Synthesis and manipulation of orthogonally protected dendrimers: Building blocks for library synthesis

Article abstract of DOI:10.1002/anie.200460031

Fruits of the laboratory: Dendrimers containing up to six different functional groups can be prepared in multistep one-pot reactions by using a selective manipulation and deprotection strategy (see example). The authors call these multifunctional dendrime

Full text of DOI:10.1002/anie.200460031

Angewandte
Chemie
Communications
The synthesis of dendrimers containing up to six different functional
groups by multistep one-pot reactions involving a selective manipu-
lation and deprotection strategy is described by E. E. Simanek and
M. B. Steffensen in their Communication on the following pages.
Angew. Chem. Int. Ed. 2004, 43, 5177
DOI: 10.1002/anie.200460031
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5177

Communications
Dendrimer Synthesis
(TBDPS, C4H9SiPh2) ether can be removed using tetrabutyl-
ammonium fluoride;[9] the thiopyridyl group (PyrS) can
participate in thiol–disulfide exchange reactions with free
thiols or be removed with dithiothreitol and subsequently
alkylated or acylated;[10] the tert-butyloxycarbonate groups
(Boc: C4H9OCO) can be removed with trifluoroacetic acid or
formic acid;[11] and the monochlorotriazine of 2 can be treated
with a variety of amine nucleophiles.[12] Here, we describe the
Synthesis and Manipulation of Orthogonally
Protected Dendrimers: Building Blocks for
Library Synthesis**
Mackay B. Steffensen and Eric E. Simanek*
Even when reduced to trivial
manipulations, the synthesis
of dendrimers still remains
laborious[1] with few excep-
tions—notably, one-pot syn-
theses.[2] As applications are
pursued for these architec-
tures, the need to execute
structure–property relation-
ships only further increases
the burden of synthesis. Not
surprisingly, the number of
reports of libraries of den-
drimers are exceedingly few,
and in many cases these libra-
ries are the result of substoi-
chiometric (with respect to
the number of reactive sur-
face groups) and statistical
reactions at the periphery to
yield cocktails of molecules as
opposed to single-molecule
chemical entities.[3] We and
others interested in these
architectures have continued
to pursue the preparations of
versatile targets that are amenable to post-synthetic substoi-
chiometric manipulation.[4] In our hands, the targets reported
here (1 and 2) are the building blocks of drug-delivery
vehicles,[5] while in other hands they may be the basis for
novel materials such as poly(block) polymers or other novel
materials.[6] Here, we describe the synthesis of these mole-
cules and their manipulation upon selective removal of
protecting groups.
Dendrimers 1 and 2 have five and six groups for
manipulation, respectively:[7] the free hydroxy group(s) are
readily acylated; the levulinic acid ester (Lev) can be
unmasked using hydrazine;[8] the tert-butyldiphenylsilyl
synthesis of these targets and establish that selective depro-
tection and subsequent reaction can be effected.
The convergent synthesis of these molecules proceeds in a
highly efficient manner. Of the four building blocks used (4–7,
Scheme 1; each indicated with a different color), three (4–6)
originate from a common intermediate (3) and are available
in yields exceeding 90%. The selective substitution of the
triazine ring reduces the synthesis of these dendrimers to a
trivial procedure. The use of diamines that react chemo-
selectively with monochlorotriazines obviates the need for
protecting groups or great excesses of reagents.[12]
Intermediate 3 is obtained in quantitative yields by
treating the bisprotected triamine[11] with cyanuric chloride
followed by the addition of 2-(2-aminoethoxy)ethanol in the
same pot. This protocol is commonly executed on a 10-gram
scale. The synthesis of 6–8 from the common intermediate 3
proceeds cleanly, starting with protection of the alcohol using
TBDPSCl to yield 4 or esterification using levulinic acid and
DCC to yield 5. These reactions are routinely run on a 5-g
scale and the products are easily purified by chromatography
on silica gel. The secondary amine of 4-aminomethylpiper-
idine (4-AMP) reacts more than 20 times faster than the
primary amine to provide the building blocks 6–8 in greater
than 90% overall yield.[12]
[*] M. B. Steffensen, Dr. E. E. Simanek
Texas A&M University
Department of Chemistry
College Station, TX 77843-3255 (USA)
Fax: (+1)979-845-9452
E-mail: simanek@tamu.edu
[**] This work was supported by a grant from the N.I.H. (NIGMS
64650). M.B.S. received a predoctoral fellowship from the Chemis-
try-Biology Training Grant (NIH TM GM08523).
Supportinginformation for this article is available on the WWW
under http://www.angewandte.org or from the author.
5178
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200460031
Angew. Chem. Int. Ed. 2004, 43, 5178 –5180

