THE EFFECT OF CARPEL AND STAMEN PRIMORDIA-TARGETED ETHYLENE
PRODUCTION AND PERCEPTION ON SEX EXPRESSION IN MELON (CUCUMIS MELO
L.)
By
Jessica Ann Taft

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Genetics - Doctor of Philosophy
2013

ABSTRACT
THE EFFECT OF CARPEL AND STAMEN PRIMORDIA-TARGETED ETHYLENE
PRODUCTION AND PERCEPTION ON SEX EXPRESSION IN MELON (CUCUMIS MELO
L.)
By
Jessica Ann Taft
Commercial melons (Cucumis melo L.) are typically andromonoecious, first producing
vegetative nodes followed by a male flower-only phase, then a male and bisexual flower phase.
Unisexuality arises from differential suppression of sex organ primordia, and ethylene is a key
factor modulating sex expression. Thus, a comprehensive model of melon sex determination
must include sex organ suppression developmental stages, known sex genes and phenotypes, and
ethylene effects. Previous work in our lab using transgenic melons expressing the ethylene
receptor mutant etr1-1 under floral primordia-targeted promoters indicated stamen primordia, not
carpel primordia, need to perceive ethylene for carpel development. Previous research reported
that the G locus, responsible for carpel suppression, encodes a WIP1 transcription factor, and the
A locus, responsible for stamen suppression in bisexual flowers, encodes an ethylene
biosynthetic enzyme gene, 1-aminocyclopropane-1-carboxylic acid synthase (ACS). However, it
is unknown how molecular ethylene production influences carpel development promotion.
To further examine the roles of floral organ primordia in promoting carpel development,
transgenic melons were produced targeting ACS expression to either stamen and petal primordia
(AP3::ACS), or carpel and nectary primordia (CRC::ACS). AP3::ACS melons showed increased
A gene expression, and decreased G expression. Increased femaleness was observed, manifested
as increased carpel-bearing buds, decreased male buds, male-only phase loss, and gain of a
bisexual-only phase not seen in wild type. Microscopic analysis of apices showed reduced

progression of floral buds into sex determination stages. In contrast, CRC::ACS melons showed
no difference in sex expression patterns or sex gene expression. These results, coupled with
knowledge of sex gene identities and sex phenotypes, led to an integrated model of melon floral
sex determination.
Increased femaleness was also observed in transgenic melons targeting etr1-1 to carpel
and nectary primordia (CRC::etr1-1). To investigate if this phenotype is useful for increased and
earlier fruit set, CRC::etr1-1 melons were examined in the field. Transgenic plants had earlier
and increased number of carpel-bearing flowers and fruit set. However, CRC::etr1-1 fruit were
smaller, resulting in equivalent kg/plot, and showed either earlier ripening (line M5), or no
obvious external ripening (line M15). Externally green M15 fruit had extensive internal ripening
with elevated internal ethylene levels, equivalent to wild type orange fruit. Expression of etr1-1
was higher in M15 exocarp compared to mesocarp, likely leading to external ripening inhibition.
It has been proposed that, in addition to the two major sex loci, one or more modifiers act
to stabilize gynoecious and hermaphrodite genotypes. Other members of the ACS or ACC
oxidase (ACO; ethylene biosynthetic enzyme) families may modulate sex determination. To
evaluate gene expression within different sex genotypes, hermaphrodite (ggaa) and monoecious
(G-A-) lines were produced from a gynoecious (ggAA) and andromonoecious (GGaa) cross. F1
progeny were monoecious as predicted and F2 sex phenotype segregation ratios were consistent
with a four gene model. Three new ACS gene members were identified using the melon genome;
expression of the 7 ACS and 3 ACO genes was analyzed in apices of different genotypes. All
ACS and ACO members showed higher expression in gynoecious and hermaphrodite apices
compared to andromonoecious and monoecious. Together, these studies provide further insight
into ethylene perception and production influences in sex expression in melon.

Copyright by
Jessica Taft
2013

DEDICATION
I dedicate this thesis to my loving family and friends who have given me the strength and
encouragement to complete this major milestone in my life. Also to my husband, who has been
the most caring and supportive influence in my graduate student career; he’s seen me at my
worst and best times, and still married me, the crazy glutton for punishment. Love to all of you
and know that I couldn’t have done it without you.

v

ACKNOWLEDGEMENTS

I would like to thank all of the members of my committee: Drs. Cornelius Barry, Brad Day, and
Amy Iezzoni, and my PI Dr. Rebecca Grumet for their continued advice and support throughout
my projects. I also want to thank past and present members of Dr. Rebecca Grumet’s lab for your
help and friendship; you all made going to lab fun and interesting. To the graduate students, postdocs, and faculty of Joint Lab; Dr. Randy Beaudry, and Dr. Ryan Warner, thank you for your
helpful advice and letting me use your room/growth chamber to finish my experiments
(especially Randy for putting up with the melon smell). A big thanks goes out to Lilia Fernando
for all her help with setting up and taking data with me for my first field. And finally, thank you
to the staff of the Horticulture Teaching and Research Center, Plant Biology Greenhouses,
Center for Advanced Microscopy (namely Melinda Frame and Alicia Pastor), and the Research
Technology Support Facility for your support; without these facilities and the wonderful staff, I
would never have completed this research.

vi

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................................... ix
LIST OF FIGURES .........................................................................................................................x
CHAPTER 1: Literature review...................................................................................................1
Sex expression in angiosperms ..............................................................................................2
Floral bud development .........................................................................................................3
Hormone involvement in sex expression ...............................................................................6
Ethylene’s role in sex expression...........................................................................................7
Genetics of sex expression ...................................................................................................11
Proposed melon sex expression model ................................................................................16
Thesis objectives ..................................................................................................................16
Literature cited .....................................................................................................................20
CHAPTER 2: Effect of floral primordia-targeted expression of the ethylene biosynthetic
enzyme gene, ACS, on sex expression and native sex genes in melon (Cucumis melo L.) .....29
Introduction ........................................................................................................................30
Materials and methods ......................................................................................................34
Results
Gene integration and expression ..........................................................................................40
Sex expression phenotypes ..................................................................................................42
Floral bud development within the apex ..............................................................................42
Native sex gene expression ..................................................................................................46
WIP1 and ACS7 promoter elements .....................................................................................50
Discussion
Ethylene evolution and native sex genes .............................................................................50
Stamen primordia-targeted expression of ACS leads to feminization .................................52
Native sex genes ..................................................................................................................55
Ethylene threshold model for sex expression in melon .......................................................56
Appendix .............................................................................................................................59
Literature cited...................................................................................................................61
CHAPTER 3: Effect of carpel primordia-targeted inhibition of ethylene perception on sex
expression, and fruit set, development, and ripening in melon (Cucumis melo L.) ...............67
Introduction ........................................................................................................................68
Materials and methods ......................................................................................................70
Results
Ethylene and sex expression ................................................................................................73
Fruit set and yield.................................................................................................................76
Fruit development and ripening ...........................................................................................76
Discussion............................................................................................................................84
Literature cited...................................................................................................................89
vii

CHAPTER 4: Expression characterization of sex genes and ethylene biosynthetic genes in
different sex genotypes of melon (Cucumis melo L.).................................................................93
Introduction ........................................................................................................................94
Materials and methods ......................................................................................................97
Results and discussion
Segregation analysis of a gynoecious x andromonoecious cross.........................................99
Analysis of sex genes in different sex genotypes ..............................................................101
Identification of ACS and ACO gene family members ......................................................103
Analysis of ACS and ACO gene family members in different sex types ...........................105
Literature cited.................................................................................................................114
CONCLUDING REMARKS ....................................................................................................118
Literature cited.................................................................................................................122

viii

LIST OF TABLES

Table 1.1 Melon sex expression patterns with corresponding genotypes and phenotypes ..........13
Table A1 PCR segregation ratios of transgenic CRC::ACS and AP3::ACS plants .....................60
Table 3.1 Greenhouse observations of the T2 generation of CRC::etr1-1plants .........................75
Table 3.2 Field observations of fruit yield parameters ................................................................79
Table 4.1 Melon sex expression patterns with corresponding genotypes and phenotypes ..........95
Table 4.2 Primers of melon ACS and ACO gene members for qRT-PCR analysis ...................100
Table 4.3 Segregation for F2 population of gynoecious x andromonoecious cross ..................102

ix

LIST OF FIGURES

Figure 1.1 Diagram of the ABCDE model of floral development .................................................5
Figure 1.2 Progression of melon floral bud development including developmental stages, sex
phenotypes, and native sex genes ..................................................................................................17
Figure 2.1 Progression of melon floral bud development including developmental stages, native
sex genes, and sex phenotypes .......................................................................................................33
Figure 2.2 qRT-PCR analysis of ACS transgene expression and ethylene evolution in lateral
apices of Control (WT and azygous), AP3::ACS, and CRC::ACS plants ......................................41
Figure 2.3 Greenhouse observations of sex expression for WT, azygous, and CRC::ACS plants
along the first 30 nodes of the main stem, and number of bisexual and male buds ......................43
Figure 2.4 Greenhouse observations of WT, azygous, and AP3::ACS plants for sex expression
along the first 30 nodes of the main stem, number of male buds and open flowers, individual WT
and AP3::ACS A5 plants, and number of bisexual buds and open flowers ...................................44
Figure 2.5 Microscopy images of melon floral buds at different developmental stages and
percentage of floral buds at different stages of development observed in microtome-sectioned
apices of WT, azygous, and AP3::ACS under light microscopy. ..................................................47
Figure 2.6 qRT-PCR analysis of the melon sex genes WIP1 and ACS7 in apices of CRC::ACS,
AP3::ACS, and AP3::etr1-1 plants relative to expression of wild type controls ...........................48
Figure 2.7 qRT-PCR analysis of ACS7 and WIP1 expression in wild type melon apices taken at
-1 hr, 2 hr, 5 hr, 8 hr and 11 hr relative to a 10 hour light cycle, and EBF1, ACS7, and WIP1
expression in ethephon treated apices at 8 hr (light) and 11 hr (dark) post-treatment relative to
untreated controls. ..........................................................................................................................49
Figure 2.8 Diagram of the promoter of WIP1 and ACS7 with positions of putative regulatory
elements and transcription factor binding sites ..............................................................................51
Figure 2.9 Ethylene threshold model of melon floral bud development including known
developmental stages, genotypes, and sex phenotypes ..................................................................57
Figure 3.1 qRT-PCR analysis of expression of the etr1-1 transgene, and ethylene evolution in
apices of Hales (WT), Azygous (Azy), and CRC::etr1-1 plants ...................................................74
Figure 3.2 Field observations of number of carpel-bearing buds, open flowers, node of first
carpel-bearing open flower on the main stem, and percentage of female vs. bisexual open flowers
of Hales (WT), and CRC::etr1-1 ...................................................................................................77
x

Figure 3.3 Average fruit set per plant for Hales (WT), azygous, and CRC::etr1-1 plants ..........78
Figure 3.4 Average fruit and open bisexual and female flower ovary length/diameter ratio on
wild type (WT) and CRC::etr1-1 ...................................................................................................81
Figure 3.5 Percentage of harvested fruit that are visibly ripe and abscission from the vine at
final harvest ....................................................................................................................................82
Figure 3.6 Mesocarp of Hales (green and ripe), and CRC::etr1-1 M15 fruit, whole fruit ethylene
evolution, and internal fruit ethylene of wild type, azygous, and CRC::etr1-1.............................83
Figure 3.7 qRT-PCR analysis of expression of the etr1-1 transgene in 10mm diameter (~3dpp),
60mm diameter (~14dpp) and 35dpp fruit exocarp and mesocarp of CRC::etr1-1fruits ..............85
Figure 4.1 Expression of the melon sex genes WIP1 and ACS7 in monoecious (G-A-),
andromonoecious (GGaa), gynoecious (ggAA), and hermaphrodite (ggaa) ...............................104
Figure 4.2 Rooted phylogenetic tree of ACS gene family members across different species ...106
Figure 4.3 Expression of ACS and ACO gene family members in monoecious (GA),
andromonoecious (Ga), gynoecious (gA), and hermaphrodite (ga) ............................................110

xi

CHAPTER 1
Literature Review
Parts of this literature review have been published as Grumet R and Taft J (2011). “Sex
Expression in Cucurbits.” In: Genetics, Genomics and Breeding in Cucurbits. C Kole, YH Wang,
TK Behera (eds). Science Publishers Inc., Enfield, NH and CRC Press (Taylor and Francis)
group, Boca Raton, FL. P: 353-375.

1

Sex expression in angiosperms:
Typical angiosperm species produce bisexual flowers that include both male and female
organs (anthers and carpels, respectively) allowing for self fertilization. The promotion of
outcrossing and heterogeneity has evolved through different mechanisms such as recognition of
‘self’ pollen and subsequent termination of fertilization (self incompatibility); changing flower
morphology to prevent physical contact of anthers and pistil; or developing unisexual flower
phenotypes (Ainsworth, 1998; Barrett, 1998). Two major classifications of sex phenotypes are
dioecious, with unisexual flowers separated on two individual plants, and monoecious, unisexual
flowers occurring on the same plant (Grant, 1994; Ainsworth, 1998; Barrett, 1998). Dioecy has
appeared in ~43% of known plant families and monoecy in ~36% of families (Renner and
Ricklefs, 1995).
Cucurbitaceae is an interesting family with ~32% dioecious species and another 36%
exhibiting a wide variety of other heritable sex expression types that can produce varying
combinations of male, female and bisexual flowers within a single species, i.e. monoecy (Roy
and Saran, 1990; Renner and Ricklef, 1995). Some cucurbit species exhibit combinations of sex
types that can occur on a single plant, including one species that has one of the rarest forms:
androdieocy (separate male and bisexual plants) (Roy and Saran, 1990). Monoecy includes
various sex expression types: andromonoecious (male and bisexual), monoecious (male and
female), gynoecious (only female), and hermaphrodite (only bisexual). The molecular and
physiological aspects of unisexuality are not well understood, but through the use of two model
species, Cucumis melo (melon) and Cucumis sativus (cucumber), these aspects are beginning to
come to light (Roy and Saran, 1990; Perl-Treves, 1999).
The typical sex expression pattern of melon begins with vegetative nodes, followed by

2

similar growth habit, however it has a mix of male and female nodes (monoecious). Depending
on the underlying genotype, both species can exhibit other sex types such as gynoecious and
hermaphrodite. Cucumber can also be androecious (all male) (Rudich, 1990), although a stable
male-only habit has not been documented in melon (Magdum et al., 1982). These expression
types can be influenced by the internal (Bai et al., 2004; Papadopoulou et al., 2005b; Little et al.,
2007, Boualem et al., 2008; Martin et al., 2009) and external plant environment (Nitsch et al.,
1952), and exogenous application of hormones (Rudich, 1990). Taken together, any bud at a
single node can be influenced by developmental stage, environment, hormone balance, and
genetics leading to the observable sex type.

Floral bud development:
Typical floral development follows a standard pattern with initiation of four sequential
whorls of organs progressing from the outside to the inside of the flower. The outermost whorl,
whorl 1, is where sepals originate; whorl 2, petals; whorl 3, stamens; and whorl 4, carpels (Coen
and Meyerowitz, 1991). In unisexuality, all four whorls are initiated, but then arrest of either
stamen or carpel primordia development leads to female or male flowers, respectively. In
cucumber buds, this arrest occurs at different stages in floral bud development, as described by
Bai et al. (2004). Developmental stage 1 is defined as the initiation of inflorescence, while stages
2-5 are sequential initiation of the whorls, beginning with sepals and ending with carpels. This
process is completed in the first five days following inflorescence initiation while the floral bud
is still microscopic (1-2mm in size) in the apex (Goffinet, 1990). Following initiation, sexual
differentiation occurs in buds destined to be male or female unisexual flowers. In male flower
buds, carpel arrest occurs at stage 5, when stamen primordia begin to enlarge. Stamen primordial
3

enlargement continues through to stage 9 when microsporocyte development is initiated, then
further maturation occurs throughout the flower ending at stage 12 (anthesis). In female flower
buds, the carpel primordia begin to elongate at stage 6 and continues until stage 9
(macrosporocyte initiation). Stamen arrest in female buds occurs at stage 8 although DNA
damage is seen in the anther at stage 7 (Hao et al., 2003). Similar stages have also been observed
in melon floral buds (Boualem et al., 2008).
Studies using a series of homeobox mutants in the model plant Arabidopsis thaliana and
Antirrhinum (snapdragon), Coen and Meyerowitz (1991) and Bowman (1997) have elucidated
the ABC model of floral development. Floral development is controlled by a set of MADS-box
transcription factors A-E, as diagramed in Figure 1.1. Development of sepals is controlled by A
genes (APETALA1), petal; A and B genes (APETALA3), stamen; B and C genes (AGAMOUS),
and carpels require C genes. Loss of the A genes causes C genes to infiltrate the expression
domain of A and vice versa, influencing sex organ development and showing their antagonistic
nature. Further studies established the roles of the E genes, which are involved in the
development of all four whorls (Pelaz et al., 2000; Honma and Goto, 2001), and D genes, which
are specific for ovule development (Figure 1.1) (Colombo et al., 1995; Favaro et al., 2003;
Pinoyopich et al., 2003). Yeast two-hybrid studies using A, B or C proteins as bait found that
they can form homo or heterodimers (Pelaz et al., 2000), and quartet structures to control organ
development (Honma and Goto, 2001; Theiben and Saedler, 2001).
Some cucumber homeobox genes have been cloned showing similar expression profiles
as observed in the homologous Arabidopsis genes (Perl-Treves et al., 1998), and mutants
demonstrated the importance of whorl position in floral development (Kater et al., 2001). A
typical B gene mutant would show a phenotype where the petals and stamens are replaced by

4

Figure 1.1. Diagram of the ABCDE model of floral development based on research performed by
Coen and Meyerowitz (1991), Pelaz et al. (2000), Honma and Goto (2001), Colombo et al.
(1995), Favaro et al. (2003), and Pinyopich et al. (2003). For interpretation of the references to
color in this and all other figures, the reader is referred to the electronic version of this thesis (or
dissertation).

5

sepals and carpels, however the equivalent mutant in cucumber, green-petals, showed a different
floral phenotype depending on the sexual fate of the bud. Female buds did have the predicted
sepals in the first two whorls, with an undeveloped third whorl and carpels in the fourth. Buds
destined to be male had sepals-sepals-carpels and the fourth whorl failed to develop. These
phenotypes showed that whorl position, not sexual identity, was the main determinant for
developmental arrest leading to unisexuality. A C-class gene of cucumber ERAF17 was cloned
and expression correlated with female flower development. After the initiation of female buds,
ERAF17 expression was found in the apex and continued in female tissues throughout
development (Ando et al., 2001). The induction of female buds and ERAF17 expression also
seemed to be correlated with higher concentrations of the phytohormone ethylene. Recent
transcriptomic analysis of apices from monoecious and gynoecious lines identified a set of
differentially expressed genes with higher expression in gynoecious versus monoecious, such as
ACC synthase, putative MYB transcription factor, auxin-related, and ABA-related genes,
providing evidence for the role of hormones in sex determination (Gu et al., 2010; Wu et al.,
2010).

Hormone involvement in sex expression:
Multiple factors influence the decision in floral buds to arrest the preferred primordia or
allow for their continued development. Use of hormone mimics or inhibitors have been used in
the field to promote male or bisexual flowers in gynoecious cucumber and melon lines,
respectively, to allow for self pollination in order to maintain these lines (Den Nijs and Visser,
1980; Owens, 1980). Early studies in melon and cucumber investigated the effect of auxin and
gibberellins. Increases in auxin level from exogenous application studies demonstrated that it had

6

a feminizing effect, as manifested by increased number of carpel- bearing buds and flowers, or
the conversion of male flowers into bisexual/female flowers (Galun 1959; Galun et al., 1963;
Rudich et al., 1969). This finding was supported by comparisons of endogenous auxin levels
within different cucumber sex phenotypes (Galun et al., 1965; Trebitsh et al., 1987).
High levels of gibberellins (GAs) have been associated with increased maleness as
manifested as an increase in male buds and/or flowers, or the conversion of female flowers into
bisexual (Atsmon et al., 1968; Perl-Treves et al., 1998). Exogenous application of GA promoted
male buds/flowers and prevented maturation of female buds. Likewise, applications of GA
inhibitors promoted femaleness (Yin and Quinn, 1995). Observations of endogenous GA
concentrations in a monoecious cucumber line versus a gynoecious line showed that higher
levels of GA were present in monoecious plants (Atsmon et al., 1968).
A later study looked at the increased femaleness phenotype observed after brassinosteriod
(BR) application in melon, cucumber and zucchini (Papadopoulou and Grumet, 2005a).
Increasing concentrations of BR applied to monoecious cucumber showed earlier onset and
increased number of female buds, however there was no effect on gynoecious cucumber,
monoecious zucchini, or andromonoecious melon. Ethylene evolution measurements taken from
apices of cucumber and zucchini after application of BR showed an increase in ethylene levels in
both monoecious and gynoecious lines.

