Plant Science 171 (2006) 546554

www.elsevier.com/locate/plantsci

Stable genetic transformation of high oleic Helianthus annuus

L. genotypes with high efficiency
Sh. Mohamed, R. Boehm, H. Schnabl *
Institute of Plant Molecular Physiology and Biotechnology, University of Bonn, Karlrobert-Kreiten-Str. 13, 53115 Bonn, Germany
Received 25 February 2006; received in revised form 29 May 2006; accepted 30 May 2006
Available online 27 June 2006

Abstract
Two previously optimized procedures for genetic transformation of two economically important high oleic sunflower genotypes (H. annuus cv.
capella and SWSR2 inbred line) were used to generate stably transformed sunflower plants of these genotypes expressing gus reporter gene:
Agrobacterium tumefaciens-mediated gene transfer to juvenile split apical meristems was applied to cv. capella, and a biolistic gene transfer
method using the same target tissue was applied to SWSR2 inbred line. Out of 120 originally used explants of cv. capella, five transgenic
regenerants could be identified by biochemical and molecular analysis. Thus, transformation efficiency amounted to 4.1%. For SWSR2 inbred line,
125 explants were used and six transgenic regenerants could be identified in an identical manner, resulting in a transformation efficiency of 4.8%.
The highest expression level of the gus gene resulted in a yield of recombinant protein of approx. 0.2% total soluble protein. Additionally, T1
progeny was tested for gus gene inheritance in both genotypes. From approx. 75% of the T0 plants, the gus gene was stably transmitted to the T1
progeny and expressed in a detectable amount. To test the reproducibility of the transformation procedures, they were used to introduce another
reporter gene, mgf5, into both genotypes. The resulting transformation efficiencies were 3.3 for both genotypes. gfp proved to be a suitable reporter
gene to select transgenic regenerants in early developing stages. The data presented here demonstrate for the first time the reproducible
transformation of high oleic sunflower genotypes with high frequency.
# 2006 Elsevier Ireland Ltd. All rights reserved.
Keywords: Gus gene; Helianthus annuus L.; Meristem transformation; Mgfp5 gene; Particle bombardment; Vacuum infiltration

1. Introduction
Sunflower (Helianthus annuus L.) is one of three most
important annual oil-bearing crops world-wide following
soybean (Glycine max L.) and rapeseed (Brassica napus L.)
[1]. The oil of high oleic sunflower genotypes can be used as
food oil or deep-frying fat [2,3]. Moreover, the high oxidative
performance of oleic acid and its very low content of
polyunsaturated fatty acids combined with low content of
stearic acid make them suitable for industrial applications like
cosmetics, pharmaceuticals, detergents, lubricants, metal
working fluids, surfactants or for chemical synthesis. However,

Abbreviations: 4-MU, 4-methylumbelliferone; nos, nopaline synthase; gus,

b-glucuronidase; nptII, neomycin phosphotransferase gene; gfp, green fluorescent protein; hptII, hygromycin phosphotransferase gene; MUG, 4-methylumbelliferyl glucuronide; BAP, 6-benzaminopurine; MS, murashige and
skoog; CaMV, cauliflover mosaic virus; rpm, rounds per minute
* Corresponding author. Tel.: +49 228 732832; fax: +49 228 731696.
E-mail address: h.schnabl@uni-bonn.de (H. Schnabl).
0168-9452/$ see front matter # 2006 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.plantsci.2006.05.012

sunflower is known as one of the most recalcitrant species for

tissue culture and genetic transformation. Therefore, progress
in sunflower transformation has been restricted for many years
by the limitations of available regeneration systems and
problems combining regeneration and transformation within
the same cells [4]. The efforts so far resulted in very low
transformation frequencies [57] and are highly genotype
dependent. It is especially desirable to optimize the genetic
transformation of high oleic sunflower genotypes.
Recently we have optimized the transformation conditions
of two economically important high oleic sunflower genotypes
(cv. capella and SWSR2 inbred line) by transient reporter gene
expression. These rely either on Agrobacterium-mediated gene
transfer [8] or a particle bombardment approach (manuscript
submitted).
Based on these previous studies, we here report the
application of the transformation methods for both genotypes
to generate stably transformed sunflower plants, using the gus
(b-glucuronidase) reporter gene. It was performed to assess the
resulting transformation efficiency and to test the inheritance of

S. Mohamed et al. / Plant Science 171 (2006) 546554

the introduced gus reporter gene to the T1 generation.

