| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Plant Physiol, February 2002, Vol. 128, pp. 696-706 Antisense RNA-Mediated Suppression of Benzophenanthridine Alkaloid Biosynthesis in Transgenic Cell Cultures of California Poppy1Department of Biological Sciences, University of Calgary, Calgary, Alberta, Canada T2N 1N4
California poppy (Eschscholzia californica
Cham.) cell cultures produce several benzophenanthridine alkaloids,
such as sanguinarine, chelirubine, and macarpine, with potent
pharmacological activity. Antisense constructs of genes encoding
two enzymes involved in benzophenanthridine alkaloid biosynthesis,
the berberine bridge enzyme (BBE) and N-methylcoclaurine
3'-hydroxylase (CYP80B1), were introduced separately into California
poppy cell cultures. Transformed cell lines expressing antisense
BBE or antisense CYP80B1 constructs and
displaying low levels of BBE or CYP80B1 mRNAs, respectively, showed
reduced accumulation of benzophenanthridine alkaloids compared with
control cultures transformed with a
Alkaloids are a diverse group of
low-Mr, nitrogenous compounds found in
about 20% of plant species. Many of the approximately 12,000 alkaloids
for which structures have been described function in the defense of
plants against herbivores and pathogens. The potent biological activity
of some alkaloids has also led to their widespread use as
pharmaceuticals. The benzylisoquinoline alkaloid class, in particular,
includes several important medicinal compounds such as morphine,
codeine, papaverine, berberine, (+)-tubocurarine, and sanguinarine. All
benzylisoquinoline alkaloids share a common biosynthetic origin
beginning with a lattice of decarboxylations, ortho-hydroxylations, and deaminations that convert
L-tyrosine into both dopamine and
4-hydroxyphenylacetaldehyde. Dopamine and 4-hydroxyphenylacetaldehyde condense to form the trihydroxylated alkaloid (S)-norcoclaurine (Fig.
1), which is the central precursor to
all benzylisoquinoline alkaloids produced in plants (Stadler et
al., 1987
(S)-Reticuline is a branch-point intermediate involved in
the biosynthesis of many structural types of benzylisoquinoline alkaloids. One such group, the benzophananthridine alkaloids, includes
the antibiotics sanguinarine, which is used in various oral hygiene
products, and marcarpine, the most highly oxidized alkaloid in its
class. The first committed step in benzophenanthridine alkaloid
biosynthesis involves conversion of the N-methyl group of
(S)-reticuline into the methylene bridge moiety of
(S)-scoulerine by the BBE (Fig. 1).
(S)-Scoulerine is then converted by two P450-dependent oxidases, an N-methyltransferase, and two additional
P450-dependent hydroxylases to dihydrosanguinarine, the first alkaloid
with the benzophenanthridine nucleus (Facchini, 2001 Plant metabolic engineering is a relatively new field of research with
the potential to create new opportunities for the improvement of plant
metabolic, cellular, and physiological processes. In recent years, a
number of impressive strategies for the genetic modification of several
important plant metabolic pathways have been reported. These include
the reduction of indole glucosinolate levels in Brassica
napus seeds (Chavadej et al., 1994 In this paper, we report the suppression of benzophenanthridine alkaloid biosynthesis in cell cultures of California poppy transformed with antisense BBE and antisense CYP80B1 constructs. The consequences of restricting metabolic flux into benzophenanthridine alkaloid biosynthesis at these two key points in the pathway were evaluated by two-dimensional 1H-NMR and in vivo 15N-NMR spectroscopy. Our results show that changes in metabolic flux through benzylisoquinoline alkaloid biosynthesis can affect the regulation of amino acid pools. Our data also provide new insight into the development of metabolic engineering strategies that target benzylisoquinoline alkaloid pathways.
