Attachment of IR-active organometallic probe groups to flavonoid inducers of nod gene expression

Article abstract of DOI:10.1016/S0022-328X(99)00398-8

Direct reaction of flavanones with electrophilic tricarbonyl[(1,2,3,4,5-η)cyclohexadienyliron(1+) complexes affords organometallic derivatives for use as probes of molecular recognition events during the induction of nodulation genes in Rhizobium leguminosarum. A more selective preparation method has been developed by deprotonation of the flavanone prior to the addition of the cyclohexadienyliron complex.

Full text of DOI:10.1016/S0022-328X(99)00398-8

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 589 (1999) 103–110
Attachment of IR-active organometallic probe groups to flavonoid
inducers of nod gene expression
Andrej V. Malkov a, Ljubica Mojovic a, G. Richard Stephenson a,*,
Andrew T. Turner a, Colin S. Creaser b,1
a School of Chemical Sciences, Uni6ersity of East Anglia, Norwich NR4 7TJ, UK
b Department of Chemistry and Physics, Nottingham Trent Uni6ersity, Clifton Lane, Nottingham NG11 8NS, UK
Received 14 May 1999
Abstract
Direct reaction of flavanones with electrophilic tricarbonyl[(1,2,3,4,5-h)cyclohexadienyliron(1+) complexes affords
organometallic derivatives for use as probes of molecular recognition events during the induction of nodulation genes in
Rhizobium leguminosarum. A more selective preparation method has been developed by deprotonation of the flavanone prior to
the addition of the cyclohexadienyliron complex. © 1999 Elsevier Science S.A. All rights reserved.
Keywords: Organometalcarbonyl probes; Flavonoids; FTIR; Rhizobium; Root nodulation
study the binding process in solution by exploiting the
environment-induced shifts in the vibrational modes in
IR spectra. This frequency-shift information can give a
qualitative guide to the polarity of the environment of
the probe, and (by polyclonal antibody analysis of
normalised antisymmetric vibrational bands [2]) the
contribution of several factors to the overall polarity
can be estimated. With more specialised probes, local
pH [7], or the p-stacking of aromatic rings [8], can be
assessed. Furthermore, we are able to detect small
quantities [3] of a protein-bound organometalcarbonyl
probe in the presence of a substantial excess of the
unbound probe molecule.
In this paper we present results from a programme of
combined synthetic and biological studies in which we
will apply these methods to the examination of the
molecular details of the control of nod gene induction
in Rhizobium leguminosarum, in the early stages of the
processes that lead to root nodulation and nitrogen
fixation in legumes. Detailed genetic investigations have
established [9] that the Rhizobium gene product NodD
is essential for chemical signalling between legume roots
and the bacteria, and that flavonoid molecules are
active as inducers of nod gene expression. This has led
to the proposal that the binding of plant-derived
flavonoids to NodD is a key step in the signalling
1. Introduction
Organometalcarbonyl probe groups offer new oppor-
tunities to explore the mode of action of biological
systems. Two recent developments have opened the way
for the general application of these methods. In Paris,
the Jaouen research group has established procedures
for the carbonylmetalloimunoassay (CMIA) method
[1], which exploits the exceptionally low detection limits
(by FT-IR spectroscopy) of vibrational modes of car-
bonylmetal complexes, while in Norwich, the interpre-
tation of shifts in the vibrational frequencies of these
bands in solution has been studied [2], and methods for
the detection of protein-bound organometalcarbonyl
complexes have been demonstrated [3]. There are now
several examples of the attachment of organometalcar-
bonyl probes to bioactive peptides [4] and oligonucle-
otides [5], following the initial investigations of the
organometallic oestradiol derivatives some years ago by
Jaouen [6].
In Norwich, our main objective is to deliver the
organometalcarbonyl bioprobe to its protein binding
site by attachment to a bioactive substance, and to
* Corresponding author.
1 Also corresponding author.
0022-328X/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(99)00398-8

104
A.V. Malko6 et al. / Journal of Organometallic Chemistry 589 (1999) 103–110
ment, the approach is attractive provided the isomeric
derivatives can be separated and identified.
2. Results and discussion
As a model to test the feasibility of this approach,
2,4-dihydroxyacetophenone (1) was chosen (Scheme 1)
for use in reactions with the cationic h5-cyclohexadi-
enyliron complexes 2 and 3. In our first experiments, a
1:1 stoichiometry of nucleophile and electrophile was
employed in acetonitrile in the presence of triethyl-
amine as a basic catalyst. The two expected products 4
and 5 (Table 1, entries 1 and 3) were obtained, but in
poor yield. Repetition with an excess of salt 2 (two
equivalents) improved the yield of 4 to 54% (entry 2).
In the nucleophile 1, the activating effect of the two
phenolic substituents on the aromatic ring is offset by
the deactivating influence of the carbonyl group of the
methyl ketone. Despite this potential problem, through
the use of an excess of the electrophile, successful
alkylation of the aromatic ring was possible. Compared
to 2,4-dihydroxyacetophone, the A ring of the commer-
cially available flavanone naringenin 6 should be more
reactive, since an additional oxygen substituent (the
oxygen link in ring C) is present. The B ring is also
activated as a nucleophile by an OH group, but it was
hoped that the high level of oxygenation of ring A
would lead to selective alkylation at this position.
Based on the model study, two equivalents of the dienyl
complexes were employed (Table 1, entries 4 and 6).
Organometallic flavonoid derivatives were successfully
obtained by this procedure (32 and 46% yield) but
examination of their NMR spectra indicated that in
both cases, no A ring protons were present in the
region 5.7–5.9 ppm. This suggested that both nucleo-
philic sites on the A ring had been alkylated (Scheme
2). This possibility was confirmed by FAB mass spec-
trometry, which indicated MH+ ions with masses 709
and 769, corresponding to the bimetallic products 11
and 12, respectively. After careful chromatographic
purification, the identity of these products was eventu-
ally confirmed by microanalysis in the case of 11 and
high-resolution mass spectrometry in the case of 12.
From these results, it was clear that the additional
activation at ring A in 6 was sufficient to make control
Fig. 1. Examples of flavanones with nodulation gene induction activ-
ity: sites for the attachment of organometalcarbonyl spectroscopic
probes.
process. By delivery of a flavonoid-bound organometal-
carbonyl probe to the NodD receptor site, we plan first
to prove that binding occurs, and then examine qualita-
tively the local environment of the binding pocket by
interpretation of the environment-induced frequency
shifts. These spectroscopic experiments await the com-
pletion of model studies with crystallographically char-
acterised protein receptors and other guest molecules.
We report here, however, the results of our first at-
tempts to synthesise organometallic derivatives of
flavonoids, which have made available a representative
set of compounds for biological evaluation. For suc-
cess, general methods are needed for the attachment
organometallic complexes to flavonoids, and the result-
ing derivatives must retain some measure of nod gene
induction activity, so demonstrating the continued ca-
pacity to enter protein binding site despite the bulk of
the attached group. Part of this work has been the
subject of a preliminary communication [10].
