MODIFYING LISIANTHUS TRAITS BY GENETIC ENGINEERING
Michele Zaccai Efraim Lewinsohn
The Institutes for Applied Research Newe Ya’ar Research Center
Ben Gurion University of the Negev ARO
Beersheva 84105 Ramat Yishay 30095
Department of Biology,
University of Michigan
Ann Arbor, MI
Keywords: Lisianthus, genetic engineering, flowering time, fragrance, LFY, BEAT
The control of flowering time is an essential issue for most ornamentals. The molecular
basis of this trait has been extensively studied in model plants, yielding useful
information on the role of various genes which could potentially be used in heterologous
systems. For instance, overexpression of the meristem-identity gene LEAFY (LFY) from
Arabidopsis has previously been shown to be sufficient to reduce flowering time in plants
as diverse as Arabidopsis itself, tobacco and aspen. We tested the effect of LFY
overexpression on flower development in lisianthus plants transformed with LFY cDNA
under constitutive control of the 35S CaMV promoter (35S::LFY, provided by Dr.
Weigel). F1 progenies from two different transformed plants flowered in average one to
two weeks earlier than control plants at two consecutive years. The number of nodes to
the first flower as well as the number of flowers per inflorescence were reduced in the
early-flowering plants, suggesting a faster flower development. Normal fertile flowers
were obtained from all plants. These results indicate that flowering time can me
manipulated by transgenic means. Lisianthus transformations with genes acting upstream
to LFY are currently being made. The possibility of introducing a fragrance to lisianthus
flowers by genetic engineering is being investigated using genes from the scented plant
Clarkia breweri: Acetyl CoA benzyl alcohol acetyl transferase (BEAT) and S-linalool
synthase (LIS). These two genes code for key enzymes in the synthesis of the aroma
compounds benzyl acetate and S-linalool respectively. Aroma modification of the fruit of
transgenic tomato plants carrying a LIS construct has already been demonstrated. The
effect of BEAT and LIS transgenes on tobacco and lisianthus fragrance is under
The beautiful flowers, long stems and extended vase-life as well as the large
variety of available cultivars have made lisianthus [Eustoma grandiflorum (Raf.) Shinn.]
an increasingly attractive ornamental crop. The extended period required for lisianthus
flowering (Halevy and Kofranek, 1984) complicates its year-round production and any
decrease in flowering time would be beneficial. The possibility of manipulating flowering
time by genetic engineering appears like a promising prospect. Overexpression of the
meristem-identity gene LEAFY (LFY) from Arabidopsis has been shown to dramatically
accelerate flowering in transgenic distant species such as tobacco, aspen (Weigel and
Nilsson, 1995) and rice (Zuhua et al., 2000). This paper investigates the flowering
Proc. XX EUCARPIA Symp. on New Ornamentals
Eds. J. Van Huylenbroeck et al.
Acta Hort. 552, ISHS 2001
characteristics of progeny from lisianthus plants transformed with LFY.
An additional target for lisianthus improvement by genetic engineering was the
introduction of a fragrance to the otherwise unscented lisianthus petals. This approach is
now conceivable due to the recent identification and characterization of genes implicated
in fragrance production (Dudareva and Pichersky, 2000). Two such genes coding for the
enzymes benzylalcohol acetyltransferases (BEAT) and linalool synthase (LIS), both
originating from Clarkia breweri regulate the biosynthesis of aromatic compounds from
precursors usually present in plants (Dudareva et al., 1996; 1998a,b). Metabolic
engineering of terpenoid pathway for the production of aroma volatiles has been
demonstrated by overexpression of the LIS in tomato fruits, causing accumulation of
linalool (Lewinsohn et al., submitted).
The possibility of obtaining a fragrance from transgenic lisianthus flowers
carrying these genes was investigated here.
2. Materials and Methods
2.1. Gene constructs
Plasmid pDW146 carrying an Arabidopsis LFY cDNA under the control of the
35S CaMV promoter (Weigel and Nilsson, 1995) in the Agrobacterium strain ASE was
kindly provided by Dr. D. Weigel.
C. breweri BEAT cDNA was cloned into a pBI binary plasmid under the control of
the 35S CaMV promoter to form the binary plasmid pBI101 for transformations using
Agrobacterium strain EHA105. C. breweri LIS cDNA under the control of the 35S
CaMV promoter was cloned into the binary vector pCGN1559, which was introduced into
the Agrobacterium strain AGLO for transformations.
2.2. Transformations and regeneration
2.2.1. Lisianthus. Agrobacterium tumefaciens mediated transformations were
performed on leaf segments of in vitro grown lisianthus plants, cultivars Pure White and
Deep Blue. The segments were soaked in liquid MS containing the bacteria, submitted to
1.5 minute vacuum and placed on MS medium for 48h. Afterwards, they were transferred
to the shoot regeneration medium (MS supplemented with 2 mg/L BA and 0.5 mg/L
IAA) containing 40 mg/L kanamycin as a selection agent and 200 mg/L cefotaxime for
Agrobacterium elimination. Regenerated shoots were rooted on 1/2 MS medium without
antibiotics. Rooted plantlets were further tested for kanamycin resistance in MS medium
and resistant plantlets were transferred in pots to the greenhouse.
