Annals of Botany 108: 1529– 1537, 2011

doi:10.1093/aob/mcr151, available online at www.aob.oxfordjournals.org

REVIEW: PART OF A SPECIAL ISSUE ON PALM BIOLOGY

Environmental regulation of sex determination in oil palm: current knowledge

and insights from other species
Hélène Adam 1, Myriam Collin 1, Frédérique Richaud 2, Thierry Beulé 2, David Cros 3, Alphonse Omoré 4,
Leifi Nodichao 4, Bruno Nouy 5 and James W. Tregear 1,*
1
IRD, 2CIRAD, IRD/CIRAD Palm Developmental Biology Group, UMR DIADE, Centre IRD, 911 avenue Agropolis, 34394
Montpellier, France, 3CIRAD, UMR AGAP, 4INRAB, Station INRAB, BP1, Pobè, Bénin and 5PalmElit SAS, Parc Agropolis Bt.
14, 2214 Boulevard de la Lironde, 34980 Montferrier sur Lez, France
* For correspondence. E-mail james.tregear@ird.fr

Received: 16 February 2011 Returned for revision: 9 March 2011 Accepted: 7 April 2011 Published electronically: 28 June 2011

† Background The African oil palm (Elaeis guineensis) is a monoecious species of the palm subfamily
Arecoideae. It may be qualified as ‘temporally dioecious’ in that it produces functionally unisexual male and
female inflorescences in an alternating cycle on the same plant, resulting in an allogamous mode of reproduction.
The ‘sex ratio’ of an oil palm stand is influenced by both genetic and environmental factors. In particular, the
enhancement of male inflorescence production in response to water stress has been well documented.
† Scope This paper presents a review of our current understanding of the sex determination process in oil palm
and discusses possible insights that can be gained from other species. Although some informative phenological
studies have been carried out, nothing is as yet known about the genetic basis of sex determination in oil palm,
nor the mechanisms by which this process is regulated. Nevertheless new genomics-based techniques, when com-
bined with field studies and biochemical and molecular cytological-based approaches, should provide a new
understanding of the complex processes governing oil palm sex determination in the foreseeable future.
Current hypotheses and strategies for future research are discussed.

Key words: Sex determination, oil palm, Elaeis guineensis, inflorescence, monoecious, temporal dioecy,
flowering.

IN T RO DU C T IO N either carpels (in the case of protandrous flowers) or stamens

The physical separation of fertile male and female reproduc-
tive organs on different flowers of the same plant or different
individuals of the same species is a condition known as
dicliny. Diclinous species are estimated to represent approx.
10 % of flowering plants (Ainsworth, 2000). It is generally
considered that dicliny confers two main advantages: (1)
better use of available resources through the specialization of
the pollen and ovule producing reproductive organs, and (2)
the enhancement of heterosis (‘hybrid vigour’) through the
promotion of outcrossing (allogamy). Dicliny can be divided
into two main categories: monoecy, in which male (staminate)
and female ( pistillate) flowers are produced on the same plant;
and dioecy, in which single-sex male and female plants exist.
Other more complex situations involving the co-occurrence of
male, female and bisexual flowers or plants are collectively
classified as polygamy.
Although spatial separation of male and female reproductive
organs is employed by many species as a means to achieve
outbreeding, a similar result can be achieved by temporal
separation. This strategy, referred to by Cruden and
Hermann-Parker (1977) as ‘temporal dioecism’, was defined
as: ‘in essence, a flowering sequence which is synchronized
so there is little or no overlap between staminate and pistillate
phases of an individual plant’. In its simplest form, temporal
dioecy can be achieved simply by the delayed maturation of
(in the case of protogynous flowers), two variants of the con-
dition known as dichogamy. It is interesting to note that many
self-incompatible plant species display dichogamous floral
development characters, implying that ‘sexual interference’
factors such as gamete wastage play a part in determining
fitness and not just the promotion of outbreeding (Barrett,
2002).
The palm family (Arecaceae) is a particularly interesting
group in which to investigate dicliny and dichogamy, as both
are prevalent within this clade (Loo et al., 2006; Weiblen
et al., 2000). As regards spatial separation of the sexes, the
most common condition in palms is monoecy, which accounts
for approx. 65 % of genera in the family (Dransfield et al.,
2008) and which dominates in the Arecoideae, the largest
palm subfamily. The majority of monoecious palms produce
bisexual inflorescences, with only a handful of exceptions,
notably in the genera Lepidorrachis, Arenga, Wallichia,
Wettinia, Marojejya, Attalea and Elaeis (Voeks, 1988; Baker
and Hutton, 2006).
The African oil palm (Elaeis guineensis) is the best studied
example of temporal dioecy in the palm family, due in particu-
lar to its economic importance as the world’s largest source of
vegetable oil. Considerable efforts have been devoted over the
last half century or so to improving the yield potential of cul-
tivated palms and to perfecting the growing conditions necess-
ary to realize this potential. Past studies have revealed that the

# The Author 2011. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved.
For Permissions, please email: journals.permissions@oup.com
1530 Adam et al. — Environmental regulation of sex determination in oil palm
‘sex ratio’ of an oil palm stand is affected by both environ-
mental and genetic factors. In this paper, we first discuss
current knowledge of the mode of action of environmental
factors on oil palm inflorescence development, and then
discuss how they might act upon the plant’s reproductive
development to regulate sex determination.