Angewandte
Chemie
Scheme 1. Synthesis of 1 and 2. Reagents and conditions: a) TBDPSCl, imidazole, THF, RT, 8 h; b) levulinic acid, DCC, DMAP, THF, RT, 8 h; c) 4-
AMP, THF, RT, 8 h; d) 7+C3N3Cl3, DIEA, THF, 08C, 2 h; then 6, RT, 24 h; then 4-AMP, RT, 18 h; e) C3N3Cl3, DIEA, THF, 08C, 2 h; then 8, RT, 18 h;
then 4-AMP, RT, 18 h; f) 11+C3N3Cl3, DIEA, THF, 08C, 2 h; then 6, RT, 18 h; g) piperazine, THF, 08C, 20 h, extraction, then 2, RT, 10 days. DIEA=
diisopropylethylamine, DCC=dicyclohexylcarbodiimide, DMAP=4-dimethylaminopyridine, 4-AMP=4-aminomethylpiperidine.
The preparation of the fourth building block (9) occurs in
a single-pot reaction that commences with monosubstitution
of cyanuric chloride with 2-pyridyldithioethylamine (PDA),
followed by addition of the bisprotected triamine, and
concluding with the addition of 4-AMP. The overall yield
for this reaction sequence is 93%, with the final product
isolated by chromatography. Synthesis of 10 occurs in a single-
pot procedure that relies on first treating 9 with cyanuric
chloride, then 8, and concluding with addition of 4-AMP. This
series of manipulations provides 10 in only 79% yield. The
uncharacteristically low yield for this series of reactions is
presumably a consequence of the condensation of the keto
group of the levulinic ester with the free primary amine upon
workup. Signals corresponding to the dimerization of 10
minus the loss of water appear in the mass spectra.
The synthesis of 11 proceeds in a manner similar to that of
10. This reaction is illustrative of the ease at which these
multistep, one-pot reactions can be monitored by thin-layer
chromatography (TLC). Addition of cyanuric chloride to a
solution of 7 (Rf = 0.3 in 9:1 CH2Cl2:CH3OH) at 08C results in
the dichlorotriazine intermediate forming rapidly to give a
single new spot by TLC (Rf = 0.6 in 9:1 CH2Cl2:CH3OH).
After addition of 6 and consumption of the dichlorotriazine, a
single spot is again visible by TLC (Rf = 0.4 in 9:1
CH2Cl2:CH3OH) which corresponds to the monochlorotri-
azine. The final reaction with 4-AMP gives 11, a material that
has an Rf value of 0.25 (9:1 CH2Cl2:CH3OH) because of the
presence of a free primary amine. Intermediate 11 is obtained
in 99% yield.
yield. This route has been executed to produce more than 10g
of 2. (The route with the lowest yield commences with 9 and
produces 2 in three steps and 70% overall yield.) Dendrimer
1 could be obtained by dimerizing 2 with piperazine, but to
avoid errors that could arise in the stoichiometry of the
reactants as a result of the substantial differences in the
molecular weight of 2 and piperazine, we adopt a two-step
procedure.[13] Initially, 2 is treated with an excess of piper-
azine. The excess piperazine is then removed and the
remaining material is treated with 1. The process proceeds
in 89% overall yield and the product is purified by
chromatography on silica gel.
We evaluated the selective manipulation and deprotection
of these scaffolds using 2 (Scheme 2). In short, nucleophilic
aromatic substitution with amino acid derivatives can be
executed to yield 12. The free hydroxy group of 12 is
amenable to modification with a chloroformate group to yield
carbonate 13. The propargyloxycarbonyl (Poc) group can
serve as either a protecting group[13] or a reagent for click
chemistry.[14] Selective deprotection of the silyl ether is
achieved with tetrabutylammonium fluoride to yield 14.
Cleavage of the levulinic acid ester is facilitated with
hydrazine to yield 15. The pyridyl disulfide group on 1 was
cleaved with dithiothreitol before coupling with the commer-
cially available biotin-PDA (an activated biotin–disulfide
conjugate) to form disulfide 16. Triscarboxylethylphosphane
reduced the disulfide with concommitent removal of the Poc
group. The generality of this reaction is unclear as previous
methods for removal of the Poc group typically required
heterogeneous solutions of tetrathiomolybdate.[13c] Excess
hydrazine can also lead to loss of the Poc group.
To obtain 2, intermediate 11 is treated sequentially with
cyanuric chloride and then 10. Overall, a total of 18 reactions
are involved in the synthesis. The longest continuous reaction
sequence is only five reactions and provides 2 in 80% overall
Targets 1 and 2 represent useful building blocks for the
preparation of dendrimer libraries for diverse purposes.
Angew. Chem. Int. Ed. 2004, 43, 5178 –5180
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5179

