全 文 ：Received 21 Aug. 2003 Accepted 18 Oct. 2003
Supported by the National Natural Science Foundation of China (30170747) and the Hi-Tech Research and Development (863) Program
of China (2001AA212161).
* Author for correspondence. Tel: +86 (0)10 62759652; E-mail: .
http://www.chineseplantscience.com
Acta Botanica Sinica
植 物 学 报 2004, 46 (8): 948-954
Effect of Matrix Attachment Regions on Resveratrol Production in Tobacco
with Transgene of Stilbene Synthase from Parthenocissus henryana
ZHONG Jin, LIU Shu-Jun, MA Si-Song, YANG Wei, HU Yuan-Lei, WU Qi, LIN Zhong-Ping*
(State Key Laboratory of Protein Engineering and Plant Genetic Engineering, Peking University, Beijing 100871, China)
Abstract: Matrix attachment regions (MARs) were used to increase resveratrol production in stilbene
synthase (STS) gene-transformed tobacco (Nicotiana tabacum L.) in this study. MARs are DNA sequences
that bind to the cell’s pertinacious nuclear matrix to form DNA loop domains. A range of effects of MARs
sequence on mean expression level and variation in expression of transgenes has been reported usually
using b-glucuronidase (GUS) gene as reporter gene. The present study investigated the effects of MARs
sequence from yeast on transgene expression of STS for the first time. The results of in vitro binding
assay showed that the MARs sequence could specifically bind to the matrix of tobacco prepared by lihium
diiodosalicylate (LIS) procedure. The stilbene synthase is the key enzyme in metabolic pathway of resveratrol
biosynthesis and a cDNA for STS was cloned from Parthenocissus henryana (Hemsl.) Diels et Gilg by RT-
PCR method with primers designed according to the conserved sequence of STS gene in grapevine. The
cloned sequence showed high nucleotide identity (93.8%) with the STS gene of Vitis vinifera cv. Optima,
and the predicted protein sequence also had Cys164 activity core and IPNSAGAIAGN motifs which were
specific for stilbene synthases. With or without flanking MARs from yeast, the STS gene under control of
CaMV35S promoter with Ω enhancer has been transferred into tobacco. Northern blot and HPLC analysis
of the leaf extracts of transgenic tobacco showed the resveratrol had been produced in STS gene-
transformed plants, while there was no resveratrol found in the untransformed tobacco. In stably trans-
formed strains, the content of resveratrol in the plants with STS gene flanking by MARs averaged about 2-
fold higher than those lacking MARs. The results proved that flanking STS by MARs was one of the
effective way to increase resveratrol production in transgenic plants, and this work could also play basic
roles on the STS gene transformed vegetables or fruits with strong resistance to disease and benefits to
human health.
Key words: matrix attachment regions (MARs); stilbene synthase gene (STS); resveratrol;
Parthenocissus henryana ; transgenic tobacco; yeast; HPLC
Matrix attachment regions (MARs) are regions of DNA
biochemically defined by their ability to bind the nuclear
matrix, a proteinaceous complex left after removal of his-
tones and most of the DNA from nuclei. MARs are gener-
ally A/T-rich DNA sequences that are several hundred base
pairs long and are localized in the noncoding regions of the
DNA, often flanking genes (Gasser and Laemmli, 1987). They
are thought to form the boundaries of transcriptionally ac-
tive DNA loop domains, and the inclusion of MAR ele-
ments at the borders of the Agrobacterium T-DNA is thus
predicted to insulate the inserted transgene within the loop
from the influences of surrounding chromatin. MAR se-
quences have been shown to be the good candidates to
increase and stabilize the expression of transgene (usually
b-glucuronidase gene (GUS)) (Bonifer et al., 1991; Holmes-
Davis and Comai, 1998).
The phenolic compound resveratrol (trans-3,4,5-
trihydroxystilbene) is a non-flavonoid phytoalexin produced
by plants usually under environmental stresses. It has pro-
voked an intense interest not only due to its antifungal
properties, but also because it is one of the main constitu-
ents of wine which is thought to confer protection against
artherosclerosis, coronary heart disease and cancer (Jang
et al., 1997). The key enzyme in metabolic pathway of
resveratrol biosynthesis is stilbene synthase, which cata-
lyzes precursors p-coumaryl-SCoA and malonyl-SCoA to
produce resveratrol. While the precursor molecules widely
exist in plant (Hain et al., 1990), the stilbene synthase genes
(STS) have been cloned and identified only from a very
small number of unrelated plant species, including peanut,
two species of pines and three varieties of cultivated grape-
vines so far. So, there appears increasing interest in trans-
ferring STS gene into plants lacking stilbene synthase in
recent years.
