Tetraarylphosphonium salts as soluble supports for the synthesis of small molecules

Article abstract of DOI:10.1002/anie.200700833

(Chemical Equation Presented) Support groups: Phosphonium salts can be used as solubility-control groups (SCGs) in small-molecule and peptide synthesis (see scheme). The concept, which is a hybrid approach between solution- and solid-phase syntheses, is i

Full text of DOI:10.1002/anie.200700833

Angewandte
Chemie
DOI: 10.1002/anie.200700833
Synthetic Methods
Tetraarylphosphonium Salts as Soluble Supports for the Synthesis of
Small Molecules**
Federica Stazi, David Marcoux, Jean-Christophe Poupon, Daniel Latassa, and
AndrØ B. Charette*
Several methods are currently available for the synthesis of
small-molecule and peptide libraries on solid supports.[1]
Since the pioneering work of Merrifield,[2] many researchers
have exploited the use of solid supports for small-library
synthesis.[3] Significant improvements have been made in this
area, especially in the design of new supports and linkers to
increase the scope of this technology.[4,5] Historically, a major
challenge associated with solid-phase synthesis has been to
merge homogeneous reaction conditions with the practical
operational advantages of solid-phase chemistry. Some con-
tributions towards this end include soluble polymer supports
(for example, linear polyethylene glycol (PEG)),[6,7] fluorous-
phase systems,[8] ionic-liquid supports,[9] and other tags.[10]
We have recently become interested in using tetraaryl-
phosphonium salts as new solubility-control groups (SCGs)
for reagents, ligands, and catalysts.[11] The main advantage of
these groups is their lower molecular weight ( ꢀ 450) when
compared to polymeric materials (ꢁ 3000). This results in
loadings that can be at least ten times higher than those with
other traditional supports. The solubility of these salts is
predictable with several types of reagents that are attached to
them. We sought to exploit the solubility properties of the
salts for the development of a complementary liquid-phase
small-molecule synthesis that is cost-effective and versatile.
We have previously shown that SCGs are soluble in polar
solvent systems (CH2Cl2, CH3CN, DMSO, DMF) and their
precipitation is induced upon adding less polar solvents
(hexane, Et2O, toluene). Herein, we demonstrate that tet-
raarylphosphonium salts can be used as new soluble supports
for small-molecule synthesis. The proof of concept will be
illustrated by the synthesis of 2-substituted piperidine deriv-
atives, which involves Grignard addition to a supported chiral
pyridinium salt. Finally, the limits of the support will be
established through peptide synthesis.
substituted dihydropyridine derivatives. As this chemistry
involves the manipulation of relatively sensitive dihydropyr-
idines, we envisioned that it would constitute a good test case
for demonstrating the proof of concept of the phosphonium-
supported chemistry. The supported chiral auxiliary[12] was
prepared from 4-bromobiphenyl-4’-carboxaldehyde and tri-
phenylphosphine (Scheme 1). Anion exchange, Jones oxida-
Scheme 1. Synthesis of supported 2-substituted dihydropyridines. DIC:
1,3-diisopropylcarbodiimide; DIPEA: N,N-diisopropylethylamine;
HOBT: 1-hydroxybenzotriazole; DCM: dichloromethane; Tf: trifluoro-
methanesulfonyl.
We have recently reported that the addition of Grignard
reagents to a chiral pyridinium imidate salt led to 2-
[*] F. Stazi, D. Marcoux, J.-C. Poupon, D. Latassa, Prof. A. B. Charette
Department of Chemistry
UniversitØ de Montreal
tion, amide formation, and protection led to the desired
supported chiral auxiliary in 69% overall yield and without
any purification by chromatography. We next examined the
amide electrophilic activation of 3 with triflic anhydride in the
presence of pyridine under the previously reported condi-
tions.[12] We were pleased to observe the complete formation
of N-imidate pyridinium species within 3 h at room temper-
ature. Subsequent nucleophilic addition of Grignard reagents
led to the formation of the corresponding 1,2-dihydropyridine
with high regio- and diastereocontrol. The products, which
PO Box 6128, Station Downtown
Montreal, QC H3C 3J7 (Canada)
Fax: (+1)514-343-5900
E-mail: andre.charette@umontreal.ca
[**] This work was supported by NSERC (Canada), VRQ, Univalor, the
Canada Research Chair Program, and the UniversitØ de MontrØal.
