A colored dendrimer as a new soluble support in organic synthesis. Part 1: Suzuki reaction

Article abstract of DOI:10.1016/S0040-4039(01)01367-3

A new strategy using a colored dendrimer as visible soluble support for organic synthesis has been developed. The efficiency of this new system has been demonstrated by the use of DR_HMPA⁹-CH₂OH as the support in a Suzuki coupling reaction. Due to the visibility of the support, following of the reaction has been rendered easier and the purification time of the crude product has been considerably shortened.

Full text of DOI:10.1016/S0040-4039(01)01367-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 6683–6686
A colored dendrimer as a new soluble support in organic
synthesis. Part 1: Suzuki reaction
Jidong Zhang,* Jozsef Aszodi, Ce´line Chartier, Nathalie L’hermite and John Weston
Medicinal Chemistry, Aventis Pharma S. A, Aventis Research Center Paris, 102 route de Noisy,
93235 Romainville Cedex, France
Received 12 July 2001; revised 23 July 2001; accepted 25 July 2001
Abstract—A new strategy using a colored dendrimer as visible soluble support for organic synthesis has been developed. The
efficiency of this new system has been demonstrated by the use of DR9HMPA–CH2OH as the support in a Suzuki coupling reaction.
Due to the visibility of the support, following of the reaction has been rendered easier and the purification time of the crude
product has been considerably shortened. © 2001 Elsevier Science Ltd. All rights reserved.
Combinatorial chemistry has become an important
component of the drug discovery process. Solid-phase
syntheses based on modified polystyrene microbeads,
originally introduced by Merrifield, have revolutionized
organic synthesis in the construction of small molecule
combinatorial libraries. The most important feature of
solid-phase synthesis is the ease with which reagents
and solvents are removed by a simple washing, which
allows an easy automation of the synthetic procedure
and therefore is transforming, together with other new
synthesis technologies like parallel synthesis, traditional
medicinal chemistry into high throughput medicinal
chemistry. Despite its big success, solid support synthe-
sis exhibits a certain number of problems due to the
heterogeneous nature of the reactions and the low
loading capacity of functional groups. Although more
and more organic reactions can be carried out with
solid supports, the task of modifying solution phase
chemistry to the solid phase remains arduous and time
consuming. The quantity of final product is often small,
which often does not satisfy the requirement of in vitro
and in vivo biological assays. The process is further
handicapped by the difficulty in routinely monitoring
solid phase reactions even though analytical technolo-
gies are improving. To address these issues, some alter-
natives using soluble polymeric supports have been
developed by different groups, which could take advan-
tage of both solution and solid-phase chemistry. (i)
Soluble support chemistry, which did not require adap-
tation of the chemistry to solid phase. (ii) Reactions
that could be followed by conventional TLC. (iii) Inter-
mediates which could be routinely characterized by
classical analytical methods, including 1H and 13C
NMR, IR and mass spectrometry. (iv) Large reagent
excess that could be used to drive reactions to comple-
tion. (v) Purifications that could be realized according
to differences of physicochemical properties of small
organic reagents and soluble polymeric supports, e.g.
by precipitation/crystallization, ultra-filtration and
especially size exclusion chromatography (SEC), which
could be automated.1
The first class of soluble supports is the linear
monomethylated
polyethylene
glycol
(typically
MPEG5000), soluble in many organic solvents and
easily precipitated in non-polar solvents (e.g. Et2O).2
However, due to their linear topology, they provide
only poor loading capacity. The second class of soluble
supports is that of the low generation dendrimers and
hyper-branched macromolecules. Dendrimers are
highly branched, globular, monodispersed macro-
molecules. They comprise a core, branching units and a
large number of terminal groups. They provide a very
high loading capacity since they can bring multiple
copies of molecules per dendrimer. They can be purified
either by precipitation or by SEC. The commercially
available polyamidoamine (PAMAM) has been success-
fully used by Kim and co-workers as a highly soluble
support in the synthesis of indoles.3 The use of other
dendrimers or hyper-branched macromolecules has also
been reported by other research groups.4 Recently, we
published an efficient divergent synthesis of generation-
1 (G1) dendrimers 1 (Scheme 1), which include dye
Red-1 in the core and nine tBoc-protected primary
* Corresponding author. Tel.: +33 1 49 91 47 31; fax: +33 1 49 91 50
87; e-mail: jidong.zhang@aventis.com
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01367-3

6684
J. Zhang et al. / Tetrahedron Letters 42 (2001) 6683–6686
Scheme 1. Reagents and conditions: (a) HCl (g), AcOEt, 100%; (b) (i) Ac2O, Py, (ii) C6F5OH, DCC, DMF, 50%; (c) TEA, DMF,
91%; (d) K2CO3, MeOH–CH2Cl2, 72%.