Communications
[2] a) M. Okaniwa, K. Takeuchi, M. Asai, M.
Ueda, Macromolecules 2002, 35, 6232 –6238;
b) S. P. Rannard, N. J. Davis, J. Am. Chem. Soc.
2000, 122, 11729—11770.
[3] a) P. Singh, Bioconjugate Chem. 1998, 9, 54 –
63; b) G. R. Newkome, C. D. Weis, C. N.
Moorefield, G. R. Baker, B. J. Childs, J. Epper-
son, Angew. Chem. 1998, 110, 318 –321;
Angew. Chem. Int. Ed. 1998, 37, 307 –310;
c) G. R. Newkome, B. J. Childs, M. J. Rourk,
G. R. Baker, C. N. Moorefield, Biotechnol.
Bioeng. 1999, 61, 243 –253; d) J. R. Baker, A.
Quintana, L. Piehler, M. Banazak-Holl, D.
Tomalia, E. Raczka, Biomed. Microdevices
2001, 3, 61 –69.
[4] a) W. Zhang, D. T. Nowlan III, L. M. Thom-
son, W. M. Lackowski, E. E. Simanek, J. Am.
Chem. Soc. 2001, 123, 8914 –8922; b) Z. Bo, A.
Schäfer, P. Franke, A. D. Schlüter, Org. Lett.
2000, 2, 1645 –1648; c) H.-F. Chow, C.-F.
Leung, K.-W. Wong, L. Xi, Angew. Chem.
2003, 115, 5069 –5073; Angew. Chem. Int. Ed.
2003, 42, 4919 –4923; d) S. Li, M. L. Szalai,
R. M. Kevwitch, D. V. McGrath, J. Am. Chem.
Soc. 2003, 125, 10516 –10517; e) K. L. Wooley,
C. J. Hawker, J. M. J. FrØchet, J. Chem. Soc.
Perkin Trans. 1 1991, 1059 –1076; f) W. Zhang,
S. E. Tichy, L. M. PØrez, G. Maria, P. A.
Lindahl, E. E. Simanek, J. Am. Chem. Soc.
2003, 125, 5086 –5094.
[5] a) S.-E. Stiriba, H. Frey, R. Haag, Angew.
Chem. 2002, 114, 1385 –1390; Angew. Chem.
Int. Ed. 2002, 41, 1329 –1334; b) F. Aulenta, W.
Hayes, S. Rannard, Eur. Polym. J. 2003, 39,
1741 –1771; c) M. J. Cloninger, Curr. Opin.
Chem. Biol. 2002, 6, 742 –748; d) R. Esfand,
D. A. Tomalia, Drug Discovery Today 2001, 6,
427 –436; e) A. K. Patri, I. J. Majoros, J. R.
Baker, Jr., Curr. Opin. Chem. Biol. 2002, 6,
466 –471.
Scheme 2. Manipulation of the core. Reagents and conditions: a) propargylchloroformate, pyridine,
THF, RT, 2 h; b) TBAF, THF, RT, 4 h; c) hydrazine, THF, RT, 2 h; d) DTT, CH2Cl2, RT, 2 h; e) biotin-
PDA, DMF, RT, 2 h. Boc-Ala-4-Amp: see SupportingInformation for preparation, TBAF =tetrabutyl-
ammonium fluoride, DTT=dithiothreitol, biotin-PDA=biotin-pyridyldithioethylamine.
[6] Materials are available in limited supply to