D.M. is gratefulto the UniversitØ de MontrØaland NSERC (ES D) for
a postdoctoral fellowship.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2007, 46, 5011 –5014
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5011

Communications
were isolated by precipitation/filtration upon adding diethyl
The main advantage of using these new supports in a
reaction sequence is the ability to follow each step by
traditional analytical techniques, in particular HPLC, mass
spectrometry (MS), TLC, and NMR spectroscopy, and isolate
the reaction product by a precipitation/filtration procedure.
In this case, completion of the coupling event could be
monitored and each intermediate of the growing peptide
chain could be characterized. Furthermore, because each
reaction was carried out under homogeneous conditions, the
use of the large excess of reagents needed in solid-phase
chemistry was not necessary. Usually a near-stoichiometric
amount of the amino acid was used (1.2 equiv).
ether, were obtained in higher yields than the corresponding
unsupported examples that were reported previously[12] (96
vs. 77% when R = Me and 92 vs. 61% when R = (Z)-
=
CH3CH CH). This improvement in yields can be accounted
for by the difficulty of isolating relatively unstable, polar,
unsupported dihydropyridine derivatives such as 4.
The cleavage of the substituted piperidine derivatives
could be accomplished under the previously described con-
ditions for the synthesis of unsupported product (Scheme 2).
For example, N-Boc (À)-coniine (6) was readily accessed
As the final cleavage was achieved under acidic con-
ditions, 9-fluorenylmethoxycarbonyl/tert-butyl (Fmoc/tBu)
chemistry was successfully applied. Anchoring of the first
Fmoc-protected amino acid (1.2 equiv) was performed in
quantitative yield with a 4-(N,N-dimethylamino)pyridine
(DMAP)/DIC[15] coupling in CH2Cl2. At the end of the
reaction, Celite (1:5 w/w) was added and the product was
precipitated upon addition of Et2O. The product (absorbed on
Celite) was then isolated by filtration, and washed with 1m
HCl, water, toluene, and Et2O. This workup gave quantitative
recovery of the product in high purity (see Figure 1). The
Scheme 2. Cleavage of the support. Boc: tert-butoxycarbonyl.
from 4b by hydrogenation using Adamꢀs catalyst followed by
cleavage of the amidine with BBr3 and final Boc protection of
the free amine (Scheme 2). The by-product oxazoline 7 could
be easily removed by precipitation upon addition of diethyl
ether and filtration. Flash chromatography of the crude
filtrate gave 6 in 54% yield from amide 3. The synthesis on
the phosphonium support is more effective than the previ-
ously reported one (47 vs. 30% overall yield).
To test the efficiency of the phosphonium group in
controlling the solubility of an attached substrate as a function
of its molecular weight, a series of small peptides was
synthesized. Two new SCG supports, based on the Wang
(9a, SCG-W)[13] and Sasrin (9b, SCG-S)[14] resin structures,
were synthesized starting from aldehyde 2 (Scheme 3).
Figure 1. Typicalreaction and workup for peptide synthesis. a) Cou-
pling reaction (Celite added); b) precipitation of phosphonium-sup-
ported peptide on Celite upon ether addition; c) isolation of phospho-
nium-supported peptide (on Celite) after filtration and washing. The
Celite-absorbed product is used directly in the next coupling step.
following steps were iterative: deprotection of the Fmoc
group with piperidine (50% in CH2Cl2), then peptide
coupling with an N-Fmoc-protected amino acid, mediated
by HOBT/DIC/DIPEA.[16] Application of the standard
workup afforded quantitative recovery of the product with
acceptable purity.[17] It should be pointed out that the product
on Celite could be carried throughout the entire sequence
without purification.