Ethylene’s role in sex expression:
Application of ethylene using ethephon or ethrel has a feminizing effect on melon and
cucumber plants (McMurray and Miller, 1968; Robinson et al., 1969; Rudich et al., 1969; Karchi,
1970; Augustine et al., 1973; Owens et al., 1980). Application at different developmental time
7

points, especially early on (i.e. at the one or two-leaf stage) can cause earlier and increased
carpel-bearing buds in monoecious and andromonoecious melon plants (Rudich et al., 1969;
Karchi, 1970). Likewise, 2-3 fold higher endogenous ethylene levels were found in gynoecious
cucumber apices compared to monoecious, however this was not observed in melon (Byers et al.,
1972a; Rudich et al., 1972; Makus et al., 1975; Rudich et al., 1976; Trebitsh et al., 1987;
Yamasaki et al., 2001, 2003). In melon, studies using hypobaric gas chambers demonstrated the
importance of endogenous ethylene levels (Byers et al., 1972b). After plants were subjected to a
reduction of internal ethylene gas within the chambers, they exhibited increased maleness that
could be reversed by application of ethylene.
The biosynthesis of ethylene begins with S-Adenosylmethionine (S-Adomet) being
converted to 1-aminocyclopropane-1-carboxylate (ACC) by the rate-limiting enzyme ACC
synthase (ACS). ACC is then oxidized to ethylene by ACC oxidase (ACO) (Johnson and Ecker,
1998; Argueso et al., 2007). Transgenic melons constitutively expressing ACS under the 35S
promoter from the Cauliflower Mosaic virus (CaMV) showed increased endogenous ethylene
levels and an increase in femaleness as exhibited by increased and earlier onset of carpel-bearing
flowers (Papadopoulou et al., 2005b). Expression of AP3::CsACO2 in Arabidopsis led to
defective stamen development, potentially linking ACO2 to stamen inhibition (Duan et al., 2008).
Inhibition of ethylene perception via silver nitrate/thiosulfate or ethylene production by
aminoethoxyvinylglycine (AVG) application leads to increased maleness in cucumber and melon
(Byers et al., 1972a; Den Nijs and Visser, 1980; Owens et al., 1980). Yamasaki and Manabe
(2011) observed that functional bisexual and male flowers were obtained from application of
silver nitrate on gynoecious cucumber, however AVG only induced male flowers. AVG
converted female flowers to males independent of concentration, while higher concentrations of

8

silver nitrate were required to bring about the same conversion. This demonstrated that inhibition
of ethylene production had a stronger effect on sex expression compared to blocking ethylene
perception in cucumber (Yamasaki and Manabe, 2011).
Although GA, auxin, and BR had an effect on sex expression in cucumber and, to a lesser
extent, melon, the effect of these hormones was not as strong as exogenous ethylene producers
and inhibitors, especially for melon and gynoecious cucumbers (Kubicki et al., 1969a; Byers et
al., 1972a; Tolla and Peterson, 1979; Trebitsh et al., 1987; Yin and Quinn, 1995). Inhibitors of
GA increased maleness in cucumber, however ethylene inhibitors were more effective (Kubicki
et al., 1969b; Tolla and Peterson, 1979). The increase in femaleness observed in monoecious
cucumbers following auxin and BR application could be a product of these hormones’ induction
of ethylene production (Trebitsh et al., 1987; Papadopoulou et al., 2005a). Application of GA
and BR was unable to influence sex expression in melon, however exogenous ethylene increased
femaleness while inhibitors increased maleness (Byers et al., 1972a; Den Nijs and Visser, 1980;
Owens et al., 1980). Based on this evidence, ethylene has been proposed as the predominant sex
hormone in melon and cucumber (Rudich, 1990).
Ethylene is involved in many different processes within plants including senescence,
abscission, pathogen responses, and responses to abiotic stresses (Abeles et al., 1992; Roman et
al., 1995; O’Donnell et al., 1996). It is also involved in developmental processes such as fruit
ripening, apical hook formation, and flower development (Abeles et al., 1992; Barry and
Giovannoni, 2007; Grumet and Taft, 2011). The ethylene perception pathway has been
extensively studied in Arabidopsis using mutants that show different responses to ethylene
(Bleeker and Kende, 2000; Steptanova and Ecker, 2000; Wang et al., 2002). Ethylene receptors
are encoded by a multigene family that resembles bacterial two-component signal transduction

9

proteins including kinase and receiver domains (Bleeker et al., 1988; Chen et al., 2005).
Perception of ethylene produces a response by inhibiting phosphorylation of the immediate
downstream component, CTR1 protein, which blocks the response pathway in the absence of
perceived ethylene (Chen et al., 2005; Kendrick and Chang, 2008). Homologs of genes involved
in the signaling pathway have been cloned in both cucumber and melon, and expression
examined in cucumber in relation to sex expression (Yamasaki et al., 2000).
Increased expression of the receptor genes CsETR1, CsETR2 and CsERS was observed in
gynoecious cucumber apices compared to monoecious, and following ethylene application
(Yamasaki et al., 2000, 2001). CsETR2 and CsERS showed stronger expression upon onset of
femaleness and in situ hybridization studies localized CsETR2 to the carpel region of female
buds, providing evidence for receptors in sex determination (Yamasaki et al., 2003). Results
from heterologous expression in Arabidopsis of the antisense cucumber homologue of ETR1,
under the control of the AP3 promoter resulted in inhibition of stamen development (Wang et al.,
2010).
Homologs of Arabidopsis ethylene receptor genes have been identified in melon
(CmETR1, CmETR2 and CmERS1) fruit (Sato-Nara et al., 1999; Takahashi et al., 2002; Owino et
al., 2007). A link between gene expression and sex types has yet to be determined, however work
with transgenics has provided evidence for ethylene’s role in melon sex expression. Constitutive
expression of an Arabidopsis dominant negative ethylene receptor mutant, etr1-1, in transgenic
melons showed an almost complete loss of carpel-bearing buds, demonstrating that ethylene
perception is critical for carpel development (Little et al., 2007). In order to further elucidate
where perception is critical for sex determination, promoters that direct etr1-1 expression to the
stamen and petals (APETALA3 or AP3) (Irish and Yamamoto, 1995), and the carpel and nectaries

10

(CRABS CLAW or CRC) (Bowman and Smyth, 1999) were used. If perception in the carpels was
required for carpel development, then inhibiting it would show a loss of carpel-bearing buds
while stamen inhibition of perception would show no change in expression. However,
CRC::etr1-1 melons did not support this hypothesis; instead an increase in femaleness was
observed. Conversely, AP3::etr1-1 melons showed an almost complete loss of carpel-bearing
buds. These results provide evidence that ethylene perception in the stamen primordia is required
for carpel development, indicated that stamens are a regulator of carpels (Little et al., 2007).

Genetics of sex expression:
While ethylene is a key regulator of melon sex expression, genetics is also has a
significant role in sex determination. Cucumber sex expression involves two major loci, Female
(F) and Monoecious (M), along with a few minor loci (Perl- Treves, 1999). The M locus acts to
inhibit stamens in carpel-bearing flowers and was found to be the ethylene biosynthetic enzyme
ACC synthase, namely CsACS2, which localizes to the ovary in carpel-bearing buds (Kamachi et
al., 1997; Yamasaki et al., 2003; Saito et al., 2007; Boualem et al., 2009, Li et al., 2009).
Sequence analysis of CsACS2 in monoecious (MMffAA), gynoecious (MMFFAA), androecious
(MMffaa), and hermaphrodite (mmFFAA) lines pointed to a single nucleotide polymorphism
(SNP) causing an amino acid change linked to the M locus (Boualem et al., 2009). Further
analysis of 28 cucumber accessions revealed a second SNP associated with the M locus in
hermaphrodite and andromonoecious lines. In a separate study, a third SNP causing a reduction
in enzymatic activity was identified within CsACS2 that also showed a link to the M locus (Li et
al., 2009). Protein structure analysis revealed that all three mutations occurred in or near the
catalytic site where SAM helps with positioning the pyridoxal 5’- phosphate (PLP) cofactor,
11

causing a reduction or loss of enzymatic activity (Huai et al., 2001; Rottman et al., 1991). This
reduction in ethylene production results in failure to inhibit stamen primordia development in
bisexual flowers.
The F locus, a promoter of carpel development, was mapped to an additional copy of
CsACS1, namely ACS1G which is the result of a gene duplication and a recombination event
involving a branched- chain amino acid transaminase gene (Trebitsh et al., 1997; Knopf and
Trebitsh, 2006). This extra ACS can account for the 2-3 fold higher ethylene levels observed in
gynoecious cucumbers (Rudich et al., 1972; Makus et al., 1975; Rudich et al., 1976; Trebitsh et
al., 1987; Yamasaki et al., 2001, 2003). Studies of ACO gene expression following ethrel
application showed increased expression of ACO3 in apices (Kahana et al., 1999). ACO3
expression was stronger in the developing stamen and carpel primordia of female buds, however
response to application was dependant on whether the plant was gynoecious or androecious (only
male flowers).
A minor locus, Androecious (A), has been linked to increased maleness in cucumbers as
aaff plants fail to make carpel-bearing buds resulting in an androecious phenotype (Rudich et al.,
1976). A series of other minor loci have been identified including In-F, gy, m-2 and Tr (Kubicki
1969b; Malepszy and Niemirowicz-Szczytt, 1991; Robinson and Decker-Walters, 1997). The m2 (andromonoecious-2) gene is linked to development of bisexual flowers with normal ovary
formation, while to the co-dominant Tr (Trimonoecious) allele allows carpel formation in male
buds (becoming bisexual) with irregular ovaries (Malepszy and Niemirowicz-Szcytt, 1991;
Robinson and Decker-Walters, 1997). In-F (Intensifier-Female) and gy (gynoecious) are both
involved in enhanced femaleness in monoecious/gynoecious cucumber (Kubicki 1969b;
Malepszy and Niemirowicz-Szcytt, 1991). Two new modifiers of femaleness, Mod-F1 and mod-

12

Table 1.1. Melon sex expression patterns with corresponding genotypes and phenotypes.
Sex expression pattern

Genotype

Phenotype

Andromonoecious

Ga

♂

Monoecious

GA

♂ ♀ - Male and female flowers

Gynoecious

gA

♀ - Female flowers only

Hermaphrodite

ga

- Male and bisexual flowers

- Bisexual flowers only

13

F2 were reported by Chen et al. (2011) to increase gynoecy independently of F in monoecious
and hermaphrodite lines.
Analogous to cucumber, melon sex expression has two major loci controlling sex
determination, Andromonoecious (A) and Gynoecious (G), along with one to two minor loci
(Perl-Treves, 1999; Grumet and Taft, 2011). Table 1.1 outlines the different sex expression
patterns in melon along with their corresponding genotype and flower phenotype. The A locus is
similar to the M locus in cucumber which is involved in stamen inhibition, and found to be a new
member of the ACS family, CmACS7 (Boualem et al., 2008). Expression of CmACS7 is localized
to the carpel primordia of carpel-bearing buds at stage 7. Dominant A plants with a fully
functional ACS7 enzyme inhibits stamens resulting in female flowers. In contrast, the recessive a
allele has a missense mutation leading to a 50% reduction in enzymatic function from analysis of
a segregating F2 population from a cross between a monoecious (AAGG) and andromonoecious
(aaGG) parents. Similar to the M locus SNPs, this missense mutation was found within the
active site of ACS7, leading to the observed reduction in ethylene production and prevention of
stamen inhibition in bisexual flowers.
The G locus, which is involved in carpel inhibition, was identified as the Wound
Inducible Protein1 (WIP1) transcription factor by positional cloning from a segregating F2
population from a monoecious (GGAA) and gynoecious (ggAA) cross (Martin et al., 2009).
Plants with recessive g were found to have a transposon from the hAT family (Gyno-hAT)
inserted in the 1.4kb non-coding region upstream of CmWIP1. The transposon insertion leads to
the downstream methylation of CmWIP1 inhibiting expression, resulting in the loss of carpel
inhibition. Dominant G plants lack this insertion and have separate staminate flowers. CmWIP1
localizes to the carpels of buds destined to be male, thus expression causes inhibition of carpels
14

in developing male buds. The identity of the G locus as neither an ethylene biosynthetic nor a
perception gene, corresponds to earlier observations for the lack of a difference in ethylene levels
in apices of gynoecious melon compared to monoecious (Byers et al., 1972a).
Interestingly, when localization of CmACS7 was studied, it was found to reside in carpel
primordia at Stages 4 and 7, and yet it influences stamen development (Boualem et al., 2008).
This suggests that crosstalk between the stamen and carpel primordia is occurs during
development. The AP3::etr1-1 results show stamen-mediated inhibition of carpel development,
also suggest a role for inter-whorl regulation during sex determination (Little et al., 2007). An
epigenetic change of CmWIP1 in melon leads to carpel inhibition, however a link to ethylene has
yet to be discovered as yeast one-hybrid studies did not show an interaction between the
CmACS7 promoter and CmWIP1 protein (Martin et al., 2009). The idea of ethylene influencing
the G locus is valid, as studies that blocked ethylene perception or biosynthesis with AVG and
silver nitrate/thiosulfate caused increased maleness to occur in dominant G genotypes (Owens et
al., 1980). Gene expression changes that occur within the stamen primordia after ethylene
perception is unknown.
Transcriptomic gene expression studies of female and bisexual flowers in cucumber have
provided information on candidates involved in carpel-bearing flower development, including
known ethylene and other hormone genes, and transcription factor genes (Guo et al., 2010; Wu et
al., 2010). While this information is a valuable resource in studying flower development, only
later developmental stages, i.e. stage 8 and beyond, were analyzed, leaving gene expression
changes at differentiation still unknown.

15

Proposed melon sex expression model:
A proposed model for sex determination in melon based on current knowledge is shown
in Figure 1.2, and begins with the typical bisexual floral bud before developmental stage 5. The
first decision point, whether to inhibit carpel development, occurs between developmental stages
5 and 6 in the floral bud (Bai et al., 2004). Previous studies suggest that if ethylene is not
perceived in the stamen primordia, carpel development is inhibited (Little et al., 2007). Upon
ethylene perception in the stamen primordia, the carpel primordia will continue developing.
Based on this proposed model, increasing endogenous ethylene concentrations perceived by the
stamen primordia potentially leads to increased femaleness. The G locus acts between stages 5
and 6 (Martin et al., 2009), however there is no current evidence for a link between ethylene and
G locus expression.
The second decision point in melon sex expression, whether to inhibit stamens, occurs
between developmental stages 6 and 7 (Bai et al., 2004). The A locus (or M locus in cucumber)
acts between stages 6 and 7 where increased ethylene produced in the carpels by CmACS7 (or
CsACS2) action leads to the inhibition of stamens in female flowers (Boualem et al., 2008, 2009;
Li et al., 2009). The presence of a mobile signaling molecule is potentially acting to cause this
regulation, which has yet to be discovered. It is unknown whether there is further involvement of
endogenous ethylene biosynthesis, perception and related genes in the melon sex determination
process.

Thesis objectives:
The overall objective for this study was to develop a comprehensive melon sex
expression model that encompasses known phenotypes, genotypes, developmental stages,
16

Figure 1.2. Progression of melon floral bud development including developmental stages (Bai et al., 2004), sex phenotypes (Goffinet,
1990), and native sex genes (Boualem et al., 2008; Martin et al., 2009).

17

ethylene influences, and gene expression. The previous etr1-1 work provided evidence for the
importance of ethylene perception in the stamens for carpel development (Little et al., 2007).
However, the location of ethylene production that is perceived by the stamens is unknown.
Utilization of transgenic melons targeting ethylene production to the stamen or carpel primordia
will provide insight into the influence on sex expression. Based on the perception data, increased
ethylene production in the stamens should show an increased femaleness phenotype if the
stamens are major players in carpel development.
Increased femaleness observed in transgenic melons with ethylene perception inhibition
targeted to the carpel primordia has been observed under greenhouse conditions (Little et al.,
2007). This increased femaleness was manifested by increased and earlier carpel-bearing buds
and flowers, and a ~50% conversion of bisexual-to-female flowers. Earlier and increased carpelbearing flowers have a possible application in the field leading to earlier and increased fruit set.
Two lines were observed in the field to determine the effect of carpel primordia-targeted
ethylene perception on sex expression, and fruit set, development and ripening.
The two major sex loci of melon have been identified, however gene models exist where
one or two minor loci are involved in bisexual development (Kubicki, 1969b; Kenigsbuch and
Cohen, 1990). It is possible that one of these minor loci could be another member of the ethylene
biosynthetic pathway or serve as a link to the interaction between ethylene and WIP1, and could
be mediated by another ethylene biosynthetic enzyme. Melon has three ACO and four ACS
members that have been published (Miki et al., 1995; Lasserre et al., 1996, 1997; Ishiki et al.,
2000; Boualem et al., 2008), however only the major sex loci ACS7 and WIP1 have had their
expression thoroughly studied (Boualem et al., 2008; Martin et al., 2009). The other members
have been characterized in vegetative or fruit tissue, but not apices, buds or flowers where sex

18

determination occurs (Miki et al., 1995; Lasserre et al., 1997; Ishiki et al., 2000).
Characterization of all of the members of ACO and ACS gene expression in andromonoecious,
monoecious, gynoecious and hermaphrodite melon genotypes may elucidate one of these minor
loci, and with the newly sequenced melon genome (Garcia-Mas et al., 2012), new members can
be identified.

19

LITERATURE CITED

20

LITERATURE CITED

Abeles FB, Morgan PW, Saltveit MEJ (1992). Ethylene in plant biology. Academic Press Inc.,
New York.
Ainsworth C, Parker J, Buchanan-Wollaston V (1998). Sex determination in plants. In Curr Top
Dev Biol 38: 167-223.
Ando S, Sato Y, Kamachi S, Sakai S (2001). Isolation of a MADS-box gene (ERA17) 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.
Argueso CT, Hansen M, Kieber JJ (2007). Regulation of ethylene biosynthesis. J Plant Growth
Regul 26: 92-105.
Atsmon D, Lang A, Light EN (1968). Contents and recovery of gibberellins in monoecious and
gynoecious cucumber plants. Plant Physiol 43: 806-810.
Augustine JJ, Baker LR, Sell HM (1973). Female flower induction on androecious cucumber,
Cucumis sativus L. J Am Soc Hort Sci 98: 197-199.
Bai SL, Peng YB, Cui JX, Gu HT, Xu LY, Li YQ, Xu ZH, 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 (1998). The evolution of plant sexual diversity. Nat Rev 3: 274-284.
Barry CS and Giovannoni JJ (2007). Ethylene and fruit ripening. J Plant Growth Regul 26: 143159.
Bleeker AB, Estelle MA, Somerville C, 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 Bi 16: 1-18.
Boualem A, Fergany M, Fernandez R, Troadec C, Martin A, Morin H, Sari M, Collin F, Flowers
J, Pitrat M et al. (2008). A conserved mutation in an ethylene biosynthesis enzyme leads to
andromonoecy in melons. Science 321: 836-838.
Boualem A, Troadec C, Kovalski I, Sari M, Perl-Treves R, Bendhamane A (2009). A conserved
ethylene biosynthesis enzyme leads to andromonoecy in two Cucumis species. PLoS ONE 4:
e6144.

21

Bowman JL (1997). Evolutionary conservation of angiosperm flower development at the
molecular and genetic levels. J Bioscience 22: 515-527.
Bowman JL and Smyth DR (1999). CRABS CLAW, 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 RE, Baker LR, Dilley DR, Sell HM (1972a). Chemical induction of perfect flowers on a
gynoecious line of muskmelon, Cucumis melo L. HortScience 9: 321-331.
Byers RE, Baker LR, Sell HM, Herner RC, Dilley DR (1972b). Ethylene: a natural regulator of
sex expression of Cucumis melo L. P Natl Acad Sci 69: 717-720.
Chen YF, Etheridge N, Schaller GE (2005). Ethylene signal transduction. Ann Bot 95: 901-915.
Chen H, Tian Y, Lu XY, Lu XH (2011). The inheritance of two novel subgynoecious genes in
cucumber (Cucumis sativus L.). Sci Hortic-Amsterdam 127: 464-467.
Coen HS and Meyerowitz EM (1991). The war of the whorls: Genetic interactions controlling
flower development. Nature 353: 31-37.
Colombo L, Franken J, Koetje E, Vanwent J, Dons HJM, Angenent GC, Vantunen AJ (1995).
The petunia MADS box gene FBP11 determines ovule identity. Plant Cell 7: 1859-1868.
Den Nijs A and Visser D (1980). Induction of male flowering in gynoecious cucumbers
(Cucumis sativus L.) by silver ions. Euphytica 29: 237-280.
Duan QH, Wang DH, Xu ZH, Bai SN (2008). Stamen development in Arabidopsis is arrested by
organ-specific overexpression of a cucumber ethylene synthesis gene CsACO2. Planta 228: 537543.
Favaro R, Pinyopich A, Battaglia R, Kooiler M, Borghi L, Ditta G, Yanofsky MF, Kater MM,
Colombo L (2003). MADS-box protein complexes control carpel and ovule development in
Arabidopsis. Plant Cell 15: 2603-2611.
Galun E (1959). The role of auxin in the sex expression of cucumber. Physiol Plant 12: 48-61.
Galun E, Yung Y, Lang A (1963). Morphogenesis of floral buds of cucumber cultured in vitro.
Dev Biol 6: 370-387.
Galun E, Ishar S, Atsmon D (1965). Determination of relative auxin content in hermaphrodite
and andromonoecious Cucumis sativus L. Plant Physiol 40: 321-326.
Garcia-Mas J, Benjak A, Sanseverino W, Bourgeois M, Mir G, Gonzalez VM, Henaff E, Camara
F, Cozzuto L et al. (2012). The genome of melon (Cucumis melo L.). P Natl Acad Sci 109:
11872-11877.
22

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, C Jeffery). Cornell
University Press, New York: 288-304.
Grant S, Houben A, Vyskot B, Siroky J, Pan WH, Macas J, Saedler H (1994). Genetics of sex
determination in flowering plants. Develop Genet 15: 214-230.
Grumet R and Taft J (2011). Sex expression in cucurbits. In: Genetics, Genomics and Breeding
in Cucurbits. (eds) C Kole, YH Wang, TK Behera. Science Publishers Inc., Enfield and CRC
Press (Taylor and Francis) group, Boca Raton: 353-375.
Guo S, Zheng Y, Joung JG, Liu S, Zhang Z, Crasta OR, Sobral BW, Xu Y, Huang S, Fei Z
(2010). Transcriptome sequencing and comparative analysis of cucumber flowers with different
sex types. BMC Genomics 11: 384-397.
Hao Y, Wang D, Peng Y, Bai S, Xu L, Li Y, Xu Z, 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.
Honma T and Goto K (2001). Complexes of MADS-box proteins are sufficient to convert leaves
into floral organs. Nature 409: 525-529.
Huai Q, Xia Y, Chen Y, Callahan B, Li N, Ke H (2001). Crystal structure of 1aminocyclopropane-1-carboxylate (ACC) synthase in complex with aminoethoxyvinylglycine
and pyridoxal-5’-phosphate provide new insight into catalytic mechanisms. J Biol Chem 276:
38210-38216.
Huang S, Li R, Zhang Z, Li L, Gu X, Fan W, Lucas WJ, Wang X, Xie B, Ni P et al. (2009). The
genome of cucumber, Cucumis sativus L. Nat Genet 41: 1275-U29.
Irish VF and Yamamoto YT (1995). Conservation of floral homeotic gene function between
Arabidopsis and Antirrhinum. Plant Cell Reprod 9: 160-164.
Johnson PR and Ecker JR (1998). The ethylene gas signal transduction pathway: A molecular
perspective. Annu Rev Genet 32: 227-254.
Ishiki Y, Oda A, Yaegashi Y, Orihara Y, Arai T, Hirabayashi T, Nakagawa H, Sato T (2000).
Cloning of an auxin-responsive 1-aminocyclopropane-1-carboxylate synthase gene (CMe-ACS2)
from melon and the expression of ACS genes in etiolated melon seedlings and melon fruits. Plant
Sci 159: 173-181.
Kamachi S, Sekimoto H, Kondo N, Sakai S (1997). Cloning of a cDNA for a 1aminocyclopropane-1-carboxylate synthase that is expressed during development of female
flowers at the apices of Cucumis sativus L. Plant Cell Physiol 38: 1197-1206.

