Flavanone 8-dimethylallyltransferase in Sophora flavescens cell suspension cultures

Article abstract of DOI:10.1016/S0031-9422(00)00198-9

Dimethylallyl diphosphate: naringenin 8-dimethylallyltransferase (EC 2.5.1) was characterized. The enzyme was enantiospecific for (-)-(2S)-naringenin and utilized 3,3-dimethylallyl diphosphate as sole prenyl donor. It required Mg²⁺ (optimum concentration, 10 mM), and has an optimum pH of 9-10. The apparent K(m) values for 3,3-dimethylallyl diphosphate and naringenin were 120 and 36 μM, respectively. The microsomal fraction prenylated several other flavanones at the C-8 position less effectively as compared with naringenin. Interestingly, when 2'-hydroxynaringenin was used as a prenyl acceptor, the 8-lavandulyl (sophoraflavanone G) and the 6-dimethylallyl derivatives were formed, together with the 8-dimethylallyl derivative, leachianone G. These results suggest that the 2'-hydroxy group of naringenin plays an important role for the formation of a lavandulyl group. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0031-9422(00)00198-9

Phytochemistry 54 (2000) 649±655
www.elsevier.com/locate/phytochem
Flavanone 8-dimethylallyltransferase in Sophora ¯avescens cell
suspension cultures
Hirobumi Yamamoto 1, Masayuki Senda, Kenichiro Inoue*
Department of Pharmacognosy, Gifu Pharmaceutical University, 6-1 Mitahora-higashi 5 chome, Gifu 502-8585, Japan
Received 14 February 2000; received in revised form 16 May 2000
Abstract
Dimethylallyl diphosphate: naringenin 8-dimethylallyltransferase (EC 2.5.1) was characterized. The enzyme was enantiospeci®c
for ( )-(2S)-naringenin and utilized 3,3-dimethylallyl diphosphate as sole prenyl donor. It required Mg2+ (optimum
concentration, 10 mM), and has an optimum pH of 9±10. The apparent Km values for 3,3-dimethylallyl diphosphate and
naringenin were 120 and 36 mM, respectively. The microsomal fraction prenylated several other ¯avanones at the C-8 position
less e€ectively as compared with naringenin. Interestingly, when 2'-hydroxynaringenin was used as a prenyl acceptor, the 8-
lavandulyl (sophora¯avanone G) and the 6-dimethylallyl derivatives were formed, together with the 8-dimethylallyl derivative,
leachianone G. These results suggest that the 2'-hydroxy group of naringenin plays an important role for the formation of a
lavandulyl group. 7 2000 Elsevier Science Ltd. All rights reserved.
Keywords: Sophora ¯avescens; Leguminosae; Cell suspension cultures; Biosynthesis; Dimethylallyltransferase; Naringenin; Prenylated ¯avanones;
Sophora¯avanone B; Sophora¯avanone G
1. Introduction
unit, at C-8 or C-6 are structural characteristics com-
mon to these ¯avanones. The production of these pre-
nylated ¯avanones with a 2'-hydroxy group was
stimulated by addition of elicitors, such as Cu2+, glu-
comannans in yeast extracts and pectin (Yamamoto et
al., 1995). Recently, we reported that cork tissues
stimulated production of ¯avanones without any in-
hibitory e€ects on cell growth (Yamamoto et al.,
1996). The studies on the mechanism for this stimu-
lation by cork tissues prompted us to investigate the
enzymes catalyzing the prenylation of ¯avanones. Sev-
eral studies have been conducted on prenyltransferases
of aromatic substrates, such as ¯avonoids of some
types (Dhillon and Brown, 1976; Schroder et al., 1979;
Biggs et al., 1987; Welle and Grisebach, 1991;
La¯amme et al., 1993; Yamamoto et al., 1997), cou-
marins (Hamerski and Matern, 1988), p-hydroxybenzo-
ate (Heide and Tabata, 1987; Muhlenweg et al., 1998)
and olivetolate (Fellermeier and Zenk, 1998). However,
nothing has so far been reported concerning the lavan-
dulylation of ¯avanones. The following two possible
Cultured cells of Sophora ¯avescens Aiton produce
prenylated ¯avanones, such as sophora¯avanone G
(SFG) and lehmannin (7,2',4'-trihydroxy-8-lavandulyl-
¯avanone) (Yamamoto et al., 1991). The cells also con-
tain kushenol F, leachianone
G (LG), 5,7,2',4'-
tetrahydroxy-6-dimethylallyl¯avanone and 7,2',4'-tri-
hydroxy-8-dimethylallyl¯avanone, in extremely small
quantities, as well as lupalbigenin [6,3'-di-(3,3-dimethy-
lallyl)-genistein] which has not yet been isolated from
the original plant (Yamamoto et al., in preparation).
