See all ›
8 CitationsSee all ›
67 ReferencesSee all ›
6 FiguresSex chromosomes and quantitative sex expression in monoecious hemp (Cannabis sativa L.)
Abstract
Hemp (Cannabis sativa) has a highly variable sexual phenotype. In dioecious hemp, the sex is controlled by heteromorphic sex chromosomes according to an X-to-autosomes equilibrium. However, in monoecious hemp, the sex determinism remains widely unknown and has never been related to a quantitative approach of sex expression. The present paper aims to contribute to the comprehension of the sex determinism in monoecious hemp by assessing the genotypic variability of its sex expression and establishing its sex chromosomes. Five monoecious and one dioecious cultivars were grown in controlled conditions under several photoperiods. The monoecy degree of 194 monoecious plants was recorded at each node by a figure ranging from 0 (male flowers only) to 6 (female flowers only). The genome size of 55 plants was determined by flow cytometry. The DNA of 115 monoecious plants was screened with the male-associated marker MADC2. The monoecy degree varied significantly among monoecious cultivars from 3.36 ± 2.28 in ‘Uso 31’ to 5.70 ± 0.81 in the most feminised ‘Epsilon 68’. The variation of monoecy degree among cultivars remained consistent across trials despite a significant “cultivar × trial” interaction and partly agreed with their earliness. The genome size of monoecious plants (1.791 ± 0.017 pg) was not different from that of females (1.789 ± 0.019 pg) but significantly lower than that of males (1.835 ± 0.019 pg). MADC2 was absent from all monoecious plants. These results strongly support that cultivars of monoecious hemp have the XX constitution and that their sex expression has a genetic basis.
Figures
Sex chromosomes and quantitative sex expression
in monoecious hemp (Cannabis sativa L.)
Anne-Michelle Faux •Alice Berhin •
Nicolas Dauguet •Pierre Bertin
Received: 8 July 2013 / Accepted: 29 October 2013 / Published online: 6 November 2013
ÓSpringer Science+Business Media Dordrecht 2013
Abstract Hemp (Cannabis sativa) has a highly
variable sexual phenotype. In dioecious hemp, the
sex is controlled by heteromorphic sex chromosomes
according to an X-to-autosomes equilibrium. How-
ever, in monoecious hemp, the sex determinism
remains widely unknown and has never been related
to a quantitative approach of sex expression. The
present paper aims to contribute to the comprehension
of the sex determinism in monoecious hemp by
assessing the genotypic variability of its sex expres-
sion and establishing its sex chromosomes. Five
monoecious and one dioecious cultivars were grown
in controlled conditions under several photoperiods.
The monoecy degree of 194 monoecious plants was
recorded at each node by a figure ranging from 0 (male
flowers only) to 6 (female flowers only). The genome
size of 55 plants was determined by flow cytometry.
The DNA of 115 monoecious plants was screened with
the male-associated marker MADC2. The monoecy
degree varied significantly among monoecious
cultivars from 3.36 ±2.28 in ‘Uso 31’ to
5.70 ±0.81 in the most feminised ‘Epsilon 68’. The
variation of monoecy degree among cultivars
remained consistent across trials despite a significant
‘‘cultivar 9trial’’ interaction and partly agreed with
their earliness. The genome size of monoecious plants
(1.791 ±0.017 pg) was not different from that of
females (1.789 ±0.019 pg) but significantly lower
than that of males (1.835 ±0.019 pg). MADC2 was
absent from all monoecious plants. These results
strongly support that cultivars of monoecious hemp
have the XX constitution and that their sex expression
has a genetic basis.
Keywords Cannabis sativa Genome size
Monoecy Photoperiod Sex chromosomes Sex
expression
Introduction
In a context of increased attention for alternative
crops, the potential of the non-food species hemp
(Cannabis sativa L.) has been raised in the production
of fibre, bio-composites or paper pulp (van der Werf
et al. 1996; Struik et al. 2000; Ranalli and Venturi
2004).
Agriculturally, growing hemp is significantly
affected by its photoperiodism and reproductive
features. Firstly, hemp is a short-day plant with a
critical photoperiod of approximately 14 h (Amaducci
A.-M. Faux (&)A. Berhin P. Bertin
Earth and Life Institute, Universite
´Catholique de
Louvain, Croix du Sud 2 – 11 (L7.05.11),
1348 Louvain-la-Neuve, Belgium
e-mail: anne-michelle.faux@uclouvain.be
P. Bertin
e-mail: pierre.bertin@uclouvain.be
N. Dauguet
de Duve Institute, Avenue Hippocrate 74 (B1.74.04),
1200 Woluwe-Saint-Lambert, Belgium
123
Euphytica (2014) 196:183–197
DOI 10.1007/s10681-013-1023-y
et al. 2008a; Borthwick and Scully 1954; Lisson et al.
2000). By modulating the flowering time, the photo-
periodic conditions can have a key influence on the
crop yield (van der Werf et al. 1994). Secondly, the
species is naturally dioecious and characterised by
sexual dimorphism, in plant size and precocity in
particular. Monoecious cultivars of hemp have been
developed from plants bearing hermaphrodite flowers
or bisexual inflorescences (Moliterni et al. 2004).
These cultivars display several agronomical advanta-
ges in comparison to the dioecious ones, such as higher
seed yields, higher crop homogeneity and easier
mechanical harvest due to synchronised maturity
(Mandolino and Carboni 2004). However, their sexual
phenotype is unstable. The monoecious state presents
a continuous distribution between the male and female
extreme phenotypes (Bocsa and Karus 1998), and the
multiplication of monoecious hemp seeds requires a
strict elimination of the sporadically occurring dioe-
cious male plants in order to prevent a gradual return
to the dioecy. In addition, the sexual phenotype of
hemp is affected by external factors, such as hormonal
treatments, the photoperiod or nitrogen status (Free-
man et al. 1980). In spite of the diversity and plasticity
of the intersexual forms of hemp, significant variations
of the sex expression were observed among monoe-
cious hemp cultivars in field trials, and higher seed
yields were obtained with the early and mid-early
feminised cultivars, suggesting that the sex expression
of monoecious plants could affect the seed yields
(Faux et al. 2013). Investigating the potential genetic
basis of the sex expression in hemp appears therefore
valuable for the improvement and cultivation of
monoecious hemp.
Hemp is a diploid species (2n=20) that includes
sex chromosomes (Hirata 1924). The chromosomes
XX are found in female plants, and XY are present in
male plants, with the Y chromosome larger than the X
one and larger than the autosomes (Yamada 1943 cited
by Sakamoto et al. 1995). The genome sizes of diploid
female and male plants have been estimated using flow
cytometry as 1,636 ±7.2 and 1,683 ±13.9 Mbp,
respectively, and the size difference between them
(47 Mbp, i.e., 2.8 % of the genome size of male plants)
has been attributed to the large long arm of the Y
chromosome (Sakamoto et al. 1998). The sex deter-
minism in dioecious hemp would be based on an X-to-
autosomes equilibrium and not on a Y-active mecha-
nism (Westergaard 1958; Ainsworth 2000). On the
opposite, the sex determinism of monoecious forms of
hemp remains widely unknown. Although Hoffman
(1961 cited by Truta et al. 2007) assumed the existence
of XX, XY and YY forms, Menzel (1964) observed
presumably XX chromosomes in monoecious hemp
plants, and recent studies did not reveal the presence of
male-associated DNA markers (MADC, for male-
associated DNA in Cannabis). However, the karyotype
was limited to the cultivar ‘Kentucky’ (Menzel 1964)
while very few monoecious plants have been screened
with MADC markers to date, i.e., 20 plants of cultivars
‘Bialobrzeskie’ and ‘Beniko’ (Mandolino et al. 1999)
and 6 plants of cultivars ‘Fibrimon’ and ‘Fleischmann’
(Torjek et al. 2002). Mandolino et al. (2002) stated that
there is no specific report on the chromosome set in
monoecious hemp. Furthermore, all of these studies
including monoecious hemp cultivars considered the
monoecious state as a qualitative trait. In this context,
assessing the genotypic variability of the sexual
phenotype expressed as a quantitative variable and
establishing the composition in sex chromosomes in a
large number of monoecious hemp plants would allow
setting preliminary information that can be of interest
for further studies on the sex determinism of monoe-
cious hemp.