ZHONG Jin et al.: Effect of Matrix Attachment Regions on Resveratrol Production in Tobacco with Transgene of Stilbene
Synthase from Parthenocissus henryana 949
The STS gene studied in this research was cloned
from Parthenocissus henryana which is also a Vitaceae
plant belonging to different genus with grapevine by
RT-PCR method for the first time. Under the control of
strong constitutive promoter, the STS gene was expected
to express constitutively and to produce resveratrols in
transgenic plants.
The MARs sequence used in this research was cloned
from yeast, which was also reported as an ARS-1
(autonomously replicating sequence) element (Amati and
Gasser, 1988). After proving its affinity to the tobacco
nuclear matrix, the MAR fragments were cloned into the
binary vector by flanking the STS gene and bar gene cas-
settes within the T-DNA borders, and the same vector with-
out MARs was used as control. The goal of this research
was to investigate the effect of MAR sequence on the ex-
pression level of the STS gene by determining the amount
of the resveratrol produced in transgenic tobaccos, and by
which to do some basic work for the STS gene-transformed
vegetables or fruits that are more disease-resistant while
having additional health protection effects.
1 Materials and Methods
1.1 Cloning of MAR sequence
MARs sequence from yeast was the PCR-amplified se-
quence of total genomic DNA from Saccharomyces
cerevisiae A364a (a gift from Prof. Lee Hartwell, Washing-
ton University). The method for isolation of yeast genomic
DNA was based on the SDS protocols. With both Hind Ⅲ
sites at the 5 end, the following two primers were designed
according to the complete sequence of the yeast EcoRⅠ
TRP1 fragment from 628 to 1 240 because of its ability of
attachment to the yeast nuclear matrix (Amati and Gasser,
1988): 5-GTAAGCTTATGTTAGCTGGTGGACTGACG-3
and 5-TGAAGCTTTGAAGGAGCATGTTCGGC-3.
Samples containing about 20 ng of genomic DNA were first
heated at 94 ℃ for 5 min followed by 30 cycles of 94 ℃ for
1 min, 52 ℃ for 40 s and 72 ℃ for 50 s, and a final extension
time at 72 ℃ for 7 min. The amplified 822 bp fragment was
inserted into pGEMR-T Easy Vector (Promega) to form pT-
MAR for sequencing and further constructions.
1.2 Matrix isolation and binding assay
Nuclei were obtained from mesophyll protoplasts iso-
lated from leaves of 4-5-week-old tobacco plants grown in
greenhouse. Nucleus isolation and matrix preparation were
performed essentially as described by Chinn and Comai
(1996), which employs lithium diiodosalicylate (LIS) to re-
move chromosomal proteins and uses restriction enzymes
for digestion. Binding assay was performed by a modification
of the method of Mirkovitch et al. (1984). The prepared
matrix containing 40-50 U appropriate restriction enzymes
(HindⅢ, EcoRⅠ and/or BamHⅠ) was incubated at 37 ℃
with shaking for 30 min. About 100 ng HindⅢ -digested
plasmid pT-MAR and 10 mg of sheared Escherichia coli
genomic DNA and 10 mg of sheared tobacco genomic DNA
acted as competitor DNA were added together and incu-
bated for 10 h at 37 ℃. Matrix was collected by centrifuga-
tion for 10 min at 12 000g and washed twice in 200 mL
digestion buffer (20 mmol/L Tris (pH 7.4), 70 mmol/L NaCl,
0.05 mmol/L spermidine, 0.125 mmol/L spermine, 10 mmol/L
MgCl2). DNA was isolated from pellet fractions by resus-
pending in 300 mL SET-K ( 0.1 mol/L Tris (pH 8.5), 1% SDS,
0.1 mol/L Na2EDTA,100 mg proteinase K) , pooled superna-
tant and wash fractions by the addition of SDS, Na2EDTA
and Tris to make the same final concentrations as above.
