IMPROVING FLORAL SCENT PRODUCTION IN FLOWERS Progress Reports - June 2001
Improving Floral Scent Production In Flowers
Natalia Dudareva,
Department of Horticulture and Landscape Architecture,
Purdue University, West Lafayette, IN 47907
INDUSTRY NEEDS AND PROJECT OBJECTIVES:
One of the major problems in floriculture industry is the lack of scent in
ornamentals, which include cut flowers, foliage and potted plants. Traditional
breeding of ornamental crops has produced cultivars with improved vase life,
shipping characteristics, and visual esthetic values (i.e., color, shape),
however it has sacrificed other important commercial trait, such as floral
scent. Flowers with new introduced or modified fragrance are in high demand by
the consumers, who are continually searching for new products.
Genetic engineering offers an exciting future for development of novel
cultivars with improved scent quality and will overcome this genetic regression.
New crops with modified composition of volatiles and newly introduced aroma
would benefit US agriculture by increasing crop productivity and the value of
ornamentals. Objective of our studies is to improve scent quality of cut flowers
by manipulation of the output of volatile compounds using recombinant DNA
technologies. To achieve this objective, we are investigating the molecular
changes that affect the level of scent emission in plants and isolating BAMT
promoter, which can be potentially used to produce transgenic cut flowers with
novel scents.
SUMMARY OF WORK COMPLETED:
Objective 1. Determine what is missing in Antirrhznum varieties
that do not emit methvlbenzoate.
Out of 50 different snapdragon cultivars analyzed, only three Sonnet White,
Potomac White and Potomac Pink do not emit methylbenzoate or emit it at very low
levels. Methylbenzoate is produced in upper and lower petal lobes of snapdragon
flowers by the action of the biosynthetic enzyme Sadenosyl-L-methionine: benzoic
acid carboxyl methyltransferase (BAMT). Analysis of BAMT gene expression in
Sonnet White, Potomac White and Potomac Pink revealed a high level of BAMT mRNA
in Potomac Pink, a low level of BAMT mRNA in Potomac White and no expression of
the BAMT gene in Sonnet White flowers.
To determine why Potomac Pink and Potomac White flowers emit methylbenzoate
at very low levels, although they contain and express the BAMT gene (despite
very low expression in Potomac White), a RT-PCR approach was used to isolate the
BAMT cDNA clones from both cultivars. cDNAs were sequenced to determine possible
changes in the coding region of the gene. Sequence analysis revealed 10 changes
in the amino acid sequence of the BAMT protein from Potomac Pink flowers and
several changes in the nucleotide sequence of the coding region that do not
result in a change in amino acid. No changes were found in the BAMT protein from
Potomac White.
To find out if these 10 changes in the amino acid sequence of Potomac Pink
BAMT (PP BAMT) affect enzymatic activity of the protein, the PP BAMT eDNA was
subcloned into an expression vector pET-28. E. coli BL21 cells were transformed
with the recombinant plasmid and the expression of PP BAMT cDNA was induced by
the addition of 0.4 mM IPTG at of 0.5 with 20 h incubation at 20¬?C. E. coli
cells expressing the PP BAMT gene produced large amounts of a 49 kDa protein,
but the protein does not have any enzymatic activity. These data indicate that
although the BAMT gene is transcribed in Potomac Pink flowers the resulting
enzyme is inactive. It suggests that changes that occurred in the gene at the
level of nucleic acid are responsible for the absence of normal emission of
methylbenzoate.
Western blot analysis using rabbit polyclonal anti-BAMT antibodies was used
to determine if this inactive BAMT protein is still being made in petal tissue
of Potomac Pink flowers and if a detectable level of BAMT protein is present in
Potomac White. Western blot analysis showed that the polyclonal antibodies
prepared against the active BAMT protein recognized PP BAMT (Fig. 1, lane 1 and
2). We found that the mutated BAMT protein is not produced in Potomac Pink
flowers (Fig. 1, lane 3 and 4), but we did find a low level of BAMT protein in
Potomac White flowers.
Fig. 1. Expression of the BAMT protein in Potomac Pink flowers as determined
by Western blot analysis. 1 and 2 ¬ó PP BAMT protein expressed in E. coli, 3
and 4 ¬ó proteins extracted from upper and lower petal lobes of Potomac Pink
flowers, 5 - BAMT protein expressed in E. coli, 6 -
proteins extracted from upper and lower petal lobes of Maryland True Pink
snapdragon flower.