Flow cytometry constitutes an indirect but advan-
tageous technique to show dissimilarities in chromo-
somal composition. Indeed, flow cytometry is able to
analyse quickly large populations of cells (Dolezel and
Bartos 2005) and highlight very small genome-size
differences (Costich et al. 1991; Vagera et al. 1994;
Dolezel and Go
¨hde 1995). Given the larger size of the
Y chromosome of hemp (Sakamoto et al. 1998), the
presence, in monoecious hemp, of a Y chromosome
similar to that found in male plants could be inferred
from the comparison of the genome sizes of monoe-
cious, female and male plants. In addition, several
DNA markers closely linked to the male phenotype
have been developed in hemp (Sakamoto et al. 1995;
Mandolino et al. 1999; Torjek et al. 2002; Peil et al.
2003; Sakamoto et al. 2005; Rode et al. 2005). Among
them, the MADC2 marker proved to be completely
associated to the male phenotype and should therefore
be located on the Y chromosome in a region excluded
from recombination during meiosis (Mandolino et al.
2002). The amplification of this marker by monoe-
cious plants would indicate the existence of a male-
associated fragment in the sex chromosomes of
monoecious hemp.
184 Euphytica (2014) 196:183–197
123
The present paper aims to contribute to the
comprehension of the sex determinism in monoecious
hemp. Two specific objectives are pursued: firstly,
assessing the genotypic variability of the sexual
phenotype expressed as a quantitative variable and,
secondly, establishing the constitution in sex chromo-
somes of monoecious hemp through the evaluation of
a large number of plants from distinct cultivars. The
characterisation of the sexual phenotype under several
photoperiodic conditions allowed to take the sensitiv-
ity of the trait to the photoperiod into account. Flow
cytometric analyses and the MADC2 DNA marker
(Mandolino et al. 1999) were used to address the
second specific objective. Finally, we discussed the
implications of the results of the present study for
future investigations on the sex determinism in
monoecious hemp.
Materials and methods
Genetic material
Six hemp cultivars were used, one dioecious, ‘Car-
magnola’, and five monoecious covering a wide range
of earliness: ‘Uso 31’ (very early), ‘Fedora 17’ (early),
‘Santhica 27’ and ‘Felina 32’ (mid-early), and ‘Epsi-
lon 68’ (late). The dioecious cultivar was obtained
from Assocanapa, Carmagnola (Italy), and all of the
monoecious ones were obtained from the Fe
´de
´ration
nationale des Producteurs de Chanvre (FNPC), Le
Mans (France).
Growth conditions
Three trials were conducted successively, the first two
in a greenhouse and the third one in a phytotron to
apply short photoperiods. Sowing was performed on
13 September 2011, 2 February 2012 and 18 may 2012
in each trial, respectively. The trial 1 included 20
plants from each of the five monoecious hemp
cultivars, and both trials 2 and 3 included 10 plants
per monoecious cultivar, with the exception of
‘Santhica 27’ in trial 1 with 15 plants and ‘Epsilon
68’ in trial 3 in which a male plant was found. In total,
95, 50 and 49 plants of monoecious hemp were grown
in trials 1, 2 and 3, respectively. In addition, trial 3
included 10 plants of the dioecious cultivar
‘Carmagnola’.
The set-point temperatures were 25/20 °C day/
night in all of the trials. The photoperiod was firstly
16/8 h during 22, 60 and 20 days and then shortened to
14/10, the natural daylength and 12/12 h in order to
promote flowering in trials 1, 2 and 3, respectively. In
trial 2, the natural daylength increased from 13 h at the
time of photoperiod shortening (2 April) to 14.5 h at
the end of the trial (27 April). In the greenhouse, the
plants received the natural daylight, and the daylength
was extended by Philips HPLR lamps (400 W). The
daylight intensity decreased from on average
580–170 lmol m
-2
s
-1
between September and
November (trial 1) and, conversely, increased from
on average 250 to 520 lmol m
-2
s
-1
between Feb-
ruary and April (trial 2). The light intensity used to
extend the natural daylength was approximately
40 lmol m
-2
s
-1
in both trials 1 and 2. In the
phytotron, lighting was performed by Philips HPI-T
plus lamps (400 W, *135 lmol m
-2
s
-1
). There
were 4 or 5 plants per lamp in both greenhouse and
phytotron.
For each trial, the sowing was performed in
dimpled germination plates. After 10 days, the seed-
lings were transplanted into 30 cm (height) by 18 cm
(diameter) pots. The substrate consisted of a 3:1:1
mixture of soil, sand and loam with 2.8 g per pot of
Osmocote
Ò
(14:13:13 NPK with slow release of
nutrients for up to six months) added just before
transplanting. The plants were watered by capillarity
twice per week.
Plant phenotyping
The sex of the dioecious hemp plants was noted. In the
monoecious hemp, the time of flowering, flowering
duration and sex expression were recorded. The
flowering time was defined as the time at which
closed male flowers or white styles were easily visible
at leaf axils. The sex expression was characterised by
the degree of monoecy modified from the scale of
Sengbusch (1952) by varying from 0 to 6 according to
the ratio between male and female flowers (Table 1).
The monoecy degree was recorded at each flowering
node independently once a week during the flowering
duration, i.e., from the first to last flower appearance.
However, the plants were checked until the end of
flowering, i.e., when male flowers were withering and
green seeds were present at most nodes. The exper-
iments stopped when this developmental stage was
Euphytica (2014) 196:183–197 185
123
reached, i.e., 63, 85 and 55 days after sowing in trials
1, 2 and 3, respectively. All of the monoecious plants
were phenotyped, i.e., 194 plants.
Estimation of the genome sizes using flow
cytometry
The nuclear genome size of monoecious and dioecious
hemp plants was measured using flow cytometry,
simultaneously with Glycine max (soybean) and Zea
mays (maize) as internal reference standards to
compare the peaks from distinct samples of hemp
nuclei. Both reference standards were chosen from
those recommended by Prac¸a-Fontes et al. (2011) for
their genome size, which is a priori relatively close and
distant from that of hemp for soybean and maize,
respectively (Sakamoto et al. 1998). Otto’s buffers
(Otto 1992; Dolezel and Go
¨hde 1995) were used for
nuclei isolation as recommended by Loureiro et al.
(2006) for plant species with low nuclear DNA content
(between 1.30 and 1.96 pg per 2C).
Fresh leaf samples of hemp (20 mg), soybean
(20 mg) and maize (50 mg) were collected from
mature leaves, mixed and chopped for 120 s with a
razor blade in a plastic Petri dish on ice containing
1 ml of the Otto I buffer used for nuclei isolation. A
larger proportion of maize leaf tissue was needed to
obtain sufficient nuclei. The resulting suspension
rested for 15 min with shaking every 5 min to
facilitate the nuclei extraction. The mixture was
thereafter filtered twice through 30 lm (Sakamoto
et al. 1998) and 10 lm nylon meshes successively
(Partec GmbH, Mu
¨nster, Germany). A volume of 200
and 500 ll of Otto I buffer was poured on the first and
second filter, respectively, to recover as many nuclei
as possible. The solution was then centrifuged for
nuclei precipitation (2,500 rpm, 5 min). The resulting
pellet was diluted in 100 ll of Otto I buffer before
being filtered through a 10 lm mesh. A volume of
800 ll of modified Otto II buffer (20 ml of 0.4 M
Na
2
HPO
4
.12H
2
O, 1 ml of PI, 1 ml of RNAse and
40 llofb-mercaptoethanol), used as staining buffer,
was added upon the filter. After 10 min, the solution
was analysed with the LSRFortessa cell analyser
(Becton–Dickinson Benelux N.V., Erembodegem)
from the de Duve Institute (Brussels, Belgium). For
each sample, a minimum of 10,000 events were
analysed at a maximum rate of 300 events s
-1
. The
fluorescence was measured using the 575/26 BP filter,
specific to the wavelengths emitted by propidium
iodide. The median, standard deviation and coefficient
of variation (CV) of the peaks were extracted using the
BD FACsDiva Software. For each sample, the genome
size of the hemp nuclei was estimated by dividing the
median of fluorescence of the hemp peak by that of
each standard peak multiplied by the genome size of
the standard. Genome sizes of 2.41 ±0.170 and
5.57 ±0.146 pg per 2C DNA for soybean and maize,
respectively, were used to calibrate the hemp peaks
(Prac¸a-Fontes et al. 2011).