Fractions were incubated for 1 h at 60 ℃, and extracted
with phenol/chloroform. DNA was precipitated by the ad-
dition of two volumes 0.6 mol/L LiCl in ethanol. Southern
blotting analysis was performed with approximately equal
counts of DNA from pellet and supernatant fractions, and
equal counts of added plasmid DNA. DNA from different
fractions were electrophoresed on a 1% agarose gel and
then transferred to Nylon membranes (Hybond N＋;
Amersham). The hybridization and detection were per-
formed following the instructions of DIG DNA Labeling
and Detection Kit (Catalog No.1093657, Boehringer
Mannheim) with DIG-dUTP labeled MAR fragments by PCR
procedure and labeled pGEMR-T Easy Vector (Promega) by
random-primed labeling as probes.
1.3 Cloning of STS coding sequence from P. henryana
Total RNA was prepared from young leaves of P.
henryana according to the method of Chang et al. (1993).
The primers for cloning are designed based on relatively
conserved nucleotide sequences of published STS gene of
three varieties of cultivated grapevines, V. vinifera, V.
labrusca and V. riparia (GenBank accession No. AF23594,
AF499024 and AF128861) respectively. The upstream primer
STSP1 was 5-CGGGATCCGCCATGGCTTCAGTTGAGAA-
ATTTAG-3 containing BamHⅠ site and the downstream
primer STSP2 was 5-GTGAGCTCGAAGGGTAAACCATTC
TCTTTTAT-3 carrying SacⅠ site. cDNA was synthesized
using primer STSP2 and PCR amplified with primers STSP1
and STSP2. The 50 mL amplification system included about
300 ng template cDNA, 200 mmol/L dNTPs, 0.5 mmol/L
primers, 5U Taq PlusⅠ DNA polymerase (Sangon). The
cycling parameters were: 94 ℃ for 8 min followed by 30
cycles at 94 ℃ for 45 s, 53 ℃ for 45 s and 72 ℃ for 1 min 30
s, and a final extension time at 72 ℃ for 10 min. The
Acta Botanica Sinica 植物学报 Vol.46 No.8 2004950
amplified fragment was cloned into pGEM T-easy Vector
(Promega) to get plasmid pT-STS for sequencing and fur-
ther constructions.
1.4 Plasmid constructions
Plasmid pGW4AB providing promoter region
CaMV35SW that included both CaMV35S and 90 bp W en-
hancer fragment of TMV was a gift from Prof. GUO San-Dui
(The Chinese Academy of Agricultural Sciences). And the
promoter region with ApaⅠ+XbaⅠ ends and STS gene
fragment with BamHⅠ+SacⅠ ends were both inserted into
the pGEM-7Z vector (Promega) to form plasmid p7Z-T1W.
Plasmid pMGBM/pGB containing 35S-GUS-Tnos and Pnos-
bar-Tnos cassettes with/without flanking MAR fragments
was constructed by our laboratory. By replacing the 35SW-
STS fragment in p7Z-T1W with 35S-GUS fragment in plas-
mid pMGBM by ApaⅠ+SacⅠ, the resultant plasmid
pMT1WBM was constructed with bar gene as selective
marker; 35SW-STS fragment in the p7Z-T1W was trans-
ferred into plasmid p5Z-T1W by SacⅠ+NotⅠ, and by re-
placing the 35SW-STS fragment in p5Z-T1W by SacⅠ+Sph
Ⅰ with 35S-GUS fragment in plasmid pGB by SacⅠ+Pst
Ⅰ, the resultant plasmid pT1WB was constructed.
1.5 Agrobacterium tumefaciens-mediated plant trans-
formation
The binary expression plasmids pMT1WBM and
pT1WB were transferred into Agrobacterium tumefaciens
strain LBA4404 by freeze-thaw method. The leaf discs of
Nicotiana tabacum L. var. W38 were co-cultured with A.
tumefaciens and the plantlets differentiated on the MS
medium supplemented with 1.0 mg/L benzyl-adenine (6-BA),
2.0 mg/L phosphinotricin (PPT) and 500 mg/L carbenicillin
(Cb). The regenerated shoots (2-3 cm) were transferred
into hormone free 1/2 MS with 2 mg/L PPT and 500 mg/L Cb
for rooting. Well-rooted plants were subsequently trans-
ferred into soil and kept under greenhouse conditions.
1.6 PCR detection and Southern blotting analysis
Total DNA prepared from regenerated tobacco by CTAB
method was used for PCR template. The amplification for
STS coding sequence was performed using primers STSP1
and STSP2, and the PCR products were checked for the
special DNA band (1.2 kb) on agarose electrophoresis gel.