It is possible that there are two alleles of the BAMT gene in Potomac Pink
plants and the normal active gene is still expressed in cells at a very low
level undetectable by Northern blot analysis and the amount of the BAMT protein
is undetectable by Western blot analysis. We are checking this hypothesis now. A
Potomac Pink plant which is an Fl hybrid was self-fertilized, and its seeds were
grown to produce an F2 population. We have now 28 individual plants from the F2
population.
The low emission of methylbenzoate in Potomac Pink plants could be due to
another carboxyl methyltransferase, for a example S-adenosyl-L-rnethionine:salicylic
acid carboxyl methytransferase (SAJvIT), which might use benzoic acid as a
substrate and produce methylbenzoate. SAMT was recently isolated from Clarkia
breweri flowers and it was shown that, although it is highly specific for
salicylic acid, it does methylate benzoic acid with relatively high efficiency
(Ross et a!., 1999). To figure out whether the SAMT gene is involved in
inethylbenzoate production in Potomac Pink flowers, first we have isolated the
SAMT gene from snapdragon flowers by using a functional genomics approach. An
ongoing snapdragon EST (expressed sequence tag) project in the P1’s laboratory
(supported by Purdue University) resulted in random sequencing of 800 clones
from a petal-specific library constructed from mRNA isolated from petals (upper
and lower lobes) of 1 to 5 day-old snapdragon flowers, a tissue highly
specialized for floral scent biosynthesis (Dudareva et al., 2000) and two clones
with sequence similarity to SAMT from C. breweri were obtained.
Objective 2
. Characterize temporal and spatial expression of BAMT
promoter using a reporter gene.
In order to create transgenic plants with modified scent
quality we need
to have gene promoters that will drive the expression of the transgenes in a
“scent specific” manner (in proper tissue (petals) and at the proper
stage of development). Thus, we proposed to isolate the BAMT promoter from
snapdragon, which can be potentially used to produce transgenic cut flowers with
novel scents. This part of the project was successfully accomplished. A genomic
clone corresponding to the BAMT gene was isolated from a snapdragon genomic DNA
library constructed from a scented variety using
the BAMT cDNA clone as a probe. The sequence analysis revealed that the gene
contains 2045 bp of the 5’ flanking region and the entire coding
region. To locate the transcription start site of the BAMT gene, primer
extension analysis was performed. Two major transcripts were found starting 47
and 40 nucleotides upstream of the first ATG codon. Examination of the 5’
flanking region revealed a putative TATA box and CAAT box located 76 and 101
nucleotides upstream of the ATG codon, respectively.
As a part of the characterization of the BAMT promoter, cell-specific
expression and also subcellular localization of BAMT was completed by light and
electron microscopic immunolocalization study at cellular and subcellular levels
using antibodies against the BAMT protein. BAMT was immunolocalized
predominantly in the conical cells of the inner epidermal layer and, to a much
lesser extent, in the cells of the outer epidermis of snapdragon flower petal
lobes. These results strongly suggest that scent biosynthetic genes are
expressed almost exclusively in the epidermal cells of floral organs. Immunogold
labeling studies reveal that BAMT is a cytosolic enzyme, suggesting cytosolic
location of methylbenzoate biosynthesis.
To determine whether the 5’ flanking sequence of the BAMT gene is
capable of driving BAMT expression in other plant species such as petunia and
tobacco, the putative full-length promoter region was amplified by PCR and
cloned into the HindIII-BamHI site of the binary vector pBI101. The pBI101
plasmid contains a prornoterless ~3-glucuronidase (GUS) coding sequence fused to
a nopaline synthase gene at 3’ end. Subcloning of the BAMT promoter region
into pBI101 was done in such way that the ATG of the BAMT gene were in frame
with the GUS coding region (translational fusion). This construct was
transferred to Agrobacteriuin tumefaciens (LBA4404) by the triparental
mating procedure. Exconjugants were used to transform Petunia Mitchell and
tobacco leaf discs. Petunia and tobacco plants transformed with a promoterless
GUS gene or a GUS gene fused to the CaMV 35S promoter were used as controls.