At the start of these investigations, no examples of
organometallic flavonoid derivatives had been reported.
Our studies require the attachment of the probe to each
of the A, B and C rings (Fig. 1) of a series of biologi-
cally active flavonoids. For ring A derivatives, a direct
addition approach is possible, since the oxygenation
needed here for gene induction activity should also
render the ring nucleophilic, permitting direct derivati-
sation by reaction with electrophilic h5-cyclohexadi-
enyliron complexes. This would be expected to give rise
to mixtures of regioisomers, but since at this stage we
require only sufficient product for the bio-assay experi-
Scheme 1.

A.V. Malko6 et al. / Journal of Organometallic Chemistry 589 (1999) 103–110
105
Table 1
Product distribution in the reaction of 2,4-dihydroxyacetophenone (1), naringenin (6), and hesperitin (19) with organoiron complexes 2 and 3
Entry
Nucleophile
Electrophile
Struct.
Products
C-6 a
Subst. R
Equiv.
C-8 a
C-6,8 a
Struct.
Struct.
Yield, %
Struct.
Yield, %
Yield, %
1
2
3
4
1
1
1
6
6
6
6
19
19
19
19
6 c
2
2
3
2
2
3
3
2
2
3
3
2
H
H
OCH3
H
H
OCH3
OCH3
H
1
2
1
2
1
2
1
2
1
2
1
1
4
4
5
–
7
–
8
–
13
–
14
7
19
54
28
–
14
–
3
–
11
–
15
88
–
–
–
–
9
–
10
–
15
–
–
–
–
–
2
–
–
–
–
–
–
–
11
11
12
12
17
17
18
18
–
32
13
46
4
28
5
25
9
–
5 b
6
7 b
8
Trace
–
5
–
3
–
9
H
10
11
12
OCH3
OCH3
H
16
–
a Numbering system for flavanones, in the case of 2,4-dihydroxyacetophenone the organoiron complex was attached to C-3 of the aromatic ring.
b Starting flavanone was recovered: entry 5; 1–4%, entry 7; 27–35%.
c NaH before addition of 2.
favouring monoalkylation difficult in the presence of an
excess of the electrophile, but that, as expected, ring A
was far more reactive than ring B. Alkylation of 6 was
repeated in each case, with one equivalent of the elec-
trophile. In the case of the electrophile 2, two major
iron-containing products were obtained under these
revised conditions. Attempts at chromatographic sepa-
ration of these two materials revealed a third, minor
metal-containing component in the product mixture. In
this way, in addition to the double adduct 11, two
further organometallic flavonoid derivatives (7 and 9)
were obtained. The minor product 9, however, was
isolated in only 2% yield. Overall, the combined yield of
the organometallic flavonoid derivatives (29%) was
comparable to the yield of the double adduct when an
excess of the electrophile was employed. Turning to salt
3 (entry 7), alkylation of the flavanone with a 1:1 ratio
of reactants, proved much less efficient. The methoxy-
substituted cyclohexadienyl complex 3 is known [11] to
be significantly less reactive than its unsubstituted
counterpart 2. Despite this difficulty, sufficient material
for biological testing could be obtained, even in this
case. At this stage, attention turned to the flavonoid
hesperetin 19, which was prepared by hydrolysis of the
commercially available hesperedin [12]. Despite the ad-
ditional oxygenation on ring B, selective reaction with
the dienyl complexes 2 and 3 at the A ring was still
observed, and similar results were obtained. With two
equivalents of the electrophiles, (Table 1, entries 8 and
10) the double adducts 17 and 18 were the only metal-
containing products identified. Under the 1:1 protocol
(entries 9 and 11) mixtures of 13, 15, 17 and 14, 16, 18
were produced in a fashion analogous to the results
obtained with naringenin, except that the reactions with
the less electrophilic salt 3 worked better in this case. In
both entries 9 and 11, significantly less double alkyla-
tion was observed.
These results demonstrate that the flavanone nucleo-
philes naringenin 6 and hesperetin 19 can be selectively
alkylated by tricarbonyliron complexes at ring A, and
that both nucleophilic sites on the A ring are capable of
participating in this reaction, though to different ex-
tents. In each case, one of the mono adducts was
formed in considerably larger amounts than the other.
Since these reactions are expected to be irreversible
under these conditions, this reflects a kinetic selectivity
between C-6 and C-8. Because reaction of the prochiral
salt 2 and the racemic salt 3 with the chiral flavanones
naringenin and hesperetin produces inseparable
diastereoisomeric products2, the NMR spectra of the
adducts, particularly the double adducts, were compli-
cated by the presence of overlapping signals, and in
many cases the characteristic coupling patterns associ-
ated with 3¦-H of 2¦-OMe-cyclohexadieneiron com-
plexes (coupling of 3¦-H to 1¦-H and 4¦-H) and 2-H of
the flavanone ring system (coupling of 2-H to 3a-H and
3b-H) were obscured. However, in each of the mono-
adducts 7–10 and 13–16, the remaining A-ring hydro-
gens appeared as the expected singlets at 5.9 or 5.7
ppm. In the case of 7, two-dimensional NMR studies
were undertaken to identify the site of alkylation and
on this basis, the structures of all four kinetically
favoured monoadducts were assigned as the result of
reaction at C-6 of flavanone.
2 Biological testing was performed (see ref. [10]) on single regioiso-
mers as mixtures of diastereoisomers.

106
A.V. Malko6 et al. / Journal of Organometallic Chemistry 589 (1999) 103–110
To improve the efficiency of production of the
ethylamine as catalyst and an excess of the electrophile,
double adducts can also be selectively obtained. The 1:1
stoichiometry, however, is the only way to prepare the
C-8 regioisomer, but difficult and time-consuming chro-
matography is necessary to obtain a pure sample. The
results of biological evaluation in an established screen
for nodulation gene induction have been reported else-
where [10], and show that the derivative 8 exhibits
significant residual biological activity (5109160 units
at 20 mM, compared to 4500 units for eriodictiol (Fig.
1)). The derivative 8, however, was inactive at 2 mM,
while eriodictiol retains its activity at this concentra-
tion. None of the double adducts 11, 12, 17 or 18 was
active even at 20 mM. The presence of a single
organoiron-bound ring, of the correct type and in the
correct situation, appears still to permit some degree of
interaction of the product with its natural biological
binding site, but attachment of two additional rings
prevents binding.
monoadducts, complete deprotonation of the phenolic
OH groups with sodium hydride, prior to the addition
of the electrophilic dienyl complex, was attempted. In
this way, starting from naringenin 6, efficient and selec-
tive production of the kinetically favoured monoadduct
7 was achieved in 88% yield (Scheme 2). It seems
probable that efficient formation of products 8, 13 and
14 in this way would also be possible, but since the
non-selective triethylamine-catalysed process had pro-
vided sufficient of each of these compounds for biolog-
ical evaluation, these reactions were not repeated using
the sodium hydride procedure. Furthermore, the only
route available to the minor C-8 adducts, 9, 10, 15 and
16 was through the use of the triethylamine-catalysed
procedure on a relatively large scale, followed by exten-
sive chromatographic purification, so the major prod-
ucts of the reactions were available at this stage in
relatively large amounts. In the course of this work, a
complete set of double adducts, 11, 12, 17 and 18
were also obtained, so all three possible substitution
patterns at the A ring were made available for biologi-
cal evaluation.