The cleavage of the peptide from both SCG supports was
considered separately. For the SCG-W, two reagents were
used: trifluoroacetic acid (TFA; either 10% in CH2Cl2 or
100%) and HCl (3m in CH2Cl2). Under these conditions, N-
Fmoc peptides were synthesized in good yields and purities
(Table 1, entries 1–10). However, only the use of HCl allowed
the isolation and recovery of a clean tetraarylphosphonium
species that could be converted back to the starting SCG-W in
> 95% yield (Scheme 4). In the case of the SCG-S support,
Scheme 3. Synthesis of Wang (9a) and Sasrin phosphonium salts
(9b). NBS: N-bromosuccinimide.
5012
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 5011 –5014

Angewandte
Chemie
Table 1: Synthesis of C-terminalpeptides on phosphonium supports.
Although the synthesis of
longer peptides was attempted, the
solubility of the resulting species
required the use of more polar
solvents. For example, the syntheses
of the phosphonium-supported
hepta(alanine) and nona(alanine)
could be accomplished, but DMSO
had to be used to dissolve the
phosphonium-supported peptide.
Moreover, the precipitation/isola-
tion sequence was not as straight-
forward, thus making this technique
Entry
Product
Cleavage condition[a]
Yield [%][b]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Ala-Phe-Val-NHFmoc
Ala-Ala-Ala-NHFmoc
10% TFA
10% TFA
10% TFA
100% TFA[d]
HCl (3m)
HCl (3m)
HCl (3m)[g]
HCl (3m)
HCl (3m)
HCl (3m)
1% TFA
80[c]
75
Ala-(Ala)3-Ala-NHFmoc
Ala-Phe-Ser-NHFmoc
Ala-Phe-Val-NHFmoc
Phe-Ala-Ala-NHFmoc
Ala-Phe-Ser-NHFmoc
Ala-Ala-Ala-NHFmoc
Ala-Ala-Ala-Val-NHFmoc
(Ala)3-Val-Gly-NHFmoc
Ala-Phe-Ser(Ot-Bu)-NHFmoc
Ala-Ala-Ala-NHBoc
50
80[e]
80[c] (70)[f]
67[e] (60)[f]
70[e] (70)[f]
95 (95)[f]
80 (73)[f]
74[h] (70)[f]
81[e]
1% TFA
60[i]
Al(a4-Ala-Ala-NH2·HClHCl
Ala-Ala-Ala-NH2·TFA
m)[j]
98[k] (80)[f] less practical for peptides contain-
100% TFA[d]
100% TFA[m]
96[k]
ing seven amino acids or more.
Locustakinin (six amino acids)[l]
80[k]
In summary, we have developed
[a] CH2Cl2 as solvent. [b] Purity >95% by HPLC. [c] After trituration with CH3CN. [d] Anisole (5%) was a new practical synthesis that uses
added as a scavenger. [e] After chromatography on silica gel. [f] The yield in parentheses refers to that of
the recovered phosphonium support. [g] Conditions for the cleavage: 408C for 36 h. [h] After
chromatography on C18. [i] Purity >95% by NMR spectroscopy. [j] Dioxane was used as solvent.
[k] Phosphonium 9b was used as soluble support. [l] Ala-Phe-Ser-Ser-Thr-Gly-NH3·TFA. [m] Anisole
tetraarylphosphonium salts as solu-
bility-control groups for small-mol-
ecule synthesis. This technique is a
hybrid approach between solution-
and solid-phase syntheses, which
exploits the main advantages of
(4%) and 1,2-diethanethiol(1%) were added as scavengers.
both methodologies without the
requirement of using special solvents. Advantages compared
to known systems include higher loadings, ease of character-
ization, reagent compatibility of the support, and predictable
solubility properties.
Experimental Section
Scheme 4. Recovery of the soluble support. AA: amino acid.