Genomic DNA of PCR positive transformants were
extracted. Single-cut enzymes SacⅠ, BamHⅠ, EcoRI in
pT1WB and SacⅠ in pMT1WBM, while BamHⅠ+EcoR
Ⅰ were used respectively to digest the total DNA of pT1WB
and pMT1WBM transformants. About 10 mg sheared ge-
nomic DNA was electrophoresed on a 0.8% agarose gel,
transferred by alkali blotting (3 mol/L NaCl, 0.4 mol/L NaOH)
to Nylon membranes (Hybond N＋; Amersham) and then
fixed by UV cross-linking. The hybridization and detection
were performed following the instructions of DIG DNA La-
beling and Detection Kit (Catalog No.1093657, Boehringer
Mannheim) with digoxigenin-dUTP labeled STS fragment
(1.2 kb) by PCR as probe.
1.7 Northern blotting analysis
Total RNA was isolated from 0.2 g leaves of
transformants by using Trizol Reagent method. RNA dena-
turation and separation on EB containing 1.2% (W/V) aga-
rose/formaldehyde gels was carried out according to stan-
dard methods (Sambrook et al., 1989). RNA was blotted
onto Hybond N+ membranes (Amersham) by capillary trans-
fer using 20× SSC and then fixed by UV cross-linking.
Prehybridization and hybridization of the membrane were
performed at 68 ℃ in a solution containing 0.5 mol/L
Na2HPO4, 0.5 mol/L NaH2PO4, 7% SDS and 1 mmol/L EDTA
(pH 7.2). The following steps were the same as that of the
Southern blotting. Digoxigenin-dUTP-labeled STS fragment
amplified by asymmetry PCR (the amount of primers was
unequal, STSP1/STSP2=1/10) was used as hybridization
probe.
1.8 HPLC analysis of resveratrol
The extraction and purification by thin layer chroma-
tography on silica gel of resveratrol from the leaves of
transgenic and control plants were based on the method of
Subikova (1991). All these steps were carried out under
light protection. HPLC was performed on a HPCHEM high
performance liquid chromatograph (Japan) using a Nucleosil
C18 column (4.6×150 mm, 5 mm) and H2O-acetonitrile as
eluent (acetonitrile:H2O=35:65, flow rate 0.6 mL/min) at room
temperature. The detection was under 306 nm. The stan-
dard sample of resveratrol was purchased from Sigma.
2 Results
2.1 Binding ability of MAR sequence to the nuclear
matrix
Sequence analysis of the 822 bp fragment cloned from
tobacco showed that it had high A/T content (62%) and
bore many DNA motifs which had been identified as being
characteristic of MARs, such as A-boxes, T-boxes and au-
tonomously replicating sequence (ARS) motif.
Procedure using lithium diiodosalicylate (LIS) and re-
striction enzymes were performed to obtain tobacco nuclear
matrix in the in vitro binding assay. To test whether the
cloned fragment could bind to the nuclear matrix of tobacco
or not, the digested plasmid pT-MAR containing this frag-
ment and pGEMR-T Easy Vector was added into the nuclear
matrix preparations in the presence of sheared bacterial and
digested tobacco genomic DNA as competitor DNA. The
ZHONG Jin et al.: Effect of Matrix Attachment Regions on Resveratrol Production in Tobacco with Transgene of Stilbene
Synthase from Parthenocissus henryana 951
inclusion of pGEMR-T Easy Vector provided a convenient
internal control of a non-MAR DNA fragment. After the
matrix-bound and solubilized DNA were separated and pu-
rified respectively, the same fraction of them and the plas-
mid added were both hybridized with digoxigenin-dUTP-
labeled MAR fragment and pGEMR-T Easy Vector (Fig.1).
The results showed that the cloned MAR fragment could
specifically bind to the matrix of tobacco and retained in
the matrix-containing pellet, whereas the pGEMR-T Easy
Vector fragment could not bind and essentially all vector
was found in the supernatant. This ability of MARs to bind
matrix from other species also demonstrated that they are
functionally conserved during evolution.
2.2 Sequence of P. henryana STS
Cloning and sequencing of the 1.18 kb STS gene from
P. henryana (GenBank accession No. AY094615) revealed
that it had 94.5% nucleotide identity to the published stil-
bene synthase of V. vinifera cv. Optima (accession No.
