Soluble polymer-supported convergent parallel library synthesis.

Article abstract of DOI:10.1039/b210696e

Soluble polymer-supported convergent synthesis has for the first time been successfully exploited for parallel library synthesis; sub-libraries of tripeptide iodoarenes and arylboronic acids reacted smoothly in a multipolymer PdII-catalyzed Suzuki coupling reaction to generate a library of bisaryl-linked hexapeptides.

Full text of DOI:10.1039/b210696e

Soluble polymer-supported convergent parallel library synthesis†
Jung-Mo Ahn, Paul Wentworth Jr.* and Kim D. Janda*
Department of Chemistry, The Scripps Research Institute and the Skaggs Institute for Chemical Biology,
10550 North Torrey Pines Road, La Jolla, CA 92037, USA. E-mail: paulw@scripps.edu; Fax: +1 858 784
2590
Received (in Cambridge, UK) 30th October 2002, Accepted 7th January 2003
First published as an Advance Article on the web 27th January 2003
Soluble polymer-supported convergent synthesis has for the
first time been successfully exploited for parallel library
synthesis; sub-libraries of tripeptide iodoarenes and ar-
ylboronic acids reacted smoothly in a multipolymer PdII-
catalyzed Suzuki coupling reaction to generate a library of
bisaryl-linked hexapeptides.
linked hexapeptides L4 (81 members) using monomethoxy
poly(ethylene glycol) MeO-PEG as a soluble polymer support
(Scheme 1). The key convergent diversity-expanding step
involves a high-yielding, homogeneous, soluble polymer–
polymer Suzuki cross-coupling reaction between sub-libraries
of PEG-supported tripeptide aryliodides L1a–i (9 members)
and PEG-supported tripeptide arylboronic acids L2a–i to give a
PEG-supported bisaryl-linked hexapeptide library L3 (81
members).‡
The choice of MeO-(PEG)-OH of 5000 average molecular
weight as the polymer support was guided by our extensive
experience with this material that we have utilized previously
both as a support for combinatorial6 and organic synthesis,7 and
as a reagent8 and catalyst support.9
A bisaryl-linked hexapeptide library was selected to illustrate
this first polymer-supported convergent strategy because of the
well-known benefits of a polymer-supported versus solution-
phase synthesis of peptide oligomers. From the outset, we
defined constraints on the reaction chosen for the key
convergent step, specifically: a) it must be a high-yielding
reaction: b) it must be amenable to modification for yield
optimization: c) there should be few side-reactions. For these
reasons the homogeneous PdII-catalyzed variant of the Suzuki
cross-coupling reaction between iodoarenes and aryl boronic
acids was selected.10 We assumed that the tripeptide component
of the library would have little effect on the cross-coupling
reaction, therefore optimization of the multipolymer Suzuki
reaction was investigated with the PEG-supported para-
iodobenzoate 5a, and PEG-supported boronic acid derivative 5b
(Scheme 2).
Formation of the bisaryl-linked PEG-dimer 6 was routinely
determined by 1H NMR using a comparison of the ratios of the
integrated peak areas of biphenyl protons (d 7.86) and the
protons attached to the methylene carbon alpha to the terminal
oxygen of the PEG-polymer support (d 4.59). The product yield
was further confirmed by cleavage of the 1,4-bisaryl ester from
the MeO-PEG support by transesterification (KCN–MeOH) to
the bis-methyl ester 7.
Polymer-supported and polymer-assisted synthetic method-
ologies are now ubiquitous throughout the fields of combinato-
rial chemistry, high-throughput- and pure organic-synthesis.1
Solid-phase chemistry, that utilizes cross-linked polymer sup-
ports, is by far the most widely accepted approach within these
new synthetic spheres.2 However, limitations imposed by the
heterogeneous nature of the chemical synthesis, have meant that
alternative soluble polymer supports are being increasingly
exploited.3,4
An inherent property of all solid-phase-based synthetic
methods is a stepwise, linear synthesis that by necessity leads to
a divergent introduction of diversity. This limitation not only
applies to the linear synthesis of oligomers, such as peptides or
oligonucleotides, but also to the sequential derivatization of
templates. Linear divergent synthesis can be performed both on
the solid-phase and in solution, however, the more powerful
convergent approach to synthesis of diversity can only be
conducted in solution because with solid-phase techniques the
combining components would be on mutually exclusive resin
beads. Boger5 and co-workers were the first to bring into sharp
focus the clear benefits of convergent combinatorial synthesis
and they applied it using solution-phase chemical methods for
the generation of diaminodiacetic acid diamide and bisaryl
libraries. However, to date, the question as to whether
convergent synthesis could be performed effectively on a
soluble polymer support had not been addressed (Fig. 1).
Herein, we have merged the power of convergent synthesis
with the synthetic benefits of polymer-supported chemistry
(vide supra). We have generated a parallel array of bisaryl-
Fig. 1 Generalized scheme of polymer-supported linear versus convergent
library synthesis. (A) Linear synthesis (B) convergent synthesis.
† Electronic supplementary information (ESI) available: detailed synthetic
procedures for synthesis of 5b–7, PEG-supported synthesis of L1a--i and
L2a–i, purification of L4, 1H-NMR spectra and LC/MS and HR-
MALDIMS data of selected hexapeptide from L4. See http://www.rsc.org/
suppdata/cc/b2/b210696e/
Scheme 1 Polymer-supported convergent approach to the bisaryl-linked
hexapeptide library L4 (81 member). G = glycine; X1, X2, X1A and X2A =
Phe, Ala or Leu.
480
CHEM. COMMUN., 2003, 480–481
This journal is © The Royal Society of Chemistry 2003

