Total synthesis of 5,5′,6,6′-tetrahydroxy-3,3′-biindolyl, the proposed structure of a potent antioxidant found in beetroot (Beta vulgaris)

Article abstract of DOI:10.1016/j.tet.2004.02.043

5,5′,6,6′-Tetrahydroxy-3,3′-biindolyl, the proposed structure of a phenolic antioxidant isolated from the red beetroot (Beta vulgaris), has been synthesised. The spectroscopic data of the synthetic material is not consistent with that reported for the natural product.

Full text of DOI:10.1016/j.tet.2004.02.043

Tetrahedron 60 (2004) 3695–3712
Total synthesis of 5,50,6,60-tetrahydroxy-3,30-biindolyl,
the proposed structure of a potent antioxidant
found in beetroot (Beta vulgaris)
b
Simon P. H. Mee,a Victor Lee,a Jack E. Baldwina and Andrew Cowley
,
*
aThe Dyson Perrins Laboratory, University of Oxford, South Parks Road, Oxford OX1 3QY, UK
bChemical Crystallography Laboratory, University of Oxford, 9 Parks Road, Oxford OX1 3PD, UK
Received 15 December 2003; revised 27 January 2004; accepted 19 February 2004
Abstract—5,50,6,60-Tetrahydroxy-3,30-biindolyl, the proposed structure of a phenolic antioxidant isolated from the red beetroot (Beta
vulgaris), has been synthesised. The spectroscopic data of the synthetic material is not consistent with that reported for the natural product.
q 2004 Published by Elsevier Ltd.
1. Introduction
5,6-Dihydroxyindole is an interesting compound, because it
plays a central role in melanogenesis8 (the process by which
eumelanin, a black intractable biopigment, is formed from
L-3,4-dihydroxyphenylalanine9). Extra interest in 5,6-dihy-
droxyindoles has arisen from the recent recognition of
their exceptional radical scavenging and photoprotective
abilities,10 which makes them among the most e0ffective
endogenous antioxidants.11 The corresponding 2,2 -linked
isomer of 1 is known,12,13 and forms under oxidative
conditions from the monomer.13 Unsubstituted 3,30-bis-
indole is also known14 and has been synthesised from
unsymmetrical coupling partners, a route that does not take
advantage of its symmetry.
Interest in phenolic antioxidants found in fruits and
vegetables has recently increased1 due to a possibility2
that they may provide nutritional benefits.3 A significant
proportion of these compounds have been found to be
more powerful antioxidants than vitamins C and E, and
b-carotene using an in vitro model for heart disease.4 The
antioxidant activity of phenols is mainly due to their
reductive properties, however, they also have the capacity
for metal chelation.5
Ninety-two different phenol-containing plant extracts were
recently screened and beetroot peel was shown to have the
second-highest dry weight concentration of total phenols.1a,6
Structural characterisation of these compounds is necessary
in order to rationalise their mode of action. Kujala et al.
recently isolated a highly unstable phenolic compound from
the peel of the red beetroot (Beta vulgaris), and proposed its
structure to be 5,50,6,60-tetrahydroxy-3,30-biindolyl 17
(Fig. 1), a dimer of 5,6-dihydroxyindole.
2. Results and discussion
2.1. Retrosynthetic analysis
Towards our goal of synthesising 5,50,6,60-tetrahydroxy-
3,30-biindolyl 1, we proposed a short symmetrical synthesis
that fully exploits its symmetry. This route features an acid-
catalysed reductive cyclisation, dehydration and deprotec-
tion in the final step that should be compatible with
the oxidative lability of the product (Scheme 1). The
Figure 1. 5,50,6,60-Tetrahydroxy-3,30-biindolyl (1).
Keywords: Antioxidant; Beetroot; Biindolyl; 5,6-Dihydroxyindole.
*
Corresponding author. Tel.: þ44-1865-275-671; fax: þ44-1865-275632;
e-mail address: jack.baldwin@chem.ox.ac.uk
Scheme 1. The proposed reductive cyclisation.
0040–4020/$ - see front matter q 2004 Published by Elsevier Ltd.
doi:10.1016/j.tet.2004.02.043

3696
S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
Scheme 2. Synthesis of 3,4-bis(3,4-dibenzyloxy-2-nitrophenyl)furan (2) from catechol. Reagents and conditions: (i) BnCl, K2CO3, acetone, 65 8C, 4 days;
(ii) NBS, CCl4, 80 8C, 1 h; (iii) 70% HNO3, AcOH, rt, 2 h; (iv) Bu6Sn2, Pd(PPh3)4, toluene, 120 8C, 48 h; (v) Pd(PPh3)4, CuBr, THF, 60 8C, 15 h, 10% of 2.
3,4-disubstituted furan 2 was envisaged as an accessible,
stable surrogate of the required dialdehyde.
homocoupling of the tin starting material 6, was obtained
from the reaction.
2.2. Synthesis of 3,4-bis(3,4-dibenzyloxy-2-nitrophenyl)-
furan 2
Homocoupling has been observed in Stille reactions21 and is
an oxidative process.22 Cu(I) alone is also capable of
catalysing the reaction, especially when electron with-
drawing substituents are present in conjugation with the
tin.23 Rigorous exclusion of oxygen accompanied by the
addition of antioxidants, such as 2,6-di-tert-butyl-4-methyl
phenol, led to no significant reduction of the unwanted
product 9.
2.2.1. Double Stille coupling using 3,4-dibromofuran 7.
Initially we envisaged that 2 could be obtained from a
double Stille coupling of 3,4-dibromofuran 7 and 3,4-di-
benzyloxy-2-tri-n-butylstannylnitrobenzene 6 (Scheme 2).
The preparation of 6 commenced with the reaction of
catechol with benzyl chloride and potassium carbonate in
acetone, providing 3,4-dibenzyloxybenzene 3 in 87%
yield.15 Bromination with N-bromosuccinamide in carbon
tetrachloride gave 89% of compound 4,15 which was
nitrated with 70% nitric acid in 85% yield to afford the
ortho-substituted aryl bromide 5.16 Palladium catalysed
tributylstannylation of ortho-substituted aryl halides is
known to be relatively difficult.17 However, by employing
elevated temperatures and extended reaction times, the
ortho-substituted tributylstannyl aryl 6 could be formed in
68% yield. The coupling partner, 3,4-dibromofuran 7, was
prepared from (E)-2,3-dibromo-2-butene-1,4-diol in 56%
yield, using a slight modification of the procedure reported
by Rewicki et al.18
Efforts to optimise24 this reaction by using different cata-
lysts (Pd(PPh3)4, Pd(PPh3)2Cl2, Pd(dppf)Cl2, Pd(MeCN)2-
Cl2, Pd2(dba)3), Pd2(dba)3 with ligands in different ratios
(PPh3, P(2-furyl)3, AsPh3, dppf, 1,3-bis(diphenylphosphino)-
propane), different solvents (toluene, dioxane, THF, NMP,
DMF, DMSO), different additives (CuI, CuBr, CuCl, LiCl)
and slow addition of the tin starting material 6 led to no
improvement. In all cases, especially0 with highly polar
solvents, the major product was 4,40,5,5 -tetrakisbenzyloxy-
2,20-dinitro-biphenyl 9. It appears that this unwanted side-
reaction is significantly faster than the Stille coupling.
2.3. Initial solutions to the unfavourable double Stille
coupling
Stille couplings using electron-withdrawing ortho-substi-
tuted aryl stannanes are rare.19 To the best of our knowledge
there is only one previously published example in which the
substituent is a nitro group.19a Initial attempts at the double
Stille coupling using Pd(PPh3)4 in dioxane gave no product,
with most of the starting materials being recovered.
Copper(I) salts have been shown to accelerate Stille
reactions,20 and with the addition of CuBr and THF as the
solvent, a small amount of disubstituted furan 2 was isolated
in 10% yield together with the monosubstituted furan 8 in
14% yield. In addition 58% of the dimer 9, which arose from
2.3.1. 3,4-Diiodofuran. In the effort to enhance the desired
reaction, synthesis of 3,4-diiodofuran as an alternative
starting material was investigated, as iodides tend to be
more reactive than bromides in Stille reactions.24 Three
syntheses of 3,4-diiodofuran have previously been
reported.25 The most recent protocol, which involves
oxidative cyclisation of (E)-2,3-diiodo-2-butene-1,4-diol
with chromic acid, could not be reproduced in our hands
as decomposition of the starting material occurred before
oxidation.26 The other two procedures were either
laborious,25c or operationally unfavourable,25b so more

S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
3697
Scheme 3. Attempted synthesis of 4,5-dibenzyloxy-2-tri-n-butylstannylaminobenzene. Reagents and conditions: (i) SnCl2, ethyl acetate, MeOH, 70 8C, 1.5 h;
or FeCl3, activated carbon, N2H4·H2O, MeOH, 70 8C, 15 h; (ii) iron powder, HCl (aq.), EtOH, 80 8C, 3 h; (iii) Bu6Sn2, Pd(PPh3)4, toluene, 120 8C, 48 h.
accessible solutions to the double Stille coupling were
investigated in preference.
might slow down the oxidative addition step in the catalytic
cycle. Next, iodide 13 was prepared in the hope that it would
be sufficiently reactive to be stannylated (Scheme 4).
2.3.2. 4,5-Dibenzyloxy-2-tri-n-butylstannylaminoben-
zene. It was anticipated that reduction of the nitro group
of 4,5-dibenzyloxy-2-tri-n-butylstannylnitrobenzene 6 to an
amino group before the Stille coupling could provide two
advantages. First, the presence of the mesomerically
electron donating amino group might reduce or prevent
homocoupling, and second it should make the tin compound
more nucleophilic, causing an enhancement of the rate-
limiting transmetallation step.27 However, efforts to reduce
the nitro group of arylstannane 6 under different conditions
(SnCl2, or FeCl3, hydrazine and activated carbon) resulted
in protodestannylation (Scheme 3). To avoid this unfavour-
able side reaction, we decided that the nitro group should be
reduced prior to the introduction of the tributylstannyl
group. Thus, the nitro group of arylbromide 5 was reduced
with iron powder and hydrochloric acid in 83%.28
Iodination of 1,2-dibenzyloxybenzene 3 was most suc-
cessful using iodine activated by mercury oxide,29 yielding
78% of 3,4-dibenzyloxyiodobenzene 11. Nitration of 11
furnished 94% of nitrobenzene 12, which was reduced using
catalytic iron(III) chloride with activated carbon and
hydrazine,30 giving 4,5-dibenzyloxy-2-aminoiodobenzene
13 in 86% yield. Disappointingly, attempts to exchange the
iodine with tributyltin using Pd(PPh3)4 and hexabutylditin
proved to be unsuccessful. Significant decomposition of 13
occurred under the reaction conditions.
It became apparent that the best option was to reverse the
functional groups in the double Stille coupling. By switch-
ing the functional groups, both coupling partners would be
electronically favoured—the halide conjugated with an
electron withdrawing nitro group and the tin with an
electron donating group. Both iodo 12 and bromo 5 versions
of the required catechol moiety had already been synthe-
sised, so only 3,4-bis(tri-n-butylstannyl)furan needed to be
obtained.
Unfortunately, 4,5-dibenzyloxy-2-aminobromobenzene 10
did not undergo the desired stannylation reaction, possibly
because the mesomeric electron-donating amino group
2.3.3. 3,4-Bis(tri-n-butylstannyl)furan (14). Wong et al.
have reported a synthesis of 3,4-bis(tri-n-butylstannyl)furan
14,31 but because it is operationally unfavourable and low
yielding, the use of 14 in our synthesis did not appeal to us
initially. Consequently, we developed an improved syn-
thesis of 3,4-bis(tri-n-butylstannyl)furan 14 (Scheme 5).32
This route to 14 involved palladium catalysed addition of
hexa-n-butylditin to 2-butyne-1,4-diol 15, followed by
oxidative cyclisation and dehydration of (Z)-2,3-bis(tri-n-
butylstannyl)-2-butene-1,4-diol 16. This gave the furan 14
in 79% overall yield. With the availability of both 12 and 14,
the coupling reaction was re-investigated.
Scheme 4. 4,5-Dibenzyloxy-2-aminoiodobenzene (13) from 1,2-dibenzyl-
oxybenzene (3). Reagents and conditions: (i) I2, HgO, DCM, rt, 15 h;
(ii) HNO3 (aq.), AcOH, rt, 2 h; (iii) FeCl3, activated carbon, N2H4·H2O,
MeOH, 70 8C, 8 h; (iv) Bu6Sn2, Pd(PPh3)4, toluene, 120 8C, 48 h.
Scheme 5. An efficient synthesis of 3,4-bis(tri-n-butylstannyl)furan. Reagents and conditions: (i) Bu6Sn2, Pd(MeCN)2Cl2, THF, rt, 48 h; (ii) IBX, DMSO,
THF, rt, 2 h.