23

Kahana A, Silberstein L, Kessler N, Goldstein RS, Perl-Treves R (1999). Expression of ACC
oxidase genes differs among sex genotypes and sex phase in cucumber. Plant Mol Biol 41: 517528.
Karchi Z (1970). Effects of 2-chloroethanephosphonic acid on flower types and flowering
sequences in muskmelon. J Am Soc Hort Sci 95: 575-578.
Kater MM, Franken J, Carney KJ, Clombo L, Angenent GC (2001). Sex determination in the
monoecious species of cucumber is confined to specific floral whorls. Plant Cell 13: 481-493.
Kendrick MD and Chang C (2008). Ethylene signaling: new levels of complexity and regulation.
Curr Opin Plant Biol 11: 479-485.
Kenigsbuch D and Cohen Y (1989). The inheritance of gynoecy in muskmelon. Genome 33:
317-320.
Knopf R and Trebitsh R (2006). The female-specific CsACS1G gene of cucumber. A case of
gene duplication and recombination between the non-sex specific 1-aminocyclopropane-1carboxylate synthase gene and a branched chain amino acid transaminase gene. Plant Cell
Physiol 47: 1217-1228.
Kubicki B (1969a). Sex determination in muskmelon (Cucumis melo L.). Genet Pol 10: 145-163.
Kubicki B (1969b). Investigations on sex determination in cucumber (Cucumis sativus L.). VI
Androecium. Genet Pol 10: 87-99.
Lasserre E, Bouquin T, Hernandez JA, Bull J, Pech JC, Balague C (1996). Structure and
expression of three genes encoding ACC oxidase homologs from melon (Cucumis melo L.). Mol
Genet Genomics 251: 81-90.
Lasserre E, Godard F, Bouquin T, Hernandez JA, Pech JC, Rody D, Balague C (1997).
Differential activation of two ACC oxidase gene promoters from melon during plant
development and in response to pathogen attack. Mol Genet Genomics 256: 211-222.
Li Z, Huang S, Liu S, Pan J, Zhang Z, Tao Q, Shi Q, Jia Z, Zhang W, Chen H et al., (2009).
Molecular isolation of the M gene suggests that a conserved-residue conversion induces the
formation of bisexual flowers in cucumber plants. Genetics 182: 1381-1385.
Little HA, Papadopoulou E, Hammar SA, Grumet R (2007). The influence of ethylene
perception on sex expression in transgenic melon as assessed by expression of the mutant
ethylene receptor gene, At-etr1-1, under control of constitutive and floral-targeted promoters.
Sex Plant Reprod 20: 123-136.
Magdum MB, Shinde NN, Seshadri VS (1982). Androecious sex form in muskmelon. Cucurbit
Genet Coop 5: 24-25.

24

Makus DJ, Pharr DM, Lower RL (1975). Some morphogenic differences between monoecious
and gynoecious cucumber seedlings as related to ethylene production. Plant Physiol 55: 352-355.
Malepszy S and Niemirowicz-Szczytt K (1991). Sex determination in cucumber (Cucumis
sativus) as a model system for molecular biology. Plant Sci 80: 39-47.
Martin A, Troadec C, Boualem A, Mazen R, Fernandez R, Morin H, Pitrat M, Dogimont C,
Bendahamane A (2009). A transposon-induced epigenetic change leads to sex determination in
melon. Nature 461: 1135-1139.
McMurray AL and Miller CH (1968). Cucumber sex expression modified by 2-chloroethane
phosphonic acid. Science 162: 1397-1398.
Miki T, Yamamoto M, Nakagawa H, Ogura N, Mori H, Imaseki H, Sato T (1995). Nucleotidesequence of a cDNA for 1-aminocyclopropane-1-carboxylate synthase from melon fruits. Plant
Physiol 107: 297-298.
Nitsch JP, Kurtz EB, Liverman JL, Went FW (1952). The development of sex expression in
cucurbit flowers. Am J Bot 39: 32-43.
O’Donnell PJ, Calvert C, Atzorn R, Wasternack C, Leyser HMO, Bowles D (1996). Ethylene as
a signal mediating the wound response of tomato plants. Science 274: 1914-1917.
Owens KW, Peterson CE, Tolla GE (1980). Production of hermaphrodite flowers on gynoecious
muskmelon by silver nitrate and aminoethyoxyvinylglycine. HortScience 15: 654-655.
Owino WO, Shoji T, Ezura H (2007). Isolation and characterization of an ethylene receptor
homolog gene, Cm-ETR2, from melon fruit. Acta Hortic 731: 427-431.
Papadopoulou E and Grumet R (2005a). Brassinosteriod-induced femaleness in cucumber and
relationship to ethylene production. HortScience 40: 1763-1767.
Papadopoulou E, Little HA, Hammar SA, Grumet R (2005b). Effect of modified endogenous
ethylene production on hermaphrodite flower development and fruit production in melon
(Cucumis melo L.). Sex Plant Reprod 18: 131-142.
Pelaz S, Ditta GS, Baumann E, Wisman E, Yanofsky MF (2001). B and C floral organ identity
functions require SEPALLATA MADS-box genes. Nature 405: 200-203.
Perl-Treves R, Kahana A, Roseman N, Yu X, Silberstein L (1998). Cloning of three cDNAs
encoding 1-aminocyclopropane-1-carboxylate oxidases from cucumber floral buds. Plant Physiol
116: 1192.
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. CC
Ainsworth (eds). BIOS Scientific, Oxford: 189-215.
25

Pinoyopich A, Ditta GS, Savidge B, Liljegren SJ, Baumann E, Wisman E, Yanofsky MF (2003).
Assessing the redundancy of MADS-box genes during carpel and ovule development. Nature
424: 85-88.
Renner SS and Ricklefs RE (1995). Dioecy and its correlates in the flowering plants. Am J Bot
82: 596-606.
Robinson RW, Shannon S, DeLaGuardia MD (1969). Regulation of sex expression in the
cucumber. BioScience 19: 141-142.
Robinson RW and Decker-Walters DS (eds). (1997). Cucurbits. CAB International, New York:
9-10, 17-19, 195-198.
Roman G, Lubarsky B, Kieber JJ, Rothenberg M, Ecker JR (1995). Genetic analysis of ethylene
signal transduction in Arabidopsis thaliana – 5 mutant loci integrated into a stress response
pathway. Genetics 139: 1393-1409.
Rottman W, Peter G, Oeller P, Keller J, Shen N, Nagy B, Taylor L, Campbell A, Theologis A
(1991). 1-aminocyclopropane-1-carboxylate synthase in tomato is encoded by a multigene
family whose transcription is induced during fruit and floral senescence. J Mol Biol 222: 937961.
Roy RP and Saran S (1990). Sex expression in the Cucurbitaceae. In: Biology and utilization of
the Cucurbitaceae. DM Bates, RW Robinson, C Jeffrey (eds). Cornell University Press, Ithaca:
251-268.
Rudich J, Halevy AH, Kedar N (1969). Increase in femaleness of three cucurbits by treatment
with ethrel, an ethylene-releasing compound. Planta 86: 69-76.
Rudich J, Halevy AH, Kedar N (1972). Ethylene evolution from cucumber plants as related to
sex expression. Plant Physiol 49: 998-999.
Rudich J, Baker LR, Scott JW, Sell HM (1976). Phenotypic stability and ethylene evolution in
androecious cucumber. J Am Soc Hort Sci 181: 48-51.
Rudich J (1990). Biochemical aspects of hormonal regulation of sex expression in cucurbits. In:
Biology and utilization of the Cucurbitaceae. DM Bates, RW Robinson, C Jeffrey (eds). Cornell
University Press, Ithaca: 288-304.
Saito S, Fujii N, Miyazawa Y, Yamasaki S, Matsuura S, Mizusawa H, Fuijta Y, Takahashi H
(2007). Correlation between development of female flower buds and expression of the CS-ACS2
gene in cucumber plants. J Exp Bot 58: 2897-2907.

26

Sato-Nara K, Yuhashi K, Higashi K, Hosoya K, Kubota M, Ezura H (1999). Stage- and tissuespecific expression of ethylene receptor homolog genes during fruit development in muskmelon.
Plant Physiol 120: 321-330.
Steptanova AN and Ecker JR (2000). Ethylene signaling: from mutants to molecules. Curr Opin
Plant Biol 3: 353-360.
Takahashi H, Kobayashi T, Sato-Nara K, Tomita K, Ezura H (2002). Detection of ethylene
receptor protein Cm-ERS1 during fruit development in melon (Cucumis melo L.). J Exp Bot 53:
415-422.
Theiben G and Saedler H (2001). Floral quartets. Nature 409: 469-471.
Tolla GE and Peterson CE (1979). Comparison of gibberellins A4/A7 and silver nitrate for
induction of staminate flowers in a gynoecious cucumber line. HortScience 14: S61-S71.
Trebitsh T, Rudich J, Rivoc J (1987). Auxin, biosynthesis of ethylene and sex expression in
cucumber (Cucumis sativus). Plant Growth Regul 5: 105-113.
Trebitsh T, Staub JE, O’Neill SD (1997). Identification of a 1-aminocyclopropane-1-carboxylic
acid synthase gene linked to Female (F) locus that enhances female sex expression in cucumber.
Plant Phyiol 113: 987-995.
Wang KLC, Li H, Ecker JR (2002). Ethylene biosynthesis and signaling networks. Plant Cell
S131-S151.
Wang DH, Li F, Duan QH, Han T, Xu ZH, Bai SN (2010). Ethylene perception is involved in
female cucumber flower development. Plant J 61: 862-872.
Wu T, Qin Z, Zhou X, Feng Z, Du Y (2010). Transcriptome profile analysis of floral sex
determination in cucumber. J Plant Physiol 167: 905-913.
Yamasaki S, Fujii N, Takahashi H (2000). The ethylene-regulated expression of CS-ETR2 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, Takahashi H (2001). The M locus and ethylenecontrolled sex determination in andromonoecious cucumber plants. Plant Cell Physiol 42: 608619.
Yamasaki S, Fujii N, Takahashi H (2003). Characterization of ethylene effects on sex
determination in cucumber plants. Sex Plant Reprod 16: 103-111.
Yamasaki S and Manabe K (2011). Application of silver nitrate induces functional bisexual
flowers in gynoecious cucumber plants (Cucumis sativus L.). J Japan Soc Hort Sci 80: 66-75.

27

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

28

CHAPTER 2
Effect of floral primordia-targeted expression of the ethylene biosynthetic enzyme gene,
ACS, on sex expression and native sex genes in melon (Cucumis melo L.)

29

Introduction:
Typical angiosperm species produce bisexual flowers that include both male and female
organs (anthers and carpels, respectively). The Cucurbitaceae family, including melon (Cucumis
melon L.) and cucumber (Cucumis sativus), is exceptional with a wide variety of heritable sex
expression types that can produce varying combinations of male, female and/or bisexual flowers
within a single species (Roy and Saran, 1990; Perl-Treves, 1999; Grumet and Taft, 2011). Melon
plants are typically andromonoecious, starting with vegetative nodes followed by a male floweronly phase, then a male and bisexual flower phase. Many factors influence the sex expression of
a developing floral bud, including developmental stage, hormones, genetics and environment.
Developing cucumber floral buds go through a series of stages including carpel or stamen
primordia arrest in male and female buds, respectively (Bai et al., 2004). Inflorescence initiation
and sequential initiation of the four floral organ whorls (sepals, petals, stamens, and carpels)
occurs during stages 1-5. Carpels are arrested in males at bud stage 6. In females, DNA
degradation was observed in anther primordia at bud stage 7 and complete arrest of stamen
development was seen at bud stage 8 (Hao et al., 2003; Bai et al., 2004).
Hormones are key factors in melon sex determination, with ethylene playing the
predominant role (Kenigsbuch and Cohen, 1989; Rudich, 1990). The biosynthesis of ethylene
begins with the conversion of S-Adenosylmethionine (S-Adomet) to 1-aminocyclopropane-1carboxylate (ACC) by the rate-limiting enzyme ACC synthase (ACS). ACC is then oxidized to
ethylene by ACC oxidase (ACO) (Johnson and Ecker, 1998). The ACS and ACO gene families
have several members, with at least three members of each in melon (Miki et al., 1995; Lasserre
et al., 1996; Ishiki et al., 2000; Boualem et al., 2008). Application of ethylene via ethephon has a
feminizing effect on melon and cucumber plants that is observed as an increase in total number

30

and/or an earlier onset of carpel-bearing buds/flowers (i.e. conversion of male nodes to carpelbearing nodes) and/or a conversion of bisexual to female buds (McMurray and Miller, 1968;
Robinson et al., 1969; Rudich et al., 1969; Karchi, 1970; Augustine et al., 1973; Owens et al.,
1980). Likewise, 2-3 fold higher endogenous ethylene levels were found in gynoecious
cucumber apices compared to monoecious (Byers et al., 1972; Rudich et al., 1972; Makus et al.,
1975; Rudich et al., 1976; Trebitsh et al., 1987; Yamasaki et al., 2001, 2003). Although high
ethylene evolution was not observed in gynoecious melon apices, hypobaric studies in melon
showed that reduced internal ethylene caused increased maleness that was reversible by ethylene
application (Byers et al., 1972).
Inhibition of ethylene biosynthesis or perception via aminoethoxyvinylglycine (AVG) or
silver nitrate/thiosulfate application leads to increased maleness in cucumber, melon, and certain
genotypes of zucchini (Cucurbita pepo) (Byers et al., 1972; Den Nijs and Visser, 1980; Owens et
al., 1980; Manzano et al., 2011). Our previous work showed that transgenic melons expressing
the dominant negative Arabidopsis thaliana ethylene receptor mutant etr1-1 under the 35S
promoter exhibited an almost complete loss of carpel-bearing buds, demonstrating that ethylene
perception is a requirement for carpel development (Little et al., 2007). To further elucidate the
critical location of ethylene perception involved in sex determination, transgenic melon plants
were produced to target etr1-1 expression to stamen and petal primordia under the Arabidopsis
APETALA3 (AP3) promoter, or carpel and nectary primordia under the CRABS CLAW (CRC)
promoter (Little et al., 2007). If perception in the carpels is required for carpel development, then
inhibition of perception would show reduced production of carpel-bearing buds while stamen
inhibition of perception would show no change in expression. CRC::etr1-1 melon plants
continued to make carpels, however AP3::etr1-1 plants showed an almost complete loss of

31

carpel-bearing buds (Little et al. 2007). These results indicated that the stamen primordia, not
carpel primordia, need to perceive ethylene in order for the carpel primordia to develop and
implicates cross talk between the sex organ whorls.
Genetic analyses show that two major loci control sex determination in melon: the A
locus which inhibits stamens in carpel-bearing flowers, and the G locus which inhibits carpels in
male flowers (Kenigsbuch and Cohen, 1989; Perl-Treves, 1999; Boualem et al., 2008; Martin et
al., 2009). The identity of the two major sex loci has recently been identified. The G locus
encodes a transcription factor named Wound Inducible Protein1 (CmWIP1) (Martin et al., 2009).
The mutant form, g, is silenced by a methylated transposon sequence inserted upstream of its
promoter, resulting in plants for which all flowers are carpel-bearing, i.e. are either gynoecious
(only female), or hermaphrodite (only bisexual) (Martin et al., 2009). Expression of the WIP1
gene is localized to the carpels of buds destined to be male, causing inhibition of carpels in
developing male buds (Martin et al., 2009). Regulation of WIP1 expression has not been
determined. The A locus encodes the ethylene biosynthetic gene, ACC synthase 7 (CmACS7) that
is expressed in carpel primordia (Boualem et al., 2008). A fully functional ACS7 can inhibit
stamen primordia to form a female flower, however the mutant ACS7, which has reduced
enzymatic activity, cannot suppress stamens resulting in bisexual flowers (Boualem et al., 2008).
Interestingly, although CmACS7 inhibits stamen development, expression occurs in carpel
primordia during sex determination stages (Boualem et al., 2008). This complements the
AP3::etr1-1 results indicating inter-whorl regulation of sex determination (Little et al., 2007).
The stages involved in melon sex determination based on our knowledge to date, are
presented in a model in Figure 2.1. Early melon floral bud development includes initiation of all
four floral organ whorls, thus providing the ability to produce both stamens and carpels (Goffinet,

32

Figure 2.1. Progression of melon floral bud development including developmental stages (Goffinet, 1990; Bai et al., 2004), native sex
genes (Boualem et al., 2008; Martin et al., 2009), and sex phenotypes (Kenigsbuch and Cohen, 1990).

33

1990; Bai et al., 2004). The first critical step in sex differentiation occurs between stages 5 and 6;
the choice to inhibit the carpel to become male, or allow the carpels to develop. In buds destined
to be male, expression of the WIP1 transcription factor in the carpel primordia (G locus), and/or
the failure of the stamen primordia to perceive ethylene, leads to carpel primordia arrest (Little et
al., 2007; Martin et al., 2009). The second critical step of sex differentiation occurs between
stages 6 and 7 for carpel-bearing buds; whether to inhibit the stamens to become female, or
remain bisexual. For both developmental paths, the carpel primordia initiate at stage 5 and the
stamen primordia continue to differentiate. In female flowers, increased ethylene produced by a
functional ACS7 (A locus) in the carpel primordia leads to stamen primordia arrest at stage 8
(Boualem et al., 2008).
At the first decision point during melon sex determination, ethylene perception by the
stamen primordia is critical for carpel primordia development. However, the location, timing,
and level of ethylene production required to promote carpel development and inhibit stamen
development is not well defined. To better understand the role of ethylene production, transgenic
melons were used to study the effect of ethylene produced by either the stamen or carpel
primordia. Sex organ primordia-targeted ethylene production was also investigated to determine
if there is an effect on native sex gene expression. Ethylene production targeted to the stamen
primordia, but not carpel primordia, led to increased femaleness and changes in sex gene
expression.

Materials and methods:
Plasmid construction and plant transformation
Construct development for the AP3::ACS and CRC::ACS plasmids was as described in

34

Little et al. (2007), for AP3::etr1-1 and CRC::etr1-1, with the following changes. The etr1-1
gene was removed from pCambia2300-CRC::etr1-1or pCambia2300-AP3::etr1-1 by BamH1
digestion and replaced with the petunia ACS from pCambia2300-35S::ACS (Papadopoulou et al.,
2005), after polymerase chain reaction (PCR) amplification to introduce a BamH1 site at the 5’
and 3’ ends using site-specific primers (Forward - 5’CCCGGATCCCTCTCTCTCTCTCCTCGTGCC (RG280); Reverse –
5’GGGGGATCCGGGCGAATTGGGTACCGGGCC (RG281)).
The AP3::ACS and CRC::ACS constructs were transferred via Agrobacterium
tumefaciens into the andromonoecious cv. Hale’s Best Jumbo (Hollar Seed, Rocky Ford,
Colorado) melon as described in Fang and Grumet (1990) with added revisions described in
Little et al (2007). Putative transgenic regenerated shoots (T0) were evaluated for ACS transgene
presence using PCR and quantitative reverse transcriptase PCR (qRT-PCR) as described below.
PCR-positive T0 were self or cross-pollinated to produce T1, BC1 and BC1S1 progeny. T1, BC1
and BC1S1 plants were grown under greenhouse conditions as described in Little et al. (2007).
T1 AP3::etr1-1-3 plants were from Little et al. (2007).

PCR and qRT-PCR analysis
Genomic DNA was extracted from melon leaf tissue using the Wizard Genomic DNA
Purification Kit (Promega, Madison, WI), and PCR performed to amplify the inserted ACS
transgene using construct specific primers. CRC::ACS was amplified with the CRC specific
primer 5’-CTTGCAATCCCTAGCCAG (RG299) and 3’ ACS primer 5’-

35

CCAGCGTTACTCTTCAATGC (RG310). AP3::ACS was amplified using the AP3-specific
primer 5’-GCTTTGGTCCCCCTCTTTTACC (RG300) and the 3’ ACS primer RG310.
For gene expression analyses by qRT-PCR, total RNA and cDNA was prepared from
200mg of melon apex tissue from T1 CRC::ACS 168 and 169, BC1S1 AP3::ACS A4 and A5, and
T1 AP3::etr1-1-3 plants. Total RNA was isolated from liquid nitrogen frozen tissue using Trizol
reagent based on manufacturer’s recommendations (Invitrogen, Carlsbad, CA), followed by
DNase treatment, and cleaned using RNeasy columns (Qiagen, Valencia CA). Quantification of
RNA was performed using a spectrophotometer (ND-2000c; Nanodrop Technologies,
Wilmington, DE); agarose gel electrophoresis was used to further assess quality. 2ug of RNA
was used for first-strand cDNA synthesis using the SuperScript II reverse transcriptase kit
(Promega, Madison, WI). Gene-specific primers were designed using Primer Express software
(Applied Biosystems). The ACS transgene-specific primers were; Forward –
5’GCTTGCAACCATGTTATCCG (RG319) and Reverse 5’CCAACCTGGTTCAGAACAATG (RG320). Reactions were carried out on the Stratagene
Mx4000 machine using Power SYBR Green PCR Master Mix (Applied Biosystems). PCR
quantification was performed via standard curve reference for each gene. C. melo Polyubiquitin
7 (ICuGI: MU43954) was used as an endogenous control for normalization with the following
primer sequences: 5’-GTTGGGGATGAGAAACGGC-3’ (RG317), and 5’GCCCACTACCACCCGAGC- 3’ (RG318). Each experiment was conducted three times with 34 biological replicates per genotype (each biological replicate was composed of 3 lateral apices
3-4mm in size, removed from branches on nodes 5-6 on 2 month-old plants), and three technical
replicates/sample.