2.2.2 Tobacco. Transformation process was similar to that of lisianthus. Tobacco
shoot regeneration medium consisted of MS supplemented with 0.05 mg/L IAA and 0.5
mg/L zeatin, and rooting medium consisted of MS. 50 mg/L kanamycin as selective agent
and 200 mg/L cefotaxime were added to both media.
2.3. Transformants analyses
2.3.1. LFY - Specific primers (5’ - GTCCCCAAACCACTACCTCC - 3’; 5’ -
ACTCTCCGCCGCTGGTGATT - 3’) were designed within the non-shared region of the
Arabodopsis LFY cDNA for molecular identification of transformed plants by PCR.
These primers amplified a 300 bp of LFY cDNA.
2.3.2. BEAT - Specific primers (5’ - CGATGCACTCCAAGAAGTTAC - 3’;
GCAGTTGGTGGTGTTGTTGG - 3’) for BEAT were used for molecular identification
of transformed lisianthus and tobacco plants. BEAT enzymatic activity was performed on
crude extracts from leaves of transformed and control plants as previously described
(Dudareva et al., 1998a; Shalit et al., 2001). Each measurement was repeated two times.
2.4. Growth of the plants in the greenhouse.
Progeny of plants showing the presence of LFY from PCR was obtained by
selfing. T1 seeds were germinated in trays and grown for 3 months under moderate
temperatures (maximum night temperature was kept below 18 C) to avoid rosette
formation. Plantlets were planted individually in pots in a greenhouse under controlled
temperature conditions (27.5 C/15.4 C, day/night) on 15.1.2000. Flowering time was
recorded when the first flower bud was fully open.
3.1. Flowering time
Following transformation and shoot regeneration in selective medium, we detected
by PCR the presence of LFY cDNA in seven plantlets, originating from independent
transformation events. Figure 1 shows PCR amplification of the expected 300 bp
fragment in three regenerated plants. No fragment was amplified in non-transformed
plants from cultivars Pure White (Fig. 1) or Deep Blue (data not shown).
The transformation efficiency (number of PCR positive plants) was estimated at
about 13% of inoculated leaf pieces and 34% of regenerated shoots. After transplantation
to the greenhouse, some of these plants developed an increased number of shoots (up to
four) compared to a single shoot in the cultivars from which they were derived. All plants
were otherwise similar to the control and flowered at the same time, about four months
after transplantation to the greenhouse. All plants were fully fertile. After selfing, the
progeny of three selected plants was studied (Table 1). None of the T1 plants developed
multiple shoots. This feature was probably related to the transformation tissue culture
conditions and was restricted to the T0 plants. P4 and 17-2 progenies flowered in average
12 and 19 days earlier than the control cultivar Pure White, respectively. In both cases,
this difference was statistically significant (p<0.05). Plants derived from 17-2 also
showed a significant reduction in the other parameters, while in P4 progenies only the
number of buds/inflorescence was decreased compared to the control (Table 1). In
contrast, T12 progenies did not flower before ‘Deep blue’ plants, but had less
buds/inflorescence and less inflorescences/plant (Table 1).
The presence of BEAT cDNA was so far detected by PCR in 18 lisianthus
regenerated plantlets. Figure 2 shows PCR analysis of two lisianthus plants transformed
with BEAT cDNA, exhibiting the expected 1.2 Kb fragment. No amplification of this
fragment was observed in any of the non-transformed plants (data not shown). All the
PCR positive plants also exhibited significantly higher BEAT enzymatic activity than
non-transformed control plants (Figure 3). High BEAT activity in the leafs of regenerated
lisianthus plants not only demonstrates the stable character of the transformation but also
indicates that the activity of this enzyme is not inhibited in lisianthus tissue. The
regenerated lisianthus plants are currently grown in the greenhouse where they develop
normally. Up to now, no fragrance was detected from the leaves of transformed plants.
The calculated rates of transformation efficiency in this experiment amounted in 18%
plants testing positive for BEAT activity from initial number of leaf pieces, representing
also 50% of total regenerated shoots.
In parallel, tobacco leaves were transformed by Agrobacterium carrying the BEAT
plasmid. Although a large number of regenerated plants were PCR positive, none of them
showed increased BEAT activity. This may suggest that an inhibiting factor of the
enzyme BEAT is present in tobacco leaves. No fragrance was detected from leaves or
petals of tobacco regenerated plants.
The efficiency of lisianthus transformation obtained in this study - up to 17% -
was significantly higher than the 1-3% usually reported in the literature (Ledger et al.,
1997). This should greatly promote the possibility of improving lisianthus by genetic
A significant decrease in fowering time was recorded in progenies of lisianthus
plants transformed with the Arabidopsis LFY cDNA. Further experiments are underway
to investigate the performance of these plants under different day length conditions.
