A convenient synthesis of unsymmetrically substituted terphenyls of biologically active stilbenes via a double Suzuki cross-coupling protocol

Article abstract of DOI:10.1016/S0040-4039(03)00447-7

A double Suzuki cross-coupling protocol has been devised as a practical route to a variety of terphenyls. Good chemoselectivity was observed. Unsymmetrically substituted triphenylenes were also easily prepared.

Full text of DOI:10.1016/S0040-4039(03)00447-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 3005–3008
A convenient synthesis of unsymmetrically substituted terphenyls
of biologically active stilbenes via a double Suzuki
cross-coupling protocol
Daniele Simoni,a,* Giuseppe Giannini,b Pier Giovanni Baraldi,a Romeo Romagnoli,a
Marinella Roberti,c Riccardo Rondanin,a Riccardo Baruchello,a Giuseppina Grisolia,a
Marcello Rossi,a Danilo Mirizzi,a Francesco Paolo Invidiata,d Stefania Grimaudoe and
Manlio Tolomeoe
aDipartimento di Scienze Farmaceutiche, Via Fossato di Mortara 17-19, Universita` di Ferrara, 44100 Ferrara, Italy
bSigma Tau, Via Pontina, Pomezia, Roma, Italy
cDipartimento di Scienze Farmaceutiche, Universita` di Bologna, Italy
dDipartimento Farmacochimico Tossicologico e Biologico, Universita` di Palermo, Italy
eDivisione di Ematologia e Servizio AIDS, Policlinico di Palermo, Italy
Received 18 November 2002; accepted 6 February 2003
Abstract—A double Suzuki cross-coupling protocol has been devised as a practical route to a variety of terphenyls. Good
chemoselectivity was observed. Unsymmetrically substituted triphenylenes were also easily prepared. © 2003 Published by Elsevier
Science Ltd.
Stilbene-based compounds are widely represented
in nature and have become of particular interest
to chemists and biologists because of their wide
range of biological activities.1 Stilbene itself does
not occur in nature, but hydroxylated stilbenes
are abundantly represented; some of these, such as
trans-resveratrol (1),2 the cis-stilbene combretastatin
A-4 (CA-4, 2),3 and the stilbene-based vitamin A ana-
logues 34 have shown unique potentialities for treat-
ment of cancer and for cancer chemoprevention (Fig.
1).
Not unexpectedly, the ability of natural and synthetic
stilbenes to elicit a wide range of physiological effects
often stems from changes in olefin configuration. It is
also well known that the natural products themselves,
at least in many cases, can not be used directly, because
of insufficient levels of activity and bad distribution
properties, which are sometimes associated with a cer-
tain degree of instability. As a result, the search on the
synthetic analogs has become of major importance and
the synthesis of double bond-locked compounds
appears particularly promising.5
Figure 1. Natural and synthetic stilbenoids.
* Corresponding author. Tel.: +39-0532-291291; fax: +39-0532-291296; e-mail: smd@dns.unife.it
0040-4039/03/$ - see front matter © 2003 Published by Elsevier Science Ltd.
doi:10.1016/S0040-4039(03)00447-7