General procedure for coupling: Support 9a or 9b (0.05–0.1 mmol,
1.0 equiv) was dissolved in CH2Cl2 (0.5–1 mL). N-Fmoc-protected
amino acid (0.06–0.12 mmol, 1.2 equiv), DMAP (1.3–2.5 mg, 0.01–
0.02 mmol, 0.2 equiv), and DIC (10–20 mL, 0.06–0.12 mmol,
1.2 equiv) were added. The solution was stirred at room temperature
very mild cleavage conditions (1% TFA in CH2Cl2) afforded
for 2 h. The reaction was monitored by electrospray MS and HPLC
fully protected N-Fmoc and N-Boc peptides in good yields
(Table 1, entries 11 and 12). Harsher conditions with N-Boc
peptides using 9a (4m HCl in dioxane or 100% TFA)
afforded peptide salts in quantitative yields of isolated
products (Table 1, entries 13–15). The peptides were charac-
terized by reversed-phase HPLC.[18]
analysis. Celite (250–500 mg) was added and the product was
precipitated by the addition of Et2O (2.5–5 mL). The solid was
isolated by filtration, and washed with Et2O (5 ” 10 mL), H2O (2 ”
10 mL), 1m HCl (2 ” 10 mL; only in the case of 9a), toluene (2 ”
10 mL), and Et2O (5 ” 10 mL). The crude product absorbed on Celite
was used directly in the next coupling step. The crude product could
be extracted from Celite with MeOH/CH2Cl2 (1:9).
Low-polarity peptides (retention time > 12 min on
reversed-phase silica) were readily isolated by selective
precipitation of the cleaved support (Table 1, entries 1, 4–7,
and 11). The peptides were quantitatively recovered upon
concentration of the filtrate by HPLC with an average purity
of 90%. Further purification by silica-gel flash chromatog-
raphy or trituration with CH3CN gave a pure final product.
Polar peptides (retention time < 12 min) were purified by
reversed-phase column chromatography after support cleav-
age (Table 1, entries 2 and 3). Finally, the amino acid salts
were isolated simply by aqueous extraction and lyophilization
(Table 1, entries 14 and 15). In many cases, the peptide
precipitated upon cleaving the support, and thus purification
by filtration was achieved (Table 1, entries 8–10 and 13). The
scope of the methodology was demonstrated by the synthesis
in high yield of the biologically active hexapeptide locusta-
kinin (Table 1, entry 15).[19]
Received: February 23, 2007
Published online: May 22, 2007
Keywords: asymmetric synthesis · chiralauxiilaries ·
.
combinatorial chemistry · peptides · soluble supports
[1] a) T. Kimmerlin, D. Seebach, J. Pept. Res. 2005, 65, 229; b) D.-S.
Shin, D.-H. Kim, W.-J. Chung, Y.-S. Lee, J. Biochem. Mol. Biol.
2005, 38, 517; c) F. Albericio, Curr. Opin. Chem. Biol. 2004, 8,
211.
[2] a) R. B. Merrifield, J. Am. Chem. Soc. 1963, 85, 2149; b) R. L.
Letsinger, M. J. Kornet, J. Am. Chem. Soc. 1963, 85, 3045.
[3] a) R. L. Letsinger, V. Mahadevan, J. Am. Chem. Soc. 1966, 88,
5319; b) J. M. FrØchet, C. Schuerch, J. Am. Chem. Soc. 1971, 93,
492; c) C. C. Leznoff, T. M. Fyles, J. Chem. Soc. Chem. Commun.
1976, 251.
Angew. Chem. Int. Ed. 2007, 46, 5011 –5014
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5013

Communications
[4] For recent reviews, see: a) A. R. Vaino, K. D. Janda, J. Comb.
Rodríguez, D. C. Horwell, R. A. Lewthwaite, M. C. Pritchard,
A. M. Reid, Angew. Chem. 2001, 113, 1088; Angew. Chem. Int.
Ed. 2001, 40, 1053.
Chem. 2000, 2, 579; b) Z. Yu, M. Bradley, Curr. Opin. Chem.
