Download −Table of Contents

Flavonoid identification and genetic analysis using PCR to analyze variegation in Dixie's
Winedot
Introduction
Flavonoids are a large and diverse class of molecules that include many important
pigment molecules en developed through classic breeding, but with modern genetics
scientists are starting to gain a deeper understanding of what contributes to flower
coloration and how it can be manipulated. This study sought to explore the mechanism
behind the variegated flower coloration of the dahlia variety Dixie’s Winedot.
Dixie’s Winedot flowers have a pale yellow background color intermixed with dark
streaks of purple, hence the classification as a variegated variety (figure 1). There are
several other variegated varieties of dahlia, which typically follow the pattern of having a
pale background color with darkly pigment streaks. The streaks can range in size from
just a thin strip to large sectors or even the entire flower. The streaks are always directed
radially from the center of the flower. The pale background color and the streak
distribution suggest that the variegation is the result of transposon activity (Chawla, 304).
Figure 1: Dixie's Winedot
The variegated pattern is likely due to the activity of a transposon.
Model
In the transposon model for variegation, the plant is homozygous for non-functional
recessive alleles for one step in the pigment pathway. One (or more) copies of the
nonfunctional alleles contain a transposon, which is disrupting the functionality of the
gene. This transposon is capable of spontaneously excising from the gene. If the
transposon excises correctly, the functionality of the allele is returned and the pigment
pathway is restored. The restored gene is passed on to subsequent daughter cells. The
daughter cells are typically arranged in a line radially from the center of the flower, due
to anticlinal cellular division in the developing petal tissue. This creates a streak of cells
that have the functional gene and hence have a darker pigmented color. If the transposon
excises very early in flower development, then a larger sector will likely develop since
the gene will be passed on to more daughter cells. Smaller sectors and streaks are caused
by later excision of the transposon. Previous studies of similar variegations in other
species have found this model to be accurate (Habu et. al.).
It was hypothesized that a single gene in the anthocyanin biosynthetic pathway was
disrupted by a transposon in the wild type Dixie’s Winedot. This experiment set out to
confirm that transposon activity was responsible for the variegated phenotype in this
dahlia variety. A full revertant plant, one with entirely purple flowers, was used as a
comparison to the variegated wild type.
Figure 2: The anthocyanin biosynthetic pathway.
Experimental Design
The two different parts of this experiment set out to characterize the variegation
mechanism and confirm this model. The first part of the experiment was to identify the
pigment that was accumulating in the recessive cells. It was assumed that the flavonoid
product accumulating would be the one before the step requiring the nonfunctional gene.
By identifying the accumulated pigment and knowing the biosynthetic pathway (figure 2),
the nonfunctional gene could be identified. Due to the pale yellow background color, it
was hypothesized that the accumulating pigment was a chalcone. To determine if this was
indeed the case, TLC was used to separate pigments and UV spectrophotometry was
performed The second part of the experiment was to use PCR and gel electrophoresis to
see if the inserted transposon could be amplified and visualized. To do this, primers were
designed for Chalcone Isomerase, (CHI). It was expected that if the transposon was
present in the gene its large size or possible secondary structure caused by inverted
repeats could prevent amplification. Therefore, it was hypothesized that there would be
little PCR product for the wild type compared to the revertant. Alternatively, a larger
PCR product for the wild type than for the revertant would also indicate that the
transposon was present.
Materials and Methods
Flavonoid Analysis
The first part of the project was to use chemical means to identify the flavonoid that was
accumulating in the wild type Dixie’s Winedot. To do this, a few petals were ground and
extracted with 2 mL of acidified methanol from both the wild type and the revertant. Both
of these extracts looked similar. To facilitate pigment identification, Thin-Layer
Chromatography (TLC) was performed using microcrystalline cellulose plates and an
acetic acid, HCl, and water solution. The TLC took approximately 45 minutes to run.
Bands of interest where scraped off the TLC plate and the pigment was resuspended in
acidic methanol. The absorbance spectrum of the extracts was taken between 250 and 600
nm. Standards were prepared and compared to the unknowns using TLC and uv-vis
spectrophotometry.