Suppression of BBE or CYP80B1 mRNAs Reduces Benzophenanthridine Alkaloid Accumulation in California Poppy Cell Cultures Several independent, paromomycin-resistant callus lines were
obtained from excised California poppy cotyledons exposed to Agrobacterium tumefaciens strain GV3101 harboring the
35S::GUS, 35S::antiBBE, or
35S::antiCYP80B1 constructs. The callus cultures were maintained on growth media containing paromomycin and timentin for
several months to promote the selection of transgenic cells and the
complete elimination of A. tumefaciens. PCR performed using
genomic DNA isolated from each paromomycin-resistant cell line and
primers specific to sequences in the NTPII selectable marker
gene consistently produced a single amplicon with the expected size of
823 bp. No amplicons were obtained using genomic DNA isolated from
wild-type California poppy cell cultures. Suspension cultures were
subsequently initiated from several cell lines and eventually transferred to growth media lacking antibiotics. This last step was
necessary because benzophenanthridine alkaloid accumulation in
California poppy cultures is inducible by antibiotics (Schumacher et
al., 1987 Northern-blot hybridization analysis using radiolabeled sense RNA
probes demonstrated the presence of antisense BBE and antisense CYP80B1
transcripts in California poppy cell cultures transformed with
35S::antiBBE and 35S:antiCYP80B1,
respectively (Fig. 2). Alternatively, the
use of antisense RNA probes showed a dramatic reduction in BBE and
CYP80B1 mRNA levels in some transgenic cell lines expressing
35S::antiBBE and
35S::antiCYP80B1, respectively, compared with
control cultures expressing 35S::GUS (Fig. 2; cell lines BBE-A1 and CYP80B1-A1). BBE mRNAs were nearly undetectable in
some transgenic cell lines, such as BBE-A1, that produced high levels
of antisense BBE transcripts. Although CYP80B1 mRNA levels were
extensively reduced in some transgenic cell lines that produced high
levels of antisense CYP80B1 transcripts, such as CYP80B1-A1, the
reduction was never as substantial as that found in BBE-A1 (Fig. 2) and
other antisense BBE cell lines. However, BBE and CYP80B1
mRNA levels were not reduced in all cell lines with abundant antisense
BBE or antisense CYP80B1 transcripts, respectively (Fig. 2; cell lines
BBE-A2 and CYP80B1-A2). As expected, BBE activity was similar in the
BBE-A2 (16 ± 8 pkat mg
GUS transcripts were abundant in control cultures transformed with 35S::GUS. Moreover, BBE and CYP80B1 mRNA levels were identical to those in wild-type cultures (data not shown). CYP80B1 transcript levels in the BBE-A1 and BBE-A2 cell lines were also identical to those in wild-type cultures. Similarly, BBE mRNA levels in the CYP80B1-A1 and CYP80B1-A2 cell lines were the same as those in wild-type cultures. Transcript levels of another alkaloid biosynthetic gene, Tyr/dopa decarboxylase, were identical in control, BBE-A1, and CYP80B1-A1 transgenic cell lines (data not shown). Cell lines transformed with 35S::GUS exhibited the same red color seen in wild-type California poppy cell cultures. However, transgenic cell lines that showed a substantial reduction in BBE and CYP80B1 mRNA levels (i.e. BBE-A1 and CYP80B1-A1) did not possess the red pigmentation found in wild-type and control cultures. HPLC analysis showed that benzophenanthridine alkaloid levels were substantially reduced in cell lines that displayed high levels of antisense BBE or antisense CYP80B1 transcripts and low levels of BBE and CYP80B1 mRNAs, respectively, compared with controls (Fig. 3). No new peaks were apparent on HPLC chromatograms of extracts from cell lines with low levels of BBE or CYP80B1 mRNAs and significantly reduced total alkaloid accumulation compared with controls (Fig. 3). In particular, a peak corresponding to (S)-reticuline was not detected on HPLC chromatograms of alkaloid extracts from the BBE-A1 cell line. The extraction of (S)-reticuline using the protocol described herein was confirmed by adding authentic (S)-reticuline to some samples as an internal standard. Moreover, (S)-reticuline was not detected in the original methanol extracts (data not shown).
The total benzophenanthridine alkaloid content of the BBE-A1 cell line was 8- to 10-fold lower than that of control cells transformed with 35S::GUS (Table I). Similarly, the total alkaloid content of the CYP80B1-A1 cell line was 3- to 4-fold lower than that of controls (Table I). The level of most individual alkaloids was reduced in both the BBE-A1 and CYP80B1-A1 cell lines compared with controls, with the exception of dihydrochelilutine, which was present at marginally higher levels in cultures exhibiting a reduced abundance of BBE and CYP80B1 mRNAs (Table I). Cell lines that produced BBE and CYP80B1 antisense mRNAs but did not show a reduction in the corresponding sense mRNA levels (i.e. BBE-A2 and CYP80B1-A2) displayed total and individual levels of alkaloid accumulation that were not significantly different from those of controls (Table I).
Suppression of Benzophenanthridine Alkaloid Biosynthesis Reduces the Growth Rate of California Poppy Cell Cultures Transgenic cell lines with low levels of BBE or CYP80B1 mRNAs, and significantly reduced total alkaloid accumulation, displayed growth rates that were considerably slower than those of control cultures (Fig. 4). The growth rates of control cultures transformed with 35S::GUS were not significantly different from those of wild-type cultures (data not shown). Similarly, the growth rates of the BBE-A1 and CYP80B1-A1 cell lines were not significantly different from each other (Fig. 4).