A hydroxy group at C-2' and either a 3,3-dimethylallyl
or a lavandulyl group, an irregular monoterpenoid
* Corresponding author. Tel.: +81-58-237-3931; fax: +81-58-237-
5979.
E-mail address: inoueken@gifu-pu.ac.jp (K. Inoue).
1 Present address: Medicinal Plant Garden, School of Pharmaceuti-
cal Sciences, Nagasaki University, 1-14 Bunkyo-mahi, 852-8521
Nagasaki, Japan.
0031-9422/00/$ - see front matter 7 2000 Elsevier Science Ltd. All rights reserved.
PII: S0031-9422(00)00198-9

650
H. Yamamoto et al. / Phytochemistry 54 (2000) 649±655
mechanisms for the formation of a lavandulyl group
are presumed: one is the direct lavandulylation of the
¯avanone skeleton and the other is the two-step
dimethylallylation, the ®rst to the ¯avanone skeleton
and the second to the resultant prenyl side chain.
In order to clarify the mechanism for the attachment
of a lavandulyl moiety onto ¯avanone skeleton, the
prenyltransferases in S. ¯avescens cell suspension cul-
tures were investigated and the results are discussed in
this paper.
DMAN) by comparison of its EIMS and 1H NMR
spectra with the data reported previously (Shirataki et
al., 1988). Interestingly, 8-DMAN obtained, whose ab-
solute con®guration was determined to be 2S from its
optical rotation, ꢀ‰aŠ
288; Lit. 268 both in MeOH)
D
(Shirataki et al., 1988). These results demonstrate that
S. ¯avescens cultured cells contain dimethylallyl dipho-
sphate: naringenin 8-dimethylallyltransferase, which is
stereospeci®c for (2S)-naringenin.
2.2. Properties of prenyltransferase
2. Results and discussion
The optimum pH of the enzymatic reaction was
about 9±10 in Tris±HCl bu€er and 10 in Glycine±
NaOH bu€er. The following experiments were carried
out at pH 9, because, in alkaline solution, naringenin
was liable to be converted into the corresponding chal-
cone. The enzyme required a divalent metal ion for the
reaction and Mg2+ was most e€ective for activity.
Mn2+ was also e€ective, though to a lesser extent
(32% of Mg2+), but addition of Ca2+ (6%), Co2+
(5%), Zn2+ (2%) or Cu2+ (1%) gave only very low
activities (data not shown). Saturation for Mg2+ was
reached at 10 mM. Optimal substrate concentrations
were investigated using the microsomal fraction. Maxi-
mal conversion rates were observed at 10 mM of
DMAPP 1 mM of naringenin, respectively and appar-
ent Kmvalues for DMAPP and naringenin were deter-
mined to be 120 and 36 mM, respectively, from the
Lineweaver-Burk plot (Fig. 1, insert).
2.1. Detection of naringenin 8-prenyltransferase activity
In examining the ¯avanone prenyltransferases in S.