3698
S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
2.4. Double Stille coupling using 3,4-bis(tri-n-butyl-
stannyl)furan 14
the presence of acetic acid gave only the deprotected
reduced aminophenyl-furan 17 in 94% yield with no
evidence of cyclisation. To investigate the conditions
required for the desired cyclisation we decided to perform
the final three transformations—reduction, cyclisation and
deprotection—separately.
Reaction of 3,4-bis(tri-n-butylstannyl)furan 14 with 4,5-di-
benzyloxy-2-nitroiodobenzene 12 using Pd(PPh3)4 and
CuBr in THF (conditions equivalent to those used in earlier
attempts), delivered 30% of the disubstituted furan 2 after
15 h. Re-optimising the Stille conditions24—Pd2(dba)3,
AsPh3 and CuI in DMF—55% of the product was isolated
after 15 h, along with recovered starting material. Investi-
gations into methods of generally accelerating this double
coupling, led to the development of a new combination of
reagents for the Stille coupling reaction. Initial studies have
shown that the presence of cesium fluoride in conjunction
with copper(I) iodide, as a co-catalyst to Pd(PPh3)4 in DMF,
produces a large acceleration of the reaction rate. Fluoride
sources have been used before in attempts to accelerate the
Stille reaction, but not in conjunction with copper(I) salts.33
Utilising these new conditions, the desired product 2 was
isolated in 92% yield after only 2 h at 40 8C (Scheme 6).
The scope of this new combination of reagents is reported
elsewhere.34
Thus, 3,4-bis(3,4-dibenzyloxy-2-nitrophenyl)furan 2 was
reduced with tin(II) chloride under non-acidic conditions35
to give 3,4-bis(3,4-dibenzyloxy-2-aminophenyl)furan 18 in
67% yield (Scheme 8). Reduction with iron powder and HCl
later proved to give a higher yield of 18. Cyclisation was
then attempted by heating this compound to reflux in
benzene with a catalytic amount of para-toluenesulfonic
acid and powdered molecular sieves for 2 days. However,
this afforded 25% of a solid compound that was clearly not
the desired product, along with most of the remaining
starting material. After detailed analysis of the spectro-
scopic data (specifically the NOSEY and HMBC spectra), it
became apparent that the product was 3-(3,4-dibenzyloxy-
2-aminophenyl)-4-(methyl-1-hydroxy)-6,7-dibenzyloxy-
quinoline 19.
Scheme 6. Successful double Stille coupling. Reagents and conditions:
(i) Pd(PPh3)4, CuI, CsF, DMF, 40 8C, 2 h.
2.5. The reductive cyclisation of 3,4-bis(3,4-dibenzyloxy-
2-nitrophenyl)furan 2
With an effective route to the 3,4-disubstitued furan 2, the
final step (reductive-cyclisation and deprotection) was
investigated (Scheme 7).
Scheme 8. Product of the cyclisation. Reagents and conditions: (i) SnCl2,
ethyl acetate, MeOH, 70 8C, 1.5 h, 67%; or iron powder, 38% HCl, EtOH,
80 8C, 3 h; (ii) p-TSA, benzene, mol. sieves, 85 8C, 2 days.
Unfortunately hydrogenation with palladium on carbon in
The structure of this product was not immediately obvious.
The proton and HMQC spectra clearly indicated the
structure contained a methylene group with two non-
equivalent protons (J¼12 Hz). So, initially we speculated
the presence of stereocenters or ring systems that were not
flat, however, none of these proposed structures fitted all
the spectroscopic data. Eventually a deductive, stepwise
approach, based on NOE interactions and long range
proton–carbon correlations provided the correct structure.
The absence of a stereocenter or a non-planar ring system
led us to suggest that the observed magnetic non-
equivalence of the methylene group is a result of restricted
rotation around the aryl–aryl bond. If this were the case
then one would expect the chiral environment to degenerate
with heating, causing the methylene protons to become
equivalent. Recording the proton spectra at increased
temperatures demonstrates that the chiral environment is
in fact temperature dependent, supporting this proposal (the
Scheme 7. Attempted cyclisation of 3,4-bis(3,4-dibenzyloxy-2-nitrophe-
nyl)furan (2). Reagents and conditions: (i) H2, Pd/C, AcOH, THF, rt, 15 h.

S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
3699
Figure 2. Variable temperature proton spectra of 3-(3,4-dibenzyloxy-2-aminophenyl)-4-(methyl-1-hydroxy)-6,7-dibenzyloxyquinoline (19) in DMSO-d6,
showing the methylene protons becoming equivalent.
hydroxyl and amino peaks shift upfield with increased
temperature as well) (Fig. 2).
The observed NOE interactions of 19 are indicated below
(Fig. 3).
A possible mechanism that explains the formation of 19
from furan 18, is for the furan to be protonated at C-3,
followed by ring-opening of the furan, which could be
Figure 3. NOE interactions in 3-(3,4-dibenzyloxy-2-aminophenyl)-4-
(methyl-1-hydroxy)-6,7-dibenzyloxyquinoline (19).
Scheme 9. Possible mechanism for the formation of 19.

3700
S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
Figure 4. Crystal structure of 20,200,30,300-tetrakisbenzyloxy-dibenzo[c,h][2,6]naphthyridine (20).
facilitated by the nitrogen’s lone pair of electrons. An
electrocyclic reaction forms the dihydroquinoline, which
aromatises to the quinoline through loss of a proton
(Scheme 9).
The tetracycle 20 is highly crystalline and is only barely
1
soluble in DMSO-d6, nonetheless a H NMR spectra was
obtained and clearly showed that 20 is not the same as the
product of the cyclisation reaction, 3-(3,4-dibenzyloxy-2-
aminophenyl)-4-(methyl-1-hydroxy)-6,7-dibenzyloxy-
quinoline 19. However, it is reasonable to conclude that the
tetracycle 20 was formed from 19 in the solution of DMSO,
over a period of 3 months. This transformation would be the
result of air oxidation of 19 followed by cyclisation and
dehydration (Scheme 10).
Recrystallisation of 19 did not yield a satisfactory crystal
for X-ray diffraction studies, despite a number of
attempts. However, we observed the formation of a
crystalline material from an aged (3 months) NMR sample
of 19 in DMSO-d6. X-ray diffraction studies of these
crystals revealed a symmetrical tetracyclic compound
(Fig. 4).
The quinoline ring system of 19 is thermodynamically
Scheme 10. Oxidation of 19 followed by cyclisation and dehydration would give 20.

S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
3701
did not lead to any product formation. The starting material
21 was recovered in almost quantitative yield. Other
possible solutions to the problem of quinoline formation
would detract from the efficiency of the original strategy, so
a simple route to the target compound was developed using
a reductive biaryl coupling of the monomer (Scheme 12).
2.6. The reductive biaryl coupling approach
2.6.1. Synthesis of the monomer. The monomer 25 was
synthesised using slight modifications of previous reactions
(Scheme 13)
Protection of 3,4-dihydroxybenzaldehyde with benzyl
chloride and potassium carbonate in DMF provided 22 in
99% yield.36 The protected benzaldehyde 22 was subjected
to a Henry reaction by heating to reflux in acetic acid with
nitromethane, producing the nitrostyrene derivative 23 in
98% yield.37 Nitration using standard conditions gave 97%
of dinitrostyrene 24, which was reductively cyclised to 25 in
61% yield with iron powder in acetic acid. The cyclisation
protocol of Borchardt et al. was employed;38 this includes a
non-polar co-solvent like benzene with flash silica. The
relatively non-polar starting material and product are
maintained in the non-polar solvent, while the silica binds
to the polar intermediates, thus minimising the inter-
molecular reactions involving these intermediates that lead
to polymerisation. Although we noticed some improvement
in yield over the standard conditions,38 in our hands the
increase was not as dramatic as has been reported
elsewhere.38
Scheme 11. Attempted cyclisation of the diacylated bisamine 21. Reagents
and conditions: (i) Ac2O, DMAP, Et3N, THF, 15 h; (ii) p-TSA, xylene, mol.
sieves, 140 8C, 4 days.
stable under the reaction conditions and was not in
equilibrium with furan 18. This problem might be
circumvented by making the quinoline formation reversible,
which could be achieved by acylating the nitrogen,
preventing aromatisation. Treatment of diamine 18 with
acetic anhydride, N,N-dimethylaminopyridine (DMAP) and
triethylamine in THF overnight, gave 3,4-bis(3,4-dibenzyl-
oxy-2-N-acetamidophenyl)furan 21 in 79% yield
(Scheme 11).
2.6.2. Coupling of the monomer. With the monomer
realised, a triisopropylsilyl protecting group was incorpo-
rated to sterically direct iodination39 and stabilise the
required iodoindole for the biaryl coupling reaction
(Scheme 14)
As before, a solution of 21 and para-toluenesulfonic acid
in benzene was heated to reflux for 2 days, however, no
reaction was observed. Presumably the N-acetyl groups
were deactivating the nucleophilicity of the nitrogen.
Switching to higher boiling xylene as the reaction medium
Deprotonation of indole 25 with n-butyllithium followed by
addition of triisopropylsilyl chloride gave protected indole
26 in 95% yield. Iodination with mercury acetate and iodine
quantitatively yielded the iodoindole 27.39 N-Iodosuccini-
mide in THF also afforded 27, but in a slightly lower yield of
83%. The position of the iodine was confirmed by X-ray
crystallography (Fig. 5).
Homocoupling of 27 was initially attempted by formation of
the organozinc compound using butyllithium and zinc
Scheme 12. The biaryl coupling approach.
Scheme 13. Synthesis of 5,6-dibenzyloxyindole (25). Reagents and conditions: (i) BnCl, K2CO3, DMF, 120 8C, 15 h; (ii) MeNO2, NH4OAc, AcOH, 120 8C,
40 min; (iii) 70% HNO3, AcOH, rt, 2 h; (iv) iron powder, AcOH, benzene, cyclohexane, SiO2, 120 8C, 30 min.

3702
S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
Scheme 14. Coupling and deprotection. Reagents and conditions: (i) nBuLi, THF, 278 8C, 15 min, then TIPSCl, 278 8C, 2 h; (ii) I2, Hg(OAc)2, DCM, 0 8C,
2 h, 100%; or NIS, THF, rt, 20 min, 83%; (iii) Pd(PhCN)2Cl2, TDAE, DMF, 50 8C, 1.5 h; (iv) TBAF, THF, rt, 10 min; (v) H2, Pd Black, THF, 18 h.
Figure 5. Crystal structure of 5,6-dibenzyloxy-3-iodo-N-triisopropylsilylindole (27).
chloride, followed by treatment with copper(I) salts.40
However, reduction of the protected iodoindole 27 to the
protected indole 26 was the main reaction. An alternative
method for biaryl coupling was investigated using a
Grignard reagent and catalytic palladium.41 Formation of
the Grignard reagent by treatment of 27 with iPrMgCl, was
followed by the addition of a solution of 27 and Pd(dppf)2-
Cl2. This time the coupled product 28 was formed in 19%
yield, but again the main product was indole 26. Eventually,
the coupling was achieved in 68% using catalytic
Pd(PhCN)2Cl2 and the mild reductant tetrakis(dimethyl-
amino)ethylene (TDAE).42 Desilylation of 28 with tetra-
butylammonium fluoride (TBAF) afforded 82% of the
benzyl protected bisindole 29. X-ray diffraction studies
showed indeed the desired product 29 was obtained in the
coupling reaction (Fig. 6).
Finally, hydrogenation with catalytic Palladium Black in
THF revealed 5,50,6,60-tetrahydroxy-3,30-bisindole 1 in 94%
yield, the proposed structure of the natural product isolated
from beetroot.
2.7. Spectroscopic analysis of 5,50,6,60-tetrahydroxy-3,30-
biindolyl 1
The spectro0scopic data we obtained from 5,50,6,60-tetra-
hydroxy-3,3 -biindolyl 1 showed subtle differences to the
data recorded from the natural product. A plausible

