Solid phase synthesis of aryl-ether dendrimers

Article abstract of DOI:10.1039/b100003i

Solid phase chemistry can be used to prepare, in excellent yield and purity, a range of Frechet-type aryl-ether dendrimers, for use in either dendrimer conjugation studies, resin loading enhancement or the wedge based synthesis of larger dendrimers.

Full text of DOI:10.1039/b100003i

View Article Online / Journal Homepage / Table of Contents for this issue
Solid phase synthesis of aryl-ether dendrimers
Andrea Basso,a Brian Evans,b Neil Peggb and Mark Bradley*a
a Department of Chemistry, University of Southampton, Highfield, Southampton, UK SO17 1BJ.
E-mail: MB14@soton.ac.uk
b Glaxo Wellcome Research and Development, Gunnels Wood Road, Stevenage, UK SG1 2NY
Received (in Cambridge, UK) 2nd January 2001, Accepted 7th March 2001
First published as an Advance Article on the web 26th March 2001
Solid phase chemistry can be used to prepare, in excellent
yield and purity, a range of Fréchet-type aryl-ether den-
drimers, for use in either dendrimer conjugation studies,
resin loading enhancement or the wedge based synthesis of
larger dendrimers.
Over the past few years we have shown that resin-bound
polyamidoamine (PAMAM) dendrimers are readily synthe-
sized1,2 and are inert under a broad range of chemical
conditions3 such as coupling reactions, nucleophilic displace-
ments or borohydride reductions. However, synthetic patterns
requiring the use of strong reducing agents or strong bases have
to be avoided. In the area of solution dendrimer synthesis
Fréchet4 has reported the construction of arylether dendrimers
via the alkylation of bromobenzylic dendrimeric fragments with
3,5-dihydroxybenzyl alcohol, with which he also introduced the
concept of convergent dendrimer synthesis. Since polyether
dendrimers are substantially more chemically inert than the
PAMAM dendrimers and because of the problems associated
with solution dendrimer synthesis, the solid-phase synthesis of
Fréchet-type aryl-ether dendrimers (Fig. 1) was targeted both as
a means of resin loading enhancement as well as a convenient
approach to dendrimer wedge4 synthesis. Since the building
blocks originally used by Fréchet were considered to be
unsuitable for solid-phase synthesis, 3,5-bis(acetoxymethyl)-
phenol (1) was used and was prepared by modification of the
literature procedure.5,6 The presence of a phenolic group and of
two acetyl-protected primary alcohols allowed the synthesis of
the dendrimer via a two-step iterative procedure, consisting of a
Mitsunobu condensation followed by ester hydrolysis. The
synthesis was first carried out using an analytical construct,7
consisting of glycine attached to the TFA-cleavable Rink
linker8 allowing ready characterisation of the intermediates and
optimisation of the Mitsunobu condensations and ester hydroly-
sis. Compound 2, obtained in two steps from 1, was coupled
onto Gly-Rink-PS resin under standard conditions (Scheme 1)
to give 3.
Scheme 1 (i) tert-Butylbromoacetate, K2CO3, CH3CN; (ii) TFA, DCM
(2+3), (iii) H-Gly-Rink-PS, DIC,† HOBt, DMF.
Fig. 2.9 Dendrimers were subsequently synthesized directly on
hydroxymethylpolystyrene 4 (Scheme 2), which was obtained
from commercial Merrifield resin (1% DVB, 0.93 mmol g21
,
90–106 mm) via displacement of the chloride with caesium
acetate in DMF and hydrolysis of the resulting ester with
Bu4NOH(aq) in THF. This was extended to give dendrimer
resin 6 as shown in Scheme 2.
The loading of resin 4 was determined to be 0.82 mmol g21
(0.44 nmol bead21). Synthesis of generation 3.0 dendrimer
resin gave a resin with a loading of 2.85 mmol g21 (3 nmol
Hydrolysis of 3 followed by Mitsunobu condensation with
(1) afforded Generation 2.0 dendrimer after TFA cleavage. The
most efficient conditions for the hydrolysis were found to be
Bu4NOH(aq)–THF 1+3, while the Mitsunobu reaction was
performed with DIAD–PPh3† in THF. Generation 3.0 and 4.0
dendrimers were synthesized using the same procedure and the
crude HPLC trace of Generation 3.0 dendrimer is shown in
Fig. 2 HPLC trace of polyether dendrimer Generation 3.0.
Scheme 2 (i) (1), DIAD, PPh3, THF; (ii) Bu4NOH (aq), THF (1:3); (iii)
Repeat twice (i) and (ii), (iv) methyl-4-hydroxybenzoate, DIAD, PPh3,
THF; (v) LiAlH4, THF.
Fig. 1 Fréchet-type aryl-ether dendrimer fragment.
DOI: 10.1039/b100003i
Chem. Commun., 2001, 697–698
This journal is © The Royal Society of Chemistry 2001
697