4. Experimental
All reactions were performed under an atmosphere of
dry, oxygen-free nitrogen in oven-dried glassware twice
evacuated and filled with nitrogen. All solvents for the
reactions were of reagent grade and were dried and
distilled immediately before use as follows: tetrahydro-
furan (THF) from sodium–benzophenone; acetonitrile
from calcium hydride. Column chromatography was
3. Conclusions
For efficient monosubstitution at C-6, the use of
sodium hydride and one equivalent of the cyclohexadi-
enyl complex is the preferred method, while with tri-
Scheme 2. Direct attachment of tricarbonyliron complexes to the A ring of flavanones.

A.V. Malko6 et al. / Journal of Organometallic Chemistry 589 (1999) 103–110
107
performed using Merck silica gel. Low-resolution EI
mass spectrometry (Kratos MS25) and elemental analy-
ses were performed at the University of East Anglia by
A.W.R. Saunders. Other mass spectra were measured at
the EPSRC National Mass Spectrometry Service Centre,
Swansea. IR spectra were recorded on a Perkin–Elmer
1720X FT-IR spectrometer. NMR spectra were recorded
on a Jeol GX400 spectrometer.
2×1H; 2,3-H), 5.69 (s, 1H, C%6-OH), 6.30 (d, J 9 Hz,
1H, 5%-H), 7.50 (d, J 9 Hz, 1H, 4%-H), 13.20 (s, 1H,
C%2-OH). 13C: l (100 MHz, CDCl3) 211 (FeCO), 203
(CꢀO), 163 and 160 (2%,4%C), 130 (5%-C or 6%C), 117 and
114 (1%,3%C), 108 (5%-C or 6%C), 85 (2,3-C, superimposed),
64.2 and 60 (1,4-C), 32 (5-C), 27 (6-C), 26 (Me).
4.1.2. Tricarbonyl{[(1,2,3,4-p)-2-methoxy-5h-(3%-ethan-
oyl-2%,6%-dihydroxyphenyl)]-1,3-cyclohexadiene}iron(0)
(5)
4.1. General procedure for the reaction of tricarbonyl-
(p5-cyclohexadienyl)iron hexafluorophosphate salts with
fla6onoid compounds and 2,4-dihydroxyacetophenone
By the general procedure, tricarbonyl[(1,2,3,4,5-h)-2-
methoxycyclohexadienyl]iron(1+)
hexafluorophos-
phate(1−) (3) (0.394 g, 1 mmol) was reacted with
2,4-dihydroxyacetophenone (0.0761 g, 0.5 mmol) and
triethylamine (0.16 ml, 1.1 mmol) in acetonitrile (6 ml)
to afford tricarbonyl{[(1,2,3,4-h)-2-methoxy-5a-(3%-
ethanoyl-2%,6%-dihydroxyphenyl)]-1,3-cyclohexadiene}-
iron(0) (5) as a dark viscous golden oil (0.055 g, 28%).
To a stirred solution at room temperature (r.t.) of the
appropriate phenolic compound in dry acetonitrile were
successively added triethylamine (from a small graduated
syringe) and the tricarbonyl[(1,2,3,4,5-h)cyclohexa-
dienyl]iron salt 2 or 3 (approximately one or two
equivalents, see below). The progress of the reaction was
followed by monitoring the disappearance of w(CO)
bands of the cation by FT-IR spectroscopy. After the salt
has been consumed (usually 0.5 h), the reaction mixture
was extracted with diethyl ether (50 ml). The extract was
washed with water (2×20 ml) and brine (20 ml). The
organic layer was dried over MgSO4 and concentrated
under reduced pressure to give a crude product. Purifica-
tion was performed by column chromatography on silica
gel (eluent petrol ether–diethyl ether) or preparative
HPLC (Dynamax C-18 column with gradient elution
with methanol–water at a flow rate of 1.5 ml min−1).
IR (CHCl3): wmax 2044, 1969 (FeCO), 1621 (CO) cm−1
.
MS (EI) m/z (rel. int. %): 344 (M+ −2CO, 13), 314 (20),
260 (25), 213 (27), 137 (30), 108 (98), 78 (65). Anal. Calc.
for C18H16O7Fe: C, 54.03; H, 4.03. Found: C, 54.3; H,
4.2%. NMR, 1H: l (400 MHz, CDC13) 2.02 (d, J 14 Hz,
1H, 6a-H), 2.16 (m, 1H, 6b-H), 2.54 (s, 3H, COMe), 2.61
(m, 1H, 4-H), 3.47 (m, 1H, 1-H), 3.69 (s, 3H, OMe), 3.81
(dm, J 8.0 Hz, 1H, 5b-H), 5.15 (m, 1H, 3-H), 6.29 (d, J
8.0 Hz, 1H, 5%-H), 7.50 (d, J 8.0 Hz, 1H, 4%-H).
4.1.3. 4%,5,7-Trihydroxy-6-{tricarbonyl[(1¦,2¦,3¦,4¦-p)-
1¦,3¦-cyclohexadien-5¦h-yl]iron(0)}fla6anone (7), 4%,5,-
7-trihydroxy-8-{tricarbonyl[(1¦,2¦,3¦,4¦-p)-1¦,3¦-
cyclohexadien-5¦h-yl]iron(0)}fla6anone (9) and 4%,5,7-
trihydroxy-6,8-bis-{tricarbonyl[(1¦,2¦,3¦,4¦-p)-1¦,3¦-
cyclohexadien-5¦h-yl]iron(0)}fla6anone (11)
4.1.1. Tricarbonyl{[(1,2,3,4-p)-5h-(3%-ethanoyl-2%,6%-
dihydroxyphenyl)]-1,3-cyclohexadiene}iron(0) (4)
By the general procedure, tricarbonyl[(1,2,3,4,5-
h)cyclohexadienyl]iron(1+) hexafluorophosphate(1−)
(2) (0.364 g, 1 mmol) was reacted with 2,4-dihydroxyace-
tophenone (0.152 g, 1 mmol) and triethylamine (0.16 ml,
1.1 mmol) in acetonitrile (6 ml) to afford tricar-
bonyl{[(1,2,3,4 - h) - 5a - (3% - ethanoyl - 2%,6% - dihydroxy -
phenyl)]-1,3-cyclohexadiene}iron(0) (4) (0.0705 g, 19%).
Similarly tricarbonyl[(1,2,3,4,5-h)cyclohexadienyl]iron
(1+) hexafluorophosphate(1−) (0.364 g, 1 mmol) was
reacted with 2,4-dihydroxyacetophenone (0.0761 g, 0.5
mmol) and triethylamine (0.16 ml, 1.1 mmol) in acetoni-
trile (6 ml) to afford tricarbonyl{[(1,2,3,4-h)-5a-(3%-
ethanoyl-2%,6%-dihydroxyphenyl)]-1,3-cyclohexadiene}-
iron(0) as a pink–yellow solid (0.100 g, 54%). M.p.
