Efficient multigram synthesis of the repeating unit of gallic acid-triethylene glycol dendrimers

Article abstract of DOI:10.1021/ol201677k

A multigram synthesis of the repeating unit of GATG (gallic acid-triethylene glycol) dendrimers is described through an efficient and cost-effective route. These conditions overcome major problems precluding scaling up and afford product in excellent overall yield and purity. Special attention has been paid in this process to green chemistry principles: atom economy, safety, and waste reduction. This scheme could be easily adapted for the preparation of similar dendritic systems.

Full text of DOI:10.1021/ol201677k

ORGANIC
LETTERS
2011
Vol. 13, No. 17
4522–4525
Efficient Multigram Synthesis of the
Repeating Unit of Gallic Acid-Triethylene
Glycol Dendrimers
Sandra P. Amaral, Marcos Fernandez-Villamarin, Juan Correa, Ricardo Riguera,* and
Eduardo Fernandez-Megia*
Department of Organic Chemistry and Center for Research in Biological Chemistry and
Molecular Materials (CIQUS), University of Santiago de Compostela,
Jenaro de la Fuente s/n, 15782 Santiago de Compostela, Spain
ricardo.riguera@usc.es; ef.megia@usc.es
Received June 22, 2011
ABSTRACT
A multigram synthesis of the repeating unit of GATG (gallic acid-triethylene glycol) dendrimers is described through an efficient and cost-effective
route. These conditions overcome major problems precluding scaling up and afford product in excellent overall yield and purity. Special attention
has been paid in this process to green chemistry principles: atom economy, safety, and waste reduction. This scheme could be easily adapted for
the preparation of similar dendritic systems.
Dendrimers are synthetic tree-like macromolecules com-
posed of repetitive layers of branching units that are pre-
pared in a controlled iterative fashion, through generations
with discrete properties. They are characterized by well-defined
structures, nil dispersity, and a globular morphology in the
nanometre scale.1 These characteristics have rendered den-
drimers with applications in a plethora of fields, including
catalysis and materials science.2 In addition, the inherent
multivalency of dendrimers allows for the controlled display
of specific ligands, drugs, and targeting and imaging agents.3
During the past decade, great efforts have been devoted
to the development of more efficient synthetic methodologies
for the facile preparation and functionalization of den-
drimers,4 a fact that has been fueled by the rapid adoption
of the Sharpless’ click concept5 in this field.6
GATG (gallic acid-triethylene glycol) is an emergent
family of dendrimers that are prepared divergently from
repeating unit 1 (Figure 1). GATG sialodendrimers and
dendronized chitosans up to the second generation (G2)
were initially described by the group of Roy as potential
microbicides.7 Despite this interesting application, the
preparation of repeating unit 1 (four steps from triethylene
(4) (a) Carlmark, A.; Hawker, C.; Hult, A.; Malkoch, M. Chem. Soc.
Rev. 2009, 38, 352. (b) Brauge, L.; Magro, G.; Caminade, A.-M.;
Majoral, J.-P. J. Am. Chem. Soc. 2001, 123, 6698. (d) Grayson, S. M.;
ꢀ
Frechet, J. M. J. Chem. Rev. 2001, 101, 3819.
€
(1) (a) Vogtle, F.; Richardt, G.; Werner, N. Dendrimer Chemistry;
Wiley-VCH: Weinheim, 2009. (b) Tomalia, D. A. Prog. Polym. Sci. 2005,
30, 294.
(2) (a) Astruc, D.; Boisselier, E.; Ornelas, C. Chem. Rev. 2010, 110,
1857. (b) Rosen, B. M.; Wilson, C. J.; Wilson, D. A.; Peterca, M.; Imam,
(5) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed.
2001, 40, 2004.
(6) (a) Antoni, P.; Hed, Y.; Nordberg, A.; Nystrom, D.; von Holst,
H.; Hult, A.; Malkoch, M. Angew. Chem., Int. Ed. 2009, 48, 2126.
(b) Franc, G.; Kakkar, A. Chem. Commun. 2008, 5267. (c) Killops, K. L.;
Campos, L. M.; Hawker, C. J. J. Am. Chem. Soc. 2008, 130, 5062.
(d) Wu, P.; Feldman, A. K.; Nugent, A. K.; Hawker, C. J.; Scheel, A.;
€
ꢀ
M. R.; Percec, V. Chem. Rev. 2009, 109, 6275. (c) Mery, D.; Astruc, D.
ꢀ
Coord. Chem. Rev. 2006, 250, 1965. (d) Liang, C.; Frechet, J. M. J. Prog.