The protocol (nuclei isolation and flow cytometry)
was developed on plants from trials 1 and 2, while the
presented results were obtained from 56 plants from
trial 3: 10 ‘Carmagnola’ (6 female and 4 male), 9 ‘Uso
31’, 9 ‘Fedora 17’, 8 ‘Santhica 27’, 10 ‘Felina 32’ and
9 ‘Epsilon 68’. The cytometric analyses were spread
out among six dates as a result of logistical constraints
linked to the use of the flow cytometer. For each plant
on each date of analysis, two distinct leaf samples
were probed independently at the flow cytometer.
PCR with male-associated DNA marker
Leaf DNA was extracted according to Murray and
Thompson (1980) with modifications as detailed
below. Frozen leaves were ground to a fine powder
in liquid nitrogen. The resultant powder was mixed
with 900 ll of extraction buffer (10 % Tris–HCl,
10 % EDTA, 2 % CTAB, 50 % NaCl, 2 % PVPP and
1%b-mercaptoethanol) and heated at 55 °C for 1 h
with upside-down shaking every 10 min. A volume of
500 ll of a 24:1 mixture of chloroform-isoamyl
alcohol (v/v) was added, shaken for 10 s and
Table 1 Sex-expression scale in monoecious hemp modified
from Sengbusch (1952)
Monoecy
degree
Sex ratio
0 100 % of male flowers
1 80–99 % of male flowers (strongly
masculinised node)
2 60–80 % of male flowers (masculinised node)
3 40–60 % of female and male flowers
4 60–80 % of female flowers (feminised node)
5 80–99 % of female flowers (strongly
feminised node)
6 100 % of female flowers
186 Euphytica (2014) 196:183–197
123
centrifuged (13,000 rpm for 10 min). The upper phase
was collected, and the extraction with chloroform-
isoamyl alcohol was repeated once. The final upper
phase was collected, mixed with 350 ll of isopropanol
at 4 °C, shaken for 10 s, rested for 30 min and
centrifuged (13,000 rpm for 10 min). The supernatant
was drained, and the resultant DNA pellet was mixed
with 500 ll of 70 % ethanol at -20 °C and centri-
fuged (13,000 rpm for 5 min). This last step was
repeated once before dissolving the pellet in 50 llTE
buffer (10 mM Tris–HCl pH8, 1 mM EDTA pH8).
The extracted DNA was kept at -20 °C.
The presence of the male-associated DNA marker
MADC2 (Mandolino et al. 1999) was tested. The
reaction volume was 12.5 ll and contained 6.25 ng of
genomic DNA, 0.4 lM of the MADC2 primer (Euro-
gentec SA, Seraing, Belgium), 80 lM of each dNTP,
1.6 mM of MgCl
2
, 1.25 U of GoTAQ and 1X of
GoTAQ buffer (Promega Corp, Madison, USA). The
PCR amplifications were performed with a PTC 100
thermal cycler (MJ Research, Waltham, Mass):
10 min at 95 °C, 40 cycles (30 s at 95 °C, 45 s at
58 °C and 2 min at 72 °C) and 4 min at 72 °C. The
amplification products and a DNA ladder (Smartlad-
der, Eurogentec SA, Seraing, Belgium) were separated
on 1 % agarose gel run at a constant voltage of 90 V.
The gels were stained with ethidium bromide and
displayed under UV light.
The MADC2 primer was preliminarily tested on ten
plants of the dioecious cultivar ‘Carmagnola’ (five of
each sex) and then applied on 115 plants of monoe-
cious hemp (from 21 to 26 plants per cultivar).
Statistical analyses
The statistical analyses were carried out to assess
(i) the effects of the cultivar and trial on the flowering
time, flowering duration and sex expression in
monoecious hemp and (ii) the effects of the sex
(monoecious, female or male), cultivar and date of
analysis on the genome size of hemp nuclei.
The effect of the experimental factors on the
flowering time and flowering duration was assessed
by using the following model:
yijk ¼lþaiþbjþðabÞij þeijk ð1Þ
where y
ijk
was the flowering time or flowering duration
of the kth hemp plant of cultivar iin trial j;lwas the
general mean response; a
i
,b
j
and (ab)
ij
were the fixed
effects of cultivar i, trial jand their interaction,
respectively; e
ijk
was the error associated to the kth
plant of cultivar iin trial j. The model (1) was tested
with the GLM procedure from the SAS statistical
package (SAS Institute Inc. 2012).
The sex expression, quantified by the degree of
monoecy (Table 1), is a discrete variable with seven
response levels and was therefore statistically ana-
lysed using the GLIMMIX procedure with a multi-
nomial response distribution (DIST =MULTI) and
cumulative logit link function (LINK =CUMLO-
GIT) (SAS Institute Inc. 2012). The analysis was
performed with the model (2) below, which integrates
random effects of the ‘‘plant’’ and ‘‘node’’ nested to
the plant in addition to the ‘‘cultivar’’ and ‘‘trial’’ fixed
effects:
yijknp ¼lþaiþbjþðabÞij þdijk þfijkn þeijknp ð2Þ
where y
ijknp
was the monoecy degree of the nth node
of the kth hemp plant of cultivar iin trial jobserved at
time p;l,a
i
,b
j
and (ab)
ij
had the same meaning as in
model (1); d
ijk
was the random effect associated to
the kth plant of cultivar iin trial j;f
ijkn
was the
random effect associated to the nth node of the kth
plant of cultivar iin trial j;e
ijknp
was the error
associated to the nth node of the kth plant of cultivar
iin trial jobserved at time p. The option MAX-
OPT =100 was used to obtain the convergence of
the model.
The statistical analysis of the genome size was
firstly simplified by incorporating the ‘‘sex’’ effect of
the dioecious cultivar into the ‘‘cultivar’’ effect, which
consequently included seven levels (one for each of
the five monoecious cultivars and two for each sex of
the dioecious cultivar), and secondly by using the
mean genome size computed for each plant on each
analysis date. The effect of the experimental factors on
the genome size was tested in the GLM procedure
(SAS Institute Inc. 2012) by using the structure of
model (1), where y
ijk
was the genome size of the kth
hemp plant of cultivar or sex ianalysed at the jth date,
a
i
,b
j
and (ab)
ij
were the fixed effects of cultivar or sex
i, date jand their interaction, respectively, and e
ijk
was
the error associated to the kth plant of cultivar or sex
ianalysed at the jth date. An entire mixed model,
which included the fixed effects of ‘‘sex’’, ‘‘cultivar’’,
‘‘date’’ and their interactions as well as the random
Euphytica (2014) 196:183–197 187
123
‘‘plant’’ effect and its interaction with ‘‘date’’, was
firstly tested, and we verified that the interpretation of
the fixed effects was the same as in model (1).
For each observed variable—flowering time,
flowering duration, monoecy degree or genome
size—the MEANS procedure and the CONTRAST
statement of the GLM or GLIMMIX procedures
were used to compute means and pairwise compar-
isons, respectively.