Moreover, we used the methods to introduce the mgf5 reporter
gene, coding for the green fluorescent protein (GFP), in an
effort to test the reproducibility of the transformation protocols
and to compare the transmission of gus and gfp genes. The data
presented here demonstrate for the first time the stable
transformation of economically important high oleic sunflower
genotypes with reproducibly high frequency.
2. Material and methods
2.1. Plant material and explant preparation
Seeds of high oleic sunflower (Helianthus annuus L.), cv.
capella and inbred line SWSR2, kindly provided by Sudwestsaat (Rastatt, Germany), were surface sterilized for 1 min with
70% (v/v) ethanol, rinsed in a 6% (w/v) sodium hypochlorite
solution plus one drop tween-20 for one hour and washed three
times with sterile water. Seeds germinated on MS medium [9]
containing MS salts 2.3 g l1, sucrose 2% (w/v), 2-(nmorpholino) ethanesulfonic acid (MES) 3.2 mM and phytoagar
7.5 g l1. The pH was adjusted to 5.7 with NaOH (1 M).
Seedlings were cultivated for 10 days in a growth chamber at
25  1 8C and a light period of 12 h (115 mE m2 s1). After
10 days, aseptic shoot apices (45 mm length) were bisected
longitudinally according to Ref. [10] and used as explants in
this study.
2.2. Agrobacterium strain, plasmids and culture
Agrobacterium tumefaciens LBA4404 strain [11] was used,
carrying either plasmid pBI121 [12] or pCAMBIA 1302 (http://
www.cambia.org.au/). pBI121 contains the gus gene under the
transcriptional control of CaMV 35S promoter and the nptII
selectable marker gene under the control of nos (nopaline
synthase gene) promoter. pCAMBIA1302 contains the mgfp5
gene [13] and the hptII selectable marker gene both under the
transcriptional control of CaMV 35S promoter.
Cultivation of the bacteria was done on YEB medium [14]
on plates supplemented with 100 mg l1 kanamycin. For
transformation, they were cultured overnight on YEB liquid
medium supplemented with 100 mg l1 kanamycin at 28 8C
with continuous shaking at 200 rpm under the same antibiotic.
Cells from overnight culture were centrifuged at 4000 rpm for
15 min at room temperature, washed once in one volume MS
medium and centrifuged again under the same conditions.
Finally, the cells were resuspended in MS medium to achieve an
OD600 value of 1 and 200 mM acetosyringone were added. The
bacterial suspension was incubated for 2 h at room temperature
without shaking prior to use.
2.3. Plasmid isolation, preparation of the gold particle and
coating with DNA

547

on LB medium containing 100 mg l1 kanamycin and the

following preparation of the plasmid DNA using the alkaline
lysis method were performed as described by Ref. [15].
Preparation of the gold particles (1.6 mm) was performed
according to the method of Ref. [16] and stored at 20 8C at a
final concentration of 60 mg ml1. The particle suspension
was thawed when required and vortexed vigorously to
resuspend the particles. 50 ml of microcarriers were taken in
1.5 ml eppendorf tube and while vortexing continuously (for
uniform DNA precipitation onto microcarriers) the following
were added sequentially: 5 ml DNA (1 mg ml1), 50 ml 2.5 M
CaCl2 and 20 ml 0.1 M spermidine. Contents were vortexed for
56 min, microcarriers allowed to settle for 1 min, pelleted by
spinning for 2 s at 14,000 rpm and the liquid removed and
replaced by 140 ml of 70% (v/v) ethanol for washing. The first
washing was followed by washing with 100% ethanol and
finally particles were resuspended in 48 ml of 100% ethanol.
These coated particles were kept at 4 8C and used within 1 h of
preparation. 6 ml of the coated particles suspension were
loaded onto the macrocarrier membrane which was allowed to
dry prior to use.
2.4. Vacuum infiltration of agrobacteria
Split shoot apices were prepared as described above, precultured on SIM2 (see below) for 3 days and divided into
groups. Each group was immersed in 2 ml Agrobacterium
suspension for 30 min and transferred to vacuum infiltration
flask (desiccator). Vacuum was applied at 150 mbar for 1 min,
then rapidly ventilated. The infiltration procedure was repeated
two times. Agrobacterium tumefaciens suspension was
removed and the explants were dried using Whatman 3 MM
filter paper, then cultivated in Petri dishes containing shoot
induction medium (SIM2, see below). After co-cultivation for 3
days, the explants were washed twice in liquid SIM2 containing
250 mg l1 cefotaxime, blotted on sterile Whatman 3 MM filter
paper and transferred to solid SIM2 supplemented with
250 mg l1 cefotaxime to eliminate the agrobacteria without
using selection agents. Dishes were sealed with parafilm and
cultured as stated below.
2.5. Particle bombardment
Split shoot apices were pre-cultured on SIM2 for one day
prior to bombardment and grouped in the center of 4 cm Petri
dishes on 2% (w/v) agarose (the cut surface facing up). The
target materials were positioned 6 cm below the microprojectile stopping screen and 1550 psi rupture discs were used.
The explants were bombarded twice according to Ref. [16]
using Biolistic1 PDS-1000/He particle delivery system
(Biorad, Hercules, USA). Explants bombarded with uncoated
particles were used as a control.
2.6. Regeneration of transgenic plants

E. coli DH5a carrying the plasmid pBI121 or pCAMBIA1302 was used as plasmid source for particle bombardment
experiments and positive controls. The culture of the bacteria