Antisense RNA-Mediated Reduction in BBE and CYP80B1 mRNA Levels Increases Free Amino Acid Pools in California Poppy Cells Two-dimensional 1H-NMR spectroscopy showed
that the cellular pools of several amino acids were larger in cultures
(i.e. BBE-A1 and CYP80B1-A1) exhibiting low levels of BBE or CYP80B1
mRNAs and significantly reduced total alkaloid accumulation (Fig.
5). Specifically, cellular pools of Ala,
Leu, Phe, Thr, and Val were larger in cultures with reduced alkaloid
levels compared with controls. In contrast, the relative abundance of
Tyr appeared to be only marginally higher in the BBE-A1 and
CYP80B1-A1 cell lines compared with controls (Fig. 5). Cellular levels
of several carbohydrates including Suc,
Constitutive expression of antisense BBE or antisense CYP80B1 genes in transgenic California poppy cell cultures yielded several lines with high levels of antisense transcripts. In approximately 50% of these lines, corresponding sense mRNA levels were substantially reduced compared with controls transformed with 35S::GUS. In the remaining cell lines, little change in sense mRNA levels was detected relative to controls. Two transgenic lines, one displaying nearly undetectable and the other relatively normal sense mRNA levels, were selected from each set of cell lines producing antisense transcripts. Two transgenic control lines were also selected at random. The BBE-A1 and CYP80B1-A1 cell lines accumulated low to undetectable levels of BBE and CYP80B1 mRNAs, whereas the BBE-A2 and CYP80B1-A2 cell lines displayed mRNA levels similar to those found in controls (Fig. 2). The suppression of BBE mRNA accumulation in the BBE-A1 cell line was also accompanied by a substantial reduction in BBE activity. CYP80B1 activity was not measured because an enzymatic substrate was not available. However, it can be assumed that CYP80B1 activity was also substantially reduced in transgenic cell lines with low levels of CYP80B1 mRNA compared with controls. The suppression of BBE and CYP80B1 mRNAs in transgenic California poppy
cell cultures resulted in a dramatic reduction in the accumulation of
benzophenanthridine alkaloids (Fig. 3; Table I). BBE activity was below
the detection limit of the assay. However, the accumulation of
alkaloids at one-tenth the control level suggests that BBE activity was
reduced by about 90% compared with controls. Alkaloid accumulation in
transgenic lines that displayed control levels of BBE or CYP80B1 mRNAs
was essentially the same as that in control cultures. Transgenic cell
lines expressing 35S::GUS were used as controls to
account for any metabolic or cellular alterations caused by the
transformation process. Previously, we showed that the transformation
of California poppy and opium poppy plants and root cultures did not
alter the growth rate, anatomy, or benzylisoquinoline alkaloid content
of these systems (Park and Facchini, 2000a Several points about the effect of the antisense RNA-mediated suppression of benzophenanthridine alkaloid biosynthesis should be noted. First, a decrease in either BBE or CYP80B1 mRNA levels produced a similar reduction in alkaloid accumulation. Second, the metabolic intermediates (S)-reticuline and (S)-N-methylcoclaurine, which serve as substrates for BBE and CYP80B1, respectively, did not accumulate in antisense-suppressed cells (Fig. 3). The antisense suppression of BBE and CYP80B1 mRNAs caused an overall reduction in benzophenanthridine alkaloid accumulation, but the absence of additional peaks on HPLC chromatograms shows that pathway intermediates did not accumulate. Third, the silencing of BBE or CYP80B1 mRNAs resulted in a significant reduction in the growth rate of the cell cultures. The reduced growth rate occurred in all antisense-suppressed cell lines (Fig. 4), but not in transgenic cell lines exhibiting control levels of BBE and CYP80B1 mRNAs (data not shown). The reduced growth rate might have resulted from the low-level accumulation of cytotoxic pathway intermediates. We suggest three hypotheses for the lack of a substantial accumulation
of pathway intermediates in antisense-suppressed cell lines. The first
involves feedback inhibition by one or more alkaloid intermediates on
early biosynthetic enzymes. However, the inhibition of enzymes involved
in benzylisoquinoline alkaloid biosynthesis by pathway intermediates or
end products has not been reported. The second concerns the putative
degradation of alkaloid intermediates by an as-yet-uncharacterized
mechanism. A third hypothesis raises the possibility that
benzylisoquinoline alkaloid biosynthetic enzymes operate as part of a
metabolon, or metabolic channel. Although direct interactions between
alkaloid biosynthetic enzymes have not yet been demonstrated, the
existence of enzyme complexes has been detected in flavonoid
biosynthesis (Burbulis and Winkel-Shirley, 1999 The expanding collection of biosynthetic genes should contribute to an
improved understanding of the metabolic mechanisms that regulate
benzylisoquinoline alkaloid biosynthesis. For example, constitutive
expression of the scoulerine 9-O-methyltransferase (SMT) gene in Coptis japonica cell cultures
resulted in 20% higher SMT activity and a small increase in the
accumulation of protoberberine alkaloids (Sato et al., 2001 Relatively little attention has been focused on the relationship
between benzylisoquinoline alkaloid and primary metabolic pathways.