Results
Flowering phenology and records of the sex
expression
The plants started to flower 34, 57 and 35 days after
sowing in trials 1, 2 and 3, respectively (as a reminder,
the photoperiod was shortened 22, 60 and 20 days
after sowing in trials 1, 2 and 3, respectively). The
010 30 50 70
Time to flowering (days)
Trial 1 Trial 2 Trial 3
0 7 14 21 28
Flowering duration (days)
0 1 2 3 4 5 6
Monoecy degree
Uso Fed San Fel Eps
(20) (20) (15) (20) (20)
aab cd bc d
(10) (10) (10) (10) (10) (10) (10) (10) (10) (9)
(20) (20) (15) (20) (20)
ab abbb
(10) (10) (10) (10) (10) (10) (10) (10) (10) (9)
abc ab cc
(77) (73) (42) (66) (59)
aa
ab ab
(26) (23) (15) (20) (12) (16) (16) (16) (17) (12)
Uso Fed San Fel Eps Uso Fed San Fel Eps
Uso Fed San Fel Eps Uso Fed San Fel Eps Uso Fed San Fel Eps
Uso Fed San Fel EpsUso Fed San Fel Eps
Uso Fed San Fel Eps
a
b
c
Fig. 1 Flowering phenology and sex expression in five
monoecious hemp cultivars. aTime to flowering, bflowering
duration and cdegree of monoecy (Table 1) (mean ±SD). The
photoperiod was 16/8 h during 22, 60 and 20 days and then
14/10, the natural daylength and 12/12 h in trials 1, 2 and 3,
respectively. In trial 2, the natural daylength increased from
13 h at the time of photoperiod shortening (2 April) to 14.5 h at
the end of the trial (27 April). ‘Uso’ to ‘Eps’ refer to the cultivars
‘Uso 31, ‘Fedora 17’, ‘Santhica 27’, ‘Felina 32’ and ‘Epsilon
68’, respectively. Significant differences among cultivars
(P\0.05) were indicated by different letters above the bars
when the ‘‘cultivar’’ effect was significant (P\0.05). The
number of data points is given between brackets for each
‘‘cultivar 9trial’’ treatment. For the monoecy degree, each data
point corresponds to the mean value computed on all flowering
nodes of a given plant at a given observation time
188 Euphytica (2014) 196:183–197
123
appearance of new flowers was noted until 55, 77
and 49 days after sowing in each trial, respectively.
The mean flowering time was 36.1, 63.0 and
37.3 days, and the mean flowering duration was
18.6, 10.6 and 7.7 days in trials 1, 2 and 3,
respectively (Fig. 1a, b).
The differences in flowering time among trials were
significant only between trial 2 and both trials 1 and 3
(P\0.001; Table 2). The cultivar significantly
affected the flowering time in trial 2 only
(P\0.001). However, the ranking of cultivars
according to their flowering time was similar across
trials, resulting in no significant ‘‘cultivar x trial’’
interaction and an overall significant ‘‘cultivar’’ effect
(P\0.001). The relative flowering times of the five
monoecious cultivars were in agreement with their
earliness as stated by the FNPC: ‘Uso 31’ was the
earliest one, followed by ‘Fedora 17’, ‘Felina 32’,
‘Santhica 27’ and ‘Epsilon 68’, the latest one.
The differences in flowering duration were sig-
nificant among all three trials (P\0.001; Table 2).
Similarly to the flowering time, the flowering
duration was affected by the cultivar in trial 2 only
(P\0.05), and no significant ‘‘cultivar x trial’’
interaction was found. The flowering duration was
on average longer in ‘Fedora 17’ and ‘Uso 31’, the
earliest cultivars, than in ‘Santhica 27’ and ‘Epsilon
68’, the latest ones, and intermediate in ‘Felina 32’
over the three trials.
As a result of the flowering phenology, the sex
expression was recorded four times in trial 1 at 7 days
intervals (34, 41, 48 and 55 days after sowing), three
times in trial 2 (57, 62 and 70 days after sowing) and
only twice in trial 3 (38 and 45 days after sowing).
From the first to the last observation time, the number
of phenotyped nodes per plant varied from 1 to 14 in
trial 1 (mean ±SD =5.8 ±2.9), from 1 to 16 in trial
2 (8.4 ±3.6) and from 1 to 12 in trial 3 (5.7 ±2.7).
Sex expression
The dioecious cultivar ‘Carmagnola’, grown in trial 3,
included six female plants, characterised by racemes
with leafy bracts and only female flowers, and four
male plants, characterised by hanging panicles with
few or no leaves and only male flowers (Fig. 2a, b). All
of the plants from the five monoecious hemp cultivars
showed inflorescences similar to those of female
plants from dioecious hemp and bore female and male
flowers arising in variable amounts (Fig. 2c).
The sex expression of the monoecious hemp plants
was affected by the cultivar and trial (P\0.001)as well
as by their interaction (P\0.01) (Table 3). The plants
were more masculinised in trial 2 (mean monoecy
degree ±SD =3.55 ±1.95), i.e., in the trial in which
the duration of the long-photoperiod treatment was
longer, than in both trials 1 (4.88 ±1.98) and 3
(5.05 ±1.34) (P\0.001; Fig. 1c). However, the
monoecy degree varied significantly among cultivars
only in trials1 (P\0.001) and 2 (P\0.05) and among
trials only in three cultivars, ‘Fedora 17’ (P\0.01) and
both ‘Felina 32’ and ‘Epsilon 68’ (P\0.001), resulting
in a significant ‘‘cultivar x trial’’ interaction. Neverthe-
less, the ranking of the cultivars according to their
monoecy degree remained consistent across the trials
(Fig. 1c). The cultivar ‘Uso 31’ was the most mascu-
linised one (mean monoecy degree ±SD =3.36 ±
2.28). It was followed by ‘Santhica 27’ (4.47 ±1.97),
‘Fedora 17’ (4.78 ±1.91), ‘Felina 32’ (5.28 ±1.39)
and ‘Epsilon 68’ (5.70 ±0.81), the most feminised
cultivar. Thus, the rankings of cultivars according to
earliness and sex expression agreed with each other,
with the exception of ‘Fedora 17’ and ‘Santhica 27’, the
former being earlier and more feminised than the latter.
The differences in monoecy degree between the culti-
vars were highly significant (P\0.001) between ‘Uso
31’ and both ‘Felina 32’ and ‘Epsilon 68’ and between
Table 2 Flowering phenology: ANOVA table for the time of flowering and flowering duration
Dependent Source DF SS MS F value Prob F
Time of flowering Cultivar 4 544.1 136.0 9.1 \0.0001
Trial 2 25,517.4 12,758.7 853.6 \0.0001
Cultivar 9trial 8 232.7 29.1 1.9 0.0560
Flowering duration Cultivar 4 239.8 60.0 4.2 0.0031
Trial 2 4,184.7 2,092.3 145.2 \0.0001
Cultivar 9trial 8 77.4 9.7 0.7 0.7160
Euphytica (2014) 196:183–197 189
123
‘Epsilon 68’ and both ‘Fedora 17’ and ‘Santhica 27’.
The differences were also significant (P\0.05)
between ‘Uso 31’ and both ‘Santhica 27’ and ‘Fedora
17’ as well as between ‘Santhica 27’ and ‘Felina 32’.
Genome size
The analysis of each sample by flow cytometry
allowed the detection of three peaks that corresponded
to hemp, soybean and maize nuclei successively. The
CVs of the peaks ranged between 1.5 and 3.4 % for
maize, 2.6 and 4.7 % for soybean and 2.6 and 5.0 %
for hemp.