Genetic Analysis
The first step of the genetic analysis was to design primers for the dahlia CHI gene. This
gene has not been sequenced. However the CHI mRNA from several related species have
been sequenced. The sequences used were all from the same family as Dahlia, Asteraceae.
Six sequences were used: one from sunflower, two from chrysanthemum, one from
Callistephus chinensis, and one from Saussurea medusa. An alignment was performed on
these sequences using ClustalX2. The PriFi tool was used in conjunction with the
sequence alignment to find suitable primers for CHI. Primers were selected in highly
conserved regions of the gene and to cover the majority of the gene. Before PCR could be
performed the genomic DNA from the dahlias had to be extracted and purifed. To do this
a CTAB extraction was performed. The particular CTAB extraction protocol used was
one from a study published in 2005 (Timmerman Vaughan et. al.). This protocol was
found to be effective at successfully extracting genomic DNA in a wide range of plant
taxa. The protocol was followed using 3g of petal material from the wild type and the
revertant flowers. The resultant DNA suspension was analyzed for quality and quantity
using gel electrophoresis. The identified primers were ordered and used to run PCR. A
high performance Taq polymerase was used to allow for amplification even if the
transposon was 5 kb in size. Because the optimal annealing temperatures for the primers
were unknown, 12 PCR reactions were run for each primer pair across a gradient from 50
to 60 C. Higher temperature allowed for more specific amplification while lower
temperature ensured strong annealing. The PCR products were ran on a 1.2% agarose gel
at 100 V for 45 minutes. 5uL of PCR product was used with 5 uL of loading dye to load
the gel. Poor resolution resulted, so the gel eletrophoresis was repeated with and E-Gel.
For the E-Gel, 5uL was used for the primer pair two products. For primer pair one, both 1
and 2uL were used. Water was added to bring the volumes up to 20 uL. A 1kb plus
standard was used as a reference.
Results
Flavonoid Analysis
A clean separation was not achieved using TLC (figure 3), however differences between
the wild type and revertant extracts were clearly visible. Two bands clearly stood out as
different between the two samples. Band 1 was yellow in color and much more prevalent
in the wild type. Band 2 was pink/purple in color and much more prevalent in the
revertant. The uv-vis spectrums for each of the bands for the two samples are shown in
figure 4. The major peaks for band one were located near 270 and 362 nm. The unknown
isolated from the wild type was more concentrated that that from the revertant for band
one. For band 2, a peak near 525 was observed, indicating an anthocyanin. The band two
unknown was more concentrated in the revertant (figure 4). The band one spectrum was
compared to that several flavonols (figure 6).
Figure 3: Thin-Layer Chromatography
A solution of acetic acid, HCl and water was used to separate the WT and revertant
extracts on a microcrystaline cellulose plate.
Figure 4. UV-Vis spec for Wild Type and Revertant Pigments.
The spectrum for each of the two bands isolated from the samples are shown above. Band
1 was much more concentrated in the WT whereas band two was more concentrated in
the revertant. The band 2 spectrum, with a peak around 525, is characteristic of an
anthocyanin.
Genetic Analysis
Good similarity was found between the CHI sequences used from the Asteraceae family.
This allowed for the design of primers for the dahlia gene (figure 9). Because the
sequences did not match up perfectly there was some degeneracy in the primers. The
primers for primer pair 1 were Fw 5'- CCTMGGWGGCGCMGGTGTGAGAGGTATGG
and Rv 5'-TTGTRAAGAGAATAGAKGMGCCTGG. For primer pair two they were Fw
5'-TAACNGGTAAGCAATACKCTGA and Rv 5'-AGTCTTGAWGCCAAACTTTG
TTT. The degeneracy is indicated by the non ATCG base pairs listed. The melting
temperatures predicted for these four primers were 71 C, 58 C, 60 C, and 61 C
respectively. Agarose gel electrophoresis was run to determine the quality and quantity of
the isolated genomic DNA. The DNA was shown to be of good quality and to be of the
concentration of approximately 100 ng / uL. The gel of all the PCR products is shown in
figure 11. The results of primer pair one are difficult to interpret because the image is
saturated. For primer pair two, two discrete fragments are visible at higher temperatures,
while more bands are visible at lower annealing temperatures. Both of the ladders did not
run run very clearly and are difficult to read. A second gel was run using a subset of the
PCR products. This gel provided better resolution than the first and allowed for better
size estimation of the fragments (figure 12). The two fragments from primer pair 2 appear
to be about 1.7 and 2.1 kb. Numerous fragments are visible for primer pair one, ranging
from 350 bp to 1.6 kb.