¯avescens cells, naringenin and 2'-hydroxynaringenin
were assumed to serve as prenyl acceptors, and 3,3-
dimethylallyl diphosphate (DMAPP), isopentenyl
diphosphate (IPP) and lavandulyl diphosphate (LPP)
as prenyl donors. Taking into account the occurrence
of leachianone
G (LG) and 7,2',4'-trihydroxy-8-
dimethylallyl¯avanone in Sophora cells, we chose
DMAPP as a prenyl donor in the present study. We
also adopted naringenin as a prenyl acceptor because
the presence of hydroxy groups on ¯avonoid skeletons
was reported not to be essential for the recognition of
substrates by prenyltransferases (Schroder et al., 1979;
La¯amme et al., 1993; Biggs et al., 1987; Yamamoto et
al., 1996). Thus, the crude cell-free extract of S. ¯aves-
cens cultured cells was incubated with l mM narin-
genin, 2 mM DMAPP and 10 mM MgCl2 for 2 h at
258C. After termination of the reaction with dil. HCl,
the phenolic compounds extracted with EtOAc were
analyzed by a reversed phase HPLC-photodiode-array
system. A new peak whose UV absorption pattern was
in complete agreement with that of 8-(3,3-dimethylal-
lyl)-naringenin (sophora¯avanone B; 8-DMAN) was
detected on the chromatogram. In the fractionation of
the crude cell-free extract by the ultracentrifugation,
more than 85% of the total prenylation activity was
found in the twice-washed 100,000 Âg pellets, indicat-
ing that the prenyltransferase was tightly bound to the
membrane fraction of the cells (data not shown). The
time course of the enzymatic reaction was linear for at
least 180 min. After storage of the enzyme solution at
858C for 2 weeks, more than 90% of the activity was
retained (data not shown). To con®rm the structure of
the prenylated product, the microsomal fraction
obtained from 300 g of the cultured cells was incu-
bated with DMAPP, Mg2+ and commercially avail-
able naringenin which was optically inactive.
Puri®cation of the EtOAc extracts of the reaction mix-
ture by prep. HPLC gave the product (14 mg), which
was identi®ed as 8-(3,3-dimethylallyl)-naringenin (8-
2.3. Substrate speci®city for prenyl diphosphate
Incubation of the microsomal fraction with Mg2+
,
naringenin and IPP at pH 9 gave 8-dimethylallylnarin-
genin (8-DMAN) at a yield of only 1.3% of that
obtained when DMAPP was used as prenyl donor. In-
cubation of the cell-free extracts or the microsomal
fraction with LPP gave no prenylated product (data
not shown). Accordingly, at pH 9, this prenyltransfer-
ase utilizes only DMAPP as the prenyl donor and
introduces one dimethylallyl group solely into C-8 of
naringenin.
2.4. Substrate speci®city for ¯avanones
The prenylation activities of the twice-washed micro-
somal fraction for several ¯avanones were also investi-
gated
(Table
1).
When
liquiritigenin
(5,7-
dihydroxy¯avanone), hesperetin (5,7,3'-trihydroxy-4'-
methoxy¯avanone) and taxifolin (3,5,7,3',4'-pentahy-
droxylfavanone) were used as substrates, the micro-
somal fraction catalyzed the transfer of
a 3,3-
dimethylallyl group to their C-8 positions less e€ec-
tively compared with the case of naringenin, suggesting
that the presence of hydroxy groups at 3-, 5- and 3'-

H. Yamamoto et al. / Phytochemistry 54 (2000) 649±655
651
Fig. 1. Dependency of naringenin 8-dimethylallyltransferase activity on the concentration of dimethylallyl diphosphate and naringenin measured
with a microsomal fraction of S. ¯avescens cultured cells. Insert: Lineweaver-Burk plot with varying concentrations (0.01±3.0 mM) of dimethylal-
lyl diphosphate and varying concentrations (0.01±1.0 mM) of naringenin to calculate the apparent Km value for DMAPP and naringenin, respect-
ively.
positions were not essential for substrate recognition.