Transformed explants were cultured on shoot induction

medium (SIM2) containing MS salts 4.3 g l1, myo-inositol

548

S. Mohamed et al. / Plant Science 171 (2006) 546554

0.56 mM, thiamineHCl 0.30 mM, glycine 26.64 mM, nicotinic

acid 4.1 mM, pyridoxineHCl 2.43 mM, sucrose 3% (w/v),
BAP (6-benzaminopurine) 0.44 mM, pH 5.7 and plantagar
6 g l1. Culture conditions were at 25  1 8C and a light period
of 12 h (115 mE m2 s1). After 3 weeks of shoot induction,
explants were subcultured on fresh SIM2 and incubated
under the same conditions. Regenerated shoots were rooted
on root induction medium (RIM1): MS salts 2.15 g l1, sucrose
1% (w/v), myo-inositol 0.28 mM, thiamineHCl 0.15 mM, 1naphtalene acetic acid (NAA) 0.54 mM, ancymidol 1.95 mM
and phytagel 4 g l1. The plantlets of cv. capella and SWSR2
inbred line with initial roots were transferred to different shoot
elongation and root development medium (SER2 medium): MS
salts 4.3 g l1, sucrose 3% (w/v), myo-inositol 0.56 mM,
thiamineHCl 1.19 mM, pH 5.7 and solidified with phytagel
4 g l1. The plantlets were grown in a chamber at 25  1 8C
and a light period of 15 h (147 mE m2 s1). Plantlets
with well-developed roots were carefully removed from the
tubes and washed to remove the agar. Then, they were
transferred to small pots containing a mixture (1:1) of soil and
garden-soil type P and cultured in a growth chamber at
25  1 8C and a light period of 12 h (115 mE m2 s1),
regularly irrigated until flowering, synthetic hybridization
and seed production [17].
2.7. Histochemical GUS activity assay
b-Glucuronidase (GUS) activity was assayed according to
Ref. [18]. Regenerated 1214 weeks old shoots were immersed
in GUS staining solution (0.1 M Na2HPO4 pH 7.0, 10 mM
NaEDTA, 0.5 mM K-ferricyanide, 0.5 mM K-ferrocyanide,
0.1% (v/v) Triton-X-100, 1 mM X-Gluc (5 bromo-4-chloro-3indolyl glucuronide) and 20% (v/v) methanol) [19]. Vacuum of
200 mbar was applied for 10 min. Then, the explants were
incubated overnight in the dark at 37 8C. Before microscopic
analysis, chlorophyll was bleached by extraction in an ethanol
series (70%, 96%) for 24 h. Untreated explants were cultured
under identical conditions and served as negative control. After
the histochemical assay the positive plants were kept for PCR
analysis.
2.8. Fluorometric GUS activity assay
GUS activity was measured quantitatively using a fluorometric assay based on Ref. [18]. 5, 10 and 1214 weeks after
co-cultivation, plant tissue (one leaf per sample) from
transformed and non-transformed samples was ground with a
pestle and mortar in the presence of liquid nitrogen. Tissue was
homogenized in microcentrifuge tube with extraction buffer
(2 ml g1 tissue), containing 50 mM NaH2PO4 pH 7.0, 10 mM
EDTA pH 8.0, 0.2% (v/v) Triton X-100 and 10 mM bmercaptoethanol. The homogenate was centrifuged to pellet
debris and the supernatant was collected as crude protein
extract. Protein content of the extract was quantified according
to Ref. [20] and mixed with 500 ml MUG solution (1 mM 4methylumbelliferyl glucuronide in 20% (v/v) methanol). The
reaction was carried out in the dark at 37 8C for 1 h and stopped

with 400 ml 0.2 M Na2CO3. The assays were analyzed in a

spectro fluorometer (Fluoro-Max, Spex, London, UK). The
fluorescence was recorded at an excitation wavelength of
365 nm and an emission wavelength of 455 nm. The readings
were compared to readings of 4-MU (4-methylumbelliferone)
standards of varying concentrations and were plotted against
time to determine the amount of MU produced. GUS activity
was calculated as micromoles of 4-MU formed per milligram
protein per minute.
2.9. Histological GFP activity assay
GFP was visualized using a fluorescence microscope
(Eclipse TE300, Nikon, Dusseldorf, Germany) fitted with a
100 W-mercury lamp, a 450490 nm excitation filter and
505 nm barrier filter.
2.10. Fluorometric GFP activity assay
Fluorometric GFP activity assay was performed according
to Ref. [6]. Plant tissue was ground with a pestle and mortar in
two volumes of extraction buffer (50 mM TrisHCl, pH 8.0,
10 mM EDTA, 10 mM dithiotreitol, 18% (v/v) glycerol) in the
presence of liquid nitrogen. The extraction was centrifuged to
pellet debris and the supernatant was collected. Protein content
of the extracts was determined by the method of Ref. [20].
Preparations were diluted with buffer to 75 mg ml1 and
analyzed in a spectro fluorometer (Fluoro-Max, Spex,
Germany) using the excitation filter at 490 nm combined with
the emission filter at 525 nm.
2.11. Plant DNA extraction and polymerase chain reaction
analysis
Genomic DNA from transformed plants, positive in the
histochemical GUS activity assay, and random non-transformed plants was extracted according to the CTAB method
[21,22]. Detection of the gus or mgfp5 gene in the samples was
conducted using PCR with the following primers: gus primer
50 -ATGTTACGTCCTGTAGAAAC-30 and 50 -CTTCACTGCCACTGACCGGA-30 , which were designed to amplify a
fragment of approximately 830 bp of the gus gene and mgfp5
primer 50 -AAAGGAGAAGAACTTTTCACT-30 and 50 -TTTGTATAGTTCATCCATGCC-30 , which amplify an approximately 800 bp fragment of the mgfp5 gene. PCR reaction was
performed in 50 ml total volume containing 50100 ng EcoRI
digested genomic DNA, 5 ml 1.5 mM MgCl2, 5 ml 10 Taq
DNA polymerase buffer, 0.75 ml 0.2 mM dNTP, 12.5 ml
0.25 mM of each primer and 2 U Taq DNA polymerase. As
a positive control, the corresponding plasmid was used as a
template. DNA samples were denatured for 5 min at 95 8C and
amplified during 32 cycles, denaturation for 1 min at 95 8C,
annealing for 1 min at 64 8C, extension for 1 min at 72 8C.
Cycling was closed with a final extension step for 10 min at
72 8C. Amplified products were electrophoresed on a 0.8% (w/
v) agarose gel and the DNA was visualized with ethidiumbromide on a UV transilluminator [15] at 305 nm.