Figure 9: Primer Selection and Gene Structure
Six known cDNA sequences for CHI from the asteraceae family were used to engineer
primers using multisequence primer selection tool PriFi. Two primer pairs were selected.
The known introns from the petunia CHI-B gene are shown.
Figure 11: Gel of PCR Products
The PCR products were visualized using gel electrophoresis. A temperature gradient
from 50 to 60 C was used for the annealing temperature during the PCR.
Figure 12: E-Gel of Selected PCR Products
An E-Gel was run to obtain better resolution of the PCR products. The primer one
products from WT and revertant were compared from high and low temperature in lanes
2-9. 1 ul and 2 ul of PCR product were ran in neighboring lanes for primer pair one.
Conclusions
Flavonoid Analysis
An identification of the unknown band 1 area was attempted using the uv-vis spectrum
obtained. The unknown spectrum did not match up well with that of chalcone because it
contained a peak around 270nm that chalcone lacks. The peaks of the unknown at 270nm
and 362nm suggest that the unknown is a flavonol (figure 5). There are several flavonols
that are part of the anthocyanin pathway. The unknown was compared to the spectrums
of three: kaempferol, resokaempferol, and kaemferide. The spectrum of the compound
isolated from band 1 appears to be similar to that of kaempherol or some other flavonol
(figure 6), with major peaks near 360 and 270 nm. However, the unknown spectrum lacks
the characteristic “shoulder” peak around 325 nm that the flavonols exhibit. Flavonols
such as quercetin, kaempherol, and myricetin are side branches to the anthocyanin
pathway. From the model proposed, these are not the products expected to be
accumulating. It is possible that these molecules are more stable that their respective
flavanone precursors due to resonance effects. This would favor the accumulation of such
flavonols if the pathway was blocked at the DFR gene (figure 2). However, studies have
shown that this is generally not the case. A study looking at Petunia hybrida mutants
lacking DFR and F3’H activity accumulated dihydrokaempherol, not kaempherol (Linn,
et. al.). While some flavonols may be present in the unknown from band 1, the main
pigment is probably not a flavonol. Other possibilities for the unknown band were
explored. From the pathway shown in Figure 1 and the hypothesis that CHI was the
defective gene, the accumulating product was expected to be chalcone. However, the
unknown spectrum does not match up well with that of chalcone. It is possible that the
accumulation of chalcone led to the synthesis of some other flavonoid. Defective CHI has
been shown in the previous studies to lead to the accumulation of isosalipurposide as well
as chalcone (Figure 7, Reuber et. al.). The broad peak of the unknown matches up well
with that of isosalipurposide (figure 8). Also, the chalcone is known undergo a slow
isomerization in vivo to the racemic flavanone. One of these enatiomers is narigenin,
which can continue the anthocyanin pathway. This suggests that a defective CHI may
lead to the accumulation of isosalipurposide, chalcone, and small amounts of the rest of
the pigments in the pathway, including the flavanones and flavanols. It is believed that
the unknown spectrum is resultant of such a mixture, however more work remains to
confirm this hypothesis. HPLC analysis of the isolated unknown from the band 1 area
would likely yield more information about the number of pigments present (Reuber et. al).
Furthermore, these pigments could be collected and analyzed by UV-VIS spectroscopy
individually. By viewing the pigments in isolation a more definitive identification could
be made.
Figure 5: Flavonoid Spectra
Here are the UV-VIS absorption spectra of the major classes of flavonoids. The unknown
matched up most closely with the flavonols.
Figure 7. Side Reactions for Chalcone.
Chalcone can be converted into isosalipurposide by CGT, or can spontaneously
isomerizes into the racemic flavanone. The S form of this flavanone can continue along
the pathway, bypassing the need for CHI.