S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
3703
Figure 6. Crystal structure of 5,50,6,60-tetrabenzyloxy-3,30-biindolyl (29).
alternative structure for the natural product is not immedi-
ately obvious, and so it remains inconclusive as to whether
or not the natural product is 5,50,6,60-tetrahydroxy-3,30-
biindolyl 1.
(THF) had been removed during work up (see Section 4).
Differences are apparent between the 13C spectra of 1 and
the natural product (Table 2).
Table 2. Comparison of 13C NMR data from compound 1 and from the
natural product
2.7.1. Mass spectrometry. The high resolution mass
spectrum of synthetic 5,50,6,60-tetrahydroxy-3,30-biindolyl
1 was obtained under electrospray (ES2, CV 230)
conditions and gave the parent ion as m/z 295.0717,
M2Hþ requires m/z 295.0719. High resolution mass
measurements on the natural product have only been
obtained for fragments corresponding to the monomer.7
The parent ion apparently was not forthcoming, although on
occasion ESIþ and ESI2 gave peaks at m/z 297 and 295,
respectively.
dC Compound 1a
Assignmentb
d
C Natural productc
Assignmentc
98.48
C7, 70
C4, 40
C3, 30
C3a, 3a0
C2, 20
C7a, 7a0
C5, 50
C6, 60
101.00
103.11
108.54
123.75
127.47
133.61
142.11
144.37
C7, 70
C3, 30
C4, 40
C3a, 3a0
C2, 20
C7a, 7a0
C5, 50
C6, 60
105.19
109.10
119.50
121.20
131.57
139.76
142.21
a
Spectra recorded in D2O at 125.8 MHz.
Assigned with the help of HMBC and HMQC experiments.
Taken from Ref. 7.
2.7.2. UV spectroscopy. The ultraviolet spectrum of 1 in
water showed one lmax at 302 nm, while the spectrum
recorded for the natural product showed two lmax at 304 and
278 nm.
b
c
In particular the peak assigned as C-3 (109.10 ppm) is
clearly observed in the spectrum of 1, but barely shows up in
the natural product spectrum. It is evidently there in the
natural product because it gives a correlation in the HMBC
spectrum (Fig. 7).
2.7.3. 1H NMR. The 1H NMR spectrum of 1 in D2O showed
three signals, but at different chemical shifts to those
reported for the natural product in D2O (Table 1).
Table 1. Comparison of 1H NMR data from compound 1 and from the
natural product
Kujala et al. explain that the virtual absence of the C-3 peak
in the 13C spectrum is due to an extremely long relaxation
time. Because the reported recycle time of the 13C
experiment was 3.8 s, we propose instead that the signal
was not detected due to broadening under the conditions
of the NMR experiment. The HMBC of 1 showed a
very similar pattern to the natural product. However, the
synthetic compound shows a strong correlation between H4
and C3, where the natural product does not. Furthermore,
the correlations from H4 to C5, H7 to C6 and H7 to C7a do
not disappear at higher thresholds, as they do in the
spectrum of the natural product.7 There is a difference
between the optimised coupling constants in the two HMBC
spectra (7.1 Hz for compound 1 vs 8.0 Hz for the natural
product), but the difference would seem to be too small to
explain the differences in the correlations outlined above.
This is because a correlation is strong when its coupling
constant is close to the optimised coupling constant used in
dH Compound 1a
Assignmentb dH Natural productc Assignmentc
6.94 (d, J¼0.3 Hz) H7, 70
7.13 (d, J¼0.3 Hz) H4, 40
7.03 (d, J¼0.3 Hz)
7.12 (d, J¼0.3 Hz)
7.22 (s)
H7, 70
H4, 40
H2, 20
7.32 (s)
H2, 20
a
Spectra recorded in D2O at 500 MHz.
Assigned with the help of HMBC and HMQC experiments.
Taken from Ref. 7.
b
c
2.7.4. 13C and HMBC NMR. 5,50,6,60-Tetrahydroxy-3,30-
biindolyl 1 does not dissolve in D2O sufficiently to allow the
preparation of an adequate sample for 13C analysis (4000
scans at 500 MHz gave no signals). The natural product,
however, is sufficiently soluble in D2O to give clear 13C
signals. To prepare a sample of 1 in D2O for 13C analysis, it
was necessary to add the D2O before all the reaction solvent

3704
S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
Figure 7. Comparision of HMBC spectra from the natural product (left) and from compound 1 (right). The correlations in the natural product spectrum marked
with a cross are unsuppressed one-bond couplings.
the HMBC experiment, so the strong correlation seen in the
HMBC of 1 between H4 and C3 will have a coupling
constant near our optimised coupling constant of 8.0 Hz.
Thus, it would be expected to be observed in an HMBC
experiment with a similar optimised coupling constant of
7.1 Hz.
the variations in pH, concentration and solvent, we
speculated that the natural compound might possibly be
an oxidation product of 5,50,6,60-tetrahydroxy-3,30-biindolyl
1. The 1H NMR and UV spectrum of 1 was recorded every
few hours over a period of 24 h. The three signals in the
proton NMR spectrum decreased in intensity, but there was
no appearance of signals corresponding to the natural
product. The UV spectrum also decreased in intensity with
no appearance of the reported lmax at 278 nm. To preclude
the possibility that the oxidation product that forms might
decompose before accumulating to a detectable level, a
sample of 1 was oxidised with H2O2. The signals slowly
disappeared and a black precipitate was formed over a
period of 1 h with no transient intermediate observed
by either 1H NMR or UV. Oxidation with 1 equiv. of
dichlorodicyanoquinone (DDQ) in deuterated acetonitrile
immediately gave a black precipitate.
2.7.5. 1H NMR and UV under different conditions.
Because an alternative structure for the natural product is
not immediately obvious, the possibility that the differences
in the spectra are due to other reasons was explored. The 1H
NMR spectrum and the UV spectrum of 1 were recorded
under a variety of different conditions.
2.7.6. 1H NMR under different conditions. As a measure
of the differences in the proton NMR spectrum, the three
signals from the natural product have an overall spread of
ca. 0.2 ppm, while the three signals from 1 have an overall
spread of ca. 0.4 ppm. Under acidic or basic conditions
(achieved by the addition of various amounts of formic
acid—used in the isolation of the natural product—or
sodium hydroxide, respectively), more concentrated con-
ditions, doping with MeCN (also used in the isolation
procedure), and recording spectra in deuterated methanol or
deuterated DMSO, the overall spread of the three signals
from 1 remained at ca. 0.4 ppm. Although, not surprisingly,
the signals shifted along the spectrum slightly when the
conditions were varied.
3. Conclusion
There are clear spectroscopic differences between the
natural product and 5,50,6,60-tetrahydroxy-3,30-biindolyl 1.
Varying the conditions of data collection for 1 did not
resolve these differences. Oxidation of 1 gave a black
precipitate and no transient intermediate corresponding to
the natural product. These observations seem0to suggest that
the0 natural product is unlikely to be 5,50,6,6 -tetrahydroxy-
3,3 -biindolyl 1. However, there are no obvious alternative
structures that are plausible. It is difficult to interpret the
HMBC spectrum from the natural product without resorting
to a 3,5,6-trisubstituted indole or benzofuran nucleus; how-
ever, a benzofuran structural isomer seems biogenetically
less likely. The 5,50- or 6,60-linked bisindole isomers require
a 3-hydroxy group, but these compounds would most likely
exist in the keto-form (Kenol(H2O) for indoxyl is 0.086).43
Thus without further investigation it remains0 unclear
whether or not the natural product is 5,50,6,6 -tetrahy-
droxy-3,30-biindolyl 1.
2.7.7. UV spectrum under different conditions. The
natural product gives two lmax at 304 and 278 nm. Under
neutral conditions 5,50,6,60-tetrahydroxy-3,30-biindolyl 1
gives one lmax at 302 nm. This lmax shifted to 300 nm in
0.1% formic acid solution and to 324 nm in 0.1% sodium
hydroxide solution. There was no appearance of the second
reported lmax at 278 nm even when the concentration of
acid or base was increased.
2.7.8. Oxidation studies. As the differences between in the
spectral data of 1 and the natural product persisted despite

S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
3705
4. Experimental
4.1. General experimental
dichloromethane (DCM) was distilled from calcium hydride
under nitrogen. PE refers to the fraction of light petroleum
ether boiling between 40 and 60 8C, and was distilled before
use. Triethylamine, dimethyl formamide (DMF), dimethyl
sulfoxide and N-methylpyrrolidine (NMP) were distilled
from calcium hydride under argon or reduced pressure
and stored over 4 A molecular sieves under argon until used.
˚
Toluene was dried over 4 A molecular sieves under argon.
All water used was distilled except where otherwise
indicated. Solvents were evaporated on a Bu¨chi R110
Rotavaporator.
Proton magnetic resonance spectra were recorded on a
Bru¨ker DPX200 (200 MHz), Bru¨ker DPX400 (400 MHz),
Bru¨ker AMX500 (500 MHz) and Bru¨ker DPX500
(500 MHz) spectrometers at ambient temperatures. Proton
spectra assignments are supported by 1H–1H COSY where
necessary. Chemical shifts (dH) are reported in parts per
million (ppm) and are referenced according to IUPAC
recommendations, 2001.44 Coupling constants (J) are
recorded to the nearest 0.5 Hz.
˚
4.2. Experimental procedure
Carbon magnetic resonance spectra were recorded on
Bru¨ker DPX200 (50.3 MHz), Bru¨ker DPX400 (100.6
MHz), Bru¨ker AMX500 (125.8 MHz) and Bru¨ker
DPX500 (125.8 MHz), spectrometers at ambient tempera-
tures. Chemical shifts (dC) are reported in parts per million
(ppm) and are referenced according to IUPAC recommen-
dations, 2001.44 Carbon spectra assignments are supported
by DEPT analysis and 13C–1H correlations were necessary.
4.2.1. 1,2-Dibenzyloxybenzene (3). A mixture of catechol
(10.0 g, 90.8 mmol), anhydrous K2CO3 (38.0 g, 272.5
mmol) and benzyl chloride (31.0 mL, 269.4 mmol) was
stirred rapidly in acetone (140 mL) and heated to reflux
under argon for 4 days. The mixture was filtered and the
solvent removed under reduced pressure. The residue was
dissolved in DCM (200 mL) and refiltered, then the solvent
was removed under reduced pressure. Recrystallisation
from DCM/PE gave 3 (22.9 g, 87%). A small amount was
further purified by column chromatography (PE/EtOAc,
9:1) to give a white solid. Rf 0.29 (PE/EtOAc, 9:1); mp 58–
59 8C, lit.15 58–59 8C; dH (400 MHz, CDCl3) 5.24 (4H, s,
CH2 of Bn), 6.95–7.00 (2H, m, H4, 5), 7.02–7.06 (2H, m,
H3, 6), 7.36–7.57 (10H, m, CH of Bn).
Low resolution mass spectra were recorded using a TRIO-1
GCMS spectrometer, a Micromass Platform (APCI)
Spectrometer, Micromass Autospec spectrometer (CIþ)
and a micromass ZAB spectrometer (CIþ, EI). Only
molecular ions (Mþ), fragments from molecular ions and
other major peaks are reported. High-resolution mass
spectra were recorded on a Micromass Autospec spectro-
meter and are accurate to ^10 ppm.
4.2.2. 3,4-Dibenzyloxybromobenzene (4). To a solution of
1,2-dibenzyloxybenzene 3 (10.0 g, 34.4 mmol) in CCl4
(35 mL) was added NBS (7.4 g, 41.6 mmol). After initiating
of the reaction by heating, the heat was removed. The
reaction boiled vigorously for 5–10 min without heating.
After the reaction subsided, the solution was heated to reflux
for 1 h, then diluted with DCM (50 mL), washed with water
(100 mL), 1 M NaOH (50 mL) and again with water
(100 mL). The organic layer was dried over Na2SO4 and
MgSO4, and the solvent removed. The residue was
recrystallised from DCM/MeOH to give 4 (11.3 g, 89%).
A small amount was further purified by column chroma-
tography (PE/EtOAc, 9:1) to give a white solid. Rf 0.32 (PE/
EtOAc, 9:1); mp 62–63 8C, lit.15 64–66 8C; m/z 386.0756,
found 386.0753; microanalysis requires C 65.05, H 4.64,
found C 65.08, H 4.59; dH (400 MHz, CDCl3) 5.15, (2H, s,
CH2 of Bn), 5.16 (2H, s, CH2 of Bn), 6.83 (1H, d, J¼8.5 Hz,
H5), 7.04 (1H, dd, J1¼8.5 Hz, J2¼2.5 Hz, H6), 7.12 (1H, d,
J¼2.5 Hz, H2), 7.33–7.50 (10H, m, CH of Bn).
Microanalyses were carried out by Elemental Microanalysis
Limited, and are quoted to the nearest 0.1% for all elements
except hydrogen, which is quoted to the nearest 0.05%.
Infrared spectra were recorded on a Perkin–Elmer Paragon
1000 Fourier Transform spectrometer as a thin film between
NaCl plates, or as KBr discs. Absorption maxima (nmax) of
the major peaks are reported in wavenumbers (cm21).
Ultraviolet spectra were recorded on a Perkin–Elmer
Lambda 2 UV/VIS spectrometer in ethanol or water as
indicated at ambient temperature. Absorption maxima
(nmax) are reported in nanometers (nm) and extinction
coefficients (1) are quoted to four significant figures.
Melting points were measured using a Cambridge Instru-
ments Gallene III hot stage melting point apparatus and are
uncorrected.
4.2.3. 4,5-Dibenzyloxy-2-nitrobromobenzene (5). 3,4-Di-
benzyloxybromobenzene 4 (9.0 g, 24.4 mmol) was dis-
solved in hot glacial acetic acid (120 mL). The solution was
cooled to 35 8C and 70% HNO3 (7.0 mL, 110.4 mmol) was
added over 5 min. A yellow solid was precipitated and the
mixture was stirred at room temperature for 2 h. Water
(200 mL) was added and the yellow solid collected was
dissolved in DCM then washed with K2CO3 solution until
the aqueous layer remained basic. The organic layer was
dried over Na2SO4/MgSO4, and the solvent was removed
under reduced pressure. The residue was recrystallised from
DCM/MeOH to give 5 (8.6 g, 85%). A small amount was
further purified by column chromatography (PE/EtOAc,
6:1) to give a pale yellow solid. Rf 0.26 (PE/EtOAc, 6:1);
Thin layer chromatography (TLC) was performed using
Merck aluminium foil backed plates pre-coated with silica
gel 60 F254 (1.05554). Visualisation was affected by
quenching of UV fluorescence (lmax¼254 nm), staining
with phosphomolybdic acid in ethanol, followed by heating.
Retention factors (Rf) are reported to two decimal places.
Column chromatography was performed using ICN silica
˚
32–63, 60 A.
Anhydrous tetrahydrofuran (THF) was distilled over
sodium/benzophenone ketyl under nitrogen and anhydrous