1
coupling reaction condition vide supra. H NMR of selected
members of L3 confirmed that the reaction proceeded in high
yield across the whole library. The L3 hexapeptides were
cleaved from the PEG support with aqueous LiOH. The
resulting bis-acids L4 were separated from the MeO-PEG by
non-covalent attachment onto a weakly basic anion-exchange
resin (IRA-67). The library of peptides L4 was then released
from the resin (1 M HCl in acetonitrile) and further purified by
reversed-phase HPLC. The purities of all crude members of L4
were determined by HPLC/MS and found to be 50–95%, with
yields ranging from 72–95%. The structure of certain L4
1
members was confirmed by H-NMR spectroscopy and HR-
MALDI MS.† The reduced purity and yield of certain library
members of L4, constructed from L1a–i and L2a–i where X1X2
≠ X1’X2’, was a result of their containing a byproduct in small
amounts (7–24%). This byproduct is the bisaryl-linked hex-
apeptide arising from a competing homocoupling of L1a–i, that
gives a X1X2-X2X1–containing hexapeptide in addition to the
required X1X2-X2’X1’-containing hexapeptide from the hetero-
dimeric coupling. This byproduct was routinely separated by
HPLC.§
In summary, we have demonstrated the first successful
merging of polymer-supported chemistry with convergent
parallel synthesis and illustrated it by producing an 81 member
bisaryl-linked hexapeptide library L4. This methodology high-
lights the unique applicability of soluble polymers as supports
in parallel library synthesis and should have universal applica-
bility within the spheres of combinatorial chemistry and high-
throughput synthesis.
Scheme 2 Multipolymer approach to biaryl 7. a) Conditions optimized Pd
catalyst, base, solvent and temperature; b) KCN–MeOH.
Solvent effects did not appear to play a significant role in the
polymer-polymer reaction as DMF, toluene, 1,4-dioxane and
THF all gave similar yields (Table 1, entry 1–4). However,
substantial improvements in the yields of 6 were observed when
inorganic bases were used instead of Et3N. Several inorganic
bases, including K2CO3, Cs2CO3, KF, and K3PO4 were
screened, and with the exception of KF (Table 1, entry 5),
provided yields of 6 between 80 and 90% with few by-products
(Table 1, entry 6–10). In addition, the reactions were faster with
inorganic bases, being complete within 24 h using K2CO3,
whereas up to 72 h was required with Et3N at the same
temperature (100 °C).
Three catalytic palladium sources were investigated,
PdCl2(dppf), Pd(PPh3)4, and Pd(OAc)2. With K2CO3 as the
base, no significant differences in yields were observed with the
three palladium catalysts (Table 1, entry 8–10), therefore
PdCl2(dppf) was chosen due to its low air-sensitivity. The
optimized reaction conditions ultimately comprised of MeO-
PEG5000-bound iodide 5a and boronic acid 5b (10 mM) with
PdCl2(dppf) (10 mol%) and K2CO3 (3 equiv.) in DMF at 100 °C
for 24 h. These reaction conditions typically gave 95% yield of
the PEG-supported biphenyl-containing product 6, with negli-
gible formation of byproducts.
This work was supported by the NIH (GM-56154) and the
Skaggs Institute for Chemical Biology.
Notes and references
‡ Experimental: MeO-PEG5000-supported tripeptide iodide library L1a–i
(20 mg, 4 mmol; 10 mM), MeO-PEG5000-supported tripeptide boronic acid
L2a–i (20 mg, 4 mmol; 10 mM), K2CO3 (1.6 mg, 12 mmol), PdCl2(dppf)
(0.33 mg, 10 mol%) and degassed DMF (0.4 mL) were added to thick-