Polym. Sci. 2005, 30, 385.
ꢀ
Voit, B.; Pyun, J.; Frechet, J. M. J.; Sharpless, K. B.; Fokin, V. V. Angew.
(3) (a) Menjoge, A. R.; Kannan, R. M.; Tomalia, D. A. Drug
DiscoveryToday2010, 15, 171. (b) Shcharbin, D.; Pedziwiatr, E.; Blasiak,
J.; Bryszewska, M. J. Controlled Release 2010, 141, 110. (c) Rolland, O.;
Turrin, C.-O.; Caminade, A.-M.; Majoral, J.-P. New J. Chem. 2009, 33,
1809. (d) Tekade, R. K.; Kumar, P. V.; Jain, N. K. Chem. Rev. 2008,
109, 49.
Chem., Int. Ed. 2004, 43, 3928.
(7) (a) Sashiwa, H.; Shigemasa, Y.; Roy, R. Macromolecules 2001, 34,
3905. (b) Meunier, S. J.; Wu, Q.; Wang, S.-N.; Roy, R. Can. J. Chem.
1997, 75, 1472. (c) Roy, R.; Park, W. K. C.; Wu, Q.; Wang, S.-N.
Tetrahedron Lett. 1995, 36, 4377.
r
10.1021/ol201677k
Published on Web 08/05/2011
2011 American Chemical Society

glycol, 23%)7 proved to be a hurdle in accessing
large quantities of GATG dendrimers and to higher
generations.
These novel applications of GATG dendrimers render
the large scale preparation of 1 of great interest. However,
attempts to scale up the synthesis shown in Scheme 1
(in black)8 to batches larger than 8 g were unfortunately
troublesome due to laborious chromatographic purifica-
tions of 5 and 7, which finally resulted in lower yields
and unfeasibility. Herein, we report a multigram synthesis
basedongreenchemistryprinciples(atom economy, safety,
waste reduction)15 which affords batches of 1 larger than
100 g in an excellent overall yield and purity (Scheme 1, in
blue). Key points for this synthesis are (i) a safer prepara-
tion of 3 by lowering the temperature of the reaction and
the use of H2O as solvent; (ii) the replacement of tosylate 5
by chloride 4, which is efficiently prepared in a solventless
process and purified by nonchromatographic means, pro-
viding a more atom-economical production of 7; and (iii) a
general reduction of waste generated.
The synthesis of 1 starts with the substitution of a chlo-
ride for an azide group in 2 (NaN3). As seen in Scheme 1
and Table 1 (entry 1), this process traditionally required
high temperatures (100 °C) and the use of DMF as solvent
(0.25 M).8 In spite of the relative thermal stability of NaN3
and 3 (see the Supporting Information, SI), we decided to
improve the safety of this step for multigram synthesis by
performing the reaction under milder conditions. Thus, a
set of experiments were designed to reduce the temperature
of the reaction. As seen in Table 1, although lowering the
temperature from 100 to 85 °C had no effect on yield (entry 2),
a further reduction to 75 °C led to incomplete conversions
(entry 3). This could be overcome by increasing the con-
centration from the original 0.25 to 1 M (entry 4), which
also presents the advantage of reducing the amount of sol-
vent used. Indeed, solvents account for the vast majority of
Aware of these limitations, in 2006 our research group
successfully reported an improved synthetic route to 1
from commercially available chlorotriethylene glycol 2
(77% overall yield, Scheme 1 in black) which has paved
the way for the preparation of GATG dendrimers and
theirblock copolymerswithpoly(ethylene glycol) (PEG) in
larger quantities and up to G4.8À10 The functionalization
of these dendrimers by means of the Cu(I)-catalyzed
azideÀalkyne cycloaddition (CuAAC)11 has been straight-
forward in our hands.8À13 The resulting functionalized
dendrimers have emerged as interesting tools in the study
of the multivalent carbohydrateÀreceptor interaction, the
dynamics of dendrimers, and the preparation of polyion
complex micelles and dendriplexes for gene therapy.10,12
More recently, we have developed GATG dendritic con-
trast agents for MRI and as inhibitors of the dimerization
of the capsid protein (CA) of HIV-1.13 In another example,
the group of Kinbara and Aida has described a guanidi-
nated GATG dendrimer of G2 as a molecular glue with the
ability to stabilize microtubules and to inhibit the sliding
motion of actomyosin.14
Table 1. Optimized Conditions for the Synthesis of 3
scale additive concn
temp time yield
entrya
(g)
(equiv)
(M)
solvent (°C)
(h)
(%)b
1
2
3
4
5
6
7
8
0.1
0.1
0.1
0.1
0.1
0.1
0.1
150
À
0.25
0.25
0.25
1
DMF
DMF
DMF
DMF
H2O
H2O
H2O
H2O
100
85
75
75
75
75
75
75
12
12
12
14
44
32
32
48
100
100
85
À
À
À
100
100
100
100
100
À
1
À
2
Figure 1. Repeatingunit andthird generationGATG dendrimer.