3706
S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
mp 104–105 8C, lit.16 105–107 8C; dH (400 MHz, CDCl3)
5.18, (2H, s, CH2 of Bn), 5.21 (2H, s, CH2 of Bn), 7.20
(1H, s, H6), 7.33–7.47 (10H, m, CH of Bn), 7.65 (1H, s,
H-3).
through celite. The organic layer was dried over Na2SO4/
MgSO4 and the solvent was removed under reduced
pressure. The residue was purified by column chroma-
tography (DCM/PE, 7:3) to give 2 (33 mg, 10%) as a pale
orange solid. Also isolated from the reaction was 3-bromo-
4-(3,4-dibenzyloxy-2-nitrophenyl)f0uran (8) (0.030 g, 14%)
as a pale orange solid and 4,40,5,5 -tetrakisbenzyloxy-2,20-
dinitro-biphenyl (9) (0.258 g, 58%) as a yellow solid.
4.2.4. 4,5-Dibenzyloxy-2-tri-n-butylstannylnitrobenzene
(6). A solution of 4,5-dibenzyloxy-2-nitrobromobenzene 5
(2.00 g, 4.83 mmol), Bu6Sn2 (3.60 mL, 7.15 mmol) and
Pd(PPh3)4 (0.28 g, 0.24 mol) was heated to reflux in toluene
(10 mL) under argon for 48 h. The cooled solution was
stirred vigorously with saturated KF solution (10 mL) for
1 h, and then filtered through celite with DCM (100 mL)
washings. The filtrate was washed with water (50 mL) and
dried over Na2SO4/MgSO4. The solvent was removed under
reduced pressure and the residue was purified by column
chromatography (PE/EtOAc, 9:1) to give 6 (2.05 g, 68%) as
a yellow oil. Rf 0.46 (PE/EtOAc, 9:1); nmax/cm21 (thin film)
2955 (s), 2921 (s), 2870 (s), 2852 (s), 1564 (m), 1515 (s, NO2
str), 1454 (m), 1321 (m, NO2 str), 1272 (s), 1205 (m),
1021 (s), 735 (m), 696 (m); m/z probe ESþ
(M(118Sn)Hþ2C4H10) 565.8 (100%), 278.4 (65%), HRMS
(M(118Sn)Hþ2C4H10) requires m/z 566.1504, found
566.1519; dH (400 MHz, CDCl3) 0.91 (9H, t, J1¼7.5 Hz,
CH3), 1.00–1.20 (6H, m, SnCH2), 1.27–1.39 (6H, sextet,
J1¼7.5 Hz, CH2CH3), 1.39–1.60 (6H, m, CH2CH2CH3),
5.25, (2H, s, CH2 of Bn), 5.33 (2H, s, CH2 of Bn), 7.08 (1H,
s, H3), 7.33–7.54 (10H, m, CH of Bn), 8.02 (1H, s, H6);
dC (100.6 MHz, CDCl3) 11.06 (SnCH2), 13.71 (CH3),
27.32, (CH2CH3), 29.06, (CH2CH2CH3), 70.97 and 71.10
(CH2 of Bn), 110.1 (C6), 120.3 (C3), 127.0, 127.4, 128.2,
128.2, 128.4, 128.7 and 128.7 (CH of Bn), 133.9 (quat. C),
136.2 and 136.2 (ipso of Bn), 146.6, 148.6 and 153.2
(quat. C).
Method B. A solution of 3,4-bis(tri-n-butylstannyl)furan 14
(1.00 g, 1.55 mmol), 4,5-dibenzyloxy-2-nitroiodobenzene
12 (1.57 g, 3.40 mmol) and CsF (1.06 g, 7.00 mmol) in
DMF (20 mL) was sonicated for a few minutes. CuI (0.29 g,
1.52 mmol) and Pd(PPh3)4 (0.36 g, 0.31 mmol) were added
and the mixture was stirred at 40 8C under argon for 2 h. The
mixture was diluted with DCM (50 mL) and water (50 mL)
then filtered through celite. The organic layer was dried over
Na2SO4/MgSO4 and the solvent removed. The residue was
purified by column chromatography (DCM/PE, 7:3) to give
2 (1.05 g, 92%) as a pale orange solid. Rf 0.44 (DCM/PE,
7:3); mp 55.5–56 8C; nmax/cm21 (KBr) 3089 (w), 3063 (w),
3031 (w), 2931 (w), 2867 (w), 1573 (m), 1519 (s, NO2 str),
1454 (m), 1342 (s, NO2 str), 1280 (s), 1203 (m), 1086 (m),
1023 (m), 868 (m), 738 (m), 696 (m); lmax (EtOH) 245 þ(1
10,390), 290 (1 5200), 340 (1 3780); m/z probe (MNH4
)
752.3 (100%), 702.2 (33%), HRMS (MNH4þ) requires m/z
752.2608, found 752.2612; dH (400 MHz, CDCl3) 5.15,
(4H, s, CH2 of Bn), 5.17 (4H, s, CH2 of Bn), 6.87 (2H, s,
H60, 600), 7.30–7.53 (24H, m, H2, H5, H30, H300, CH of Bn);
dC (100.6 MHz, CDCl3) 71.1 and 71.4 (CH2 of Bn), 110.4
(C30, 300), 117.0 (C60, 600), 120.8 and 123.7 (quat. C), 127.2,
127.3, 127.4, 127.5, 128.2, 128.2, 128.3, 128.6 and 128.(CH
of Bn), 135.7 and 135.9 (ipso of Bn), 140.0 (C2, 5), 141.6,
147.9 and 152.3 (quat. C) (Fig. 8).
4.2.5. 3,4-Dibromofuran (7).15 (E)-2,3-Dibromo-2-butene-
1,4-diol (20.0 g, 81.3 mmol) and 7% H2SO4 (50 mL) was
added to a flask with distillation apparatus attached. The
mixture was rapidly stirred at 110 8C to begin distillation.
A solution of K2Cr2O7 (25.1 g, 85.4 mmol) and H2SO4
(16.1 mL, 300.4 mmol) in water (160 mL) was then added
over 1 h using a dropping funnel while distillation
continued. After the chromic acid solution had been
added, the mixture was further distilled for one more
hour. The product was extracted from the distillate with PE
(2£100 mL) and the organic layers were washed with sat.
Na2CO3 solution, dried over Na2SO4/MgSO4 and the
solvent was removed under reduced pressure. Purification
by column chromatography (PE) gave 7 (10.3 g 56%) as a
colourless liquid. Rf 0.46 (PE); nmax/cm21 (thin film) 3150
(m), 1795 (w), 1639 (w), 1543 (m), 1330 (m), 1215 (m),
1140 (m) 1037 (m), 972 (s), 865 (s), 783 (s), 589 (s); m/z
probe CIþ (M(79,79Br)zþ) 223.9, HRMS (M(79,79Br)zþ)
requires m/z 223.8472, found 223.8483; dH (200 MHz,
CDCl3) 7.46 (2H, s, H2, 5); dC (50.3 MHz, CDCl3) 104.0
(C3, 4), 141.6 (C2, 5).
Figure 8.
4.2.7. 3-Bromo-4-(3,4-dibenzyloxy-2-nitrophenyl)furan
(8). Rf 0.30 (DCM/PE, 6:4); mp 108–109 8C; nmax/cm21
(KBr) 3147 (w), 3033 (w), 2921 (w), 1588 (m), 1514 (s,
NO2 str), 1452 (m), 1330 (s, NO2 str), 1292 (s), 1270 (s),
1222 (s), 1063 (s), 1044 (s), 1007 (m), 872 (s), 812 (m), 734
(s), 695 (s), 591 (m); m/z probe CIþ (MHþ) 480.1 (50%),
452.1 (15%), 353.2 (20%), 336.1 (25%), 279.1 (100%),
263.1 (35%), 108.1 (53%), 91.0 (72%), HRMS (MHþ)
requires m/z 480.0447, found 480.0454; dH (400 MHz,
CDCl3) 5.25, (2H, s, CH2 of Bn), 5.26 (2H, s, CH2 of Bn),
6.85 (1H, s, H60), 7.33–7.52 (12H, m, H2, H5, CH of Bn),
7.77 (1H, s, H30); dC (100.6 MHz, CDCl3) 71.2 and 71.4
(CH2 of Bn), 102.3 (CH), 110.7 (C60), 116.9 (C30), 119.7
and 124.3 (quat. C), 127.3, 127.4, 128.3 and 128.7 (CH of
Bn), 135.7 and 135.8 (ipso of Bn), 140.3 and 141.5 (C2, 5),
141.6, 148.4 and 152.2 (quat. C) (Fig. 9).
4.2.6. 3,4-Bis(3,4-dibenzyloxy-2-nitrophenyl)furan (2).
Method A. A solution of 3,4-dibromofuran 7 (0.100 g,
0.443 mmol),
4,5-dibenzyloxy-2-tri-n-butylstannylnitro-
benzene 6 (0.829 g, 1.328 mmol), CuBr (13 mg, 0.089
mmol) and Pd(PPh3)4 (0.102 g, 0.089 mmol) in THF 5 mL
was stirred at 40 8C under argon for 15 h. The mixture was
diluted with DCM (50 mL) and water (50 mL) then filtered