I NF LO R E S C E N C E D E VE LO P M E N T IN OI L PA L M
As mentioned earlier, the oil palm normally produces func-
tionally unisexual male and female inflorescences in an alter-
nating cycle, thereby minimizing the chances of
self-pollination. Macroscopic views of the normal male and
female inflorescences are shown in Fig. 1. Male and female
inflorescences are branched and easily distinguishable from
their overall structure, the male inflorescence bearing a
larger number of rachillae (branches) inserted into the rachis
(central axis). Another easily recognizable difference
between the two sexes is in the form of the rachillae them-
selves; in the case of the female inflorescence they bear a
distal sterile region that develops into a characteristic spine
not seen in the male rachilla. The early differentiation of oil
palm inflorescences was studied by scanning electron
microscopy (SEM) by Van Heel et al. (1987) and in our
group using SEM and light microscopy on stained tissue sec-
tions (Adam et al., 2005). The earliest clear anatomical differ-
ence between the two sexes is the greater number of ‘early
bracts’ arising at the base of ‘spikelet primordia’ in the male
inflorescence compared with its female equivalent (Van Heel
et al., 1987). This reflects the larger number of flowers
borne on the male rachilla and confirms the earlier conclusions
of Corley (1976a). A later key difference between the male and
female inflorescences concerns the arrangement of flowers on
the rachillae, the staminate flowers being borne singly whereas
pistillate flowers are flanked by a pair of abortive accompany-
ing staminate flowers. The female inflorescence thus produces
floral triads, a structural arrangement that is dominant
(although not universal) in the subfamily Arecoideae. The
A B
F I G . 1. Macroscopic views of oil palm inflorescences: (A) mature female
inflorescence, (B) mature male inflorescence. Scale bars ¼ 10 cm.
accompanying staminate flowers develop pollen sacs with pro-
liferating microsporocytic cells; however, the latter display
incomplete development and the flower is shed soon after,
before the pistillate flower of the triad has reached maturity
(Adam et al., 2005). Both types of inflorescence are protected
during their development by the prophyll and first penduncular
bract, which are ruptured just before flower maturity is
reached. Other peduncular bracts are small, triangular struc-
tures which do not envelope the developing inflorescence
(Dransfield et al., 2008).
I N F LO R E S C E N C E D E V E LO P M E N T I N
RE LATI ON TO VE GETATIV E GROW TH
In favourable growing conditions, an inflorescence is initiated
in the axil of each leaf of the palm. The rate of leaf emission
varies with age, with typically three leaves produced per
month in young palms and two per month in the case of
older individuals. The development of an oil palm inflores-
cence between the stages of initiation and flower maturity
lasts 2 – 3 years for both sexes, the process being initiated
soon after seedling establishment.
In tropical humid climates with regular rainfall, inflores-
cence and fruit production is spread evenly throughout the
year. From an experimental point of view, this allows the
possibility of dissecting mature palms so as to visualize a com-
plete series of inflorescence developmental stages within the
same individual. As a means of cross-referencing between
palms, inflorescences are usually numbered according to
their subtending leaves. Thus, the youngest expanding leaf
with separating leaflets (and its associated inflorescence) is
designated as +1, the immediately older leaves being +2,
+3, etc., and younger leaves 0, – 1, etc. A summary of the
key stages of inflorescence development in relation to leaf
number is shown in Fig. 2.
I NS I G HT S I NTO T H E CH RO N OLO GY O F S E X
D E T E R M I N AT I O N F RO M F I E L D
OBS E RVAT I ON S
Typically, a mature palm will alternate between male and
female inflorescence production during its lifetime; however,
the proportion of time spent in each phase will vary consider-
ably depending on both environmental and genetic factors.
The sex ratio is normally defined as the ratio of female to
total inflorescences in a given group of palms. In regions
with high and regular rainfall (e.g. Malaysia and Indonesia),
oil palm sex ratios tend to vary little throughout the year, in
contrast to areas experiencing a marked dry season such as
in West Africa, where the sex ratio undergoes extensive fluctu-
ations. Corley (1976a) noted that in the latter case, the period
of lowest sex ratio (high male inflorescence production) occurs
during the rainy season and speculated that this character is an
adaptation against the reduction in airborne pollen density
caused by high atmospheric humidity. This illustrates the well-
established observation that oil palm sex determination is
strongly influenced by climatic factors, with male inflores-
cence production being promoted by water deficit.
Male inflorescence production can also be induced artifi-
cially by defoliating adult plants, a technique which has
 Adam et al. — Environmental regulation of sex determination in oil palm
Leaf +15
Floral development
Leaf +18
Flower maturity
Sex
determination
Leaf –26
Initiation of prophyll and first
peduncular bract
Individualization of
Leaf –20
Initiation of rachilla
bract ( ) Vegetative
meristem
Leaf –6
Initiation of rachilla
meristem (•)
F I G . 2. Key stages of inflorescence development in relation to leaf number.
proved invaluable for increasing pollen production from pisi-
fera palms (Durand-Gasselin et al., 1999). The latter are
used as the male parent in most breeding programmes as
they are homozygous for the Sh– thin shell thickness allele
and give rise to high oil-yielding tenera palms (Sh– /Sh + )
when crossed with a thick-shelled dura (Sh + Sh + ) female
parent. It is interesting to note that pisifera palms produce
abortive fruit, which place lower demands on the plant’s
resources. As a result, pisifera palms usually remain female
when grown in conditions of high and regular water avail-
ability, thus creating the problem of limited pollen production
for breeding, which can be overcome by defoliation as men-
tioned earlier. The exact nature of the stress response under-
lying the induction of male inflorescence production by
defoliation is not yet clear, but current hypotheses will be dis-
cussed in more detail below.
Beyond the obvious agronomic interest in using defoliation as
a means of controlling pollen production, the same approach can
also provide a useful experimental system by which to investi-
gate the process of sex determination in oil palm. In the
context of the studies described here, defoliation can be
defined as the pruning of all expanded leaves at their bases so
1531
Leaf +6
Individualization of
floral meristem ( )
Leaf –27
rachis meristem
Leaf –40
Initiation of rachis meristem
Leaf +2
Initiation of floral bracts
Leaf –2
Individualization of rachilla
meristem ( )
as to remove the entire lamina, the unopened leaves in the
spear being cut around 1 m from their base (Durand-Gasselin
et al., 1999). Field observations of sex ratio changes in response
to defoliation have provided clear indications of the chronology
of sex determination and differentiation. The timeline of oil
palm inflorescence development, as inferred from the effects
of defoliation, are summarized in Fig. 3, as reproduced from
Hartley (1988) by Durand-Gasselin et al. (1999). It can be
seen that the time intervals between the different stages of inflor-
escence development vary according to the age of the palm, in
accordance with the lower rate of leaf emission seen in mature
palms compared with young individuals. One important point
to note is that sex determination is not the only effect of leaf
pruning; the latter treatment also induces inflorescence abortion.