Nal (0.1)
2
À
2
a In all cases 2 equiv of NaN3 were used. b Yields refer to conversions
determined by 1H NMR (D2O), except entry 8.
(8) Fernandez-Megia, E.; Correa, J.; Rodrıguez-Meizoso, I.; Riguera,
R. Macromolecules 2006, 39, 2113.
(9) Fernandez-Megia, E.; Correa, J.; Riguera, R. Biomacromolecules
2006, 7, 3104.
~
(10) Ravina, M.; de la Fuente, M.; Correa, J.; Sousa-Herves, A.;
Pinto, J.; Fernandez-Megia, E.; Riguera, R.; Sanchez, A.; Alonso, M. J.
ꢀ
(13) (a) Fernandez-Trillo, F.; Pacheco-Torres, J.; Correa, J.; Ballesteros,
ꢀ
P.; Lopez-Larrubia, P.; Cerdan, S.; Riguera, R.; Fernandez-Megia, E.
ꢀ
Biomacromolecules DOI: 10.1021/bm2004466. (b) Domenech, R.; Abian,
O.; Bocanegra, R.; Correa, J.; Sousa-Herves, A.; Riguera, R.; Mateu, M. G.;
ꢀ
Fernandez-Megia, E.; Velazquez-Campoy, A.; Neira, J. L. Biomacromole-
Macromolecules 2010, 43, 6953.
(11) (a) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless,
K. B. Angew. Chem., Int. Ed. 2002, 41, 2596. (b) Tornøe, C. W.;
Christensen, C.; Meldal, M. J. Org. Chem. 2002, 67, 3057.
cules 2010, 11, 2069.
(14) (a) Okuro, K.; Kinbara, K.; Takeda, K.; Inoue, Y.; Ishijima, A.;
Aida, T. Angew. Chem., Int. Ed. 2010, 49, 3030. (b) Okuro, K.; Kinbara,
K.; Tsumoto, K.; Ishii, N.; Aida, T. J. Am. Chem. Soc. 2009, 131, 1626.
(15) (a) Anastas, P.; Eghbali, N. Chem. Soc. Rev. 2010, 39, 301. (b) Li,
C.-J.; Trost, B. M. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 13197.
€ ꢀ
(12) (a) Novoa-Carballal, R.; Sawen, E.; Fernandez-Megia, E.; Correa,
J.; Riguera, R.; Widmalm, G. Phys. Chem. Chem. Phys. 2010, 12, 6587.
(b) Munoz, E. M.; Correa, J.; Fernandez-Megia, E.; Riguera, R. J. Am.
Chem. Soc. 2009, 131, 17765. (c) Sousa-Herves, A.; Fernandez-Megia, E.;
Riguera, R. Chem. Commun. 2008, 3136.
Org. Lett., Vol. 13, No. 17, 2011
4523

Scheme 1. Synthetic Route for the Multigram Preparation of the GATG Repeating Unit (1) (in blue) Compared to Previously Reported
Conditions (in black; refs 7,8); CC Refers to Column Chromatography
mass wasted in synthesis, which has forced chemists to
consider greener alternatives.15,16 In this respect, H2O, the
only natural solvent on earth, is recognized as a safe,
nontoxic, nonflammable, environmentally benign, and
inexpensive alternative with great potential for large scale
processes(highheatcapacityand convenient boilingpoint-
).5,15,16 Accordingly, we decided to explore the use of H2O
as solvent for the conversion of 2 to 3. Application to H2O
of the best conditions encountered for DMF (Table 1,
entry 4) revealed slower reaction rates,17 which could be
efficiently counterbalanced with slightly longer reaction
times and higher concentrations (Table 1, entries 5À6). No
accelerating effect was observed when using NaI as a
catalyst (entry 7). Conditions in entry 6 proved to be easily
scalable, affording 3 quantitatively in 158 g scale (entry 8).
Upon completion of the reaction, the mixture was cooled
down to rt and H2O was removed under reduced pressure.