S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
3707
further purified by column chromatography (PE/EtOAc,
9:1) to give a white solid. Rf 0.33 (PE/EtOAc, 9:1); mp 63–
65 8C, lit.29 65–67 8C; nmax/cm21 (KBr) 3062 (w), 3033
(w), 2935 (w), 2870 (w), 1575 (m), 1499 (s), 1454 (m), 1394
(m), 1382 (m), 1247 (s), 1206 (s), 1133 (s), 998 (s), 798 (s),
751 (s), 698 (s); m/z probe CIþ (MNH4þ) 434.1 (100%),
(MNH4þ2I) 308.2 (55%), 290.2 (20%), 108.1 (12%), 91.1
(22%), HRMS (MNH4þ) requires m/z 434.0617, found
434.0611; dH (400 MHz, CDCl3) 5.13 (2H, s, CH2 of Bn),
5.15 (2H, s, CH2 of Bn), 6.70 (1H, d, J¼8.5 Hz, H5), 7.22
(1H, dd, J1¼8.5 Hz, J2¼2.0 Hz, H6), 7.27 (1H, d, J¼2.0 Hz,
H2), 7.31–7.50 (10H, m, CH of Bn); dC (100.6 MHz,
CDCl3) 71.30 and 71.46 (CH2 of Bn), 83.26 (C1), 117.0
(C5), 124.0 (C2), 127.3, 127.4, 127.9, 128.0, 128.6 and
128.6 (CH of Bn), 130.5 (C6), 136.7 and 136.9 (ipso of Bn),
149.1 and 149.9 (quat. C).
Figure 9.
4.2.8. 4,40,5,50-Tetrakisbenzyloxy-2,20-dinitro-biphenyl
(9). Rf 0.30 (DCM/PE, 8:2); mp 189–190 8C; nmax/cm21
(KBr) 3063 (w), 3034 (w), 2915 (w), 1574 (m), 1518 (s,
NO2 str), 1454 (m), 1382 (m), 1333 (s, NO2 str), 1270 (s),
1219 (s), 106þ6 (m), 1021 (m), 736 (m), 696 (m); m/z probe
CIþ (MNH4 ) 686.0 (33%), (MHþ) 669.0 (8%), 639.0
(16%), þ623.0 (40%), 106.2 (100%), 91.2 (96%), HRMS
(MNH4 ) requires m/z 686.2502, found 686.2534; dH
(400 MHz, CDCl3) 5.14, (4H, d, J¼12.0 Hz, CH2 of Bn),
5.26 (4H, d, J¼12.0 Hz, CH2 of Bn), 6.66 (2H, s, H6, 60),
7.33–7.53 (20H, m, CH of Bn), 7.89 (2H, s, H3, 30); dC
(100.6 MHz, CDCl3) 71.22 and 71.41 (CH2 of Bn), 110.6
(C3, 30), 114.8 (C6, 60), 127.3, 127.5, 128.3 and 128.7 (CH
of Bn), 129.1 (quat. C), 135.6 and 135.9 (ipso of Bn), 139.8,
147.9 and 152.8 (quat. C).
4.2.11. 4,5-Dibenzyloxy-2-nitroiodobenzene (12). 3,4-Di-
benzyloxyiodobenzene 11 (50 g, 120 mol) was dissolved in
hot glacial acetic (600 mL). The solution was cooled to
35 8C and 70% HNO3 (35 mL, 552 mmol) was added over
30 min. A yellow solid precipitated and the mixture was
stirred at room temperature for 2 h. Water (1 L) was added
and the yellow solid collected was dissolved in DCM then
washed with sat. K2CO3 solution until the aqueous layer was
basic. The organic layer was dried over Na2SO4/MgSO4 and
the solvent was removed under reduced pressure. The
residue was recrystallised from DCM/MeOH to give 12
(52 g, 94%). A small amount was further purified by column
chromatography (PE/EtOAc, 6:1) to give a yellow solid. Rf
0.27 (PE/EtOAc, 6:1); mp 108.5–109 8C; nmax/cm21 (KBr)
3056 (w), 3027 (w), 2899 (w), 2858 (w), 1588 (w), 1575
(m), 1514 (s, NO2 str), 1498 (s), 1450 (m), 1316 (s), 1266
(s), 1210 (m), 1196 (m), 1011 (m), 859 (þm), 749 (m), 732
(m), 699 (m); m/z probe CIþ (MNH4 ) 479.0 (20%),
4.2.9. 4,5-Dibenzyloxy-2-aminobromobenzene (10). To a
solution of 4,5-dibenzyloxy-2-nitrobromobenzene 5 (3.0 g,
7.23 mmol) and 35% HCl (3.00 mL) in ethanol (25 mL),
was added iron powder (1.21 g, 21.67 mmol). The mixture
was heated to reflux under argon for 3 h. The cooled
solution was diluted with DCM (100 mL) and water
(100 mL), and then filtered through celite, washing with
DCM (100 mL). The organic layer was separated and again
water (100 mL) was added. K2CO3 was added until no more
bubbling occurred and the aqueous layer was basic. The
organic layer was separated, dried over Na2SO4/MgSO4 and
the solvent removed under reduced pressure. Purification by
column chromatography (PE/EtOAc, 8:2) gave 10 (2.3 g,
83%) as a tan solid. Rf 0.19 (PE/EtOAc, 8:2); mp 98–99 8C;
(MHþ) 462.0 (7%), 432.0 (100%), 340.0 (30%), 305.2
þ
(67%), 91.1 (80%), HRMS (MNH4
) requires m/z
479.0468, found 479.0452; microanalysis requires C
52.08, H 3.50, N 3.04, found C 52.40, H 3.46, N 3.05; dH
(400 MHz, CDCl3) 5.19 (2H, s, CH2 of Bn), 5.20 (2H, s,
CH2 of Bn), 7.33–7.47 (10H, m, CH of Bn), 7.49 (1H, s,
C6), 7.68 (1H, s, C3); dC (100.6 MHz, CDCl3) 71.31 and
71.41 (CH2 of Bn), 77.60 (C1), 111.7 (C3), 125.5 (C6),
127.3, 127.4, 128.4, 128.5, 128.7 and 128.8 (CH of Bn), 135.3
and 135.6 (ipso of Bn), 145.1, 148.5 and 152.6 (quat. C).
n
max/cm21 (KBr) 3448 and 3366 (w, N–H str), 3063 (w),
3033 (w), 2926 (w), 2868 (w), 1614 (m), 1596 (w), 1508 (s),
1452 (m), 1409 (m), 1381 (m), 1259 (m), 1225 (m), 1210
(m), 1178 (s), 1023 (m), 1007 (m), 982 (m), 917 (m), 873
(m), 846 (m), 763 (m), 750 (m), 734 (m), 700 (s); m/z probe
ESþ (M(79Br)Hþ) 384.1 (85%), HRMS (M(79Br)Hþ)
requires m/z 384.0599, found 384.0595; dH (400 MHz,
CDCl3) 3.64 (2H, br, NH2), 5.04, (2H, s, CH2 of Bn),
5.09 (2H, s, CH2 of Bn), 6.43 (1H, s, H3), 7.06 (1H, s, H6),
7.30–7.45 (10H, m, CH of Bn); dC (100.6 MHz, CDCl3)
71.31 and 73.07 (CH2 of Bn), 99.49 (C1), 103.4 (C3), 121.0
(C6), 127.3, 127.7, 127.9, 127.9, 128.4 and 128.5 (CH of Bn),
136.9 and 137.2 (ipso of Bn), 139.1, 142.0 and 149.9 (quat. C).
4.2.12. 4,5-Dibenzyloxy-2-aminoiodobenzene (13). A
mixture of 4,5-dibenzyloxy-2-nitroiodobenzene 12 (2.00 g,
4.33 mmol), activated carbon (0.21 g, 17.50 mmol), and
FeCl3 (70 mg, 0.43 mmol) in MeOH (15 mL) was heated to
reflux under argon for 10 min. N2H2·H2O (0.84 mL,
17.23 mmol) was added slowly and the mixture heated to
reflux for 8 h. The cooled solution was diluted with DCM
(50 mL) and water (50 mL), then filtered through celite,
washing with DCM (100 mL). The organic layer was
separated, dried over Na2SO4/MgSO4, and the solvent was
removed under reduced pressure. The residue was purified
by column chromatography (PE/EtOAc, 8:2) to give 13
(1.60 g, 86%) as a pale yellow solid. Rf 0.19 (PE/EtOAc,
8:2); mp 92–94 8C; nmax/cm21 (KBr) 3437 and 3353 (m,
NH str), 3061 (w), 3032 (w), 2926 (w), 2868 (w), 1609 (m),
1506 (s), 1404 (m), 1381 (m), 1255 (s), 1226 (m), 1210 (m),
1177 (s), 1018 (m), 1006 (m), 982 (m), 847 (m), 765 (m),
4.2.10. 3,4-Dibenzyloxyiodobenzene (11). 1,2-Dibenzyl-
oxybenzene 3 (50 g, 172 mmol), HgO (41 g, 189 mmol),
and I2 (48 g, 189 mmol), were stirred in DCM (700 mL) at
room temperature for 24 h. The solution was filtered then
washed with sat. Na2S2O3 solution (200 mL). The organic
layer was dried over Na2SO4/MgSO4 and the solvent was
removed under reduced pressure. Recrystallisation from
DCM and MeOH gave 11 (56 g, 78%). A small amount was

