Wang linker free of side reactions

Article abstract of DOI:10.1021/ol303367s

A new resin for the solid-phase synthesis of peptide acids was developed. It was based on a linker with two unique features: methoxy groups as the only activating groups of the phenyl ring and a copper(I)-catalyzed Click chemistry reaction to anchor it to

Full text of DOI:10.1021/ol303367s

ORGANIC
LETTERS
2013
Vol. 15, No. 2
246–249
Wang Linker Free of Side Reactions
Vida Castro,†,‡ Hortensia Rodriguez,*,†,‡ and Fernando Albericio*,†,‡,§,
Institute for Research in Biomedicine, 08028-Barcelona, Spain, CIBER-BBN,
Networking Centre on Bioengineering, Biomaterials and Nanomedicine, PCB,
08028-Barcelona, Spain, Department of Organic Chemistry, University of Barcelona,
08028-Barcelona, Spain, and School of Chemistry, University of KwaZulu-Natal,
4001 Durban, South Africa
hortensia.rodriguez@irbbarcelona.org; albericio@irbbarcelona.org
Received October 16, 2012
ABSTRACT
A new resin for the solid-phase synthesis of peptide acids was developed. It was based on a linker with two unique features: methoxy groups as
the only activating groups of the phenyl ring and a copper(I)-catalyzed Click chemistry reaction to anchor it to the solid support. The efficiency of
this new resin in solid phase peptide synthesis was compared with that of Wang resin.
Linkers for the solid phase synthesis of peptides and
other organic molecules are based mostly on the presence
of alkoxy phenyl moieties, the latter commonly bound to
an amine-based solid support through an amide bond. At
the end of the synthetic process, these resins are treated
with TFA to release the target molecules, thereby render-
ing carboxylic acids or amides. However, this acid treat-
ment often leads to side reactions. Thus, additional
cleavages through the amide bond used to anchor the
linker to the solid support and/or through the phenoxy
moiety present in the linker cause the presence of side
products accompanying the target.1 These can be difficult
to remove from the crude product, or worse still, they may
back-alkylate the cleaved peptide atsensitive residues, such
as Trp or Tyr. In addition to jeopardizing the purification,
this back-alkylation decreases the final yield.2 To circum-
vent this problem, here we developed some nonacid de-
gradable linkers. These linkers are characterized by
activation of the benzyl alcohol by a noncleavable electron-
donating group in either the ortho or para position,3 and/or
by two phenyl rings attached by a Suzuki reaction.4 Here
we introduce a new concept of linker. In this case the linker
is anchored to the solid support through a stable triazole
moiety and with methoxy groups as the only activating
groups of the phenyl ring (Figure 1). The efficiency of the
new resin in solid phase synthesis (SPS) is compared with
that of Wang resin.
† Institute for Research in Biomedicine.
‡ CIBER-BBN.
§ University of Barcelona.
University of KwaZulu-Natal.
(1) (a) Giraud, M.; Cavalier, F.; Martinez, J. J. Pept. Sci. 1999, 5,
457–461. (b) Yraola, F.; Ventura, R.; Vendrell, M.; Colombo, A.;
Fernandez, J.-C.; de la Figuera, N.; Fernandez-Forner, D.; Royo, M.;
Forns, P.; Albericio, F. QSAR Comb. Sci. 2004, 23, 145–152. (c) Cironi,
P.; Tulla-Puche, J.; Barany, F.; Albericio, F.; Alvarez, M. Org. Lett.
2004, 6, 1405–1408. (d) Stathopoulos, P.; Papas, S.; Tsikaris, V. J. Pept.
Sci. 2006, 16, 227–232. (e) Stanger, K. J.; Krchnack, V. J. Comput. Chem.
2006, 8, 652–654.
(2) The þ107 peak corresponding to the Wang linker is a purification
nightmare for most peptides produced using this support.
(3) Gu, W.; Silverman, R. B. Org. Lett. 2003, 5, 415.
ꢀ
ꢁ
(4) Colombo, A.; De la Figuera, N.; Fernandez, J. C.; Fernandez-
Forner, D.; Albericio, F.; Forns, P. Org. Lett. 2007, 9, 4319.
r
10.1021/ol303367s
Published on Web 12/27/2012
2012 American Chemical Society