Despite absolute values that were very close to each
other, the genome size of the hemp nuclei was
significantly affected by the sex of the plant, defined
as monoecious, female or male, regardless of the
reference standard (Fig. 3;Table4). The genome size
of the male ‘Carmagnola’ plants was significantly larger
(P\0.001) than that of all of the other plants, i.e., both
female ‘Carmagnola’ plants and the plants from all of
the monoecious cultivars. The female plants were not
significantly different from the monoecious plants, nor
were the monoecious cultivars significantly different
from each other. The mean genome sizes estimated with
soybean as the reference standard were 1.791 ±0.017,
1.789 ±0.019 and 1.835 ±0.019 (pg) in monoecious,
female and male plants, respectively, and the corre-
sponding sizes estimated with maize as the reference
standard were 1.838 ±0.020 (pg), 1.835 ±0.017 and
1.883 ±0.018, respectively. The date of analysis
significantly affected the genome size of the nuclei
(Table 4). However, no interaction was observed
between the fixed effects (‘‘cultivar or sex’’ and ‘‘date’’).
Male-associated DNA marker
The MADC2 primer amplified one male-associated
band of 390 bp—the MADC2 marker—in the male
plants of the dioecious cultivar ‘Carmagnola’, as
expected (Mandolino et al. 1999). This male-associ-
ated band was absent from all of the female ‘Car-
magnola’ plants (Fig. 4a). Moreover, the MADC2
marker was absent from all of the plants belonging to
monoecious hemp cultivars (Fig. 4b for cultivar
‘Fedora 17’).
Discussion
Flowering phenology
The variation of flowering time among trials was in
agreement with the short-day photoperiodic reaction
Fig. 2 Inflorescences found in dioecious and monoecious hemp: afemale and bmale plants from the dioecious cultivar ‘Carmagnola’
and ca plant from the monoecious cultivar ‘Uso 31’
Table 3 Sex expression in monoecious hemp: test of fixed
effects on the monoecy degree (Table 1)
Effect Num DF Den DF F value Prob F
Cultivar 4 159.99 8.90 \0.0001
Trial 2 170.51 15.05 \0.0001
Cultivar 9trial 8 157.79 2.66 0.0092
190 Euphytica (2014) 196:183–197
123
of hemp: the plants started to flower earlier in the trials
in which the photoperiod was shortened earlier (trials
1 and 3). The late flowering observed in trial 2 should
be due to a longer photoperiod-induced phase (PIP) of
development, the duration of which is determined by
the number of hours exceeding the critical photoperiod
[approximately 14 h in hemp (Amaducci et al. 2008a;
Borthwick and Scully 1954; Lisson et al. 2000)] and
the genotypic sensitivity to the photoperiod (Lisson
et al. 2000). Thus, the genes conferring the photope-
riod sensitivity were likely further stimulated by the
long 16-h day treatment of trial 2, so that the flowering
was delayed and the variation of flowering time among
cultivars was more significant in this trial.
The differences in flowering duration observed
among trials (on average 18.6, 10.6 and 7.7 days,
respectively) could be due to the photoperiod experi-
enced by the plants after the flowering time, which was
14 and 12 h in trials 1 and 3, respectively, and
increased from 13 to 14.1 h during the flowering
period in trial 2 (between 60 and 77 days after
sowing). In field trials, Amaducci et al. (2008b)
observed that the flowering duration of hemp plants
was shorter in late sown treatments, which flowered
under decreasing daylengths. However, a relatively
high difference in flowering duration was observed
between trials 1 and 2, which experienced relatively
close photoperiods. It is possible that the flowering
duration was affected by the time of photoperiod
shortening—which occurred 14 days before and close
to the flowering time in trials 1 and 2, respectively—or
that the response of the flowering duration to the
photoperiod is not linear. In soybean, a short-day
species, the flowering duration is affected by pho-
toperiods above a critical threshold, which varies
according to the genotype (Summerfield et al. 1998).
1.70 1.75 1.80 1.85 1.90
Genome size of hemp nuclei (pg)
Uso Fed San Fel Eps CarF CarM
abaaaaa
(9) (9) (8) (10) (9) (6) (4)
abaaaaa
(9) (9) (8) (10) (9) (6) (4)
Uso Fed San Fel Eps CarF CarM
ab
Fig. 3 Genome size of monoecious and dioecious hemp nuclei
(mean ±SD) estimated using asoybean and bmaize as the
reference standard. ‘Uso’ to ‘Eps’ refer to the five monoecious
cultivars ‘Uso 31, ‘Fedora 17’, ‘Santhica 27’, ‘Felina 32’ and
‘Epsilon 68’, respectively, and ‘CarF’ and ‘CarM’ refer to
female and male plants of the dioecious cultivar ‘Carmagnola’.
The values are the overall mean genome size to which the
residual values obtained from an ANOVA including ‘‘date’’ as
cofactor were added (Costich et al. 1991). Different letters
above the bars indicate significant differences (P\0.001). The
number of data points is given between brackets for each
cultivar
Table 4 Flow cytometry: ANOVA table for the genome size (GS) of hemp nuclei estimated using soybean or maize as the reference
standard
Dependent variable Source
a
DF SS MS F value Prob F
GS—soybean Cultivar or sex 6 0.017 0.003 8.78 \0.0001
Date 5 0.009 0.002 5.89 0.0003
(Cultivar or sex) 9date 16 0.004 0.000 0.86 0.6157
GS—maize Cultivar or sex 6 0.019 0.003 7.98 \0.0001
Date 5 0.010 0.002 5.33 0.0006
(Cultivar or sex) 9date 16 0.006 0.000 0.98 0.4949
a
The analysis was performed considering seven levels for the ‘‘cultivar or sex’’ factor: one for each of the five monoecious cultivars
and two for each sex of the dioecious cultivar
Euphytica (2014) 196:183–197 191
123
In addition, the flowering duration was longer in the
earliest cultivars (‘Uso 31’, ‘Fedora 17’) than in the
latest ones (‘Santhica 27’, ‘Epsilon 68’). Amaducci
et al. (2008b) observed a large genotypic variation in
the flowering duration among hemp cultivars; how-
ever, no link between earliness and flowering duration
was drawn. Additional trials in controlled conditions
are needed to explain the response of the flowering
duration to the photoperiod and genotype.
The sex expression in response to the genotype
Both previous field trials (Faux et al. 2013) and the
present study revealed (i) a significant effect of the
cultivar on the sex expression of monoecious hemp
plants and (ii), despite a significant ‘‘cultivar x
trial’’ interaction in the present study, the same
ranking of cultivars according to their sex expres-
sion. The consistency of the genotypic effect on the
sex expression across environments supports that
the sex expression in monoecious hemp has a
genetic basis.
The rankings of cultivars according to their sex
expression and earliness partly agreed with each other.
In dioecious hemp, the association of earliness and
maleness is known: the male plants flower generally
earlier than the female ones (Bocsa and Karus 1998;
Struik et al. 2000). Borthwick and Scully (1954)
attributed the excess of male over female plants
observed in long photoperiods to the lower sensitivity
to the photoperiod of the male plants. In monoecious
hemp, Amaducci et al. (2008b) suggested that early
plants have male characteristics. These observations
support the hypothesis that both traits share a common
basis in their genetic determinism. Effects of exoge-
nous hormones on sex have been reported in hemp:
gibberellin induces predominantly male plants, while
auxin, cytokinin and abscisic acid induce the femini-
sation of the plants (Heslop-Harrison 1956; Mohan
Ram and Jaiswal 1972; Chailakhyan and Khryanin
1978,1979; Freeman et al. 1980). In addition,
Chailakhyan and Khryanin (1978,1979) observed
that plants treated with gibberellin produced floral
buds earlier than the controls, whereas the auxin
Fig. 4 PCR products
obtained with the male-
associated DNA marker
MADC2 applied on aten
plants of the dioecious hemp
cultivar ‘Carmagnola’, male
(m) and female (f), and
btwelve plants of the
monoecious hemp cultivar
‘Fedora 17’. The arrows
indicate the male-specific
band of 390 bp long. L:
Smartladder (bp)
192 Euphytica (2014) 196:183–197
123
treatment delayed the flowering. In the long-day
model plant Arabidopsis thaliana, gibberellin is
required for early flowering and flowering under
non-inductive short days (Wilson et al. 1992). Inves-
tigating the role of gibberellin in hemp flowering
appears therefore to be an interesting pathway to
dissect the relation between earliness and maleness.