3708
S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
750 (m), 731 (m), 700 (s); m/z probe ESþ (MHþ) 432.1
(100%), (MHþ2NH2) 415.0 (15%), 391.3 (17%), HRMS
(MHþ) requires m/z 432.0461, found 432.0461; dH
(400 MHz, CDCl3) 3.82 (2H, br, NH2), 5.04, (2H, s, CH2
of Bn), 5.09 (2H, s, CH2 of Bn), 6.43 (1H, s, H3), 7.27 (1H,
s, H6), 7.32–7.50 (10H, m, CH of Bn); dC (100.6 MHz,
CDCl3) 71.12 (CH2 of Bn), 72.28 (C1), 73.20 (CH2 of Bn),
102.3 (C3), 126.8 (C6), 127.3, 127.8, 127.9, 128.0, 128.5
and 128.6 (CH of Bn), 136.9 and 137.3 (ipso of Bn), 142.3,
142.3 and 151.0 (quat. C).
and the solvent was removed under reduced pressure.
Purification by column chromatography (PE/EtOAc, 1:1)
gave 18 (0.61 g, 88%) as a pale brown solid. Rf 0.38 (PE/
EtOAc, 1:1); mp 137–138.5 8C; nmax/cm21 (KBr) 3455,
3397, 3364 and 3324 (m, NH str), 3061 (w), 3032 (w), 2917
(w), 2867 (w), 1616 (m), 1552 (m), 1505 (s), 1454 (m), 1418
(m), 1370 (m), 1354 (m), 1262 (s), 1204 (s), 1171 (s), 1146
(s), 1054 (m), 1024 (m), 994 (m), 854 (m), 826 (m), 735 (s),
696 (s); lmax (EtOH) 308 (1 5560); m/z probe ESþ (MHþ)
674.0 HRMS (MHþ) requires m/z 675.2859, found
675.2864; dH (400 MHz, CDCl3) 3.37 (4H, br, NH2), 4.87,
(4H, s, CH2 of Bn), 5.08 (4H, s, CH2 of Bn), 6.32 (2H, s,
H30, 300), 6.56 (2H, s, H60, 600), 7.24–7.50 (20H, m, CH of
Bn), 7.56 (2H, s, H2, 5); dC (100.6 MHz, CDCl3) 71.05 and
72.70 (CH2 of Bn), 103.3 (C30, 300), 109.6 (quat. C), 119.4
(C60, 600), 123.0 (quat. C), 127.2, 127.5, 127.6, 127.8, 128.3
and 128.5 (CH of Bn), 137.2 and 137.7 (ipso of Bn), 139.4
(quat. C), 141.2 (C2, 5), 141.4 and 149.9 (quat. C) (Fig. 11).
4.2.13. 3,4-Bis(3,4-dihydroxy-2-aminophenyl)furan (17).
3,4-Bis(3,4-dibenzyloxy-2-nitrophenyl)furan
2
(51 mg,
0.07 mmol) and AcOH (4 mL, 0.07 mmol) were stirred in
THF (1 mL) with 10% palladium on carbon (20 mg) under
an atmosphere of H2 for 15 h. The solution was filtered and
the solvent removed to give 17 (21 mg, 94%) as a pale
yellow solid. The lability of the compound prevented
purification. Mp decomposed .190 8C; nmax/cm21 (KBr)
3374 (s, br, OH str), 1557 (m), 1514 (s), 1451 (m), 1308 (m),
1243 (m), 1052 (m), 876 (m), 804 (m); lmax (EtOH) 313 (1
45,850); dH (500 MHz, CD3OD) 6.33 (2H, s, H30, 300), 6.49
(2H, s, H60, 600), 7.60 (2H, s, H2, 5); dC (125.8 MHz,
CD3OD), 105.0 (C30, 300), 110.2 (quat. C), 117.8 (C60, 600),
123.9 (C3, 4), 137.5 and 138.3 (quat. C), 141.4 (C2), 146.0
(quat. C) (Fig. 10).
Figure 11.
4.2.15. 3-(3,4-Dibenzyloxy-2-aminophenyl)-4-(hydroxy-
methyl)-6,7-dibenzyloxyquinoline (19). A mixture of
3,4-bis(3,4-dibenzyloxy-2-aminophenyl)furan 18 (500 mg,
0.74 mmol), p-toluenesulfonic acid (70 mg, 0.37 mmol) and
˚
powdered molecular sieves (100 mg, 4 A) in benzene
(10 mL) was heated to reflux for 2 days. DCM (20 mL)
was added, the mixture filtered, then washed with sat.
Na2CO3 solution (10 mL). The organic layer was dried over
Na2SO4/MgSO4 and the solvent was removed under
reduced pressure to give a residue that was purified by
column chromatography (PE/EtOAc, 3:7). 19 (125 mg,
25%) was obtained as a white solid, along with recovered
starting material (332 mg, 64%). Rf 0.42 (PE/EtOAc, 3:7);
mp 183.5–184 8C; nmax/cm21 (KBr) 3447 and 3355 (m, NH
str), 3217 (w), 3028 (m), 2926 (w), 2870 (w), 1623 (s), 1546
(m), 1506 (s), 1454 (m), 1411 (m), 1385 (m), 1358 (m),
1260 (s), 1222 (s), 1208 (s), 1170 (m), 1150 (m), 1058 (m),
1014 (m), 866 (m), 735 (s), 696 (s); lmax (EtOH) 242 (1
56,550), 300 (1 8840), 337 (1 10,040); m/z probe APCIþ
(MHþ) 675.7 (40%), HRMS (MHþ) requires m/z 675.2859,
found 675.2861; dH (500 MHz, CDCl3) 3.35 (2H, br, NH2),
4.55 (1H, br, OH), 4.58 (1H, d, J¼12.0 Hz, a-CH2OH), 4.80
(1H, d, J¼12.0 Hz, b-CH2OH), 5.08 (1H, d, J¼12.0 Hz,
a-CH2 of Bn), 5.10 (1H, d, J¼12.0 Hz, b-CH2 of Bn), 5.18
(1H, d, J¼12.0 Hz, a-CH2 of Bn), 5.20 (1H, d, J¼12.0 Hz,
b-CH2 of Bn), 5.28–5.38 (4H, m, CH2 of Bn), 6.50 (1H, s,
H3), 6.72 (1H, s, H6), 7.28–7.56 (21H, m, H8, CH of Bn),
7.61 (1H, s, H5), 8.46 (1H, s, H2); dC (125.8 MHz, CDCl3)
60.0 (CH2OH), 70.3, 70.7, 71.2 and 72.5 (CH2 of Bn), 104.4
(C5), 104.6 (C3), 110.2 (C8), 117.8 (quat. C), 119.3 (C6),
122.6 (quat. C), 127.0, 127.1, 127.3, 127.6, 127.7, 127.8,
127.8, 127.9, 128.3, 128.4, 128.5 and 128.5 (CH of Bn),
136.3, 136.4, 136.9, 137.2, 138.4, 142.1, 142.4 and 145.3
Figure 10.
4.2.14. 3,4-Bis(3,4-dibenzyloxy-2-aminophenyl)furan
(18). Method A. A solution of 3,4-bis(3,4-dibenzyloxy-2-
nitrophenyl)furan 2 (240 mg, 0.33 mmol) and SnCl2
(743 mg, 3.92 mmol) in ethyl acetate (1.0 mL) and MeOH
(0.5 mL) was heated to reflux for 1.5 h. The solution was
poured into a slurry of ice and water (20 mL), and then
DCM was added (20 mL). EDTA.2Na was added until the
aqueous layer was basic. The mixture was shaken
vigorously for a few minutes, and then filtered through
celite, washing with DCM (100 mL). The organic layer was
dried over Na2SO4/MgSO4 and the solvent was removed
under reduced pressure. The residue was purified by column
chromatography (PE/EtOAc, 1:1) to give 18 (148 mg, 67%)
as a pale brown solid.
Method B. Iron powder (0.46 g, 8.24 mmol) was added to a
solution of 3,4-bis(3,4-dibenzyloxy-2-nitrophenyl)furan 2
(0.76 g, 1.03 mmol) and 35% HCl (1.20 mL) in EtOH
(8 mL). The mixture was heated to reflux under argon for
3 h. The cooled solution was diluted with DCM (20 mL) and
water (20 mL), and then filtered through celite, washing
with DCM (100 mL). The organic layer was separated and
again water (20 mL) was added. K2CO3 was added until no
more bubbling occurred and the aqueous layer was basic.
The organic layer was separated, dried over Na2SO4/MgSO4

S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
3709
(quat. C), 149.3 (C2), 149.7, 150.1, 151.6 (quat. C);
dH(400 MHz, DMSO) 4.41 (2H, s, NH2), 4.62 (1H, dd,
J1¼11.5 Hz, J2¼5.0 Hz, a-CH2OH), 4.70 (1H, dd, J1¼
11.5 Hz, J2¼5.0 Hz, b-CH2OH), 4.98 (2H, s, CH2 of Bn),
5.11 (2H, s, CH2 of Bn), 5.28 (1H, t, J¼5.0 Hz, OH), 5.32
(2H, s, CH2 of Bn), 5.37 (2H, s, CH2 of Bn), 6.61 (1H, s,
H3), 6.75 (1H, s, H6), 7.28–7.62 (21H, m, H8, CH of Bn),
7.82 (1H, s, H5), 8.37 (1H, s, H2) (Fig. 12).
7.25–7.48 (22H, m, H2, H5, CH of Bn), 7.52 (2H, s, NH),
7.59 (2H, s, H60, 600); dC (100.6 MHz, CDCl3) 23.94 (CH3),
71.01 and 71.87 (CH2 of Bn), 110.5 (C60), 116.6 (quat. C),
117.2 (C30, 300), 122.5 (quat. C), 127.2, 127.5, 127.8, 127.9,
128.4 and 128.5 (CH of Bn), 129.4 (quat. C), 136.8 and
137.1 (ipso of Bn), 141.1 (C2, 5), 145.9 and 148.8 (quat. C),
168.7 (CvO) (Fig. 14).
Figure 12.
Figure 14.
4.2.16. 20,200,30,300-Tetrakisbenzyloxy-dibenzo[c,h][2,6]-
naphthyridine (20). 3-(3,4-Dibenzyloxy-2-aminophenyl)-
4-(hydroxymethyl)-6,7-dibenzyloxyquinoline 19 (20 mg)
was dissolved in DMSO-d6 (0.8 mL), and left standing
open to the air for 3 months to give pale yellow crystals of
4.2.18. 3,4-Dibenzyloxybenzaldehyde (22). A mixture of
3,4-dihydroxybenzaldehyde (5.00 g, 36.20 mmol), K2CO3
(25.02 g, 181.03 mmol) and benzyl chloride (10.00 mL,
86.89 mmol) in DMF (25 mL), was stirred rapidly at 120 8C
under argon for 15 h. The solution was filtered and the
solvent removed under reduced pressure. The residue was
purified by column chromatography (PE/EtOAc, 7:3) to
give 22 (11.4 g, 99%) as a white solid. Rf 0.30 (PE/EtOAc,
7:3); mp 84–85 8C, lit.36 84–85 8C; dH (400 MHz, CDCl3)
5.22 (2H, s, CH2 of Bn), 5.26 (2H, s, CH2 of Bn), 7.04 (1H,
d, J¼8.0 Hz, H2), 7.31–7.51 (11H, m, H6, CH of Bn), 7.52
(1H, d, J¼2.0 Hz, H5), 9.83 (1H, s, HCvO).
1
20. Due to the crystals being highly insoluble, only a H
NMR spectra was obtainable. Mp 264–265 8C; nmax/cm21
(KBr) 3428 (br,s), 1620 (w), 1516 (m), 1479 (w), 1448 (m),
1380 (w), 1287 (s), 1242 (m), 1184 (w), 1152 (w), 1026 (s),
824 (w), 727 (m), 693 (m); m/z probe ESþ (MHþ) 655.1
(30%), 413.0 (20%), 334.9 (68%), 256.8 (80%), 226.7
(94%), 175.6 (100%), HRMS (MHþ) requires 665.2597, m/z
found 665.2602; dH (500 MHz, DMSO-d6) 5.43 (4H, s, CH2
of Bn), 5.53 (4H, s, CH2 of Bn), 7.31–7.65 (20H, m, CH of
Bn), 7.80 (2H, s, H40, 400), 8.59 (2H, s, H10, 100), 10.22 (2H, s,
H1, H5) (Fig. 13).
4.2.19. (E)-3,4-Dibenzyloxy-b-nitrostyrene (23). A solu-
tion of 3,4-dibenzyloxybenzaldehyde 22 (10.0 g, 31.4 mmol),
MeNO2 (10.2 mL, 188.8 mmol), and NH4OAc (9.7 g,
125.8 mmol) in AcOH (100 mL) was heated to reflux for
40 min. Most of the AcOH was removed under vacuum and
the residue was dissolved in DCM (100 mL). The solution
was washed with sat. K2CO3 solution until the aqueous layer
was basic and then with water (100 mL). The organic layer
was dried over Na2SO4/MgSO4 and the solvent was
removed under reduced pressure to give 23 (11.10 g,
98%) as a bright yellow solid. A small amount was purified
by column chromatography (PE/DCM, 3:7) for spectro-
scopic analysis. Rf 0.38 (PE/DCM, 3:7); mp 115–116.5 8C,
lit.37 117–1188; dH (400 MHz, CDCl3) 5.20, (2H, s, CH2 of
Bn), 5.23 (2H, s, CH2 of Bn), 6.97 (1H, d, J¼8.0 Hz, H5),
7.10 (1H, d, J¼2.0 Hz, H2), 7.12 (1H, dd, J1¼8.0 Hz,
J2¼2.0 Hz, H6), 7.33–7.50 (11H, m, b H, CH of Bn), 7.90
(1H, d, J¼13.5 Hz, a H).
Figure 13.
4.2.17. 3,4-Bis(3,4-dibenzyloxy-2-N-acylaminophenyl)-
furan (21). A mixture of 3,4-bis(3,4-dibenzyloxy-2-amino-
phenyl)furan 18 (266 mg, 0.39 mmol), acetic anhydride
(82 mL, 0.87 mmol), triethylamine (166 mL, 1.19 mmol)
and a catalytic amount of DMAP in THF (6 mL) was stirred
at room temperature for 15 h. The solvent was removed and
the residue was purified by column chromatography (PE/
EtOAc, 1:9) to give 21 (236 mg, 79%) as a pale yellow
solid. Rf 0.33 (PE/EtOAc, 1:9); mp 64–65 8C; nmax/cm21
(KBr) 3271 (m, br), 3031 and 2930 (w), 1658 (s, CvO str),
1511 (s), 1454 (m), 1411 (m), 1369 (m), 1251 (s), 1205 (m),
1165 (m), 1146 (m), 1014 (m), 863 (m), 737 (m), 696 (s);
m/z probe ESþ (MNaþ) 780.0 (38%), (MHþ) 759.0 (70%),
HRMS (MHþ) requires m/z 759.3070, found 759.3078;
microanalysis requires C 75.97, H 5.58, found C 75.70, H
5.54; dH (400 MHz, CDCl3) 1.83 (6H, s, CH3), 4.90 (4H, s,
CH2 of Bn), 5.13 (4H, s, CH2 of Bn), 6.68 (2H, s, H30, 300),
4.2.20. (E)-4,5-Dibenzyloxy-2, b-dinitrostyrene (24). A
mixture of (E)-3,4-dibenzyloxy-b-nitrostyrene 23 (10.9 g,
30.2 mmol), was dissolved in hot glacial acetic (150 mL).
The solution was cooled to 35 8C and 70% HNO3 (8.7 mL,
137.2 mmol) was added over 15 min. A yellow solid
precipitated and the slurry was stirred at room temperature
for 2 h. Water (300 mL) was added, and then the yellow
solid collected was dissolved in DCM (200 mL) then
washed with sat. K2CO3 solution until the aqueous layer
was basic. The organic layer was dried over Na2SO4/MgSO4
and the solvent was removed under reduced pressure. The
residue was recrystallised from DCM/MeOH to give 24