Scheme 2. Chloro Merrifield Resin Conversiona
a NaN3 (3 equiv), DMF, 24 h, rt.
Figure 1. Linker concept.
Intermediate 4 was loaded onto the Merrifield azide
resin 5 (Scheme 3) via CuAAC13 to render CHO-BTL
resin. This reaction was carried out in ACN, ascorbate of
sodium, 2,6-lutidine, DIEA, and DMF as solvent for 6 h at
room temperature. If only catalytic amounts of CuBr are
used, reaction times should increase. The progress of
CuAAC was monitored using IR spectroscopy, through
the loss of the azide band and the appearance of new
vibrational modes at 1673 cmꢀ1 and 1546 cmꢀ1, indicative
of the carbonyl group and the alkene of the triazol moiety,
respectively.
The copper-catalyzed azideꢀalkyne cycloaddition
(CuAAC) has been extensively used in SPS to obtain cyclic
peptides,5 triazolyl aminoacyl (peptidyl) penicillins,6 and
aldehyde-functionalized resins.7 For efficient and stable
coupling of the linker to a solid support, this reaction
requires an alkyne- and an azide-functionalized resin asthe
solid support.8
First, 5-ethynyl-2,4-dimethoxybenzaldehyde (4) was
synthesized from commercially available 2,4-dimethoxyben-
zaldehyde (1) (Scheme 1), which was iodated with iodine
monochloride at C5 of 1.9 The alkyne moiety was intro-
duced by a Sonogashira coupling reaction,10 exchanging the
iodine atom by TMSA, to obtain 3. Subsequent deprotec-
tion gave the intermediate 4 in an overall yield of 28%.
To generate the Merrifield azide resin 5, ClꢀMerrifield
resin (0.7 mmol Cl/g) was reacted with sodium azide in DMF
(Scheme 2) following a previous reported method with slight
modifications (conversion was carried at room temperature
rather than at 60 °C).11 The reduction/ninhydrin test was
used to monitor the incorporation of azide function on the
resin.12 Treatment with the standard PPh3, ninhydrin, pyr-
idine, and phenol solutions followed by warming gave a
positive result (solution turned dark purple). This step was
also monitored using IR spectroscopy to corroborate the
The OH-BTL resin was finally obtained by reducing
CHO-BTL resin using NaBH4 in THF/MeOH.
The OH-BTL resin was characterized by IR spectro-
metry, and its stability was evaluated under standard acid
conditions by treatment with the cleavage cocktail [TFA/
TIS/H2O (95:2.5:2.5)] during 1, 2, 3, and 24 h. Filtrates
after each treatment were analyzed by HPLC and HPLC-
MS. No cleavage from the resin was detected. Also, no
change in the IR spectra of any of the residual resins was
detected. Once the stability of the linker anchored to the
resin was validated, the scope and limitations of this new
resin in comparison with a standard Wang resin were
examined. Thus, three model peptides were synthesized
by the Fmoc/tBu strategy: Leu-enkephalin (H-Tyr-Gly-
Gly-Phe-Leu-OH) 6, RGD pentapetide (H-Arg-Gly-Asp-
Gly-Trp-OH) 7, and Indolicidin (H-Ile-Leu-Pro-Trp-Lys-
Trp-Pro-Trp-Trp-Pro-Trp-Arg-Arg-OH) 8.
successful conversion of 5 (νN = 2092 cmꢀ1).
3
The first Fmoc-amino acid was incorporated into both
resins using the symmetrical anhydride in the presence of
catalytic amounts of DMAP, while DIPCDI-OxymaPure
was used for the rest of the residues.14 The symmetrical
anhydride was a satisfactory method for introducing
the first amino acid on the OH-BTL resin (Table 1),
as demonstrated by means of UV analysis of the
Scheme 1. Synthesis of 5-Ethynyl-2,4-dimethoxy-
benzaldehyde (4)a
(7) Lober, S.; Rodiguez-Loaiza, P.; Gmeiner, P. Org. Lett. 2003, 5,
1753.
(8) Azide resins, which are prepared in a straightforward manner
from commercially available chloromethyl or amino resins, are more
convenient that alkyne resins for this purpose.
(9) Meng, C. Q.; Ni, L.; Worsencroft, K. J.; J. Ye, Z.; Weingarten,
M. D.; Simpson, J. E.; Skudlarek, J. W.; Marino, E. M.; Suen, K.;
Kunsch, C.; Souder, A.; Howard, R. B. J. Med. Chem. 2007, 50, 1304.
(10) Esser, B.; Bandyopadhyay, A.; Rominger, F.; Gleiter, R.
Chem.;Eur. J. 2009, 15, 3368.
(11) (a) Kumar, S. J. Phys. Chem. 2010, 114, 11395. (b) Riente, P.
Org. Lett. 2012, 14, 3668. Janout, V.; Jing, B.; Regen, S. L. Bioconjugate
Chem. 2002, 13 (2), 356.
a (i) ICl (1.2 equiv), MeOH, 3 h, rt; (ii) Pd(PPh3)4 (0.03 equiv), CuI
(0.005 equiv), Piperidine (1.9 equiv), TMSA (1.2 equiv), THF, 1 h, rt; (iii)
K2CO3 (2 equiv), DCM/MeOH (9:1), 1 h, rt.
(12) Punna, S.; Finn, M. G. Synlett 2004, 1, 99.
(13) Turner, R., A.; Oliver, A., G.; Lokey, R., S. Org. Lett. 2007, 9,
5011.
(5) Metaferia, B. B.; Rittler, M.; Gheeya, J. S.; Lee, A.; Hempel, H.;
Plaza, A.; Stetler-Stevenson, G.; Bewley, C. A.; Khan, J. Bioorg. Med.
Chem. Lett. 2010, 20, 7337.
ꢁ
(6) Cornier, P. G.; Boggain, D. B.; Mata, E. G. Tetrahedron Lett.
2012, 53, 632.
(14) Subiros-Funosa, R.; Prohens, R.; Barbas, R.; El-Faham, A.;
Albericio, F. Chem.;Eur. J. 2009, 15, 9394.
Org. Lett., Vol. 15, No. 2, 2013
247

