant Cell Physiol. 45: 427-435.

O’Donnell, P.J., Jones, J .B., Antoine, F.R., Ciardi, J ., and Klee, H.J. 2001. Ethylene-
dependent salicylic acid regulates an expanded cell death response to a plant
pathogen. Plant Journal 25: 315-323.

Oeller, P.W., Min-Wong, L., Taylor, L.P., Pike, D.A., and Theologis, A. 1991.
Reversible inhibition of tomato fruit senescence by antisense RNA. Science 254:
437-439.

Padgette, S.R., Kolacz, K.H., Delannay, X., Re, D.B., Lavallee, B.J., Tinius, C.N.,
Rhodes, W.K., Otero, Y.T., Marry, G.F., Eichholtz, D.A., Peschke, V.M., Nida,
D.L,. Taylor, N.B., and Kishore, GM. 1995. Development, identification, and
characterization of a glyphosate-tolerant soybean line. Crop Science 35: 1451-
1461.

138

 

Peeters, A.J.M., Cox, M.C.H., Benscop, J .J ., Vreeburg, R.A.M., Bou, J ., and Voesenek,
L.A.C.J. 2002. Submergence research using Rumex palustris as a model; looking
back and going forward. Journal of Experimental Botany 53: 391-398.

Peferon, M. 1997. Progess and prospects for field use of Bt genes in crops. Trends in
Biotechnology 15: 173-177.

Penmetsa, RV. and Cook, DR. 1997. A legume ethylene-insensitive mutant
hyperinfected by its rhizobial symbiont. Science 275: 527-530.

Penninckx, I.A.M.A., Eggermont, K., Terras, F .R.G., Thomma, B.P.H.J., De Samblanx,
G.W., Buchala, A., Manners, J .M., and Broekaert, W.F. 1996. Pathogen-induced
systemic activation of a plant defensin gene in Arabidopsis follows a salicylic aid-

dependent pathway. Plant Cell 8: 2309-2323.

Pierik, R., Cuppens, M.L.C., Voesenek, L.A.C.J., and Visser, E.J.W. 2004. Interactions
between ethylene and gibberellins in phytochrome-mediated shade avoidance
responses in tobacco. Plant Phys. 136: 2928-2936.

Pierek, R., Visser, E.J.W., de Kroon, H., and Voesenek, L.A.C.J. 2003. Ethylene is
required in tobacco to successfully compete with proximate neighbors. Plant,
Cell and Environment 26: 1229-1234.

Poppy, G. 2000. GM crops: environmental risks and non-target effects. Trends in Plant
Science 5: 4-6.

Papadopoulou, E. 2002. Sex expression in cucurbits: the role of ethylene synthesis and
perception and sex determination genes. PhD Dissertation. Michigan State
University

Papadopoulou, E., Little, H.A., Hammar, S.A., and Grumet, R. 2005. Effect of modified
endogenous ethylene production on hermaphrodite flower development and fruit
production in melon (Cucumis melo L.). in submission.

Rahman, A., Hosokawa, S., Oono, Y., Amakawa, T., Goto, N., and Tsururrri, S. 2002.
Auxin and ethylene response interactions during Arabidopsis root hair
development dissected by auxin influx modulators. Plant Physiology 130: 1908-
1917.

Rao, M.V., Lee, H., and Davis, KR. 2002. Ozone-induced ethylene production is
dependent on salicylic acid, and both salicylic acid and ethylene act in concert to
regulate ozone-induced cell death. Plant Journal 32: 447-456.

Rojo, E., Solano, R., and Sanchez-Serrano, J .J . 2003. Interactions between sigraling
compounds involved in plant defense. J. Plant Growth Regul. 22: 82-98.

139

 

Schenk, P.M., Kazan, K., Wilson, I., Anderson, J .P., Richmond, T., Somerville, SC, and
Manners, J .M. 2000. Coordinated plant defense responses in Arabidopsis
revealed by rrricroarray analysis. PNAS 97: 11655-11660.

Shaw, J .F ., Chen, H.H., Tsai, M.F., Kuo, C., and Huang, LC. 2002. Extended flower
longevity of Petunia hybrida plants transformed with boers, a mutated ERS gene
Brassica oleracea. Molecular Breeding 9: 211-216.

Shibuya, K., Barry, K.G., Ciardi, J .A., Loucas, H.M., Underwood, B.A., Nourizadeh, S.,
Ecker, J .R., Klee, H.J., and Clark, D.G. 2004. The central role of PhEIN2 in
ethylene responses throughout plant development in petunia. Plant Physiology

136: 2900-2912.

Smalle, J. and Van Der Straeten, D. 1997. Ethylene and vegetative development.
Phyiologia Plantarum 100: 593-605.