The resulting residue was suspended in Et2O and solids
(NaN3 and NaCl) filtered through a sintered funnel. Our
previously described use of CH2Cl2 and Celite for this
filtration8 is discouraged for large scale synthesis to avoid
formation of diazidomethane or other potentially explo-
sive compounds by NaN3 in contact with traces of heavy
metals in Celite (see the SI).18
tions were envisioned as more convenient than those for
entries 1À4 and our previously reported process (TsCl,
pyridineÀCH2Cl2).8 In addition to a nonchromatographic
purification and theavoidanceof toxic solvents, a dramatic
reduction in waste would result in agreement with the prin-
ciple “the best solvent is no solvent”.15,16 Indeed, treat-
ment of 3 with 2 equiv of SOCl2 in the presence of catalytic
amounts of benzyltriethylammonium chloride (BTEAC,
0.3 mol %)19 afforded 4 quantitatively (1 h, 65 °C) after an
aqueous workup to recover the catalyst (Table 2, entry 5).
This transformation proved to be highly practical and scal-
able, allowing the production of 4 in 99% yield in amounts
larger than 170 g (entry 6).
Table 2. Optimized Conditions for the Synthesis of 4
scale
(g)
temp
time
(h)
yield
(%)
entrya
additive
solvent
(°C)
1
2
3
4
5
6
0.5
0.5
0.5
0.5
1
py
CH2Cl2
CHCl3
Toluene
CH2Cl2
À
rt
48
48
48
48
1
49
62
py
reflux
100
rt
py
59
DMAP
BTEAC
BTEAC
64
65
100
99
160
À
65
3
With a method in hand for the multigram preparation of
azidoalcohol 3, we faced the incorporation of the chlo-
ride group in 4. With this aim, thionyl chloride (SOCl2)
was selected as a chlorinating agent based on its easy re-
moval by distillation. In addition, the volatility of HCl and
SO2 produced in the process simplifies purifications. Initial
studies were performed with pyridine/DMAP in different
solvents (CH2Cl2, CHCl3, toluene) and temperatures (rtÀ
100 °C), which unfortunately led to 4 in moderate yields
(Table 2, entries 1À4). We then explored the possibility of
using neat SOCl2 in a solventless process. These condi-
a In all cases 2 equiv of SOCl2 were used.
When a reaction between methyl gallate 6 and chloride 4
was performed under the conditions described for tosylate
5 (K2CO3, DMF, 80 °C; Scheme 1),7,8 the lower reactivity
of 4 was soon revealed, leading to 7 in a moderate 48%
yield (Table 3, entry 1). As a result, a set of experiments
were performed to study the effect of the concentration,
temperature, solvent, amount of base, and catalysts [KI
and 18-crown-6 (18C6)] on the kinetics and yield of the
reaction. The influence of KI and 18C6 was analyzed first.
The use of 0.1 equiv of each one resulted in conversions
around 70% (Table 3, entry 2), which could be increased
up to 93% at concentrations higher than the original 0.1 M
(16) Sheldon, R. A. Green Chem. 2005, 7, 267.
(17) (a) Shi, W.; Dolai, S.; Averick, S.; Fernando, S. S.; Saltos, J. A.;
L’Amoreaux, W.; Banerjee, P.; Raja, K. Bioconjugate Chem. 2009, 20,
1595. (b) Liu, L.-H.; Dietsch, H.; Schurtenberger, P.; Yan, M. Biocon-
jugate Chem. 2009, 20, 1349.
€
(18) (a) Brase, S.; Gil, C.; Knepper, K.; Zimmermann, V. Angew.
Chem., Int. Ed. 2005, 44, 5188. (b) Hassner, A.; Stern, M.; Gottlieb,
H. E.; Frolow, F. J. Org. Chem. 2002, 55, 2304.
(19) Eubanks, R. J. I.; Pacifici, J. G. “Process for the preparation of
ether-containing chlorides”. US Patent 4605784, 1986.
4524
Org. Lett., Vol. 13, No. 17, 2011

(entry 3). The effect of the temperature was then analyzed,
with temperatures lower than 80 °C resulting in much
lower conversions, while higher temperatures resulted in
partial decomposition (entries 4À5). The use of acetone as
solvent or less than 10 equiv of K2CO3 was unsuccessful
(entry 6). When analyzing the separate influence of KI and
18C6, a negligible effect of KI on reactivity was shown
(entry 7). Finally, the scaling up of conditions in entry 7
gave us the opportunity to reduce the amount of added
chloride 4 from 4 equiv to 3.06 (1.02 equiv per hydroxyl
group in 6), with no adverse effect on yield. This represents
a savings of 23% in added 4 and ensures an easier
purification of 7. Workup conditions involved solvent
evaporation followed by filtration through a pad of alu-
mina to remove solid residues. Aqueous workup was left
aside due to the partial solubility of 7 in H2O. Purification
of the crude product by automatic medium pressure liquid
chromatography (MPLC) was revealed to be less time-
consuming than conventional column chromatography,
while affording 7 with higher purity. When these condi-
tions were applied to 44 g of gallate 6, 130 g of pure 7 were
obtained in 87% yield (Table 3, entry 8; purification by
MPLC: 5 portions, 30 min each portion).