However, the two effects occur at different phases of develop-
ment: whereas abortion is mostly observed for inflorescences
at stages +4 to +10 (4 – 6 months prior to flower maturity),
sex determination changes are typically seen in the region of
stage –20, approx. 22 months before flower maturity in adult
palms (summarized in Durand-Gasselin et al., 1999). As with
most aspects of flowering, the exact chronology of these
effects varies according to palm age. Although it can be
1532 Adam et al. — Environmental regulation of sex determination in oil palm
–40 –30 –20
Adult palms (two leaves emitted per month)
11 months
Young palms (three leaves emitted per month)
7·5 months
LEAF SEX
INITIATION DETERMINATION
F I G . 3. Development and sex determination of the oil palm inflorescence in relation to palm age [after Hartley (1988), as reproduced by Durand-Gasselin et al.
promoted artificially by leaf pruning, inflorescence abortion is in
fact a commonly observed phenomenon in oil palms. Some evi-
dence has been obtained to suggest that inflorescence abortion is
sex-dependent, by examining the effects of moderate leaf
pruning (Corley, 1973). In the latter study, no effect was
observed on the sex of inflorescences reaching maturity during
the first 3 months after pruning; however, over the following 6
months, the sex ratio was significantly lowered. This indicates
that in the specific case of leaf-pruning-related stress at least,
female inflorescences undergo a higher rate of abortion than
male inflorescences. It should be borne in mind that the more
extreme stress induced by pruning the laminae of all expanded
leaves results in the abortion of inflorescences of both sexes.
The ability of oil palms to change the sex of the inflores-
cences they produce in response to environmental factors pro-
vides the basis for a useful experimental system to manipulate
reproductive development in palms and study its molecular
determinants. To date, the character of sex determination
per se has not been used by oil palm breeders in genetic
improvement programmes, so nothing is known regarding
the number or type of genetic loci which might be involved
in determining maleness or femaleness. Nevertheless, the
selection of genotypes conferring high oil yield has inevitably
been accompanied by a general increase in sex ratios, which
provides indirect evidence that genetic factors also come into
play. Indeed, with palms of highly selected genotypes
planted in favourable growing conditions, it is commonly
observed that male inflorescence production is close to zero
in the first year or two of flowering, thus jeopardizing fruit
production. For this reason, seed batches of high-yielding
genetic crosses are now sometimes supplied mixed with a
small proportion of seeds derived from ‘super-male’ individ-
uals. The fact that oil palm sex ratios may be enhanced or
reduced depending on genotype, coupled with the fact that a
genetic map has been produced for this species (Billotte
et al., 2001), raises the interesting possibility of identifying
sex determination loci in the future.
LEAF POSITION
–10 0 +10 +20
22 months 5·5 – 6 months
14·5 months 5·5 – 6 months
INFLORESCENCE FLOWER FRUIT
ABORTION MATURITY MATURITY
LEAF
OPENING
(1999)].
E N V I RON M E N TA L RE G U L AT I O N O F S E X
DE T E R M IN AT I ON W IT H IN AN D B E YO N D
T H E AR E CAC E A E
The ability to vary sexual expression in response to the
environment provides plants with a means of adjusting
resources allocated to ovule and pollen production, thereby
conferring adaptive flexibility (Korpelainen, 1998). Gender
plasticity has been observed in plant species belonging to a
wide range of phylogenetic groups. In the case of androdioe-
cious populations of Mercurialis annua (Euphorbiaceae), in
which male and hermaphrodite plants coexist, progeny sex
ratios were found to depend on the density of the parental
plants (Dorken and Pannell, 2008). This phenomenon was
explained in part by the increased pollen production of her-
maphrodite plants when growing at low density, leading to
proportionally fewer male plants in the subsequent progeny.
Other environmental factors that have been shown to affect
sex ratios in plant populations include daylength and light
intensity, as illustrated by the monoecious (or sometime poly-
gamous) species Atriplex halimus (Chenopodiaceae), which
displays increased femaleness under short days and low light
irradiance (Talamali et al., 2001, 2003). In the case of the
sexually dimorphic woody pioneer species Antirhea borbonica
(Rubiaceae), a polliniferous morph was identified and studied
(Litrico et al., 2005). It was found to display considerable
gender plasticity, almost no fruit-bearing female flowers
being produced in pioneer populations, whereas in late succes-
sion, high fruit-sets were observed (although they were lower
than for the female morph). In a clonal context, gender plas-
ticity may operate not only in relation to environmental con-
ditions but also in relation to the maturation of the plant, as
seen in the andromonoecious species Pennisetum typhoides
(Poaceae; Sandmeier and Dajoz, 2000) or the monoecious
aquatic Sagittaria sagittifolia (Alismataceae; Dorken and
Barrett, 2003). In the case of the subdioecious species
Urtica dioica (Urticaceae), in which monoecious plants
 Adam et al. — Environmental regulation of sex determination in oil palm
coexist with female and male individuals, only the former
were found to display a variable sex ratio, with different
clones of the same genotype varying in the percentage of
male flowers produced (Glawe and de Jong, 2005).
Subsequent studies revealed that the maternal parent strongly
contributes to sex ratios, the exact genetic mechanism involved
remaining as yet unresolved (Glawe and de Jong, 2007, 2009).