235945), while the deduced amino acid sequence identity
between them was 96.9%.
The examination of the predicted protein sequence of
the P. henryana also showed that the sequence motif,
IPNSAGAIAGN, which is specific for stilbene synthase,
was present; while the sequence motif, GVLFGFGPGLT,
which is the family signature sequence for stilbene
synthase, just appeared as GVLFGFGSGLT with P turned
to S. Furthermore, the conserved cys164 that was es-
sential for the activity of stilbene synthase was also
found in the predicted protein sequence of the P.
henryana (Fig.2). These results suggested that the STS
gene was rather conserved even between different ge-
nus plants.
2.3 Plasmid constructs used for transformation
The coding region of the STS gene from P. henryana
was under control of promoter region 35SW (Prof. GUO
San-Dui, personal communication) that contained
CaMV35S, a 90-bp W enhancer sequence from TMV and
a Kozak sequence. Such strong promoter was very ben-
eficial to the constitutive expression of transgene in the
plants. bar gene was used as selective marker. Figure 3
shows the constructions of plasmid pMT1WBM contain-
ing flanking MARs and pT1WB without MARs for
transformations.
2.4 STS gene-transformed plants
STS gene was introduced into tobacco by leaf disc trans-
formation method. Transgenic plants were first identified
by PCR amplification of STS gene, bar gene and MAR se-
quence (Figures not shown). The integration of STS cod-
ing region in transformants was demonstrated by Southern
blotting analysis which showed 1.5 kb and 3.0 kb of hybrid-
ization bands in pT1WB and pMT1WBM transformants
respectively (total DNA digested by BamHⅠ+EcoRⅠ)
probed by DIG-labeled STS gene (Fig.4). By further experi-
ments on copy-number determination of transferred STS
gene (Fig.4), nine lines of pT1WB and nine lines of
pMT1WBM transformants which all had two copies were
Fig.1. Specific in vitro binding of matrix attachment region
(MAR) sequence to the tobacco nuclear matrix. Tobacco nuclear
matrix was extracted by lithium diiodosalicylate (LIS) procedure.
The MAR fragment was cloned from yeast. M, 1 kb plus DNA
Ladder; P, aliquot of the DNA found in the pellet, i.e. bound to
the nuclear matrix; S, aliquot of the DNA found in the supernatant,
i.e. not bound to the nuclear matrix; T, aliquot of total Hind Ⅲ-
digested plasmid pT-MAR put into the binding reaction. Bands
marked “V” is the DNA fragments of the vector.
Fig.2. Deduced amino acid sequence of coding region of stilbene synthase of Parthenocissus henryana.Total amino acid number is 392.
Signature motifs of stilbene synthase are shaded in bold letter. The conserved cys164 is underlined.
Acta Botanica Sinica 植物学报 Vol.46 No.8 2004952
chosen for further comparison and study. Northern blot
results of 18 lines of transformants (nine lines of
pMT1WBM and nine lines of pT1WB, respectively)
showed that the introduced STS gene could be transcribed
and expressed normally in transgenic plants (Fig.5), and it
seemed that the STS mRNA accumulations in pMT1WBM-
transformed plants were overall higher than that of
pT1WB transformants. This result was also in accor-
dance with the quantitative analysis of resveratrol pro-
duction in transformants by HPLC to some extent.
2.5 HPLC analysis of resveratrol in transformants
To better compare the production of resveratrol in
pT1WB and pMT1WBM group, two trains of each line
of transformants were got. HPLC analysis of the leaf ex-
tracts of both transgenic plants and untransformed plants
showed that while the control plant had not any peaks at
the corresponding retention time of the standard sample
resveratrol, the transgenic lines had obvious correspond-
ing peaks, demonstrating the production of the resveratrols
and the alike substances by the expression of the intro-
duced STS gene. The quantitative analysis of resveratrol
production by HPLC also showed that, comparing to the
transformants lacking MARs, the content of resveratrol
appeared higher in pMT1WBM transformants having
MARs (Fig.6, Table 1).