169–171°C (dec.). IR (CHCl3): wmax 2053, 1973 (FeCO),
1631 (CꢀO) cm−1. MS (EI) m/z (rel. int. %): 342
(M+ −CO, 1), 314 (M+ −2CO, 24), 286 (M+ −3CO,
19), 284 (31), 268 (18), 266 (24), 229 (16), 213 (24), 137
(18), 115 (12), 78 (78). Anal. Calc. for C17H14O6Fe: C,
By the general procedure, tricarbonyl[(1,2,3,4,5-
h)cyclohexadienyl]iron(1+) hexafluorophosphate(1−)
(2) (0.363 g, 1 mmol) was reacted with 4%,5,7-trihydroxy-
flavanone (0.272 g, 1 mmol) and triethylamine (0.42 ml,
2.9 mmol) in acetonitrile (10 ml) to afford 4%,5,7-tri-
hydroxy-6-{tricarbonyl[(1¦,2¦,3¦,4¦-h)-1¦,3¦-cyclohexa-
dien-5¦a-yl]iron(0)}flavanone (7) (0.07 g, 14%), 4%,5,7-tri-
hydroxy - 8 - {tricarbonyl[(1¦,2¦,3¦,4¦ - h) - 1¦,3¦ - cyclo-
hexadien-5¦a-yl]iron(0)}flavanone (9) (0.0105 g, 2%) and
4%,5,7-trihydroxy-6,8-bis-{tricarbonyl[(1¦,2¦,3¦,4¦-h)-1¦,-
3¦-cyclohexadien-5¦a-yl]iron(0)}flavanone (11) (0.0917
g, 13%) as dark viscous golden oils. Similarly, tri-
carbonyl[(1,2,3,4,5-h)cyclohexadienyl]iron(1+) hexa-
fluorophosphate(1−) (0.400 g, 1.1 mmol) was reacted
with 4%,5,7-trihydroxyflavanone (0.136 g, 0.5 mmol) and
triethylamine (2.2 ml, 1.5 mmol) in acetonitrile (10 ml)
to afford 4%,5,7-trihydroxy-6,8-bis-{tricarbonyl[(1¦,2¦,3¦,
4¦-h)-1¦,3¦-cyclohexadien-5¦a-yl]iron(0)}flavanone (11)
(0.125 g, 32%).
1
55.14; H, 3.78. Found: C, 55.1; H, 3.8%. NMR, H: l
4%,5,7-Trihydroxy-6-{tricarbonyl[(1¦,2¦,3¦,4¦-h)-1¦,3¦-
cyclohexadien-5¦a-yl]iron(0)}flavanone (7). IR (CHCl3):
wmax 2053, 2040, 1973 (FeCO), 1636 (CꢀO) cm−1. MS
(FAB) m/z (rel. int. %): 491 (MH+, 50), 434 (M+
(400 MHz, CDCl3) 1.82 (d, J 11 Hz, 1H, 6a-H), 2.18 (m,
1H, 6b-H), 2.54 (s, 3H, COMe), 3.04 and 3.27 (2×m,
2×1H; 1,4-H), 4.09 (m, 1H; 5b-H), 5.44 and 5.56 (2×m,

108
A.V. Malko6 et al. / Journal of Organometallic Chemistry 589 (1999) 103–110
−2CO, 100), 406 (M+ −3CO, 80), 391 (17), 362 (17),
285 (22). Anal. Calc. for C24H18O8Fe: C, 58.78; H, 3.67.
Found: C, 58.4; H, 3.8%. NMR, 1H: l (400 MHz,
CDCl3) 1.81 (m, 1H, 6¦a-H), 2.10 (m, 1H, 6¦b-H), 2.98
(m, 1H, 3a-H), 3.00–3.22 (m, 3×1H; 1¦,3b, 4¦-H); 3.93
(m, 1H, 5¦b-H); 5.27 (m, 1H, 2-H); 5.39 and 5.53
(2×m, 2×1H; 2¦,3¦-H), 5.92 (s, 1H, 8-H), 6.94 (d, J
6.0 Hz, 2H; 3%,5%-H), 7.32 (d, J 6.0 Hz, 2H; 2%,6%-H);
12.50 (s, 1H, 5-OH).
ly tricarbonyl[(1,2,3,4,5 - h) - 2 - methoxycyclohexadi-
enyl]iron(1+) hexafluorophosphate(1−) (0.867 g, 2.2
mmol) was reacted with 4%,5,7-trihydroxyflavanone
(0.272 g, 1 mmol) and triethylamine (0.42, 2.9 mmol) to
afford 4%,5,7-trihydroxy-6,8-bis-{tricarbonyl[(1¦,2¦,3¦,
4¦-h)-2¦-methoxy-1¦,3¦-cyclohexadien-5¦a-yl]iron(0)}
flavanone (12) (0.357 g, 46%).
4%,5,7-Trihydroxy-6-{tricarbonyl[(1¦,2¦,3¦,4¦-h)-2¦-
methoxy-1¦,3¦-cyclohexadien-5¦a-yl]iron(0)}flavanone
(8). IR (CHCl3): wmax 2051, 2042, 1971 (FeCO), 1622
(CꢀO) cm−1. MS (FAB) m/z (rel. int. %): 521 (MH+,
16), 464 (M+ −2CO, 100), 434 (63), 315 (22), 136 (24).
HRMS (FAB) Calc. for C25H21O9Fe (MH+), 521.0535.
4%,5,7-Trihydroxy-8-{tricarbonyl[(1¦,2¦,3¦,4¦-h)-1¦,3¦-
cyclohexadien-5¦a-yl]iron(0)}flavanone (9). IR (CHCl3):
wmax 2051, 2041, 1973 (FeCO), 1632 (CꢀO) cm−1. MS
(FAB) m/z (rel. int. %): 491 (MH+, 2), 447 (23), 391
(100), 362 (17), 289 (16), 279 (52). HRMS (FAB) Calc.
for C24H19O8Fe (MH+), 491.0429. Found: 491.0429.
1
Found: 521.0535. NMR, H: l (400 MHz, CDCl3) 1.99
(m, 1H, 6¦a-H), 2.14 (m, 1H, 6¦b-H), 2.59 (dd; J 5.0,
3.0 Hz; 1H, 4¦-H), 2.76 (m, 1H, 3a-H), 3.06 (m, 1H;
3b-H), 3.45 (m, 1H, 1¦-H), 3.68 (s, 3H, 2¦-OMe), 4.22
(m, 1H, 5¦b-H); 5.15 (m, 1H, 2-H); 5.30 (d, J 6.0 Hz,
1H, 3¦-H), 5.90 (s, 1H, 8-H), 6.80 (d, J 8.0 Hz, 2H;
3%,5%-H), 7.34 (d, J 8.0 Hz, 2H; 2%,6%-H); 12.50 (s, 1H,
5-OH).