However, this does not rule out the involvement of
other hormones. Besides, given the sensitivity to the
photoperiod of both hemp flowering and sex expres-
sion, it would be worthwhile to integrate the photo-
period in the study of the hormonal regulation of hemp
flowering.
The sex expression in response to the environment
The pattern of variation of the sex expression among
trials suggested that the ‘‘trial’’ effect on the sex
expression was primarily due to the photoperiod
(Fig. 1c). Indeed, a masculinising effect of long
photoperiods has been observed in hemp (Borthwick
and Scully 1954; Arnoux 1966; Freeman et al. 1980).
In the present study, the duration of the 16-h day
treatment was the longest in trial 2 (60 days vs. 22 and
20 days in trials 1 and 3, respectively), resulting in a
higher production of male flowers in this trial.
However, the significant ‘‘cultivar x trial’’ interac-
tion pointed out, firstly, that the ‘‘cultivar’’ effect on
the monoecy degree decreased from trial 1 to trial 3
(Fig. 1c; Table 3). This observation could be
explained by the flowering duration or by both the
photoperiodic and light conditions. Indeed, the ‘‘cul-
tivar’’ effect on the monoecy degree varied among
trials similarly to the flowering duration, suggesting
that a long flowering duration would allow a higher
differentiation of the sex expression among the
monoecious hemp cultivars. In addition, according to
Borthwick and Scully (1954), a high light intensity
induces a greater production of male flowers on female
plants. The light conditions of trials 1 and 2—
performed in the greenhouse under natural light
artificially extended—would therefore be more
favourable to the production of male flowers than
those of trial 3—performed in phytotron under artifi-
cial light only. However, the photoperiodic conditions
were less masculinising in trials 1 and 3 than in trial 2,
as a result of their shorter duration of 16-h day
treatment (Borthwick and Scully 1954; Arnoux 1966;
Freeman et al. 1980). Therefore, the light and
photoperiodic conditions would have opposite effects
on the sex expression in trial 1. It would be possible
that the cultivars were affected differentially by these
conditions, resulting in a relatively high genotypic
variability of the sex expression in this trial. On the
opposite, the light and photoperiodic conditions would
be both inductive for the production of male flowers in
trial 2, and, conversely, both relatively little inductive
in trial 3, inducing thereby less variable genotypic
responses. Further studies are clearly necessary to
elucidate how the genotypic response of the sex
expression is affected by the environment.
Secondly, the sex expression varied significantly
among trials only in the most feminised cultivars
(‘Fedora 17’, ‘Felina 32’ and ‘Epsilon 68’). On the
opposite, in field trials performed with the same five
present cultivars (Faux et al. 2013), the variation of sex
expression with sowing date was the highest in the
most masculinised cultivars (‘Uso 31’ and ‘Santhica
27’). In view of the effect of the light intensity on the
sex expression in hemp (Borthwick and Scully 1954),
the inconsistency observed between field and con-
trolled conditions could be due to the light quality and
intensity. In the field, the plants grew under natural
daylight only, while they received both natural and
artificial light in the greenhouse and artificial light
only in the phytotron. In addition, the daylight
intensity is higher during the field growth season of
hemp (from April to September) than during the
periods of the present greenhouse trials (from Sep-
tember to April). According to Borthwick and Scully
(1954), the cultural conditions could also affect the sex
expression of the plants, since differences in the
production of male flowers by female hemp plants
were observed between field and greenhouse experi-
ments conducted simultaneously under natural day-
light and photoperiod.
Flow cytometry: practical considerations
The quality of the nuclei suspension prepared for the
purpose of flow cytometry is properly judged by
analysing the histogram of the relative fluorescence
intensity, which accounts for the relative DNA content
(Dolezel and Bartos 2005; Loureiro et al. 2006). The
CVs of the peaks obtained in the present study were
acceptable according to the 5 % threshold given by
Dolezel and Bartos (2005).
Euphytica (2014) 196:183–197 193
123
Flow cytometry has been widely used to determine
the absolute nuclear DNA content of various plant
species (Dolezel and Bartos 2005; Prac¸a-Fontes et al.
2011). In this context, the choice of an internal
reference standard is considered to be one of the main
issues (Dolezel and Bartos 2005). Here, the genome
size of hemp nuclei was function of the reference
standard—soybean or maize. This observation was
due to a difference between the peak ratio between
both reference standards (average ratio =0.44) and
their genome size ratio given by Prac¸a-Fontes et al.
(2011) (2.41/5.57 =0.43) and used here to calibrate
the hemp peaks. Such discrepancies among laborato-
ries are common issues in flow cytometry (Dolezel
et al. 1998; Prac¸a-Fontes et al. 2011). According to
Dolezel and Bartos (2005), the genome sizes estimated
using soybean as a reference standard were more
accurate that those obtained using maize since
soybean and hemp have closer genome sizes.
The effect of the date of analysis on the genome size
observed in this study (Table 4) was similar to the
‘‘run’’ effect reported by Costich et al. (1991) and
Taliaferro et al. (1997). According to Taliaferro et al.
(1997), the differences in genome size among dates of
analysis could result from nuances associated with
sample preparation or machine calibration. Moreover,
cytosolic compounds might interact with PI fluores-
cence, as noted with anthocyanin in Euphorbia
pulcherrima by Bennett et al. (2008). Hemp produces
a variety of secondary metabolites, such as cannabi-
noids, unique to the species (de Meijer et al. 2003),
flavonoids, stilbenoids, terpenoids, alkaloids and
lignans (Flores-Sanchez and Verpoorte 2008), which
could cause significant variations in the genome-size
estimation. In the present study, the significant ‘‘date’’
effect was removed according to Costich et al. (1991)
for the computation of the mean genome size of hemp
nuclei (Fig. 3).
Genome size of the monoecious hemp
The use of flow cytometry revealed significant differ-
ences in genome size between male hemp plants vs.
both monoecious and female ones. To our knowledge,
the present study constitutes the first report on the
genome size of monoecious hemp. Considering a
conversion factor of 1 bp =0.978 910
9
DNA con-
tent (pg) (Dolezel and Greilhuber 2010), the genome
sizes calculated using soybean as the reference
standard were 1,751 ±16, 1,750 ±18 and 1,795 ±
18 Mbp for monoecious, female and male hemp
plants, respectively. Thus, the genome size of monoe-
cious hemp was similar to that of female plants.
Though higher, the values of genome size obtained
in the present study are in the range of those found by
Sakamoto et al. (1998) for male and female hemp
plants (1,683 ±13.9 and 1,636 ±7.2 Mbp, respec-
tively). The difference between both studies could
firstly be explained by the use of distinct reference
standards (Dolezel and Bartos 2005), i.e., Arabidopsis
thaliana by Sakamoto et al. (1998)vs. soybean and
maize in the present study. Secondly, Sakamoto et al.
(1998) used the DAPI fluorochrome, which preferen-
tially binds to AT bases and therefore leads to DNA-
content estimations that should be interpreted with
caution (Dolezel et al. 1992). Thirdly, the genome size
of 130 Mbp that was considered for Arabidopsis
thaliana by Sakamoto et al. (1998) has been reassessed
at 157 Mbp by the Arabidopsis Genome Initiative
(2000). Nevertheless, the difference in percentage of
the genome size between male and female plants
found in our study (2.7 %) was very close to the 2.8 %
reported by Sakamoto et al. (1998) and to the 2.7 %
found by J. Suda (Charles University, Czech Republic,
‘pers. comm.’).
Male-associated DNA marker
The male-associated band amplified by the MADC2
primer corresponded to the marker described by
Mandolino et al. (1999). The adequacy of this marker
to discriminate male from female plants in the
dioecious cultivar ‘Carmagnola’ as observed here
confirmed the results of Mandolino et al. (1999;2002).