3710
S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
(11.9 g, 97%) as a yellow solid. A small amount was further
purified by column chromatography (PE/DCM, 3:7) for
spectroscopic analysis. Rf 0.32 (PE/DCM, 3:7); mp 161–
162 8C, lit.37 162–163 8C; dH (400 MHz, CDCl3) 5.27 (2H,
s, CH2 of Bn), 5.30 (2H, s, CH2 of Bn), 6.97 (1H, s, H6),
7.25 (1H, d, J¼13.5 Hz, b H), 7.34–7.50 (10H, m, CH of
Bn), 7.82 (1H, s, H3), 8.55 (1H, d, J¼13.5 Hz, a H).
5.81 mmol) and Hg(OAc)2 (2.04 g, 6.40 mmol) in DCM
(100 mL) at 0 8C. The mixture was stirred for an additional
hour at room temperature, then filtered through celite and
washed with sat. Na2S2O5 solution (50 mL). The organic
layer was dried over Na2SO4/MgSO4 and the solvent was
removed under reduced pressure. Purification by column
chromatography (PE/DCM, 6:4) gave 27 (3.55 g, 100%) as
a cream solid.
4.2.21. 5,6-Dibenzyloxyindole (25). A mixture of (E)-4,5-
dibenzyloxy-2, b-dinitrostyrene 24 (5.00 g, 12.30 mmol),
iron powder (13.74 g, 246.02 mmol) and SiO2 (18.50 g) in
AcOH (75 mL), benzene (90 mL) and cyclohexane (30 mL)
was heated to reflux for 30 min under argon. The mixture
was filtered through celite with DCM/EtOAc (1:1) until the
washings contained no product by TLC. The solution was
washed with K2CO3 solution until the aqueous layer was
basic then with water (100 mL). The organic layer was dried
over Na2SO4/MgSO4 and the solvent was removed under
reduced pressure. The residue was purified by column
chromatography (PE/EtOAc, 3:1) to give 25 (2.47 g, 61%)
as a cream solid. Rf 0.30 (PE/EtOAc, 3:1); mp 111–113 8C,
lit.37 113–114 8C; dH (400 MHz, CDCl3) 5.14 (2H, s, CH2
of Bn), 5.20 (2H, s, CH2 of Bn), 6.43 (1H, m, H3), 6.92 (1H,
d, J¼0.5 Hz, H7), 7.02 (1H, dd, J1¼2.5 Hz, J2¼3.0 Hz, H2),
7.24 (1H, s, H4), 7.30–7.54 (10H, m, CH of Bn), 8.03 (1H,
br, NH).
Method B. NIS (0.36 g, 1.60 mmol) was added to a solution
of 5,6-dibenzyloxy-N-triisopropylsilylindole 26 (0.63 g,
1.30 mmol) in THF (10 mL) at 0 8C and stirred for
30 min. The solution was diluted with DCM (50 mL),
washed with sat. Na2S2O5 solution (20 mL), dried over
Na2SO4/MgSO4 and the solvent was removed under
reduced pressure. Purification by column chromatography
(PE/DCM, 6:4) gave 27 (0.66 g, 83%) as a cream solid. Rf
0.33 (PE/DCM, 6:4); mp 105–106 8C; nmax/cm21 (KBr)
2942 (m), 2864 (m), 1496 (m), 1474 (s), 1465 (s), 1454 (s),
1386 (m), 1308 (m), 1200 (s), 1168 (s), 1016 (s), 989 (m),
883 (m), 866 (s), 697 (s); lmax (EtOH) 230 (1 21,970), 297
(1 6130); m/z probe ESþ (MHþ) 610.7 (40%), (MHþ2I)
484.4 (75%), HRMS (MHþ) requires m/z 612.1795, found
612.1791; dH (400 MHz, CDCl3) 1.07 (18H, d, J¼7.5 Hz,
CH3 of TIPS), 1.49 (3H, heptet, J¼7.5 Hz, CH of TIPS),
5.25, (2H, s, CH2 of Bn), 5.30 (2H, s, CH2 of Bn), 6.98 (1H,
s, H7), 7.03 (1H, s, H4), 7.16 (1H, s, H2), 7.29–7.48 (8H, m,
CH of Bn), 7.59 (1H, s, CH of Bn), 7.61 (1H, s, CH of Bn);
dC (100.6 MHz, CDCl3) 12.62 (CH of TIPS), 17.97 (CH3 of
TIPS), 59.46 (C3), 71.49 and 72.44 (CH2 of Bn), 102.4 (C7),
105.4 (C4), 127.0, (CH of Bn), 127.0 (quat. C), 127.6, 127.8,
128.5 and 128.5 (CH of Bn), 134.0, (C2), 134.5 (quat. C),
137.6 and 137.7 (ipso of Bn), 146.0 and 146.4 (quat. C).
4.2.22. 5,6-Dibenzyloxy-N-triisopropylsilylindole (26).
n-BuLi in hexanes (3.90 mL of a 1.62 M solution,
6.32 mmol) was added slowly to a solution of 5,6-di-
benzyloxyindole 25 (1.74 g, 5.28 mmol) in THF (20 mL) at
278 8C under argon. After 15 min, TIPSCl (1.58 mL,
7.4 mmol) was added and the solution was stirred for 2 h
at 278 8C then 30 min at room temperature. Most of the
THF was removed under reduced pressure and replaced
with DCM (50 mL). The solution was washed with water
(50 mL) dried over Na2SO4/MgSO4 and the solvent
removed under reduced pressure to give a residue that was
purified by column chromatography (PE/DCM, 6:4). 26
(2.43 g, 95%) was obtained as a white solid. Rf 0.18 (PE/
DCM, 6:4); mp 84–85 8C; nmax/cm21 (KBr) 3090 (w), 3062
(w), 3031 (w), 2944 (s), 2865 (s), 1514 (m), 1497 (m), 1482
(s), 1452 (s), 1309 (m), 1263 (m), 1220 (s), 1188 (s), 1161
(s), 1016 (s), 884 (s), 868 (s), 736 (s), 722 (s), 690 (s), 652
(m); lmax (EtOH) 206 (1 54,080), 228 (1 40,910), 268 (1
6960), 300 (1 7420); m/z probe ESþ (MHþ) 485.0 (57%),
HRMS (MHþ) requires m/z 486.2828, found 486.2826;
microanalysis requires C 76.65, H 8.09, found C 76.61, H
8.22; dH (400 MHz, CDCl3) 1.08 (18H, d, J¼7.5 Hz, CH3 of
TIPS), 1.52 (3H, heptet, J¼7.5 Hz, CH of TIPS), 5.24 (4H,
s, CH2 of Bn), 6.50 (1H, dd, J1¼3.0 Hz, J2¼1.0 Hz, H3),
7.01 (1H, d, J¼1.0 Hz, H7), 7.12 (1H, d, J¼3.0 Hz, H2),
7.19 (1H, s, H4), 7.28–7.57 (10H, m, CH of Bn); dC
(100.6 MHz, CDCl3) 12.66 (CH of TIPS), 18.05 (CH3 of
TIPS), 71.96 and 72.50 (CH2 of Bn), 102.5 (C7), 104.4 (C3),
105.6 (C4), 125.3 (quat. C), 127.1, 127.4, 127.5, 127.6 and
128.4 (CH of Bn), 130.3 (C2), 135.3, 138.0, 145.0 and 145.5
(quat. C).
4.2.24. 5,50,6,60-Tetrabenzyloxy-N,N0-triisopropylsilyl-
3,30-biindolyl (28). A mixture of 5,6-dibenzyloxy-3-iodo-
N-triisopropylsilylindole 27 (0.50 g, 0.82 mmol), tetrakis-
(dimethylamino)ethylene (0.38 mL, 1.63 mmol) and
Pd(PhCN)2Cl2 (30 mg, 0.08 mmol) was stirred in DMF
(4 mL) at 50 8C for 1.5 h under argon. The solution was
diluted with DCM (20 mL), washed with water (20 mL),
dried over Na2SO4/MgSO4, and the was solvent removed
under reduced pressure. Purification by column chromatog-
raphy (PE/DCM, 3:7) gave 28 (269 mg, 68%) as a white
gum. Rf 0.44 (PE/DCM, 3:7); nmax/cm21 (thin film) 2946
(s), 2866 (s), 1506 (m), 1470 (s), 1314 (m), 1192 (s), 1151
(s), 1016 (s), 883 (m), 733 (m), 695 (s), 652 (m); lmax
(EtOH) 304 (1 14,000); m/z probe (MHþ) 968.5 (100%),
506.7 (13%), 485.3 (25%), HRMS (MHþ) requires m/z
969.5422, found 969.5508; microanalysis requires C 76.81,
H 7.90, found C 76.76, H 8.03; dH (400 MHz, CDCl3) 1.16
(36H, d, J¼7.5 Hz, CH3 of TIPS), 1.58 (6H, heptet, J CH of
TIPS), 5.23, (2H, s, CH2 of Bn), 5.30 (2H, s, CH2 of Bn),
7.09 (2H, s, CH of Ar), 7.31–7.56 (24, m, Ar CH, Ar CH
and CH of Bn); dC (100.6 MHz, CDCl3) 12.77 (CH of
TIPS), 18.17 (CH3 of TIPS), 71.78 and 72.55 (CH2 of Bn),
102.8 and 104.8 (Ar CH), 112.5 and 124.3 (quat. C), 127.1,
127.3, 127.5, 127.6, 127.6, 128.4 and 128.5 (Ar CH, CH of
Bn), 135.8 (quat. C), 137.9 and 138.0 (ipso of Bn), 145.2
and 145.7 (quat. C).
4.2.23. 5,6-Dibenzyloxy-3-iodo-N-triisopropylsilylindole
(27). Method A. A solution of I2 (1.62 g, 6.38 mmol) in
DCM (200 mL) was added slowly over 1 h to a mixture
of 5,6-dibenzyloxy-N-triisopropylsilylindole 26 (2.82 g,
4.2.25. 5,50,6,60-Tetrabenzyloxy-3,30-biindolyl (29). TBAF
in THF (0.24 mL of a 1.0 M solution, 0.24 mmol) was