1
NMR, H: l (400 MHz, CDCl3) 1.81 (m, 1H, 6¦a-H),
2.20 (m, 1H, 6¦b-H), 2.76 (m, 1H, 3a-H), 2.92–3.23 (m,
4×1H; 1¦,3b,4¦,5¦b-H); 5.30 (m, 1H, 2-H); 5.46 and
5.52 (2×m, 2×1H; 2¦,3¦-H), 5.90 (s, 1H, 6-H), 6.88
(d, J 6.0 Hz, 2H; 3%,5%-H), 7.31 (d, J 6.0 Hz, 2H;
2%,6%-H); 12.54 (s, 1H, 5-OH).
4%,5,7-Trihydroxy-6,8-bis-{tricarbonyl[(1¦,2¦,3¦,4¦-h)-
1¦,3¦-cyclohexadien-5¦a-yl]iron(0)}flavanone (11). IR
(CHCl3): wmax 2053, 2041, 1974 (FeCO), 1623 (CꢀO)
cm−1. MS (FAB) m/z (rel. int. %): 709 (MH+, 42), 652
(M+ −2CO, 85), 624 (M+ −3CO, 100), 556 (70), 536
(30), 362 (45), 219 (35), 136 (37). Anal. Calc. for
C33H24O11Fe2: C, 55.93; H, 3.39. Found: C, 55.8; H,
4%,5,7-Trihydroxy-8-{tricarbonyl[(1¦,2¦,3¦,4¦-h)-2¦-
methoxy-1¦,3¦-cyclohexadien-5¦a-yl]iron(0)}flavanone
(10). IR (CHCl3): wmax 2050, 2041, 1970 (FeCO), 1620
(CꢀO) cm−1. MS (FAB) m/z (rel. int. %): 521 (MH+,
8), 464 (M+ −2CO, 34), 448 (16), 435 (26), 391 (41),
149 (100), 136 (49). HRMS (FAB) Calc. for
C25H21O9Fe (MH+), 521.0535. Found: 521.0535.
1
1
3.2%. NMR, H: l (400 MHz, CDCl3) 1.57–1.77 (m,
NMR, H: l (400 MHz, CDCl3) 1.98 (m, 1H, 6¦a-H),
2×1H, 6¦a-H), 2.08 and 2.21 (2×m, 2×1H, 6¦b-H),
2.77 (m, 1H, 3a-H), 2.85–3.26 (m, 5×1H; 1¦,4¦- and
3b-H); 3.91 and 4.03 (2×m, 2×1H, 5¦b-H); 4.97–5.11
(m, 3×1H; 2 and 3¦-H), 5.47 and 5.67 (2×m, 2×1H;
2¦-H); 6.92 (d, J 7.0 Hz, 2H; 3%,5%-H), 7.33 (d, J 7.0 Hz,
2H; 2%,6%-H); 12.58 (s, 1H, 5-OH).
2.15 (m, 1H, 6¦b-H), 2.58 (dd; J 6.0, 3.0 Hz; 1H, 4¦-H),
2.77 (dd; J 15.0, 4.0 Hz, 1H, 3a-H), 3.06 (dd; J 15.0, 6.0
Hz; 1H; 3b-H), 3.45 (m, 1H, 1¦-H), 3.68 (s, 3H, 2¦-
OMe), 3.69 (m, 1H, 5¦b-H); 5.11 (m, 1H, 2-H); 5.30 (d,
J 11.0 Hz, 1H, 3¦-H), 5.86 (s, 1H, 6-H), 6.88 (d, J 8.0
Hz, 2H; 3%,5%-H), 7.31 (d, J 8.0 Hz, 2H; 2%,6%-H); 12.50
(s, 1H, 5-OH).
4.1.4. 4%,5,7-Trihydroxy-6-{tricarbonyl[(1¦,2¦,3¦,4¦-p)-
2¦-methoxy-1¦,3¦-cyclohexadien-5¦h-ylliron(0)}fla6a-
none (8), 4%,5,7-trihydroxy-8-{tricarbonyl[(1¦,2¦,3¦,4¦-
p)-2¦-methoxy-1¦,3¦-cyclohexadien-5¦h-yl]iron(0)}-
fla6anone (10), and 4%,5,7-trihydroxy-6,8-bis-{tricar-
bonyl[(1¦,2¦,3¦,4¦-p)-2¦-methoxy-1¦,3¦-cyclohexadien-
5¦h-yl]iron(0)}fla6anone (12)
4%,5,7-Trihydroxy-6,8-bis-{tricarbonyl[(1¦,2¦,3¦,4¦-h)-
2¦-methoxy-1¦,3¦-cyclohexadien-5¦a-yl]iron(0)}flava-
none (12). IR (CHCl3): wmax 2051, 2040, 1971 (FeCO),
1623 (CꢀO) cm−1. MS (FAB) m/z (rel. int. %): 769
(MH+, 35), 712 (M+ −2CO, 52), 683 (100), 653 (17),
629 (29), 596 (55), 543 (45), 421 (36), 249 (76). HRMS
(FAB) Calc. for C35H29O13Fe2 (MH+), 769.0317.
1
By the general procedure, tricarbonyl[(1,2,3,4,5-h)-2-
Found: 769.0317. NMR, H: l (400 MHz, CDCl3) 1.98
methoxycyclohexadienyl]iron(1+)
hexafluorophos-
(m, 2×1H, 6¦a-H), 2.14 (m, 2×1H, 6¦b-H), 2.59 (m,
2×1H, 4¦-H), 2.76 (m, 1H, 3a-H), 3.06 (m, 1H; 3b-H),
3.45 (m, 2×1H, 1¦-H), 3.68 (s, 2×3H, 2¦-OMe), 3.69
and 4.22 (2×m, 2×1H, 5¦b-H); 5.08 (m, 2×1H, 2-H);
5.30 (m, 2×1H, 3¦-H), 6.85 (d, J 8.0 Hz, 2H; 3%,5%-H),
7.32 (d, J 8.0 Hz, 2H; 2%,6%-H); 12.50 (s, 1H, 5-OH).
phate(1−) (3) (0.394 g, 1 mmol) was reacted with
4%,5,7-trihydroxyflavanone (0.272 g, 1 mmol) and tri-
ethylamine (0.42 ml, 2.9 mmol) in acetonitrile (10 ml)
to afford 4%,5,7-trihydroxy-6-{tricarbonyl[(1¦,2¦,3¦,4¦-
h)-2¦-methoxy-1¦,3¦-cyclohexadien-5¦a-ylliron(0)}flava-
none (8) (0.154 g, 3%), 4%,5,7-trihydroxy-8-{tricar-
bonyl[(1¦,2¦,3¦,4¦-h)-2¦-methoxy-1¦,3¦-cyclohexadien-
5¦a-yl]iron(0)}flavanone (10) (0.038 g, 0.7%) and 4%,5,7-
trihydroxy-6,8-bis-{tricarbonyl[(1¦,2¦,3¦,4¦-h)-2¦-meth-
oxy-1¦,3¦-cyclohexadien-5¦a-yl]iron(0)}flavanone (12)
(0.029 g, 4%) as a dark viscous golden oil. Similar-
4.1.5. 3%,5,7-Trihydroxy-4%-methoxy-6-{tricarbonyl[(1¦,-
2¦,3¦,4¦-p)-1¦,3¦-cyclohexadien-5¦h-yl]iron(0)}fla6a -
none (13), 3%,5,7-trihydroxy-4%-methoxy-8-{tricarbonyl-
[(1¦,2¦,3¦,4¦-p)-1¦,3¦-cyclohexadien-5¦h-yl]iron(0)}fla-
6anone (15), and 3%,5,7-trihydroxy-4%-methoxy-6,8-bis-

A.V. Malko6 et al. / Journal of Organometallic Chemistry 589 (1999) 103–110
109
100), 586 (52), 566 (26), 514 (26), 361 (29). HRMS
(FAB) Calc. for C34H27O12Fe2 (MH+), 739.0201.