Sinn, J .P., Schlagnhaufer, C.D., Arteca, RN, and Pell, E.J. 2004. Ozone-induced
ethylene and foliar injury responses are altered in l-aminocyclopropane-l -

carboxylate synthase antisense potato plants. New Phytologist 164: 267-277.

Sisler, EC. and Serek, M. 1997. Inhibitors of ethylene responses in plants at the
receptor level. Physiologia Plantarum 100: 577-582.

Snow, A.A. and Palma, RM. 1997. Commercialization of transgenic plants: potential
ecological risks. Bioscience 47: 86-96.

Snow, A.A. 2002a. Possible phenotypic effects of genetically modified pathways on
gene flow from field tests. In LL Wolfenbarger, ed, Proceedings of a Workshop
on: Criteria for Field Testing of Plants with Engineered Regulatory, Metabolic,
and Sigraling Pathways, June 3-4, 2002. Information Systems for Biotechnology,
Blacksberg, Virgina, pp 63-67.

Snow, A.A. 2002b. Transgenic crops—why gene flow matters. Nature Biotechnology
20: 542.

Steams, J .C. and Glick, BR. 2003. Transgenic plants with altered ethylene biosynthesis
or perception. Biotechnology Advances 21: 193-210.

Steptanova, AN. and Ecker, JR. 2000. Ethylene sigraling: fi'om mutants to molecules.
Current Opinion in Plant Biology 3: 353-360.

Tanirnoto, M., Roberts, K., and Dolan, L. 1995. Ethylene is a positive regulator of root
hair development in Arabidopsis thaliana. Plant Journal 8: 943-948.

140

 

Tholen, D., Laurentius, A.C.J., and Poorter, H. 2004. Ethylene insensitivity does not
increase leaf area or relative gowth rate in Arabidopsis, Nicotiana tabacum, and
Petunia x hybrida. Plant Physiology 134: 1803-1812.

Thomma, B.P.H.J., Penninckx, I.A.M.A., Broekaert, W.F., and Carnmue, B.P.A. 2001.
The complexity of disease sigraling in Arabidopsis. Current Opinion in
Immunology 13: 63-68.

Trebitsh, T., Staub, J .E. and O’Neill, SD. 1997. Identification of a 1-
aminocyclopropane-l -carboxylic acid synthase gene linked to the Female (F)
locus that enhances female sex expression in cucumber. Plant Physiology 113: L
987-995.

Tuominen, H., Overmyer, K., Keinanen, M., Kollist, H., and Kangasjarvi, J. 2004.
Mutual antagonism of ethylene and jasmonic acid regulates ozone-induced
spreading cell death in Arabidopsis. Plant Journal 39: 59-69.

 

Van der Graaff, E.E., Auer, C.A., and Hooykaas, P.J.J. 2001. Altered development of
Arabidopsis thaliana carrying the Agrobacterium tumefaciens ipt is partially due
to ethylene effects. Plant Growth Regulation 34: 305-315.

Van der Straeten, D., Zhou, Z.Y., Prinsen, E., Van Onckelen, H.A., and Van Montagu
M.C. 2001. A comparative molecular-physiological study of submergence
response in lowland and deepwater rice. Plant Physiology 125: 955-968.

Van Zhong, G. and Burns, J .K. 2003. Profiling ethylene-related gene expression in
Arabidopsis thaliana by microarray analysis. Plant Molecular Biology 53: 117-
131.

Verbeme, M.C., Hoekstra, J ., Bol, J .F., and Linthorst, H.J.M. 2003. Sigraling of
systemic acquired resistance in tobacco depends on ethylene perception. Plant
Journal 35: 27-32.

Voesenek, L.A.C.J., Benschop, J .J ., Bou, J ., Cox, M.C.H., Groeneveld, H.W., Millenaar,
F .F ., Vreeburg, R.A.M., and Peeters, A.J.M. 2003. Interactions between plant
hormones regulate submergence-induced shoot elongation in the flooding-tolerant
dicot Rumex palustris. Annals of Botany 91: 205-211.

Wang, K.L.C., Li, H., and Ecker, J .R. 2002. Ethylene biosynthesis and sigraling
networks. Plant Cell S131-S151.

Whitelaw, C.A., Lyssenko, N.N., Chen, L., Zhou, D., Matteo, A.K., and Tucker, M.L.
2002. Delayed abscission and shorter intemodes correlate with a reduction in the

ethylene receptor LeE T R I transcript in transgenic tomato. Plant Physiology 128:
978-987.

141

Wilkinson, J.Q., Lanahan, M.B., Yen, H.C., Giovannoni, J .J ., and Klee, H.J. 1995. An
ethylene-inducible component of sigral transduction encoded by Never-ripe.
Science 270: 1 807-1 809.

Wilkinson, M.J., Sweet, J ., and Poppy, GM. 2003. Risk assessment of GM plants:
avoiding gidlock? Trends in Plant Science 8: 208-212.