S. Mohamed et al. / Plant Science 171 (2006) 546554

2.12. Southern blot analysis

20 mg of genomic DNA were digested for 3 h at 37 8C with
either EcoRI or HindIII. Both enzymes are equal single-cutters
within the T-DNA region of pBI121. Digested samples were
separated on 0.8% (w/v) agarose gel. After incubation of the
gel in 0.25 M HCl for 30 min, it was immersed in denaturation
solution (1.5 M NaCl, 0.5 M NaOH) for 30 min and in
neutralizing solution (3 M NaCl, 0.5 M Tris/HCl pH 7.5) for
another 30 min. All steps were performed at room temperature.
Finally, the gel was blotted on a nitrocellulose membrane
(Hybond-N+, Amersham, Freiburg, Germany) overnight in a
capillary blot system using 20 SSC buffer (3 M NaCl, 0.3 M
Na-citrate). After DNA transfer, the membrane was exposed to
UV light for 5 min and baked for 2 h at 80 8C. The DNA probe
was labeled with [a-32P]-dCTP using the Random primed
hexalabeling DNA Kit (Fermentas, St. Leon-Rot, Germany)
following the manufacturers instructions. The pre-hybridization was carried out by using 50100 ml hybridization solution
at 65 8C for at least 3 h without labeled DNA. The
hybridization solution was refreshed and the hybridization
process performed overnight at 65 8C with labeled DNA probe.
The membrane was washed three times with 50 ml washing
buffer for 20 min each at 65 8C. The membrane was then
exposed to X-ray film (Hyper-film, Kodak, Stuttgart, Germany) for 35 days at 70 8C.
3. Results
An earlier developed Agrobacterium infiltration method
without tissue wounding proofed to be optimal for the genotype
cv. capella, but the same method applied to SWSR2 inbred line
resulted in approx. 50% less transformation efficiency [8]. In
contrast, an alternatively developed particle bombardment
approach resulted in approx. 20% higher transient gus
expression with SWSR2 inbred line compared to cv. capella
(manuscript submitted). Now, these both transformation
procedures were applied to the respective genotypes to
generate stably transformed gus expressing plants, to determine
the transformation efficiency, and to test the inheritance to the
next generation. Moreover, the transformation protocols were
applied to another reporter gene, mgfp5, to test the
reproducibility of the transformation methods.

549

Table 1
Summary of stable transformation events of T0 plants of high oleic H. annuus L.
genotypes, cv. capella using gus gene with Agrobacterium infiltration and
SWSR2 using biolistic gene transfer method
Genotype

cv. capella

SWSR2

Total number of used explants

Number of regenerated plants
after transformation
Number of plants tested with
fluorometric GUS activity assay
Number of positive plants in
fluorometric GUS activity assay
Number of plants tested with PCR
Number of PCR-positive plants
Transformation frequency (%) a

120
77

125
82

77

82

40
5
4.1

40
6
4.8

a
Transformation frequency was calculated on the basis of positive PCR
plants in relation to the total number of used explants. PCR was performed with
gus-specific primer 1416 weeks after co-cultivation or bombardment.

3.1. Transformation of shoot apices with gus reporter gene

120 split shoot apices of cv. capella were used for
transformation using the optimized Agrobacterium infiltration
protocol, and for SWSR2 inbred line, a total of 125 split shoot
apices was used for transformation using the optimized biolistic
protocol (see Table 1).
The initial screening was done by the fluorometric GUS
activity assay 5 weeks after co-cultivation for all regenerants.
Out of all regenerants, five and six plantlets of cv. capella and
SWSR2 inbred line, respectively, showed high enzyme activity
values (Fig. 1). The highest product formation (mean of
2154 mmol MU mg1 min1) was observed in plant no. 3 of cv.
capella. Based on this value and appropriate standard curves,
the expression level of recombinant GUS protein was
calculated to be approx. 0.2% total soluble protein.
The GUS positive plantlets have been further tested for gus
expression 10 and 1214 weeks after co-cultivation. In this time
frame, a detectable variation in the gus expression within the
same plant could be observed (Fig. 1). However, these effects
were not very pronounced, ranging from 10% to 20% loss of
original GUS activity.
The histochemical GUS activity assay of selected regenerants showed a uniform dark blue color in the transformed
tissues (data not shown). On the other hand, non-transformed

Fig. 1. Variation of GUS activity among gus expressing T0 plants of H. annuus (A) cv. capella and (B) SWSR2 inbred line over a period of 1214 weeks on different
development media. One leaf was ground each time with extraction buffer and the crude protein extract was used in fluorometric GUS activity assay. For details see
materials and methods. Data represent mean of three parallel measurements. The level of variation was <5%.

550

S. Mohamed et al. / Plant Science 171 (2006) 546554

Table 2
Analysis of gus gene transmission to the T1 progeny in high oleic H. annuus L. cv. capella ands SWSR2 inbred line. GUS expression was recorded by the
histochemical GUS activity assay using a random leaf of each T1 plant
Genotype and plant number

No. of seeds

Capella
Capella
Capella
Capella
Capella

1
2
3
4
5

3
5
7
10
9

SWSR2
SWSR2
SWSR2
SWSR2
SWSR2
SWSR2

1
2
3
4
5
6

11
5
9
13
7
10

GUS activity

No. of PCR positive plants

Transgenic T1 plants recovered

3
4
5
6
2

+
+
+
+


1
4
4
2
0

1/3
4/4
4/5
2/6
0/2

9
1
6
10
7
7

+

+
+



3
0
2
5
3
0

3/9
0/1
2/6
5/10
3/7
0/7

No. of germinated seeds

PCR was performed with digested DNA of all plants using gus specific primer.