other investigators by contacting Professor
Simanek.
These molecules suffer the limitation that the deprotection
[7] Inspiration from: C.-H. Wong, X.-S. Ye, Z. Zhang, J. Am. Chem.
Soc. 1998, 120, 7137 –7138.
[8] a) J. H. van Boom, P. M. J. Burgers, Tetrahedron Lett. 1976, 17,
4875 –4878; b) N. Jeker, C. Tamm, Helv. Chim. Acta 1984, 67,
190.
[9] S. Hanessian, P. Lavallee, Can. J. Chem. 1977, 55, 562 –565.
[10] J. Connor, A. J. Schroit, Biochemistry 1988, 27, 848 –851.
[11] D. A. Westerberg, P. L. Carney, P. E. Rogers, S. J. Kline, D. K.
Johnson, J. Med. Chem. 1989, 32, 236 –243.
[12] M. B. Steffensen, E. E. Simanek, Org. Lett. 2003, 5, 2359 –2361.
[13] a) S. Sinha, P. Ilankumaran, S. Chandrasekaran, Tetrahedron
Lett. 1999, 40, 771 –774; b) R. G. Bhat, S. Sinha, S. Chandrase-
karan, Chem. Commun. 2002, 812 –813; c) P. R. Sridhar, S.
Chandrasekaran, Org. Lett. 2002, 4, 4731 –4733.
steps occur in yields of about 90% and release the protecting
group as a by-product into solution, thus necessitating
purification. The incorporation of these groups onto a solid
support, a current aim of our research, addresses these
limitations. We refer to these dendrimers as molecular fruit-
salad trees, the nanoscale variant of a tree produced by
grafting branches that yield different fruits on a common
trunk.[15] Further studies on these architectures will be
reported in due course.
Received: March 17, 2004
Published Online: August 12, 2004
[14] H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. 2001, 113,
2056 –2075; Angew. Chem. Int. Ed. 2001, 40, 2004 –2021.
[15] Fruit-salad trees are commercially available: http://www.fruitsal-
adtrees.com/.
Keywords: amines · dendrimers · protectinggroups ·
synthesis design
.
[1] a) S. M. Grayson, J. M. J. FrØchet, Chem. Rev. 2001, 101, 3819 –
3867; b) G. R. Newkome, F. Vögtle, C. N. Moorefield, Dendri-
mers and Dendrons: Concepts, Syntheses, Applications, VCH,
New York, 2001; c) J. M. J. Fr›chet, D. A. Tomalia, Dendrimers
and Other Dendritic Polymers, Wiley, New York, 2002.
5180
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 5178 –5180