36

Diurnal gene expression analysis and ethephon treatments
Hale’s Best Jumbo seeds were pre-germinated on dampened Whatman #2 filter paper for
24 hours in the dark at 37°C and then sown into 4.8cm x 3.8cm plastic pots filled with Suremix
Perlite soil medium (Michigan Grower Products Inc., Galesburg, MI). Seedlings were grown in a
growth chamber with a 10 hour/14 hour light dark cycle at 24°C and 66.5% humidity. Apices
were removed from 10 day-old plants (second true leaf beginning to emerge) with a razor blade
one hour before the beginning of the light cycle (-1 hour) and at 3 hr, 6 hr, 9 hr and 12 hr time
points for diurnal experiments. For the ethephon experiments, 20uL of 300ppm ethephon or
water with 1% Tween (control) was dropped onto apices and collected at 0 hour (beginning of
light cycle), 8 hours and 11 hours post treatment. RNA extraction and qRT-PCR analysis of
native sex gene expression followed the above procedure. Each experiment was conducted twice
with four apices/time point (one biological replicate) and three biological replicates using a
completely random design.

Ethylene evolution measurements
Two lateral apices (~3-4mm in size) were excised from each 2 month-old melon plant
using a razor blade from node positions 5 and 6 (1 replicate), weighed, and placed into a 10mL
plastic syringe containing a 5mm diameter Whatman filter paper #2 disc moistened with distilled
water (dH2O). A rubber septum was fitted on the opposite end and wrapped with Parafilm to
create a tight seal before the plunger was inserted to the 3mL mark after 1 hour rest period to
allow for dissipation of wound ethylene. Headspace gas samples were taken after 2.5 hours with
a 1mL syringe and analyzed on a gas chromatograph (HACH CARLE series 100 AGC, Linear

37

1200 recorder) with an activated alumina column and flame ionization detector with a 1ppm
ethylene control. The experiment was conducted three times with six replicates/genotype. A
subset of 3 replicates/genotype was selected for qRT-PCR analysis as described above to verify
expression of the transgene after headspace samples were taken. Gene expression trends did not
change during the period of time for sampling of ethylene evolution.

Promoter analysis
The WIP1 promoter sequence has been published (Martin et al., 2009) and was used for
regulatory element and putative transcription factor binding site analysis. The 3kb sequence
upstream of the transcription start site of ACS7 was kindly provided by Jordi Garcia-Mas at the
Universitat de Barcelona in Barcelona, Spain and was used for identification of promoter
sequences using the TSSP software (SoftBerry:
http://linux1.softberry.com/berry.phtml?topic=tssp&group=programs&subgroup=promoter).
Regulatory element and transcription binding site analysis for the promoter regions utilized
PlantPAN (Chang et al., 2008) and PLACE (Higo et al., 1999; Prestridge, 1991).

Sex expression and floral bud microscopy
The T1/BC1 or BC1S1 generation of AP3::ACS and CRC::ACS transgenic melon plants
and their parental and azygous controls were grown under greenhouse conditions as described in
Little et al. (2007). Each plant was observed for presence of staminate and/or carpel-bearing
buds, flowers at anthesis, and first node of appearance, for 30 nodes on the main stem.
Experiments were completed 2-3 times with 4-6 replicates per genotype per experiment in a
randomized complete block design.

38

Examination of floral bud development via microscopy was completed on apices
collected from greenhouse grown melon plants at the 1 true leaf (TL), 2 TL, 3 TL and 4 TL
stages under greenhouse conditions as described in Little et al. (2007) with 5-6
replicates/genotype/plant age. Fixation, dehydration, and paraffin-embedding of collected apical
tissue were performed based on procedures adapted from Takahashi et al. (2010), Ohtsu et al.
(2007), Kidner and Timmermans (2006), and Jackson (1991), and briefly described here. Apices
were collected with a razor blade and placed in an ice-cold mixture of 75% ethanol/25% acetate
fixative solution in 15mL Falcon tubes and subjected to vacuum infiltration for 15 minute
intervals, three to four times, followed by replacement of the fixative with fresh solution and
placed on a rotator overnight at 4⁰C. The next morning samples were placed in fresh fixative for
30 minutes, followed by an ethanol dehydration series: 30% ethanol for 1 hour, 50% ethanol for
1 hour, and 70% ethanol overnight, all steps were completed on a rotator at 4oC. Tissues were
then placed in 85% ethanol for 1 hour then 95% ethanol with 2% eosin added overnight; both
were completed on a rotator at 4oC. The following morning, tissues were moved into glass
bottles with 100% ethanol at room temperature on a rotator for 30 minutes then subjected to
xylene series at room temperature and shaking: 25% xylene/75% ethanol for 30 minutes, 50%
xylene/50% ethanol for 30 minutes, 75% xylene/25% ethanol for 30 minutes, and 100% xylene
for 1 hour. The xylene was removed from the bottles and 100% xylene and ¼ volume paraplast
chips (Sigma-Aldrich, St. Louis, MO) were added and let sit overnight. Samples were then
placed at 42oC until chips melted and moved to 60oC with an addition of another ¼ volume of
chips. Freshly melted paraplast wax was added to the bottles twice a day for three days before
embedding into plastic molds for sectioning and stored at 4oC. Paraffin sections were cut at
10um using a Leica rotary microtome (2125RT; Leica Microsystems, Buffalo Grove, IL) and

39

imaged using a light microscope with Spot RT3 Digital Camera System (SPOT Imaging
Solutions, Diagnostic Instruments Inc., MI). Developing floral buds were characterized using
floral bud stages described in Bai et al. (2004) based on stamen and carpel primordia
morphology, and the number of buds observed at specific stages was recorded for each apex.

Statistical analyses
Data were analyzed by analysis of variance (ANOVA) using Proc GLM or mixed
protocol in SAS (SAS Institute, Cary, NC). Different letters represent significant differences at
P<0.05 using LSD.

Results:
Gene integration and expression
The CRC::ACS and AP3::ACS transgenes were introduced into andromonoecious (cv.
Hale’s Best Jumbo) melon plants via Agrobacterium-mediated transformation, verified to
contain the respective transgenes using PCR, and self-pollinated or back-crossed to wild type
Hales to produce T1, BC1 and BC1S1 progeny (Appendix). ACS transgene expression was
detected within apical tissue of PCR-positive progeny in all lines tested (Figure 2.2a-b). Apices
from AP3::ACS plants had a 3-8 fold increase in ethylene evolution compared to controls (Figure
2.2c). Apical ethylene evolution from CRC::ACS line 169 T1 plants showed two-fold higher
ethylene, but not in the 168 line (Figure 2.2d).

40

c

a
Control

12
10
8
6
4
2
0

Ethylene (pmol/g)

5
4
3
2
1
0

b.
Relative expression
(UBQ7)

c.

ACS expression in apices

b
AP3::ACS AP3::ACS
A4
A5

b

b
a
Control CRC::ACS CRC::ACS
168
169

Ethylene evolution in apices
400

c

300
200
100

b
a

0
Control

d.

Ethylene (pmol/g)

Relative expression
(UBQ7)

a.

350
300
250
200
150
100
50
0

AP3::ACS AP3::ACS
A4
A5

b
a
Control

a
CRC::ACS CRC::ACS
168
169

Figure 2.2. qRT-PCR analysis of ACS transgene expression (a,b) and ethylene evolution (c,d) in lateral apices of Control (WT and
azygous), AP3::ACS A4 and A5 BC1S1 plants (a,c), and CRC::ACS 168 and 169 T1 plants (b,d) generations. Values shown for qRTPCR are the average of three biological replicates with three technical replicates/biological replicate + SE. Values presented for
ethylene evolution are the average of two experiments with 5-6 replicates with 1-2 plants/rep and 1-2 lateral
apices/plant/genotype/experiment + SE. Different letters represent significant differences by LSD at P<0.05.

41

Sex expression phenotypes
The typical sex expression phenotype of melon is a vegetative phase, followed by a phase
with only male flowers, and then a mix of male and bisexual flowers. This pattern was observed
in wild type Hale’s Best Jumbo and azygous segregant progeny from the CRC::ACS and
AP3::ACS lines (Figure 2.3a). No significant difference was observed in the sex expression
pattern of transgenic CRC::ACS 168 and 169 plants (Figure 2.3a), or the number of male or
bisexual buds and open flowers compared to wild type and azygous plants (Figure 2.3b).
In contrast, sex expression along the main stem of AP3::ACS plants showed an increased
femaleness phenotype as manifested by a loss of the male-only phase, and gain of a bisexualonly phase, not observed in wild type and azygous siblings (Figure 2.4a). Onset of first carpelbearing bud was not affected (Figure 2.4a). BC1S1 A4 and A5 AP3::ACS plants showed a
marked decrease in the number of male buds and open flowers in PCR verified plants (Figure
2.4c-d). Variation in duration of the bisexual-only phase was observed in A5 plants, where
approximately half of the plants with verified transgene presence completely lost their male buds,
essentially becoming hermaphrodite plants (Figure 2.4b). The AP3::ACS A4 line also showed a
significant increase in the number of carpel-bearing buds and open flowers (Figure 2.4e-f).
Equivalent results were observed in both the BC1 and BC1S1 generations.

Floral bud development within the apex
A possible explanation for the loss of male buds is that higher ethylene produced by
stamen primordia in the AP3::ACS transgenic plants led to premature stamen primordia arrest
and a subsequent failure of the bud to develop. Paraffin sectioned apices of wild type, azygous
and both transgenic lines were examined to determine floral bud developmental stages (Figure
42

a.

Sex expression on mainstem
30

node on mainstem

25
20
Male&Bisexual
Male
Vegetative

15
10
5
0
WT

Azy

CRC::ACS CRC::ACS
168
169

Buds on mainstem

b.
80

70

# buds

60

50
40

Bisexual
Male

30
20

10
0
WT

Azy

CRC::ACS CRC::ACS
168
169

Figure 2.3. Greenhouse observations of sex expression for WT, azygous, and T1 CRC::ACS 168
and 169 plants along the first 30 nodes of the main stem (a), and number of carpel-bearing
(bisexual) and male buds (b). Values presented are the average of two experiments with 2-3
replicates/experiment + SE.

43

Sex expression on main stem

a.
30

node on mainstem

25
b

b

20
15
10
a, a’

a, a’

5

a’

a’

0

WT
b.

Azy

AP3::ACS A4 AP3::ACS A5

Male&Bisexual
Bisexual
Male
Vegetative

Sex expression on main stem
30

node on main stem

25
20

15
10
5
0
1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1011121314151617181920212223

WT

AP3::ACS A5

Figure 2.4. Greenhouse observations of WT, azygous, and BC1S1 AP3::ACS A4 and A5 plants
for sex expression along the first 30 nodes of the main stem (a), individual WT and AP3::ACS
A5 plants (b), number of male buds (c) and open flowers (d), and number of carpel-bearing
(bisexual) buds(e) and open flowers (f). Values presented for a,c,d,e and f are the average of two
experiments with 2-3 plants/genotype/replicate and 6 rep/experiment + SE. Equivalent results
were observed in both the BC1 and BC1S1 generations. Different letters represent significant
differences by LSD at P<0.05. In panel (a), a,b notation refers to onset of male and bisexual
phase; a,a’ notation refers to first node of carpel-bearing buds.
44

Figure 2.4 (cont’d)
c.

Male buds on main stem
70

a

Open carpel-bearing flowers on main stem

e.

a

# open flowers

60

# buds

50
40
30

20

b

b

10
0
WT

Azy

AP3::ACS
A4

AP3::ACS
A5

Open male flowers on main stem

d.
35

b

a
a
a

WT

Azy.

AP3::ACS
A4

35

a

b

30

25

# buds

15
10

b

5

a
a

a

WT

25

a

20

AP3::ACS
A5

Carpel-bearing buds on main stem

f.

30

# open flowers

4.5
4
3.5
3
2.5
2
1.5
1
0.5
0

Azy

20
15

10
5

b

0

0
WT

Azy.

AP3::ACS
A4

AP3::ACS
A5

45

AP3::ACS
A4

AP3::ACS
A5

2.5a-h). A lower percentage of buds were seen in the post-sex determination stages (5-9) in the
AP3::ACS A4and A5 lines with an average of 37% in the post-sex determination for wild type
and azygous versus 15% for AP3::ACS lines (P= 0.055) (Figure 2.5i).

Native sex gene expression
qRT-PCR analysis of the two major native sex loci in melon, G (WIP1) and A (ACS7),
was performed on apical tissue of CRC::ACS T1, and AP3::ACS BC1S1transgenic plants to
investigate the effect of the introduced gene. No significant difference in gene expression for the
two loci was detected in CRC::ACS plants; however AP3::ACS A4 showed a significant downregulation of WIP1, and both AP3::ACS lines showed up-regulation of ACS7 (Figure 2.6).
Conversely, transgenic AP3::etr1-1 melons, which showed an almost complete loss of carpelbearing buds (Little et al., 2007), showed increased WIP1 expression and decreased expression
of ACS7 (Figure 2.6).
To obtain baseline information about WIP1 and ACS7 expression in wild type plants,
Hales was grown under controlled conditions over a time course of 12 hours. Control apices
showed a diurnal pattern of WIP1 expression, peaking late in the light cycle (Figure 2.7a). ACS7
expression was generally constant throughout the cycle (Figure 2.7a).
The effect of increased ethylene on the G and A sex genes also was tested by application
of ethylene application to apices at the beginning of the light cycle (0 hour). G and A expression
was subsequently examined at 8 hours (diurnal peak for WIP1) and 11 hours (dark) posttreatment. Expression of the known ethylene responsive gene, Ethylene Response Factor1
(Solano et al., 1998) from melon (CmEBF1), increased with ethephon treatment (Figure 2.7b).

46

a.

b.

c.

d.

S

S

Inf

P

S

P

L

St

e.

f.

S

g.

P
P
C

St

h.

P
St

St
S

S

St

C

C
C

%

i.

% buds atat developmental stages
% buds developmental stages
100
90
80
70
60
50
40
30
20
10
0

WT
Azygous
AP3::ACS A4
AP3::ACS A5

Stages 2-4
Stages 5-9
pre-sex determination
post-sex determination
Developmental stages
Figure 2.5. Microscopy images of melon floral buds at inflorescence stage 1(a), stage 2 (b), stage
3 (c), stage 4 (d), stage 5 (e), stage 6 (f), stage 7 (g), and bisexual bud at stage 8 (h) based on Bai
et al. (2004) cucumber stages. L= leaf primordia, Inf= inflorescence meristem, S= sepal
primordia, P= petal primordia, St= stamen primordia, and C= carpel primordia. Black bars
represent 50 µm at 100x magnification (a-g), and 40x magnification (h). Percentage of floral
buds at different stages of development observed in microtome-sectioned apices of WT, azygous,
and BC1S1 AP3::ACS A4 and A5 under light microscopy (i).Values presented are an average of
5-6 replicates/time point + SE.

47

WIP1 expression in apices
WIP1 expression in apices

a.
a.

Log2 expression (relative to control)

4

3

CRC::ACS
168

AP3::ACS
A4

AP3::ACS
A5

AP3::etr1-1

*

2

1
0
-1

-2
-3

-4
-5

*

-6
b.
b.
4

Log2 expression (relative to control)

CRC::ACS
169

3

ACS7 expression in apices
ACS7 expression in apices
CRC::ACS
168

CRC::ACS
169

AP3::ACS
A4

AP3::ACS A5

*

AP3::etr1-1

*

2
1
0

-1
-2

*

-3

Figure 2.6. qRT-PCR analysis of the melon sex genes WIP1 (a) and ACS7 (b) in apices of T1
CRC::ACS 168 and 169, BC1S1 AP3::ACS A4 and A5, and T1 AP3::etr1-1 plants relative to
expression of wild type controls. Values shown are the average of three biological replicates with
three technical replicates/biological replicate + SE. * represent significant differences by t test at
P<0.05.
48

a.
a.

Sex Sex gene expression in apices
gene expression in apices

Relative expression (UBQ7)

16
14
12
10
8

WIP1

6

ACS7

4
2
0
-1

0

1

2

b.
b.

3

4
5
6
7
Daylength (hours)

8

9

10

11

Expression in apices
Expression in apices
Log2 expression (relative to control)

6

EBF1
EBF1

*

4

2

ACS7
ACS7

WIP1
WIP1

*

*
8 hours

0

11 hours

-2

*

-4

*

-6

Figure 2.7. qRT-PCR analysis of ACS7 and WIP1 expression (a) in wild type melon apices taken
at -1 hr, 2 hr, 5 hr, 8 hr and 11 hr relative to a 10 hour light cycle, and EBF1, ACS7, and WIP1
expression (b) in ethephon treated apices at 8 hr (light) and 11 hr (dark) post-treatment relative to
untreated controls (log2). Values shown are the average of two experiments with three biological
replicates with three technical replicates/biological replicate + SE. Trends were equivalent for
both experiments. * represent significantly different from untreated control by t test at P<0.05.
49

As was observed in the apices of AP3::ACS transgenic melons (Figure 2.6), WIP1 expression
decreased in response to ethylene; ACS7 expression showed a transient increase in expression
(Figure 2.7).

WIP1 and ACS7 promoter elements
Analysis of the promoter areas of WIP1 for known regulatory elements and transcription
factor binding sites showed a densely packed set of ethylene and light responsive elements in the
first 500bp of sequence. There was an ethylene response element (ERE) at -340bp (Itzhaki et al.,
1994); a putative ERFBP binding site at -442bp (Riechmann and Meyerowitz, 1998); and
multiple light-regulatory elements: GATA box, GT1 motif, and circadian motif, at positions 83bp, -90bp, -359bp, and -485bp (Figure 2.9a) (Terzaghi and Cashmore, 1995; Arguello-Astorga
and Herrera-Estrella, 1998). Three promoter regions were predicted for ACS7at positions -263bp,
-2386bp, and -2859bp. The longest predicted promoter was used for analysis based on average
predicted promoter lengths of CsACS1G and AtACS7 (Wang et al., 2004; Wu et al., 2012). The
ACS7 promoter had ERE elements at -2733bp and -1409, and a putative binding site for a
tobacco EIN3-like protein (NbEIL) at -2279 (Figure 2.9b)(Kosugi and Ohashi, 2000). No lightregulatory elements or putative light-regulated protein binding sites were predicted for the ACS7
promoter.

Discussion:
Ethylene evolution and native sex genes
Ethylene has long been identified as the predominant hormone involved in sex

50

ERFBP ERE
a. WIP1 -442 -340

- 620
b. ACS7
- 2859

ERE
-2733

NbEIL site
-2279

ERE
-1409

- 2386

GATA box
-83
TS

0
GT1 GATA Circadian
motif box
motif
-485 -359
-90 TS
- 263

0

Figure 2.8. Diagram of the promoter of WIP1 (a) and ACS7 (b) with positions of putative regulatory elements and transcription factor
binding sites for ethylene responsive element (ERE), tobacco EIN3-like binding site (NbEIL), ERF binding site (ERFBP), GATA box,
and GT1 motif (Itzhaki et al., 1994; Terzaghi and Cashmore, 1995; Riechmann and Meyerowitz, 1998; Arguello-Astorga and HerreraEstrella, 1998). Arrows indicate predicted promoter start sites.

51

determination in cucurbit species as evidenced by effects of exogenous application, sex
genotype-related differences, and endogenous ethylene levels (Rudich, 1990; Perl- Treves, 1999;
Grumet and Taft, 2011). Cloning of the key sex expression genes, F and M in cucumber, and A
in melon (Trebitsh 1997; Knopf and Trebitsh, 2006; Boualem et al., 2008, 2009) which all
encode ACS genes, and the effect of increased internal ethylene level via 35S::ACS expression in
transgenic melons (Papadopoulou et al., 2005) further show the importance of ethylene
production in promoting femaleness, either by stimulating carpel production or suppressing
stamens. Here we produced transgenic melons expressing ACS under floral primordia-targeted
promoters to study the effect of location and timing of ethylene production during the sex
determination process, and to investigate the role of cross-talk between the sex organ primordia
in regulation of development.
The CRC::ACS and AP3::ACS melon plants showed a 2-8 fold increase in apical ethylene
relative to wild type plants. Numerous studies measuring ethylene evolution from gynoecious
cucumber apices have observed 2-3 fold differences between gynoecious and monoecious
ethylene levels, providing evidence for the biological significance of two-fold differences (Byers
et al., 1972; Rudich et al., 1972, 1976; Yamasaki et al., 2001, 2003).

Stamen primordia-targeted expression of ACS leads to feminization
Transgenic melons expressing ACS in the stamen and petal primordia showed increased
femaleness manifested by gain of a bisexual-only phase not typically seen in wild type, a
decrease in male buds and open flowers, and loss of the male-only phase along the main stem. In
the extreme case, plants became phenotypically hermaphrodite rather than andromonoecious.