3 Discussion
This work allowed us to study the effects of flanking
yeast MARs on transgene (STS) expression and through
which to increase the resveratrol production in transgenic
plants. Of the various roles suggested for MARs, the role
of boundary element and enhancer element are among the
most important when these sequences are used to flank
transgene. When a gene flanked by MARs is integrated
into the chromosome, it can presumably form an indepen-
dent domain which could insulate the transgene from the
effects of neighboring DNA sequences, and it is easily ac-
cessible to the transcription machinery (Holmea-Davis and
Comai, 1998; Allen et al., 2000). Thus, both increases at
transgene expression levels and reduction of variations
Fig.3. Construction of binary vectors of pT1WB and pMT1WBM. STS is stilbene synthase coding region from Parthenocissus
henryana; bar is bialaphos resistance gene; 35SW represent CaMV35S and W enhancer; matrix attachment region (MAR) is a
matrix attachment region from yeast; Pnos and Tnos are the promoter and the terminator of nos gene from Agrobacterium
tumefaciens, respectively.
Fig.4. Southern blotting analysis of transformants. 1-4,
pT1WB transformant digested by EcoRⅠ, BamHⅠ, SacⅠ and
EcoRⅠ+BamHⅠ, respectively; 5, untransformed W38 negative
control; 6-7, pMT1WBM transformant digested by SacⅠ,
EcoRⅠ+BamHⅠ, respectively; 8, 1.2 kb positive control band.
Fig.5. Northern blotting analysis of transformants. A. pT1WB
transformants. B. pMT1WBM transformants. 1, untransfor-
med W38 negative control; 2-10, nine lines of independent
transformants.
ZHONG Jin et al.: Effect of Matrix Attachment Regions on Resveratrol Production in Tobacco with Transgene of Stilbene
Synthase from Parthenocissus henryana 953
among transformants are expected from such a domain
model.
However, though the main tendency of the MAR ef-
fects of above has been demonstrated, results from studies
with different MARs sequences flanking transgenes in dif-
ferent transgenic systems have shown variations (Cheng
et al., 2001). These differences in the effects of MARs se-
quence on transgene expression might come from the dif-
ferent MARs themselves, the different transformation meth-
ods or the different transgenic systems. We also studied
the effects of the same MAR sequence (the yeast MARs in
present study) on transgene of b-glucuronidase (GUS) re-
porter gene in transgenic tobacco system and transgenic
fungus (Ganoderma lucidum) system, and found the ef-
fects of the same MARs differ slightly (3.13-fold in tobacco
and 3.56-fold in G. lucidum; results will be published
elsewhere). Thus, the effect of MAR sequences on
transgenic expression should be tested on a case-by-case
basis. In this research we brought the MARs into practical
use to increase the expression level of transgenic STS gene,
and studied the effect of MARs on resveratrol production
in transgenic tobacco.
The effects of resveratrol in disease resistance and
health protection have been demonstrated in recent years,
and the interests of studies on stilbene synthase, the key
enzyme to synthesis resveratrol in plant, have also in-
creased gradually. We isolated a stilbene synthase gene
from Parthenocissus henryana, which had not been re-
ported before (GenBank accession No. AY094615). By
transferring the cloned STS gene into the tobacco lack-
ing of stilbene synthase, the resveratrol and the alike
substances could be produced in transgenic plants. Since
the precursors of stilbene synthase, malonyl-CoA and p-
coumaroyl-CoA, were both present in plant including to-
bacco (Hain et al., 1990), the integration and expression of
the introduced STS gene could be further confirmed by the
production of resveratrol, while the control plant which
was not transformed by the STS gene had no stilbene syn-
thase in itself and could not synthesis resveratrol in the
plant. Constitutive expression of STS in transgenic plant
also avoided the common situation that the resveratrol
was only produced under the plant stresses to some
extent. The same construction (pMT1WBM) could also
be transferred into certain vegetables or fruits, in order to
get vegetables or fruits that are more disease-resistant while
having additional health protection effects, and the rela-
tive experiments are being carried out so far.
In our previous study also on the yeast MARs’ ef-
fects on expression level of transgenic b-glucuronidase
(GUS) reporter gene, this MARs fragment could increase
the GUS expression to 3.13 folds, comparing to the 2.02
folds of STS gene expression increases in the absolute same
transformation systems. This difference might essentially
come from the transgene itself. In fact, to study the ef-
fects of MAR sequence on expression level of transgenic
STS gene further and to make better use of MAR se-
quence in increasing resveratrol production in
transgenic plants, more relative research work still needs
to be carried out, for the process of resveratrol produc-
tion is very complicated. Many factors including the pre-
cursors synthesis and the growth conditions of plant es-
pecially the environmental elements might also affect the
synthesis of resveratrol. For example, the content of p-
coumaryl-SCoA and malonyl-ScoA, the precursors of stil-
bene synthase from the phenylpropanoid pathway, was
not consistent and fixed in the plant; meanwhile, the envi-
ronmental factors like UV might also influence the synthe-
sis of the precursors. Besides, the metabolic process would
be influenced by developmental stage of plant and it would
lead to change of the resveratrol biosynthesis rate. The
developmental and environmental factors related to the
resveratrol synthesis remained to be studied further.