amines.5 The fact that the dye Red-1 was incorporated
into the dendrimer confers to them an advantage over
most dendrimers, which is their color. This new addi-
tional property together with the chemically functional-
izable terminal tBoc-protected primary amines make
them interesting targets in our research to find new
carriers, for example, for active entities and/or antibod-
ies. Herein we would like to report our studies on the
use of this visible soluble support in organic synthesis,
which simplifies the follow-up of the reaction and
shortens purification times.
The attachment of 2-iodobenzoyl and 4-iodobenzoyl to
DR9HMPA–CH2OH 6 was realized in pyridine with,
respectively, 2-iodobenzoyl chloride 7a and 4-iodoben-
zoyl chloride 7b to afford after purification on a Sep-
hadex LH-20 SEC cartridge dendrimers 8a,b (Scheme
2). The Suzuki reaction with six commercial available
boronic acids was performed in DMF in the presence
of Pd(PPh3)4 and Na2CO3 (2 M in H2O). The reaction
mixture was extracted with AcOEt and washed with
water to eliminate inorganic salts. The organic phase
was concentrated under vacuum, and purified on a
Sephadex LH-20 SEC cartridge to give the correspond-
ing dendrimers 9a–f and 10a–f (Table 1). It was worth
noticing that purification aided by the color indication
took less than 30 min and all the products were charac-
terized with conventional analytical methods (e.g.
TLC).
For this purpose, the Suzuki reaction has been chosen
as an example. The red dendrimer 1 (DR9–NHtBoc:
D=dendrimer, R=Red and nine terminal groups) was
treated with gaseous HCl in EtOAc at 0°C for 1 h to
give dendrimer 4. The commercially available hydroxy-
methylphenoxyacetic acid 2 was acetylated by treat-
ment with Ac2O in pyridine, and treated with pen-
tafluorophenol in the presence of DCC in DMF to
afford the activated ester 3. Coupling of 4 and 3 in the
presence of TEA in DMF yielded dendrimer 5, which
was then deprotected with K2CO3 in a mixture of
MeOH/CH2Cl2 (60/40). The desired dendrimer 6 with
the HMPA-linker (DR9HMPA–CH2OH) was obtained as
a red solid (Scheme 1).
To cleave the final compound from the dendrimer
support, the commonly used TFA/CH2Cl2 method was
adopted. The dendrimers 9a–f and 10a–f were treated
by TFA in CH2Cl2 at rt for 1 h, when the deprotection
was complete, as indicated by TLC, the reaction mix-
ture was evaporated under vacuum, the residue was
solubilized in a mixture of CH2Cl2/MeOH (60/40),
treated with K2CO3 overnight, and taken into a mixture
Scheme 2. Reagents and conditions: (a) 7a,b, Py, rt, 97% (8a) and 90% (8b); (b) ArB(OH)2, Pd(PPh3)4, Na2CO3 (2 M), DMF,
120°C, 59–100%; (c) TFA, CH2Cl2, 33–73%.