tissue did not exhibit any blue color under identical assay
conditions.
The presence of respective T-DNA in 40 randomly chosen
regenerants, including the hitherto identified GUS positive
plantlets, was tested by PCR 1416 weeks after cocultivation,
using gus-specific primer. An amplification of the expected
830 bp fragment corresponding to the gus gene was observed in
all GUS positive plants, whereas no amplification was detected
in the samples from non-transformed plants (data not shown).
The resulting transformation frequency of cv. capella was
calculated on the basis of positive PCR plants and recorded as a
percentage from the total number of co-cultivated or
bombarded explants (Table 1). The frequency amounted to
4.1% using Agrobacterium infiltration method whereas this was
4.8% in SWSR2 inbred line using direct gene transfer method.
Finally, southern blot analysis of seven randomly chosen T0
plants confirmed the presence and integration of the gus
fragment gene into the sunflower genome of both genotypes
(Fig. 2).
Eventually, single or two hybridizing bands were observed
in cv. capella plants transformed with Agrobacterium infiltra-

tion method (lanes 1, 3, 7 and 9). Meanwhile, insertion events of

the gus gene into 24 loci of SWSR2 inbred line genome were
detected in the transformed plants using biolistic gene transfer
(lanes 2, 4 and 8). This proves the stable integration of the gus
gene into the two high oleic H. annuus L. genotypes.
Conversely, no hybridising band was observed in nontransformed cv. capella or SWSR2 inbred line (lanes 5 and 6).
3.2. Transgene expression and inheritance in the T1
progeny
After manual self-hybridization of the GUS positive T0
plants, the resulting seeds were collected and germinated to test
the inheritance and expression of the gus gene in the T1 plant
population. In approx. 75% of the recovered T1 plants of both
genotypes, the gus gene was stably transmitted and expressed in
a measurable amount, as tested by histochemical GUS activity
assay and PCR (Table 2). Plant number 5 of cv. capella and
plants numbers 2 and 6 of SWSR2 inbred line, however, showed
non-detectable expression levels in the subsequent generation
in histochemical GUS assay and were negative in PCR analysis.

Fig. 2. Southern blot hybridization of random independent T0 transgenic plants of two high oleic H. annuus L. genotypes, cv. capella and SWSR2 inbred line.
Genomic DNA was digested with EcoRI and hybridized with 32P-labelled gus probe. Lanes 1, 3, 7, 9 transgenic cv. capella plants (plants 4, 3, 2, and 1 respectively,
from Fig. 1), lanes 2, 4, 8 transgenic SWSR2 inbred line plants (plants 1, 4 and 6, respectively, from Fig. 1), lanes 5 and 6 non-transformed sunflower and lane 10
marker DNA. Numbering of T0 plants is according to Fig. 1.

S. Mohamed et al. / Plant Science 171 (2006) 546554

551

Fig. 3. Southern blot hybridization of DNA from T1 plants from two randomly chosen T0 plants of genotype cv. capella and SWSR2 inbred line transformed with the
gus gene. Genomic DNA was digested with HindIII and hybridized with 32P-labelled gus probe. (A) Lanes 14 transgenic cv. capella T1 plants from T0 plant 2, lane 5
non-transformed cv. capella and lane 6 marker DNA. (B) Lanes 1, 3, 4 transgenic SWSR2 inbred line T1 plants from T0 plant 1 and lane 2 non-transformed SWSR2
inbred line. Numbering of T0 plants is according to Fig. 1.

In SWSR2 inbred line, although the progeny of plant number 5

did not show any gus expression, three of them were positive in
PCR analysis (Table 2).
The progeny of two randomly chosen T0 plants were further
analysed by southern hybridization: plant 2 of genotype cv.
capella and plant 1 of SWSR2 inbred line (Fig. 3, numbering of
T0 plants according to Fig. 1). Two loci of gus gene insertion were
observed in the progeny of all four T1 plants of cv. capella T0
plant 2 (Fig. 3A lanes 14). However, the size of the two bands
was different from the respective size in the T0 plant (Fig. 2, lane
7). Since this difference is found equally in all T1 plants, it might
be explained with genomic rearrangements in early flower
development of the T0 plant. The three SWSR2 T1 plants
transformed by biolistic gene transfer showed three insertion loci
of the gus gene (Fig. 3B lanes 1, 3, 4) which is the same pattern as
in the respective T0 plant (Fig. 2, lane 2). The different band
intensities might be due to different gus gene copy numbers
within the loci. In one of these plants, the restriction pattern is
slightly different what might be a result of chromosomal
rearrangement nearby the integration site (Fig. 3B lane 3). No
hybridizing band was observed in non-transformed cv. capella or
SWSR2 inbred line (lanes 5 and 2, respectively).

3.3. Transformation of shoot apices using gfp gene

The optimized transformation methods for cv. capella and
SWSR2 inbred line were applied to introduce another reporter
gene, mgfp5 (coding for green fluorescent protein, GFP), in an
effort to monitor transgene delivery to plant tissue in early
developing stages and to test the reproducibility of the
transformation methods. A total of 120 explants of each
genotype was used for transformation. After regeneration, the
plantlets were subjected to histological, fluorometric and
molecular analysis.
Expression of GFP could be visually observed 2025 days
after co-cultivation or bombardment in the leaves of the
regenerated shoots. GFP-expressing tissue presented green
fluorescence, while non-expressing leaf tissue appeared dark
and did not fluoresce (data not shown). The intensity of the
green fluorescence detected ranged from faint to strong green.
Plantlets expressing GFP could be separated at this time and
grown until maturity. Out of 120 explants four plants from each
genotype were transgenic.
Fluorometric GFP activity assay was performed for both
the GFP positive plants in the histological assay and non-

Fig. 4. Relative fluorescence of crude protein extracts from leaves of positive gfp transgenic plants of high oleic H. annuus L. genotypes, cv. capella and SWSR2
inbred line, after background substraction. One leaf was ground each time with extraction buffer and the crude protein extract was used in fluorometric GFP activity
assay. For details see materials and methods. Data represent mean of three parallel measurements. The level of variation was <8%.