Woeste, K.E., Vogel, J .P., and Kieber, J .J . 1999. Factors regulating ethylene
biosynthesis in etiolated Arabidopsis thaliana seedlings. Physiologia Plantarum
105: 478-484. F

Wolfenbarger, LL. and Grumet, R. 2002. Executive summary. In LL Wolfenbarger, ed,
Proceedings of a Workshop on: Criteria for Field Testing of Plants with
Engineered Regulatory, Metabolic, and Sigraling Pathways, June 3-4, 2002.
Information Systems for Biotechnology, Blacksberg, Virgina, pp 5-12. L-

 

Wolfenbarger, LL. and Phifer, PR. 2000. The ecological risks and benefits of
genetically engineered plants. Science 290: 2088-2093.

Yamasaki, S, Fujii, N and Takahashi, H. 2003. Characterization of ethylene effects on
sex determination in cucumber plants. Sex Plant Reproduction 16: 103-111.

Yang, SF. and Oetiker, J.H. 1998. Molecular biology of ethylene biosynthesis and its
application in horticulture. J. Japan. Hort. Soc. 67: 1209-1214.

Yi, H.C., Joe, S., Nam, K.H., Lee, J.S., Kang, B.G., and Kim, W.T. 1999. Auxin and
brassinosteroid differentially regulate the expression of three members of the 1-
aminocyclopropane-l-carboxylate synthase gene family in mung bean (Vigna
radiata L.) Plant Molecular Biology 41: 443-454.

Zarembinski, T.I. and Theologis, A. 1994. Ethylene biosynthesis and action: a case of
conservation. Plant Molecular Biology 26: 1579-1597.

Zhou, Z.Y., Vriezen, W., Van Caeneghern, W., Van Montagu, M., and Van der Straeten,
D. 2001. Rapid induction of a novel ACC synthase in deepwater rice seedlings
upon complete submergence. Euphytica 121: 137-143.

Zhou, Z.Y., Engler, J .D., Rouan, D., Michiels, F., Van Montagu, M., and Van der
Straeten, D. 2002. Tissue localization of a submergence-induced 1-
aminocyclopropane-l-carboxy1ic acid synthase in rice. Plant Physiology 129: 72-
84.

142

CONCLUSIONS AND FUTURE RESEARCH

The objectives of this research included tissue targeted modification of ethylene
perception of melon plants and study of the effects of increased endogenous ethylene
production of sex determination and fruit set in field gown 35S::A CS melon. Another
objective was to evaluate secondary effects associated with genetic engineering to modify
traits regulated by ethylene within the context of ecological risk assessment.

For the first objective, andromonoecious melon was transformed with gene
constructs of the Arabidopsis etrI-I mutant driven by heterologous floral targeted
promoters. The Crab ’s Claw promoter from Arabidopsis was used to target carpels and
nectaries, and the Apetala3 promoter, also from Arabidopsis was used to target stamens
and petals. Greenhouse characterization of sex expression and floral morphology of
AP3::etrI-I and CRC::etrI-1 transgenic melon plants provided unexpected results.
AP3::etr1-I plants showed increased maleness as demonstrated by largely abolished
pistillate bud formation and, of the few pistillate buds produced, many showed poorly
developed carpels. In contrast, CRC.°:etrI-I plants produced the opposite phenotype,
with increased femaleness as demonstrated by earlier and increased pistillate bud
production of the main stem and the observation of many female buds, in place of the
expected hermaphrodite bud. Collectively, these observations suggest that a reciprocal
relationship exists between stamens and carpels in the hermaphrodite bud development
such that stamens must perceive ethylene to promote carpel formation and vice versa.
While pistillate buds initiated on CRC::etrI-1, most aborted early in development,

providing support for a role of ethylene in sustained carpel maturation. In order to firrther

143

 

establish the proposed role of ethylene perception by the stamens and carpels, it will be
necessary to determine the timing and localization of etrI-I expression in the developing
buds. Support for the model could be provided by expression of the etrI-I transgene at
the time critical for sex determination. In situ hybridization of apices containing floral
buds at various ages would allow a detailed analysis of etrI-I expression. Interest for
firture research also includes the evaluation of transgenic monoecious cucumber L
harboring AP3::etrI-I and CRC::etrI-I to allow further insight into the role of ethylene I

in sex determination. i

 

Therefore, future research on the role of ethylene perception in sex determination
should focus on trying to fill in the gaps of the proposed or other established models.
Questions that remain unsolved include: 1. Is the expression of the etrI-I in transgenic
AP3::etr1-I and CRC::etrI-1 within the developing floral primordia consistent with the
proposed role of ethylene in the development of carpels and stamens? 2. Is the proposed
model of ethylene role in sex determination for melon consistent with its role in
cucumber? 3. Is the M locus in cucumber analogous to the A locus in melon? 4. Is the
inhibition of stamens in pistillate buds of mm cucumber the result of a constitutive
response to ethylene or of a reduced ability to respond to ethylene? 5. Is a more specific
carpel-targeted promoter necessary to confidently support the proposed model?