Biol. 2002, 6, 347; c) F. Guillier, D. Orain, M. Bradley, Chem.
Rev. 2000, 100, 2091.
[11] J.-C. Poupon, A. A. Boezio, A. B. Charette, Angew. Chem. 2006,
118, 1443; Angew. Chem. Int. Ed. 2006, 45, 1415.
[12] A. B. Charette, M. Grenon, A. Lemire, M. Pourashraf, J. Martel,
J. Am. Chem. Soc. 2001, 123, 11829.
[5] For a recent approach, see: R. A. Houghten, Y. Yu, J. Am. Chem.
Soc. 2005, 127, 8582.
[6] For pioneering work using a PEG support, see: a) M. Mutter, H.
Hagenmaier, E. Bayer, Angew. Chem. 1971, 83, 883; Angew.
Chem. Int. Ed. Engl. 1971, 10, 811; b) E. Bayer, M. Mutter,
Nature 1972, 237, 512.
[7] a) P. M. Fischer, D. I. Zheleva, J. Pept. Sci. 2002, 8, 529; b) D. J.
Gravert, K. D. Janda, Chem. Rev. 1997, 97, 489; c) P. H. Toy,
K. D. Janda, Acc. Chem. Res. 2000, 33, 546.
[8] a) M. Mizuno, K. Goto, T. Miura, D. Hosaka, T. Inazu, Chem.
Commun. 2003, 972; b) D. P. Curran, Angew. Chem. 1998, 110,
1230; Angew. Chem. Int. Ed. 1998, 37, 1174; c) W. Zhang, Chem.
Rev. 2004, 104, 2531.
[9] a) W. Miao, T.-H. Chan, J. Org. Chem. 2005, 70, 3251; b) P.
Wasserscheid, T. Welton, Ionic Liquids in Synthesis, VCH,
Weinheim, 2003; c) P. Wasserscheid, W. Keim, Angew. Chem.
2000, 112, 3926; Angew. Chem. Int. Ed. 2000, 39, 3772.
[10] a) J. Yoshida, K. Itami, Chem. Rev. 2002, 102, 3693; b) D. L.
Flynn, J. Z. Crich, R. V. Devraj, S. L. Hockerman, J. J. Parlow,
M. S. South, S. Woodard, J. Am. Chem. Soc. 1997, 119, 4874; c) T.
Bosanac, J. Yang, C. S. Wilcox, Angew. Chem. 2001, 113, 1927;
Angew. Chem. Int. Ed. 2001, 40, 1875; d) S. V. Ley, A. Massi, F.
[13] S. S. Wang, J. Am. Chem. Soc. 1973, 95, 1328.
[14] a) M. Mergler, R. Tanner, J. Gosteli, P. Grogg, Tetrahedron Lett.
1988, 29, 4005; b) M. Mergler, R. Nyfeler, R. Tanner, J. Gosteli,
P. Grogg, Tetrahedron Lett. 1988, 29, 4009.
[15] a) N. Sewald, H.-D. Jakubke, Peptides: Chemistry and Biology,
VCH, Weinheim, 2002; b) C. A. G. N. Montalbetti, V. Falque,
Tetrahedron 2005, 61, 10827.
[16] To avoid the formation of diketopiperazinine, a dipeptide was
directly coupled to the first phosphonium-supported amino acid
in the second coupling step: E. Pedroso, A. Grandas, X.
de las Heras, R. Eritja, E. Giralt, Tetrahedron Lett. 1986, 27, 743.
[17] Formation of the amide bond occurred quantitatively. Although
the presence of DIPEA and DIC was occasionally observed, no
residual Fmoc amino acid could be detected (liquid chromatog-
raphy–MS).
[18] See the Supporting Information for details.
[19] a) L. Schoofs, G. M. Holman, P. Proost, J. Van Damme, T. K.
Hayes, A. De Loof, Regul. Pept. 1992, 37, 49; b) L. Schoofs, D.
Veelaert, J. Vanden Broeck, A. De Loof, Peptides 1997, 18, 145.
5014
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 5011 –5014