Although relatively few data are available for the palm family,
sexual expression has been shown to be influenced by environ-
mental factors not only in oil palm but also in two different
species of the monoecious Arecoid genus Attalea. In
A. funifera, a study was carried out in Brazil by Voeks (1988)
on the relative abundance of the three different inflorescence
types produced: male with staminate flowers, bisexual with a
few pistillate and many staminate flowers, and female with
pistillate and sterile staminate flowers. Although bagging exper-
iments had revealed that autogamy was possible in the bisexual
inflorescences, the latter contribute only very weakly to fruit-set,
given their low abundance at the population level (5 % compared
with 80 % male and 15 % female inflorescences). The data
obtained in this study suggested that sexual expression in
A. funifera is related to energetic factors, with palms apparently
producing only male inflorescences in the crowded understorey
whereas female expression was observed when they had grown
and entered the energy-rich canopy. Reproductive cost measure-
ments (estimated via dry weights of inflorescence and infructes-
cence components) revealed the higher energetic demands of
mature female flowers and fruit compared with male flowers
(1.6 and 13.2 versus 1.0 in relative units, respectively), with
the production of an ‘initial female inflorescence’ requiring an
energy input estimated as over 10 times the annual budget of
the youngest mature palms. A second species of Attalea,
A. speciosa, was also studied in terms of its gender plasticity
(Barot et al., 2005). In this case, the onset of reproduction and
sex allocation were compared in three contrasting environments:
forest, partially invaded pasture and a pure stand (babassual).
Reproduction was found to start at a smaller size in the pasture
and pure stand compared with forest. Male flowering started at
a lower height than female flowering in the pasture and pure
stand; however, this was not the case in the forest environment.
Although several hypotheses were raised to explain these data, it
seems probable that the earlier flowering observed in more open
environments was made possible by the much greater light (and
therefore energy resources) available to palms in these habitats.
As regards the earlier male flowering in pasture/babassual, it was
postulated that the lower levels of stored energy in younger
palms (compared with larger mature palms in the forest) might
be stretched beyond viability by the higher demands of female
flowering and fruiting. Thus, the delayed female flowering of
mature pasture/babassual palms might be considered as an adap-
tation conferring more efficient energy use and therefore
survival.
At a more general level, although there is some scope for the
optimization of resource allocation to male and female repro-
duction in hermaphrodite species by the modulation of pollen
and ovule development, the greatest flexibility is probably
afforded by the monoecious and polygamous conditions,
which allow the adjustment of the relative numbers of special-
ized male and female flowers both within the same plant and
between individuals. Within the palm family it seems likely
1533
that the gender plasticity of at least some of the other five ‘tem-
porally dioecious’ genera producing unisexual inflorescences
(see above) might be subject to environmental regulation.
One intriguing case is that of the Madagascan Marojejya
darianii: a number of specimens of this species were success-
fully grown to maturity on the island of Hawaii, but as yet,
after several years of flowering, only male inflorescence pro-
duction has been observed (Marcus, 2010). Although environ-
mental factors may play a causal role in cases such as this, it
can also be speculated that endogenous factors within the
plant that change during its life-cycle progression (e.g.
hormone composition), or indeed a combination of the two
elements, may be involved.
M O L E C U L A R D E T E R M I N A N T S O F F LOW E R
A N D IN F LO R E S C E NCE S E X: CA N W E L E A R N
F RO M OT H E R S P E C I E S ?
Dioecy and monoecy appear to have arisen on a large number
of occasions during the evolution of flowering plants, as
witnessed by their wide and fragmented distribution within
the Angiosperm tree of life. This suggests that the genetic
changes that resulted in the monoecious and dioecious habits
are different for the most part in different evolutionary
lineages (Ainsworth, 2000). Nevertheless, thanks to compara-
tive genomics and evo-devo approaches, it is clear that a wide
range of signalling elements that regulate reproductive devel-
opment are conserved throughout flowering plants. An early
indication of this conservation was obtained by comparing
the first two Angiosperms to have their genomes sequenced,
Arabidopsis thaliana and rice (Izawa et al., 2003). The latter
authors carried out a survey of the structural and functional
conservation of genes between the two model plants, using
flowering pathways as an example. This study revealed that a
wide range of flowering-related genes are structurally con-
served between monocots and eudicots. Nevertheless, func-
tional divergence appears to have frequently occurred to give
novel regulatory mechanisms which distinguish the two
model species, this degree of divergence providing a glimpse
of the situation in flowering plants as a whole. As regards
sex determination, although it seems likely that a generalized
signalling cascade for flower development has been partially
maintained during evolution, the genetic basis of sexual
dimorphism per se is likely to vary considerably. From a
developmental point of view, this hypothesis is supported by
the great variety of ways in which unisexual flower develop-
ment is achieved. Examples of the latter, provided by the
palm family, include non-initiation of the undesired organ
(e.g. the pistil of staminate flowers in certain Phytelephas
species), early developmental arrest of the undesired organ
(e.g. the androecium of pistillate flowers in Elaeis guineensis)
or incomplete development of the gametophyte (e.g. the
androecium of pistillate flowers in Rhapidophyllum hystrix)
(Uhl and Moore, 1977; Adam et al., 2005; Dransfield et al.,
2008). As discussed by Weiblen et al. (2000), dioecy and
monoecy have probably arisen independently a number of
times since the palm family first originated; for this reason,
the Arecaceae provide an interesting microcosm for studying
the evolution of sexual dimorphism in flowering plants as a
whole.