For the second goal, a field trial was conducted to study sex expression and fruit
set on transgenic andromonoecious melon constitutively over-expressing an ACC
synthase (A CS) gene to confer increased endogenous ethylene production. Consistent
with geenhouse observations, field gown 35S::A CS melon plants showed earliness and

increase of mature hermaphrodite flower production indicating ethylene is important for

144

firll hermaphrodite development. 35S: :A CS melon plants also showed earlier and
increased fruit set on the main stem. The observation of increased sequential fruit set on
the main stem suggests that ethylene may play a role in the sigraling associated with fruit
set patterns. The earliness of hit set could have horticultural sigrificance in areas, such
as Michigan, with a short gowing season. However, pleiotropic effects, which may be
associated with constitutively elevated ethylene such as accelerated ripening or altered
stress response, would be agonomically undesirable.

Future research could focus on achieving temporal and tissue localized ACS l

 

expression to enhance femaleness produce earliness in fruit set without deleterious
secondary effects. As described in Appendix A, CRC::A CS and AP3::A CS constructs
have already been prepared. Unfortunately, as was demonstrated in CRC::etrI-I melon
plants and in preliminary evaluation of CRC::A CS melon, the CRC promoter in this
heterologous system lacked tissue specificity. The effect of AP3 : :A CS on sex expression
is yet to be determined as the construct has not yet been transformed into melon.
Additional ACS constructs to target floral tissue include the promoter region of a melon
homologue of CRC, although this would require its cloning. Additional options might be
the promoter of the Arabidopsis stamen and carpel targeted AGAMOUS MADS box
gene (Y anofsky et al., 1990), or the cucumber carpel restricted AGAMOUS-Like gene
(CAG2) (Perl-Treves et al., 1998), if its promoter was available.

For the final objective, a case study was conducted to describe many of the
possible pleitropic effects associated with altered ethylene sigraling to demonstrate how
modification of sigraling pathways can affect the activities of many genes and result in

unexpected phenotypes. Secondary effects which could alter gene flow or fitness in the

145

event of gene flow were emphasized with regard to their relevance from a perspective
ecological risk assessment. Continuation of this work might include description of
possible methodology to implement in ecological risk assessment strategies to identify
relevant pleiotropic effects. In addition, genetic engineering strategies to minimize

unexpected phenotypes could also be discussed.

References

Perl-Treves, R., Kahana, A., Rosenman, N., Xiang, Y., and Silberstein, L. 1998.
Expression of multiple AGAMOUS-Like genes in male and female flowers of
cucumber (Cucumis sativus L.). Plant Cell Physiol. 39: 701-710.

Yanofsky, M.F., Ma, H., Bowman, J.L., Drews, G.N., Feldmann, K.A., and Meyerowitz,

EM. 1990. The protein encoded by the Arabidopsis homeotic gene agarnous
resembles transcriptional factors. Nature 346: 35-39.

146

 

Appendix A:

Floral targeted expression of a heterologous ACS in andromonoecious melon

Introduction

Analysis of 35S::A CS transgenic melon plants allowed the evaluation of increased
endogenous ethylene on sex expression in melon (Papadopoulou et al., 2005). However,
constitutive expression of a transgene with the potential to affect many processes
throughout plant gowth and development, such as ACS, is undesirable from a practical
and regulatory perspective; therefore, from an applied standpoint, constitutive ACS
expression should be avoided to minimize secondary effects on other plant
characteristics. In addition, tissue specific or temporal expression of the ACS transgene
may permit further eludication of the critical timing or site of ethylene synthesis required
to impact sex expression. Therefore, constructs were prepared to evaluate increased ACS
expression targeted to developing petals and stamens via the promoter of the Arabidopsis
class B floral identity gene Apetala3 (Irish and Yamarnoto, 1995) and to developing
carpels and nectaries via the promoter of Arabidopsis carpel and nectary development

regulator Crab ’s Claw (Bowman and Smyth, 1999).

Materials and Methods
Plasmid Construction

CRC::A CS: The petunia ACS cDNA (kindly provided by Dr R. Woodson, Purdue
University) was PCR amplified from pBS-A CS to introduce a BamHI site at the 5’ and 3’

ends using specific primers 5’-CCCGGATCCCTCTCTCTCTCTCCTCGTGCC (RG280)

147

 

for the 5’ end and 5’GGGGGATCCGGGCGAATTGGGTACCGGGCC (RG281) for the
3’ end, and the resulting PCR fragnent cloned with pGEM®-T Easy Vector System
(Promega Corporation, Madison, Wisconsin). The etrI-I fi'om the previously prepared
pCambia2300-CRC::etrI-I (Chapter 2) was removed as a BamHI fragnent and replaced
with the ACS BamHI fragnent.