52

This increased femaleness phenotype is consistent with our previous data showing that ethylene
perception in the stamens is required to promote femaleness (Little et al., 2007). However, in
those experiments, increased femaleness was manifested by increased carpel development (Little
et al., 2007); here the primary effect was loss of male buds. Studies with exogenous ethylene
have indicated that timing and concentration are key factors determining whether carpel or
stamen development is affected. Loss of male buds was more frequently observed in
andromonoecious and monoecious melons treated with ethrel at a very young age (second leaf
stage) (Rudich et al., 1969; Karchi, 1970). Increase in carpel-bearing buds occurred more
frequently when plants were treated at later stages (i.e. visual flower bud and first flower at
anthesis). Inhibition of stamens within carpel-bearing buds causing female rather than bisexual
flowers also occurred more frequently when plants were treated at later stages, consistent with a
time when carpel-bearing buds would have been already developing (Karchi, 1970). Thus, it has
been suggested that there are critical windows in floral bud development when sex organs are
sensitive to the effects of ethylene (Galun et al., 1963; Yamasaki et al., 2003).
Furthermore, higher concentrations (300ppm-600ppm) of ethrel were required to inhibit
stamens in carpel-bearing buds than to promote production of carpel-bearing buds (Karchi, 1970).
Andromonoecious and monoecious melons require concentrations of ethrel at 500ppm or higher
to inhibit staminate buds (Karchi, 1970; Rudich et al., 1969). When low (150ppm)
concentrations of ethrel were used, carpel-bearing buds were promoted (Rudich et al., 1969;
Karchi, 1970).
The loss of male buds on the AP3::ACS plants suggested the possibility that bud abortion
may be occurring, rather than continued development of asexual flowers that would result from
premature inhibition of stamen primordia. If increased endogenous ethylene levels in the stamen

53

primordia were sufficiently high early in development to cause inhibition of stamen primordia
prior to successful establishment of carpel development, this would result in asexual flowers.
Consistent with this possibility that abortion occurred at an early stage of bud development,
fewer AP3::ACS buds transitioned from the pre-sexual stages into the sex differentiation stages.
We also did not observe the production of any asexual buds or flowers. Similarly, exogenous
application of ACC to monoecious cucumber led to abortion of staminate buds before female
flowering (Trebitsh et al., 1997; George, 1971). Loss of male flowers was also observed in the
Cucurbita pepo ‘Vegetable Spaghetti’ genotype treated with ethephon (Manzano et al., 2011),
and different melon genotypes treated with ethrel (Karchi, 1970).
Despite increased ethylene production, the AP3::ACS plants did not cause a conversion
of bisexual flowers to female. This may result from the higher threshold required for stamen
suppression than carpel promotion. Only those buds whose ethylene levels were not sufficiently
high to suppress stamens would be able to proceed further in development, thus resulting in
bisexual flowers.
In contrast to the effects of AP3::ACS, targeting of ethylene production to the carpel and
nectary primordia using the CRC promoter did not cause a significant difference in number of
male or carpel-bearing buds and open flowers, or sex expression pattern along the main stem.
These results are consistent with the previous report by Little et al. (2007) showing that lack of
ethylene perception by the carpel primordia did not prevent subsequent carpel development, and
the role of stamen primordia in promoting carpel development.

54

Native sex genes
The ACS transgene also influenced native sex gene expression. Apices of AP3::ACS
melons demonstrated a decrease in expression of the native sex gene, WIP1, the suppressor of
carpel development, consistent with a phenotype of increased femaleness. Conversely, inhibition
of ethylene perception in the AP3::etr1-1 plants, which show a near loss of carpel-bearing buds,
was associated with increased WIP1 expression. WIP1 has not previously been reported to be
affected by ethylene, and yeast one-hybrid studies showed no direct interaction between the
WIP1 transcription factor and ACS7 promoter (Martin et al., 2009). These results provide
evidence that ethylene acts to regulate WIP1 gene expression, consistent with increased
exogenous and endogenous ethylene levels causing increased carpel-bearing buds and flowers
(McMurray and Miller, 1968; Robinson et al., 1969; Rudich et al., 1969; Karchi, 1970;
Shimotsuma and Jones, 1972; Augustine et al., 1973; Owens et al., 1980; Papadopoulou et al.,
2005). WIP1 gene expression was also reduced by exogenous ethylene. The WIP1 promoter was
found to include the known ethylene response element (ERE) and ERFBP found within the
promoters of the ethylene-responsive genes carnation glutathione-S-transferase (GST1) and
tomato E4 (Montgomery et al., 1993; Itzhaki et al., 1994; Xu et al., 1996).
Interestingly, the WIP1 gene showed a clear diurnal pattern of expression with a peak
occurring mid-day. Promoter analysis for cis-regulatory sites predicted multiple light regulatory
motifs found within known light regulated genes, i.e. RuBisCo (rbcS) (Gidoni et al., 1989) and
chlorophyll a/b-binding proteins of photosystem II (Lhcb) (Giuliano et al., 1988). Studies in
cucumber have observed that sex expression can be influenced by day length; short days increase
femaleness and long days increase maleness (Takahashi et al., 1983; Yamasaki et al., 2003b).
These promoter elements may contribute to day length related effects on sex expression.
55

Ethylene levels were also shown to follow a diurnal pattern with a peak at mid-day, which was
enhanced under short day conditions and correlated with increased femaleness (Yamasaki et al.,
2003b).
Boualem et al. (2008) reported that increased ethylene produced by the fully functional
ACS7 enzyme leads to stamen inhibition in carpel-bearing flowers. AP3::ACS plants showed an
increase in ACS7 sex gene expression in the apex. However, the mutant ACS7 present in
andromonoecious melon is unable to suppress stamens due to reduced enzymatic activity
(Boualem et al., 2008).

Ethylene threshold model for sex expression in melon
Our objective for this study was to examine the effects of location and timing of ethylene
production during floral bud development. Yin and Quinn et al. (1995) proposed an ethylene
threshold model in which differing levels of sensitivity in stamen or carpel primordia could allow
the primordia to react to different ranges of ethylene concentrations, and thus inhibit or promote
each sex independently. The ethylene results presented here provide further evidence for
ethylene thresholds acting within the apex during floral bud development, and indicate critical
developmental stages and levels for sex differentiation for promotion or inhibition of sex organs
(Figure 2.9).
At floral bud stage 4, the stamen, but not carpel, whorls have initiated. At stage 5, at low
levels of ethylene, carpels are suppressed, as would occur in monoecious (GA) or
andromonoecious (Ga) plants. At intermediate levels, G expression is suppressed in
andromonoecious or monoecious plants, or not expressed in gynoecious (gA) or bisexual (ga)

56

Stage of
WIP1 (G)
action

Stage of
ACS7 (A)
action
gA
GA

AP3::ACS

AP3::ACS

AP3::ACS
ga (bisexual)
gA (gyno.)

AP3::ACS
ga, gA

AP3::ACS
ga

CRC::ACS

CRC::ACS

Ethylene
level

GA
Ga
CRC::ACS
Ga (andromono.)
GA (mono.)

4
Stamen whorl formation
Pre-sex

CRC::ACS
Ga, GA
5-6
Carpel whorl formation
Carpel arrest
Developmental stage

Stamen
inhibition

Carpel
promotion

Ga
CRC::ACS
Ga, GA
7
Stamen arrest

Figure 2.9. Ethylene threshold model of melon floral bud development including known developmental stages (stamen whorl
formation/pre-sex, 4; carpel whorl formation/carpel arrest, 5-6; stamen arrest, 7), stages of sex gene action (WIP1/G; ACS7/A),
genotypes (GA, Ga, gA, ga), and sex phenotypes (monoecious (mono.), andromonoecious (andromono.), gynoecious (gyno.), and
bisexual).

57

plants, and carpels are promoted. If ethylene levels at this stage are too high, the stamen
primordia are inhibited, leading to bud abortion. As the floral bud moves into stages 5-6, if
sufficient ethylene is perceived by the stamen primordia, the carpel primordia continue to
develop; levels that are too low to promote carpels lead to a staminate bud. Carpel-bearing buds
on hermaphrodite and gynoecious plants continue to develop. At stage 7, if ethylene levels in
carpel-bearing buds are sufficiently high, as occurs in gynoecious or monoecious plants, the
stamen primordia will fail to develop producing female flowers. If ethylene levels do not exceed
this threshold, the stamens continue to develop to become a bisexual flower.
This information makes evolutionary sense when building a unisexual flower. After
initiation of the whorls, the stamen or carpel primordia must be inhibited to become staminate or
carpel-bearing, respectively. If development of the stamen primordia was arrested before the
initiation of the carpels, energy to sustain fruit set would be wasted to produce an asexual flower.
Regulation of stamens by the carpels prevents this occurrence. Similarly, regulation of carpels by
the stamens, which develop first, minimizes resources allocated to carpels in flowers destined to
be male. Finally, a higher ethylene threshold for suppression of stamens than carpels, also
ensures that stamens are not inhibited prior to establishment of carpels.

58

APPENDIX

59

PCR segregation of the two lines of CRC::ACS (168 and 169) matched a single-gene
insertion. One line of AP3::ACS (A4) showed a PCR segregation ratio matching a two gene
insertion model, the other line (A5) was consistent with a three-gene insertion.

Table A1. PCR segregation ratios of transgenic CRC::ACS T1 and AP3::ACS BC1/BC1S1 lines.
Line

Generation

Segregation (+ : -) Expected ratios

2

CRC::ACS 168

T1

17:9

3:1

X
0.82 ns

CRC::ACS 169

T1

23:6

3:1

0.10 ns

AP3::ACS A4

BC1

26:8

3:1

0.29 ns

AP3::ACS A5

BC1

23:0

15:1

0.22 ns

AP3::ACS A4

BC1S1

15:2

15:1

0.05 ns

AP3::ACS A5

BC1S1

16:3

225:1

0.44 ns

60

LITERATURE CITED

61

LITERATURE CITED

Arguello-Astorga G and Herrera-Estrella, L (1998). Evolution of light regulated plant promoters.
Annu Rev Plant Phys 49: 525-555.
Augustine JJ, Bakier LR, Sell HM (1973). Female flower induction on androecious cucumber,
Cucumis sativus L. J Amer Soc Hort Sci 98: 197-199.
Bai SL, Peng YB, Cui JX, Gu HT, Xn LY, Li YQ, Xu ZH, 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.
Boualem A, Fergany M, Fernandez R, Troadec C, Martin A, Morin H, Sari M, Collin F, Flowers
J, et al. (2008). A conserved mutation in an ethylene biosynthesis enzyme leads to
andromonoecy in melons. Science 321: 836-838.
Bowman JL and Smyth DR (1999). CRABS CLAW, 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 RE, Baker LR, Sell HM, Herner RC, Dilley DR (1972). Ethylene: a natural regulator of
sex expression in Cucumis melo L. P Natl Acad Sci USA 69: 717-720.
Chang WC, Lee TY, Huang HD, Huang HY, Pan RL (2008). PlantPAN: Plant Promoter
Analysis Navigator, for identifying combinatorial cis-regulatory elements with distance
constraint in plant gene group. BMC Genomics 9: 561.
Den Nijs A, Visser D (1980). Induction of male flowering in gynoecious cucumbers (Cucumis
sativus L.) by silver ions. Euphytica 29:237-280.
Fang G and Grumet R (1990). Agrobacterium tumefaciens mediated transformation and
regeneration of muskmelon plants. Plant Cell Rep 9: 160-164.
Galun E, Jung Y, Lang A (1963). Morphogenesis of floral buds of cucumber cultured in vitro.
Dev Biol 6: 370-387.
Gidoni D, Brosio P, Bond-Nutter D, Bedbrook J, Dunsmuir P (1989). Novel cis-acting elements
in petunia Cab gene promoters. Mol Genet Genomics 215: 337-344.
Giuliano G, Pichersky E, Malik VS, Timko MP, Scolnik PA (1988). An evolutionarily conserved
protein-binding sequence upstream of a plant light-regulated gene. P Natl Acad Sci USA 85:
7089-7093.

62

Goffinet M (1990). Comparative ontology of male and female flowers of Cucumis sativus. In:
Biology and Utilization of the Cucurbitaceae. Bates DM, Robinson RW, Jefferey C (eds).
Cornell Univ Press, NY: 288-304.
Grumet R and Taft JA (2011). Sex expression in cucurbits In: Genetics, Genomics and Breeding
in Cucurbits. Kole C, Wang YH, Behera TK, Kole C (eds). Science Publishers Inc. Enfield NH
and CRC Press (Taylor & Francis) group, Boca Raton, FL. P. 353-375.
Hao YJ, Wang DH, Peng YB, Bai SL, Xu LY, Li YQ, Zu ZH, Bai SN (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.
Higo K, Ugawa Y, Iwamoto M, Korenaga T (1999). Plant cis-acting regulatory DNA elements
(PLACE) database: 1999. Nucleic Acids Res 27: 297-300.
Ishiki Y, Oda A, Yaegashi Y, Orihara Y, Arai T, Hirabayashi T, Nakagawa H, Sato T (2000).
Cloning of an auxin-responsive 1-aminocyclopropane-1-carboxylate synthase gene (CMe-ACS2)
from melon and the expression of ACS genes in etiolated melon seedlings and melon fruits. Plant
Sci 159: 173-181.
Itzhaki H, Maxson JM, Woodson WR (1994). An ethylene-responsive enhancer element is
involved in the senescence-related expression of the carnation glutathione-S-transferase (GST1)
gene. P Natl Acad Sci 91: 8925-8929.
Jackson D (1991). In-situ hybridization in plants. In: Molecular Plant Pathology: A Practical
Approach. Bowles DJ, Gurr SJ and McPherson M (eds). Oxford University Press.
Johnson PR and Ecker JR (1998). The ethylene gas signal transduction pathway: A molecular
perspective. Annu Rev Genet 32: 227-254.
Karchi Z (1970). Effects of 2-chloroethanephosphonic acid on flower types and flowering
sequences in muskmelon. J Amer Soc Hort Sci 95: 575-578.
Kidner C and Timmermans M (2006). In situ hybridization as a tool to study the role of miRNAs
in plant development. Method Mol Cell Biol 342: 159-179.
Kosugi S and Ohashi Y (2000). Cloning and DNA-binding properties of a tobacco EthyleneInsensitive3 (EIN3) homolog. Nucleic Acids Res 28: 960–967.
Lasserre E, Bouquin T, Hernandez JA, Bull J, Pech JC, Balague C (1996). Structure and
expression of three genes encoding ACC oxidase homologs from melon (Cucumis melo L.). Mol
Genet Genomics 251: 81-90.

63

Little HA, Papadopoulou E, Hammar SA, Grumet R (2007). The influence of ethylene
perception on sex expression in transgenic melon as assessed by expression of the mutant
ethylene receptor gene, At-etr1-1, under control of constitutive and floral-targeted promoters.
Sex Plant Reprod 20: 123-136.
Makus DJ, Pharr DM, Lower RL (1975). Some morphogenic differences between monoecious
and gynoecious cucumber seedlings as related to ethylene production. Plant Physiol 55: 352-355.
Manzano S, Martinez C, Megias Z, Gomez P, Garrido D, Jamilena M (2011). The role of
ethylene and brassinosteriods in the control of sex expression and flower development in
Cucurbita pepo. Plant Growth Regul 65: 213-221.
Martin A, Troadec C, Boualem A, Mazen R, Fernandez R, Morin H, Pitrat M, Dogimont C,
Bendhamane A (2009). A transposon-induced epigenetic change leads to sex determination in
melon. Nature 461: 135-1139.
McMurray AL and Miller CH (1968). Cucumber sex expression modified by 2chloroethanephosphonic acid. Science 162: 1397-1398.
Miki T, Yamamoto M, Nakagawa H, Ogura N, Mori H, Imaseki H, Sato T (1995). Nucleotidesequence of a cDNA for 1-aminocyclopropane-1-carboxylate synthase from melon fruits. Plant
Physiol 107:297-298.
Montgomery J, Goldman S, Deikman J, Margossian L, Fischer RL (1993). Identification of an
ethylene-responsive region in the promoter of a fruit ripening gene. P Natl Acad Sci 90: 53395343.
Ohtsu K, Smith MB, Emrich SJ, Borsuk LA, Zhou B, Chen T, Zhang X, Timmermans MCP,
Beck J, Buckner B et al. (2007). Global gene expression analysis of the shoot apical meristem of
maize (Zea mays L.). Plant J 52: 391-404.
Owens KW, Peterson CE, Tolla GE (1980). Production of hermaphrodite flowers on gynoecious
muskmelon by silver nitrate and aminoethoxyvinylglycine. HortScience 15:654-655.
Papadopoulou E, Little HA, Hammar SA, Grumet R (2005). Effect of modified endogenous
ethylene production on sex expression, bisexual flower development, and fruit production in
melon (Cucumis melo L.). Sex Plant Reprod 18: 131-142.
Prestridge DS (1991). SIGNAL SCAN: A computer program that scans DNA sequences for
eukaryotic transcriptional elements. CABIOS 7: 203-206.
Riechmann JL and Meyerowitz EM (1998). The AP2/EREBP family of plant transcription
factors. Biol Chem 379: 633-646.
Robinson RW, Shannon S, La Guardia MD (1969). Regulation of sex expression in the
cucumber. BioScience 19: 141-142.
64

Roy RP and Saran S (1990). Sex expression in the Cucurbitaceae. In: Biology and utilization of
the Cucurbitaceae. Bates DM, Robinson RW, Jeffrey C (eds). Cornell University Press, Ithaca,
NY: 251-268.
Rudich J (1990). Biochemical aspects of hormonal regulation of sex expression in Cucurbits. In:
Bates DM, Robinson RW, Jeffrey C (eds). Biology and utilization of the Cucurbitaceae. Cornell
University Press, Ithaca, NY: 269-280.
Rudich J, Baker LR, Scott JW, Sell HM (1976). Phenotypic stability and ethylene evolution in
androecious cucumber. J Amer Soc Hort Sci 101: 48-51.
Rudich J, Haley AH, Kedar N (1972). Ethylene evolution from cucumber plants related to sex
expression. Plant Physiol 49: 998-999.
Rudich J, Haley AH, Kedar N (1969). Increase in femaleness of three cucurbits by treatment
with ethrel, an ethylene-releasing compound. Planta 86: 69-76.
Solano R, Stepanova A, Chao Q, Ecker JR (1998). Nuclear events in ethylene signaling: a
transcriptional cascade mediated by ETHYLENE-INSENSITIVE3 and ETHYLENERESPONSE-FACTOR1. Genes and Dev 12: 3703-3714.
Takahashi H, Saito T and Suge H (1983).Separation of the effects of photoperiod and hormones
on sex expression in cucumber. Plant Cell Physiol 24: 147-154.
Takahashi H, Kamakura H, Sato Y, Shiono K, Abiko T, Tsutsumi N, Nagamura Y, Nishizawa
NK, Nakazomo M (2010). A method for obtaining high quality RNA from paraffin sections of
plant tissues by laser microdissection. J Plant Res 126: 807-813.
Terzaghi WB and Cashmore AR (1995). Light regulated transcription. Annu Rev Plant Phys 46:
445–474.
Trebitsh T, Rudich J, Riov J (1987). Auxin, biosynthesis of ethylene and sex expression in
cucumber (Cucumis sativus). Plant Growth Regul 5: 105-113.
Wang NN, Shih MC, Li N (2004). The GUS reporter-aided analysis of the promoter activities of
Arabidopsis ACC synthase genes AtACS4, AtACS5, and AtACS7 induced by hormones and
stresses. J Exp Bot 56: 909-920.
Wu T, Qin Z, Feng Z, Zhou X, Xin M, Du Y (2012). Functional analysis of the promoter of a
female-specific cucumber CsACS1G gene. Plant Mol Biol Rep 30: 235-241.
Xu RL, Goldman S, Coupe S, Deikman J (1996). Ethylene control during tomato fruit ripening
involves two cooperative cis-elements. Plant Mol Biol 31: 1117-1127.

65

Yamasaki S, Fujii N, Matsura S, Mizusawa H, Takahashi H (2001). The M locus and ethylenecontrolled sex determination in andromonoecious cucumber plants. Plant Cell Physiol 42: 608619.
Yamasaki S, Fujii N, Takahashi H (2003a). Characterization of ethylene effects on sex
determination in cucumber plants. Sex Plant Reprod 16:103-111.
Yamasaki S, Fujii N, Takahashi H (2003b). Photoperiodic regulation of CS-ACS2, CS-ACS4
and CS-ERS gene expression contributes to the femaleness of cucumber flowers through diurnal
ethylene production under short-day conditions. Plant Cell Environ 26: 537-546.

66

CHAPTER 3
Effect of carpel primordia-targeted inhibition of ethylene perception on sex expression and
fruit set, development, and ripening in melon (Cucumis melo L.)

67

Introduction:
Angiosperms typically produce bisexual flowers expressing both male and female organs
(stamens and carpels, respectively). The Cucurbitaceae family is unique in that it contains
species exhibiting a variety of heritable sex phenotypes. The Cucumis species cucumber (C.
sativus) and melon (C. melo) are model organisms for studies of sex determination (Roy and
Saran, 1990; Grumet and Taft, 2011). The common sex expression pattern in melon is
andromonoecious. Progressing along the main stem, vegetative nodes are followed by nodes
bearing male flowers, and then by nodes with a mixture of male and bisexual flowers.
Ethylene is a key factor in the regulation of sex determination in cucurbits (Perl-Treves,
1999; Grumet and Taft, 2011). Exogenous application of ethylene-releasing chemicals leads to
increased femaleness in both melon and cucumber. Transgenic melons over-expressing the
ethylene biosynthetic enzyme ACS, under the 35S promoter showed increased femaleness
(Papadopoulou et al., 2005). Inhibition of ethylene biosynthesis with aminoethoxyvinylglycine
(AVG), or ethylene perception with silver ions leads to increased maleness (Den Nijs and Visser,
1980; Owens et al., 1980). Similarly, constitutive expression of the Arabidopsis thaliana
ethylene receptor mutant etr1-1, conferring loss of ethylene perception in transgenic melons,
resulted in almost complete loss of carpel-bearing nodes (Little et al., 2007). These results
demonstrated that ethylene perception is required for carpel primordia development during
melon sex determination (Little et al., 2007). Furthermore, our previous studies targeting
expression of etr1-1 to either the stamen and petal primordia with the APETALA3 (AP3)
promoter, or carpel and nectary primordia with the CRABS CLAW (CRC) promoter, indicated
that ethylene perception by stamen primordia, but not carpel primordia, is essential for the
production of carpel bearing buds (Little et al., 2007). Stamen targeted expression of etr1-1

68

caused a near-complete loss of carpel-bearing buds. In contrast, transgenic andromonoecious
melon plants expressing etr1-1 under the control of the carpel and nectary primordia-directed
CRABS CLAW (CRC) promoter showed increased and earlier carpel bearing buds and flowers in
the greenhouse (Little et al., 2007).
Ethylene is also a key factor regulating fruit ripening in fleshy fruits such as tomato,
apple and melon (Barry and Giovannoni, 2007). Fruit can be classified based on whether they
undergo a rise in ethylene levels and respiration at early ripening stages (climacteric), or not
(non-climacteric). In climacteric fruit, ethylene is a major factor in controlling fruit firmness,
color, taste, and aroma (Barry and Giovannoni, 2007; Bapat et al., 2010; Paul et al., 2012).
Studies in melon, which has climacteric and non-climacteric members, have elucidated aspects
of ethylene’s role in the ripening process (Paul et al., 2012). Research by Guis et al. (1997) and
Pech et al. (2008) using ACO-antisense melons under the constitutive 35S promoter have
characterized ethylene-dependent and independent pathways of ripening. Loss of chlorophyll,
carotenoid increase in the rind, peduncular abscission, aroma production, and a majority of the
flesh softening process were found to be ethylene-dependent processes, while acidity level
changes, and flesh carotenoid and sucrose accumulation were ethylene-independent (Pech et al.,
2008).
In this work, we studied transgenic melons targeting expressing of etr1-1 to the carpel
and nectary primordia under control of the CRABS CLAW (CRC) promoter (CRC::etr1-1) to
determine if the increased femaleness phenotype previously observed in the greenhouse allows
for earlier or increased fruit set. Fruit development was also assessed to determine if ethylene
perception inhibition in the carpels has an effect on fruit growth and ripening. We observed
increased femaleness and a conversion of bisexual-to-female flowers, not seen in wild type

69

plants. The CRC::etr1-1 plants also exhibited increased and earlier fruit set, and modified fruit
ripening. These results demonstrate the potential to modulate fruit set and ripening by targeted
ethylene inhibition to the carpel and nectary primordia.