S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
3711
added dropwise to a solution of 5,50,6,60-tetrabenzyloxy-
References and notes
N,N0-triisopropyl-3,30-biindolyl 28 (105 mg, 0.11 mmol) in
THF (1 mL) and stirred for 10 min. Cold MeOH (5 mL) was
added and the white precipitate was collected, then washed
with cold MeOH (10 mL) to give 29 (58 mg, 82%) as a
white powder. Mp 225–225.5 8C; nmax/cm21 (KBr) 3399
(m, br), 3281 (s, NH str), 3063 (w), 3031 (w), 2935 (w),
2877 (w), 1629 (m), 1508 (m), 1475 (s), 1466 (s), 1454 (m),
1312 (s), 1244 (s), 1189 (s), 1137 (s), 1010 (m), 990 (m),
975 (m), 917 (m), 879 (m), 741 (s), 698 (s); lmax (EtOH)
300 (1þ6480); m/z probe ESþ (MNaþ) 679.3 (57%),
(MNH4 ) 674.3 (56%), (MHþ) 657.3 (100%), HRMS
(MHþ) requires m/z 657.2753, found 657.2765; dH
(400 MHz, DMSO) 5.14, (4H, s, CH2 of Bn), 5.19 (4H, s,
CH2 of Bn), 7.07 (2H, s, H7), 7.28–7.53 (24H, m, H2, H4,
CH of Bn), 10.84 (2H, d, J¼2.0 Hz, NH); dC (100.6 MHz,
DMSO) 71.40 and 72.20 (CH2 of Bn), 98.71 (C7), 106.4
(C4), 110.7 and 120.2 (quat. C), 121.2 (C2), 128.3, 128.4,
128.5, 128.5, 129.1 and 129.2 (CH of Bn), 132.0 (quat. C),
138.6 and 138.9 (ipso of Bn), 144.6 and 146.7 (quat. C).
¨
¨
1. (a) Kahkonen, M. P.; Hopia, A. I.; Vuorela, H. J.; Rauha, J.-P.;
Pihlaja, K.; Kujala, T. S.; Heinonen, M. J. Agric. Food Chem.
1999, 47, 3954–3962. (b) Kujala, T. S.; Loponen, J. M.; Klika,
K. D.; Pihlaja, K. J. Agric. Food Chem. 2000, 48, 5338–5342.
2. (a) Temple, N. J. Nutr. Res. 2000, 20, 449–459. (b) Block, G.
Nutr. Rev. 1992, 50, 207–213.
3. Ames, B. N.; Shigenaga, M. K.; Hagen, T. M. Proc. Natl.
Acad. Sci. U.S.A. 1993, 90, 7915–7922.
4. Vinson, J. A.; Dabbagh, Y. A.; Serry, M. M.; Jang, J. J. Agric.
Food Chem. 1995, 43, 2800–2802.
5. Rice-Evans, C.-A.; Miller, N. J.; Bowell, P. G.; Bramley, P.;
Pridham, J. B. Free Radical Res. 1995, 22, 375–383.
6. Vinson, J. A.; Yong, H.; Su, X.; Ligia, Z. J. Agric. Food Chem.
1998, 46, 3630–3634.
7. Kujala, T.; Klika, K.; Ovcharenko, V.; Loponen, J.; Vienola,
M.; Pihlaja, Z. Naturforsch. 2001, 56c, 714–718.
8. Prota, G. Melanins and melanogenesis; Academic: San Diego,
1992.
9. Mason, H. S. In Pigment cell growth; Gordan, M., Ed.;
Academic: New York, 1953.
4.2.26. 5,50,6,60-Tetrahydroxy-3,30-biindolyl (1). 5,50,6,60-
Tetrabenzyloxy-3,30-biindolyl 29 (30 mg, 0.04 mmol) was
stirred in THF (1 mL) with Palladium Black (3 mg) under
an atmosphere of H2 for 18 h. The solution was quickly
filtered and the solvent removed at 30 8C under reduced
pressure to give the target compound 1 (13 mg, 94%) as a
greyish orange solid. The solid product did not dissolve in
10. Memoli, S.; Napolitano, A.; d’Ischia, M.; Misuraca, G.;
Palumbo, A.; Prota, G. Biochim. Biophys. Acta 1996, 1346, 61.
11. Prota, G.; Misuraca, G. In Melanogenesis and malignant
melanoma: biochemistry, cell biology, molecular biology,
pathophysiology, diagnosis and treatment; Hori, Y., Hearing,
V. J., Nakayama, J., Eds.; Elsevier: Amsterdam, 1996; p 49.
12. (a) Napolitano, A.; Pezzella, A.; Rosaria, M.; Prota, V.; Prota,
G. Tetrahedron 1995, 51, 5913–5920. (b) Bergman, J.; Koch,
E.; Pelcman, B. Tetrahedron 1995, 51, 5631–5642.
13. Napolitano, A.; Corradini, M.; Prota, G. Tetrahedron Lett.
1985, 26, 2805–2808.
D2O sufficiently to allow the preparation of a sample for 13
C
NMR analysis (14,000 scans at 500 MHz gave no signals),
although it was readily soluble in DMSO-d6. To prepare a
sample in D2O, the reaction solution was quickly filtered
and most of the THF was removed at 30 8C under reduced
pressure. D2O (0.8 mL) was added giving a pale orange
solution. The remaining THF was removed under reduced
pressure leaving an aqueous solution with a milky
precipitate (the milky precipitate was soluble in DMSO-d6
and gave clean spectra of the compound showing that the
heteroatom protons had been deuterated). This was quickly
filtered to give a pale orange aqueous solution. lmax (H2O)
302; m/z probe ES2 (M2Hþ) 295.07 (100%), HRMS
(M2Hþ) requires m/z 295.0719, found 295.0717; dH
(500 MHz, D2O) 6.94 (2H, s, H7, 70), 7.13 (2H, s, H4, 40),
7.32 (2H, s, H2, 20); dC (125.8 MHz, D2O) 98.47 (C7), 105.2
(C4), 109.1 (C3), 119.5 (C3a), 121.2 (C2), 131.6 (C7a),
139.8 (C5), 142.2 (C6); dH (500 MHz, DMSO-d6) 6.78 (2H,
s, H7, 70), 7.03 (2H, s, H4, 40), 7.16 (2H, d, J¼2.0 Hz, H2,
20), 8.23 (2H, s, OH-5), 8.50 (2H, s, OH-6), 10.40 (2H, d,
J¼2.0 Hz, NH); dC (125.8 MHz, DMSO-d6) 98.10 (C7),
105.2 (C4), 110.6 (C3), 119.5 (C2), 119.8 (C3a), 131.5
(C7a), 141.2 (C5), 143.5 (C6).
14. (a) Desarbre, E.; Bergman, J. J. Chem. Soc., Perkin Trans. 1
1998, 2009–2016. (b) Berens, U.; Brown, J.; Long, J.; Selke,
R. Tetrahedron: Asymmetry 1996, 7, 285–292.
15. Pines, S. H.; Karady, S.; Sletzinger, M. J. Org. Chem. 1968,
33, 1758–1761.
16. Lee, F. G. H.; Suzuki, J.; Dickson, D. E.; Manian, A. A.
J. Heterocycl. Chem. 1972, 387–392.
17. (a) Marshall, J. A. Chem. Rev. 2000, 100, 3163–3185. (b) Li,
G.; Bittman, R. Tetrahedron Lett. 2000, 41, 6737–6741.
(c) Kosngi, M.; Shimuzu, K.; Ohtami, A.; Migita, T. Chem.
Lett. 1981, 829. (d) Kosugi, M.; Ohya, T.; Migita, T. Bull.
Chem. Soc. Jpn 1983, 56, 3855–3856.
18. Gorzynski, M.; Rewicki, D. Liebigs Ann. Chem. 1986,
625–637.
¨
19. (a) Scott, T. C.; Soderberg, B. C. G. Tetrahedron Lett. 2002,
´
´
43, 1621–1624. (b) Albeniz, A. C.; Espinet, P.; Martın-Ruiz,
B. Chem. Eur. J. 2001, 7, 2481–2489. (c) Furness, M. S.;
Robinson, T. P.; Goldsmith, D. J.; Bowen, J. P. Tetrahedron
Lett. 1999, 40, 459–462. (d) Cuevas, J.-C.; Pat, I. P.; Snieckus,
V. Tetrahedron Lett. 1989, 30, 5841–5844.
20. (a) Farina, V.; Kapadia, S.; Krishnan, B.; Wang, C.;
Liebeskind, L. S. J. Org. Chem. 1994, 59, 5905–5911.
(b) Liebeskind, L. S.; Fengi, R. W. J. Org. Chem. 1990, 55,
5359–5364.
Acknowledgements
This research was supported by the Overseas Research
Student awards scheme and Professor Sir Jack Baldwin. We
would like to thank Dr. Barbara Odell and Dr. Tim Claridge
for their help with NMR processing, Dr. Peter Rutledge for
his advice with the manuscript preparation and Professor
Tytti Kujala for allowing us to reproduce his spectroscopic
data.
21. (a) Rossi, K.; Bellina, F.; Raugei, E. Synlett 2000, 1749–1752.
(b) Farina, V.; Krishnan, B.; Marshall, D. R.; Roth, G. P.
J. Org. Chem. 1993, 58, 5434–5444. (c) Friesen, R. W.;
Sturino, C. F. J. Org. Chem. 1990, 55, 2572–2574. (d) Dubois,
E.; Beau, J.-M. Tetrahedron Lett. 1990, 31, 5165–5168.
(e) Farina, V.; Roth, G. P. Tetrahedron Lett. 1991, 32,

3712
S. P. H. Mee et al. / Tetrahedron 60 (2004) 3695–3712
4243–4246. (f) Tolstikov, G. A.; Miftakhov, M. S.; Danilova,
N. A.; Vel’der, Y. L.; Spririkhin, L. V. Synthesis 1989, 625.
31. Yang, Y.; Wong, H. N. C. Tetrahedron 1994, 50, 9583–9608.
32. Mee, S. P. H.; Lee, V.; Baldwin, J. E. Synth. Commun. 2003,
33, 3205–3209.
22. (a) Shirakawa, E.; Murota, Y.; Nakao, Y.; Hiyama, T. Synlett
1997, 1143–1144. (b) Alcaraz, L.; Taylor, R. J. K. Synlett
1997, 791–792.
33. Littke, A. F.; Schwarz, L.; Fu, G. C. J. Am. Chem. Soc. 2002,
124, 6343–6348.
23. Piers, E.; Gladstone, P. L.; Yee, J. G. K.; McEachern, E. J.
Tetrahedron 1998, 54, 10609–10626.
34. Mee, S. P. H.; Lee, V.; Baldwin, J. E. Angew. Chem., Int. Ed.
2004, 43, 1132–1136.
24. (a) Farina, V.; Krishnamarthy, V.; Scott, W. J. Organic
reactions; Paquette, L. A., Ed.; Wiley: New York, 1997; Vol.
50. (b) Farina, V.; Krishnan, B. J. Am. Chem. Soc. 1991, 113,
9585–9595. (c) Farina, V.; Buker, S. R.; Benigni, D. A.;
Hausk, S. I.; Sapino, C. J. Org. Chem. 1990, 55, 5833–5847.
(d) Beletskaya, I. P. J. Organomet. Chem. 1983, 250,
551–564.
35. Bellamy, F. D.; Ou, K. Tetrahedron Lett. 1984, 25, 839–842.
36. McElhanon, J. R.; Wu, M.-J.; Escobar, M.; Chaudhry, U.; Hu,
C.-L.; McGrath, D. V. J. Org. Chem. 1997, 62, 908–915.
37. Benigni, J. D.; Minnis, R. L. J. Heterocycl. Chem. 1965,
387–392.
38. Sinhahabu, A. K.; Borchardt, R. T. J. Org. Chem. 1983, 48,
3347–3349.
25. (a) Kraus, G. A.; Wang, X. Synth. Commun. 1998, 28,
1093–1096. (b) Yang, Y.; Wong, H. N. C. Tetrahedron 1994,
50, 9583–9608. (c) Zaluski, M.-C.; Robba, M.; Bonhomme,
M. Bull. Soc. Chim. Fr. 1970, 1838–1846.
39. Bray, B. L.; Mathies, P. H.; Solas, D. R.; Tidwell, T. T.; Artis,
D. R.; Muchowski, J. M. J. Org. Chem. 1990, 55, 6317–6328.
40. (a) Kabir, H.; Miura, M.; Sasaki, S.; Harada, G.; Kuwatani, Y.;
Yoshida, M.; Iyoda, M. Heterocycles 2000, 52, 761–774.
(b) Rajca, A.; Wang, H.; Bolshov, P.; Rajca, S. Tetrahedron
2001, 57, 3725–3735.
26. Preparation of 3,4-dibromofuran using this method resulted in
less than 5% yield in all variations attempted.
27. (a) Farina, V.; Krishnan, B. J. Am. Chem. Soc. 1991, 113,
9585–9595. (b) Negishi, E.; Takahashi, T.; Baba, S.; Van
Horn, D. E.; Okukado, N. J. Am. Chem. Soc. 1987, 109,
2393–2401.
41. (a) Bumagin, N. A.; Nikitina, A. F.; Beletskaya, I. P. Russ.
J. Org. Chem. 1994, 30, 1619–1629. (b) Haung, J.; Nolan, S. P.
J. Am. Chem. Soc. 1999, 121, 9889–9890.
42. Kuroboshi, M.; Waki, Y.; Tanaka, H. Synlett 2002, 637–639.
43. (a) Capon, B.; Kwok, F.-C. J. Am. Chem. Soc. 1989, 111,
5346–5356. (b) Capon, B.; Kwok, F.-C. Tetrahedron Lett.
1986, 27, 3275–3278.
28. Tidwell, J. H.; Buchwald, S. L. J. Am. Chem. Soc. 1994, 116,
11797–11810.
29. Orito, K.; Hatakeyama, T.; Takeo, M.; Suginome, H. Synthesis
1995, 1273–1277.
44. Harris, R. K.; Becker, E. D.; De Menezes, S. M. C.;
Goodfellow, R.; Granger, P. Pure Appl. Chem. 2001, 73,
1795–1818.
30. Hine, J.; Halm, S.; Miles, D. E.; Ahn, K. J. Org. Chem. 1985,
50, 5092–5096.