Certainly, an important consideration for any metabolic engineering
strategy is the availability of primary metabolic precursors. The role
of Tyr as the precursor to the benzylisoquinoline alkaloid pathway
suggests that the regulation of the shikimate and aromatic amino acid
pathways might contribute significantly to the ability of California
poppy cell cultures to produce substantial quantities of products, such
as sanguinarine and macarpine. NMR spectroscopy is a technique that can
provide valuable insights into the integration and regulation of plant
metabolism through both in vitro and in vivo measurements. We used
two-dimensional 1H-NMR and noninvasive
15N-NMR to obtain a broad profile of metabolic
alterations caused the antisense mRNA-mediated suppression of
benzophenanthridine alkaloid biosynthesis. Recent applications of
1H-NMR include an analysis of the ligands of
barley root exudates (Fan et al., 1997 We have shown that the transformation of California poppy cell cultures
with antisense BBE or antisense CYP80B1
constructs can suppress the accumulation of benzophenanthridine
alkaloids, which in turn affects cellular growth rate and amino acid
metabolism. Our data provide insight into the complex regulation of
benzylisoquinoline alkaloid biosynthesis and the integral relationship
between amino acid and alkaloid metabolism. Much remains to be learned
about the control of benzylisoquinoline alkaloid pathways before we can
routinely design rational metabolic engineering strategies. A proposed
model suggesting that a plant cell can be engineered to accumulate
valuable benzylisoquinoline alkaloid intermediates using an antisense
mRNA-mediated approach (Kutchan, 1995
PCR Plant genomic DNA was extracted as described by Edwards
et al. (1991) Construction of Transformation Vectors BBE and CYP80B1 coding regions
were re-amplified by PCR using primers designed to add
KpnI and SalI restrictions sites to the
5' and 3' ends, respectively, of each clone. The PCR products were
inserted into the pBI102 binary vector (Facchini et al., 1996 Production of Transgenic Cell Cultures Seeds of California poppy (Richters Herbs, Goodwood, Canada)
were surface-sterilized with 70% (v/v) ethanol for 30 s and 2% (v/v) sodium hypochlorite solution for 10 min, rinsed three times in
sterilized water, and germinated on basal media consisting of B5 salts
and vitamins (Gamborg et al., 1968 RNA Gel-Blot Hybridization Total RNA for gel-blot hybridization analysis was isolated using
the method of Logemann et al. (1987) HPLC Analysis California poppy cell cultures were frozen in liquid
N2, ground to a fine powder using a mortar and pestle, and
extracted with methanol in a boiling water bath for 15 min. Extracts
were reduced to dryness under vacuum, dissolved in 1.0 M
sodium carbonate/bicarbonate (3:2, w/w), pH 10.0, and extracted
three times with ethyl acetate. Pooled ethyl acetate fractions were
reduced to dryness and the residue taken up in 1 mL of methanol.