The present study showed the absence of the
MADC2 marker in 115 plants obtained from monoe-
cious hemp cultivars. This result amplified the absence
of this marker observed in the monoecious cultivars
‘Bialobrzeskie’ and ‘Beniko’ by Mandolino et al.
(1999).
General discussion
The species Cannabis sativa is characterised by
heteromorphic sex chromosomes. Female plants of
dioecious hemp have XX chromosomes, and male
ones have XY chromosomes. The present study
showed a similitude of genome size between
194 Euphytica (2014) 196:183–197
123
monoecious plants and female plants of dioecious
hemp and the absence of the male-associated DNA
marker MADC2 in monoecious hemp, which is likely
located on the Y chromosome in a region excluded
from recombination (Mandolino et al. 2002). These
findings suggest that monoecious hemp has basically
the same sex chromosomes than female plants of
dioecious hemp, i.e., XX chromosomes, and confirm
the cytological observations made in the ‘Kentucky’
cultivar by Menzel (1964).
The sex expression in monoecious hemp varies
quantitatively among plants as well as, according to
our results, among cultivars and should therefore be a
heritable quantitative trait. Considering that monoe-
cious hemp has the XX constitution regarding sex
chromosomes, the genetic determinism of sex expres-
sion in monoecious hemp should necessarily involve
genes that promote the production of male flowers by
cytologically female hemp plants. Consistent varia-
tions of sex expression and earliness among cultivars
were found, and both traits are sensitive to the
photoperiod. Therefore, it is possible that genes
responsible for the flowering response to the photo-
period are involved in the determinism of the sex
expression in monoecious hemp. A very high degree
of polymorphism has been noted in Cannabis (Faeti
et al. 1996; Forapani et al. 2001). Regarding the
structure of the genetic diversity, Forapani et al.
(2001) suggested the existence of a single and widely
shared gene pool, with a proportion of among-cultivar
variation strongly dependent upon the examined
cultivars. Given the polymorphism revealed in the
species and that sex expression in monoecious hemp
appears to be a heritable quantitative trait, investigat-
ing the sex determinism of monoecious hemp through
the identification of quantitative trait loci (QTL)
linked to the sex expression seems relevant. To this
purpose, the characterisation of distinct hemp cultivars
made in the present study can be useful to select
contrasting parents for the creation of a segregating
population.
In conclusion, the present study provided new
insights and established fundamental information for
further studies on the sex determinism of monoecious
hemp. Firstly, the sex expression in monoecious hemp
varies quantitatively and significantly among cultivars.
The ranking of cultivars according to their sex expres-
sion was maintained across environments and partly
consistent with their earliness. Secondly, the genome
size of monoecious hemp plants is not significantly
differentfrom that of female plants but lowerthan that of
male plants. Thirdly, the male-associated DNA marker
MADC2 is absent from monoecious hemp. These
results support that monoecious hemp has the XX
constitution in sex chromosomes and that its sex
expression has a genetic basis. In addition, they suggest
that the sex determinism in monoecious hemp could be
approached through the identification of quantitative
trait loci linked to the sex expression.
Acknowledgments We thank Prof. Bernadette Govaerts and
the platform Statistical Methodology and Computing Support
(UCL, Belgium) for their valuable advice on statistics, Dr. Olivier
Hardy (ULB, Belgium) and Prof. Jan Suda (Charles University,
Czech Republic) for their advice on flow cytometry, and the
Fe
´de
´ration Nationale des Producteursde Chanvre (France) for the
genetic material. This work was supported by the Fonds de la
Recherche Scientifique (Grant Number 1.5.279.08).
References
Ainsworth C (2000) Boys and girls come out to play: the
molecular biology of dioecious plants. Ann Bot 86:
211–221
Amaducci S, Colauzzi M, Bellocchi G, Venturi G (2008a)
Modelling post-emergent hemp phenology (Cannabis sa-
tiva L.): theory and evaluation. Eur J Agron 28:90–102
Amaducci S, Colauzzi M, Zatta A, Venturi G (2008b) Flowering
dynamics in monoecious and dioecious hemp genotypes.
J Ind Hemp 13:5–19
Arabidopsis Genome Initiative (2000) Analysis of the genome
sequence of the flowering plant Arabidopsis thaliana.
Nature 108:796–815
Arnoux M (1966) Influence des facteurs du milieu sur
l’expression de la sexualite
´du chanvre monoı
¨que (Can-
nabis sativa L.). II: Action de la nutrition azote
´e. Ann
Ame
´lior Plantes 16:123–134
Bennett MD, Price HJ, Johnston S (2008) Anthocyanin inhibits
propidium iodide DNA fluorescence in Euphorbia pulch-
errima: implications for genome size variation and flow
cytometry. Ann Bot 101:777–790
Bocsa I, Karus M (1998) The cultivation of hemp: botany,
varieties, cultivation and harvesting. Hemptech,
Sebastopol
Borthwick HA, Scully NJ (1954) Photoperiodic responses in
hemp. Bot Gaz 116:14–29
Chailakhyan MK, Khryanin VN (1978) The influence of growth
regulators absorbed by the root on sex expression in hemp
plants. Planta 138:181–184
Chailakhyan MK, Khryanin VN (1979) The role of leaves in sex
expression in hemp and spinach. Planta 144:205–207
Costich D, Meagher T, Yurkow E (1991) A rapid means of sex
identification in Silene latifolia by use of flow cytometry.
Plant Mol Biol Rep 9:359–370
de Meijer EPM, Bagatta M, Carboni A, Crucitti P, Moliterni
VMC, Ranalli P, Mandolino G (2003) The inheritance of
Euphytica (2014) 196:183–197 195
123
chemical phenotype in Cannabis sativa L. Genetics
163:335–346
Dolezel J, Bartos J (2005) Plant DNA flow cytometry and
estimation of nuclear genome size. Ann Bot 95:99–110
Dolezel J, Go
¨hde W (1995) Sex determination in dioecious
plants Melandrium album and M. rubrum by using high-
resolution flow cytometry. Cytometry 19:103–106
Dolezel J, Greilhuber J (2010) Nuclear genome size: are we
getting closer? Cytometry Part A 77A:635–642
Dolezel J, Sgorbati S, Lucretti S (1992) Comparison of three
DNA fluorochromes for flow cytometric estimation of
nuclear DNA content in plants. Physiol Plantarum 85:
625–631
Dolezel J, Greilhuber J, Lucretti S, Meister A, Lysak MA, Nardi
L, Obermayer R (1998) Plant genome size estimation by
flow cytometry: inter-laboratory comparison. Ann Bot
82(Suppl. A):17–26
Faeti V, Mandolino G, Ranalli P (1996) Genetic diversity of
Cannabis sativa L. germplasm based on RAPD markers.
Plant Breed 115:367–370
Faux AM, Draye X, Lambert R, d’Andrimont R, Raulier P,
Bertin P (2013) The relationship of stem and seed yields to
flowering phenology and sex expression in monoecious
hemp (Cannabis sativa L.). Eur J Agron 47:11–22
Flores-Sanchez IJ, Verpoorte R (2008) Secondary metabolism
in Cannabis. Phytochem Rev 7:615–639
Forapani S, Carboni A, Paoletti C, Moliterni VMC, Ranalli P,
Mandolino G (2001) Comparison of hemp varieties using
random amplified polymorphic DNA markers. Crop Sci
41:1682–1689
Freeman DC, Harper KT, Charnov EL (1980) Sex change in
plants: old and new observations and new hypotheses.
Oecologia 47:222–232
Heslop-Harrison J (1956) Auxin and sexuality in Cannabis sa-
tiva. Physiol Plantarum 9:588–597
Hirata K (1924) Cytological basis of the sex determination in
Cannabis sativa L. Jpn J Genet 4:198–201
Lisson SN, Mendham NJ, Carberry PS (2000) Development of a
hemp (Cannabis sativa L.) simulation model. 2. The
flowering response of two hemp cultivars to photoperiod.