AP3::A CS: Construct preparation was analogous to CRC::A CS, with the
exception that ACS was transferred as a BamHI fragment into the previously prepared
pCambia2300-AP3::etr]-] (Chapter 2), with etrI-I dropped out as a BamHI fragnent.
Plant Transformation

The pCambia-CRC-A CS and —AP3-A CS constructs were introduced into
Agrobacterium tumefaciens strain C58 (Deblaere et al.,l985) for subsequent
transformation into melon. Andromonoecious muskmelon, Hale’s Best Jumbo (Hollar
Seed, Rocky Ford, Colorado), was transformed via Agrobacterium-mediated
transformation of cotyledon tissue via the method of Fang and Grumet (1990) with
revisions based on the cucumber transformation method of Tabei et a1. (1998) (Chapter
2). Regenerated shoots (T0) were selected on kanamycin (200 mg/L) and evaluated for
the presence of neomycin phosphotransferase by NPTII ELISA (Agia®, Elkhart, Indiana)
as well as presence of the ACS gene by polymerase chain reaction (PCR) (described
below). PCR-Positive To plants were transferred to the geenhouse for self-pollination
and production of T1 progeny for further evaluation.

PCR analysis
The Wizard Genomic Plant DNA Purification Kit (Promega Corporation,

Madison, WI) was used to extract genomic DNA from leaf tissue of melon. PCR was

148

 

used to amplify the inserted transgene using 5’ primer RG280 and 3’ primer RG281 for
amplification of A CS, and 5’ primer 5’-CTTGCAATCCCTAGCCAG (RG 299) and 3’
primer 5’-CCAGCGTTACTCTTCAATGC (RG310) for amplification of CRC-A CS. The
amplification reactions of ACS were carried out with Taq DNA polymerase (Invitrogen,
Carlsbad, California) in 3.0 mM MgCl2 for 1 cycle of 2 minutes at 95°C followed by 40
cycles of 95°C for 1 minute, 62°C for 1 rrrinute, and 72°C for 1 minute, then 1 cycle of 5
minutes at 72°C (Sambrook and Russell, 2001). Amplication reactions of CRC-A CS were
carried out as with ACS but with the following progarn: 1 cycle of 2 minutes at 95°C
followed by 40 cycles of 95°C for 50 seconds, 58°C for 50 seconds, and 72°C for 2:20
minutes, then 1 cycle of 5 minutes at 72°C.
Northern Analysis
Leaf, male bud, and female bud tissue of geenhouse gown T1 CRC::A CS lines 168 and
169, 35S::A CS line 4, and wild type Hale’s Best Jumbo were sampled for RNA isolation.
Petals were excised from the male buds with a razor blades and desigrated as “petals”
while the remaining portion of the male bud containing stamens, nectaries, and floral cup
was desigrated “lower bud.” Female buds were cut with a razor blade such that the top
portion of contained stamen, stigna, and petal tissue (designated “sti, pet, sta”) and the
bottom portion contained the tissue of the ovary (designated “ovary”). RNA was isolated
and hybridization carried out as described in Chapter 3.
Ethylene Experiments

Young leaves (approximately 4-7 cm wide) were removed fi'om lateral branches
of geenhouse gown plants with a razor blade, weighed, and placed in a 50 ml syringe

containing ‘A of a Whatrnan #2 filter paper (90 mm) moistened with 200 uL distilled

149

 

water (dH2O). The syringe plunger was adjusted to 30 mL, and the end of the syringe
was fitted with a rubber septum and sealed with parafilm. Gas samples were removed via
an airtight syringe (lmL) inserted through the rubber septa into the airspace of the tissue
containing 50mL syringe. Samples were analyzed by gas chromatogaphy (Hach Carle
series 100GC, Linear 1200 recorder) with an activated alumina column and flame
ionization detector. Samples were collected three times from 4 randomly sampled plants L
ri‘

per genotype.

 

Results

While both AP3::A CS and CRC::A CS transformation constructs were prepared.
Only two families of TI transgenic melon plants harboring CRC::A CS were available in
time for evaluation. T1 seed has been produced for seven additional CRC::A CS melon
lines; however, no AP3::A CS melon lines were produced. Preliminary geenhouse
characterization of two T1 CRC::A CS lines included PCR, Northern blot hybridization,
and measurement of ethylene production, but no evaluation of sex expression was
conducted. PCR verified the presence of the CRC::A CS construct. Segegation ratios are
based on PCR results and are consistent with one transgene insertion event in CRC168
and two in CRC169 (Table 1).