When 2'-hydroxynaringenin was used, 8-dimethylallyl
derivative, LG, was detected only in a very small
quantity (3% of the yield of the corresponding deriva-
tive formed from naringenin). However, 8-lavandulyl-
2'-hydroxynaringenin (SFG), along with the 6-
dimethylallyl derivative (6-DMAHN), were produced
four times more e€ectively than LG. This observation
was quite di€erent from that with the other ¯avanones
mentioned above.
2.5. Substrate speci®city for ¯avonoids other than
¯avanones
The speci®city of the prenyltransferase for prenyl
acceptors was examined with several ¯avonoids at 0.3
Table 1
Prenylation of ¯avonoids by microsomal fraction from S. ¯avescens cultured cells
Substrate (0.3 mM)
Relative activity (%)
8-Dimethylallyl
8-Lavandulyl
6-Dimethylallyl
Flavanones
Naringenin
Liquiritigenin
Hesperetin
b
100.0a
35.3
57.1
42.2
3.1
±
±
±
±
±
±
Taxifolin
±
12.8
±
16.1
2'Hydroxynaringenin
Other ¯avonoids
Kaempferol
Quercetin
16.5
18.7
1.8
±
±
±
±
±
±
±
±
Apigenin
Genistein
Maackiain
±
80.6
11.6c
±
a Naringenin 8-prenyltransferase activity was 5.94 nkat/mg protein.
b Not detected.
c C-6 and C-3' were prenylated.

652
H. Yamamoto et al. / Phytochemistry 54 (2000) 649±655
mM concentration and pH 9 (Table 1). In this exper-
iment, the microsomal fraction washed twice with
Tris±HCl bu€er were used to remove the di€erent bio-
transformation of the substrates and any further
metabolism of the products. The formation of preny-
lated ¯avonoids was monitored using a HPLC-photo-
diode-array system. Not only naringenin, but also
ase, such as naringenin and kaempferol 8-
prenyltransferases.
2.6. Activity for the formation of lavandulyl group in
microsomal fraction
In the above-mentioned experiments, we adjusted
the pH of the reaction mixture to 9 because it was the
optimum for naringenin 8-prenyltransferase. However,
it is reported that the optimum pH of many prenyl-
transferases from a variety of sources were around 7.5
(Dhillon and Brown, 1976; Heide and Tabata, 1987;
La¯amme et al., 1993; Schroder et al., 1979; Biggs et
al., 1987; Welle and Grisebach, 1991; Hamerski and
Matern, 1988; Fellermeier and Zenk, 1998; Muhlenweg
et al., 1998). Thus, we re-examined the prenylation ac-
tivity for naringenin and 2'-hydroxynaringenin at pH
7.5. As shown in Fig. 2, when naringenin was used as
the prenyl acceptor, only 8-DMAN was formed
whereas when 2'-hydroxynaringenin was used, three
products, LG, 6-DMAHN and SFG, were obtained.
These observations, at pH 7.5, were similar to those at
pH 9 except for the ratio of products. The total
amounts of the prenylated derivatives formed from 2'-
kaempferol
(5,7,4'-trihydroxy¯avonol),
quercetin
(5,7,3',4'-tetrahydroxy¯avonol), apigenin (5,7,4'-trihy-
droxy¯avone) were prenylated at their C-8 position.
Genistein (5,7,4'-trihydroxyiso¯avone) was also preny-
lated to a€ord wighteone (6-dimethylallylgenistein)
and lupalbigenin [6,3'-di(dimethylallyl)-genistein], but
maackiain, a pterocarpan was not prenylated. Taking
into account that the microsomal fraction used in this
experiment recognized the stereochemistry of C-2 of
naringenin rigidly as mentioned above and the pre-
vious observation that kaempferol 8-prenyltransferase
obtained from Epimedium diphyllum cultured cells did
not accept naringenin as substrate (Yamamoto et al.,
1997), it seems to be unlikely that naringenin 8-prenyl-
transferase catalyzes the prenylation reaction for all
¯avonoids examined. It is probable that the micro-
somal fraction contains more than one prenyltransfer-
Fig. 2. Formation of 8-dimethylallyl-, 8-lavandulyl- and 6-dimethylallyl ¯avanones by a microsomal fraction of S. ¯avescens cultured cells. 0.3
mM of naringenin or 2'-hydroxynaringenin were incubated with the enzyme at pH 7.5 or 9 in the presence of 10 mM Mg2+ and 2 mM
DMAPP.