Additional transformations are currently performed in lisianthus with Arabidopsis floral
genes acting upstream of LFY (Simpson et al., 1999) which may have a more substantial
effect on flowering time of transgenic lisianthus plants.
Although no scent was detected in the leaves of small transgenic lisianthus plants
showing BEAT activity, it may be possible that some fragrance will be obtained from the
petals and stigmata due to their unique morphological structure. BEAT expression at the
RNA level as well as analyses of aromatic compounds in transgenic lisianthus are
currently performed and will help elucidate the mode of action of the transgene. Further
transformations are planned with genes involved in fragrance production, such as LIS,
which efficiency has already been demonstrated in tomato (Lewinsohn et al., submitted)
aiming to introduce a scent to the petals of lisianthus and other ornamental plants.
The authors thank Dr. D. Weigel for providing LFY cDNA as well as many helpful
comments. This work was partly supported by a funding from the Israeli Chief Scientist,
Ministry of Agriculture.
Dudareva N., Cseke L., Blanc V.M., and Pichersky E., 1996. Evolution of floral scent in
Clarkia: novel patterns of S-linalool synthase gene expression in the C. breweri flower.
Plant Cell 8: 1137-1148.
Dudareva N., and Pichersky E., 2000. Biochemical and molecular genetic aspects of floral
scents. Plant Physiol 122: 627-633.
Dudareva N., D’Auria J.C., Hee Nam K., Raguso R.A., and Pichersky E., 1998a. Acetyl-
CoA:benzylalcohol acetyltransferase – an enzyme involved in floral scent production
in Clarkia breweri. The Plant Journal 14, 297-304.
Dudareva N., R.A. Raguso, J. Wang, J.R. Ross, and Pichersky E., 1998b. Floral scent
production in Clarkia breweri. III Enzymatic synthesis and emission of benzenoid
esters. Plant Physiol. 116: 599-604.
Halevy, A.H., and Kofranek A.M., 1984. Evaluation of lisianthus as a new flower crop.
Ledger S.E., Deroles S.C., Manson D.G., Bradley J.M., and Given
N.K.,1997.Transformation of lisianthus (Eustoma grandiflorum). Plant Cell Reports 16
Lewinsohn E., Tadmor Y., Schalechet F., Amar O., Lastochlkin E., Larkov O., Ravid U.,
Gepstein S., and Pichersky E., (submitted). Enhanced S-linalool levels in metabolically
engineered tomato fruits.
Shalit M., Katzir N., Tadmor Y., Larkov O., Burger Y., Shalechet F., Lastochkin E.,
Ravid U., Amar O., Edelstein M., Karchi Z., and Lewinsohn E., 2001. Acetyl CoA:
alcohol acetyl transferase activity and aroma formation in ripening melon fruits. J.
Agric. Food Chem. 49:794-799.
Simpson G.G., Gendall A.R., and Dean C., 1999. When to switch to flowering. Annu.
Rev. Cell Dev. Biol. 99:519-550.
Weigel D., and Nilsson O., 1995. A developmental switch sufficient for flower initiation
in diverse plants. Nature 337: 495-500.
Zuhua H., Zhu Q., Dabi T., Weigel D., and Lamb C., 2000. Transformation of rice with
the Arabidopsis floral regulator LEAFY causes early heading. Transgenic Research 9:
Table 1. Growth and flowering parameters of T1 progenies obtained from T0 plants
transformed with LFY cDNA: P4 and 17-2 derived from cv. Pure White; T12 derived
from cv. ‘Deep Blue’. Results are the means (± SE) of at least 20 plants.
Height # nodes Days # buds/ # of
(cm) from base from inflores- inflores-
to first planting cence cences/plant
‘Pure White’ (control) 40.0±2.1 8.3±0.4 100.3±3.3 9.0±1.0 2.9±0.3
Progeny from P4 39.5±1.3 8.1±0.2 88.6±0.6 7.16±0.4 2.3±0.1
Progeny from 17-2 26.6±2.4 6.4±0.7 81.1±3.4 4.3±1.1 1.6±0.4
‘Deep Blue’ (control) 44.9±1.9 8.1±0.3 84.5±1.2 9.2±0.9 2.3±0.2
Progeny from T12 43.6±1.5 8.0±0.2 89.3±1.5 5.3±0.4 1.7±0.1
Pure P4 17-2 T12 LFY
Figure 1. PCR analysis of three lisianthus plants transformed with LFY cDNA (P4, 17-2
and T12) and the control cultivar Pure White. LFY: Arabidopsis LFY cDNA.
Control BEAT F3 F7
Figure 2. PCR analysis of two lisianthus plants transformed with BEAT cDNA (F3 and
F7) and a non transformed plant (control). BEAT: C. breweri BEAT cDNA.
F3 F7 Control 1 Control 2
Figure 3. Enzymatic activity of BEAT (expressed as cpm/mg protein) in crude extracts
from two lisianthus plants transformed with BEAT cDNA and two non-transformed
control plants. Bars represent SE for each measurement.