Materials and methods:
Plant material and growth conditions
Melon genotypes used for these experiments were wild type, andromonoecious Hale’s
Best Jumbo (Hales; Hollar Seed, Rocky Ford, CO) and two CRC::etr1-1 transgenic lines, M5
and M15, that were originally produced from Hales as described by Little et al. (2007). T1 M5
and M15 plants were self-pollinated in the greenhouse to obtain T2 seed. Segregating progeny
were verified for transgene presence by PCR analysis using a CRC specific primer (5’ CTTGCAATCCCTAGCCAG -3’, RG299) and etr1-1 specific primer (5’GCTCATGGGACACACAACTCGG-3’, RG306), and for transgene expression as described
below. PCR positive T2 M5 and M15 plants were grown under greenhouse conditions as
described by Little et al. (2007). Ovary diameter and length was recorded on M5 and M15 plants
in the greenhouse.

Field experiments
Field trials were performed in East Lansing, MI during the summer of 2009 with Hales
and T2 CRC::etr1-1 M5 and M15 plants in a randomized complete block design with 6
replications with 6 plants/plot, and in the summer of 2011using 4 replications and 10 plants/plot.
Three week-old greenhouse-grown plants were transplanted into the field at 1m intervals into
70

1.5m wide black plastic rows with 3m between rows. The plot was surrounded on all sides by a
border of Hales plants. Standard practices were used for fertilization and pesticide application. In
2009, the main stems of 4 plants/plot were scored for number of male and carpel-bearing buds
and open flowers, and first node appearance of carpel-bearing buds and open flowers. In 2011,
the plants were observed for first node on the main stem of set fruit and total fruit set on the
whole plant throughout the growing season. Fruit development and ripening was monitored on
selected fruit by tagging fruit of equivalent diameter from each of the genotypes at 26, 40, and 46
days after planting. At final harvest (10 weeks after transplanting), ripening stage for all fruit was
characterized by rind color as: green, turning (mix of green and orange), or orange. Fruit length,
diameter, and weight were recorded for all fruit.

qRT-PCR analysis
Total RNA was prepared from 200mg of liquid nitrogen frozen melon apex tissue (~34mm in size) removed from plants at the 2 leaf stage, or from lateral branches (5-6 nodes in
length) using Trizol (Invitrogen, Carlsbad, CA). cDNA preparation was performed using
Superscript II reverse transcriptase (Promega, Madison, WI) following manufacturer’s
recommendations. RNA was also prepared from 200-300mg of fruit mesocarp or exocarp
(minus rind) from 10mm and 60mm diameter field grown fruit (~3 days post pollination (dpp),
and ~14 dpp), and greenhouse grown 35 dpp fruit (~107mm diameter, turning stage). Genespecific primers for the etr1-1transgene were designed using Primer Express 3.0 software
(Applied Biosystems); the primer sequences are Forward -GTTGGGGATGAGAAACGGC-3’
(RG317), and Reverse -GCCCACTACCACCCGAGC- 3’ (RG318). qRT-PCR analysis was
carried out using a Stratagene Mx4000 system with Power SYBR Green PCR Master Mix

71

(Applied Biosystems) for PCR quantification via standard curve reference. C. melo Polyubiquitin
7 (ICuGI: MU43954) was used as an endogenous control for normalization; the primer
sequences are Forward -GACCGGGAAGGAAATTGAGAT-3’ (RG541) and Reverse CCGCTCCTTAATACGATCAATTGG-3’ (RG543).

Ethylene measurements
Ethylene measurements were performed on apices of greenhouse grown Hales,
CRC::etr1-1 M5 and M15 plants. Lateral apices (~3-4mm in size) were excised from branches 810 nodes long using a razor blade, and weighed to record fresh weight. Apices were placed in a
10mL plastic syringe with a 5mm diameter Whatman filter paper #2 disc moistened with distilled
water, and placed at room temperature for 1 hour to allow for diffusion of wound ethylene.
Ethylene measurements were performed by gas chromatography as described in Papadopoulou et
al. (2005). The experiment was repeated twice with six plants/genotype in a randomized
complete block design. Three replicates/plant were taken for ethylene analysis with two lateral
apices/replicate.
To perform fruit ethylene measurements, fruit harvested from the field were classified by
rind color as green (green rind, unripe), turning (mixture of green and orange), or orange (fully
orange rind, ripe) and weighed. Three fruit/rind color category (one replicate) were placed in
separate 10 gallon buckets and sealed for 1 hour with air tight covers with a rubber septum prior
to sampling for ethylene concentration. Internal fruit ethylene was measured from each fruit after
whole fruit head space was analyzed. An 18 gauge needle attached to a 1mL syringe was inserted
in each fruit to draw air from the cavity. The experiment was repeated on three harvest dates,
with three replicates/fruit rind category/genotype/harvest date.

72

Statistical analysis
Data were analyzed by analysis of variance (ANOVA) using Proc GLM or Mixed
protocol in SAS (SAS Institute, Cary, NC).

Results:
Ethylene and sex expression
Introduction of the CRC::etr1-1 transgene into Hales via Agrobacterium-mediated
transformation was described in Little et al. (2007). qRT-PCR analysis of T2 transgenic plants
detected etr1-1 transgene expression in apical tissue of M5 and M15 plants (Figure 3.1a).
Inhibition of ethylene perception is frequently associated with increased ethylene production
(Wilkinson et al., 1997; Little et al., 2007). A 2-6 fold increase in apical ethylene evolution was
observed in transgenic PCR positive plants (Figure 3.1b). Studies of ethylene evolution in
gynoecious cucumber apices have shown two-fold greater ethylene compared to their
monoecious counterparts, providing evidence for the biological significance of two-fold
differences in modulating sex determination (Rudich et al., 1972; Yamasaki et al., 2003).
Our previous greenhouse studies of T1 plants (Little et al., 2007) observed an earlier
onset and increased production of carpel-bearing flowers. These phenotypes were also observed
in greenhouse trials with T2 progeny. The CRC::etr1-1 plants showed a significant increase in
number of open carpel-bearing flowers on the main stem, and earlier onset by an average of 6
nodes (Table 3.1). A second phenotype observed in the T1 progeny and again in the T2
generation, was conversion of the CRC::etr1-1 carpel-bearing flowers from bisexual to female

73

etr1-1 expression in apices
Relative expression (UBQ7)

a.
180
160
140
120
100
80
60
40
20
0

b

c
a

Hales

b.

a

Azy

CRC::etr1-1 CRC::etr1-1
M5
M15
Ethylene evolution in apices

Ethylene (pmol/g)

250

b

200
150

ab
100

50

a

0
Control
CRC::etr1-1 M5 CRC::etr1-1 M15
Figure 3.1. qRT-PCR analysis of expression of the etr1-1 transgene (a), and ethylene evolution
(b) in apices of Hales (WT), Azygous (Azy), and CRC::etr1-1 M5 and M15 plants (Control in
(b) is a combination of Hales and Azygous samples). Values in qRT-PCR are the average of
three biological replicates with three technical replicates/biological rep + SE. Values presented in
ethylene evolution are the average of two experiments with 5-6 replicates with 1-2 plants/rep and
1-2 samples/plant/genotype/experiment + SE. Different letters represent significant differences
by LSD at P<0.05.

74

Table 3.1. Greenhouse observations of T2 generation of CRC::etr1-1 M5 and M15.

Genotype

# carpelbearing buds

First node of
carpel-bearing
bud

# open carpelbearing
flowers

First node of
open carpelbearing flower

% open
bisexual
flowers

% open female
flower

Hales

22.7 + 0.7a

9.6 + 0.6a

3.0 + 1.2a

19.5 + 4.2a

100 + 0.0a

0 + 0.0a

Azygous

23.3 + 0.7a

9.0 + 0.8a

1.7 + 0.32a

17.8 + 4.6a

100 + 0.0a

0 + 0.0a

CRC::etr1-1
M5

24.5 + 0.7a

10.0 + 0.9a

5.8 + 1.2b

11.8 + 0.9b

59.6 + 7.4b

40.4 + 7.4b

CRC::etr1-1
M15

26.0 + 1.6a

11.2 + 1.2a

7.0 + 0.3b

13.5 + 1.1b

57.3 + 4.9b

42.7 + 4.9b

Values presented are the average of two experiments with 6 replicates/genotype/experiment + SE. Different letters represent
significant differences by LSD at P<0.05.

75

(i.e. suppression of stamens). Approximately 40% of the CRC::etr1-1 carpel-bearing flowers
were female in both lines, a phenotype that is not observed in the andromonoecious melons
(Table 3.1).
Earlier and increased carpel-bearing flowers could be of possible value for melon
production if it is manifested in the field and results in increased and earlier fruit set. The first
field experiment was conducted to examine sex expression phenotypes. Both the M5 and M15
CRC::etr1-1 lines showed an increase in the number of carpel-bearing buds and open flowers,
and earlier onset of the first open carpel-bearing flower on the main stem by approximately 7-10
nodes, or 8-10 days (Figure 3.2a-c). A ~50% conversion of the typically bisexual flowers to
female flowers also was observed in the field, coinciding with our greenhouse studies (Figure
3.2d). The plants did not exhibit obvious differences in vegetative characteristics.

Fruit set and yield
The second field experiment was conducted to assess the influence of the CRC::etr1-1
transgene on fruit set, development, and yield. M5 and M15 transgenic fruit showed increased
average fruit set, especially early in the season, consistent with the earlier appearance of carpelbearing flowers (Figure 3.3). Total number of fruit at harvest was increased in the CRC::etr1-1
lines compared to wild type (Table 3.2). Total yield, as measured by kg/plot, did not significantly
differ from wild type plants (Table 3.2). The transgenic fruit was generally smaller, i.e. had
lower fruit weight (Table 3.2).

Fruit development and ripening
M5 and M15 transgenic fruit were more elongated, as evidenced by a higher

76

Carpel-bearing buds on main stem
80
b
60
b
a
40
20
0

CRC::etr1-1CRC::etr1-1
M5
M15
c. Carpel-bearing open flowers on main
d.
stem
10
b
8
b
6
40
4
a
20
2
a
0
0
Hales
CRC::etr1-1 CRC::etr1-1
Hales
CRC::etr1-1CRC::etr1-1
M5
M15
M5
M15
Figure 3.2. Field observations of number of carpel-bearing buds (a), open flowers (b), node of first carpel-bearing open flower on the
main stem (c), and percentage of female vs. bisexual open flowers (d) of Hales (WT), and CRC::etr1-1 M5 and M15. Values for sex
expression data are presented as the average of 6 replicates with 2-3 plants/rep + SE. Different letters represent significant differences
by LSD at P<0.05.
%

# carpel-bearing open
flowers

Hales

Node of first carpel-bearing open
flower
a
30
25
b
20
b
15
10
5
0
Hales
CRC::etr1-1 CRC::etr1-1
M5
M15
Percentage of female open flowers
80
b
60
b

b.

node position on main
stem

# carpel-bearing buds

a.

77

c

9
8
7
6
5
4
3
2 a
1
0

b
b

WT
Azy

a
a

M5

8/8

8/6

8/4

8/2

7/31

7/29

7/27

7/25

7/23

M15

7/21

7/19

# fruit

Average fruit set per plant

Date
Figure 3.3. Average fruit set per plant for Hales (WT), azygous, and CRC::etr1-1 M5 and M15
plants in the summer of 2011. Values are the average of 4 replicates with 3 plants/rep + SE.
Different letters represent significant differences by LSD at P<0.05.

78

Table 3.2. Field observations of fruit yield parameters.
Parameter
Hales
Azygous
(WT)
Total fruit/plot
42 + 4.2z
40 + 3.8

CRC::etr1-1
M5
M15
52.5 + 4.6
51.5 + 4.1

Nontransgenic
41 + 2.6ay

CRC::etr1-1
52 + 2.9b

Total fruit yield
(kg/plot)

60.3 + 4.6

56.0 + 4.7

63.2 + 6.8

60.5 + 6.1

58.2 + 3.1a

61.9 + 4.3a

Average fruit
weight (kg)

1.5 + 0.0

1.40 + 0.1

1.20 + 0.0

1.17 + 0.0

1.45 + 0.1a

1.19 + 0.0b

z

Values are presented are the average + SE of 4 replicates with 10 plants/genotype/rep.

y

Comparison of non-transgenic (Hales and azygous) vs. CRC::etr1-1 transgenic plants by orthogonal contrast. Different letters
represent significant differences at P<0.05.

79

length/diameter ratio (LD) (Figure 3.4a). Examination of carpel-bearing flowers showed that this
difference was already apparent in ovaries of female and bisexual flowers (Figure 3.4b). No
difference however was seen between the LD of female versus bisexual flowers (Figure 3.4b).
To determine effects on fruit ripening, fruit exocarp color at final harvest was categorized
as: green, turning (mix of green and orange), or orange. Fruit at full slip were orange, but could
be easily removed from the vine. At the final harvest date, 100% of the of the CRC::etr1-1 M5
fruit showed external ripening (classified as turning, orange, or full slip), compared to ~50% of
wild type, and ~30% of the M15 line (Figure 3.5a). Despite equivalent timing of fruit set, a
majority of the M15 fruit did not show external ripening signs. Examination of abscission
showed that a significantly reduced portion of M15 fruits had reached full slip compared to both
M5 and wild type fruit (Figure 3.5b). Although there was a lack of external ripening phenotypes,
when full size M15 green fruits were cut open, the mesocarp was orange and frequently
accompanied by over-ripe characteristics (i.e., mesocarp degradation, septum separation from
seeds, and/or septum degradation)(Figure 3.6a). Measurements of ethylene evolution from
whole fruit showed that orange fruit of all lines had higher ethylene evolution than green fruit
(Figure 3.6b). Similarly, samples of internal ethylene levels from taken from the seed cavity of
green fruit showed that Hales, azygous, and M5 lines all had low ethylene levels. However, M15
green fruit showed high ethylene levels comparable to those of orange fruit (Figure 3.6c).
A possible explanation for the green exterior, but overly ripe interior, is that expression of
the etr1-1 transgene was inhibiting perception of ethylene in the exocarp but not the mesocarp.
There was a general trend of increased etr1-1 expression as the transgenic fruit developed

80

a.

Average fruit L/D
2.05

b

2

b

L/D

1.95
1.9

a

a

1.85
1.8
1.75
1.7
0

Hales

b.
2.4
2.3

Azy

CRC::etr1-1 CRC::etr1-1
M5
M15

Average open flower ovary L/D
b b
b b

L/D

2.2
2.1

a

Bisexual

2

Female

1.9
1.8

a

1.7
0
Hales

CRC::etr1-1 CRC::etr1-1
M5
M15

Figure 3.4. Average fruit (a) and open bisexual and female flower ovary (b) length/diameter ratio
on wild type (WT) and CRC::etr1-1 M5 and M15. Values for fruit L/D are presented as the
average + SE of 35-40 fruit/genotype. Values for ovary L/D are the average of data combined
from two experiments with 6 plants/genotype/experiment + SE. Different letters represent
significant differences by LSD at P<0.05.

81

%

a.

% visibly ripe fruit at harvest
% visibly ripe tagged at finalfinal harvest
100
90
80
70
60
50
40
30
20
10
0

b

a

a
a

WT

Azy

M5

b.
70

a

60
50
%

M15

% abcission from vine at final harvest
% abscission from vine at final harvest

40

b

b

30

c

20

10
0
WT

Azy

CRC::etr1-1 CRC::etr1-1
M5
M15

Figure 3.5. Percentage of harvested fruit that are visibly ripe (visibly ripe fruit were classified as
turning, orange and full slip) (a), and abscission from the vine (b) at final harvest. Values are
average + SE of four replicate plots (35-40 fruit/genotype). Different letters represent significant
differences by LSD at P<0.05.

82

a.

Whole fruit ethylene

Green Hales

Ripe Hales

Ethylene (pmol/g)

b.
700
600
500
400
300
200
100
0

b b
b
b
a

a a

a
Green

Ripe

Internal fruit ethylene

c.

Green Seed separation and
exocarp septum degradation
CRC::etr1-1 M15

Ethylene (uL/L)

200

Mesocarp
degradation

WT
Azy
CRC::etr1-1 M5
CRC::etr1-1 M15

b

b

150

b
b

100
50

a a

a

b

WT
Azy
CRC::etr1-1 M5
CRC::etr1-1 M15

0
Green
n=12

Ripe
n=12

Figure 3.6. Mesocarp of Hales (green and ripe), and CRC::etr1-1 M15 fruit (a), whole fruit ethylene evolution (b) and internal fruit
ethylene (c) of wild type, azygous, and CRC::etr1-1 M5 and M15. Values for whole and internal fruit ethylene are the average of data
combined from three sample dates with three replicates of 3 fruit/rep + SE. Different letters represent significant differences by LSD
at P<0.05.
83

(Figure 3.7). Fruit approaching the turning stage (35dpp) showed an increase in transgene
expression in the exocarp, especially in M15 fruit (Figure 3.7).

Discussion:
Greenhouse and field observations of the T2 generation of CRC::etr1-1 M5 and M15
plants showed increased and earlier onset of carpel-bearing flowers, increased early fruit set and
modified fruit ripening characteristics. The phenotypes observed for the CRC::etr1-1 lines were
specific to floral and fruit tissue, unlike 35S::etr1-1 melons, which showed a variety of
vegetative and developmental phenotypes (Little et al., 2007). Typically increased femaleness
has been associated with exogenous application or increased endogenous ethylene (Rudich et al.,
1969; Karchi 1970; Papadopoulou et al., 2005). Transgenic melons constitutively expressing the
ethylene biosynthetic enzyme, ACS, or ACS targeted to stamen primordia, showed higher
ethylene evolution, and increased and earlier onset of bisexual flowers (Papadopoulou et al.,
2005; Taft et al., 2010). Here we observed increased femaleness in the CRC::etr1-1 plants,
despite inhibition of ethylene perception by carpel primordia. A possible explanation is that
failure of the carpel primordia to perceive ethylene results in increased ethylene production due
to lack of feedback inhibition, as has been observed frequently in non-ethylene perceiving
mutants and transgenics (Barry and Giovannoni, 2007), and was observed in the CRC::etr1-1
apices. The increased ethylene produced by the CRC::etr1-1 apices, could still be perceived by
the stamen primordia, which are the organs that regulate carpel development in
andromonoecious melons (Little et al., 2007). Ethylene perception by stamen primordia is
required to promote carpel development, and would result in increased production of carpelbearing buds.
84

etr1-1 expression in fruit tissue

140

Relative expression (UBQ7)

c
120
100
80
b

60

b
b

40
20

b
a
a

0

Mesocarp
Exocarp

a

a

a a

a

10mm 60mm 110mm 10mm 60mm 110mm
CRC::etr1-1 M5
CRC::etr1-1 M15
Figure 3.7. qRT-PCR analysis of expression of the etr1-1 transgene in 10mm diameter (~3dpp),
60mm diameter (~14dpp) and 35dpp fruit exocarp and mesocarp of CRC::etr1-1 M5 and M15
fruits. Values are the average of three biological replicates with three technical
replicates/biological replicate + SE. Different letters represent significant differences by LSD at
P<0.05.