H. Yamamoto et al. / Phytochemistry 54 (2000) 649±655
653
hydroxynaringenin at pH 7.5 were almost equal to
those at pH 9. However, at pH 7.5, the amount of the
8-prenylated (8-dimethylallyl and 8-lavandulyl) deriva-
tives was superior to that of the 6-dimethylallyl deriva-
tive. On the contrary, at pH 9, the latter was
predominantly formed. The possibility of the direct
transfer of the lavandulyl group to 2'-hydroxynarin-
genin also was re-examined at pH 7.5 using LPP as
the prenyl donor, but the 8-lavandulyl derivative was
not detected (data not shown). These ®ndings suggest
that a lavandulyl group at C-8 of naringenin is formed
by two-step dimethylallylations and that subsequent
dimethylallylation at the prenyl group of C-8 and
direct dimethylallylation at C-6 of ¯avanone skeleton
requires hydroxylation at C-2'. Thus, the activities of
hydroxylation at C-2' of naringenin and 8-DMAN in
the microsomal fraction of S. ¯avescens cultured cells
were tentatively investigated using NADPH, NADH,
FAD and FMN as cofactor. The hydroxylation ac-
tivity of 8-DMAN was observed (data not shown), in-
dicating the presence of a route from naringenin via 8-
DMAN (naringenin±8-DMAN±LG±SFG) for the bio-
synthesis of sophora¯avanone G. However, the activity
for naringenin was not detected although the micro-
somal fraction had the activities of prenylations at C-8
or C-6 of 2'-hydroxynaringenin. Nevertheless, it seems
to be likely that the activity of hydroxylation at C-2'
of naringenin is present in S. ¯avescens cells, although
very low, because of the low content of 6-prenylated
2'-hydroxynaringenin in S. ¯avescens cell cultures
(Yamamoto et al., in preparation) and of the neccesity
for the hydroxy group at C-2' for prenylation at C-6.
The possible biosynthetic pathways from naringenin to
sophora¯avanone
G and the enzymatic reactions
detected in the present paper were summarized in
Scheme 1. Detailed experiments concerning the
hydroxylation at C-2' in ¯avanones are now in
progress.
3. Experimental
3.1. Chemicals
Dimethylallyl diphosphate and lavandulyl dipho-
sphate were synthesized according to the method of
Cornforth and Popjak (1969). 2'-Hydroxynaringenin
was synthesized according to the method of He et al.
(1988). Isopentenyl diphosphate was a kind gift from
Dr. K. Yazaki, Graduate School of Agriculture,
Kyoto University. Liquiritigenin was a kind gift from
Dr. H. Hayashi, Niigata College of Pharmacy. Narin-
genin, hesperetin, taxifolin, kaempferol, apigenin, quer-
cetin, and genistein were purchased from Funakoshi,
Japan. Maackiain was isolated from cultured cells of
S. ¯avescens (Yamamoto et al., 1991).
Scheme 1. Possible biosynthetic pathways leading to sophora¯avanone G in S. ¯avescens cultured cells.

654
H. Yamamoto et al. / Phytochemistry 54 (2000) 649±655
3.2. Plant material and culture method
3.7. Isolation and identi®cation of the reaction product
The origin and subculturing of callus cultures
(Yamamoto et al., 1991) and the establishment of cell
suspension cultures (Yamamoto et al., 1996) of S. ¯a-
vescens were described in previous papers.