Northern analyses were done to evaluate expression in the leaf, male bud, and
female bud tissue of CRC::A CS lines 168 and 169 (CRC168 and CRC169). While the
CRC promoter was expected to confer transgene expression to floral regions of the carpel
and nectary, northems of leaf tissue showed ACS transgene expression in the leaf of

CRC168 and CRC169 melon comparable to that of the 358 line ACS4 (Figure 1A).

150

 

Table A.1. Segegation analysis of CRC::A CS T1 transgenic melon.

 

 

 

 

Segregation Expected
Line Ploidy 2’
Observed Ratio
CRC168 9:4 2n 3:1 1.12ns
CRC169 16:1 2n 15:1 0.02 ns

 

Progeny were evaluated for the presence of the ACS transgene by PCR amplification.

ns- )8 value not sigrificantly different at p=0.05

151

A. Leaf
WT 35S CRC168 CRC169

 

B. Male Bud

Lower bud
WT 358 168 169 WTa Is35$ 168 169

    

C. Female Bud
Pet, sti, sta
WI" 358 168 169 Ma ry35$ 168 169

 

Figure A.1. Northern hybridization of heterologous ACS gene expression in tissue from
greenhouse gown T1 transgenic CRC::A CS melon. The top panel of each pair shows
hybridization of total RNA from non-transgenic plants (WT), 35S::A CS plants (35S), and
CRC::A CS plants ( lines 168 and 169) with DIG-labeled ACS probe. The bottom panel
shows rRNA stained with ethidium bromide. (A) leaf tissue. (B) Male buds (upper bud,
including stamen, nectary, and floral cup, and petal region). (C) F ernale Buds

(petal/stigna/stamen region and ovary region)

152

Northern blots of male bud tissue showed similar lack of tissue specificity from the CRC
promoter in this heterologous system (Figure 13); however, hybridization to RNA
isolated from female buds suggested higher expression levels in CRC::A CS lines than the
35S::A CS lrne (ACS4) used as a positive control, taking into account the higher loading
of RNA from ACS4 than CRC168 and CRC169 (Figure 1C).

Gas chromatogaphy was used to measure ethylene production from leaves of
CRC168 and CRC169 in comparison to ACS4 and WT. Both CRC168 and CRC169
showed the elevated ethylene evolution relative to wild type Hale’s Best Jumbo,

consistent with leaky expression from CRC promoter (Figure 2).

153

 

Leaf Ethylene
600

500
400

300

Ethylene (pmollg)

200

100

 

WT ACS4 M1 68 M1 69

Figure A.2. Ethylene production fi'om leaves of wild type (WT) Hale’s Best Jumbo,
transgenic 35S::ACS (ACS4), and CRC::A CS (M168 and M169) Hale’s melon. Samples
were collected three times from randomly sampled plants, 4 times per genotype. (All

data are means :t standard error.)

154

 

References

Bowman, J .L. and Smyth, DR. 1999. CRABS CLA W, a gene that regulates carpel and
nectary development in Arabidopsis, encodes a novel protein with zinc finger and
helix-loop-helix domains. Development 126: 2387-2396.

Fang, G. and Grumet, R. 1990. Agrobacterium tumefasciens mediated transformation
and regeneration of muskmelon plants. Plant Cell Rep. 9: 160-164.

Irish, V.F. and Yamamoto, Y.T. 1995. Conservation of floral homeotic gene firnction
between Arabidopsis and Antirrhinum. Plant Cell Rep. 9:160-164.

Papadopoulou, E., Little, H.A., Hammar, S.A., Grumet, R. (2005) Effect of modified
endogenous ethylene production on hermaphrodite flower development and mu
production in melon (Cucumis melo L.). in submission.

Tabei, Y., Kitade, S., Nishizawa, Y., Kikuchi, N., Kayano, T., Hibi, T., and Akustu, K.
1998. Transgenic cucumber plants harboring a rice chitinase gene exhibit
enhanced resistance to gay mold (Botrytis cinerea). Plant Cell Reports 17: 159-
164.

155

 

 

Appendix B:

Cloning of a partial genomic sequence of CS-A CS3

Introduction:

The objective this project was to gain a better understanding of the feminizing
effect of ethylene and its relationship to the F -locus (Female) conferring gynoecy in L"
cucumber. Analysis of the near isogenic monoecious (fl) and gynoecious (FF) cucumber

lines demonstrated complete co-segegation of the F allele with an additional copy of the

 

CS-A CS] gene, designated as ACS] G (Trebitsh et al., 1997). Southern hybridization
analysis suggested that ACS] and ACS] G were closed linked, residing on a 19.4 kb
BamH] fragnent. I sought to clone genomic sequences of CS-A CS] and CS-A CS] G to
compare coding sequences as well as non-coding regulatory regions to look for sequence
elements that may confer expression specificities (e. g. hormone response elements, tissue
or developmental targeting).