Scheme 3. Click Chemistry Reaction on Solid Support and Solid Phase Peptide Synthesis (SPPS)a
a (i) CuBr (1 equiv) in ACN, Na ascorbate (1 equiv), 2,6-lutidine (10 equiv), DIEA (10 equiv), DMF, 6 h, rt; (ii) NaBH4 (3 equiv), THF/MeOH (1:1),
1 h, rt; (iii) SPPS with Fmoc/tBu strategy; (iv) cleavage.
Table 1. Results of Resin Substitution
resin
peptide
yields (%)a,b
OH-BTL
6
7
8
6
7
8
92, 100
82, 42
80, 32
100, 80
78, 36
70, 23
Wang
a Yields calculated by Fmoc UV determination taking as initial
loading 0.9 mmol/g for Wang resin and 0.6 mmol/g, for OH-BTL resin.
b Yield determined by AAA of hydrolyzed peptides attached to resin
(HCl/propionic acid (1:1), PhOH).
Table 2. SPPS Results
Figure 2. HPLC profile comparison of peptide 7 synthesized on
OH-BTL or Wang resin.
%
%
d
resins
cleavagea
peptide
6
yieldb
purityc
%
OH-BTL
I
86
100
83
83
44
48
80
61
35
98.2
97.4
94.0
93.9
95.3
89.8
97.4
77.5
74.7
ꢀ
II
I
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
7
8
II
I
II
II
II
II
Wang
6
7
8
0.8
15.8
11.1
a Cleavage conditions: I, TFA/DCM/TIS/H2O (50:45:2.5:2.5); II,
TFA/TIS/H2O (95:2.5:2.5). b Yields were calculated by weight, taking as
a reference the amount of peptide obtained when the resin was cleaved.
c Purity measured by HPLC at 220 nm. d Byproduct percent correspond-
ing to (M þ 107) þ, detected by HPLC at 220 nm.
piperidineꢀdibenzofulvene adduct.15 The peptide content
of the final peptide resin calculated by amino acid analysis
(AAA) was shown to be superior for the OH-BTL resin for
all three peptides (Table 1).
Two cleavage treatments at different acid concentra-
tions (I and II, Table 2) were tested to release peptides 6ꢀ8
from the resins. Both cocktails rendered the three peptides
with similar yields. The purities of the peptides were
slightly higher with the less acidic cleavage cocktail (I)
Figure 3. HPLC profile comparison of peptide 8 synthesized on
OH-BTL or Wang resin.
(Table 2). Use of a more diluted cleavage cocktail (<10%
of TFA) was not efficient as expected (data not shown).
(15) Ma, Y.; Souveaux, E. Biopolymers 1989, 28, 965.
248
Org. Lett., Vol. 15, No. 2, 2013