85

The conversion of approximately 50% of the bisexual flowers to female which was
observed in CRC::etr1-1 plants, may be a further manifestation of this phenotype. One of the
major sex genes in melon, the A locus, which is responsible for stamen inhibition in carpelbearing flowers is an ACS enzyme (Boualem et al., 2008), and inhibition of stamens within
bisexual flowers has been observed as a response to application of ethylene in melon (Rudich et
al., 1969; Karchi, 1970). Increased apical ethylene evolution that can be perceived by the
stamens during carpel bud development could subsequently lead to their own inhibition.
Increased or earlier carpel-bearing flower production was associated with increased fruit
set on the CRC::etr1-1 plants, especially early in the season. Total yield, however, was not
different from wild type. It appears that the increased initial fruit set on the CRC::etr1-1 lines
was compensated for by a combination of abscission of young fruit, and smaller final fruit size.
This suggests that higher competition for photosynthetic resources of the plant might have
occurred in the CRC::etr1-1 plants, either due to increased fruit number or earlier fruit set on
younger, smaller plants. Competition for resources has been observed to regulate the balance
between number of carpel-bearing flowers and fruit set in Cucurbita species (Stephenson et al.
1988; Krupnick et al. 1999; Avila-Sakar et al. 2001). Cucumber thinning has been demonstrated
to increase fruit size and decreased assimilates within the plant, leading to fruit abortion of young
fruits (Hikosaka and Sugiyama, 2005).
At final harvest, all of the M5 transgenic fruit showed a ripe phenotype. The earlier
occurrence of ripe fruit in this line is likely due to earlier fruit set. In contrast, the M15 line did
not show this phenotype; instead a higher proportion did not appear ripe based on external rind
color and failure to slip from the vine. However, when M15 fruit classified as green were cut
open, a majority showed that over-ripening had occurred within the mesocarp. Externally

86

evolved ethylene levels were consistent with those of green fruit for wild type and M5 fruit but,
internal ethylene measurements of these fruit were on the level of ripe (orange) fruit of wild type
and M5. The elevated ethylene would be expected during climacteric ripening (Barry and
Giovannoni, 2007; Paul et al., 2012), and is consistent with the accelerated mesocarp ripening
that was not externally visible.
The non-ripening rind (green) and elongated phenotype is similar to the spontaneous
tomato mutant Never-ripe 2 which was linked to ethylene perception inhibition (Barry et al.,
2005). The external and internal ripening phenotypes observed could be explained by expression
of the etr1-1 transgene under the CRC promoter, targeting ethylene insensitivity to the carpel and
ultimately to the exocarp/rind. A study by Fernandez et al. (2009) looking at tomato plants with
the CRC promoter fused to GUS observed that expression was specifically localized to the fruit
epidermis, but not mesocarp, throughout the ripening process. The expression of the etr1-1
transgene was elevated in the exocarp of M15 fruit approaching the age that would be turning
stage in wild type fruit, possibly explaining the observed external non-ripening. Earlier work
with transgenic 35S::ACO-antisense melons inhibiting the ethylene ripening processes showed
that rind yellowing, abscission, and a majority of flesh softening processes are ethylenedependent (Ayub et al., 1996; Pech et al., 2008). Although elevated expression was also
observed in the exocarp of the M5 line at 35dpp, it was comparable to levels within mesocarp of
M15 fruit. This suggests that etr1-1 expression within the M5 exocarp and the M15 mesocarp
was not high enough to inhibit ripening, while the greater expression within the M15 exocarp
was sufficient to inhibit ethylene-dependent external ripening.
In summary, these studies show that inhibition of ethylene perception targeted to the
carpel and nectary primordia modifies sex expression in the field as manifested by earlier onset

87

and increased carpel-bearing flowers, and a ~50% loss of stamens in carpel-bearing flowers
(conversion from bisexual-to-female). Fruit set and total fruit number was increased, with earlier
ripening apparent in the M5 line, however, M15 fruit showed inhibited external ripening. These
results provide insight into the role of ethylene production and perception in carpel development,
translating into earlier and increased fruit set. If regulated properly, manipulation of the ethylene
signaling pathway could provide a shorter field growing season with increased and earlier fruit
set, however a balance needs to be struck to maintain desired fruit size. These studies also
provide further evidence for ethylene’s involvement in external fruit ripening (i.e. rind
degreening and abscission from the vine) in melon, and coordinated regulation of exocarp and
mesocarp during normal ripening processes. Collectively, these results demonstrate potential for
moderating melon fruit set and ripening by targeted regulation of ethylene perception.

88

LITERATURE CITED

89

LITERATURE CITED

Avila-Sakar G, Krupnick GA, Stephenson AG (2001). Growth and resource allocation in
Cucurbita pepo spp. texana: effects of fruit removal. Int J Plant Sci 162: 1089-1095.
Ayub R, Guis M, BenAmor M, Gillot L, Roustan JP, Latche A, Bouzayen M, Pech JC (1996).
Expression of ACC oxidase antisense gene inhibits ripening of cantaloupe melon fruits. Nat
Biotechnol 14: 862-866.
Bapat VA, Trivedi PK, Ghosh A, Sane VA, Ganapathi TR, Nath P (2010). Ripening of fleshy
fruit: Molecular insight and the role of ethylene. Biotechnol Adv 28: 94-107.
Barry CS, McQuinn RP, Thompson AJ, Seymour GB, Grierson D, Giovannoni JJ (2005).
Ethylene insensitivity conferred by the Green-ripe and Never-ripe 2 ripening mutants of tomato.
Plant Physiol 138: 267-275.
Barry CS and Giovannoni JJ (2007). Ethylene and fruit ripening. Plant Growth Regul 26: 143159.
Boualem A, Fergany M, Fernandez R, Troadec C, Martin A, Morin H, Sari M, Collin F, Flowers
J, et al. (2008). A conserved mutation in an ethylene biosynthesis enzyme leads to
andromonoecy in melons. Science 321: 836-838.
Den Nijs A, Visser D (1980). Induction of male flowering in gynoecious cucumbers (Cucumis
sativus L.) by silver ions. Euphytica 29:237-280.
Fernandez AI, Viron N, Alhagdow M, Karimi M, Jones M, Amsellem Z, Sicard A, Czerednik A,
Angenent G, Grierson D, May S et al. (2009). Flexible tools for gene expression and silencing in
tomato. Plant Physiol 151:1729-1740.
Grumet R and Taft JA (2011). Sex expression in cucurbits In: Genetics, Genomics and Breeding
in Cucurbits. Kole C, Wang YH, Behera TK, Kole C (eds). Science Publishers Inc. Enfield NH
and CRC Press (Taylor & Francis) group, Boca Raton, FL. p. 353-375.
Guis M, Botondi R, BenAmor M, Ayub R, Bouzayen M, Pech JC, Latche A (1997). Ripeningassociated biochemical traits of cantaloupe Charentais melons expressing an antisense ACC
oxidase transgene. J Amer Soc Hort Sci 122: 748-751.
Hikosaka S and Sugiyama N (2005). Effect of fruit-load on growth patterns of fruit at the middle
nodes of gynoecious-type cucumbers. J Hortic Sci Biotech 80:130-134.
Karchi Z (1970). Effects of 2-chloroethanephosphonic acid on flower types and flowering
sequences in muskmelon. J Amer Soc Hort Sci 95: 575-578.

90

Krupnick GA, Brown KM, Stephenson AG (1999). The influence of fruit on the regulation of
internal ethylene concentration and sex expression in Cucurbita texana. Int J Plant Sci 160: 321330.
Little HA, Papadopoulou E, Hammar SA, Grumet R (2007). The influence of ethylene
perception on sex expression in melon (Cucumis melo L.) as assessed by expression of the
mutant ethylene receptor, At-etr1-1, under the control of constitutive and floral targeted
promoters. Sex Plant Reprod 20: 123-136.
Owens KW, Peterson CE, Tolla GE (1980). Production of hermaphrodite flowers on gynoecious
muskmelon by silver nitrate and aminoethoxyvinylglycine. HortScience 15:654-655.
Papadopoulou E, Little HA, Hammar SA, Grumet R (2005). Effect of modified endogenous
ethylene production on sex expression, bisexual flower development, and fruit production in
melon (Cucumis melo L.). Sex Plant Reprod 18: 131-142.
Paul V, Pandey R, Srivastava, GC (2012). The fading distinction between classical patterns of
ripening in climacteric and non-climacteric fruit and the ubiquity of ethylene – An overview. J
Food Sci Tech Mys 49: 1-21.
Pech JC, Bouzayen M, Latche A (2008). Climacteric fruit ripening: Ethylene-dependent and
independent regulation of ripening pathways in melon fruit. Plant Sci 175: 114-120.
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 (eds. CC
Ainsworth). BIOS Scientific, Oxford: 189-215.
Roy RP and Saran S (1990). Sex expression in the Cucurbitaceae. In: Biology and utilization of
the Cucurbitaceae. Bates DM, Robinson RW, Jeffrey C (eds). Cornell University Press, Ithaca,
NY: 251-268.
Rudich J, Halevy AH, Kedar N (1969). Increase in femaleness of three cucurbits by treatment
with ethrel, an ethylene-releasing compound. Planta 86: 69-76.
Rudich J, Halevy AH, Kedar N (1972). Ethylene evolution from cucumber plants related to sex
expression. Plant Physiol 49:998-999.
Stephenson AG, Devlin B, Horton JB (1988). The effects of seed number and prior fruit
dominance on the pattern of fruit production in Cucurbita pepo (zucchini squash). Annu Bot 62:
653-661.
Taft J, Little H, Hammar S, Grumet R (2010). Effect of floral primordia-targeted ethylene
production on sex determination in melon (Cucumis melo L). In: Cucurbitaceae 2010
Proceedings. Thies JA, Kousik S, Levi A (eds). American Society of Horticultural Science,
Alexandria VA: 159-162.

91

Wilkinson JQ, Lanahan MB, Clark DG, Bleeker AB, Chang C, Meyerowitz EM, Klee HJ (1997).
A dominant mutant receptor from Arabidopsis confers ethylene insensitivity in heterologous
plants. Nat Biotechnol 15: 444-447.
Yamasaki S, Fujii N, Takahashi H (2003). Characterization of ethylene effects on sex
determination in cucumber plants. Sex Plant Reprod 16:103-111.

92

CHAPTER 4
Expression characterization of sex genes and ethylene biosynthetic genes in different sex
genotypes of melon (Cucumis melo L.)

93

Introduction:
Melon (Cucumis melo L.) plants have the capacity for a variety of sex expression
phenotypes including andromonoecious (male and bisexual flowers), monoecious (male and
female flowers), gynoecious (female-only flowers), and hermaphrodite (bisexual-only flowers).
Differentiation to produce unisexual flowers occurs by suppression of either carpel primordia or
stamen primordia at key development stages (Goffinet, 1990; Bai et al., 2004; see chapter 2).
Control of melon (Cucumis melo L.) sex expression involves two major loci, Gynoecious (G) and
Andromonoecious (A), as well as possible modifier loci (Kubicki, 1969; Kenigsbuch and Cohen,
1990; Roy and Saran, 1990). The G locus is involved in inhibition of early carpel development
within male floral buds, allowing for separate male flowers to occur in dominant G genotypes, i.e.
andromonoecious (GGaa) and monoecious (GGAA) (Kenigsbuch and Cohen, 1990; Roy and
Saran, 1990). Stamen inhibition in carpel-bearing flowers is controlled by the A locus, wherein
dominant A leads to suppression of stamens to produce a female flower instead of bisexual
(Kenigsbuch and Cohen, 1990; Roy and Saran, 1990). Table 4.1 outlines the different sex
genotypes of melon and their corresponding phenotypes.
Ethylene is a key factor regulating melon sex determination (Rudich, 1990; Grumet and
Taft, 2011). Increased exogenous and endogenous ethylene levels are known to increase
femaleness as manifested by increased carpel-bearing bud production and/or suppression of
stamens (Rudich et al., 1969; Karchi, 1970; Papadopoulou et al., 2005). In contrast, inhibition of
ethylene production or perception causes increased maleness as shown by increased male bud
number or the conversion of female flowers to bisexual flowers (Den Nijs and Visser, 1980;
Owens, 1980; Little et al., 2007). The ethylene biosynthetic pathway begins

94

Table 4.1. Melon sex expression patterns with corresponding genotypes and phenotypes.
Sex expression pattern

Genotype

Phenotype

Andromonoecious

Ga

♂

Monoecious

GA

♂ ♀ - Male and female flowers

Gynoecious

gA

♀ - Female flowers only

Hermaphrodite

ga

- Male and bisexual flowers

- Bisexual flowers only

95

with the S-AdoMet precursor acted upon by ACC synthase (ACS) to produce ACC that is then
oxidized by ACC oxidase (ACO) to form ethylene (Argueso et al., 2007).
The molecular identity of the A and G loci has recently been determined. The A locus
encodes an ethylene biosynthetic gene, ACC synthase 7 (CmACS7) that is expressed in carpel
primordia (Boualem et al., 2008). A fully functional ACS7 can inhibit stamen primordia to
consequently form a female flower, however mutant ACS7 (reduced enzymatic activity) cannot
suppress stamens, thereby forming bisexual flowers (Boualem et al., 2008).
The G locus encodes a transcription factor named Wound Inducible Protein1 (CmWIP1).
An insertion of a transposon into the WIP1 promoter leads to methylation and thus silencing of
the G locus, resulting in plants that make only female (gynoecious) or only bisexual
(hermaphrodite) flowers (Martin et al., 2009). Our studies have shown that expression of WIP1 is
regulated by ethylene. Increased ethylene causes a decrease in WIP1 transcript levels (see
chapter 2). The source of ethylene regulating WIP1 has not been determined, and may potentially
act as a modifier of sex expression. The melon G locus may be contrasted with the corresponding
F locus in cucumber (Trebitsh et al., 1997; Knopf and Trebitsh, 2006), Unlike G, F encodes an
ACC synthase, further indicating importance of ethylene in preventing suppression of carpel
primordia and potential involvement of ethylene biosynthetic enzymes in both sex determination
steps (Trebitsh et al., 1997; Knopf and Trebitsh, 2006).
Melon has several members in each of the ACS and ACO gene families that have variable
expression across plant tissues. Several of the ACS members (ACS1, ACS2, and ACS3) and
ACO1 were first identified in melon fruit as potential factors involved in the ripening process, or
in response to hormone application on seedlings (Miki et al., 1995; Lasserre et al., 1996; Ishiki et
al., 2000). Further characterization looked at expression of ACS and ACO in other plant tissues

96

such as seedlings, leaf and flowers, or response to stress (Lasserre et al., 1997; Ishiki et al., 2000).
A majority of these members have not been tested for expression in apex, floral tissue, or in
different sex genotypes of melon. With the recent sequencing of the melon genome (Garcia-Mas,
et al., 2012) and transcriptome data available for the closely related cucumber, we can also ask if
there are previously unreported members of the ACS and ACO family with potential roles in sex
determination.
In this study, we sought to characterize the ACS and ACO gene family members in
different melon sex genotypes to provide gene expression profiles within apices. Monoecious
and hermaphrodite lines were developed through a gynoecious x andromonoecious cross, and
segregation ratios matched a four gene model implicating two major loci and two modifying loci.
Three new ACS gene family members were identified from the available melon genome:
CmACS5, CmACS10, and CmACS12. Gene expression analysis of apices showed that transcript
levels of a majority of the ACS members and all of the ACO members were higher in the
gynoecious and hermaphrodite lines.

Materials and methods:
Plant genotypes and growing conditions
The gynoecious breeding line, Wisconsin 998 (WI998; ggAA) (kindly provided by Dr. J.
Staub, Univ Wisconsin) was crossed to the commercial andromonoecious line Hale’s Best
Jumbo (Hales; GGaa) to develop monoecious and hermaphrodite lines. The F1 hybrid progeny
(GgAa) were self pollinated to produce the F2 generation. Two plantings of the F2 were
performed in the greenhouse, each with 200- 400 seeds, to determine segregation ratios of sex
phenotypes. Plants were selected for those that only produced female flowers (gynoecious;
97

ggAA), or bisexual flowers (hermaphrodite; ggaa) along the first 10-15 nodes of the main stem.
To allow room for seed production, F2 plants that produced male flowers were removed from the
greenhouse at the appearance of the first male bud ; thus numbers for andromonoecious and
monoecious were pooled when testing F2 segregation ratios. The hermaphrodite plants were self
pollinated to produce F3 families, which were screened for homozygosity. Andromonoecious
(GGaa), monoecious (G-A-), gynoecious (ggAA), and hermaphrodite (ggaa) genotypes were
grown under greenhouse conditions described in Little et al. (2007), and used for analysis of
apical gene expression as described below.

Identification of ACS gene family members
Previously published sequences for ACS1, ACS2, ACS3, ACS7, ACO1, ACO2, and ACO3
were available at NCBI or ICuGI and used for primer design described below. To identify new
members of the ACS family, available cucumber 454 pyrosequencing data from Ando and
Grumet (2010) was searched for ACC synthases that had homology to other known ACS genes.
A corresponding melon EST was identified in ICuGI (MU48956; referred to here as CmACS10)
using the cucumber sequence to BLAST the melon EST database. Two additional members were
identified with homology to Arabidopsis ACS4, 5, 8, and ACS12, respectively, referred to here as
CmACS5 and CmACS12, using the newly sequenced melon genome (Garcia-Mas et al., 2012),
and provided to us by Jordi Garcia-Mas and Walter Sanseverino (ILTA, Spain).

qRT-PCR analysis
Total RNA was prepared from 200mg of liquid nitrogen frozen melon apex tissue (~3-

98

4mm in size) removed from plants at the 2 leaf stage (developing floral buds just entering stage
5; beginning of sex differentiation) using the Trizol method (Invitrogen, Carlsbad, CA). cDNA
preparation was performed using Superscript II reverse transcriptase (Promega, Madison, WI)
following manufacturer’s recommendations. Gene-specific primers for the ACS and ACO gene
family members were designed using Primer Express 3.0 software (Applied Biosystems);
the primer sequences are listed in Table 4.2. qRT-PCR analysis was carried out using a
Stratagene Mx4000 system with Power SYBR Green PCR Master Mix (Applied Biosystems). C.
melo Polyubiquitin 7 (ICuGI: MU43954) was used as an endogenous control for normalization.
qRT-PCR experiments were repeated twice with three biological replicates and three technical
replicates/biological replicate.

Statistical analysis
WIP1 and ACS member gene expression data were log2 transformed. Data were analyzed
by analysis of variance (ANOVA) using Proc Mixed protocol in SAS (SAS Institute, Cary, NC).

Results and discussion:
Segregation analysis of a gynoecious x andromonoecious cross
To study natural variation of gene expression in melon, four sex genotypes were
produced by a cross between andromonoecious Hales (GGaa) and gynoecious WI998 (ggAA).
The resulting F1 progeny (GgAa) showed the expected monoecious phenotype. Segregation
analysis of the sex phenotypes of the F2 progeny were tested for a 2 gene, 3 gene, or 4 gene
model. The ratios were not consistent with a two gene model as the expected number of the
99

Table 4.2. Primers of melon ACS and ACO gene members for qRT-PCR analysis.
Primer

Forward sequence

Reverse sequence

qRT ACS1 3F(RG742)
qRT ACS1 3R (RG743)

5’-AAGACATGCAAGGTTCACAGAAGC-3’

5’-TCCTGCATTGTATTCAAGCA-3’

Cm-ACS2-3F qRT (RG634)
Cm-ACS2-3R qRT (RG635)

5’-GCAGAGCCCGGTTTCATTAG-3’

5’-CGCATTCAACGTCGTTCATG-3’

Cm-ACS3-2F qRT (RG636)
Cm-ACS3-2R qRT (RG637)

5’-TTGGTTACGAAGTGGGTTTGAATAT-3’

5’-ACCCTGGCTCAGTGCAATG-3’

qRT ACS5 1F (RG748)
qRT ACS5 1R (RG749)

5’-GGTTGGTTTAGAATGTGCTTTGC-3’

5’-ACGATGCATGGCCACCTT-3’

qRT-ACS7F (RG465)
qRT-ACS7R (RG464)

5’-TTTAGCGTCCATGTTGTCGAACCG-3’

5’-TCATACCGTTTCTTGAGCCTGTCC-3’

newACS-2F (RG644)
(ACS10)
newACS-2R (RG645)

5’-AGAGTGCAAACGGTTTCCGA-3’

5’-AGGTTAAGCTTTTGAGCTTGCTCGT-3’

qRT ACS12 F (RG744)
qRT ACS12 R (RG745)

5’-CAAAGGGAATGGGCATGGTG-3’

5’-CCACGAATTCAAAAGAAAGCTGGT-3’

qRT WIP1F (RG680)
qRT WIP1R (RG681)

5’-TAGGGCTTCCAACTCCTTCCTCT-3’

5’-CTTGCAATTGATGGGTGTGATCTTCTTG-3’

qRT ACO1F (RG477)
qRT ACO1R (RG476)

5’-GCTGCACAAGCTGAAGTTAATGA-3’

5’-TCAGGAAGATGGCGTAAGAAAAA-3’

qRT ACO2F (RG475)
qRT ACO2R (RG474)

5’-GACGGCAGATGGTTCAAAATC-3’

5’-TCGCCGATGTTGACGAAAA-3’

qRT ACO3F (RG473)
qRT ACO3R (RG472)

5’-AAGGAATTTGCAGATGAATTGGA-3’

5’-AGCCCAAGATTCTCACACAACAA-3'

CmUBQ7-3F (RG541)
CmUBQ7-3R (RG543)

5’-GACCGGGAAGGAAATTGAGAT-3’

5’-CCGCTCCTTAATACGATCAATTGG-3’

100

double recessive hermaphrodite phenotype in a total of 600 plants would be 37.5, however only
5 plants were hermaphrodite (Table 4.3). F3 progeny testing of the F2 progeny verified that the
hermaphrodite F2 plants were true-breeding for sex type, as expected for homozygous recessive
2

individuals. X values support a four gene model implicating two modifiers acting along with the
two major sex loci to produce a hermaphrodite plant. Prior analyses have led to sex expression
models for melon including only two genes, or as many as four genes (2 major and 2 modifying
sex loci) (Kubicki, 1969; Kenigsbuch and Cohen, 1990). Differences among the results likely
reflect different modifier alleles present in the starting parental genotypes.

Analysis of sex genes in different sex genotypes
Expression of the sex genes, G (WIP1) and A (ACS7), was examined in apices of the
different sex genotypes. Apices were collected from plants at the two leaf stage where the most
developed buds are just entering developmental stage 5 (Taft chapter 2).