Figure 6. Comparison with Known Flavonols.
The unknown spectrum was compared with that of several known flavonols. The
unknown had similar peak positioning as the flavonols. However the distinctive shoulder
of the flavonols around 325 nm was missing in the unknown.
Genetic Analysis
Primer pair one seemed to result in significant non specific amplifaction. A higher
annealing temperature may improve the specificity, particularly since the melting
temperature for the forward primer was predicted to be 71C. However, if the petunia
intron locations are representative of those in dahlia, primer pair one is likely to be
ineffective. The intron positions and sizes for Petunia hybrida CHI-B are shown in figure
9. The CHI protein sequences for petunia and asteraceae were shown to be similar, so
such a comparison is valid (figure 10). The intron predicted in the forward primer of
primer pair one would prevent annealing and successful amplification of the desired
sequence. Primer pair two was more successful at achieving specific amplification. It is
likely that these fragments are the result of the amplification of the desired gene sequence.
However, the fragments were sized at about 2.1 and 1.7 kb (figure 12). This is
significantly larger than the expected band of 0.5 kb. This could be explained by the
dahlia gene having larger introns than the petunia one. Primer pair two resulted in the
amplification of two fragment lengths. There are several possible explanations for this.
The first is that dahlia could have two different CHI genes of different sizes. Petunia is
known to have two different CHI genes, CHI-A and CHI-B. Amplification of these
different genes could explain why two different fragment lengths were produced. A
second explanation is that there are two different alleles of the same gene present that
have different sizes. If this explanation is correct, the different sizes of these alleles could
be explained by one of the alleles having a 0.4 kb transposon inserted into it. Because the
wild type flower has streaks on purple, some of this allele should be present in the WT.
Even though the revertant flower is entirely purple, it is likely that one or more copies of
the transposon allele remain and are masked by the functional revertant allele. The results
of the PCR with primer pair 2 are consistent with this hypothesis but they do not rule out
other explanations. Future work could be conducted to provide more evidence to either
support or discredit the hypothesis proposed. The PCR products from primer pair two
could be sequenced. If the two sequences were identical except for a transposon-like
insert this would be strong evidence that the hypothesis was correct. Alternatively, the
PCR with primer pair two could be repeated on other dahlia varieties without variegation
to see if the two band result is specific to Dixie’s Winedot. Even if the was not a
transposon present in the area of the gene covered by primer pair 2, there could be a
transposon located somewhere else on the gene.
Appendix
References
Chawla H S. Introduction to Plant Biotechnology. p 304. Science Publishers. 2000.
Froemel S, de Vlaming P, Stotz G, Wiering H, Forkmann G, Schram A W. Genetic and
biochemical studies on the conversion of flavanones to dihydroflavonols in flowers of
Petunia hybrida. 1985.
Habu Y, Hisatomi Y, Lida S. Molecular Characterization of the mutable flaked allele for
flower variegation in the common morning glory. The Plant Journal. 1998.
Linn F, Heidmann I, Saedler H, Meyer P. Epigenetic changes in the expression of the
maize A1 gene in Petunia hybrida: Role of numbers of integrated gene copies and state of
methylation. April 1990.
Mabry T J, Markham K R, Thomas M B. The Systematic Identification of Flavonoids.
1970.
Markham K R. Techniques of Flavonoid Identification. 1982.
Ogata J, Sakamoto T, Yamaguchi M, Kawanobu S. Yoshitama K. Isolation and
characterization of anthocyanin 5-O-glucosyltransferase from flowers of Dahlia
variabilis . Journal of Plant Physiology. November 2004
Reuber S, Jende-Strid B, Wray V, Weissenbock G. Accumulation of chalcone
isosalipurposide in primary leaves of barley flavonoid mutants indicates a defective
chalcone isomerase. Physiologia Plantarum. 1997.
Timmerman-vaughan Gail M, Pither-Joyce Meeghan, Clover Gerard. A positive-control
PCR assay that is based on chloroplast ndhB sequences and applicable to nucleis acid
extractions from diverse plant taxa. Plant Molecular Biology Reporter. March 2005.