Table 1 Statistics of data of resveratrol production in pT1WB and pMT1WBM transformants
Number of transformants in
Mean SE Variance
Coefficient of
P Fold
population variation
pT1WB 18 2.075 1.346 1.712 0.649 *** -
pMT1WBM 18 4.192 2.556 6.169 0.610 - 2.020
P, probability according to F-test for homogeneity of variance compared with the corresponding control population pT1WB; Fold, increased
resveratrol content with respect to corresponding control population pT1WB; ***, significant at P < 0.001.
Fig.6. Comparison of resveratrol product content in pT1WB
and pMT1WBM transformants.
Acta Botanica Sinica 植物学报 Vol.46 No.8 2004954
Since the main goal in molecular breeding are to maxi-
mize expression levels, especially with respect to transgene-
coded protein, and to minimize variations in gene expres-
sion among transgenic lines, inclusion of the MAR se-
quence has been demonstrated to be useful in present
research. Our results showed that the yeast MARs have
positive effect in transgenic tobacco when STS gene/se-
lectable marker gene cassette is flanked by this fragment,
and the higher content of resveratrol has been produced in
transgenic tobacco containing this MARs than that of the
transformants lacking this MARs. The application of MARs
in STS gene-transformed plants has proved to be one of
the effective ways to get plants more disease-resistant while
having additional health protection effects.
References:
Allen G C, Spiker S, Thompaon W F. 2000. Use of matrix attach-
ment regions (MARs) to minimize transgene silencing. Plant
Mol Biol, 43: 361-176.
Amati B B, Gasser S M. 1988. Chromosomal ARS and CEN
elements bind specifically to the yeast nuclear scaffold. Cell,
54: 967-978.
Bonifer C, Hecht A, Saueressing H, Winter D M, Sippel A E.
1991. Dynamic chromatin: the regulatory domain organiza-
tion of eukaryotic gene loci. J Cell Biochem, 47: 99-108.
Chang S, Puryear J, Cairney J. 1993. A simple and efficient method
for isolating RNA from pine trees. Plant Mol Biol Rep, 11:
113-116. (Managing editor: ZHAO Li-Hui)
Cheng Z Q, Targolli J, Wu R. 2001. Tobacco matrix attachment
region sequence increased transgene expression levels in rice
plants. Mol Breeding, 7: 317-327.
Chinn A M, Comai L. 1996. The heat shock cognate 80 gene of
tomato is flanked by matrix attachment regions. Plant Mol
Biol, 32: 959-968.
Gasser S M, Laemmli U K. 1987. A glimpse at chromosomal
order. Trends Genet, 3: 16-22.
Hain R, Bieseler B, Kindl H, Schröder G, Stöcker R. 1990. Ex-
pression of a stilbene synthase gene in Nicotiana tabacum
results in synthesis of the phytoalexin resveratrol. Plant Mol
Biol, 15: 325-335.
Holmes-Davis R, Comai L. 1998. Nuclear matrix attachment re-
gions and plant gene expression. Trends Plant Sci, 3: 91-96.
Jang M, Cai L, Udeani G O, Slowing K W, Thomas C F, Beecher
C W W, Fong H H S, Farnsworth N R, Kinghorn A D, Mehta
R G, Moon R C, Pezzuto J M. 1997. Cancer chemo-preven-
tive activity of resveratrol, a natural product derived from
grapes. Science, 275: 218-220.
Mirkovitch J, Mirault M E, Laemmli U K. 1984. Organization of
the higher-order chromatin loop: specific DNA attachment
sites on the nuclear scaffold. Cell, 39: 223-232.
Sambrook J, Fritsch E F, Maniatis T. 1989. Molecular Cloning: a
Laboratory Manual. 2nd ed. New York: Cold Spring Harbor
Laboratory Press.
Subikova V. 1991. Resveratrol accumulation in grapevine infected
with grapevine vein necrosis disease. Biol Plantarum (PRAHA),
33: 287-290.