Extracts were analyzed using a System Gold 126 HPLC and 128 photodiode
array detector (Beckman-Coulter, Mississauga, Canada). Alkaloids were
separated at a flow rate of 0.75 mL min Enzyme Assay Two grams of cultured cells was ground to a fine powder under liquid N2 using a mortar and pestle and extracted in 50 mM Gly-NaOH, pH 8.9. Polyvinylpolypyrrolidone was added to remove phenolic compounds and the extract was desalted on a PD-10 column (Amersham-Pharmacia, Uppsala). Reaction mixtures consisting of 1 mM (S)-reticuline and 500 µL of the enzyme extract were incubated at 30°C for 1 to 2 h. Reactions were stopped by the addition of 10 µL of 1.0 N NaOH, and 100 µL was fractionated on a C18 reverse phase column (4.6 × 250 mm, Ultrasphere, Beckman-Coulter) using methanol:water (1:1, v/v) containing 0.1% (v/v) triethylamine as the mobile phase. The (S)-scoulerine peak was identified by comparison of its UV spectrum and retention time with those of an authentic standard. 15N- and Two-Dimensional 1H-NMR Spectroscopy Whole-cell extracts for two-dimensional 1H-NMR analysis were prepared by grinding 100 mg of lyophilized California poppy cell cultures in three 10-mL aliquots of 80% (v/v) ethanol. After centrifugation of the pooled extracts, the supernatant was reduced to 5 mL under vacuum, de-ionized using 1 mL of Chelex-100, and lyophilized. The sample was dissolved in 100% (v/v) D2O to reduce the resonance for water. Nitrogen-15 NMR spectra were determined using cell cultures grown for 3 d in media containing 99.9% (w/w) K15NO3 rather than the equivalent 14N-containing compound. NMR(NMR) studies were performed with a Bruker AM400 widebore NMR spectrometer equipped with a 10-mm multinuclear broadband probe. NMR spectra were recorded at the following frequencies: 1H (400 MHz) and 15N (40 MHz). For in vivo measurements, cells were perfused in the NMR magnet by recirculating oxygen-saturated media. Amino Acid Analysis Cultured cells (1 g) were lyophylized and ground in 100%
methanol (10:1 [v/w]). The homogenate was incubated at 60°C for 30 min and then centrifuged for 15 min at 12,000g. The
supernatant was collected, and the pellet was extracted once more with
50% (v/v) methanol. The combined extracts were reduced to
dryness and redissolved in 75 µL of dilution buffer containing 100 mM NaHCO3 and 100 mM
H3BO3, pH 8.5. Twenty microliters of the
resuspended solution was mixed with 20 µL of 9-fluorenylmethyl
chloroformate (20.7 mg mL
We thank Dean McIntyre for assistance with the NMR spectroscopy, Ikhlas Khan for the liquid chromatography-mass spectroscopy identification of benzophenanthridine alkaloids, and Fumihiko Sato for the gift of (S)-scoulerine. We also thank Tasmanian Alkaloids Pty Ltd for kindly providing a generous supply of (S)-reticuline.
Received August 14, 2001; returned for revision September 17, 2001; accepted October 26, 2001. 1 This work was supported by the Natural Sciences and Engineering Research Council of Canada (grant to P.J.F.). S.U.P. was the recipient of Bettina-Bahlsen Memorial, Graduate Faculty Council, and J.B. Hyne Graduate Scholarships and a Dean's Special Doctoral Scholarship offered through the University of Calgary.
2 Present address: Agriculture and Agri-Food Canada, Saskatoon Research Centre, 107 Science Place, Saskatoon, Saskatchewan, Canada S7N 0X2.
* Corresponding author; e-mail pfacchin{at}ucalgary.ca; fax 403-289-9311.
Article, publication date, and citation information can be found at www.plantphysiol.org/cgi/doi/10.1104/pp.010741.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ASPB Publications | PLANT PHYSIOLOGY | THE PLANT CELL | |
|---|---|---|---|
TOP
ABSTRACT
INTRODUCTION
-glucuronidase gene. Pathway
intermediates were not detected in any of the transformed cell lines.
The suppression of benzophenanthridine alkaloid biosynthesis using BBE
or CYP80B1 antisense RNA constructs also reduced the growth rate of the
cultures. Two-dimensional 1H-nuclear magnetic resonance and
in vivo 15N-nuclear magnetic resonance spectroscopy showed
no difference in the abundance of carbohydrate metabolites in the
various transgenic cell lines. However, transformed cells with reduced
benzophenanthridine alkaloid levels contained larger cellular pools of
several amino acids including alanine, leucine, phenylalanine,
threonine, and valine compared with controls. The relative abundance of
tyrosine, from which benzophenanthridine alkaloids are derived, was
less than 2-fold higher in antisense-suppressed cells relative to
controls. These results show that alterations in the metabolic flux
through benzophenanthridine alkaloid biosynthesis can affect the
regulation of amino acid pools. These data provide new insight into the
metabolic engineering of benzophenanthridine alkaloid pathways.
-tocopherol (vitamin E) content of Arabidopsis seeds (Shintani and
DellaPenna, 1998
1 protein) and
control (19 ± 7 pkat mg
-aminobutyrate (GABA) in transgenic cells
lines with a reduced accumulation of benzophenenthridine alkaloids
compared with controls (Figs. 
80°C for 24 h.