Scheme 4. Cleavage of H-Trp-OH on Wang Resina
a Cleavage conditions: II, TFA/TIS/H2O (95:2.5:2.5). Side products:
(A) alkylated Trp derivative; (B) undesired cleavage product.
Figure 4. HPLC profiles comparison of H-Trp-OH synthesized
on OH-BTL or Wang resin. “A” corresponds to alkylated Trp
derivative; “B” corresponds to undesired cleavage product.
In conclusion, here we developed a new acid-labile resin
for the preparation of acid peptides with two unique
features: methoxy groups as the only activating groups of
the phenyl ring and a CuAAC to anchor the linker to the
solid support. OH-BTL-resin was more acid-stable and
more efficient for SPPS than Wang resin. Using this new
resin, three model peptides were obtained with higher
yields and purities when compared with those achieved
with Wang resin. The final proof of concept of the super-
iority of OH-BTL-resin was obtained after cleavage of Trp
from resins. The same concept, methoxy groups for acti-
vating and CuAAC for anchoring, is being extended to
other alkoxybenzyl resins such as Rink and BAL.
For all syntheses, the yields with OH-BTL resin were
higher than those with Wang resin. These two resins
showed similar purities for peptide 6. However, even for
this peptide, which did not contain Trp, the crude product
obtained with the Wang resin showed a peak of (M þ 107)þ
(0.8%), as determined by HPLC analysis. This peak was
probably caused by the 4-hydroxymethyl p-hydroxybenzyl
ester analog (see Supporting Information). In the case of
the RGD pentapeptide 7 and indolicidin 8, the purities on
OH-BTL were higher than those on Wang resin. The
HPLC and HPLC-MS analysis show that impurities
on the latter resin were mainly due to Trp alkylation
(Table 2), in both cases showing the peak corresponding to
(M þ 107)þ (Figures 2, 3).
Acknowledgment. This work was partially supported
by Fellowship Marie Curie Initial Training Networks
(ITN) MEMTIDE Project: FP7-PEOPLE_ITN08, CICYT
(CTQ2009-07758), and the Generalitat de Catalunya
(2009SGR 1024).
In order to prove the absence of side reactions on OH-
BTL resin, the Fmoc-Trp-OH was coupled on both resins
(OH-BTL and Wang). Fmoc removal and cleavage under
the acid conditions II [TFA/TIS/H2O (95:2.5:2.5)] in both
resins rendered H-Trp-OH at 3.6 min with purities of 100
and 73.6% for OH-BTL and Wang resin respectively
(Figure 4). In Wang resin, two peaks at 5.6 and 6.3 min
corresponded to a back-alkylated product (peak A,
Figure 4) and to an ester caused by the undesired cleavage
product respectively (peak B, Figure 4) (Scheme 4).
Supporting Information Available. Additional informa-
tion regarding the synthesis and characterization of or-
ganic compounds by NMR, IR, HPLC, HPLC-MS, and
HRMS. This material is available free of charge via the
Internet at http://pubs.acs.org.
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 2, 2013
249