Found: 739.0201. NMR, H: l (400 MHz, CDCl3) 1.80
{tricarbonyl[(1¦,2¦,3¦,4¦-p)-1¦,3¦-cyclohexadien-5¦h-yl]-
iron(0)}fla6anone (17)
1
By the general procedure, tricarbonyl[(1,2,3,4,5-
h)cyclohexadienyl]iron(1+) hexafluorophosphate(1−)
(2) (0.363 g, 1 mmol) was reacted with 3%,5,7-trihy-
droxy-4%-methoxyflavanone (0.302 g, 1 mmol) and tri-
ethylamine (0.42 ml, 2.9 mmol) in acetonitrile (10 ml)
to afford 3%,5,7-trihydroxy-4%-methoxy-6-{tricarbonyl-
[(1¦,2¦,3¦,4¦-h) - 1¦,3¦ - cyclohexadien - 5¦a - yl]iron(0)}-
flavanone (13) (0.0603 g, 11%), 3%,5,7-trihydroxy-4%-
methoxy-8-{tricarbonyl[(1¦,2¦,3¦,4¦-h)-1¦,3¦-cyclohexa-
dien-5¦a-yl]iron(0)}flavanone (15) (0.0247 g, 5%) and
3%,5,7-trihydroxy-4%-methoxy-6,8-bis-{tricarbonyl[(1¦,-
2¦,3¦,4¦-h)-1¦,3¦-cyclohexadien-5¦a-yl]iron(0)}flavanone
(17) (0.0341 g, 5%) as a dark viscous golden oil. Simi-
larly tricarbonyl[(1,2,3,4,5-h)cyclohexadienyl]iron(1+)
hexafluorophosphate(1−) (0.404 g, 1.1 mmol) was re-
acted with 3%,5,7-trihydroxy-4%-methoxyflavanone (0.151
g, 0.5 mmol) and triethylamine (0.22 ml, 1.5 mmol) in
acetonitrile (10 ml) to afford 3%,5,7-trihydroxy-4%-
methoxy-6,8-bis-{tricarbonyl[(1¦,2¦,3¦,4¦-h)-1¦,3¦-cyclo-
hexadien-5¦a-yl]iron(0)}flavanone (17) (0.105 g, 28%).
3%,5,7-Trihydroxy-4%-methoxy-6-{tricarbonyl[(1¦,2¦,-
3¦,4¦-h)-1¦,3¦-cyclohexadien-5¦a-yl]iron(0)}flavanone
(13). IR (CHCl3): wmax 2051, 2040, 1968 (FeCO); 1634
(CꢀO) cm−1. MS (FAB) m/z (rel. int. %): 521 (MH+,
60), 464 (M+ −2CO, 90), 436 (M+ −3CO, 100), 362
(21), 285 (22). Anal. Calc. for C25H20O9Fe: C, 57.69; H,
(m, 2×1H, 6¦a-H), 2.08 (m, 2×1H, 6¦b-H), 2.76 (dd;
J 17.0, 9.0 Hz; 1H, 3a-H); 2.94–3.20 (m, 5×1H; 3b-
and 1¦,4¦-H); 3.90 (s, 3H, 4%-OMe); 3.91 and 4.02
(2×m, 2×1H, 5¦b-H); 4.90–5.25 (m, 3×1H, 2- and
2¦-H); 5.32 and 5.38 (2×m, 2×1H, 3¦-H), 6.95 (m,
2×1H; 5%,6%-H), 7.01 (s, 1H, 2%-H), 12.40 (s, 1H, 5-
OH).
4.1.6. 3%,5,7-Trihydroxy-4%-methoxy-6-{tricarbonyl[(1¦,-
2¦,3¦,4¦-p)-2¦-methoxy-1¦,3¦-cyclohexadien-5¦h-yl]iron-
(0)}fla6anone (14), 3%,5,7-trihydroxy-4%-methoxy-8-{tri-
carbonyl[(1¦,2¦,3¦,4¦-p)-2¦-methoxy-1¦,3¦-cyclohexa-
dien - 5¦h - yl]iron(0)}fla6anone (16) and 3%,5,7 - trihydr -
oxy-4%-methoxy-6,8-bis-{tricarbonyl[(1¦,2¦,3¦,4¦-p)-2¦-
methoxy - 1¦,3¦ - cyclohexadien - 5¦h - yl]iron(0)}fla6anone
(18)
By the general procedure, tricarbonyl[(1,2,3,4,5-h)-2-
methoxycyclohexadienyl]iron(1+) hexafluorophosph-
ate(1−) (3) (0.394 g, 1 mmol) was reacted with 3%,5,7-
trihydroxy-4%-methoxyflavanone (0.302 g, 1 mmol) and
triethylamine (0.42 ml, 2.9 mmol) to afford 3%,5,7-tri-
hydroxy-4%-methoxy-6-{tricarbonyl[(1¦,2 ¦,3¦,4¦-h)-2¦-
methoxy-1¦,3¦-cyclohexadien-5¦a-yl]iron(0)}flavanone
(14) (0.0847 g, 15%), 3%,5,7-trihydroxy-4%-methoxy-8-
{tricarbonyl[(1¦,2¦,3¦,4¦-h)-2¦-methoxy-1¦,3¦-cyclohexa-
dien-5¦a-yl]iron(0)}flavanone (16) (0.0169 g, 3%) and
3%,5,7-trihydroxy-4%-methoxy-6,8-bis-{tricarbonyl[(1¦,-
2¦,3¦,4¦-h)-2¦-methoxy-1¦,3¦-cyclohexadien-5¦a-yl]-
iron(0)}flavanone (18) (0.0246 g, 3%) as a dark viscous
golden oil. Similarly tricarbonyl[(1,2,3,4,5-h)-2-meth-
1
3.85. Found: C, 57.8; H, 3.8%. NMR, H: l (400 MHz,
CDCl3) 1.80 (m, 1H, 6¦a-H), 2.12 (dq; J 11.0, 4.0 Hz;
1H, 6¦b-H), 2.76 (dd; J 17.0, 2.0 Hz; 1H, 3a-H), 3.04
(m, 2×1H; 3b,4¦-H); 3.21 (m, 1H, 1¦-H), 3.90 (s, 3H,
4%-OMe), 3.91 (m, 1H, 5¦b-H), 5.00 (m, 1H, 2-H); 5.25
and 5.38 (2×m, 2×1H; 2¦,3¦-H), 5.72 (s, 1H, 8-H),
6.87 (m, 2×1H; 5%,6%-H), 7.01 (s, 1H, 2%-H); 12.48 (s,
1H, 5-OH).
oxycyclohexadienyl]iron(1+)
hexafluorophosphate-
(1−) (0.433 g, 1.1 mmol) was reacted with 3%,5,7-trihy-
droxy-4%-methoxyflavanone (0.151 g, 0.5 mmol) and
triethylamine (0.42 ml, 2.9 mmol) in acetonitrile (10 ml)
to afford 3%,5,7-trihydroxy-4%-methoxy-6,8-bis-{tricar-
bonyl[(1¦,2¦,3¦,4¦-h)-2¦-methoxy-1¦,3¦-cyclohexadien-
5¦a-yl]iron(0)}flavanone (18) (0.100 g, 25%).