The microsomal fraction obtained from 300 g of
fresh cells was incubated with 400 mmol naringenin,
800 mmol DMAPP and 4 mmol MgCl2 in 100 mM
Tris±HCl bu€er (pH 9, total vol. 400 ml) at 258C for 4
h. The reaction was terminated by the addition of 40
ml of 6 M HCl, and the products were extracted with
EtOAc ꢀ3 Â 200 ml). The organic layers were com-
bined and concd. in vacuo. The residue was dissolved
in MeOH and puri®ed by use of prep. HPLC using the
following conditions: column, Hikarisil C18-2E ꢀ20 Â
250 mm, Asahi Chemical, Japan); solvent, 45%
CH3CN/H2O; ¯ow rate, 9.5 ml/min; oven temp., 408C;
detection, 294 nm. The fraction around Rt 49 min was
collected and evaporated in vacuo to give 14 mg of
sophora¯avanone B, identical with the authentic
3.3. Enzyme preparation
For the enzyme preparation, all procedures were
carried out at 48C. S. ¯avescens cells (5 g) cultured for
8±10 days, were homogenized in 100 mM K±Pi bu€er
(pH 6.5) containing 10 mM DTT and 0.5 g PVPP
using a te¯on homogenizer. The homogenate was cen-
trifuged at 12,000 Âg for 20 min, then the supernatant
was passed through a Sephadex PD-10 (Pharmacia)
column equilibrated with 100 mM Tris±HCl bu€er
(pH 9) containing 10 mM DTT, and the eluate was
used as the crude enzyme preparation. For the prep-
aration of the microsomal fraction, the 12,000 Âg
supernatant was centrifuged at 100,000 Âg for 20 min.
The pellet was washed twice with the same bu€er and
then resuspended in 5 ml of the same bu€er.
1
sample in the MS and H NMR spectra. ‰aŠD25 288
ꢀc ˆ 1:0, MeOH).
Acknowledgements
We thank Dr. K. Yazaki, Lab. Molecular and Cellu-
lar Biology of Totipotency, Div. Integrated Life
Sciences, Graduate School of Biostudies, Kyoto Uni-
versity, for a generous gift of isopentenyl diphosphate
and the H NMR data of dimethylallyl diphosphate.
We also thank Dr. H. Hayashi, Niigata College of
Pharmacy, for a kind gift of liquiritigenin.
3.4. Protein content
Protein contents were determined according to the
method of Bradford (1976).
1
3.5. Enzyme reaction
The incubation mixture contained 0.5 mmol narin-
genin, 1 mmol DMAPP, 5 mmol MgCl2, and enzyme
extract, 200 ml (ca. 0.35 mg protein) in a total vol. of
500 ml. DMAPP and MgCl2 were dissolved in H2O, re-
spectively, and naringenin was dissolved in EtOH. The
reaction was initiated by the addition of naringenin to
the mixture and, after the incubation for 60 min at
258C, terminated by the addition of 50 ml of 6 M HCl.
The reaction mixture was extracted with 200 ml of
EtOAc containing 0.1 mg/ml naphthalene as an in-
ternal standard. The EtOAc extract was analyzed by
HPLC.
References
Biggs, D.R., Welle, R., Visser, F.R., Grisebach, H., 1987.
Dimethylallylphosphate:3,9-dihydroxypterocarpan 10-dimethylal-
lyltransferase from Phaseolus vulgaris. Identi®cation of the reac-
tion product and properties of the enzyme. FEBS Letters 220 (1),
223±226.
Bradford, M.M., 1976. A rapid and sensitive method for the quanti-
tation of microgram quantities of protein utilizing the principle of
protein-dye binding. Analytical Biochemistry 72 (1), 248±254.
Cornforth, R.H., Popjak, G., 1969. Chemical synthesis of substrates
of sterol biosynthesis. Methods in Enzymology 15, 359±390.
Dhillon, D.S., Brown, S.A., 1976. Localization, puri®cation, and
characterization of dimethylallylpyrophosphate: umbelliferone
dimethylallyltransferase from Ruta graveolens. Archives of
Biochemistry and Biophysics 177 (1±2), 74±83.