View Article Online
Table 1 Bead diameter (mm) in different solvents (average of 10 beads)
on hydroxymethyl polystyrene resins and that the final loading
is sufficient for multiple single-bead screenings and is inert
under severe chemical conditions. Moreover we have shown
that polyether dendrimers can be conveniently synthesised on
the solid phase. No purification of the intermediates (usually a
non-trivial step, taking in account the molecular weight and
character of these molecules) is required to obtain highly pure
compounds. The wedges synthesised on solid phase could be
used to assemble bigger fragments according to the convergent
dendrimer methodology outlined by Fréchet.
Dry
DCM
DMF
MeOH
H2O
THF
4
6
90
110
115
165
145
195
155
185
190
95
105
110
95
110
115
175
150
190
(Ac)8-6
bead21) (7 times that of the initial resin). Since the synthesis of
this dendrimer involved 7 distinct Mitsunobu condensations and
14 distinct ester hydrolyses and the final loading was 86% of
that expected this implies that each reaction had proceeded to
99.3% completion. The swelling properties of this resin were
analyzed and compared with those of hydroxymethylpolystyr-
ene and are summarized in Table 1.
The diameter of the dry resin beads increased by about 25%
after the dendrimerisation process. Acetyl-protected dendrimer
resin had the same swelling trend as the hydroxymethylpoly-
styrene 4, while resin 6 presented some interesting features with
apolar solvents like DCM and THF. In these solvents the resin
swelled much less than the parent acetyl-protected resin and
also less than resin 4, due to the high density of hydroxy groups.
The two resins 4 and 6 were also compared in terms of reaction
kinetics. Fmoc-Ala-OH was coupled onto the two resins and
Fmoc and quantitative ninhydrin tests were carried out with
samples removed over 2 h. Differences in reactivity were found
to be negligible.
To prove the versatility of this new resin, methyl 4-hydroxy-
benzoate was coupled via a Mitsunobu condensation on to resin
6 and the resin-bound methyl ester was then reduced with
LiAlH4 (Scheme 2). This two-step procedure was found to be an
efficient alternative method to introduce the Wang linker onto a
polystyrene support. The polyether dendrimer resin was
perfectly stable to LiAlH4 reduction and the final loading of the
dendrimer–Wang resin 7 was found to be 2.3 nmol bead21. The
utility of resin 7 was demonstrated by synthesising the
hexapeptide Leu-Enkephaline-Lys. Following cleavage with
TFA–H2O 95+5 and purification, the peptide was isolated in
66% yield, relative to the loading of resin 4.
We thank the BBSRC for funding (ROPA).
Notes and references
† DIAD
imide.
= diisopropylazodicarboxylate. DIC = diisopropylcarbodi-
1 V. Swali, N. J. Wells, G. J. Langley and M. Bradley, J. Org. Chem., 1997,
62, 4902.
2 N. J. Wells, A. Basso and M. Bradley, Biopolymers (Peptide Science),
1998, 47, 381.
3 (a) A. Basso, B. Evans, N. Pegg and M. Bradley, Tetrahedron Lett., 2000,
41, 3763; (b) A. Basso, B. Evans, N. Pegg and M. Bradley, Eur. J. Org.
Chem., 2000, 23, 3887.
4 C. J. Hawker and J. M. J. Fréchet, J. Chem. Soc., Chem. Commun., 1990,
1010.
5 3,5-Bis(methoxycarbonyl)phenol was protected as the tert-butyldime-
thylsilyl ether before reduction with LiAlH4. After acetylation of the
primary alcohols, silyl protection was removed with TFA–H2O 9+1.
6 P. R. Ashton, D. W. Anderson, C. L. Brown, A. N. Shipway, J. F.
Stoddart and M. S. Tolley, Chem. Eur. J., 1998, 4, 781.
7 S. C. Mckeown, S. P. Watson, R. A. E. Carr and P. Marshall, Tetrahedron
Lett., 1999, 40, 2407.
8 For a review on linkers and cleavage strategies, see: F. Guiller, D. Orain
and M. Bradley, Chem. Rev., 2000, 100, 2091.
9 NMR of Generation 3.0 dendrimer: DMSO-d6: d(1H) = 7.30 (s, 1H, ArH
Gen 1.0), 7.24 (s, 2H, ArH Gen 2.0), 7.19 (br s, 6H, ArH Gen 1.0 and 2.0),
7.06 (br s, 2H, NH2), 6.97 (s, 4H, ArH Gen 3.0), 6.95 (s, 8H, ArH Gen
3.0), 5.22 (s, 4H, CH2 Gen 1.0), 5.17 (s, 8H, CH2 Gen 2.0), 4.68 (s, 2H,
O-CH2-CO), 4.55 (s, 16H, CH2 Gen 3.0), 3.85 (br s, 2H, NH-CH2-CO);
d(13C): 171.1, 168.3, 159.0, 158.8, 158.4, 144.3, 139.5, 139.3, 120.0,
119.3, 117.4, 113.8, 113.5, 111.4, 69.5, 69.3, 67.4, 63.2, 41.8; m/z (TOF
LD+): 1107.5 (100%, (M + Na)+), 1123.4 (30%, (M + K)+).
In conclusion, we have demonstrated that the polyether
dendrimer resin can be conveniently and efficiently synthesised
698
Chem. Commun., 2001, 697–698