1534 Adam et al. — Environmental regulation of sex determination in oil palm

S E X D E T E R M I N AT I O N M O D E L S I N For this species, a number of tasselseed mutants displaying

AN G IO S P E R M S feminized male flowers have been isolated and characterized
and a signalling pathway has emerged (Calderon-Urrea and
Although approx. 10 % of flowering plants produce unisexual
Dellaporta, 1999). It was found that the sex determination
flowers, relatively few data are available on the genetic basis of
genes TASSELSEED1 and TASSELSEED2 are required for
sexual dimorphism, whether in dioecious or in monoecious
the programmed death of pistil cells in the male flower,
plants. Genome sequences have already been obtained for
TASSELSEED1 being a putative upstream regulator of
the dioecious species poplar (Tuskan et al., 2006) and the
TASSELSEED 2. These two genes were cloned and found to
trioecious papaya (Ming et al., 2008). In the latter case a
encode, respectively, a lipoxygenase involved in jasmonic
small male-specific region (MSY) was identified on the Y
acid synthesis and a steroid dehydrogenase (DeLong et al.,
chromosome and it was proposed that two sex determination
1993; Acosta et al., 2009). A third gene, SILKLESS1, of as
genes controlled the sex determination: a feminizing or
yet unknown sequence, blocks TASSELSEED pathway-
stamen suppressor gene causing stamen abortion before or at
induced cell death in the pistil primordia of primary ear
flower inception; and a carpel suppressor gene causing
florets. Other studies have implicated the RNA-dependent
carpel abortion later in flower development (Ming et al.,
RNA polymerase gene RMR6 as a regulator of SILKLESS1
2007). The most extensively studied dioecious model is that
(Parkinson et al., 2007), suggesting crosstalk with epigenetic
of Silene latifolia, which uses an X/Y-type sex determination
regulatory pathways. It is also interesting to note that crosstalk
mechanism. For this species, a study was made of the pheno-
was observed with the signalling mechanism determining
typic effects of X-ray-induced deletions (Lebel-Hardenack
inflorescence branching, itself a sex-linked character (Chuck
et al., 2002), in conjunction with an amplified fragment
et al., 2007). The latter authors demonstrated that the sex
length polymorphism approach to identify sex determination
determination gene TASSELSEED4 encoded a micro-RNA
mutants that had lost male traits. This approach provided
that is able to bind to and regulate the turnover of mRNAs
data to confirm the classical genetic prediction of three dis-
of the TASSELSEED6 gene related to APETALA2 of
persed male-determining loci on the Y chromosome, one for
A. thaliana. Further observations by Hultquist and Dorweiler
carpel suppression, one for early stamen development and
(2008) led to a wider model which also included the
another for late stamen development. A similar approach was
micro-RNA miR156 and target genes of the
used by Koizumi et al. (2010). In this case C-ion beam radi-
SQUAMOSA-binding protein box (SPB-box) family.
ation was used to obtain a Y chromosome lesion that resulted
In an evolutionary context, following the discovery of the
in the production of hermaphrodite flowers with a mature
tasselseed sex determination pathway in maize, several
gynoecium. Although the identity of the gene or genes respon-
studies were performed with the aim of determining whether
sible for each of these activities remains to be determined,
all or some of the signalling elements identified might be func-
complementary studies of sex-specific expression have
tionally conserved in other plant groups. Overall, the data
enabled the identification of a number of genes of interest
obtained suggest that although orthologues of the
(Robertson et al., 1997; Scutt et al., 2002; Zluvova et al.,
TASSELSEED genes can be easily identified in other species,
2006) and microdissection techniques have allowed the iso-
their functions probably vary even within the same plant
lation of Y chromosome-located genes (Delichère et al.,
family. Thus, although the maize pistil abortion mechanism
1999; Matsunaga et al., 2003).
appears to be conserved in its wild relative Tripsacum dacty-
Two plant species for which sex determination gene func-
loides (Li et al., 1997), there is little evidence for conservation
tions have been clearly attributed are melon and maize. In
beyond the subfamily Panicoideae of the Poaceae (Zaitchik
the case of the polygamous species melon (Cucumis melo),
et al., 2000; Malcomber and Kellogg, 2006). It is interesting
sex ratios are controlled by allelic compositions at two differ-
to note that homologues of TASSELSEED2 identified in
ent loci. Boualem et al. (2008) showed that the a allele present
Silene latifolia and A. thaliana displayed male-specific
at the androdioecious locus encoded an ethylene biosynthesis
expression; however, in both cases, this expression was
enzyme, CmACS-7, that repressed stamen development in
observed in the tapetum rather than in the gynoecium primor-
female flowers when active. Martin et al. (2009) later
dium as in maize (Lebel-Hardenack et al., 1997).
showed that the transition from male to female flowers in
In addition to the tasselseed pathway, maize has also yielded
gynoecious lines resulted from the insertion of a transposon
informative data on the likely role of plant hormones in the
(and consequent DNA methylation changes) in the promoter
sexual differentiation process. Bensen et al. (1995) character-
of a zinc finger transcription factor gene, CmWIP1.
ized the ANTHER EAR1 (AN1) gene, mutations of which
Expression of CmWIP1, which was abolished in the transpo-
result in both reduced plant height and the development of
son insertion mutant, led to carpel abortion and also indirectly
perfect flowers on normally pistillate ears. It was postulated
repressed the expression of CmACS-7 to allow stamen develop-
that the AN1 gene encoded a kaurene synthase A enzyme
ment. Collectively, these data provide an interesting insight
involved in gibberellin biosynthesis, which explained the
into the sex determination process in the Cucurbitaceae and
observed modulation of phenotype severity in response to gib-
also reveal some of the genetic basis of hormone-dependent
berellin application. A further demonstration of the involve-
sex determination, a common phenomenon in diclinous plant
ment of hormones in the sexual differentiation process in
species.
maize was obtained by Young et al. (2004) and Pineda Rodo
Another functionally characterized model is that of the
et al. (2008), who demonstrated that increases in biologically
monoecious species maize, which produces morphologically
active cytokinin levels achieved by genetic transformation
distinct male and female inflorescences on the same plant.
 Adam et al. — Environmental regulation of sex determination in oil palm
led to the abolition of pistil abortion and the appearance of the
tasselseed phenotype.
As regards the question of sex determination in oil palm, the
lack of a widely conserved pathway for this key developmental
process amongst flowering plants suggests that at best only
part of the knowledge gained in model plants can be transposed
directly to the Arecaceae. In terms of phylogenetic relationships,
the Poaceae are clearly the closest family for which in-depth
knowledge has been obtained, through studies on maize. The
latter species is therefore a useful starting point from which to
initiate searches for palm orthologues of key genes previously
shown to be involved in sex differentiation, notably those ident-
ified via the tasselseed mutants. The increasing body of genomic
sequence data now becoming available for palms, including the
date palm draft sequence (Al-Dous et al., 2011), will prove very
useful in this connection.
TOWA RDS A M OD E L TO E XP L A IN SE X
DE T E R M I NAT IO N IN OI L PA L M
The fact that oil palm sex determination is a complex process
is amply demonstrated by the complexity of phenological
information obtained from plants growing in the field.