To define the cis-acting elements responsible for the temporal and spatial
expression of BAMT gene a series of four 5’ deletions derived from the
full-length BAMT promoter were generated by PCR and translationally fused to the
GUS-NOS reporter gene. These chirneric genes were transformed into Petunia
Mitchell and Nicotiana tabacurn. The GUS gene under control of the
CaMV 35 S promoter was used as a control. Deletions that fail to express GUS in
a BAMT-specific manner will identify sequences responsible for regulation of the
biosynthesis of methylbenzoate in a flower.
OBJECTIVES FOR THE COMING YEAR:
1. We will continue to evaluate what is missing in Antirrhinum varieties
that do not emit methylbenzoate. This will provide needed information about
the molecular changes that affect the level of scent emission in plants.
Scent production will be analyzed in 28 individual Potomac Pink plants from
the F2 population as well as BAMT activity, BAMT mRNA expression and amount of
BAMT protein. Also RT-PCR will be used to isolate the BAMT cDNA clones from
individual plants. cDNAs will be sequenced to determine the type of BAMT cDNA
(mutated or normal) present. If we find both types of BAMT eDNA, quantitative
RT-PCR with type-specific primers will be used to find the ratio between normal
and mutated forms.
We will also check for the presence of inhibitor(s) or possible enhancer(s)
in petal tissue of Potomac Pink plants. In these experiments pure active BAMT
protein will be added to crude extracts from petal tissue of individual plants
and assayed for BAMT activity.
To investigate genetic inheritance of methylbenzoate emission, scent
production in the individual plants obtained in crosses between Potomac Pink and
Potomac White cultivars will be analyzed.
To find out whether the SAMT gene is involved in methylbenzoate production in
Potomac Pink flowers, first we will demonstrate the enzymatic activity of this
protein. The full-length SAMT clones will be obtained by a 5-rapid amplification
of eDNA ends method (5-RACE) or by screening a cDNA library with SAMT probe(s).
For functional expression, the coding region will then be amplified by PCR,
subcloned into an expression vector, and, for convenient purification, expressed
as a 6-His fusion protein in E. coli. The recombinant protein will then
be purified and assayed for potential enzymatic functions. Assays will be
performed under standard physiological conditions. When the function of the gene
is proven, a PCR-based strategy using oligonucleotides specific for the SAMT
gene will be used to figure out whether the SAMT gene is involved in
methylbenzoate production in Potomac Pink flowers.
2. We will continue to characterize temporal and spatial expression of
BAMT promoter in petunia and tobacco plants using a reporter gene.
We are continuing to transform/regenerate plants for the BAMT-GUS. BAMT with
deletionsGUS. promoterless GUS, and 35S-GLS containing leaf segments. After
shoot and root induction on kanamycin-containing media, plants will be put in
the soil and kept in a greenhouse. The presence of the promoter-GUS transgene in
individual transformed plants will be confirmed by PCR using appropriate
primers. l)NA blot analysis will be performed on individual transgenic plants to
determine if they represent independent transformation events and the copy
number of the introduced chimeric gene. The pattern of expression directed by
BAMT promoter will be monitored by histochemical and fluorometric analysis of
~3-glucuronidase.
These studies will provide important information about BAIMT promoter
including:
a. organ and tissue specific expression of BAMT promoter in the
heterologous system;b. regulatory regions, which control the spatial and temporal distribution
of expression;c. positive enhancer element(s), if any; and
d. relative promoter transcriptional strength when compared with CaMV 35S
promoter.
3. Genetically engineer floral scent production in non-scented plants via
gene transfer.
Once BAMT promoter is characterized it will be used to generate transgenic
tobacco plants with modified scent quality. BAMT gene responsible for
rnethylbenzoate biosynthesis and SAMT responsible for methylsalicylate
biosynthesis will be our first and second choices to achieve a new scent in
plants.
PROFESSIONAL/PUBLISHED INFORMATION
Results obtained in this project were presented at the invited seminars at:
1. Ball Seed and Ball Helix, January 16, 2001
2. Western Michigan University, March 16, 2001
3. Midwest ASPB Meeting, Knox College, March 23-24, 2001
4. Institute for Applied Research, Beer-Sheva, Israel, May 7, 2001
5. Hebrew University of Jerusalem, Rehovot, Israel, May 8, 2001
6. Plant Research international at Wageningen, Holland, May 14, 2001