The strategy for this cloning project was to use inverse PCR to amplify the
desired unknown regions flanking the partial ACS] sequence reported by Trebitsh et a1.
(1997) in GenBank (accession number U598l3). Inverse PCR utilizes the polymerase
chain reaction to amplify unknown flanking regions of a known template that is within a
restriction fi'agnent that has been self-ligated to form a monomer circle using primers
oriented in opposing directions. Restriction site information concerning the two genes of

interest was available from Trebitsh et a1. (1997).

156

Materials and methods

High molecular weight nuclear DNA was isolated from gynoecious (GP-14) and
monoecious cucumber using Biol 01 Floraclean nuclear DNA extraction kit. DNA was
fully digested with Xba], HindIII, or MspI. Southern analysis of Trebitsh et a1. (1997)
indicated that digestion with Xba] should result in the presence of 10 kb and 12 kb
fragnents, which contained the CS-A CS] and CS-A CSIG genes, respectively. Similarly,

Southern analysis indicated that digestion with HindIII should result in the presence of

 

two Skb fragnents, each corresponding to one of A CS] genes and that Msp] digestion
should result in 0.75 and 2.5 kb fragnents corresponding to CS-ACS] and the a 3.25 kb
fragnent corresponding to CS-A CS] G (Trebitsh et al., 1997).

The fully digested genomic DNA (verified by gel electrophoresis) was subjected to
conditions favoring intramolecular ligation (ie larger volume, overnight at 16°C). The
precipitated and resuspended ligation product was used as template for inverse PCR using
primers desigred from sequence information for CS-A CS] (GenBank accession number
U598l3) and homologous CS-A CS3 (GenBank accession number AB006805, Shiomi et
al., 1998) (Figure 1). The resulting PCR products were analyzed by gel electrophoresis;
bands corresponding to the desired fragnent size based on the above restriction

information were isolated fi'om the gel and cloned for sequence analysis.

157

 

 

 

 

 

 

210 149329
—> ->—>
I ' 'l .
202 5787 M

— CS-ACS1/3 (A8006805)
[:| cs-Acs1 (U598513)

 

Inverse PCR Primers:
RG149: 5’-GGCTCCCAGGTTT'I’CGAGTTGGTGC-3’
RG1 29: 5 ’ -GGTCGTGGCGGCTGCTACTAAAATG-3 ’
RG1 87: 5 ’ -GGAAGCCGTTGGAGCTAGTGCAATG-3 ’
RG201 : 5 ’ -GCCTAGGGATGGATGGAAGAAG-3 ’

RG202: 5 ’ -CAGCTGGCGCTACATCAGCC-3 ’

Figure 8.]. Layout of primers desigred for inverse PCR based on CS-A CS1 and CS-

ACS3 sequences. Numbered arrows indicate relative location of primer sequence. M

denotes the relative location MspI restriction sites.

158

 

Results and Discussion

Initial efforts on this project focused on amplification of circularized Xbal and
HindIII fragnent by inverse PCR. However, many attempts at optimizing the PCR
system which included nested primers, varying concentrations of template, and various
polymerases failed to warrant amplification of DNA fiagnents near the desired size.
Therefore, efforts were shifted to the Msp] digest in order to obtain to smaller self-1i gated
product for PCR amplification.

Sequential PCR amplification with primers 3’ of the Msp] site in the Trebitsh et
a1. (1997) sequence (RG187/RG149 followed by RG187/RG129) led to identification of a
600 bp fragnent. Subsequent cloning and sequence analysis indicated perfect homology
to CS-ACS3 (GenBank accession number AB006805, Shiorrri etal., 1998). The CS-ACS3
cDNA, which was found to be identical to ACS] (this work; Mibus and Tatlioglu, 2004)
cloned in a study which characterized differential expression of A CC synthase and ACC
oxidase genes in cucumber fi'uit in response to auxin and wounding.

A 3’ primer (RG201) was desigred to a region at the 5’ end of ACS1/3 coding
sequence to amplify non-coding sequence further upstream by inverse PCR. This 3’
primer in combination with a 5’ primer just before the Msp] site (RG202) in the Trebitsh
sequence was added to the ligated Msp] digestion for PCR. A 2.0 kb fragnent was
amplified and cloned. Sequence evaluation showed perfect homology to the 5’ and 3’
ends ACS3 sequence as well as new non-coding 5’ sequence.

Non-coding sequence was evaluated for auxin response element (ARE) sequence
TGTC and ethylene response element sequence (ERE) GCCGCC. The search for ARES

yielded eight matches; six are upstream of the putative TATA box, and three are the

159

 

 

sequence complement GACA (Figure 2). However, no putative EREs were identified.
Therefore, this sequence analysis provides support for the modulation of ethylene levels

by auxin.