Although the currently available data do not allow a precise
model to be formulated to explain how oil palm sex determi-
nation is regulated, we can nevertheless identify various
factors of importance and propose a possible framework for
their interactions. At least four different types of factor can
be identified that might participate in sex determination and
differentiation: abiotic factors (e.g. water stress), metabolic
factors (e.g. carbon reserves), hormone status and genetic
factors. The masculinizing effects of drought conditions are
already well known; however, the physiological and genetic
bases of this induced developmental change remain to be elu-
cidated. In one of the earliest publications to mention the
subject of gender plasticity in oil palm, Beirnaert (1935) pos-
tulated that the ratio of carbon assimilation to nitrogen absorp-
tion was the key factor determining the sex ratio. This
hypothesis can be reconciled with the observed effects of
drought if the lowering of photosynthesis in conditions of
water stress is taken into account. Other factors affecting the
metabolic status of palms, including the effects of high-density
planting, fruit pruning and the defoliation of neighbouring
palms, are all compatible with the notion that higher sex
ratios are associated with higher levels of carbon assimilates
(Sparnaaij, 1960). Compared with flowering plants as a
whole, oil palm exhibits relatively low phenotypic plasticity
in terms of its overall architecture; transitory carbon reserves
therefore provide a more flexible means by which to buffer
source –sink imbalances associated with climate variability
in this species (Legros et al., 2009c). The latter authors per-
formed a comparative study between oil palms at two different
sites in Indonesia with or without a dry season; data were also
obtained on the effects of fruit pruning and photoperiod vari-
ation (Legros et al., 2009a, b). One of the main elements
revealed by this work was that source – sink imbalances in
oil palm were buffered principally by fluctuations in non-
structural carbohydrates in the stem, which consisted mostly
of glucose and starch. Of these two components, starch was
identified as a major physiological buffer of seasonal
1535
source – sink imbalances, but glucose was not (Legros et al.,
2009c). Interestingly, however, glucose levels were found to
decrease progressively in response to fruit pruning over the
22-month period of study (Legros et al., 2009b).
As regards the possible involvement of hormones in oil palm
sex determination, some indications were obtained by using the
experimental strategy of applying growth regulators to young
palms grown in large polythene bags (Corley, 1976b). In the
latter study, the auxin NAA (a-naphthyleneacetic acid) was
found to promote female inflorescence production; however,
this effect was not observed until 24 months after the end of treat-
ment, whereas sex differentiation was expected to take place
16– 22 months before anthesis. For this reason, the feminizing
effect was assumed to be indirect. In the case of ethephon, the
predominant effect was to delay flowering, albeit with a statisti-
cally insignificant lowering of the sex ratio. Treatment with gib-
berellic acid, however, was shown to significantly increase the
production of male inflorescences and reduce production of
female inflorescences within the expected time frame; gibberel-
lins are thus likely to play a role in the oil palm sex determination
process. It is interesting to note that a masculinizing effect of
gibberellic acid treatment has also been observed in cucumber
(Fuchs et al., 1977), whereas in maize, a feminizing effect is
observed with this type of treatment (Dellaporta and
Calderon-Urrea, 1994). These opposing effects observed in
different species no doubt reflect the diversity of sex-
determining mechanisms that have evolved in flowering
plants, even if some regulatory modules appear to be structurally
conserved. Given the relatively distant phylogenetic relation-
ships between the Arecaceae and the various species for which
Photosynthesis
Water
availability
Glucose Starch
Hormone status
Genetic factors
Female
inflorescence
Indeterminate
inflorescence
Male
inflorescence
F I G . 4. Possible modes of interaction of factors affecting inflorescence sex in
oil palm.
1536 Adam et al. — Environmental regulation of sex determination in oil palm
functional sex determination studies have been performed, it is
clear that prior conceptions need to be avoided when tackling
the complex issue of how this process is regulated in oil palm.
By combining the various data discussed above, a basic fra-
mework to explain the interactions between the various par-
ameters of importance in oil palm sex determination can be
proposed (Fig. 4). The schema shown draws on mostly pheno-
logical and physiological data and suggests points of inter-
action with genetic factors that might be associated with the
contrasting behaviour of palms of different genotype in the
same environmental conditions. A number of questions
remain, regarding notably the modes of interaction between
metabolism, hormone status and the sex determination
process per se. The available evidence, along with general per-
spectives on plant responses to environmental change, point
towards a role for hormones as an intermediary signalling
element between water balance and sex determination, with
metabolic factors possibly acting upstream of the hormone
molecule(s). The model shown is a first attempt to draw all
these elements together and will now need to be validated,
adjusted and refined by combining new functional genomic
and metabolomic approaches with phenological observations
in the field. As a first step in this direction, we are currently
characterizing the transcriptome of the oil palm inflorescence,
with a view to identifying points of divergence in the gene
expression patterns associated with male and female develop-
ment. This work will be combined with hormone studies on
the same material, and ultimately the genes of interest display-
ing expression patterns related to male or female differen-
tiation will be studied in a more complex phenological
context. This overall approach should also help to shed light
on events further upstream (i.e. mediating the response to
environmental conditions) and on the role of metabolic
factors in the sex determination process.
CO NCL USI ON S
In this review, we have presented and discussed the wide
ranging but often fragmentary data currently available on the
regulation of sex determination in oil palm. New efforts cur-
rently being devoted to this field will no doubt shed light on
how oil palm regulates inflorescence sex in response to its
environment and reveal to what extent the genetic and physio-
logical elements identified in other species are conserved. Due
to various characteristics that typify palms in general (notably
their large size and long life cycle), research on oil palm flow-
ering is complicated but nevertheless vital to understand the
biology of this important crop plant. The knowledge gained
should also in the longer term improve our understanding of
the evolution of sex determination in the Arecaceae as a whole.
AC KN OW LED GEMEN T S
We gratefully acknowledge financial support for this work via
a grant from PalmElit SAS and institutional funding from IRD
and Cirad.
L I T E R AT U R E C I T E D
Acosta IF, Laparra H, Romero SP, et al. 2009. tasselseed1 is a lipoxygenase
affecting jasmonic acid signaling in sex determination of maize. Science
323: 262–265.
Adam H, Jouannic S, Escoute J, Duval Y, Verdeil JL, Tregear JW. 2005.