aq KOH (2.2 equiv) in boiling methanol. After neutralization
with Amberlite IR-120, acid 1 was obtained quantitatively
(Scheme 1). Similar results were obtained by using EtOH as a
cosolvent, or at rt with a 2-fold concentration of KOH.
Interestingly, these conditions represent a 5 times reduction
in the amount of organic cosolvent compared to previous
conditions.7,8 When acidification was performed with aq
HCl instead of Amberlite IR-120, lower mass recoveries
(around 70%) were typically obtained by extraction.
In conclusion, conditions for the multigram preparation
of the repeating unit of GATG dendrimers (1) have been
developed that afford batches larger than 100 g in excellent
overall yield and purity (4 steps, 86%, one chromato-
graphic step). Special attention has been paid in this
process to observe green chemistry principles: atom econ-
omy, safety, and waste reduction. This scheme could be
easily adapted for the multigram preparation of similar
systems with varying numbers and lengths of spacer arms
and interesting building blocks for the construction of
functional nanostructures.20
Acknowledgment. This work was financially supported
by the Spanish MICINN (CTQ2009-10963 and CTQ-
2009-14146-C02-02) and the Xunta de Galicia (10CSA-
209021PR). M.F.-V. thanks the Spanish Ministry of Edu-
cation for an FPU fellowship. We thank Prof. Steven V.
Ley (University of Cambridge) for helpful discussions.
Table 3. Optimized Conditions for the Synthesis of 7
scale
(g)
K2CO3 concn temp time yield
entrya
additiveb (equiv)
(M)
(°C)
(h)
(%)c
Supporting Information Available. Materials and meth-
ods, WARNING notice on azides, experimental proce-
dures, characterization, and spectra of 1, 3, 4, and 7. This
material is available free of charge via the Internet at
http://pubs.acs.org.
1
2
3
4
5
6
7
8
0.15
À
10
10
10
10
10
6
0.1
0.1
0.5
0.5
0.5
0.5
0.5
0.5
80
80
80
70
60
80
80
80
24
24
24
24
24
24
24
32
48
72
93
58
32
85
93
87
0.15 KI, 18C6
0.15 KI, 18C6
0.15 KI, 18C6
0.15 KI, 18C6
0.15 KI, 18C6
0.15 18C6
€
(20) (a) Muller, M. K.; Brunsveld, L. Angew. Chem., Int. Ed. 2009, 48,
10
10
2921. (b) Abbel, R.; van der Weegen, R.; Meijer, E. W.; Schenning, A. P.
H. J. Chem. Commun. 2009, 1697. (c) Li, W.; Zhang, A.; Schluter, A. D.
Macromolecules 2008, 41, 43. (d) Baranoff, E. D.; Voignier, J.; Yasuda,
T.; Heitz, V.; Sauvage, J.-P.; Kato, T. Angew. Chem., Int. Ed. 2007, 46,
130
18C6
a In all cases 4 equiv of 4 were used, except 3.06 equiv for entry 8. b 0.1
equiv of KI and 0.1 equiv of 18C6 were used. c Yields were determined by
1H NMR (CDCl3), except those for entries 1 and 8.
ꢀ
4680. (e) Oar, M. A.; Serin, J. M.; Dichtel, W. R.; Frechet, J. M. J.;
Ohulchanskyy, T. Y.; Prasad, P. N. Chem. Mater. 2005, 17, 2267.
(f) Chen, C.-T.; Pawar, V. D.; Munot, Y. S.; Chen, C.-C.; Hsu, C.-J.
Chem. Commun. 2005, 2483. (g) Middel, O.; Verboom, W.; Reinhoudt,
D. N. Eur. J. Org. Chem. 2002, 2587. (h) Baars, M. W. P. L.; Kleppinger,
R.; Koch, M. H. J.; Yeu, S.-L.; Meijer, E. W. Angew. Chem., Int. Ed.
2000, 39, 1285.
The final step in the preparation of 1 involves the hydro-
lysis of 7. This was efficiently performed in a 130 g scale with
Org. Lett., Vol. 13, No. 17, 2011
4525