552

S. Mohamed et al. / Plant Science 171 (2006) 546554

Fig. 5. Southern blot analysis of two mgfp5 expressing T0 plant of two high
oleic H. annuus L. genotypes, cv. capella and SWSR2 inbred line. Genomic
DNA was digested with EcoRI and hybridized with 32P-labelled probe corresponding to mgfp5 gene. Lane 1: transgenic cv. capella plant, lane 2: transgenic
SWSR2 inbred line plant, lane 3: non-transformed sunflower, and lane 4:
marker DNA.

transformed plants. The presented results in Fig. 4 confirm mgfp5

expression in the histological positive plantlets. The intensity of
the gene expression strongly varied among the transformed
plants and between the two genotypes. In comparison, the
maximum fluorescence value recorded in cv. capella amounted to
777.8 relative fluorescence and resulted from plant number 3,
while the corresponding value in SWSR2 inbred line was 298
relative fluorescence and recorded from plant number 2.
Eventually, all the histological and fluorometric GFP positive
plantlets were also positive in PCR analysis whereas nonexpressing GFP plants showed negative results. The predicted
amplified bands appeared at 800 bp using primer specific to
mgfp5 gene (data not shown). Transformation frequency was
calculated on the basis of PCR analysis and recorded as
percentage from the total number of co-cultivated or bombarded
explants. This frequency amounted to 3.3% for both genotypes
using the optimized transformation method for each.
To confirm that the mgfp5 gene had stably integrated into the
genome of the two tested sunflower genotypes, southern blot
analysis was performed using one plant with positive PCR
results of each genotype in addition to non-transformed sample
(Fig. 5). At least one and two insertion loci of the mgf5 gene
could be detected in cv. capella and SWSR2 inbred line
genome, respectively. In contrast, no band was obtained from
the DNA of non-transformed sunflower.
These results clearly show that GFP has proved to be a
suitable reporter protein to select transgenic regenerants in
early developing stages.
4. Discussion
Up to date, all published sunflower transformation protocols
suffered from low overall transformation efficiencies which

varied between 0.1% with public inbred line HA300B [6],

0.52% with R105 [7] and 0.22% with RHA266 [5]. Thus, low
transformation efficiency is one of the main drawbacks in
todays application of genetic engineering to sunflower and
efficient protocols for economically important genotypes are
highly appreciated. Therefore, we developed genotype-specific
transformation protocols for a high oleic hybrid genotype (cv.
capella) and a high oleic inbred line (SWSR2) based on
transient reporter gene expression [8, manuscript submitted].
Here, the transformation protocols were successfully
applied to shoot meristems of both genotypes to generate
stably transformed regenerants for the first time. In agreement,
shoot meristems have been previously used to produce
genetically transformed plants in a range of plant species such
as sunflower [2326,10], soybean [27], maize [28], and cotton
[29].
Histochemical, fluorometric as well as molecular analysis
confirmed that the T-DNA was integrated into the sunflower
genome resulting in a transformation efficiency of 3.2% for cv.
capella and 4.8% for SWSR2 inbred line. This is much higher
than described earlier in comparable studies with other public
inbred lines [57] and maybe an effect of the genotypes used
but maybe also a result of the comprehensive optimization of
several parameters in the course of transformation protocol
development. Resulting recombinant protein yield was approx.
0.2% total soluble protein. Further factors influencing the yield
of active transgene product like choice of promoter, avoiding
gene silencing by reducing copy number or avoiding
homologous sequences can be considered in further applications but were not of primary interest in this study.
With regard to fluorometric GUS analysis after different
times, the obtained results showed the similarity of the gene
expression in each case. There were only slight variations in the
gus expression in the same plant during a period of 1214 weeks.
However, since only one leaf could be used per test, the origin of
the variations observed remains questionable. It can be due to
normal variation of transgene expression level in different parts
of the plant, influenced by the physiological state of the leaf (e.g.
degree of senescence) or may also resulting from transgene
methylation as it was repeatedly observed [3032].
The transgene also was successfully transmitted to the T1
progeny in most cases. However, the expected mendelian
fashion of inheritance was not observed, what probably, at least
partially, is due to the multiple insertion events of the gus gene
into the plant genome or simply due to the low number of
germinated seeds in some cases.
Southern blot analysis showed that the gus gene inserted into
a single or two loci of the cv. capella genome of selected T0 and
T1 plants using Agrobacterium infiltration method. This pattern
of insertion is very similar to the DNA insertion pattern
documented for Brassica napus plants [33] using the same
transformation method. However, biolistic gene delivery has
achieved up to four insertion loci of the gus gene into SWSR2
inbred line. These multiple insertions were also previously
reported in many plant species transformed via particle
bombardment such as cultivated jute [34], orchid [35] and
soybean [36].

S. Mohamed et al. / Plant Science 171 (2006) 546554

The transformation frequency reported here for high oleic H.