Reproductive developmental complexity in the African oil palm (Elaeis
guineensis). American Journal of Botany 92: 1836– 1852.
Ainsworth C. 2000. Boys and girls come out to play: the molecular biology of
dioecious plants. Annals of Botany 86: 211–221.
Al-Dous EK, George B, Al-Mahmoud ME, et al. 2011. De novo genome
sequencing and comparative genomics of date palm (Phoenix dactyli-
fera). Nature Biotechnology (in press) doi:10.1038/nbt.1860.
Baker WJ, Hutton I. 2006. Lepidorrhachis. Palms 50: 33–38.
Barot S, Mitja D, Miranda I, Meija GD, Grimaldi M. 2005. Reproductive
plasticity in an Amazonian palm. Evolutionary Ecology Research 7:
1 –15.
Barrett SCH. 2002. Sexual interference of the floral kind. Heredity 88:
154– 159.
Beirnaert A. 1935. Introduction à la biologie florale du palmier à huile (Elaeis
guineensis Jacquin). Série scientifique de Institut National pour l’Etude
Agronomique du Congo Belge 5: 3– 42.
Bensen RJ, Johal GS, Crane VC, et al. 1995. Cloning and characterisation of
the maize An1 gene. The Plant Cell 7: 75–84.
Billotte N, Risterucci AM, Barcelos E, Noyer JL, Amblard P, Baurens FC.
2001. Development, characterisation and across-taxa utility of oil palm
(Elaeis guineensis Jacq.) microsatellite markers. Genome 44: 413 –425.
Boualem A, Fergany M, Fernandez R, et al. 2008. A conserved mutation in
an ethylene biosynthesis enzyme leads to andromonoecy in melons.
Science 321: 836– 838.
Calderon-Urrea A, Dellaporta SL. 1999. Cell death and cell protection genes
determine the fate of pistils in maize. Development 126: 435– 441.
Chuck G, Meeley R, Irish E, Sakai H, Hake S. 2007. The maize tasselseed4
microRNA controls sex determination and meristem cell fate by targeting
Tasselseed6/indeterminate spikelet1. Nature Genetics 39: 1517–1521.
Corley RHV. 1973. Oil palm physiology: a review. In: Wastie RL, Earp DA.
eds. Advances in oil palm cultivation. Kuala Lumpur, Malaysia:
Incorporated Society of Planters.
Corley RHV. 1976a. Inflorescence abortion and sex differentiation. In: Corley
RHV, Hardon JJ, Wood BJ. eds. Developments in crop science 1: oil palm
research. Amsterdam: Elsevier, 37– 54.
Corley RHV. 1976b. Sex differentiation in oil palm: effects of growth regula-
tors. Journal of Experimental Botany 27: 553– 558.
Cruden RW, Hermann-Parker SM. 1977. Temporal dioecism: an alternative
to dioecism? Evolution 31: 863–866.
Delichère C, Veuskens J, Hernould M, et al. 1999. SlY1, the first active gene
cloned from a plant Y chromosome, encodes a WD-repeat protein. Embo
Journal 18: 4169– 4179.
Dellaporta SL, Calderon-Urrea A. 1994. The sex determination process in
maize. Science 266: 1501– 1505.
DeLong A, Claderon-Urrea A, Dellaporta S. 1993. Sex determination gene
TASSELSEED2 of maize encodes a short-chain alcohol dehydrogenase
required for stage-specific floral organ abortion. Cell 74: 757– 768.
Dorken ME, Barrett SCH. 2003. Gender plasticity in Sagittaria sagittifolia
(Alismataceae), a monoecious aquatic species. Plant Systematics and
Evolution 237: 99– 106.
Dorken ME, Pannell JR. 2008. Density-dependent regulation of the sex ratio
in an annual plant. The American Naturalist 171: 824 –830.
Dransfield J, Baker WJ, Harley MM, Asmussen C, Lewis CE. 2008.
Genera Palmarum: the evolution & classification of palms. London:
Kew Publishing.
Durand-Gasselin T, Noiret JM, Kouamé RK, Cochard B, Adon B. 1999.
Availability of quality pollen for improved oil palm (Elaeis guineensis
Jacq.) seed production. Plantations, Recherche, Développement 6:
264– 276.
Fuchs E, Atsmon D, Halevy AH. 1977. Adventitious staminate flower for-
mation in gibberellin treated gynoecious cucumber plants. Plant and
Cell Physiology 18: 1193– 1201.
Glawe GA, de Jong TJ. 2005. Environmental conditions affect sex expression
in monoecious, but not in male and female plants of Urtica dioica. Sexual
Plant Reproduction 17: 253–260.
Glawe GA, de Jong TJ. 2007. Inheritance of progeny sex ratio in Urtica
dioica. Journal of Evolutionary Biology 20: 133–140.
 Adam et al. — Environmental regulation of sex determination in oil palm
Glawe GA, de Jong TJ. 2009. Complex sex determination in Urtica dioica.
Evolutionary Ecology Research 23: 635.
Hartley CWS. 1988. The oil palm, 3rd edn. London: Longman.
Hultquist JF, Dorweiler JE. 2008. Feminized tassels of maize mop1 and ts1
mutants exhibit altered levels of miR156 and specific SBP-box genes.
Planta 229: 99– 113.
Izawa T, Takahashi Y, Yano M. 2003. Comparative biology comes into
bloom: genomic and genetic comparison of flowering pathways in rice
and Arabidopsis. Current Opinion in Plant Biology 6: 113– 120.
Koizumi A, Yamanaka K, Nishihara K, Kazama Y, Abe T, Kawano S.
2010. Two separate pathways including SlCLV1, SlSTM and SlCUC
that control carpel development in a bisexual mutant of Silene latifolia.
Plant and Cell Physiology 51: 282– 293.
Korpelainen H. 1998. Labile sex expression in plants. Biological Reviews 73:
157– 180.
Lebel-Hardenack S, Koutnikova DY, Saedler H, Grant SR. 1997.