3%,5,7-Trihydroxy-4%-methoxy-8-{tricarbonyl[(1¦,2¦,-
3¦,4¦-h)-1¦,3¦-cyclohexadien-5¦a-yl]iron(0)}flavanone
(15). IR (CHCl3): wmax 2051, 2040, 1968 (FeCO); 1635
(CꢀO) cm−1. MS (FAB) m/z (rel. int. %): 521 (MH+,
45), 464 (M+ −2CO, 40), 436 (M+ −3CO, 100), 290
(15), 232 (15). HRMS (FAB) Calc. for C25H21O9Fe
3%,-5,7-Trihydroxy-4%-methoxy-6-{tricarbonyl[(1¦,2 ¦,-
3¦,4¦-h)-2¦-methoxy-1¦,3¦-cyclohexadien-5¦a-yl]iron-
(0)}flavanone (14). IR (CHCl3): wmax 2050, 2041, 1971
(FeCO), 1641 (CꢀO) cm−1. MS (FAB) m/z (rel. int. %):
1
(MH+), 521.0535. Found: 521.0535. NMR, H: l (400
MHz, CDCl3) 1.74 (dd; J 16.0, 6.0 Hz; 1H, 6¦a-H), 2.06
(m, 1H, 6¦b-H), 2.76 (dd; J 16.0, 9.0 Hz; 1H, 3a-H);
2.94–3.20 (m, 3×1H; 1¦,3b,4¦-H); 3.90 (m, 1H, 5¦b-
H), 3.96 (s, 3H, 4%-OMe), 4.98 (m, 1H, 2-H); 5.13 and
5.32 (2×m, 2×1H; 2¦,3¦-H), 5.70 (s, 1H, 6-H), 6.95
(m, 2×1H; 5%,6%-H), 7.08 (s, 1H, 2%-H); 12.06 (s, 1H,
5-OH).
3%,5,7-Trihydroxy-4%-methoxy-6,8-bis-{tricarbonyl-
[(1¦,2¦,3¦,4¦-h)-1¦,3¦-cyclohexadien-5¦a-yl]iron(0)}flava-
none (17). IR (CHCl3): wmax 2051, 2039, 1971 (FeCO),
1624 (CꢀO) cm−1. MS (FAB) m/z (rel. int. %): 739
(MH+, 28), 682 (M+ −2CO, 51), 654 (M+ −3CO,
551 (MH+, 31), 494 (M+ −2CO, 100), 466 (M+
−
3CO, 98), 451 (20), 315 (19), 136 (33). HRMS (FAB)
Calc. for C26H23O10Fe (MH+), 551.0641. Found:
1
551.0641. NMR, H: l (400 MHz, CDCl3) 1.99 (dm, J
12.0 Hz, 1H, 6¦a-H), 2.14 (dq; J 12.0, 4.0 Hz; 1H,
6¦b-H), 2.59 (dd; J 6.0, 3.0 Hz; 1H, 4¦-H), 2.76 (dd; J
17.0, 2.0 Hz; 1H, 3a-H), 3.45 (m, 1H; 3b-H), 3.49 (m,
1H, 1¦-H), 3.68 (s, 3H, 2¦-OMe), 3.69 (m, 1H, 5¦b-H),
3.90 (s, 3H, 4%-OMe), 5.15 (m, 1H, 2-H); 5.27 (m, 1H,
3¦-H), 5.70 (s, 1H, 8-H), 6.88 (m, 2×1H; 5%,6%-H), 7.02
(s, 1H, 2%-H); 12.49 (s, 1H, 5-OH).

110
A.V. Malko6 et al. / Journal of Organometallic Chemistry 589 (1999) 103–110
3%,5,7-Trihydroxy-4%-methoxy-8-{tricarbonyl[(1¦,2¦,-
Mass Spectrometry Service Centre, Swansea, for FAB
mass spectrometric data. We thank Dr J. Alan Downie
and Karen E. Wilson, Department of Genetics, John
Innes Centre, Colney Lane, Norwich NR4 7UH, UK,
for discussions and for the screening data presented in
reference [10].
3¦,4¦-h)-2¦-methoxy-1¦,3¦-cyclohexadien-5¦a-yl]iron-
(0)}flavanone (16). IR (CHCl3): wmax 2050, 2042, 1970
(FeCO), 1640 (CꢀO) cm−1. MS (FAB) m/z (rel. int. %):
551 (MH+, 50), 494 (M+ −2CO, 65), 466 (M+
−
3CO, 100), 315 (25), 249 (11), 136 (20). HRMS (FAB)
Calc. for C26H23O10Fe (MH+), 551.0641. Found:
1
551.0641. NMR, H: l (400 MHz, CDCl3) 1.90 (m, 1H,
6¦a-H), 2.09 (dq; J 13.0, 5.0 Hz; 1H, 6¦b-H), 2.56 (m,
1H, 4¦-H), 2.77 (m, 1H, 3a-H), 3.02 (m, 1H; 3b-H),
3.25 (m, 1H, 1¦-H), 3.35 (m, 1H, 5¦b-H), 3.41 (s, 3H,
2¦-OMe), 3.94 (s, 3H, 4%-OMe), 4.99 (m, 1H, 2-H); 5.28
(m, 1H, 3¦-H), 5.73 (s, 1H, 6-H), 6.91 (m, 2×1H;
5%,6%-H), 7.02 (s, 1H, 2%-H); 12.79 (s, 1H, 5-OH).
3%,5,7-Trihydroxy-4%-methoxy-6,8-bis-{tricarbonyl-
[(1¦,2¦,3¦,4¦-h)-2¦-methoxy-1¦,3¦-cyclohexadien-5¦a-yl]-
iron(0)}flavanone (18). IR (CHCl3): wmax 2050, 2041,
1968 (FeCO), 1635 (CꢀO) cm−1. MS (FAB) m/z (rel.
int. %): 799 (MH+, 12), 742 (M+ −2CO, 68), 714
(M+ −3CO, 100), 683 (20), 646 (45), 626 (70), 598
(46), 421 (42), 361 (61). HRMS (FAB) Calc. for
C36H31O14Fe2 (MH+), 799.0400. Found: 799.0400.