At stage 5 all four

floral whorls have been initiated and sex differentiation begins; carpel primordia can be either
inhibited or promoted based on ethylene levels and WIP1 gene expression (Martin et al., 2009;
Taft chapter 2). WIP1 expression was highest in monoecious (G-A-) apices compared to apices
of andromonoecious (G-aa), gynoecious (ggAA) and hermaphrodite (ggaa) plants (Figure 4.1a).
This result matched the gene expression analysis in Martin et al. (2009) where monoecious buds
showed higher expression compared to gynoecious and hermaphrodite buds; andromonoecious
plants were not used for comparison in their study. This is expected as the G locus, which
encodes the transcription factor WIP1, is silenced in recessive genotypes due to a transposon
inserted upstream of the gene, however dominant genotypes do not have this insertion allowing

101

Table 4.3. Segregation for F2 population of gynoecious (WI998) x andromonoecious (Hales)
cross.
Genotype

Flower sex
phenotype

Models: Expected
Two gene

Exp.1

Three gene Four gene Observed Observed

36

Male,
female

9

Ga (P2)

Male,
bisexual

3

gA(P1)

Female1

3

112.5

9
3

112.5 3

ga

Bisexual

1

37.5

4

37.5 1

GA

450

P

0.01

Total

189
450

12

X 2 : 35.53**

Exp.2

196

393

589

8

3

3

6

2

1

4

5

590

63

29.93**

0.50

2.00

3.22

0.01

0.80

0.40

0.20

1

May also include gynomonoecious (female and bisexual) sex phenotype. ** denotes statistical
significance at P<0.01.

102

for expression of WIP1 (Martin et al., 2009). WIP1expression level was lower in
andromonoecious than monoecious, however the reason for this difference is not known.
Expression of the A gene, ACS7, which suppresses stamens in carpel-bearing buds
(Boualem et al., 2008) was nearly exclusively expressed in gynoecious apices compared to
andromonoecious, monoecious, and hermaphrodite apices, respectively (Figure 4.1b). This
nearly exclusive expression is expected since expression of the A locus would be higher in a
female genotype compared to bisexual, and buds developing within the apex at the two leaf stage
in monoecious plants are mostly male (Taft chapter 2). This profile matches the results for
gynoecious and hermaphrodite plants observed by Boualem et al. (2008) and Martin et al. (2009),
where ACS7 showed higher expression in female buds compared to bisexual buds. The
equivalent locus in cucumber, M locus, which is involved in stamen inhibition in carpel-bearing
buds, also showed the same expression profile for gynoecious and hermaphrodite lines (Boualem
et al., 2009).

Identification of ACS and ACO gene family members
There are 4 melon ACS and 3 ACO gene family members that have been previously
identified; ACS1, ACS2, ACS3, ACS7, ACO1, ACO2, and ACO3 (Miki et al., 1995; Lasserre et al.,
1996, 1997; Ishiki et al., 2000; Boualem et al., 2008; Martin et al., 2009). Examination of our
available cucumber 454 pyrosequencing data set from Ando and Grumet (2010) identified a new
member of the ACS family with homology to Arabidopsis ACS10 and 12, and a corresponding

103

WIP1 expression in apices
WIP1 expression in apices

c

Relative expression (UBQ7)

a.
a.

0.07

a

0.06
0.05
0.04
0.03
0.02

b

c

GGaa

ggAA

ggaa

0.01
0

G-A-

ACS7 expression in apices
ACS7 expression inapices

b.
b.

Relative expression (UBQ7)

b

a

120

100
80
60
40
20

b

b

G-A-

b

GGaa

0
ggAA

ggaa

Figure 4.1. Expression of the melon sex genes WIP1 (a) and ACS7 (b) in monoecious (G-A-),
andromonoecious (GGaa), gynoecious (ggAA), and hermaphrodite (ggaa). Values are the
average of three biological replicates with three technical replicates/biological rep + SE.
Different letters represent significant differences by LSD at P<0.05.

104

melon EST was identified in ICuGI (MU48956; referred to here as CmACS10). A phylogenetic
tree of the ACS gene family in melon was constructed by Sanserevino et al. (personal
communication) using sequences from the newly available melon genome (Figure 4.2) (GarciaMas et al., 2012). Two additional members were identified with homology to Arabidopsis ACS4/
5/ 8 clade, and ACS12, referred to here as CmACS5 and CmACS12. Examination of the ACS
phylogenetic tree shows that the melon ACS members are most closely related to cucumber ACS
members, as would be expected; the same set of family members was identified in both species.
No new ACO gene family members were identified from the melon genome beyond the three
already known.

Analysis of ACS and ACO gene family members in different sex types
Expression of the different ACS and ACO gene family members was examined in young
apices of the four sex genotypes (Figure 4.3). ACS2, ACS5, and ACS10 were detectable within
the apex (Figure 4.3a). Apices from gynoecious (ggAA) plants showed the highest expression of
ACS2, ACS5, and ACS10, while hermaphrodite showed higher expression of ACS5 and ACS10
compared to monoecious and andromonoecious (Figure 4.3a).Similarly, expression of all three
ACO members also was highest in the gynoecious apices with second highest values in
hermaphrodite buds compared to andromonoecious and monoecious genotypes (Figure 4.3b).
Expression of ACS5, ACS10, and all the ACO members (ACO1, ACO2 and ACO3) were
higher in recessive g genotypes (gynoecious and hermaphrodite) compared to dominant G
(Figure 4.3b). A possible explanation is that the increased expression of ACS and ACO members
increases ethylene levels, reducing WIP1 (G) expression in order to inhibit male bud production

105

Figure 4.2. Rooted phylogenetic tree of ACS gene family members across different species provided by Jordi Garcia-Mas and Walter
Sanseverino (personal communication).
106

Figure 4.2 (cont’d)

107

Figure 4.2 (cont’d)

108

Figure 4.2 (cont’d)

109

ACS gene family members expression in apices

a.
80

a

75

70

a

65

Relative expression (UBQ7)

60

a

55
G-AGGaa
ggAA
ggaa

50
45
40
25

ab

20
b

15
10

b

b

5
a
0 ab b ab
ACS1

c

c

b

ACS2

bb

aa

ACS3

c
ACS5

c
ACS10

b ab

a

ACS12

Figure 4.3. Expression of ACS (a) and ACO (b) gene family members in monoecious (GA),
andromonoecious (Ga), gynoecious (gA), and hermaphrodite (ga). Values are the average of
three biological replicates with three technical replicates/biological rep + SE. Different letters
represent significant differences by LSD at P<0.05.

110

Figure 4.3 (cont’d)

ACO gene family members expression in apices

b.

Relative expression (UBQ7)

60
a

50

a

a

40

G-AGGaa
ggAA
ggaa

30

20
b

10

b
c

c

c

c

0
ACO1

ACO2

111

b
c

c
ACO3

to facilitate all carpel-bearing phenotypes (g genotypes). However, the g locus in all carpelbearing genotypes is recessive due to silencing of the WIP1 transcription factor and loss of the
inhibition of carpel development, thus ethylene would have to interfere with methylation of the
WIP1 transcription factor in order to cause male flowers and change to a G genotype.
Alternatively, it is possible that the reduced expression of the ACS and ACO genes in
andromonoecious and monoecious genotypes is an effect of the dominant G locus, leading to
suppressed expression of ethylene related.
Unlike the recessive g allele in melon, the cucumber allele responsible for preventing
suppression of carpel development, is the dominant F, F encodes an extra ACS1, namely ACS1G
and was a product of gene duplication (Trebitsh et al., 1997; Knopf and Trebitsh, 2006). The
homolog to ACS1G in melon is CmACS3; unlike CsACS1 and 1G which are located in tandem in
cucumber, CmACS1 and CmACS3 in melon are located on linkage groups 11 and 8, respectively.
Expression of CmACS3 had not been examined in apices to determine if it is related to sex
determination, however it is a possible modifier candidate. ACS3 expression was higher in
gynoecious and hermaphrodite, however compared to other ACS members, the gene expression
was very low.
In summary, gene expression profiles of the ACS and ACO gene family members in
different melon sex genotypes were examined within apices to determine if these ethylene
production genes were implicated in sex determination. A cross between gynoecious x
andromonoecious plants was used to develop monoecious and hermaphrodite lines. Segregation
of the F2 generation matched a four gene model with two modifying loci in order to form the
hermaphrodite genotype. Expression profiles of the ethylene production gene family members of
ACS and ACO showed that genotypes that produce all carpel-bearing buds (gynoecious and
112

hermaphrodite) tended to show higher expression of all family members within the apex. It is
possible that higher ethylene produced by higher gene expression of the biosynthetic enzymes
suppresses a functional G (WIP1), leading to promotion of carpels and a greater number of
carpel-bearing buds. An alternative hypothesis is that a functional G effects gene expression of
the ACS and ACO genes within the apex to reduce their expression and increase the number of
male buds. This suggests that one of more ACS or ACO members may act as modifiers in sex
determination.

113

LITERATURE CITED

114

LITERATURE CITED

Ando K and Grumet R (2010). Transcriptional profiling of rapidly growing cucumber fruit by
454-pyrosequencing analysis. J Amer Soc Hort Sci 135: 291-302.
Argueso CT, Hansen M, Kieber JJ (2007). Regulation of ethylene biosynthesis. J Plant Growth
Regul 26: 92-105.
Bai SL, Peng YB, Cui JX, Gu HT, Xn LY, Li YQ, Xu ZH, 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.
Boualem A, Fergany M, Fernandez R, Troadec C, Martin A, Morin H, Sari M, Collin F, Flowers
J, et al. (2008). A conserved mutation in an ethylene biosynthesis enzyme leads to
andromonoecy in melons. Science 321: 836-838.
Boualem A, Troadec C, Kovalski I, Sari M, Perl-Treves R, Bendhamane A (2009). A conserved
ethylene biosynthesis enzyme leads to andromonoecy in two Cucumis species. PLoS ONE 4:
e6144.
Den Nijs A and Visser D (1980). Induction of male flowering in gynoecious cucumbers
(Cucumis sativus L.) by silver ions. Euphytica 29: 237-280.
Duan QH, Wang DH, Xu ZH, Bai SN (2008). Stamen development in Arabidopsis is arrested by
organ-specific overexpression of a cucumber ethylene synthesis gene CsACO2. Planta 228: 537543.
Garcia-Mas J, Benjak A, Sanseverino W, Bourgeois M, Mir G, Gonzalez VM, Henaff E, Camara
F, Cozzuto L et al. (2012). The genome of melon (Cucumis melo L.). P Natl Acad Sci 109:
11872-11877.
Goffinet MC (1990). Comparative ontogeny of male and female flowers in Cucumis sativus. In:
Biology and Utilization of the Cucurbitaceae. DM Bates, RW Robinson, C Jeffery (eds). Cornell
University Press, New York: 288-304.
Grumet R and Taft J (2011). Sex expression in cucurbits. In: Genetics, Genomics and Breeding
in Cucurbits. C Kole, YH Wang, TK Behera (eds). Science Publishers Inc., Enfield and CRC
Press (Taylor and Francis) group, Boca Raton: 353-375.
Ishiki Y, Oda A, Yaegashi Y, Orihara Y, Arai T, Hirabayashi T, Nakagawa H, Sato T (2000).
Cloning of an auxin-responsive 1-aminocyclopropane-1-carboxylate synthase gene (CMe-ACS2)
from melon and the expression of ACS genes in etiolated melon seedlings and melon fruits. Plant
Sci 159: 173-181.

115

Karchi Z (1970). Effects of 2-chloroethanephosphonic acid on flower types and flowering
sequences in muskmelon. J Amer Soc Hort Sci 95: 575-578.
Kenigsbuch D and Cohen Y (1990). The inheritance of gynoecy in muskmelon. Genome 33:
317-320.
Knopf R and Trebitsh R (2006). The female-specific CsACS1G gene of cucumber. A case of
gene duplication and recombination between the non-sex specific 1-aminocyclopropane-1carboxylate synthase gene and a branched chain amino acid transaminase gene. Plant Cell
Physiol 47: 1217-1228.
Kubicki B (1969). Sex determination in muskmelon (Cucumis melo L.) Genet Pol 10: 145-163.
Lasserre E, Bouquin T, Hernandez JA, Bull J, Pech JC, Balague C (1996). Structure and
expression of three genes encoding ACC oxidase homologs from melon (Cucumis melo L.). Mol
Genet Genomics 251: 81-90.
Lasserre E, Godard F, Bouquin T, Hernandez JA, Pech JC, Rody D, Balague C (1997).
Differential activation of two ACC oxidase gene promoters from melon during plant
development and in response to pathogen attack. Mol Genet Genomics 256: 211-222.
Little HA, Papadopoulou E, Hammar SA, Grumet R (2007). The influence of ethylene
perception on sex expression in transgenic melon as assessed by expression of the mutant
ethylene receptor gene, At-etr1-1, under control of constitutive and floral-targeted promoters.
Sex Plant Reprod 20: 123-136.
Martin A, Troadec C, Boualem A, Mazen R, Fernandez R, Morin H, Pitrat M, Dogimont C,
Bendhamane A (2009). A transposon-induced epigenetic change leads to sex determination in
melon. Nature 461: 135-1139.
Miki T, Yamamoto M, Nakagawa H, Ogura N, Mori H, Imaseki H, Sato T (1995). Nucleotidesequence of a cDNA for 1-aminocyclopropane-1-carboxylate synthase from melon fruits. Plant
Physiol 107:297-298.
Owens KW, Peterson CE, Tolla GE (1980). Production of hermaphrodite flowers on gynoecious
muskmelon by silver nitrate and aminoethoxyvinylglycine. HortScience 15:654-655.
Papadopoulou E, Little HA, Hammar SA, Grumet R (2005). Effect of modified endogenous
ethylene production on hermaphrodite flower development and fruit production in melon
(Cucumis melo L.). Sex Plant Reprod 18: 131-142.
Roy RP and Saran S (1990). Sex expression in the Cucurbitaceae. In: Biology and utilization of
the Cucurbitaceae. DM Bates, RW Robinson, C Jeffrey (eds). Cornell University Press, Ithaca:
251-268.
Rudich J, Halevy AH, Kedar N (1969). Increase in femaleness of three cucurbits by treatment
with ethrel, an ethylene-releasing compound. Planta 86: 69-76.
116

Rudich J (1990). Biochemical aspects of hormonal regulation of sex expression in Cucurbits. In:
Bates DM, Robinson RW, Jeffrey C (eds). Biology and utilization of the Cucurbitaceae. Cornell
University Press, Ithaca, NY: 269-280.
Trebitsh T, Staub JE, O’Neill SD (1997). Identification of a 1-aminocyclopropane-1-carboxylic
acid synthase gene linked to Female (F) locus that enhances female sex expression in cucumber.
Plant Physiol 113: 987-995.

117

CONCLUDING REMARKS

Melon and cucumber are model species of unisexual flower development that can be
governed by developmental stage, hormones, and genetics in order to cause a variety of sex
expression phenotypes (Roy and Saran, 1990; Perl-Treves, 1999; Grumet and Taft, 2011). The
development of a comprehensive model for melon sex determination that includes known
phenotypes, developmental stages, ethylene influences, and gene expression will provide a
greater understanding of the molecular aspects involved in this process.
To examine the influence of ethylene production that is perceived by the stamen
primordia during early flower development (Little et al., 2007), transgenic plants targeting
production to the stamen and anther (AP3::ACS), or carpel and nectary primordia (CRC::ACS)
were used. Melon plants with ethylene production targeted to the stamen primordia showed high
levels of ethylene evolution in the apex, and increased femaleness, as manifested by gain of a
bisexual-only phase not seen in wild type, and loss of male buds and open flowers along the
main stem. The most dramatic phenotype observed was a conversion to hermaphrodite instead of
andromonoecious. Microscopic analysis of apical tissue found that a smaller proportion of presex determination buds transitioned into the sex determination stages within the transgenic
AP3::ACS melons. In contrast, carpel primordia-targeted plants did not show a difference in sex
expression or gene expression for the native sex genes, providing further evidence in support of
stamen’s involvement in carpel development.
Native sex gene expression was also influenced in AP3::ACS plants as transcript levels of
the A locus (involved in stamen inhibition in carpel-bearing flowers; ACS7; Boualem et al.,
2008) were increased, and expression of the G locus (involved in carpel inhibition in male buds;

118

WIP1; Martin et al., 2009) was decreased. Exogenous application of ethylene also showed a
reduction in WIP1 gene expression, and promoter analysis identified ethylene-related motifs.
This provides evidence for the regulation of WIP1 expression by ethylene, and a link to genetic
control of sex expression by the G locus and ethylene. Taken together, these results provide
further evidence for ethylene thresholds acting within the apex during floral bud development,
and indicate critical developmental stages and levels for sex differentiation for promotion or
inhibition of sex organs.
Previous studies conducted in our lab using carpel and nectary primordia-targeted
ethylene perception inhibition (CRC::etr1-1) showed increased femaleness under greenhouse
conditions (Little et al., 2007). CRC::etr1-1 melons were then tested in a field setting to
determine if the floral sex phenotype would also be observed in the field, and if so, if there was
an effect on fruit set, development, and ripening. Increased femaleness was observed in the field
as manifested by earlier onset and increased carpel-bearing flowers, and a conversion of
bisexual-to-female flowers, consistent with greenhouse observations.
Transgenic plants set earlier and higher numbers of fruit compared to wild type, however
the fruit was smaller resulting in similar total kg/plot yields, most likely caused by resource
competition in the plants leading to young fruit abscission. Two different ripening phenotypes
were observed in CRC::etr1-1 fruit; one line (M5) showed 100% ripe fruit at final harvest, likely
due to earlier fruit set, and the other (M15) showed a comparable percentage to wild type,
however less M15 fruit abscised from the vine compared to wild type and M5. Externally, these
fruit had a green rind, similar to unripe melon, however internally the mesocarp showed signs of
over-ripening (mesocarp degradation, seed separation from septum, and/or septum degradation).
Internal ethylene levels of externally green M15 were comparable to those seen in orange (ripe)

119

fruit of M5 and wild type. Expression of the etr1-1 transgene using the CRC promoter could
explain this phenotype as expression was higher in the exocarp, targeting ethylene perception
inhibition to the exocarp, not mesocarp, and thus leading to the observable external non-ripening.
These results show potential for regulating melon fruit set and ripening by targeting ethylene
perception to the gynoecium.
Gene models of melon sex determination have been proposed with two major sex loci
and one or two modifier loci (Kubicki, 1969; Kenigsbuch and Cohen, 1980). One of the major
sex genes in melon was identified as a member of the ACS ethylene biosynthetic enzyme family,
ACS7 (Boualem et al., 2008). Melon has multiple ACS and ACO members (Miki et al., 1995;
Lasserre et al., 1996; Ishiki et al., 2000; Boualem et al., 2008) which have the potential to be
modifiers of sex determination. In order to examine this possibility, monoecious and
hermaphrodite sex genotypes from a gynoecious and andromonoecious cross were developed to
examine gene expression in the apices of these four genotypes. Segregation ratios of the F2
matched a four gene model, and three new ACS members were identified using available
cucumber sequences and the melon genome. Gene expression analysis revealed that transcript
levels of a majority of the ACS members and all of the ACO members were higher in recessive g
genotypes (gynoecious and hermaphrodite). However, gynoecious showed higher expression
compared to hermaphrodite. These results implicate that dominant G genotypes, namely a
functional WIP1 transcription factor (Martin et al., 2009), can suppress gene expression of
ethylene production genes.
Cumulatively, the results presented here provide insight into the molecular aspects of
unisexual floral development in melon and its link to ethylene. The molecular mechanism of
ethylene regulation of the G locus (WIP1) still needs to be investigated. Promoter analysis
120

identified putative sites for regulation by an ethylene-regulated protein, and examination of this
activity will be essential to solidify this connection. Another question is how does the stamen’s
perception of ethylene pass this signal to the carpels to allow for continued carpel development?
If it isn’t ethylene itself, what could this signal be? How does it affect WIP1 gene expression to
inhibit carpel development, or inhibit WIP1 to promote carpel development? What is the actual
identity of the two modifying genes involved in sex differentiation? Are the ethylene perception
genes also a player in sex expression? These questions should be addressed in the future to better
understand the molecular intricacies of carpel development.

121

LITERATURE CITED

122

LITERATURE CITED

Boualem A, Fergany M, Fernandez R, Troadec C, Martin A, Morin H, Sari M, Collin F, Flowers
J, et al. (2008). A conserved mutation in an ethylene biosynthesis enzyme leads to
andromonoecy in melons. Science 321: 836-838.
Grumet R and Taft JA (2011). Sex expression in cucurbits In: Genetics, Genomics and Breeding
in Cucurbits. Kole C, Wang YH, Behera TK, Kole C (eds). Science Publishers Inc. Enfield NH
and CRC Press (Taylor & Francis) group, Boca Raton, FL. p. 353-375.
Ishiki Y, Oda A, Yaegashi Y, Orihara Y, Arai T, Hirabayashi T, Nakagawa H, Sato T (2000).
Cloning of an auxin-responsive 1-aminocyclopropane-1-carboxylate synthase gene (CMe-ACS2)
from melon and the expression of ACS genes in etiolated melon seedlings and melon fruits. Plant
Sci 159: 173-181.
Kenigsbuch D and Cohen Y (1990). The inheritance of gynoecy in muskmelon. Genome 33:
317-320.
Kubicki B (1969). Sex determination in muskmelon (Cucumis melo L.) Genet Pol 10: 145-163.
Lasserre E, Bouquin T, Hernandez JA, Bull J, Pech JC, Balague C (1996). Structure and
expression of three genes encoding ACC oxidase homologs from melon (Cucumis melo L.). Mol
Genet Genomics 251: 81-90.
Little HA, Papadopoulou E, Hammar SA, Grumet R (2007). The influence of ethylene
perception on sex expression in transgenic melon as assessed by expression of the mutant
ethylene receptor gene, At-etr1-1, under control of constitutive and floral-targeted promoters.
Sex Plant Reprod 20: 123-136.
Martin A, Troadec C, Boualem A, Mazen R, Fernandez R, Morin H, Pitrat M, Dogimont C,
Bendahamane A (2009). A transposon-induced epigenetic change leads to sex determination in
melon. Nature 461: 1135-1139.
Miki T, Yamamoto M, Nakagawa H, Ogura N, Mori H, Imaseki H, Sato T (1995). Nucleotidesequence of a cDNA for 1-aminocyclopropane-1-carboxylate synthase from melon fruits. Plant
Physiol 107: 297-298.
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 (eds. CC
Ainsworth). BIOS Scientific, Oxford: 189-215.
Roy RP and Saran S (1990). Sex expression in the Cucurbitaceae. In: Bates DM, Robinson RW,
Jeffrey C (eds). Biology and utilization of the Cucurbitaceae. Cornell University Press, Ithaca,
NY: 251-268.
123