J. Zhang et al. / Tetrahedron Letters 42 (2001) 6683–6686
6685
Table 1. Yields (%) of Suzuki coupling and cleavage
of CH2Cl2/NaOH (1N). The insoluble red dendrimer
was filtered, the aqueous solution was washed twice
with CH2Cl2, acidified with HCl (1N), and extracted
with AcOEt. The desired acids 11a–f and 12a–f were
obtained as white solids. All products gave satisfactory
analytical spectra (Table 1).6
5. Zhang, J.; Drugeon, G.; L’hermite, N. Tetrahedron Lett.
2001, 42, 2599.
6. (a) Preparation of DR9HMPA–CH2OH 6: Step 1: To a red
suspension of dendrimer 1 (2.57 g, 1.02 mmol) in 10 ml of
AcOEt, was added at 0°C under nitrogen, 20 ml of HCl (g,
6 M) in AcOEt. The reaction mixture was allowed to reach
rt slowly, and then evaporated to dryness under vacuum to
give 1.97 g of the desired dendrimer 4 (yield: 100%). Step
2: To 60 ml of DMF, under nitrogen and at rt, was added
2.0 g of dendrimer 4 (1.03 mmol), 2.44 ml of TEA (17.5
mmol) and 4.84 g of activated ester 3 (12.4 mmol). The
mixture was allowed to stand at rt for 48 h and diluted in
AcOEt (200 ml), the organic layer was washed with NaOH
(1N), H2O (saturated NaCl), and evaporated to dryness
under vacuum to yield 3.26 g of dendrimer 5 (91%). Step
4: To a mixture of MeOH–CH2Cl2 (250 ml, 6:4), was
added dendrimer 4 (3.26 g, 0.94 mmol) and anhydrous
K2CO3 (2.34 g, 16 mmol), at rt and under a nitrogen. The
reaction mixture was stirred for 5 h and evaporated to
eliminate the solvents. The residue was taken into of
mixture of H2O–CH2Cl2 (100 ml:50 ml), the precipitate
formed at the interface was filtered, washed with H2O,
CH2Cl2 and Et2O to afford, after drying, compound 6
In conclusion, the colored dendrimer DR9HMPA–CH2OH
proved to be a good choice for supporting an acid
derivative, as demonstrated with the Suzuki reaction.
The reaction could be easily carried out in an organic
solvent in a homogenous way and it could be followed
by TLC. The purification could be efficiently performed
on Sephadex LH-20 SEC cartridges, and due to the
coloration, it is easy to collect the red fraction. This
shortened the purification time by SEC (less than 30
min), and the time of TLC analysis. We are currently
studying the introduction of other linkers to the den-
drimer DR9–NHtBoc, application to other reactions
and the way to automate the system for parallel synthe-
sis and parallel purification.
1
(2.09 g, 72%). Data for 6: H NMR (300 MHz, DMSO-d6)
Acknowledgements
l 1.15 (t, J=7 Hz, 3H), 1.81–1.90 (m, 18H), 1.93–12.02
(m, 6H), 3.30–3.35 (m, 18H), 3.47–3.49 (m, 8H), 3.51 (m,
2H), 3.59 (m, 2H), 3.99–407 (m, 24H), 4.38 (br d, J=5.5
Hz, 18H), 4.41 (br s, 18H), 5.04 (br t, J=5.5 Hz, 9H), 6.87
(app.d, J=8.5 Hz, 18H), 6.96 (app.d, J=9 Hz, 2H), 7.17
(br s, 8H), 7.19 (app.d, J=8.5 Hz, 18H), 7.83 (app.d, J=9
Hz, 2H); 7.92 (app.d, J=9 Hz, 2H), 8.08 (br t; J=5.5 Hz,
3H), 8.17 (br t; J=5.5 Hz, 6H), 8.18 (br s; 1H), 8.34
(app.d, J=9 Hz, 2H), 8.52 (br s, 2H), 8.64 (br s; 4H); MS
(LSIMS): [M+H]+=3083.9+; IR (Nujol) 3150–3500, 1653,
1600, 1583, 1541, 1506 cm−1; (b) General procedure for
attachment of acyl chloride to 6, ex. 8a: To a solution of
pyridine (20 ml), under nitrogen at rt, was added succes-
sively 7a (780 mg, 2.91 mmol), TEA (0.42 ml, 2.90 mmol),
DMAP (17 mg, 0.145 mmol) and 6 (200 mg, 0.065 mmol).