3.6. HPLC analysis
The amount of 8-prenylnaringenin was determined
by HPLC using a Capcellpak C18 AG-120 column
ꢀ4:6 Â 250 mm, Shiseido, Japan) in an oven at 408C,
with CH3CN/H2O linear gradient solvent system con-
taining 1% AcOH, from 40 to 58% CH3CN in 24
min, at a ¯ow rate of 0.9 ml/min, monitoring the
absorption at 294 nm or using a photodiode-array
SPDM-6A system (Shimadzu). The quantities were cal-
culated from the peak area at 294 nm recorded by
Chromatopac C-R4A (Shimadzu).
Fellermeier, M., Zenk, M.H., 1998. Prenylation of olivetolate by a
hemp transferase yields cannabigerolic acid, the precursor of tet-
rahydrocannabinol. FEBS Letters 427 (2), 283±285.
Hamerski, D., Matern, U., 1988. Elicitor-induced biosynthesis of
psoralens in Ammi majus L. suspension cultures. Microsomal con-
version of demethylsuberosin into (+)-marmesin and psoralen.
European Journal of Biochemistry 171 (2), 369±375.
He, X., Yang, F., Lei, X., Chen, J., Min, Y., 1988. Synthesis of B-
ring substituted 5,7-dihydroxy¯avanones and their e€ects on gui-
nea pig hearts. Yiyao Gongye 19 (10), 447±451.
Heide, L., Tababta, M., 1987. Geranylpyrophosphte: p-hydroxy-

H. Yamamoto et al. / Phytochemistry 54 (2000) 649±655
655
benzoate geranyltransferase activity in extracts of Lithospermum
erythorhizon cell cultures. Phytochemistry 26 (6), 1651±1655.
La¯amme, P., Khouri, H., Gulick, P., Ibrahim, R., 1993. Enzymatic
prenylation of iso¯avonoids in white lupin. Phytochemistry 34
(1), 147±151.
prenyltransferase of iso¯avonoid phytoalexin biosynthesis in soy-
bean. Phytochemistry 30 (2), 479±484.
Yamamoto, H., Kawai, S., Mayumi, J., Tanaka, T., Iinuma, M.,
Mizuno, M., 1991. Prenylated ¯avanone production in callus cul-
tures of Sorphora ¯avescens var. angustifolia. Zeitschrift fur
Naturforschung 46c (3±4), 172±176.
Muhlenweg, A., Melzer, M., Li, S.-M., Heide, L., 1998. 4-
Hydroxybenzoate 3-geranyltransferase from Lithospermum ery-
throrhizon: puri®cation of a plant membrane-bound prenyltrans-
ferase. Planta 205 (3), 407±413.
Yamamoto, H., Ichimura, M., Inoue, K., 1995. Stimulation of pre-
nylated ¯avanone production by mannans and acidic polysac-
charides in callus culture of Sophora ¯avescens. Phytochemistry
40 (1), 77±81.
Schroder, G., Zahringer, U., Heller, W., Ebel, J., Grisebach, H.,
1979. Biosynthesis of antifungal iso¯avovoids in Lupinus albus.
Enzymatic prenylation of genistein and 2'-hydroxygenistein.
Archives of Biochemistry and Biophysics 194 (2), 635±636.
Shirataki, Y., Yokoe, I, Noguchi, M., Tomimori, T., Komatsu, M.,
1988. Studies on the constituents of Sophora species. XXII.
Constituents of the root of Sophora moorcroftiang Benth. ex
Baker. (1). Chemical Pharmaceutical Bulletin 36 (6), 2220±2225.
Welle, R., Grisebach, H., 1991. Properties and solubilization of the
Yamamoto, H., Yamaguchi, M., Inoue, K., 1996. Absorption and
increase in the production of prenylated ¯avanones in Sophora
¯avescens cell suspension cultures by cork pieces. Phytochemistry
43, 603±608.
Yamamoto, H., Kimata, J., Senda, M., Inoue, K., 1997.
Dimethylallyl diphosphate: kaempferol 8-dimethylallyltransferase
in Epimedium diphyllum cell suspension cultures. Phytochemistry
44 (1), 23±28.