3006
D. Simoni et al. / Tetrahedron Letters 44 (2003) 3005–3008
In this context, our attention was recently drawn to the
idea that terphenyl-based architectures, which incorpo-
rate a phenyl ring as bioisosteric substitution of the
alkenyl bridge, could constitute interesting carbon
frameworks in obtaining conformationally restricted
stilbene mimetics. With these criteria in mind we set out
to search for a general synthetic approach to terphenyl
analogues of CA-4 (2) (compounds 4a and 13a,b) as
well as of the stilbene-based arotinoids 3 (compounds
4b) (Fig. 2).
73% yields (ortho-, meta- and para-isomers, respec-
tively) and with chemoselectivity the bromobiphenyl
derivatives 8, in toluene/ethanol as solvent and tetra-
kis(triphenylphosphine)-palladium(0) as catalyst for the
reaction, whereas aqueous Na2CO3 provided the basic
environment required. The reaction was completed in 2
h at reflux. Flash column chromatography, followed by
recrystallization afforded the desired product. The sub-
sequent Suzuki cross-coupling between 8 and the
boronic acid 9 gave, after removal of the benzyl pro-
tecting group, the desired 4a in 52/85/64% yields (ortho-,
meta- and para-isomers, respectively). As expected,
treatment of the benzylated ortho-terphenyl derivative
4a with iron(III) chloride in dichloromethane in the
presence of traces of sulphuric acid easily gave the
triphenylene 5 (70% yield). O-Debenzylation occurred
just during the cyclization, thus avoiding a further
hydrogenation reaction.
As palladium-catalyzed cross-coupling reactions
between arylboronic acids and aryl halides or triflates
(Suzuki coupling reaction) have become a common
method for the synthesis of biaryls,6,7 we envisaged the
possibility that a highly convergent and flexible double
Suzuki coupling approach could be used to prepare a
variety of unsymmetrically substituted terphenyls. It is
worth noting that such a procedure has been used only
sporadically and has not yet received full recognition as
a general route to terphenyls.8
In the next step, commercially available 6 and 10a were
envisaged to be easily incorporated as the B- and
C-rings of the terphenyl architectures 4b, whereas 10b
the residual A-ring (Fig. 3). Somewhat surprisingly,
Suzuki coupling between 6 and 10b allowed only in
very low yields the desired 11a, because of the concomi-
tant dimerization of 6. Assuming that the relatively
high steric hindrance given by the adamantane group
could be responsible for the low yields of the coupling
procedure, allowing a preferential self-condensation of
the bromophenyl boronic acids, we opted to examine a
different approach, where the iodo-bromo intermedi-
ates 12a could be incorporated as the B-ring of the
terphenyl system, while intermediates 10c and 10a as
the A- and C-rings, respectively.
The value of our approach is exemplified in a four-step
synthetic route to terphenyls 4a of CA-4. Additionally,
the easy access to the combretastatin’s ortho-terphenyl
derivative 4a, also allowed an easy entry to the more
restricted triphenylene 5. We speculated that a double
Suzuki cross-coupling between 6 and intermediates 7
and 9 could accomplish the desired terphenyls 4a
(Scheme 1). A subsequent rational route to the
triphenylene 5 could involve iron(III) chloride oxidative
cyclotrimerisation of the newly synthesized ortho-ter-
phenyl derivative 4a.9,10 Practically, the Suzuki coupling
between 6 and the iodo-derivative 7 afforded in 65/81/
Figure 2. Terphenyl and triphenylene analogues of natural and synthetic stilbenoids.
Scheme 1.