annuus L. genotypes, cv. capella (using Agrobacterium
infiltration method) and SWSR2 inbred line (using biolistic
gene transfer) was 4.1% and 4.8%, respectively. This is much
higher than those which have been reported for sunflower
transformation so far. However, the higher transformation
frequencies could be obtained routinely under a more optimal
set of conditions [37]. In agreement, using Agrobacterium
infiltration method with Petunia hybrida has also achieved high
transformation efficiency (10% in T1 plants) [38]. Moreover,
particle bombardment method has been successfully employed
and achieved high transformation efficiencies amounting to
3.25% for Hordeum vulgare L. [39], 8.6% for Triticum
aestivum var. CPAN1676, 7.5% for T. aestivum var. PBW343
and 4.9% for Triticum dicoccum DDK1001, respectively [40]
and 12% for Dendrobium phalaenopsis [35].
The introduction of mgfp5 gene into cv. capella and SWSR2
inbred line facilitated the monitoring of the transgene in the
plant tissue in early developing stages (2025 days after cocultivation or bombardment). This result agreed with several
reports such as [4143] who reported that GFP proved to be an
excellent reporter of early transformation events. Using a
fluorescence microscope, shoots expressing mgfp5 gene were
quickly and easily distinguished and four transgenic plants
were detected of each genotype.
The transformation frequency was calculated on the basis of
PCR analysis and recorded as a percentage of the total number
of co-cultivated or bombarded explants. This frequency
amounted to 3.3% for both genotypes and thus were
comparable to those recorded by using the gus gene. This
proves the reproducibility of the transformation methods in
connection with the genotypes used.
To our knowledge, the first user for gfp gene in sunflower
transformation was [6] who achieved stable transformation
using non-meristematic hypocotyl explants of public inbred
HA300B with efficiency of 0.1%. In this study, for the first time,
gfp gene was used with high efficiency (3.3%) in transformation
of shoot apices of high oleic H. annuus L. genotypes, cv. capella
and SWSR2 inbred line.
The results obtained can be used for biotechnological
applications of these genotypes, e.g. optimizing the fatty acid
spectrum for specific applications.
Acknowledgements
The authors want to thank Mrs. Grohne (Sudwestsaat) for
the provision of seeds of cv. capella and SWSR2 inbred line.
This project was supported by the Egyptian government (grants
to S.M.).
References
[1] S. Weber, W. Friedt, N. Landes, Molinier F J., Improved Agrobacteriummediated transformation of sunflower (Helianthus annuus L.): assessment of
macerating enzymes and sonication, Plant Cell Rep. 21 (2003) 475482.
[2] M. Fick, F. Miller, Sunflower breeding, in: A.A. Schneiter (Ed.), Sunflower Technology and Production, Agronomy, Madison, Wiscosin, USA,
1997, pp. 395495.

553

[3] G. Dorrell, A. Vick, Properties and processing of oilseed sunflower, in:

A.A. Schneiter (Ed.), Sunflower Technology and Production, Agronomy,
Madison, Wiscosin, USA, 1997, pp. 709745.
[4] I. Potrykus, Gene transfer to cereals: an assessment, Biotechnology 8
(1990) 535542.
[5] T. Hewezi, J. Kallerhoff, G. Alibert, Genetic transformation of a poor
tissue-culture responding genotype of sunflower, in: Proceedings of the
Fifth European Conference on Sunflower Biotechnology, San Giuliano
Therme, Italy, November 48, (2001), p. 31.
[6] A. Muller, M. Iser, D. Hess, Stable transformation of sunflower (H. annuus
L.) using a non-meristematic regeneration protocol and green fluorescent
protein as a vital marker, Transgen. Res. 10 (2001) 435444.
[7] O. Lucas, J. Kallerhoff, G. Alibert, Production of stable transgenic sunflowers (Helianthus annuus L.) from wounded immature embryos by
particle bombardment and co-cultivation with Agrobacterium tumefaciens, Mol. Breed. 6 (2000) 476487.
[8] S. Mohamed, R. Boehm, P.C. Binsfeld, H. Schnabl H, Agrobacteriummediated transformation of two high oleic sunflower (Helianthus annuus
L.) genotypes: assessment and optimization of important parameters,
Helia 27 (2004) 2549.
[9] T. Murashige, F. Skoog, A revised medium for rapid growth and bioassay
with tobacco tissue cultures, Physiol. Plant 15 (1962) 473497.
[10] J. Malone-Schoneberg, C.J. Scelonge, M. Burrus, L. Bidney L, Stable
transformation of sunflower using Agrobacterium and split embryonic axis
explants, Plant Sci. 103 (1994) 199207.
[11] A. Hoekema, P.R. Hirsch, P.J.J. Hooykaas, R.A. Schilperoort, A binary
plant vector strategy based on separation of vir- and T-region of the
Agrobacterium tumefaciens Ti-plasmid, Nature 303 (1983) 179180.
[12] P.-Y. Chen, C.-W. Wang, S.-C. Soong, K.-Y. To, Complete sequence of the
binary vector pBI121 and its application in cloning T-DNA insertion from
transgenic plants, Mol. Breed. 11 (2003) 287293.
[13] K.R. Siemering, R. Golbik, R. Sever, J. Haseloff, Mutations that supress
the thermosensitivity of green fluorescent protein, Curr. Biol. 6 (1996)
16531663.
[14] G. Vervliet, M. Holsters, H. Teuchy, M. van Montagu, J. Schell, Charaterization of different plaque-forming and defective temperate phages in
Agrobacterium strains, J. Gen. Virol. 26 (1975) 3348.
[15] J. Sambrook, D.W. Russell, Molecular Cloning: A Laboratory Manual,
3rd ed., Cold Spring Harbor Laboratory Press, New York, 2001.
[16] J.C. Sanford, F.D. Smith, J.A. Russell, Optimizing the biolistic process for
different biological application, Methods Enzymol. 217 (1993) 483509.
[17] S. Mohamed, P.C. Binsfeld, C. Cerboncini, H. Schnabl, Regeneration
systems at high frequency from high oleic Helianthus annuus L. genotypes, J. App. Bot. 77 (2003) 8589.
[18] R.A. Jefferson, T.A. Kavanagh, M.W. Bevan, GUS fusions: b-glucuronidase as sensitive and versatile gene fusion marker in higher plants,
EMBO J. 6 (1987) 39013907.
[19] S. Kusogi, Y. Ohashi, K. Nakajima, Y. Arai, An improved assay for bglucuronidase in transformed cells: methanol almost completely suppresses a putative endogenous b-glucuronidase activity, Plant Sci. 70
(1990) 133140.
[20] M.M. Bradford, A rapid and sensitive method for the quantitation of
microgram quantities of protein utilizing the principle of protein-dye
binding, Anal. Biochem. 72 (1976) 248254.
[21] J.J. Doyle, J.L. Doyle, A rapid DNA isolation procedure for small
quantities of fresh leaf tissue, Phytochem. Bull. 19 (1987) 1115.
[22] K.W. Cullings, Design and testing of a plant-specific PCR primer for
ecological and evolutionary studies, Mol. Ecol. 1 (1992) 233240.
[23] N. Knittel, V. Gruger, G. Hahne, P. Lenee, Transformation of sunflower
(Helianthus annuus L.): a reliable protocol, Plant Cell Rep. 14 (1994) 81
86.
[24] B. Schrammeijer, P.C. Sijmons, P.J.M. van den Elzen, A. Hoekema,
Meristem transformation of sunflower via Agrobacterium, Plant Cell
Rep. 9 (1990) 5560.
[25] M. Burrus, J. Molinier, C. Himber, R. Hunold, R. Bronner, P. Rousselin, G.
Hahne, Agrobacterium-mediated transformation of sunflower (Helianthus
annuus) shoot apices: transformation patterns, Mol. Breed. 2 (1996) 329
338.