Aust J Exp Agr 40:413–417
Loureiro J, Rodriguez E, Dolezel J, Santos C (2006) Compari-
son of four nuclear isolation buffers for plant DNA flow
cytometry. Ann Bot 98:679–689
Mandolino G, Carboni A (2004) Potential of marker-assisted
selection in hemp genetic improvement. Euphytica 140:
107–120
Mandolino G, Carboni A, Forapani S, Faeti V, Ranalli P (1999)
Identification of DNA markers linked to the male sex in
dioecious hemp (Cannabis sativa L.). Theor Appl Genet
98:86–92
Mandolino G, Carboni A, Bagatta M, Moliterni VMC, Ranalli P
(2002) Occurrence and frequency of putatively Y chro-
mosome linked DNA markers in Cannabis sativa L. Eu-
phytica 126:211–218
Menzel MY (1964) Meiotic chromosomes of monoecious
Kentucky hemp (Cannabis sativa). Bull Torrey Bot Club
91:193–205
Mohan Ram HY, Jaiswal VS (1972) Induction of male flowers
on female plants of Cannabis sativa by gibberellins and its
inhibition by abscisic acid. Planta 105:263–266
Moliterni VMC, Cattivelli L, Ranalli P, Mandolino G (2004)
The sexual differentiation of Cannabis sativa L.: a mor-
phological and molecular study. Euphytica 140:95–106
Murray MG, Thompson WF (1980) Rapid isolation of high
molecular weight plant DNA. Nucleic Acids Res 8:4321–4325
Otto F (1992) Preparation and staining of cells for high-reso-
lution DNA analysis. In: Radbruch A (ed) Flow cytometry
and cell sorting. Springer-Verlag, Berlin, pp 101–104
Peil A, Flachowsky H, Schumann E, Weber WE (2003) Sex-
linked AFLP markers indicate a pseudoautosomal region in
hemp (Cannabis sativa L.). Theor Appl Genet 107:102–109
Prac¸a-Fontes M, Carvalho C, Clarindo W, Cruz C (2011)
Revisiting the DNA C-values of the genome size-standards
used in plant flow cytometry to choose the ‘‘best primary
standards’’. Plant Cell Rep 30:1183–1191
Ranalli P, Venturi G (2004) Hemp as a raw material for indus-
trial applications. Euphytica 140:1–6
Rode J, In-Chol K, Saal B, Flachowky H, Kriese U, Weber WE
(2005) Sex-linked SSR markers in hemp. Plant Breed
124:167–170
Sakamoto K, Shimomura K, Komeda Y, Kamada H, Satoh S
(1995) A male-associated DNA sequence in a dioecious
plant, Cannabis sativa L. Plant Cell Physiol 36:1549–1554
Sakamoto K, Akiyama Y, Fukui K, Kamada H, Satoh S (1998)
Characterization, genome sizes and morphology of sex
chromosomes in hemp (Cannabis sativa L.). Cytologia
63:459–464
Sakamoto K, Abe T, Matsuyama T, Yoshida S, Ohmido N,
Fukui K, Satoh S (2005) RAPD markers encoding retro-
transposable elements are linked to the male sex in Can-
nabis sativa L. Genome 48:931–936
SAS Institute Inc (2012) SAS
Ò
9.3 help and documentation.
SAS Institute Inc., Cary, NC
Sengbusch RV (1952) Ein weiterer beitrag zur vererbung des
geschlechts bei hanf als grundlage fu
¨r die zu
¨chtung eines
mono
¨zischen hanfes. Z Pflanzenzu
¨chtung 31:319–338
Struik PC, Amaducci S, Bullard MJ, Stutterheim NC, Venturi G,
Cromack HTH (2000) Agronomy of fibre hemp (Cannabis
sativa L.) in Europe. Ind Crop Prod 11:107–118
Summerfield RJ, Asumadu H, Ellis RH, Qi A (1998) Charac-
terization of the photoperiodic response of post-flowering
development in maturity isolines of soyabean [Glycine max
(L.) Merrill] ‘Clark’. Ann Bot 82:765–771
Taliaferro CM, Hopkins AA, Henthorn JC, Murphy CD,
Edwards RM (1997) Use of flow cytometry to estimate
ploidy level in Cynodon species. Int Turfgrass Soc Res J
8:385–392
Torjek O, Bucherna N, Kiss E, Homoki H, Finta-Korpelova Z,
Bocsa I, Nagy I, Heszky L (2002) Novel male-specific
molecular markers (MADC5, MADC6) in hemp. Euphy-
tica 127:209–218
Truta E, Olteanu N, Surdu S, Zamfirache MM, Oprica L (2007)
Some aspects of sex determinism in hemp. Analele Stiin-
tifice ale Universitatii ‘‘Alexandru Ioan Cuza’’, Sectiunea
Genetica si Biologie Moleculara VIII:31–39
Vagera J, Paulikova D, Dolezel J (1994) The development of
male and female regenerants by in vitro androgenesis in
dioecious plant Melandrium album. Ann Bot 73:455–459
van der Werf HMG, Hassken HJ, Wijlhuizen M (1994) The
effect of daylength on yield and quality of fibre hemp
(Cannabis sativa L.). Eur J Agron 3:117–123
196 Euphytica (2014) 196:183–197
123
van der Werf HMG, Mathijssen EWJM, Haverkort AJ (1996)
The potential of hemp (Cannabis sativa L.) for sustainable
fibre production: a crop physiological appraisal. Ann Appl
Biol 129:109–123
Westergaard M (1958) The mechanism of sex determination in
dieocious flowering plants. Adv Genet 9:217–281
Wilson RN, Heckman JW, Somerville CR (1992) Gibberellin is
required for flowering in Arabidopsis thaliana under short
days. Plant Physiol 100:403–408
Euphytica (2014) 196:183–197 197
123
- ... The plasticity in this trait has led to some confusion regarding the genetic basis of sex, and the role of the Y chromosome and autosomes in sex determination in Cannabis. Some authors list Cannabis as having XY sex chromosomes with an X:0 sex determination system (Faux et al., 2014), which means that the Y chromosome does not contribute to sex determination, and the differences between males and females are produced by the number of X chromosomes present. Others argue that sex is governed by an active Y system and autosomes are not involved (Sakamoto et al., 1998), similar to the sex determination system present in humans. ...
- ... Furthermore, hemp has a highly variable sexual phenotype. In a study by Faux et al. 2014, the authors demonstrate the expression in monoecious hemp varies quantitatively and significantly among cultivars. This variation can be attributed to the karyotype polymorphisms detected in our study. ...
Article
Hemp (Cannabis sativa L.) is a diploid species including both dioecious and monoecious cultivars with hetero- and homomorphic sex chromosomes, respectively. It displays a high plasticity of sex expression, i.e., the ratio of female and male flowers. In this study, we investigated the role of sex chromosomes in the genetic determinism of sex expression in dioecious and monoecious hemp. The... [Show full abstract]
Article
Flowering phenology and sexual dimorphism are two major features that affect stem and seed production in hemp (Cannabis sativa L.), a short-day naturally dioecious plant. The sowing time is of primary importance because it affects flowering time and thereby influences stem yield. In spite of their unstable sexual phenotype, monoecious cultivars facilitate the harvest of both stems and seeds by... [Show full abstract]
Article
Hemp (Cannabis sativa L., 2n = 20) is a dioecious plant. Sex expression is controlled by an X-to-autosome balance system consisting of the heteromorphic sex chromosomes XY for males and XX for females. Genetically monoecious hemp offers several agronomic advantages compared to the dioecious cultivars that are widely used in hemp cultivation. The male or female origin of monoecious maternal... [Show full abstract]
Article
Sex expression is of primary importance for the genetic improvement and production of monoecious hemp: masculinized phenotypes are associated with higher fungal sensitivity, and feminized phenotypes with higher seed yields. However, sex expression varies quantitatively among plants and nodes and with time. Here, we developed eight variables that characterize the sex expression in monoecious... [Show full abstract]
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.
This publication is from a journal that may support self archiving.
Learn more