References

Mibus, H. and Tatlioglu, T. 2004. Molecular characterization and isolation of the F/f
gene for femaleness in cucumber (Cucumis sativus L.). Theor. Appl. Genet. 109:
1669-1676.

Shionri, S., Yamamoto, M., Ono, T., Kakiuchi, K., Nakamoto, J ., Nakatasuka, A., Kubo,
Y., Nakarnura, R., Inaba, A., and Irnaseki, H. 1998. cDNA cloning of ACC
synthase and ACC oxidase genes in cucumber fruit and their differential
expression by wounding and auxin. J Jpn Hortic Sci 67: 685-692.

Trebitsh, T., Staub, J .E., and O’Neill, SD. 1997. Identification of a 1-
aminocyclopropane-l -carboxylic acid synthase gene linked to the Female (F)
locus that enhances female sex expression in cucumber. Plant Physiology 113:
987-995.

160

.85; E
on as as: as x3 «5: 3523.5 ea £5. .56 3.8 :6 $8 no 8883 382.8 .m e. 2 .3 as»...—

n no N gggfivg

~41-1uuuaaflnqa<<<W¢1<—-¢-u—u~14—<.¢uu-¢aq—<<dldi

 

—uu14—:-qacaau<~ddaq—14<uHaaaQ-d444-uuna—A!!l—alaI-a441

 

 

 

  

 

 

 

 

OOHN 09°“ 906“ 656“ GUON omen ovafl anon ONON Oflon
coon 914.4444": 4 nienJhlddl lrlli 11 . . . : p.14gnaunathaqutng
00mm 1:)‘4e .4114 . . 3114 41 qu’n‘e?) .l a
8.: g3§69§52§ - ll 14 is: 1111:
OOBH 1 , , 11:34 .14 ..... ill 1 15.- “Hnarhvldlr ll illrr lr_.lul.,1..'
coca gggsh. .l 4411 4 : 41 nubeuunnSJE
..... 12...... 1.......4_..44~4...14:—42.4

~IIAl-Il11.1114-llaudda‘1lHand.qaauudquii—«qua—-q-fi41l1‘ll.r

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

coo." oanu coma crud 00nd Band. avnu anmu sand can”
r

can." 4.3.9:: «541143.35: iii: 1: corpus—)1.-unIaJuUUUgflchuQRJiuJuud illilu.
Dov“ an lliialri.n-nieue§u ll . 111141.14 l 114 it 1114 all: til 1
con." 44 ll 4 lam—3. 3’: 4.4144 444: rrrrrrrrr ind 4 4r 1 41111 441.4
coma 2.4.—3:54:34 4 . 4.111 4 444444 1 iii
OOH." n :3 ill. iiiiiiii 4 4 .ll 41 ii :1 11:11.6
—I¢II-‘111-4C.l.‘l‘fl-4Ql¢1-Ididr-‘d.‘lqlidn.lllI—rIIICq‘IIl‘C‘Id.‘GII.4‘I‘dfllird.‘i‘fi‘d‘dd-ll¢IqNI¢1—q‘44

06.: aqua coca chad anon anon ovoH anon anon Odo."
coca aaufiUfiM-‘uia 4444444444 r 4 4:44. 4:554 444 r 4. 4 4 4114 ll:
00ml «no aqnuglhignunuinqnqasflr-flnnu q.- rllril 4144...
can E£E§3§E4§§E¢§h§4nhgggggd
och r4531 «I‘IIIIl-lll'l- delindllll: Ililli nIfiidflJoaoliidi I It“
000 4 .141 ll 4 «indignant: 1:335: qJJu:n443.5:a«if:1J5‘u414u99454qunqgniu54«net-End“: a)
‘QIII-‘flli-Idifl.Iul‘didq‘didl-‘1‘4-fi4l“‘ld<1q‘l‘1-.I4‘_liflidli1l-ICII—I¢IG.-d‘d_rd1drl‘I-QI-I-IN—l.l‘

can can can Ohm can can ovm can can OHm
con Efigiggdgflnh‘ngg‘ggEggggtgfigdg

oov kghggggt¢§§4ggmgzgaggqen 3.54 :54

 

8m figgoggfigfismggfi 4 l- . 3......4
8n «E396.85efifiigfiagggéefiogoggfiogggfiagflfififiiflfiéi

OOH F _ - u a. F- n. F.- I: II! 3': d’i‘fld at! -

 

“dune—«uuu—quua—H<.<-<du¢—--¢aid-—1a41—<.14—<a--u-1-aquuufidn<qdqiqn—-¢-acan-q.aqnud-I-qd-id-unaud

com on o. . or cm on av on on ca

161

   

Iffilllljflliljlljfll11115131