walled glass vials. The vials were sealed under Ar and the reaction mixture
was stirred at 100 °C for 24 h. The reaction mixture was then cooled to room
temperature and added into cold ether (10 mL). The precipitated PEG-
bound hexapeptides L3 were separated by centrifugation, washed with cold
ether (10 mL) and dried under reduced pressure.
§ While the homodimer was separated by LC to increase the purity of each
member of the L4 array, it should be noted that the byproduct is also a
member of L4. Therefore, given that we know its constitution and the
amount present there is no need for it to be removed.
MeO-PEG-bound tripeptides were prepared by standard Na-
Boc peptide chemistry. Each coupling reaction was monitored
by the Kaiser ninhydrin test.11 Each MeO-PEG-supported
peptide coupling proceeded to > 98% based on routine 1H NMR
analysis. The first amino acid coupled to MeO-PEG was fixed
as Gly to minimize potential epimerization during the coupling
reaction in the presence of DMAP. The second and third
residues were randomized, either as Ala (A), Phe (F) or Leu (L).
After tri-peptide synthesis, the PEG-supported tripeptides were
split into two portions and the Na-termini of the tripeptides were
coupled with either 4-iodobenzoic acid or 4-carboxyphe-
nylboronic acid, giving the PEG-supported iodide sub-library
L1a–i (9 members) and PEG-supported boronic acid sub-
library L2a–i (9 members).
1 R. J. Booth and J. C. Hodges, Acc. Chem. Res., 1999, 32, 18; S. J.
Shuttleworth, S. M. Allin and P. K. Sharma, Synthesis, 1997, 1217.
2 J. S. Fruchtel and G. Jung, Angew. Chem., Int. Ed. Engl., 1996, 35, 17;
L. A. Thompson and J. A. Ellman, Chem. Rev., 1996, 96, 555.
3 B. Yan, Acc. Chem. Res., 1998, 31, 621; B. Yan, J. Fell and G.
Kumaravel, J. Org. Chem., 1996, 61, 7467.
4 P. H. Toy and K. D. Janda, Acc. Chem. Res., 2000, 33, 546; P.
Wentworth, and K. D. Janda, Chem. Commun., 1999, 1917; D. J.
Gravert and K. D. Janda, Chem. Rev., 1997, 97, 489.
5 D. L. Boger and W. Chai, Tetrahedron, 1998, 54, 3995; D. L. Boger, W.
Chai and Q. Jin, J. Am. Chem. Soc., 1998, 120, 7220; D. L. Boger, J.
Goldberg and C.-M. Andersson, J. Org. Chem., 1999, 64, 2422.
6 H. Han and K. D. Janda, J. Am. Chem. Soc., 1996, 118, 2539.
7 J. A. Lopéz-Pelegrín, P. Wentworth, F. Sieber, W. A. Metz and K. D.
Janda, J. Org. Chem., 2000, 65, 8527.
The hexapeptide-library L3 (81 members) was then prepared,
in parallel, using the optimized polymer–polymer Suzuki cross-
Table 1 Optimization of the polymer–polymer Suzuki cross-coupling
reaction between 5a and 5b
Entry
Catalyst
Base
Solvent
Yield (%)
8 A. D. Wentworth, P. Wentworth, U. F. Mansoor and K. D. Janda, Org.
Lett., 2000, 2; P. Wentworth, A. M. Vandersteen and K. D. Janda, Chem.
Commun., 1997, 759; F. Sieber, P. WentworthJr, J. D. Toker, A. D.
Wentworth, W. A. Metz, N. N. Reed and K. D. Janda, J. Org. Chem.,
1999, 64, 5188.
9 H. Han and K. D. Janda, Angew. Chem., Int. Ed. Engl., 1997, 36, 1731;
H. Han and K. D. Janda, J. Am. Chem. Soc., 1996, 118, 7632; H. Han and
K. D. Janda, Tetrahedron Lett., 1997, 38, 1527.
1
2
3
4
5
6
7
8
9
PdCl2(dppf)
PdCl2(dppf)
PdCl2(dppf)
PdCl2(dppf)
PdCl2(dppf)
PdCl2(dppf)
PdCl2(dppf)
PdCl2(dppf)
Pd(PPh3)4
Et3N
Et3N
Et3N
Et3N
THF
60
50
40
60
55
85
80
90
85
90
Toluene
1,4-Dioxane
DMF
DMF
DMF
DMF
DMF
DMF
DMF
KF
Cs2CO3
K3PO4
K2CO3
K2CO3
K2CO3
10 A. Suzuki, Pure Appl. Chem., 1985, 57, 1749.
11 E. Kaiser, R. L. Colescott, C. D. Bossinger and P. J. Cook, Anal.
Biochem., 1970, 34, 595.
10
Pd(OAc)2
CHEM. COMMUN., 2003, 480–481
481