D. Simoni et al. / Tetrahedron Letters 44 (2003) 3005–3008
3007
Figure 3. Synthetic intermediates.
Figure 4. Terphenyl and triphenylene analogues of natural combretastatin A-4.
a superior manner thus providing an easy preparation
of a variety of terphenyls. The oxidative cyclization of
the benzylated ortho-terphenyl derivative of combretas-
tatin A-4, using ferric chloride, allowed the preparation
of its triphenylene derivative. Finally, the simplicity of
the reaction could suggest a possible application of this
methodology for generation of combinatorial terphenyl
or triphenylene libraries. The antiproliferative and
apoptosis-inducing effects on HL60 cell line demon-
strated an interesting activity for the terphenyl deriva-
tive 13a.11
As a matter of fact, couplings between 10a and 12a
were simply accomplished by the standard Suzuki pro-
cedure used above and completed in appreciable yields
(52/85/64% for the ortho-, meta- and para-isomer,
respectively) allowing 11b within a few hours with
chemoselectivity for the aryl iodide. The aldehydes 11b
were in turn oxidized by means of MnO2 to the corre-
sponding carboxylic acids and the subsequent Suzuki
cross-coupling with the boronic acid 10c were per-
formed under the conditions used above for 11b, allow-
ing the benzylated 4b (33/45/67% yields, respectively,
for the ortho-, meta- and para-isomer) in turn subjected
to standard O-debenzylation procedure (Pd/C catalyzed
hydrogenation, 66% yield).
Acknowledgements
Finally, it is worth noting that the pyridine derivatives
13a and 13b were also prepared by means of the
synthetic scheme described for 4b employing the 2-
bromo-3-iodo pyridine (12b) as the B-ring and interme-
diates 9 and 10d for the other rings. Pd/C-catalyzed
hydrogenation of the benzylated 13a,b furnished the
desired terphenyls in 86% yields (Fig. 4). The cyclocou-
pling procedure performed on the benzylated 13a,b
with iron(III) chloride gave principally the O-debenzyl-
ation. The triphenylene derivative 14 was obtained in
only 10% yield.
This work was supported by the Ministero dell’Univer-
sita` e della Ricerca Scientifica e Tecnologica (Rome)
and by Sigma Tau (Rome), Italy.
References
1. Becher, K. B. Synthesis 1983, 341–368.
2. Dorrie, J.; Gerauer, H.; Wachter, Y.; Zunino, S. J.
Cancer Res. 2001, 61, 4731–4739.
3. Lin, C. M.; Ho, H. H.; Pettit, G. R.; Hamel, E. Biochem-
istry 1989, 28, 6984–6995.
4. Simoni, D.; Roberti, M.; Invidiata, F. P.; Rondanin, R.;
Baruchello, R.; Malagutti, C.; Mazzali, A.; Rossi, M.;
Grimaudo, S.; Dusonchet, L.; Meli, M.; Raimondi, M.
V.; D’Alessandro, N.; Tolomeo, M. J. Med. Chem. 2001,
44, 2308–2318.
5. Ohsumi, K.; Hatanaka, T.; Fujita, K.; Nakagawa, R.;
Fukuda, Y.; Nihei, Y.; Suga, Y.; Morinaga, Y.;
Akiyama, Y.; Tsuji, T. Bioorg. Med. Chem. Lett. 1998, 8,
3153–3158.
In summary, the purpose of this study was to investi-
gate methods for the preparation of terphenyl architec-
tures as bioisosters of natural and synthetic stilbenes. In
this context we have demonstrated the utility of a
double Suzuki cross-coupling protocol to obtain a facile
entry to terphenyls and triphenylenes. We have found
that halogenated boronic acids or, alternatively,
bromo-iodo aryls, could be utilized as the B ring of the
terphenyl system, being the double cross-coupling pro-
cedure endowed with good chemoselectivity. The
devised double Suzuki protocol works in many cases in
6. Miyaura, N.; Suzuki, N. Chem. Rev. 1995, 95, 2457–2483.
7. Stanforth, S. P. Tetrahedron 1998, 54, 263–303.

3008
D. Simoni et al. / Tetrahedron Letters 44 (2003) 3005–3008
antiproliferative activity IC50=0.3 0.07 mM; apoptosis
8. Sutton, A. E.; Clardy, J. Tetrahedron Lett. 2001, 42,
inducing activity AC50=0.7 0.03 mM) similar to other
chemotherapeutic drugs usually used in the clinical prac-
tice (cis-platinum: IC50=0.6 0.08 mM, AC50=1.2 0.1
mM; Etoposide: IC50=0.12 0.07 mM, AC50=0.45 0.08
mM; Citarabine: IC50=0.08 0.01 mM, AC50=0.15 0.02
mM; 5-Fluorouracil IC50=1.6 0.3 mM, AC50=6.5 0.6
mM).
547–551 and references cited therein.
9. Borner, R. C.; Jackson, F. W. J. Chem. Soc., Chem.
Commun. 1994, 845–846.
10. Boden, N.; Bushby, R. J.; Cammidge, A. N. J. Chem.
Soc., Chem. Commun. 1994, 465–466.
11. The most effective compound was the terphenyl deriva-
tive 13a that showed an activity (in HL60 cell line,