NMR, 1H: l (400 MHz, CDCl3) 1.76 (m, 2×1H,
6¦a-H), 2.13 (m, 2×1H, 6¦b-H), 2.57 (m, 2×1H,
4¦-H), 2.75 (m, 1H, 3a-H), 2.99 (m, 1H; 3b-H); 3.26
and 3.40 (2×m, 2×1H, 1¦-H); 3.47 and 3.70 (2×m,
2×1H, 5¦b-H); 3.48 and 3.71 (2×s, 2×3H, 2¦-OMe);
3.91 (s, 3H, 4%-OMe), 5.12 (m, 1H, 2-H), 5.24 (m,
2×1H, 3¦-H), 6.90 (m, 2×1H; 5%,6%-H), 7.00 (s, 1H,
2%-H); 12.62 (s, 1H, 5-OH).
References
[1] (a) M. Salmain, A. Vessieres, G. Jaouen, I.S. Butler, Anal.
Chem. 63 (1991) 2323. (b) A. Varenne, A. Vessieres, M. Salmain,
S. Durand, P. Brossier, G. Jaouen, Anal. Biochem. 242 (1996)
172.
[2] C.S. Creaser, M.A. Fey, G.R. Stephenson, Spectrochim. Acta
50A (1994) 1295.
[3] C.E. Anson, C.S. Creaser, O. Egyed, G.R. Stephenson, Spec-
trochim. Acta 53A (1997) 1867.
[4] (a) F. Le Borgne, J.P. Beaucourt, Tetrahedron Lett. 29 (1988)
5649. (b) A. Varenne, M. Salmain, C. Brisson, G. Jaouen,
Bioconjugate Chem. 3 (1992) 471. (c) J.A. Carver, B. Fates,
L.A.P. Kane-Maguire, J. Chem. Soc. Chem. Commun. (1993)
928. (d) M. Salmain, M. Gunn, A. Gorfti, S. Top, G. Jaouen,
Bioconjugate Chem. 4 (1993) 425. (e) C.E. Anson, C.S. Creaser,
O. Egyed, M.A. Fey, G.R. Stephenson, J. Chem. Soc. Chem.
Commun. (1994) 39. (f) A. Gorfti, M. Salmain, G. Jaouen, J.
Chem. Soc. Chem. Commun. (1994) 433. (g) K.L. Malisza, S.
Top, J. Vaissermann, B. Caro, M.C. Senechaltocquer, D.
Senechal, J.Y. Saillard, S. Triki, S. Kahlal, J.F. Britten, M.J.
McGlinchey, G. Jaouen, Organometallics 14 (1995) 5273. (h) A.
Gorfti, M. Salmain, G. Jaouen, M.J. McGlinchey, A. Bennouna,
A. Mousser, Organometallics 15 (1996) 142. (i) D. Osella, M.
Ravera, M. Vincenti, B. Malezieux, G. Jaouen, Tetrahedron
Lett. 37 (1996) 6561. (j) D. Osella, M. Ravera, M. Vincenti, M.
Salmain, G. Jaouen, Organometallics 15 (1996) 3037. (k) A.
Kazimierczak, J. Zakrzewski, M. Salmain, G. Jaouen, Bioconju-
gate Chem. 8 (1997) 489.
4.1.7. Preparation of 4%,5,7-trihydroxy-6-{tricarbonyl-
[(1¦,2¦,3¦,4¦-p)-1¦,3¦-cyclohexadien-5¦h-yl]iron(0)}-
fla6anone (7) using NaH for deprotonation
[5] Z. Wang, B.A. Roe, K.M. Nicholas, R.L. White, J. Am. Chem.
Soc. 115 (1993) 4399.
To a stirred solution of 4%,5,7-trihydroxyflavanone
(0.272 g, 1.0 mmol) in THF (40 ml) was added NaH (0.
145 mg, 6 mmol) at −78°C. The resulting solution was
stirred for 1 h at −78°C. Tricarbonyl[(1,2,3,4,5-
h)cyclohexadienyl]iron(1+) hexafluorophosphate(1−)
(2) (0.364 g, 1 mmol) was added in one portion and the
mixture was stirred for a further 15 min at −78°C. The
reaction mixture was allowed to warm to r.t. and left to
stir overnight. Water (50 ml) was added and the mix-
ture was extracted with diethyl ether (3×30 ml). The
combined extracts were dried over MgSO4 and concen-
trated under reduced pressure. Further purification was
performed as described in the general procedure to
afford 4%,5,7-trihydroxy-6-{tricarbonyl[(1¦,2¦,3¦,4¦-h)-
1¦,3¦-cyclohexadien-5¦a-yl]iron(0)}flavanone (7) (0.43 g,
88%).
[6] (a) G. Jaouen, A. Vessieres, I.S. Butler, Acc. Chem. Res. 26
(1993) 361. (b) A. Pioko, R.G. Sutherland, A. Vessieres, G.
Jaouen, J. Organomet. Chem. 512 (1996) 79. (c) D. Osella, G.
Dutto, C. Nervi, M.J. McGlinchey, A. Vessieres, G. Jaouen, J.
Organomet. Chem. 533 (1997) 97.
[7] C.E. Anson, T.J. Baldwin, C.S. Creaser, M.A. Fey, G.R.
Stephenson, Organometallics 15 (1996) 1451.
[8] C.E. Anson, C.S. Creaser, G.R. Stephenson, Spectrochim. Acta
52A (1996) 1183.
[9] (a) J.W. Redmond, M. Batley, D.C. Yuan, M.W. Sutherland,
Nature 323 (1986) 632. (b) J.L. Firmin, K.E. Wilsom, L. Rossen,
A.W.B. Johnston, Nature 324 (1986) 90. (c) B. Horvath, C.W.
Bachem, J. Schell, A Kondorosi, EMBO J. 6 (1987) 841. (d)
S.A.J. Zaat, C.A. Wijffelman, H.P. Spaink, A.A.N. van Brussel,
R.J.H. Okker, B.J.J. Lugtenberg, J. Bacteriol. (1987) 198. (e)
S.A.J. Zaat, J. Scripsema, C.A. Wijffelman, A.A.N. van Brussel,
B.J.J. Lugtenberg, Plant Mol. Biol. 13 (1989) 175. (f) K. Re-
court, J. Scripsema, J.W. Kijne, A.A.N. van Brussel, B.J.J.
Lugtenberg, Plant Mol. Biol. 16 (1991) 841.
[10] C.E. Anson, C.S. Creaser, J.A. Downie, O. Egyed, A.V. Malkov,
L. Mojovic, G.R. Stephenson, A.T. Turner, K.E. Wilson,
Bioorg. Med. Chem. Lett. 8 (1998) 3549.
Acknowledgements
[11] A.J. Birch, D. Bogsanyi, J. Organomet. Chem. 214 (1981) C39.
[12] S. Bubavari (Ed.), The Merck Index, 12th ed., 1996, pp. 4704–
4705.
A.V.M., L.M. and A.T.T. thank the BBSRC for
financial support and we thank the EPSRC National