Conserved expression of a TASSELSEED2 homolog in the tapetum of
the dioecious Silene latifolia and Arabidopsis thaliana. Plant Journal
12: 515 –526.
Lebel-Hardenack S, Hauser E, Law TF, Schmid J, Grant SR. 2002.
Mapping of sex determination loci on the white campion (Silene latifolia)
Y chromosome using amplified fragment length polymorphism. Genetics
160: 717–725.
Legros S, Mialet-Serra I, Caliman JP, Siregar FA, Clément-Vidal A,
Dingkuhn M. 2009a. Phenology and growth adjustments of oil palm
(Elaeis guineensis) to photoperiod and climate variability. Annals of
Botany 104: 1171– 1182.
Legros S, Mialet-Serra I, Caliman JP, et al. 2009b. Phenology, growth and
physiological adjustments of oil palm (Elaeis guineensis) to sink limit-
ation induced by fruit pruning. Annals of Botany 104: 1183–1194.
Legros S, Mialet-Serra I, Clément-Vidal A, et al. 2009c. Role of transitory
carbon reserves during adjustment to climate variability and source-sink
imbalances in oil palm (Elaeis guineensis). Tree Physiology 29:
1199–1211.
Li D, Blakey CA, Dewald C, Dellaporta SL. 1997. Evidence for a common
sex determination mechanism for pistil abortion in maize and in its wild
relative Tripsacum. Proceeding of the National Academy of Sciences USA
94: 4217–4222.
Litrico I, Pailler T, Thompson JD. 2005. Gender variation and primary suc-
cession in a tropical woody plant, Antirhea borbonica (Rubiaceae).
Journal of Ecology 93: 705–715.
Loo AHB, Dransfield J, Chase MW, Baker WJ. 2006. Low-copy nuclear
DNA, phylogeny and the evolution of dichogamy in the betel nut
palms and their relatives (Arecinae; Arecaceae). Molecular
Phylogenetics and Evolution 39: 598–618.
Malcomber ST, Kellogg EA. 2006. Evolution of unisexual flowers in grasses
(Poaceae) and the putative sex-determination gene, TASSELSEED2 (TS2).
New Phytologist 170: 885–899.
Marcus J. 2010. Observations on the flowering of Marojejya darianii. Palms
54: 189 –192.
Martin A, Troadec C, Boualem A, et al. 2009. A transposon-induced epige-
netic change leads to sex determination in melon. Nature 461:
1135–8113.
1537
Matsunaga S, Isono E, Kejnovsky E, et al. 2003. Duplicative transfer of a
MADS box gene to a plant Y chromosome. Molecular Biology and
Evolution 20: 1062– 1069.
Ming R, Yu Q, Moore PH. 2007. Sex determination in papaya. Seminars in
Cell and Development Biology 18: 401– 408.
Ming R, Hou S, Feng Y, et al. 2008. The draft genome of the transgenic tro-
pical fruit tree papaya (Carica papaya Linnaeus). Nature 452: 991– 996.
Parkinson SE, Gross SM, Hollick JB. 2007. Maize sex determination and
abaxial leaf fates are canalized by a factor that maintains repressed epige-
netic states. Developmental Biology 308: 462–473.
Pineda Rodo A, Brugière N, Vankova R, et al. 2008. Over-expression of a
zeatin O-glucosylation gene in maize leads to growth retardation and tas-
selseed formation. Journal of Experimental Botany 59: 2673–2686.
Robertson SE, Li Y, Scutt CP, Willis ME, Gilmartin PM. 1997. Spatial
expression dynamics of Men-9 delineate the third floral whorl in male
and female flowers of dioecious Silene latifolia. Plant Journal 12:
155–168.
Sandmeier M, Dajoz I. 2000. Flowering phenology and gender variation in
Pennisetum typhoides. International Journal of Plant Sciences 161:
81–87.
Scutt CP, Jenkins T, Furuya M, Gilmartin P. 2002. Male specific genes
from dioecious white campion identified by fluorescent differential
display. Plant and Cell Physiology 43: 563– 572.
Sparnaaij LD. 1960. The analysis of bunch production in the oil palm.
Journal of the West African Institute of Oil Palm Research 3: 109– 180.
Talamali A, Dutuit P, Le Thomas A, Gorenflot R. 2001. Polygamy in
Atriplex halimus L. (Chenopodiaceae). Comptes rendus de l’Académie
des Sciences 324: 107–113.
Talamali A, Bajji M, Le Thomas A, Kinet JM, Dutuit P. 2003. Flower
architecture and sex determination: how does Atriplex halimus play
with floral morphogenesis and sex genes? New Phytologist 157: 105– 113.
Tuskan G, Difazio S, Jansson S, et al. 2006. The genome of black cotton-
wood, Populus trichocarpa (Torr. & Gray). Science 313: 1596–1604.
Uhl NW, Moore HE. 1977. Centrifugal stamen initiation in phytelephantoid
palms. American Journal of Botany 64: 1152–1161.
Van Heel WA, Breure CJ, Menendez T. 1987. The early development of
inflorescences and flowers of oil palm (Elaeis guineensis Jacq.) seen
through the scanning electron microscope. Blumea 32: 67–78.
Voeks RA. 1988. Changing sexual expression of a Brazilian rain forest palm
(Attalea funifera Mart.). Biotropica 20: 107– 113.
Weiblen GD, Oyama RK, Donoghue MJ. 2000. Phylogenetic analysis of
dioecy in monocotyledons. The American Naturalist 155: 46–58.
Young TE, Giesler-Lee J, Gallie DR. 2004. Senescence-induced expression
of cytokinin reverses pistil abortion during maize flower development.
Plant Journal 38: 910–922.
Zaitchik BF, LeRoux LG, Kellogg EA. 2000. Development of male flowers
in Zizania aquatica (North American wild-rice; Gramineae).
International Journal of Plant Sciences 161: 345–351.
Zluvova J, Nicolas M, Berger A, Negrutiu I, Monéger F. 2006. Premature
arrest of the male flower meristem precedes sexual dimorphism in the
dioecious plant Silene latifolia. Proceeding of the National Academy of
Sciences USA 103: 18854–18859.