After 36 h of stirring at rt, the mixture was evaporated and
extracted with CH2Cl2 and H2O. The organic layer was
concentrated under vacuum and purified through a Sep-
hadex LH20 cartridge to give after evaporation of CH2Cl2
8a as red solid (325 mg, 97%). Data for 8a: MS (ESI):
[M+H]+=5154.64+; (c) Suzuki coupling, ex. 9a: To a solu-
We thank the Structural Analysis Department (Aventis,
Romainville) for performing the spectral analysis.
References
1. (a) Kim, R. M.; Chang, J. Comb. Chem. 2000, 353; (b)
Haag, R. Chem. Eur. J. 2001, 7, 327.
2. (a) Han, H.; Wolfe, M. M.; Brenner, S.; Janda, K. D.
Proc. Natl. Acad. Sci. USA 1995, 92, 6419; (b) Gravert, D.
J.; Janda, K. D. Chem. Rev. 1997, 97, 489.
3. Kim, R. M.; Manna, M.; Hutchins, S. M.; Griffin, P. R.;
Yates, N. A.; Bernick, A. M.; Chaman, K. T. Proc. Natl.
Acad. Sci. USA 1996, 93, 10012.
4. (a) Hovestad, N. J.; Ford, A.; Jastrzebski, J. T. B. H.; von
Koten, G. J. Org. Chem. 2000, 65, 6338; (b) Kantchev, A.
B.; Parquette, J. R. Tetrahedron Lett. 1999, 40, 8049.

6686
J. Zhang et al. / Tetrahedron Letters 42 (2001) 6683–6686
tion of DMF (10 ml) were added, under nitrogen, 8a
K2CO3 (0.09 g, 0653 mmol) overnight. The mixture was
then extracted with CH2Cl2 and NaOH (1N), the insolu-
ble red dendrimer was filtered, the aqueous layer was
washed with CH2Cl2, acidified with HCl (1N), and re-
extracted with AcOEt. The organic layer was dried over
MgSO4, filtered and evaporated to dryness to give 11a as
(0.142 g, 0.027 mmol), Pd (PPh3)4 (0.014 g, 0.012 mmol),
2-formylphenylboronic acid (0.074 g, 0.496 mmol) and
Na2CO3 (0.31 ml, 2 M in H2O). The mixture was heated
at 110°C for one night and extracted with AcOEt and
H2O. The organic layer was evaporated and purified
through a Sephadex LH20 cartridge with CH2Cl2 to yield
9a as red solid (0.11 g, 82%). Data for 9a: MS (FAB):
[M+H]+=4959.3+; (d) Cleavage procedure, ex. 11a: To a
mixture of TFA/CH2Cl2 (20:80, 10 ml) was added 9a
(0.090 g, 0.018 mmol) under nitrogen. After 1 h at rt, the
solvent was evaporated, the residue was taken up with a
mixture of MeOH/CH2Cl2 (60:40, 10 ml), treated with
1
a white solid (0.018 g, 50%). Data for 11a: H NMR (300
MHz, DMSO-d6) l 7.30 (br d, J=7.5 Hz, 1H), 7.34 (dd,
J1=7.5 Hz, J2=1.5 Hz, 1H), 7.56 (m, 2H), 7.65 (td,
J1=7.5 Hz, J2=1.5 Hz, 1H), 7.69 (td, J1=7.5 Hz, J2=
1.5 Hz, 1H), 7.89 (dd, J1=7.5 Hz, J2=1.5 Hz, 1H), 7.95
(dd, J1=7.5 Hz, J2=1.5 Hz, 1H), 9.71 (s, 1H), 12.82 (br
s, 1H); MS (FAB): [M−H]−=225−.