554

S. Mohamed et al. / Plant Science 171 (2006) 546554

[26] J. Molinier, C. Thomas, M. Brignou, G. Hahne, Transient expression of

ipt gene enhances regeneration and transformation rates of sunflower
shoot apices (Helianthus annuus L.), Plant Cell Rep. 21 (2002) 251
256.
[27] D.E. McCabe, W. Swain, B. Martinell, P. Christou, Stable transformation
of soybean (Glycine max) by particle bombardment, Biotechnology 6
(1988) 923926.
[28] K. Lowe, B. Bowen, G. Hoerster, M. Ross, D. Bond, D. Pierce, B. GordonKamm, Germline transformation of maize following manipulation of
chimeric shoot meristems, Biotechnology 13 (1995) 677682.
[29] G. Keller, L. Spatola, D.E. McCabe, B. Matinell, W. Swain, E. Maliyakal
E, Transgenic cotton resistant to herbicide bialaphos, Trans. Res. 6 (1997)
385392.
[30] F. Linn, I. Heidmann, H. Saedler, P. Meyer, Epigenetic changes in the
expression of maize A1 gene in Petunia hybrida: role of numbers of
integrated gene copies and state of methylation, Mol. Gen. Genet. 222
(1990) 329336.
[31] N.J. Kilby, H.M. Lyser, I.J. Furner, Promoter methylation and progressive
transgene inactivation in Arabidopsis, Plant Mol. Biol. 20 (1992) 103
112.
[32] P. Lambe, M. Dinant, R. Matagne, Differential long-term expression and
methylation of the hygromycin phosphotransferase (hpt) and b-glucuronidase (GUS) genes in transgenic pearl millet (Pennisetum glaucum)
callus, Plant Sci. 108 (1995) 5162.
[33] W.C. Wang, G. Menon, G. Hansen, Development of a novel Agrobacterium-mediated transformation method to recover transgenic Brassica
napus plants, Plant Cell Rep. 22 (2003) 274281.

[34] M. Ghosh, T. Saha, P. Nayak, S.K. Sen, Genetic transformation by particle

bombardment of cultivated jute, Corchorus capsularis L, Plant Cell Rep.
20 (2002) 936942.
[35] S. Men, X. Ming, Y. Wang, R. Lui, W. Wei, Y. Li, Genetic transformation
of two species of orchid by biolistic bombardment, Plant Cell Rep. 21
(2003) 592598.
[36] M.S. Reddy, R.D. Dinkins, G.B. Collins, Gene silencing in transgenic
soybean plants transformed via particle bombardment, Plant Cell Rep. 21
(2003) 676683.
[37] Z. Zhao, T. Cai, L. Tagliani, M. Miller, N. Wang, H. Pang, M. Rudert, S.
Schroeder, D. Hondred, J. Seltzer, D. Pierce, Agrobacterium-mediated
sorghum transformation, Plant Mol. Biol. 44 (2000) 789798.
[38] D. Tjokrokusumo, T. Heinrich, S. Wylie, R. Potter, J. McComb, Vacuum
infiltration of Petunia hybrida pollen with Agrobacterium tumefaciens to
achieve plant transformation, Plant Cell Rep. 19 (2000) 792797.
[39] M. Manoharan, L.S. Dahleen, Genetic transformation of the commercial
barley (Hordeum vulgare L.), Plant Cell Rep. 21 (2002) 7680.
[40] A. Chugh, P. Khurana, Regeneration via somatic embryogenesis from leaf
basal segments and genetic transformation of bread and emmer wheat by
particle bombardment, Plant Cell Tissue Org. Cult. 74 (2003) 151161.
[41] M.C. Jordan, Green fluorescent protein as a visual marker for wheat
transformation, Plant Cell Rep. 19 (2000) 10691075.
[42] C.N. Stewart, The utility of green fluorescent protein in transgenic plants,
Plant Cell Rep. 20 (2001) 376382.
[43] F. Murray, R. Brettell, P. Matthews, D. Bishop, J. Jacobsen, Comparison of
Agrobacterium-mediated transformation of four barley cultivars using
GFP and GUS reporter genes, Plant Cell Rep. 22 (2004) 397402.